plant microbe interaction research paper

Microbial Diversity, Interventions and Scope pp 23–49 Cite as

Recent Advances in Plant-Microbe Interaction

• Jayakumar Pathma 4 ,
• Gurusamy Raman 5 ,
• Rajendiran Kamaraj Kennedy 6 &
• Laxman Sonawane Bhushan 7  
• First Online: 26 June 2020

820 Accesses

2 Citations

The association of plants and microbes has begun since their evolution. Microbes and plants have coevolved and interacted with each other to meet their demands. Their relationship might be cordial symbiotic as in case of interaction between plants and beneficial microbes or detrimental as in case of interaction between plants and phytopathogens. Numerous genera of microbes are known to be associated with the plants and their rhizosphere. The interaction among these diverse microbial communities and their ability to excel the competition decides the overall plant health. In the past decades, agricultural microbiologists had given more emphasis to plant growth-promoting rhizosphere microbes and soilborne phytopathogens and their interactions, which has resulted in the identification and use of promising microbial strains with biocontrol and biofertilizing properties. With recent advancement in molecular diagnostics, it is evidenced that in addition to rhizosphere microbes, the interactions between plant microbiomes, viz. epiphytes and endophytes, colonizing the entire plant and the plant genome (holobiont) significantly affect the fitness of the plant. Scientific studies evidence that the plant genotype, biostage, soil biogeochemistry and microbe-microbe interaction decide the nature of associated microbiomes. Recent research shows that artificial inoculation of beneficial microbiomes instead of a single or a consortium of microbial strains would improve the success rate of establishment and functioning of the introduced microbial community. This chapter highlights the recent advancements in plant-microbe interaction and ways it could be explored and exploited to enhance plant health, thereby improving crop production qualitatively and quantitatively supporting sustainable agriculture.

This is a preview of subscription content, log in via an institution .

Buying options

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Agler MT, Ruhe J, Kroll S et al (2016) Microbial hub taxa link host and abiotic factors to plant microbiome variation. PLoS Biol 14:e1002352

Article   PubMed   PubMed Central   CAS   Google Scholar  

Ali SZ, Sandhya V, Grover M et al (2009) Pseudomonas sp. strain AKM-P6 enhances tolerance of sorghum seedlings to elevated temperatures. Biol Fertil Soils 46:45–55

Article   CAS   Google Scholar  

Ali SZ, Sandhya V, Grover M et al (2011) Effect of inoculation with a thermotolerant plant growth promoting Pseudomonas putida strain AKMP7 on growth of wheat ( Triticum spp.) under heat stress. J Plant Interact 6:239–246

Ardebili ZO, Ardebili NO, Hamdi SMM (2011) Physiological effects of Pseudomonas fluorescens CHA0 on tomato ( Lycopersicon esculentum Mill.) plants and its possible impact on Fusarium oxysporum f. sp. lycopersici . Aust J Crop Sci 5:1631–1638

CAS   Google Scholar  

Armanhi JSL, de Souza RSC, Damasceno NDB et al (2018) A community-based culture collection for targeting novel plant growth-promoting bacteria from the sugarcane microbiome. Front Plant Sci 8:2191

Article   PubMed   PubMed Central   Google Scholar  

Bacilio-Jimenez M, Aguilar-Flores S, Ventura-Zapata E et al (2003) Chemical characterization of root exudates from rice (Oryza sativa) and their effects on the chemotactic response of endophytic bacteria. Plant Soil 249:271–277

Badri DV, Chaparro JM, Zhang R et al (2013) Application of natural blends of phytochemicals derived from the root exudates of arabidopsis to the soil reveal that phenolic related compounds predominantly modulate the soil microbiome. J Biol Chem 288:4502–4512

Article   CAS   PubMed   PubMed Central   Google Scholar  

Bai Y, Muller DB, Srinivas G et al (2015) Functional overlap of the Arabidopsis leaf and root microbiota. Nature 528:364–369

Article   CAS   PubMed   Google Scholar  

Barka EA, Nowak J, Clement C (2006) Enhancement of chilling resistance of inoculated grapevine plantlets with a plant growth-promoting rhizobacterium, Burkholderia phytofirmans strain PsJN. Appl Environ Microbiol 72:7246–7252

Baudoin E, Benizri E, Guckert A (2002) Impact of growth stage on the bacterial community structure along maize roots, as determined by metabolic and genetic fingerprinting. Appl Soil Ecol 19:135–145

Article   Google Scholar  

Berendsen RL, Pieterse CM, Bakker PA (2012) The rhizosphere microbiome and plant health. Trends Plant Sci 17:478–486

Berg G (2009) Plant – microbe interactions promoting plant growth and health: perspectives for controlled use of microorganisms in agriculture. Appl Microbiol Biotechnol 84:11–18

Berg G, Martina K, Daria R et al (2017) Plant microbial diversity is suggested as the key to future biocontrol and health trends. FEMS Microbiol Ecol 93:1–9

Bharti N, Pandey SS, Barnawal D et al (2016) Plant growth promoting rhizobacteria Dietzia natronolimnaea modulates the expression of stress responsive genes providing protection of wheat from salinity stress. Sci Rep 6:34768

Bitocchi E, Bellucci E, Giardini A et al (2013) Molecular analysis of the parallel domestication of the common bean ( Phaseolus vulgaris ) in Mesoamerica and the Andes. New Phytol 197:300–313

Bodenhausen N, Bortfeld-Miller M, Ackermann M et al (2014) A synthetic community approach reveals plant genotypes affecting the phyllosphere microbiota. PLoS Genet 10:e1004283

Bouffaud ML, Poirier MA, Mulle D et al (2014) Root microbiome relates to plant host evolution in maize and other Poaceae. Environ Microbiol 16:2804–2814

Article   PubMed   Google Scholar  

Boughton AJ, Hoover K, Felton GW (2006) Impact of chemical elicitor applications on greenhouse tomato plants and population growth of the green peach aphid, Myzus persicae . Entomol Exp Appl 120:175–188

Bowen G (1969) Nutrient status effects on loss of amides and amino acids from pine roots. Plant Soil 30:139–142

Bradacova K, Weber NF, Morad-Talab N et al (2016) Micronutrients (Zn/Mn), seaweed extracts, and plant growth-promoting bacteria as cold-stress protectants in maize. Chem Biol Technol Agric 3:19

Brundrett MC (2002) Coevolution of roots and mycorrhizas of land plants. New Phytol 154:275–304

Bulgarelli D, Rott M, Schlaeppi K et al (2012) Revealing structure and assembly cues for Arabidopsis root-inhabiting bacterial microbiota. Nature 488:91–95

Bulgarelli D, Schlaeppi K, Spaepen S et al (2013) Structure and functions of the bacterial microbiota of plants. Annu Rev Plant Biol 64:807–838

Bulgarelli D, Garrido-Oter R, Munch PC et al (2015) Structure and function of the bacterial root microbiota in wild and domesticated barley. Cell Host Microbe 17:392–403

Burd GI, Dixon DG, Glick BR (1998) A plant growth-promoting bacterium that decreases nickel toxicity in seedlings. Appl Environ Microbiol 64:3663–3668

Buyer JS, Roberts DP, Cohen RE (1999) Microbial community structure and function in the spermosphere as affected by soil and seed type. Can J Microbiol 45:138–144

Cardinale M, Grube M, Erlacher A et al (2015) Bacterial networks and co-occurrence relationships in the lettuce root microbiota. Environ Microbiol 17:239–252

Chaparro JM, Badri DV, Vivanco JM et al (2014) Rhizosphere microbiome assemblage is affected by plant development. ISME J 8:790–803

Chen HY, Huh JH, Yu YC et al (2015) The Arabidopsis vacuolar sugar transporter SWEET2 limits carbon sequestration from roots and restricts Pythium infection. Plant J83:1046–1058

Google Scholar  

Chisholm ST, Coaker G, Day B et al (2006) Host-microbe interactions: shaping the evolution of the plant immune response. Cell 124:803–814

Coltharp C, Xiao J (2012) Super resolution microscopy for microbiology. Cell Microbiol 14:1808–1818

Copeland JK, Yuan L, Layeghifard M et al (2015) Seasonal community succession of the phyllosphere microbiome. Mol Plant-Microbe Interact 28:274–285

Costa R, Gotz M, Mrotzek N et al (2006) Effects of site and plant species on rhizosphere community structure as revealed by molecular analysis of microbial guilds. FEMS Microbiol Ecol 56:236–249

Davidson IA, Robson MJ (1986) Effect of contrasting patterns of nitrate application on the nitrate uptake, N 2 -fixation, nodulation and growth of white clover. Ann Bot 57:331–338

De Werra P, Baehler E, Huser A et al (2008) Detection of plant-modulated alterations in antifungal gene expression in Pseudomonas fluorescens CHA0 on roots by flow cytometry. Appl Environ Microbiol 74:1339–1349

Article   PubMed   CAS   Google Scholar  

Dixon RA (1995) Stress-induced phenylpropanoid metabolism. Plant Cell 7:1085–1097

Downing KJ, Leslie G, Thompson JA (2000) Biocontrol of the sugar-cane borer Eldana saccharina by expression of the Bacillus thuringiensis cry1Ac7 and Serratia marcescens chiA gene in sugar-cane associated bacteria. Appl Environ Microbiol 66:2804–2810

Ebert D, Hamilton WD (1996) Sex against virulence: the coevolution of parasitic diseases. Trends Ecol Evol 11(2):79–82

Edwards CA, Burrows I (1988) The potential of earthworm composts as plant growth media. In: Edwards CA, Neuhauser E (eds) Earthworms in waste and environmental management. SPB Academic Press, The Hague, pp 21–32

Ehrenreich A (2006) DNA microarray technology for the microbiologist: an overview. Appl Microbiol Biotechnol 73:255–273

Erkel C, Kube M, Reinhardt R, Liesack W (2006) Genome of rice cluster I archaea – the key methane producers in the rice rhizosphere. Science 313:370–372

Farre-Armengol G, Filella I, Llusia J et al (2016) Bidirectional interaction between phyllospheric microbiotas and plant volatile emissions. Trends Plant Sci 21:854–860

Fierer N, Jackson RB (2006) The diversity and biogeography of soil bacterial communities. Proc Natl Acad Sci 103:626–631

Fierer N, Ladau J, Clemente JC et al (2013) Reconstructing the microbial diversity and function of pre-agricultural tall grass prairie soils in the United States. Science 342:621–624

Fisher MM, Triplett EW (1999) Automated approach for ribosomal intergenic spacer analysis of microbial diversity and its application to freshwater bacterial communities. Appl Environ Microbiol 65:4630–4636

Fitzpatrick CR, Copeland J, Wang PW et al (2018) Assembly and ecological function of the root microbiome across angiosperm plant species. Proc Natl Acad Sci 115:1157–1165

Gans J, Woilinsky M, Dunbar J (2005) Computational improvements reveal great bacterial diversity and high metal toxicity in soil. Science 309:1387–1390

Garcia-Palacios P, Milla R, Delgado-Baquerizo M et al (2013) Side-effects of plant domestication: ecosystem impacts of changes in litter quality. New Phytol 198:504–513

Germida JJ, Siciliano SD (2001) Taxonomic diversity of bacteria associated with the roots of modern, recent and ancient wheat cultivars. Biol Fertil Soils 33:410–415

Girvan MS, Bullimore J, Pretty JN et al (2003) Soil type is the primary determinant of the composition of the total and active bacterial communities in arable soils. Appl Environ Microbiol 69:1800–1809

Glazebrook J (2005) Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu Rev Phytopathol 43:205–227

Glick BR (1995) The enhancement of plant growth by free-living bacteria. Can J Microbiol 41:109–117

Gopal M, Gupta A, Thomas GV (2013) Bespoke microbiome therapy to manage plant diseases. Front Microbiol 4:355

Grayston SJ, Wang SQ, Campbell CD et al (1998) Selective influence of plant species on microbial diversity in the rhizosphere. Soil Biol Biochem 30:369–378

Grover M, Ali SZ, Sandhya V et al (2011) Role of microorganisms in adaptation of agriculture crops to abiotic stresses. World J Microbiol Biotechnol 27:1231–1240

Hacquard S, Garrido-Oter R, Gonzalez A et al (2015) Microbiota and host nutrition across plant and animal kingdoms. Cell Host Microbe 17:603–616

Haichar FE, Marol C, Berge O et al (2008) Plant host habitat and root exudates shape soil bacterial community structure. ISME J 2:1221–1230

Handelsman J (2004) Metagenomics: application of genomics to uncultured microorganisms. Microbiol Mol Biol Rev 68:669–685

Hartmann A, Rothballer M, Schmid M (2007) Lorenz Hiltner, a pioneer in rhizosphere microbial ecology and soil bacteriology research. Plant Soil 312:7–14

Hartmann M, Niklaus PA, Zimmermann S et al (2014) Resistance and resilience of the forest soil microbiome to logging-associated compaction. ISME J 8:226–244

Hassan S, Mathesius U (2012) The role of flavonoids in root-rhizosphere signalling: opportunities and challenges for improving plant-microbe interactions. J Exp Bot 63:3429–3444

Haudry A, Cenci A, Ravel C et al (2007) Grinding up wheat: a massive loss of nucleotide diversity since domestication. Mol Biol Evol 24:1506–1517

Hetrick BAD, Wilson GWT, Cox TS (1993) Mycorrhizal dependence of modern wheat cultivars and ancestors: a synthesis. Can J Bot 71(3):512–518

Ho YN, Hsieh JL, Huang CC (2013) Construction of a plant–microbe phytoremediation system: combination of vetiver grass with a functional endophytic bacterium, Achromobacter xylosoxidans F3B, for aromatic pollutants removal. Bioresour Technol 145:43–47

Ho YN, Mathew DC, Huang C (2017) Plant-microbe ecology: interactions of plants and symbiotic microbial communities. In: Yousaf Z (ed) Plant ecology – traditional approaches to recent trends. https://doi.org/10.5772/intechopen.69088

Horner NR, Grenville-Briggs LJ, van West P (2012) The oomycete Pythium oligandrum expresses putative effectors during mycoparasitism of Phytophthora infestans and is amenable to transformation. Fungal Biol 116:24–41

Horner-Devine M, Lage M, Hughes J et al (2004) A taxa-area relationship for bacteria. Nature 432:750–753

Houlden A, Timms-Wilson TM, Day MJ et al (2008) Influence of plant developmental stage on microbial community structure and activity in the rhizosphere of three field crops. FEMS Microbiol Ecol 65:193–201

Hugerth LW, Andersson AF (2017) Analysing microbial community composition through amplicon sequencing: from sampling to hypothesis testing. Front Microbiol 8:1561

Husain S, McKeen W (1963) Interactions between strawberry roots and Rhizoctonia fragariae. Phytopathology 53:541–545

Inceoglu O, Al-Soud WA, Salles JF et al (2011) Comparative analysis of bacterial communities in a potato field as determined by pyrosequencing. PLoS One 6:e23321

Jacobs S, Zechmann B, Molitor A et al (2011) Broad-spectrum suppression of innate immunity is required for colonization of Arabidopsis roots by the fungus Piriformospora indica . Plant Physiol 156:726–740

James EK, Gyaneshwar P, Mathan N et al (2002) Infection and colonization of rice seedlings by the plant growth-promoting bacterium Herbaspirillum seropedicae Z67. Mol Plant-Microbe Interact 15:894–906

Jha Y, Subramanian R, Patel S (2011) Combination of endophytic and rhizospheric plant growth promoting rhizobacteria in Oryza sativa shows higher accumulation of osmoprotectant against saline stress. Acta Physiol Plant 33:797–802

Jin CW, Ye YQ, Zheng SJ (2014) An underground tale: contribution of microbial activity to plant iron acquisition via ecological processes. Ann Bot 113:7–18

Kapulnik Y, Kushnir U (1991) Growth dependency of wild, primitive and modern cultivated wheat lines on vesicular–arbuscular mycorrhiza fungi. Euphytica 56:27–36

Katznelson H, Rouatt J, Payne T (1955) The liberation of amino acids and reducing compounds by plant roots. Plant Soil 7:35–48

Kempster VN, Scott ES, Davies KA (2002) Evidence for systemic, cross-resistance in white clover ( Trifolium repens ) and annual medic ( Medicago truncatula var. truncatula) induced by biological and chemical agents. Biocontrol Sci Tech 12:615–623

Kim DH, Kaashyap M, Rathore A et al (2014) Phylogenetic diversity of Mesorhizobium in chickpea. J Biosci 39:513–517

Kloepper JW, Ryu CM (2006) Bacterial endophytes as elicitors of induced systemic resistance. In: Schulz BJE, Boyle CJC, Sieber TN (eds) Microbial root endophytes. Springer, Berlin/Heidelberg, pp 33–52

Chapter   Google Scholar  

Kravchenko LV, Azarova TS, Shaposhnikov AI et al (2003) Root exudates of tomato plants and their effect on the growth and antifungal activity of Pseudomonas strains. Microbiology 72:37–41

Kyselkova M, Kopecky J, Frapolli M et al (2009) Comparison of rhizobacterial community composition in soil suppressive or conducive to tobacco black root rot disease. ISME J 3:1127–1138

Lanoue A, Burlat V, Henkes GJ et al (2010) De novo biosynthesis of defense root exudates in response to Fusarium attack in barley. New Phytol 185:577–588

Lebeis SL, Paredes SH, Lundberg DS et al (2015) Salicylic acid modulates colonization of the root microbiome by specific bacterial taxa. Science 349:860–864

Lee HC, Lai K, Lorenc MT et al (2012) Bioinformatics tools and databases for analysis of next-generation sequence data. Brief Funct Genomics 11:12–24

Lehmann A, Barto EK, Powell JR et al (2012) Mycorrhizal responsiveness trends in annual crop plants and their wild relatives – a meta-analysis on studies from 1981 to 2010. Plant Soil 355:231–250

Leveau JHJ (2007) The magic and menace of metagenomics: prospects for the study of plant growth-promoting rhizobacteria. Eur J Plant Pathol 119:279–300

Levy A, Salas Gonzalez I, Mittelviefhaus M et al (2018) Genomic features of bacterial adaptation to plants. Nat Genet 50:138

Lindow SE, Brandl MT (2003) Microbiology of the phyllosphere. Appl Environ Microbiol 69:1875–1883

Liu WT, Marsh TL, Cheng H et al (1997) Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Appl Environ Microbiol 63:4516–4522

Lozupone CA, Stombaugh JI, Gordon JI et al (2012) Diversity, stability and resilience of the human gut microbiota. Nature 489:220–230

Lukow T, Dunfield PF, Liesack W (2000) Use of the T-RFLP technique to assess spatial and temporal changes in the bacterial community structure within an agricultural soil planted with transgenic and non-transgenic potato plants. FEMS Microbiol Ecol 32:241–247

Lundberg DS, Lebeis SL, Paredes SH et al (2012) Defining the core Arabidopsis thaliana root microbiome. Nature 488:86–90

Marasco R, Rolli E, Ettoumi B et al (2012) A drought resistance-promoting microbiome is selected by root system under desert farming. PLoS One 7:e48479

Mark GL, Dow JM, Kiely PD et al (2005) Transcriptome profiling of bacterial responses to root exudates identifies genes involved in microbe–plant interactions. Proc Natl Acad Sci 102:17454–17459

Marques JM, da Silva TF, Vollu RE et al (2014) Plant age and genotype affect the bacterial community composition in the tuber rhizosphere of field-grown sweet potato plants. FEMS Microbiol Ecol 88:424–435

Marschner P, Crowley D, Lieberei R (2001) Arbuscular mycorrhizal infection changes the bacterial 16S rDNA community composition in the rhizosphere of maize. Mycorrhiza 11:297–302

Mathew DC, Ho Y-N, Gicana RG et al (2015) A rhizosphere-associated symbiont, Photobacterium spp. strain MELD1, and its targeted synergistic activity for phytoprotection against mercury. PLoS One 10:e0121178

Matthews A, Pierce S, Hipperson H et al (2019) Rhizobacterial community assembly patterns vary between crop species. Front Microbiol 10:581

Mayak S, Tirosh T, Glick BR (2004) Plant growth-promoting bacteria that confer resistance to water stress in tomatoes and peppers. Plant Sci 166:525–530

Mendes R, Kruijt K, de Bruijn I et al (2011) Deciphering the rhizosphere microbiome for disease-suppressive bacteria. Science 332:1097–1100

Mendes LW, Kuramae EE, Navarrete AA et al (2014) Taxonomical and functional microbial community selection in soybean rhizosphere. ISME J 8:1577–1587

Micallef SA, Channer S, Shiaris MP et al (2009) Plant age and genotype impact the progression of bacterial community succession in the Arabidopsis rhizosphere. Plant Signal Behav 4:777–780

Mougel C, Offre P, Ranjard L et al (2006) Dynamic of the genetic structure of bacterial and fungal communities at different developmental stages of Medicago truncatula Gaertn. cv. Jemalong line J5. New Phytol 170:165–175

Mueller UG, Sachs JL (2015) Engineering microbiomes to improve plant and animal health. Trends Microbiol 10:606–617

Mutch LA, Young JP (2004) Diversity and specificity of Rhizobium leguminosarum biovar viciae on wild and cultivated legumes. Mol Ecol 13:2435–2444

Muyzer G, de Waal EC, Uitterlinden AG et al (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59:695–700

Naeem M, Aslam Z, Khaliq A et al (2018) Plant growth promoting rhizobacteria reduce aphid population and enhance the productivity of bread wheat. Braz J Microbiol 49:9–14

Nemergut DR, Knelman JE, Ferrenberg S et al (2016) Decreases in average bacterial community rRNA operon copy number during succession. ISME J 10:1147–1156

Ofek M, Voronov-Goldman M, Hadar Y et al (2014) Host signature effect on plant root-associated microbiomes revealed through analyses of resident vs. active communities. Environ Microbiol 16:2157–2167

Oldroyd GED, Harrison MJ, Paszkowski U (2009) Reprogramming plant cells for endosymbiosis. Science 324:753–754

Osbourn AE, Barber CE, Daniels MJ (1987) Identification of plant-induced genes of the bacterial pathogen Xanthomonas campestris pathovar campestris using a promoter-probe plasmid. EMBO J 6:23–28

Pace NR, Stahl DA, Lane DJ et al (1985) Analyzing natural microbial populations by rRNA sequences. ASM News 51:4–12

Panstruga R, Kuhn H (2015) Introduction to a virtual special issue on cell biology at the plant–microbe interface. New Phytol 207:931–938

Parniske M (2008) Arbuscular mycorrhiza: the mother of plant root endosymbioses. Nat Rev Microbiol 6:763–775

Pathma J, Sakthivel N (2013) Molecular and functional characterization of bacteria isolated from straw and goat manure based vermicompost. Appl Soil Ecol 70:33–47

Pathma J, Rahul GR, Kamaraj Kennedy R, Subashri R, Sakthivel N (2011) Secondary metabolite production by bacterial antagonists. J Biol Control 25:165–181

Pathma J, Raman G, Sakthivel N (2019) Microbiome of rhizospheric soil and vermicompost and their applications in soil fertility, pest and pathogen management for sustainable agriculture. In: Panpatte DG, Jhala YK (eds) Soil fertility management for sustainable development. Springer Nature Singapore, Singapore, pp 189–210

Peiffer JA, Spor A, Koren O et al (2013) Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proc Natl Acad Sci 110:6548–6553

Pel MJ, Pieterse CM (2013) Microbial recognition and evasion of host immunity. J Exp Bot 64:1237–1248

Penrose DM, Glick BR (2002) Methods for isolating and characterizing ACC deaminase containing plant growth promoting rhizobacteria. Physiol Plant 118:10–15

Poza-Carrion C, Suslow T, Lindow S (2013) Resident bacteria on leaves enhance survival of immigrant cells of Salmonella enterica . Phytopathology 103:341–351

Quecine MC, Araujo WL, Tsui S et al (2014) Control of Diatraea saccharalis by the endophytic Pantoea agglomerans 33.1 expressing cry1Ac7. Arch Microbiol 196:227–234

Rainey PB, Heithoff DM, Mahan MJ (1997) Single-step conjugative cloning of bacterial gene fusions involved in microbe-host interactions. Mol Gen Genet 256:84–87

Ram SG, Thiruvengadam V, Vinod KK (2007) Genetic diversity among cultivars, landraces and wild relatives of rice as revealed by microsatellite markers. J Appl Genet 48:337–345

Ramirez KS, Craine JM, Fierer N (2012) Consistent effects of nitrogen amendments on soil microbial communities and processes across biomes. Glob Chang Biol 18:1918–1927

Ramirez-Flandes S, Gonzalez B, Ulloa O (2019) Redox traits characterize the organization of global microbial communities. Proc Natl Acad Sci 116:3630–3635

Raynaud X, Nunan N (2014) Spatial ecology of bacteria at the microscale in soil. PLoS One 9:e87217

Rediers H, Rainey PB, Vanderleyden J et al (2005) Unravelling the secret lives of bacteria: use of in vivo expression technology and differential fluorescence induction promoter traps as tools for exploring niche-specific gene expression. Microbiol Mol Rev 69:217–261

Roberts E, Lindow S (2014) Loline alkaloid production by fungal endophytes of fescue species select for particular epiphytic bacterial microflora. ISME J 8:359–368

Rodrigues JL, Pellizari VH, Mueller R et al (2013) Conversion of the Amazon rainforest to agriculture results in biotic homogenization of soil bacterial communities. Proc Natl Acad Sci 110:988–993

Roesch LFW, Fulthorpe RR, Riva A et al (2007) Pyrosequencing enumerates and contrasts soil microbial diversity. ISME J 1:283–290

Rolli E, Marasco R, Vigani et al (2015) Improved plant resistance to drought is promoted by the root-associated microbiome as a water stress-dependent trait. Environ Microbial 17:316–331

Rosenblueth M, Martinez-Romero E (2006) Bacterial endophytes and their interactions with hosts. Mol Plant-Microbe Interact 19:827–837

Rosenzweig N, Tiedje JM, Quensen JF et al (2012) Microbial communities associated with potato common scab-suppressive soil determined by pyrosequencing analyses. Plant Dis 96:718–725

Sanchez-Lopez A, Pintelon I, Stevens V et al (2018) Seed endophyte microbiome of Crotalaria pumila unpeeled: identification of plant-beneficial methylobacteria. Int J Mol Sci 19:291

Article   PubMed Central   CAS   Google Scholar  

Sangabriel-Conde W, Maldonado-Mendoza IE, Mancera-Lopez ME et al (2015) Glomeromycota associated with Mexican native maize landraces in Los Tuxtlas, Mexico. Appl Soil Ecol 87:63–71

Sanguin H, Remenant B, Dechesne A et al (2006) Potential of a 16S rRNA-based taxonomic microarray for analyzing the rhizosphere effects of maize on Agrobacterium spp. and bacterial communities. Appl Environ Microbiol 72:4302–4312

Santos-Medellin C, Edwards J, Liechty Z et al (2017) Drought stress results in a compartment-specific restructuring of the rice root-associated microbiomes. MBio 8:e00764-17

Saravanakumar D, Kavino M, Raguchander T et al (2011) Plant growth promoting bacteria enhance water stress resistance in green gram plants. Acta Physiol Plant 33:203–209

Schenk PM, Carvalhais LC, Kazan K (2012) Unravelling plant–microbe interactions: can multi-species transcriptomics help? Trends Biotechnol 30:177–184

Scheuring I, Yu DW (2013) How to assemble a beneficial microbiome in three easy steps. Ecol Lett 15:1300–1307

Sen K, Sengupta C, Saha J (2014) PGPR consortium alleviating downy mildew of cucumber. Int J Plant Anim Environ Sci 4:150–159

Sessitsch A, Weilharter A, Gerzabek MH et al (2001) Microbial population structures in soil particle size fractions of a long-term fertilizer field experiment. Appl Environ Microbiol 67:4215–4224

Shade A, Handelsman J (2012) Beyond the venn diagram: the hunt for the core microbiome. Environ Microbiol 14:4–12

Shang Q, Yang G, Wang Y et al (2016) Illumina-based analysis of the rhizosphere microbial communities associated with healthy and wilted Lanzhou lily ( Lilium davidii var. unicolor) plants grown in the field. WJMB 32:95

Sindhu SS, Rakshiya YS, Verma MK (2011) Biological control of termites by antagonistic soil microorganisms. In: Singh A, Parmar N, Kuhad RC et al (eds) Bioaugmentation, biostimulation and biocontrol, soil biology. Springer, Berlin/Heidelberg, pp 261–310

Singh P, Gobbi A, Santoni S et al (2018) Assessing the impact of plant genetic diversity in shaping the microbial community structure of Vitis vinifera phyllosphere in the Mediterranean. Front Life Sci 11:35–46

Souza RC, Mendes IC, Reis-Junior FB et al (2016) Shifts in taxonomic and functional microbial diversity with agriculture: how fragile is the Brazilian Cerrado? BMC Microbiol 16:42

Staley JT, Konopka A (1985) Measurements of in situ activities of non-photosynthetic microorganisms in aquatic and terrestrial habitats. Annu Rev Microbiol 39:321–346

Stanley HF, Fagan WF (2002) Fungal endophytes: common host plant symbionts but uncommon mutualists. Integr Comp Biol 42:360–368

Stout MJ, Zehnder GW, Baur ME (2002) Potential for the use of elicitors of plant defense in arthropod management programs. Arch Insect Biochem 51:222–235

Subramanian P, Kim K, Krishnamoorthy R et al (2016) Cold stress tolerance in psychrotolerant soil bacteria and their conferred chilling resistance in tomato ( Solanum lycopersicum Mill.) under low temperatures. PLoS One 11:e0161592

Sugiyama A, Ueda Y, Zushi T et al (2014) Changes in the bacterial community of soybean rhizospheres during growth in the field. PLoS One 9:1–9

Szoboszlay M, White-Monsant A, Moe LA et al (2016) The effect of root exudate 7,4′-dihydroxyflavone and naringenin on soil bacterial community structure. PLoS One 11:e0146555

Tan Y, Cui Y, Li H et al (2017) Rhizospheric soil and root endogenous fungal diversity and composition in response to continuous Panax notoginseng cropping practices. Microbiol Res 194:10–19

Tank N, Saraf M (2010) Salinity-resistant plant growth promoting rhizobacteria ameliorates sodium chloride stress on tomato plants. J Plant Interact 5:51–58

Timonen S, Sinkko H, Sun H et al (2017) Ericoid roots and mycospheres govern plant-specific bacterial communities in boreal forest humus. Microb Ecol 73:939–953

Toljander JF, Santos-Gonzalez JC, Tehler A et al (2008) Community analysis of arbuscular mycorrhizal fungi and bacteria in the maize mycorrhizosphere in a long-term fertilization trial. FEMS Microbiol Ecol 65:323–338

Trda L, Boutrot F, Claverie J et al (2015) Perception of pathogenic or beneficial bacteria and their evasion of host immunity: pattern recognition receptors in the frontline. Front Plant Sci 6:219

Tringe SG, Mering CV, Kobayashi A (2005) Comparative metagenomics of microbial communities. Science 308:554–557

Tunlid A, Baird BH, Trexler MB et al (1985) Determination of phospholipid ester-linked fatty acids and poly-b-hydroxybutyrate for the estimation of bacterial biomass and activity in the rhizosphere of the rapa plant Brassica napus . Can J Microbiol 31:1113–1119

Turner TR, James EK, Poole PS (2013a) The plant microbiome. Genome Biol 14:209

Turner T, Ramakrishnan K, Walshaw J et al (2013b) Comparative metatranscriptomics reveals kingdom level changes in the rhizosphere microbiome of plants. ISME J 7:2248–2258

van der Heijden MG, Hartmann M (2016) Networking in the plant microbiome. PLoS Biol 14:e1002378

Van Peer R, Niemann G, Schippers B (1991) Induced resistance and phytoalexin accumulation in biological control of Fusarium wilt of carnation by Pseudomonas sp. strain WCS 417r. Phytopathology 81:728–734

Van Wees SCM, Ent SV, Pieterse CMJ (2008) Plant immune responses triggered by beneficial microbes. Curr Opin Plant Biol 11:443–448

Vandeputte OM, Kiendrebeogo M, Rasamiravaka T et al (2011) The flavanone naringenin reduces the production of quorum sensing-controlled virulence factors in Pseudomonas aeruginosa PAO1. Microbiology 157:2120–2132

Velicer GJ, Raddatz G, Keller H, Deiss S, Lanz C, Dinkelacker I, Schuster SC (2006) Comprehensive mutation identification in an evolved bacterial cooperator and its cheating ancestor. Proc Natl Acad Sci 103:8107–8112

Vellend M, Agrawal A (2010) Conceptual synthesis in community ecology. Q Rev Biol 85:183–206

Verma P, Yadav AN, Khannam KS et al (2015) Assessment of genetic diversity and plant growth promoting attributes of psychrotolerant bacteria allied with wheat ( Triticum aestivum ) from the northern hills zone of India. Ann Microbiol 65:1885–1899

Vestergard M, Hendry F, Castro JIR et al (2008) Rhizosphere bacterial community composition responds to arbuscular mycorrhiza, but not to reductions in microbial activity induced by foliar cutting. FEMS Microbiol Ecol 64:78–89

Vogel C, Bodenhausen N, Gruissem W et al (2016) The Arabidopsis leaf transcriptome reveals distinct but also overlapping responses to colonization by phyllosphere commensals and pathogen infection with impact on plant health. New Phytol 212:192–207

Vogl A (1898) Mehl und die anderen mehl produkte der cerealien und leguminosen. Zeitschrift Nahrungsmittle untersuchung, Hgg Warlenkunde 21:25–29

Vorholt JA (2012) Microbial life in the phyllosphere. Nat Rev Microbiol 10:828–840

Walder F, Schlaeppi K, Wittwer R et al (2017) Community profiling of Fusarium in combination with other plant-associated fungi in different crop species using SMRT sequencing. Front Plant Sci 28:8

Wang G, Xu Y, Jin J et al (2009) Effect of soil type and soybean genotype on fungal community in soybean rhizosphere during reproductive growth stages. Plant Soil 317:135–144

Wang Y, Li C, Tu C et al (2017) Long-term no-tillage and organic input management enhanced the diversity and stability of soil microbial community. Sci Total Environ 609:341–347

Watrous J, Roach P, Alexandrov T et al (2012) Mass spectral molecular networking of living microbial colonies. Proc Natl Acad Sci 109:1743–1752

Weese DJ, Heath KD, Dentinger BT et al (2015) Long-term nitrogen addition causes the evolution of less-cooperative mutualists. Evolution 69:631–642

Weinert N, Piceno Y, Ding GC et al (2011) PhyloChip hybridization uncovered an enormous bacterial diversity in the rhizosphere of different potato cultivars: many common and few cultivar-dependent taxa. FEMS Microbiol Ecol 75:497–506

Willers C, Jansen van Rensburg PJ, Claassens S (2015) Microbial signature lipid biomarker analysis–an approach that is still preferred, even amid various method modifications. J Appl Microbiol 118(6):1251–1263

Woese CR, Fox GE (1977) Phylogenetic structure of the prokaryotic domain: the primary kingdoms. Proc Natl Acad Sci 74:5088–5090

Wu CH, Bernard SM, Andersen GL et al (2009) Developing microbe–plant interactions for applications in plant-growth promotion and disease control, production of useful compounds, remediation and carbon sequestration. Microb Biotechnol 2:428–440

Xing X, Koch AM, Jones AM et al (2012) Mutualism breakdown in breadfruit domestication. Proc Biol Sci 22:1122–1130

Xiong W, Li Z, Liu H et al (2015a) The effect of long-term continuous cropping of black pepper on soil bacterial communities as determined by 454 pyrosequencing. PLoS One 10:e0136946

Xiong W, Zhao Q, Zhao J, Xun W et al (2015b) Different continuous cropping spans significantly affect microbial community membership and structure in a vanilla-grown soil as revealed by deep pyrosequencing. Microb Ecol 70:209–218

Xu Y, Wang G, Jin J et al (2009) Bacterial communities in soybean rhizosphere in response to soil type, soybean genotype, and their growth stage. Soil Biol Biochem 41:919–925

Yan Y, Kuramae EE, de Hollander M, Klinkhamer PG et al (2017) Functional traits dominate the diversity-related selection of bacterial communities in the rhizosphere. ISME J 11:56–66

Yuan Z, Liu P, Saenkham P et al (2008) Transcriptome profiling and functional analysis of Agrobacterium tumefaciens reveals a general conserved response to acidic conditions (pH 5.5) and a complex acid-mediated signalling involved in agrobacterium–plant interactions. J Bacteriol 190:494–507

Zachow C, Muller H, Tilcher R et al (2014) Differences between the rhizosphere microbiome of Beta vulgaris ssp. maritima -ancestor of all beet crops-and modern sugar beets. Front Microbiol 5:415

Zhang J, Subramanian S, Stacey G et al (2009) Flavones and flavonols play distinct critical roles during nodulation of Medicago truncatula by Sinorhizobium meliloti . Plant J 57:171–183

Zhang N, Wang D, Liu Y (2014) Effects of different plant root exudates and their organic acid components on chemotaxis, biofilm formation and colonization by beneficial rhizosphere-associated bacterial strains. Plant Soil 374:689–700

Zhu YG, Smith SE, Barritt AR et al (2001) Phosphorus (P) efficiencies and mycorrhizal responsiveness of old and modern wheat cultivars. Plant Soil 237:249–255

Zolla G, Badri DV, Bakker MG et al (2013) Soil microbiomes vary in their ability to confer drought tolerance to Arabidopsis . Appl Soil Ecol 68:1–9

Download references

Acknowledgement

The authors wish to thank Dr. R. Nagarajaprakash, group leader, Chemical Sciences Research Group, Lovely Professional University, Punjab, India, for his constructive criticism and support in preparing this book chapter.

Author information

Authors and affiliations.

Department of Entomology, School of Agriculture, Division of Research and Development, Lovely Professional University, Phagwara, Punjab, India

Jayakumar Pathma

Department of Life Sciences, Yeungnam University, Gyeongsan, Gyeongsangbuk-do, South Korea

Gurusamy Raman

Department of Biotechnology, St. Joseph’s College, Trichy, Tamil Nadu, India

Rajendiran Kamaraj Kennedy

Department of Entomology, School of Agriculture, Lovely Professional University, Phagwara, Punjab, India

Laxman Sonawane Bhushan

You can also search for this author in PubMed   Google Scholar

Editor information

Editors and affiliations.

School of Bioengineering and Biosciences, Lovely Professional Univeristy, Phagwara, Punjab, India

Shiwani Guleria Sharma

Neeta Raj Sharma

Molecular Genetics Laboratory, Dayanand Medical College and Hospital, Ludhiana, Punjab, India

Mohit Sharma

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Singapore Pte Ltd.

About this chapter

Cite this chapter.

Pathma, J., Raman, G., Kennedy, R.K., Bhushan, L.S. (2020). Recent Advances in Plant-Microbe Interaction. In: Sharma, S., Sharma, N., Sharma, M. (eds) Microbial Diversity, Interventions and Scope. Springer, Singapore. https://doi.org/10.1007/978-981-15-4099-8_2

Download citation

DOI : https://doi.org/10.1007/978-981-15-4099-8_2

Published : 26 June 2020

Publisher Name : Springer, Singapore

Print ISBN : 978-981-15-4098-1

Online ISBN : 978-981-15-4099-8

eBook Packages : Biomedical and Life Sciences Biomedical and Life Sciences (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

• Search Menu
• Special Issues
• Advance articles
• High Impact Research
• Why Publish with AoB?
• Author Guidelines
• Submission Site
• Open Access Policies
• Self-Archiving Policy
• Benefits of Publishing Open Access
• Quarterly Newsletter
• About Annals of Botany
• About the Annals of Botany Company
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Introduction, a mycorrhizal paradigm shift: abundant, underexplored and ecologically relevant symbiotic relationships, funding statement, conflict of interest, acknowledgements, author contributions, literature cited.

• < Previous

Plant–microbe interactions through a lens: tales from the mycorrhizosphere

Alex Williams, Besiana Sinanaj and Grace A Hoysted contributed equally to the manuscript

• Article contents
• Figures & tables
• Supplementary Data

Alex Williams, Besiana Sinanaj, Grace A Hoysted, Plant–microbe interactions through a lens: tales from the mycorrhizosphere, Annals of Botany , Volume 133, Issue 3, 1 March 2024, Pages 399–412, https://doi.org/10.1093/aob/mcad191

• Permissions Icon Permissions

The soil microbiome plays a pivotal role in maintaining ecological balance, supporting food production, preserving water quality and safeguarding human health. Understanding the intricate dynamics within the soil microbiome necessitates unravelling complex bacterial–fungal interactions (BFIs). BFIs occur in diverse habitats, such as the phyllosphere, rhizosphere and bulk soil, where they exert substantial influence on plant–microbe associations, nutrient cycling and overall ecosystem functions. In various symbiotic associations, fungi form mycorrhizal connections with plant roots, enhancing nutrient uptake through the root and mycorrhizal pathways. Concurrently, specific soil bacteria, including mycorrhiza helper bacteria, play a pivotal role in nutrient acquisition and promoting plant growth. Chemical communication and biofilm formation further shape plant–microbial interactions, affecting plant growth, disease resistance and nutrient acquisition processes.

Promoting synergistic interactions between mycorrhizal fungi and soil microbes holds immense potential for advancing ecological knowledge and conservation. However, despite the significant progress, gaps remain in our understanding of the evolutionary significance, perception, functional traits and ecological relevance of BFIs. Here we review recent findings obtained with respect to complex microbial communities – particularly in the mycorrhizosphere – and include the latest advances in the field, outlining their profound impacts on our understanding of ecosystem dynamics and plant physiology and function.

Deepening our understanding of plant BFIs can help assess their capabilities with regard to ecological and agricultural safe-guarding, in particular buffering soil stresses, and ensuring sustainable land management practices. Preserving and enhancing soil biodiversity emerge as critical imperatives in sustaining life on Earth amidst pressures of anthropogenic climate change. A holistic approach integrates scientific knowledge on bacteria and fungi, which includes their potential to foster resilient soil ecosystems for present and future generations.

Soil is essential for supporting life on Earth. The biodiversity of soil, however, is a threatened resource that is vulnerable to both climate change and intensive agroecosystem management ( Lehmann et al ., 2020 ; Guerra et al ., 2021 ). Soil quality is critical to ecological balances and resource provision, so implementing measures to preserve and maintain soil is fundamental for the health of both managed and natural systems. With hundreds of thousands of taxa per gram of soil, complex microbial communities dominate soil biodiversity ( Allison and Martiny, 2008 ). This consortia of microorganisms, also known as the soil microbiome, collectively perform key ecosystem processes including nutrient cycling and maintaining soil health.

The soil microbiome comprises bacteria, archaea, fungi, algae and protists and frequently consists of microhabitats that are linked by diverse intra- and inter-kingdom interactions ( Braga et al ., 2016 ; Berg et al ., 2020 ). Due to their critical significance in the operation of terrestrial ecosystems, particularly plant growth, development and function, bacterial–fungal interactions (BFIs) are drawing increasing research attention. BFIs occur within complex microbial communities, existing in various plant and soil niches including the phyllosphere (the aerial part of the plant), the rhizosphere (the region surrounding the plant roots) and the bulk soil (the soil outside of the rhizosphere and not penetrated by plant roots) ( Fig. 1 ). Under fluctuating environmental conditions, bacteria and fungi engage in complex interactions ranging from antagonism to mutualism. As a result, BFIs have significant impacts on the plant hosts and can augment growth, reproduction, transport/movement, nutrition, stress resistance and pathogenicity of the partners involved ( Deveau et al ., 2018 ). Understanding plant–microbe associations and their implications for plant and ecosystem functioning relies on a solid knowledge base regarding the dynamics between eukaryotic and prokaryotic organisms ( Frey-Klett et al ., 2011 ). It is noteworthy that, like plants, microorganisms in the soil such as mycorrhizal fungi can also recruit distinct bacteria forming a ‘mycorrhizosphere’, which is defined as the soil zone influenced by both roots and mycorrhizal fungi ( Barea et al ., 2002 ). Furthermore, another distinct suite of microbes, which provide enhanced nutrient mobilization and turnover, can be recruited directly to the ‘hyphosphere’, which is defined as the narrow region around fungal hyphae where conditions are different from the bulk soil due to hyphal exudates ( Wang and Feng, 2021 ). In this review, we describe these BFIs with a focus on endosymbiotic fungi [specifically arbuscular mycorrhiza (AM) and Mucoromycotina ‘fine root endophytes’ (MFRE)] and collate recent findings obtained towards BFIs in the mycorrhizosphere. We include the latest advances regarding the role of interactions and mechanisms involved, the different outcomes for each partner and, ultimately, how these impact plant function. Finally we discuss knowledge gaps, emerging opportunities and applications of BFIs with respect to the mycorrhizal hyphosphere and function.

Step-wise overview of putative multipartite interactions between plant roots, mycorrhizosphere microbial communities and arbuscular mycorrhizal (AM) fungi and Mucoromycotina fine root endophyte (MFRE) fungi. 1. A sessile plant. 2. Exclusive MFRE fungi (white hypha) interactions are rare, often forming simultaneous associations alongside AM fungi (red hypha); diverse bacterial populations associate with fungi in and around the hyphae and this is termed the mycorrhizosphere. 3. Putative/unknown overview (dashed lines) of complex plant root and fungal hyphal exudate interactions. 4. Overview of potential bacterial interactions with both AM and MFRE fungi in terms of nutrient mineralization and transfer to host plant. 5. Summary panel of fundamental questions from mycorrhizal bacterial–fungal interactions (BFIs) including analysis of putative biofilms, effects of bacterial populations on nutrient mineralization and mycorrhizal function, bacterial–bacterial (B-B) interactions, fungal–fungal (F-F) interactions and BFIs.

Characterizing the mycorrhizosphere

Terrestrial plants can obtain vital nutrients through a number of pathways: directly through root uptake or through symbiotic interactions, such as mycorrhizal hyphal associations ( Smith and Smith, 2011 ) or via N 2 -fixing bacteria (reviewed elsewhere, see Pii et al ., 2015 and references therein). Based on their structure and function, four major mycorrhizal types have been identified: AM, ectomycorrhiza (EM), orchid mycorrhiza and ericoid mycorrhiza ( van der Heijden et al ., 2015 ). Due to their capacity to establish symbiotic interactions with ~80 % of land plant families ( Tedersoo et al ., 2020 ), AMs formed by the obligately symbiotic fungi of the phylum Glomeromycota are a crucial functional group of soil biota that can generally contribute to crop productivity and ecosystem sustainability ( Smith and Read, 2008 ; Köhl et al ., 2016 ). AM colonization of plant roots begins when the AM fungi sense chemical signals released by plant roots. The hyphae penetrate the root tissues and develop inter- and intra-cellularly, forming running hyphae, coils and arbuscules. The AM fungi also form vesicles that contain lipid reserves and which can be used by the fungus during times of scarcity ( van Aarle and Olssen, 2003 ). In contrast to AM symbiosis, comparatively few plants live in symbiosis with EM fungi. EM symbiosis is established predominantly by basidiomycetes but also ascomycetes ( Martin et al ., 2016 ). The spores derived from the sporocarps of EM fungi germinate and inhabit specific types of lateral roots known as fine or short roots. These fungi’s hyphae extend into the surrounding soil, some passing through the epidermal and cortical cells and forming the Hartig net, which serves as the site for nutritional exchange between the fungal and plant cells. In addition to enhancing plant growth by facilitating the uptake of soil phosphorus (P) and other vital mineral nutrients, both AM and EM fungi possess ‘non-nutritional’ effects. These effects include stabilizing soil aggregates, preventing erosion, and mitigating plant stress caused by environmental factors such as drought and extreme temperatures ( Smith and Read, 2008 ; Gianinazzi et al ., 2010 ).

The acquisition and utilization of nutrients by plants, whether obtained directly through the root system or facilitated by mycorrhizal and/or other symbiotic associations, rely on the recruitment of specific groups of soil microbes. Previous research has predominantly examined the interactions between beneficial free-living soil bacteria and plants, commonly known as plant-growth-promoting rhizobacteria (PGPR). These bacteria have been the primary focus of most studies conducted so far. However, mycorrhizosphere and hyphosphere bacteria are likely to play a role in the success and outcome of plant-growth-promoting fungi. The interaction between mycorrhizal fungi and associated bacteria is complex and highly specific. The composition and function of the bacterial community can vary depending on the plant species – these compositions also differ between controlled laboratory and natural field conditions ( Wang et al. , 2016 ; Zhang et al ., 2022 ). The relationships between AM fungi and hyphosphere microbiome members, and the outcome of their interactions on organic nutrient utilization and nutrient cycling, have only been investigated in a few studies so far, primarily using in vitro experimental setups assessing the effects of single or multiple bacterial genotypes ( Jiang et al ., 2021 ). For example, in in vitro culture, where AM fungi were manipulated, fructose secreted by hyphae served to increase bacterial phosphatase activity and shift bacterial community structure ( Zhang et al ., 2018 a , b ). To date, some well-designed mesocosm studies using root-free compartments, aiming to characterize AM fungal hyphosphere microbial communities, have been carried out. Using a range of non-sterile soils, Emmett et al . (2021) exposed root-associated AM fungi to preferentially recruit bacteria to their hyphosphere. Complementary techniques such as 16S rRNA gene amplicon sequencing, terminal fragment restriction length polymorphism (T-RFLP) analysis and 13 CO 2 pulse labelling have also led to the recognition of hyphal-associated phosphate-solubilizing bacteria (PSB; Wang et al ., 2016 ). However, how these results translate to complex natural or agro-ecosystems in terms of plant growth and fitness is yet to be explored. Thus, our current knowledge of the functioning of the hyphospheric microorganisms and the underlying mechanisms of interactions between AM fungi and their associated microbiome is still largely incomplete ( Jansa and Hodge, 2021 ; Zhang et al ., 2022 ). In addition, little consideration has been taken regarding interaction with new emerging groups of mycorrhizal fungi such as MFRE fungi ( Bidartondo et al ., 2011 ; Field and Pressel, 2018 ; Hoysted et al ., 2023 ). It is clear that BFIs play a critical role in plant host health, along with soil nutritional status, but research focus is required to identify further key players and their functional mechanisms.

Evolutionary origin of plant–fungal–bacterial interactions

In aid of continued identification of important BFIs, evolutionary insights can be gleaned from extant fungal endosymbiotic bacteria that live within hyphae. These endobacteria are widespread, occurring throughout the phyla Mucoromycota, Ascomycota and Basidiomycota, and range from obligate symbionts to facultative endofungal associates ( Steffan et al ., 2020 ; Table 1 ). Given that endobacteria are found in all three subphyla of the Mucoromycota, it is likely that the common ancestor of this early phylum evolved bacterial hosting abilities and thus also harboured endobacteria ( Bonfante and Venice, 2020 ). This is supported by the Rhynie Chert, a fossil-rich sedimentary deposit from the early Devonian where putative bacterial colonies and swellings have been observed in plants which harbour AM-like fungal structures, e.g. Nothia ( Taylor et al ., 2003 ). The ability of putative bacteria and fungi to penetrate the cell walls of these Devonian plants and possibly overcome plant defences suggests it is likely that fungal endobacteria would have existed >410 million years ago. The function of early endobacteria may have been critical to fungal fitness, enough so that they have persisted to this day.

Endobacteria known to be involved in bacterial–fungal interactions (BFIs).

The diversity and functional significance of intimate BFIs is largely unknown. To date, three groups of endobacteria have been identified in AM fungi: CaGg ( Candidatus Glomeribacter gigasporarum), which are rod-shaped Betaproteobacteria that inhabit members of the Gigasporaceae fungal family; CaMg ( Candidatus Moeniiplasma glomeromycotorum), which are widespread coccoid-shaped Mollicutes-related endobacteria; and Burkholderia -related bacteria that appear phylogenetically similar to Mycoavidus cysteinexigens , an endobacterium of the soil fungus Mortierella elongata ( Lastovetsky et al. , 2018 ). Mycoavidus sp. has been shown to rely on M. elongata for carbon (C) and nitrogen (N), though it is unclear what the fungal host receives in return ( Li et al ., 2017 ). In a study coupling microfluidics with metabolomics, Burkholderia bacteria were shown to engage in metabolite exchange with M. elongata , obtaining organic acids from their fungal partners, and resulting in a higher growth rate for both organisms ( Uehling et al ., 2019 ). Similarly, the Burkholderia -related bacteria interacting with AM fungi may also rely on their hosts for nutrients.

There are several potential ways that the fungi may benefit from endobacteria. The few known functions of fungal endobacteria include the production of mycotoxins to confer fungal pathogenicity and promotion of fungal reproductive success ( Table 1 ). Of the endobacteria that occur in mycorrhizal fungi, CaGg bacteria have the full operon for B12 vitamin synthesis ( Ghignone et al ., 2012 ), a potential currency for profiting from osmotically constant conditions and a supply of C compounds within their fungal hosts. CaGg and CaMg bacteria can coexist in the same spore ( Desirò et al ., 2014 ), which brings into question whether they provide fungi with two different benefits or whether they compete for fungal hosts.

The extent that BFI involving endobacteria in mycorrhizal fungi are necessary for downstream plant–fungal symbioses remains to be explored. Several of these endobacteria have reduced genomes containing host-colonization genes such as vacB ( Ruiz-Lozano and Bonfante, 2000 ), which imply they have integrated closely within their fungal hosts and may have adapted to persist by promoting fungal fitness and consequently the fungi’s ability to form associations with plants. On the other hand, there are mycorrhizal endobacteria, such as Rhizobium radiobacter , which do not exhibit reduced genomes and appear to be facultative in their interactions with their host fungi. Rhizobium radiobacter has been reported to aid Piriformspora indica in establishing plant–fungal symbioses ( Guo et al ., 2017 ). The evolution of the mechanisms for these facultative bacterial symbionts is yet to be determined ( Steffan et al ., 2020 ) and understanding their functional significance in the context of plant–fungal interactions under different abiotic and biotic stresses will be important for future efforts to selectively increase the ecological fitness of mycorrhizal fungi. We now know that BFI have been occurring at least as long as plants have existed, and were central to plant evolution. These interactions are not always positive for the plant host. Predicting their function could partly be informed by studying the chemical signatures of these enigmatic systems.

Under the hood: the mechanics of plant–microbial perception

Although a strategy that benefits both plant and microbe has clear evolutionary advantages, it is by no means the only route available to microbes to gain plant-derived resources. In fact, the evolution of plant fungal pathogens and plant defences is rooted in the divergence of microbial trophic strategy and the co-evolutionary arms race ( Anderson et al ., 2010 ). Plant–microbe symbioses are ancient associations that facilitated early plant terrestrialization and have arisen multiple times through evolutionary history ( Puginier et al ., 2022 ). Plant recognition of symbionts shares molecular machinery with pathogenic responses. For example, symbiotic fungal colonization of Medicago truncatula showed that the plant has similar pre-infection nuclear repositioning responses to both the AM symbiont and fungal pathogens; these responses involve dmi3 which encodes a calcium/camodulin-dependent protein kinase ( Genre et al ., 2009 ). This finding indicates common primary responses occur in plants to interacting microbes, and are further modified depending on the nature of the interacting organism. All plant–microbial interactions are dependent on mutual recognition through effectors (e.g. fungal polysaccharides, microbe-associated molecular patterns, chitin) or via plant/hyphal exudates ( Fig. 1 , panel 3). Importantly, the composition of plant-exuded chemicals can be altered during plant–microbe interactions. For instance, plant volatiles are exuded very early on in the interaction between M. truncatula and Rhizophagus irregularis and contain large amounts of limonene ( Dreher et al ., 2019 ). In contrast, when infected with the pathogenic oomycete Aphanomyces euteiches , exudation of sesquiterpene predominated. Thus, exudates are highly dependent on the specific microbe present. Although the exact function of many different exuded chemical remains to be established, changing chemical ‘flavours’ is a strategy by the plant to distinguish ‘friend’ from ‘foe’. During early symbiosis, and interaction with a pathogen, the common symbiotic pathway (CSP) is activated. However, the mechanisms of interaction between plant and specific microbes are nuanced, and similarities usually hold at the top levels. For instance, in rhizobia, activation of nitrous oxide precedes transient reactive oxygen species (ROS) accumulation, activation of the auxin pathway and the suppression of pathogen-elicited phytohormone responses (such as ethylene, salicylic acid and jasmonic acid), resulting in sustained ROS production and plant defence responses ( Anderson et al ., 2010 ). The fact that nodulation (NOD) factors in rhizobia can also elicit a defence response in non-hosts ( Liang et al ., 2013 ) suggests effector and NOD factor recognition share a similar pathway with pathogen response in plants, despite their potentially beneficial function. Most endophytic and necrotrophic lifestyles are not evolutionarily stable traits, with evidence for multiple ‘switches’ between these modes within a single lineage of origin ( Delaye et al ., 2013 ). It is perhaps this shifting of trophic strategies in interacting microbes over large evolutionary timescales that has led to the development of varied and elaborate plant response mechanisms. Biotrophy (whereby pathogens can benefit nutritionally from living host cells) seems to emerge and persist, indicating it is a much more stable strategy. Although they have multiple and separate origins, biotrophs, necrotrophs and endosymbionts all share commonalities, such as host immune circumvention ( Andrew et al ., 2013 ).

The development of fungal effectors probably first appeared to ‘mask’ invading pathogens from the plant host to prevent a defence response (e.g. the Pleidies effectors in the maize pathogen Ustilago maydis ; Navarette et al ., 2021 ). The nutritional benefit to the microbe through access to plant resources could be curtailed when the host plant developed additional recognition and protection strategies (mechanical impedance, chemical disarmament) and when the hyper-sensitive response (HR) evolved to effectively kill the infected cell and pathogen. Development of cytotoxicity and self-defence within pathogens will have then provided a benefit for a necrotrophic organism, where cell destruction allows access to resources (indeed, many necrotrophic effectors have evolved specifically to induce HR and cell death; Shao et al ., 2021 ). During a necrotrophic challenge, the plant immune system requires different strategies to defend its tissue; for instance, camalexin production in Arabidopsis thaliana is an effective defence against many necrotrophs ( Khare et al ., 2017 ). Pathogen resistance to host defence, for instance via metabolic degradation of these antimicrobial molecules, is a key facet of the ongoing adaptation and survival of these pathogens ( Newman and Derbyshire, 2020 ). Hence, effector-led infection/defence is locked in an evolutionary cycle, where development of novel effectors leads to development of novel defences: a cat-and-mouse between pathogens masking their existence and plant detection of their presence. The occasional emergence of hemi-biotrophy, where pathogens can ‘switch’ their trophic modes, may occur to gain the upper hand and most efficient nutrient acquisition, as suggested is the case for the fungal hemi-biotroph Sclerotinia sclerotiorum ( Kabbage et al ., 2015 ). These multifaceted evolutionary events have generated the complexity we currently observe in defence-related plant metabolism, and an immune machinery that must balance interaction with multiple symbiotic/pathogenic microbes of differing origins.

The similarities between initiation of microbial symbioses and immune suppression strategies are remarkable. However, most molecular interactions are highly specialized (i.e. specific host–microbe machinery), presenting difficulty in unpicking the fundamental aspects of early symbioses. In these early scenarios, as well as in contemporary ecologies, microbes would not have interacted with plants in isolation but instead in complex, trophically competitive situations. Niche exploitation and specialization are likely to have appeared rapidly where many competing microbes are vying for an edge over access to plant-derived resources. In certain circumstances partnerships between two or more microbes would have provided more successful niche exploitation. For example, it has been suggested that Oxalobacteraceae species, which are found living abundantly within AM hyphae, provide the fungi with greater colonization ability, probably through production of unique metabolites ( Scheublin et al ., 2010 ). In contrast, pathogens can also utilize symbioses to help infect hosts. Strains of the plant beneficial bacteria Pseudomonas fluorescens appear to be susceptible to colonization by pathogens such as Escherichia coli and Staphylococcus aureus ( Zarei et al ., 2022 ). Considering certain opportunistic pathogenic bacteria (e.g. S. aureus ) have the ability to hijack the mammalian immune response via inoculation of avirulent bacteria ( Boldock et al ., 2018 ), it is not unreasonable to suggest a similar hijacking approach can also occur in plants. Likewise, when multiple beneficial organisms interact with the same host, selective niche specialization allows a competitive edge to the individual but where niche overlap occurs, nutritional cross-feeding has been shown to promote coexistence ( Jacoby and Kopriva 2019 ). For example, in nodulating plants, such as clover, both the plant and rhizobia stand to gain a nutritional advantage when a third-party AM symbiosis also forms ( Ossler et al ., 2015 ). These benefits are dependent on the developmental stage of the symbiosis, where initial growth penalties are succeeded by much greater benefits when the tripartite interaction is fully established ( Mortimer et al ., 2008 ). However, specific host genetics and abiotic environments will also determine the extent of collegiality between these symbiotic partners ( Ossler et al ., 2015 ). During their evolution, plant-associated microbes have diversified to exploit a range of interaction strategies. While complex, the outcome tends to be positive or negative for the plant host. It is due to this fact that plant recognition mechanisms contain so much overlap and redundancy, albeit characterizing the extent to which these mechanisms differ among plant groups, and with specific interacting microbes, is an ongoing task.

Microbial consortia are the functional workforce of the rhizosphere

Understanding the roles of individual rhizosphere microbes within an ecological consortium is much more difficult than for one or two competing root-colonizers. The rhizosphere, much like the gut, is a melting pot of functionally active microbes, which can act to alter host physiology in various ways ( Selosse et al ., 2014 ). Modern cataloguing approaches (meta-genomics) can identify the microbe species present, or the types of functional genes that exist in the system, but cannot directly infer function ( Hornung et al ., 2019 ). This is a particular problem considering that many species with assigned ‘functions’ (e.g. pathogen, symbiont, saprotroph, wood-decayer) are only studied in certain contexts, and their functional adaptability is not recorded. For example, Pseudomonas simiae generally acts as a beneficial and growth-promoting bacterium in A. thaliana ; yet when CO 2 is elevated and nutrient availability is low, A. thaliana root development becomes repressed under P. simiae colonization ( Williams et al ., 2018 ). Considering these points, the traditional reductionist experimental systems we use to study interaction (i.e. 1 plant, 1 microbe, 1 set of conditions) could be inadequate to describe the biological function of the organisms in question (see Spanu and Panstruga, 2017 ). Considering entire microbial assemblages of a system, such as a rhizosphere, as a tool to predict functional traits is equally problematic, largely due to the complexity of these interactions ( Munoz-Ucros et al ., 2021 ). More recent use of synthetic communities coupled with high-resolution -omics technologies permits unprecedented insight into the molecular foundations of niche partitioning across different species. This approach was recently taken in A. thaliana ( Mataigne et al ., 2022 ) where use of genome-scale metabolic models indicated that metabolic asymmetry of the rhizosphere (driven by nutrient availability and root exudation) predetermines microbial niche partitioning and overlap.

Most microorganisms operate as part of larger multi-species consortia, including within specialized multicellular complexes such as biofilms ( Yang et al ., 2011 ). These communities cohabit a specialized slurry consisting of extracellular polymeric substances (EPS). They are present throughout nature in most conceivable habitats (such as water pipes, bones and prosthetics after trauma or surgery; Bar-On and Milo 2019 ), anywhere a suitable surface and microbes are present. Pertinently, biofilms develop within the rhizosphere, particularly along the root surface ( Fig. 1 , panel 5). The biochemistry of biofilms is seldom described, principally due to their mechanical sensitivity and the difficulty extracting a root system from soil without interrupting them. It has been demonstrated that Bacillus subtilis biofilms on plant roots are reliant on plant-derived polysaccharides that are used as both a cue and an energy source to produce the EPS of the biofilm ( Beauregard et al ., 2013 ). Biofilms are chemically rich environments and, along with the mucilage of the roots, may coexist as an interface between plant and microbes for both communication and nutrient exchange via exuded metabolites. Further, it has been postulated that the evolution and exudation of C-rich metabolites that constitute the mucilage act as a de facto phytogenic scaffold, not dissimilar in properties to the EPS, which rhizosphere microbes can make use of to rapidly develop their biofilms ( Nazari et al ., 2022 ). Microbe to microbe communication occurs within biofilm environments too, and it is likely that mutually beneficial interactions occur here such as between AMs and PGPR bacteria, such as P. simiae ( Pandit et al., 2020 ). Biofilms also act as a potentially attractive cache of C-rich resources for invading organisms. Fascinatingly, these rich communities exhibit elements of self-preservation, as well as providing a genetic reservoir with high potential for adaptation, for example through horizontal gene transfer between bacteria ( Lyons and Kolter, 2015 ; Berthold et al ., 2016 ). It is remarkable then that their study is still so limited, and there is much that their formation, function and metabolism can tell us about early life and microbial [bacteria–fungi (B–F), bacteria–bacteria (B–B), fungi–fungi (F–F)] interactions.

Bacterial trafficking via hyphal highways

In addition to harbouring genetic and chemical diversity, microbial biofilms act as a medium through which microbes can move within the rhizosphere. In soil, bacteria are heterogeneously dispersed and are mobile through swimming, swarming, twitching, sliding or gliding, all of which are more efficient in the presence of a liquid ( Henrichsen, 1972 ; Dechesne et al ., 2010 ). Although pores filled with water are present between soil particles, the rugged landscape of substrates can mean bacterial movement is limited, even in the presence of chemotaxis ( Zhang et al ., 2020 ). In fungal networks, the thin water film which surrounds hyphae is used like a ‘highway’, allowing particular bacteria to translocate along it with ease ( Jansa and Hodge, 2021 ). For example, the saprotrophic fungus Lyophyllum sp. carries bacteria, which travel distances as great as 1 cm per day ( Warmink and Van Elsas, 2009 ). Bacterial movement via hyphae has also been demonstrated in vitro for AM fungi and the phosphorus-mineralizing bacterium Rahnella aquatilis (Jiang et al ., 2021). These authors present a model for the recruitment and transport of bacteria by mycorrhizal fungi lacking the genetic machinery to mineralize nutrients in complex organic material. The implications of this model for the function of BFIs have real ecological significance, as they amplify the influence of fungi in the rhizosphere; that is, they not only shape the composition of the microbial community around them, but also the spatial distribution of bacteria. It is likely that there is some level of selectivity to the types of bacteria that are translocated on the hyphae, for example through size exclusion, the release of deterrent or attractant compounds, and through microbe–microbe communication.

Microbial hub-bub: the molecular language of the rhizosphere

Microbes possess an innate ability to secrete and recognize different chemical cues in their environment. These semiochemicals are either ‘self’ specific, an example being small-peptide pheromonic signals ( Calcagnile et al ., 2019 ), or shared between different species, examples including allelochemicals such as coumarins, flavonoids and organic acids. Self and neighbour recognition allows microbes to form mixed communities as well as secure nutrient resources by limiting competition, for instance production of siderophores with both iron-sequestering and growth-inhibitory functions ( Gu et al ., 2020 ). While self-recognition cues are more straightforward to identify in controlled conditions, it is only recently that the vast suite of secondary metabolite signals microbes use have started to be unravelled in mixed communities ( Chevrette et al ., 2022 ). This is largely due to development of high-throughput, high-resolution mass spectrometry and imaging techniques such as DESI-ESI and MS-MS networking ( Stasulli and Shank, 2016 ). Similarly, monocultured microbes seldom produce signals in the absence of other microorganisms ( Liu and Kakeya, 2020 ) limiting the knowledge we can garner from traditional culturing approaches. More intricate, tri- or multi-partite interactions can provide us with much more relevant information, for instance the components and activity of microbial allelochemical exudation. A captivating illustration of competitive allelopathy can be found in the fungal endophyte of Japanese plum-yew, Paraconiothyrium variabile , which can negatively modulate the production of the mycotoxin beauvericin by phytopathogenic Fusarium oxysporum ; this occurs through elevated secretion of P. variabile secondary metabolites including 13-oco-9, 11-octadecdienoic acid ( Combès et al ., 2012 ). In this example resource guarding by the endophyte is tailored towards host protection – if the host is damaged C allocation to the endophyte is reduced.

The examples above assume cell-to-cell communication between individual microbial species, but if considering a microbial consortium, it is logical to discuss whole community chemical communication. Quorum sensing (QS) is the process by which exuded signals can synchronize a microbial community response to a trigger, a classic example being population density ( Darch et al ., 2012 ). QS relies on autoinducers that are produced and recognized between cells. This phenomenon is generally studied within species (e.g. B. subtilis , which utilize a multi-protein complex, ComQXPA, to achieve QS; Kalamara et al ., 2018 ), but sensing and response between species also occurs ( Whitehead et al ., 2001 ), for instance in a biofilm where the production of both granular structures and EPS are determined by the QS metabolite N -acyl-homoserine-lactone (AHL; Tan et al ., 2014 ). Although multi-species QS mechanisms are less well understood, combined modelling and genetic cataloguing approaches suggest that both long-range and species-specific signals can move in this way ( Silva et al ., 2017 ; Abisado et al ., 2018 ). More recently the use of similar modelling techniques and knowledge of AHL-directed QS have been postulated as a method to engineer QS-like behaviours across industrially valuable microbes, such as biofuel-producing cyanobacteria ( Kokarakis et al ., 2023 ). Although these synchronizing behaviours are incredibly useful in the rhizosphere context, a dearth of technology directed toward their study means this is still an emerging field of research. QS certainly has a role in many microbes, including rhizobia (for nodulation/N fixation; Gosai et al ., 2020 ), PGPR ( Hartmann, 2020 ), and plant-growth-promoting fungi (rhizobia within AM spores; Palla et al ., 2018 ). Importantly, plant hosts have evolved both to recognize and to respond to some of these QS signals, allowing them a greater selective control over microbial partners (such as QS quenching to disrupt activity of various pathogens; Rodríguez et al ., 2020 ), highlighting the complexity of chemical communication among mixed plant-interacting microbial communities in the rhizosphere. Hence, the interface through which BFI and plant communication takes place is important for recognition and function. It is by identifying and characterizing this chemical language that we can start to develop a toolbox for a much more in-depth understanding of rhizosphere communication at large.

Mycorrhiza helper bacteria: the fungal facilitators

Mycorrhiza helper bacteria (MHB) are the suite of microbes associated with mycorrhizal fungi and which serve as the third crucial participant in plant–BFI associations ( Sangwan and Prasanna, 2022 ). MHB actively contribute to, and promote, various aspects within the mycorrhizal symbiosis including the functioning of AM symbiosis, stimulation of hyphal growth and spore germination, root colonization, and the improvement of metabolic fitness of AM fungi. MHB are not specifically associated with a particular type of mycorrhizal symbiosis, or taxonomic classification of bacterial strain. MHB were first discovered when Douglas fir was co-inoculated with the EM fungus Laccaria laccata and P. fluorescens BBc6 ( Duponnois and Garbaye, 1991 ). Since their discovery, MHB have been classified into two distinct groups based on their mode of action: mycorrhiza helper bacteria influence the functioning of an existing AM fungal symbiosis; and mycorrhization helper bacteria actively stimulate the formation of the symbiotic association between AM fungi and the roots of their host plants ( Frey-Klett et al. , 2007 ). MHB have been isolated from diverse environments: the mycorrhizosphere, hyphosphere, sporocarps and fruiting bodies of EM fungi.

To date, many bacterial strains have been reported to be able to promote either AM or EM symbioses ( Table 2 ; Garbaye, 1994 ; Barea et al ., 2002 ; Johansson et al ., 2004 ; Artursson et al ., 2006 ; Duponnois, 2006 ) with the first MHB being described in association with the AM fungus Rhizophagus ( Glomus ; Mosse, 1962 ). Being inhabitants of the mycorrhizosphere, MHB are not governed directly by the plant host, but they exhibit a gradient of specificity in relation to the fungus ( Hameeda et al ., 2007 ). However, it is important to note that the composition and function of the bacterial community can vary depending on the plant species and prevailing environmental conditions. The MHB strains that have been identified to date belong to many bacterial groups and genera ( Table 2 ), such as gram-negative Proteobacteria, gram-positive Firmicutes and gram-positive Actinomycetes. The co-evolutionary history between MHB and their fungal partners has probably influenced the specificity of their interactions, enabling finely tuned cooperation in nutrient acquisition and other essential functions. However, there are additional groups of MHB that exhibit more generalized effects on plant growth promotion, highlighted in numerous studies ( Deveau and Labbe, 2017 ). The effects encompass the production of phytohormones, such as auxins, cytokinins and gibberellins, as well as play pivotal roles in regulating various aspects of plant growth, including root development, shoot growth and flowering. Therefore, beneficial activities of MHB are not solely limited to their role in mycorrhizal formation and function.

Bacteria known to be involved in bacterial–fungal interactions (BFIs) on the exterior of fungal structures.

MHB: engineers of mycorrhizal development and function

MHB have been reported to facilitate spore germination and promote mycelial growth in mycorrhizal fungi. The stimulatory effect of MHB and their culture filtrates on spore germination was first documented in the endomycorrhizal fungi, Glomus mossae (Mosse, 1962). Subsequently, the effects of MHB associated with Glomus clarum , where direct contact between the spores and bacteria was necessary for the induction of spores, were observed ( Xavier and Germida, 2003 ), indicating a ligand–receptor interaction between the two microbes. However, the interaction was suggested to be much more complex as these spore germination stimulatory bacteria were accompanied by other bacterial isolates on the G. clarum spore surface, producing antagonistic volatiles, which regulated spore germination. In contrast, volatile compounds produced by different species of Streptomyces were shown to promote the germination of G. mossae spores ( Tylka et al ., 1991 ). Similarly, it was shown that the otherwise obligately symbiotic Glomus intraradices could grow and sporulate without a plant in fungal–bacterium co-cultures with Paenibacillus validus ( Hildebrandt et al ., 2002 ), with raffinose being detected as the specific C source in the cultures supporting mycelial growth ( Hildebrandt et al ., 2006 ). An increase in mycelial biomass and promotion of mycorrhiza establishment has also been shown in the presence of MHB ( Garbaye and Bowen, 1989 ; Gryndler and Vosátka, 1996 ; Founoune et al ., 2002 ). The colonization of roots with Glomus fistulosum and the growth rate of the hyphae in the soil substrate were significantly higher when the fungus was co-inoculated with Pseudomonas putida or with the low-molecular-weight fraction of the bacterial culture supernatant ( Vosátka and Gryndler, 1999 ), indicating that the effective substances were in this fraction. MHB may contribute to AM fungal hyphal growth through their anti-pathogenic efficacy ( Lioussanne et al. , 2006 ). The MHB Paenibacillus sp. strain B2 was found to stimulate colonization of Funneliformis mossae in sorghum, while simultaneously suppressing soil-borne pathogens such as Phythophora parasitica , Fusarium oxysporum and Rhizoctonia solani ( Budi et al ., 1999 ). MHB have also been implicated in having a strong positive impact on spore germination and on presymbiotic fungal growth under toxic concentrations of heavy metals; bacterial inoculation not only reduced damage to G. mossae hyphae but even resulted in increased mycelial growth and mycorrhiza formation ( Vivas et al ., 2005 ). MHB may also play a vital role in promoting the proliferation of plant root systems. Morphological changes in roots during mycorrhiza formation have been attributed to phytohormones, including auxins and ethylene ( Kaska et al ., 1999 ; Horii and Ishii, 2006 ). These include the formation of lateral roots and dichotomous branching of short roots ( Barker and Tagu, 2000 ), which leads to an increase in potential points at which plant and fungus can interact. MHB have also been indirectly implicated in facilitating root colonization and directing mycelial growth of both AM and EM fungi towards fine roots through the release of chemotropic signals such as flavonoids from plants ( Lagrange et al ., 2001 ; Akiyama et al ., 2002 ). Xie et al . (1995) demonstrated that the NOD factors produced by an MHB Bradyrhizobium japonicum strain stimulated the production of flavonoids by soybean ( Glycine max ) seedlings and subsequently promoted mycorrhiza formation.

MHB interacting with AM fungi also have the ability to enhance the functionality of already established mycorrhizal symbioses, by contributing to mineral weathering and their ability to solubilize phosphorus ( Viveganandan and Jauhri, 2000 ). Within the soil and rhizosphere, various microbes, including fungi and bacteria, actively participate in mineral weathering by secreting protons and complexing agents such as low-molecular-weight organic anions or siderophores. The collective influence of the entire mycorrhizal complex, comprising the root, the symbiotic fungus and other associated microorganisms, on mineral weathering ultimately leads to better plant growth ( Hameeda et al ., 2007 ). In an onion crop, the MHB strains Enterobacter sp. and B. subtilis expedited the uptake of phosphorus from rock phosphate when inoculated with R. irregularis ( Toro et al ., 1997 ). This characteristic of MHB was also demonstrated by Jayasinghearachchi and Seneviratne (2005) , who found that mixed biofilms of phosphate-solubilizing fungi and Bradyrhizobium elkanii accelerated the process of rock phosphate solubilization. Hence, MHB probably play a central functional role within BFI and plant interactions that we are only just beginning to uncover.

The unfolding awareness of emerging groups of mycorrhizal fungi has had major ramifications for terrestrial mycorrhizal research ( Bidartondo et al ., 2011 ; Field et al ., 2015 , 2016 , 2019 ; Field and Pressel, 2018 ; Hoysted et al ., 2019 , 2023 ). Mucoromycotina fine root endophyte (MFRE) fungi are a facultatively biotrophic and saprotrophic lineage ( Fig. 1 , panels 2–5; Bidartondo et al ., 2011 ; Field et al ., 2015 , 2016 ), opening up a realm of undiscovered ecosystem functions alongside the strictly biotrophic Glomeromycota AM fungi, with which MFRE fungi were classified together until recently ( Stürmer, 2012 , Orchard et al ., 2017 a ). Exclusive plant–MFRE symbioses seem to be rare, with the majority of land plants forming simultaneous associations alongside AM fungi ( Rimington et al ., 2015 ), with recent research highlighting a potential functional complementarity between the two groups ( Field et al ., 2019 ). Traditionally, MFRE fungi have been functionally associated with the facilitation of plant P uptake and occur commonly in soils with low P availability ( Smith et al ., 2015 ; Albornoz et al ., 2016 ; Orchard et al ., 2017 b ). However, recent exploration of MFRE function in wild-collected plant-based systems showed that MFRE symbionts transferred significantly more 15 N tracer compared to 33 P tracer to an early-diverging vascular plant than to Haplomitropsida liverworts in comparable experiments ( Field et al ., 2016 ; Hoysted et al ., 2019 ). Furthermore, using analysis of natural abundance 15 N signatures, it was shown that MFRE-associated Lycopodiella inundata and Juncus bulbosus were 15 N enriched compared to co-occurring reference plants with different mycorrhizal partners, supporting a hypothesis that plants hosting MFRE symbionts benefit from fungally acquired N ( Hoysted et al ., 2019 ). While it has been shown that some Glomeromycota AM fungi can transfer N to their associated host ( Ames et al ., 1983 ; Hodge et al ., 2001 , Leigh et al ., 2009 ), the ecological relevance has been widely debated ( Smith and Smith, 2011 ). Coupling this with emerging evidence that MFRE fungi may transfer more N to host liverworts from organic sources compared to Glomeromycota AM fungi ( Field et al ., 2019 ) further highlights their saprotrophic capabilities. Nonetheless, it is crucial to highlight that the majority of MFRE research has been carried out using unpasteurized soil culture-based experimental systems ( Orchard et al. , 2017 b ; Albornoz et al ., 2020 ) or wild-collected plants ( Field et al. , 2015 , 2016 , 2019 ; Hoysted et al. , 2019 , 2021 a , b ), and so probably included other soil microorganisms such as bacteria, alongside MFRE fungi. Recently, an experimental system has been developed to overcome this challenge and can test the function of MFRE fungi in the absence of other microorganisms ( Hoysted et al ., 2023 ).

Using a novel and tractable in vitro experimental system designed to manipulate axenic MFRE cultures ( Hoysted et al ., 2023 ) we can begin to address fundamental questions regarding the mycorrhizosphere and its involvement in MFRE lifestyles (i.e. facultative biotrophy and/or saprotrophy and function). These fundamental questions give rise to exciting possibilities and questions particularly regarding the different ecosystems in which we find diverse lineages of mycorrhizal fungi and their associated bacteria. Microbe complementarity compared to associations with only one type of fungus may be hugely beneficial for the host plant ( Field et al ., 2019 ) as there may be differences in the ability of plants to utilize soil nutrients when associating with different groups of biotrophic and/or saprotrophic fungi, and/or their bacterial associates. With several environmental factors potentially impacting the activity of saprotrophic microbes (both bacteria and fungi), including temperature, pH, substrate additions and soil moisture ( Rousk and Bååth, 2011 ), it is imperative to study such tripartite interactions further including fungus, B-B interactions and BFIs.

Perspectives on plant–bacterial–fungal legacies

The relationship between mycorrhizal fungi and the mycorrhizosphere carries significant ecological implications. Despite plants and mycorrhizal fungi interacting for >450 million years, the functional significance of mycorrhizosphere and hyphosphere bacterial partners are only starting to be appreciated. Such BFIs play pivotal roles in shaping the functioning of ecosystems and influencing plant community dynamics. The soil bacterial community can adversely affect mycorrhizal fungal activities in soil by competing for different nutrient sources, reducing plant growth (i.e. pathogens), interacting with other soil microbes and producing unfavourable chemicals ( Nehl et al ., 1997 ; Glick, 2005 ). However, soil microbial communities can also exert positive effects on mycorrhizal fungi through: the production of plant hormones, solubilization of soil nutrients, improvement of soil structure, control of plant pathogens and stimulation of plant growth ( Glick, 2005 ; Rillig and Mummey, 2006 ).

Among the most important implications of microbial interactions in ecosystem conservation is the alleviation of different soil stresses including salinity, drought, acidity, compaction and heavy metals. Microbes are constantly interacting in the soil, and therefore an understanding of how mycorrhizal fungi and other soil bacteria can alleviate the unfavourable effects of heavy metals on plant growth or how the specific manipulation and engineering of microbes can be used for the remediation of polluted soils ( Berg, 2009 ; Joner and Leyval, 2009 ) is highly desirable. AM fungi are able to alleviate stress by enhancing plant growth, storing heavy metals in the vacuoles of their vesicles and binding them by the production of the glomalin complex ( Khan, 2005 ; Holátko et al ., 2021 ). Byproducts of glomalin interactions are able to bind soil particles and form soil aggregates, resulting in improved soil structure ( Andrade et al ., 1998 ; Barea et al ., 2005 ). Consequently, the advantages of engineering the mycorrhizosphere with a view to enhancing soil structural properties and fertility are attractive. Further research is needed to investigate how stress tolerance can be improved when using AM fungi in combination with other soil microbes, including new emerging groups of mycorrhizal fungi, such as MFRE fungi ( Bidartondo et al ., 2011 ; Field and Pressel, 2018 ; Hoysted et al ., 2023 ).

Engineering of the mycorrhizosphere with respect to enhancing crop plant growth and yield without the use of chemical fertilizers has been highlighted in the literature ( Giovanni et al ., 2020 ; Santos and Olivares, 2021 ). However, BFIs are highly complex and include a wide range of mechanisms where diverse molecules are involved. In agriculture, BFIs are often explored in the frame of the development of biocontrol strains, yet the response, often carried out in glasshouses, is dependent on many factors including the species of each organism involved in the tripartite interaction as well as the prevailing environmental conditions. While AM fungi have the potential to contribute to cereal nutrient assimilation ( Thirkell et al ., 2019 ), limitations surrounding the need for high colonization and a diverse fungal community to maximize crop yield, and failure to apply rigorous agronomic methodology, have been raised ( Ryan and Graham, 2018 ). Such caveats include the small number of agricultural field studies and the ability to manage AM fungi likely to be compromised by rudimentary knowledge underlying mycorrhizal symbiosis. Therefore, there is a call for further field-scale experimentation of what role AM fungi may be playing in crop growth ( Rillig et al. , 2019 ; Thirkell et al. , 2019 ). This is further complicated by the complexity of interactions between mycorrhizal fungi and other soil microbes including bacterial communities, which we have extensively highlighted in this review. Across 11 host species from three different habitats, AM fungi were reported to benefit most plants similarly, but the response of the plant differed when soil filtrates from habitats in which they did not occur were applied ( Pizano et al. , 2017 ). This implies that certain environments accumulate species-specific soil pathogens, whereas other habitats accumulate soil microbes that support native plant species ( Pizano et al ., 2017 ). Further, a meta-analysis, incorporating datasets from both laboratory and field studies, found that the beneficial impact of AM fungi on grain yields was less pronounced in ‘field-inoculation’ conditions, displaying an overall effect of 16 %. The findings exhibited variability, crop-specificity, were lower for new cultivars introduced after 1950 and were additionally influenced by soil pH ( Zhang et al ., 2018 ). Currently, the application of bacterial–fungal bioinoculants to agricultural systems is limited by a variety of factors, including unculturability, low survival rate and low colonization ability of introduced concoctions, as well as lowered expression of key antagonistic traits under field conditions. Nevertheless, their full potential will probably be captured during future work, with the use of comprehensive multi-omics approaches. Therefore, understanding of detailed dynamic inter-kingdom relationships, whether synergistic or antagonistic, will allow scientific exploration and, potentially, agricultural exploitation of these BFIs.

Within the soil microbiome, BFIs play a crucial role in ecosystem functioning and plant health. Here, we emphasize the complexity of BFIs, which can vary depending on plant species, mycorrhizal type and environmental conditions. Overall, we call for critical insights into the modulation of microbiome assembly of emerging groups of mycorrhizal fungi. Filling knowledge gaps and exploring the functional details of microbial interactions in the soil microbiome will advance our holistic interpretation of the true complexity of tripartite interactions in natural and agricultural settings. Using cytological, molecular, proteomic and metabolomic comparisons between inoculated host plants (singly or co-colonized, with both MFRE and AM fungi, in addition to mycorrhizosphere bacteria) we can begin to tease out the finer functional details, in terms of nutrient transfer. Furthermore, investigating how mycorrhizosphere microbial communities accelerate mineralization of essential nutrients will provide invaluable insight into their ecological role within plant–fungal life-systems.

Continued research into the functional capacity of soil microbes and the comparison of sterile and non-sterile experimental systems will contribute to our understanding of the complex interplay between plants, fungi and bacteria in the soil ecosystem. Recognizing the glaring caveats discussed above, this research may also provide opportunities for translation and exploitation into vulnerable agricultural systems. The ecological significance of plant–fungal–bacterial interactions highlights the potential for improving synergistic interactions between mycorrhizal fungi and soil microbes to contribute towards a sustainable future, and therefore we suggest increased efforts to disentangle positive, negative or neutral outcomes in tri-/multi-partite interactions in both laboratory and field settings. The use of soil microbes, including MFRE fungi, stands as a potential solution for alleviating soil stresses and improving soil structure and fertility, such as with AM fungi. Fundamentally, soil biodiversity plays a critical role, particularly interactions within the soil microbiome, in sustaining life on Earth. The vulnerability of soil biodiversity to climate change underlines the urgency of maintaining and improving soil quality for the health of managed and natural ecosystems.

The authors received no financial support for the preparation, authorship and/or publication of this article.

The authors report no known conflicts of interest that impact the validity of this work.

We thank Dr Zoë Popper for inviting G.A.H. to contribute the present review to Annals of Botany . We thank Dr Nicholas Brereton and the anonymous reviewers for their constructive comments that helped to develop our manuscript. A.W. and B.S. were supported by an ERC CoG “MYCOREV” (865225). We also thank Research Figures, Ireland, for the scientific illustration.

G.A.H., A.W. and B.S. conceived and designed the outline and contributed equally to the review article. G.A.H., A.W. and B.S. contributed to the first draft of the manuscript. G.A.H., A.W. and B.S. led the development and editing of the manuscript.

Abisado RG , Benomar S , Klaus JR , Dandekar AA , Chandler JR. 2018 . Bacterial Quorum sensing and microbial community interactions . mBio 9 : e02331 – e02317 .

Google Scholar

Akiyama K , Matsouka H , Hayashi H. 2002 . Isolation and identification of a phosphate deficiency-induced C-glycosylflavanoid that stimulates arbuscular mycorrhiza formation in melon roots . Molecular Plant-Microbe Interactions 15 : 334 – 340 .

Albornoz FE , Lambers H , Turner BL , Teste FP , Laliberté E. 2016 . Shifts in symbiotic associations in plants capable of forming multiple root symbioses across a long-term soil chronosequence . Ecology and Evolution 6 : 2368 – 2377 .

Albornoz FE , Hayes PE , Orchard S , et al.  . 2020 . First cryo-scanning electron microscopy images and X-ray microanalyses of Mucoromycotinian fine root endophytes in vascular plants . Frontiers in Microbiology 11 : 2018 .

Allison SD , Martiny JBH. 2008 . Resistance, resilience and redundancy in microbial communities . Proceedings of the National Academy of Sciences 105 : 11512 – 11519 .

Ames RN , Reid CPP , Porter LK , Cambardella C. 1983 . Hyphal uptake and transport of nitrogen from two 15 N-labelled sources by Glomus mossae , a vesicular-arbuscular mycorrhizal fungus . New Phytologist 95 : 381 – 396 .

Anderson JP , Gleason CA , Foley RC , Thrall PH , Burdon JB , Singh KB. 2010 . Plants versus pathogens: an evolutionary arms race . Functional Plant Biology 37 : 499 – 512 .

Andrade G , Mihara KL , Linderman RG , Bethlenfalvay GJ. 1998 . Soil aggregation status and rhizobacteria in the mycorrhizosphere . Plant and Soil 202 : 89 – 96 .

Andrew RL , Bernatchez L , Bonin A , et al.  . 2013 . A road map for molecular ecology . Molecular Ecology 22 : 2605 – 2626 .

Artursson V , Finlay RD , Jansson JK. 2006 . Interactions between arbuscular mycorrhizal fungi and bacteria and their potential for stimulating plant growth . Environmental Microbiology 8 : 1 – 10 .

Barea JM , Azcón R , Azcón-Aguilar C. 2002 . Mycorrhizosphere interactions to improve plant fitness and soil quality . Antonie Van Leeuwenhoek 81 : 343 – 351 .

Barea JM , Azcón R , Azcón-Aguilar C. 2005 . Interactions between mycorrhizal fungi and bacteria to improve plant nutrient cycling and soil structure . In: Buscot F , Varma A . eds. Microorganisms in soils: roles in genesis and functions . Berlin : Springer , 195 – 212 .

Google Preview

Barker S , Tagu D. 2000 . The roles of auxins and cytokinins in mycorrhizal symbioses . Journal of Plant Growth Regulation 19 : 144 – 154 .

Bar-On YM , Milo R. 2019 . Towards a quantitative view of the global ubiquity of biofilms . Nature Reviews Microbiology 17 : 199 – 200 .

Beauregard PB , Chai Y , Vlamakis H , Losick R , Kolter R. 2013 . Bacillus subtilis biofilm induction by plant polysaccharides . Proceedings of the National Academy of Sciences of the United States of America 110 : E1621 – E1630 .

Berg G. 2009 . Plant–microbe interactions promoting plant growth and health: perspectives for controlled use of microorganisms in agriculture . Applied Microbiology and Biotechnology 84 : 11 – 18 .

Berg GA , Rybakova D , Fischer D , et al.  . 2020 . Microbiome definition re-visited: old concepts and new challenges . Microbiome 8 : 103 .

Berthold T , Centler F , Hübschmann T , et al.  . 2016 . Mycelia as a focal point for horizontal gene transfer among soil bacteria . Scientific Reports 6 : 36390 .

Bidartondo MI , Read DJ , Trappe JM , Merckx V , Ligrone R , Duckett JG. 2011 . The dawn of symbiosis between plants and fungi . Biology Letters 7 : 574 – 577 .

Boldock E , Surewaard BGJ , Shamarina D , et al.  . 2018 . Human skin commensals augment Staphylococcus aureus pathogenesis . Nature Microbiology 3 : 881 – 890 .

Bonfante P , Venice F. 2020 . Mucoromycota: going to the roots of plant-interacting fungi . Fungal Biology Reviews 34 : 100 – 113 .

Braga RM , Dourado MN , Araújo WL. 2016 . Microbial interactions: ecology in a molecular perspective . Brazilian Journal of Microbiology 47 : 86 – 98 .

Budi SW , Tuinen Dv , Martinotti G , Gianinazzi S . 1999 . Isolation from the Sorghum bicolor mycorrhizosphere of a bacterium compatible with arbuscular mycorrhiza development and antagonistic towards soilborne fungal pathogens. Applied and Environmental Microbiology 65 : 11 .

Calcagnile M , Tredici SM , Talà A , Alifano P. 2019 . Bacterial semiochemicals and transkingdom interactions with insects and plants . Insects 10 : 441 .

Chevrette MG , Thomas CS , Hurley A , et al.  . 2022 . Microbiome composition modulates secondary metabolism in a multispecies bacterial community . Proceedings of the National Academy of Sciences of the United States of America 119 : e2212930119 .

Combès A , Ndoye I , Bance C , et al.  . 2012 . Chemical communication between the endophytic fungus Paraconiothyrium variabile and the phytopathogen Fusarium oxysporum . PLoS One 7 : e47313 .

Darch SE , West SA , Winzer K , Diggle SP. 2012 . Density-dependent fitness benefits in quorum-sensing bacterial populations . Proceedings of the National Academy of Sciences of the United States of America 109 : 8259 – 8263 .

Dechesne A , Wang G , Gülez G , Or D , Smets BF. 2010 . Hydration-controlled bacterial motility and dispersal on surfaces . Proceedings of the National Academy of Sciences of the United States of America 107 : 14369 – 14372 .

Delaye L , García-Guzmán G , Heil M. 2013 . Endophytes versus biotrophic and necrotrophic pathogens—are fungal lifestyles evolutionarily stable traits ? Fungal Diversity 60 : 125 – 135 .

Desirò A , Salvioli A , Ngonkeu EL , et al.  . 2014 . Detection of a novel intracellular microbiome hosted in arbuscular mycorrhizal fungi . The ISME Journal 8 : 257 – 270 .

Deveau A , Labbe L. 2017 . Mycorrhiza helper bacteria . In: Martin F , ed. Molecular mycorrhizal symbiosis . Hoboken : John Wiley & Sons Inc , 437 – 450 .

Deveau A , Bonito G , Uehling J , et al.  . 2018 . Bacterial–fungal interactions: ecology, mechanisms and challenges . FEMS Microbiology Reviews 42 : 335 – 352 .

Dreher D , Baldermann S , Schreiner M , Hause B. 2019 . An arbuscular mycorrhizal fungus and a root pathogen induce different volatiles emitted by Medicago truncatula roots . Journal of Advanced Research 19 : 85 – 90 .

Duponnois R. 2006 . Bacteria helping mycorrhiza development . In: Mukerji KG , Manoharachary C , Singh J , eds. Microbial activity in the rhizosphere . Berlin : Springer-Verlag , 297 – 310 .

Duponnois R , Garbaye J. 1991 . Effect of dual inoculation of Douglas fir with the ectomycorrhizal fungus Laccaria laccata and mycorrhization helper bacteria (MHB) in two bare forest nurseries . Plant and Soil 138 : 169 – 176 .

Emmett BD , Lévesque-Tremblay V , Harrisson MJ. 2021 . Conserved and reproducible bacterial communities associate with extraradical hyphae of arbuscular mycorrhizal fungi . ISME Journal 15 : 2276 – 2288 .

Field KJ , Pressel S. 2018 . Unity in diversity: structural and functional insights into the ancient partnerships between plants and fungi . The New Phytologist 220 : 996 – 1011 .

Field KJ , Rimington WR , Bidartondo MI , et al.  . 2015 . First evidence of mutualism between ancient plant lineages (Haplomitriopsida liverworts) and Mucoromycotina fungi and its response to simulated Palaeozoic changes in atmospheric CO 2 . The New Phytologist 205 : 743 – 756 .

Field KJ , Rimington WR , Bidartondo MI , et al.  . 2016 . Functional analysis of liverworts in dual symbiosis with Glomeromycota and Mucoromycotina fungi under simulated Palaeozoic CO 2 decline . The ISME Journal 10 : 1514 – 1526 .

Field KJ , Bidartondo MI , Rimington WR , et al.  . 2019 . Functional complementarity of ancient plant–fungal mutualisms: contrasting nitrogen, phosphorus and carbon exchanges between Mucoromycotina and Glomeromycotina fungal symbionts of liverworts . The New Phytologist 223 : 908 – 921 .

Founoune H , Duponnois R , Meyer JM , Ba AM , Chotte JL , Neyra M. 2002 . Interactions between ectomycorrhizal symbiosis and fluorescent Pseudomonads on Acacia holosericea : isolation of mycorrhization helper bacteria (MHB) from a Soudano-Sahelian soil . FEMS Microbiology and Ecology 41 : 37 – 46 .

Frey-Klett P , Garbaye J , Tarkka M. 2007 . The mycorrhiza helper bacteria revisited . The New Phytologist 176 : 22 – 36 .

Frey-Klett P , Burlinson P , Deveau A , Barret M , Tarkka M , Sarniguet A. 2011 . Bacterial–fungal interactions: hyphens between agricultural, clinical, environmental, and food microbiologists . Microbiology and Molecular Biology Reviews 75 : 583 – 609 .

Garbaye J. 1994 . Mycorrhization helper bacteria: a new dimension to the mycorrhizal symbiosis . Acta Botanica Gallica 141 : 517 – 521 .

Garbaye J , Bowen GD. 1989 . Stimulation of mycorrhizal infection of Pinus radiata by some microorganisms associated with the mantle of ectomycorrhizas . New Phytologist 112 : 383 – 388 .

Genre A , Ortu G , Bertoldo C , Martino E , Bonfante P. 2009 . Biotic and abiotic stimulation of root epidermal cells reveals common and specific responses to Arbuscular mycorrhizal fungi . Plant Physiology 149 : 1424 – 1434 .

Ghignone S , Salvioli A , Anca I , et al.  . 2012 . The genome of the obligate endobacterium of an AM fungus reveals an interphylum network of nutritional interactions . The ISME Journal 6 : 136 – 145 .

Gianinazzi S , Golotte A , Binet MN , van Tuinen D , Redecker D , Wipf D. 2010 . Agroecology: the key role of Arbuscular mycorrhizas in ecosystem services . Mycorrhiza 20 : 519 – 530 .

Giovanni L , Palla M , Agnolucci M , et al.  . 2020 . Arbuscular mycorrhizal fungi and associated microbiota as plant biostimulants: research strategies for the selection of the best performing inocula . Agronomy 10 : 106 .

Glick BR. 2005 . Modulation of plant ethylene levels by the bacterial enzyme ACC deaminase . FEMS Microbioloy Letters 251 : 1 – 7 .

Gosai J , Anandhan S , Bhattacharjee A , Archana G. 2020 . Elucidation of quorum sensing components and their role in regulation of symbiotically important traits in Ensifer nodulating pigeon pea . Microbiological Research 231 : 126354 .

Gryndler M , Vosátka M. 1996 . The response of Glomus fistulosum –maize mycorrhiza to treatments with culture fractions from Pseudomonas putida . Mycorrhiza 6 : 207 – 211 .

Gu S , Wei Z , Shao Z , et al.  . 2020 . Competition for iron drives phytopathogen control by natural rhizosphere microbiomes . Nature Microbiology 5 : 1002 – 1010 .

Guerra CA , Bardgett RD , Caon L , et al.  . 2021 . Tracking, targeting, and conserving soil biodiversity . Science 371 : 239 – 241 .

Guo H , Glaeser SP , Alabid I , et al.  . 2017 . The abundance of endofungal bacterium Rhizobium radiobacter (syn. Agrobacterium tumefaciens ) increases in its fungal host Piriformospora indica during the tripartite sebacinalean symbiosis with higher plants . Frontiers in Microbiology 8 : 629 .

Hameeda B , Srijana M , Rupela OP , Reddy G. 2007 . Effect of bacteria isolated from composts and macrofauna on sorghum growth and mycorrhizal colonisation . World Journal Microbiology Biotechnology 23 : 883 – 887 .

Hartmann A. 2020 . Quorum sensing N-acyl-homoserine lactone signal molecules of plant beneficial Gram-negative rhizobacteria support plant growth and resistance to pathogens . Rhizosphere 16 : 100258 .

Henrichsen J. 1972 . Bacterial surface translocation: a survey and a classification . Bacteriological reviews 36 : 478 – 503 .

Hildebrandt U , Janetta K , Bothe H. 2002 . Towards growth of arbuscular mycorrhizal fungi independent of a plant host . Applied and Environmental Microbiology 68 : 1919 – 1924 .

Hildebrandt U , Ouziad F , Marner FJ , Bothe H. 2006 . The bacterium Paenibacillus validus stimulates the growth of the arbuscular mycorrhizal fungus Glomus intraradices up to the formation of fertile spores . FEMS Microbiology Letters 254 : 258 – 267 .

Hodge A , Campbell CD , Fitter AH. 2001 . An arbuscular mycorrhizal fungus accelerates decomposition and acquires nitrogen directly from organic material . Nature 413 : 297 – 299 .

Holátko J , Brtnický M , Kucerík J , et al.  . 2021 . Glomalin–truths, myths, and the future of this elusive soil glycoprotein . Soil Biology and Biochemistry 153 : 108116 .

Horii S , Ishii T. 2006 . Identification and function of Gigaspora margarita growth-promoting microorganisms . Symbiosis 41 : 135 – 141 .

Hornung BVH , Zwittink RD , Kuijper EJ. 2019 . Issues and current standards of controls in microbiome research . FEMS Microbiology Ecology 95 : fiz045 .

Hoysted GA , Jacob AS , Kowal J , et al.  . 2019 . Mucoromycotina fine root endophyte fungi form nutritional mutualisms with vascular plants . Plant Physiology 181 : 565 – 577 .

Hoysted GA , Bidartondo MI , Duckett JG , et al.  . 2021a . Phenology and function in lycopod–Mucoromycotina symbiosis . New Phytologist 229 : 2389 – 2394 .

Hoysted GA , Kowal J , Pressel S , Duckett JG , Bidartondo MI , Field KJ. 2021b . Carbon for nutrient exchange between Lycopodiella inundata and Mucoromycotina fine root endophytes is unresponsive to high atmospheric CO 2 . Mycorrhiza 31 : 431 – 440 .

Hoysted GA , Field KJ , Sinanaj B , Bell CA , Bidartondo MI , Pressel S. 2023 . Direct nitrogen, phosphorus and carbon exchanges between Mucoromycotina ‘fine root endophyte’ fungi and a flowering plant in novel monoxenic cultures . New Phytologist 238 : 70 – 79 .

Jacoby RP , Kopriva S. 2019 . Metabolic niches in the rhizosphere microbiome: new tools and approaches to analyse metabolic mechanisms of plant–microbe nutrient exchange . Journal of Experimental Botany 70 : 1087 – 1094 .

Jansa J , Hodge A. 2021 . Swimming, gliding or hyphal riding? On microbial migration along the arbuscular mycorrhizal hyphal highway and functional consequences thereof . The New Phytologist 230 : 14 – 16 .

Jayasinghearachchi HS , Seneviratne G. 2005 . Fungal solubilisation of rock phosphate is enhanced by forming fungal–rhizobial biofilms . Soil Biology and Biochemistry 38 : 405 – 408 .

Jiang F , Zhang L , Zhou JC , George TS , Feng G. 2021 . Arbuscular mycorrhizal fungi enhance mineralisation of organic phosphorus by carrying bacteria along their extraradical hyphae . The New Phytologist 230 : 304 – 315 .

Johansson JF , Paul LR , Finlay RD. 2004 . Microbial interactions in the mycorrhizosphere and their significance for sustainable agriculture . FEMS Microbiology Ecology 48 : 1 – 13 .

Joner E , Leyval C. 2009 . Phytoremediation of organic pollutants using mycorrhizal plants: a new aspect of rhizosphere interactions . In: Lichtfouse E , Navarrete M , Debaeke P , Veronique S , Alberola C , eds. Sustainable agriculture . Dordrecht : Springer , 885 – 894 .

Kabbage M , Yarden O , Dickman MB. 2015 . Pathogenic attributes of Sclerotinia sclerotiorum : Switching from a biotrophic to necrotrophic lifestyle . Plant Science 233 : 53 – 60 .

Kalamara M , Spacapan M , Mandic-Mulec I , Stanley-Wall NR. 2018 . Social behaviours by Bacillus subtilis : quorum sensing, kin discrimination and beyond . Molecular Microbiology 110 : 863 – 878 .

Kaska DD , Myllyla R , Cooper JB. 1999 . Auxin transport inhibitors act through ethylene to regulate dichotomous branching of lateral root meristems in pine . New Phytologist 142 : 49 – 58 .

Khan A. 2005 . Role of soil microbes in the rhizospheres of plants growing on trace metal contaminated soils in phytoremediation . Journal of Trace Elements in Medical Biology 18 : 355 – 364 .

Khare D , Choi H , Huh SU , et al.  . 2017 . Arabidopsis ABCG34 contributes to defense against necrotrophic pathogens by mediating the secretion of camalexin . Proceedings of the National Academy of Sciences of the United States of America 114 : E5712 – E5720 .

Köhl L , Lukasiewicz CE , van der Heijden MG. 2016 . Establishment and effectiveness of inoculated arbuscular mycorrhizal fungi in agricultural soils . Plant, Cell and Environment 39 : 136 – 146 .

Kokarakis EJ , Rillema R , Ducat DC , Sakkos JK. 2023 . Developing cyanobacterial quorum sensing toolkits: toward interspecies coordination in mixed autotroph/heterotroph communities . ACS Synthetic Biology 12 : 265 – 276 .

Lagrange H , Jay-Allgmand C , Lapeyrie F. 2001 . Ryutin, the phenylglycoside from eucalyptus root exudates, stimulates Pisolithus hyphal growth at picomolar concentrations . New Phytologist 149 : 349 – 355 .

Lastovetsky OA , Ahn E , Mondo SJ , et al.  . 2018 . Distribution and population structure of endobacteria in arbuscular mycorrhizal fungi at North Atlantic dunes . The ISME Journal 12 : 3001 – 3013 .

Lehmann J , Bossio DA , Kögel-Knabner I , Rillig MC. 2020 . The concept and future prospects of soil health . Nature Reviews Earth & Environment 1 : 544 – 553 .

Leigh J , Hodge A , Fitter AH. 2009 . Arbuscular mycorrhizal fungi can transfer substantial amounts of nitrogen to their host plant from organic material . The New Phytologist 181 : 199 – 207 .

Li Z , Yao Q , Dearth SP , et al.  . 2017 . Integrated proteomics and metabolomics suggests symbiotic metabolism and multimodal regulation in a fungal – endobacterial system . Environmental Microbiology 19 : 1041 – 1053 .

Liang Y , Cao Y , Tanaka K , et al.  . 2013 . Nonlegumes respond to rhizobial nod factors by suppressing the innate immune response . Science 341 : 1384 – 1387 .

Liousanne L , Jolicoeur M , St-Arnaud M. 2006 . Is the inhibition of Phythophora nicotianae in tomato roots induced by arbuscular mychorrizae fungi related to changes in root exudation ? Canadian Journal of Plant Pathology-Revue Canadienne de Phytopathologie 28 : 357 .

Liu C , Kakeya H. 2020 . Cryptic chemical communication: secondary metabolic responses revealed by microbial co-culture . Chemistry, an Asian Journal 15 : 327 – 337 .

Lupini S , Peña-Bahamonde J , Bonito G , Rodrigues DF. 2022 . Effect of endosymbiotic bacteria on fungal resistance toward heavy metals . Frontiers in Microbiology 13 : 822541 .

Lyons NA , Kolter R. 2015 . On the evolution of bacterial multicellularity . Current Opinion in Microbiology 24 : 21 – 28 .

Martin F , Kohler A , Murat C , Veneault-Fourrey C , Hibbett DS. 2016 . Unearthing the roots of ectomycorrhizal symbioses . Nature Reviews Microbiology 14 : 760 – 773 .

Mataigne V , Vannier N , Vandenkoornhuyse P , Hacquard S. 2022 . Multi-genome metabolic modeling predicts functional inter-dependencies in the Arabidopsis root microbiome . Microbiome 10 : 217 .

Mortimer PE , Pérez-Fernández MA , Valentine AJ. 2008 . The role of arbuscular mycorrhizal colonization in the carbon and nutrient economy of the tripartite symbiosis with nodulated Phaseolus vulgaris . Soil Biology and Biochemistry 40 : 1019 – 1027 .

Mosse B. 1962 . The establishment of vesicular–arbuscular mycorrhiza under aseptic conditions . Journal of General Microbiology 27 : 509 – 520 .

Munoz-Ucros J , Zwetsloot MJ , Cuellar-Gempeler C , Bauerle TL. 2021 . Spatiotemporal patterns of rhizosphere microbiome assembly: from ecological theory to agricultural application . Journal of Applied Ecology 58 : 894 – 904 .

Navarrete F , Grujic N , Stirnberg A , et al.  . 2021 . The Pleiades are a cluster of fungal effectors that inhibit host defenses . PLoS Pathogens 17 : e1009641 .

Nazari M , Bickel S , Benard P , Mason-Jones K , Carminati A , Dippold MA. 2022 . Biogels in soils: Plant mucilage as a biofilm matrix that shapes the rhizosphere microbial habitat . Frontiers in Plant Science 12 : 798992 .

Nehl D , Allen S , Brown J. 1997 . Deleterious rhizosphere bacteria: an integrating perspective . Applied Soil Ecology 5 : 1 – 20 .

Newman TE , Derbyshire MC. 2020 . The evolutionary and molecular features of broad host-range necrotrophy in plant pathogenic fungi . Frontiers in Plant Science 11 : 591733 .

Orchard S , Hilton S , Bending GD , et al.  . 2017a . Fine endophytes ( Glomus tenue ) are related to Mucoromycotina, not Glomeromycota . The New Phytologist 213 : 481 – 486 .

Orchard S , Standish RJ , Dickie I , et al.  . 2017b . Fine root endophytes under scrutiny: a review of the literature on arbuscule-producing fungi recently suggested to belong to the Mucoromycotina . Mycorrhiza 27 : 619 – 638 .

Ossler JN , Zielinski CA , Heath KD. 2015 . Tripartite mutualism: Facilitation or trade-offs between rhizobial and mycorrhizal symbionts of legume hosts . American Journal of Botany 102 : 1332 – 1341 .

Palla M , Battini F , Cristani C , Giovannetti M , Squartini A , Agnolucci M. 2018 . Quorum sensing in rhizobia isolated from the spores of the mycorrhizal symbiont Rhizophagus intraradices . Mycorrhiza 28 : 773 – 778 .

Pandit A , Adholeya A , Cahill D , Brau L , Kochar M. 2020 . Microbial biofilms in nature: unlocking their potential for agricultural applications . Journal of Applied Microbiology 129 : 199 – 211 .

Partida-Martinez LP , Hertweck C. 2005 . Pathogenic fungus harbours endosymbiotic bacteria for toxin production . Nature 437 : 884 – 888 .

Pii Y , Mimmo T , Tomasi N , Terzano R , Cesco S , Crecchio C. 2015 . Microbial interactions in the rhizosphere: beneficial influences of plant-growth promoting rhizobacteria on nutrient acquisition processes – A review . Biology and Fertility of Soils 51 : 403 – 415 .

Pizano C , Mangan SA , Graham JH , Kitajima K. 2017 . Host-specific effects of soil microbial filtrates prevail over those of arbuscular mycorrhizae in a fragmented landscape . Ecological Applications 27 : 1946 – 1957 .

Puginier C , Keller J , Delaux P-M. 2022 . Plant–microbe interactions that have impacted plant terrestrializations . Plant Physiology 190 : 72 – 84 .

Rillig MC , Mummey DL. 2006 . Mycorrhizas and soil structure . The New Phytologist 171 : 41 – 53 .

Rillig MC , Aguilar-Trigueros CA , Camenzind T , et al.  . 2019 . Why farmers should manage the arbuscular mycorrhizal symbiosis . The New Phytologist 222 : 1171 – 1175 .

Rimington WR , Pressel S , Duckett JG , Bidartondo MI. 2015 . Fungal associations of basal vascular plants: reopening a closed book ? The New Phytologist 205 : 1394 – 1398 .

Rodríguez M , Torres M , Blanco L , Béjar V , Sampedro I , Llamas I. 2020 . Plant growth-promoting activity and quorum quenching-mediated biocontrol of bacterial phytopathogens by Pseudomonas segetis strain P6 . Scientific Reports 10 : 4121 .

Rousk J , Bååth E. 2011 . Growth of saprotrophic fungi and bacteria in soil . FEMS Microbiology Ecology 78 : 17 – 30 .

Ruiz-Lozano JM , Bonfante P. 2000 . A Burkholderia strain living inside the Arbuscular mycorrhizal fungus Gigaspora margarita possesses the vacB gene, which is involved in host cell colonization by bacteria . Microbial Ecology 39 : 137 – 144 .

Ryan MH , Graham JH. 2018 . Little evidence that farmers should consider abundance or diversity of arbuscular mycorrhizal fungi when managing crops . The New Phytologist 220 : 1092 – 1107 .

Sangwan S , Prasanna R. 2022 . Mycorrhizae helper bacteria: Unlocking their potential as bioenhancers of plant–arbuscular mycorrhizal fungal associations . Microbial Ecology 84 : 1 – 10 .

Santos LF , Olivares FL. 2021 . Plant microbiome structure and benefits for sustainable agriculture . Current Opinion in Plant Biology 26 : p100198 .

Scheublin TR , Sanders IR , Keel C , van der Meer JR. 2010 . Characterisation of microbial communities colonising the hyphal surfaces of arbuscular mycorrhizal fungi . The ISME Journal 4 : 752 – 763 .

Selosse M-A , Bessis A , Pozo MJ. 2014 . Microbial priming of plant and animal immunity: symbionts as developmental signals . Trends in Microbiology 22 : 607 – 613 .

Shao D , Smith DL , Kabbage M , Roth MG. 2021 . Effectors of plant necrotrophic fungi . Frontiers in Plant Science 12 : 1 .

Silva KPT , Chellamuthu P , Boedicker JQ. 2017 . Quantifying the strength of quorum sensing crosstalk within microbial communities . PLoS Computational Biology 13 : e1005809 .

Smith SE , Read DJ. 2008 . Mycorrhizal symbiosis , 3rd edn. London : Academic Press .

Smith SE , Smith FA. 2011 . Roles of arbuscular mycorrhizas in plant nutrition and growth: new paradigms from cellular to ecosystem scales . Annual Review of Plant Biology 62 : 227 – 250 .

Smith SE , Anderson IC , Smith FA. 2015 . Mycorrhizal associations and phosphorus acquisition: from cells to ecosystems . Annual Plant Reviews 48 : 409 – 439 .

Spanu PD , Panstruga R. 2017 . Editorial: biotrophic plant–microbe interactions . Frontiers in Plant Science 8 : 192 .

Spraker JE , Sanchez LM , Lowe TM , Dorrestein PC , Keller NP. 2016 . Ralstonia solanacearum lipopeptide induces chlamydospore development in fungi and facilitates bacterial entry into fungal tissues . The ISME Journal 10 : 2317 – 2330 .

Stasulli NM , Shank EA. 2016 . Profiling the metabolic signals involved in chemical communication between microbes using imaging mass spectrometry . FEMS Microbiology Reviews 40 : 807 – 813 .

Steffan BN , Venkatesh N , Keller NP. 2020 . Let’s get physical: bacterial–fungal interactions and their consequences in agriculture and health . Journal of Fungi 6 : 243 .

Stürmer SL. 2012 . A history of the taxonomy and systematics of arbuscular mycorrhizal fungi belonging to the phylum Glomeromycota . Mycorrhiza 22 : 247 – 258 .

Tan CH , Koh KS , Xie C , et al.  . 2014 . The role of quorum sensing signalling in EPS production and the assembly of a sludge community into aerobic granules . The ISME Journal 8 : 1186 – 1197 .

Taylor TN , Klavins SD , Krings M , Taylor EL , Kerp H , Hass H. 2003 . Fungi from the Rhynie chert: a view from the dark side . Transactions of the Royal Society of Edinburgh: Earth Sciences 94 : 457 – 473 .

Tedersoo L , Bahram M , Zobel M. 2020 . How mycorrhizal associations drive plant population and community biology . Science 367 : eaba1223 .

Thirkell TJ , Pastok D , Field KJ. 2019 . Carbon for nutrient exchange between Arbuscular mycorrhizal fungi and wheat varies according to cultivar and changes in atmospheric carbon dioxide concentration . Global Change Biology 26 : 1725 – 1738 .

Toro M , Azcón R , Barea JM. 1997 . Improvement of arbuscular mycorrhiza development by inoculation of soil with phosphate-solubilising rhizobacteria to improve rock phosphate bioavailability (P 32 ) and nutrient cycling . Applied and Environmental Microbiology 63 : 4408 – 4412 .

Tylka GL , Hussey RS , Roncadori RW. 1991 . Axenic germination of vesicular–arbuscular mycorrhizal fungi: Effects of selected Streptomyces species . Phytopathology 81 : 754 – 759 .

Uehling JK , Entler MR , Meredith HR , et al.  . 2019 . Microfluidics and metabolomics reveal symbiotic bacterial–fungal interactions between Mortierella elongata and Burkholderia include metabolite exchange . Frontiers in Microbiology 10 : 21163 .

van Arle IM , Olssen PA. 2003 . Fungal lipid accumulation and development of mycelial structures by two arbuscular mycorrhizal fungi . Applied and Environmental Microbiology 69 : 6762 – 6767 .

van der Heijden MGA , Martin FM , Selosse MA , Sanders IA. 2015 . Mycorrhizal ecology and evolution: the past, the present, and the future . New Phytologist 205 : 1406 – 1423 .

Vivas A , Barea JM , Azcón R. 2005 . Brevibacillus brevis isolated from cadmium- or zinc-contaminated soils improves in vitro spore germination and growth of Glomus mosseae under high Cd or Zn concentrations . Microbial Ecology 49 : 416 – 424 .

Viveganandan G , Jauhri KS. 2000 . Growth and survival of phosphate-solubilizing bacteria in calcium alginate . Microbiological Research 155 : 205 – 207 .

Vosátka M , Gryndler M. 1999 . Treatment with culture fractions from Pseuudomonas putida modifies the development of Glomus fistulosum mycorrhiza and the response of potato and maize plants to inoculation . Applied Soil Ecology 11 : 245 – 251 .

Wang F , Feng G. 2021 . Arbuscular mycorrhizal fungi interactions in the rhizosphere . In: Gupta VVSR , Sharma AK , eds. Rhizosphere Biology: interactions between microbes and plants . Singapore : Springer , 217 – 235 .

Wang F , Shi N , Jiang RF , Zhang FS , Feng G. 2016 . In situ stable isotope probing of phosphate-solubilising bacteria in the hyphosphere . Journal of Experimental Botany 67 : 1689 – 1701 .

Warmink JA , Van Elsas JD. 2009 . Migratory response of soil bacteria to Lyophyllum sp strain Karsten in soil microcosms . Applied and Environmental Microbiology 75 : 2820 – 2830 .

Whitehead NA , Barnard AML , Slater H , Simpson NJL , Salmond GPC. 2001 . Quorum-sensing in Gram-negative bacteria . FEMS Microbiology Reviews 25 : 365 – 404 .

Williams A , Pétriacq P , Beerling DJ , Cotton TEA , Ton J. 2018 . Impacts of atmospheric CO 2 and soil nutritional value on plant responses to rhizosphere colonization by soil bacteria . Frontiers in Plant Science 9 : 1493 .

Xavier LJC , Germida JJ. 2003 . Bacteria associated with Glomus clarum spores influence mycorrhizal activity . Soil Biology and Biochemistry 35 : 471 – 478 .

Xie ZP , Staehelin C , Vierheilig H , et al.  . 1995 . Rhizobial nodulation factors stimulate mycorrhizal colonisation of nodulating and nonnodulating soybeans . Plant Physiology 108 : 1519 – 1525 .

Yang L , Liu Y , Wu H , Høiby N , Molin S , Song Z. 2011 . Current understanding of multi‐species biofilms . International Journal of Oral Science 3 : 74 – 81 .

Zarei M , Rahimi S , Saris PEJ , Yousefvand A. 2022 . Pseudomonas fluorescens group bacterial strains interact differently with pathogens during dual-species biofilm formation on stainless steel surfaces in milk . Frontiers in Microbiology 13 : 1 – 9 . doi: 10.3389/fmicb.2022.1053239

Zhang L , Feng G , Declerck S. 2018a . Signal beyond nutrient, fructose, exuded by an arbuscular mycorrhizal fungus triggers phytate mineralisation by a phosphate solubilising bacterium . ISME Journal 12 : 2339 – 2351 .

Zhang L , Shi N , Fan J , Wang F , George TS , Feng G. 2018b . Arbuscular mycorrhizal fungi stimulate organic phosphate mobilisation associated with changing bacterial community structure under field conditions . Environmental Microbiology 20 : 2639 – 2651 .

Zhang S , Lehmann A , Zheng W , You Z , Rillig MC. 2018c . Arbuscular mycorrhizal fungi increase cereal grains: a meta-analysis . New Phytologist 222 : 543 – 555 .

Zhang W , Li XG , Sun K , et al.  . 2020 . Mycelial network-mediated rhizobial dispersal enhances legume nodulation . The ISME Journal 14 : 1015 – 1029 .

Zhang L , Zhou J , George TS , Limpens E , Feng G. 2022 . Arbuscular mycorrhizal fungi conducting the hyphosphere bacterial orchestra . Trends in Plant Science 27 : 402 – 411 .

Author notes

Email alerts, citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1095-8290
• Print ISSN 0305-7364
• Copyright © 2024 Annals of Botany Company
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Shaping the leaf microbiota: plant-microbe-microbe interactions

Affiliations.

• 1 Department of Microbial Interactions, IMIT/ZMBP, University of Tübingen, Tübingen, Germany.
• 2 Max Planck Institute for Plant Breeding Research, Köln, Germany.
• 3 Institute for Plant Sciences and Cluster of Excellence on Plant Sciences (CEPLAS), University of Cologne, Center for Molecular Biosciences, Cologne, Germany.
• PMID: 32910810
• PMCID: PMC8210630
• DOI: 10.1093/jxb/eraa417

The aerial portion of a plant, namely the leaf, is inhabited by pathogenic and non-pathogenic microbes. The leaf's physical and chemical properties, combined with fluctuating and often challenging environmental factors, create surfaces that require a high degree of adaptation for microbial colonization. As a consequence, specific interactive processes have evolved to establish a plant leaf niche. Little is known about the impact of the host immune system on phyllosphere colonization by non-pathogenic microbes. These organisms can trigger plant basal defenses and benefit the host by priming for enhanced resistance to pathogens. In most disease resistance responses, microbial signals are recognized by extra- or intracellular receptors. The interactions tend to be species specific and it is unclear how they shape leaf microbial communities. In natural habitats, microbe-microbe interactions are also important for shaping leaf communities. To protect resources, plant colonizers have developed direct antagonistic or host manipulation strategies to fight competitors. Phyllosphere-colonizing microbes respond to abiotic and biotic fluctuations and are therefore an important resource for adaptive and protective traits. Understanding the complex regulatory host-microbe-microbe networks is needed to transfer current knowledge to biotechnological applications such as plant-protective probiotics.

Keywords: Biofilm; innate immunity; microbe–microbe interaction; microbial colonization; phyllosphere; quorum sensing.

© The Author(s) 2020. Published by Oxford University Press on behalf of the Society for Experimental Biology.

Publication types

• Research Support, Non-U.S. Gov't
• Microbial Interactions
• Microbiota*
• Plant Leaves

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Int J Mol Sci

Plant–Microbe Interaction: Aboveground to Belowground, from the Good to the Bad

Associated data.

Not applicable.

Soil health and fertility issues are constantly addressed in the agricultural industry. Through the continuous and prolonged use of chemical heavy agricultural systems, most agricultural lands have been impacted, resulting in plateaued or reduced productivity. As such, to invigorate the agricultural industry, we would have to resort to alternative practices that will restore soil health and fertility. Therefore, in recent decades, studies have been directed towards taking a Magellan voyage of the soil rhizosphere region, to identify the diversity, density, and microbial population structure of the soil, and predict possible ways to restore soil health. Microbes that inhabit this region possess niche functions, such as the stimulation or promotion of plant growth, disease suppression, management of toxicity, and the cycling and utilization of nutrients. Therefore, studies should be conducted to identify microbes or groups of organisms that have assigned niche functions. Based on the above, this article reviews the aboveground and below-ground microbiomes, their roles in plant immunity, physiological functions, and challenges and tools available in studying these organisms. The information collected over the years may contribute toward future applications, and in designing sustainable agriculture.

1. Introduction

Plants are colonized aboveground and belowground by mutualistic and parasitic organisms. These organisms can be categorized into groups based on areas of colonization; for instance, microorganisms that colonize the external parts of the plant are generally known as epiphytes, while those that colonize the inside of the plants are endophytes. Furthermore, there are phyllosphere organisms that colonize the leaf surface; and the most abundant group of them would be the rhizosphere inhabitants, which colonize regions closest to the root system [ 1 , 2 , 3 ]. This region is teaming with microbes, which are attracted to the root systems due to their exudates. The exudates depend on the developmental stages and physiological statuses of the plants [ 4 , 5 ]. Although the recruitment of microbes to the root region may be a consequence of plant exudation, the microorganisms that colonize this region have diverse roles in supporting plant growth, development, and inhibition of host pathogens. This implies interdependency between the host and microbes in the aboveground and belowground interactions [ 6 , 7 ].

The microbial diversification, speciation, structural complexity, and interactions that surround the root systems make it essential to understand the microbial population as well as the root architecture, to have a clear view on how these interactome associate [ 8 ]. Due to the high levels of interaction between the plant and microbes, these components are observed as holobionts or metaorganisms [ 9 , 10 ]. In addition to the intertwining plant and microbe associations (plant-microbe–plant), there is the microbe–microbe and microbe–soil association. The complexity of the microbes in soil is not just circumnavigated by the plant, but by the environment and other constituents in the soil. The physicochemical and biological components of the soil largely influence the microbiome. For instance, climate change and its effect on agriculture, such as drought and flooding, severely impact soil microbiomes. Furthermore, physical changes in temperature, pH, oxygen level, and soil structure also affect its inhabitants [ 11 ]. In addition, the chemical compounds derived through the cycling of materials in the soil or from agricultural practices also affects soil biology. Microbes that adapt to a particular stress condition might be beneficial to plants, since beneficial microbes are shown to increase soil health and fertility. This is not inclusive of the role played by macro- and microorganisms belowground and aboveground in influencing the plant-microbe interaction (such as animal grazing, etc.) [ 5 ].

Beneficial microorganisms can be inoculated in soil or used as input to improve agricultural practices. Microbial inoculants are administered to the plant or the soil to boost crop productivity and health, and mitigate the negative effects of agrochemicals. It is a viable alternative to chemical treatment and is capable of promoting plant development, controlling pests and diseases, and stabilizing soil structure. These inputs may be employed as biocontrol agents, biopesticides, bioherbicides, and biofertilizers. Over the last few decades, significant developments have been made in manufacturing, marketing, and use of inoculants [ 12 ]. Nowadays, the use of inoculants is more widespread, owing to the availability of excellent and multifunctional strains in the market, improving yield at a lower cost than synthetic fertilizers. Rhizobia are the most extensively utilized microbes as inoculants [ 13 ]. The legume–rhizobia symbiosis influences the mechanism of biological nitrogen fixation (BNF), which satisfies the plant’s N needs [ 12 ]. Plant growth-promoting bacteria (PGPB) can support a plant in a range of areas on its own or in combination with other factors. PGPB influences plants through phytohormones and siderophores synthesis, phosphate solubilization, and elicitation of a plant’s internal defense against biotic and abiotic stressors [ 14 , 15 ]. Various microorganisms are increasingly being employed in agriculture for ecological pest and disease management [ 16 ].

The recent surge in new technologies in genome studies has enabled us to further characterize the microbial diversity, genome, and proteome of microorganisms living in association in soil or on plants. DNA/RNA, genome analysis, transcriptome, proteome, metagenome, and all other omics-based technologies have provided a means to dissect beneficial and non-beneficial plant-microbe interactions at depths and speeds that were not possible decades ago. These technologies enable us to understand the dynamics belowground and aboveground, to further utilize this information to improve growth, yield, and disease reduction [ 17 ]. If we are able to decipher the factors responsible for the establishment of the microbial communities in the rhizosphere, we will be able to utilize this information in designing sustainable ecosystems that are beneficial, stable, and productive for the long haul [ 18 , 19 ]. Despite the fact that the ecosystem was sustainable prior to interference, this approach restores the environment to its original state before human intervention. Hence, given the above background, this current review focuses on the aboveground and belowground microbial interactions, the development of diseases and emerging threats, the beneficial uses of microbes, and the available new tools to study them at greater depth (see Supplementary Figure S1 for the methodology of Systematic Review for Plant Microbe Interactions).

2. The Two-Phase Microbial Communities

2.1. aboveground microbes.

Endophytic and epiphytic groups of microorganisms colonize and inhabit plant tissues, such as leaves and flowers [ 20 ]. The phyllosphere organisms (those that are on external plant surfaces) will generally be influenced by the environment and may be commensal-like organisms, or organisms that can cause disease. The phyllosphere is a severe and unstable environment with oligotrophy-like features, such as nutritional restriction in carbon and nitrogen, as well as numerous, highly variable physicochemical limitations (light penetration, UV radiation, temperature, desiccation) [ 21 ]. Microbial adjustment to the phyllosphere environment appears to be dependent on a number of factors linked to a range of physicochemical and biotic limitations, such as exposure to air, water, soil, animal, or insect borne microbes [ 1 , 22 , 23 ]. Meanwhile endophytic organisms systematically obtain their microbial nutrients through the xylem and aerial tissue, such as fruits and flowers [ 22 , 24 ]. The distribution of the endophytes within the plant tissue will largely depend on the nutritional source within the organ to support the growth and development of the endophytes. There will be observable differences in the genera between endophytes and the phyllosphere inhabitants [ 25 ]. For example, in a study conducted on tomato plants, an Acinetobacter dominant community was reported in stems and leaves, while tissues of stems and leaves were colonized by Xanthomonas , Rhizobium , Methylobacterium , Sphingomonas , and Pseudomonas [ 2 ]. When the tomato was compared to other host plants, Bacillus and Pantoea dominated the lettuce phyllosphere. In potato phyllosphere, Devosia, Dyadobacter , and Pedobacter were dominant, while Pseudomonas dominated the spinach phyllosphere [ 2 ]. In another study conducted on maize, the phyllosphere was dominated by Sphingomonas and Methylobacteria [ 26 ]. From this observation, we can conclude that the microbes dominating a particular plant depends on the host tissue, geographical location, tissue nutrient content, and the physicochemical characteristics of the soil [ 27 ]. While endophytes in tomatoes are diverse, Acinetobacter , Enterobacter , Pseudomonas, and Pantoea were identified as the most dominant, with varying densities in different vegetative tissues. Enterobacter , Pseudomonas, and Pantoea were also commonly found in other host plants [ 2 , 28 ]. However, in grapes, the phyllosphere was inhabited by Pseudomonas , Sphingomonas , Frigoribacterium , Curtobacterium , Bacillus , Enterobacter , Acinetobacter , Erwinia , Citrobacter , Pantoea , and Methylobacterium, while the endophytes were dominated by Ralstonia , Burkholderia , Pseudomonas , Staphylococcus , Mesorhizobium , Propionibacterium , Dyella , and Bacillus [ 2 , 29 ]. Aleklett et al. [ 30 ] identified Pseudomonas and Enterobacteriaceae taxa on apple flowers. Pseudomonads were the most abundant genus in floral organs of several fruits, such as apples, grapefruits, and pumpkins.

Resistance and tolerance responses towards antibacterial and immunological chemicals generated by plant tissues, as well as competing microbes, can be developed by epiphytic microorganisms [ 31 ]. In the phyllosphere of tobacco, epiphytic bacteria with enzymes, which degrade N-acylhomoserine lactone (AHL) and quorum-sensing signals, have been discovered; thus, it was proposed that signaling circuits may be associated with the formation of complex epiphyllic microbial communities [ 32 ]. Epiphytic microbes could also evolve pathways for aggregation or exopolysaccharide production, to increase adherence or resistance to desiccation [ 33 ]. Epiphytic organisms could also produce and release phyto-hormonal chemicals, such as indole-3-acetic acid (IAA), by stimulating the loosening of plant cell walls and releasing saccharides from plant cell walls [ 34 ]. In conclusion, the relationship between the plant host and aboveground microorganisms is host dependent and largely influenced by the environment and signaling circuits associated to the microbial communities. These organisms may provide beneficial or detrimental relationships with the host, resulting in either enhanced growth and defense or the elicitation of disease and losses, respectively [ 25 , 35 ].

2.2. Microbes from Belowground

The rhizosphere is a highly ubiquitous region, with plant exudates, which recruits microbes, creating a microbial reservoir [ 8 ]. The microorganisms in the rhizosphere constantly interact with one another, resulting in commensalism, parasitism, amensalism, saprophytism, and symbiotic associations. These organisms affect the aboveground activities and are part of the bulk soil species. Plants and the environment of the soil determine the soil microbial communities [ 36 ]. However, nothing is known about how the rhizosphere composition is chosen from bulk soil. Two mechanisms might explain the population—the neutral or niche-based mechanism. The neutral mechanism is based on the fact that most organisms are able to exploit most soil niches and, therefore, are limited by the distance among plants, recruitment parameters, and hampered dispersal [ 37 ]. However, for the niche-based mechanism, environmental changes alter the microbial communities [ 38 ]. Plants have the tendency to recruit microorganisms to the rhizosphere that will assist with biological functions, such as nutrient uptake, growth, and development. The example seen in cereals is the niche community, where richness and microbial abundance is strongly dependent on the ecological space of the rhizosphere [ 39 ]. Similarly, the orchid microflora studies, starting from seed germination through establishment, reproduction, and survival of orchids are heavily reliant on orchid mycorrhizal fungi (OMF). Corollary, changes in OMF composition and abundance can have a substantial influence on dispersion and fitness of orchids [ 40 ], providing another example of a niche-based mechanism.

The recruitment agent, exudates/mucilages from the roots, release amino acids, cutin monomers, flavonoids, hormones, organic acids, polyphenols, sugars, and nutrients that are involved in moderating the plant-microbe interactions and microbial gene expression [ 23 , 41 , 42 ]. These chemicals act as signals; and are deployed to initiate the microbial colonization of roots. While some of these compounds are elicited to enhance plant growth and development, secondary metabolites, such as benzoxazinoids in maize roots, are produced to specifically inhibit Actinobacteria and Proteobacteria [ 8 , 43 ]. The recruitment of microbiota to the root is mobilized when the machinery involved in biofilm formation, chemotaxis, detoxification, mobility, polysaccharide degradation, and secondary metabolism is switched on [ 44 ]. The microbiome begins to expand, establishes niches, and recruits additional microbes through a cross-feeding approach, resulting in new niche groups being developed within the population [ 45 ]. Once the microbial population establishes a community around the root, the plant’s exudate shifts its focus toward enhancing the formation of biofilms around the roots [ 46 ].

Besides the variations that may be observed between plant genera, the differences in variety and genotype will also affect the chemical constituents in the root exudate [ 47 , 48 ]. As mentioned above, these exudates are a blend of molecules, which are influenced by plant size, genotype, photosynthetic activity, and soil conditions. These phytochemicals influence the diversity and the composition of the microorganism around the root [ 49 ], where the amalgamated exudate modifies bacterial assemblage in the rhizosphere. When the influence of species was tested in the angiosperms, variation was observed in the Pseudomonas species occurring in the soil. This variation in soil species was also influenced by the spatiotemporal and physiochemical organization of the rhizosphere [ 25 , 50 ].

In addition to the microbes found surrounding the root system, there are also the endophytic microbes in plants. The root source components play deeply into the colonization of endophytes in the plants. Some of these endophytes have important symbiotic uses in agriculture. One example is the Piriformospora indica , which causes elevated phosphorous [P] uptake and protects against various stressors in plants [ 51 ]. Further, Gill et al. [ 51 ] reported that cyclophilin A–like from P. indica was overexpressed in protecting against salt stress in tobacco plants. When working in concert, Azotobacter chroococcum and P. indica help with nutrient acquisition and synergy in action [ 52 ]. Some of these endophytic organisms were also responsible in chemotaxis activities. In tomatoes, the non-pathogenic Fusarium oxysporum reduced the occurrence of nematodes [ 53 ]. When biochar was used in tomato plants, it absorbed the exudates and created a strong chemotactic signal towards Ralstonia solanacearum , suppressing its swarming ability [ 54 ]. Collectively, the mechanisms, functions, and communication signals in root–microbe interactions were reviewed in other publications, detailing the intricacies of these interactions [ 25 , 46 ].

2.2.1. Root-Root Interaction

Plant species and genotypes have strong specificity of exudates that are able to influence the neighboring plants. Little is known on how these signals are transmitted and received by both root and microorganism. These root exudates have been implicated in several processes, including influencing nutrient availability [ 49 ] and mediating nutrient competition. Plant exudates have been reported to increase mineralization. The presence of certain acids in the soil (phosphatases, carboxylase) improved ion availability to the plant [ 55 ], and indirectly influenced N 2 -cycling. The absence of these acids in exudates will influence N availability in soils [ 56 ] explaining the difference in nutrient acquisition between plants. In addition, plants are not only influenced by its own exudates but are influenced by their neighboring plants. For instance, in intercropping where leguminous and non-leguminous plants are grown together, the release of carboxylates by legumes resulted in enhanced P nutrition and growth to neighboring plants [ 57 ]. In phosphorous deprived soils, Zemunik et al. [ 58 ] reported that the phosphorous levels were replenished through the influx of carboxylate exudates by arbuscular mycorrhizal fungi (AMF) and plant roots. In plants such as cucumber, citric acids from the roots attracted Bacillus amyloliquefaciens , and fumaric acid from banana roots attracted Bacillus subtilis resulting in biofilm production [ 59 ].

The root–root interaction is not limited to the regulation of nutrient acquisitions. These interactions also influence the root growth of neighboring plants through allelopathy, where released phytotoxins are able to reduce the growth or survival of neighbors and, therefore, reduce competition of resources. A classic example of toxins is catechin, which inhibits germination, root growth, and development [ 60 , 61 ]. Volatile organic compounds (VOCs) also function as allelochemicals to regulate rhizosphere signaling by mycorrhiza networks [ 62 , 63 ]. Allelopathic plants are mostly resistant to their own phytotoxins and, therefore, act specifically on other plant species at different levels of effectiveness. However, there are certain non-allelopathic neighbors that can be resistant [ 64 ]. The exudates around the roots are controlled by the rhizobiome, which affects the quantity, composition, and possibility of feedback regulation between the plant and microbiome. From the above observation, it would appear that the root–root interaction is one that is competitive and not niche [ 65 ]. Competition between neighboring plants is seen in species that have root systems spread across a wider region horizontally than those that have vertical and deep root systems. Furthermore, when legumes, non-legumes, and interspecies studies were conducted to evaluate root–root interactions, these plant exudates influenced each other and the microbial communities within their vicinity [ 66 , 67 ].

The root–root interactions may show the presence of major bacterial communities and AMF in the soil [ 36 ]. The bacteria may largely be responsible for hormone induced growth and antibiotics based on inhibition of negative organisms. The AMF, on the other hand, is an important component that plays an important role in water and nutrient absorption. This organism has the ability to maintain some level of nutrient absorption despite competition in soil; thus, it maintains the plant community composition [ 68 , 69 ]. In addition to the positive interactions found in the soil, the ecosystem is home to soilborne pathogens. These pathogens affect the plant soil feedback by affecting the plant growth, nutrient accumulation, and other processes, resulting in a prolonged negative impact on the soil microbial population due to the presence of the pathogen and the application of fungicide [ 70 , 71 ]. This will cause a shift on the hierarchies of different soil microbial communities, depending on the changes in the soil from environment, human activity, and disease. While it is still unclear on how pathogens affect nutrient uptake, it is possible that it negatively affects plants through root growth and resource uptake per unit root [ 72 ].

2.2.2. Root–Microbe Interactions

The root–microbe interactions can be addressed as symbiotic and parasitic interactions. In this section, we explore the beneficial interactions established between the root–microbe, such as: [ 1 ] rhizobacteria–legume; [ 2 ] actinobacteria–root; [ 3 ] mycorrhizal–root, and [ 4 ] other root–microbe interactions; of these, the most widely studied relationship is between legumes and rhizobacteria. These organisms produce Nod factors, perceived by the plant receptors as inducing activation of the pathway, resulting in nodule formation from the differentiation of pericycle and cortical cells [ 25 , 73 ]. The bacterium then uses this organ as a processing site of atmospheric nitrogen into ammonia, which is then used in protein synthesis. This process is regulated by feedback inhibition to reserve energy and inhibit N 2 -fixation when the supply of nitrogen is sufficient [ 74 ].

As Rhizobium interactions are not established with every plant, certain model organisms have been used in understanding the changes that occur within the rhizobium and nodule. Larrainzar and Wienkoop [ 75 ], Lorite et al. [ 76 ], and Wan et al. [ 77 ] conducted a proteome analysis on the genome sequence of Sinorhizobium meliloti, a symbiont of M. truncatula, Mesorhizobium loti , the symbiont of L. japonicus, and Bradyrhizobium japonicum , the symbiont of soybean, and compared it with the free-living organisms. These analyses enabled them to answer some very important questions on the recognition of the host by the rhizobacteria, nutrient exchange, and the control of nodulation. It is believed that the flavonoids released by the legumes into the soil are able to activate the NodD protein, which in turn sets into motion a series of nodulation genes encoded by Sym plasmids. From the studies conducted by the above researchers and others, it is concluded that flavonoids are responsible for the varied responses of several bacterial genes.

The rhizobacteria provides signaling chemicals that act on their host. A study on Medicago truncatula disclosed that there was change in protein content within the nodule during the formation of leghemoglobin and enolase isoforms [ 31 ]. In other legumes, R. leguminosarum induced ethylene responsive proteins. Ethylene is a major regulator of plant defense response. In the ethylene-insensitive mutant of M. truncatula ( skl ), hypernodulation was observed in the roots, likely due to a compromised immune system. It was reported that the skl mutant had a defective ethylene pathway. Therefore, we hypothesize that ethylene is responsible for the symbioses and nodulation of the host. Further, when the root system in soybean was examined post inoculation with B. japonicum , there were rhizobial proteins that were necessary for the induction of the Nod factors detected in the roots [ 77 ]. Elevated levels of calcium-dependent protein kinase (CDPK) were observed and were expected to trigger the activation of symbioses [ 78 ]. The presence of peroxidases, lipoxygenases, phospholipases, and lectins indicate a possible role for lectins in attachment, and lipids in the early infection process of rhizobacteria in the root system [ 79 ]. While the success of the nodulation process is reliant on the lack of defense response against the rhizobacteria, a proteomic study of the M. truncatula root colonized by S. meliloti identified pathogenesis related protein (PR10) isoforms [ 80 ]. These proteins were implicated in phytohormone interactions, and ligand binding, influencing specifically auxin and cytokinin activity in plant meristem [ 79 , 81 ].

Nutrient exchange is another factor that influences the rhizobacteria-legume relationship. Rhizobacteria are present as bacteroids in the plant symbiosomes where all nutrient exchange is controlled by the composition of bacteroids and peribacteroid membranes. When the proteins in these membranes were analyzed, heat shock proteins, proteases, nodulins, transporters, receptor kinases, plant defense, and signaling proteins were identified, indicating that nodulation is an ongoing, complex process, where the plant’s defense mechanism is continuously regulated to allow for the nodulation in the roots [ 82 ]. The nodules contain enzymes required for C utilization, N 2 -fixation, heme synthesis, transporters, and stress related proteins. The differences shown by the ABC transporter in free-living and nodule inhabiting bacteria imply that they are responsible for specialized functions in nutrient transfer and nodulation [ 83 ]. The proteome studies are indicative that bacteroids enhanced nitrogen and carbon metabolism while suppressing fatty acid and nucleic acid metabolism [ 84 ]. However, transcriptome studies observed expression of high levels of aquaporins, ATPases, metal binding proteins, nutrient transporters (carbon, nitrogen, potassium, and sulfate), osmoregulators, and regulatory proteins in the nodules [ 85 , 86 ]. All of these components are useful in maintaining homeostasis within the nodule, to facilitate the transmembrane transport of nutrients and proteins.

The proteomes of nodules have also been studied under stress circumstances. Drought is a primary stressor that prevents nodules from fixing nitrogen. The metabolic enzymes, such as sucrose synthase, amino acid synthesis enzymes, and leghemoglobin, which regulates oxygen levels within the nodule, were reduced in drought-stressed M. truncatula nodules [ 87 ]. Drought results in an increase in protein accumulation in the bacteroid fractions, including enhanced protein synthesis components, in contrast to the decreased protein accumulation in the host [ 88 ]. As nodulation is a costly process for the plant, it is presumed that this process is auto-regulated through signaling mechanisms. A suggested mode of control involves auxin transport from the shoot to the root to regulate nodule numbers as seen from the differential expressions of auxin inducible proteins in mutant and wild type M. truncatula [ 89 , 90 ]. The second possibility is the regulation of the plant defense mechanisms that may arrest or inhibit nodulation [ 91 ].

Actinomycetes are also able to form symbiotic relationships with host plants. Some symbiotic relationships of this group of bacteria have been reported in Angiosperms, such as Alnus, Casuarina, and Datisca genera [ 92 ]. A study of proteomes in Frankia alni and Alnus sp. identified secreted proteins, which were generally hydrolytic enzymes believed to play a role in the formation of this symbiotic relationship [ 92 , 93 ]. Another group of organisms involved in the symbiotic relationship with plants is the mycorrhizal fungi. These fungi invade the root systems and establish the arbuscular in the cortical cells or extracellular hyphal structures (ectomycorrhiza or EM). The AMF are the most dominant of these fungal interactions [ 93 , 94 ]. Like in the rhizobacteria, AMF remains separated in the plant by a membrane, which does not hamper the nutrient exchange between the host and fungus. AMF provides the phosphorus to the plant in exchange for carbon and lipids [ 95 ]. The carbon supply to the symbiont is feedback-regulated to limit excessive loss of nutrients from the host [ 96 ]. A proteome analysis on the root of M. truncatula colonized by Glomus mosseae exhibited redox, stress, respiration, and cell wall modifications, all necessary changes to facilitate the colonization of the host root system by Glomus mosseae.

A differential expression of proteins was observed in wild type and mutant [ dmi3 ] M. truncatula proteins inoculated with G. intraradices [ 97 ]. Proteins, such as lipoxygenases, thioredoxins, and ATPases, were identified through proteomic and transcriptome analyses. Further, studies on these proteins showed the importance of transporters (nutrient and water) [ 98 ], and metabolism (amino acids, fatty acids, and carotenoids) in AMF infected roots [ 96 ]. When the G. intraradices infected wild type, dim3, and sunn mutants of M. truncatula were analyzed, proteins that were specifically induced or reduced were chalcone reductase, a 2,4-D-inducible glutathione transferase, a glutathione-dependent dehydroascorbate reductase and a cyclophilin [ 97 ]. These proteins postulate the importance of this symbiotic relationship in redox and defense mechanism facilitation for healthy plant growth and stress management. With the presence of annexins, alcohol dehydrogenases, and profucosidases—there is the possibility of the mycorrhizal infection playing a role in detoxification, in addition to stress response [ 99 ].

One area of symbiotic relationship that is extensively studied is the relationship between free-living organisms, such as Trichoderma and its positive impact on plant host protection from disease, induction of immune response, and improved plant growth. Trichoderma has become an important biological control agent as this genus has the ability to parasitize other fungi through diverse mechanisms. T. harzianum , an extensively studied species, produces proteases that are able to degrade fungal cell walls. T. asperellum was known to induce the production of proteins related to disease and the defense response pathway [ 5 , 43 , 63 , 100 ]. Additionally, these organisms resulted in an increase in levels of isoprenoid, ethylene biosynthesis, energy metabolism, and protein folding. T. atroviride, T. harzianum , and T. asperellum were reported to have elevated levels of disease resistance proteins, such as chitinases and cyclophilins, which provides heightened resistance in the plants [ 101 ].

2.2.3. Microbe–Microbe

Communication between microbe–microbe can be addressed as (1) between pathogenic microbes; (2) pathogens and endophytes; (3) succession by microbes; and (4) lifestyle changes in the environment. Pathogens are able to affect microbial community on plant surfaces and soil. In maize exhibiting SLB infection, it was observed that the resident microbial community was reduced in richness [ 102 ]. When infected by a pathogen, the host is open to infection by others, even non-pathogenic microbes due to increased susceptibility. When infected with white rust, Brassicaceae were more susceptible to mildew pathogens and, hence, easily succumbed to white rust [ 103 , 104 ]. In A. thaliana , Albugo laibachii , was observed to have increased susceptibility to non-host pathogen Phytophthora infestans [ 105 ]. Bacterial populations utilized quorum sensing (QS) and biofilm formation as a means to establish beneficial plant-microbe interactions [ 106 , 107 ]. QS mediated by QS signals between pathogenic organisms were implicated in increasing the pathogenicity and virulence of these microbes on the host [ 108 ]. One such example is when QS resulted in Phytophthora nicotianae zoospore aggregation, which resulted in heightened pathogenicity [ 64 ]. However, while interaction between the species of microbes in the infection process is evident, each microbial cell is responsible for the successful colonization and disease progression in a host [ 109 ]. Many genes have been identified in bacteria, responsible for the formation of biofilm, colonization of the roots, and improved growth [ 110 ]. The QS signals produced by bacteria are able to effect plant transcriptome and proteome [ 79 ] by adhering to the environment and plant surfaces and, thus, impacting the processes within the plant [ 111 ].

Busby et al. [ 112 ] observed that foliar pathogens might be inhibited by endophytes through hyperparasitism, competition, and/or antibiosis. These endophytes produce a list of chemicals that are toxic to microbes and can prevent pathogen infiltration [ 113 ]. Some of these interactions between endophytes and pathogens are direct. Jakuschkin et al. [ 114 ] reported in his study that fungal endophytes acted antagonistically against powdery mildew of Erysiphe sp. The presence of chemical constituents, such as polyketide synthase, are natural antibiotics in endophytes that enable these organisms to act as biocontrol agents against pathogens [ 115 ]. T. atroviride , Ulocladium atrum , Stachybotrys sp., and Truncatella angustata were shown to generate quantitative disease resistance in P. trichocarpa against Melampsora rust pathogen by Raghavendra and Newcombe [ 116 ]. The afore-mentioned fungi, on the other hand, were found to be relatively uncommon in wild P. trichocarpa [ 112 ], suggesting that disease-modifying effects of foliar fungus differs between wild and experimental settings. Endophytes also employ QS to inhibit harmful bacteria through the expression of QS inhibitors (QSIs) or quorum-quenching (QQ) enzymes to prevent signaling molecules from working. The plant pathogens Erwinia carotovora , Bacillus thuringiensis , and Enterobacter asburiae have all been inhibited by the AHL lactonase enzyme (a powerful QQ) found in endophytic bacteria [ 117 , 118 ]. Enzymes produced by bacteria protect plants against environmental and biotic stressors. In drought, trehalose helps stabilize the membranes and enzymes. Surplus supply of trehalose by bacteria not only helps alleviate environmental stresses but also helps with eliciting disease response and induction of systemic resistance (ISR) [ 119 ]. Further research is needed to decipher how these interactive chemicals impact the plant microbiome structure and function and influence the plant health [ 120 ].

Microbes that colonize a host will always compete for nutrient, space, and survival. It was observed that the order of infiltration decides the resistance of the host against the infection. For instance, if an endophyte is present within a host before the arrival of a pathogen, the resistance will be stronger compared to when the pathogen and endophyte infiltrate the plant together or if the pathogen arrives slightly before the endophyte [ 121 ]. The biotrophic pathogen Ustilago maydis was inhibited by co-inoculation with Fusarium verticillioides . The endophyte did not protect when applied prior to the infection, indicating that the endophyte inhibited U. maydis by direct interaction. U. maydis did not affect the endophyte community, and it did not relate to the differences in the levels of resistance in the maize lines [ 122 ].

U. maydis is an interesting organism that has the ability to exhibit different lifestyles in different niche environments. There are other organisms that display such characteristics, for instance Moesziomyces sp. and Ustilaginales act as biocontrol agents in certain niches [ 108 ] through the secretion of hydrolase that antagonizes A. laibachii [ 123 ]. However, it was reported that some of these Ustilaginales could switch between being plant pathogens or beneficial epiphytes in different niches. This is also observed in instances where certain Fusarium oxysporum can act as antagonists to other F. oxysporum strains [ 124 ]; this was linked to the plethora of effector molecules produced. However, effector molecules have not been identified from endophytes and cannot be linked to any host specificity [ 125 ]. Therefore, this suggests that the anamorphs of Ustilaginales may produce filamentous structures [ 126 ], but there is no clear indication as to what the different adaptations in these organisms are that make them switch between pathogenic and epiphytic lifestyles. Figure 1 provides the plant-microbe interactions that are generally observed aboveground and belowground.

Shows the plant-microbe interactions aboveground and belowground observed in a plant’s natural environment. The pink box covers endophytic microorganisms, whereas the green box describes phyllosphere and epiphytic microorganisms. The purple column provides the explanation for microbe–microbe interaction; the orange column shows the explanation for root–microbe contact, and the blue column for root–root interaction. 3. Microbes and plant immunity.

Plants are an important source of nutrient for microorganisms. When microbes form non-beneficial interactions with the host, the immune system of plants would be triggered, either strongly or weakly, depending on the host and the pathogen. Unlike animals, plants have a defense mechanism, with structural, chemical, and protein-based components, to defend against attacks. A good understanding of the plant immune systems will allow us to develop better disease resistant varieties. Unlike our mobile and adaptive immune systems, plants depend on innate immunity, efficient signaling pathways, and beneficial microbes [ 127 ]. The initial step in triggering a defense mechanism lies in the invasion of the host cells. Pathogens, such as bacteria, infiltrate the host through various mechanisms, e.g., trichomes, lenticels, stomata, and other openings. However, in fungi, the infiltration process depends on the formation of the penetration pegs, while viruses are opportunistic pathogens that enter through injuries or locations of infection to cause disease in the plants [ 128 , 129 , 130 ].

When the primary defense is breached, the microbe associated molecular pattern (MAMPs/PAMPs) activates both the MAMP-triggered immunity/PAMP-triggered immunity (MTI/PTI) and the effector-triggered immunity (ETI). MTI/PTI is the horizontal immunity, while ETI is the vertical immunity. Some pathogens may trigger the ETI without the PTI through the interaction of effector molecules and the nucleotide-binding site-leucine-rich repeat (NB-LRR) found in the R genes, resulting in hypersensitive cell death [HR] [ 130 ]. While PTI and ETI share some common chemical components, they are viewed as separate evolutionary pathways [ 131 ] that are responsible for the plant’s immunity. A single NB-LRR receptor (directly or indirectly) provides immunity against pathogens once activated by pathogen effector molecules. The PTI involves protein recognition receptors (PRRs) that are present on the cell surface that act as binding sites for PAMPs/MAMPs. Consequently, the bound complex elicits a signaling cascade that is responsible for inhibiting the growth of the pathogens/microbes [ 130 , 132 ]. While plants have the PTI and ETI, microbes have evolved mechanisms that are able to overcome the PTI, by releasing effector molecules into the plant, triggering plant susceptibility.

Previously, it was assumed that the presence of the R gene was necessary for the perception of the pathogen. This was alluded to as the guard model. Recent research has shown that the indirect recognition of the effectors is inconsistent with the guard model. Presently, it seems that multiple recognition sites are available for different microbe effectors. It is now well-established that multiple targets in hosts are present for different pathogen effectors and the classical Guard Model does not explain this when lacking the R protein [ 133 ]. What is observed above involves evolution and, therefore, would be better explained by a decoy model [ 134 ]. The decoy is explained as a concept where the effector target is the decoy that acts on pathogen perception, even when the R protein is absent [ 133 , 135 ]. At the point of infection, systemic acquired resistance (SAR) is activated to prevent further proliferation of the pathogen to neighboring cells through the activation of the defense pathway, which results in the activation and expression of pathogenesis-related (PR) proteins [ 136 , 137 ].

Through the advent of the genomic tools, a better understanding of the interactions between plant and pathogens is obtained. Transcriptomics have enabled us to identify genes that are enhanced or inhibited in the plant-microbe interaction, providing a clearer picture of what may be happening in the regulation at the molecular level [ 25 , 138 , 139 , 140 ]. These studies also implied important roles for microRNAs in plant response against the pathogens, the plants innate immunity, as well as the triggered defenses in plants [ 141 , 142 , 143 ]. Beyond the effectors, receptors, and models described, it was postulated that the immune system in plants is moderated by systemic and local elicitation of phytohormones. These hormones are involved in the activation of induced systemic resistance (ISR) and SAR. For the above responses to take effect, there has to be interactions between the plant and the microbe [ 37 , 137 , 144 , 145 , 146 ].

SAR is split into several steps, where the most significant stage of SAR response is signal production and amplification at the site of infection and signal transduction to distal organs [ 147 ]. Numerous mobile chemicals were discovered as potential SAR signals or significant contributors in the mobility of long-distance SAR signals. Among them are methyl salicylate (MeSA) [ 148 ], glycerol-3-phosphate dependent factor (G3P) [ 149 ], azelaic acid (AzA) [ 150 ], dehydroabietinal (DA) [ 151 ], the lipid transfer protein known as defective induced resistance 1 (DIR1) [ 152 ], and pipecolic acid (PIP), a lysine catabolite amino acid [ 153 , 154 ]. Following signal detection, the emergence of SAR (defense priming) in the distal organ is linked with extensive metabolic and transcriptional remodeling [ 149 , 155 ].

In Arabidopsis, the important molecules that must be present in the distal pathogen free leaves are the buildup of PIP and SA, followed by the expression of flavin-dependent-monoxygenase1 ( FMO1 ), enhanced disease susceptibility ( EDS1 ), flowering locus D1 ( FLD ), isochorismate synthase 1 ( ICS1 ), phytoalexin deficient 4 ( PAD4 ), AGD2-like defense response protein 1 ( ALD1 ), and SNF1-related protein kinases 2.8 ( SnRK2.8 ) genes. The majority of these components are parts of the SA-amplification chain [ 156 ]. The activation of a transcription factor, a non-expressor of PR genes 1 ( NPR1 ) by SA is also required for defense priming [ 154 , 157 , 158 ]. It must be emphasized that defense priming and signal amplification are interdependent with systemic PIP formation and PIP facilitated SA-independent and SA-dependent priming of plant defenses in an FM01-dependent manner [ 154 , 155 ].

Previous findings also showed that the SAR-inducing action in cucumber and Arabidopsis phloem sap caused by various phytopathogens proved efficacious in other plants [ 149 , 151 ], suggesting that the mobility of SAR signal(s) are not unique to plants or pathogens [ 156 ]. The SAR signaling transmission is as seen in Figure 2 .

The process of the SAR mechanism. The generation of signals in the diseased organ. The signal is sent to other parts of the plant that are not affected. After transcriptional and metabolic remodeling, the essential molecule is present in the healing organ as immunity. The black arrow represents the movement of the SAR signal to the distal organ.

While SAR happens at the site of pathogen infiltration, ISR happens from its site of trigger in the rhizosphere. The association between the plant and microbe in the soil can be used to improve plant defense and crop productivity. When disease is present in the soil, microbes with the ability to inhibit the activity of these pathogens may be introduced to manipulate the environment into one community that ensures the health of the soil. This may be achieved with a rich inoculum and the management of environments [ 136 , 159 , 160 , 161 , 162 ]. Finding the right balance in plant-microbe and microbe–microbe interaction is important in establishing chemical communication in the rhizosphere. This association plays an important role in engaging the signaling cascade that prompts the resistance or defense in the plant against pathogens and facilitates activities that improve yield and growth of the crop [ 131 , 163 , 164 ].

The microbial assemblage in the soil secretes molecules that are able to induce gene expression in the plant species. Some of these signals (VOCs, for example: alcohols, alkanes, ketones, terpenoids, etc.) operate as communicators within the microbial communities in the rhizosphere [ 25 , 131 , 165 ]. These compounds promote several functions, such as disease inhibition, nutrient acquisition, improved growth and development, mineralization, and other processes. These compounds are also responsible for triggering alterations in the plant’s transcriptome. While phytohormones, such as auxins, abscisic acid (ABA), cytokinins, gibberellins, jasmonic acid (JA), and salicylic acid (SA) are at play in plants; the same hormones are also secreted by beneficial microbes [ 145 , 166 , 167 , 168 ]. The beneficial chemicals exuded by microbes that activate plant defense mechanisms are listed in Table 1 . In addition to the induction of ISR by the plant-microbe interaction in the rhizosphere, immunity or plant defense may also be elicited through a phenomenon called trans -generational. This form of immune memory is transferred to the following generations in the plant in response to pathogens [ 131 , 169 ]. For instance, when an avirulent Pseudomonas syringae was applied on Arabidopsis, it resulted in the next generation of plants, showing increased levels of salicylic acid [SA], which resulted in heightened disease resistance [ 131 , 162 , 169 ].

The beneficial chemicals exuded by microbes and the benefit to plant defense mechanism.

Despite many remarkable discoveries in the field of plant immunity, many mysteries remain unresolved, such as the identification of avirulence ( Avr ) genes in plant–pathogen interactions, plant root immune mechanisms, molecular mechanisms of pathogen colonization in plants, regulation of cellular activity and gene expression, and signaling mechanisms involved in plant immunity. As a result, progress in post-genomic era technologies will open the door for a deeper understanding of plant–pathogen interactions and plant immunity.

3. Functions of Rhizosphere Consortia

As mentioned in the previous section, soil microbes are involved in three main processes—growth and development, nutrient acquisition, and stress management against biotic and abiotic stressors. These processes are managed through an interplay or chemical signaling that facilitate these functions.

3.1. Hormones and Their Promotion of Growth and Development

Firstly, organisms, known as plant growth promoters, achieve their purpose in plant growth and development through an array of phytohormones secreted into the soil. The main players, such as auxin, cytokinin, ethylene, and gibberellin influence plant growth, and are extensively studied in cereal root systems [ 8 , 136 , 203 ]. Pseudomonas , Burkholderia , and Pantoea are involved in biological processes, such as P solubilization, N 2 -fixation, auxin, and ACC deaminase production. Beneficial and non-beneficial organisms produce auxin. Auxin is responsible for root growth and formation, elongation of nodular cells, and response against stressors [ 204 , 205 , 206 ]. In pathogens, auxin is linked to its virulence. One good example of auxin’s role in virulence has been studied in the Agrobacterium tumefaciens where the expression of tumors in plants depends on the secretion of IAA [ 207 , 208 ]. As tryptophan is required for auxin production, aminocyclopropane-1-carboxylic acid [ACC] is needed for the production of ethylene and microbial growth. ACC deaminase-producing PGPRs help in the utilization of ACC to equilibrate the levels of ACC inside and outside the plant [ 209 , 210 ].

Ethylene is involved in the regulation of growth, elicitation of defense, and the management of plant stressors [ 211 ]. The elicitation of defense response by singular or consortia is dependent on ethylene. The role of ethylene in affecting the community structure was determined using ethylene mutants. These studies showed that the mutants affected the bacterial community structure, but these studies were not able to correlate the abundance of species to ethylene due to variable ethylene levels, and the cross-talk with other hormones [ 211 , 212 , 213 ]. Moreover, the original colonizers will have the ability to influence the microbial population. Hence, if the microbial populations are linked to ethylene regulation, they are likely to shape the microbial structure in the soil and control the regulation of stress in the plants [ 36 , 136 ].

Jasmonic acid (JA) and its methyl ester (MeJA) have been associated with defense and wound response in plants [ 214 , 215 ]. Recent studies have also alluded to JA being involved in the recruitment of microbial communities around the roots [ 8 , 163 ]. JA regulates the components in the root exudates, such as benzoxazinoids, known to improve herbivore resistance. These compounds contribute to the allelopathic and chemotactic nature of root exudates. These exudates recruit miscellaneous microorganisms that cater to specific niches in the soil as well as the specific needs of the plant itself [ 164 , 216 ]. However, while JA is responsible for the recruitment of microorganisms, we are unable to correlate JA and the population structure due to too many variables in the environment.

Salicylic acid (SA) is another signal molecule that is involved in plant defense. However, unlike JA and ethylene, SA is directly related to SAR. Together with JA and ethylene, SA forms the core defense hormone in the plant. The role played by SA has been studied using A. thaliana mutants, where the knockout mutants showed lower levels of survival and less prolific colonization [ 136 , 217 , 218 ]. Further, the study by Lebeis et al. [ 219 ] observed that SA linked pathways were required for the colonization of endophytes and the shaping of soil microbial structure. However, phytohormones (ABA, cytokinin, ethylene, JA, SA IAA, brassinosteroid, and others) may show synergistic or antagonistic effects against plant related processes. For instance, ABA is a major player in moderating abiotic stresses. ABA negatively interacts with SA mediated defenses and works either positively or negatively with JA and ethylene related biotic responses, respectively [ 41 , 220 , 221 ]. Therefore, in either biotic or abiotic stressors, phytohormones play their specific roles in shaping the soil microbial structures [ 25 ].

3.2. Biological Processes in Nutrient Acquisition

Microorganisms are involved in nutrient cycling and acquisition from the soil. Therefore, organisms, such as plant growth promoting rhizobacteria (PGPRs), are studied extensively for use as biofertilizers [ 222 ]. The compounds exuded by these microbes and plants work together to facilitate processes, such as nodulation, quorum sensing, N 2 -fixation, mineralization, and others [ 223 ]. Some of these processes have already been discussed under plant–microbe interaction Nod factors bind lysin motif-containing receptor-like kinases (LysM RLKs) and initiate signalling cascades, resulting in nodulation by bacteria in exchange for photosynthetic carbon [ 25 ]. Further, bacteria that establish IAA secretion in and around the root area enable the development of root hairs [ 223 , 224 ].

There is evidence that the biochemical constituents of rhizobacteria are able to elicit defense as well as facilitate symbiotic relations. Iron, for instance, when secreted by certain B. subtilis strains, is able to activate the host defense mechanism [ 35 , 46 , 136 ]. Bacterial volatiles activate the Fe deficiency transcription factor that, in turn activates a series of enzyme that results in iron accumulation. Freitas et al. [ 225 ] observed that when G03 was used to treat cassava plants, iron content increased substantially in the leaves. Similarly, certain organisms, such as Bacillus paramycoides KVS27, and Bacillus thuringiensis KVS25, increased growth of Brassica juncea through P solubilization, N 2 assimilation, IAA, siderophore, and HCN production. The observed activities were attributed to the synergism among these organisms that resulted in the secretion of multiple chemicals, collectively resulting in plant growth. Therefore, the effects that are seen on plant defenses may involve a consortium rather than singular microbes. The interaction between microbe–microbe, microbe–plant, and microbe–environment collectively influences growth and development of plants and the microbial community [ 226 ].

3.3. Microbial Defense Mechanisms

The manifestation of disease depends on various factors, such as host range, susceptibility of host, environment, pathogen population, agricultural practices, and various biotic stressors [ 227 , 228 ]. The resistance towards any pathogen produced by a host depends on the roles played by aboveground and belowground microbes that are able to modify the defense responses in plants [ 73 ]. Though the control of disease has been largely through chemicals, the effort to go green has directed research in the identification of biocontrols in disease suppression [ 35 , 71 , 229 , 230 ]. The use of beneficial microbial population is slowly gaining popularity worldwide, where enzymes, antibiotics, siderophores, volatile compounds, and inhibitory chemicals control the spread of disease [ 231 , 232 , 233 ]. These biocontrol agents have a myriad of activities that enable them to suppress the pathogens. Whether it is antagonistic, competitive, or triggering of the defenses—all work well in keeping disease in check. The antibiotics that are expressed by microbes promote growth and suppress pathogens. This is achieved through the activation of certain hormones, such as auxins, which enable changes to root architecture to improve nutrient absorption and improve growth [ 234 ]. Pseudomonads are widely known to produce DAPG, which induces ISR, while cyclic lipopeptides (cLPs-surfactin, fengycin, and iturin) from Bacillus spp. and Pseudomonas spp. produce surfactants that are able to inhibit pathogens [ 44 , 235 , 236 ]. In addition to the antimicrobials, and lipopeptides, QS enzymes play a role in suppression of disease, and induction of ISR [ 35 , 237 ]. Some of these organisms also play a role in regulating defense through the control of hormones in plants [ 238 ]. While certain taxa, such as Actinobacteria, Serratia, and Enterobacter are able to control several soil-borne diseases. These groups of organisms are able to induce action through ISR and SAR, protecting the plant systemically through the involvement of hormones, signal molecules, and the activation of pathways in the plant [ 239 , 240 ].

4. Challenges of Emerging Plant Pathogens and Their Impacts on Plant–Microbe Interaction

As addressed in sections above, plant–microbe interactions can be either positive or negative. However, an immediate challenge to agriculture is the new and emerging pathogens that continue to plague the industry. While plant defense mechanisms are in place to protect the plants from the exposure to pathogens, new and emerging pathogens may have evolved mechanisms that enable them to evade the host’s innate immune system [ 241 ]. Since it has been observed that pathogens co-evolve with their host, it is likely that, to avoid this “arms race”, the pathogens expand their host ranges. These organisms evolve their virulence or pathogenicity factors to enable them to elicit disease in the same susceptible host and new ones [ 242 ]. These new and rapidly evolving microbes pose a threat to the agricultural industry. A better understanding of the pathogen’s invasion and infiltration strategies will enable better control over disease through strategic heightening of defenses or breeding [ 127 ].

Several possibilities for the emergence of new harmful organisms (such as bacteria and fungus) are (i) the bacteria may be endemic in agricultural land, but a novel host has just been found, (ii) After becoming endemic, the microorganism turns pathogenic, owing to a rise to its pathogenicity or a loss in the host’s defenses, (iii) The microbe may have just been introduced into a new environment with unknown hosts, and the organism might be harmful to novel plants, and (iv) Insect vectors feed on a new host, containing harmful organisms, and spreading the organism to succeeding plants [ 243 ]. Diseases arise due to a variety of causes, including interactions between pathogenic organisms, plant–pathogen interactions, plant–insect–pathogen interactions, and unfavorable environmental circumstances. According to Deberdt et al. [ 244 ], climate change, can modify the character of microbes, transforming them into opportunistic diseases. It is well known that when plants are weakened or stressed by external conditions, microbes may easily colonize them, resulting in plant mortality. Certainly, various abiotic stress, such as drought, heat factors, and so forth affect the plant, inflicting great damage to the forest and agriculture [ 245 ].

Further, trade has become an agent of disease transmission globally. Though scrutiny of migration pathways and quarantines have been imposed, new and emerging diseases are constantly being reported [ 246 ]. The advent of omics tools has enabled us to obtain new information on emerging populations. There is also a flood of databases of phytopathogens and plant genomes that has made it possible for us to study the plant-microbe interaction more closely [ 247 ]. The current information derived from the sequence databases show that there is an accelerated genome adaptation in pathogens to their environment. This high rate of evolution has further compounded the problem of disease in host–pathogen interactions. Population genomics studies is a good way to study the adaptive evolution of plant pathogens and design better disease management strategies. In the following section, we will deal with the technologies used to study the microorganisms and the plant-microbe interactions [ 248 , 249 ].

5. Unraveling Plant–Microbe Interaction at the Molecular Level

The underlying theme in this review involves the three main interactions observed between plants and microbes. While symbiosis and mycorrhizae are two main facets of this interaction, the aspect of disease has garnered interest, especially with the losses incurred by current pathogens and the threat of emerging diseases. The exploitation of this interaction provides for the development of sustainable disease management strategies [ 35 ].

While our current method of addressing disease in plants is through resistance breeding via conventional or molecular techniques, the advent of new genome platforms has enabled us to acquire large amounts of big data on plants and pathogens through a series of sequencing and re-sequencing of these genomes. The method of identification, such as 16S rRNA sequencing, WGS, or classic culture techniques, can potentially have an impact on the reporting of the discovered microbiome. Delmont’s [ 250 ] 2009–2012 survey of Park Grass, for example, employed at least six distinct techniques of DNA extraction to produce an accurate representation of the soil microbiome. The genomic and post genomic era is upon us, and we are now faced with this large amount of data that needs to be deciphered and utilized in the development of disease resistance in plants, as well as in improving our understanding of ISR and SAR [ 249 , 251 , 252 , 253 ]. It is now possible to dissect and scrutinize the plant-microbe interaction at a molecular level through the utilization of platforms of genomics, proteomics, transcriptomics, and metabolomics. The genome data on microbe and plants, the various proteins that are secreted in the plant–microbe interaction, and the differentially expressed genes in the host and the metabolomes involved helps us understand these complex relationships [ 254 , 255 ]. While the genome information may be utilized to develop resistant plants through breeding or genetic engineering, the protein information may be utilized to identify key proteins in plant growth and development that controls various physiological and biochemical pathways [ 79 ]. The transcriptome data enables us to observe the variations in the expression of genes in response to the environment, growth, and development, while the metabolome data provides us with the metabolic changes incurred through the interaction between the plant and the microbe. Collectively the post-genomic era data have enlightened us in the area of gene discovery, beneficial microbes, and proteins that may be used in crop improvement, growth improvement, and heightened disease resistance [ 256 ]. Below, we will briefly go through the techniques that are useful in deciphering the biological functions and benefits of plant-microbe interaction.

5.1. Genome Sequencing

The various genome-sequencing platforms that have been developed over the years have made studying the interactions between plants and pathogens, at the molecular level, possible, and more informative. The availability of genome sequences of plants and microbes and the ability to conduct genome wide annotation of proteins and genes through bioinformatics platforms has further advanced the field of plant–microbe interactions [ 257 ]. The very first contribution to the bacterial genome was first obtained in 1995, which resulted in the ability of computational modeling in envisioning the entire operation of this organism from the sequence structure alone [ 258 ]. In 2000, the sequencing and annotation of Arabidopsis paved the way for better understanding of the sequence to operations through the use of genome scale modeling and annotations [ 258 ]. Through genome informatics, we are able to understand the microbe and plant systems better at the molecular level. Genome sequencing also allows us to connect the multifaceted signaling pathways that regulate the defense mechanism in the plant. However, despite the large amount of data available from the genomic- and post-genomic era, there are still gaps in the knowledge due to the high complexity of the interaction between plants and microbes, complicated further by internal and external regulatory factors [ 259 ].

Generally, the genomics and transcriptomics data allows us to draw information necessary for the metabolic network modeling of plants and pathosystems. By merging the metabolic pathways of the plant and pathogen, we are better able to study the positive and negative effects of these interactions [ 249 , 257 ]. The initial study of plant–microbe interactions and understanding of the relationship at the genome level was taken one gene at a time or one protein at a time. Over time, a more holistic approach was taken where the entire plant and pathogen genome was elucidated together. In the early 21st century, transcriptomic tools, such as the cDNA microarray and SuperSAGE, were used to profile gene expression and signaling in Arabidopsis thaliana–P. syringae and rice– Magnaporthe oryzae interactions [ 95 , 260 , 261 ]. As the sequencing platforms became more advanced, the RNAseq technology was developed, and the differential expression profiles of plant−pathogen interactions were elucidated. Combined with the transcriptomics data, the proteome data of plant–microbe interactions were also derived through 2D gels, MS/MS, GC/MS, LC/MS, and iTRAQ [ 262 ].

One of the important outcomes of the post-genomic era is the utilization of the sequencing data in annotations, making sense of how the organisms operated through metabolic modeling [ 258 ]. Through these modeling activities, we are able to investigate the capabilities and inefficiencies of an organism through studies from the genes, to proteome and transcriptome [ 258 , 259 ]. Through these network models, we are able to address all possible interactions between plants and pathogens. Genome-scale reconstruction models (GSRMs) were developed for many organisms and are useful in understanding the multi-cellular community interactions for phenotype−genotype gap bridging, and to investigate the functional evolution of metabolic and regulatory networks [ 263 ].

5.2. Amplicon Sequencing

High-throughput sequencing of marker gene amplicons is commonly used to clarify the composition, structure, and geographic dispersion of microbial populations in the environment, and remains a popular method in plant microbiome research [ 251 ]. Amplicon sequencing has the benefit of being very precise, identifying specific groups of microorganisms or functional genes [ 264 ]. Amplicon sequencing specificity enables it to accurately identify many rare species; yet, its sensitive characteristics makes it susceptible to contamination [ 265 ]. Hence, any analysis that depends significantly on amplicon sequencing must include both positive and negative controls [ 266 ]. This technique involves the sequencing of PCR products, obtained by using primers for the taxon-specific variable regions [ 267 ].

When studying bacterial populations, the 16S rRNA gene is the target used for amplification, sequencing, and identification of the targeted microbiome [ 268 ]. Several different primer sets were developed for the 16S rRNA genes of bacteria and the 18S rRNA genes and ITS segments that surround their regions of diversity. The universal primers used in amplicon sequencing amplify genes from various taxonomic groups with varying degrees of effectiveness [ 269 , 270 , 271 ]. Given their length, 16S genes with large introns may be overlooked by standard PCR design [ 272 , 273 ]. The quantity of rRNA gene clusters per genome has a direct influence on determining the total relative abundance of specific bacterial species [ 274 ]. The amplified product is then subjected to any one of the sequencing platforms that are available [ 267 ].

The sequence information obtained from the amplified products can be used in phylogenetic studies of the organisms within the sample. The phylogenetic relationship derived can be used in inferring taxonomic information. This taxonomic identification is largely dependent on how extensive the reference databases are. While 18S rRNA and ITS is available for the identification of fungi, the ITS is preferred as there are good reference databases and the sequences from ITS show a higher level of variance [ 275 ]. Detailed categorization of observed reads to the genus or species level is sometimes challenging because the amplicon sequence lacks the necessary sequence diversity to identify closely related genera or species with 18S rDNA primers, [ 276 ]. As a result, the ITS region was recommended over the 18S rRNA gene because of the greater sequence diversity seen in the ITS region and the availability of a much more curated and extensive reference database [ 275 ]. Nonetheless, unequal ITS fragment lengths may encourage PCR amplification of shorter ITS sequences as an alternative. This might result in a skewed estimation of the relative abundances of fungal species. However, to make sure that there are no biases of relative abundance of fungal taxa based on ITS sequence identification, non-ITS based targets may also be included to provide robustness to the data derived [ 277 ]. Following amplicon sequencing, the microbiome is analyzed through clustering of OTUs based on the defined sequence similarity thresholds. Sequences with similarity are assigned to the same taxa by OTU. These microbes are assumed to share origins.

Although amplicon sequencing may be used to infer community function, it is not an ideal method unless particular functional genes are utilized, where the function and phylogeny are congruent [ 278 ]. The following issues are linked with the amplicon sequencing approach: (i) During DNA amplification, sequencing mistakes and chimaeras can occur [ 279 ]. (ii) It is possible that primer coverage will not cover the necessary microbial diverse populations [ 269 ]. (iii) The relative abundance of operational taxonomic units (OTUs) may be skewed due to differences in amplification efficiency across the target genes [ 280 , 281 ]. (iv) Variability in gene copy numbers may have an impact on conclusions based on the relative abundance of the OTUs [ 282 ].

5.3. Metagenomics

Metagenomic analysis provides a variety of approaches that are based on biomolecules, such as lipids, DNA, RNA, and proteins [ 283 ] that researchers may use to uncover plant microbiome activity and diversity to identify microbial participants in soil. The shotgun genome sequencing method of metagenomics, as opposed to the amplification of targets in the amplicon sequencing, provides more information. This method provides sequences from bacteria, viruses, archaea, phages, and fungi. However, in comparison to the 16S rRNA method, this technique will require higher information depth to distinguish the uncommon/rare members of the microbiome, and quality control to trim and filter the reads using bioinformatic tools [ 284 ]. The online-based tools are easily used for any sequence information and can be easily utilized to map the reads obtained against any reference databases. These mapped reads are then functionally annotated using various online resources [ 285 ].

The shotgun metagenome sequencing makes it possible to study greater structure of microbial communities while also providing an unbiased perspective of the phylogenetic and functional makeup of environmental microbial populations [ 286 ]. Through metagenomics, the level of identification can go right down to the strain level, which is at higher efficiency compared to amplicons, which are more likely to provide characterization to the taxonomic levels of the amplicons [ 287 ]. However, while the identification is more precise, this method would require additional bioinformatic tools to reconstruct the genome based on the short reads obtained, or the utilization of higher resolution sequencing platforms. The metagenomic method is a useful tool to find and characterize microbes at the strain level, where the algorithm used will enable the system to overcome the intergenomic repetitive elements and detect small differences in the genetics of the organisms [ 288 , 289 , 290 ]. Further, the gene sequences in metagenomics may be functionally annotated to provide a clearer picture of the microbial characterization compared to the amplicon survey. The functional annotation will include gene prediction and annotation, where, firstly, the protein coding sequences are identified, followed by matching this predicted protein to a protein function [ 291 , 292 ]. However, the identification of genes from the metagenome analyses does not ensure that all genes identified are expressed. While both amplicon sequencing and metagenome use the sequencing platforms, these methods have their limitations. This is precisely why sequencing platforms and bioinformatic tools are constantly updated and upgraded to improve the quality of reads and informatics obtained [ 293 ]. Therefore, to gain a better understanding of the total microbial diversity, studies may employ one or a combination of methods to acquire as much information as possible while adhering to their sample size.

5.4. Soil Proteomic

Proteomics is used in the study of the function and control of biological systems based on the prediction of protein profiles. Considering that soils have the capability to restore extracellular proteins through a variety of ways, the effectiveness of protein retrieval from diverse sources must be evaluated as part of the progress of soil proteomics [ 294 ]. Although metagenomics enables the identification of microbes in the rhizosphere, metaproteomics enables the investigation of rhizosphere biological activities [ 295 ]. It is feasible to relate ecological function to microbial community composition when these two techniques are used for the same problem [ 295 ]. Previous studies that used this technique attempted to comprehend the truffle brûlé mechanism in its specific niche, where other symbiotic fungi were driven away by this fungus once it formed symbiotic relations with the plant [ 296 ]. Recent research conducted on soil microbes have combined various omics methods (culturomics, metaproteomics, and 16S rRNA sequencing) to identify microbial communities and elucidate microbial population roles in the glacial ecosystem [ 297 ].

The idea behind mass soil protein analyses is that having a full proteomic profile of a microbial community would make it easier to find distinctive polypeptides whose syntheses are influenced by certain ecological factors [ 298 ]. Due to a huge number of unique proteins synthesized by various species, molecular characterization of soil proteins, for revealing species composition and metabolic activity, has been challenging. Amidst this drawback, advances in immunological methods, as well as a spike in the range of accessible enzyme analyses, have been utilized to complement precise molecular resolutions in situations where a specific polypeptide was of interest [ 299 ]. Understanding ecological activities by measuring molecular diversity in soil settings requires the understanding of protein structural complexities in comparison to other identifiable compounds, such as fatty acids and nucleic acids. One- or two-dimensional polyacrylamide gel electrophoresis (PAGE) can be used to create comparison protein profiles, relying on electric charge and physical size attributes [ 298 ]. Concerns of evolutionary variety within particular groups of species inhabiting comparable ecological niches may also be addressed using amino acid sequence analysis. Despite the fact that the metaproteomics approach has been around for more than a decade, it is still constrained by computational and technical support [ 295 ]. First, contamination by humic acid and other pollutants that impedes protein extraction makes it extremely reliant on soil type. Secondly, various extraction techniques might have an impact on the detected metaproteomics [ 300 ]. This limitation can be circumvented by utilizing several extraction techniques simultaneously and pooling all extracted proteins prior to further analysis [ 295 ]. Thirdly, protein identification is hampered by the lack of a comprehensive protein database [ 301 , 302 ]. Building in-house libraries largely depends on the metagenomics data acquired from comparable settings in previous studies [ 303 ]. The current availability of low-cost high-throughput sequencing has undoubtedly aided the integration of metagenomics with metaproteomics. However, by using next generation sequencing (NGS), it is feasible to get more reads in less time, allowing the species to be identified, and at the same time establishing an optimized databases for protein identification [ 295 ].

Overall, metaproteomics is a strong tool used for studying biological functions of a microbial community, and this information is used to correlate functional and taxonomic soil makeup in the ecosystem [ 304 , 305 ]. Furthermore, soil protein analysis might provide relevant data on the biogeochemical capacity of the soil and pollutant decomposition, as well as operate as a predictor of soil health and restoration [ 306 ]. This might help us comprehend organic contaminants and organic compound degradation, nutrient cycles, and plant–plant and plant–microbial interaction at the molecular level.

6. Microbes in Sustainable Agriculture

Microorganisms have a huge impact on the physical, chemical, and biological processes in the soil that are directly and indirectly important for plant and animal growth and development. While extensive studies have been carried out on a global scale to identify suitable microbes for use in the agricultural industry, more can be done in the continuous isolation and characterization of future biocontrol and growth promoting organisms that are suitable for specific applications. In this section, we will look at how bacteria can be used in agriculture.

Nutrient cycling: microbes recycle several nutrients, such as carbon, nitrogen, phosphorus, potassium, zinc, calcium, manganese, and silicon on a constant basis. Nutrient recycling is critical, not only for plants, but also for all forms of life, as it provides essential components for the synthesis of amino acids, proteins, DNA, and RNA required by all living organisms. The contribution of microbes in this regard is largely undervalued. Identifying and maintaining the density and community of essential microorganisms in each cycle will be of utmost importance. Further, to boost the organism’s cycle abilities, key genes, such as the Nod factors, can be genetically modified to improve nitrogen-fixing ability, for instance. The same can be done with any other nutrient cycling process [ 36 , 307 ].

Bioremediation: industrialization and current agricultural techniques increase the negative impacts on agricultural land and water by releasing vast amounts of hazardous waste, heavy metals, and organic contaminants, all of which are severe problems, not just for agriculture, but also for human health. Although trace amounts of heavy metals, such as lead (Pb), cadmium (Cd), mercury (Hg), chromium (Cr), zinc (Zn), uranium (Ur), selenium (Se), silver (Ag), gold (Au), nickel (Ni), and arsenic (As) are beneficial to plants, excessive uptake reduces plant growth by interfering with photosynthesis, mineral nutrition, and essential enzyme activities. Industrialization and contemporary farming methods are putting increasing amounts of pressure on the environment. Bioremediation is a process that uses algae, bacteria, fungi, or plants to remove heavy metal ions from a polluted environment. Bioremediation with microorganisms is long-term and sustainable since it helps to restore the natural state of the damaged environment while being cost-effective. Heavy metal detoxification by microorganisms can occur spontaneously, by the addition of native microbial strains or through genetic manipulation. To reduce the active concentration of metal ions present in polluted environments, microorganisms use biosorption, adsorption, compartmentalization of heavy metals into intracellular molecules, metal binding, vacuolar compartmentalization, extracellular mobilization, or immobilization of metal ions [ 307 , 308 ]

Growth and development: microorganisms use a variety of processes to enhance plant development and growth in both normal and stressful settings, including nitrogenase enzyme activity, nitrate reductase activity, siderophore generation, and phytohormone synthesis. Major plant hormones include auxin, cytokinin, gibberellin, abscisic acid, and ethylene, with more phytohormones being discovered. Phytohormones are produced by a variety of microbial species, and they are frequently used in agriculture to improve plant growth and stress tolerance. Plant growth-promoting bacteria (PGPB), also known as rhizobacteria, have been genetically modified to increase the synthesis of stress-induced hormones, antibiotics, antifreeze proteins, trehalose, and lytic enzymes, all of which help plants develop and cope with stress. In order to compete with the already-adapted indigenous microorganisms, PGPR must develop and sustain a biologically active population. Genes that promote growth have been shown to improve strains. As a result, efforts have been undertaken to vary the timing or level of their expression, as well as transfer and express them in different hosts, in order to improve plant growth and development [ 307 ].

Genes involved in growth promotion were shown to be effective tools for strain improvements by altering their expression timing and level, or by transferring and expressing them in different hosts to increase plant growth and fitness. Microorganisms modified through genetic engineering have improved specific characteristics, such as the ability to degrade a wide range of contaminants for bioremediation of soil, water, and activated sludge, improved plant biotic and abiotic stress tolerance, and increased phytohormone production, among other things. In a hostile environment, the modified strain can survive and remain active [ 13 ]

Stress management: as sessile organisms, plants are subject to abiotic and biotic stress. Diseases, drought, submergence, metal toxicity, salinity, and various other stressors are faced by crops throughout the seasons. Microbes are key regulators of stress through the various biomolecules that are exuded in the form of antibiotics and hormones. The chemical exudates from the microbes activate ISR and induce the resistance mechanism in plants. Genes, such as chitinases and glucanases, have been effectively used to engineer both crops and microbes to enhance the expression of these genes in planta or in microbes for enhanced resistance towards pathogens [ 309 ].

While the wild type and mutant microbes have the general functions as stated above, the transition from laboratory to field and market is slow. Transition to market would require the optimization of concoction, determination of concentration, frequency of application, and selection of carriers for these organisms. All of this requires time and funding to fine-tune. While some microbial concoctions have made their way into the market as biofertilizers, biocontrols, soil amendments, and biostimulants, many are still in the laboratory and greenhouse phase, working on optimization. Another factor that has impeded the transition to the utilization of microbes is the efficacy of these compounds compared to chemicals agents, as chemical have wider spectrums of efficacy and and result in more consistent effects on plants and detrimental microbes.

7. Future Prospects and Challenges in Plant–Microbe Interactions

Most current and past studies have either selected one of more of the methods employed above to determine the soil microbial structure, density, and function. However, while these techniques do provide some insight on the plant–microbial interactions, they by no means provide a complete picture of the microbial interactions that occur in reality between the plant and microbes in a variety of conditions. Further, rather than identifying the microbes that are present in a particular environment, it would be beneficial for us to know the roles that they play in the environment, singly and in combination with others. Therefore, based on the above listed interactions, it is necessary to “put the pieces together” based on: (1) The microbes present aboveground; (2) The microbes present belowground; (3) How the belowground microbes affect the host and interaction aboveground; (4) The processes and interactions between the host and microbe and root and root; (5) The exudates produced by the plants, the microbes, and their functions; (6) How these affect a plant’s gene expression and immune system; (7) How these microbes affect plant growth, development, and immunity; and (8) Whether there are specific chemical compounds involved in microbe recruitment (and many more).

The above-mentioned information is a consequence of direct or indirect effects of microbes on the host. In recent decades, with the arrival of next generation sequencing platforms, we were able to observe these interactions at the molecular level. The depth of information made available from genomic, proteomics, transcriptomics, and metabolomics has shed some light on the intricacies of the plant–microbe interaction, enabling us to understand the process of disease development, growth and development, immune response, nutrient cycling and absorption, disease suppression, and others. Most of the studies have been directed towards identifying dominant taxa in a particular environment, the effects of plants and microbe exudates on the recruitment of microbes, and the structural architectures of the diversity and communities. There is still much that needs to be studied on the mechanism of recruitment, on the communities influence the plants and each other. Little is known on how the microbial factors influence root exudation and architecture. This information may be manipulated to optimize the microbial communities in the soil and improve overall performance of plants. The following are some applications of metagenome studies conducted in plant–microbe interactions.

• (1) Identified productive microbiomes by creating conducive environments for the rhizosphere microbiome to communicate with the plant and surrounding environment.
• (2) Applied comparative genomics and metabolomics studies to identify specific rhizobacteria that were naturally selected based on root exudates; optimized utilization of these cultures to increase growth and development.
• (3) Identified microbes and their proteomes, able to trigger ISR and SAR across monocots and dicots.
• (4) Applied transcriptome profiling to identify defense-associated transcripts involved in innate immunity and plant resistance scenarios.
• (5) Identified microbes used in seeding of disease suppressive soil to enhance plant fitness and productivity.
• (6) Identified plant-associated microbiomes that influenced different plant traits including abiotic stress tolerance, flowering, growth, and disease suppression. Host co-evolution with the microbiome could be utilized in future crop breeding strategies for low-input sustainable agriculture.
• (7) Mapped microbiomes in the soil through all developmental stages, the differences in the proteins exuded. This information may be used to generate microbial concoctions for soil amendments to support growth and yield in all stages.
• (8) Exploited beneficial microorganisms and identified emerging pathogens.

With further research and more information being provided from omics-based studies, we expected that more clarity will be obtained concerning plant–microbe interactions.

Supplementary Materials

The following are available online at https://www.mdpi.com/article/10.3390/ijms221910388/s1 .

Author Contributions

Conceptualization, K.N.; writing, K.N. (main) and N.S.N.A.R.; figuration and formatting, N.S.N.A.R.; editing K.N. All authors have read and agreed to the published version of the manuscript.

This paper was funded by the Ministry of Education Malaysia through grants awarded to Kalaivani Nadarajah (FRGS/1/2019/STG03/UKM/01/2 and FRGS/2/2014/SG05/UKM/02/1). Thanks also to Universiti Kebangsaan Malaysia for financial assistance and provision of facilities.

Institutional Review Board Statement

Informed consent statement, data availability statement, conflicts of interest.

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

METHODS article

Ognnmda: a computational model for microbe-drug association prediction based on ordered message-passing graph neural networks.

• 1 School of Computer Science and School of Cyberspace Science, Xiangtan University, Xiangtan, China
• 2 Hunan Institute of Engineering College of textile and clothing, Xiangtan, China

In recent years, many excellent computational models have emerged in microbe-drug association prediction, but their performance still has room for improvement. This paper proposed the OGNNMDA framework, which applied an ordered message-passing mechanism to distinguish the different neighbor information in each message propagation layer, and it achieved a better embedding ability through deeper network layers. Firstly, the method calculates four similarity matrices based on microbe functional similarity, drug chemical structure similarity, and their respective Gaussian interaction profile kernel similarity. After integrating these similarity matrices, it concatenates the integrated similarity matrix with the known association matrix to obtain the microbe-drug heterogeneous matrix. Secondly, it uses a multi-layer ordered message-passing graph neural network encoder to encode the heterogeneous network and the known association information adjacency matrix, thereby obtaining the final embedding features of the microbe-drugs. Finally, it inputs the embedding features into the bilinear decoder to get the final prediction results. The OGNNMDA method performed comparative experiments, ablation experiments, and case studies on the aBiofilm, MDAD and DrugVirus datasets using 5-fold cross-validation. The experimental results showed that OGNNMDA showed the strongest prediction performance on aBiofilm and MDAD and obtained sub-optimal results on DrugVirus. In addition, the case studies on well-known drugs and microbes also support the effectiveness of the OGNNMDA method. Source codes and data are available at: https://github.com/yyzg/OGNNMDA .

1 Introduction

The human microbiome consists of trillions of microbes that reside inside and outside the human body, and these microbes play an essential role in maintaining the overall health of the human body ( Ogunrinola et al., 2020 ). The host-microbe plays a crucial role in several physiological processes in the human body, such as energy collection and storage ( Amato et al., 2019 ), facilitating carbohydrate absorption, and protecting the body from foreign microorganisms and pathogens ( Hajiagha et al., 2022 ). Moreover, the changes in microbiota composition can significantly affect human health Kim et al. (2018) ; Partula et al. (2019) ; Catinean et al. (2018) . Many studies have shown that the dysbiosis or unbalance of microbiota is closely related to disease, and the microbiota is an important causative factor for many diseases. Therefore, microbes are considered new therapeutic targets for precision medicine ( Cullin et al., 2021 ), and the research on the relationship between microbes and drugs not only aids in drug development but also the diagnosis and treatment of human diseases. However, the popularization and widespread use of antibiotics in modern medicine have led to the emergence of an increasing number of drug-resistant microbes, which seriously threaten human health ( Pugazhendhi et al., 2020 ). Although many researchers have provided extensive evidence on the association between microbes and drugs, traditional biomedical experiments are time-consuming, labor-intensive, and costly ( Paul et al., 2010 ). These reasons hinder the efficiency of drug development and hardly satisfy the massive demands for novel drugs. Therefore, it is necessary to explore the microbe-drug associations at a large-scale level for drug development.

To overcome the above challenges, computational models have emerged as an effective method for identifying microbe-drug associations, and these models are used to predict microbe-drug associations by integrating different genomic information, including genomics, macro genomics, and metabolomics. With the rapid development of high-throughput sequencing technology and advanced genomics techniques, the research on microbe-drug association prediction has developed rapidly, generating a large amount of valuable research data. To further investigate the potential association between microbes and drugs, a series of microbe-drug association databases have been constructed in recent years, such as aBiofilm ( Rajput et al., 2018 ), MDAD ( Sun et al., 2018 ) and DrugVirus ( Andersen et al., 2020 ), which have immensely promoted the development of microbe-drug association prediction models. Over the past few years, many computational models have emerged that utilize the above databases to infer potential associations between microbes and drugs. As an illustration, Zhu et al. proposed a computational method, HMDAKATZ, which applied the KATZ measure to predict inherent associations between microbes and drugs ( Zhu et al., 2019b ). Long et al. (2020) proposed a computational method called GCNMDA, which combined graph convolutional networks (GCNs) and conditional random fields (CRFs) with an attentional mechanism aiming to identify the hidden associations between microbes and drugs. In 2021, GATMDA was proposed, which utilized inductive matrix completion and graph attention networks (GNNs) to predict associations between microbes and diseases ( Long et al., 2021 ). The Graph2MDA model combined the constructed multimodal attribute graphs and variational graph autoencoder (VGAE) to predict microbe-drug associations accurately ( Deng et al., 2022 ). GSAMDA is likewise a microbe-drug association prediction model, which primarily applies graph attention networks (GATs) and sparse autoencoders ( Tan et al., 2022 ). The computational model NIRBMMDA ( Cheng et al., 2022 ) combines neighborhood-based inference (NI) and restricted Boltzmann machine (RBM) methodologies to predict Microbe-Drug Associations (MDA). By leveraging NI, it extracts proximity information from microbes or drugs, while RBM is used to learn the latent probability distribution inherent in the known association data. This integrative approach harnesses the strengths of both components, resulting in a more robust predictive framework. In the study of Tian et al. (2023) , they proposed the SCSMDA model, which was based on GCN and integrated structure-enhanced contrast learning and self-paced negative sampling strategies to improve the accuracy in microbe-drug association prediction. In addition, the GACNNMDA model integrated a GTA-based autoencoder and a CNN-based classifier, which transforms multiple attribute combinations of the microbes and drugs into two feature matrices to predict the associations of the microbes and drugs ( Ma et al., 2023 ). Qu et al. (2023) proposed MHBVDA to predicts virus-drug associations by integrating multiple biological data sources and employing integrating two matrix decomposition-based methods. And it innovatively applies Bounded Nuclear Norm Regularization (BNNR) with regularization terms to mitigate the impact of noisy data and overfitting issues, thereby enhancing prediction accuracy. However, these methods based on graph neural networks still have room for improvement in prediction performance. When multi-layer networks are stacked, there is some confusion between different orders of neighborhood information, the node representations become indistinguishable, and the network performance decreases, which tends to prevent GNN with multiple layers from effectively utilizing the higher-order neighborhood information ( Li et al., 2018 ).

Therefore, to achieve better prediction performance, inspired by the work of Song et al. (2023) , this paper proposed an ordered gating mechanism-based ordered message-passing GNN method to infer potential microbe-drug associations, called OGNNMDA. In OGNNMDA, the known association data are preprocessed to compute Gaussian interaction profile kernel similarity and additional biomedical information similarity (microbe functional similarity, drug structural similarity) for drugs and microbes, respectively. Then, the multiple similarity matrices are fused and stitched together to obtain the heterogeneous networks. The heterogeneous network was fed into the encoder consisting of the two-layer fully connected network and the 12-layer ordered message-passing GNN to derive embedding representations of the drugs and microbes, respectively. Finally, the bilinear decoder was adopted to reconstruct the microbe-drug association matrix to infer possible associations between the microbes and drugs. Furthermore, to evaluate the predictive performance of OGNNMDA, in-depth comparative experiments, ablation experiments, and case studies are conducted in this paper. The results demonstrate that OGNNMDA outperforms current representative existing methods and achieves satisfactory results in potential drug-microbe association prediction.

All the aBiofilm, MDAD and DrugVirus datasets provide important insights into the complex interactions between the drugs and the microbes, providing researchers in the fields of bioinformatics and graphical neural networks with a wealth of information to analyze and utilize to advance their studies and methods. The basic statistical information of the three datasets is presented in Table 1 .

Table 1 . Statistical information about the datasets.

2.1 aBiofilm

In 2018, Rajput et al. introduced the aBiofilm ( http://bioinfo.imtech.res.in/manojk/abiofilm/ ) dataset, which is of great significance for the development of the bioinformatics and graph neural network fields ( Rajput et al., 2018 ). Over the last three decades, many anti-biofilm agents have been experimentally verified to disrupt biofilms. aBiofilm organizes these data, which contain a database, a predictor, and a data visualization module. The database contains biological, chemical, and structural details of 5,027 anti-biofilm agents (1720 different ones) reported from 1988 to 2017. After eliminating redundant associations among them, a total of 2,884 known interaction associations of 1720 drugs and 140 microbes were finally obtained.

MDAD ( https://github.com/Sun-Yazhou/MDAD/ ) is also a valuable microbe-drug association dataset, which was proposed by Sun et al. based on a variety of drug-related databases as well as a large amount of literature ( Sun et al., 2018 ). Specifically, MDAD contains 5,505 associations between 180 microbes and 1,388 drugs collected from 993 documentation. After filtering out redundant information, a total of 2,470 microbe-drug associations were obtained, involving 173 microbes and 1,373 drugs.

2.3 DrugVirus

DrugVirus ( https://drugvirus.info/ ) compiles interactions involving 118 virus-targeting drugs and 83 human viruses, encompassing SARS-CoV-2 (2019-nCoV) ( Andersen et al., 2020 ). Building upon this foundation, Lond et al. systematically extracted and curated 57 drug-virus associations from pertinent drug databases and scholarly publications, which involved 76 unique drugs and 12 distinct viruses. Ultimately, they assembled a dataset comprising 175 drugs and 95 viruses, yielding a total of 933 documented drug-virus interaction records.

3 Preprocessing

In this section, firstly, the definition of the association adjacency matrix is given, secondly, the similarity calculation of drugs and microbe based on the adjacency matrix is given, and finally, the heterogeneous network is obtained based on multiple similarities.

For simplicity, for each dataset, let D = d 1 , d 2 , … , d N d denote the set of different drugs, and M = m 1 , m 2 , … , m N m denote the set of different microbes. Therefore, a primitive known microbe-drug association network N e t = D ∪ M , E can be constructed: for each given drug d i 1 ≤ i ≤ N d and microbe m j 1 ≤ j ≤ N m there exists a unique edge corresponding to it in E if and only if there is a known association between them. Based on the above definition, the adjacency matrix A ∈ R N d × N m can be obtained as shown in Eq. 1 .

That is, for any given d i 1 ≤ i ≤ N d and m j 1 ≤ j ≤ N m , there is A i , j = 1 if and only if there is an edge between them in E . Otherwise, A i , j = 0.

3.1 Constructing drug-drug similarity networks

First, considering that the functions of drugs are determined by their microstructures, and drugs with similar structures have similar chemical properties. So, the SIMCOMP2 tool based on the maximum common substructure between drugs is used in this paper to calculate the drug structure similarity ( Hattori et al., 2010 ). For two drugs d i and d j respectively, their structure-based similarity can be expressed as DSS ( d i , d j ). After calculating all the similarities between all drug pairs, an N d × N d matrix D S S ∈ R N d × N d can be obtained to represent the chemical structure similarities between N d different drugs.

Next, for any two given drugs or microbes, the Gaussian interaction profile kernel similarity between them is calculated herein by utilizing a Gaussian kernel function based on known microbe disease associations as shown in Eq. 2 :

where A ( i , :) and A ( j , :) denote the ith and jth rows of the adjacency matrix A , respectively, and γ d denotes the drug-normalized kernel bandwidth, which can be calculated by Eq. 3 .

3.2 Constructing microbe-microbe similarity networks

Also, this paper measures microbe similarity in two ways. The first one is the functional similarity of microbe proposed by Kamneva (2017) . This computational method is mainly based on the microbial gene family information kernel protein-protein interaction association network. The second similarity between microbes is the Gaussian interaction profile kernel similarity MGS. similar to the drug similarity based on the Gaussian interaction profile kernel, for any given microbe pair m i and m j , it is computed using the Gaussian kernel function based on the known microbe drug associations as shown in Eq. 4 .

where A (:, i ) and A (:, j ) denote the i th and j th columns of the adjacency matrix A , respectively, and γ m denotes the microbe normalized kernel bandwidth that can be computed according to Eq. 5 .

3.3 Constructing the heterogeneous network

Considering that not all drugs have their structures retrieved from databases, it is not possible to obtain all chemical structure similarities between drugs lacking structural information and other drugs. Therefore, in this paper, a comprehensive similarity is constructed to estimate the similarity between drugs and microbes by integrating Gaussian interaction profile nuclear similarity, microbe functional similarity, and drug chemical structure similarity. Specifically, for any two given drugs d i and d j , the integrated similarity between them is calculated as shown in Eq. 6 :

In addition, for any given microbe pair m i and m j , the combined similarity between them is calculated as shown in Eq. 7 :

Then, the heterogeneous network H ∈ R ( N d + N m ) × ( N d + N m ) , shown in Eq. 8 , can be constructed by combining the above integrated microbe similarity network D S ∈ R N d × N d , the integrated disease similarity network M S ∈ R N m × N m and the known drug-microbe association network A ∈ R N d × N m .

Next, the model uses above newly constructed heterogeneous network H as an input to the GNN-based encoder to learn the low dimensional embedding representations of the drugs and microbes.

Figure 1 illustrates the framework of OGNNMDA, comprising three primary modules: the input module, encoder module, and decoder module. The input module is responsible for extracting multiple biomedical information features to be utilized as inputs for OGNNMDA. The encoder module focuses on learning the node embedding representation of the microbes and drugs. Lastly, the decoder module employs bilinear decoders to predict new drug-microbe associations.

Figure 1 . Flowchart of the OGNNMDA.

4.1 Encoder

OGNNMDA is a graph neural network that directly processes the graph as input, effectively utilizing both node information and structural characteristics. Graph neural networks have gained significant popularity in link prediction tasks ( Zhang and Chen, 2018 ), showcasing their widespread adoption. By leveraging the adjacency matrix H obtained earlier, Eq. 9 defines the specific formulation of the GNN.

Here, l ∈ 1 … L conv , h v l ∈ R 1 × k is the embedding feature of the layer l , N v denotes the set of neighboring nodes for the node v , L conv corresponds to the number of layers in the GNN network and the number of message-passing rounds. The dimension of the node’s embedding feature is denoted by k . In this study, the final embedding dimension is set to match the embedding dimensions used across the GNN layers. H is the microbe-drug heterogeneous network graph defined in Eq. 8 , which is processed for embedding and provides edge information for the GNN. The node representation h 0 ∈ R N d + N m × k is obtained by a two-layer MLP defined by Eq. 10 and 11 . The trainable variables W f c 1 , W f c 2 ∈ R N d + N m × k and B f c 1 , B f c 2 ∈ R k are involved in this process. Additionally, H init ∈ R N d + N m × N d + N m represents the initial node representation, and σ denotes the ReLU activation function.

The function ϕ calculates the messages transmitted between nodes, where the edge attribute is directly used as the message. The symbol □ represents the message aggregation function, and in this paper, the mean method is employed ( Huan et al., 2021 ). This means that messages received from multiple neighboring nodes are aggregated by taking their average, resulting in message characteristics used for updating node representations. Finally, γ represents the node representation update function, which implements the ordered message-passing mechanism discussed in this paper.

In the message-passing process of a single-level GNN, a node only exchanges messages with its immediate neighbors. This pattern of neighbor message transmission at different orders aligns with the structure of the node root tree in a multi-layer GNN ( Liu et al., 2020 ). As illustrated in Figure 2 , for a node v , N v l represents the neighbor information of node v at the l th layer, and the nesting relationship of its neighbor messages at each layer can be described using Eq. 12 .

Figure 2 . Taking a two-layer GNN as an example, layer 0 represents the initial node embedding, and the adjacency of nodes between layers forms multiple trees. In the figure, u is a neighbor node of v . N v ( 2 ) and N u ( 1 ) are shown in the image with two colors respectively. The right side shows the tree structure of neighbor information with v node as the viewpoint, and the arrow represents the direction of neighbor information transfer.

In single-layer message passing, direct-neighbor node messages and higher-order neighbor node messages are differentially encoded to ensure orderly message delivery. Specifically, the neuron rows are aligned with the node root tree at each layer, enabling the acquisition of node feature representations with consistent nesting relationships. To implement this alignment encoding method, the neurons can be ordered by linearly arranging the neurons of each layer and considering a segmentation point, denoted as s . The information of the neighbors of the current node v , at order one or higher, can be encoded as s v l ( Song et al., 2023 ). The segmentation point s corresponds to the nested nature of node v , and its size relationship is determined by Eq. 13 .

Next, we describe the node feature update function γ , which is exemplified below for a specific node v . The function can be divided into three distinct steps.

1. Compute the aggregated message representation m l ∈ R N d + N m × k for layer l .

2. For node v , this paper utilizes the gating vector g ̂ v l of dimension N d + N m to describe the segmentation point s v l . Specifically, the value of the left part 0 , s v l − 1 is set to 1, indicating the neighboring neurons of node v that are higher than the first order. Conversely, the value of the right part s v l , N d + N m − 1 is set to 0, denoting direct neighboring neurons. This is achieved by calculating the cumulative sum of the probability that each position in the node servers as a split point s v l . The expectation gating vector g ̂ v l is obtained through a biased linear projection of the node representation vector in layer l − 1 and the message vector in layer l , as shown in Eq. 15 .

In Eq. 15 , the trainable parameters W g l ∈ R 2 k × k and B g l ∈ R k are utilized. Additionally, h v l − 1 ; m v l represents the concatenation of two vectors h v l − 1 and m v l . To ensure that the predicted gated vector g ̂ v l adheres to the relative size relationship of the splitting points mentioned earlier, the operation described in Eq. 16 . This operation yields the final gated vector g v l .

3. Equation 17 demonstrates the utilization of the gating vector g v l to regulate the integration of the layer l − 1 node representation h v l − 1 with the layer l aggregated context m v l . This process results in the acquisition of the new node representation h v l .

In Eq. 17 , the symbol ⋅ represents element-by-element multiplication, and LN refers to the layer normalization operation ( Chen et al., 2022 ).

4.2 Decoder

After the previous rounds of the ordered message passing process, the final node embedding representation h L conv ∈ R N d + N m × k is obtained. This representation can be considered as the concatenation of the final embedding features of the drugs, h d ∈ R N d × k , and the microbes, h m ∈ R N m × k . In this paper, the final embedding features h d and h m are obtained separately using the matrix splicing approach defined in Eq. 18 .

To reconstruct the adjacency matrix A ′ representing possible microbe-disease associations, the bilinear decoder is employed. It is a structural component employed for predicting the probability of potential edges or links based on node embedding vectors. These decoders commonly integrate the embedding vectors of node pairs within a graph to generate a score function that assesses the likelihood of a link between two nodes. The key characteristic of bilinear decoders lies in their utilization of bilinear transformations to capture the interaction effects among nodes. Specifically, for a drug node and microbe node pair (u, v) with their respective embedding vectors h d ( u ) and h m ( v ), a bilinear decoder might compute the score by Eq. 19 .

Where W is a learnable weight matrix. This score can be interpreted as the probability of link occurrence after a nonlinear activation function transformation, so that A ′ can be obtained by the bilinear decoder as shown in Eq. 20 .

In the above formula, where W B ∈ R k × k represents a trainable matrix and σ x = 1 / 1 + e − x is the sigmoid function. Overall, the complete computational flow of OGNNMDA can be seen in Algorithm 1 .

Algorithm 1. OGNNMDA.

Require: Known associations matrix A ∈ R N d × N m , drug similarity matrix D S ∈ R N d × N d , microbe similarity matrix M S ∈ R N m × N m and α = 600 is the number of iterations for OGNNMDA

Ensure: The constructed drug-microbe associations matrix A ′ ∈ R N d × N m

  1:  Construct the heterogeneous network H according to formula ( 8 )

  2:  Initialize the embedding feature matrix H init according to formula ( 11 ).

  3:  Initialize the gate vector = 0

  4:   for i = 1 → α do

  5:   calculate h 0 according to formula ( 10 )

  6:    for l = 1 → L conv do

  7:    calculate message matrix m l formula ( 14 ).

  8:    calculate g ̂ l by formula ( 15 )

  9:    calculate g ̃ l formula ( 16 )

  10:    calculate h l formula ( 17 )

  11:    end for

  12:   get the embedding feature for drugs and microbes with h d and h m according to formula ( 18 )

  13:   get the reconstruction matrix A ′ by formula ( 20 )

  14:   end for

4.3 Optimization

During the experiment, positive samples were the drug-microbe pairs with known associations, while negative samples were the drug-microbe pairs without known associations. These sets of positive and negative samples are denoted as Ω + and Ω − , respectively, for ease of description. It is important to note that the number of pairs with known associations in both the aBiofilm dataset and the MDAD dataset is significantly smaller than the number of pairs without known associations. Therefore, when training OGNNMDA, the loss function incorporates a weighted cross-entropy loss, as defined in Eq. 21 .

In the above formula, ( i , j ) represents a pair of the drug d i and microbe m j . λ is introduced as a balancing factor, calculated as the ratio of the number of samples in Ω − to the number of samples in Ω + . This factor helps attenuate the impact of data imbalance and emphasizes the reinforcement of known correlation information.

In this paper, the Xavier initialization method ( Duong et al., 2019 ) is employed to initialize the trainable parameter matrices in various components of the model. These include the 2-layer fully connected layer, the ordered message-passing graph neural network layer, the bilinear decoder, and others, denoted as W f c l , B f c l | W f c l ∈ R N d + N m × k , B f c l ∈ R k , 1 ≤ l ≤ K f c , W g l , B g l | W g l ∈ R 2 * k × c s , B g l ∈ R c s , 1 ≤ l ≤ K conv , and the bias matrix W B ∈ R k × k . Furthermore, the Adam optimizer ( Wang et al., 2023 ) is utilized to minimize the loss function. Adam combines the benefits of momentum optimization and adaptive learning rate, enabling quick convergence and adaptation to different parameter learning rates during the training process. This optimization technique enhances the training effectiveness of the deep learning model.

To prevent overfitting, the paper introduces node dropout ( Piotrowski et al., 2020 ) and regularized dropout ( Berg et al., 2017 ) schemes in the graph convolution layer. Node dropout can be seen as training multiple models on various sub-nodes, and the combination of these sub-nodes is used to predict unknown microbe-drug pairs ( Tan et al., 2020 ).

This paper begins by providing a brief overview of the experimental setup and the analysis and selection of certain hyperparameters. The aim is to validate the predictive performance advantages of OGNNMDA through intensive comparison experiments. These experiments involve 6 representative microbe-drug association prediction models, including state-of-the-art approaches. The evaluation is conducted on three well-known public datasets, namely, aBiofilm, MDAD and DrugVirus, within a 5-fold cross-validation framework. Furthermore, ablation experiments are performed to investigate the effectiveness of the ordered message-passing mechanism employed in OGNNMDA. Finally, a case study is presented to validate OGNNMDA using two commonly used drugs, ciprofloxacin and moxifloxacin, along with two common oral microbes, Actinobacillus aggregatum and Clostridium nucleatum.

5.1 Experimental parameter setting

In this paper, all experimental evaluations are conducted within a five-fold cross-validation setup. To ensure statistical robustness, each method is executed ten independent times for every experiment, thereby enabling the calculation of the mean value for each performance metric across these repetitions. In detail, this involves dividing all known associations in the dataset equally into 5 parts, denoted as t e s t p = t p 1 , t p 2 , t p 3 , t p 4 , t p 5 . Additionally, a subset of the same size as the known associations is randomly selected from the unknown association set. This subset is divided equally into 5 parts, denoted as t e s t n = t n 1 , t n 2 , t n 3 , t n 4 , t n 5 .

During the i − th (1 ≤ i ≤ 5) cross-validation iteration, the training set is defined as t r a i n i = t e s t p − t p i , and the test set is defined as t e s t i = t p i ∪ t n i . The final test result of the 5-fold cross-validation experiment is calculated based on the combined test set, test = test p ∪ test n .

Based on the previous description of the model structure, OGNNMDA incorporates several hyperparameters, including the dimension size ( k ) of embedded features, the number of fully-connected layers ( L fc ), the number of ordered message-passing GNN layers ( L conv ), the initial learning rate ( r ) of Adam’s optimizer, the total training period ( α ), the node dropout metrics ( β ), and the regularized dropout parameter ( γ ).

To establish initial values for these parameters, we set L fc = 2, r = 0.008, α = 600, β = 0.6, and γ = 0.4. Subsequently, we examine the effects of different values for parameters k and L conv through experimental analysis.

To investigate the impact of different hyperparameter values on the model, this paper performed 5-fold cross-validation (5 cv) experiments on the aBiofilm and MDAD datasets. The results for the AUROC were plotted in Figure 3 , showcasing the outcomes for various combinations of the parameters L conv and k .

Figure 3 . (A) Model hyperparameter analysis on the aBiofilm dataset. (B) Model hyperparameter analysis on the MDAD dataset.

From Figures 3A, B , it is evident that the optimal combination of L conv and k is L conv = 12 and k = 512. Therefore, this parameter setting will be utilized for OGNNMDA in subsequent experiments.

5.2 Comparison experiments

In this study, we replicate the code and data based on publicly accessible resources of these six methodologies, with all competing methods’ parameter configurations set according to their optimal values as reported in their respective publications. The 6 methods we compared OGNNMDA with are HMDAKATZ ( Zhu et al., 2019a ), GCNNMDA ( Long et al., 2020 ), GSAMDA ( Tan et al., 2022 ), SCSMDA ( Tian et al., 2023 ), LAGCN ( Yu et al., 2021 ), and NTSHMDA ( Luo and Long, 2018 ), which are widely used in linkage prediction problems across various bioinformatics domains. However, due to GSAMDA not having performed experiments on DrugVirus in their paper nor specifying the construction process for the microbe-disease associations and drug-disease associations used to derive disease-based microbial and drug-Hamming similarities, comparative evaluations on DrugVirus are limited to the remaining five competing approaches.

To train and evaluate these methods, a 5-fold cross-validation experimental framework was employed. Performance evaluation was based on metrics such as AUC, AUPR, accuracy, and F1 score, chosen for their effectiveness in assessing performance. The experimental results, including the performance metrics, are presented in Tables 2 – 4 . Additionally, ROC curves (see Figure 4A , 5A , 6A ) and PR curves (see Figure 4B , 5B , 6B ) were plotted to facilitate comparison among the different methods on the respective datasets.

Table 2 . Comparison of AUC, AUPR, Acc, and F1-score obtained by each method based on aBiofilm dataset at 5-cv.

Table 3 . Comparison of AUC, AUPR, Acc and F1-score obtained by each method based on MDAD dataset at 5-cv.

Table 4 . Comparison of AUC, AUPR, Acc and F1-score obtained by each method based on DrugVirus dataset at 5-cv.

Figure 4 . (A) ROC curves for each modeling approach based on the aBiofilm dataset 5-cv. (B) PR curves for each modeling approach based on the aBiofilm dataset 5-cv.

Figure 5 . (A) ROC curves for each modeling approach based on the MDAD dataset 5-cv. (B) PR curves for each modeling approach based on the MDAD dataset 5-cv.

Figure 6 . (A) ROC curves for each modeling approach based on the DrugVirus dataset 5-cv. (B) PR curves for each modeling approach based on the DrugVirus dataset 5-cv.

Based on the experimental results from Table 2 , it is evident that OGNNMDA achieves the highest AUC values on the aBiofilm dataset, with an average AUC of 0.9693 ± 0.0008. This is 0.65% higher than the next highest AUC value of 0.9628 ± 0.0021 obtained by SCSMDA. OGNNMDA also outperforms other methods in terms of AUPR, Accuracy, and F1-Score, with values of 0.9690 ± 0.0009, 0.9141 ± 0.0031, and 0.9151 ± 0.0026, respectively.

Similarly, in Table 3 , which presents the results on the MDAD dataset, OGNNMDA exhibits superior performance across all four evaluation metrics. The comparison between the two tables suggests that OGNNMDA performs better on the aBiofilm dataset compared to MDAD. This disparity can be attributed to the sparser nature of the data in MDAD, resulting in a smaller ratio of positive to negative samples and a more pronounced sample imbalance issue.

Finally, we examine the results from Table 4 , which presents the performance of all methods on the DrugVirus dataset. OGNNMDA achieved the highest AUPR score with a mean value of 0.8633 ± 0.0078; however, SCS-MDA outperformed others in terms of the AUC (0.8810 ± 0.0053), Accuracy (0.8098 ± 0.0071), and F1-score (0.8201 ± 0.0038). Notably, OGNNMDA did not maintain its leading position on the DrugVirus dataset as it did on the aBiofilm and MDAD datasets. This relative underperformance may be attributed to the smaller scale of the DrugVirus dataset compared to aBiofilm and MDAD, potentially limiting OGNNMDA’s ability to effectively train its more complex weighting parameters for optimal prediction.

5.3 Ablation experiment

To evaluate the efficacy of the ordered message-passing mechanism, this section presents ablation experiments, the results of which are presented in Table 5 . In this context, GNN refers to a simple graph neural network model utilizing a mean aggregator as an encoder, while OGNN represents an enhanced ordered message-passing graph neural network model based on GNN, specifically the model proposed in this paper, OGNNMDA. The evaluation entails 5-fold cross-validation experiments on the aBiofilm and MDAD datasets, with specific parameter settings described in previous sections.

Table 5 . Results of ablation experiments.

Based on the data presented in Table 5 , the underlying GNN encoder exhibits poor performance on both datasets, showing a significant gap in all metrics compared to the OGNNMDA model utilizing OGNN as the encoder. Therefore, it is reasonable to conclude that the ordered message-passing mechanism effectively enhances the embedding performance of GNN, leading to improved prediction results in microbe drug association prediction.

5.4 Case study

To validate the prediction performance of OGNNMDA, case study experiments were conducted using two popular drugs and two microbes as targets. First, OGNNMDA was trained on the complete aBiofilm dataset to obtain the predicted association information neighbor matrix. Then, the top 20 most relevant objects for each target microbe and drug were filtered out. Finally, the relevant published PubMed literature was searched to validate the predicted microbe-drug association pairs against existing references. The first drug selected for the case study was ciprofloxacin, a fluorinated quinolone antibiotic, which has been extensively studied and shown to be associated with a wide range of human microbiome ( Yayehrad et al., 2022 ). For instance, Rehman et al. (2019) demonstrated the effectiveness of amphotericin-B and 5% ciprofloxacin in blocking the growth mechanisms of Pseudomonas aeruginosa and Candida albicans. Ciprofloxacin has also shown susceptibility against Staphylococcus aureus , Staphylococcus epidermidis, Mycobacterium subspecies, Escherichia coli , and Mycobacterium tuberculosis ( Smirnova and Oktyabrsky, 2018 ). The second drug chosen for the case study is moxifloxacin, a fluoroquinolone antibiotic ( Rodríguez-López et al., 2020 ), known to be associated with antibiotic-resistant bacteria (ARB) ( Loyola-Rodriguez et al., 2018 ) and Listeria monocytogenes ( Rodríguez-López et al., 2020 ). The specific experimental results for the two drugs are presented in Tables 6 , 7 , respectively. These tables provide supporting literature information for the top 20 predicted microbes associated with ciprofloxacin and moxifloxacin. Upon observing Tables 6 , 7 , it is evident that 20 and 17 out of the top 20 predicted microbes associated with ciprofloxacin and moxifloxacin, respectively, have been validated by the available literature.

Table 6 . Top 20 related microbes to Ciprofloxacin predicted by OGNNMDA.

Table 7 . Top 20 related microbes to Moxifloxacin predicted by OGNNMDA.

Furthermore, the first microbe selected for the case study was Aggregate Actinobacteria Accompanying Bacteria, a Gram-negative bacterium belonging to the family Pasteuriaceae ( Krueger and Brown, 2020 ). It is primarily found in the oral cavity and is associated with various oral diseases and systemic infections ( Jensen et al., 2019 ). In terms of its impact on human health, aggregates of Actinobacillus companionis are commonly linked to periodontal diseases, particularly aggressive forms of periodontitis. This bacterium has the ability to invade and colonize periodontal tissues, leading to inflammation, destruction of the periodontal ligament, and bone loss. Consequently, it is often found at a higher rate in individuals with severe periodontal disease. Sol et al. demonstrated that sub-killer concentrations of LL-37, Cathelicidin, and Scrambled LL-37 inhibit the biofilm formation of Actinobacillus actinomycetemcomitans and act as conditioning agents and lectins, greatly enhancing clearance by neutrophils and macrophages ( Sol et al., 2013 ). Basavaraju et al. found that AHL lactonase hydrolyzes the lactone ring in the high serine portion of AHL, without affecting the rest of the signaling molecular structure. This inhibitory effect of AHL lactonase on group sensing of actinomycete aggregates has been observed ( Basavaraju et al., 2016 ). The second microbe chosen for the case study was Clostridium nucleatum, a bacterium known for causing opportunistic infections and recently associated with colorectal cancer ( Brennan and Garrett, 2019 ). In this study, Tables 8 , 9 present the top 20 predicted drugs that are most relevant to Aggregate Actinobacteria Accompanying Bacteria and Clostridium nucleatum, respectively. Based on the information in the tables, 17 out of the top 20 predicted drugs for Aggregate Actinobacteria Accompanying Bacteria and 18 out of the top 20 predicted drugs for Clostridium nucleatum have been validated in the existing literature. Therefore, it can be concluded that OGNNMDA achieves satisfactory predictive performance in both microbe and drug case studies.

Table 8 . Top 20 drugs associated with the microbe Aggregatibacter actinomycetemcomitans predicted by OGNNMDA.

Table 9 . Top 20 drugs associated with the microbe Fusobacterium nucleatum as predicted by OGNNMDA.

6 Conclusion and discussion

This paper proposes OGNNMDA, a novel deep learning model for predicting potential microbe-drug associations, based on graph neural networks (GNNs) with an ordered message-passing mechanism. OGNNMDA utilizes multiple sources of biological data to construct similarity features for drugs and microbes, which are combined to form a heterogeneous network containing association and similarity information. To obtain drug and microbe embeddings, a multilayer GNN with ordered message passing is employed to differentiate node neighborhood messages during the message passing stage. A bilinear decoder is then used to generate association prediction scores. The OGNNMDA methodology was subjected to a rigorous evaluation regimen, encompassing comparative experiments on the aBiofilm and MDAD datasets as well as the DrugVirus dataset, where it utilized a 5-fold cross-validation scheme. The empirical outcomes revealed that OGNNMDA surpassed the current state-of-the-art performance benchmarks on both the aBiofilm and MDAD datasets. However, in the context of the DrugVirus dataset, OGNNMDA demonstrated a commendable yet second-best performance compared to existing methods. For clarity, while comprehensive experimental evaluations including comparative analyses were conducted for the DrugVirus dataset, the ablation experiments and case studies were confined to the aBiofilm and MDAD datasets alone. Despite this, the overall results affirm OGNNMDA’s robustness and competitive advantage in predicting potential microbe-drug associations across different datasets. The main contributions of this model can be summarized as follows.

1. It fully leverages additional biomedical data, such as microbe functional similarity based on microbial genomic information and drug molecular structural phase-based feature similarity.

2. It introduces an improved GNN model with an ordered message-passing mechanism, which achieves better embedding performance by distinguishing node neighbor messages.

3. The overall model outperforms existing state-of-the-art methods for predicting potential microbe-drug associations.

However, OGNNMDA is not without its limitations. The model’s performance is contingent upon the scale of the accessible dataset; with a relatively modest-sized corpus, the inherent sparsity in the microbial-drug association adjacency matrix can potentially impede the exhaustive exploitation of the graph’s structural information and limit the expressiveness of the learned embeddings. Furthermore, OGNNMDA homogenously handles microbial and drug nodes within the network without explicitly accounting for their distinctive patterns of interaction. In light of these challenges, future research directions can be directed towards:

1. Expanding Feature Representation: Augmenting the existing feature space by integrating supplementary biomedical data such as genomic sequences of microbes ( Deng et al., 2022 ) and pharmacological similarity based on side effect profiles ( Zheng et al., 2019 ). This enrichment could provide deeper insights into the intrinsic properties of both microorganisms and drugs, thereby enhancing the quality of the representations learned.

2. Addressing Sparsity Issues: Investigating innovative techniques to tackle the issue of sparse associations, which might involve adopting advanced link prediction strategies or devising specialized regularization methods that are tailored for sparse graphs. These approaches could ensure more efficient utilization of available relational information.

3. Adaptation of Graph Contrastive Learning: Exploring the potential benefits of incorporating graph contrastive learning (GCL) paradigms to improve the robustness and generalizability of the learned embeddings. GCL has shown promise in other domains by extracting meaningful node or graph representations from limited or unlabeled data, hence it could be a viable avenue to mitigate the impact of small datasets on OGNNMDA’s performance ( Cai et al., 2023 ).

4. Refinement of Message-Passing Mechanisms: Examining alternative graph neural network architectures like Graph Attention Networks (GATs) and Graph Convolutional Networks (GCNs), and refining their message-passing processes to better suit the unique characteristics of the microbial-drug association problem.

By systematically addressing these limitations and venturing into new methodological frontiers, future iterations of OGNNMDA and similar models are poised to achieve heightened accuracy and resilience in predicting microbe-drug associations, thus contributing significantly to this burgeoning research domain.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

JZ: Data curation, Software, Writing–original draft, Writing–review and editing. LK: Writing–review and editing. AH: Writing–review and editing. QZ: Writing–review and editing. DY: Writing–review and editing. CW: Data curation, Writing–review and editing.

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was partly sponsored by the National Natural Science Foundation of China (No. 62272064). This work was carried out in part using computing resources at the High Performance Computing Platform of Xiangtan University.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Amato, K. R., Mallott, E. K., McDonald, D., Dominy, N. J., Goldberg, T., Lambert, J. E., et al. (2019). Convergence of human and old world monkey gut microbiomes demonstrates the importance of human ecology over phylogeny. Genome Biol. 20, 201–212. doi:10.1186/s13059-019-1807-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Andersen, P. I., Ianevski, A., Lysvand, H., Vitkauskiene, A., Oksenych, V., Bjørås, M., et al. (2020). Discovery and development of safe-in-man broad-spectrum antiviral agents. Int. J. Infect. Dis. 93, 268–276. doi:10.1016/j.ijid.2020.02.018

Basavaraju, M., Sisnity, V. S., Palaparthy, R., and Addanki, P. K. (2016). Quorum quenching: signal jamming in dental plaque biofilms. J. Dent. Sci. 11, 349–352. doi:10.1016/j.jds.2016.02.002

Berg, R. V. D., Kipf, T. N., and Welling, M. (2017). Graph convolutional matrix completion. arXiv preprint arXiv:1706.02263.

Google Scholar

Brennan, C. A., and Garrett, W. S. (2019). Fusobacterium nucleatum—symbiont, opportunist and oncobacterium. Nat. Rev. Microbiol. 17, 156–166. doi:10.1038/s41579-018-0129-6

Cai, X., Huang, C., Xia, L., and Ren, X. (2023). Lightgcl: simple yet effective graph contrastive learning for recommendation. arXiv preprint arXiv:2302.08191.

Catinean, A., Neag, M. A., Muntean, D. M., Bocsan, I. C., and Buzoianu, A. D. (2018). An overview on the interplay between nutraceuticals and gut microbiota. PeerJ 6, e4465. doi:10.7717/peerj.4465

Chen, Y., Tang, X., Qi, X., Li, C.-G., and Xiao, R. (2022). Learning graph normalization for graph neural networks. Neurocomputing 493, 613–625. doi:10.1016/j.neucom.2022.01.003

CrossRef Full Text | Google Scholar

Cheng, X., Qu, J., Song, S., and Bian, Z. (2022). Neighborhood-based inference and restricted Boltzmann machine for microbe and drug associations prediction. PeerJ 10, e13848. doi:10.7717/peerj.13848

Cullin, N., Antunes, C. A., Straussman, R., Stein-Thoeringer, C. K., and Elinav, E. (2021). Microbiome and cancer. Cancer Cell 39, 1317–1341. doi:10.1016/j.ccell.2021.08.006

Deng, L., Huang, Y., Liu, X., and Liu, H. (2022). Graph2mda: a multi-modal variational graph embedding model for predicting microbe–drug associations. Bioinformatics 38, 1118–1125. doi:10.1093/bioinformatics/btab792

Duong, C. T., Hoang, T. D., Dang, H. T. H., Nguyen, Q. V. H., and Aberer, K. (2019). On node features for graph neural networks. arXiv preprint arXiv:1911.08795 .

Hajiagha, M. N., Taghizadeh, S., Asgharzadeh, M., Dao, S., Ganbarov, K., Köse, Ş., et al. (2022). Gut microbiota and human body interactions; its impact on health: a review. Curr. Pharm. Biotechnol. 23, 4–14. doi:10.2174/1389201022666210104115836

Hattori, M., Tanaka, N., Kanehisa, M., and Goto, S. (2010). Simcomp/subcomp: chemical structure search servers for network analyses. Nucleic acids Res. 38, W652–W656. doi:10.1093/nar/gkq367

Huan, Z., Quanming, Y., and Weiwei, T. (2021). “Search to aggregate neighborhood for graph neural network,” in 2021 IEEE 37th International Conference on Data Engineering (ICDE) ( IEEE ), 552–563.

Jensen, A. B., Haubek, D., Claesson, R., Johansson, A., and Nørskov-Lauritsen, N. (2019). Comprehensive antimicrobial susceptibility testing of a large collection of clinical strains of aggregatibacter actinomycetemcomitans does not identify resistance to amoxicillin. J. Clin. Periodontology 46, 846–854. doi:10.1111/jcpe.13148

Kamneva, O. K. (2017). Genome composition and phylogeny of microbes predict their co-occurrence in the environment. PLoS Comput. Biol. 13, e1005366. doi:10.1371/journal.pcbi.1005366

Kim, N., Yun, M., Oh, Y. J., and Choi, H.-J. (2018). Mind-altering with the gut: modulation of the gut-brain axis with probiotics. J. Microbiol. 56, 172–182. doi:10.1007/s12275-018-8032-4

Krueger, E., and Brown, A. C. (2020). Aggregatibacter actinomycetemcomitans leukotoxin: from mechanism to targeted anti-toxin therapeutics. Mol. oral Microbiol. 35, 85–105. doi:10.1111/omi.12284

Li, Q., Han, Z., and Wu, X.-M. (2018). Deeper insights into graph convolutional networks for semi-supervised learning. Proc. AAAI Conf. Artif. Intell. 32. doi:10.1609/aaai.v32i1.11604

Liu, M., Gao, H., and Ji, S. (2020). “Towards deeper graph neural networks,” in Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery and data mining , 338–348.

Long, Y., Luo, J., Zhang, Y., and Xia, Y. (2021). Predicting human microbe–disease associations via graph attention networks with inductive matrix completion. Briefings Bioinforma. 22, bbaa146. doi:10.1093/bib/bbaa146

Long, Y., Wu, M., Kwoh, C. K., Luo, J., and Li, X. (2020). Predicting human microbe–drug associations via graph convolutional network with conditional random field. Bioinformatics 36, 4918–4927. doi:10.1093/bioinformatics/btaa598

Loyola-Rodriguez, J. P., Ponce-Diaz, M. E., Loyola-Leyva, A., Garcia-Cortes, J. O., Medina-Solis, C. E., Contreras-Ramire, A. A., et al. (2018). Determination and identification of antibiotic-resistant oral streptococci isolated from active dental infections in adults. Acta Odontol. Scand. 76, 229–235. doi:10.1080/00016357.2017.1405463

Luo, J., and Long, Y. (2018). Ntshmda: prediction of human microbe-disease association based on random walk by integrating network topological similarity. IEEE/ACM Trans. Comput. Biol. Bioinforma. 17, 1341–1351. doi:10.1109/TCBB.2018.2883041

Ma, Q., Tan, Y., and Wang, L. (2023). Gacnnmda: a computational model for predicting potential human microbe-drug associations based on graph attention network and cnn-based classifier. BMC Bioinforma. 24, 35. doi:10.1186/s12859-023-05158-7

Ogunrinola, G. A., Oyewale, J. O., Oshamika, O. O., and Olasehinde, G. I. (2020). The human microbiome and its impacts on health. Int. J. Microbiol. 2020, 8045646. doi:10.1155/2020/8045646

Partula, V., Mondot, S., Torres, M. J., Kesse-Guyot, E., Deschasaux, M., Assmann, K., et al. (2019). Associations between usual diet and gut microbiota composition: results from the milieu intérieur cross-sectional study. Am. J. Clin. Nutr. 109, 1472–1483. doi:10.1093/ajcn/nqz029

Paul, S. M., Mytelka, D. S., Dunwiddie, C. T., Persinger, C. C., Munos, B. H., Lindborg, S. R., et al. (2010). How to improve R&D productivity: the pharmaceutical industry's grand challenge. Nat. Rev. Drug Discov. 9, 203–214. doi:10.1038/nrd3078

Piotrowski, A. P., Napiorkowski, J. J., and Piotrowska, A. E. (2020). Impact of deep learning-based dropout on shallow neural networks applied to stream temperature modelling. Earth-Science Rev. 201, 103076. doi:10.1016/j.earscirev.2019.103076

Pugazhendhi, A., Michael, D., Prakash, D., Krishnamaurthy, P. P., Shanmuganathan, R., Al-Dhabi, N. A., et al. (2020). Antibiogram and plasmid profiling of beta-lactamase producing multi drug resistant staphylococcus aureus isolated from poultry litter. J. King Saud University-Science 32, 2723–2727. doi:10.1016/j.jksus.2020.06.007

Qu, J., Song, Z., Cheng, X., Jiang, Z., and Zhou, J. (2023). A new integrated framework for the identification of potential virus–drug associations. Front. Microbiol. 14, 1179414. doi:10.3389/fmicb.2023.1179414

Rajput, A., Thakur, A., Sharma, S., and Kumar, M. (2018). abiofilm: a resource of anti-biofilm agents and their potential implications in targeting antibiotic drug resistance. Nucleic acids Res. 46, D894–D900. doi:10.1093/nar/gkx1157

Rehman, A., Patrick, W. M., and Lamont, I. L. (2019). Mechanisms of ciprofloxacin resistance in pseudomonas aeruginosa: new approaches to an old problem. J. Med. Microbiol. 68, 1–10. doi:10.1099/jmm.0.000873

Rodríguez-López, P., Rodríguez-Herrera, J. J., and Cabo, M. L. (2020). Tracking bacteriome variation over time in listeria monocytogenes-positive foci in food industry. Int. J. food Microbiol. 315, 108439. doi:10.1016/j.ijfoodmicro.2019.108439

Smirnova, G. V., and Oktyabrsky, O. N. (2018). Relationship between escherichia coli growth rate and bacterial susceptibility to ciprofloxacin. FEMS Microbiol. Lett. 365, fnx254. doi:10.1093/femsle/fnx254

Sol, A., Ginesin, O., Chaushu, S., Karra, L., Coppenhagen-Glazer, S., Ginsburg, I., et al. (2013). Ll-37 opsonizes and inhibits biofilm formation of aggregatibacter actinomycetemcomitans at subbactericidal concentrations. Infect. Immun. 81, 3577–3585. doi:10.1128/IAI.01288-12

Song, Y., Zhou, C., Wang, X., and Lin, Z. (2023). Ordered gnn: ordering message passing to deal with heterophily and over-smoothing. arXiv preprint arXiv:2302.01524.

Sun, Y.-Z., Zhang, D.-H., Cai, S.-B., Ming, Z., Li, J.-Q., and Chen, X. (2018). Mdad: a special resource for microbe-drug associations. Front. Cell. Infect. Microbiol. 8, 424. doi:10.3389/fcimb.2018.00424

Tan, S. Z. K., Du, R., Perucho, J. A. U., Chopra, S. S., Vardhanabhuti, V., and Lim, L. W. (2020). Dropout in neural networks simulates the paradoxical effects of deep brain stimulation on memory. Front. Aging Neurosci. 12, 273. doi:10.3389/fnagi.2020.00273

Tan, Y., Zou, J., Kuang, L., Wang, X., Zeng, B., Zhang, Z., et al. (2022). Gsamda: a computational model for predicting potential microbe–drug associations based on graph attention network and sparse autoencoder. BMC Bioinforma. 23, 492. doi:10.1186/s12859-022-05053-7

Tian, Z., Yu, Y., Fang, H., Xie, W., and Guo, M. (2023). Predicting microbe–drug associations with structure-enhanced contrastive learning and self-paced negative sampling strategy. Briefings Bioinforma. 24, bbac634. doi:10.1093/bib/bbac634

Wang, L., Yang, X., Kuang, L., Zhang, Z., Zeng, B., and Chen, Z. (2023). Graph convolutional neural network with multi-layer attention mechanism for predicting potential microbe-disease associations. Curr. Bioinforma. 18, 497–508. doi:10.2174/1574893618666230316113621

Yayehrad, A. T., Wondie, G. B., and Marew, T. (2022). Different nanotechnology approaches for ciprofloxacin delivery against multidrug-resistant microbes. Infect. Drug Resist. 15, 413–426. doi:10.2147/IDR.S348643

Yu, Z., Huang, F., Zhao, X., Xiao, W., and Zhang, W. (2021). Predicting drug–disease associations through layer attention graph convolutional network. Briefings Bioinforma. 22, bbaa243. doi:10.1093/bib/bbaa243

Zhang, M., and Chen, Y. (2018). Link prediction based on graph neural networks. Adv. neural Inf. Process. Syst. 31. doi:10.48550/arXiv.1802.09691

Zheng, Y., Peng, H., Ghosh, S., Lan, C., and Li, J. (2019). Inverse similarity and reliable negative samples for drug side-effect prediction. BMC Bioinforma. 19, 554–104. doi:10.1186/s12859-018-2563-x

Zhu, L., Duan, G., Yan, C., and Wang, J. (2019a). “Prediction of microbe-drug associations based on katz measure,” in 2019 IEEE international conference on bioinformatics and biomedicine (BIBM) ( IEEE ), 183–187.

Zhu, L., Hong, Z., and Zheng, H. (2019b). “Predicting gene-disease associations via graph embedding and graph convolutional networks,” in 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) ( IEEE ), 382–389.

Keywords: graph neural network, ordered message-passing mechanism, microbe-drug association, multi-similarities, prediction model

Citation: Zhao J, Kuang L, Hu A, Zhang Q, Yang D and Wang C (2024) OGNNMDA: a computational model for microbe-drug association prediction based on ordered message-passing graph neural networks. Front. Genet. 15:1370013. doi: 10.3389/fgene.2024.1370013

Received: 13 January 2024; Accepted: 14 March 2024; Published: 16 April 2024.

Reviewed by:

Copyright © 2024 Zhao, Kuang, Hu, Zhang, Yang and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Linai Kuang, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.