Comparative analysis of fungal genomes reveals different plant cell wall degrading capacity in fungi

Most people picture fungi as the quiet recyclers of the world — the mushrooms on a log, the mold on last week's bread, breaking down what's already dead. But when Zhao and colleagues sat down with one hundred three fungal genomes and asked which fungi carry the largest enzymatic arsenals for dismantling plant cell walls, the answer wasn't the decomposers. It was the pathogens. Fungi actively invading living wheat and rice were outpacing the rotters, enzyme for enzyme. That reversal is the entry point into what this study actually found. Plant cell walls are built from cellulose, hemicellulose, and pectin — layered, interlocking polysaccharides that are among the most chemically resistant structures in biology. To get through them, whether to infect a living plant or to extract nutrition from a dead one, fungi deploy what are called carbohydrate-active enzymes, or CAZymes. Zhao and colleagues describe four main classes. Glycoside hydrolases, or GHs, cleave the bonds between sugar molecules — they're the workhorses for breaking down cellulose and hemicellulose. Polysaccharide lyases, or PLs, tackle acidic polysaccharides like pectin. Carbohydrate esterases, or CEs, strip off chemical groups that would otherwise block access to the chain. And glycosyltransferases, or GTs, run the process in reverse — building and remodeling carbohydrate structures. Many of these enzymes also carry carbohydrate-binding modules, non-catalytic domains that help the enzyme grip onto its insoluble target. Together, the CE, GH, and PL classes are what researchers mean when they talk about cell-wall-degrading enzymes. The logic of the study is simple and powerful: if you can catalog which CAZymes a fungus carries, you have a molecular fingerprint of how it makes a living. Zhao and colleagues screened predicted proteomes from one hundred three representative fungi across four major phyla — Ascomycota, Basidiomycota, Chytridiomycota, and Zygomycota — using a computational tool called Hmmscan with family-specific hidden Markov models, which are essentially statistical templates for each enzyme family, drawn from a database called dbCAN. The search identified one hundred eighty-seven distinct CAZyme families across the dataset. More than half the fungi examined carry over three hundred CAZyme modules. At the extremes, Fusarium oxysporum, a plant pathogen, carried seven hundred thirty CAZyme modules. The yeast Schizosaccharomyces cryophilus managed with one hundred twenty-five. That's a six-fold difference, and it's not random. Some families turned out to be near-universal. CE1, GH5, GH47, GT2, and GH16 were present in all one hundred three species or nearly all of them. These are the shared toolkit — the enzymes every fungus apparently needs regardless of lifestyle. But the distribution of rarer families tells a more interesting story. Twenty-eight families appeared only in Ascomycetes, and fifteen only in Basidiomycota. Ascomycetes carry significantly more GH2, GH72, and GH76, while Basidiomycetes have more GH5 and GH79. These are not subtle differences — they reflect deep evolutionary divergences in how different fungal lineages have tuned their digestive chemistry. And then there's this: cellulases from several GH families showed up in fungi with no known ability to degrade cellulose. The genome is carrying enzymes for a job we didn't know they had. Now here is where the study's central surprise lands. Zhao and colleagues organized their fungi by lifestyle — biotrophs, necrotrophs, hemibiotrophs, and saprophytes — and counted. Biotrophic fungi are the most restrained: they colonize living tissue without killing it, and they carry the fewest CAZymes, with notable absences in families like GH6, GH61, GH78, PL1, and PL3. That makes biological sense — you don't want to dissolve the host you're living inside. Necrotrophic fungi, which kill host cells and feed on the wreckage, carry more. Hemibiotrophic fungi, which start alive and switch to killing, sit close to necrotrophs. Plant pathogens overall have the highest CAZyme numbers in the dataset. Statistically, saprophytic fungi carry fewer enzymes in the CE, GH, and PL classes than plant pathogens do — each comparison significant with a p-value below 0.05, with reductions in specific families including CE5, GT1, PL1, and PL3 significant with a p-value below 0.01. Let that land for a second. The fungi we see visibly decomposing wood and leaf litter, the ones we think of as the planet's biochemical recyclers, are genomically outgunned by the fungi quietly invading living crops. Rhizopus oryzae, an opportunistic pathogen known as a strong degrader in laboratory cultures, shows a relatively sparse lignocellulose-degrading CAZyme profile in the genomic comparison. What a fungus looks like it's doing in the world doesn't always match what its genome says it's equipped to do. The enzyme toolkit also records something else: what kind of plant a fungus attacks. Zhao and colleagues found that fungi infecting dicots — broad-leaved plants like tomato and soybean — carry significantly more pectinases than fungi infecting monocots like wheat, rice, and corn. Specifically, pathogens of dicots have more enzymes from GH families GH28, GH88, and GH105, with each comparison significant with a p-value below 0.01. The reason is structural: pectin is a major component of dicot cell walls, but it's much less prominent in the walls of grasses. So the pectinase arsenal reflects the target. The PL1 family — polysaccharide lyase family 1, whose members break down pectin through a specific chemical mechanism called lyase activity — shows this pattern with particular clarity. Zhao and colleagues performed phylogenetic analysis of PL1 sequences across their dataset and found a complicated evolutionary history: the ancestral fungus likely carried numerous PL1 paralogs, different lineages then lost different subsets, and multiple independent expansions occurred in plant pathogens. Basidiomycetes appear to have shed most of their ancestral PL1 genes, while Sordariomycetes, a class within Ascomycota, retained more. Fusarium species show clusters of closely related PL1 genes consistent with recent duplication — the genome copying a useful tool and diversifying it. Twenty-one of the one hundred three fungi surveyed carry no polysaccharide lyase at all, and nearly a third of the Ascomycetes examined lack all nine pectinase-associated CAZyme families the authors surveyed. The pectinase toolkit is both phylogenetically ancient and highly variable — shaped by which hosts different lineages ended up attacking. Comparative genomics gives you the blueprint. But does the fungus actually build what the blueprint describes? To answer that, Zhao and colleagues turned to Fusarium graminearum, the wheat scab fungus and one of the most economically damaging cereal pathogens in the world, and examined which CAZyme genes were actually expressed during infection. They used public microarray data from three experiments — spike infection of barley, wheat head infection, and conidium germination — processed with a statistical normalization method called robust multi-array RMA treatment, and clustered the resulting expression profiles by k-means analysis. The answer was clear: most CAZyme genes related to plant cell wall degradation were up-regulated during plant infection. The genome encodes the arsenal, and infection pulls the trigger. The expression patterns during barley spike infection and wheat head infection were similar to each other and distinct from germination — the fungus activates a different gene program when it hits living plant tissue than when it’s simply growing from a spore. Some CAZyme genes were actually down-regulated during infection but up-regulated during germination, suggesting they serve functions specific to early spore development rather than host attack. A minority cluster — genes encoding certain CEs and enzymes that decompose the fungal cell wall itself — were down-regulated during infection, hinting at a deliberate restraint of self-digestion while the fungus is invading. The genome-wide picture and the expression data tell the same story from two different angles. Zhao and colleagues show that CAZyme repertoires are simultaneously evolutionary records and ecological fingerprints. The PL1 phylogeny alone encodes a history of gene duplication, differential loss across lineages, and repeated specialization in pathogens. The finding that some fungi carry cellulase families without any known role in cellulose degradation points to functions not yet understood — or to enzyme activities that haven't been looked for yet, which is an open door for industrial enzyme discovery. And the central reversal — pathogens encoding more cell-wall-degrading machinery than many saprophytes — challenges a simple intuition about who needs the biggest toolkit. It turns out that quietly invading a living plant, triggering its defenses, and extracting nutrients before the host can respond may require a more elaborate molecular toolkit than decomposing wood in peace. You can only see that when you look at the whole kingdom at once. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.