Ustilago maydis can utilize corn stover
The ability of the fungus U. maydis to grow on increasing concentrations of milled corn stalks was investigated (Fig. 1A). The performance of U. maydis was assessed online using a BioLector® system, where scattered light was quasi-continuously monitored to measure the media's turbidity. This measurement served as a proxy for U. maydis growth as scattered light correlates with cell density (38). In the control with fungal inoculum but without corn stover, only a slight uptick in the scattered light was observed, reaching a maximum of 13.9 ± 0.1 a.u. after 22 h. This increase is likely attributable to residual nutrients still present in the inoculum media. In contrast, when the media was supplemented with corn stover, there was a notable increase in the scattered light, indicative of fungal biomass production. This surge began around 6 h after inoculation in all corn stover concentrations tested. While the maximum scattered light values were observed after 23 h using 3 g/L or 10 g/L corn stover, increasing the corn stover material to 20 g/L resulted in a maximum signal after 14 h (Fig. 1A). Microscopic inspection of the cultures after 16 h confirmed the presence of dividing fungal cells in the yeast form, mirroring cells grown in glucose (Glc) as sole carbon source (Additional File 1: Figure S1).
U. maydis growth performance on 20 g/L corn stover was further characterized by measuring metabolic activity parameters such as pH and oxygen transfer rate (OTR) in parallel to scattered light (Fig. 1B and C) combining BioLector® with µRAMOS technologies (32). The employed U. maydis strain expressed the green fluorescence protein (Gfp) in the cytoplasm. Hence, as an additional parameter to characterize fungal growth, we used the fluorescence read-out for online monitoring. In concomitance with the initiation of the exponential growth phase after 6 h, the pH of the media decreased from its initial pH of 5.8 to a minimum of 5.4 after 12 h. This initial drop in pH may be indicative of increased metabolic activity involving for example the production of organic acids, or the release of acetic acid resulting from the breakdown of plant cell walls. Aligning with the onset of the stationary growth phase, the pH of the media began to rise, and reached a 6.3 value by the end of the cultivation period (Fig. 1B). A similar shift in pH in U. maydis fermentations have been interpreted as a fungal response to nutrient limitations characterized by a transition to less acidic metabolic pathways (39, 40).
The OTR rapidly increased during the exponential growth phase consistent with an increase in metabolic activity of the fungus, as U. maydis metabolizes substrates and actively replicates, consuming oxygen. The OTR reached its maximum after 14 h, coinciding with the maximum fungal cell density determined by Gfp fluorescence and scattered light measurements. After a short plateau, the OTR decreased rapidly suggesting a decline in the metabolic activity (32, 33). OTR values reached a basal level after 20 h, which remained constant until the end of cultivation (Fig. 1C). The Gfp fluorescence signal increased correlating with the increment in scattered light signal, further confirming fungal proliferation (Fig. 1B). Collectively, these results show that U. maydis can utilize corn stover as a carbon source. Moreover, the developed screening method allows a detailed characterization of the fungal performance on corn stover in a microtiter scale by simultaneously recording diverse growth and metabolic activity parameters online.
U. maydis exhibits differential utilization of the various carbohydrate substrates present in corn stover
To identify the corn stover components utilized by U. maydis, we conducted a comprehensive compositional analysis of the residue remaining after incubation in the presence and absence of U. maydis. First, the suspensions were collected from the BioLector® plate at the end of the cultivation and the liquor was separated from the solid and dried before performing the analyses. The unfermented liquor fraction contained large amounts of soluble sugars, mainly glucose, sucrose and fructose (Table 1).
Table 1
Carbohydrate composition of the liquor fraction [% of dry weight].
Condition
|
Glucose
|
Sucrose
|
Fructose
|
Total
|
B73 - U. maydis
|
9.0 ± 0.9
|
2.4 ± 0.9
|
9.6 ± 0.6
|
21.0 ± 0.8
|
B73 + U. maydis
|
0.3 ± 0.04
|
0.1 ± 0.03
|
0.04 ± 0.03
|
0.4 ± 0.05
|
bm3 - U. maydis
|
14.6 ± 0.5
|
3.0 ± 0.8
|
13.7 ± 0.8
|
31.2 ± 1.4
|
bm3 + U. maydis
|
0.3 ± 0.1
|
0.03 ± 0.01
|
0.2 ± 0.1
|
0.5 ± 0.1
|
Soluble sugar quantification [% of dry weight] of unfermented and fermented B73 and bm3 material. Data are shown as mean ± SD of n = 4 (B73) and n = 6 (bm3) biological replicates. Bold values indicate statistically significant differences between -/+ U. maydis conditions determined by a two-tailed students t-Test at p-value < 0.05.
However, after fermentation, only traces of these carbohydrates were detected, indicating that U. maydis is able to utilize them as a carbon source. As corn stover was routinely autoclaved to avoid microbial contaminations during fermentation, we explored if this process alters the composition and availability of soluble sugars, but no significant differences were detected in non-autoclaved corn stover (Additional file 1: Table S1).
The compositional analysis of the solid residue needs to consider the presence of fungal biomass as a result of the extensive generation of U. maydis biomass during fermentation. Quantification of glucosamine (GlcN), a monosaccharide present in U. maydis chitinaceous cell walls but absent in plant tissues, was used to estimate the amount of U. maydis derived material in the residue. After subtraction of the estimated fungal biomass, the relative abundance of plant-derived material within the solid residue was determined. This fraction was analyzed for the presence of the main insoluble corn stover constituents i.e., starch, lignin, crystalline cellulose and/or hemicellulosic polysaccharides (Table 2).
Table 2
Relative biomass composition [% of AIR] of the pre- and postfermentation residue.
Condition
|
Arabinose
|
Galactose
|
Glucose
|
Xylose
|
HC
|
CC
|
Lignin
|
Acetate
|
Starch
|
U. maydis
|
Total
|
B73 - U. maydis
|
2.5 ± 0.1
|
0.7 ± 0.0
|
2.5 ± 0.4
|
23.1 ± 0.7
|
28.9 ± 0.7
|
35.1 ± 0.6
|
15.8 ± 0.9
|
5.0 ± 0.3
|
0.5 ± 0.2
|
n.d.
|
85.3 ± 1.7
|
B73 + U. maydis
|
2.5 ± 0.1
|
1.0 ± 0.1
|
2.8 ± 0.3
|
18.3 ± 0.7
|
24.6 ± 0.6
|
32.6 ± 1.2
|
16.4 ± 0.2
|
3.9 ± 0.2
|
0.8 ± 0.0
|
11.2 ± 1.4
|
89.5 ± 1.2
|
bm3 - U. maydis
|
2.4 ± 0.2
|
0.6 ± 0.0
|
2.1 ± 0.3
|
23.6 ± 0.8
|
28.7 ± 0.8
|
35.4 ± 1.5
|
15.4 ± 1.2
|
5.0 ± 0.2
|
0.4 ± 0.2
|
n.d.
|
84.9 ± 2.9
|
bm3 + U. maydis
|
2.2 ± 0.1
|
1.0 ± 0.1
|
3.0 ± 0.2
|
18.0 ± 1.0
|
24.2 ± 1.1
|
33.3 ± 0.8
|
14.2 ± 0.8
|
3.5 ± 0.2
|
0.7 ± 0.0
|
15.4 ± 1.5
|
94.6 ± 1.7
|
B73 - Celluclast®
|
1.7 ± 0.2
|
0.9 ± 0.1
|
2.5 ± 0.6
|
18.2 ± 1.6
|
23.2 ± 2.2
|
32.2 ± 2.7
|
15.3 ± 0.7
|
4.4 ± 0.3
|
0.7 ± 0.0
|
12.6 ± 1.5
|
88.4 ± 1.8
|
B73 + Celluclast®
|
1.2 ± 0.1
|
0.6 ± 0.1
|
0.3 ± 0.3
|
14.7 ± 1.0
|
16.9 ± 1.2
|
27.5 ± 1.8
|
16.6 ± 0.9
|
4.5 ± 0.7
|
0.8 ± 0.1
|
24.1 ± 1.9
|
90.4 ± 2.1
|
bm3 - Celluclast®
|
1.2 ± 0.1
|
0.8 ± 0.1
|
2.2 ± 0.4
|
15.9 ± 0.8
|
20.2 ± 1.0
|
31.7 ± 1.9
|
12.8 ± 0.8
|
4.4 ± 0.5
|
0.7 ± 0.1
|
18.2 ± 2.1
|
87.9 ± 2.2
|
bm3 + Celluclast®
|
0.9 ± 0.1
|
0.5 ± 0.1
|
n.d.
|
12.9 ± 1.1
|
13.6 ± 1.7
|
25.1 ± 2.3
|
15.3 ± 0.5
|
4.6 ± 0.4
|
0.9 ± 0.1
|
37.4 ± 5.7
|
96.9 ± 5.1
|
n.d. = not detected; HC = sum of hemicellulosic monosaccharides; CC = crystalline cellulose; Total = Sum of plant and fungal components
|
Comparison before and after digestion with Ustilago maydis and without/with addition of Celluclast®. Data are shown as the mean ± SD of 4 biological replicates for B73 and 6 biological replicates for bm3. Bold values indicate statistically significant differences between the material determined by a two-tailed students t-Test at p-value < 0.05.
The results did not reveal any significant decrease in the proportion of lignin between fermented and non-fermented residues, indicating that U. maydis is not able to degrade this aromatic polymer. Likewise, it appears that U. maydis does not utilize starch, although it should be noted that only trace amounts are found in senescent corn stem tissue. Only a modest decrease in the relative abundance of crystalline cellulose (-7.1%) and total hemicellulose (-14.9%) was detected in fermented samples. Further determination of the hemicellulosic monosaccharide composition revealed that the reduction in total hemicellulose was likely caused by a decrease in xylose (-20.8%), indicating partial xylan degradation. Additionally, the strongest decrease was observed in the proportion of wall-bound acetate (-22%), which is mostly found as a substituent on the xylan backbone in corn stover (Table 2). The utilized method can only determine relative abundances, as the quantitative harvest of the solid fraction from the BioLector® plate is virtually impossible. Hence, to validate our relative estimations, parallel cultures in glass shake flasks, compatible with mass balance calculations, were used. The analysis of the residue confirmed that the largest remaining fraction of the corn stover consisted of solid material. In fact, the percentage of solid residue increased from 67.8 ± 2.7 wt% in samples without U. maydis to 80.1 ± 1.2 wt%.in fermented samples. This considerable difference can be explained by the conversion of virtually all soluble sugars present in the liquor fraction into insoluble fungal materials. Further dissection of the solid fraction showed that the predominant portion in both conditions corresponded to alcohol insoluble residue (AIR), which mainly encompasses polysaccharides and other large polymers contained in plant and microbial cell walls. 11.2 ± 0.1% of the AIR corresponded to U. maydis biomass, which matches the fungal biomass estimated in the BioLector® samples (11.2 ± 1.4%) (Table 2, Additional file 1: Figure S2). This parallel analysis further confirmed the main results observed in the microtiter plates. The amount of crystalline cellulose in fermented samples decreased by 11.1%. Although the reduction in total hemicellulose was not statistically significant, xylose and acetate content decreased by 15.8% and 22.3%, respectively, further suggesting degradation of O-acetylated xylan. Interestingly, neither of the two applied methods could account for the small but significant increase in galactose. One interpretation is that this increase is derived from the fungal cell wall, suggesting potential changes in the fungal wall composition if growing on corn stover compared to our reference glucose substrate.
Overall, these results indicate that U. maydis is mainly feeding on easily accessible soluble carbohydrates present in the plant biomass. U. maydis is able to directly assimilate these substrates without the need of enzymatic transformation (41), explaining the quick growth and metabolic activity rise during the initial stages of fermentation. Utilization of larger, insoluble lignocellulosic polymers seems to be more limited. Lignin remains unaffected and only a small fraction of crystalline cellulose and O-acetylated GAX appears to be susceptible to degradation and further utilization by the fungus under the used conditions. This restricted availability of carbohydrate substrates from the lignocellulosic composite would explain the growth halt and the drop in the fungal metabolic activity in the late fermentation stages observed by online monitoring.
Growth performance of U. maydis is enhanced on corn stover derived from a maize lignin mutant
We proceeded to assess whether alterations in the lignocellulose structural attributes of the corn stover feedstock could enhance the utilization of this residue by U. maydis. Our initial investigation focused on the impact of a plant wall material with altered composition. Therefore, we examined U. maydis performance growing on corn stover derived from a lignin-deficient maize mutant, since modification of the lignin composition has been shown to impact the enzymatic degradation of corn stover (42). The brown midrib 3 (bm3) maize mutant exhibits an increased ratio of guaiacyl to syringyl moieties in its lignin, resulting in enhanced enzymatic degradation (28, 43). When bm3 corn stover was used as a substrate for U. maydis growth in our screening platform, a significant increase in the maximum scattered light signal was observed compared to the reference corn stover previously used i.e., B73 (Fig. 1B and C). Although there were no discernible differences in initial fungal growth between B73 and bm3 material, the exponential phase was prolonged by 2 h on bm3 material and the scattered light reached a maximum value 17.4% higher than in B73. These findings were further corroborated by an increase in maximum Gfp fluorescence when bm3 corn stover was used. Similarly, the maximum OTR was higher for bm3 corn stover material than for B73. During the first 12 h, the OTR curves are indistinguishable. After that period, while B73 material reached a plateau, the OTR for bm3 continued to rise, peaking 14 h after inoculation. Subsequently, the OTR decreased rapidly, ultimately reaching a final plateau after 19 h, similar to the pattern observed for B73. Notably, the pH profile also differed between the two different substrates. During the initial incubation period, the pH levels decreased similarly in both conditions. However, the pronounced pH increase observed for B73 material during the exponential growth phase of U. maydis was delayed and of a lesser magnitude in the case of growth on bm3 material. Both OTR and pH measurements suggest a prolonged fungal metabolic activity when growing on bm3 material. This indicates that U. maydis performance is enhanced in corn stover obtained from the lignin-deficient bm3 mutant compared to B73.
The composition of the bm3 residue after incubation in the presence or absence of U. maydis was analyzed, following a similar approach as the one used for wildtype corn stover (B73). The undigested bm3 liquor contained 10% more soluble carbohydrates than B73. Consistent with the observations for B73 material, only traces of glucose, sucrose and fructose were detected in the samples incubated with U. maydis, further confirming that the fungus efficiently converts these soluble carbohydrates (Table 1, Additional file 1: Table S1).
After incubation with U. maydis, the bm3 solid fraction contained 15.3 ± 1.5% of fungal material, representing a 37.8% increase compared to cultivations on B73 (Table 2) confirming that U. maydis can grow more efficiently on bm3 corn stover. Compositional analyses of the residue revealed a reduction of 6% in the crystalline cellulose relative content due to fermentation, slightly reduced compared to the 7.1% shown by B73. Notably, utilization of GAX in bm3 corn stover seems to be improved as higher reductions in total hemicellulose, xylose, and acetate content compared to B73 were observed (15.7%, 23.7%, and 30%, compared to 14.9%, 20.8%, and 22%, respectively).
Both the online measurements and the analytical quantification of fungal biomass present in the residue after fermentation substantiate the superior growth of U. maydis in bm3 material compared to B73 (Fig. 1B and C; Table 2). A higher abundance of soluble sugars in bm3 corn stover appears to account for most of this enhanced fungal performance. However, an improved utilization of O-acetylated xylan may also contribute to this phenomenon suggesting that the changes in lignocellulosic properties of this maize variety might be beneficial for fungal performance.
Performance of engineered lignocellulolytic U. maydis strains growing on corn stover
Expression of most genes encoding lignocellulose hydrolytic enzymes in U. maydis remains low during non-pathogenic yeast-like growth, becoming active primarily during the plant infection phase (22, 24). A collection of engineered U. maydis strains secreting intrinsic enzymes with diverse hydrolytic activities during non-pathogenic growth was tested in our conditions (24). Among them, two showed enhanced secreted cellulase activity. The Pomabgl1 strain displayed exo-β1-4-glucanase activity, enabling it to grow on cellobiose as a sole source of carbon and to release small amounts of glucose from microcrystalline cellulose (24). Supernatants of the second strain, Pomaegl1, harbor endo-β1-4-glucanase activity on carboxymethylcellulose and regenerated amorphous cellulose (24, 44). The third selected strain, Pomaxyn11A, exhibits enhanced xylanase activity in culture supernatants, generating smaller oligomers from birch-wood xylan compared to the control (24, 45).
We assessed the performance of the three GfpPomabgl1, Pomaegl1 and Pomaxyn11A U. maydis strains in our BioLector® screening platform comparing them to the control MB215Gfp strain (Fig. 2A). All four strains showed similar overall performance when growing on corn stover. However, growth acceleration was observed in each of the engineered strains, characterized by a shorter lag phase and faster reach of the maximum scattered light value compared to the control strain (Fig. 2A). The control strain entered the exponential growth phase around 6.5 h and reached maximum growth after 15.5 h. Meanwhile, Pomaxyn11A, Pomaegl1 and GfpPomabgl1 strains initiated the exponential phase around 5 h and reached their respective maxima after 14.5 h, 13.5 h and 13 h. Notably, the GfpPomabgl1 strain not only exhibited the fastest growth, but also achieved higher cell densities according to the scattered light measurements, indicating the best overall growth performance. While the use of these strains did not show a clear improvement in lignocellulose utilization despite enhanced lytic activities, the results show the potential of the established BioLector®-based screening platform to assist in the evaluation of engineered U. maydis strains. Future work will need to enlarge the collection of engineered strains and to address the synergic activity of diverse lignocellulolytic enzymes as a likely requirement for the efficient degradation of the corn stover lignocellulosic composite.
Assisted enzymatic lignocellulose degradation improves U. maydis performance
After observing the limited utilization of the lignocellulosic fraction in our system, we tested whether the addition of an enzyme cocktail with lignocellulolytic activity improves U. maydis performance. Hence, we evaluated the growth of U. maydis on corn stover (variety B73), comparing cultures with and without the addition of Celluclast® (Fig. 2B and C). Scattered light and Gfp fluorescence were monitored for up to 36 h allowing the detection of the plateau under these conditions. The initial exponential growth phase was identical independent of the addition of Celluclast®, reaching the stationary phase at 137 ± 9 a.u. after 15 h. However, the addition of Celluclast® resulted in a notable shift in fungal growth dynamics, characterized by a second, less rapid growth phase that led to a substantial increase in the maximum scattered light, reaching 242 ± 12 a.u. after 22 h. This corresponds to a 176,6% increase in the fungal cell density directly attributable to the addition of Celluclast®. Measurement of Gfp fluorescence showed comparable results, and Celluclast®-supplemented samples exhibited a secondary boost in signal intensity mirroring the scattered light profile (Fig. 2C).
The analysis of the solid residues confirmed the online data, as the amount of fungal material present in the solid residue from samples with Celluclast® reached 24.1%, almost doubling the amount observed in the non-supplemented control (Table 2). Celluclast® addition significantly decreased the relative abundances of all cell wall carbohydrate sources present in corn stover. The proportions of crystalline cellulose and total hemicellulose decreased by 14.6 and 27.1%, respectively. All hemicellulosic monosaccharides were reduced, particularly glucose where only trace amounts were found in samples supplemented with Celluclast®.
Using bm3 corn stover as substrate combined with Celluclast® addition resulted in synergistic effects (Fig. 2B and C). A first peak in maximum fungal growth was reached after 15.5 h, exhibiting higher scattered light and Gfp fluorescence compared to B73 as expected from better fungal performance. Celluclast® addition to bm3 corn stover also resulted in a second, less rapid growth phase, but a higher maximum value was obtained and at a later time (27 h) compared to B73. Similar distinctions were identified when analyzing the Gfp fluorescence (Fig. 2C). Confirming the online measurements, the quantity of fungal biomass detected in bm3 solid residue supplemented with Celluclast® reached 37.4% representing a 105.5% increase compared to non-supplemented bm3 (Table 2). This increase in fungal biomass production resulting from Celluclast® addition was higher in bm3 compared to B73. Similarly, the solid residue composition showed bigger reductions in the proportion of crystalline cellulose and hemicellulose components (Table 2). This suggests that the altered wall structure of bm3 mediates enhanced substrate accessibility and/or hydrolytic activity of the enzymes present in the Celluclast® cocktail.
Unexpectedly, we also observed increments in the relative abundance of lignin and starch correlating with an increasing abundance of fungal biomass. One possibility is that these values reflect the relative decrease of the other components of the solid residue. Alternatively, we cannot discard the possibility that certain fungal components interfere with our determinations, primarily when large amounts of fungal biomass are present in the solid residue as the case of bm3 samples supplemented with Celluclast®. Fungal cell membranes contain sterols which may interfere with our spectrophotometric quantification of lignin (46). Similarly, U. maydis might accumulate glycogen during corn stover fermentation which is indiscernible from starch in our assays.
Together, these results indicate that Celluclast® treatment during cultivation results in additional substrates derived from the lignocellulose fraction in corn stover for U. maydis to continue growing upon consumption of the soluble sugars. Combining the lignocellulolytic activity supplementation with the use of bm3 corn stover results in a synergistic effect, allowing a 3-fold build-up of fungal biomass compared to the initial conditions set for B73.