High G lignin xylan hypo acetylated lines does not affect plant growth or biotic stress responses
The fah1-2 point mutation ferulate 5-hydrxylase (F5H) gene leads to over-accumulation of guaicyl (G) lignin (Chapple et al., 1992) whereas the overexpression of acetyl xylan esterase from Aspergillus niger (35S:AnAXE1) depletes xylan acetylation level in Arabidopsis (Pawar et al., 2016). To understand the effect of simultaneous G-lignin enrichment and acetyl-hypo-accumulation on plant cell wall properties and saccharification, we crossed the fah1-2 mutant and the single-insert containing line D with 35S:AnAXE1 construct. The parent plants and their homozygous progeny were indistinguishable from the wild-type plants (Fig. 1a). To further confirm the presence of fah1-2 mutation in the F5H gene in the obtained hybrid plants, we amplified the F5H fragment with gene-specific PCWL-58 and PCWL-59 primers (Additional file: Table S1). The amplicon obtained by amplifying F5H gene was digested with MseI restriction enzyme and the banding pattern after restriction showed the expected difference between wild type and AnAXE1 showing 400 bp and 300 bp bands and fah1-2 and fah1-2/35S:AnAXE1 plants showing 300 bp, 275 bp and 125 bp bands (Additional file: Fig. S1a and S1b). Another typical chemotype of fah1-2 mutant is the absence of sinapoyl malate (Chapple 1992). Using a high-performance liquid chromatography coupled with diode array detector (HPLC-DAD) analysis of soluble metabolites from leaves revealed undetectable levels of sinapoyl malate in fah1-2 and fah1-2/35S: AnAXE1 whereas the wild-type and 35S:AnAXE1 line D plants had detectable levels of this compound (Additional file: Fig. S1c). To further confirm the expected change in lignin composition in the hybrid, we exposed four weeks old rosette leaves to ultraviolet (UV) light and we found that fah1-2 and fah1-2/35S:AnAXE1 plants showed red colour indicative of lack of sinapoyl malate level whereas the wild-type and 35S:AnAXE1 line D plants showed blue color (Fig. 1b). To demonstrate the lack of S-lignin in hybrid plants, we stained transverse sections of their inflorescence stem with Mȧule stain. We found that the interfascicular fibres stained red (positive for S-lignin) in the wild-type and 35S:AnAXE1 plants and brown (absence of S-lignin) in fah1-2 and fah1-2/35S:AnAXE1 (Pradhan Mitra & Loqué, 2014) (Fig. 1c).
Similarly, the presence of AnAXE1 in the hybrid was confirmed by amplifying expected size band using AnAXE1 - specific primers PCWL-33 and PCWL-34 (Additional file: Table S1) whereas these amplicons were absent in for fah1-2 and wild-type (Additional file: Fig. S1d). The further analysis was performed on F2 and F3 generation plants which were positive for AnAXE1 amplicon and also confirmed its expression by qPCR (Additional file: Fig. S1e and S1f). AnAXE1 overexpressed lines showed an increase in esterase activity (Pawar et al., 2016). To evaluate this, we performed esterase activity assays on the leaves of the parent lines and the selected homozygotic hybrid line using p-nitrophenyl acetate as a substrate. Esterase activity in the wall-bound protein fraction was increased in the parent 35S:AnAXE1 line and in fah1-2/35S:AnAXE1 homozygotic plants compared to the wild type and fah1-2 mutants. (Fig. 1d). To further confirm a reduction in polysaccharide acetylation in the hybrid line because of the increase in esterase activity, we analysed cell wall acetyl content in alcohol-extracted wall residues (AIR) after saponification. As expected, we observed decreased acetyl content in 35S:AnAXE1 and fah1-2/35S:AnAXE1 plants as compared to wild type and fah1-2 mutant (Fig. 1e). These experiments confirmed expected genotypes and phenotypes in parents and fah1-2/35S:AnAXE1. We hereafter referred to these crossed plants as HyperG Hypo Acetylated line (HrGHypAc).
Morphological studies revealed that HrGHypAc plants were indistinguishable from wild-type plants. The xylem vessel cells were not collapsed (Fig. 1c) as in irx cell wall mutants with compromised growth (Turner and Somerville 1997). Often, changes in cell wall composition affect plant responses to biotic stress (Miedes et al. 2014). More specifically, cell wall-associated components are known to function as damage-associated molecular patterns (DAMPs) to induce or aggravate immunity. To evaluate innate immune responses of the HrGHypAc plants, we performed a quantitative pathogenesis assay. Individual parents were also tested similarly, with wild-type plants as controls. Fully expanded rosette leaves from the investigated plants were first infiltrated with Pseudomonas syringae DC 3000 pv. tomato strain (PstDC3000) suspension at a density of 5 x 104 cfu/mL. The accumulation and growth of PstDC3000 in the infiltrated leaves were evaluated at day 0 and day 3, post-infiltration. Simultaneously, we also determined the expression changes in defence-related marker genes namely WRKY53, PR1, PAD3, LORE, and SERK in the infiltrated leaves at 0 and 3 dpi (days post-infection). Neither fah1-2, nor HrGHypAc plants showed any differences in bacterial accumulation at 3 dpi in comparison to the wild-type plants (Additional file: Fig. S1g). Interestingly, fah1-2 plants showed higher upregulation of WRKY53, PR1 and PAD3 genes at 3 dpi compared to the wild-type. Contrastingly, WRKY53 was down-regulated in 35S:AnAXE1 at day 3 dpi. In the HrGHypAc plants, the tested immune markers were wild-type comparable indicating that genetically FAH1 and over-expressed AnAXE1 may function antagonistically on WRKY53 expression (Additional file: Fig. S1h). However, since overall pathogenesis outcomes were unaffected in the tested plants, it is implied that the contribution of these genes towards defence is not significant. Taken together, these data suggest that the HrGHypAc or the parental plants are not compromised in growth or defence.
The total sugar content and Updegraff cellulose content is higher in HrGHypAc plants
To study the effect of simultaneous modification of lignin and xylan structure, we performed detailed cell wall characterization of alcohol insoluble residue (AIR) isolated from seven weeks old inflorescence stems of above plants. Using the phenol sulphuric method, we analysed the total cell wall polysaccharide composition which was higher (+ 20%) in HrGHypAC as compared to wild-type or parent plants. (Fig. 2a). Further, the content of crystalline cellulose by the Updegraff method was increased by 13% in HrGHypAc line as compared to either wild type or parent plants (Fig. 2b). Non-crystalline glucose originated from amorphous cellulose and hemicellulosic polysaccharide and other hemicellulosic sugars such as xylose, arabinose, rhamnose, fucose, galactose and mannose were similar in analysed plants (Fig. 2c). These findings suggests that the increase in total sugar content could be because of higher cellulose content. Another important component in the stem cell wall is lignin which was unchanged between investigated plants (Fig. 2d). Overall, these data suggest that HrGHypAc line contained increased total sugar, and crystalline cellulose content and with no major changes in lignin content or matrix sugar composition as compared to parent and wild type cell wall.
HrGHypAc plants show improvement in xylan and cellulose digestibility
To understand the effect of simultaneous modification in lignin monomer and xylan acetylation on biomass recalcitrance, we performed a polysaccharide digestibility assay with and without pre-treatment using AIR from the stems of parent genotypes, progeny and wild-type. After digesting the untreated AIR of wild type, fah1-2, 35S:AnAXE1 and HrGHypAc with commercial enzyme blend containing cellulases and hemicellulases and by measuring the glucose release by GOD-POD assay (Acker et al. 2016), we detected moderately higher (+ 4.38%) in the HrGHypAc compared to wild-type whereas the parent was comparable to wild type (Fig. 3a). Furthermore, after hot water pre-treatment known to increase accessibility of polysaccharide degrading enzymes, the saccharification efficiency was increased in both the parents and HrGHypAc by 19.85%, 17.41%, and 19.25% respectively as compared to the wild type (Fig. 3b). The glucose release was comparable in HrGHypAc and parent plants. Deacetylated xylan is more accessible to xylanases (Pawar et al. 2016). Therefore, AIR was digested with glycosyl hydrolase family 11 (GH11) without treatment, a higher level of xylose release was observed in parent and HrGHypAc plants as compared to the wild type. 35SAnAXE1 line exhibited a higher amount of xylose release as compared to the fah1-2 and HrGHypAc plants (Fig. 3c). This was further confirmed by running digested product on TLC as more xylose was detected in parents and cross as compared to wild type (Additional file: Fig. S2a) (Additional file: Fig. S2b). Moreover, xylobiose or conjugated xylobiose were also in higher amounts and longer chain xylotetraose or conjugated xylotetraose were detected in a lower amount in the cross than wild type. Since glucuronoyl esterase (GE) breaks the γ-ester linkage between xylan and lignin (Hüttner et al. 2017), AIR was treated with GE followed by endoxylanases and xylose released by these treatments was increased by 45%, 26% and 43% in fah1-2, 35S:AnAXE1 and HrGHypAc, respectively compared to wild-type (Fig. 3d). Also, similar amount of xylose release was detected in fah1-2 and HrGHypAc line but was higher in 35S:AnAXE1 parent as compared to above plants.
We also digested AIR with a GH67 alpha glucuronidase (which hydrolyses glucuronic acid side-group on xylan chain) followed by digestion with a GH11 endoxylanase and found the xylose release increased by 34%, 53% and 49.5% in fah1-2, 35S:AnAXE1 and HrGHypAc, respectively, compared to wild-type (Fig. 3e). We also measured glucose release in GE and alpha glucuronidase digested pellet and it was only increased in fah1-2 mutant as compared to wild type plants (Additional file: Fig. S2c).
In summary, whereas only the HrGHypAc line showed an increase in glucose yield in saccharification without pre-treatment, all three lines, both parents and the HrGHypAc line, showed improved xylan digestibility as compared to wild type by a GH11 endoxylanase after pre-treatment with hot water and either GE or alpha glucuronidase.
Altered extractability and digestibility of xylan, cellulose and lignin in cell wall-modified lines
Lignin and xylan modification can alter its interaction with pectin and consequently can influence polysaccharide digestibility (Liu et al. 2022; Broxterman and Schols 2018; Ralet et. al., 2016). To study further the effect of the genotypes (fah1-2, 35S:AnAXE1 and HrGHypAc) on the accessibility of xylan to a GH11 endoxylanase, the AIR was treated with ammonium formate to solubilise pectin and the remaining pellet was digested with pectate lyase to completely remove the residual homogalacturonan. The galacturonic acid was analysed in both pectin-rich ammonium formate extract and in pectate lyase fraction. HrGHypAc plants showed a decrease in galacturonic acid level by 14% as compared to the wild type in the ammonium formate extract but it was similar in wild type, parent lines and their progeny in the pectate lyase fraction (Fig. 4a). The de-pectinated pellet obtained in above assay was digested with a GH11 endoxylanase and the total xylose released by the enzyme was measured in the supernatant. Significantly higher xylose content than wild type (5%, 6% and 2.5% in fah1-2, 35S:AnAXE1 and the HrGHypAc respectively) were detected in this assay (Fig. 4b). Subsequently, we measured cellulose content in de-pectinized xylanase digested samples, found it significantly higher by 49%, 41% and 55.5% in fah1-2, 35S:AnAXE1 and the HrGHypAc, respectively, as compared to wild-type (Fig. 4c). To understand the effect of the genotypes on lignin content in de-pectinized samples and in de-pectinized xylanase-digested samples, we measured acetyl bromide soluble lignin (ABSL) in these samples. In de-pectinized samples, ABSL was significantly increased in the HrGHypAc line by 7% as compared to wild-type and parent lines (Additional file: Fig. S3a). In the de-pectinized xylanase-treated samples, fah1-2 and 35S:AnAXE1 lines showed 13% and 9.5% decreased lignin content, respectively, and the HrGHypAc line showed similar lignin content as wild-type plants (Additional file: Fig. S3b). We also measured ABSL content after saccharification and found that it was decreased by 5.7%, 10.4% and 15% respectively in fah1-2, 35S:AnAXE1 and HrGHypAc lines (Additional file: Fig. S3c). To further understand the effect of genotypes on xylan extractability, we sequentially extracted cell walls with ammonium formate and KOH and analysed the xylan content in 1M and 4 M KOH fractions using LM10 antibody. The xylan signal was significantly increased in fah1-2, AnAXE1, and the HrGHypAc plants, respectively, as compared to wild-type in combined 1M and 4 M KOH fractions (Additional file: Fig. S3d). The extractability was higher in 35S:AnAXE1 among all tested genotypes. All these cell wall digestibility and extractability experiments revealed that both HrGHypAc and parental plants show altered xylan digestibility and lignin extractability after sequential extraction.
HrGHypAc plants show increased expression of CESA, THESUS1 and WAK genes
To correlate the changes in plant cell wall properties in parental and HrGHypAc plants, we evaluated the expression of the key cell wall-associated genes. Cellulose Synthases (CESAs) in the rosette complexes present in the plasma membrane are responsible for the cellulose polymer biosynthesis (Li et al. 2014). When tested for relative expression levels of various CESAs transcripts, secondary cell wall-specific CESA6 and CESA8 (+ 2.6 and + 2.5-fold) mRNAs were significantly upregulated whereas CESA1 was lower (-19.64%) in HrGHypAc plants than wild type (Fig. 5). No significant change in fold change expression was found for CESA2 among tested plants. The expression of secondary cell wall biosynthesis transcriptional activator MYB83 (Ko et al. 2014) and the lignin biosynthesis pathway transcriptional activator MYB63 (Zhou et al. 2009) were also higher in the HrGHypAc line when compared to wild type (Fig. 5). Expression levels of xylan modifying genes xylan modifying enzymeβ-xylosidase 1 (BXL1) (Goujon et al. 2003) and xyloglucan-modifying enzyme XTH4 (Xyloglucan endotransglucosylase/hydrolase) (Kushwah et al., 2020) were unchanged except XTH4 in fah1-2 compared to the wild-type (Fig. 5). THESUS1 (THE1) is a receptor-like kinase involved in sensing defective cell wall growth and development, and a related gene FERONIA (FER) implicated maintenance of in pollen tube cell wall integrity sensing (Cheung and Wu 2011) were upregulated in the HrGHypAc by 78.61%, or unchanged compared to wild-type (Fig. 5). Similar to THE1 and FER, Wall Associated Kinases (WAKs) function as cell wall integrity sensing receptors associated with pectin and have an affinity for pectin oligomers which may be released during cell wall degradation (Amsbury 2020). HrGHypAc showed upregulation of several WAKs (WAK1, 3 and 4) (Fig. 5). This transcriptomic data suggested secondary cell wall-specific genes and receptor-like kinases are differentially regulated in mainly in HrGHypAc plants.