pHBMT1, a BAHD-family monolignol acyltransferase, mediates lignin acylation in poplar

Abstract

Poplar (Populus) lignin is naturally acylated with p-hydroxybenzoate ester moieties. However, the enzyme(s) involved in the biosynthesis of the monolignol-p-hydroxybenzoates have remained largely unknown. Here, we performed an in vitro screen of the Populus trichocarpa BAHD acyltransferase superfamily (116 genes) using a wheatgerm cell-free translation system and found five enzymes capable of producing monolignol-p-hydroxybenzoates. We then compared the transcript abundance of the five corresponding genes with p-hydroxybenzoate concentrations using naturally occurring unrelated genotypes of P. trichocarpa and revealed a positive correlation between the expression of p-hydroxybenzoyl-CoA monolig-nol transferase (pHBMT1, Potri.001G448000) and p-hydroxybenzoate levels. To test whether pHBMT1 is responsible for the biosynthesis of monolignol-p-hydroxybenzoates, we overexpressed pHBMT1 in hybrid poplar (Populus alba × P. grandidentata) (35S::pHBMT1 and C4H::pHBMT1). Using three complementary analytical methods, we showed that there was an increase in soluble monolignol-p-hydroxybenzoates and cell-wall-bound monolignol-p-hydroxybenzoates in the poplar transgenics. As these pendent groups are ester-linked, saponification releases p-hydroxybenzoate, a precursor to parabens that are used in pharmaceuticals and cosmetics. This identified gene could therefore be used to engineer lignocellulosic biomass with increased value for emerging biorefinery strategies.

Figure 1

Model of a p-hydroxybenzoylated lignin, and scheme showing how pHB enters lignification. (A) Model of a typical poplar lignin, showing pHB esters (dark blue) acylating primary hydroxyl groups on the lignin polymer. (B) Biosynthesis of monolignol–pHB conjugates and their incorporation into the lignin polymer. p-Hydroxybenzoyl-CoA and pHB are in dark blue.

Figure 2

Activity of putative pHBMT enzymes towards different CoA donors and monolignol acceptors. For each enzyme, the five different CoA donors were simultaneously tested with three canonical monolignol acceptors (p-coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol) in the multiplexed assay. A “+” indicates the formation of detectable levels of monolignol conjugate products when the respective enzyme was added. A “−” indicates that no detectable monolignol conjugate products were formed in the presence of the respective enzyme. The negative control consisted of a blank wheat germ translation reaction supplemented with the indicated CoA donor and the three monolignol acceptors.

Figure 3

Cell-wall-bound pHB (expressed as mg pHBA/g xylem tissue) of 316 9-year-old P. trichocarpa genotypes. The pHB amounts are ranked from low to high. Blue: P. trichocarpa genotypes originating from 44°N < latitude < 49.1°N, gray: P. trichocarpa genotypes originating from 49.12°N < latitude < 52.72°N, red: P. trichocarpa genotypes originating from 52.77°N < latitude < 54.18°N.

Figure 4

Spearman correlation matrix of xylem RNA expression levels of the BAHD clade containing the five putative pHBMTs and pHB content (expressed as mg pHBA/g xylem tissue). Potri.001G448000 (bold and star) is the only gene that shows significant correlation with pHB. Gene names in bold have in vitro activity towards p-hydroxybenzoyl-CoA, but do not show a significant correlation with pHB amount (except for Potri.001G448000). The other genes have no activity towards p-hydroxybenzoyl-CoA, but belong to the same phylogenetic clade (see Supplemental Figure S1). “x” marks correlations with P > 0.001389 (i.e., not statistically significant, Bonferroni correction).

Figure 5

Height, diameter, and expression of 35S::pHBMT1 and C4H::pHBMT1 poplars. (A) Height (cm) and (B) diameter (mm) of 4-month-old hybrid WT poplar (P39), 35S::pHBMT1, and C4H::pHBMT1 lines (n = 5 biological replicates for each line). (C) Representative 4-month-old WT line (left) and 4-month-old 35S::pHBMT1-23 line (right). (D) Expression analysis of pHBMT1 in transgenic lines, normalized to the highest expressing line (35S::pHBMT1-23). PtEF1β was employed as a reference gene, n = 3 biological replicates for each line (each with three technical replicates). All error bars represent sem.

Figure 6

Cell-wall-bound pHB (expressed as mg pHBA/g xylem tissue post-saponification; left y-axis, blue) and soluble pHB (expressed as mg pHBA/g xylem tissue post-saponification; right y-axis, yellow) of WT (P39), 35S::pHBMT1, and C4H::pHBMT1 poplars. The methanol extract of 4-month-old xylem tissue was saponified and quantified via HPLC to determine the amount of soluble pHBA. The remaining cell wall fraction was also saponified and analyzed on a HPLC to determine the amount of cell-wall-bound pHB. n = 3 biological replicates per line (each with two technical replicates), error bars represent sem. Statistical differences determined via Student’s t-test: *0.05 > P > 0.01; **0.01 > P > 0.001; and ***P < 0.001.

Figure 7

Amount of S-pHB released by DFRC of WT (P39), 35S::pHBMT1, and C4H::pHBMT1 poplars. Values are expressed in mg released S–p-hydroxybenzoate (S–pHB/g) cell wall. n = 3 biological replicates per line (each with two technical replicates), error bars represent sem. Statistical differences determined via Student’s t-test: *0.05 > P > 0.01.

Figure 8

Lignin analysis by 2D-NMR. A–C, Representative heteronuclear single quantum coherence spectra showing the aromatic region of enzyme lignin of (A) WT poplar (P39), (B) 35S::pHBMT1, and (C) C4H::pHBMT1. Integrated values of G, S, and pHBA units are given on an S + G = 100% basis. (D) Aromatic substructures colored to correspond with the signals in the heteronuclear single-quantum coherence spectra.