Compensatory Guaiacyl Lignin Biosynthesis at the Expense of Syringyl Lignin … · The lignin...

Compensatory Guaiacyl Lignin Biosynthesis at theExpense of Syringyl Lignin in 4CL1-KnockoutPoplar1[OPEN]

Chung-Jui Tsaiabcd23 Peng Xuab4 Liang-Jiao Xueabd5 Hao Huab6 Batbayar Nyamdaria Radnaa Narana

Xiaohong Zhoua7 Geert Goeminneef8 Ruili Gaogh Erica Gjersingij9 Joseph Dahlena

Sivakumar Pattathil jk10 Michael G Hahncdjk Mark F Davisdij John Ralphgh Wout Boerjanef andScott A Hardingab

aWarnell School of Forestry and Natural Resources University of Georgia Athens Georgia 30602bDepartment of Genetics University of Georgia Athens Georgia 30602cDepartment of Plant Biology University of Georgia Athens Georgia 30602dCenter for Bioenergy Innovation Oak Ridge National Laboratory Oak Ridge Tennessee 37831eDepartment of Plant Biotechnology and Bioinformatics Ghent University 9052 Ghent BelgiumfVlaams Instituut voor Biotechnologie UGent Center for Plant Systems Biology 9052 Ghent BelgiumgDepartment of Biochemistry University of Wisconsin-Madison Madison Wisconsin 53706hGreat Lakes Bioenergy Research Center Wisconsin Energy Institute University of Wisconsin-MadisonMadison Wisconsin 53726iNational Renewable Energy Laboratory Golden Colorado 80401jBioEnergy Science Center Oak Ridge National Laboratory Oak Ridge Tennessee 37831kComplex Carbohydrate Research Center University of Georgia Athens Georgia 30602

ORCID IDs 0000-0002-9282-7704 (C-JT) 0000-0001-8007-9879 (PX) 0000-0003-1766-5298 (L-JX) 0000-0003-4986-2034 (HH)0000-0002-0337-2999 (GG) 0000-0002-2073-7224 (EG) 0000-0003-3870-4137 (SP) 0000-0003-2136-5191 (MGH) 0000-0003-4541-9852(MFD) 0000-0002-6093-4521 (JR) 0000-0003-1495-510X (WB) 0000-0001-5098-2370 (SAH)

The lignin biosynthetic pathway is highly conserved in angiosperms yet pathway manipulations give rise to a variety of taxon-specific outcomes Knockout of lignin-associated 4-coumarateCoA ligases (4CLs) in herbaceous species mainly reduces guaiacyl(G) lignin and enhances cell wall saccharification Here we show that CRISPR-knockout of 4CL1 in poplar (Populus tremula 3 alba)preferentially reduced syringyl (S) lignin with negligible effects on biomass recalcitrance Concordant with reduced S-lignin wasdownregulation of ferulate 5-hydroxylases (F5Hs) Lignification was largely sustained by 4CL5 a low-affinity paralog of 4CL1typically with only minor xylem expression or activity Levels of caffeate the preferred substrate of 4CL5 increased in line withsignificant upregulation of caffeoyl shikimate esterase1 Upregulation of caffeoyl-CoA O-methyltransferase1 and downregulation of F5Hsare consistent with preferential funneling of 4CL5 products toward G-lignin biosynthesis at the expense of S-lignin Thustranscriptional and metabolic adaptations to 4CL1-knockout appear to have enabled 4CL5 catalysis at a level sufficient tosustain lignification Finally genes involved in sulfur assimilation the glutathione-ascorbate cycle and various antioxidantsystems were upregulated in the mutants suggesting cascading responses to perturbed thioesterification in lignin biosynthesis

The angiosperm lignin biosynthetic pathway hasbeen studied for decades and continues to be a topic ofinterest thanks to its plasticity (Sederoff et al 1999Boerjan et al 2003 Ralph et al 2004) A host of factorslikely contribute to the plasticity of lignification in-cluding enzyme redundancy and taxon-dependentproperties during development or in response to en-vironmental pressures (Weng and Chapple 2010Vanholme et al 2019) 4-CoumarateCoA ligase (4CL)catalyzes ATP-dependent CoA-thioesterification of vari-ous cinnamic acid derivatives (Knobloch and Hahlbrock1975) in arguably the most promiscuous step of mono-lignol biosynthesis In all sequenced angiosperm ge-nomes 4CL is encoded by multiple genes belonging totwo distinct phylogenetic classes (Ehlting et al 1999

Saballos et al 2012 Chen et al 2014b) AlthoughClass-II 4CLs involved in the biosynthesis of flavonoids andother soluble phenolics exhibit a substrate preferencefor 4-coumaric acid many lignin-associated Class-Imembers show similar in vitro affinities for caffeicacid 4-coumaric acid and sometimes ferulic acid(Ehlting et al 1999 Harding et al 2002 Lindermayret al 2002 Gui et al 2011 Chen et al 2013)The poplar (Populus trichocarpa Nisqually-1) genome

contains four Class-I 4CL members as two paralogouspairs derived from Salicoid whole-genome duplication(Supplemental Fig S1 Tsai et al 2006) Among them4CL1 (Potri001G036900) is known to be involved inlignin biosynthesis based on molecular and reverse ge-netic characterization in Populus tremuloides and Populus

Plant Physiology May 2020 Vol 183 pp 123ndash136 wwwplantphysiolorg 2020 American Society of Plant Biologists All Rights Reserved 123

httpsplantphysiolorgDownloaded on November 15 2020 - Published by Copyright (c) 2020 American Society of Plant Biologists All rights reserved

tremula 3 alba INRA 717-1B4 (Hu et al 1998 1999Voelker et al 2010) The genome duplicate 4CL5(Potri003G188500) is the only other 4CL gene familymember expressed in lignifying xylem but its tran-script levels are much lower than those of 4CL1(Supplemental Fig S1 Hefer et al 2015 Swamy et al2015 Hu et al 2016 Xue et al 2016Wang et al 2018)4CL1 and 4CL5were proposed to encode 4CL proteinsthat form heterotetramers in a 31 ratio (referred to asPtr4CL3 and Ptr4CL5 in Chen et al [2013 2014a]) Theactivity of individual isoforms as well as the tetramericcomplex is sensitive to inhibition by hydroxycinnamicacids and their shikimate esters (Harding et al 2002Chen et al 2014a Lin et al 2015) Such complexity of 4CLcatalysis is consistent with multiple cellular strategies for

directing hydroxycinnamic acids toward lignin biosyn-thesis Furthermore silencing lignin-associated 4CLs canhavedifferent effects on syringyl-to-guaiacyl lignin (SG)ratio depending on the species ranging from increasesin tobacco (Nicotiana tabacum Kajita et al 1996) Arabi-dopsis (Arabidopsis thaliana Lee et al 1997) and switch-grass (Panicum virgatum Xu et al 2011) to little change inalfalfa (Medicago sativa Nakashima et al 2008) to de-creases in rice (Oryza sativa Gui et al 2011) Variableeffects on SG ratio have even been reported amongclosely related poplar species (Hu et al 1999 Voelkeret al 2010 Chanoca et al 2019 and references therein)Such variation has been attributed both to differing de-grees of 4CL downregulation as well as to 4CL multi-plicity (Boerjan et al 2003 Saballos et al 2012) Analysisof knockout (KO) mutants should eliminate the uncer-tainty caused by partial gene silencing

Arabidopsis sorghum (Sorghum bicolor) and maize(Zea mays) 4cl mutants showed preferential reductionsin G-lignin resulting in increased SG ratios (Saballoset al 2012 van Acker et al 2013 Li et al 2015 Xionget al 2019) Though rare until recently transgenic nullscan now be efficiently obtained for genetically lesstractable systems like woody perennials or polyploidsusing CRISPRCas9 technology (Voytas and Gao2014 Bewg et al 2018) We previously reported thatCRISPR-KO of the predominant lignin 4CL in poplarled to a reduced SG ratio (Zhou et al 2015) whereassimilar KO in tetraploid switchgrass increased the SGratio (Xu et al 2011) This study aims to furthercharacterize the poplar 4cl1 mutants by more com-prehensive cell wall analysis biomass saccharificationphenolic profiling and RNA sequencing (RNA-seq)We present data to show distinct effects of 4CL-KO onlignin biosynthesis and enzymatic hydrolysis of cellwalls relative to 4cl mutants of Arabidopsis and otherherbaceous species Our findings suggest that alteredcaffeic acid homeostasis along with altered expressionof key lignin biosynthetic genes cooperatively sustainproduction of G-enriched lignin via the minor 4CL5pathway in the Populus mutants

RESULTS

4CL1-KO Preferentially Reduces S-Lignin Biosynthesis

We reported previously that CRISPR editing of 4CL1in P tremula 3 alba INRA 717-1B4 (hereafter ldquo4cl1mutantsrdquo) led to uniformly discolored wood and a 23reduction in lignin content relative to wild-type andCas9-only transgenic controls (hereafter ldquocontrolsrdquo)based on pyrolysis molecular beam mass spectrometry(pyMBMS) analysis (Zhou et al 2015) Total lignincontent determined by the Klason method revealeda 19 reduction in transgenic wood (Table 1) ThepyMBMS analysis also detected a 30 decrease in theSG ratio (Zhou et al 2015) attributable to a sharp (34)reduction of S-units and a small (5) though significantreduction of G-units (Table 1) Thioacidolysis showed asimilar response in S-lignin but a greater reduction (18)

1This work was supported in part by the Institute of Food andAgriculture United States Department of Agriculture (grant no2015ndash67013ndash22812) the Division of Integrative Organismal Systems(grant nos IOSndash1546867 and IOSndash09223992) the Division of Biolog-ical Infrastructure (grant no DBIndash0421683) of the National ScienceFoundation the BioEnergy Science Center (grant no WPNndashERKP695)the Center for Bioenergy Innovation (grant no WPNndashERKP886) and theGreat Lakes Bioenergy Research Center (grant no DEndashSC0018409)funded by the Office of Biological and Environmental Research Officeof Science of the United States Department of Energy

2Senior author3Author for contact cjtsaiugaedu4Present address Department of Genetics and Genomics Sciences

Icahn School of Medicine at Mount Sinai New York 100295Present address Key Laboratory of Forest Genetics and Biotech-

nology Co-Innovation Center for Sustainable Forestry in SouthernChina College of Forestry Nanjing Forestry University NanjingJiangsu 210037 China

6Present address Department of Plant Biology Ecology and Ev-olution Oklahoma State University Stillwater Oklahoma 74078

7Present address State Key Laboratory of Subtropical SilvicultureZhejiang Agriculture amp Forestry University Hangzhou Zhejiang310058 China

8Present address Vlaams Instituut voor Biotechnologie Metabolo-mics Core UGent-VIB Research Building 9052 Ghent Belgium

9Present address Los Alamos National Laboratory Los AlamosNew Mexico 87545

10Present address Mascoma LLC (Lallemand Inc) Lebanon NewHampshire 03766

The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (wwwplantphysiolorg) isChung-Jui Tsai (cjtsaiugaedu)

XZ generated transgenic plants and measured Klason lignin PXperformed histology and amplicon sequencing JD measured spe-cific gravity and acoustic velocity HH SP and MGH performedglycome profiling analysis BN RN and SAH performed glyco-syl composition analysis SAH measured crystalline cellulose con-tent GG and WB performed phenolic profiling analysis EG andMFD performed saccharification analysis through the BioEnergyScience Center RG and JR performed nuclear magnetic resonanceanalysis and coordinated other cell wall analysis through the GreatLakes Bioenergy Research Center L-JX performed bioinformaticanalysis C-JT conceived the project analyzed data and wrote thearticle with PX and SAH with contributions from other authors

[OPEN]Articles can be viewed without a subscriptionwwwplantphysiolorgcgidoi101104pp1901550

124 Plant Physiol Vol 183 2020

Tsai et al


of G-lignin in addition to a 52 increase of the minorcomponent H-lignin (Table 1) which is poorly pyro-lyzed and hence undetectable using pyMBMS (Sykeset al 2015) To probe the structural modifications inwalls of the mutant further two-dimensional nuclearmagnetic resonance (NMR) analysis was performedusing both whole cell wall residues and cellulolyticenzyme lignins The results were highly concordantwith those of pyMBMS and thioacidolysis showinga 30 decrease in SG ratio along with increasedH-lignin in the mutants (Fig 1) The frequency ofp-hydroxybenzoate units also increased in the mutants(Fig 1) but we do not see evidence for the appearanceof (higher levels of) bndashb-coupling products derivedfrom monolignol (largely sinapyl) p-hydroxybenzoateconjugates to produce tetrahydrofurans rather than thefamiliar resinols derived from unconjugated mono-lignols as has been observed in palms (Elaeis guineensisLu et al 2015) Taken together all analytical methodsemployed in this study showed consistent decreases ofSG ratio by 20 to 30 in the 4cl1 mutants resultingfrom a preferential reduction of S-ligninPhloroglucinol staining of stemcross sections confirmed

reduced lignification in the mutants (Fig 2 A and B)

In agreement with previous findings (Coleman et al2008 Voelker et al 2010) reduced lignin accrual ledto collapsed xylem vessels (Fig 2 A and C versus Band D) Accordingly wood specific gravity was sig-nificantly reduced in 4cl1 mutants (Fig 2E) The mu-tants also exhibited significantly lower acoustic velocity(Fig 2F) which is correlated with the microfibril angleof the S2 layer (Schimleck et al 2019) No other ap-parent growth anomaly was observed under green-house conditions It appears that the effects on poplargrowth by targeted KO of 4CL1were smaller than thosereported previously by inherently less specific antisensedownregulation (Voelker et al 2010) Further evalua-tion under field conditionswould be necessary to assessthe impact of lignin alteration on mutant growth

4CL1-KO Has No Obvious Effects onBiomass Saccharification

Reduction of lignin content regardless of SG changehas been shown to improve enzymatic sugar releasein alfalfa Arabidopsis and poplar (Chen and Dixon2007 Mansfield et al 2012a van Acker et al 2013

Table 1 Wood chemical properties of control and 4cl1-KO poplars

Data represent means 6 SE Statistical significance was determined by two-sample Studentrsquos t test and indicated by bold-faced P values

Cell Wall Composition Control 4cl1 P Value Change

Klason lignin ( dry weight n 5 7ndash9)Total lignin content 1594 6 040 1293 6 037 0001 219

pyMBMSa (arbitrary units n 5 7ndash12)G 1808 6 017 1718 6 019 0006 25S 3276 6 021 2160 6 028 0001 234SG ratio 181 6 002 126 6 001 0001 230

Thioacidolysis (mmolg Klason lignin n 5 5)H 2915 6 101 4419 6 120 0001 52G 114637 6 2962 94408 6 2645 0001 218S 179843 6 5337 118051 6 4108 0001 234SG ratio 157 6 002 125 6 001 0001 220

Crystalline cellulose ( dry weight n 5 8)Glc 4726 6 042 4585 6 030 0016 23

Hemicelluloses ( dry weight n 5 5)Ara 029 6 001 033 6 001 0032 15Rha 048 6 001 049 6 000 0062 3Xyl 1412 6 019 1509 6 013 0003 7Man 095 6 002 070 6 002 0001 226Gal 065 6 003 060 6 002 0190 28Glc 403 6 017 377 6 016 0290 26

Glycosyl composition (mol n 5 5ndash6)Ara 102 6 006 112 6 004 0169 10Rha 197 6 008 200 6 010 0796 2Fuc 012 6 002 017 6 002 0092 42Xyl 6005 6 142 6165 6 196 0471 3GlcA 198 6 007 230 6 011 0035 16OMe-GlcA 058 6 004 083 6 009 0019 43GalA 350 6 014 447 6 030 0013 28Man 137 6 003 103 6 009 0002 225Gal 347 6 038 280 6 012 0123 219Glc 2588 6 134 2365 6 200 0320 29

apyMBMS data were from Zhou et al (2015)

Plant Physiol Vol 183 2020 125

4CL Redundancy in Poplar Lignification


Wang et al 2018) Unexpectedly we found no differ-ence between control poplar and 4cl1 mutants in en-zymatic Glc and Xyl release after hydrothermalpretreatment (Fig 2G) using the high-throughput re-calcitrance assay developed by the US Department ofEnergy-funded BioEnergy Science Center (BESC Seliget al 2010) Independent analyses using the high-throughput method of the Department of Energy-Great Lakes Bioenergy Research Center (GLBRCSantoro et al 2010) also detected no differences in Glcrelease (Fig 2G) Mutant samples released slightlyhigher levels of Xyl but only after less severe (hotwater and dilute acid not alkaline) pretreatments(Fig 2G) The overall lower sugar yields from theGLBRC assays are similar to those reported elsewhere(Wilkerson et al 2014) and reflect the milder pre-treatment and enzymatic hydrolysis conditions thanthose applied with the BESC methods (Santoro et al2010 Selig et al 2010)

Altered Lignin-Carbohydrate Interactions in the Cell Wallof 4cl1 Mutants

As lignin is purportedly cross linked with cell wallpolysaccharides we next examined the 4CL1-KO effects

on cell wall glycans and their interaction with ligninWe detected a small decrease in crystalline cellulosecontents of the mutants (Table 1) Monosaccharidestypically associated with hemicelluloses showed slightchanges with Man decreasing significantly and themost abundant Xyl increasing slightly (Table 1) Gly-cosyl residue composition analysis further revealedincreased mol of GlcA its 4-O-methylated form andGalUA the main component of pectins (Table 1) Thesedata suggest altered cell wall polysaccharide composi-tion in lignin-reduced 4cl1 mutants

Extractive-free wood meals were then subjected tocell wall fractionation to provide a finer resolutionanalysis of matrix polysaccharide organization by gly-come profiling using a panel of cell wall glycan-directedmonoclonal antibodies (mAbs Pattathil et al 2010 2012)The amounts of extractable cellwallmaterial differed littlebetween control and 4cl1 mutants (Supplemental FigS2A) The glycan epitope profiles from cell wall fractionsenrichedwith pectins (oxalate and carbonate extracts) andhemicelluloses (1 and 4 M KOH extracts) were largelysimilar between genotypes with a few epitopes exhib-iting slightly lower signals in the mutants (Fig 3Supplemental Fig S2B) In contrast we observed sig-nificantly increased polysaccharide extractability in thechlorite fraction of mutant cell walls (Fig 3) especially

Figure 1 NMR analysis of 4cl1 mutant and control poplar wood Representative 1Hndash13C heteronuclear single-quantum co-herence correlation spectra of the aromatic region of enzyme lignins from ball-milled wood samples A to C Wild type (A)Cas9 (B) and 4cl1mutant (C) The main lignin structures and linkages identified are illustrated below and color-coded to matchtheir assignments in the spectra Volume integrals (with the same color coding) were measured using the a-CH correlation peaksfrom A- B- and C-units and S26 1 S926 G2 and H26 (corrected for Phe) aromatics (with the integrals halved as usual for theS- H- and C-units) are noted as the mean6 SE of biological replicates (n 5 3 for wild type 2 for Cas9 and 5 for 4cl1 Kim andRalph 2010 Mansfield et al 2012b) Note that the interunit linkage distribution (ABC) is determined from volume integrals ofjust those units and made to total 100 small differences (such as the apparently enhanced b-ether A-level in the 4cl1 line)should not be overinterpreted aswewere unable to delineate authenticated peaks for nor therefore obtain reliable accounting ofvarious tetrahydrofurans (from bndashb-coupling of monolignol p-hydroxybenzoates) nor of the 4ndashOndash5- and 5ndash5-linked units thatrequire one or two G-units (and would therefore logically be higher in the 4cl1 lines with their higher G-units) Also note thattheH-unit (H26) correlation peak overlapswith another peak fromPhe protein units (Kim et al 2017) integrals were corrected bysubtracting the integral from the resolved Phe peak below it to obtain the best estimate available


Tsai et al


for epitopes recognized by mAb groups 4-O-methylGlcA-substituted xylans (Xylan-5) linear unsubstitutedxylans (Xylan-6Xylan-7) rhamnogalacturonan I ara-binogalactan and pectic (rhamnogalacturonan I andhomogalacturonan) backbones (Pattathil et al 2010Schmidt et al 2015 Ruprecht et al 2017) Becausechlorite degrades lignin and liberates lignin-boundglycans the enhanced carbohydrate extractability inthe chlorite fraction is consistent with less interactionbetween matrix polysaccharides and lignin in the mu-tant cell wall Whether the increased release of thoseglycans reflects reduced lignin abundance andor al-tered lignin composition was not determined

Altered Phenylpropanoid Metabolism in 4cl1 Mutants

Ultra performance liquid chromatography-massspectrometry profiling of methanolic extracts fromdeveloping xylem revealed a major shift in phenyl-propanoid metabolism in the mutants Consistentwith decreased lignin biosynthesis various oligoli-gnols and their hexosides were detected at drasti-cally reduced levels in the mutants (Table 2) Alsosignificantly reduced were lignin pathway interme-diates 4-coumaroyl and caffeoyl shikimate esters(Table 2) whose synthesis depends on 4CL-activated

hydroxycinnamoyl-CoAs as acyl donors (Hoffmannet al 2003) Among potential 4CL substrates caffeicacid is the only hydroxycinnamic acid consistentlydetected in control xylem under our experimentalconditions (Table 2) The 4cl1 mutants accrued ele-vated levels of caffeic acid and its sulfate ester alongwith very high levels of phenolic hexose esters andhexosides (Table 2) In particular a caffeic acid 34-O-hexoside increased by nearly 80-fold and becamethe most abundant soluble phenylpropanoid identi-fied in the mutant xylem extracts (Table 2) Mutantsalso contained detectable levels of 4-coumaric acidand ferulic acid sulfate as well as highly increasedhexose conjugates of 4-coumarate ferulate andsinapate (Table 2) Chlorogenic acid (5-O-caffeoylquinate) the predominant hydroxycinnamoyl qui-nate ester in poplar changed little in the mutantswhereas the less abundant 3-O-caffeoyl quinate 3-O-feruloyl quinate and 4-O-feruloyl quinate increasedsignificantly (Table 2) It appears that unlike forthe shikimate conjugates synthesis of hydrox-ycinnamoyl quinate esters in poplar is largely inde-pendent of 4CL1 Together the phenolic profilingdata revealed a buildup of hydroxycinnamates andtheir diversion into nonstructural phenylpropanoidpools at the expense of lignin precursors in the 4cl1mutants

Figure 2 Histology physical properties andsaccharification of mutant and control poplarwood A and B Stem cross sections (10th inter-node) stained with phloroglucinol C and D Stemcross sections (8th internode) stained with tolui-dine blue Representative images from control(wild type or Cas9) and 4cl1 mutant lines areshown Collapsed vessels are marked with aster-isks Scale bars 5 50 mm E and F Wood-specificgravity (E) and acoustic velocity (F) of control (n59) and 4cl1 (n5 12) samples G Enzymatic sugarrelease of control and mutant samples after BESChydrothermal (n 5 6) and GLBRC hydrothermaldilute acid and dilute alkaline pretreatments(n 5 5) Data in E to H represent means 6 SE Sta-tistical significance was determined by Studentrsquost test (P 0001 P 001 and P 005)




Transcriptional Adjustments of Lignin Biosynthesis

RNA-seq analysis of developing xylem tissues fromsix representative mutants and six controls (three wild-type and three Cas9 controls) identified 1752 geneswith significantly altered transcript abundance in re-sponse to 4CL1-KO approximately two-thirds of whichwere upregulated (1153) and the remaining (599)downregulated Because guide RNA (gRNA)-directedCRISPRCas9 mutations are located in the first exon ofPta4CL1 (Zhou et al 2015) aberrant transcripts con-taining premature stop codons resulting from frame-shift indels (Supplemental Table S1) would producenonfunctional proteins in the mutants In addition theaberrant transcripts are potential targets for nonsense-mediated mRNA decay (Conti and Izaurralde 2005)Accordingly 4CL1 was among the most significantlydownregulated genes in themutants with its (aberrant)transcript levels decreased by 92 (Fig 4A) Our datathus underscore the efficacy of CRISPRCas9 muta-genesis at multiple levels affecting steady-state tran-script accumulation as well as protein function

Transcript levels of many lignin biosynthetic genesincluding 4CL1 paralog 4CL5 were not significantlychanged in the 4cl1 mutants However the genomeduplicates F5H1 and F5H2 encoding ferulatecon-iferaldehyde 5-hydroxylases were significantly down-regulated and they are the only known monolignolpathway genes besides 4CL1 to show such patterns(Fig 4A) As F5Hs occupy the branch-point into S-lignin

biosynthesis their downregulation is consistent withthe preferential reduction of S-lignin in 4cl1 mutantsOther genes implicated in lignification including aUDP-glycosyltransferase (Potri014G096100) orthologous toArabidopsis UGT72B1 involved in monolignol glycosy-lation (Lin et al 2016) and a laccase (Potri010G193100)involved in oxidative coupling of monolignols (Ranochaet al 2002 Lu et al 2013) were also downregulated inthemutants (Fig 4A) In contrast transcript levels of twolignin genes increased significantly and they encodeenzymes acting immediately upstream (caffeoyl shiki-mate esterase CSE1) and downstream (caffeoyl-CoAO-methyltransferase CCoAOMT1) of the 4CL reactionwith caffeic acid Both enzymes are encoded by genomeduplicateswith similar expression levels in poplar xylemand in both cases only one copy eachwas affected in the4cl1 mutants (Fig 4A) This suggests involvement ofCSE1 and CCoAOMT1 in homeostatic regulation ofcaffeic acid and caffeoyl-CoA in response to 4CL1-KORecently a cytosolic ascorbate peroxidase (APX) wasshown to possess 4-coumarate 3-hydroxylase (C3H) ac-tivity and to convert 4-coumaric acid to caffeic acid inboth monocots and dicots supporting an alternativeroute in lignin biosynthesis involving free phenolic acids(Barros et al 2019) Two putative orthologs of this dual-function APX-C3H are present in the poplar genomeand one of them (Potri009G015400 APX-C3H1) wassignificantly upregulated in the mutants (Fig 4A) Al-though catalytic properties remain to be confirmedupregulation of the putative poplarAPX-C3H1may alsocontribute to elevated accumulation of caffeic acid andits derivatives

Cell Wall Remodeling and Detoxification Responses in the4cl1 Mutants

Genes downregulated in the KO mutants showed anoverrepresentation of gene-ontology (GO) terms asso-ciated with cell wall biogenesis and polysaccharide bi-osynthesis (Fig 4C) Examples include transcriptionfactors (TFs) such as top-level wood-associated NAC-domain (WND) proteins WND1A and WND5A anddownstream TFs NAC154 NAC156 MYB10 MYB128MYB167 and MYB221 that regulate lignin celluloseand xylan biosynthesis (Ye and Zhong 2015 Fig 4A)There was widespread downregulation of genes in-volved in the biosynthesis of all major cell wall gly-cans including cellulose xylans and pectins (Fig 4A)In contrast genes encoding cell wall-modifying en-zymes or cell wall-loosening proteins were signifi-cantly upregulated (Fig 4B) The results are consistentwith the altered cell wall polysaccharide compositionand extractability and providemolecular bases for cellwall remodeling as a result of impaired lignification inthe mutants

GO categories associated with detoxification sulfurassimilation and oxidation-reduction processes wereover-represented among genes upregulated in themutants (Fig 4C) Of particular interest are genes

Figure 3 Glycome profiling of 4cl1 mutant and control poplar woodHeatmap depiction of signal intensities resulting from binding of cellwall glycan-directed mAbs to two cell wall fractions extracted bysodium carbonate and chlorite from control and mutant samples mAbsare arranged in rows by the cell wall glycan epitope groups (denoted byvertical color bars on the right) and biological replicates (n 5 6) areshown in columns Asterisks indicate significant differences betweenplant groups determined by Studentrsquos t test (P 001 fold-change$ 15)


Tsai et al


involved in sulfur assimilation into Cys for synthesisof other sulfur-containing compounds includingglutathione Upregulation of genes encoding gluta-thione S-transferases (GSTs) redox-active proteinsand enzymes in the glutathione-ascorbate cycle un-derscores the likelihood of redox adjustments in themutants (Fig 5)

Conditional Involvement of 4CL5 in Lignin Biosynthesis

The 4cl1mutants somehow accrue80 of wild-typelignin levels (Table 1) even though 4CL1 transcriptsnormally comprise 80 of xylem-expressed 4CL tran-script levels in poplar (Supplemental Fig S1) Theonly other xylem-expressed 4CL isoform 4CL5 must

therefore sustain lignin biosynthesis in the absence of4CL1 Interestingly CRISPR-KO of 4CL5 did not changeeither lignin content or SG ratio of the stem wood(Supplemental Fig S3) suggesting only a minor or con-ditional role in lignification under normal growth condi-tions As 4CL5 transcript abundance was not significantlyaltered in the 4cl1 mutants (Fig 4A) other mechanism(s)must exist for in vivo enhancement of its function Weconstructed genotype-specific xylem gene coexpressionnetworks (see ldquoMaterials and Methodsrdquo) to compare theexpression contexts of 4CL1 and 4CL5 and their shifts inresponse to 4CL1-KO (Supplemental Fig S4) The wild-type network placed PAL2 and 4CL1 along withNAC154andMYB128 in the brown module 4CL5 and APX-C3H1in the yellow module and PAL5 CSE1 CCoAOMT1F5H1 and F5H2 along with MYB167 andMYB221 in the

Table 2 Xylem phenolic metabolites with significantly altered abundance in 4cl1 mutants

Data represent mean signal intensities 6 SE of n 5 10 control (three wild-type and seven Cas9 vector control lines) or 14 4cl1-KO lines Only datawith signal intensities $ 1000 in at least one group (except ) significant differences (P 0002) between groups based on Studentrsquos t test(except ) and confirmed peak annotation are shown The exceptions are included for reference

NameRetention

Timemz Ion Control 4c1

Fold

Change

OligolignolG(8-O-4)G 9-O-Hexoside 715 5371995 M-H 1820 6 123 47 6 22 239G(8-O-4)G Hexoside 744 5371995 M-H 7153 6 430 1406 6 167 251G(8-O-4)G(8-O-4)G 994 5712202 M-H 3242 6 190 3 6 2 21064Gox(8-O-4)G 1192 3731307 M-H 1321 6 70 12 6 4 2110G(8-5)G [2H2O] 1324 3391250 M-H 2049 6 181 23 6 10 289G(8-O-4)S(8-5)G 1495 5832209 M-H 6137 6 402 65 6 49 294G(8-O-4)S(8-5)G 1577 5832205 M-H 1264 6 154 2 6 1 2641

Hydroxycinnamoyl shikimate4-Coumaroyl shikimate 972 3190841 M-H 17617 6 4276 4658 6 734 238Caffeoyl shikimate 628 3350822 M-H 79601 6 16211 16973 6 1928 247Caffeoyl shikimate 778 3350791 M-H 16415 6 2815 2844 6 324 258Feruloyl shikimate 1031 3490942 M-H 682 6 140 23 6 10 230

Phenolic acid4-Coumaric acid 796 1630399 M-H 41 6 13 760 6 114 19Caffeic acid 504 1790352 M-H 1110 6 121 3404 6 389 31Caffeic acid sulfate 182 2589947 M-H 1392 6 138 4650 6 254 33Ferulic acid sulfate 507 2730085 M-H 0 6 0 3912 6 315 infin

Phenolic acid hexosehexoside4-Coumaroyl hexose 464 3250996 M-H 2876 6 431 212766 6 15741 744-Coumaroyl hexose 521 3250985 M-H 1994 6 334 88055 6 5699 44Caffeoyl hexose 343 3410931 M-H 224 6 103 13415 6 936 60Caffeic acid 34-O-hexoside 413 3410946 M-H 7950 6 649 611832 6 22149 77Caffeic acid 34-O-hexoside 561 3410937 M-H 6539 6 302 205906 6 12360 32Ferulic acid 4-O-hexoside 383 7112309 2M-H 117 6 56 209459 6 10490 1798Feruloyl hexose 546 3551104 M-H 1539 6 415 184710 6 11173 120Feruloyl hexose 554 7112215 2M-H 0 6 0 32748 6 3070 infinFeruloyl hexose 596 3551093 M-H 266 6 144 38777 6 3127 146Sinapic acid 4-O-hexoside 449 7712552 2M-H 1463 6 217 263304 6 8064 180Sinapic acid 4-O-hexoside 621 3851179 M-H 1091 6 89 78135 6 5624 724-Hydroxybenzoyl hexose 274 2990824 M-H 4356 6 426 95399 6 9034 22Vanillic acid 4-O-hexoside 230 6591967 2M-H 9322 6 313 363392 6 13301 39

Hydroxycinnamoyl quinate3-O-Caffeoyl quinate 311 3530889 M-H 2001 6 284 7236 6 833 363-O-Feruloyl quinate 483 3671046 M-H 1026 6 127 2665 6 341 264-O-Feruloyl quinate 683 3671047 M-H 393 6 85 1746 6 194 445-O-Caffeoyl quinate 432 3530911 M-H 21688 6 2000 28126 6 2621 134-O-Caffeoyl quinate 469 3530893 M-H 2283 6 220 2630 6 266 12




turquoise module (Fig 6 Supplemental Dataset S1) Inthe 4cl1-KO network 4CL1 CCoAOMT2 F5H1 F5H2NAC154 MYB128 and MYB167 were assigned to theturquoise module and PAL2 PAL5 4CL5 CSE1 andCCoAOMT1 to the blue module (Fig 6 SupplementalDataset S1) The data support distinct regulation of4CL1 and 4CL5 and suggest transcriptional rewiringof 4CL5 CSE1 and CCoAOMT1 to the same coex-pressionmodule as part of a compensatory response tosustain lignification in the 4cl1 mutants

DISCUSSION

4CL-KO Effects on Lignin Composition and EnzymaticHydrolysis Differ between Poplar and Herbaceous Species

The cell-wall polymer lignin provides structural in-tegrity and protection to plants (Weng and Chapple2010) Its abundance and structural complexity how-ever render it a negative factor in forage qualityand inhibit its removal in cellulosic biomass utilizationfor pulping and biofuels production (Boerjan et al2003) Studies with wild poplar populations (Studeret al 2011) sorghum and maize brown midrib mutants

Figure 4 Transcriptional responses of 4cl1 mutants A Expression re-sponse heatmaps of genes involved in cell wall biogenesis B Expres-sion response heatmaps of genes involved in cell wall remodelingLogFC log2-transformed fold-change (mutantcontrol) values frag-ments per kilobase of transcript per million mapped reads (FPKMs)average transcript abundances of control samples Only genes withsignificant differences (Q 001 boldface and underlined) and withcontrol FPKM $ 10 are shown except for genome duplicates markedwith asterisks C GO enrichment of differentially up- or downregulatedgenes in the mutants Representative GO terms are shown with thenegative log10-transformed P values Heatmaps are visualized accord-ing to the color scales at the bottom AAT arabinosyltransferaseAGP arabinogalactan protein ARK ARBORKNOX ASD arabinofur-anosidase BXL b-xylosidasesa-arabinofuranosidase CCoAOMTcaffeoyl-CoA O-methyl-transferase CEL cellulase CesA cellulosesynthase CSI cellulose synthase interacting protein CSLA mannansynthase F5H ferulate 5-hydroxylase GATL galacturonosyl-transferasendashlike GAUT galacturonosyltransferases GH glycosylhydrolase GH9B class B endoglucanase GMPP GDP-Man pyro-phosphorylase GT glycosyltransferase GT43 xylan xylosyltransferaseGUX glucuronyltransferase GXM glucuronoxylan O-methyltransferaseKOR KORRIGAN MAN mannase PME(I) pectin methyl-esterase (in-hibitor) XOAT xylan O-acetyltransferase XTH xyloglucan endo-transglucosylase

Figure 5 Transcriptional responses of detoxification genes in themutants Expression response heatmaps of genes associated withglutathione-ascorbate metabolism sulfur assimilation and antioxidantsystems Data presentation is the same as Figure 4 Only genes withcontrol fragments per kilobase of transcript per million mappedreads (FPKM) $ 5 (or $10 in the case of GSTs and PRDs) are shownAPK adenosine-phosphosulfate kinase APS ATP sulfurylase APSRadenosine-phosphosulfate reductase CLT chloroquine-resistance-like(glutathione) transporter CSD CuZn superoxide dismutase DHARdehydroascorbate reductase GPP Gal-1-phosphate phosphataseGRX glutaredoxin GulLO gulonolactone oxidase LSU response tolow sulfur MSR Met sulfoxide reductases PRD peroxidase PSKphytosulfokine SAT Ser acetyltransferase SIR sulfite reductase TRXthioredoxin


Tsai et al


(Saballos et al 2008 Xiong et al 2019) ArabidopsisT-DNAmutants (van Acker et al 2013) and transgenicalfalfa and poplar (Chen and Dixon 2007 Mansfieldet al 2012a Wang et al 2018) have reported a majorand negative effect of lignin content on enzymaticsugar release with lignin SG ratio playing a minorrole For instance KO of lignin-associated 4CL ledto reduced lignin content and improved enzymaticsaccharification in Arabidopsis maize sorghum andswitchgrass (Saballos et al 2008 Van Acker et al 2013Park et al 2017 Xiong et al 2019) However for thepoplar 4cl1 mutants investigated here similar degreesof lignin reduction had at best negligible effects onsugar release by multiple pretreatment methods (Fig 2G and H) The SG ratio was reduced by 30 inpoplar mutants due to drastically reduced S-lignin a-bundance (Table 1) but was increased in Arabidopsisswitchgrass sorghum and maize mutants owing tostrong reductions of G-lignin contents (Saballos et al2008 van Acker et al 2013 Park et al 2017 Xionget al 2019) Poplars and other angiosperm trees withsignificant secondary growth typically contain S-richlignin (SG 15) whereas herbaceous species areusually G-rich (eg Sun et al 2012) The biased effectsof 4CL-KO on S- and G-lignin flux of woody and her-baceous species therefore seem to track with canalizedtaxonomic determinants of lignin SG ratio S-lignin isknown to be associated with more chemically labilelinkages and to possess a higher reactivity duringpulping as well as various biomass pretreatment re-gimes (Studer et al 2011 Mansfield et al 2012a) Itthus appears that the potential gain in enzymatic hy-drolysis brought about by reduced lignin was offset by

decreased SG ratio in the poplar 4cl1 mutants Weinterpret these results to suggest that lignin reductionpredominantly in the form of S-units does not improvebiomass saccharification

Transcriptional and Metabolic Coordination Enables4CL5 Compensation

Plasticity in the lignin biosynthetic pathway as evi-denced by multiple possible routes (Fig 6 Barros et al2019) may accommodate developmental environmen-tal or species-dependent variations in lignification 4CLsfor instance can utilize multiple hydroxycinnamate sub-strates in vitro but how substrate utilization is regulatedin vivo is largely unknown 4CL activation of 4-coumarateis obligatory for production of the downstream interme-diates 4-coumaroyl and caffeoyl shikimate esters in lig-nin biosynthesis (Schoch et al 2001 Hoffmann et al2004 Coleman et al 2008) and this was confirmed bygreatly depleted shikimate ester levels in the 4cl1 mu-tants (Table 2) Like many Class-I 4CLs involved in lig-nin biosynthesis poplar 4CL1 also utilizes caffeic acidwhich is the most abundant free hydroxycinnamate inpoplar xylem (Harding et al 2002 Chen et al 2013Table 2) The significant increases of free caffeic acid andits hexose conjugates in the mutant xylem are consistentwith a substrate buildup due to impaired 4CL1 conver-sion Caffeic acid accrual can also be attributed to con-version from surplus 4-coumaric acid another 4CL1substrate by the upregulated APX-C3H1 as well asfrom residual caffeoyl shikimate via CSE1 No incorpo-ration of caffeates or caffeoyl alcohol into the lignin

Figure 6 Schematic of the transcrip-tional and metabolic adjustments oflignin biosynthesis in 4cl1 mutants AThe major monolignol biosyntheticroutes are shown with relevant path-way steps and intermediates discussedin this study Red and blue fonts indi-cate higher and lower abundances re-spectively in the 4cl1 mutants Theslightly reduced G-lignin is shown inblack to contrast with the drasticallyreduced S-lignin in blue Orangebackground shading depicts the path-way flows with thickness representingrelative transcript and metabolite levelchanges detected in the mutants B andC Coexpression module assignmentsfor lignin pathway genes and TFs incontrol (B) and mutant (C) networksGenes are arranged and color-coded bymodules they were assigned to and redand blue text in C indicates up- ordownregulation in the mutants re-spectively Abbreviations are the sameas in Figure 4




polymer was evident in the NMR spectra (even at levelsnearer the baseplane than those plotted in Fig 1) Besidesthe reduced SG ratio and consequent changes in theinterunit linkage distribution that were largely as ex-pected and the elevation of p-hydroxybenzoate levelstherewere no other evident structural changes in the 4cl1lignin and no obviously new components as has oftenbeen observed in mutants and transgenics (Ralph andLanducci 2010)

Several lines of evidence suggest that increased caf-feic acid accrual in the 4cl1mutants was not merely dueto precursor buildup but to a tightly regulated com-pensatory mechanism to sustain lignification (1) The4cl1 mutants exhibit 20 lignin reduction that thereduction is not more severe means that substantiallevels of lignin biosynthesis must be upheld by 4CL5the remaining xylem isoform (2) Xylem transcript andprotein abundances of 4CL5 are several-fold lower thanthose of 4CL1 (Fig 4A Wang et al 2018) (3) 4CL1 and4CL5 exhibited distinct coexpression patterns withother genes and TFs implicated in lignin biosynthesis(Fig 6) These along with the silent lignin phenotype of4cl5mutants during normal growth (Supplemental FigS2) all point to 4CL59s conditional involvement in lig-nin biosynthesis under specific circumstances (4) Allthree lignin genes upregulated in the 4cl1 mutantsCSE1 APX-C3H1 and CCoAOMT1 occupy pathwaysteps immediately upstream and downstream of caffeicacid-to-caffeoyl-CoA activation (Fig 6) This is in sharpcontrast to Arabidopsis 4cl1 mutants in which signifi-cant upregulation of early pathway genes AtPAL2AtC4H and AtC3ʹH was reported (Vanholme et al2012) Our finding that 4CL5 CSE1 and CCoAOMT1belonged to the same coexpression module in the 4cl1-specific network highlights a specific involvement ofCSE1 and CCoAOMT1 in the distinct lignin composi-tional changes in the poplar 4cl1 mutants ElevatedCSE1 expression would be expected to direct residuallignin biosynthetic pathway fluxes toward caffeic acidwhile also relieving the buildup of potential 4CL in-hibitors 4-coumaroyl and caffeoyl shikimates (Linet al 2015) in the process The increased supply ofcaffeic acid would be necessary for 4CL5 catalysis tocommence because its estimated Km value is several-fold higher than that of the predominant 4CL1 (Chenet al 2013) The observed 3-fold increase in free caffeicacid levels in the 4cl1 mutants is in general agreementwith this scenario Upregulation of the downstreamCCoAOMT1 could further drive the caffeate-to-caffeoyl-CoA flux toward lignification This along with caffeicacid glycosylation could prevent overaccumulation ofcaffeic acid that may cause substrate self-inhibition of4CL5 (Chen et al 2013) Thus transcriptional coac-tivation of CSE1 and CCoAOMT1 in effect altered thebiochemical milieu enabling the low-affinity 4CL5 toengage in a compensatory function to sustain lignifi-cation even though 4CL5 expression was not increasedin the 4cl1mutants Regardless the 4CL5 pathway wasnot very efficient and could only partially restore lig-nification in the 4cl1 mutants We further argue that

the 4CL5 compensatory pathway is primarily chan-neled toward G-lignin biosynthesis due to down-regulation of F5H paralogs The data thus provide amechanistic basis for the biased effect on S-lignin ac-crual in the poplar 4cl1 mutants

Crosstalk among Lignification Redox Regulation andSulfur Metabolism

The two lignin biosynthetic enzymes APX-C3H1 andCSE1 involved in caffeic acid synthesis both possessadditional functions associated with detoxification ofhydrogen peroxide lipid peroxides and superoxide(Gao et al 2010 Barros et al 2019) Their upregulationin 4cl1 mutants coincided with induction of GSTsPRDsCSDsMSRs and genes encoding enzymes of theglutathione-ascorbate cycle all of which have beenimplicated in scavenging of reactive oxygen species(Rouhier et al 2008 Molina-Rueda et al 2013 Rey andTarrago 2018) Coordinated regulation of multipleantioxidant systems may be crucial to maintain redoxbalance under the presumably highly oxidative cellularenvironment resulting from disturbed lignin biosyn-thesis GSTs also mediate conjugation of glutathioneto xenobiotics or electrophilic natural products in-cluding phenylpropanoids for sequestration andortransport (Dean et al 1995 Dixon et al 2010) Thusthe widespread upregulation of GSTs in the 4cl1 mu-tants could also be linked to detoxification of excessphenylpropanoids

Besides its roles in redox balance and detoxificationglutathione as an abundant low-molecular-weight thiolmetabolite is intimately associated with sulfur homeo-stasis (Takahashi et al 2011) Many sulfur assimilationpathway genes can be induced by sulfur starvation orupon Cd exposure which triggers synthesis of gluta-thione and its oligomeric phytochelatins for metalchelation resulting in sulfur depletion (Scheerer et al2010 Honsel et al 2012) Interestingly two genesbelonging to the plant-specific LSU family were sig-nificantly upregulated in the mutant xylem (Fig 5)hinting at potential sulfur limitation there Sulfurassimilation also supplies sulfate donors for sulfationof diverse peptides proteins and metabolites (Jezet al 2016) Indeed we observed increased abun-dance in the mutants of two unusual sulfate esterscaffeic acid sulfate and ferulic acid sulfate thoughwith unknown function (Table 2) Taken togetherelevated levels of glutathione conjugation phenolicacid sulfates and other sulfated compounds mighthave depleted available pools of reduced sulfur in the4cl1 xylem which in turn activated the sulfur assim-ilation pathway

CONCLUSIONS

Our data support differences in regulation and per-turbation of lignin biosynthesis between Populus and


Tsai et al


herbaceous species We identified the juncture betweencaffeic acid and caffeoyl-CoA as being highly sensitiveto 4CL1-KO while also key to modulating the com-pensatory function of 4CL5 Transcriptional metabolicand biochemical coordination of the compensatorypathway likely underscores the species-specific ligninperturbation response reported here The work raisesthe enticing possibility that such lineage-specific capabil-ity might have originated from taxon-dependent devel-opmental plasticity that warrants future investigation

MATERIALS AND METHODS

Plant Materials

The 4cl1mutants and Cas9-only vector controls were reported in Zhou et al(2015) All mutant lines harbor frameshift mutations in the gRNA target site of4CL1 (GAGGATGaTaAAaTCTGGAGGGG the protospacer adjacent motif isunderlined and sequence polymorphisms with 4CL5 are in lowercase) whichhave been reconfirmed using leaves from plants that have been vegetativelypropagated and maintained in the greenhouse for over 4 years (Bewg et al2018 Supplemental Table S1) 4CL5-KO plants were produced by targeting thehomologous site (GAGGATGtTgAAgTCTGGAGGG) The construct was pre-pared according to Jacobs and Martin (2016) except two oligos tailed withMedicago truncatula U6 (AACTCCAGACTTCAACATCCTCAAGCCTACTGGTTCGCTTGA) or scaffold sequence (GAGGATGTTGAAGTCTGGAGTTTTAGAGCTAGAAATAGCAAGTT underlined) were used in a half (10 mL) reac-tion containing 0005 pmol of linearized p201N vector and 01 pmol of inserts(U6 scaffold and the two gRNA oligos) using the Gibson AssemblyMaster Mix(New England Biolabs)Agrobacterium-mediated transformation of P tremula3alba INRA 717-1B4 was performed according to Leple et al (1992) Editingpatterns were determined by amplicon-sequencing using 4CL14CL5 consen-sus primers (GTTCAGACGTGTGCTCTTCCGATCTAGCACCGGTTGTHCCAand CCTACACGACGCTCTTCCGATCTGAGGAAACTTRGCTCTGAC) tailedwith Illumina sequencing primers (underlined) to check for both on-target andoff-target cleavage as described in Zhou et al (2015) Indexed samples werepooled and sequenced as a spike-in on a NextSeq500 (Illumina) at the GeorgiaGenomics and Bioinformatics Core University of Georgia and the data wereanalyzed by the open-source software AGEseq (Xue and Tsai 2015) For char-acterization primary transformants in 1-gallon pots were grown to 2 m inheight before harvesting Developing xylem scrapings were collected at 15 to30 cm from the top and snap-frozen in liquid nitrogen for RNA and metaboliteanalyses The rest of the stemwas debarked and air-dried for wood chemical andphysical analyses (Supplemental Table S1) Vegetatively-propagated plants wereused in histological analysis

Histology

Stem cross sections (100 mm) were prepared from young shoots using aVibratome (Ted Pella) Sectionswere stainedwith 005 (wv) toluidine blue or2 (wv) phloroglucinol in acid ethanol (1 M HCl95 ethanol5 11 vv) andimages were acquired using a Zeiss Axioskop-50 microscope equipped with aLeica DFC500 digital camera

Wood Specific Gravity and Acoustic Velocity

Three8-mm longitudinal stemwood sections were obtained from the baseof each sample for nine control (three wild-type and six Cas9 lines) and 12 in-dependent mutant lines Specific gravity (wood density divided by the densityof water) was measured by water displacement after the sections were satu-rated with water (ASTM International Committee 2017) Acoustic velocity wasmeasured on the samples before water saturation using a time of flight in-strument (SoniSys) equipped with two 1-MHz transducers to send and receivethe acoustic signal The specific gravity and acoustic velocity values from thethree sections were averaged for each sample and statistical differences be-tween groups were determined by Studentrsquos t test

Wood Chemistry Saccharification and NMR Analysis

Thewood samples were ground to pass a 40-mesh screen using aWileymillfollowed by 99 cycles of ethanol extraction using a Soxhlet extraction unit(Buchi) and air-dried Extractive-free wood meal was used for Klason lignin(Swamy et al 2015) and crystalline cellulose content determination accordingto Updegraff (1969) Lignin composition was determined by thioacidolysis(Foster et al 2010a) matrix polysaccharide composition by trifluoroacetic acidhydrolysis (Foster et al 2010b) and glycosyl composition by methanolysisfollowed by trimethylsilylation (Harding et al 2018) High-throughput sac-charification assays were performed using the BESC method with hot waterpretreatment at 180degC for 40 min (Selig et al 2010) as well as the GLBRCmethods using hot water dilute acid (2 [vv] H2SO4) or dilute base (625 mM

NaOH) pretreatment at 90degC for 3 h (Santoro et al 2010) NMR analysis usingball-milled whole cell wall residues or enzyme lignin after cellulase digestionwas performed as detailed in Kim and Ralph (2010) and Mansfield et al(2012b) Unless otherwise noted these analyses were performed for fivecontrols (three wild-type and two vector controls) and five independent 4cl1mutants (Supplemental Table S1)

Glycome Profiling

Extractive-free wood meals were sequentially extracted with 50 mM am-monium oxalate 50 mM sodium carbonate 1 M KOH 4 M KOH acid chloriteand 4 M KOH to obtain cell wall fractions for screening with a panel of plantglycan-directed mAbs (Pattathil et al 2010) Statistical analysis of mutant ef-fects was performed with the program Limma (Smyth 2005) after excludingmAbs with hybridization intensities01 in all samples (three wild-type threevector controls and six independent 4cl1 mutants Supplemental Table S1)mAbs that showed significantly different signal intensities betweenmutant andcontrol samples (P 001 and fold-change $ 15) were visualized by heatmapsusing R software

Phenolic Profiling

Developing xylem scrapings were ground to a fine powder under liquidnitrogen aliquoted and stored at280degC One aliquot per sample was freeze-dried and 10 mg (dry weight) of tissue were extracted with 400 mL chlor-oformmethanol (11 vv) in an ultrasonic bath with pre-icendashchilled water for30 min followed by addition of 200 mL water vortexing and centrifugationfor phase separation The aqueous phase was transferred to a new tube andan aliquot of 100 mL was evaporated to dryness in a Centrivap (Labconco) forshipment to the Vlaams Instituut voor Biotechnologie The residues wereresuspended in 200 mL of cyclohexanewater (11 vv) and 15 mL were injectedonto an Acquity ultra performance liquid chromatography equipped with aSynapt Q-TOF (Waters) for chromatographic separation and mass spectrometricdetection of phenolicmetabolites following the conditions and settings detailed inSaleme et al (2017) A total of 10 control (three wild-type and seven Cas9) and 14independent 4cl1 samples were analyzed (Supplemental Table S1) Significantdifferences between the two groups were declared for peaks that had averagesignal intensities $1000 counts in at least one group with P 001 and fold-change $ 2 Compound annotation was based on matching mz retention timeand tandem mass spectroscopy fragmentation (Supplemental Fig S5) against anin-house library as well as literature data

RNA-Seq Analysis and Sample-Specific CoexpressionNetwork Construction

RNA was extracted from frozen xylem powder using the Direct-zol RNAMiniprep kit (Zymo) with PureLink Plant RNA Reagent (Life Technologies) forIllumina RNA-Seq library preparation and NextSeq 500 sequencing as de-scribed in Harding et al (2018) Six control (three wild-type and three Cas9)and six 4cl1 independent mutant samples (Supplemental Table S1) were se-quenced to obtain 83 to 121 million paired-end 75-bp reads per sampleexcept vector controlCas9-39 for which97M reads were obtained and only16-million randomly sampled reads were used for data analysis After pre-processing for quality control reads were mapped to the variant-substitutedPopulus tremula 3 alba genome v11 as detailed in Xue et al (2015) Genessatisfying the criteria of FPKM $ 5 in at least four out of six replicates of atleast one group average FPKM$ 10 in at least one group Q 005 and fold-change $ 15 were used for GO enrichment analysis by the software topGO




(Alexa and Rahnenfuhrer 2010) with Fisherrsquos exact test and the negativelog10-transformed P values were visualized as heatmaps For coexpressionnetwork analysis relaxed thresholds (P 005 FPKM $ 5) were used toobtain 5512 genes that differed between control and 4cl1 plants The averageexpression values of the 5512 genes were calculated for the control and 4cl1mutant groups We then adapted the approach of Liu et al (2016) for con-struction of genotype-specific networks Briefly a group of reference sampleswas assembled from our previous studies unrelated to lignin pathway genemanipulation (Swamy et al 2015 Xue et al 2016) Only unstressed xylemsamples were included (n 5 35) Expression values of the 5512 genes wereextracted from the 35 samples and the averaged expression values of controlor 4cl1 mutant samples were added separately to create two datasets (n 5 36each) for construction of the control- and mutant-specific networks Forsimplicity we refer to the former containing the control group (wild-type andCas9 lines) as the wild-type network Pairwise Spearman correlation coef-ficients of both datasets followed a normal distribution (Supplemental FigS4) and were used for weighted gene coexpression network analysis usingthe WGCNA R package (Langfelder and Horvath 2008) as detailed in Xueet al (2016) with a soft threshold of 10 The coexpression relationships wereassessed by hierarchical clustering using the topological overlap measureand modules were determined with a dynamic tree cutting algorithm(Supplemental Fig S4) The list of 5512 genes their expression differencesbetween control and 4cl1 mutants and sample-specific network moduleassignments are provided in Supplemental Dataset S1

Accession Numbers

The RNA-seq data has been deposited to the National Center for Bio-technology Information Sequence Read Archive under accession NoPRJNA589632

Supplemental Data

The following supplemental materials are available

Supplemental Figure S1 Class-I 4CL gene family members in Populus

Supplemental Figure S2 Glycome profiling analysis of all six cell wallfractions

Supplemental Figure S3 Characterization of 4cl5 KO mutants

Supplemental Figure S4 Genotype-specific coexpression networkanalysis

Supplemental Figure S5 Tandem mass spectroscopy spectra of metabo-lites listed in Table 2

Supplemental Table S1 List of mutant and control plant lines used inthis study

Supplemental Dataset S1 List of differentially expressed genes and theirnetwork module assignments

ACKNOWLEDGMENTS

The authors thank Gilles Pilate of The Institut National de la RechercheAgronomique France for providing poplar clone INRA 717-1B4 Yenfei Wang(University of Georgia) for assistance with transgenic plant production EliMcKinney (University of Georgia) for greenhouse plant care Nicholas Rohrand Bob Schmitz (University of Georgia) for Illumina RNA library constructionJacob Reeves (University of Georgia) for assistance with data processing andSequence Read Archive submission the Georgia Genomics and BioinformaticsCore for Illumina NextSeq sequencing the Complex Carbohydrate ResearchCenter Analytical Services and Training Laboratory for pyMBMS analysis andCliff Foster of the Great Lakes Bioenergy Research Center for cell wall analyses

Received December 20 2019 accepted February 26 2020 published March 62020

LITERATURE CITED

Alexa A Rahnenfuhrer J (2010) topGO Enrichment Analysis for GeneOntology R package version 20 wwwbioconductororg

ASTM International Subcommittee (2017) ASTM D2395-17 Standard testmethods for density and specific gravity (relative density) of wood andwood-based materials In Book of Standards Vol Vol 0410 ASTM In-ternational West Conshohocken PA

Barros J Escamilla-Trevino L Song L Rao X Serrani-Yarce JC PalaciosMD Engle N Choudhury FK Tschaplinski TJ Venables BJ et al(2019) 4-Coumarate 3-hydroxylase in the lignin biosynthesis pathway isa cytosolic ascorbate peroxidase Nat Commun 10 1994

Bewg WP Ci D Tsai C-J (2018) Genome editing in trees From multiplerepair pathways to long-term stability Front Plant Sci 9 1732

Boerjan W Ralph J Baucher M (2003) Lignin biosynthesis Annu Rev PlantBiol 54 519ndash546

Chanoca A de Vries L Boerjan W (2019) Lignin engineering in forest treesFront Plant Sci 10 912

Chen F Dixon RA (2007) Lignin modification improves fermentable sugaryields for biofuel production Nat Biotechnol 25 759ndash761

Chen H-C Song J Wang JP Lin Y-C Ducoste J Shuford CM Liu J Li QShi R Nepomuceno A et al (2014a) Systems biology of lignin biosyn-thesis in Populus trichocarpa Heteromeric 4-coumaric acidcoenzyme Aligase protein complex formation regulation and numerical modelingPlant Cell 26 876ndash893

Chen H-C Song J Williams CM Shuford CM Liu J Wang JP Li Q ShiR Gokce E Ducoste J et al (2013) Monolignol pathway 4-coumaricacidcoenzyme A ligases in Populus trichocarpa Novel specificity meta-bolic regulation and simulation of coenzyme A ligation fluxes PlantPhysiol 161 1501ndash1516

Chen H-Y Babst BA Nyamdari B Hu H Sykes R Davis MF HardingSA Tsai C-J (2014b) Ectopic expression of a loblolly pine class II 4-coumarateCoA ligase alters soluble phenylpropanoid metabolism butnot lignin biosynthesis in Populus Plant Cell Physiol 55 1669ndash1678

Coleman HD Park J-Y Nair R Chapple C Mansfield SD (2008) RNAi-mediated suppression of P-coumaroyl-CoA 39-hydroxylase in hybridpoplar impacts lignin deposition and soluble secondary metabolismProc Natl Acad Sci USA 105 4501ndash4506

Conti E Izaurralde E (2005) Nonsense-mediated mRNA decay Molecularinsights and mechanistic variations across species Curr Opin Cell Biol17 316ndash325

Dean JV Devarenne TP Lee IS Orlofsky LE (1995) Properties of a maizeglutathione S-transferase that conjugates coumaric acid and other phe-nylpropanoids Plant Physiol 108 985ndash994

Dixon DP Skipsey M Edwards R (2010) Roles for glutathione transferasesin plant secondary metabolism Phytochemistry 71 338ndash350

Ehlting J Buumlttner D Wang Q Douglas CJ Somssich IE Kombrink E(1999) Three 4-coumaratecoenzyme A ligases in Arabidopsis thalianarepresent two evolutionarily divergent classes in angiosperms Plant J19 9ndash20

Foster CE Martin TM Pauly M (2010a) Comprehensive compositionalanalysis of plant cell walls (lignocellulosic biomass) part I Lignin J VisExp 11 1745

Foster CE Martin TM Pauly M (2010b) Comprehensive compositionalanalysis of plant cell walls (lignocellulosic biomass) part II Carbohy-drates J Vis Exp 12 1837

Gao W Li H-Y Xiao S Chye M-L (2010) Acyl-CoA-binding protein 2 bindslysophospholipase 2 and lysoPC to promote tolerance to cadmium-induced oxidative stress in transgenic Arabidopsis Plant J 62 989ndash1003

Gui J Shen J Li L (2011) Functional characterization of evolutionarilydivergent 4-coumaratecoenzyme a ligases in rice Plant Physiol 157574ndash586

Harding SA Hu H Nyamdari B Xue L-J Naran R Tsai C-J (2018) Tu-bulins rhythms and cell walls in poplar leaves Itrsquos all in the timing TreePhysiol 38 397ndash408

Harding SA Leshkevich J Chiang VL Tsai CJ (2002) Differential sub-strate inhibition couples kinetically distinct 4-coumaratecoenzyme aligases with spatially distinct metabolic roles in quaking aspen PlantPhysiol 128 428ndash438

Hefer CA Mizrachi E Myburg AA Douglas CJ Mansfield SD (2015)Comparative interrogation of the developing xylem transcriptomes oftwo wood-forming species Populus trichocarpa and Eucalyptus grandisNew Phytol 206 1391ndash1405

Hoffmann L Besseau S Geoffroy P Ritzenthaler C Meyer D Lapierre CPollet B Legrand M (2004) Silencing of hydroxycinnamoyl-coenzyme Ashikimatequinate hydroxycinnamoyltransferase affects phenylpropanoidbiosynthesis Plant Cell 16 1446ndash1465


Tsai et al


Hoffmann L Maury S Martz F Geoffroy P Legrand M (2003) Purifica-tion cloning and properties of an acyltransferase controlling shikimateand quinate ester intermediates in phenylpropanoid metabolism J BiolChem 278 95ndash103

Honsel A Kojima M Haas R Frank W Sakakibara H Herschbach CRennenberg H (2012) Sulphur limitation and early sulphur deficiencyresponses in poplar Significance of gene expression metabolites andplant hormones J Exp Bot 63 1873ndash1893

Hu H Gu X Xue L-J Swamy PS Harding SA Tsai C-J (2016) TubulinC-terminal post-translational modifications do not occur in woodforming tissue of Populus Front Plant Sci 7 1493

Hu WJ Harding SA Lung J Popko JL Ralph J Stokke DD Tsai CJChiang VL (1999) Repression of lignin biosynthesis promotes celluloseaccumulation and growth in transgenic trees Nat Biotechnol 17 808ndash812

Hu WJ Kawaoka A Tsai C-J Lung J Osakabe K Ebinuma H Chiang VL(1998) Compartmentalized expression of two structurally and func-tionally distinct 4-coumarateCoA ligase genes in aspen (Populus trem-uloides) Proc Natl Acad Sci USA 95 5407ndash5412

Jacobs TB Martin GB (2016) High-throughput CRISPR vector constructionand characterization of DNA modifications by generation of tomatohairy roots J Vis Exp 2016 53843

Jez JM Ravilious GE Herrmann J (2016) Structural biology and regulationof the plant sulfation pathway Chem Biol Interact 259(Pt A) 31ndash38

Kajita S Katayama Y Omori S (1996) Alterations in the biosynthesis oflignin in transgenic plants with chimeric genes for 4-coumaratecoen-zyme A ligase Plant Cell Physiol 37 957ndash965

Kim H Padmakshan D Li Y Rencoret J Hatfield RD Ralph J (2017)Characterization and elimination of undesirable protein residues in plantcell wall materials for enhancing lignin analysis by solution-state nuclearmagnetic resonance spectroscopy Biomacromolecules 18 4184ndash4195

Kim H Ralph J (2010) Solution-state 2D NMR of ball-milled plant cell wallgels in DMSO-d6pyridine-d5 Org Biomol Chem 8 576ndash591

Knobloch KH Hahlbrock K (1975) Isoenzymes of p-coumarateCoA ligasefrom cell suspension cultures of Glycine max Eur J Biochem 52 311ndash320

Langfelder P Horvath S (2008) WGCNA An R package for weightedcorrelation network analysis BMC Bioinformatics 9 559

Lee D Meyer K Chapple C Douglas CJ (1997) Antisense suppression of 4-coumaratecoenzyme A ligase activity in Arabidopsis leads to alteredlignin subunit composition Plant Cell 9 1985ndash1998

Leple JC Brasileiro ACM Michel MF Delmotte F Jouanin L (1992)Transgenic poplars Expression of chimeric genes using four differentconstructs Plant Cell Rep 11 137ndash141

Li Y Kim JI Pysh L Chapple C (2015) Four isoforms of Arabidopsis4-coumarateCoA ligase have overlapping yet distinct roles in phenyl-propanoid metabolism Plant Physiol 169 2409ndash2421

Lin C-Y Wang JP Li Q Chen H-C Liu J Loziuk P Song J Williams CMuddiman DC Sederoff RR et al (2015) 4-Coumaroyl and caffeoylshikimic acids inhibit 4-coumaric acidcoenzyme A ligases and modulatemetabolic flux for 3-hydroxylation in monolignol biosynthesis of Populustrichocarpa Mol Plant 8 176ndash187

Lin J-S Huang X-X Li Q Cao Y Bao Y Meng X-F Li Y-J Fu C Hou B-K(2016) UDP-glycosyltransferase 72B1 catalyzes the glucose conjugationof monolignols and is essential for the normal cell wall lignification inArabidopsis thaliana Plant J 88 26ndash42

Lindermayr C Moumlllers B Fliegmann J Uhlmann A Lottspeich FMeimberg H Ebel J (2002) Divergent members of a soybean (Glycinemax L) 4-coumaratecoenzyme A ligase gene family Eur J Biochem 2691304ndash1315

Liu X Wang Y Ji H Aihara K Chen L (2016) Personalized characterizationof diseases using sample-specific networks Nucleic Acids Res 44 e164

Lu F Karlen SD Regner M Kim H Ralph SA Sun R-C Kuroda K-IAugustin MA Mawson R Sabarez H et al (2015) Naturally P-hy-droxybenzoylated lignins in palms BioEnergy Res 8 934ndash952

Lu S Li Q Wei H Chang M-J Tunlaya-Anukit S Kim H Liu J Song JSun Y-H Yuan L et al (2013) Ptr-miR397a is a negative regulator oflaccase genes affecting lignin content in Populus trichocarpa Proc NatlAcad Sci USA 110 10848ndash10853

Mansfield SD Kang K-Y Chapple C (2012a) Designed for deconstruc-tionmdashpoplar trees altered in cell wall lignification improve the efficacyof bioethanol production New Phytol 194 91ndash101

Mansfield SD Kim H Lu F Ralph J (2012b) Whole plant cell wall char-acterization using solution-state 2D NMR Nat Protoc 7 1579ndash1589

Molina-Rueda JJ Tsai C-J Kirby EG (2013) The Populus superoxide dis-mutase gene family and its responses to drought stress in transgenicpoplar overexpressing a pine cytosolic glutamine synthetase (GS1a)PLoS One 8 e56421

Nakashima J Chen F Jackson L Shadle G Dixon RA (2008) Multi-sitegenetic modification of monolignol biosynthesis in alfalfa (Medicagosativa) Effects on lignin composition in specific cell types New Phytol179 738ndash750

Park J-J Yoo CG Flanagan A Pu Y Debnath S Ge Y Ragauskas AJ WangZ-Y (2017) Defined tetra-allelic gene disruption of the 4-coumarateco-enzyme A ligase 1 (Pv4CL1) gene by CRISPRCas9 in switchgrass re-sults in lignin reduction and improved sugar release BiotechnolBiofuels 10 284

Pattathil S Avci U Baldwin D Swennes AG McGill JA Popper ZBootten T Albert A Davis RH Chennareddy C et al (2010) A com-prehensive toolkit of plant cell wall glycan-directed monoclonal anti-bodies Plant Physiol 153 514ndash525

Pattathil S Avci U Miller JS Hahn MG (2012) Immunological approachesto plant cell wall and biomass characterization Glycome profilingMethods Mol Biol 908 61ndash72

Ralph J Landucci LL (2010) NMR of lignins In C Heitner DR Dimmel andJA Schmidt eds Lignin and Lignans Advances in Chemistry CRCPress Boca Raton FL pp 137ndash234

Ralph J Lundquist K Brunow G Lu F Kim H Schatz PF Marita JMHatfield RD Ralph SA Christensen JH et al (2004) Lignins Naturalpolymers from oxidative coupling of 4-hydroxyphenyl-propanoidsPhytochem Rev 3 29ndash60

Ranocha P Chabannes M Chamayou S Danoun S Jauneau A BoudetAM Goffner D (2002) Laccase down-regulation causes alterations inphenolic metabolism and cell wall structure in poplar Plant Physiol 129145ndash155

Rey P Tarrago L (2018) Physiological roles of plant methionine sulfoxidereductases in redox homeostasis and signaling Antioxidants 7 114

Rouhier N Lemaire SD Jacquot J-P (2008) The role of glutathione inphotosynthetic organisms Emerging functions for glutaredoxins andglutathionylation Annu Rev Plant Biol 59 143ndash166

Ruprecht C Bartetzko MP Senf D Dallabernadina P Boos I AndersenMCF Kotake T Knox JP Hahn MG Clausen MH et al (2017) Asynthetic glycan microarray enables epitope mapping of plant cell wallglycan-directed antibodies Plant Physiol 175 1094ndash1104

Saballos A Sattler SE Sanchez E Foster TP Xin Z Kang C PedersenJF Vermerris W (2012) Brown midrib2 (Bmr2) encodes the major 4-coumaratecoenzyme A ligase involved in lignin biosynthesis in sor-ghum (Sorghum bicolor (L) Moench) Plant J 70 818ndash830

Saballos A Vermerris W Rivera L Ejeta G (2008) Allelic associationchemical characterization and saccharification properties of brownmidrib mutants of sorghum (Sorghum bicolor (L) Moench) BioEnergyRes 1 193ndash204

Saleme MLS Cesarino I Vargas L Kim H Vanholme R Goeminne Gvan Acker R Fonseca FCA Pallidis A Voorend W et al (2017) Si-lencing CAFFEOYL SHIKIMATE ESTERASE affects lignification andimproves saccharification in poplar Plant Physiol 175 1040ndash1057

Santoro N Cantu SL Tornqvist C-E Falbel TG Bolivar JL Patterson SEPauly M Walton JD (2010) A high-throughput platform for screeningmilligram quantities of plant biomass for lignocellulose digestibilityBioEnergy Res 3 93ndash102

Scheerer U Haensch R Mendel RR Kopriva S Rennenberg HHerschbach C (2010) Sulphur flux through the sulphate assimilationpathway is differently controlled by adenosine 59-phosphosulphate re-ductase under stress and in transgenic poplar plants overexpressinggamma-ECS SO or APR J Exp Bot 61 609ndash622

Schimleck L Dahlen J Apiolaza LA Downes G Emms G Evans RMoore J Pacircques L van den Bulcke J Wang X (2019) Non-destructiveevaluation techniques and what they tell us about wood property var-iation Forests 10 728

Schmidt D Schuhmacher F Geissner A Seeberger PH Pfrengle F (2015) Au-tomated synthesis of arabinoxylan-oligosaccharides enables characterizationof antibodies that recognize plant cell wall glycans Chemistry 21 5709ndash5713

Schoch G Goepfert S Morant M Hehn A Meyer D Ullmann P Werck-Reichhart D (2001) CYP98A3 from Arabidopsis thaliana is a 39-hydroxylase of phenolic esters a missing link in the phenylpropanoidpathway J Biol Chem 276 36566ndash36574




Sederoff RR MacKay JJ Ralph J Hatfield RD (1999) Unexpected varia-tion in lignin Curr Opin Plant Biol 2 145ndash152

Selig MJ Tucker MP Sykes RW Reichel KL Brunecky R Himmel MEDavis MF Decker SR (2010) Lignocellulose recalcitrance screening byintegrated high-throughput hydrothermal pretreatment and enzymaticsaccharification Ind Biotechnol (New Rochelle NY) 6 104ndash111

Smyth GK (2005) Limma Linear models for microarray data In R Gentleman VCarey S Dudoit R Irizarry and W Huber eds Bioinformatics and Com-putational Biology Solutions using R and Bioconductor Springer New Yorkpp 397ndash420

Studer MH DeMartini JD Davis MF Sykes RW Davison B Keller MTuskan GA Wyman CE (2011) Lignin content in natural Populus vari-ants affects sugar release Proc Natl Acad Sci USA 108 6300ndash6305

Sun L Varanasi P Yang F Loqueacute D Simmons BA Singh S (2012) Rapiddetermination of syringylguaiacyl ratios using FT-Raman spectroscopyBiotechnol Bioeng 109 647ndash656

Swamy PS Hu H Pattathil S Maloney VJ Xiao H Xue L-J Chung J-DJohnson VE Zhu Y Peter GF et al (2015) Tubulin perturbation leads tounexpected cell wall modifications and affects stomatal behaviour inPopulus J Exp Bot 66 6507ndash6518

Sykes RW Gjersing EL Foutz K Rottmann WH Kuhn SA Foster CEZiebell A Turner GB Decker SR Hinchee MAW et al (2015) Down-regulation of P-coumaroyl quinateshikimate 3ʹ-hydroxylase (C3ʹH) andcinnamate 4-hydroxylase (C4H) genes in the lignin biosynthetic path-way of Eucalyptus urophylla 3 E grandis leads to improved sugar releaseBiotechnol Biofuels 8 128

Takahashi H Kopriva S Giordano M Saito K Hell R (2011) Sulfur as-similation in photosynthetic organisms molecular functions and regu-lations of transporters and assimilatory enzymes Annu Rev Plant Biol62 157ndash184

Tsai C-J Harding SA Tschaplinski TJ Lindroth RL Yuan Y (2006)Genome-wide analysis of the structural genes regulating defense phe-nylpropanoid metabolism in Populus New Phytol 172 47ndash62

Updegraff DM (1969) Semimicro determination of cellulose in biologicalmaterials Anal Biochem 32 420ndash424

van Acker R Vanholme R Storme V Mortimer JC Dupree P Boerjan W(2013) Lignin biosynthesis perturbations affect secondary cell wallcomposition and saccharification yield in Arabidopsis thaliana BiotechnolBiofuels 6 46

Vanholme R de Meester B Ralph J Boerjan W (2019) Lignin biosynthesisand its integration into metabolism Curr Opin Biotechnol 56 230ndash239

Vanholme R Storme V Vanholme B Sundin L Christensen JH GoeminneG Halpin C Rohde A Morreel K Boerjan W (2012) A systems biology

view of responses to lignin biosynthesis perturbations in Arabidopsis PlantCell 24 3506ndash3529

Voelker SL Lachenbruch B Meinzer FC Jourdes M Ki C Patten AMDavin LB Lewis NG Tuskan GA Gunter L et al (2010) Antisensedown-regulation of 4CL expression alters lignification tree growth andsaccharification potential of field-grown poplar Plant Physiol 154874ndash886

Voytas DF Gao C (2014) Precision genome engineering and agricultureOpportunities and regulatory challenges PLoS Biol 12 e1001877

Wang JP Matthews ML Williams CM Shi R Yang C Tunlaya-Anukit SChen H-C Li Q Liu J Lin C-Y et al (2018) Improving wood propertiesfor wood utilization through multi-omics integration in lignin biosyn-thesis Nat Commun 9 1579

Weng J-K Chapple C (2010) The origin and evolution of lignin biosyn-thesis New Phytol 187 273ndash285

Wilkerson CG Mansfield SD Lu F Withers S Park J-Y Karlen SDGonzales-Vigil E Padmakshan D Unda F Rencoret J et al (2014)Monolignol ferulate transferase introduces chemically labile linkagesinto the lignin backbone Science 344 90ndash93

Xiong W Wu Z Liu Y Li Y Su K Bai Z Guo S Hu Z Zhang Z Bao Yet al (2019) Mutation of 4-coumarate coenzyme A ligase 1 gene affectslignin biosynthesis and increases the cell wall digestibility in maizebrown midrib5 mutants Biotechnol Biofuels 12 82

Xu B Escamilla-Trevintildeo LL Sathitsuksanoh N Shen Z Shen H ZhangYH Dixon RA Zhao B (2011) Silencing of 4-coumaratecoenzyme Aligase in switchgrass leads to reduced lignin content and improvedfermentable sugar yields for biofuel production New Phytol 192611ndash625

Xue L-J Alabady MS Mohebbi M Tsai C-J (2015) Exploiting genomevariation to improve next-generation sequencing data analysis and ge-nome editing efficiency in Populus tremula 3 alba 717-1B4 Tree GenetGenomes 11 82

Xue L-J Frost CJ Tsai C-J Harding SA (2016) Drought response tran-scriptomes are altered in poplar with reduced tonoplast sucrose trans-porter expression Sci Rep 6 33655

Xue L-J Tsai C-J (2015) AGEseq Analysis of genome editing by sequenc-ing Mol Plant 8 1428ndash1430

Ye Z-H Zhong R (2015) Molecular control of wood formation in trees J ExpBot 66 4119ndash4131

Zhou X Jacobs TB Xue L-J Harding SA Tsai C-J (2015) Exploiting SNPsfor biallelic CRISPR mutations in the outcrossing woody perennialPopulus reveals 4-coumarateCoA ligase specificity and redundancyNew Phytol 208 298ndash301


Tsai et al


tremula 3 alba INRA 717-1B4 (Hu et al 1998 1999Voelker et al 2010) The genome duplicate 4CL5(Potri003G188500) is the only other 4CL gene familymember expressed in lignifying xylem but its tran-script levels are much lower than those of 4CL1(Supplemental Fig S1 Hefer et al 2015 Swamy et al2015 Hu et al 2016 Xue et al 2016Wang et al 2018)4CL1 and 4CL5were proposed to encode 4CL proteinsthat form heterotetramers in a 31 ratio (referred to asPtr4CL3 and Ptr4CL5 in Chen et al [2013 2014a]) Theactivity of individual isoforms as well as the tetramericcomplex is sensitive to inhibition by hydroxycinnamicacids and their shikimate esters (Harding et al 2002Chen et al 2014a Lin et al 2015) Such complexity of 4CLcatalysis is consistent with multiple cellular strategies for

directing hydroxycinnamic acids toward lignin biosyn-thesis Furthermore silencing lignin-associated 4CLs canhavedifferent effects on syringyl-to-guaiacyl lignin (SG)ratio depending on the species ranging from increasesin tobacco (Nicotiana tabacum Kajita et al 1996) Arabi-dopsis (Arabidopsis thaliana Lee et al 1997) and switch-grass (Panicum virgatum Xu et al 2011) to little change inalfalfa (Medicago sativa Nakashima et al 2008) to de-creases in rice (Oryza sativa Gui et al 2011) Variableeffects on SG ratio have even been reported amongclosely related poplar species (Hu et al 1999 Voelkeret al 2010 Chanoca et al 2019 and references therein)Such variation has been attributed both to differing de-grees of 4CL downregulation as well as to 4CL multi-plicity (Boerjan et al 2003 Saballos et al 2012) Analysisof knockout (KO) mutants should eliminate the uncer-tainty caused by partial gene silencing

Arabidopsis sorghum (Sorghum bicolor) and maize(Zea mays) 4cl mutants showed preferential reductionsin G-lignin resulting in increased SG ratios (Saballoset al 2012 van Acker et al 2013 Li et al 2015 Xionget al 2019) Though rare until recently transgenic nullscan now be efficiently obtained for genetically lesstractable systems like woody perennials or polyploidsusing CRISPRCas9 technology (Voytas and Gao2014 Bewg et al 2018) We previously reported thatCRISPR-KO of the predominant lignin 4CL in poplarled to a reduced SG ratio (Zhou et al 2015) whereassimilar KO in tetraploid switchgrass increased the SGratio (Xu et al 2011) This study aims to furthercharacterize the poplar 4cl1 mutants by more com-prehensive cell wall analysis biomass saccharificationphenolic profiling and RNA sequencing (RNA-seq)We present data to show distinct effects of 4CL-KO onlignin biosynthesis and enzymatic hydrolysis of cellwalls relative to 4cl mutants of Arabidopsis and otherherbaceous species Our findings suggest that alteredcaffeic acid homeostasis along with altered expressionof key lignin biosynthetic genes cooperatively sustainproduction of G-enriched lignin via the minor 4CL5pathway in the Populus mutants

RESULTS

4CL1-KO Preferentially Reduces S-Lignin Biosynthesis

We reported previously that CRISPR editing of 4CL1in P tremula 3 alba INRA 717-1B4 (hereafter ldquo4cl1mutantsrdquo) led to uniformly discolored wood and a 23reduction in lignin content relative to wild-type andCas9-only transgenic controls (hereafter ldquocontrolsrdquo)based on pyrolysis molecular beam mass spectrometry(pyMBMS) analysis (Zhou et al 2015) Total lignincontent determined by the Klason method revealeda 19 reduction in transgenic wood (Table 1) ThepyMBMS analysis also detected a 30 decrease in theSG ratio (Zhou et al 2015) attributable to a sharp (34)reduction of S-units and a small (5) though significantreduction of G-units (Table 1) Thioacidolysis showed asimilar response in S-lignin but a greater reduction (18)

1This work was supported in part by the Institute of Food andAgriculture United States Department of Agriculture (grant no2015ndash67013ndash22812) the Division of Integrative Organismal Systems(grant nos IOSndash1546867 and IOSndash09223992) the Division of Biolog-ical Infrastructure (grant no DBIndash0421683) of the National ScienceFoundation the BioEnergy Science Center (grant no WPNndashERKP695)the Center for Bioenergy Innovation (grant no WPNndashERKP886) and theGreat Lakes Bioenergy Research Center (grant no DEndashSC0018409)funded by the Office of Biological and Environmental Research Officeof Science of the United States Department of Energy

2Senior author3Author for contact cjtsaiugaedu4Present address Department of Genetics and Genomics Sciences

Icahn School of Medicine at Mount Sinai New York 100295Present address Key Laboratory of Forest Genetics and Biotech-

nology Co-Innovation Center for Sustainable Forestry in SouthernChina College of Forestry Nanjing Forestry University NanjingJiangsu 210037 China

6Present address Department of Plant Biology Ecology and Ev-olution Oklahoma State University Stillwater Oklahoma 74078

7Present address State Key Laboratory of Subtropical SilvicultureZhejiang Agriculture amp Forestry University Hangzhou Zhejiang310058 China

8Present address Vlaams Instituut voor Biotechnologie Metabolo-mics Core UGent-VIB Research Building 9052 Ghent Belgium

9Present address Los Alamos National Laboratory Los AlamosNew Mexico 87545

10Present address Mascoma LLC (Lallemand Inc) Lebanon NewHampshire 03766

The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (wwwplantphysiolorg) isChung-Jui Tsai (cjtsaiugaedu)

XZ generated transgenic plants and measured Klason lignin PXperformed histology and amplicon sequencing JD measured spe-cific gravity and acoustic velocity HH SP and MGH performedglycome profiling analysis BN RN and SAH performed glyco-syl composition analysis SAH measured crystalline cellulose con-tent GG and WB performed phenolic profiling analysis EG andMFD performed saccharification analysis through the BioEnergyScience Center RG and JR performed nuclear magnetic resonanceanalysis and coordinated other cell wall analysis through the GreatLakes Bioenergy Research Center L-JX performed bioinformaticanalysis C-JT conceived the project analyzed data and wrote thearticle with PX and SAH with contributions from other authors

[OPEN]Articles can be viewed without a subscriptionwwwplantphysiolorgcgidoi101104pp1901550


Tsai et al



























Tsai et al



















Tsai et al








NameRetention


Fold

Change










DISCUSSION






Tsai et al
















CONCLUSIONS



Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al



























Tsai et al



















Tsai et al








NameRetention


Fold

Change










DISCUSSION






Tsai et al
















CONCLUSIONS



Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al



















Tsai et al








NameRetention


Fold

Change










DISCUSSION






Tsai et al
















CONCLUSIONS



Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al








NameRetention


Fold

Change










DISCUSSION






Tsai et al
















CONCLUSIONS



Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al



DISCUSSION






Tsai et al
















CONCLUSIONS



Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al
















CONCLUSIONS



Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al




Plant Materials


Histology







Glycome Profiling


Phenolic Profiling








Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al



Accession Numbers


Supplemental Data









ACKNOWLEDGMENTS



LITERATURE CITED

























Tsai et al










































































Tsai et al










































































Tsai et al


Compensatory Guaiacyl Lignin Biosynthesis at the Expense of Syringyl Lignin … · The lignin...

Documents

Transcript of Compensatory Guaiacyl Lignin Biosynthesis at the Expense of Syringyl Lignin … · The lignin...