Hemicellulose is a major component of primary plant cell walls. Many of the glycosyl residues found in hemicellulose are derived from the sugar precursor UDP-glucuronic acid, which can be con- verted into UDP-arabinose, UDP-apiose, UDP-galacturonic acid, and UDP-xylose. The enzyme controlling the biosynthesis of UDP- glucuronic acid, UDP-glucose dehydrogenase (EC 1.1.1.22), was cloned from soybean (Clycine max [L.] Merr.) by an antibody screening procedure. This enzyme is surprisingly homologous to the bovine sequence, which is the only other eukaryotic UDP-glucose dehydrogenase sequence known. The characteristic motifs of the enzyme, the catalytic center, a NAD-binding site, and two proline residues for main chain bends, are conserved within the prokaryotic and eukaryotic sequences. The soybean full-length cDNA clone encodes a protein of 480 amino acids with a predicted size of 52.9 kD. The enzyme is highly expressed in young roots, but lower expression levels were observed in expanding tissues of the epicotyl and in young leaves. The expression pattern of the enzyme in different developmental stages strengthens the argument that UDP- glucose dehydrogenase is a key regulator for the availability of hemicelluose precursors.
Growing plant cells continuously synthesize new plant cell components. The hemicellulose polysaccharides are synthesized by various enzymes in the Golgi apparatus (Carpita and Gibeaut, 1993; Driouich et al., 1993). The precursors for polysaccharides of the cell wall are the cor- responding nucleoside-diphosphate sugars, in particular the UDP-sugars (Feingold, 1982). Whereas cellulose com- ponents are synthesized from UDP-Glc, many of the hemi- cellulose sugars ( e g UDP-Ara, UDP-Api, UDP-GalUA, and UDP-Xyl) are derived from UDP-GlcUA as shown in the scheme in Figure 1. The uronic acid can be epimerized to UDP-GalUA, the most prominent precursor of pectic compounds. Decarboxylation of UDP-GlcUA leads to UDP-Xyl, which can be further epimerized to UDP-Ara (for review, see Bolwell, 1988; Feingold and Barber, 1990; Iiyama et al., 1991; Gibeaut and Carpita, 1994). Thus, UDP- GlcUA is a precursor for sugar nucleotides, which are needed for the biosynthesis of many components of hemi- cellulose, including arabinans, arabinogalactans, glucu- ronoarabinoxylans, rhamnogalacturonans, xylans, and xy- loglucans. Various biochemical studies have suggested that the production of UDP-GlcUA may be the rate-limiting
step in providing precursors for the expanding cell wall. Therefore, it was hypothesized that UDP-GlcDH, the en- zyme that converts UDP-Glc to UDP-GlcUA, is the key regulator in this pathway (Amino et al., 1985; Robertson et al., 1995). Nevertheless, the precise function of UDP-GlcDH in providing sugar nucleotide precursors for hemicellulose biosynthesis has not yet been elucidated.
The detailed analysis of UDP-GlcDH is difficult because the enzyme has not yet been purified from plant sources. Enzyme activity can be measured only in crude or partially purified extracts, with an important limitation being that competing reactions can degrade the substrate UDP-Glc at a high rate and the product undergoes further reactions. UDP-GlcDH is also strongly feedback-inhibited by UDP- Xyl (Davies and Dickinson, 1972), a metabolite derived from UDP-GlcUA by a decarboxylase. This biochemical regulation may distort the true enzyme activity in a crude in vitro assay.
The analysis of the role of UDP-GlcDH is further com- plicated by the existence of a second biosynthetic route for UDP-GlcUA called the inositol oxidation pathway (Loewus and Dickinson, 1982) (Fig. 1, right side). Within this path- way Glc-6-P is converted to GlcUA via inositol. Subse- quently, GlcUA is also activated to give UDP-GlcUA. To our knowledge it is unknown at present which route for the synthesis of UDP-GlcUA is the most important one in plants. Data for the inositol oxidation pathway were ob- tained by feeding radioactive precursors (Roberts and Loewus, 1973; Verma and Dougall, 1979). The rate-limiting step of this pathway is not known. There is evidence for the coexistence of both pathways with changing importance of either route during plant development (Rubery, 1972; Da- lessandro and Northcote, 1977a, 197%; Witt, 1992).
Here we report the cloning of UDP-GlcDH from soybean (Glycitze max L.). The enzyme is highly expressed in various tissues but is almost absent in others. Most interestingly, the enzyme structure of UDP-GlcDH is highly conserved between plants and animals, although the product of the reaction is used for totally different polymers and further reactions. In animals UDP-GlcUA is used for proteogly- cans, dermatan sulfate, heparan sulfate, and heparin (Hempel et al., 1994). Animals also use UDP-GlcUA for the detoxification of xenobiotics by glucuronosylation, which
Abbreviations: EST-clone, expressed sequence tags from the dbest database of GenBank; UDP-Api, UDP-apiose; UDP-GlcDH, UDP-Glc dehydrogenase (EC 1.1.1.22).
Figure 1. Scheme of the pathways for hemicellulose precursors. UDP-GlcUA is produced by oxidation of UDP-Glc through the en- zyme UDP-GlcDH (left side). An alternative route for the production of UDP-GlcUA via inositol (right side) can be found in seedlings (see text for details).
is not well established in plants. However, Schulz and Weissenbock (1988) reported about transferases from rye using UDP-GlcUA for the glucuronosylation of plant flavones.
MATERIALS A N D METHODS
Construction of the c D N A Library
Total RNA was isolated from mid-log-phase soybean (Glycine max L.) cell suspension cultures by the method of Chomczynski and Sacchi (1987) and mRNA was further purified from total RNA by the PolyAtract system (Pro- mega). A cDNA library was synthesized from 5 pg of mRNA using the A-uni-Zap kit (Stratagene). About 30% of the ligated phage cDNA was packaged into A phages with the Gigagold system (Stratagene), yielding more than 107 plaque-forming units in the primary library. Half of this library was amplified once on large Petri dishes and used for this study.
Screening of the cDNA, Sequencing, and Computer Analysis
About 5 X 105 plaque-forming units were screened with a rabbit polyclonal antiserum assumed to be directed against p47phox (obtained from A. Cross, The Scripps Clinic Research Foundation, La Jolla, CA). Screening was carried out using standard procedures (Ausubel et al., 1995). In brief, 30,000 plaque-forming units per 145-mm dish were grown for 4 h at 42°C. Plates were overlaid with a nitrocellulose filter, impregnated with 0.2 M isopropyl- thio-P-galactoside, and further incubated for 5 h at 37°C. The nitrocellulose filters were removed and washed exten- sively in TBS + Tween 20 to remove cell debris. After
blocking the membrane in TBS + Tween 20 + 3% BSA, the filters were incubated with the p47phox antiserum in a 1:3000 dilution for 1 h and washed with TBS + Tween 20 four times for 5 min. Primary antibodies were visualized by incubation with a secondary antibody, conjugated with alkaline phosphatase (Bio-Rad), nitroblue tetrazolium, and X-phosphate staining typically for 15 to 30 min. Phages were purified to homogeneity using standard procedures. The cDNA inserts of the isolated A phages were subcloned into pBluescript (Stratagene) by the in vivo excision method.
Ten individual clones were identified by the antiserum. Hybridization experiments and sequencing proved that a11 clones were derived from a single gene. The insert of the longest clone was sequenced by the dideoxy-chain termi- nation method using the Sequenase system (Amersham). Both strands were sequenced from various subclones using standard primers and synthesized oligonucleotide primers.
Analysis of the cDNA sequence was performed with the Blast program tool (Altschul et al., 1990) and on a local computer system using the Clone and Align programs (Scientific and Educational Software, State Line, PA).
Expression of the c D N A in Escherichia coli and the Generation of N e w Antibodies
A full-length cDNA insert was cloned into the pQE31 expression vector (Qiagen, Chatsworth, CA) and trans- formed into Escherichia coli XL-1. Expression of the His- tagged fusion protein was carried out in 100-mL scale by induction of the bacteria with 1 mM isopropylthio-p-galac- toside for 5 h at 30°C. The fusion protein was purified under denaturing conditions on nitrilotriacetic acid-agar- ose according to the Qiagen protocol. After extensive dial- ysis of the purified protein against 10 mM sodium phos- phate buffer (pH 7.4), the enzyme was used to immunize two rabbits (white New Zealand, Thomae Pharma, Biber- ach, Germany). Antibodies were collected from a bleeding rabbit 11 d after the second boost injection (750 k g of protein per animal and boost). IgGs were purified from the antiserum using a ProteinA column (Hitrap, Pharmacia) and used throughout this study.
Northern Blot Analysis
Total RNA (10 pg per lane) was separated on a 1.1% formaldehyde agarose gel and transferred onto a nylon membrane (Hybond Nt, Amersham) via downward cap- illary blotting. Hybridization was carried out according to the protocol of Church and Gilbert (1984) with a random primed probe (Ready To Go system, Pharmacia) of the soybean cDNA for UDP-GlcDH. The blot was rehybridized with H1, a cDNA probe from bean, believed to be consti- tutively expressed (Wingate et al., 1988).
Southern Blot Analysis
Genomic DNA from soybean and Arabidopsis was iso- lated as described by Taylor et al. (1993). Restricted DNA was separated on a 0.7% agarose gel, transferred to a nylon membrane (Hybond Nt), and hybridized with a 32P-
labeled probe at 65°C for stringent conditions or at 58°C forheterologous hybridization using the buffer system ofChurch and Gilbert (1984).
ImmunoprecipitationSoybean cell-suspension cultures were homogenized in
50 mM potassium-phosphate buffer (pH 7.5) containing 2HIM EDTA, 5 mM DTT, 0.5 mM PMSF, and 0.5% (w/v)polyvinylpolypyrrolidone. The protein extract was frac-tionated by (NH4)2SO4 precipitation. The pellet corre-sponding to the 25 to 50% saturation fraction was desaltedon a PD10 column (Pharmacia) and used for the immuno-precipitation experiment.
The IgG fraction was diluted from 1:80 to 1:1200 into theprotein extract and incubated on ice for 3 h. Precipitatedprotein was collected by 15 min of centrifugation at 4°C.The pellet was washed once, resuspended in assay buffer,and used for the enzyme assay.
Enzyme Assay
UDP-GlcDH was measured spectroscopically in a 1-mLassay adapted from Roberts and Cetorelli (1973) with somemodifications. In brief, the assay consists of 20 mM Tris-Cl(pH 8), 1 mM EDTA, 1 mM NAD+, 0.4 mM UDP-Glc, and 1mM NaN3. The reduction of NADH was measured as anincrease at 340 nm. Controls were performed withoutNAD+ or UDP-Glc, showing a slight decrease at 340 nm,and then were subtracted.
BC S M
RESULTS
Isolation of the GeneThe gene for UDP-GlcDH was isolated during our work
on the oxidative burst in plant-pathogen interactions. It isthought that reactive oxygen species are formed via aplasma-membrane-bound NAD(P)H-oxidase with featuressimilar to those of the NADPH-oxidase from neutrophilcells of the immune system (Mehdy, 1994; Tenhaken et al.,1995). Using an antibody assumed to be directed againstthe p47phox subunit of the mammalian NADPH-oxidase(p47phox; Lomax et al., 1989) we identified a single proteinband of about 50 kD in soybean protein extracts (Fig. 2). Asexpected, the protein was not associated with membranes(Fig. 2A). Since the size of the protein recognized by theantibody was almost identical in human and soybean pro-tein extracts, we decided to clone the respective gene froma cDNA expression library. A library was constructed fromsoybean cell culture mRNA. Poly(A)+ RNA was isolatedfrom 4-d-old cell culture (mid-log-phase) and used for thesynthesis of a directional cDNA expression library. Of 5 X105 plaques screened with the p47phox antiserum, 10 im-mune positive phages were identified and further purified.All of the positive clones cross-hybridized under stringentconditions, indicating the same cDNA for all identifiedclones.
The longest cDNA insert was sequenced and it containeda putative full-length clone of 1.7 kb. The open readingframe of 1440 bases encodes a protein of 480 amino acids
Mr(kD)
68-
45-
29-
Figure 2. Western blot with different antibodies that bind to thesoybean UDP-ClcDH. Protein extracts from soybean were separatedby 10% SDS-PACE and transferred to a PVDF membrane. A, Crudeextract (C), soluble proteins (S), and microsomal proteins (M) wereanalyzed with a presumed anti-p47phox antibody (see text for de-tails). B, Crude extract was analyzed with an anti-UDP-ClcDH anti-body directed against the expressed soybean UDP-GlcDH fusionprotein in E. coli. Both antibodies bind to the same protein.
(Fig. 3), and the predicted molecular mass is 52.9 kD. Thissize is very close to the observed molecular mass of theprotein in western blots (Fig. 2).
Sequence Analysis
A database search of the amino acid sequence revealedno homology to the p47phox protein from the neutrophilsuperoxide-generating oxidase, but it did show a strikinglyhigh identity to the bovine enzyme UDP-GlcDH (Fig. 4). Toclarify the surprising result, we used other anti-p47phoxsera from the same rabbit (taken on a different day) as wellas anti-p47phox antibodies from a different rabbit. None ofthem detected the 50-kD band in soybean protein extracts(data not shown). Therefore, it was clear that we have notcloned the p47phox-homolog from the soybean NAD(P)H-oxidase, but that it is likely a plant UDP-GlcDH.
Analysis of the soybean sequence for UDP-GlcDH re-vealed that all of the biochemically identified structuralfeatures of the bovine enzyme are present (Fig. 3). Theconserved sequence motifs include the NAD-binding site,the catalytic Cys residue, and two Pro residues, whichrepresent the main chain bends in the protein structure(Hempel et al., 1994). A computer search of the dbestdatabase of GenBank identified several unknown Arabidop-sis thaliana EST-clones with sequences almost identical tothose of the soybean protein. These sequences were assem-bled using the soybean protein sequence for UDP-GlcDHas a guideline. Using this computer approach more than90% of the Arabidopsis sequence could be predicted. The 3'end of the sequence was obtained by sequencing two of theEST-clones. Part of the 5' sequence of the Arabidopsis genewas confirmed from sequencing a genomic clone for UDP-GlcDH that we have isolated by a heterologous screen of a www.plantphysiol.orgon May 16, 2020 - Published by Downloaded from
GCACGAGGCAAAATGGTGAAGATTTGCTGCATTGGTGCTGGATATGTGGGGGGTCCTACT M V K I C C I 1 Q A O Y V O Q l P T
ATGGCAGTCATTGCACTTAAGTGCCCATCCATTGAAGTCGCTGTTGTTGATATCTCTAAA M A V I A L K C P S I E V A V V D I S K TCCCGCATTGCAGCCTGGAACAGCGACCAGCTTCCTATCTATGAACCTGGCCTTGATGGT S R I A A W N S D Q L P I Y E P G L D G GTTGTGAAGCAATGCCGTGGCAAGAACCTCTTCTTCAGCACTGATGTTGAAAAGCATGTC V V K Q C R G K N L F F S T D V E K H V TTTGAGGCTGACATAGTGTTTGTCTCTGTCAACACCCCAACCCTCAGGGTCTTGGA F E A D I V F V S V N T H T K T Q G L G GCTGGCAAAGCAGCAGATTTGACATATTGGGAGAGTGCAGCTCGCATGATTGCTGATGTC A G K A A D L T Y W E S A A R M I A D V TCGAAGTCTGACAAGATCGTGGTGGAGAAATCCACAGTCCACAGTCCCTGTCAAAACTGCTGAAGCC S K S D K I V V E K S T V P V K T A E A ATAGAGAAGATTCTGACCCACAATAGCAAGGGAATC~TTCCAGATTCTATC~CCCT I E K I L T H N S K G I K F Q I L S N H G A A T T C C T T G C A G A G G G A A C T G C A A T C T C T T T A C C G T G T T C T T A T T E F L A E G T A I K D L F N P D R V L I GGAGGCAGGGAAACCCCAGAGGGCC~GCTATTCAAATCCACAGCATTGAAGGATGTTTATGCT G G R E T P E G Q K A I Q T L K D V Y A CAATGGGTCCCCGAAGAAAGAATACTGACCACAAATCTTTGGTCGGCAGAACTGTCTAAG Q W V P E E R I L T T N L W S A E L S K CTTGCTGCTAATGCCTTCTTAGCACAAAGGATTTCATCTGTCAATGCCATGTCAGCACTT L A A N A F L A Q R I S S V N A M S A L TGTGAGGCTACTGGGGCAAACGTTCAACAGGTGTGTCCTATTCTGTTGGTACAGACTCAAGG C E A T G A N V Q Q V S Y S V G T D S R ATTGGACCCAAGTTTCTCAATGCCAGTGTTGGTTTTGGTGGATCCTGCTTCCAGAAGGAT I G P K F L N A S V Q F O Q S ~ F O K D ATCTTGAACCTTGTTTACATCTGTGAGTGCAATGGCCTTCCAGAGGTGGCTGAGTATTGG I L N L V Y I C E C N G L P E V A E Y W AAACAAGTGATCAAGATCAATGATTATCAGAAGAGCCGATTTGTGAACCGTGTTGTTGCA K Q V I K I N D Y Q K S R F V N R V V A TCAATGTTCAACACAGTTTCAAACAAAAAGATTGCTATTCTGGGATTTGCCTTCAAGAAA S M F N T V S N K K I A I L G F A F K K GACACTGGTGACACAAGGGAGACTCCTGCCATTGATGTATGCCAGGGGCTACTAGGTGAT D T G D T R E T P A I D V C Q G L L G D AAGGCCAACCTGAGCATATACGACCCGCAAGTAACCGAGGACCAAATCCACAGTCCAGAGGGATCTA K A N L S I Y D P Q V T E D Q I Q R D L TCCATGAACAAGTTTGATTGGGATCATCCTATCCACTTGCAGCCCACAAGTCCTACTACT S M N K F D W D H P I H L Q P T S P T T GTGAAGAAGGTCAGTGTTGTTTGGGATGCCTATGCCTATG~GCAACAAAGGATGCACATGGCCTT V K K V S V V W D A Y E A T K D A H G L TGCATTCTAACCGAGTGGGATGAGTTCAAGACTCTTGATTACCAGAAGATATTTGACAAC C I L T E W D E F K T L D Y Q K I F D N ATGCAAAAACCAGCATTTGTTTTTGATGGCAGAAACATTGTGGATGCTGATAAGTTGCGT M Q K P A F V F D G R N I V D A D K L R GAGATTGGCTTCATAGTTTACTCAATTGGTAAGCCACTGGACCCATGGCTCAAAGACATG E I G F I V Y S I G K P L D P W L K D M CCTGCTGTGGCATAAAATAGATCATTGATCAAGTACAATGCAATATCAGTCAAGTGTTGA P A V A * GGCAGTCAGTTTGATTCTTTATTATTTTTTGTAAGAGTTTTTCTTTTCCTCTTTAACGTT CTCATAATAGTTGACATAATTGGGGGTATAAATTGTGGGAACTGGGGAGGGTTTATGTTCT ATGTTGTTTTGTACTCCAATCTTTTATGAATGTACTGGGG~CTTGTTTTGTTTGATTA A T C T T C T C T T A A T A T C P
Figure 3. Sequence of the UDP-GlcDH from soybean. This clone was identified from an expression library, so a few base pairs of the 5’ untranslated promoter region are probably missing. The sequence encodes a protein with 480 amino acids with a predicted size of 52.9 kD. The stop codon is indicated by the asterisk at bp 1453. The NAD-cofactor binding site (amino acid positions 8-14) is boxed and shaded in gray. The catalytic site (amino acid positions 267-276) with a centered Cys residue (amino acid position 272) is underlined. Pro at amino acid positions 89 and 156 are printed in black boxes and believed to present main chain bends in the protein structure. These features are 100% conserved within the bovine sequence and also within prokaryotic sugar nucleotide dehydrogenase (see Hempel et al. [1994] for details).
genomic Arabidopsis library with the soybean UDP-GlcDH cDNA as a probe (data not shown).
For further analysis of the enzyme we generated anti- bodies against the soybean UDP-GlcDH. The enzyme was expressed as a His-tagged fusion protein in E. coli and used as an antigen after purification of the fusion protein. Two rabbits were immunized and both showed specific high- titer antibodies after two boost injections. These anti-UDP- GlcDH antibodies recognized the soybean protein initially identified with the p47phox antibody (Fig. 2B).
lmmunoprecipitation and Enzyme Assay
A crude protein extract corresponding to the protein precipitated by 25 to 50% (NH,),SO, saturation shows readily measurable activity of UDP-GlcDH (data not shown). The enzyme assay was linear over at least 1 h. Controls lacking UDP-Glc as a substrate did not show any increase of reduced NADH. A boiled enzyme control was also totally inactive.
The antibodies were used to immunoprecipitate UDP- GlcDH from the crude protein extracf. A seria1 dilution of the IgG fraction was added to the protein extract from soybean and incubated on ice for severa1 hours. After centrifugation enzyme activity for UDP-GlcDH was mea- sured in the supernatant and in the resuspended pellet. The enzyme activity can be completely precipitated by the polyclonal antibody. A fraction of this activity is measur- able in the pellet. The enzyme in the precipitate is partially inhibited by the antibodies, presumably by direct interac- tion with epitopes significant for enzymatic activity (Fig. 5 ) . The immunoprecipitation of UDP-GlcDH by the anti- body, together with the high sequence homology to bovine UDP-GlcDH, clearly proves that the cloned gene encodes the soybean UDP-GlcDH. So far, we have been unable to demonstrate enzyme activity of the UDP-GlcDH fusion protein expressed in E. coli. The fusion protein could only be purified under denaturing conditions. Attempts to re- nature the enzyme have as yet been unsuccessful.
soy 4 0 A r n 4 0 Bov 5 1
soy 9 0 Ara 9 0 BOV 1 0 1
Soy 1 4 7 A r n 1 4 7 BOV 1 5 1
Soy 1 9 7 A r n 1 9 7 BOV 2 0 0
soy 2 4 7 Ara 2 4 7 BOV a s a
Soy 2 9 7 Ara 2 9 7 Bov 3 0 0
soy 3 4 7 A r n 3 4 7 Bov 3 5 0
soy 3 9 7 Ara 3 9 7 Eov 3 9 1
SK
Figure 4. Alignment of UDP-ClcDH from soybean, Arabidopsis, and bovine liver. The amino acid sequence from Arabidopsis was assem- bled from various ESTs available in the dbest database of GenBank. The 3’ end of the Arabidopsis gene was obtained by sequencing two EST-clones. The bovine sequence is from Hempel et al. (1994). ldentical sequences are white in black boxes, homolog exchanges are shaded in gray.
Figure 5. Immunoprecipitation of UDP-GlcDH from soybean. Anti-serum directed against the soybean UDP-ClcDH was added to acrude protein fraction (200 .̂L of a [NH4]2SO4 precipitation [25-50% saturation]). The enzyme activity is immunoprecipitated by theantibodies (O). The remaining activity in the supernatant (D) de-creases after antibody addition. Total enzyme is partially inhibited bythe polyclonal antibody, presumably due to binding of antibodies toepitopes significant for enzymatic activity.
Expression of the Enzyme during Plant Development
The expression of UDP-GlcDH was analyzed at themRNA level by northern hybridization. Total RNA wasseparated on an agarose gel and transferred to a nylonmembrane. Northern hybridization with a 32P-labeledprobe showed a high expression in root tips and lateralroots and a moderate expression in the epicotyl and inexpanding leaves (Fig. 6). In contrast, the expression in the
12 10 11
1 2 345 6 78 9 10 11 12Figure 6. Expression of UDP-GlcDH in soybean seedlings andplants. Total RNA (10 /ig) from different developmental stages of theplant was separated on a denaturing agarose gel, transferred to anylon membrane, and hybridized with a J2P-labeled cDNA probe forUDP-GlcDH. RNA was prepared from plant organs as indicated indrawings C to E. C represents a 2-d-old seedling; D represents a7-d-old plant; and E shows a 2-week-old plant. The expression ofmRNA for UDP-GlcDH is shown in A. The blot was reprobed withHI for loading (B).
upper part of the main root, in the hypocotyl, and inmature leaves was much lower.
Genome Structure
To test whether UDP-GlcDH is encoded by a single copygene or by a small gene family, we hybridized restrictedgenomic DNA from soybean with the cDNA clone. Understringent conditions (65°C) only two or three fragmentswere labeled, indicating a single-copy gene (Fig. 7A). Aninternal EcoRI site was present in the cDNA clone at posi-tion 481 so that at least two labeled fragments were ex-pected. Hybridization under less stringent conditions (58°C)also detected a few other bands of lower intensity (Fig. 7B),which indicates that related dehydrogenases with differentsubstrate specificity might be labeled by the UDP-GlcDHclone. We performed the same experiment using genomicDNA from Arabidopsis, which has a less complex genome.Only a single band of XM-restricted Arabidopsis DNA hy-bridized to the soybean cDNA clone (Fig. 7C). An internalEcoRI site was present in the Arabidopsis gene, and at least inArabidopsis, UDP-GlcDH is a single-copy gene. These resultswere also confirmed by the analysis of more than 10 differentArabidopsis EST-clones, which all belong to the same genebased on sequence identity (Fig. 4).
DISCUSSION
Many of the important precursors for hemicellulose bio-synthesis are formed via UDP-GlcUA, which is subse-quently converted into UDP derivatives of GalUA, Xyl, andAra for pectic polymers and hemicellulose. Regulation ofthe deposition of these diverse polysaccharides is thoughtto be controlled by various synthases in the Golgi appara-tus (Carpita and Gibeaut, 1993; Driouich et al., 1993). Thesesynthases use different UDP-sugars as substrates and the
19S mm ¥•w- p.
kBp
—23.1
— 6.5
— 4.3
— 2.3
— 0.56
Figure 7. Cenomic Southern blot with soybean and ArabidopsisDMA, restricted with FcoRI (E) or Xba\ (X). The membrane washybridized with the !2P-labeled soybean cDNA probe. A, SoybeanDNA (high stringency, 65°C); B, soybean DNA (lower stringency,58°C); and C, Arabidopsis DNA (lower stringency, 58°C). The sizesof the fragments are indicated in kilobase pairs (kBp). www.plantphysiol.orgon May 16, 2020 - Published by Downloaded from
1132 Tenhaken and Thulke Plant Physiol. Vol. 112, 1996
maintenance of UDP-sugar pools is a likely prerequisite for normal cell-wall biosynthesis.
The enzyme UDP-GlcDH, which oxidizes UDP-Glc to the corresponding uronic acid and thereby controls the pool of the common precursor UDP-GlcUA, was cloned for the first time from a plant species. The sequence of the soybean gene is highly homologous to the only other known eu- karyotic sequence from bovine liver (Fig. 4) (Hempel et al., 1994). Both sequences are identical to 61% and homolo- gous, including conserved amino acid exchanges, to more than 77%. Since the enzyme was initially identified as UDP-GlcDH based on this high homology to the bovine sequence, we have confirmed the identity by biochemical means. The cDNA was expressed as a His-tagged fusion protein. Antibodies directed against the purified fusion protein recognize a single band in soybean protein extracts of the predicted size of approximately 50 to 52 kD (Fig. 2B). The antibody is able to specifically inhibit UDP-GlcDH enzyme activity (Fig. 5). In addition, UDP-GlcDH can be immunoprecipitated by the antibody (Fig. 5 ) . Taken to- gether, all the data prove that the cloned soybean gene is
The UDP-GlcDH sequence contains a cofactor binding site for NAD (Fig. 3, amino acid positions 8-14), which is identical to the bovine sequence (Hempel et al., 1994) and is also conserved in other NAD-linked dehydrogenases. The catalytic center of the bovine enzyme was initially identified by chemical modifications of amino acids and subsequent peptide sequencing (Franzen et al., 1981). It contains a Cys residue at position 275 (amino acid position 272 in soybean; Fig. 3). The surrounding amino acids are strictly conserved in prokaryotic and eukaryotic sequences of sugar-nucleotide dehydrogenases. Two Pro residues (amino acid positions 92 and 159 in bovine liver, amino acid positions 89 and 156 in soybean; Fig. 3) that are thought to present turns in the structure of the protein main chain are also maintained in both sequences. Since various motifs of the enzymes are absolutely conserved, one can expect a very similar structure of the plant and the animal enzyme for UDP-GlcDH.
Using the dbest database of GenBank (Newman et al., 1994) we were able to assemble the nearly complete Ara- bidopsis sequence of UDP-GlcDH. The 3’ end of the gene was obtained with identical results by sequencing two different EST-clones. All of the available homologous se- quences belong to the same gene, indicating a single-copy gene for UDP-GlcDH in this weed. The Southern blot data with soybean DNA (Fig. 7A) as well as with Arabidopsis DNA (Fig. 7C) also point to a single-copy gene.
UDP-GlcUA can be synthesized in plants via two differ- ent pathways (see Fig. 1). The simple route is the direct conversion of the ubiquitous UDP-Glc to the uronic acid by the enzyme described in this paper. Alternatively, UDP- GlcUA can also be formed within the inositol oxidation pathway (Loewus et al., 1973). This route involves severa1 enzymatic steps. There is clear evidence for the existence of both pathways in plants but the relative contribution of both routes to the UDP-GlcUA pool is largely unknown. In cambium cells from sycamore, UDP-GlcDH rather than the
,
UDP-GlcDH.
inositol pathway seems to contribute the major portion to the UDP-GlcUA pool (Dalessandro and Northcote, 1977b). In germinating Liliunz longiflorum pollen tubes, Maiti and Loewus (1978) were able to demonstrate the flow of Glc into hemicellulose via inositol. Nevertheless, the same preparation contains significant amounts of UDP-GlcDH, since this tissue was used to partially purify the enzyme. A 12-fold enrichment of specific enzyme activity for UDP- GlcDH was obtained by Davies and Dickinson (1972).
In contrast, Roberts and Cetorelli (1973) were unable to measure significant activity of UDP-GlcDH in various monocotyledons and dicotyledons and concluded that the inositol oxidation pathway must be the major route for hemicellulose sugars. The incorporation of [14C]G1c into detached root tips from corn was reduced by simultaneous addition of inositol (Roberts and Loewus, 1973). The au- thors concluded that part of the UDP-GlcUA must be syn- thesized via inositol, as additional inositol diluted the I4C label in the hemicellulose fraction. One can speculate that the inositol oxidation pathway is predominantly active in seedlings to metabolize the liberated inositol from phytic acid, a common storage compound in seeds. Taken to- gether, most data are conflicting in some points and the application of radioactive precursors changes pool sizes of metabolites and should be interpreted with caution.
Cloning of UDP-GlcDH provides a new tool with which to dissect the expression pattern of the two pathways. From the present study it becomes clear that at least during particular steps of the seedling’s developmental program, UDP-GlcDH is highly expressed. This implies that the en- zyme plays an important role for providing hemicellulose precursors in roots and expanding leaves. The low expres- sion leve1 in other parts of soybean plants (upper root, hypocotyl, and leaves) can be explained by the lack of demand for UDP-GlcUA-derived sugars in differentiated mature cells. Alternatively, the inositol oxidation pathway takes over in a11 other tissues. Although we cannot rule out the latter possibility, this explanation is somehow unlikely. Enzyme activity of UDP-GlcDH is strictly correlated with growing and expanding tissues that have a demand for hemicellulose precursors. At the moment we do not know if UDP-GlcDH is expressed in all types of cells; a careful examination of this question is underway. In addition, a probe for the inositol oxidation pathway recently became available when the gene for inositol-1-phosphate phos- phatase was cloned from tomato (Gillaspy et al., 1995). The enzyme exists in three different isoforms, and at least one of them is involved in the generation of inositol phosphates as second messengers. Another isoform is believed to be a part of the inositol oxidation pathway leading to UDP- GlcUA. By using gene-specific probes for individual iso- forms of inositol-1-phosphate phosphatase, it should be possible to analyze the expression for genes in both path- ways that fill the UDP-GlcUA pool. This will definitely clarify if both routes are inversely expressed or if they contribute jointly to the pool of hemicellulose precursors.
As far as we know, UDP-GlcDH has not been fully purified from any plant species, with the possible excep- tion of an enzyme preparation from French bean contain-
ing UDP-GlcDH activity, which was recently largely pu- rified (Robertson et al., 1996). Those authors describe the co-purification of UDP-GlcDH and alcohol dehydroge- nase. Severa1 chromatographic procedures were unable to separate the two enzyme activities. An antibody raised against a partially purified enzyme fraction, how- ever, recognized a 40-kD band in bean protein extracts. The predicted size of UDP-GlcDH in soybean is 52 kD and is identical with the size of the bovine enzyme. The sequences from soybean and Arabidopsis are conserved to more than 95% (Fig. 4).
In addition, our antibody raised against UDP-GlcDH cross-reacts with a 52-kD band in bean plants as expected (data not shown). Therefore, it is highly unlikely that the characterized protein(s) from bean is UDP-GlcDH, espe- cially since soybean and bean plants are closely related. The reported K , value for UDP-Glc for the bean enzyme is 5.5 mM (Robertson et al., 1996). In contrast, the soybean enzyme has a K , of about 0.2 mM, which agrees with a report from Davies and Dickinson (1972) in which they estimated the K , for the Lilium enzyme to be 0.3 mM. The pool size of UDP-Glc in soybean cell cultures is about 0.1 mM (Hayashi and Matzuda, 1981).
Robertson et al. (1996) emphasized the similarity of their partially purified enzyme fraction to alcohol dehy- drogenase. Since this enzyme is cloned from various species, we compared the molecular mass and the amino acid sequence of alcohol dehydrogenase from Arabidop- sis (GenBank accession no. X77943) with those of UDP- GlcDH. The sequence shows no significant homology to UDP-GlcDH. Hempel et al. (1994) showed that there is no obvious homology between UDP-GlcDH and alcohol dehydrogenase except for the NAD-binding site. It is interesting that the molecular mass of the Arabidopsis alcohol dehydrogenase is about 40 kD, and taken to- gether, it seems likely that Robertson et al. (1996) puri- fied an alcohol dehydrogenase rather than UDP-GlcDH from bean plants.
The detailed expression pattern of UDP-GlcDH in plants remains to be analyzed within our transgenic plants carry- ing a promotor-GUS fusion. The function of UDP-GlcDH needs further clarification that we hope to provide from the current analysis of various transgenic plants, in which the UDP-GlcDH is expressed in sense or anti-sense orientation.
Modified cell walls with a changed composition in hemicellulose are very helpful in elucidating the com- plex structural network of plant cell walls. A genetic approach for cell-wall composition mutants with Arabi- dopsis was recently described by Reiter et al. (1993). Our transgenic plants with manipulated levels of UDP- GlcDH might have similar phenotypes to particular mu- tants. Since we currently do not understand why cell- wall matrix polysaccharides from grasses are so different from those from dicotyledonous plants (Carpita and Gibeaut, 1993), transgenic plants with lower or altered hemicellulose content are promising tools to further ad- dress this problem. In addition, plants with such genet- ically modified cell walls are of potential interest in the area of biotechnology.
ACKNOWLEDCMENTS
We thank Heinrich Kauss for comments on the paper and for continuous support of the project. We thank Christine Rübel for technical help and Ralf Kaps for taking care of the animals. We also thank Andrew Cross for the generous gift of p47phox anti- bodies and Chris Lamb for the H1 cDNA probe.
Received May 22, 1996; accepted August 16, 1996 Copyright Clearance Center: 0032-0889/96/ 112/ 1127/08. The GenBank accession number for the soybean UDP-GlcDH se-
quence reported in this article is U53418.
LITERATURE ClTED
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mo1 Biol 215: 403-410
Amino S, Takeuchi Y, Komamine A (1985) Changes in the en- zyme activities involved in formation and interconversion of UDP-sugars during cell cycle in a synchronous culture of Catha- ranthus Yoseus. Physiol Plant 64: 111-117
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (1995) Current Protocols in Molecular Biol- ogy . Greene Publishing Associates / Wiley Interscience, New York
Bolwell GP (1988) Synthesis of cell wall components: aspects of control. Phytochemistry 27: 1235-1253
Carpita NC, Gibeaut DM (1993) Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J
Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Ana1 Biochem 162: 156-159
Church GM, Gilbert W (1984) Genomic sequencing. Proc Natl Acad Sci USA 81: 1991-1995
Dalessandro G, Northcote DH (1977a) Possible control sites of polysaccharide synthesis during cell growth and wall expansion of pea seedlings (Pisum sntivum L.). Planta 134: 39-44
Dalessandro G, Northcote DH (1977b) Changes in enzymatic activities of nucleoside diphosphate sugar interconversions dur- ing differentiation of cambium to xylem in sycamore and poplar. Biochem J 162: 267-279
Davies MD, Dickinson DB (1972) Properties of uridine diphos- phoglucose dehydrogenase from pollen of Lilium longiflouum. Arch Biochem Biophys 152: 53-61
Driouich A, Faye L, Staehelin LA (1993) The plant Golgi appara- tus: a factory for complex polysaccharides and glycoproteins. Trends Biochem Sci 18: 210-214
Feingold DS (1982) Monomeric and oligomeric sugars and sugar derivatives-occurrence, metabolism, function. In FA Loewus, W Tanner, eds, Plant Carbohydrates. Springer-Verlag, Berlin,
Gillaspy GE, Keddie JS, Oda K, Gruissem W (1995) Plant inositol monophosphatase is a lithium-sensitive enzyme encoded by a multigene family. Plant Cell 7: 2175-2185
Hayashi T, Matsuda K (1981) Sugar nucleotides from suspension- cultured soybean cells. Agric Biol Chem 45: 2907-2908
Hempel J, Perozich J, Romovacek H, Hinich A, Kuo I, Feingold DS (1994) UDP-glucose dehydrogenase from bovine liver: pri- mary structure and relationship to other dehydrogenases. Pro- tein Science 3: 1074-1080
1134 Tenhaken and Thulke Plant Physiol. Vol. 11 2, 1996
Iiyama K, Lam TBT, Meikle PJ, Ng K, Rhodes DI, Stone BA (1991) Cell wall biosynthesis and its regulation. In H Jung, D Buxton, R Hatfield, J Ralph, eds, Forage Cell Wall Structure and Digestibility. American Society of Agronomy, Madison, WI, pp
Loewus F, Chen MS, Loewus MW (1973) The myo-inositol oxida- tion pathway to cell wall polysaccharides. In F Loewus, ed, Biogenesis of Plant Cell Wall Polysaccharides. Academic Press, New York, pp 1-27
Loewus FA, Dickinson DB (1982) Cyclitols. In FA Loewus, W Tanner, eds, Plant Carbohydrates. Springer-Verlag, Berlin, pp
Lomax KJ, Leto TL, Nunoi H, Gallin JI, Malech HL (1989) Re- combinant 47-kilodalton cytosol factor restores NADPH oxidase in chronic granulomatous disease. Science 245: 409412
Maiti IB, Loewus FA (1978) Evidence for a functional myo-inositol oxidation pathway in Lilium long9orum pollen. Plant Physiol62:
Mehdy MC (1994) Active oxygen species in plant defense against pathogens. Plant Physiol 105: 467-472
Newman T, de Bruijn FJ, Green P, Keegstra K, Kende H, McIn- tosh L, Ohlrogge J, Raikhel N, Somerville S, Thomashow M, Retzel E, Somerville C (1994) Genes galore: a summary of the methods for accessing the results of a large scale partia1 sequenc- ing of anonymous Avabidopsis thaliana cDNA clones. Plant Physiol 106: 1241-1255
Reiter WD, Chapple CCS, Somerville CR (1993) Altered growth and cell walls in a fucose-deficient mutant of Arabidopsis. Science 261: 1032-1035
Roberts RM, Cetorelli JJ (1973) UDP-D-glucuronic acid pyrophos- phorylase and the formation of UDP-D-glucuronic acid in plants. In F Loewus, ed, Biogenesis of Plant Cell Wall Polysac- charides. Academic Press, New York, pp 49-68
Roberts RM, Loewus FA (1973) The conversion of ~-glucose-6-
621-684
193-21 6
280-283
[14C] to cell wall polysaccharide material in Zea mays in the presence of high endogenous levels of myo-inositol. Plant Physiol 52: 646-650
Robertson D, Beech I, Bolwell GP (1995) Regulation of the en- zymes of UDP-sugar metabolism during differentiation of French bean. Phytochemistry 39: 21-28
Robertson D, Smith C, Bolwell GP (1996) Inducible UDP-glucose dehydrogenase from French bean (Phaseolus vulgauis L.) locates to vascular tissue and has alcohol dehydrogenase activity. Bio- chem J 313: 311-317
Rubery PH (1972) The activity of uridine diphosphate-D-glucose: nicotinamide-adenine dinucleotide oxidoreductase in cambial tissue and differentiating xylem isolated from sycamore trees. Planta 103: 188-192
Schulz M, Weissenbock G (1988) Three specific UDP-glucuronate: flavone-glucuronosyl-transferases from primary leaves of Secale cevenle. Phytochemistry 27: 1261-1267
Taylor BH, Manhart JR, Amasino RM (1993) Isolation and char- acterization of plant DNAs. In BR Glick, JE Thompson, eds, Methods in Plant Molecular Biology and Biotechnology. CRC Press, Boca Raton, FL, pp 37-47
Tenhaken R, Levine A, Brisson LF, Dixon RA, Lamb C (1995) Function of the oxidative burst in hypersensitive disease resis- tance. Proc Natl Acad Sci USA 92: 4158-4163
Verma DC, Dougall DK (1979) Biosynthesis of myo-inositol and its role as a precursor of cell-wall polysaccharides in suspension cultures of wild-carrot cells. Planta 146: 55-62
Wingate VPM, Lawton MA, Lamb CJ (1988) Glutathione causes a massive and selective induction of plant defense genes. Plant Physiol 87: 206-210
Witt HJ (1992) UDP-glucose metabolism during differentiation and dedifferentiation of Riella helicopkylla. J Plant Physiol 140: 276-281
![Ulvan,aSulfatedPolysaccharidefromGreenAlgae,Activates ... · 2019. 7. 31. · Total uronic acid content was determined according to Blumenkrantz and Asboe-Hansen [18] using glucuronic](https://static.documents.pub/doc/80x56/6119f1d718590323f955d0e3/ulvanasulfatedpolysaccharidefromgreenalgaeactivates-2019-7-31-total-uronic.jpg)