Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2001 Sep;127(1):360–371. doi: 10.1104/pp.127.1.360

The Catalytic Site of the Pectin Biosynthetic Enzyme α-1,4-Galacturonosyltransferase Is Located in the Lumen of the Golgi1

Jason D Sterling 1, Heather F Quigley 1, Ariel Orellana 1, Debra Mohnen 1,*
PMCID: PMC117991  PMID: 11553763

Abstract

α-1,4-Galacturonosyltransferase (GalAT) is an enzyme required for the biosynthesis of the plant cell wall pectic polysaccharide homogalacturonan (HGA). GalAT activity in homogenates from pea (Pisum sativum L. var. Alaska) stem internodes co-localized in linear and discontinuous sucrose gradients with latent UDPase activity, an enzyme marker specific for Golgi membranes. GalAT activity was separated from antimycin A-insensitive NADH:cytochrome c reductase and cytochrome c oxidase activities, enzyme markers for the endoplasmic reticulum and the mitochondria, respectively. GalAT and latent UDPase activities were separated from the majority (80%) of callose synthase activity, a marker for the plasma membrane, suggesting that little or no GalAT is present in the plasma membrane. GalAT activities in proteinase K-treated and untreated Golgi vesicles were similar, whereas no GalAT activity was detected after treating Golgi vesicles with proteinase K in the presence of Triton X-100. These results demonstrate that the catalytic site of GalAT resides within the lumen of the Golgi. The products generated by Golgi-localized GalAT were converted by endopolygalacturonase treatment to mono- and di-galacturonic acid, thereby showing that GalAT synthesizes 1→4-linked α-d-galacturonan. Our data provide the first enzymatic evidence that a glycosyltransferase involved in HGA synthesis is present in the Golgi apparatus. Together with prior results of in vivo labeling and immunocytochemical studies, these results show that pectin biosynthesis occurs in the Golgi. A model for the biosynthesis of the pectic polysaccharide HGA is proposed.

Pectins are a family of polysaccharides present in all plant primary walls (O'Neill et al., 1990). Homogalacturonan (HGA) accounts for approximately 60% of the total pectin in plants (O'Neill et al., 1990; Mohnen et al., 1996). HGA is a linear polymer composed of 1→4-linked α-d-galactosyluronic acid residues (GalA) that are often methyl esterified at the C-6 carboxyl group (Mort et al., 1993) and may also be O-acetylated at O-2 and O-3 (Ishii, 1997). The GalA residues may, in some plants, be substituted at O-3 with a β-linked xylosyl residue (Schols et al., 1995; Yu and Mort, 1996).

An enzyme that catalyzes the transfer of GalA from UDP-GalA onto endogenous HGA was first identified in particulate suspensions from Phaseolus aureus and was named polygalacturonate 4-α-galacturonosyltransferase (EC 2.4.1.43; Villemez et al., 1965, 1966). A similar enzyme was characterized in microsomal membranes from suspension-cultured tobacco (Nicotiana tabacum L. cv Samsun) cells (Doong et al., 1995) and solubilized from tobacco membranes (Doong and Mohnen, 1998). The solubilized enzyme was shown to catalyze the addition of a single α-GalA residue onto O-4 of the nonreducing end (Scheller et al., 1999) of exogenous oligomeric HGA acceptors (Doong and Mohnen, 1998) and thus was named homogalacturonan 4-α-galacturonosyltransferase (HGA-GalAT).

No direct evidence has yet been reported for the subcellular localization of the glycosyltransferases responsible for pectin biosynthesis. However, the results of in vivo labeling and immunocytochemical analyses have suggested that pectin is synthesized in the Golgi apparatus (Northcote and Pickett-Heaps, 1966; Harris and Northcote, 1971; Staehelin and Moore, 1995). Immunocytochemical studies using antibodies that recognize the pectic polysaccharides HGA and rhamnogalacturonan I suggest that these pectic polysaccharides are synthesized within different compartments of the Golgi apparatus (Zhang and Staehelin, 1992) and are transported from the Golgi in vesicles that migrate to, and fuse with, the plasma membrane. The polysaccharides are then released into the extracellular space and incorporated into the wall (Northcote and Pickett-Heaps, 1966).

We now provide evidence that an α-1,4-galacturonosyltransferase (GalAT) from pea (Pisum sativum L. var. Alaska) epicotyls, with properties similar to the previously described HGA-GalAT from tobacco, is located in the Golgi apparatus and that the catalytic site of this enzyme is in the Golgi lumen. The third internodes of etiolated pea epicotyls were used for this study since the different subcellular membranes from this tissue can be separated using Suc density gradients, making such tissue useful for topological and localization studies of enzymes involved in cell wall polysaccharide biosynthesis (Brummell et al., 1990; Orellana et al., 1997; Wulff et al., 2000). Our data provide strong evidence that HGA is synthesized in the Golgi apparatus and a model for the biosynthesis of this pectic polysaccharide is proposed.

RESULTS

Subcellular Localization of GalAT Activity in Pea Membranes

Subcellular membrane fractions obtained by linear Suc gradient centrifugation of homogenates from etiolated pea third internodes were analyzed for GalAT activity and for organelle-specific enzyme activity (Fig. 1). GalAT activity was detected almost exclusively in fractions 26 through 30 (1.17–1.14 g cm−3; Fig. 1a). The presence of GalAT activity in such a limited number of fractions suggests that it is associated with an organelle of a distinct density. Enzyme marker assays for the ER and for the mitochondria (antimycin A-insensitive NADH:Cyt c reductase and Cyt c oxidase, respectively) showed that the ER (fractions 30–40, 1.14–1.10 g cm−3) and mitochondria (fractions 19–22, 1.20–1.18 g cm−3) were present in a different subset of fractions (Fig. 1b) than those that contained the greatest amount of GalAT activity.

Figure 1.

Figure 1

Open in a new tab

Location of GalAT activity and enzyme markers in pea membranes fractionated by linear Suc density gradient centrifugation. a, GalAT activity (▾) and Suc density (▿). The data were obtained from a single analysis of each fraction. Similar results were obtained in three separate gradients. b, Cytochrome (Cyt) c oxidase (▪) and antimycin A-resistant NADH Cyt c reductase (⋄), enzyme markers for mitochondria and endoplasmic reticulum (ER), respectively. Activities are expressed as the percentage of the highest total activity (191.2 nmol Cyt c reduced min−1 and 397.6 nmol Cyt c oxidized min−1 for Cyt c reductase and Cyt c oxidase, respectively) obtained in the peak fraction. The data were obtained from a single analysis of each fraction. Similar results were obtained in two independent gradients. c, Latent UDPase activity. Latent UDPase activity, a Golgi marker, is detected as the increase in A690 obtained in fractions permeabilized with 0.1% (v/v) Triton X-100 compared with the absorbance obtained in fractions in the absence of detergent. The position of greatest latent activity is depicted with a double-headed arrow (Golgi). Data are the average of duplicate samples from each fraction. Similar results were obtained in three separate gradients.

Latent UDPase activity is a well-characterized Golgi-specific enzyme marker (Orellana et al., 1997) that is detected following the permeabilization of membranes with detergent. The greatest amount of latent UDPase activity was present in fractions 24 through 32 of a density of 1.17 to 1.14 g cm−3 (Fig. 1c). These fractions coincide with the fractions that contain the greatest amount of GalAT activity, suggesting that GalAT is a Golgi-localized enzyme.

To confirm that GalAT is present in a Golgi-enriched fraction, pea homogenates were separated using discontinuous Suc gradient centrifugation. Each fraction (Fig. 2) was analyzed for GalAT activity and for membrane marker activities (Table I). Latent UDPase activity was highest in the membranes that banded at the 8% to 33% (w/w) Suc interface suggesting that these membranes were Golgi enriched (Tanford and Reynolds, 1976). The Golgi-enriched fraction also contained approximately 90% of the total GalAT activity (Table I). Some (21% of total) ER-specific enzyme marker activity was also detected in the Golgi-enriched fraction. However, the ER-enriched fraction (33%–38% [w/w] Suc interface) contained no GalAT activity (Table I). These results, when taken together with the results shown in Figure 1, strongly suggest that the GalAT from pea membranes is a Golgi-localized enzyme.

Figure 2.

Figure 2

Open in a new tab

Illustration of the gradient fractions collected following discontinuous Suc gradient centrifugation. The collected gradient fractions were analyzed for enzyme marker activity and for the presence of GalAT activity (see Table I). The membranes that banded at the 8% to 33% (w/w) Suc interface are defined as the Golgi vesicle-enriched fraction and the membranes that banded at the 33% to 38% (w/w) Suc interface are defined as the ER-enriched fraction (see Table I).

Table I.

Distribution of enzyme activities within pea homogenates fractionated by discontinuous Suc gradient centrifugation (see Fig. 2)

Gradient Fraction Latent UDPase Cyt c Oxidase NADH:Cyt c Reductase GalATa
μmol PO4 released min−1 μmol Cyt c oxidized min−1 ± se μmol Cyt c reduced min−1 ± se pmol GalA incorporated min−1 ± se
GVb 33.9 0 3.38 ± 0.12 1.46 ± 0.08
33% NDc 0 0 0
ERd 11.3 2.19 ± 0.15 6.21 ± 0.60 0
38% NDc 4.07 ± 0.20 3.98 ± 0.26 0
Pellet 5.87 2.17 ± 0.47 2.74 ± 0.16 0.12 ± 0.08
Open in a new tab

Discontinuous Suc gradients were made and enzyme assays conducted as described in “Materials and Methods.” Total enzyme activity in each gradient fraction is shown. The data are the average of two replicates ± se from one gradient. Similar results were obtained in two independent gradients.

a

 Assayed using 1.1 μm UDP-[14C]GalA. 

b

 The interface between the 8% and the 33% (w/w) Suc layers which is enriched in Golgi membranes. 

c

 ND, Not determined due to the presence of phosphate in the buffer. 

d

 The interface between the 33% and the 38% (w/w) Suc layers that is enriched in ER membranes. 

The biosynthesis of cellulose and callose has been shown to occur at the plasma membrane (Frederikson and Larsson, 1989; Delmer, 1999), and the possibility could not be discounted that GalAT may also be localized to the plasma membrane. Thus, experiments were conducted, using discontinuous Suc gradient centrifugation, to separate Golgi membranes from plasma membranes. The gradient fractions were assayed for GalAT activity, latent UDPase, and callose synthase activity, a known marker for plasma membrane vesicles in plants (Morre et al., 1987; Dhugga and Ray, 1994; Table II). Callose synthase activity was present in all of the discontinuous Suc gradient fractions assayed. These results indicate that the Golgi-enriched membranes obtained using discontinuous Suc gradients also contained some plasma membranes. Thus, the possibility cannot be ruled out that GalAT is localized to a subset of plasma membrane vesicles that co-localize with Golgi membranes. Nevertheless, because approximately 80% of the callose synthase activity is located in fractions that are distinct from those that contain the majority of the GalAT and latent UDPase activities (Table II), these results most strongly support the conclusion that GalAT is Golgi localized and is not a component of the plasma membrane.

Table II.

Distribution of latent UDPase, callose synthase, and GalAT activities in membranes collected by discontinuous density gradient centrifugation

Gradient Fraction Latent UDPase Callose Synthase GalATa
μmol PO4 released min−1 nmol Glc incorporated min−1 ± se pmol GalA incorporated min−1 ± se
GV 145.5 463.9 ± 70.8 1,453.3 ± 150.9
33% 49.6 585.5 ± 98.2 0
ER 31.6 465.6 ± 43.3 200.5 ± 48.2
38% 15.7 768.2 ± 52.0 0
Pellet 11.6 404.0 ± 70.8 0
Open in a new tab

Total enzyme activity in each gradient fraction is shown. The data are the average of duplicate samples ± se from one gradient. Similar results were obtained in two separate gradients.

a

 Assayed with 0.92 μM UDP-[14C]GalA and 100 μM UDP-GalA. 

Characterization of the Products Made by GalAT in Enriched Golgi Membranes from Pea

The subcellular fractionation of GalAT activity with the Golgi-enriched fraction from pea suggested that a pectin biosynthetic enzyme was localized in the Golgi. To test whether HGA is formed by the GalAT in pea Golgi, the [14C]-labeled products generated by incubating UDP-[14C]GalA with intact pea Golgi were analyzed by size exclusion chromatography (SEC) and by high performance anion-exchange chromatography (HPAEC). The products were analyzed both before and after treatment with base (to remove methyl and other esters), with endopolygalacturonase (EPGase; to fragment HGA), and with base plus EPGase. Fractions were collected, analyzed by scintillation counting, and the elution times of the radioactive products compared with the elution times of GalA, oligogalacturonides (OGAs), and dextran molecular mass standards.

The SEC elution profiles of untreated and the EPGase-treated products from intact pea Golgi are shown in Figure 3. The untreated product gave a peak (20% of the total radioactivity) with a retention time earlier than the 500-kD dextran standard, and a peak (42% of the total radioactivity) that co-eluted with GalA. Some radioactivity eluted between these two peaks and most likely corresponds to variously sized HGA. EPGase treatment of the intact product resulted in the conversion of the high-molecular mass material into products that have the same retention time as mono- and diGalA (Fig. 3). These results suggest that most of the [14C]GalA-labeled product made using intact Golgi membranes is α-1,4-linked HGA.

Figure 3.

Figure 3

Open in a new tab

SEC of products made by the GalAT in intact pea Golgi membranes. The untreated (▾) and EPGase-treated (□) products recovered following incubation of intact Golgi membranes with 4.7 μm UDP-[14C]GalA in the absence of exogenous acceptor were separated by SEC. The amount of radioactivity in each fraction was determined by scintillation counting. The cpm of each fraction minus the background is shown. Base (0.1 n NaOH) was included in the eluant solutions to enable electrochemical detection of GalA and dextran molecular mass standards by pulsed electrochemical detection.

The untreated and EPGase-treated products, and the products that were formed upon base and base plus EPGase treatment, were analyzed by HPEAC (Fig. 4, a–d). The untreated product contained a peak (27% of total radioactivity) that had a retention time similar to GalA, as well as a series of smaller peaks (32% of total radioactivity) that eluted at retention times comparable with medium-sized OGAs (DPs of 12 and 15–20; Fig. 4a). Base treatment of the untreated product did not alter the retention time of the radioactive peaks, suggesting that the products formed by intact pea Golgi were not extensively methoxylated or otherwise esterified (Fig. 4c). EPGase treatment of the untreated (Fig. 4b) or base-treated (Fig. 4d) products resulted in the appearance of diGalA, an increase in the amount of GalA, and the disappearance of the mid-sized OGAs, thereby confirming that the oligomeric peaks in the untreated and the base-treated products were HGA.

Figure 4.

Figure 4

Open in a new tab

HPAEC of the product made by the GalAT in intact pea Golgi membranes. The untreated (a), EPGase-treated (b), base-treated (c), and base plus EPGase-treated (d) products were analyzed by HPAEC. The amount of radioactivity present in each fraction was determined by scintillation counting. The cpm of each fraction minus the background is shown. Base (0.1 n NaOH) was included in the eluant solutions to enable electrochemical detection of the size standards GalA, diGalA, triGalA, UDP-GalA, and OGA degree of polymerization (DP) of 14 by pulsed electrochemical detection. A smaller amount (20% less) of the base + EPGase-treated products was injected into the column than the amount injected of the other products and therefore the cpm in d have been adjusted to correct for this difference.

The amount of radiolabeled mono- and diGalA recovered after EPGase digestion and HPAEC was not equal to the amount of radioactivity detected as the series of small peaks in the untreated product (compare Fig. 4, a with b). Furthermore, the absence of a peak corresponding to the high-molecular mass product (see Fig. 3) in either Figure 4, a or c, suggests either that the high-molecular mass material was unable to elute as a defined peak on the anion-exchange column or that the material was unable to pass through the 0.2-μm filters used prior to HPAEC analysis (see “Materials and Methods”). Analysis of the 0.2-μm filters revealed that approximately 47% and 26% (for untreated and base-treated products, respectively) of the total label incorporated into product remained on the filter following filtration.

In a preliminary study, we found that the amount of radiolabeled product formed by membrane-bound GalAT from pea increased with the addition of exogenous OGA acceptors in the presence of 0.1% (v/v) Triton X-100 (data not shown). Therefore, the product formed by Golgi membranes in the presence of 0.1% (v/v) Triton X-100 and an OGA acceptor (DP 14) was analyzed to confirm that the GalAT produced HGA under these conditions.

Several differences were observed between the products made by intact and detergent-permeabilized Golgi membranes. For example, approximately 47% of untreated product made by intact Golgi bound to the HPLC prefilter, whereas only 12% of the untreated product from Triton X-100-permeabilized Golgi bound to the filter. Comparable amounts of radioactive product (83% for intact Golgi and 98% for permeabilized Golgi) passed through the filter after base and EPGase treatment. These results suggested that the products formed using detergent-permeabilized Golgi and exogenous oligomeric HGA acceptors were smaller than, and/or had different characteristics than, the products made by intact Golgi vesicles.

The [14C]-labeled products formed using 0.1% (v/v) Triton X-100-permeabilized pea Golgi in the presence of an OGA acceptor (DP 14) were analyzed by SEC and HPAEC (Figs. 5 and 6) before and after base treatment, EPGase treatment, and base plus EPGase treatment. SEC of the untreated product (Fig. 5) gave two peaks that eluted with the same retention time as an authentic OGA of DP 15 (15 mer, 32% of total) and GalA (22% of total), respectively. The peak that eluted before GalA was shown (see below) to be GalA-1-phosphate (GalA1-P; 44% of total; Fig. 5). The radioactive 15-mer peak was completely converted by EPGase treatment into peaks that had the same retention time as GalA and diGalA (Fig. 5). No large molecular mass peak was detected in the untreated product formed by Triton X-100-treated pea Golgi, suggesting that in the presence of exogenous acceptor the GalAT in Triton X-permeabilized membranes does not synthesize a large molecular mass product. Similar results have been obtained with a GalAT solubilized from tobacco membranes (Doong and Mohnen, 1998).

Figure 5.

Figure 5

Open in a new tab

SEC of products produced by the GalAT in detergent-permeabilized Golgi in the presence of exogenous acceptor. Untreated (▾) and EPGase-treated (□) products made by Triton X-100-permeabilized Golgi vesicles incubated with 5 μm UDP-[14C]GalA and OGA of DP 14 as acceptor were analyzed by SEC. The collected fractions were analyzed by scintillation counting. The cpm in each fraction minus background is plotted. GalA, OGAs of DP 7 (7 mer) and 15 (15 mer), and dextran molecular mass standards were detected by pulsed electrochemical detection.

Figure 6.

Figure 6

Open in a new tab

HPAEC of products produced by the GalAT in detergent-permeabilized Golgi in the presence of exogenous acceptor. HPAEC analysis of the untreated (a), EPGase-treated (b), base-treated (c), and base plus EPGase-treated (d) products. The collected fractions were analyzed by scintillation counting and the cpm in each fraction minus background is plotted. GalA, diGalA, triGalA, tetraGalA, and OGA of DP 14 were detected by pulsed electrochemical detection. The untreated product was diluted 1:1 with 10 mm EDTA before it was spin filtered and separated by HPAEC and therefore the cpm in a have been adjusted to correct for this dilution.

HPEAC analysis of the products made by Triton X-100-treated pea Golgi in the presence of an OGA acceptor are consistent with the results obtained by SEC. The untreated product (Fig. 6a) eluted as three predominant peaks of radioactivity that co-eluted with GalA, GalA-1-P (elutes slightly earlier than tetraGalA, see below), and the OGA 15 mer, respectively. A small amount of larger sized material was also present.

Two methods were used to establish that the radiolabeled peak that elutes between tri-GalA and tetra-GalA is GalA-1-P. Authentic GalA-1-P was chromatographed by HPAEC, fractions collected, and the uronic acid content of each fraction determined colorimetrically. The retention time of the radiolabeled peak and GalA-1-P were the same (data not shown). To confirm that the putative radiolabeled GalA-1-P was GalA-1-P, radiolabeled product generated by permeabilized Golgi in the presence of OGA of DP 12 was recovered by extraction of the product with chloroform:methanol and a portion of the aqueous phase was treated with alkaline phosphatase. The untreated (Fig. 7a) and phosphatase-treated products (Fig. 7b) were separated by HPAEC. As anticipated, the putative GalA-1-P peak disappeared after treatment with phosphatase and the amount of GalA increased. Phosphatase treatment did not change the retention time of OGA (DP 13) or GalA. These results establish that the radioactive product that elutes by HPEAC between tri-GalA and tetra-GalA is phosphorylated GalA.

Figure 7.

Figure 7

Open in a new tab

Alkaline phosphatase treatment of an aqueous extract of the radiolabeled products produced in pea Golgi. Permeabilized Golgi vesicles were incubated for 60 min at 30°C with an exogenous OGA acceptor of DP 12 and 2.5 μm UDP-[14C]GalA. The reaction was stopped by the addition of chloroform:methanol (5:1, v/v) and the aqueous phase was extracted, washed, and treated for 60 min at 37°C with: a, water (▪); or b, 10 units of alkaline phosphatase (□). The reaction was terminated by incubation at 85°C for 10 min, filtered, and analyzed by HPAEC. Fractions were collected and analyzed by scintillation counting. The cpm in each fraction minus background is shown.

Base treatment of the intact product made by detergent-permeabilized Golgi did not shift the retention time of the peaks (Fig. 6c), suggesting that the products were not methoxylated or otherwise esterified. Treatment of the untreated or base-treated products with an EPGase resulted in the disappearance of the OGA 15 mer and larger peaks, the appearance of diGalA, and an increase in the amount of GalA (Figs. 6, b and d). These results show that most of the [14C]GalA-labeled product made in Triton X-100-permeabilized Golgi membranes in the presence of exogenous OGA 14 mer is OGA of DP 15.

The GalAT in permeabilized Golgi vesicles and the GalAT solubilized from tobacco synthesize similar products. Therefore, it was possible that 0.1% (v/v) Triton X-100 was solubilizing the enzyme from Golgi membranes rather than permeabilizing the membranes. To test this, Golgi vesicles were treated with 0.1% (v/v) Triton X-100, centrifuged at 100,000g for 1 h, and the GalAT activity in the supernatant and in the pellet was determined (data not shown). All of the activity was present in the pellet, suggesting that 0.1% (v/v) Triton X-100 alone was not sufficient to solubilize GalAT activity from Golgi membranes and that the changes in product formation upon Triton X-100 treatment were due either to the permeabilization of the Golgi membranes or to the effect of the detergent on the GalAT itself. Nevertheless, the characteristics of the products formed by GalAT from pea (Figs. 3–6) are consistent with the characteristics of the membrane bound (Doong et al., 1995) and solubilized forms of GalAT from tobacco (Doong and Mohnen, 1998).

Topography of GalAT in Pea Golgi

Pectin biosynthesis is believed to occur in the lumen of the Golgi endomembrane system (Northcote and Pickett-Heaps, 1966; Zhang and Staehelin, 1992) and it is likely that the glycosyltransferases, or at least their catalytic sites, face the lumen of the Golgi. To test this hypothesis, isolated Golgi membranes were treated with Proteinase K (a protease that cannot pass through intact membranes) in the presence and in the absence of 0.1% (v/v) Triton X-100 and then assayed for GalAT activity. GalAT activity was similar in intact Golgi vesicles, Triton X-100-treated vesicles, and proteinase K-treated vesicles (Fig. 8). However, GalAT activity in Golgi membranes treated with both Triton X-100 and proteinase K was reduced to levels comparable with that of boiled Golgi vesicles. These results demonstrate that the intact Golgi membranes protect the catalytic site of GalAT from Proteinase K digestion and strongly suggest that the catalytic domain of GalAT is located within the lumen of the Golgi.

Figure 8.

Figure 8

Open in a new tab

Proteinase K treatment of pea Golgi membranes. Golgi vesicles (100 μg protein) were incubated for 30 min at 30°C in the presence or absence of 0.1% (v/v) Triton X-100 and/or 4 μg proteinase K under the following conditions: BGV, Golgi vesicles boiled for 5 min prior to incubation; GV, intact Golgi vesicles; +TX, Golgi vesicles incubated in the presence of 0.1% (v/v) Triton X-100; +PK, intact Golgi vesicles incubated with 4 μg Proteinase K; +TX and PK, Golgi vesicles incubated with both 0.1% (v/v) Triton X-100 and 4 μg Proteinase K. An aliquot (10 μL) of the Golgi vesicle mixture from each treatment was brought to a final concentration of 0.1% (v/v) Triton X-100 and assayed for GalAT activity (see “Materials and Methods”). The amount of radioactivity recovered in each of the products is shown. Data are the average of triplicates from one experiment. Similar results were obtained in three separate experiments.

DISCUSSION

The results presented here provide the first direct enzymatic evidence that GalAT is located in the Golgi apparatus. GalAT activity localizes exclusively to fractions that contain the majority of the latent UDPase activity, a well-characterized Golgi-specific enzyme marker. A combination of detergent and proteinase K treatments show that pea GalAT is a membrane-bound protein whose catalytic site faces the lumen of the Golgi apparatus. The catalytic site of tobacco HGA methyltransferase has also been localized to the lumen of the Golgi apparatus (Goubet and Mohnen, 1999). Taken together, our results show that pectin biosynthetic enzymes are present in the Golgi and confirm that pectin biosynthesis occurs in the Golgi endomembrane system (Northcote and Pickett-Heaps, 1966; Harris and Northcote, 1971; Zhang and Staehelin, 1992).

A model summarizing our current understanding (Doong et al., 1995; Liljebjelke et al., 1995; Doong and Mohnen, 1998) of the biosynthesis of HGA is shown in Figure 9. UDP-GalA, the substrate for GalAT, is believed to be formed on the cytosolic side of the Golgi by the epimerization of UDP-GlcA catalyzed by a UDP-glucuronate-4-epimerase (EC 5.1.3.6; Feingold et al., 1960; Feingold and Avigad, 1980; Mohnen, 1999) that is associated with Golgi membranes (K. Adams and D. Mohnen, unpublished data). The UDP-GalA is thought to be transported into the lumen of the Golgi apparatus by a proposed membrane-bound UDP-GalA:UMP antiport transport protein. Similar nucleotide sugar transporters involved in protein glycosylation have been identified, purified, and cloned from several mammalian and yeast sources (Hirschberg et al., 1998). GalAT, a Golgi-resident protein with its catalytic site in the lumen of the Golgi, binds UDP-GalA and catalyzes the transfer of GalA onto a growing HGA chain and the production of UDP. The UDP is hydrolyzed in the lumen into inorganic phosphate and UMP by a Golgi-resident membrane-bound nucleoside diphosphatase (NDPase/UDPase) (Orellana et al., 1997). The UMP is exported out of the Golgi in exchange for UDP-GalA by the proposed UDP-GalA:UMP antiport transporter.

Figure 9.

Figure 9

Open in a new tab

A model for HGA biosynthesis in planta. 1, UDP-GalA is formed by the epimerization of UDP-GlcA on the cytosolic side of the Golgi. 2, UDP-GalA is transported by a proposed UDP-GalA:UMP antiporter into the lumen of the Golgi apparatus. 3, GalA is transfered from UDP-GalA onto a growing HGA chain by a GalAT that has its catalytic site facing the lumen of the Golgi. The UDP released is hydrolyzed to UMP and inorganic phosphate by a Golgi-localized NDPase (UDPase). The UMP is then exported by a UDP-GalA:UMP antiporter in exchange for UDP-GalA.

The products synthesized by the GalAT from pea are comparable with the products synthesized by the GalAT from tobacco (Doong et al., 1995; Doong and Mohnen, 1998). The GalAT in detergent-permeabilized pea Golgi membranes produces a product that is predominantly only one GalA residue longer than the exogenous acceptor. This result is similar to that obtained with the solubilized GalAT from tobacco, where the product generated is also only one residue longer than the exogenous acceptor (Doong and Mohnen, 1998). In contrast, the GalAT in intact pea Golgi and in tobacco membranes both use endogenous acceptors to generate high-molecular- mass products, with the product in pea (>500 kD) being larger than that produced in tobacco (approximately 105 kD; Doong et al., 1995). The number of GalA residues added onto the endogenous acceptor(s) in the pea and tobacco membranes in vitro has not yet been determined.

There are a number of possible reasons why the pea detergent-permeabilized and the tobacco-solubilized GalATs add only a single GalA to the end of the exogenous acceptors. For example, treating pea Golgi vesicles with detergent and exogenous OGA acceptors may cause the GalAT to preferentially use the exogenous OGAs as acceptors, whereas the GalAT in intact Golgi uses an endogenous acceptor. The exogenous OGA (e.g. DP 14) acceptor may not be the optimal substrate for the GalAT, whereas the endogenous product found in intact Golgi membranes presumably contains the necessary structural information for the enzyme to synthesize a polymeric product. Such acceptor-mediated control of enzyme processivity has been demonstrated for a mammalian α-2,8-polysialic acid synthase (Kojima et al., 1996), where the placement of an α-1,6-linked Fuc residue in the acceptor was shown to be necessary for polysialylation. It is also possible that treating Golgi membranes with Triton X-100 disrupts an enzyme complex (protein-protein or other interaction) or removes an element that is required for polymer synthase activity. Protein-protein complexes involved in glycan biosynthesis have been demonstrated for glycogenin, a self-glucosylating protein that initiates glycogen biosynthesis (Lin et al., 1999).

The presence of GalA among the products formed by both intact and detergent-permeabilized pea Golgi membranes, and the presence of GalA1-P in the products formed by detergent-permeabilized pea Golgi membranes, are a major difference between the reaction products formed by pea and tobacco membranes (Doong and Mohnen, 1998). The presence of GalA and GalA1-P most likely results from the hydrolysis of UDP-[14C]GalA by a phosphodiesterase (Feingold et al., 1958) and a phosphatase (Neufeld et al., 1961). Such hydrolases have been reported to be released during the homogenization and isolation of membranes from some plant tissues (Villemez et al., 1966; Rodgers and Bolwell, 1992; Brickell and Reid, 1996) and it is likely that such enzymes were released during the preparation of the pea Golgi.

Several enzymes including latent UDPase (Orellana et al., 1997) and mammalian glycosyltransferases (Paulson and Colley, 1989) such as the polysialyltransferases that function in the polysialylation of neural cell adhesion molecule and other N-linked glycoproteins (for review, see Tsuji, 1996) have their catalytic sites located in the lumen of the Golgi. Most Golgi-localized glycosyltransferases are type II integral membrane proteins with a single N-terminal membrane-spanning region and a lumenal C-terminal catalytic domain (Paulson and Colley, 1989; Tsuji, 1996). On the other hand, plasma membrane-localized polymer synthases such as hyaluronan synthase (Weigel et al., 1997) and the putative cellulose synthase (Delmer, 1999) are predicted to have multiple transmembrane-spanning regions at the amino and carboxy termini and a large central domain containing the catalytic site. The purification of the GalAT, the determination of its amino acid sequence, and the identification of its gene will be required to elucidate the structure of GalAT and will allow the comparison of GalAT structure with that of known glycosyltransferases and polymer synthases.

MATERIALS AND METHODS

Chemicals

Uridine diphosphate-α-d-[14C]GlcA (UDP-[14C]GlcA, 293.6 mCi mmol−1) and uridine diphosphate-α-d-[14C]Glc (UDP-[14C]Glc, 286.2 mCi mmol−1) were purchased from DuPont-New England Nuclear (Boston). Uridine diphosphate-α-d-[14C]GalA (UDP-[14C]GalA) was prepared by the epimerization of UDP-[14C]GlcA according to Liljebjelke et al. (1995). OGAs with a DP of 14 and a mixture of OGAs of DP 7 to 23 were prepared according to Doong et al. (1995). Proteinase K was purchased from Boehringer Mannheim (Indianapolis). Homogeneous EPGase (748 units mL−1, 1 unit releases 1 μmol GalA per min) from Fusarium moniliforme was a gift from Carl Bergmann (Complex Carbohydrate Research Center, Athens, GA). All other reagents were purchased from Sigma (St. Louis).

Plant Materials

Pea (Pisum sativum L. var. Alaska) seeds were purchased from the Green Seed Company (Athens, GA). Seedlings were grown in moist vermiculite for 7 to 8 d at 25°C in the dark.

Linear Suc Gradient Fractionation of Pea Homogenates

Third internodes (from the cotyledon, 1 cm in length, 5 g total weight) from 7- to 8-d-old etiolated peas (Pisum sativum L. var Alaska) were excised and homogenized in 50 mm HEPES [4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid]/KOH, pH 7.0, containing 0.4 m Suc, 1 mm MgCl2, 10 mm KCl, 1 mm EDTA, and 1 mm dithiothreitol using a razor blade and a mortar and pestle. The homogenate was filtered through two layers of Miracloth (Calbiochem, San Diego) and the filtrate centrifuged for 10 min at 1,000g. The 1,000g supernatant (5 mL) was overlaid onto 24 mL of a 20% to 50% Suc (w/w) gradient (Munoz et al., 1996) in 50 mm HEPES/KOH, pH 7.0 containing 0.1 mm MgCl2, 1 mm EDTA, and 1 mm dithiothreitol and centrifuged for 3.5 h at 100,000g. Fractions (0.5 mL) were collected and analyzed for marker enzyme activity. All manipulations were done at 4°C. Suc concentrations were determined using an Abbe refractometer (model 3L, Milton Roy Company, Rochester, NY) and the densities were calculated using an ambient temperature of 20°C.

Isolation of Golgi Membranes

Enriched Golgi membranes were isolated using discontinuous Suc gradients according to the method of Munoz et al. (1996). All manipulations were done on ice or at 4°C. The third internodes (1 cm, 40–70 g total mass) of 7- to 8-d-old etiolated pea stem epicotyls were excised and homogenized in 1 volume of buffer A (0.1 m KH2PO, pH 6.65, containing 5 mm MgCl2, 1 mm dithioerythritol, and 16% [w/w] Suc) using a razor blade and a mortar and pestle. The homogenate was filtered through two layers of Miracloth and centrifuged for 2 min at 1,000g. The supernatant was layered onto 38% (w/w) Suc in buffer A (8 mL) and centrifuged for 90 min at 100,000g. The upper phase was removed without disturbing the interface layer and replaced with 33% (w/w) Suc (15 mL) and 8% (w/w) Suc (5 mL) in buffer A. The gradient was centrifuged for a further 100 min at 100,000g. The pellet was resuspended in 10 mm Tris, pH 7.5, containing 0.25 m Suc (ST buffer) and stored at −80°C. The membranes that equilibrated at each of the interfaces were collected, diluted with an equal volume of water, and centrifuged for 50 min at 100,000g. These pellets were resuspended in ST buffer and stored at −80°C. In some experiments (Table II), membranes from the 33% and 38% (w/w) Suc phases were also collected and stored as described above. Otherwise, the 33% and 38% Suc phase with no prior membrane collection was used for enzyme analysis (Table I).

Enzyme Assays

Cyt c oxidase activity was measured as described by Briskin et al. (1987) with some modifications. Cyt c oxidase was assayed spectrophotometrically at 550 nm using a scanning spectrophotometer (model UV-2101; Shimadzu, Columbus, MD). A portion (35 μL) of each gradient fraction was added to 30 mm K2HPO4, pH 7.4 (735 μL), containing 1 mm EDTA, 1 mm NaHCO3, 0.00045% (v/v) Triton X-100, and 50 μm Cyt c (reduced with sodium dithionite). The total activity of Cyt c oxidase was calculated for each fraction using the initial linear rate of Cyt c oxidation (decrease in A550) and the extinction coefficient of Cyt c (18.5 mm−1 cm−1).

Antimycin A-resistant NADH:Cyt c reductase was measured according to a modification of Vannier et al. (1992). A portion (0.1 mL) of each gradient fraction was combined with 50 mm KH2PO4, pH 7.5 (0.9 mL), containing 50 μm Cyt c, 0.3 mm NADH, 5 mm KCN, and 1 μm antimycin A. The linear increase in A550 was taken as the rate of Cyt c reduction and was used to calculate the total activity in each fraction.

Latent UDPase activity was measured according to Briskin et al. (1987) with some modifications. A reaction mixture (15 μL) containing 6 mm UDP, 6 mm MgSO4, and 60 mm Tris/MES [2-(N-morpholino)-ethanesulfonic acid], pH 6.5, with or without 0.1% (v/v) Triton X-100, was added to 15 μL of gradient fraction in a microtiter plate (Nunc, Suwanee, GA) and kept for 15 min at 25°C. The reaction was stopped by the addition of Ames reagent (270 μL) and processed as in Ames (1966). The A690 of the sample was measured using a Tikertek Multiscan ELISA plate reader (model MCC/340 MKII, Flow Laboratories, McLean, VA) and was converted to the amount of Pi released from UDP (UDPase activity) using a standard curve of KH2PO4. Latent UDPase activity was considered to be present in those fractions in which an increase in A690 was observed when fractions were incubated in the presence 0.1% (v/v) Triton X-100. The latent UDPase activity reported in Tables I and II and in Figure 1 was calculated by subtracting the amount of Pi released from UDP by enzyme in the absence of Triton X-100 from the amount of Pi released from UDP by enzyme in the presence of Triton X-100.

Callose synthase activity was measured according to a method by Dhugga and Ray (1994) with some modifications. Gradient fractions (10 μL of one-tenth-diluted sample) were incubated for 5 min in 10 mm MOPS [3-(N-morpholino)-propanesulfonic acid], pH 7.0, containing 1 mm UDP-Glc, 10 mm cellobiose, 2 mm CaCl2, 0.3 mm spermine, 0.02% (v/v) Triton X-100, and 0.32 μm UDP-[14C]Glc. The reaction was stopped by the addition of 70% (v/v) ethanol (2 mL) and the insoluble products were collected by filtration through a GF/A glass microfiber disc (Whatman, Maidstone, UK). The discs were washed with 70% (v/v) ethanol (15 mL), dried, and the amount of radioactivity on the discs measured by scintillation counting.

GalAT activity was measured as described by Doong et al. (1995) with some modifications. Samples (10 μL) were incubated for 2 min in 50 mm HEPES, pH 7.8, containing 100 μg of OGAs (DP of 7–23), 1.1 μm UDP-[14C]GalA (specific activity 293.6 mCi mmol−1), 0.25 mm MnCl2, 25 mm KCl, and 0.1% (v/v) Triton X-100 in a total volume of 30 μL (unless otherwise stated). The reaction was stopped by the addition of 700 μL chloroform:methanol (3:2, v/v) and centrifuged to collect the precipitable products. The product was washed twice with 500 μL of 65% (v/v) ethanol and the amount of radioactivity quantified by scintillation counting.

All enzyme assays were conducted at room temperature with the exception of the GalAT assay, which was conducted at 30°C. Protein concentrations were determined by the protein assay (Bio-Rad, Hercules, CA) using bovine serum albumin as a standard. Uronic acids were analyzed colorimetrically using the meta-hydroxybiphenyl assay (Blumenkrantz and Asboe-Hansen, 1973).

Product Characterization

The products generated by GalAT in pea Golgi vesicles were analyzed according to Doong et al. (1995). The products were formed using either detergent-permeabilized Golgi in the presence of OGAs of DP 14 (14 mer) as exogenous acceptor or using intact Golgi without exogenous acceptor.

Four separate GalAT reactions containing 12 μL of Golgi vesicles (approximately 43 μg of protein) and 5 μm UDP-[14C]GalA in 50 mm HEPES, pH 7.8, 0.25 mm MnCl2, and 25 mm KCl were incubated with or without 0.1% (v/v) Triton X-100 and 15 μg of OGA 14 mer for 35 min at 30°C in a total volume of 60 μL. The products were recovered, washed as described above, resuspended in 0.5 mL of 50 mm NaOAc, pH 6.0, containing 5 mm EDTA, and pooled (total volume 2 mL). Individual pools were analyzed as follows by either HPAEC or SEC. For example, a 500-μL aliquot was analyzed directly by HPAEC. A second aliquot of 500 μL was treated for 8 h at 30°C with a homogeneous EPGase (2 units) and the products generated were analyzed by HPAEC. A third aliquot of the pooled products (1 mL) was adjusted to pH 12 by the addition of 1 n NaOH and kept at 4°C for 8 h. The solution was adjusted to pH 5.0 with 1 n HCl and a portion (500 μL) was analyzed by HPAEC. The remaining base-treated product (500 μL) was treated with EPGase for 8 h and analyzed by HPAEC.

Alkaline Phosphatase Treatment of GalA-1-P

Aqueous extracts from a GalAT reaction were prepared by incubating 12 μL of Golgi vesicles (approximately 43 μg of protein) for 60 min at 30°C in 50 mm HEPES, pH 7.8, containing 2.5 μm UDP-[14C]GalA, and 15 μg of OGAs of DP 12, 0.25 mm MnCl2, 25 mm KCl, and 0.1% (v/v) Triton X-100 in a total volume of 60 μL. The reaction was terminated by the addition of chloroform:methanol (5:1, v/v, 300 μL) and 240 μL 50 mm NaOAc, pH 6.0, containing 5 mm EDTA was added. The mixture was centrifuged at 13,000g for 2 min. The aqueous upper phase was mixed with 300 μL of chloroform:methanol (5:1, v/v), centrifuged, and removed. The organic phase from the first centrifugation was mixed with 50 mm NaOAc, pH 6.0 (300 μL), containing 5 mm EDTA, centrifuged, and the aqueous phase was combined with the previously obtained aqueous phase. An aliquot (80 μL) of the combined aqueous phases was adjusted to pH 8.5 with Tris (10 mm). Water or 10 units of alkaline phosphatase was added, and the mixture kept for 60 min at 37°C. The reaction was stopped by heating at 85°C for 10 min. The resulting mixture was filtered (0.2 μm) and chromatographed by HPAEC. Fractions (1 mL) were collected and the amount of radioactivity in each fraction was quantified by liquid scintillation counting.

SEC of the [14C]-Labeled Product Generated Using Intact and Permeabilized Golgi

The products made by GalAT in intact Golgi without exogenous acceptor, and the products formed by detergent-permeabilized (0.1% [v/v] Triton X-100) Golgi in the presence of exogenous OGA 14 mer (15 μg) were separated using a Superose12 HR10/30 SEC column (Pharmacia, Uppsala). The column was eluted with 50 mm sodium acetate containing 5 mm EDTA at 0.35 mL min−1. Fractions (0.5 mL) were collected and the radioactivity in a portion (400 μL) of each fraction was determined by scintillation counting. The SEC elution profiles of dextran molecular mass standards, GalA, triGalA, and OGAs of a DP of 7, 12, 15, and 16 were detected using a pulsed electrochemical detector (Dionex, Sunnyvale, CA) with post-column addition of 400 mm NaOH at a flow rate of 0.2 mL min−1 (see below).

HPAEC of the [14C]-Labeled Product Made Using Intact and Permeabilized Golgi

[14C]GalA-labeled product generated by GalAT in pea Golgi vesicles, and GalA and OGA standards, were filtered (0.2-μm nylon microfilterfuge tubes, Rainin Instrument Co., Inc., Woburn, MA) and chromatographed on a CarboPac PA100 column (4 × 250 mm) fitted with a CarboPac PA100 guard column (3 × 25 mm) using a DX 500 LC (Dionex, Sunnyvale, CA). The column was eluted at 1 mL min−1 with a gradient formed from ultra-pure deionized water (eluant A) and 1 m sodium acetate containing 5 mm EDTA (eluant B) as follows: 0% (v/v) B at 0 to 2 min, 40% (v/v) B at 17 min, 80% (v/v) B at 49 min, 100% (v/v) B at 50 to 52 min, and 0% (v/v) B at 57 min (modified from Doong et al., 1995). The non-radiolabeled standards were detected by pulsed electrochemical detection after post-column addition of 400 mm NaOH at 0.5 mL min−1. The pulse sequence for the electrochemical detector was 0.05 V at 0 s, 0.05 V at 0.2 s (begin integration), 0.05 V at 0.4 s (end integration), 0.75 V at 0.41 to 0.6 s, and −0.15 V at 0.61 to 1.0 s. Fractions (0.75 mL) were collected, an aliquot (500 μL) of each fraction was combined with 10 volumes of scintillation cocktail (Scintiverse, Fisher Scientific, Norcross, GA), and the amount of radioactivity was measured using a scintillation counter. Quenching by salt and/or base was found to be negligible.

Proteinase K Digestion

Proteinase K treatment of Golgi vesicles was conducted according to Orellana et al. (1997). In brief, Golgi vesicles (100 μg protein) were incubated for 30 min at 30°C in 10 mm Tris, pH 7.5, containing 0.25 m Suc and 1 mm MgCl2 (100 μL total volume) in the presence and absence of 0.1% (v/v) Triton X-100 and 4 μg Proteinase K. Proteolysis was stopped by the addition of 1 mm phenylmethylsulfonylfluoride and 10 μL of each treatment was brought to a final concentration of 0.1% (v/v) Triton X-100 and assayed for GalAT activity in the presence of exogenous OGA acceptors.

ACKNOWLEDGMENTS

We thank Lorena Norambuena for her assistance with the Suc gradients and the marker enzyme assays and Malcolm O'Neill for critical reading of the manuscript.

Footnotes

1

This work was supported by the National Science Foundation (grant no. INT–9722509 to D.M.), by the National Research Initiative (competitive U.S. Department of Agriculture award no. 98–35304–6772 to D.M.), by Fondecyt (grant no. 1000675 to A.O.), by the Programa de Cooperacion Internacional from Conicyt, Chile (grant to A.O.), by a Department of Energy-funded center grant (no. DE–FG05–93–ER20097), and by the Eastman Chemical Company, Kingsport, TN (fellowship to J.D.S.).

LITERATURE CITED

  1. Albersheim P, Darvill AG, O'Neill MA, Schols HA, Voragen AGJ. A hypothesis: the same six polysaccharides are components of the primary cell walls of all higher plants. In: Visser J, Voragen AGJ, editors. Pectins and Pectinases. Amsterdam: Elsevier Sciences; 1996. pp. 47–53. [Google Scholar]
  2. Ames BN. Assay of inorganic phosphate, total phosphate and phosphatases. Methods Enzymol. 1966;8:115–118. [Google Scholar]
  3. Blumenkrantz N, Asboe-Hansen G. New method for quantitative determination of uronic acids. Anal Biochem. 1973;54:484–489. doi: 10.1016/0003-2697(73)90377-1. [DOI] [PubMed] [Google Scholar]
  4. Brickell LS, Reid JSG. Biosynthesis in vitro of pectic (1–4)-β-D-galactan. In: Visser J, Voragen AGJ, editors. Pectins and Pectinases. Amsterdam: Elsevier Sciences; 1996. pp. 47–53. [Google Scholar]
  5. Briskin DP, Leonard RT, Hodges TK. Isolation of the plasma membrane: membrane markers and general prinicipals. Methods Enzymol. 1987;148:542–558. [Google Scholar]
  6. Brummell DA, Camirand A, Maclachlan GA. Differential distribution of xyloglucan glycosyl transferases in pea Golgi dictyosomes and secretory vesicles. J Cell Sci. 1990;96:705–710. [Google Scholar]
  7. Delmer D. Cellulose biosynthesis: exciting times in a difficult field of study. Annu Review Plant Physiol Plant Mol Biol. 1999;50:245–276. doi: 10.1146/annurev.arplant.50.1.245. [DOI] [PubMed] [Google Scholar]
  8. Dhugga KS, Ray PM. Purification of 1,3-β-d-glucan synthase activity from pea tissue: two polypeptides of 55 kDa and 70 kDa copurify with enzyme activity. Eur J Biochem. 1994;220:943–953. doi: 10.1111/j.1432-1033.1994.tb18698.x. [DOI] [PubMed] [Google Scholar]
  9. Doong RL, Liljebjelke K, Fralish G, Kumar A, Mohnen D. Cell free synthesis of pectin: identification and partial characterization of polygalacturonate 4-α-galacturonosyltransferase and its products from membrane preparations of tobacco (Nicotiana tabacum L. cv samsun) cell suspension cultures. Plant Physiol. 1995;109:141–152. doi: 10.1104/pp.109.1.141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Doong RL, Mohnen D. Solubilization and characterization of a galacturonosyltransferase that synthesizes the pectic polysaccharide homogalacturonan. Plant J. 1998;13:363–374. [Google Scholar]
  11. Feingold DS, Avigad G. Sugar nucleotide transformations in plants. In: Preiss J, editor. The Biochemistry of Plants. Vol. 3. New York: Academic Press; 1980. pp. 101–170. [Google Scholar]
  12. Feingold DS, Neufeld EF, Hassid WZ. Enzymatic synthesis of uridine diphosphate glucuronic acid and uridine diphosphate galacturonic acid with extracts from Phaseolus aureus seedlings. Arch Biochem Biophys. 1958;78:401–405. doi: 10.1016/0003-9861(58)90365-5. [DOI] [PubMed] [Google Scholar]
  13. Feingold DS, Neufeld EF, Hassid WZ. The 4-epimerization and decarboxylation of uridine diphosphate d-glucuronic acid by extracts from Phaseolus aureus seedlings. J Biol Chem. 1960;235:910–913. [PubMed] [Google Scholar]
  14. Frederikson K, Larsson C. Activation of 1,3-β-glucan synthase by Ca2+, spermine, and cellobiose: localization of activator sites using inside out plasma membrane vesicles. Physiol Plant. 1989;77:196–201. [Google Scholar]
  15. Goubet F, Mohnen D. Subcellular localization and topology of homogalacturonan methyltransferase in suspension-cultured Nicotiana tabacum cells. Planta. 1999;209:112–117. doi: 10.1007/s004250050612. [DOI] [PubMed] [Google Scholar]
  16. Harris PJ, Northcote DH. Polysaccharide formation in plant golgi bodies. Biochim Biophys Acta. 1971;237:56–64. doi: 10.1016/0304-4165(71)90029-8. [DOI] [PubMed] [Google Scholar]
  17. Hirschberg CB, Robbins PW, Abeijon C. Transporters of nucleotide sugars, ATP, and nucleotide sulfate in the endoplasmic reticulum and Golgi apparatus. Annu Rev Biochem. 1998;67:49–69. doi: 10.1146/annurev.biochem.67.1.49. [DOI] [PubMed] [Google Scholar]
  18. Ishii T. O-Acetylated oligosaccharides from pectins of potato tuber cell walls. Plant Physiol. 1997;113:1265–1272. doi: 10.1104/pp.113.4.1265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kojima N, Tachida Y, Yoshida Y, Tsuji S. Characterization of mouse ST8Sia II (STX) as a neural cell adhesion molecule-specific polysialic acid synthase: requirement of core α1,6-linked fucose and a polypeptide chain for polysialylation. J Biol Chem. 1996;271:19457–19463. doi: 10.1074/jbc.271.32.19457. [DOI] [PubMed] [Google Scholar]
  20. Liljebjelke K, Adolphson R, Baker K, Doong RL, Mohnen D. Enzymatic synthesis and purification of uridine diphosphate [14C]galacturonic acid: a substrate for pectin biosynthesis. Anal Biochem. 1995;225:296–304. doi: 10.1006/abio.1995.1158. [DOI] [PubMed] [Google Scholar]
  21. Lin A, Mu J, Yang J, Roach PJ. Self-glucosylation of glycogenin, the initiator of glycogen biosynthesis, involves an inter-subunit reaction. Arch Biochem Biophys. 1999;363:163–170. doi: 10.1006/abbi.1998.1073. [DOI] [PubMed] [Google Scholar]
  22. Mohnen D. Biosynthesis of pectins and galactomannans. In: Barton D, Nakanishi K, Meth-Cohn O, Pinto BM, editors. Comprehensive Natural Products Chemistry. Vol. 3 1999. , Carbohydrates and Their Derivatives including Tannins, Cellulose, and Related Lignins. Elsevier, Oxford, pp 497–527. [Google Scholar]
  23. Mohnen D, Doong RL, Liljebjelke K, Fralish G, Chan J. Cell free synthesis of the pectic polysaccharide homogalacturonan. In: Visser J, Voragen AGJ, editors. Pectins and Pectinases. Amsterdam: Elsevier Science; 1996. pp. 109–126. [Google Scholar]
  24. Morre DJ, Brightman AO, Sandelius AS. Membrane fractions from plant cells. In: Findlay JBC, Evans WH, editors. Biological Membranes. Oxford: IRL Press; 1987. pp. 37–72. [Google Scholar]
  25. Mort AJ, Qiu F, Maness NO. Determination of the pattern of methyl esterification in pectin: distribution of contiguous nonesterified residues. Carbohydr Res. 1993;247:21–35. doi: 10.1016/0008-6215(93)84238-2. [DOI] [PubMed] [Google Scholar]
  26. Munoz P, Norambuena L, Orellana A. Evidence for a UDP-glucose transporter in Golgi-apparatus derived vesicles from pea and its possible role in polysaccharide biosynthesis. Plant Physiol. 1996;112:1585–1594. doi: 10.1104/pp.112.4.1585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Neufeld EF, Feingold DS, Ilves SM, Kessler G, Hassid WZ. Phosphorylation of D-galacturonic acid by extracts from germinating seeds of Phaseolus aureus. J Biol Chem. 1961;236:3102–3105. [PubMed] [Google Scholar]
  28. Northcote DH, Pickett-Heaps JD. A function of the Golgi apparatus in polysaccharide synthesis and transport in the root-cap cells of wheat. Biochem J. 1966;98:159–167. doi: 10.1042/bj0980159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. O'Neill M, Albersheim P, Darvill A. The pectic polysaccharides of primary cell walls. In: Dey PM, editor. Methods in Plant Biochemistry. Vol. 2. London: Academic Press; 1990. pp. 415–441. [Google Scholar]
  30. Orellana A, Neckelmann G, Norambuena L. Topography and function of Golgi uridine-5′-diphosphatase from pea stems. Plant Physiol. 1997;114:99–107. doi: 10.1104/pp.114.1.99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Paulson JC, Colley KJ. Glycosyltransferases: structure, localization, and control of cell type-specific glycosylation. J Biol Chem. 1989;264:17615–17618. [PubMed] [Google Scholar]
  32. Rodgers MW, Bolwell GP. Partial purification of Golgi-bound arabinosyltransferase and two isoforms of xylosyltransferase from French bean (Phaseolus vulgaris L.) Biochem J. 1992;288:817–822. doi: 10.1042/bj2880817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Scheller HV, Doong RL, Ridley BL, Mohnen D. Pectin biosynthesis: a solubilized galacturonosyltransferase from tobacco catalyzes the transfer of galacturonic acid from UDP-galacturonic acid onto the non-reducing end of homogalacturonan. Planta. 1999;207:512–517. [Google Scholar]
  34. Schols HA, Bakx EJ, Schipper D, Voragen AGJ. A xylogalacturonan subunit present in the modified hairy regions of apple pectin. Carbohydr Res. 1995;279:265–279. [Google Scholar]
  35. Staehelin LA, Moore I. The plant Golgi apparatus: structure, functional organization and trafficking mechanisms. Annu Rev Plant Physiol Plant Mol Biol. 1995;46:261–288. [Google Scholar]
  36. Tanford C, Reynolds JA. Characterization of membrane proteins in detergent solutions. Biochim Biophys Acta. 1976;457:133–170. doi: 10.1016/0304-4157(76)90009-5. [DOI] [PubMed] [Google Scholar]
  37. Tsuji S. Molecular cloning and functional analysis of sialyltransferases. J Biochem. 1996;120:1–13. doi: 10.1093/oxfordjournals.jbchem.a021369. [DOI] [PubMed] [Google Scholar]
  38. Vannier MP, Thoiron B, Morvan C, Demarty M. Localization of methyltransferase activities throughout the endomembrane complex system of flax (Linum usitatissimum L.) hypocotyls. Biochem J. 1992;286:863–868. doi: 10.1042/bj2860863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Villemez CL, Lin T-Y, Hassid WZ. Biosynthesis of the polygalacturonic acid chain of pectin by a particulate enzyme preparation from phaseolus aureus seedlings. Proc Natl Acad Sci USA. 1965;54:1626–1632. doi: 10.1073/pnas.54.6.1626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Villemez CL, Swanson AL, Hassid WZ. Properties of a polygalacturonic acid-synthesizing enzyme system from Phaseolus aureus seedlings. Arch Biochem Biophys. 1966;116:446–452. doi: 10.1016/0003-9861(66)90051-8. [DOI] [PubMed] [Google Scholar]
  41. Weigel PH, Hascal VC, Tammi M. Hyaluronan synthases. J Biol Chem. 1997;272:13997–14000. doi: 10.1074/jbc.272.22.13997. [DOI] [PubMed] [Google Scholar]
  42. Wulff C, Norambuena L, Orellana A. GDP-fucose uptake into the Golgi apparatus during xyloglucan biosynthesis requires the activity of a transporter-like protein other than the UDP-glucose transporter. Plant Physiol. 2000;122:867–877. doi: 10.1104/pp.122.3.867. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Yu L, Mort AJ. Partial characterization of xylogalacturonans from cell walls of ripe watermelon fruit: inhibition of endopolygalacturonase activity by xylosylation. In: Visser J, Voragen AGJ, editors. Pectins and Pectinases. Amsterdam: Elsevier; 1996. pp. 79–88. [Google Scholar]
  44. Zhang GF, Staehelin LA. Functional compartmentation of the golgi apparatus of plant cells: immunocytochemical analysis of high-pressure frozen- and freeze-substituted sycamore maple suspension culture cells. Plant Physiol. 1992;99:1070–1083. doi: 10.1104/pp.99.3.1070. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES