Comparative characterization of human and rat liver glycogen synthase

ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 292, No. 2, February 1, pp. 479-486, 1992

Comparative Characterization Glycogen Synthase Sydney

A. Westphal

and Frank

of Human and Rat Liver

Q. Nuttalll

Metabolic Research Laboratory and Section of Endocrinology, Metabolism and Nutrition, Veterans Administration Medical Center, Minneapolis, Minnesota 55417; and Department of Medicine, University of Minnesota, Minneapolis, Minnesota 55455

Received July 17, 1991, and in revised form September 23, 1991

Glycogen synthase from human liver was studied, and its properties were compared with those of rat liver glycogen synthase. The rat :and human liver glycogen synthases were similar in tbeir pH profile, in their kinetic constants for the substrate UDP-glucose and the activator glucose 6-phosphate, and in their elution profiles from Q-Sepharose. The apparent molecular weight of the human synthase subunit w;as 80,000-85,000 by gel electrophoresis, which is similar to that of the rat enzyme. In addition, antibodies to rat liver glycogen synthase recognized human liver glycogen synthase, indicating that the enzymes of these two species share antigenic determinants. However, there were significant differences between the two enzymes. In particular, the total activity of the human enzyme w:as higher than that of the rat. The human enzyme, purified to near homogeneity, had a specific activity of 40 U,‘mg protein compared with 20 U/mg protein for the rat enzyme. The active forms of the rat enzyme had greater thermal stability than those of the human enzyme, but tbe human enzyme was more stable on storage in various buffers. Last, amino acid analysis indicated differences between the enzymes of the two species. Since glycogen synthase is an important enzyme in liver glycogen synthesis, the characterization of this enzyme in the human willi help provide insight regarding human liver glycogen synthesis. (o 1992 Academic PESS. IN.

solute dependence on the allosteric modifier glucose 6-P.2 However, its affinity for glucose 6-P is relatively poor, and the enzyme has little or no activity at a physiologic pH (6.7-7.0), at least in rat liver (1, 3, 4). Therefore, this form of the enzyme

is not likely

to be active

with an intermediate state of phosphorylation, referred to as synthase R (4-7). The various forms of synthase are

interconvertible

by phosphorylation

and dephosphory-

lation

potentially

by several

reactions

catalyzed

lation

has been shown to be a very dynamic

’ To whom correspondence should be addressed at Section of Endocrinology, Metabolism and Nutrition, V.A. Medical Center, One Veterans Drive (lllG), Minneapolis, MN 55417. Fax: (612)725-2093. 0003.9861/92 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

process

(10).

Whereas liver glycogen synthase has been studied extensively in different animals, there is a paucity of information on human liver glycogen synthase. In fact, all of our knowledge with respect to liver glycogen synthase and its regulation has been obtained from animals. It has been similar

step in the synthesis of glycogen is catalyzed by glycogen synthase (1). Glycogen synthase is a highly phosphorylated enzyme, and its activity is regulated by its state of phosphorylation (2). The most phosphorylated form, referred to as synthase D, has an ab-

kinases

and at least one phosphatase (8, 9). The proportion of synthase in the inactive and active forms depends on the relative activity of the kinase and phosphatase enzymes. In isolated hepatocytes, phosphorylation-dephosphory-

assumed that the human enzyme functions

The rate-limiting

in uiuo. Less

phosphorylated forms of synthase do not have an absolute dependence on glucose-6-P, but their activity is stimulated by this modifier. They also have near maximal activity at a physiological pH. The least phosphorylated form is referred to as synthase I. In liver tissue extracts, there is little synthase I present even when synthase has been activated. The major form (or forms) present is a form

to its animal

in a manner

counterparts.

In view of a lack of specific knowledge regarding this enzyme in human liver, we considered it important to characterize human liver glycogen synthase and to com-

* Abbreviations used: P, phosphate; EGTA, ethylene glycol bis(@aminoethyl ether) N,N’-tetraacetic acid; DTT, dithiothreitol; Mes, 4morpholineethanesulfonic acid; Mopso, 3-(N-morpholino)-2-hydroxypropanesulfonic acid; Mops, 4-morpholinepropanesulfonic acid; Pipes, 1,4-piperazineethanesulfonic acid; Epps, 4-(2-hydroxyethylj-l-piperazinepropanesulfonic acid; fl-GP, P-glycerol phosphate; SDS, sodium dodecyl sulfate. 479

Inc. reserved.

480

WESTPHAL

AND

NUTTALL

pare it with the rat enzyme, one of the species in which the enzyme has been thoroughly studied.

Synthase I

Synthase

D

METHODS Tissue sources. Human liver specimens from a regional organ procurement program were the primary source of human liver. This program harvests the organs of people serving as donors for organ transplantation. The specimens obtained were available either because a transplant recipient was not available or because only part of the liver was used for the transplantation. The tissue was processed by the procurement team immediately upon death by placing it in liquid nitrogen. It was transported frozen in dry ice and was stored at -80°C until needed. Initially, human liver specimens were obtained from hospital autopsies. The tissue was frozen in liquid nitrogen and stored at -80°C until needed. This was not an ideal source. Often the autopsies were not done until 12-24 h or more after death. As this was a comparative study, rat liver glycogen synthase was studied simultaneously with the human form. Rat livers were from fed male Sprague-Dawley rats weighing 180-250 g. The rats were anesthetized with intraperitoneal Seconal (50 mg/kg) and allowed a quiet period of about 15 min while being warmed with a heating lamp. As preliminary studies had shown that the human enzyme was nearly all in the inactive or D form, glucagon, which converts the enzyme essentially all to the D form, was injected intraperitoneally (100 pg/kg) into the rats 2 min prior to opening the abdominal cavity and removing the liver.

0.4’

’ 6

’ 9

0’ 6

’ 7

PH

’ 8

’ 9

PH

Enzyme purification. All procedures were performed on ice unless otherwise noted. In addition, precautions were taken to prevent the enzyme from becoming oxidized, which decreases its activity. These included bubbling nitrogen gas through the sample at every step and degassing all the buffers before their use. The liver tissue was homogenized in 50 mM P-glycerol phosphate, 5 mM EDTA, 10 mM DTT, 100 mM NaF, pH 7.6 (Buffer B), and centrifuged at 8OOOgfor 10 min. (Similar results were obtained whether or not protease inhibitors were added, therefore, we discontinued their use.) The pellet was discarded and the extract collected. Polyethylene glycol (PEG; 50%) was added to give a 6.5% solution. After 30 min, this was centrifuged at SOOOgfor 10 min. The supernatant was discarded. The pellet was collected and resuspended in Buffer B using one-third the extract volume. This was centrifuged at 104,OOOgfor 90 min. The glycogen pellet was collected, washed, resuspended in the same buffer, and recentrifuged at 104,OOOgfor 90 min. The glycogen pellet was collected and resuspended in Buffer B. The suspended glycogen and attached enzymes were incubated with salivary amylase at room temperature for l-2 h in order to digest the glycogen. It then was applied to a Q-Sepharose column equilibrated with 25 mM @-glycerol phosphate, 5 mM DTT, 2 mM EGTA, 2% glycerol, pH 7.6. The column was washed with 2 bed volumes of this column buffer, and the proteins were eluted with a step gradient of 2 bed volumes per step of the column buffer containing 0.1, 0.25, 0.4, and 0.5 M NaCl.

Conversion. Conversion of synthase D to the I + R forms was done using a glycogen particle preparation as described by Gilboe and Nuttall (11). The glycogen pellet was resuspended to one-half the supernatant volume in 50 mM imidazole, pH 7.0, after centrifugation at 104,OOOg.A 0.4.ml aliquot of resuspended glycogen particle was mixed with 0.1 ml of imidazole buffer. The conversion reaction was started by warming the mixture to 25’C. At intervals up to 60 min, 0.025-ml aliquots were withdrawn, diluted in 0.2 ml of cold stopping reagent (200 mM KF, 10 mM K,HPO,, 7.6 mM EGTA, pH 8.8, for measuring synthase D or pH 7.0 for measuring synthase I + R), and kept on ice before assay. At the end of the incubation, samples to be used for kinetic studies were diluted with KF to make a final concentration of 150 mM KF.

Western blot. Electrophoresis was done using 8% polyacrylamide gels in the presence of sodium dodecyl sulfate based on the method of Laemmli (12). The separated proteins were then either stained with silver by the method of Nielsen and Brown (13) or electrotransferred to nitrocellulose sheets (14). The transferred proteins were blotted using polyclonal antibodies raised in rabbits with rat liver glycogen synthase as the antigen (10,15). The nitrocellulose was then washed and incubated with an alkaline phosphatase-conjugated antibody to rabbit IgG. After

TABLE

Rat n = 8 Human (autopsy) n=8 Human (donor)

.

FIG. 1. Effect of pH on synthase activity. Extracts of rat and human liver were prepared as described under Methods. The I form was obtained by incubating rat and human tissue extracts at 25’C for 60 min.

Tissuepreparation. Many of the experiments were done using either a tissue extract or a glycogen particle preparation. Liver tissue was homogenized in 50 mM imidazole, 250 mM sucrose, 5 mM EGTA, 2 mM DTT, pH 7.0 (1:2, wt/vol) (Buffer A). The homogenate was centrifuged at low speed (SOOOg)for 10 min and the pellet discarded. The resulting supernatant, designated “extract,” was collected and poured through gauze. To obtain a glycogen particle preparation, the SOOOgsupernatant was centrifuged at 104,OOOgfor 30 min at 6°C. The supernatant and the microsomal layer overlying the glycogen pellet were removed. The pellet was washed, recentrifuged, and then resuspended in 50 mM imidazole, 5 mM EGTA, 2 mM DTT, pH 7.0 (resuspension buffer), using one-third of the volume used for tissue homogenization. This was used directly for assay of synthase and/or phosphorylase.

Enzyme

’ 7

6

Activity

I (U/g

wet wt)

Active

Synthase total

% Active

Active

0.11 f 0.04

0.99 + 0.02 0.38 f 0.02

11.0 X!Z0.3 7.4 f 0.6

17.0 f .9 0.4 f .02

28.0 -+ 0.4 4.0 f 0.2

70.0 f 2

0.02 IL 0.002

0.19 f 0.02

2.20 f 0.04

11.0 f 0.8

9.2 f .02

15.0 f 0.2

64.0 -+ 2

n=6

Note. One unit = pmol/min

of glucose transferred

from UDP glucose to glycogen.

Phosphorylase

total

% Active

9.1 2 0.5

HUMAN

LIVER

GLYCOGEN

481

SYNTHASE

TABLE II K, for UDPG (mM) G-6-P (mM): Synthase D Rat Human Synthases I + R Rat Human

0

0.05

32.0 f 0.7 33.0 f 0.7 1.4 + 0.05 1.5 k 0.06

0.07

0.20

7.2

17 + 0.9 20 k 1.0

11.0 + 0.5 8.9 f 0.5

0.4 f 0.01 0.3 f 0.01 0.2 f 0.03 0.2 f 0.02

1.0 * 0.02 1.1 f 0.02 K. for G-6-P (mM) 0.5

UDPG (mM): Synthase D Rat Human Synthase I + R Rat Human

1.4 1.5

f 0.5 + 0.6

6.7

0.5 f 0.03 0.3 f 0.01

0.007 f 0.001 0.014 f 0.003

Note. n = 3.

the final wash, the membrane was incubated with nitroblue tetrazolium substrate. Amino acid analysis. Highly purified lyophilized enzyme from the Q-Sepharose column was hydrolyzed by heating at 110°C for 24 h in constant boiling HCl (-6 M) in a sealed glass ampule. The resulting amino acids were determined using a Beckman 6300 amino acid analyzer. Materials. Glucose 6-P, UDP-glucose, and rabbit liver glycogen were obtained from Sigma Chemical Co. (St. Louis, MO). Glycogen was further purified by passage through a mixed bed ion-exchange resin (Amberlite MG-3) as described previously (16). [%]UDP-glucose was prepared by a method developed in our laboratory (17). Human salivary cu-amylase was purified using the method of Bernfeld up to the second ammonium sulfate step (18). All other chemicals were the highest grade available from commercial sources. Assay methods. Glycogen was determined as glucose after hydrolysis with amyloglucosidase as described by Huijing (19). Protein content was determined by the dye binding method of Bradford (20). Glycogen synthase activity was determined as indicated in Ref. (4). Total synthase activity is measured at pH 8.5 in the presence of a saturating concentration of glucose 6-P and a near saturating concentration of UDP-glucose. Synthase R activity is measured at pH 7.0 at the same UDPG concentration but in the absence of glucose 6-P. Glycogen phosphorylase a and total phosphorylase were determined by the method of Gilboe et al. (21) as modified by Tan and Nuttall for the liver enzyme (22). One unit of activity for both synthase and phosphorylase is defined as 1 pmol of [i4C]glucose incorporated into glycogen per minute.

RESULTS

Glycogen content. The glycogen content was similar in the rat and human livers. In the rats, it was 23.3 + 1 mg/g liver (n = 3). In the human autopsy specimens, it was 20.7 f 2 mg/g liver (1%= 3), and in the human donor specimens, it was 23.7 f 2 mg/g liver (n = 3). Total glycogen Synthme and phmphmylase activities. synthase activity in the human donor liver was nearly 2.5

times that in the rat on a gram liver basis (Table I). In both tissues, the proportion in the active form was about 11%. Total phosphorylase activity was considerably less in human liver. In fact, it was only 54% of that in rat liver. The proportion in the active form again was similar in the two tissues. Thus, while the synthase activity of the human donor livers was twice that of the rat, the phosphorylase activity was only one-half that of the human liver. In the human autopsy liver, the synthase activity was very low (0.38 U/g wet wt). Phosphorylase activity also was low. Therefore, further studies were not done using this tissue source. pH profiles. A pH profile of synthase D activity was done using dilute extracts from both human and liver tissues (Fig. 1, right). The pH profiles were very similar. As reported previously, the pH maximum was approximately 8.8 in both species and in both there was little activity at pH 7.0. The activity measured could be accounted for by the small amount of synthase R present. The pH profile for the I form was determined by incubating dilute rat and human tissue extracts at 25°C for 60 min. This allows endogenous phosphatase activity to convert synthase largely to its least phosphorylated form. The I form of both the rat and human enzyme showed a broad pH optimum, from 7 to 9, and the profiles again were very similar (Fig. 1, left). As with the D form, the activity of the human enzyme was greater than that for the rat, as expected. We did not attempt to determine the profile for the R form (4). However, it is likely to be similar as the profiles for the I and D forms were the same. Kinetics. The kinetic constants for the D and I + R forms of the enzyme were determined from Lineweaver-

482

WESTPHAL

Svnthase

AND

NUTTALL

Synthase

D

I 0 Rat . Human

0 Rat . Human 80 60 40 -

I 40 Temperature

Svnthase

40 Temperature

60

50 (“C)

Synthase

D

60

50 (“C)

I

120 100 l

Human

0 Rat l Human

80 /

40 Temperature

50 (“C)

60

40 Temperature

50 (“C)

60

FIG. 2. Effect of temperature on activity (A and C) and stability (B and D) of glycogen synthases D and I. Glycogen particles of rat and human were prepared as described under Methods. The predominantly I form was obtained as in the legend to Fig. 1. To determine the effect of temperature on enzyme activity, the samples were incubated at 30, 40, 50, and 60°C for 30 s and then assayed at their incubation temperature (A and C). To determine the effect of temperature on enzyme stability, the samples were incubated at 30, 40, 50, and 60°C for 60 min, allowed to cool, and then assayed at the standard assay temperature of 30°C (B and D).

Burk plots using initial rate conditions (Table II). For the D form, in the absence of glucose 6-P, the apparent K,,, was 32 mM for the rat and 33 mM for the human. The addition of 0.07 mM glucose 6-P lowered the apparent K,,, to about 17 mM for the rat and 20 mM for the human. At 0.20 mM glucose 6-P, it was reduced to 11 mM for the rat and 9 mM for the human. At a saturating concentration of glucose 6-P (7.2 mM), the Km was reduced to 0.4 mM for the rat and 0.3 mM for the human. Thus, the apparent K,,,‘s of synthase D for the rat and human were essentially the same and were strongly dependent on the glucose 6P concentration. The K, for glucose 6-P was determined in the presence of 0.5 and 6.7 mM UDP-glucose (Table II). At 0.5 mM UDP-glucose, the apparent K, was 1.4 mM for the rat and 1.5 mM for the human enzyme. In the presence of 6.7 mM UDP-glucose, the apparent K, was reduced to 0.5 and 0.3 mM for the rat and for the human, respectively. Thus, the apparent K,‘s for rat and human were essentially the same and responded to changes in UDP-glucose in the same fashion. The physiologic concentration of glucose 6-P in liver is about 0.1 mM, while that of UDP-glucose is about 0.4 mM. Thus, these results indicate that synthase D should have little activity in uivo in either species. In the absence of glucose 6-P, the apparent Km for the active forms of synthase was 1.4 mM for the rat and 1.5

mM for the human. Addition of 0.05 mM glucose 6-P lowered this to about 1 mM for both species; 7.2 mM glucose 6-P lowered it further to about 0.2 mM (Table II). Thus, again, the kinetic constants of the rat and human were essentially identical. The apparent K, for glucose 6-P of synthase I at the approximate physiologic concentration of 0.5 mM UDP-glucose was difficult to determine. It was estimated to be about 0.01 mM for the enzyme from each species. Stability. The temperature dependence of the synthase D reaction was determined using glycogen particle preparations. The activity was similar at both 30 and 40°C for enzyme from both species. They were essentially inactivated at 50°C (Fig. 2A). Enzyme stability was determined by incubating the preparations for 60 min at the temperature indicated. They were then cooled and assayed at 30°C. There was a sharp decline in stability at 50°C for both (Fig. 2B). With respect to the active forms of synthase (I + R), activity was less at 40°C than at 30°C for the rat enzyme but there was little further change at 50 and 60°C (Fig. 2C). At higher temperatures, the human enzyme was much more labile than the rat enzyme. At 50 and 60°C the activity was very low. The rat enzyme also was somewhat more stable than the human enzyme at 50°C or greater when incubated for a longer period of time (Figs. 2D). In

HUMAN

LIVER

GLYCOGEN

fact, it was more stable than the D form from either species. The stability of rat and human synthase activity in glycogen particle preparations (1:l dilution) also was determined in the presence of different buffers at 1 h at room temperature and at 4°C for 3 days, and 7 days (Table III). At 7 days, in pH 7.0, 50 InM P-glycerol phosphate, P04, Mes, Mopso, and Mops, the mean human synthase activity as a percentage of baseline was 85 compared with that of the rat which was 72. In 50 mM pH 7.0, Tes, imidazole, glycylglycine, or Tris, the difference was striking. The human enzyme had a mean of 81% of its baseline activity whereas the rat had 11%. In pH 7.0, 50 mM piperazine buffers, Pipes, Hepes, and Epps, the enzyme from both species was much less stable. For the human enzyme, the mean activity was 14% of its baseline activity. For the rat, it was only 1%. Thus, generally, the human form appeared to be more stable than the rat form in these buffers. Enzyme purification. Liver synthase was purified as indicated under Methods. Comparative elution profiles from the Q-Sepharose column obtained from rat and human preparations are shown in Fig. 3. For both species, the phosphorylase was eluted from the column with 0.1 M NaCl, and the synthase was eluted with 0.4 M NaCl. The phosphorylase peak of the human was smaller than that of the rat, while the synthase peak of the human was greater than that of the rat, as expected. Human phosphorylase eluted slightly ahead of the rat phosphorylase; otherwise, the elution profiles were identical. The specific activity and recovery of human synthase at each step of the purification are shown in Table IV. Total recovery was 15%. The specific activity increased

TABLE Activity

10000

6000 c

1

Human

Mean

133 122 111 134 126 125

Mean

PGP PO4 Mes Mopso Mops Tes Imidazole Glycylglycine Tris Pipes Hepes EPP~

;

.

R phosphotylase

6000 w-s-1

H phosphovlase

4000

2000

0

0

lb

3b 26 Column Fraction

4’0

FIG. 3. Elution profile of enzymes from rat and human liver preparations. The liver tissue was prepared as described under Methods and was applied to a column of Q-Sepharose. The column was eluted with a step gradient of buffer containing 0.1, 0.25, 0.40, and 0.50 M NaCl (broken line) as described under Methods. Fractions were collected from the column and assayed for glycogen phosphorylase and glycogen synthase activity. R, rat; H, human.

4000-fold from 0.01 U/mg of protein in the extract to 40 U/mg of protein from the column. The rat preparation in these experiments had a specific activity of 15-20 U/ mg protein. Thus, the highly purified human enzyme had a specific activity that was twice that of the rat enzyme. The synthase enzyme isolated from the column was quite pure as judged by SDS-polyacrylamide gel electrophoresis and silver staining (Fig. 4). The apparent molecular mass was 80,000-85,000. Western blots of the gel confirmed that this was glycogen synthase (Fig. 5). This

III

as % Baseline I days

Rat

Human

Rat

100 101 95 100 97

167 137 139 137 131 142

93 90 100 99 105 97

87 113 58 51 49 72

135 106 117 108 117

110 94 94 98 99

106 67 58 44 69

108 87 87 88 93

23 10 6

138 118

96 94 96 95

10 5 106 40

92 67 39 66

129

Mean

II

3 days

Rat

0.5

I I

lh Buffer

483

SYNTHASE

128

90

5 11 1 0 3 1

484

WESTPHAL

AND

NUTTALL

TABLE IV

Extract Extract pellet Glycogen pellet Glycogen pellet Q-Sepharose

Protein (mdml)

Total units

42.3 28.8 1.8 0.9 0.004

66 40 32 23 10

Specific activity W/w)

Recovery

0.01 0.04 0.45 0.99 40.00

100% 61% 48% 35% 15%

also indicated that antiserum raised against rat liver glycogen synthase recognized the enzyme from human liver. The amino acid composition of a single purified human glycogen synthase preparation was compared with those deduced from the rat liver synthase cDNA (23) and from rabbit liver (Table V). The only major differences in mole

FIG. 5.

Western blot analysis of human liver glycogen synthase. See Methods for procedure details.

ratios observed were for cysteic acid, glycine, methionine, and tyrosine. DISCUSSION

FIG. 4. Polyacrylamide gel electrophoresis of human liver glycogen phosphorylase and glycogen synthase. Glycogen phosphorylase (left lane) and glycogen synthase (right lane) obtained from the Q-Sepharose elution were analyzed by 8% polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and stained with silver as described under Methods. Standard proteins were phosphorylase b (95,500), glutamate dehydrogenase (55,000), and ovalbumin (43,000). Note: on SDS-polyacrylamide gel electrophoresis synthase usually is present as two closely associated bands.

We (4,23,24), as well as others (25-27), have previously characterized liver glycogen synthase from rats and rabbits. We also have studied the regulation of the rat enzyme in uiuo and in isolated hepatocyte preparations (1, 24). To our knowledge, the human liver enzyme has not been characterized nor has its regulation been studied. In the present study, we have directly compared several characteristics of the human liver enzyme with that of the rat liver enzyme. Human liver obtained at autopsy was found to be an unsuitable source. Living donor liver was an excellent source. Unfortunately, only very limited amounts of tissue were available from the human organ donor program. The latter tissue was used for characterizing the enzyme. The human and rat liver enzymes were found to be similar in several respects. The kinetic constants for the substrate, UDP-glucose, and for the modifier, glucose 6P, were essentially identical for both the D (most phosphorylated) and the I (least phosphorylated) forms of the enzyme (Table II) as were the pH profiles for both forms

HUMAN TABLE Amino

Amino acid Cysteic acid Aspartate Asparagine Threonine Serine Glutamate Glutamine Proline Glycine Alanine Valine Methionine Isoleucine Leucine Tyrosine Phenylalanine Histidine Lysine Arginine Tryptophan

Acid Mole

LIVER

GLYCOGEN

V Ratios

(%)

Human liver (mol%)

Rabbit liver (mol%)

Rat liver (mol%)

0.6 11.1

1.2 11.1

6.6 7.7 10.0

5.4 4.3 9.4 8.4 7.2 6.4 5.1 2.5 3.9 8.9 1.8 3.7 2.3 5.5 5.9

1.9 6.4 4.4 6.3 6.7 6.3 3.0 5.1 6.1 5.3 6.1 1.9 5.3 8.8 3.6 5.6 3.7 5.6 6.4 1.7

6.3 10.9 6.8 8.0 0.7 3.8 9.4 0.0 5.4 2.2 3.3 8.5

Note. Preparation of the protein for amino acid analysis results in asparagine and glutamine being deamidated. They are measured as aspartate and glutamate, respectively. Tryptophan also is destroyed. With the method used, the cysteic acid content (cysteine) also is somewhat unreliable. There were no cysteine residues present.

(Fig. 1). As with the D form of the rat enzyme, the poor affinity for UDP-glucose and glucose 6-P, in association with the near absence of activity at a physiological pH (7.0), strongly suggest that the D form, in both rats and humans, has little catalytic activity in. uiuo unless an as yet unidentified allosteric activator is present. Synthase R, a form having a phosphorylation state intermediate between the D and I forms, has been shown to be the major active form in uiuo in the rat. This form cannot be isolated as such. However, liver extracts from rats enriched in synthase R have been studied and the kinetic constants and pH profile of this form determined (4). This was not done for the human enzyme but since the kinetic and pH profiles for the D and I forms were similar to that of the rat enzyme, it is lilkely that they would be similar for the R form as well. As pointed out by Mersmann and Segal(30), the I form of synthase in rat liver is not independent of glucose 6-P for activity in uiuo. This also is true for the R form. The concentration of glucose 6-P will determine the affinity of the active forms of the enzyme for its substrate UDPglucose in both species. In the absence of glucose 6-P, the K,,, for UDP-glucose is considerably higher than the intracellular concentration of this substrate for all forms of the enzyme. The human enzyme elution profile from a Q-Sepharose column was the same as that for the rat enzyme.

SYNTHASE

485

On SDS-gel electrophoresis, two broad bands were identified with an M, of approximately 85 and 80 kDa. This also was similar to that for the rat enzyme. The reason two bands are present is not clear. It may be due to different phosphorylation states of the enzyme (2). More likely, it is due to different conformational states of the protein (31). The C-terminal region is hydrophilic, contains several phosphorylation sites, and may exist in a somewhat extended form remote from the remaining core protein. This could result in an aberrant behavior of the entire protein on SDS-gel electrophoresis (23). It is likely not due to proteolysis. Use of a variety of proteolytic enzyme inhibitors during tissue processing has not resulted in maintenance of only a single band on electrophoresis. It also is not likely to be due to different gene products. A single gene appears to be present in rat liver, and alternate gene processing has not been observed (23). Western blots using polyclonal antibodies raised in rabbits using rat liver synthase as antigen indicated that the human enzyme shares epitopes with the rat enzyme and suggests a considerable degree of structural homology between the two. However, several major differences in the characteristics of the rat and human liver were identified. First, the total synthase activity in human liver was about twice as great on a tissue weight basis as that in rat liver. Our data indicate that this is due to a greater specific activity of the human enzyme, i.e., greater catalytic efficiency, although this needs to be confirmed more rigorously. Second, the I form, but not the D form, of the rat liver enzyme had greater thermal stability than the human enzyme (Fig. 2). Thus, the rat I form is likely to have a more rigid tertiary structure than the human enzyme. We previously have speculated that the least phosphorylated form of the rat and rabbit enzymes should have a less extended tail region and thus should be more stable (2, 24). The thermal stability data for the I and D forms are compatible with this. However, it would appear to be less true for the human enzyme. The stability of the rat and human enzymatic activity when stored at 4°C at pH 7.0 in various buffers also was different. In general, the human enzyme activity was more stable. This was particularly apparent when the buffers were Tes, Tris, imidazole, or glycylglycine (Table III). The reason for the difference is unclear, but it again indicates a difference in structure of the enzyme from the two species. Last, although amino acid analysis could only be done on one human synthase preparation, it suggests modest differences in the mole ratio of some amino acids when compared with the amino acid content of the rat liver enzyme deduced from the rat liver cDNA (23) (Table V). Further characterization of the human enzyme will be facilitated by isolation of the cDNA for the human enzyme and deduction of the amino acid sequence.

486

WESTPHAL

AND

ACKNOWLEDGMENTS Supported by Merit Review Funds to F.Q.N. from the Department of Veteran Affairs; and in part by the Liver Tissue Procurement And Distribution System (LTPADS), Grant NOl-DK-6-2270. The authors thank Claudia Durand for typing the manuscript. Dr. Westphal was a Fellow in Endocrinology and Metabolism.

REFERENCES 1. Nuttall, F. Q., Gilboe, D. P., Tan, A. W. H., Doorneweed, D. D., Theen, J. W., Gannon, M. C., and Chou, B. B. (1981) in The Regulation of Carbohydrate Formation and Utilization in Mammals (Veneziale, C. M., Ed.) pp. 315-343, University Park Press, Baltimore. 2. Tan, A. W. H., and Nuttall, F. Q. (1983) J. Biol. Chem. 258,96249630. 3. Hizukuri, S., and Larner, J. (1964) Biochemistry 3, 1783-1788. 4. Nuttall, F. Q., and Gannon, M. C. (1989) Anal. Biochem. 178,311319. 5. Tan, A. W. H. (1981) Biochem. J. 200,169-172. 6. Tan, A. W. H. (1982) J. Biol. Chem. 257(g), 5004-5007. I. Esmann, V., Solling, H., and Juhl, H. (1981) Znt. J. Biochem. 13(11), 1123-1126. A. A., and Roach, P. J. 8. Camici, M., Ahmed, Z., DePaoli-Roach, (1984) J. Biol. Chem. 259(4), 2466-2473. 9. Schelling, D., Leader, D. P., Zammit, V. A., and Cohen, P. (1988) Biochem. Biophys. Acta 927,221-231. 10. Tan, A. W. H., and Nuttall, F. Q. (1985) J. Biol. Chem. 260(8), 4751-4757. 11. Gilboe, D. P., and Nuttall, F. Q. (1986) Arch. Biochem. Biophys. 249,34445. 12. Laemmli, U. K. (1970) Nature 227, 680-685. 13. Nielsen, B. L., and Brown, L. R. (1984) Anal. Biochem. 141, 311315.

NUTTALL

14. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76(g), 4350-4354. 15. Renart, J., Reiser, J., and Stark, G. R. (1979) Proc. Natl. Acad. Sci. USA 76,3116. 16. Huijing, F., Nuttall, F. Q., Villar-Palasi, L., and Larner, J. (1969) Biochim. Biophys. Acta 177, 204-212. 17. Tan, A. W. H. (1979) Biochim. Biophys. Acta 582, 543-547. 18. Bernfeld. P.. and Nisselbaum. 860. ’ ’ 19. Huijing, 20. Bradford,

J. S. (1956) J. Biol. C&em. 220. R51Y-L

F. A. (1970) Clin. Chim. Acta 30, 567-572. M. M. (1976) Anal. Biochem. 72, 248-254.

21. Gilboe, D. P., Larson, K. L., and Nuttall, 47(l), 20-27. 22. Tan, A. W. H., and Nuttall,

F. Q. (1972) Anal. Biochem.

F. Q. (1975) Biochim.

Biophys. Acta

410,45-60. 23. Bai, G., Zhang, Z., Werner, R., Nuttall, F. Q., Tan, A. W. H., and Lee, E. Y. C. (1990) J. Biol. Chem. 265(14), 7843-7848. 24. Nuttall, F. Q., Gilboe, D. P., Gannon, M. C., Niewoehner, C. B., and Tan, A. W. H. (1988) Am. J. Med. 85(Suppl. 5A), 77-85. 25. McVerry, P. H., and Kim, K. H. (1974) Biochemistry 13(17), 35055 3511. 26. Killilea, S. D., and Whelan, W. J. (1976) Biochemistry 15(6), 13491356. 27. Jett, M. F., and Soderling, 6743-6745.

T. R. (1979) J. Biol. Chem. 254(14),

28. Lin, D. C., and Segal, H. L. (1973) J. Biol. Chem. 248,7007-7011. 29. Camici, M., DePaoli-Roach, A. A., and Roach, P. J. (1984) J. Biol. Chem. 259, 3429-3434. 30. Mersmann, H. J., and Segal, H. L. (1967) Proc. Natl. Acad. Sci. USA 58, 1688-1695. 31. Francis, R. T., Bailey, T. J., and Becker, R. R. (1985) Camp. Biochem. Physiol. B 81, 199-206.