Use of an Antibody Probe to Study Regulation of Glycogen Phosphorylase by Its NH,-terminal Region*

Antibodies that are specific for the NHz-terminal region of rabbit muscle glycogen phosphorylase were isolated. Studies, using synthetic peptides representing different segments of the NHz-terminal region of muscle phosphorylase, indicated the antibodies are highly specific for the first 4 NHz-terminal residues of the enzyme. The molecular weight of the complex formed between dimeric phosphorylase and the antibodies estimated by gel filtration suggests that only 1 molecule of antibody binds per dimer of phosphorylase. The antibodies were strongly inhibitory to both phosphorylase kinase and phosphorylase phosphatase. Apparent binding con- stants for glucose l-phosphate and AMP and inhibition by compounds that bind at or near the glucose l-phos- phate and AMP sites were not affected by the antibod- ies. The apparent K,,, for the high molecular weight substrate, glycogen, was lowered 2-fold by the presence of the antibodies. The primary binding site for malto-heptaose, for has Fletterick,


Antibodies
that are specific for the NHz-terminal region of rabbit muscle glycogen phosphorylase were isolated.
Studies, using synthetic peptides representing different segments of the NHz-terminal region of muscle phosphorylase, indicated the antibodies are highly specific for the first 4 NHz-terminal residues of the enzyme. The molecular weight of the complex formed between dimeric phosphorylase and the antibodies estimated by gel filtration suggests that only 1 molecule of antibody binds per dimer of phosphorylase. The antibodies were strongly inhibitory to both phosphorylase kinase and phosphorylase phosphatase. Apparent binding constants for glucose l-phosphate and AMP and inhibition by compounds that bind at or near the glucose l-phosphate and AMP sites were not affected by the antibodies. The apparent K,,, for the high molecular weight substrate, glycogen, was lowered 2-fold by the presence of the antibodies.
The primary binding site for maltoheptaose, and presumably for glycogen, recently has been shown to be a site separate from the active site (Kasvinsky, P. J., Madsen, N. B., Fletterick, R. J., and Sygusch, J. (1978) J. Biol. Chem 253, 1290-1296). The improved binding affinity for glycogen, induced by the antibodies, is consistent with regulation of this glycogen site by the NHz-terminal region.
The binding of the antibodies to phosphorylase b completely stabilized the enzyme to loss of its cofactor, pyridoxal 5'-phosphate, under conditions in which the cofactor is normally completely resolved. Because the antibodies did not affect the apparent binding affinities for compounds (glucose, glucose l-phosphate, and caffeine) that bind in the same hydrophobic active site crevice as pyridoxal phosphate, the suggestion is made that the dramatic effect of the antibodies on the pyridoxal5'-phosphate site is quite specific.
The specific antibodies against muscle phosphorylase were able to bind to the liver isozyme of phosphorylase. When antibodies were bound to the liver isozyme, the apparent affinity (K,,,) for glucose l-phosphate was improved by 4.1-fold at saturating AMP. At concentrations of glucose l-phosphate lower than the Km, the antibodies increased enzyme activity by more than lofold. The conclusion is made that the structural character of the NH&erminal regions of liver and muscle * This work was supported by Research Grant GM-09587 and GM-06474.01 from the National Institutes of Health, United States Public Health Service. This is Journal Paper ,J-9269 of the Iowa Agriculture and Home Economics Experiment Station, Ames, IA; Project 2120. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
4 Present address, Laboratory of Metabolism, National Institute of Alcohol Abuse and Alcoholism, 12501 Washington Ave., Rockville, MI) 20852.
5 To whom requests for reprints should be addressed. phosphorylase b isozymes may be, at least partially, responsible for their differing affinities for glucose lphosphate.
The NHa-terminal region of rabbit skeletal muscle glycogen phosphorylase is a primary locus for dictating the catalytic potential of this enzyme. The physical and catalytic properties of the enzyme are heavily dependent on whether serine-14 is phosphorylated or dephosphorylated (1). A variety of approaches has been utilized in attempts to gain some understanding of the mechanism involved in regulation by phosphorylation.
Comparisons of the physical parameters and structural features of phosphorylase a (phosphorylated form) and phosphorylase b (dephosphorylated form) have indicated only subtle differences in the monomeric unit of these two forms (1). Fletterick et al. (2) used a difference Fourier electron density map to compare the crystalline structures of phosphorylase a and b. They found only localized differences in the protein chain. The most dramatic structural difference was in the NHz-terminal segment itself. Although the NH?terminal region of phosphorylase a is easily discernible by xray diffraction (3), the first 17 NHa-terminal residues of phosphorylase b are not visible on the x-ray map (4). Helmreich and co-workers used an immunochemical approach to the study of the regulatory phenomena of glycogen phosphorylase (5,6). The antibodies used in their experiments were directed against many antigenic determinants on phosphorylase; thus, the effects of antibodies on properties of phosphorylase could not be related to their binding at specific regions of the enzyme molecule.
Utilization of antibodies that bind specifically to a given region of an enzyme molecule should greatly improve the usefulness of an immunochemical approach to studying regulation of enzymes. Antibodies to specific regions of protein molecules have been used to predict structural conformation (7). In an earlier publication, we described the preparation of antibodies that specifically interact with the NH?-terminal region of muscle phosphorylase (8). In the present communication, we describe the use of this antibody probe for the NHn-terminal region to study the involvement of this important site in regulating the properties of the enzyme.
It was our hope that the binding of the antibody would induce changes in those physical and catalytic properties most intimately controlled by the NHy-terminal region.
Antibodies have been an important tool for studying structural homology of different glycogen phosphorylases (9-15). There is a wide range in the degree of immunochemical crossreactivity between glycogen phosphorylases from different species (g-12), as well as phosphorylases from different organs of the same species (13-15). In spite of major differences in immunochemical, in addition to physical and catalytic, prop-  (22). The most dramatic effect of the specific antibody on the liver isozyme was on its AMP-dependent activity.

EXPERIMENTAL PROCEDURES
Pre~~~ration of Anlihorf~-Hyperinlnlurle serum, containing antiphosphorylase n antibodies, was obtained from the same goat used in a previous study (8). After the initial immunization (81, the goat was boosted with soluble phosphorylase (I six times over a pcrlod of 7 months. A 2000-ml blood sample was extracted at the end of this period; experiments described in this paper were performed by using antibodies isolated from this blood sample. The y-globulin fraction was prepared (8), and this y-globulin fraction should contain primarily IgG because it was isolated from hyperimmune serum (2X). Antihodies, anti-(I-ll))p, specific for the NH?-terminal region, were isolated from the y-globulin fraction by affinity chromatogra~~h~ on a column (1.3 x 21 cm) of Sepharosr 4B (I'harmacia) contammg covalentI> bound phosphopeptide, (l-18)1) (8). The peptide, (l-18)1). corresponds to the first 18 NH2-terminal residues of phosphorylase u and has thr following amino acid sequence: NHL-Ser.Arg-Pro-I,eu-Srr-Asl,-(;lrl-Glu-Lys-Arg-Lys-Gln-Ile-Ser(I'~~)-Val-Arg-G~y-Leu-COOH.
A gradient of 0.0 to 4.0 M guanidine HCI nas used to elute anti-(I-1X)p from the affinity resin. Two major protein components were eluted from the column, and both were dialyzed against 0.15 M NaCl to remove guanidine HCI and stored III the freezer. The first (1.1 to 2.3 M guanidine HCl) contained 8% of the eluted anti-phosphorylase a activity, as measured by radioimmunoassay (8), and 55% of the eluted protein. The second (3.0 to 3.6 M guanidine HCl) contained only 8': of the eluted anti-phosphorylase c~ activity and 26'; of the eluted protein. The first component, anti-(I-18)p, was used for the experiments described in this paper because of its greater specific immunochemical activity.
Normal goat u-globulin was obtained from a goat that had not heen injected with phosphorylase.
Antibody was prepared against goat yglobulin in rabbit (rabbit anti-goat y-globulin) as previously described (8). Preparation of Enzymes-Rabbit skeletal muscle glycogcn phosphorylase h and rabbit liver glycogen phosphorylase h were isolated as by Fischer and Krebs (24) and Appleman et (I/. (21), respectively. Muscle phosphorylase n and liver phosphorylase a were prepared from their respective phosphorylase h forms by phosphorylation with rabbit muscle phosphorylase kinase (25). [ "I'lPhosphorylase u was obtained in an identical manner except [ "P]ATl', prepared by the method of Glynn and Chappell (X), was used as substrate for kinase. I'hosphorylase h' was obtained from rabbit muscle phosphorylase u by limited proteolysis with trypsin (27). Rabbit muscle phosphorylase kinase and rabbit liver phosphorylase phosphatase were purified as by Brostrom et nl. (28) and Brandt ct a/. (29), respectively. The protein concentration of phosphorylase was measured spectrophotometrically by using the extinction coefficient Ej;;,,,,,, at 280 nm of 13.2 (30). For calculation of moles of phosphorylase, a monomer of phosphorylase was assumed to have a molecular weight of 97,400 (31).
Assuy of Clycogen 1Dhos~horyl[l,se-The catalytic activity of glycogen phosphorylase was determined in the direction of glgcogen synthesis by measuring incorporation of "C from [IY-"C]glucosr Iphosphate (Amersham) into glycogen. A filter paper assay similar to that described by Parrish et al. (32) was employed. After the desired period of incubation, an aliquot of the reaction mixture was removed spotted on a piece (2 X 2 cm) of Whatman 1'81 cellulosr phosphatrl paper, and the glvcogen was precipitated on the paper by immediate transfer to a bath of ice cold 60"; (v/v) ethanol. The papers wer'ti washed in a second bath of 6(X (v/v) ethanol, drird, and ass;iyrtl fat radioactivity.
Normal goat v-globulin, isolated from a goat not ~njrcted with phosphorylase, was added to stabilize phosphoryhisc act ivith when the concentration of phosphorylase was less than I &tld. Normal goat y-globulin was also added in place in anti-( I-lH)p in control experiments. anti-(1-18)p to Ac(l-18)~. In addition, Ac( l-18)p caused 90 to 1OOV inhibition of binding of anti-( l-18)p to phosphorylase n and h. Therefore, anti-(l-18)p is highly specific for determinants located within the NHZ-terminal region. 'l'he amount of binding of anti-( 1-18)~ to phosphorylase and peptides was not affected by 2-mercaptoethanol, Titers for 1gM are known to be sensitive to 2-mercaptoethanol, but titers for IgG havr been shown previously to be insensitive to 2-nlercal)toethanol (33); thus, anti-(1-18)p is most likely IgG. Precipitin c'otnplexes of anti-(l-18)p with either phosphorylase or peptitles were never observed.
[,'H]Ac(l-18)p is the same as peptide Ac(l-18)~ except its N-acetylated NH, terminus was labeled with tritiated acetic anhydride, as previously described (8). The NH? terminus of native phosphorylase is N-acetylated (16). For the first preparation of anti-(1-18)p, data suggested that a phosphate group on the NH?-terminal serine-14 was not necessary for interaction with anti-( l-18)p (8). This UXelusion also can be made for the new preparation of anti-(l-18)~ from the data in Table I The acetylated analog of peptide (5-18)p, Ac(5-18)p, was also a very poor inhibitor.
An equally high concentration of another short peptide fragment, peptide Ac(9-II)Gly, was not a good inhibitor. An experiment was conducted to ascertain whether residues 1 to 4 also are important in the binding of anti-(1-18)p to phosphorylase (Table II). Peptide (5-18)~ was ineffective, relative to Ac(I-18)p, in displacing ['"Plphosphorylase from anti-(1-18)p.
Peptide Ac(l-4) was able to cause significant displacement.
From these studies, we concluded that anti-(l-18)~ has a high degree of specificity for the fast 4 NHzterminal residues.
Complex-At the concentrations (3 pg/ ml) of phosphorylase used to study the effects of anti-(l-18)p on the properties of the muscle enzyme, both phosphorylase b and phosphorylase a were present as dimers (35). It is possible that anti-(1-18)p could bind to both NH2-terminal regions of a phosphorylase dimer. In addition, each molecule of anti-(1-18)p has two combining sites for interaction with phosphorylase.
Therefore, high molecular weight complexes of anti-(1-18)p and phosphorylase could form. Experiments were conducted to estimate the molecular weight of these complexes ( Fig. 1). ["'PlPhosphorylase a was preincubated with anti-(1-18)~ under conditions similar to those used to study the effects of anti-(1-18)p on the properties of the enzyme. The preincubation mixture was chromatographed on a Sepharose 4B column to estimate the molecular weight of the complexes generated.
Of the radioactivity that was applied, 88% was recovered. A small amount (7.9% of recovered radioactivity) of high molecular weight material was eluted as a peak just after the void volume (molecular weight greater than 10'). A peak containing 71% of the recovered radioactivity was eluted in a position corresponding to a relatively low molecular weight. Arrow B, Fig. 1  In the absence of inhibiting /3-glycerophosphate, and 0.03 M 2-mercaptoethanol.
[3H]Ac(l-18)p was prepared as described in a previous communication (8). After incubation for 30 min at 3O"C, 0.75 mg (10 ~1) of rabbit anti-goat y-globulin was added and the mixture was incubated at 4°C for 1 day. The amount of radioactivity in the precipitates was determined as described in Table I.

Abbreviation
Sequence nmol % 1 2 3 456 78 9 10 11 12 13 14 15 16 17 18 [""P]Phosphorylase a (5 pg, 12,000 cpm) was preincubated for 10 min at room temperature with 240 pg of anti-(l-18)p in 75 mM NaCl and column buffer (0.25 mg/ml of normal goat y-globulin, 40 mM ,L?-glycerophosphate, 30 mM 2-mercaptoethanol, pH 6.8). The preincubation mixture (0.40 ml) was applied to a Sepharose 4B (Pharmacia) column (0.7 x 56 cm) equilibrated at room temperature in column buffer. Fractions (0.4 ml) were collected, immediately, at a flow rate of 6 ml/h. Arrow B indicates the position of elution of the peak of radioactivity when normal goat y-globulin was substituted for anti-(1-18)p in the sample mixture. Arrow B also indicates the position of elution of radioactivity when no y-globulin was present in either the sample mixture or the column buffer. Arrow A indicates the position of elution of the peak of radioactivity when 930 pg of phosphorylase a was substituted for 5 pg of phosphorylase a and when normal goat y-globulin was substituted for anti-(I-l8)p in the sample mixture. The above experiments were repeated at least three times each, and the elution profiles were always the same. phosphorylase a. The major peak of radioactivity, representing the complex of anti-(1-18)p and dimeric phosphorylase a, eluted at a position corresponding to a molecular weight slightly less than the molecular weight of a tetramer of phosphorylase a. Since anti-(1-18)p is probably IgG (Mr = 160,000), it is likely this peak represents a complex of one dimer of phosphorylase a and 1 molecule of anti-(1-18)p. Addition of larger quantities of anti-(1-18)p to the same quantity of phosphorylase a used in Fig. 1 did not change the positioning of this major peak of radioactivity.

Resolution of Pyridoxal
ii'-Phosphate-Pyridoxal 5'-phosphate can be resolved from phosphorylase b in the presence of the deforming buffer, imidazole, and the trapping agent, cysteine (36). The effect of anti-(1-18)p on resolution is presented in Fig. 2  Step 1: preincubation with anti-(I-18)p; a preincubation mixture (0.3 ml, pH 6.8), containing 6.0 pg of phosphorylase b and either (i) 240 pg of anti-(1-18)p and 500 pg of normal goat y-globulin or (ii) 740 pg of normal goat y-globulin, in 13 mM P-glycerophosphate, 10 mM 2mercaptoethanol, and 100 mM NaCl, was incubated 10 min at 30°C.
To completely prevent resolution of 6 pg of phosphorylase 6, 120 pg of antibody was required (data are not presented). When lesser amounts of anti-(1-18)p were used, some resolution was observed. Assuming anti-( l-18)p is primarily IgG (M, = 160,000), this represents a molar ratio of anti-(1-18)p to phosphorylase (M,. = 97,400) of 12.5. The molar ratio of anti- (1-18)~ to phosphorylase in the experiments described by Fig.  2 (1)  activity in the absence of AMP (i.e. phosphorylase a activity) was measured by addition of 50 ;LI of a solution, containing 2.W glycogen and 32 mM [ IJ-"Clglucose l-phosphate, and incubation for 15 min at 30°C. Incorporation of '"C into glycogen was determined as described in the text. B, a preincubation mixture (20 ~1, pH 7.4) containing [ "I'lphosphorylase n (0.3 pg, 20,000 cpm), 25 mM Tris, 0.5 mM dithioerythritol, 75 mM NaCI, and either (i) 12 pg of anti-(1-18)~ and IO lg of normal goat y-globulin (0) or (ii) 22 wg of normal goat y-globulin (O), was incubated at 30°C. After 30 min. a solution (IO ~1, pH 7.4), containing 0.016 /Lg of phosphorylase phosphatase, 0.25 mg/ml of normal goat y-globulin, 25 mM Tris, and 0.5 mM dithioerythritol, was added and incubated at 30°C. After different periods of dephosphorylation, the protein was precipitated by addition 10 ~1 of 15'; trichloroacetic acid and 10 ~1 of 1 mg/ml bovine serum albumin, and the amount of acid-soluble "I' was measured.
of' Fig. 3H, dephosphorylation was more than 90'; complete with or without anti-( l-18)1'. In contrast, longer incubation of phosphorylase h with kinase or addition of more kinase than indicated in the legend of Fig. 3A   anti-(l-18)1) from phosphorylase. In a radioimmunoassay similar in protocol to that described in Table II,  AMP, as well as glucose l-phosphate, glucose, glucose 6phosphate, and ATP had no effect on the binding of anti-(l-18)~ to phosphorylase.
Because the AMP-dependent activity of phosphorylase h was less inhibited by anti-(1-18)p than the AMP-independent activity of phosphorylase a, the presence of AMP might be expected to decrease the inhibition of phosphorylase a. Possible interpretations of the inhibition of V,,,;,, are presented in the discussion. Effect ofAnti-(I-18)p on Rabbit Lir$er Glycogen Phosphorylase-The liver isozyme of glycogen phosphorylase differs from the muscle isozyme in immunological (38) as well as other physical and catalytic properties (1). However, muscle phosphorylase kinase will catalyze phosphorylation of the NHZ-terminal region of the liver phosphorylase b isozyme (21). Using a radioimmunoassay similar to that described in Table II, we have found that anti-(1-18)p binds equally well to both the phosphorylated (a form) and dephosphorylated (b form) forms of liver phosphorylase. However, a given quantity of anti-( 1-18)~ bound lo-fold less liver phosphorylase than muscle phosphorylase a. Liver phosphorylase b has a much lower affinity for glucose l-phosphate than muscle phosphorylase b (22). Liver phosphorylase b has, relatively, little enzymatic activity in the presence of 16 mM glucose l-phosphate.
The effect of different amounts of anti-(l-18)p on the activity of liver phosphorylase b in the presence of 16 mM glucose l-phosphate is presented in Fig. 4. At the highest level of anti-( 1-18)~ tested, a greater than 4-fold increase in enzyme activity was observed. When normal goat y-globulin was substituted for anti-( l-18)1,, there was no change in activity.
When liver phosphorylase b was converted to phosphorylase a by rabbit muscle phosphorylase kinase and assayed under the same conditions as in Fig. 4 the specific activity was 32 pmol/min/mg. Unlike liver phosphorylase b, when anti-(1-18)p was added to the liver phosphorylase a, there was little or no effect on the activity of liver phosphorylase a , I / -I (1.0 pg) was preincuhated at 30°C in a mixture (0.05 ml, pH 6.8), containing 4.0 rnM EDTA, 40 rnM P-glycerophosphate, 30 mix 2-mercaptoethanol, 90 rnM NaCl, 25 pg of normal goat y-globulin, and either (i) varying amounts of anti-(I-18)p (0) or (ii) varying amounts of additional normal goat y-globulin (0). After 15 min, a substrate mixture (0.05 ml, pH 6.8), containing 10 mM AMP, 2.0% glycogen, and 32 mM [U-'%]glucose l-phosphate, was added, and phosphorylase activity was measured as described in the text.
A common population of antibodies may bind to both live! phosphorylase and muscle phosphorylase since the liver isozyme displaced the muscle isozyme from anti-( l-18)1) in radioimmunoassays.
The population of anti-(1-18)p that catalytically activates liver phosphorylase is probably specific for the NHZ-terminal region because the muscle phosphorylase phosphopeptide Ac(l-18)p completely blocked the activation by anti-(1-18)p.
In the absence of anti-(1-18)p, Ac(l-lH)p slightly stimulated liver phosphorylase b. Anti-(1-18)p greatly improved the affinity for glucose lphosphate and removed most of the cooperativity seen in the double reciprocal plot (Fig. 5). The ratio of the apparent K,,, for glucose l-phosphate in the absence of anti-( 1-18)~ to the K,,, in the presence of anti-(1-18)p was 4.1 (Table IV). At the lowest concentrations of glucose l-phosphate tested, the a('tivity was increased by anti-(1-18)p by more than 1%fold. The apparent K,, for AMP was dependent on the concentration of glucose l-phosphate.
At saturating levels of glucose l-phosphate, the ratio of the K,, for AMI' in the absenc,e of anti-( I- When the kinetic constants for a given compound were determined, the concentration of that compound was varied. The V,,,,,, was estimated from double reciprocal plots and the K,,, and K,, were estimated from Hill plots.

Compound
Concentration of Plucose-1-P AMP 100 -0.9 I 18)~ to the K,, in the presence of anti-(1-18)p was 1.6 ( Table  IV). The maximal velocity was not affected by anti-(1-18)p. Thus, the primary reason for catalytic activation of liver phosphorylase b is an increased affinity for glucose l-phosphate. In the absence of AMP, anti-(1-18)p had no effect on the activity of liver phosphorylase h.

DISCUSSION
The difference in the effect of anti-(1-18)p on the interconversion reactions may be explained by differences in the degree of steric interference by anti-(l-18)p with the binding of the interconverting enzymes to phosphorylase. Binding of phosphorylase phosphatase (MY = 35,000) (29) probably would be affected less than binding of phosphorylase kinase (Mr = 1,300,OOO) (39) by steric interference from anti-(1-18)p. Phosphorylase kinase may not bind to phosphorylase b containing bound anti-(1-18)p; the 12% conversion that did occur may reflect the percentage of NH?-terminal regions that were not bound to anti-(1-18)p.
The phosphorylated seryl residue (serine-14) seems to be accessible to phosphatase when anti-(l-18)~ is combined with phosphorylase. This is consistent with the observed high specificity of anti-( 1-18)~ for a determinant located within the first 4 NHP-terminal residues of the enzyme. An estimation of the molecular weight of the complex of muscle phosphorylase and anti-(1-18)p was made by gel filtration chromatography.
The complex eluted from the column in a position that corresponded to a molecular weight slightly less than tetrameric phosphorylase a (Mr = 390,000). Anti-(l-18)~ is most probably IgG (M,. = 160,000) because of the injection schedule used to prepare the goat hyperimmune serum and its insensitivity to 2-mercaptoethanol. . From x-ray crystallographic data, the distance between the variable region combining sites of a human myeloma IgG was found to be 142 A (41). Therefore, it is certainly a physical possibility for anti-(1-18)~ to span the two NH%-terminal regions of a phosphorylase dimer. This interpretation of the data would also explain why precipitin complexes are not formed when anti-(1-18)p combines with phosphorylase. Tzartos and Evangelopoulos (9) studied the effects of antibodies, directed against multiple determinants of pig muscle glycogen phosphorylase b, on the kinetic constants for the enzyme.
They also found that the apparent affinities for glucose l-phosphate and AMP were unaffected by their nonspecific antibodies.
However, the apparent affinity for glycogen was reduced when the pig enzyme was complexed with antibodies.
When glycogen fragments were used instead of the intact macromolecular substrate, glycogen, the antibodies did not affect the apparent affinity for these lower molecular weight substrates. They suggested that this observation was consistent with steric hindrance by the antibodies.
In contrast, anti-(l-18)p improved the apparent affinity for the macromolecular substrate, glycogen, for both rabbit muscle phosphorylase a and 6. Therefore, steric hindrance to binding of glycogen did not exist when anti-(1-18)p was bound to the NHz-terminal region. The apparent K, for glycogen may be influenced by binding of glycogen at the "glycogen storage site" (3), which is located 25 8, from the active site (3,42). The improved affinity for glycogen in the presence of anti-(l-18)~ may be related to binding of glycogen at either the glycogen storage site or the active site, or both. The dissociation constant for oligosaccharide at the active site is at least 20-fold greater than that at the storage site (42). The dissociation constant at the storage site is similar to that reported for glycogen (43). Because of the better binding affinity at the "storage site" and the observation that anti-(1-18)p does not affect apparent binding constants for other compounds that bind in or near the active site, the improved apparent affinity for glycogen by anti-(1-18)p could be due to an effect on the glycogen storage site. Therefore, the NH&erminal region may regulate what occurs at the glycogen storage site, as well as its well known regulation of activities at the catalytic site. The reduction in V,,, of phosphorylase could be due to steric interference, by anti-(1-18)p, with the interaction of phosphorylase with glycogen during catalysis, after binding of glycogen at the storage site. In the experiments for determination of kinetic constants we observed that the percentage of inhibition by anti-( l-18)p was the same at all concentrations of AMP and glucose l-phosphate, including concentrations that gave less than half the maximal velocity.
In contrast, as the concentration of glycogen decreased below its Km, the percentage of inhibition by anti-(1-18)p decreased. Activation by anti-( l-18)p of the initial velocity was observed at concentrations of glycogen that were less than half the apparent K,,,. Presumably, at these glycogen concentrations, the improved affinity for glycogen becomes more important t,han the reduced catalytic turnover when anti-(1-18)p is bound. Anti-(I-18)p may lower the V,,,,, because of some effect on the structure or conformational mobility of phosphorylase. An antibody molecule bound to phosphorylase could interfere with necessary movements of the protein chain during catalysis. Freedom of movement of the NH%-terminal region may be important for catalysis, and the binding of large antibody molecules may hinder that movement. This latter explanation is supported by a larger reduction in the V,,,,, for phosphorylase a (70% inhibition) in the absence of AMP (Table III). In the case of phosphorylase a, the NH*-terminal phosphate has a definite role in catalytic activation.
The NHt-terminal region of phosphorylase b is not directly involved in activation by AMP, but some effect by anti-(1-18)p might be expected because of the close proximity of a segment of the NH,terminal region to the AMP site (4). The pyridoxal5'-phosphate site is located very close to the binding site for glucose l-phosphate (32,44). This important cofactor site was dramatically affected by anti-(1-18)p. Under conditions in which complete loss of activity could be obtained in the absence of anti-(1-18)p, the binding of anti-(1-18)p to the NH,-terminal region of phosphorylase b provided complete protection against loss of enzymatic activity due to loss of pyridoxal 5'phosphate. The AMP binding site and the NHz-terminal region are close to one another, but they are about 30 A from the glucose l-phosphate site (catalytic site) and the pyridoxal Y-phosphate site (44). Nevertheless, phosphorylation of the NHZ-terminal region (45) or binding of AMP (46) will also protect the enzyme from loss of pyridoxal 5'-phosphate, as well as increase affinity for substrates and by guest on March 23, 2020 http://www.jbc.org/ Downloaded from differential changes in the affinity for inhibitors (caffeine, activation of enzymes. Antibodies prepared against native, glucose, glucose &phosphate, and AMP) (1). Because anti-(l-active enzymes have been used to enhance enzyme activity in 18)~ has little or no effect on the apparent affinity for sub-partially active enzymes (50,51) or to induce enzyme activity strates and inhibitors, its stabilization of the pyridoxal 5'-in inactive enzymes (52)(53)(54). Because of the rigid specificity of phosphate site is particularly interesting. The data support antibody combining sites, antibodies may induce activation the hypothesis that close communication may exist between by causing a conformation change in that region of the protein the NHZ-terminal region and the pyridoxal5'-phosphate site. molecule that interacts with the antibody, which may lead to Graves et al. (47) showed that the pyridoxal 5'-phosphate in conformation changes in other regions of the protein molecule. phosphorylase had an important effect on the enzymatic This mechanism of antibody-induced effects on specific prointerconversion and tryptic attack of the NH2-terminal region. teins has been reviewed by Crumpton (55) and Melchers el Pyridoxal 5'-phosphate is locat,ed in the same hydrophobic al. (54). Anti-(1-18)p could activate by inducing and stabilizpocket as the binding site for glucose l-phosphate (44). Phos-ing a conformation of the NH,-terminal region of liver phosphorylase is completely inactive when pyridoxal5'-phosphate phorylase h that is similar to that of muscle phosphorylase. is removed (48). Because anti-(1-18)p had no effect on the Anti-(1-18)~ was unable to affect the If,,,;,, for liver phosbinding affinity for compounds (glucose-l-P, glucose, and phorylase h (Table IV), but the V,,,;,, for muscle phosphorylase caffeine) that bind in the same hydrophobic active site crevice h and a was reduced. As previously discussed, the amount of as pyridoxal 5'-phosphate, but had a major effect on the reduction of V,,,, was very different for the a and h forms of stability of the pyridoxal 5'-phosphate site, communication the muscle isozyme. The variation in the degree of effect of between the NHZ-terminal region and the hydrophobic active anti-( 1-18)~ on the Vn,,x for different phosphorylases could be site region may be mediated through pyridoxal 5'-phosphate. related to differences in the nature of the interaction of anti-Another interpretation of the effect of anti-(1-18)p on res-(1-18)~ with the different phosphorylases. olution of pyridoxal Y-phosphate should be considered. When dimeric phosphorylase b was subjected to a deforming buffer (imidazolium citrate, pH 6.0) used for resolution, the enzyme was reversibly dissociated into monomers (48). During this treatment enzyme-bound pyridoxal5'-phosphate may become more exposed, since it will now exchange with free radiolabeled pyridoxal 5'-phosphate (47). As previously discussed, anti-(I-18)p may cross-link the monomeric units of dimeric phosphorylase.
Monomerization could be a necessary step for exposure of pyridoxal 5'-phosphate by the deforming buffer, and anti-( 1-18)~ might prevent exposure of pyridoxal5'-phosphate by linking the monomers of dimeric phosphorylase 6.