dimer-tetramer equilibrium of glutathione reductase from the cyanobacterium spirulina maxima

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 268, No. 1, January, pp. 255-263,1989

Dimer-Tetramer Equilibrium of Glutathione Reductase from the Cyanobacterium Spirulina maxima

JUAN L. RENDbNl AND GUILLERMO MENDOZA-HERNANDEZ

Departamento de Bioquimica, Facultad de Medic&a, Universidad National Autbrwma de M&ico. Apartado Postal 70.159,04510 Mexico City, D.F. Mexico

Received May 11,1988, and in revised form September 1,1988

Glutathione reductase [NAD(P)H:GSSG oxidoreductase; EC 1.6.4.21 from cyanobacterium Spirulina maxima exists as an equilibrium system between a dimer (.~a~,~ = 5.96) and a tetramer (s~~,~ = 8.49) which has a very slow interconversion rate at neutral pH. Our results showed that the apparent dissociation constant (kd) was 4.61 X lo-? M. The proportion of both forms at pH 7.0 did not alter at either 4 or 25°C. However, electropho- retie analysis at various pH values showed that at 25°C a gradual transition takes place between oligomers with an apparent pK, of 7.55. When dimers aggregate to form tetramers, the reaction involves the uptake of eight protons (K = 1.58 X lo-” M’). At pH 7.7, the equilibrium shifts completely from dimers-tetramers to dimers when temperature is increased, which would suggest that the dissociation is an endothermic process. Ther- modynamic parameters obtained from the temperature study show that the dissociation of glutathione reductase is characterized by positive entropy and enthalpy changes. Nei- ther NADPH nor GSSG have any effect on the dimer-tetramer equilibrium. Measure- ments of reductase activity indicate that the tetramer is almost certainly active, whereas the dimer is either less active or inactive. Q 1989 Academic Press, Ine

The existence of equilibria between oligomers in proteins is a well-documented fact (1). This phenomenon has a wide oc- currence in the living world, ranging from bacteria to mammals (2-4). The equilibrium involving a dimer and a tetramer is of particular biological significance because, as has been demonstrated in some cases, a particular component of the system (either a dimer or a tetramer) has a higher catalytic activity as compared to its counterpart (5,6) and thereby confers reg- ulatory properties to the system. The only published information on the existence of an equilibrium between oligomers for glutathione reductase [NAD(P)H:GSSG oxidoreductase; EC 1.6.4.21 is on sea-urchin eggs (7). The enzyme in this organism ex-

1 To whom correspondence should be addressed.

ists as a principal form of M, 100,000 in apparent equilibrium with higher molecular weight aggregates. On the other hand, Kalt-Torres et al. (8) have reported the existence of a tetrameric glutathione reductase, from pea chloroplasts, although no equilibrium between oligomers was observed.

In a previous paper (9), we reported the purification and general properties of Spirulina maxima glutathione reductase. In this cyanobacterium, the enzyme exists in two different oligomeric states, a dimer and a tetramer, with sedimentation coefficients of 5.96 and 8.49, respectively. In that paper, preliminary evi- dence on the existence of a slow equilibrium interrelating both forms was presented. In this work we report on the dependence of equilibrium upon both pH and temperature; we also present evi-

255 0003-9861/89 $3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.

256 RENDbN AND MENDOZA-HERNANDEZ

dence suggesting that the tetramer is the more active species.

MATERIALS AND METHODS

Materials. Electrophoresis reagents, as well as GSSG and NADPH, were purchased from Sigma Chemical Co. (St. Louis). All other chemicals used were of analytical grade and were obtained from Baker de Mexico.

Enzyme. Glutathione reductase was purified to ho- mogeneity from S maxima as previously described (9). Enzyme purity was routinely assessed by PAGE’ under native and denaturing conditions. Enzyme assays were run as described elsewhere (9) using satu- rating concentrations (at least 10 times K,) of both NADPH and GSSG.

Protein concentration, Enzyme concentration was quantified by the absorbance coefficient at 274 nm of a 10 mg/ml enzyme solution in 0.1 M potassium phosphate buffer (pH 7.0) plus 1 mM EDTA according to the gravimetric method of Kupke and Dorrier (10). A value of 9.47 was obtained which is in accordance with what has previously been reported for yeast glutathione reductase (11). Ultracentrifugation protein profiles were determined by the Bradford dye-binding assay (12).

Ultracentrifugation. Ultracentrifugation experiments were carried out in a L5-65 Beckman prepara- tive ultracentrifuge using a SW 50.1 rotor. The enzyme samples were always placed on top of a linear sucrose gradient (5 to 20%) prepared in 0.1 M potassium phosphate buffer (pH 7.0) containing 1 mM EDTA (buffer A) and ultracentrifuged at 290,OOOgfor 7 h at 2-4°C. Tubes were then punctured at the bottom and fractions of about 100 ~1 were collected.

Polyacrylamide gel electrophoresis. Electrophoresis under nondenaturing conditions was performed in 90 x 6-mm cylindrical gels prepared in buffer A at a constant acrylamide concentration (2’ = 0.08, C = 0.04); the same buffer was used in both cathode and anode reservoirs. This method was used rather than pore- gradient PAGE due to its greater technical simplic- ity. Figure 1 shows the electrophoretic separation of the dimer and the tetramer as well as its corresponding densitometry in a constant acrylamide concentration gel. It is evident that both forms have very different electrophoretic mobilities, a characteristic which favors the analysis of enzyme mixtures by this method. Assignment of zones to either dimer or tetramer was made by comparing with a calibrated pore- gradient gel run under identical conditions of buffer

’ Abbreviations used: PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate.

FIG. 1. Electrophoretic pattern of glutathione re- duetase at neutral pH in a constant concentration polyacrylamide gel (7’ = 0.08). An enzyme aliquot (30 pg) was placed on the top of a gel prepared in buffer A and run during 12 h as described under Materials and Methods. Direction of migration was from cathode to anode. After destaining, a densitometric scan was recorded.

composition, pH, and temperature (9). In both sys- tems, densitometric analysis showed identical dimer- tetramer proportions. Electrophoreses were carried out, unless otherwise indicated, at 25°C for 5 to ‘7 h at 5 mA/gel. Each gel was stained overnight in 0.25% (wt/vol) Coomassie brilliant blue in methanol:acetic acid:water (5:1:5), and destained exhaustively with the same solution without the dye. Clear gels were analyzed by densitometry in a quick scan densitome- ter (Helena Laboratories).

Analysis of poteinprojiles. From the densitometric scan or protein profile obtained in ultracentrifugation experiments, the relative proportion of each form was quantified by cutting out and weighing the corresponding zone profile and then relating each of these to the initial protein concentration. In those calculations involving molar concentrations of both dimer and tetramer, we have assumed a molecular weight of 96,000 and 192,000, respectively. Dissocia- tion degree of enzyme ((Y) represents the molar fraction of the dimer in the dimeric state and equals its fractional area in the densitometric traces,

All results reported here are expressed as means f standard deviation and represent the average of at least two experiments. When more than two experiments were performed, the number is indicated in pa- renthesis. When necessary, data were fitted into a linear trace by the least-squares method.

GLUTATHIONE REDUCTASE FROM Spirulina maxima 257

0.6

FRACT I ON NUMBER

FIG. 2. Sedimentation pattern of glutathione reductase at neutral pH followed by electrophoretic analysis of separated zones. An enzyme sample (280 rg) was layered on top of a sucrose gradient and ultracentrifuged at 4°C. Immediately after determination of the protein profile, the composition of the separated dimer and tetramer was analyzed by electrophoresis at pH 7.0. To eliminate possible zone overlapping, only the shadowed fractions were used. Ultracentrifugation and electrophoretic conditions were performed as described under Materials and Methods. Numbering of fractions is from the top of gradient. Thus, the bottom of the tube is at the right. Inset, electrophoretic patterns of (A) control (30 pg), (B) dimer zone (20 pg), and (C) tetramer zone (40 pg).

RESULTS

Intercmversim at neutral pH. Separa- tion of both oligomers in the ultracentrifuge, followed by electrophoretic analysis of each form under nondenaturing conditions, revealed that equilibrium is very slow. Several hours after separation, the zone containing the tetramer remains un- contaminated by the dimer, while dimer fractions contained only a partial tetramer content (Fig. 2). However, when separated fractions corresponding to dimer and tetramer were incubated several days at room temperature, the resulting electrophoretic patterns were identical in both cases to that shown in Fig. 1, suggesting that equilibrium was achieved.

Analysis of enzyme samples at pH 7.0, either by ultracentrifugation or electro-

phoresis, revealed that the tetramer, which constitutes up to 86% (in mass) of the total protein, was the predominant species (Figs. 1 and 2). Protein profiles obtained from ultracentrifugation runs, assessed by either the Bradford assay method or their absorbance at 280 nm, were identical. Taking into account the slow interconversion rate an equilibrium constant can be calculated from the relative areas of dimer and tetramer obtained in either ultracentrifugation or electrophoresis experiments for the reaction

D d$ZD,

where D4 and Dz represent tetramer and dimer, respectively. The value obtained was 4.61 X lop7 M starting from an initial protein concentration of 1 mg/ml.

258 RENDbN AND MENDOZA-HERNANDEZ

P”..

A B C D E

FIG. 3. Electrophoretic patterns of glutathione reductase as a function of pH. Enzyme samples were dialyzed overnight at 25°C against 0.1 M potassium phosphate buffer previously adjusted to the indicated pH with KOH. An enzyme aliquot (40 pg each) was then applied on a 8% polyacrylamide gel prepared in the same buffer at the corresponding pH. Other conditions are described under Materials and Methods. (A) pH 7.5; (B) pH 7.55; (C) pH 7.63; (D) pH 7.68; (E) pH 7.7.

The protein profile obtained when the enzyme was preincubated overnight at pH 7.0 in the presence of either NADPH (100 PM) or GSSG (1.2 mM), and followed by ultracentrifugation under the same conditions, was identical to that shown in Fig. 2, which suggests that no changes occurred in the equilibrium proportions.

Under the conditions used in this work no aggregates higher than a tetramer were observed, although the long-term storage of enzyme (over six months), gave a minor slow-migrating band in electrophoresis or a fast-migrating particle in the ultracentrifuge. The nature of this aggregate remains to be elucidated.

Efect of pH on equilibrium, Our previous work (9) proved the outstanding dependence of equilibrium on pH. In order to fully characterize such a dependence, an electrophoretic study of the equilibrium state over a broad pH range was performed. Figure 3 shows some representa- tive gels. It is clear that the noncoincidence of protein zones at different pH’s is due to

a higher mobility of the enzyme as a consequence of its gradual dissociation from the isoelectric point. The densitometric traces were used to plot the enzyme’s degree of dissociation (a) against pH (Fig. 4). It was evident that a sharp transition occurred in the pH ranges 7.5 to 7.7 at 25°C and 8.8 to 9.0 at 4”C, resulting in a fully dimerized enzyme (CV = 1). Because of the sensitivity to pH of the electrophoretic patterns, the aggregation of dimers to form tetramers suggests a proton-dependent system according to the reaction

D 4=2Dz+nH+.

In order to calculate n, which is the stoi- chiometric coefficient of the number of protons involved in the transition, a ther- modynamic equilibrium constant may be defined as

After rearranging and obtaining loga- rithms we have

logP212 - -- Wil

npH+logK.

Thus, from a linear plot of log[D,y/[D,] versus pH, the values of TL and K can be calculated. The inset to Fig. 4 shows such a plot, in which n = 7.73 +- 0.6 and K = 1.58 + 1.3 X lO-‘j“ M’. Therefore, a total of eight protons appears to be involved in the dimer-tetramer transition.

We have previously demonstrated that the interconversion at different pH’s is freely reversible (9). In order to gain insight about the time scale of the equilibrium when pH is changed, the following experiment was performed. A 20-pg sample of enzyme which had been stored at neutral pH in buffer A was placed on the top of a polyacrylamide gel prepared at pH 9.0 in 0.1 M Tris-HCl containing 1 mM EDTA. Electrophoresis was then run immediately under conditions described under Materials and Methods. When the gel was destained, a pattern identical to that shown in Fig. 3E was revealed. A similar experiment was carried out by placing an


75 76 77 PH

4// I I I 1 5 7 8 9 IO

PH

FIG. 4. Dissociation degree (CY) of enzyme samples as a function of pH at two different tempera- tures. Each point was calculated from the relative area of dimer in the densitometric traces. Data for the curve at 4°C was taken from previous work (9) and supplemented by additional experiments at this temperature. Inset, plot of log [D,y/[D,] versus pH. Intercept and slope values were -63.8 and 7.73, respectively (T = 0.80).

enzyme aliquot at pH 9.0 on the top of a gel prepared in buffer A and then run at neutral pH. The profile obtained was identical to that shown in Fig. 1. These results demonstrate that the achievement of a new equilibrium state when pH is changed is at most in the range of minutes. Thus, the relative proportions of dimer and tetramer in Fig. 4 represent equilibrium conditions.

Eflect of temperature on equilibrium. Ul- tracentrifugations were carried out at 20°C and neutral pH in order to elucidate the possibility that the proportion of dimer and tetramer could be different at this temperature as compared with the pattern obtained at 4°C. The resulting profile (not shown), as far as the relative dimer-tetramer proportions were concerned, was identical to that shown in Fig. 2 at 4°C although a partial overlap of zones as a consequence of the higher diffusion rate was observed. Electrophoretic runs performed under identical conditions gave the same result as shown in Fig. 4 (CX = 0.14 f 0.017). At pH 7.7 and 25”C, however, all enzyme molecules were in the dimer state (o( = 1.0,

Fig. 4), while at 4”C, the tetramer was the more common species (LY = 0.14). When enzyme samples were analyzed at pH 7.7 over a temperature range of 4 to 25”C, a gradual change in the relative proportions of both forms, resulting in a fully dimerized enzyme (Fig. 5a), was observed. These electrophoretic patterns were not modified with longer incubation times, thus suggesting that a new equilibrium state was reached when enzyme was dialyzed overnight at a new temperature.

From the slope of a van’t Hoff plot (Fig. 5b) a value of $67 * 8 kcal mall’ for the AH” of the dissociation reaction was calculated.

Table I summarizes some thermody- namic parameters of the dimer-tetramer equilibrium.

Activity of both forms. When both forms were separated in the ultracentrifuge, and assayed for enzymatic activity, it became clear that GSSG reductase activity was present in those fractions containing either dimer or tetramer (Fig. 6). However, there is no exact correspondence between protein and activity profiles, since the frac-

260 RENDCN AND MENDOZA-HERNANDEZ

b -10

- 14

A 0 c D f/T K-‘( x 10~)

FIG. 5. (a) Electrophoretic patterns of glutathione reductase as a result of temperature changes. Enzyme samples were dialyzed overnight against buffer A at pH 7.7 and the indicated temperature and then analyzed by PAGE. Forty micrograms was applied in each lane. Electrophoretic runs were made at the corresponding temperature. (b) van’t Hoff plot showing the temperature dependence of the dissociation constant (/cd). Each point represents the dissociation constant calculated from the densitometric traces. A value of -33981 was obtained for the slope of plot (n = 3, T = 0.98). (A) 4°C; (B) 8°C; (C) 11°C; (D) 22°C.

tion in the tetramer zone which corre- from the highest peak of the dimer and tet- sponds to the highest point of the peak has ramer fractions, was 161.5 + 14 and 253.6 a protein concentration about sixfold that +- 20 U mg-‘, respectively (n = 3). of the equivalent tube in the dimer zone, while the activity in the latter is less than DISCUSSION

the 11th part of the highest value of the The results reported here demonstrate tetramer. The specific activity, calculated that glutathione reductase from S. max-

TABLE I

THERMODYNAMICPARAMETERSOFTHEDIMER-TETRAMEREQUILIBRIUM

Temperature

c-7 kaa

(10’ X M) Kb (M’)

AGo” (kcal mall’)

AH0 (kcal mall’)

ASO (eUld

4 4.61 _+ 1.23 +8.03 f 0.15 8 11.29+- 4.8 +7.65 f 0.20 +212”

11 101.3 * 38 +6.49 f 0.22 +67&8 (+190 i 28)f 15 423.5 f 160 +5.76 f 0.23 25 1.58 + 1.3 x lo-@ +87

a Calculated from the relative dimer-tetramer proportions at pH 7.7 (n = 3). * Calculated from the intercept of the linear plot in Fig. 4. ‘Calculated from the relation AGo = -RTln K,. d Entropic units (cal mall’ K-l). e Calculated from the relation AS’ = (AH” - AG”)/!i”. fCalculated from the intercept of a Van’t Hoff plot (n = 3).

GLUTATHIONE REDUCTASE FROM Spirulina ma&ma 261

FRACTION NUMBER

FIG. 6. Protein and activity profiles in sedimentation experiments. An enzyme aliquot (131 pg) was ultracentrifuged at 4°C as described under Materials and Methods. Protein concentration and enzymatic activity were then determined. Protein content in each fraction was measured by relating its fractional area in the protein profile to the initial sample concentration. GSSG reductase activity was quantified by using a l-p1 aliquot of each fraction. Maxima of protein and activity were traced at the same level in the plot. (0) Protein profile; (0) enzyme activity.

ima exists as a pH-dependent equilibrium system between a dimer and a tetramer. The latter are interrelated by a very slow interconversion rate at neutral pH; this is suggested by: (a) The excellent resolution of zones when the enzyme is subjected to mass-transport experiments according to Nichol et al. (13); and (b) the fact that several hours after their separation in the ultracentrifuge, the tetramer zone still remains free of dimeric species, while the zone corresponding to the dimer shows only a partial tetramer content, the proportion of which does not correspond to the equilibrium condition. Because of the latter, the possibility that the tetramer could be an artifact produced by the interaction of dimers with an uncharged component of the buffer system, as described by Cann (14,15), can be eliminated. In addition, the resulting profile at neutral pH is identical

to that observed at 4°C both in acetic acid- acetate buffer at pH 5.2 (unpublished results) and in Tris-HCl buffer at pH 7.5 (9). The high stability of the tetramer enables the elucidation of its subunit arrangement by the use of crosslinking agents, which would be unfeasible if the interconversion rate were fast (16).

The calculation of the equilibrium constant from the relative proportions of oligomers obtained in the mass-transport experiments is justified in the case of very slow transitions (17). From the value obtained, a free energy change of t8.03 Kcal mol-’ can be calculated for the dissociation reaction D 4 + 2 Dz; however, this equilibrium does not consider the fact that protons are involved, as is suggested by the present work. The above results can be rec- onciled if we assume the scheme

D 4t2Dp-4H+=2Dz+8H+.

In this reaction the dissociation of the tetramers to dimers is followed by the release of protons from the dimers. Therefore, the value of 1.58 X 10-‘j4 Mg calculated from Fig. 4 for the equilibrium constant (K) corre- sponds to the overall reaction, while the value of 4.61 X 10m7 M involves only the equilibrium in the absence of proton binding (kd), at constant pH. These values can be used to determine what effect protonation plays in the overall equilibrium. Thus, the AGo calculated from K, is made of contributions from the protonation/de- protonation and the dissociation of the tetramer to dimers, and equals

AG;; = AG& + AGfi+

where AG,& is the observed free-energy change for the dissociation reaction without contributions from protons, and AG;;+, is the free-energy change for the release of protons from the dimer. Using the corresponding values shown in Table I we obtain +78.96 kcal mol-’ for AGfi+.

The high degree of dependence of the equilibrium state upon pH suggests that dissociable groups could be involved in the transition. We have noted an important change between dimer and tetramer pro-

262 RENDdN AND MENDOZA-HERNANDEZ

portions in two temperature-dependent pH ranges, which suggests that the pK,‘s of the groups involved are clearly dependent on temperature. Although the chemical nature of such groups remains to be elucidated it is interesting to note that in the presence of acrylamide monomers the enzyme behaves as a dimer (9). In this con- text, it has been reported that acrylamide can be covalently bound to proteins by re- acting with the E-amino group of lysine residues (18); if such groups are involved in the dimer-tetramer transition of glutathione reductase, the effect of acrylamide on the dissociation state of the enzyme could be easily explained. On the other hand, the high stability of dimers to 6 M urea (9) and to pH changes, as our results show, suggest that the monomer-monomer contacts are of a very different nature, probably hydrophobic. This possibility is supported by the fact that a very strong dissociating agent such as SDS is required to completely dis- sociate dimers to form monomers (9).

In order to gain insight into the possible nature of the interactions involved in the dimer-tetramer equilibrium, the intrinsic change in entropy, i.e., without solvent contributions, was calculated and compared with the experimental value of A.‘?. Following Doty and Myers (19), who assume losses of translational and rotational degrees of freedom and no entropy recover upon association, we obtain a value of -102 eu for the Spirulina enzyme. As compared to the experimental value shown in Table I, this result suggests that an additional of about -100 eu derived from changes in the interaction between protein and solvent as a consequence of the association must be considered. This contribution could well arise from the formation of a number of hydrogen bonds. In this sense, Nemethy et al. (20) calculate -2 to -5 eu for the formation of hydrogen bond, depending on the involved side chains. Although the partici- pation of hydrophobic binding cannot be excluded, the notable dependence of the association on pH supports the possibility that hydrogen-bond forming groups are involved.

It is interesting to note that when the pH is changed, the time required for the achievement of a new equilibrium is several orders of magnitude faster as compared with the reequilibration at neutral pH. Although this seems to be a contradic- tory result, there exists the possibility that a high-activation energy barrier at pH 7.0 could be the rate-limiting step in the dimer-tetramer transition. Clearly, more work is needed to fully characterize this observation.

Finally, we must consider the question of the enzymatic activity of both forms. Al- though the reducing activity is associated with both dimer and tetramer zones in sedimentation experiments (Fig. 6), it is evident that the latter is the more active species of the two, since GSSG reductase activity in the dimer fractions could be associated with the existing tetramers as a consequence of a partial reestablishment of the equilibrium (see Fig. 2). Therefore, the low specific activity of 161.5 U mgP1 calculated for the dimer (as compared with 238 U mg-’ for the pure enzyme) can be explained as the result of a mixture of active and inactive (or less active) protein, whereas the value of 253.6 U mg-’ for dimer-free tetramer must be considered as the intrinsic activity of this aggregation state. In addition, we have demonstrated that at neutral pH temperature-raising does not alter the dimer-tetramer proportions; this means that at 25”C, the temperature at which enzymatic assays are carried out, the tetramer remains as the predominant species.

ACKNOWLEDGMENT

The authors thank Ing. 0. Zempoalticatl from Sosa Texcoco Co. for the gift of Spirulina maxima.

REFERENCES

1. FRIEDEN, C. (1971) Anwu. Rev. B&hem 40,653- 696.

2. BOEKER, E. A., AND SNELL, E. E. (1968) J Biol Chem. 243,1678-1684.

3. MORIMOTO, K., AND KEGELES, G. (1971) Arch. Bic- them. Biophys. 142.247-257.


4. UNGEWICKELL, E., AND GRATZER, W. (1978) Eur. J. B&hem. S&379-385.

5. WAGNER, R., GONZALEZ, D. H., PODESTA, F. E., AND ANDREO, C. S. (198’7) Eur. J. B&hem. 164, 661-666.

12. BRADFORD, M. M. (1976) And Biochem 72,248-T&4. 13. NICHOL, L. W., BETHUNE, J. L., KEGELES, G., AND

HESS, E. L. (1964) in The Proteins (Neurath, H., Ed.), Vol. 2, pp. 305-403, Academic Press, New York.

6. VERON, M., GUILLOU, Y., FAZEL, A., AND COHEN, G. N. (1985) Eur. J. B&hem. 151,521-524.

7. II, I., AND SAKAI, H. (1974) B&him. Biophys. Acta 350,151-161.

14. CANN, J. R., AND GOAD, W. B. (1965) J. BioL Chem. 240,148-155.

15. CANN, J. R., AND GOAD, W. B. (1965) J. BioL Chem. 240,1162-1164.

8. KALT-TORRES, W., BURKE, J. J., AND ANDERSON, J. M. (1984) Physiol Plant. 61,271-278.

9. REND~N, J. L., CALCAGNO, M., MENDOZA-HERNAN- DEZ, G., AND ONDARZA, R. N. (1986) Arch. Birr them. Biophys. 248,215-223.

10. KUPKE, D. W., AND DORRIER, T. E. (1978) in Meth- ods in Enzymology (Hirs, C. H. W., and Timas- heff, S. N., Eds.), Vol48, pp. 155-162, Academic Press, New York.

16. HUCHO, F., M~LNER, H., AND SUND, H. (1975) Eur. J. B&hem. 59,79-87.

17. NICHOL, L. W. (1981) in Protein-Protein Interac- tions (Frieden, C., and Nichol, L. W., Eds.), pp. 15, Wiley, New York.

18. GEISTHARDT, D., AND KRUPPA, J. (1987) Anal. Bic- them. 160,184-191.

19. DOTY, P., AND MYERS, G. E. (1953) Discuss. Faru- day Sot. 13,51-58.

11. COLMAN, R. F., AND BLACK, S. (1965) J. BioL Chem. 20. N~METHY, G., STEINBERG, I. Z., AND SCHERAGA, 240,1796-1803. H. A. (1963) Biopolymers 1,43-56.

dimer-tetramer equilibrium of glutathione reductase from the cyanobacterium spirulina maxima

Documents