Correlation between enzyme activity and hinge‐bending domain ...

Eur. J. Biochem. 180,61-66 (1989) FEBS 1989

Correlation between enzyme activity and hinge-bending domain displacement in 3-phosphoglycerate kinase Michael A. SINEV ’,Oleg I. RAZGULYAEV ’,Maria VAS ’, Alexander A. TIMCHENKO’ and Oleg B. PTITSYN ’ Institute of Protein Research, Academy of Sciences of the USSR, Pushchino Institute of Enzymology, Biological Research Center, Hungarian Academy of Sciences, Budapest (Received August 19, 1988) - EJB 88 0993

Diffuse X-ray-scattering data give evidence for large-scale structural change in pig muscle 3-phosphoglycerate kinase upon substrate binding. Simultaneous binding of 3-phosphoglycerate and MgATP either to the unmodified enzyme or to its active methylated derivative leads to about an 0.1-nm decrease in radius of gyration. These data coincide well with the previous data for yeast 3-phosphoglycerate kinase. When, instead of methylation, the two reactive thiol groups of pig muscle 3-phosphoglycerate kinase are carboxamidomethylated, the enzyme becomes inactive and the radii of gyration of its ‘apo’ and ‘holo’ forms do not differ within limits of experimental error. Thus, a correlation exists between the activity of 3-phosphoglycerate kinase and its substrate-induced largescale conformational change. This correlation is a strong argument in favor of the functional importance of domain locking in the reaction catalyzed by 3-phosphoglycerate kinase. substrate binding [12]. This locking would bring the two substrates together in a water-free microenvironment favourable for catalysis [12]. The validity of this ‘hinge-bending’ hypothesis has been qualitatively confirmed by small-angle X-ray-scattering experiments where a considerable decrease (of about 0.1 nm) of the radius of gyration for yeast 3-phosphoglycerate kinase has been detected upon simultaneous binding of 3-phosphoglycerate and MgATP [17, 181. The results have been interpreted in terms of domain locking. More detailed information about the mode of this locking has been obtained [19] by comparison of experimental diffuse X-ray-scattering curves (in a wide-scattering-angle region) with scattering curves calculated [20, 211 on the basis of the known three-dimensional structure of the ‘apo’ form of yeast 3-phosphoglycerate kinase. For instance, differences between the experimental scattering curves in the absence and in the presence of substrates have been explained by the rotation of the ATP-binding domain around the ‘short’ axis of the enzyme molecule [19]. To investigate whether the domain locking has indeed a functional importance as suggested by Blake et al. [12] (see also Ptitsyn [22]) we have compared pig muscle 3-phosphoglycerate kinase with its active and inactive derivatives. A previous study [23] has shown that the activity of these derivatives strongly depends on the size of the modifying reagent: methylation of the two fast-reacting thiols of the enzyme does not affect the activity, whereas their carboxamidomethylation is accompanied by a loss of enzymic activity. The loss of activity in the latter case is not due to the absence of substratebinding ability [23, 241. Moreover, the two reactive thiol groups are very likely close to the interdomain cleft [25]. Therefore, a breakdown of the hinge-bending, domainCorrespondence to 0. B. Ptitsyn, Institute of Protein Research, locking mechanism has been proposed [23] as a very likely Academy of Sciences of the USSR, SU-142292 Pushchino, Moscow explanation for the inactivation of carboxamidomethylated Region, USSR enzyme. Abbreviation. R,, radius of gyration. The most adequate method for checking this suggestion is Enzymes. 3-Phosphoglycerate kinase or ATP: 3-phospho-~-glycdiffuse X-ray scattering. Using this method we have compared erate 1-phosphotransferase (EC 2.7.2.3).

X-ray-diffraction analysis has shown the existence of largescale displacements in the structure of some enzymes upon substrate binding [l - 31. These displacements involve a change of the relative orientation of protein domains, e. g. the ‘locking’ of domains in hexokinase [l] or the mutual ‘sliding’ of domains in liver alcohol dehydrogenase [2]. Large-scale displacements may screen the active centers from water, increasing the specificity of transfer reactions [4,5] or facilitating transfer of hydride ion in the reaction catalyzed by dehydrogenases [2]. 3-Phosphoglycerate kinase catalyzes the reversible conversion of 1,3-bisphosphoglycerate to 3-phosphoglycerate with synthesis of an ATP molecule in the glycolytic pathway [6]. The enzyme from various species exhibits a highly conservative monomeric structure and catalytic properties [6 - 81. Xray-diffraction analyses of yeast [9,10] and horse muscle [l 1 131 enzymes demonstrate a high similarity in their threedimensional structure. The most characteristic feature of the molecule is the presence of two widely separated domains with a deep cleft between them. The construction of an active human-yeast chimeric 3-phosphoglycerate-kinase molecule produced by domain interchange [14] is an excellent illustration of the high structural similarity. There are some reasons to suggest that 3-phosphoglycerate kinases from various sources exhibit similar structural changes upon substrate binding [lo, 12, 13, 15, 161. The large distance (about 1 nm) between the ATP (or ADP)-binding site and the site proposed for 3-phosphoglycerate binding allowed Blake and his colleagues to suggest that 3-phosphoglycerate kinase is a ‘hinge-bending’enzyme whose domains ‘lock’ upon

62 the effect of substrates on unmodified and alkylated (methylated or carboxamidomethylated) pig muscle 3-phosphoglycerate kinases. Structural changes in pig muscle and yeast 3-phosphoglycerate kinases have also been compared.

that 3-phosphoglycerate does not absorb at 250 - 350 nm) and that of phosphoglycerate kinase were found to be additive as observed earlier [30, 311. Diffuse X-ray scattering and data treatment

MATERIALS AND METHODS Enzymes and chemicals

Yeast 3-phosphoglycerate kinase was purchased from Sigma as a crystalline suspension in 3.0 M ammonium sulfate. Pig muscle 3-phosphoglycerate kinase was isolated according to the method described for the preparation of rabbit muscle enzyme [26] and crystallized by ammonium sulfate and stored in the presence of dithiothreitol. Methylation and carboxamidomethylation of the pig muscle enzyme were carried out according to the procedure described previously [23]. NADH, ATP and dithiothreitol were Reanal preparations. 3-phosphoglycerate was a Boehringer product, methyliodide was an Aldrich preparation. Ellman’s reagent [27] was purchased from Serva. Iodoacetamide (Fluka) was recrystallized from carbon tetrachloride. All other chemicals were reagent grade commercial preparations. Yeast 3-phosphoglycerate kinase was studied in a 50 mM imidazole/HCl buffer, pH 7.0, containing 0.1 mM dithiothreito1 and 0.1 mM NaN 3. Pig muscle 3-phosphoglycerate and its modified derivatives were studied in 0.1 M Tris/HCI buffer, pH 7.5, containing 1 mM EDTA and 5 mM dithiothreitol. The homogeneity of pig muscle and yeast enzymes was checked by sodium dodecyl sulfate/polyacrylamide gel electrophoresis [28].The enzymatic activity of 3-phosphoglycerate kinase was tested according to Cartier et al. [29]. The specific activity of the pig muscle and yeast enzymes was about 550 600 IU/mg (25°C). The methylated pig muscle enzyme had the same activity. Preparation of samples f o r X-ray measurements

The appropriate aliquot of crystalline protein suspension was centrifuged at 4000 x g for 10 min. The pellet was dissolved in a small amount of buffer. The solution was centrifuged again (20000 xg, 20 min) and dialyzed against the buffer to remove ammonium sulfate. To investigate the ternary complex an appropriate protein solution was diluted by a stock substrate solution (32 mM MgC12, 32 mM ATP and 130 mM 3-phosphoglycerate) in a ratio of 9: 1 [17, 181. As a result, at least 95% of protein was liganded by substrates [30, 311. All the samples were prepared at 4°C. Each protein solution was centrifuged (I0000 xg, 20 min) before X-ray measurements. Samples of methylated and carboxamidomethylated pig enzymes contained about five unreacted thiols/enzyme molecule which is in agreement with the previous data [23]. The thiol content was determined in the presence of 3 M guanidinium hydrochloride by Ellman’s reaction [27]. Alkylation efficiencies of the two fast-reacting thiols in methylated and carboxamidomethylated enzymes were about 90%. Carboxamidomethylated 3-phosphoglycerate kinase had an activity of about 50 IU/mg, i.e. 8 - 9% of the original activity. The concentration of the enzyme solutions was determined on the basis of absorption coefficients of A;& = 4.9 for the yeast enzyme [26,30] and A t ? , = 6.9 both for the muscle enzyme and for its modified derivatives [23, 261. The molar absorption coefficient of ATP was taken as E~~~ nm = 15400 M-’ cm-’ [32]. The absorbance of ATP (it is known

X-ray-scattering intensity was measured by a Kratky small-angle camera on the installation described previously [33]. The absolute intensities were determined via a calibrated lupolen sample [34]. X-ray scattering intensities from solutions of different forms of phosphoglycerate kinases were obtained at protein concentrations between about 5 - 40 mg/ mi. The scattering intensitics were measured in the scatteringangle region corresponding to the scattering-vector module (p)O.l2-5.5 nrn-l(p = 4nsin0//2,where;listhewavelength, 20 is the scattering angle) for higher enzyme concentrations and 0.1 2 - 2.0 nm- for lower concentrations. To compensate the scattering of unbound substrates, the scattering of solvents with appropriate amounts of substrates was measured. The relative statistical errors of difference scattering intensities were 0.5% at the start of the scattering curves and z 35 % at p = 2.0-5.5 nm-l. The absolute scale of scattering intensity (i.e. the scattering/electron) was used for treatment and demonstration of the obtained data. For each protein form studied the scattering curves on the considered absolute scale did not depend on protein concentration at p > 0.8 - 1.0 nmThe extrapolation of scattering curves at smaller p to zero protein concentration was made by linear regression of scattering intensities. To study the differences between the ‘apo’ (unliganded protein) and the ‘holo’ (ternary complex) 3-phosphoglycerate kinases the measurements were done in a bisectorial cell. All measurements were made at 20 1 “C. No activity loss was detected for any active phosphoglycerate kinase after Xray measurements. Two methods were used for determination of values of radii of gyration ( R , values). The first, the so-called ‘standard’ method [35], is based on the application of Guinier approximation [36] for desmeared experimental scattering curves at small-angle region:

’

’.

lgZ(p)

= lgZ(0)

-

R,”lge P2 3

~

( I is the scattering intensity) with subsequent extrapolation

of the R, values obtained to zero protein concentration. To desmear the experimental scattering curves the program ‘Syrena’ was used, which is based on algorithms described previously [37, 381. Guinier approximation and extrapolation of R, values to zero protein concentration were made by appropriate linear regressions. Guinier approximation was made for p < 0.6 nm-’ (see below). The second method is essentially indirect Fourier transformation of scattering indicators, suggested by Glatter [39]. The Glatter algorithm is based on the following equations [39]: dmax

I ( p ) = 471

sin p r

j p(r)-dr

CLr

0

dmax

@

dmax

p(r)r2

= 0

j p(r)dr 0

where p ( r ) is the pair distance distribution function of the scattering centers of the particle and d,,,, is the maximal

63

I

I 0

I

1

I

0.2

I

0.4

I

1

I

0.6

Fig. 1 . Guinier plots 1gI versus p2for yeast and different types of pig musclephosphoglycerate kinase. I is the desmeared scattering intensity (absolute scale) after extrapolation to zero protein concentration. The angle range for Guinier approximation (indicated by arrow) has been chosen on the basis of that for the calculated scattering curve for the crystalline structure of ‘apo’-yeast enzyme (+), which corresponds to R, = 2.49 nm. (For explanation of the symbols see Fig. 2.)

Fig. 2. Concentration dependences of radii of gyration f o r yeast and different t.ypes of pig muscle phosphoglycerate kinases. ‘Apo’ ( 0 ) and ‘h010’- ( 0 )yeast enzyme; ‘apo’- (0)and ‘ho1o’- (m) pig muscle enzyme; ‘apo’- (0)and ‘holo’- (+) methylated pig enzyme; ‘apo’( A ) and ‘ho1o’- (A)carboxamidomethylated pig enzyme

distance in the particle. The Glatter program permits calculation of the p ( r ) function and is thus a way to obtain the R, value of the protein molecule. To determine R, values by the Glatter program, experimental scattering curves, extrapolated to zero protein concentration, were used in the range from p l = 0.12-0.20 nm-’ t o p 2 = 2.2 nm-’.Thechoiceoflarger values for p2 did not affect the results of calculation. The values of d,, were chosen according to Glatter’s criteria [40].

was eliminated by extrapolation of these values to zero protein concentration (see Fig. 2). The R, values obtained from Fig. 2 are listed in Table 1 . R, values calculated by the Glatter algorithm are also presented in Table 1 (marked by asterisks) and they agree well with those obtained by the ‘standard’ method. The table shows that simultaneous binding of 3phosphoglycerate kinase and MgATP to pig muscle 3phosphoglycerate kinase leads to a considerable decrease of R, (0.12 nm), which is practically the same as for the yeast enzyme (0.11 nm). A similar decrease of R, (0.10 nm) has been obtained for the SUbStrdteS binding to the active methylated pig muscle enzyme. In contrast to the active enzymes, the substrate binding to the inactive carboxamidomethylated pig muscle 3-phosphoglycerate kinase does not lead to a measurable decrease of its R, value. The decrease of R, values (0.10-0.12 nm) obtained for the active phosphoglycerate kinases upon simultaneous substrate binding can be due only to a large-scale structural change of the enzyme molecule. In fact, a direct calculation of R, decrease due to substrate binding to the ‘apo’ form of the yeast enzyme (based on protein atomic coordinates and the proposed [lo, 411 substrate positions) gives a value of only 0.012 nm. It follows that substrate binding leads to large-scale structural changes both in pig muscle 3-phosphoglycerate kinase and its active methylated derivative. Thus, the results on R, values strongly support the previous suggestion based on chemical modification data [23- 251. It has been shown for yeast 3-phosphoglycerate kinase [18, 191 that the substrate-induced large-scale structural change of the enzyme affects the X-ray scattering at middle and large scattering angles as well. Scattering curves for ‘apo’ and ‘holo’ forms both of yeast and pig muscle 3-phosphoglycerate kinases obtained for this scattering angle region are shown in Fig. 3. Fig. 4 presents the differences between scattering of ‘apo’ and ‘holo’ forms for all the investigated proteins. From Fig. 4A and 4B one can see that substrate-binding influences

Calculation of scattering curves

Scattering curves for the crystalline structure of yeast 3-phosphoglycerate kinase (with and without substrates) were calculated according to Pavlov and Fedorov [20] on the basis of atomic coordinates [lo, 411. The molecular mass of 3-phosphoglycerate kinase obtained from diffuse X-ray scattering were 43 - 44 kDa which are close to the values calculated from the amino acid contents of yeast [42] and horse muscle [43] enzymes (44.5 kDa). To estimate the molecular mass of yeast and pig muscle enzymes, their partial-specific-volume values of 0.749 cm3/g and 0.744 cm3/g, respectively, were calculated on the basis of their chemical compositions [44].

RESULTS AND DISCUSSION To check the suggestion on the functional importance of substrate-induced large-scale structural changes in phosphoglycerate kinase we have compared the radii of gyration of the ‘apo’ and ‘holo’ forms for each investigated protein. To determine R, values by Guinier approximation the interval of p < 0.6 nm-’ was used, which was chosen from the consideration of the scattering curve calculated for a crystal structure of ‘apo’-(yeast phosphoglycerate kinase) (see Fig. 1). The effect of intermolecular scattering on R, values

64 Table 1. Correlation hetween enzymic activity of3-phosphoglycerate kinase and substrate-induced decreases of’their rudii ofgyralion

Type of 3-phosphoglycerate kinase

Radius ofgyration (R,) AR,

Substrate binding

Enzymic activity

0.11 & 0.04 0.12a

+

+

0.12 & 0.03 0.12”

+

+

0.10 I 0.04 0.1 1a

+

t

0.04

+

-

nm ‘apo’

2.53 f 0.03 2.54“ (7.8 nm)

Yeast ‘holo’

2.42 0.03 2.42“ (7.0 nm)

‘apo’

2.54 I 0 . 0 2 2.54” (8.0 nm)

Pig muscle ‘holo’

2.42 I 0.02 2.42“ (7.2 nm)

‘apo’

2.49 0.03 2.50” (7.8 nm)

Methylated pig muscle ‘holo’

2.39 & 0.03 2.39” (7.0 nm)

‘apo’

2.50 & 0.03 2.49“ (7.6 nm) 0.01

Carboxamidomethylated pig muscle ‘holo’

-0.01

a

2.49 0.03 2.50“ (7.8 nm)

The R, values obtained according to the Glatter algorithm [39,40] with d,,,

shown in brackets.

and ’ho1o’- ( @ ) Fig. 3. Comparison of experimental (smeared) scattering curves extrapolated to zero protein concentration f o r ‘apo’- (0) 3-phosphoglycerute kinuses. (A) and (B) correspond to yeast and pig muscle enzymes, respectively

the scattering curves of yeast and pig muscle 3phosphoglycerate kinases in a similar way: the scattering of ‘holo’ forms is larger than that of ‘apo’ forms for 0.12 < p < 1.0 nm-’ and predominantly smaller for p > I .O nm- The effect of substrate binding on the scattering curve of the active methylated pig inuscle enzyme is very similar (Fig. 4C). On the other hand, the influence of substrate binding on the scattering curve of the inactive carbox-

amidomethylated pig muscle enzyme (Fig. 4D) is markedly different: the scattering of ‘apo’ and ‘holo’ forms coincide up to about p = 1.O nm- while at larger y the scattering of the ‘bolo' form is predominantly larger. The influence of substrate binding on the scattering curve at middle and large angles can be due both to conformational changes in a protein and to the scattering of the bound substrate molecules. To check the second possibility we have

’,

65

I

A

-0.11 /

-

O .0i Looo ooooooo

onoo “oo“00 0

oooo 00 0

0 0 -

0 00

ooo

Fig. 4. Comparison ofsubstrate-induced effects on the scattering curves for yeast ( A ) , pig muscle ( B ) , methylatedpig muscle ( C ) and carbox; amidometh;lated pig muscle (0) 3-phosphoglycerate kinases. A Ig I (p) = lg cYo(p)-ig p20(p), where CY,, ( p ) and $”Jo ( p ) are the experimental (smeared) scattering intensities extrapolated to zero protein concentration for the ‘apo’- and ‘bolo'-forms of the enzyme, respectively

A

calculated the scattering curve for the ‘apo’ form of yeast 3-phosphoglycerate kinase from its X-ray atomic coordinates [lo, 411 and compared it with the scattering curve for the same enzyme conformation with bound MgATP and 3-phosphoglycerate (whose coordinates were taken from [lo]). The difference between two curves is actually close to the difference between the experimental curves for carboxamidomethylated pig muscle enzyme (see Fig. 4D). This suggests that the effect of substrate binding in the latter case is due mainly to the scattering molecules rather than to the conformational changes of the protein. However we cannot exclude a small ‘sliding’ of the domains which does not influence the radius of gyration but can affect scattering intensity at middle and large angles [19, 211. On the other hand, the fact that the change of the scattering curves of all active 3phosphoglycerate kinases is different from that of the carboxamidomethylated enzyme confirms that this change cannot be attributed only to the scattering of bound substrate molecules. Fig. 5 presents p(r)-functions calculated from experimental scattering curves for ‘ap0’- and ‘ho1o’-forms of the investigated enzymes. One can see that the simultaneous binding of substrates leads to the increase of middle distances and the decrease of large distances in yeast and pig muscle 3-phosphoglycerate kinases as well as in the active methylated pig muscle enzyme. On the other hand p(r)-functions for the ‘apo’- and ‘ho1o’-forms of the inactive carboxamidomethylated pig enzyme practically coincide. Thus, the real-space representation of the experimental data visualizes large-scale

0

I

0

h

L u

L

Q

D

I

0

I

Fig. 5. p(r) .functions.for ‘app0’- (0)and 'bolo'- ( 0 )forms ofyeast ( A ) .pig muscle ( B ) , methylatedpig muscle ( C ) and carboxamidomethyluted pig muscle ( D ) 3-phosphoglycerate kinases calculuted,from the corresponding experimental scattering curves. See Materials and Methods

66 structural changes in the active enzymes upon the substrate binding and the absence of these changes for the inactive carboxamidomethylated enzyme. What is the nature of the structural changes in pig muscle 3-phosphoglycerate kinase? Unfortunately, up to now there is no X-ray structural analysis of this enzyme. However, as pointed out in the Introduction, 3-phosphoglycerate kinases from different sources are structurally very similar. We have shown earlier [19] by diffuse X-ray scattering that substrate binding leads to the ‘locking’ of two domains of yeast 3phosphoglycerate kinase. As the changes of both radii of gyration and scattering curves at middle and large angles for yeast and pig muscle 3-phosphoglycerate kinases are very similar, it is reasonable to conclude that a similar ‘locking’ of domains takes place also in the case of pig muscle enzyme (as well as in its active methylated derivative). A locking movement should lead to the decrease of the distance between the inertia centers of the domains and hence to the decrease in the radius of gyration [19, 211. In contrast, the absence of the above differences both between the radii of gyration and the scattering curves for the ‘apo’- and ‘ho1o’forms of the inactive carboxamidomethylated pig muscle 3-phosphoglycerate kinase permits one to exclude domain closure, i.e. the ‘locking’ type of domain displacements. It follows that it is likely that the inter-domain cleft remains open upon substrate binding to this modified enzyme. Thus, a correlation exists between the activity of 3-phosphoglycerate kinase and the domain ‘locking’ ability. This is a strong argument in favour of the functional importance of the ‘hinge-bending’ structural motion postulated previously [121. The authors are indebted to G.S. Polubesov for providing the precision X-ray measurements, to Ms. I. Szamosi for excellent technical assistance and to Dr D. I. Svergun (Institute of Crystallography, Academy of Sciences of the USSR, Moscow) for presentation of the ‘Syrena’ program.

11. 12. 13. 14. 15. 16. 17.

~

18.

19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.

REFERENCES 1. Bennett, W. S. & Steitz, T. A. (1978) Proc. Natl Acad. Sci. U S A 75,4848-44852. 2. Eklund, H., Samama, J.-P., Wallen, L., Branden, C.-I., Akeson, A. & Joncs, T. A. (1981) J . Mol. B i d . 146, 561 -587. 3. Remington, S., Wiegand, G. & Huber, R. (1982) J . Mol. B i d . 158, 111-152. 4. Koshland, D. E. 81 Neet, K. E. (1968) Annu. Rev. Biochem. 37, 359-410. 5. Anderson, C. M., Zucker, F. H. & Steitz, T. A. (1979) Science (Wash. D.C.) 204, 375-380. 6. Scopes, R. K. (1973) in The Enzymes, 3rd Edn (Boyer, P. D., ed.) vol. 8, pp. 335 - 352, Academic Press, New York. 7. Fifis, T. & Scopes, R. K. (1978) Biochem. J . 175, 311-319. 8. Mori, N., Singer-Sam, J. & Riggs, A. D. (1986) FEBS Lett. 204, 31 3-317. 9. Bryant, T. N., Watson, H. C. & Wendell, P. L. (1974) Nature (Lond.) 247,14-17. 10. Watson, H. C., Walker, N. P. C., Shaw, P. J., Bryant, T. N , Wendell, P. L., Forthergill, L. A., Perkins, R. E., Conroy, S.

C., Dobson, M. J., Tuite, M. E., Kingsman, A. I. & Kingsman, S. M. (1982) EMBO J . 1, 1635-1640. Blake, C. C. F. &Evans, P. R. (1974) J . Mol. Biol. 84, 585-601. Banks, R. D., Blake, C. C. F., Evans, P. R., Haser, R., Rice, D. W., Hardy, G. W., Merrett, M. &Phillips, A. W. (1979) Nature (Lond.) 279,773-777. Blake, C. C. E. & Rice, D. W. (1981) Philos. Trans. R. Soc. Lond. A , 293,93 - 104. Mas, M. T., Chen, C. Y., Hitzeman, R. A. & Riggs. A. D. (1986) Science (Wash. D.C.) 233, 788 - 790. Blake, C. C. F., Rice, D. W. & Cohen, F. E. (1986) Int. J . P e p . Protein Res. 27, 443-448. Watson, H. C . & Gambilln, S. J . (1985) J. (Biosci. Bungalore) SUPPI.8,499 - 506. Pickover, C. A,, McKay, D. B., Engelman, D. M. & Steitz, T. A. (1979) J . Biol. Chem. 254, 11323 1 1 329. Timchenko, A. A. &Tsyuryupa, S. N. (1982) Biofizika27,10171021. Ptitsyn, 0. B., Pavlov, M. Yu., Sinev, M. A. & Timchenko, A. A. (1986) in Multidomainproteins (Patthy, L. & Friedrich, P., eds) pp. 9-25, Academiai Kiado, Budapest. Pavlov, M. Yu. & Fedorov, B. A. (1983) Biopolymers 22, 15071522. Pavlov, M. Yu. (1985) Dokl. Acad. Nauk S S S R 281,458-462. Ptitsyn, 0. B. (1978) FEBS Lett. 93, 1 -4. Dtkany, K. & Vas, M. (1984) Eur. J . Biochem. 139, 125- 130. Tompa, P., Hong, P. T. & Vas, M. (1986) Eur. J . Biochem. 154, 643 - 649. Vas, M. & Csanady, G. (1987) Eur. J . Biochem. 163,365-368. Krietsch, W. K. G. & Biicher, T. (1970) Eur. J . Biochem. 17,568580. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70-77. D’Alessio, V. M., Perna, P. J. & Paule, M. R. (1979) J . Biol. Chem. 254, 11282- 11 287. Cartier, P. P., Leroux, J. P. & Marchand, J. (1967) Ann. Biol. Clin. 25, 109-136. Scopes, R. K. (1978) Eur. J . Biochem. 91, 119-129. Vas, M. & Batke, J. (1984) Eur. J . Biochem. 139, 115-123. Bock, R. M., Ling, N.-S., Morell, S. A. & Lipton, S. H . (1956) Arch. Biochem. Biophys. 62,253 -264. Serdyuk, I. N. & Grenader, A. K. (1977) Eur. J . Biochem. 79, 495 - 504. Kratky, 0. & Pilz, I. (1972) Q . Rev. Biophys. 5, 481 -537. Asadchikov, V. E., Lamzin, V. S., Egorov, A. M. & Popov, V. 0. (1986) Stud. Biophys. 112,197-200. Guinier, A. & Fournet, G. (1955) Small-angle scattering of X rays, J. Wiley & Sons, New York; Chapman & Hall, London. Schedrin, B. M. & Feigin, L. A. (1966) Kristallographiya (USSRj 11, 159-163. Rolbin, Y. A., Svergun, D. I. & Schedrin, B. M. (1980) Kristallographiya ( U S S R ) 25, 23 1 - 239. Clatter, 0. (1977) J . Appl. CrystaNogr. 10,415-421. Clatter, 0. (1982) in Small-angle X-ray scattering (Clatter, 0. & Kratky, O., eds) pp. 119- 196, Academic Press, New York. Bernstein, F. C., Koetzle, T. F., Williams, G. J. B., Meyer, E. F., Brice, M. D., Rodgers, J. R., Kennard, O., Shimanouchi, T. & Tasumi, M. (1977) Eur. J . Biochem. 80, 319-324. Perkins, R. E., Conroy, S. C., Dunbar, B., Fothergill, L. A,, Tuite, M. F., Dobson, M. J., Kingsman, S. M. & Kingsman, A. J. (1983) Biochem. J . 211, 199-218. Merrett, M. (1981) J . Biol. Chem. 256, 10293-10305. Lee, J. C. & Timasheff, S. N. (1974) Arch. Biochem. Biophys. 165, 268 - 273.

34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.