Biochimica et Biophysica Acta, 379 (1975) 147-156 © Elsevier Scientific Publishing Company, Amsterdam - - Printed in The Netherlands

BBA 36909

HYDROGEN-ION TITRATION CURVE OF OVOMUCOID

ABDUL WAHEED and A. SALAHUDDIN Department of Biochemistry, Jawaharlal Nehru Medical College, A.M.U., Aligarh, 202001 (India)

(Received June 24th, 1974)

SUMMARY

A systematic study of the H ÷ titration curve of purified ovomucoid was made at three temperatures (15, 25 and 35 °C) and three ionic strengths (0.05, 0.15 and 1.0). In all, 49 protons were dissociated reversibly in the pH range, 2.0-12.0. From the analysis of the results up to pH 12.0, the numbers of different dissociable groups per 28 300 g protein, together with their intrinsic pK values in parentheses were found to be; 27 side-chain carboxyl (pKint = 4.0), four imidazole (pKlnt == 6.5), one a-amino ( p g i n t = 7 .5 ) , 12 e-amino (pKint = 9.6), one guanidino (pKlnt = 11.8) and one a- carboxyl group with abnormally low pK. The total number of basic nitrogens per mole of the protein was 22 so that four guanidino groups remained untitrated up to pH 12.0. Spectrophotometric titration showed that three out of five phenolic groups were titrated reversibly up to pH 11.9 with an intrinsic pK of 10.25; the remaining two groups became accessible only on protein denaturation. Viscosity results suggest- ed absence of conformational change in the pH range 2.0-11.2. This explains the constancy of the pK values of carboxyl groups in the pH range 2.0-5.0. The empirical value of the electrostatic interaction factor, w, was 0.04, both in the carboxyl and phenolic regions.

INTRODUCTION

The indispensable prerequisite to a proper understanding of the pH dependence of the biological [1] and physical properties [1, 2] of ovomucoid is the study of its H + equilibria, which, to our knowledge, has not been systematically investigated so far. In this paper we report our titration results on a purified ovomucoid [3]. The data on the reduced viscosity of the protein as a function of pH have also been obtained to detect conformational change, if any, during titration.

MATERIALS AND METHODS

Ovomucoid was isolated and purified by salt precipitation and by subsequent chromatography on sulphoethyl-Sephadex using a linear salt gradient as described elsewhere [3]. The protein was homogeneous with respect to both size and charge as evident from a single symmetrical peak in gel chromatography, a single band on sodium dodecylsulphate-polyacrylamide gel electrophoresis and a single band in

148

polyacrylamide gel electrophoresis in presence of 9 M urea [3]. The molecular weight of the protein as determined [3] by sodium dodecylsulphate-polyacrylamide gel electrophoresis was 28 300.

Constant-boiling HCI [4], CO3~--free KOH [5] and pure nitrogen [6] were prepared. The concentration of NaOH was determined by actual potentiometric titration using HC1. Strong cation (Dowex 50W-X8) and anion (Dowex l-X4) ex- change resins were from John T. Baker Co., (U.S.A.). These were used in the prepa- ration of resins in hydrogen, hydroxyl, ammonium and acetate cycles to be used in Dintzis column [7]. KC1, potassium acid phthalate, and sodium tetraborate were recrystallized from water. Other reagents were of analytical grade. Glass-double- distilled water was used throughout.

Determination of protein concentration The protein concentration was determined by heating a known weight of iso-

ionic ovomucoid at 105-107 °C to constant weight. The protein concentration thus determined in quadruplicate agreed within 4- 0.1 ~.

pH measurements The pH was measured on an Elico pH meter model LI-10 (Electronics,

Hydrabad, India) in conjunction with the Elico glass and saturated calomel electrodes in a locally fabricated titration assembly of the type used by Tanford [8]. The liquid junction was carefully established in a capillary near the stop-cock and the temper- ature of the titration vessels was maintained within 4- 0.2 °C by circulating water from ultrathermostat type U 10 (G.D.R.). The pH meter was standardized before and after each titration using 0.05 M potassium acid phthalate or 0.01 M sodium tetra- borate. In alkaline region, the pH was measured in an atmosphere of nitrogen.

Viscosity measurement The reduced viscosity of ovomucoid was measured using a calibrated Cannon

Fenske viscometer with a flow time of 400 s for distilled water at 25 °C as described earlier [9]. The times of fall of the solvent, to, and that of protein solution, t, were measured and the reduced viscosity, ~red, calculated by the equation,

~]red = ( t - - to)/toc -- (1 -- v2Q0)/~0 (1)

where c is the protein concentration which was 0.21 g/100 ml, O0 is the density of the solvent (1.0003 g/ml) and ~2 is the partial specific volume. The value of vz was taken [2] to be 0.697 ml/g.

The isoionic preparation of ovomucoid was obtained by passing an extensively dialyzed aqueous solution of the protein through a mixed-bed ion-exchange column prepared according to Dintzis [7]. The pH of the emerging solution was, by definition, isoionic pH.

Ovomucoid was titrated by the discontinuous method at different ionic strengths and temperatures. The protein concentration was generally in the range 0.3~.9 ~ ; higher protein concentration was used for the titration in the pH range in which the number of titratable groups was smaller. The isoionic pH was taken as the reference point in the titration. The solutions were prepared by taking ovomucoid,

149

HC1 or KOH and KC1 by weight in calibrated 5-ml volumetric flasks and the total volume made up to 5 ml in each case. The pH of each solution was measured within 4-5 h after addition of the acid or alkali.

The activity coefficients of H ÷ and O H - were determined in the extreme acidic and basic pH ranges from the blank titration of the solvent using appropriate equa- tions [8] and then the concentrations of free H ÷ or O H - were calculated from the measured pH. The amount of H ÷ bound to or dissociated from 28 300 g of ovomucoid was computed from the difference of the amount of H + (or O H - ) present in the solution. The results were expressed in terms of the number of H ÷ dissociated, ?, per mole of ovomucoid. The experimental uncertainty in the determination of ? was about 4~o in the pH ranges 2-5 and 8-12. However, in the pH range 5-8 the average error was higher, as expected (6 ~).

Heat of ionization The apparent heat of ionization, AH, was calculated according to Wyman [10],

using the expression,

AH = --2.303 R [dpH/d(I/T)] (2)

where dpH is the change in pH caused by the change in temperature, d(1/T), for the same value of ~. Below pH 10, dpH was calculated from the actual data obtained in the temperature intervals 15-25 °C and 25-35 °C. Above pH 10, dpH was read from the titration curves.

Spectrophotometric titration Spectrophotometric titration of the phenoxyl groups of ovomucoid was per-

formed according to Crammer and Neuberger [11]. The absorbance of the protein solution in the ultraviolet region was measured on a Carl Zeiss Jena spectrophoto- meter model VSU2-P (Germany) using appropriate controls, at 30 °C. Protein solu- tions of different pH values, in the pH range 7.0-12.75, were prepared on weight basis, by taking KCI, KOH and about 30/ ,moles of ovomucoid in calibrated 5 ml volumetric flasks. The increase in the molar extinction of the ovomucoid at a given pH, over the value at pH 7.0, gave the difference in molar extinction coefficient, Ae, which was taken to be a measure of the ionization of phenolic groups of the protein.

RESULTS

The isoionic solution of ovomucoid was clear and the pH of its 1 ~o solution was 4.55 ± 0.05 at 25 °C which expectedly increased to pH 4.70 on lowering the protein concentration to 0.6 ~. The variations in isoionic pH with increase in ionic strength from 0 to 1.0 were random and experimentally insignificant. The net charge, 7, on the ovomucoid molecule was calculated from the measured isoionic pH by the help of Eqn. 31 of ref. 12 and was very close to zero (0.07): it remained unaltered either by increasing the ionic strength from 0 to 1.0 or by lowering the protein concen- tration to 0 .6~. These results are interesting, for they indicate that ovomucoid does not bind ions other than H ÷ in acid pH range and that the charge on the protein

150

molecule is due to the proton binding so that Z can be replaced by ZTH which repre- sents the net charge on the protein molecule due to protonation.

H ÷ titration curves of ovomucoid were obtained in the pH range 1.6-12.0 al three ionic strengths (0.05, 0.15 and 1.0) and at three temperatures (15, 25 and 35 °C) The results found at one ionic strength, i.e. 0.15 and at 15, 25 and 35 °C are shown in Fig. 1 ; the data at other ionic strengths (0.05 and 1.0) were omitted for clarity. The reversibility of the titration was tested by first exposing the protein to pH values 1.8 6.7 and 12.0 and then adjusting the pH to the values indicated in Fig. l by the additio~ of acid or alkali. A good agreement between the experimental points obtained in the forward and reverse titrations showed the reversibility of the titration curve in the pH range 2.0-12.0. Further, the pH of ovomucoid solution (ionic strength 0.15) was determined to be the same at 25 °C both during heating from 15 to 35 °C and cooling back from 35 to 15 °C, suggesting that the effect of temperature on the titration was reversible.

0

"a

"~ 1(] o

-~ 20

~ 3c ~ o

~ 4G

~ 5 0

6 i i i i t i

3 5 7 9 11 13

?H

Fig. I. H ÷ titration curve of ovomucoid at different temperatures and at 0.15 ionic strength: O 15 ° C; O, 25 °C; L 35 °C. The experimental points for the reverse titration at 25 °C, ionic strength 0.15 were obtained by adjusting the pH: O, pH 1.8; I , pH 6.7; D, pH 12.0.

The absorption spectra of ovomucoid at pH 7.0 showed one trough at 255 nm and two maxima at 278 and 280 nm (see Fig. 2). On raising the pH to 12.0 the two maxima shifted to 282 and 285 nm, respectively. The observed features in the alkaline difference spectra, i.e. two maxima at 245 and 295 nm and the absence of a shoulder at 285 nm, are known to be characteristics of the ionization of tyrosine [13]. Although Ae at 245 nm is markedly higher than at 295 nm, the latter was selected for monitoring the ionization of tyrosine groups due to the fact that, in addition to tyrosine ioni- zation, the possible alkaline hydrolysis of disulphide bonds will also make a significant contribution to Ae at 245 nm [14]. The maximum value of Ae at 295 nm was 11 791 cm2.mole -1 at 0.15 ionic strength, 30 °C. Since Ae at 295 nm for tyrosine [l l] is 2300 cm 2"mOle -~ the number of phenolic residues per mole of ovomucoid was com- puted to be 5.1.

From Fig. 1 it is clear that 28 protons were titrated in the pH range 1.6-6.1.

151

%

~E

60

50

40

30

2C

10

! 60

40 , %

20 C 320

o.0

k "',, a--~o a~o

240 260 280 300 320 340

WaVelenf;t h, nm

Fig. 2. U l t r av io l e t absorp t ion spectra o f ovomuco id at ionic strength 0.15, and 30 ° C . - - - - - - 7.0; - - - - - , pH 12.0. The inset shows the alkaline difference spectrum.

, p H

The break in the titration curve at pH 6.1 (f = 28.4), and the fact that titration of carboxyl groups in proteins would be completed at pH 6.0 [12], suggested that the 28 dissociated protons belong to the carboxyl groups of the protein. This should include a-, fl- and 7-carboxyl groups derived 'from C-terminal, aspartyl and glutamyl residues of the protein, respectively. The single proton which dissociated at pH 2.0 was possibly derived from the a-carboxyl group. Five additional groups were dissociated in the pH range 6.1-8.4, in which range imidazole and a-amino groups of proteins would undergo deprotonation [12]. Thus it seems that the five protons are dissociated from the four imidazole and one a-amino groups of the protein. It should be noted that 16 additional protons were titrated in the pH range 8.6-11.9. The ionizable groups of proteins which are expected to contribute protons in this pH range are e-amino, phenolic and guanidino groups [12]. By separate spectrophotometric titration, the number of phenoxyl groups available for titration up to pH 12.0 in the native protein was three. Thus, the remaining 13 titratable groups would include e-amino and guani- dino groups. The pK value of e-amino groups in proteins has been found to be 10.4 [12], so that probably all the e-amino groups in ovomucoid must have been titrated up to pH 11.9. The value of ~ corresponding to the isoionic pH of 4.7 was 22 (see Fig. 1). This gives the number of the basic nitrogen groups of the protein. Out of the total 22 basic groups, only 18 were available for reversible titration and the remaining four were presumably guanidino groups which could not be titrated up to pH 11.9. This is because the pK value of guanidino group in proteins is generally greater than 12.0 [12].

From the data on the heat of ionization, AH~o., illustrated in Fig. 3 it is clear that the value of f in the AH~o, region of -t= 2 kcal/mole, which characterizes the ionization of carboxyl groups in proteins [12], is 28. Thus, the number of titratable carboxyl groups in ovomucoid would be 28. Because of the overlap of the ZlHlo, values for imidazole and phenoxyl groups further analysis of the data was not done [121.

152

8

4 c

0 CIrO-O-OO- I I

0 10 20 30 40

Number of protons dissociated per 28300 l of ovomucoid

Fig. 3. Apparent heat of ionization of ionizable groups of ovomucoid at ionic strength 0.15. Th, values of AH~on were calculated in the temperature range: O, 15-25 °C; and 0, 25-35 °C.

The phenolic titrations carried out at 25 °C and at 0.05, 0.15 and 1.0 ioni( strengths are reversible up to pH 12.0 at which the Ae value corresponded to th( ionization of three phenolic groups: further increase in pH to 12.75 caused the ex posure of two more phenolic groups due to alkaline denaturation.

12 ?1 10

6

4

2

0 8 9 10 11 12

pH

Fig. 4. Spectrophotometric titration of ovomucoid, at 295 nm, 30 °C. Ionic strengths were: O, 0.05; 0 , 0.15; [-1, 1.0. The reversibility of the titration was studied by lowering the pH from: U, pH 12.0. The broken lines show time dependence of the phenolic titration.

The amino acid analysis of ovomucoid preparation has not been done. How- ever, this preparation was identical [3] to the IIR ovomucoid obtained by Beeley [15] in (i) anti-tryptic activity; (ii) elution profile on sulphoethyl-Sephadex; (iii) hexose content; (iv) glucosamine content; and (v) sialic acid content. The number of ionizable groups were deduced from amino acid composition data of Beeley [15] for IIR ovomucoid and are listed in Table I.

153

TABLE I

TITRATION DATA FOR OVOMUCOID AT 25 °C, IONIC STRENGTH 0.15

Values in parentheses are those which have been used to generate the titration curve. The analytical data were obtained from Beeley [15] and the expected values of pgint are taken from the literature [12].

Groups Number of groups Intrinsic pK

Predicted by Found by amino acid content titration

Expected Observed w

a-Carboxyl I a fl- and ),-carboxyl 32 b Imidazole 4 a-Amino 1 ¢ e-Amino 12 Phenolic 5 Guanidino 5

28 4.6 4.04 4 6.3 (6.5) 1 7 .5 (7 .5 )

(12) 10.4 (9.60) 3 + 2 d 9.6 10.25

(5) 12 (11.80)

0.040

0.040

a Ref. 16. b Ref. 17. c Ref. 18. a Inaccessible.

Viscosity data The reduced viscosity of ovomucoid was measured in the pH range 1.2-12.0

at 25 °C, ionic strength 0.15 and the results are graphically shown in Fig. 5. The reduced viscosity remained unaltered in the pH range 1.9-11.2 but increased as the pH was lowered below pH 1.9 or raised above 11.9.

0 2

ooo---o

I I

4 6 pH

8 I0 I

Fig. 5. Reduced viscosity of ovomucoid as a function of pH. Ionic strength 0.15; temperature 25 -¢- 0.05 °C.

Thermodynamic analysis Thermodynamic analysis of the titration results was done following Linder-

strom-Lang [19] using the equation:

pH - - l o g P i / ( n i - - r i ) = p K i . t - - 0.868 wZ

where w = eZ/2DkT (lib -- K/(1 + ra)

(3)

(4)

In Eqn 3, ni refers to the total number of the groups with an intrinsic pK of Pgin t , ri is the number of such ionized groups, Z 7 is the mean net charge on the protein

154

molecule, w is electrostatic interaction factor, b is the radius of the protein sphere, a is b + 2.5/~; other terms have the usual significance. In Eqn 3, Z can be replaced by ZH which can be determined by subtracting ~ from (ZH)max. The latter represents the number of protons dissociated from ovomucoid at its isoionic pH and was 22.

The titration results in the carboxyl region (pH 2.0-3.65) as well as in the phe- nolic region (pH 9.5-12.0) were analyzed according to Eqn 3, by plotting pH -- log Pj / (n~ - - r~) against ZH; straight lines were drawn by the method of least squares (see Fig. 6). The values of nj for carboxyl and phenolic groups used in these calculations were 28 and 3, respectively. The values of pKin~ and w calculated from carboxyl and phenolic regions are listed in Table II. It should be noted that plot of log ~ / ( n ~ - - Pi )

vs pH was linear both for carboxyl and phenolic regions indicating the constancy of the pK~nt of the two groups. The electrostatic interaction factor, w, was expectedly influenced by the ionic strength in the carboxyl and phenolic regions.

4

3

4

3

4!

12

k 10

121

A III

3o

. . . . . . I PH

5 10 15 20

II

- - ° ' " i. 0 -10 - 2 0 -50 - 4 0

2H

27--

2 9 10 11 pH

Fig. 6. Titration data for the carboxyl (A, 25 °C) and phenolic (B, 30 °C) groups of ovomucoid, at ionic strengths 0.05 (Curve I), 0.15 (Curve II) and 1.0 (Curve Ill) plotted according to Eqn 3. The insets show the plot of log f j / (n~ - - f,) vs pH.

The theoretical titration curve shown by the solid line in Fig. 1 was drawn with the observed values of pH, ri, ni, Z , , w and the assumed values of PKint (shown in parentheses in Table I). A good agreement was generally obtained between the experimental and calculated points. Thus, by implication, the values of pKl,t of imi- dazole, a-amino, and e-amino groups appeared to be 6.5, 7.5 and 9.6, respectively. The theoretical titration curve was consistent with the titration of one guanidino group having a pKlnt ofl 1.9.

155

TABLE II

EFFECT OF IONIC STRENGTH ON w AND pglnt FOR THE CARBOXYL AND PHENOLIC GROUPS OF OVOMUCOID AT 25 °C

Group Ionic pKlnt w strength

Carboxyl 0.05 3.98 0.043 0.15 4.04 0.039 1.00 4.10 0.032

Phenolic 0.05 10.27 0.046 0.15 10.25 0.042 1.00 10.14 0.025

DISCUSSION

From the titration results presented above, several important conclusions of interest concerning the structure of ovomucoid can be reached.

A total of 49 ionizable groups were available for titration up to pH 12.0 in the native ovomucoid. These included 28 carbosxyl (one a-, 27 fl- and 7-carboxyl groups), four imidazole, one a-amino, three phenolic and 13 e-amino and guanidino groups. However, the titration curve passing through the experimental points could be generated by assuming the dissociation of 12 e-amino groups and one guanidino group up to pH 12.0. Since the titration data showed 22 basic nitrogen groups per mole, four guanidino groups still remained to be titrated up to pH 12.0. The accessibility of all the e-amino groups [20] and the existence of 22 basic groups [20] in ovomucoid were also noted earlier. The number of titratable cnrboxyl groups is five short of the analytical value. This was confirmed by our titration data in 9 M urea [3] in which 32 carboxyl groups could be titrated. I t is conceivable that the carboxyl groups are forced into contact with centres of positive charge in the native protein.

Out of the five phenolic groups in ovomucoid, only three were titrated up to pH 12.0, the remaining two became exposed only by alkaline denaturation or by 9 M urea. The exposure of 60 ~o tyrosine residues in ovomucoid has been revealed by our unpublished results on the solvent perturbation of the fluorescence of ovomucoid as well as by solvent perturbation data of Donovan [21].

The electrostatic interaction factor, w, for ovomucoid would depend, among others, on the geometry and hence on the overall native protein conformation. The latter remains unaltered in the pH range 2-12 as judged by viscosity data. This is also indicated, although indirectly, by the linearity of the plot of log ~ l / ( n l - - Fl) vs pH in the two pH ranges, i.e. pH 2.0-5.5 and pH 9.5-11.9. It seems that the change in hydrodynamic volume of the protein even near pH 11.9 is too small (see Fig. 5) to have any measurable effect o n p g i n t and it is certainly insufficient to cause exposure of the two buried tyrosine residues. The electrostatic interaction factor of ovomucoid was identical (w ---- 0.04), within experimental error, both in the carboxyl and phenolic regions and was not very different from the value calculated from Eqn 4.

ACKNOWLEDGEMENTS

Thanks are due to Council of Scientific and Industrial Research for the award

156

of a fellowship to one of us (A.W.). Part of the material in this paper was submit ted by A.W. in the thesis for the Ph.D. degree of Biochemistry at A.M. University, Aligarh, April 1974.

REFERENCES

1 Melamed, M. D. (1966) in Glycoproteins (Gottschalk, A., ed.), Vol. V, pp. 317-334, Elsevier, Amsterdam

2 Davis, J. G., Mapes, C. J. and Donovan, J. W. (1971) Biochemistry 10, 39-42 3 Waheed, A. (1974) Ph.D. Thesis, Aligarh Muslim University, Aligarh, India 4 Foulk, C. W. and Hollingsworth, M. (1923) J. Am. Chem. Soc. 45, 1220-1228 5 Powell, J. E. and Miller, M. A. (1957) J. Chem. Ed. 34, 330 6 Salahuddin, A. (1962) Ph.D. Thesis, Aligarh Muslim University, Aligarh, India 7 Dintzis, H. M. 0952) Ph.D. Thesis, Harvard University, Cambridge, U.S.A. 8 Tanford, C. 0955) in Electrochemistry in Biology and Medicine (Shedlovsky, T., ed.)pp. 248-

265, John Wiley, New York 9 Ahmad, F. and Salahuddin, A. (1974) Biochemistry 13, 245-249

l0 Wyman, Jr, J. (1939) J. Biol. Chem. 127, 1-13 11 Crammer, J. L. and Neuberger, A. (1943) Biochem. J. 37, 302-310 12 Tanford, C. (1962) Adv. Protein Chem. 17, 69-165 13 Tanford, C., Hauestein, J. D. and Rands, D. G. 0956) J. Am. Chem. Soc. 77, 6409-6413 14 Donovan, J. W. (1967) Biochem. Biophys. Res. Commun. 29, 734-740 15 Beeley, J. G. (1971) Biochem. J. 123, 399-405 16 Penasse, L., Jutisz, M. Fromageot, C. and Fraenkel-Conrat, H. (1952)Biochim. Biophys. Acta

9, 551-556 17 Monsigny, M., Adam-Chosson, A. and Montreuil, J. (1968) Bull. Soc. Chim. Biol. 50, 857-886 18 Osuga, D. T. and Feeney, R. E. (1968) Arch. Biochem. Biophys. 124, 560-574 19 Linderstrom-Lang, K. (1924) C.R. Trav. Lab. Carlsberg 15, No. 7 20 Lewis, J. C., Snell, N. S., Hirschmann, D. F. and Fraenkel-Conrat. H. (1950)J. Biol. Chem. 186,

23-35 21 Donovan, J. W. (1967) Biochemistry 6, 3918-3930