Sulfhydryl groups of rabbit liver arylsulfatase A

Sulfhydryl groups of rabbit liver arylsulfatase A

Biochimica et Biophysica Acta 831 (1985) 67-73 Elsevier 67 BBA 32288 S u l f h y d r y l groups o f rabbit liver arylsuifatase A A b d u l W a h e ...

503KB Sizes 3 Downloads 17 Views

Biochimica et Biophysica Acta 831 (1985) 67-73 Elsevier


BBA 32288

S u l f h y d r y l groups o f rabbit liver arylsuifatase A A b d u l W a h e e d a n d R o b e r t L. V a n E t t e n Department of Chemistry, Purdue University, West Lafayette, IN 47907 (U.S.A.)

(Received April 10th, 1985)

Key words: Aryl sulfatase; Chemical modification; Fluorescence quenching: Sulfhydryl group: (Rabbit liver)

Rabbit liver arylsulfatase A (aryl-sulfate sulfhydrolase, EC monomers of 130 kDa contain two free sulfhydryl groups as determined by spectrophotometric titration using 5,5'-dithiobis(2-nitrobenzoate) and by labeling with the fluorescent probe 5-(iodoacetamidoethyl)aminonaphthalene-l-sulfonic acid. Fluorescence quenching data indicate that the reactive sulfhyd~l is present in proximily to one or more tryptophan residues. Chemical modification of the sulfhydryl groups does not alter the distinctive pH-dependent aggregation property of the arylsulfatase A. The free sulfhydryls of the enzyme react with numerous sulfhydryi reagents. Based on the reactions of iodoacetic acid, methyl methanethiosulfonate, 5,5'-dithiobis(2nitrobenzoate) and 5-(iodoacetamidoethyl)aminonaphthalene-l-sulfonic acid with the sulfhydffl groups of arylsulfatase A, it is concluded that free sulfhydryls are not essential for the enzyme activity.. In contrast, the observed inactivation of the enzyme by p-hydroxymercuribenzoate or p-hydroxymercuriphenylsulfonate is probably due to a modification of a histidine residue, consistent with previous reports that histidine is near the active site of arylsulfatase A. p-Hydroxymercuribenzoate and p-hydroxymecuriphenylsulfonate are able to react both with cysteine and with histidine residues of the protein molecule.

Introduction Arylsulfatase A (aryl-sulfate sulfhydrolase, EC from rabbit liver is a 130 kDa glycoprotein monomer at pH 7.5, and polymerizes to a 257 kDa dimer at pH 4.5 [1]. Early studies on crude preparations of the rabbit liver enzyme indicated that sulfhydryl-modifying chemicals inactivated the enzyme molecule [2]. Similar results have been reported for several human enzymes [3-5]. However, Jerfy and Roy [6] found that the sulfhydryl reagent, p-chloromercuribenzoate did not inactivate homogeneous ox-liver arylsulfatase A at pH 7.5 in Abbreviations: pHMB, p-hydroxymercuribenzoate; pMPS, p-hydroxymercuriphenylsulfonate; DTNB, 5,5'-dithiobis(2nitrobenzoate); IAENS, 5-(iodoacetamidoethyl)aminonaphthalene-l-sulfonic acid; MMTS, methylmethanethiosulfonate; AENS, 5-(acetamidoethyl)aminonaphthalene-l-sulfonic acid.

10 mM diethylbarbiturate buffer, although the enzyme was inactivated at pH 5.0 in 10 mM acetate buffer. Apparently, conflicting results were also reported for studies of the effects of p-chloromercuribenzoate on the human urine enzyme. Stumpf and Austin [7] observed no effect of pchloromercuribenzoate on the activity of human urine arylsulfatase A, whereas Neuwelt et al. [4] observed enzyme inactivation. On the basis of amino acid composition of rabbit liver arylsulfatase A, there are 20 half-cystine residues per tool of 130 kDa protein monomer [8]. In view of the potential structural complexity and the conflicting results on enzyme inactivation, it became important to establish how many free sulfhydryl groups are present in the arylsulfatase A monomer and whether they are really essential for the enzyme activity. We describe experiments which establish the number of free sulfhydryl groups and

0167-4838/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)


we demonstrate that the free sulfhydryl groups are not essential for the enzyme activity of homogeneous rabbit liver enzyme. In addition, we find that the modification of the sulfhydryl groups with various sulfhydryl reagents does not alter the unique association-dissociation property of the arylsulfatase A.

spectra were measured using an IBM 9420 UVvisible spectrophotometer. The number of bound AENS molecules per tool of enzyme monomer was determined using the molar extinction of 6.0.10 ~ M ~ for the N-acetylcysteine-AENS conjugate at 340 nm [12].

lnactiuation studies Experimental procedures Homogeneous rabbit liver arylsulfatase A was prepared and the enzyme activity was determined as previously described [9]. A molecular weight of 130 kDa was used to calculate the molar concentration of the arylsulfatase A protein monomer. pHMB, MMTS, DTNB, IAENS and L-cysteine were purchased from Sigma Chemical Co., St. Louis. p H M B was purified according to a published procedure [10]. Other reagents were of analytical grade.

Determination of free sulfhydryl groups Spectrophotometric titration of arylsulfatase A with 5,5'-dithiobis(2-nitrobenzoate) was carried out using 1 ml of a 3.1 p,M enzyme solution in 0.1 M Tris-HC1 (pH 7.5). After each addition of 1/~1 of 1 mM DTNB in 0.1 M Tris-HCl (pH 7.5), the reaction mixture was allowed to stand at room temperature for 10 min and then the absorbance at 412 nm was recorded against a blank, having no enzyme. The titration was continued until there was no further absorbance change at 412 nm. The number of free sulfhydryls was determined from the amount of D T N B required to achieve the breakpoint in the titration curve, or by using the absorbance at the breakpoint together with the molar extinction of the sulfhydryl-DTNB derivative o f 1 3 6 0 0 M 1 . c m I [11].

Labeling of rabbit lieer arylsulfatase A with the fluorescent probe IA ENS The enzyme sample, 1 ml in 50 mM Hepes (pH 7.5) containing 3.1 btM enzyme protein, was mixed with I mM IAENS at 37°C for 4 h in the dark. The reaction mixture was dialyzed extensively against 0.1 M Tris-HCl (pH 7.5) buffer. The stoichiometry of labeling of arylsulfatase A was determined by spectrophotometric measurements of the enzyme-AENS conjugate. The absorption

Enzyme inactivation was followed by incubation of a known amount of enzyme in 100 mM Tris-HCl or 100 mM Hepes (pH 7.5) or 100 mM sodium acetate (pH 5.0) buffer, with a 3000-fold molar excess of the sulfhydryl reagent at 37°C. At different time intervals, an aliquot of the reaction mixture was removed and assayed for residual arylsulfatase A activity. Inactivation of the enzyme by p H M B was followed at different concentrations of the reagent by incubating identical amounts of enzyme at 37°C for 30 rain. In order to measure the extent of protection from inactivation that was afforded by competitive inhibitors such as sulfate or phosphate, the enzyme was first incubated with the inhibitor ions for 10 15 min and then the sulfhydryl reagents were added.

Fluorescence measurements Fluorescence spectra were recorded using a Perkin-Elmer MPF-44A fluorescence spectrometer at room temperature. Excitation and emission slit widths were kept constant at 6 nm. Protein samples were exicted at 280 nm for intrinsic fluorescence emission measurement and at 352 nm to measure fluorescence emission spectra of bound AENS. Fluorescence polarization was measured at room temperature by a MPF-44A polarizer accessory following the manufacturer's instructions.

Gel filtration chromatography A Sephacryl S-300 column (1.5 x 87 cm) was equilbrated with 0.15 M Tris-HCI (pH 7.5) or 0.1 M sodium acetate (pH 5.0)/0.1 M NaC1 buffer. Standard protein samples (ferritin, aldolase and bovine serum albumin; 440, 158 and 68 kDa, respectively) in 2 ml were applied to the column and eluted with buffer, collecting 3 ml fractions. Each fraction was monitored for protein at 280 nm and, when arylsulfatase A was applied on the column, fractions were also monitored for enzyme activity.

69 07' :


] A



The rabbit liver arylsulfatase A monomer of 130 kDa contains two free sulfhydryl groups which react readily with DTNB as well as with IAENS (a sulfhydryl-group-specific fluorescent reagent). A typical result of the titration of arylsulfatase A with DTNB in 0.1 M Tris-HC1 (pH 7.5) is shown in Fig. 1. From the break point of the titration curve, the mol of reactive sulfhydryl-groups per mol of 130 kDa aryisulfatase A was determined to be 2.0. When an enzyme preparation was first treated with excess iodoacetic acid, extensively dialyzed, and then titrated with DTNB, no change was observed in the absorbance at 412 nm, consistent with the conclusion that the reactive free sulfhydryl groups are first blocked by reaction with iodoacetic acid and no longer react with DTNB. The reactivity of these sulfhydryl groups was further explored by modification with IAENS. AENS-conjugated rabbit liver arylsulfatase A monomer was prepared (pH 7.5) by reaction of the enzyme with an excess of IAENS. The absorption and fluorescence spectra of the AENS-enzyme were recorded (Fig. 2). The absorption spectrum was characterized by a shoulder near 340 nm, a peak near 278 nm and some fine structure around 255 nm. The absorption at 340 nm was used to calculate mol free sulfhydryl groups per mol arylsulfa-

°~2I olo!







004 o











EDTNB]/ [ENZYME] Fig. 1. Spectrophotometrictitration of rabbit liver arylsulfatase A sulfhydrylgroups. The enzyme was reacted with 5,5'-dithiobis(2-nitrobenzoate) in 0.1 M Tris-HC1 (pH 7.5) buffer. After each addition of the reagent, the sample was allowed to stand for 10 rain at room temperature and its absorbance at 412 nm was recorded against a blank, having no enzyme.

~ 050

~_ o6 w


03 0 5 W

1 025

05 I

d W

,, \(

01 ~ 0

0 300




J ~ L ~ 250




WAVELENGTH,nrn Fig. 2. (A), Fluorescence emission spectra of native and AENS-modified arylsulfatase A. Fluorescence spectra of 0.3 /zM native ( ) and AENS-arylsulfatase A ( . . . . . . ) in 0.l M Tris-HC1 (pH 7.5) buffer and AENS-arylsulfatase A ( - - - - - - ) i n 0.l M sodium acetate (pH 4.5) were recorded with sample excitation at 280 nm. (B), Absorption spectrum of 3.1 ,aM AENS-arylsulfatase A in 0.t M Tris-HCI (pH 7.5).

tase A monomer, which was found to be 2.3. Within experimental error, this value is in agreement with that determined by DTNB titration. In order to study the effect of sulfhydryl group modification on the tertiary structure of arylsulfatase A, fluorescence emission spectra of native and AENS-conjugated arylsulfatase A were recorded during sample excitation at 280 nm. Typical emission spectra are shown in Fig. 2A. The native and AENS-conjugated enzyme showed identical emission maxima near 320 nm, a value that is a characteristic of tryptophan-containing proteins. The fluorescence quantum yield of tryptophan in the AENS-conjugated enzyme was quenched by 40%. When the pH of the solution was changed to 4.5, where rabbit liver arylsulfatase A exists a dimer [1], the fluorescence spectrum of the AENS-modified enzyme was not affected. However, there was a small amount of quenching of the fluorescence quantum yield tryptophan. The fluorescence emission spectra of the AENS-conjugated protein molecule was also recorded at pH 7.5 and 4.5 with sample excitation at 352 nm. The native enzyme did not show any fluorescence. The AENS-conjugated arylsulfatase A showed a fluorescence emission maximum near 480 nm at pH 7.5 and 4.5, similar results reported for other AENS-protein systems [13]. The spectra at pH 7.5 and 4.5 were similar, with only a slight quenching of the

70 fluorescence q u a n t u m yield of the A E N S probe at p H 4.5. These results suggest that during dimerization of the AENS-conjugated arylsulfatase A the electronic properties of the fluorescent probe did not change significantly. In such a case, one would expect similar results for fluorescence polarization of the enzyme-coupled fluorescent probe at both p H 7.5 and 4.5, since at these p H values the enzyme exists as a m o n o m e r and dimer, respectively. Therefore, we measured fluorescence polarization of A E N S coupled to arylsulfatase A at these pH values. The values of fluorescence polarization were found to be 0.27 and 0.33 at pH 7.5 and 4.5, respectively. This result further suggests that the electronic properties of the fluorescent probe are similar whether AENS-arylsulfatase A is in a monomeric or dimeric form. Using gel filtration chroamtography, we examined the elution behavior of native and of several modified forms of rabbit liver arylsulfatase A at p H 7.5 and 4.5. A typical chromatographic profile is shown in Fig. 3. The elution profiles of the native enzyme at p H 7.5 and 4.5 (as a mono-




mer and a dimer, respectively) were superimposable on the elution profiles of pHMB-, AENS-, I A A - and D T N B - m o d i f i e d arylsulfatase A at the same p H values. These results, together with the fluorescence studies, establish that the modification of the two free sulfhydryls of the enzyme m o n o m e r does not alter the characteristic pHdependent dimerization of rabbit liver arylsulfatase A. Several earlier studies [5,6] using the classical sulflaydryl reagent p H M B yielded apparently conflicting results regarding the role of sulfhydryl groups in arylsulfatase A activity. In the present studies the chemical modification of rabbit liver arylsulfatase A was carried out using several sulfhydryl reagents including iodoacetic acid, p H M B , p r i M P S , I A E N S , M M T S and D T N B . The enzyme (0.3 ffM) was incubated with a 3000-fold molar excess of the sulfhydryl reagent at p H 7.5 or pH 5.0 and the enzyme activity was determined at various times. The results are shown in Fig. 4. Iodoacetic acid, D T N B , M M T S and I A E N S at 1 m M concentration did not significantly inactivate the enzyme in 30 min. However, 1 mM p H M B or p r i M P S caused 70-75% inactivation of the enzyme in 30 min. Since rabbit liver arylsulfatase A that was modified by 1 m M p H M B or p r i M P S did show approx. 30% residual activity, we further






08[ O

li i'


04 F




O0 L 0


i/ ! \




< lo








Fig. 3. Sephacryl chromatographic profile of native and several modified arylsulfatase A preparations. The Sephacryl S-300 column (1.5×87 cm) was equilibrated with 0.15 M Tris-HC1 (pH 7.5) or with 0.1 M sodium acetate (pH 5.0)/0.1 M NaCI buffer at 4°C. Standard proteins ((1) ferritin, (2) aldolase and (3) bovine serum albumin) were applied on the column at both pH values. The elution profiles of arylsulfatase A monomer (M) and dimer (D) forms of native and iodoacetate-, pHMB-, DTNI~- and IAENS-conjugated derivatives of arylsulfatase A ( . . . . . . ) were obtained. Fractions were moinitored for protein concentration at 280 nm and for arylsulfatase A activity at 515 nm.


0 L- - -



80 !




i i i

~ ~o rr



20 ~ 0



12 16 20 24, 28 52 TIME (rain)

Fig. 4. Time-dependence of enzyme activity during the reaction of arylsulfatase A with different sulfhydryl reagents in 0.1 M Tris-HCl (pH 7.5) at 37°C. (A) Control (O), enzyme treated with iodoacetate (zx), IAENS (m) or pHMB (~). (B) Enzyme control (O), enzyme treated with methyl methanethiosulfonate (1), DTNB ([:3)or pMPS (IIQ-


examined the inactivation of the enzyme as a function of the starting concentration of pHMB. With a 300-fold molar excess concentration of p H M B , only a 7% loss of enzyme activity was observed, while at less than a 200-fold molar excess of p H M B the enzyme remained fully active. A maximal inactivation of only about 70% was observed, even when a 15000-fold molar excess of p H M B was employed. When arylsulfatase A was incubated at pH 5.0 with 1 mM p H M B in the presence of 10 mM sulfate or 1 mM phosphate ions, respectively, partial protection of the enzyme was observed. However, sulfate ions did not protect the enzyme from inactivation when incubated with p H M B at pH 7.5 (results not shown). The observed inactivation of arylsulfatase A activity by p H M B or priMPS, and absence of inactivation with serveral other well-known sulfhydryl reagents, iodoacetic acid, DTNB, IAENS and MMTS, suggested to us that there might be another amino acid residue whose modification with p H M B or p r i M P S led to inactivation of the enzyme. In order to check this, we first modified the enzyme with iodoacetic acid and then studied the effect of added pHMB. Fig. 5 shows that the rate of inactivation of the native enzyme by p H M B was virtually identical to the rate of inactivation of arylsulfatase A in which the free sulfhydryl groups were first modified by reaction with iodoacetic acid. This indicates that some amino acid group other than the reactive sulfhydryl is modified with



1 ~






TI ME (rain) Fig. 5, Inactivation of native and iodoacetate-modified arysulfatase A by p H M B and reactivation by cysteine. Enzyme samples in 0.1 M Tris-HCl (pH 7.5) at 37°C were incubated with 1 m M p H M B (O), or first with 1 m M iodoacetate for 30 min and then with 1 m M p H M B (zx). After 45 rain of incubation of the enzyme with pHMB, 20 m M cysteine was added and the reactivation of the enzyme followed (D). When arylsulfatase A was first exposed to 10 m M histidine and then 1 m M p H M B was added, no inactivation occurred ( ~ ) .

pHMB. The inactivation of the enzyme could be reversed when excess (20 mM) L-cysteine was added to the reaction mixture at 45 min. Reactivation of the inactive enzyme was rapid and at least 85% of the activity was recovered with in 2 min. Significantly, the enzyme was not inactivated when 10 mM L-histidine was present at zero time in the reaction mixture with pHMB. The nature of the buffer used in these types of experiment also proved to be important. We compared the extent of inactivation of rabbit liver arylsulfatase A by I mM p H M B at p H 7.5 when the reaction was carried out in 10 mM diethylbarbiturate versus that in 10 mM Tris buffer. In contrast to the strong inactivation we observed with p H M B in Tris, there was no significant inactivation when the barbital buffer was used. The protective effect of the barbital buffer provides an explanation for the fact that Jerfy and Roy [6] saw no inactivation at p H 7.5, since they used that buffer system. Discussion Rabbit liver arylsulfatase A has been shown to contain 20 half-cystines per monomer of 130 kDa protein, as determined from amino acid analysis [8]. The present studies indicate that only two cysteine residues are present as reactive, exposed thiol groups. The other half cystines may be present as buried sulfhydryl or as intrachain disulfide bridges [14]. Since rabbit liver arylsulfatase A monomer contains two identical subunits [15], one may reasonably conclude that each subunit has one free, exposed sulfhydryl group. Interestingly, the unique, pH-dependent aggregation behavior of arylsulfatase A was not altered upon modification of sulfhydryl groups with a variety of reagents, as was evident from the minor effects on the intrinsic fluorescence of the protein, the fluorescence quantum yield, the fluorescence polarization of the fluorescent probe or the gel chromatographic profiles. The fluorescence due to tryptophan residues in the protein molecule was quenched when free sulfhydryls were modified by the fluorescent probe, IAENS. From these results one may conclude that the free sulfhydryl groups are highly accessible to the solvent, they are not at the monomer-monomer contact region and they


are close to tryptophan residues in the arylsulfatase A monomer. We expected that modification of free sulfhydryl groups in arylsulfatase A would result in a loss in the enzyme activity [2-5]. Surprisingly, modification of the enzyme, with four well-known sulfhydryl reagents, iodoacetic acid [16], DTNB [17], MMTS [18] and IAENS [12], do not cause any loss of the enzyme activity. From these results we conclude that cysteine is not essential for arylsulfatase A activity. Despite this, the reagents pHMB or priMPS are moderately strong inhibitors of rabbit liver arylsulfatase A, consistent with results reporeted earlier for some other arylsulfatase A enzymes [2-5]. When we examined the concentration dependence of the modification of the enzyme by pHMB, very little inactivation was observed at a reagent concentration of 0.1 mM (which was a 300-fold molar excess over protein). Therefore, the ratio of the molar concentration of pHMB to protein is an important parameter that must be considered in order to compare meaningfully the inactivation studies conducted by the different research groups. In fact, pHMB and priMPS do inhibit arylsulfatase A at high concentrations of the reagent (1-5 mM). The enzyme is protected from inactivation by pHMB in the presence of the competitive inhibitors sulfate or phosphate ions, suggesting that the amino acid residues modified by pHMB are near the active site. Furthermore, pHMB modification is completely reversed by the addition of cysteine, suggesting that the reaction of pHMB with the affected amino acid residue(s) is less favorable than the subsequent reaction with cysteine. Our findings raised some questions about the tacitly assumed specificity of pHMB and priMPS towards sulfhydryl groups, in view of the fact that iodoacetic acid-modified arylsulfatase A, which no longer has a readily reactive free sulfhydryl group, could be subsequently inactivated by pHMB (Fig. 5). From this result, it appeared that pHMB reacts not only with free sulfhydryl groups but some other amino acid residue as well. Because histidine can complex various transition metal ions [19] and a histidine is at or near the active site of rabbit liver and ox liver arylsulfatase A [20,21], we performed a simple experiment in which histidine was added to the enzyme solution before addition of

the pHMB. Although histidine is not a competitive inhibitor of the enzyme, its presence in the reaction mixture with pHMB served to protect the enzyme. From these results we conclude that pHMB or priMPS reacts with histidine residues of the arylsulfatase A, consistent with a possibility advanced in a review [22]. Such a reaction is consistent with the fact that the logarithms of the stability constants of mercuric ion complexes with histidine [19,23] and with cysteine are 21 and 44, respectively. Thus, mercuric compounds may react with both cysteine and histidine, although the reaction with cysteine is generally much more favorable than that with histidine. Therefore, results obtained in the inactivation and derivatization of enzymes and proteins using high concentration of mercuric reagents such as pHMB and priMPS should be interpreted with caution. The nature of the buffer is also potentially important in such experiments. In conclusion, these studies provide the following important results: (1) the 130 kDa monomer of rabbit liver arylsulfatase A contains two free sulfhydryl groups; (ii) when a fluorescent probe is placed on the reactive sulfhydryls it causes quenching of tryptophan fluorescence; (iii) the free sulfhydryls are not present at the monomer-monomer contact region of the protein dimer; (iv) sulfhydryl groups are not essential for the enzyme activity; and (v) the partial inactivation of the enzyme by pHMB is probably due to a modification of the active-site histidine residue of the enzyme.

Acknowledgement This work was supported by Research Grant G M 22933 from the U.S. National Institute of General Medical Sciences.

References 1 Waheed, A. and Van Etten, R.L. (1979) Arch. Bicohem. Biophys. 194, 215-225 2 Meangwyn-Davies, G.D. and Friedenwald, J. (1954) Arch. Biochem. Biophys. 53, 29-43 3 Ba|asubramanian, A.S. and Bachhawat, B.K. (1963) J. Neurochern. 10, 201-211 4 Neuwelt, E., Stumpf, D., Austin, J. and Kohler, P. (1972) in Sphingolipids, Sphingolipidoses and Allied Disorders (Volk,


5 6 7 8 9 10 11 12 13 14

B.W. and Aronson, St.M., eds.), pp. 415-428 Plenum Press, New York Gniot-Szulzycka, J. (1974) Acta Biochim. Pol. 21, 247-254 Jerry, A. and Roy, A.B. (1969) Biochim, Biophys. Acta 175, 355 364 Stumpf. D. and Austin, J. (1971) Arch. Neurol. 24, 117-124 Waheed, A. and Van Etten, R.L. (1980) Biochim. Biophys. Acta 614, 92-101 Van Etten, R.L. and Waheed, A. (1980) Arch. Biochem. Biophys. 202, 366-373 Boyer, P.D. (1954) J. Am. Chem. Soc. 76, 4331-4337 Ellman, G.L. (1959) Arch. Biochem. Biophys. 82, 70-77 Hudson, E.N. and Weber, G. (1973) Biochemistry 12, 4154-4161 Jullien, M. and Garel, J.R. (1981) Biochemistry 20, 7021 7026 Laidler, P.M., Waheed, A. and Van Etten, R.L. (1985) Biochim. Biophys. Acta 827, 73-83

15 Lee, G.D. and Van Etten, R.L. (1975) Arch. Biochem. Biophys. 166, 280-294 16 Taga, E.M. and Van Etten, R.L. (1982) Arch. Biochem. Biophys. 214, 505-515 17 Laidler, P.M., Taga, E.M. and Van Etten, R.L. (1982) Arch. Biochem. Biophys. 216, 512-521 18 Smith, D.J. and Kenyon, G.L. (1974) J. Biol. Chem. 249, 3317-3318 19 Sundberg, R.J. and Martin, R.B. (1974) Chem. Rev. 74, 471-512 20 Lee, G.D. and Van Etten, R.L. (1975) Arch. Biochem. Biophys. 171,424-434 21 Jerry, A. and Roy, A.B. (1974) Biochim. Biophys. Acta 371, 76-88 22 Nicholls, R.G. and Roy, A.B. (1971) The Enzymes 5, 21-41 23 Ringbom, A. (1963) in Complexation in Analytical Chemistry pp. 332 Wiley Interscience, New York