Ternary hydroxide complexes in neutral solutions of A13+ and F−

Vol. 155, No. 3, 1988

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

September 30, 1988

Pages ll94-1200

TERNARY HYDROXIDE COMPLEXES IN NEUTRAL SOLUTIONS OF A]3+ AND F-

R. Bruce Martin Department of Chemistry University of Virginia Charlottesville, VA 22903 Received July 12, 1988

SUMMARY. Especially in G prbtein systems AIF4- has been claimed as an activating species serving as a tetrahedral phosphate analog. However, in aqueous solutions (H20)2AIF4- is hexacoordinate with two bound water molecules. In neutral solutions five different mixed OH- and F- complexes of Al 3+ comprise the main species under usual experimental conditions. Comparison of the mole fraction distribution curves with limited results on the activity as a function of ambient F- concentrations suggests an activating complex composed of Al 3+ with three F- and uncertain geometry. Even fewer activity data suggest a tetrahedral Be2+ complex with three F-. © 1988 A c a d e m i c

Press,

Inc.

Since the discovery that fluoride activation of adenylate cyclase requires trace amounts of Be2+ or Al 3+ (1), there has been a plethora of papers on the effects of Al 3+ and F- on other G-protein systems especially transducin (2-4).

In the last system T~ activation by a small amount of

Al 3+ peaks at about 3 mM F- (3).

Since in the Al 3+ + F- system the mole

fraction of the complex AIF4- also peaks in the same F- concentration region, AIF4- looks like the active species (1,3,4).

Many papers offer the

supposedly tetrahedral AIF4- as a surrogate for tetrahedral phosphate. Despite designation of tetrahedral AIF4- as the active species structures are drawn with only three fluorides bound to a tetrahedral Al 3+ in turn bound to the B phosphate of GDP in the nucleoside site of T~ (3,4). However, the apparent active species in solution should correspond to what is diagrammed in structures.

I t is inconsistent to claim that a supposed

tetrahedral AIF4- acts as a phosphate analog and draw bound complexes with only three fluorides.

This paper points out that AIF4- is hexacoordinate in

aqueous solutions with two bound water molecules and at pH 7.5 this complex does not peak at 3 mM F- as mixed hydroxo complexes intervene. RESULTS

The scheme in Figure 1 represents the equilibria involved in mixed fluoride and hydroxide complexes of Al 3+ or Be2+. 0006-291X/88 $1.50 Copyright © 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.

1194

Across the top are the

Vol. 155, No. 3, 1988

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

F, Ko~,

x~x.Fo~ ,~o\h\" K

a 'hl /'

~

Fz

Ko~

F3

Ko~

F4

~

F~

F6

'=K,& h!F2K,3 LI ~ K,4 LI ~11 hv1___~ ~ IIir - 3 - ~ - n ir4

haft"z__~zhzFa Ka3!3K~khJF1 Ka4I h4 ha

Figure 1.

z~

Scheme for formation of mixed complexes of OH- and F- with a metal ion. F- addition occurs horizontally and hydroxide addition vertically downward. F signifies fluoride and h hydroxide. Equilibrium constants on horizontal arrows designate F- stability constants and on vertical arrows represent acidity constants for complex deprotonation.

successive e q u i l i b r i a for substitution of water by F- at the aqueous metal ion with the designated equilibrium constants, Kon, where n = 1 to 4 (Be2+) or 6 (Al3+).

At the l e f t the hm, with m = 0 to 4, specify the number of OH-

bound to the metal ion.

Since these equilibrium constants are expressed,

not as addition of hydroxide but as loss of a proton with a PKam value, (m = 1 to 4), the arrows are drawn in an upward direction.

The equilibrium

constants for fluoride substitution of water, Kon, and the PKam values for deprotonation, both originating with the aqueous metal ion, constitute the inputs from the two sets of binary complexes, either OH- or F- and the metal ion.

Literature values (5-8) of the Kon and PKam constants appear above the

horizontal line in Table I. Ternary or mixed complexes containing both OH and F- as well as the metal ion are represented by hmFn in Figure i .

Thus hlF 3 represents the

aluminum complex (HO)AIF3-. Seven ternary complexes appear in Figure 1, and the system becomes completely defined i f we can evaluate the seven independent equilibrium constants on the horizontal arrows in the center of Figure 1.

Equilibrium Constants on the unlabeled vertical arrows are not

independent and may be calculated later from the properties of a cyclic system. Additional complexes beyond those depicted in Figure 1 occur in negligible amounts for both Be2+ and Al 3+. The purely s t a t i s t i c a l method for evaluating the ternary equilibrium constants is based on the known equilibrium constant values for the binary complexes.

Since OH- and F- bear identical charges and are similarly sized,

this approximation works well.

The relevant equations for evaluating the

seven unknown constants take on an especially concise form in terms of the commonly used ~ symbolism for products of equilibrium constants; e.g., B02 : 1195

Vol. 155, No. 3, 1988

Table I.

BIOCHEMICAL AND BIOPHYSICAL RESEARCHCOMMUNICATIONS

Stability Constant Logarithms for Mixed Hydroxide and Fluoride Complexes

K01 K02 K03 K04 K05 K06 PKal PKa2 PKa3 PKa4 Kll K12 K13 K14 K21 K22 K31 a.

Ko1-K02.

Ref. 5.

Be2+

A13+

5.0 a 3.8 2.8 1.4

6.4 b 5.2 3.9 2.8 1.1 O.4 5.5 d 5.6 5.7 6.4

5.8 c 8.4 9.4 13.7 3.4 2.7 0.7

6.1 4.6 3.3 < 1 5.5 3.9 4.6

2.8 O. 0 -0.5

b. Ref. 6.

c. Ref. 7.

The seven equations are as follows:

d. Refs. 5 and 8.

~11 = 2(~20~02 )1/2, / B21 =

3(~302~03) 1/3, ~12 = 3(~30~032)I/3, ~31 = 4(~403~04 )1/4, ~13 = 4(~40~043) 1/4, ~22 = 6(~40~04) 1/2, and ~14 = 5(~50B054)I/5"

The numerical

f a c t o r s before the parentheses arise because of the greater numbers of ways of making mixed complexes and correspond to the numbers in Pascal's triangle.

In other systems mixed complexes form to a greater extent than

predicted s t a t i s t i c a l l y

(9,10).

Thus t h i s analysis presents a minimum

standard f o r the formation of ternary complexes:

they are l i k e l y to be more

important than shown by the calculations in t h i s paper. Results of the analysis performed separately f o r Be2+ and A13+ appear at the bottom of Table I, which l i s t s e q u i l i b r i u m constants in Figure 1.

values f o r the seven t e r n a r y

Two of these s t a t i s t i c a l l y

derived

values in Table I may be checked with an experimental study f o r A13+.

The

study (11) performed at 25°C and 0.1 ionic strength reports log K l l = 6.1 (identical

to t h a t in Table I) and log K12 = 4.2 (0.4 log units l e s s ) .

Another analysis (12) y i e l d s r e s u l t s f o r A13+ closely s i m i l a r to those in Table I a f t e r adjustment f o r the very low ionic strength. 4.8, so t h a t the s t a t i s t i c a l l y

Thus log K12 =

derived value in Table I nears the average of

the two l i t e r a t u r e values. A c i d i t y constant pKa values for the unlabeled upward arrows in Figure 1 may be calculated from the constants given in Table I and the p r o p e r t i e s of a c y c l i c system.

Thus f o r A13+ the successive pKa values across the f i r s t

row of unlabeled arrows in Figure 1 are 5.5 ( l a b e l e d ) , 5.8, 6.4, 7.0, and 1196

Vol. 155, No. 3, 1988

>9.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

The l a s t pKa > 9 v a l u e r e f e r s to the r e a c t i o n (H20)2AIF 4- ~ H+ +

(H20)(HO)AIF42- and i n d i c a t e s t h a t the l a s t complex remains i n s i g n i f i c a n t

in

neutral solutions.

DISCUSSION Figures 2 and 3 show distribution curves for Be2+ and A13+ in the presence of varying amounts of ambient F- at pH 7.5.

Comparedto the

distribution without considering hydroxide (13), Figure 2 for Be2+ shows the almost total eclipse of BeF+ by BeOH+ and (HO)BeF. At F- concentrations greater than 1 mM, hydroxide has l i t t l e impact, and the traditional binary complexes dominate. As Figure 3 for Al 3+ illustrates, inclusion of hydroxide has a major impact on the species distribution at pH 7.5 and less than 30 mM F-.

In the

region of 0.3 to 6 mM ~- where eight different complexes appear, none attains 40% of the total Al 3+ present.

Traditional distribution curves

(8,13,14) drawn for Al 3+ and F- apply only at pH 4-5: at lower pH F- becomes HF, and at higher pH aqueous Al 3+ deprotonates and also forms ternary complexes with OH- and F-.

Without consideration of hydroxide the AIF4-

mole fraction exhibits a maximum at about 6 mM F- while in Figure 3 i t s maximum appears at 23 mM F-.

Instead, the maximum appearing at 4 mM F- in

Figure 3 belongs to (HO)AIF3-, but i t rises to only 0.34 mole fraction. Since the pH = 7.5 is constant in Figures 2 and 3, curves corresponding to complexes with the same number of fluorides have the same shape and f a l l on the same position on the pF axis; only their height varies. curve for AIF3

That the

rises 1/3 as high as that for (HO)AIF3- depends upon the

"kO

,

i

I

I

'

Be~++ F - pH 7.5 0.8 - ~

5 Figure 2.

BeF3-

2

~Be

(OH)+

4

~

3 pF

c,^ c 2 ~='4

/

2

Mole f r a c t i o n Be2+ basis as a function of - l o g [ F ] at pH 7.5. For the mixed complexes the number of bound hydroxides are indicated by the subscript on h; e.g., hF represents (HO)AIF+.

1197

Vol. 155, No. 3, 1988

BIOCHEMICAL AND BIOPHYSICAL RESEARCHCOMMUNICATIONS

1.0

i

I

l

I

I

AI3++-FpH7.5

0.8 ~

O H )

2

AIFZ

-~ 0.6 tL

h3F

o 0.4

:E

0.2iII(OH)3

~

~

AIF~

pF Figure 3. Mole fraction A13+ basis versus ambient F- as -log[F-] at pH 7.5. For ternary complexes the number of bound hydroxides is indicated by the subscript on h; e.g., hF3 represents (HO)AIF3-.

d i f f e r e n c e between pKa = 7.0 for the reaction AIF 3 ÷ H+ + (OH)AIF 3- and the p l o t t e d pH of 7.5.

At pH 7.0 the two curves would coincide.

comparing the a c t i v i t y

Therefore, in

of a system at constant pH against pF with curves

such as those in Figures 2 and 3, only the number of bound F- is indicated: the number of bound OH- is not s p e c i f i e d . The hydroxide induced s h i f t of the maximum for the AIF 4- complex to higher pF values requires a r e i n t e r p r e t a t i o n of the active species postulated in several systems with added A13+ and F-.

A c t i v a t i o n r e s u l t s in

the transducin system with A13+ and variable amounts of F- most closely trace the (HO)xAIF 3 curves in Figure 3 (2,3). complex should contain three f l u o r i d e ions. the number of OH- or H20 ligands.

Thus the protein bound This technique cannot specify

Less d e f i n i t i v e

r e s u l t s e x i s t f o r Be2+,

but most studies suggest a correspondence with the BeF3- curve in Figure 2. The r e l i a b i l i t y

of the analysis rests on the absence of other strong

complexing agents that compete with F- as a ligand. GTP4-, and other phosphates a l l

EDTA, c i t r a t e ,

ATP4-,

bind A13+ much more s t r o n g l y than F- does.

In the adenylate cyclase study, a c t i v a t i o n by Be2+ and A13+ but not Sc3+ (which binds F- as s t r o n g l y as A13+) in the presence of EDTA (1) has been a t t r i b u t e d to the much f a s t e r rate of Sc3+ sequestration by EDTA (8). e f f e c t s may also impact other systems.

Rate

The strong competition of phosphate

and ADP3- f o r A13+ in the presence of F- has received emphasis (15). However, t h i s paper does not consider ternary OH-, F- complexes and reports an A13+ s t a b i l i t y

constant order of HP042- > ADP3- > ATP4-, the reverse of

that f o r other metal ions (16).

The AIF 3 and AIF 4- species are described as 1198

Vol. 155, No. 3, 1988

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

octahedral and i t is assumed that this form binds to proteins (15). Mg2+ also binds F- (14). Tetrahedral AIF4- does not exist in aqueous solutions.

There the

presumed tetrahedral AIF4- is actually hexacoordinate (H20)2AIF4-. Aqueous r

Al 3+, AlF63- and all binary complexes of Al 3+ and F- are hexacoordinate.

In

contrast all binary and ternary complexes of Be2+ with H20, OH-, and Fligands are tetrahedral.

WhenOH- is bound to Al 3+, a more complex

situation occurs. The t i g h t l y spaced four pKa values for Al 3+ in Table I spanning only 6.4 - 5.5 = 0.9 log units strongly indicate a cooperative interaction and conversion from hexacoordinate aqueous Al 3+ to tetrahedral AI(OH)4- (14).

Contrast the more normal pKa spacing for Be2+ of 13.7 - 5.8

= 7.9 log units, where there is no decrease in coordination number. Also for F- binding compare the identical spacing of 3.6 log units for log K01 log K04 for both Be2+ and Al 3+ in Table I, again indicating no change in coordination number, 4 for Be2+ and 6 for Al 3+, as F- substitutes for H20 ligands.

At pH > 6.2 the dominant Al 3+ species in aqueous solutions is

tetrahedral AI(OH)4-, and this species, not aqueous AI 3+, should be the starting point for thinking about Al 3+ speciation in neutral solutions. Addition of sufficient F- in neutral solutions converts tetrahedral AI(OH)4through a range of mixed complexes and eventually to hexacoordinate AIF63-. The denticity of the intermediate ternary complexes containing both OH- and F- is unknown, but several of the complexes probably possess both 4 and 5 or 5 and 6 ligands about the Al 3+.

An additional variable is the geometry that

an intermediate aqueous species such as (H20)x(HO)AIF3- takes up on a protein surface and next to a GDP molecule. Metal ions frequently adopt lesser coordination numbers on sterically crowded proteins, and both the Be2+ and Al 3+ complexes with three F- may adopt a tetrahedral geometry with H20, OH-, or a phosphate oxygen as the fourth ligand.

ACKNOWLEDGHENT. This research was supported by a research grant from the NIEHS (ES04446).

REFERENCES

1. 2. 3. 4. 5.

Sternweiss, P.C. and Gilman, A.G. (1982) Proc. Natl. Acad. Sci. USA 79, 4888-4891. Cook, N.J., Nullans, G., and Virmaux, N. (1985) Biochem. Biophys. Res. Commun. 131, 146-151. Bigay, J., Deterre, P., Pfister, C., and Chabre, M. (1985) FEBS Lett. 191, 181-185. Bigay, J., Deterre, P., Pfister, C., and Chabre, M. (1987) EMBOJ. 6, 2907-2913. Mesmer, R.E. and Baes, Jr., C.F. (1969) Inorg. Chem. 8, 618-626. 1199

Vol. 155, No. 3, 1988

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Smith, R.M. and Martell, A.E. (1976) C ri t i cal S t a b i l i t y Constants, volume 4, Plenum, New York. Baes, J r . , C.F. and Mesmer, R.E. (1976) The Hydrolysis of Cations, John Wiley, New York. Martin, R.B. (1986) Clin. Chem. 32, 1797-1806. Sigel, H. (1975) Angew. Chem. Internat. Edit. 14, 394-402. Martin, R.B. and Prados, R. (1974) J. Inorg. Nucl. Chem. 36, 1665-1670. Couturier, Y. (1986) Bull. Soc. Chim. France, 171-177. Dyrssen, D. (1984) Vatten 40, 3-9. Goldstein, G. (1964) Anal. Chem. 36, 243-244. Martin, R.B. (1988) Metal Ions Biol. Syst. 25, 1-57. Jackson, G.E. (1988) Inorg. Chim. Acta 151, 273-276. Martin, R.B. (1979) Metal Ions Biol. Syst. 8, 57-123.

1200