Binding of colchicine to renal tubulin at 5°C

Vol. 116, No. 3,1983

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 866-872

November 15, 1983

BINDING OF COLCHICINE TO RENAL TUBULIN AT 5°C Larry D. Barnes a, Angela K. Robinson a, Robert F. Williams b, and Paul M. Horowitz a aThe Department of Biochemistry, The University of Texas Health Science Center at San Antonio. San Antonio, Texas 78284 bDivision of Earth and Physical Sciences, The University of Texas at San Antonio, San Antonio, Texas 78285 Received September 15, 1983 Previous reports in the literature state that the binding of colchicine to soluble tubulin is negligible at 0 - 40C as measured by non-equilibrium binding methods. In contrast, we have detected significant binding of colchicine to tubulin at 5°C. Furthermore, for the first time equilibrium dialysis has been used to measure colchicine binding. The value of the dissociation constant was 1.8 ~M at 5°C, and the stoichiometry of colchicine binding at 5°C equaled that at 37°C. Dissociation at 5°C of bound colchicine was negligible over a period of 23 h, and the estimated minimal half-time of dissociation was 150 h.

Colchicine is unique among ligands which interact with tubulin.

Its

binding has been reported to be negligible around 4°C, its rate of equilibration with tubulin is apparently slow at 37°C, and its binding has been reported to be nearly irreversible (1-7).

The binding of colchicine to tubulin from

grasshopper embryos (8), HeLa and KB cells (i), chick embryo brain (9), rat brain (i0), and goat b~ain (ii) at 0 - 4°C is reported to be less than nine per cent, and generally less than five percent, of the corresponding binding at 37°C.

Vinblastine-induced tubulin crystals from sea urchin eggs bind colchi-

cine at 4°C after incubation for 24 to 36 h, and the association constant for colchicine binding at 4°C is about four times less than the maximum association constant at 20°C (12).

In contrast, podophyllotoxin, which is an apparent

competitive inhibitor of colchicine binding (3,10,12-14) and which has a binding region on tubulin in cow,non with colchicine (15), readily binds to soluble tubulin at 4°C (3,15).

Similarly, desacetamido-colchicine and 2-methoxy-5-

(2',3',4'-trimethoxyphenyl) tropone, both apparent competitive inhibitors of colchicine binding, bind to tubulin at 4°C (ii). Previously, we purified tubulin from bovine renal medulla and determined its binding properties for colchicine at 37°C (14). Renal tubulin is similar

0006-291X/83 $1.50 Copyright © 1983 by Academ~ Press, Inc. AH righ~ o f reproduction m any form reserved.

866

Vol. 116, No. 3, 1983

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

to brain tubulin with respect to the apparently slow rate of colchicine action,

tight binding,

decay of binding activity,

activity by colchicine and vinblastine.

of binding

Since renal tubulin can be readily

purified by polymerization-depolymerization

in the absence of microtubule-

associated proteins and since its ligand-binding those of other soluble tubulins

and stabilization

inter-

properties

are similar to

(14), we chose it for this study.

Here we report

the finding that colchicine binds to renal tubulin at 5°C as measured by equilibrium dialysis. MATERIALS AND METHODS [ring A-4-3H] Colchicine (specific activity 5.7 - 8.1 Ci/mmol) was purchased from Amersham and unlabeled c o l c h i c i n e f r o m Aldrich Chemical Company. Purity of [4-3H]colchicine was at least 98% as determined by thin-layer chromatography on silica gel in chloroform:acetone:diethylamine (5:4:1) (17). Stock [3H]colchicine was stored at -20°C in ethanol and the solvent was evaporated under a stream of N 2 prior to dissolving for use in the assay. Concentrations of colchicine solutions were calculated from the absorbance at 350 nm using an extinction coefficient equal to 16,000 em-iM -I (18). Tubulin was purified from bovine renal medulla by four cycles of in vitro assembly and disassembly of microtubules (14). Renal medullary tubulin polymerized in the presence of dimethyl sulfoxide and glycerol does not contain microtubule-associated proteins. Renal medulla tubulin used for all of the binding experiments was at least 95% pure as determined by electrophoresis on polyacrylamide gels containing sodium dodecyl sulfate (14). Equilibrium dialysis measurements of colchicine binding were done in Lucite cells each composed of two 500 ~i cavities separated by a membrane of dialysis tubing. Dialysis tubing was prepared for equilibrium dialysis as previously described (19). [3H]Colchicine was added to i00 mM sodium phosphate containing 2 m~l MgS04 and 0.2 mM EGTA, pH 6.8, in a final volume of 400 ~I in one cavity. Colchicine was added such that the concentration ranged from 0.25 to i00 ~M at equilibrium if no tubulin was present. The other cavity contained the same buffer solution and 300-350 ~g of renal tubulin in a final volume of 400 ~i. In the presence of tubulin, the final concentrations of colchicine differed due to binding. In some experiments, tubulin and [ H]eolchicine were initially added to the same cavity. Dialysis cells were covered with foil and were rotated at 21 rpm in a cold box. The solution temperature, measured with a thermistor inside the dialysis cell, was 4.6°C to 5.6°C. Duplicate samples of 20 to i00 ~i were removed from each cavity for measurement of radioactivity. Samples were counted in a Beckman LS-230 or 7000 counter at 30-35% efficiency. The total radioactivity from the tubulin and ligand cavities after equilibration was 97% of the radioactivity initially added. The radioactivity in the cavity containing only colchicine was used to calculate the free colchicine concentration. The difference in radioactivity between the cavity containing both tubulin and colchicine and the cavity containing only colchicine was assumed to represent tubulin-bound colchicine. Binding data were graphically represented in two forms. In the first form, the mol of eolchicine bound per mol of tubulin were plotted as a function of the logarithm of the free colchieine concentration to estimate the approach to saturation (20). In the second form, data were presented as a Scatchard plot (21). The equilibrium constant and stoichometry were calculated from the slope and abscissa intercept, respectively, of the Scatchard plot. The concentration of tubulin was measured according to Lowry et al. (22) using bovine serum albumin as the standard protein. A mole-

867

Vol. 116, No. 3, 1983

BIOCHEMICAL A N D BIOPHYSICAL RESEARCH COMMUNICATIONS

cular w e i g h t of i00,000 for tubulin (23,24) was used to calculate mol The ionic strength of the buffer was considered sufficient to obviate for D o n n a n effects during e q u i l i b r i u m dialysis.

of tubulin. corrections

RESULTS [3H]Colchicine of tubulin attained tional

(Fig.

reached

IA)°

Equilibration

as a function

and eighty-fold

range

at e q u i l i b r i u m of b i n d i n g

(Fig.

sites

1.84 ± 0.37 ~M, bound per mol

i

in about

of

J

(Fig.

Binding

i

of colchicine

concentration

2B).

The value

These values

i

i

A two hundred was attained

w i t h a single

of the e q u i l i b r i u m was

0.50 ± 0.07 mol

are the means

i.

I

class

constant

was

colchicine

and standard

,

2.0

tubulin was

to tubulin was

of free colchicine

The data are compatible

with

for at least an addi-

concentration.

and the stoichiometry

i

constant

of the colchicine

2A).

tubulin.

i

IB).

4 h at 5°C in the absence

[3H]colchicine

in 6 to 8 h at 5°C and remained

i0 to 12 h (Fig.

measured

equilibrium

I

devia-

l

,

,

u

o

i

i

z~

l

:=L

o

o 5 0J

1.5

OJ

.8 1.0

-8 2 "6

o ~

0

~ ~--~

'-E'

~

o

u

a

o

0.5

A I

0

2

I

4

I

B

I

I

0

6

4

8

I

12

I

I

16

Time (hr)

Time (hr)

FIGURE i. Time courses of equilibration of [3H]colchicine in equilibrium dialysis cells in the presence and absence of tubulin at 5°C. A. Equilibration in the absence of tubulin. [3H]Colchicine was initially present in one cavity at a concentration of 3,86 nmol/20 ~i (8.8 x 104 dpm/20 ~i) in i00 mM sodium phosphate, 2 mM MgSO4, and 0.2 mM EGTA, pH 6.8. The opposite cavity contained buffer. Duplicate 20~I samples were removed from each cavity as a function of time. 0, [3H]colchicine in the cavity to which [3H]colchicine was initially added; 0, [3H]colchicine in the cavity initially containing only buffer; A, total [3H]colchicine in both cavities. B. Equilibration in the presence of tubulin. [3H]Colchicine was initially present in one cavity at a concentration of 2.07 nmol/20 ~i in buffer and the opposite cavity contained 0.33 mg of renal tubulin in 400 ~i of buffer. Duplicate 20 ~i samples were removed from each cavity as a function of time. 0, [3H]colchicine in the cavity to which [3H]colchicine was initially added; 0, [3H]colchicine in the cavity containing tubulin; &, total [3H]colchicine in both cavities. About 3% of the total [3H]colchicine was absorbed to the membranes in the absence and presence of tubulin.

868

VOI. 116, No. 3, 1983

BIOCHEMICAL A N D BIOPHYSICAL RESEARCH COMMUNICATIONS

i

~0.5

~

0.3

0.4 O.2

0.3

m

.~ 0.2 .__

o 0.1 E

0

0.I 0

A

E

'

O]

1.0

tO

0

I

0

I

0.1

I

I

I

0.2

I

I

I

0.3

I

I

0.4

0.5

mol ColchicineBound/mol Tubulin

Colchicine Free (gM)

FIGURE 2. Equilibrium binding of colchicine to renal tubulin at 5°C. One cavity of the dialysis cell contained 0.32 mg of renal tubulin and [3H]colchicine in 400 ~i of i00 mM sodium phosphate, 2 mM MgS04, and 0.2 mM EGTA, pH 6.8. The initial concentration of [JH]colchicine ranged from 0.41 to 51.5 ~M. The opposite cavity contained 400 pl of buffer. Equilibrium dialysis was done for 18 h at 5°C. A. Mol of colchicine bound per mol of tubulin as a function of the logarithm of the free colchicine concentration at equilibrium. B. Scatchard plot of the data shown in panel A.

tions for ten experiments with four different preparations of renal tubulin.

Values for these binding parameters were not significantly

different whether

[3H]colchicine initially was loaded in the same

cavity as tubulin (five experiments) experiments).

or in the opposite cavity (five

For comparison, binding of colchicine to renal tubulin

was also measured at 37°C using the filter assay on DEAE-paper as previously described

(14).

The tubulin concentration was 0.i mg/ml and the

duration of binding was 3 h.

The values of the dissociation constant and

stoichiometry at 37°C were 0.42 ± 0.04 ~M and 0.46 ± 0.04 mol colchicine bound per mol tubulin, respectively

(mean ± S.D.; n=4).

Reversibility of colchicine binding to tubulin at 5°C was tested by dilution and by displacement of bound colchicine.

[3H]Colchicine was equili-

brated with tubulin for 18 h to final concentration of 5 ~M

for free colchi-

cine such that 0.47 mol of colchicine was bound per mol of tubulin. chicine in the ligand cavity was replaced with buffer.

[3H]Col-

After 23 h of additional

equilibrium dialysis, 0.49 mol of colchicine was bound per mol of tubulin. experiment was repeated in the same manner except

[3H]colchicine in the ligand

cavity was replaced with 50 ~M unlabeled colchicine after 18 h. 869

The

Before replace-

Vol. 116, No. 3, 1983

ment of the

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

[3H]colchicine,

0.49 mol of colchicine was bound per mol of tubulin.

The same stoichiometry was obtained after replacement

of [3H]colchicine.

Thus,

colchicine bound to renal tubulin at 5°C for 18 h did not detectably dissociate at 5°C after an additional

23 h. DISCUSSION

These results demonstrate

for the first time that colchicine significantly

binds to soluble tubulin at 5°C.

The decay of colchicine binding capacity of

tubulin at 37°C (16) coupled with the time required for equilibrium dialysis prevent this method from being used to measure binding at 37°C. ditions and non-equilibrium methods used in previous detect colchicine binding to tubulin at 5°C.

The assay con-

studies are inadequate to

In preliminary

studies we have

detected colchicine binding at 5°C using the filter assay on DEAE-paper provided we increase the tubulin concentration eight-fold and lengthen the incubation time six-fold in comparison to conditions normally used to measure binding at 37°C (14).

Since tubulin is a highly conserved protein,

the similarities between

the other binding properties of renal tubulin and tubulins isolated from other mammalian tissues indicate that binding of colchicine at 5°C is a common property. The method of equilibrium dialysis theoretically permits equilibrium analysis of colchicine-tubulin accordingly.

However,

5°C is difficult

interaction and the data have been analyzed

actual equilibrium binding of colchicine to tubulin at

to demonstrate experimentally.

No significant

of bound colchicine was detectable within 23 h at 5°C. previously

demonstrated

Wilson and co-workers

that at 37°C colchicine binds almost irreversibly

tubulin from grasshopper experimentally

dissociation

embryos and brain of chick embryos

(2,8,9).

to

Garland

measured the dissociation rate constant of the colchicine-tubulin

(porcine brain) complex at 370C (7) from which a half-time of 36 h can be calculated.

Extrapolating

from our experimental data, we estimate that the minimum

half-time

for dissociation of bound colchicine at 50C is 150 h assuming first-

order dissociation and detectability

of i0 percent dissociation.

Our results are compatible with the model proposed by Garland extend the temperature

limit to 5°C.

Garland proposed,

870

(7) and

based on kinetics of

Vol. 116, No. 3, 1 9 8 3

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

colchicine binding, that the binding of colchicine induces a conformational change in tubulin.

A slow rate of conformational change could account for the

overall slow rate of equilibration of maximally-bound colchicine to tubulin and for the negligible dissociation of colchicine bound at 37°C. Binding of desacetamido-colchicine and 2-methoxy-5(2',3',4'-trimethoxyphenyl) tropone to brain tubulin at 4°C (ii) demonstrates that the capacity of tubulin to bind colchicine at 4 ° should be an inherent structural property of the tubulin. Furthermore, Andreu and Timasheff (25) demonstrated that tropolone methyl ether binds to brain tubulin more tightly and more slowly at 0°C than at 37°C.

The

binding of N-acetylmescaline, an analog which mimics the trimethoxyphenyl moiety of colchicine, at 0°C was estimated to be much weaker than the binding of

tro-

polone methyl ether (25). Their results suggest that colchicine should bind at 5°C.

Our data demonstrate that colchicine does bind at 5°C, and this binding

probably reflects an initial binding of the tropolone moiety. These results show for the first time that colchicine binds to soluble tubulin without a temperature discontinuity. ACKNOWLEDGEMENTS We thank Dr. Neal C. Robinson for use of the equilibrium dialysis cells. This research has been supported by Grants AG-774, AQ-723, and AX-769 from the Robert A. Welch Foundation to L.D.B., P.M.H., and R.F.W., respectively. REFERENCES i. 2. 3. 4. 5. 6. 7. 8. 9. i0. 11. 12. 13. 14. 15.

Borisy, G.G. and Taylor, E.W. (1967) J. Cell Biol. 34, 525-533. Wilson, L., Bamburg, J.R°, Mizel, S.B., Grisham, L.M. and Cresswell, K.M., (1974) Fed. Proceed. 33, 158-166. Wilson, L. (1975) Ann. N.Y. Acad. Sci. 253, 213-231. Sherline, P., Leung, J.T., and Kipnis, D.M. (1975) J. Biol. Chem. 250, 5481-5486. Garland, D. and Teller, D.C. (1975) Ann. N.Y. Acad. Sci. 253, 232-238. McClure, W.O. and Paulson, J.C. (1977) Molec. Pharmacol. 13, 560-575. Garland, D.L. (1978) Biochemistr7 17, 4266-4272. Wilson, L. and Friedkin, M. (1967) Biochemistry 6, 3126-3135. Wilson, L. (1970) Biochemistry 9, 4999-5007. Zweig, M.H. and Chignell, C.F. (1973) Biochem. Pharmacol. 22, 2141-2150. Ray, K., Bhattacharyya, B., and Biswas, B.B. (1981) J. Biol. Chem. 256, 6241-6244. Bryan, J. (1972) Biochemistry ii, 2611-2616. Bhattacharyya, B. and Wolff, J. (1974) Proc. Natl. Acad. Sci. USA 71, 2627-2631. Barnes, L.D. and Roberson, G.M. (1979) Arch. Biochem. Biophys. 196, 511524. Cortese, F., Bhattacharyya, B. and Wolff, J. (1977) J. Biol. Chem. 252, 1134-1140.

871

Vol. 116, No. 3, 1983

16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Ludue~a, R.F. (1979) in Microtubules (Roberts, K. and Hyams, J.S., Eds.) pp 65-116, Academic Press, NY. Ertel, N.H. and Wallace, S.O. (1970) Biochem. Med. 4, 181-192. Horowitz, R.M. and Ullyot, G.E. (1952) J. Amer. Chem. Soc. 74, 587-592. Kuehn, G.D., Barnes, L.D. and Atkinson, D.E. (1971) Biochemistry iO, 39453951. Thompson, D.J. and Klotz, I.M. (1971) Arch. Biochem. Biophys. 147, 178-185. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. Ponstingl, H., Krauhs, E., Little, M., and Kempf, T. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 2757-2761. Krauhs, E., Little, M., Kempf, T., Hofer-Warbinek, R., Abe, W., and Ponstingl, H. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 4156-4160. Andreu, J.M. and Timasheff, S.M. (1982) Biochemistry 2_~i, 534-543.

872