Affinity labeling of the exchangeable nucleotide binding site in platelet tubulin

Vol. 123, No. 1, 1984

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages

August30, 1984

92-99

AFFINITY LABELING OF THE EXCHANGEABLE NUCLEOTIDE BINDING SITE IN PLATELET TUBULIN Manfred

Steiner

From the Division of Hematology/Oncology The Memorial Hospital, Pawtucket, and Brown University, Providence, R.I. Received

July

13,

1984

SUMMARY: The exchangeable nucleotide binding site of platelet tubulin was labeled with [ 14C]p-fluorosulfonyl benzoylguanosine (FSBG). FSBG promoted polymerization of tubulin but depolymerization did not occur in the presence of this nucieoside analogue. GTP was able to block FSBG [14C]I o d oacetamide-treated binding to tubulin. tubulin which -4as first reacted with FSBG was digested with trypsin. The resultant peptides were analyzed by reverse phase high pressure liquid chromatography. One FSBG-labeled peptide could be identified both by its radioactivity and the characteristic UV absorbance spectrum associated with it. This may represent the exchangeable nucleotide site. A second peptide with a distinct nucleotide absorbance peak was found both in FSBG-treated and untreated tubulin preparations. This evidence is suggestive of the nonexchangeable nucleotide binding site.

The role

of nucleotides

and the stabilization terest.

While

molecules there

there

(1).

important this

nucleotides about

sites

to their

binding

available

(2,3)

irreversibly

sites

guanosine

Copyright Ail rights

with

their

function

analogues

on tubulin.

through (FSBG)

are

that

heterodimer,

number

that

utilized

at

of GTP at sites

would

the be an

Studies bind

while

others

of

covalently

of such compounds

are

are

Fluorosulfonyl the

benzoylguanosine;PIPES,

92

two

tubulin

preferable

i.nteraction.

$1.50

of the

in microtubules.

photoactivatable

is a compound

in-

of the nucleotide

binding

A growing

electrophilic

0 1984 by Academic Press, Inc. of reproduction in any form reserved.

the

protein

of great

function

ligation

nucleotide

FSBG, p-fluorosulfonyl Abbreviations: N,I\Il-bis (2-ethane sulfonic acid). 0006-291X/84

exact

exchangeability

of the

some of which

bound

on the

and the unexchangeable

nucleotide

of microtubule

to be a subject

associated the

in clarifying

require

continues

questions

Identification step

nature

benzoyl

remain

no uncertainty

one of the binding second

the polymerization

of tubulin

of guanine

is

in

latter

mechanism piperazine-

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Vol. 123, No. 1, 1984 to bind

to tubulin.

[ 14C] labeled

the nucleotide

binding

tide

supports

analogues

an absorption thus

site.

its

which

This

the site

separated

by high

second

nucleotide

treated

tubulin

ing

different

other

used

to study

guanine

In addition, from that

whose binding binding

to a specific

performance absorbance

as several

of tubulin.

quite

nucleotide

be localized

which

compound

and then

nucleo-

FSBG has

of peptides

and

and detection.

analogue,

exchangeable

could

is

measurement

I have used this identify

This

polymerization

spectrum

facilitates

FSBG was prepared

liquid

could site

on platelet

tryptic

peptide

chromatography.

peak was seen both

may represent

be inhibited

by GTP, to tubulin.

which

was

In addition,

a

in FSBG-treated

the nonexchangeable

nucleotide

and unbind-

site. MATERIALS

Synthesis

of p-Fluorosulfonyl

AND METHODS

[14C]Benzoyl-5-Guanosine

The method of Esch and Allison (4) was followed for the synthesis of [14C]p-fluorosulfony1 benzoyl chloride starting with carboxy-p-amino [ 14C]benzoic acid. The latter was obtained fron ICN; Irvine, CA (spec. act. 56 mCi/mmol) . The guanosine ester of p-fluorosulfonyl benzoic acid was prepared essentially according to the method described by Pal et al. (3) using guanosine instead of adenosine. All reaction products were checked by thin layer chromatography. FSBG had Rfs of 0.13 and 0.25 when the silica gel TLC plates were developed with isopropyl ether:acetone: H20 (65:20:15). Radiochromatograph scanning revealed high radioactivity After elution with methanol identical UV absorption specin both bands. tra were obtained. Maxima were found at 253 and 275 nm. For these studies I used the band with Rf 0.13 as it could be shown to be 5’-p-fluorosulfonyl benzoyl guanosine. The other band was a structural isomer. Preparation

of Platelet

Tubul in

Microtubule protein was isolated from human platelets according to a previously published method (5) using temperature-dependent polymerization and depolymerization. Two cycle microtubule protein was depolymerized at 4’ C and then freed of exchangeable nucleotide by the method of Maccioni It was then repolymerized in the presence of FSBG. The proet al. (6). tein was dialyzed exhaustively against 0.1 M piperazine-N, N’-bis (Z-ethane sulfonic acid) (PIPES) buffer, pH 6.9 containing 2 mM MgSD4 and 8 mM EDTA. Protein concentration was determined by the method of Lowry et al. (7) using bovine serum albumin as standard. Isolation

of Covalently

Modified

Peptide

Platelet microtubule protein isolated dependent polymerization-depolymerization phere of nitrogen against several changes HCl containing 3 mM EDTA, pH 8.5 and 0.12 concentrating tubulin to approximately 2-3 93

by 3 cycles of temperaturewas dialyzed under an atmosof 8 M urea in 0.35 M Tris/ M mercaptoethanol. After mg/ml by ultrafiltration,

Vol. 123, No. 1, 1984

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

the protein solution was treated with 6.25 PCi [14C]iodoacetamide per ml (spec. act. 23.6 Ki/mmol) and nonradioactive iodoacetamide whose concentration was adjusted to 2.7 mM. After 60 min at 25'C the reaction was stopped by dialysis against distilled water. This solution was lyophilized and then redissolved in distilled Hz0 (1 ml) at pH 8.8. To this solution was added ammonium bicarbonate at a final concentration of 25 mM and tosylphenylalanyl chloromethyl ketone treated trypsin (10 pg/ml microtubule protein. After proteolytic digestion for 4 hr at 37"C, the pH of the mixture was decreased to 2 and the mixture lyophilized. The residue was dissolved in 0.01 M ammonium acetate, pH 6.07, passed through a 0.2 v Nylon filter (Rainin Instrument Co., Woburn, MA) and an aliquot corresponding to 4-6 nmols of tubulin injected into a Bondapak Cl8 (Waters Associates, Inc. Milford, MA) high pressure liquid chromatograph column. The peptides were eluted by the following program: 1) 0.01 M ammonium acetate, pH 6.07 (buffer A); 2) 40% acetonitrile in 0.01 M ammonium acetate buffer, pH 6.07 (buffer B). Starting with 100% buffer A, the program proceeds to 50% buffer B in 50 min, then to 100% buffer B in another 80 min. The optical absorbance was measured at 220 nm and individual peaks were scanned from 210 - 320 nm. Major peaks were collected directly into scintillation vials to which Aquasol (New England Nuclear, Boston, MA) was added and then counted. Contamination by microtubule associated proteins (MAPS) was determined by sodium docecyl sulfate (SDS) polyacrylamide gel electrophoresis. Quantitative evaluation of the amount of contaminants by densitometric scanning of the Coomassie blue stained gels showed that 3 cycle tubulin contained < 15% MAPS. The modified peptide was identified by-scanning the optical absor210-320 nm as the peak eluted from bance in the wavelength range from the HPLC column. Measurement

of Tubulin

Assembly

Microtubule protein dissolved in PIPES buffer, pH 6.9 containing 2 mM MgS04 and 8 mM EGTA at a concentration of > 1 mg/ml was polymerized at 37'C either with 1 mM GTP or 1 mM FSBG in thz medium. Reassembly was monitored by measuring the change in optical absorbance at 580 nm which was recorded as a function of time. Disassembly of microtubules was evaluated by measuring the reversibility of the optical changes on cooling the solutions to 4°C. Polyacrylamide

Gel Electrophoresis

To separate e- and B-tubulin SDS-urea polyacrylamide gel phoresis was utilized essentially as described by Eipper (8). disc gel electrophoresis was performed on 7.5% polyacrylamide

electroRegular gels (5).

RESULTS [14C]FSBG

was able

site

of attachment

primary gel

electrophoresis

against

l-l.5

for

B -subunit.

the

The e-tubulin ratio

of 0.14.

to bind

(Fig.

to tubulin.

as demonstrated 1).

.4fter

mM FSBG the molar At this

band

isolated

by gel

by SDS-urea

exhaustive

ratio

ratio

The B-subunit

of the tubulin

appeared

electrophoresis

GTP in concentrations

ranging 94

polyacrylamide

dialysis analogue

of tubulin was O-62-0.68

to be saturated. showed

from

was the

a molar

0.2 to 1.0 mM

Vol. 123, No. 1, 1984

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

0

4

8

12 Fraction

Figure 1. SDS-urea polyacrylamide microtubule protein treated with gel was scanned at 580 om (solid [14C]activity (dots).

effectively

this

inhibited

concentration Assembly

(Fig.

2).

range

FSBG by tubulin.

equimolar

of FSBG was reduced

of FSBG-treated

polymerized

deviation

was unable

of

incorporation

FSBG-tubulin

from

28

24

gel electrophoresis of platelets [l"C]FSBG. The Coomassie blue stained line), dried, sliced and counted for

binding

and disassembly

same maximal tubulin

the

16 20 Number

without

the baseline

to depolymerize

tubulin delay

when exposed

by 74-81%.

was tested

and

as normal

In

produced

GTP-tubulin.

the

FSBG-

to 4 “C.

E g

0.15

____/--

d 5 p

-*

,.--

5:

*.

/-/’

‘.

-.s_ ---em_

/ /

O.lC

/

1’ /

ii 9

/ 4’

2 0.05 ‘p 0

/

/

/ / / /‘\/ C

2

4

6 Time,

.,

16

20

min

Figure 2. Tubulin assembly measured by turbidimetric assay. Platelet tubulin was isolated by 2 cycles of temperature-dependent polymerization and depolymerization and resuspended in PIPES buffer containing 1 mM FSBG. After exposure to the nucleoside derivative for 60 min at 1O'C the temperature of the tubulin solution was raised to 37OC and the optical absorbance changes at 580 nm recorded as a function of time (interrupted line). Controls (solid line) were exposed to dimethylsulfoxide (the solvent for FSBG) equal in volume to the FSBG solution added and were polymerized in the presence of GTP at 37'C. After complete the temperature was lowered to 4'C and the resultant changes assembly, in optical absorbance recorded.

95

BIOCHEMICAL

Vol. 123, No. 1, 1984

I

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

’ 220

I 300

260 Wavelength,

Figure line).

3.

Absorbance

The absorbance found

spectrum

Maxima

were

maxima

at 245 and 274.

The tryptic pressure protein major

liquid showed peaks

spectra

of FSB (solid

(Fig.

number

of radioactivity.

Time

Figure 4. Soluble tryptic platelet tubulin isolated

peptides

by reverse

were

3.

had absorbance

analyzed

by high

[14C]-iodoacetamide-alkylated

of radioactive Comparing

in Figure

comparison

protein 4).

and FSBG (interrupted

iis shown

FSB for

of microtubule

chromatography a large

line)

of FSBG-tubulin

at 254 and 274 nm.

peptides

nm

peptides. their

There

were

radioactivity,

it

16 can be

, min.

of

3 cycle

phase HPLC.

[14C]carboxamidomethylated

Optical

absorbance

was

measured at 220 nm (solid line). Each peptide peak was scanned from 220320 nm as they eluted from the column and was collected into a liquid scintillation vial for measurement of radioactivity (dotted curve). Details of the procedure are given under METHODS. The arrow indicates the peptide modified by FSBG. The inset shows the absorbance scan of the FSBG-peptide.

%

BIOCHEMICAL AND BIOPHYSKAL

Vol. 123, No. 1, 1984 recognized the

that

the

radioactive

peptide

counts

In addition,

there

double

of the

those

correlates the

the

were

4 other

known

amino

acid

sequence

Most

of the

in the

another

peak

absorbance

200-220

peaks

When[14C]FSBG

which

This

distribution

nm range,

spectra

a trough

nm corresponding

had additional

maxima

radioactivity

eluted

at 58.3

ahead

each

separation

recognizable tion

of this of peptides.

at 12.4

was pretreated

nucleotide

peptide.

area

min irrespective with

of peptides

with

240-250

from

from a steep

nm range,

to the

aromatic

amino

in the

252-255

nm area.

very

and

acids.

to be associated an absorbance

early

(within

nucleotides

nucleotide

with

peak the

were

first

found

with

peak was always

of whether

FSBG or not.

of radioactivity

eluted

in the

Unbound

A second

approximately

as they

min and showed

5 min),

peptides.

(9,lO).

was found

Free FSBG, GTP or GDP eluted

3 times

can be predicted

of the peaks

at 252 nm. far

that

of CY- and B tubuion

the typical

was used,

peptide

had counts

peptides

scans

revealed

at 270-280

Two distinct

tryptic

labeled

that

type.

the

min had approximately

of the other

peaks

prevailing

with

descent

at 41.9

of the majority

well

HPLC column

the

eluting

RESEARCH COMMUNICATIONS

the tubulin

The absorbance

prepara-

maximum in the

was at 254 min.

DISCUSSION Fluorosulfonyl affinity

label

(3,11,12).

benzoy for

a series

In the

binding

present

of FSBG to the

was obtained.

Both

nucleotide of different study

exchangeable

the

of tubulin

amide

gel

electrophoresis

provide

It

not

surprising

of different

to do this

(13).

nucleotide

strong

FSBG can promote

nonhydrolyzable FSBG provided

analogues a convenient 97

for

binding

support

as an effective binding

evidence

as demonstrated

a-subunit

number

convincing

by GTP and the

the

that

nucleotide

inhibition

FSBG with

is

has been used

enzymes preferential

site

of tubulin

association

of [ 14C]

by SDS-urea for

microtubule

this

affinity

hypothesis.

assembly.

of GTP have label

polyacryl-

A

been found that

was

BIOCHEMICAL

Vol. 123, No. 1, 1984 easily

synthesized.

sorbance

peak

zable

Bound to tubulin,

in the

Maccioni

(14)

recently

reported

an azidobenzoyl

to both

displacement

FSBG had a characteristic

ab-

UV region.

and Seeds

GTP analogues,

GTP bound

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

2 different

and a periodate

a- and %-subunits

by GTP and support

that

of lamb brain

hydroly-

oxidatibn

product

tubulin.

of polymerization

of

Competitive

by the analogues

could

be shown. The alkylation the previously

described

estimated

a total

agreement

with

the

of free

SH groups

abundance

%-subunit

sequence

eluted

and was clearly

far

eluted.

This

result

distinct

from

that

The consistent peptide which

with were

that

this

tide

binding

eluting that

It

removed

from

the

at 58.3

of aromatic

and the

amino

of 12.4

were

amino

exchangeable

acids

yet that

binding

identified bind sites.

peptide

to bind

to the

in had

tryptic

maximum at 252 nm free

GTP or GDP peak

in tubulin which the

a different

no radioactive tubulin the

One tryptic

absorbance

as those

containing

peak when [“Cl-FSBG-treated

8 SH groups

was quite

peak at 254 nm in the

min both

was clearly

There

in good

acids.

FSBG as well

the peptide

are

We

showing

where

of an absorbance

time

This min.

position

(15).

findings

had an absorbance

was reproducible

with

site.

I have not the

min.

retention

represents

FSBG was found

at 58.3

presence

treated

( 9,lO).

reaffirmed

in tubulin

These

of tubulin

and 12 in the cc-monomer

that

thiols

residues.

3,4 had 2 and 9 had 1 SH group. peptide

[ 14C]iodoacetamide

of free

of 20 cysteinyl the published

with

exact

the nucleotide Such studies

preparations

were

not,

suggests

nonexchangeable nucleotide counts

nucleofrom

associated

with

peptides

and

was analyzed. location

of the

at the

exchangeable

are

in progress

and nonnow.

ACKNOWLEDGEMENTS This study was supported by research grant HL 19323 of the Nat ional Heart, Lung and Blood Institute. The expert assistance of Ms. Linda Case and Susan Berry is gratefully acknowledged. 98

that

Vol. 123, No. 1, 1984

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Weisenberg, R.C., Borisy, G.G. & Taylor, E.W. (1968) Biochemistry 9, 4466-4479. Geahlen, R.L. & Haley, B.E. (1979) J. Biol. Chem. 254, 11, 982-11, 987. Pal, K., Wechter, W.J. 6 Colman, R.F. (1975) J. Biol. Chem. 250, 8140-8147. Esch, F.S. & Allison, W.S. (1978) Anal. Biochem. 84, 642-645. Ikeda, Y.8 Steiner, M. (1976) J. Biol. Chem. 251, 6135-6141. Maccioni, R.B. & Seeds, N.W. (1977) Proc. Natl. Acad. Sci. USA 74, 462-466. Lowry, O.H., Rosebrough, N.Y.,Farr, A.L. & Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. Eipper, B.A. (1974) J. Biol. Chem. 249, 1407-1417. Ponstingl, H., Krauhs, E. fr Kempf, T. (1981) Proc. Acad. Sci. USA 78, 2757-2761. Krauhs, E., Little, M., Kempf, T., Hofer-Warbinek, R., Ade, W. 6 Ponstingl, H. (1981) Proc. Natl. Acad. Sci. USA 78, 4156-4160. Wyatt, J.L. & Colman, R.F. (1977) Biochemistry 16, 1333-1342. Zoller, M.J. & Taylor, S.S. (1979) J. Biol. Chem. 254, 8363-8368. Weisenberg, R.C. 8 Deery, W.J. (1976) Nature 263, 792-793. Maccioni, R.B. 8 Seeds, N.W. (1983) Biochemistry 22, 1572-1579. Ideda, Y. & Steiner, M. (1978) Biochemistry 17, 2454-2459.

99