Trifluoperazine binding to calmodulin: A shift reagent 43Ca NMR study

Vol. 122, No. 3, 1984

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

August 16, 1984

Pages 1350-l

TRIFLUOPERAZINE

BINDING

Hans

J.

Vogel, Torbjorn

Department

Received

July

5,

TO CALMODULIN:

of

A SHIFT

Thomas Andersson, Drakenberg and Physical Lund,

REAGENT

William H. Sture ForsPn

43CA

NMR

STUDY

Braunlin,

Chemistry 2, University S-220 07 Lund, Sweden

of

Lund

1984

43Ca NMR experiments of Ca2+ binding to calmodulin (CaM) were performed in the presence and absence of the calmodulin antagonist trifluoperazine (TFP). By making use of the shift reagent Dy(PPP)z(a I:2 complex of DyC13 and Na5P3010) we have succeeded in separating the "3Ca resonances of protein-bound Ca2+ and free Ca2+ in the otherwise unresolved spectra. This experimental strate$y has allowed us to demonstrate unequivocally that the affinity of CaM for Ca + is markedly increased in the presence of TFP. Thus Ca'+ is not liberated from the protein upon addition of TFP as had been suggested based on earlier 43Ca NMR experiments (Shimuzu, T., Hatano, M., Nagao, S. and Nozawa, Y. (1982), Biochem. Biophys. Res. Cotnm. 106, 1112-1118).

Calmodulin tein

in

all

(CaM)'

eukaryotic

that

the

activation

lin

takes

place

measurements

ternary

complexes.

the

for

these

substances creases

that

CaM

upon In

ported

'I3

to

that on their

'Abbreviations troponin

of

"...Ca 43 Ca

compete

of these 2+

with

agreed

TnI

and

CaM's

be of with

is set

other

an

free

similar

of

used are: CaM, calmodulin, TFP, W7, N-(6-aminohexyl)-5-chloro-l-naphtalenesulfonamide.

CaM

CaM

by

Ca

binding CaM

that

CaM. 43

such

to

(5-7).

hydrophobic

from

in

binding

by

affinity hydrophobic ions

compounds

the

These

Ca measurements

trifluoperazine,

the

unexpectedly,

metal

Shimizu

as

combining

Not 2+

of

for

direct

such

increased

(2).

enhanced

provided

enzymes

the

surprising

Tetrahymena

apo-CaM

substances,

affinity

or it

is 45ca2+

melittin,

target

to

to

2+

surfaces

rise

the

(10,ll)

Ca of

concerning

that

to

for

Ca2+

pro-

have shown 2+ Ca -calmodu-

by

of

hydrophobic

gives

results seems

binding

hydrophobic

also

TFP

phosphodiesterase the

as

Cd NMR experiments

NMR measurements

conclusions

I,

can

regulator

laboratories

experiments

Several

Ca2+ -induced .

various

affinity

such

demonstrated

binding

view

proteins,

substances

Moreover,

CaM's

(3,4).

TFP,

CaM's

in

enzyme cyclic-AMP 2+ -levels than

dialysis

notion

calcium-binding

Studies

that

model

this

of

(8,9).

(1,Z).

indicated

with

addition

Ca

multifunctional

Equilibrium

drug

reversibly

the

lower

of

antipsychotic

based

of

presence

evidence

the

cells

at

These

in

is

in-

(12).

et

al.

addition

(13) of

authors

TFP..."

suggested which

TnI

we

per-

re-

356

Vol.

122,

No.

formed

on

bovine

cative

of

increased

controversy on the

CaM

of

lished

anionic

shift

elsewhere

the

(17).

resonances

for

resonances

that

MATERIALS

we

AND

introduced

reagents.

The

use

of

of

to

CaM by 43 Ca NMR

details

shift 2+

the

of

reagents

Ca

and

free

Ca

analysis

has 2+

data

this 43 Ca

NMR.

indi-

To

improve based

method

will

allowed

us

without

Ca

as

apparent

experiment

this

43

of

COMMUNICATIONS

our

resolve

a novel

the

complicates

to

TFP

full

RESEARCH

interpreted

order

binding

protein-bound

normally

In

have

The

the

BIOPHYSICAL

we had

(14). the

studies,

AND

eventhough affinity

reinvestigated

earlier

use

(14-16), 2+

Ca

we

our

BIOCHEMICAL

3, 1984

to

the

NMR

be

on pub-

follow

overlap

of

spectra.

METHODS

43Ca NMR spectra were obtained an two different NMR spectrometers, a homebuilt 6 Tesla instrument (16) and a Nicolet 360 WB. The details of the sample preparations and the spectral parameters used for these studies have been presented elsewhere (15,17). Isotopically enriched 43Ca (60 i:) was obtained from Techsnabexport, Moscow, USSR through their Swedish sales agent Matreco AB, Sodertalje, Sweden. Bovine brain and bovine testis CaM were purified as previously described (11). All other chemicals used were of the highest quality commercially available.

RESULTS Upon resonance 2+ of Ca width

titration is

CaM.

the

a spectrum from

composed the

this

this

Note

that

spectra, and

shape

since

this

of

CaM

in

resonance

goes

exchange

(15,16).

observed

and

of

have

absence

brain

and

presence

However, slow

in

the

exchange

conditions 1351

of

are

not

followed

the prevail.

Since

narrow

(18)

presence protein in

signal

the

temperature

no

Addition 43 Ca reso-

the

long

via drug

shift.

the

was

When

1).

overlooked

shown.

testis

slow

with figure

easily

of

the

TFP.

B,

rather

and

from

fitted

proteins.

TFP

3 compares of

presence

is

linewidth

prohibitively

temperature

be panel

chemical

ions

is

resonance

is

indicates

resonance

Figure

increasing

thus

directly

in

required

concentration.

the

curves

broad

(see

resonance

the

line-

spectrum

nonetheless

with in

4 eqv.

the

this

calcium-binding

a titration

wide

almost

narrower

a similar

most

the

that

line

a broad

decrease

both the

can

43Ca2+ of

titration

of

it

with

studies

sites

would

with

that

resonance

a marked

the

binding

protein

A),

Lorentzian

results in

of

panel broad

Hz

Figure 1 shows such 2+ Ca . Although we know

additional

1,

protein-bound

in

CaM

Hz

apparent

for

behaviour

a higher

dence

2 the

the

is

arise

results

TFP

365

range

figure

The strong

figure

of

of

an

of

increased

6 eqv. levels

700-800

a ratio

(15,16).

containing

and

to

one

further

rapidly

component

a narrower

will

drug

two

is

of shift

situation

broad

it

up is

subsaturating

a single

presence

nance.

Hz

only

spectra

concentration

a sample

in

to

chemical

In of

800

NMR

normally

decreases

at

comparison

the

this

for

43Ca2+ Ca

Ca

resonance

simulation ease

From

the

an

;ith

the 2+

the

obtained

of

relative

in

major

data

CaM

in

As

obtained

our

(see

observed

per

of

of

drug

is

these acquisition depen-

added

intermediate no

change

of (10,14).

the to

in

linewidth

fast

Vol.

122,

No.

3, 1984

BIOCHEMICAL

AND

BIOPHYSICAL

PPM

RESEARCH

COMMUNICATIONS

1000

0

-1000

HZ

Simulations using Lorentzian lines for a 43Ca resonance measured .3 MHz for a sample containing 1 mM CaM, 6 mM CaZt, 100 mM NaClOs, pH 7.0. Panel A shows (from top to bottom) the measured spectrum, the simulated spectrum and the two Lorentzian lines (750 and 280 Hz) that were used for the fitting. Panel B shows the fitting of the measured data points to a single Lorentzian resonance of 365 Hz. The measured spectrum was broadened by 25 Hz in both cases by exponential multiplication of the free induction decay.

In this

a final

the

shift

final

concentration

shift

reagent

ions

up

to

experiment reagent

-125

ppm

used

Dy(PPP)g of

which

we

1.2

(a

mM.

can

displace

(17).

As

This

2

w-3.0

sample

I:2

is

from

complex

complex

resonances

of

is

a very

for

free

demonstrated

by

1

I

0

the

1

2

TFP /

3

4

DyC13

I and and

Na5P30,D)

effective or

comparing

added

3

to to

a

anionic

weakly

bound

figures

Ca

2+

4B and

I

0

CaM

figure

4C

I 20

60

60

T:rtlP. ( w

The 43Ca NMR linewidth as a function of the [TFPl/[CaM] ratio for mM CaC12 solution, pH 7.0 containing 0.40 mM bovine brain-CaM (0) a 3.0 mM CaC12 solution, pH 7.0 containing 0.44 mM bovine testis-CaM. The spectra for this figure were obtained at 17.2 MHz. The linewidth of the free Ca2+ resonance (- 12 Hz) is not substracted from those given in this figure.

Figure 4 3 310 The the this

The temperature dependence of the *3Ca NMR linewidth for: (0) a 3. mM CaClp solution at pH 7.0 containing 0.9 mM bovine testis-CaM, (0) a mM CaC12 at pH 7.0 containing 0.4 mM bovine testis-CaM and 1.3 mM TFP. spectra for this figure were obtained at 17.2 MHz. The linewidth of free Ca2+ resonance (- 12 Hz) is not substracted from those given in figure. 1352

Vol.

122,

No.

BIOCHEMICAL

3, 1984

,

1

I

>

50

Figure 4. containing additions

we obtain in

the

43 Ca The

This

downfield

improved

reagent

as

resonance

of the

the

can

is It

be

upon -50

broadened

paramagnetic of

-50

4C

the

-100

not

been

to

is

additional reagent spectrum

of

that of

COMMUNICATIONS

for

figures

exchange

with

amount (17). 4D was

free

of Upon obtained.

1353

obtained pH 7.0 with mM Dy(PPP):-,

the

excess

for a sample the following D) 3 mM TFP.

shift

the

of

Ca over 43Ca2+

addition

4A and

4C.

reagent 2+

protein-bound by

the

PPM

of

an the

affected

comparison

ppm

addition

containing

corresponds

because an shift

figure

after

solutions

ppm has

seen at

experiences

sample

of +I0

resonance

shifted

(16).

at

RESEARCH

VT-7

0

resolution

NMR spectrum

resonance

resonance

T

BIOPHYSICAL

“3Ca NMR spectra (24.3 MHz, 106 transients) 1 M bovine testis CaM, 100 fl NaCIOk, A) 2.0 evq. Ca2+, B) 6 eqv. Ca'+,C) 1.2

much

CaM.

AND

of

.

the

The

shift

upfield

Ca 2+ ions. The latter 2+ Ca bound to the protein

broadening addition

due of

The

most

to

the

3 eqv.

of

noticeable

presence TFP

to change

Vol.

in

122,

the

No.

spectrum

Similar W7,

3, 1984

results that

also

BIOCHEMICAL

is

a considerable

as

those

binds

BIOPHYSICAL

narrowing

shown

to

AND

in

of

figure

calmodulin

in

a Ca

RESEARCH

the

upfield

COMMUNICATIONS

shifted

4 were obtained with 2+ -dependent manner

2.

resonance

another (9)

drug,

(data

not

shown).

DISCUSSION The

results

nance

for

upon

addition

been

suggested

nance the

Ca

for 2+

of

the

the

of (< that

(16).

On

the

Ca'+.

Thus of

linewidths

(16)) off-rate

too

stopped contained ionic in

free

in

figure

43 Ca

other

equilibrium

that

to

flow 100 strength

the

various

sensitive

to

free

those

the

two

measurable or

studies

'H

whereas

(15,19).

The thus strength

virtually

on

at

23'C

the

in

agreement

(2-4,

8,9). Ca

Ca2+

From

have

the

at

NMR

one 2+

con-

exchange

rate

changed,

this

affinity

for

outcome

in

of

a in

figures

This

number

Ca NMR

23'C

the

time-

can

difference

43

using

in

the

not

the

resonances -1 800 set

set -I

further no change

increased with

earlier 700

Ca

2t

of

(ca.

the

an

is

between

temperature

rate

for

had reso-

(19)).

the 4C and is

in

(1000

set

This

Ca

-1 2+

low-affinity sites of CaM, as we have discussed 2+ Ca off-rate for the two high affinity sites 43 by Ca NMR but it can be estimated using

The

mM NaC104

This

There is 2+ Ca with

linewidth

determined

fluorescence

solution.

as the rate

exchange

upfield-shifted 2t a minimum Ca off-rate with

of

exchange

reflecting

are

protein

in

on-rates as

the

the

reso-

intensity

narrowing

decreased

the

in

low

the

fluorescence

ionic

the

2.

markedly

studies

for

for

be

has

NMR results

(11,15,16,19). slow

a very

TFP

in

of

differences

interpreted

these

is

resonance

that

directly

stopped-flow

elsewhere is

that

of

the in

that

decrease

from

Rather,

and

assumption

agreement and

liberated

shown

the

or

a decrease

suggesting

calculate

excellent

not (13).

indicates

From

measured

we can

is

al.

clearly

disappear

protein

addition

be

4 demonstrate not

results 43 the Ca

drug,

the

can

variety

the the of

the

Ca

2+.

figure

does 2+

et

Ca

10 set-'j.

result

Thus

Shimizu

to by

linewidth

clude

4D

TFP.

excess

bound

presence scale

in 2+

Ca

by

substantiated in

depicted

protein-bound

NMR

(19,ll).

our

earlier

good

agreement

indicates in

The

contrast

samples

work for

that

these to

recent

used

for

was the

performed 2-t off-rate Ca

numbers allegations

are

at

figure lower measured

not

extremely

(20).

'A second change is that the relative integrated intensities of the protein-bound and the free resonance alter. This is related to the fas,t that the shift reagent used in this study has an affinity for Ca which approximates that of the two weak2$ites of CaM (17). bound to the protein. Hence it would pull out a fraction of the Ca The increase observed for the protein-bound resonance upon addition of TFP provides thus additional evidence for an increased Ca2+ affinity in the CaM-drug complex. 1354

4

,

Vol.

122,

No,

The lnents

3, 1984

conclusions

performed

this

experimental

tion

of

ion

binding

bound

the

to

23Na+

cations,

ments

such

at

and

from

This

al.

CaM its

"excess

as implicit

It

is

our

itrarmally

of

concerning

metal

ion

problems

similar

to

off 2+ Ca

binding those

the

as

to

protein. to

Ca NMR

authors

works

seem

the

binding for method

is

free

and

break

exemplified the

Shimizu

proteinas

35c1 -

tighter

down, by

and

divalent leading

this

to

study,

time-consuming

by

study

such

other

measure-

Unfortunately

proteins

favour

and

ions 2+

to

to

becomes Ca

experi-

in the collecwell

between

monovalent

excess

perform

43

on

for

case

to

COMMUNICATIONS

time-savings rates

when

outlined

based

usually

case

the

RESEARCH

These

exchange

experience

pays level

ratio. obvious

often in

were

method"

the

is

BIOPHYSICAL

(13)

to of

when

decrease,

assumptions

cases

et 2+

This is usually the 39 + K (23). However,

a subsaturating

NMR papers suffer

Ca

AND

because

(13).

fast.

conclusions. in

high

macromolecules

rates

erroneous that

data

(22)

the

a very

Shimizu

strategy

are

exchange

of

at

NMR

ions

(211, the

BIOCHEMICAL

several and

Hatano

other (24)

above.

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

15. 16. 17. 18.

19. 20.

Klee, C.B., and Vanaman, T.C. (1982) Adv. Protein Chem. 28, 214-321. Rasmussen, H. (1983) Ann. Int. Medic. 98, 809-816. Keller, C.H., Olwin. B.B.. LaPorte, D.C.. and Storm. D.R. (1982) Biochemistry-21, 156-162.. Maulet, Y., and Cox, J.A. (1983) Biochemistry 22, 5680-5686. LaPorte, D.C., Wierman, B.M., and Storm, D.R. (1980) Biochemistry 9, 3814-3819. Vogel, H.J., Lindahl, L., and Thulin, E. (1983) FEBS Lett. 157, 24 -246. Krebs, J., Buerkler, J., Guerini, D., Brunner, J., and Carafoli, E (1984) Biochemistry 23, 400-403. Keller, C.H., Olwin, B.B., Heideman, W., and Storm, D.R. (1982) in Calcium and Cell Function (Cheung, W.Y. Ed) Academic Press, New York, 10 3-127. , . Inagaki, M., Tanaka, T., and Hidaka, H. (1983) Pharmacology 27, 125-129. Forsen, S., Thulin, E., Drakenberg, T., Krebs, J., and Seamon, K. (1980) FEBS Lett. 117, 189-194. Thulin, E., Andersson, A., Drakenberg, T., Forsen, S., and Vogel, H.J. (1984) Biochemistry 23, 1862-1870. Andersson, A., Drakenberg, T., Forsen, S., and Thulin, E. (1983) Eur. J. Biochem. 134, 459-465. Shimizu, T., Hatano, M., Nagao, S., and Nozawa, Y. (1982) Biochem. Biophys. Res. Comm. 106, 1112-1118. Thulin, E., Forsen, S., Drakenberg, T., Andersson, T., Seamon, K.B., and Krebs, J. (1980) in Calcium-binding proteins: Structure and Function (Siegel F.L. et al. Eds.), Elsevier, Amsterdam, 243-244. Andersson, T., Drakenberg, T., For&, S., and Thulin, E. (1982) Eur. J. Biochem. 126, 501-505. Drakenberg, T., ForsPn, S., and Lilja, H. (1983) J. Magn. Reson. 53, 412421. Vogel, H.J., and Braunlin, W.H. (1983) J. Magn. Reson. Vogel, H.J., Drakenberg, T., and Forsen, S. (1983) in NMK of newly accessible nuclei (Laszlo, P., Ed.) Vol I, Academic Press, New York, 157-192. Bayley, P., Ahlstriim, P., Martin, R., and Forsen, S. (1984). Biochem. Biophys. Res. Comm. 120, 185-191. Burger, D., Cox, J.A., Comte, M.A., and Stein, E.A. (1984) Biochemistry 23, 1966-1971. 1355

Vol.

21. 22. 23. 24.

122,

No.

3, 1984

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Forsen, S., and Lindman, B. (1981) in Methods of Biochemical Analysis (Glick, D., Ed.), Vol. 27, 284-486. Braunlin, W.H., Drakenberg, T., and For&, S. (1985) in "Current Topics ir Bioenergetics" (Lee, P.C. Ed), (in press). Braunlin, W-H., Vogel, H.J., and Forsen, S. (1984) Eur. J. Biochem. (in press). Shimizu, T., and Hatano. M. (1982) Biochem. Biophys. Res. Comm. 104, 13561362, ibid 104, 720-726, (1983) Inorg. Chem. Acta 80, L37-L39.

1356