PMR studies of the substrate induced conformational change of glutamine binding protein from E. coli

BIOCHEMICAL

Vol. 53, No. 1,1973

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

PMRSTUDIESOF THE SUBSTRATE INDUCEDCONFORMATIONAL CHANGE OF GLUTAMINEBINDING PROTEINFROME.

COLI

George P. Kreishman, Dan E. Robertson, and Chien Ho, Department of Biophysics and Microbiology, University of Pittsburgh, Pittsburgh, Pennsylvania

Received

April

30,

15260

1973 Summary

The substrate induced conformational change of glutamine binding protein isolated from E. coZi has been studied by high resolution proton magnetic resonance spectroscopy. The addition of L-glutamine to a protein solution caused a marked change in the proton magnetic resonance spectrum. The chemical shifts of several resonances were considerably different for the free and complexed protein. The line width of the methyl protons decreased considerably with the addition of substrate indicating that the environment of a sizeable percentage of the methyl groups is different. The kinetics of binding as well as a possible mode of action of the binding proteins will be discussed.

The isolation

of glutamine binding protein

others (2), released from

coli

E.

(GLn-BP) (l),

as well as

cells by the osmotic shock procedure of

Neu and Heppel (3) has been described and experiments have been reported which implicate

these proteins

in the transport

of amino acids, sugars, in-

organic ions, and vitamins across cell membranesin gram negative bacteria cells

The exact mode of action of the binding proteins

(2,4).

transport

process has yet to be established (4-6).

Several studies, utilizing rotary

in the

dispersion,

circular

fluorescence,

gel electrophoresis,

dichroism, infrared

spectroscopy, denaturants,

and proton magnetic resonance (PMR) on the conformational this class of proteins have been reported with conflicting Whereas minimal substrate induced conformational reported utilizing electrophoresis two distinct

optical

properties

changes upon binding were

and binding data for galactose binding protein

Copyright 0 1973 by Academic Press, Inc. AN rights of reproduction in any form reserved.

states (9-10). 18

of

results (1,7-10).

and PMRexperimental methods (1,7-9),

conformational

optical

the gel have shown

Wehave extended these studies

BIOCHEMICAL

Vol. 53, No. 1,1973

to Gln-BP utilizing

AND BIOPHYSICAL

high resolution

marked conformational

RESEARCH COMMUNICATIONS

PMRspectroscopy and have observed a

change in the protein upon substrate binding.

sible mechanismof action of the binding proteins in the transport is discussed in light

A posprocess

of these results.

EXPERIMENTAL

A mutant of E. cozi strain 7 (GLNP-1) which can grow on glutamine as the sole carbon source and which was supplied by Dr. Leon A. Heppel was used for isolation

of Gin-BP.

plimented with glucose. fluid

The cells were grown in minimal mediumsup-

The Gln-BP was isolated

and assayed for glutamine binding activity

lished method (1).

The binding protein

from the osmotic shock by the previously

was exchanged with D20 (99.7%)

from Merck, Sharpe & Dohmeof Canadaby repeated concentration ultrafiltration

pub-

utilizing

with an Amicon UM-10 membranefollowed by dilution

with

D20. The L-glutsmine (A grade) was purchased from Calbiochem. Unless otherwise noted, all

samples were prepared in 0.05M NaCl, O.OlM potassium

phosphate pD 7.1 buffer.

The pD of solutions was determined by adding 0.4

pH unit (11) to the value read from a Radiometer Model 26 pH meter equipped with a Beckman39030 combination electrode.

The protein

concentration

was determined by the Lowry method (12). The PMRspectra were obtained on the MPC-HF250 MHz superconducting Spectrometer (13).

The signal-to-noise

ratio

was improved by the

correlation

spectroscopic method of Dadok et aZ. (14) utilizing

a Sigma 5

computer.

The probe temperature of the spectrometer was 31'C.

Chemical

shifts

are reported in parts per million

(ppm) from the residual HODres-

onance in each sample. The HODsignal is -4.72 ppm downfield from the proton resonance of 2,2-dimethyl-2-silapentane-5-sulfonate. sign of the chemical shifts that of HODand the positive

indicates

The negative

that the resonance is downfield from

sign indicates

from the HODsignal. 19

that the resonance is upfield

Vol. 53, No. 1,1973

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

PHE

A

N-H Protons

Fig. 1.

Aromatic Protons

Aliphmc

Protons

The 250 MHz spectra-$f Gln-BP: A, 6x10-'M protein denatured with 7M ds-urea; B, 2x10 M protein in O.Ozv NaCl, O.OlM potassium phosphate, pD 7.1 buffer; and C, 2x10 M protein with 2O:l excess Gln in 0.05M NaCl, O.OlM potassium phosphate, pD 7.1 buffer. The aromatic region is the accumulation of 1000 scans and is shown at a higher attenuation than the aliphatic region for which only 200 scans were accumulated. The symbol (*) in spectrum C denotes resonances due to free Gin. The symbol (+) in spectra B and C denotes ring-current shifted methyl groups. The position of the HODresonance in A relative to B and C is different due to exchange with the urea.

RESULTS ANDDISCUSSION --

The 250 MHz PMHspectrum of Gln-BP which has been denatured by 7M dkurea is shown in Figure 1A. fairly

The resonances of the various protons are

sharp and each type of residue is magnetically

be expected for a random coil polypeptide however, the linewidths

chain (15).

as would

In the natural

state,

are much broader and nonequivalence between various

types of residues becomesapparent (Figure 1B). changeable protons,

equivalent,

In addition

to the nonex-

a broad resonance -4 ppm downfield from the HODreso-

20

BIOCHEMICAL

Vol. 53, No. 1,1973

nance

is

tons

observed

do not

exposure, not

exchange these

both

with

the

at this

from

HOD can be attributed

methyl

groups.

The resonances

methyl

groups is

tire

1C.

Gross

spectral

marked ing

region.

dipolar

of the methyl

current

shifted

indicating These

that

results

of residues induces

methyl

relative

clearly

indicate

that

different

and the

change linewidths

PMR time

exchange

to that

this 5

leads

where

The additional lifetime

scale.

of ~0.1

to the magnetic

absence

the

free

broadening

the

of protein, and bound

of ~3Hz for

the

second.

21

free

spectrum

is

the

This

line

of the

sharpen-

residues

from a of the

addition

has changed.

the

spectrum

free

over

limit

individual

Gln yields

free

substrate bind-

a range is

slow

L-glutamine,

the rate In the

number

of the

of exchange

Gln in the

of Gln

of a large

unchanged

states.

of the

en-

that

of the

(16).

the

residues

kinetics

upfield

over

upon the

the

to

shown in

Several

reflects

broadening

the mean lifetime

methyl

Gln is

broadening

to an additional

ppm.

and therefore,

to 1:20,

is

of the

environments

of the excess

of the

groups

solution

aromatic

Since

ppm

and the

of these

shifted

in Gln-BP.

of 1:2

in the

‘c is

are

The additional

of Gln between

change, $

ratio

residues

environment.

upon Gln binding

a conformational

as compared

of some of the

position

of protein-substrate on the

at s+3.8

their

are

protein

resonance

resonances

at ~3.8

can be observed feature

to a more mobile

to specific

can be assigned

anisotropy

interesting

as a change

environment

peak

of aromatic

in the PMR spectrum

and are

and exhibit

resonance

of Gln to a protein

The most

can be interpreted

ring

by

addition

weeks

of the majority

this

pro-

of nonexchangeable

resonances

to a composite

ring-current

several

are broad

The broad

from

these

of the protein

of the

in the vicinity

changes

sharpening

rigid,

ing

of the

even after

regions,

time.

upfield

due to the magnetic

The effect Figure

are

Since

the resonances

assignment

upfield

shift

Since

and aliphatic

specific

which

solvent

in the interior

solvent.

possible

of N-H protons.

deuterated

aromatic

overlap, was not

expected

are clearly

the

in the

considerable residues

with

protons

in contact

protons

in the region

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

of chemical of slow resonances

state

(17).

a pre-exchange

ex-

Vol. 53, No. 1,1973

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Although it has been reported that Gln-BP can be stored for several months at 3' and -9O'C without

loss of activity

line broadening in the PMHspectra, ple was refrigerated

(l),

indicative

we have observed severe

of aggregation,

at 3'C for several weeks or after

fer to remove the Gln.

when the sam-

dialysis

Although the aggregation of the protein

due to the high concentration

against bufis probably

needed for PMH, extreme care must be exercised

in the handling of this protein.

The PMHspectrum was reproducible

for sev-

eral weeks if the sample was stored frozen at +40°C and only these samples were used in this study. The conformational with the postulate mobility

change as observed by PMRfor Gln-BP is consistent

of Boos and Gordon (10) that the change in electrophoretic

of galactose binding protein

structural

upon substrate binding indicates

change and perhaps a change in the surface charge.

must be exercised in comparing the structural proteins

with its in vivo transport

essentially Gln binding, can favorably protein,

of purified

If the binding protein

the conformation of Gln-BP is sufficiently interact

with the lipid

to the P protein

mone2la typhinruriwn ( 18 ).

If,

properties

bilayer

binding is

that it

or bind to a membranebound

in the histidine

transport

studies on the effect

of binding proteins

conclusions can be drawn.

different

however, the binding protein

surface of the membrane, further

are currently

function.

Care, however,

free in the periplasmic space then one can speculate that upon

similar

formational

properties

a

system in Salis bound to the

of lipids

on the con-

are required before meaningful

These possible modes of action of binding proteins

under investigation.

Acknowledgement: This work is supported by research grants from the National Institutes

of Health (GM-18698 and H&00292) and from the National

Science Foundation (GB-37096X). HEFEHENCES 1.

2.

Weiner, J. H. and Heppel, L. A., J. Biol. Berger, E. A. and Heppel, L. A., J. Biol. references (2-40) therein. 22

Chem. 246, 6933 (1971). Chem. m 7684 (1972) and -'

Vol. 53, No. 1,1973

Z: 2: 7. 8. 9.

13. 14.

15. 16. 17. 18.

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Neu, H. C. and Heppel, L. A., J. Biol. Chem. 240, 3685 (1965). Kaback, H. R., Ann. Rev. Biochem. 39, 561 (19m. Pardee, A. B., Science l&, 632 (1x8). Heppel, L. A., J. Gen. Physiol. 54, 95s (1969). Langridge, R., Shinagawa, H. andTardee, A. B., Science 169, 59 (1970). Berman, K. and Boyer, P. D., Biochemistry z, 4650 (1972r Boos, w., Gordon, A. S., Hall, R. E. and Price, H. D., J. Biol. Chem. 247,

10. 11. 12.

BIOCHEMICAL

917 (1972).

621 (1971). KS, W. and Gordon, A. S., J. Biol. Chem.z, Glasoe, P. D. and Long, F. A., J. Phys. Chem.6& 188 (1960). Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., J. Biol. Chem. 193, 265 (1951). Dadok, J., SFcher, R. F., Bothner-by, A. A. and Link, T., Abstracts, Section C-2, Eleventh Experimental NMRConference, Pittsburgh, Pa. (1970). Dadok, J., Sprecher, R. F. and Bothner-by, A. A., Abstracts of 13th Experimental NMRConference, Pacific Grove, Calif. McDonald, C. C. and Phillips, W. D., J. Am. Chem. Sot. 91, 1513 (1969). McDonald, C. C. and Phillips, W. D., J. Am. Chem. Sot. %, 6332 (1967). Pople, J. A., Schneider, W. G., and Bernstein, H, J., "Ffigh-resolution Nuclear Magnetic Resonance" (McGraw-Hill, NewYork, 1959), p. 221. Ames, G. F. L. and Lever, J., Proc. Natl. Acad. Sci. U.S.A. 66, 1096 (1970).

23