BIOCHEMICAL

Vol. 75, No. 3, 1977

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

CIRCULAR DICHROISM OF BIOLOGICAL

MEMBRANES:

PURPLE MEMBRANE OF HALOBACTERIUM HALOBIUM M. M. Long*,

D. W. Urry*

and W. Stoeckenius**

*Laboratory

of Molecular Biophysics and the Cardiovascular Research and Training Center University of Alabama Medical Center Birmingham, Alabama 35294 **University of California School of Medicine Cardiovascular Research Institute and Department of Biochemistry and Biophysics San Francisco, California 94143

Received

February

24,1977

SUMMARY: Circular dichroism spectra of purple membrane suspensions exhibit distortions which are characteristic of particulate a-helical systems with ellipticities at 224 nm of the order of 10000. We show here that the pseudo reference state approach to the correction of CD data, when applied to the purple membrane system, results in a CD spectrum which, in its entirety, is typical of a protein with about 75% a-helix, e.g. the corrected ellipticity at 224 nm is 24000 and at 190 nm is 45000. This result is consistent with electron microscopy and diffraction data on the purple membrane which indicates 70-80% a-helix. Application to the purple membrane system validates the pseudo reference state approach. We also note that 80% trifluoroethanol solubilizes the membrane but results in a similar average secondary structure of the protein. Circular average

dichroism

membrane

if,

corrections

are

common to all

membrane

compared

cantly

and variably

membrane system

and indeed

0 1977

distortions,

by Academic Press, Inc. in any form reserved.

of reproduction

a-helical

can provide

Typically

the

but display and the

cross

proteins.

on the

insight

architecture,

if,

and flattening

structure,

depending

The anomalies are

patterns

to estimate

dampened,

suspension.

(1).

the two minima,

of soluble

any attempt

in this

scattering

systems

show a-helical

to the pattern for

light

particulate

of the maximum,

of biomembranes

and changes

are made for

importantly

Copyright All rights

architecture

suspensions

wavelengths

(CD) spectra

CD spectra shifts

point

In addition,

state

suspension documented

which of bio-

to longer

over

particulate

as has been well

and only

distortions

the CD curves

are due to the

into

when and

are signifiof the

nature

of the

(l-10).

A

725 ISSN

0006-291X

Vol. 75, No. 3, 1977

relatively

simple

of membranes, (1).

BIOCHEMICAL

It

bility

is

and practical

called the

of this

purpose

described.

published

x-ray

protein

pseudo

with Recently

halobium

therefore,

reference

state

Materials

and Methods

report

for

appro

whose

and Unwin

that

contains

distorted

has been developed

the

validity

(1 -13)

and Blaurock

(14)

membrane

a-helix.

validity

of CD data

inde-

of

the single

as much as 70-80%

the correction

has been

bacteriorhodopsin,

the

and applica-

structure

of the purple

to corroborate

CD spectra

ch,

protein

studies

indicated

(15),

these

to demonstrate

microscopy

was chosen

method

state

Henderson

which

membrane

membrane,

reference

a membrane

and electron

in this

means of correcting

of this

approach

pendently

Halobacterium

the

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

The purple

of the pseudo

from

particulate

systems.

CD spectra were obtained at ambient temperature on a Cary 60 spectropolarimeter adapted with a model 6001 attachment for circular dichroism using the 100 mdeg range. UV spectra were recorded on a Cary 14 spectrophotometer with the sample cell placed close to the phototube in the cell compartment (16). The effective pathlengths of the cells were 0.233 mn for the suspensions and 1.7 mn for the membrane solutions. The final protein concentration, as determined by the Lowry method (17), was 0.53 mg/ml for the suspension and 0.053 mg/ml for the solution. Sonication was performed on ice with a Branson Sonic power sonifier at 7.5 amp for two 30 second intervals. Solubilization for the pseudo reference state was achieved by addition of stock sodium dodecyl sulfate, SDS (Eastman Organic Chemicals), to the membrane suspension first, followed by dilution to final volume with trifluoroethanol (TFE). The 80% TFE sample without SDS was filtered through a mitrex 5 urn Millipore filter giving a spurious high absorption due to absorbing material extracted by TFE from the filter itself. Trifluoroethanol, purchased from Halocarbon Products, Hackensack, N. J., was redistilled in glass with a Duflon column of glass beads (lower 3/4) and glass helices (upper l/4). The first lo-15% of distillate and last 20-30% of pot residue were discarded. NaHC03 was added to the remainder to remove traces of acid. The purple membrane was prepared as outlined by Oesterhelt and Stoeckenius (15). Results

and Discussion

The spectrum

of the

purple

is

characterized

by red

shifting

all

particulate

systems,

are

tions

(1).

distortions the magnitude is

apparent

membrane

suspension

and dampening. due to light is

are still

value

of a soluble that

corrections

the

protein

which

are

required.

726

These

scattering

Even when the suspension apparent:

(see

sonicated of is

[O]'~~

70-80%

Figure

1, Curve

features,

cotmnon to

and flattening (Figure is

1, Curve still

a-helical.

a)

only

distorb),

the

one-half

Therefore

it

Vol. 75, No. 3, 1977

BIOCHEMICAL

:. t3

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

a.

Purple membrane suspension (frozen)

b.

Purple membrane sonicated I

c.

Purple membrane TFE.li20.4:I

d.

Corrected

e.

Metmyoglobin

Ellipticities

220

200

mm

240

X (nm)

FIGURE 1:

With

CD spectra of the purple membrane suspension which had been previously frozen (a) and after sonication (b) Curve d is the same corrected ellipticity as for Figure 2, Curve a. Note the marked correspondence between d and c, the membrane solubilized with 80% TFE and filtered to remove non-protein particles. Curve e is the spectrum of metmyoglobin which is 70% a-helix as determined by x-ray, reprinted with permission from Quadrifoglio, F., and Urry, D.W., J. Am. Chem. Sot. 90, 27552760, 1968. Copyright by the American Chemical Society.

the

was calculated, and Figure reference

pseudo

wavelength

3, Curve state

reference

(or

d.

state

approach

by wavelength, In this

solution)

calculation absorbances

727

the corrected

and is

shown

in Figure

the membrane (see

Figure

membrane

2),

1, Curve

suspension the

spectrum

former's

d

and pseudo CD

Vol. 75, No. 3, 1977

BIOCHEMICAL

0. Purple membrane sonicated I men.

I.0 h .5 $)0.6 0 6.u

x 0

0.6

-

0.4

-

0.2

-

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

b. Pseudo reference TFE:+O-SDS.

\

stats I:4 (0.4%)

_._ 200

220

260

240 A (nm)

FIGURE 2:

curve,

Absorbance data of the purple membrane suspension, protein/ml, 0.233 mm cell) and the purple membrane (0.053 mg protein/ml, 1.72 mm effective pathlength 4.

the

latter's

ORD curve

["torrected is evaluated

the

suspension

pseudo

a better

fit

reference than

Duysens'

correction

factor

[ml1

correction for

3, Curve

[m15”

[o]~,,~~

-k

-

Q;

ellipticity

at one wavelength

of the

Q, is

=

(Figure

(19). factor absorption

10

[ml

is

for

the

flattened

absorption

obscuring

728

used.

(1) X, k (=0.52)

the mean residue

has been shown

AF is

1 were

-AF

at wavelength

It

and Equation

Q,

.

(1,16,18),

state.

c),

(a): (0.53 m solution, (b 4 : relative to

molar

empirically absorbance

flattening

due to light

is

that

a constant rotation

[ml 5/k

at wavelength (20)

scattering.

and Q, is

is A, the

BIOCHEMICAL

Vol. 75, No. 3, 1977

AND BiOPHYSlCAL

RESEARCH COMMUNICATIONS

7

6 6

o Purple membrane sonicated I min

d I

5

-

: I ’ !

4-

b. Pseudo stole

reference - TFE:,+O-SD! 1'4 10.4%)

5

c. Pseudo State

reference - ORD c*w*

4

d. Corrected

* b-3

Ellipticities 3

e b x

x 52

2

-

I

-

/

2-

T

I

0

I

2

200

220

240

A (nm)

FIGURE 3:

CD spectra of the purple membrane sonicated suspension (this is a different sample from Figure 1, Curve b), the purple membrane solution (a), and pseudo reference state (b), the pseudo reference state ORD (c), and the corrected ellipticity (d). The calculated percent of o-helix using Curve d is 73% for the purple membrane - in good agreement with the x-ray data indicating between 70 and 80% a-helix.

The TFE - SDS molecularly meets

the

Equations

criteria

for

an appropriate

state pseudo

(mds)

of the

reference

state

purple

membrane

by obeying

2 and 3.

224

[@Imds

In the

dispersed

initial

>

224

[OICl

approximation

224 [@Isusp

= (Qi2')'

indicated

by the

729

.

(2)

Q;,,

subscript,

CI,

to [CII’~~,

Q, is

Vol. 75, No. 3, 1977

BIOCHEMICAL

taken

as 1, making

about

10 to 20% less

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

a conservative than

[o];::

I

.

I

value,

In this

which

case it

generally

should

be

was 9% less.

(3)

In this

approximation

conservative

value;

be within

(Figure

is the

d).

its

With pseudo

the

corrected

o-helix

a calculated indicated

Becher

(22),

and Cassim

CD data

of membranes.

extrema

match

spectrum

is

spectrum

(Figure

well.

the

uncorrected

which

The correspondence

is

similar

quantity

calculated and the

effective

soluble

is

striking

of a-helix protein

with

of the

purple

importance

the

by comparing curve

the

estimate

membrane

shape

at the

of the

corrected

the metmyoglobin for

bacteriorhodopsin. result

protein

in

a

bacterior-

metmyoglobin.

membranes

with

detergents

and other

730

has proven particulate

especially

systems

by

raw

magnitudes

corrections

the membrane

membrane's

of correcting

absolute

but also

for

gives

(11-13).

to be compared

the

This

70 and 80% a-helix

corrected

that

of an

x 10'.

purple

d of

studies

from

the calculated

showing

ellipticity

the

the

evident

e) with

the

Curve

microscopy

CD spectrum

do the

and in combination

in solubilizing

is

73% o-helix,

This

1, Curve

This

underscores

with

appropriate.

between

1 and 3,

spectrum;

-3.3

(224

3c) were

ellipticity

and electron

was calculated

agrees

Not only

a protein

is

should

absorption

(Figures

with

is

it

(Figure

suspension

CD spectrum

X-ray

which

curve

a

wavelengths

The characteristic

of the

Usually

ORD curve

membrane

b).

.

also

suspension

in good agreement

--in situ

of 73% a-helix

TFE alone

is

state

ellipticity

a is the

(Curve

bacteriorhodopsin

from

and membrane

corrected

which

[O],$l

[@Ii:"

met at critical

reference

of 73% a-helix.

of 45% calculated

hodopsin

pseudo

state

CD spectrum

in 25% of

state

224 nm extremum

value

The CD value corrected

reference

3, Curve

reference

as 1, making

the CD requirements

spectrum

at the

was within

complete

In Figure

pseudo

it

2) and the

used to obtain Curve

here

lo-30%.

and 190 nm) the curves

at 190 nm, Q, is taken

(1,23-27).

BIOCHEMICAL

Vol. 75, No. 3, 1977

It

is

interesting

TFE without the

spectrum.

secondary

in the membrane. details

that

SDS and filtered

corrected

average

to note

of their

ACKNOWLEDGMENTS: of Health Program

This

the membrane (Figure

This

structural

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

suggests

1, Curve that,

properties

conformation

c)

solubilized

has a CD spectrum

in a membrane

of the

does not necessarily

membrane

suspension

proteins

mean that

are

with

80%

close

TFE solution, similar

polypeptides

to the

to those retain

the

in TFE.

This work was supported in part by the National Project Grant numbers HL-11310 and HL-06285.

Institutes

REFERENCES: 1. :: 4. z: 7. 8. 1;: i:: 13. Z: 16. 17. 18. 19.

2 2 24. 25. 26. 27.

Urry, D.W. (1972) Biochim. Biophys. Acta 265, 115-168. Urry, D.W. and Ji, T.H. (1968) Arch. Biochem. Biophys. 128, 802-807. Schneider, A.S., Schneider, M.T. and Rosenheck, H. (197rProc. Natl. Acad. Sci. 66-, 793-798. Kornguth, S.E., Flangas, A., Perrin, J., Geison, R. and Scott, G. (1972) Prep. Biochem. 2, 167-192. Litman, B.J. (1972) B&hem. 11, 3243-3247. Singer, J.A. and Morrison, M.T972) Biochim. Biophys. Acta 274, 64-70. Juliano, R.L. (1973) Biochim. Biophys. Acta 300, 341-378. Rottem, S. and Hayflick, L. (1973) Can. J. Biochem. 51, 632-636. Storey, B.T. and Lee, C.P. (1973) Biochim. Biophys. Eta 292, 554-565. i$ni;;; B.J. and Barenholz, Y. (1975) Biochim. Biophys. Acta 2, Henderson, R. and Unwin, P.N. (1975) Nature (London) 257, 28-32. Unwin, P.N. and Henderson, R. (1975) J. Mol. Biol. %325-440. Henderson, R. (1975) J. Mol. Biol. 93, 123-138. Blaurock, A.E. (1975) J. Mol. Biol. 93, 139-158. Oesterhelt, D. and Stoeckenius, W. (m74) Methods in Enzymology, Volume 31 (Colowick, S.P. and Kaplan, N.O., eds.) p. 667-678, Academic Press, New York. Urry, D.W. and Long, M.M. (1974) Methods in Membrane Biology, Volume 1 (Korn, E.D., ed.) p. 105-141, Plenum Press, New York. Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. Urry, D.W. (1974) Methods in Enzymology, Volume 32 (Colowick, S.P. and Kaplan, N.O., eds) p. 220-234, Academic Press, New York. Urry, D.W. and Long, M.M. (1977) in The Physiological Basis for Disorders of Biomembranes (Andreoli, T.E., Hoffman, J.F., Fanestil, D.D., eds.) (in press) Plenum Press, New York. Duysens, L.N.M. (1956) Biochim. Bioph s. Acta 19, 1-12. Quadrifoglio, F. and Urry, D.W. (1968 Y J. Am. fiem. Sot. 90, 2755-2760. Becher, B. and Cassim, J.Y. (1976) Biophys. J. l&, 1183-lzb0. Long, M.M. and Urry, D.W. (1976) In Membrane Spectroscopy (Grell, E., ed.) (in press) Springer-Verlag, Heidelberg. Masotti, L., Urry, D.W. and Llinas, R. (1973) Acta Vitamin Enzymol. 2, 154-158. Masotti, L., Urry, D.W., Krivacic, J.R. and Long, M.M. (1972) Biochim. Biophys. Acta 266, 7-17. Urry, D.W., Masotti, L. and Krivacic, J.R. (1971) Biochim. Biophys. Acta 241, 600-612. Urry, D.W., Masotti, L. and Krivacic, J.R. (1970) Biochem. Biophys. Res. Commun. 4l-, 521-524.

731