Oxygen equilibrium of emulsified solutions of normal and sickle hemoglobin

Vol. 88, No. 3, 1979

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

June 13, 1979

Pages 980-987

OXYGEN EQUILIBRIUM

OF EMULSIFIED

J.G.

Pumphrey',

April

30,

tina A.P.

K. mbya',

National Institute Chemistry, National Institutes of Health,

Labbratory of Biophysical and Digestive Diseases, Received

SOLUTIONS OF NORMAL AND SICKLE HEMOGLOBIN Minton of Arthritis, Metabolism Bethesda, MD, 20014

1979

Summary: Emulsions containing microdroplets of concentrated solutions of normal or sickle hemoglobin dispersed in a continuous oil phase have been prepared, and the aggregation of sickle hemoglobin within microdroplets in a deoxygenated emulsion demonstrated. The equilibrium oxygen saturation of hemoglobin in the emulsions has been measured as a function of the partial pressure of oxygen by a novel spectrophotometric technique which corrects for the scattering of light in the emulsion. The half-saturation oxygen pressure and cooperativity of oxygen binding are substantially greater in concentrated (27 g/al) solutions of The shape sickle hemoglobin than in solutions of non-aggregating hemoglobin. of the oxygen equilibrium curve of concentrated sickle hemoglobin is qualitatively discussed in terms of a previously proposed model (1).

INTRODUCTION Analysis tions

of the difference

of normal

stantial

hemoglobin

information

of sickle highly

concentrated

oxygen

equilibrium

powerful

The oxygen

tions

utilized

by Gill

cell, 1. 2.

in

with

and highly

viscous,

aa

has remained of whole readily oxygen

for

et al.

(8),

concentrated

a thin

Present Present Streets,

solutions

with

One technique

equilibrium

"gelled"

layer

solutions

sickle

who recently

reported

HbS solutions

in (tens

@ I979

by Academic Press, Inc. in any form reserved.

of reproduction

methods

980

of measuring

Hence,

indicating surface

surface

this

of sickle that

a specially of microns

HkS solu-

to volume volume

was

of oxygen

constructed thick)

ratio

ratio

the measurement

address: D-477, Abbott Laboratories, North address: Box 746, Johns Hopkins University, Baltimore, Md 21218

0006-291X/79/110980-08$01.00/0 Copyright All rights

or "gelation"

and suspensions

(4,5,6,7),

an adequate

sub-

unexploited.

blood

when a sufficient

of HbS solution

can provide

are inapplicable. essentially

of solu-

of HbS are both

conventional

measured

attaining

curves

(HbS)

However,

however,

exists.

hemoglobin

(1,2,3).

method

do equilibrate

and sickle

equilibrium

of aggregation,

equilibrium

is,

the oxygen

the thermodynamics

in hemoglobin

analytical

erythrocytes

(HbA)

about

hemoglobin

between

optical

bounded

Chicago, Illinois 34th and Charles

on the 60064

Vol. 88, No. 3, 1979

BIOCHEMICAL

surface

oxygen-permeable

by a transparent

method

for

in which sion

measuring the necessary

of the hemoglobin

the

oxygen large solution

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

membrane.

equilibrium

surface

in concentrated

to volume

as microdroplets

We report

ratio in

is

here solutions

obtained

a continuous

another of HbS,

by disperoil

phase.

METHODS Preparation of HbS and HbA Solutions. Fresh whole blood was obtained from heterozygous (sickle trait) donors. All succeeding steps were carried out in the cold. The erythrocytes were washed three times in 0.9% NaCl and the buffy coat removed. The cells were hemDlysed by three cycles of freezing/ thawing followed by the addition of an equal volume of distilled water, and the henolysate dialyzed overnight against 0.05 E Tris, 5 m- EDTA, pH 8.0 (23'C). The hemolysate was centrifuged at 30,000 RPM for 1 hr and the supernatant decanted. This step was repeated until no more precipitate was obtained. The clarified hemolysate was applied to a DEAE-Sephadex column equilibrated in 0.05 g Tris, pH 7.8 (23'C), and eluted with this same buffer. Pooled fractions containing either HbS or HbA were dialyzed against 0.05 M phosphate, 0.1 M KCl, 10 ti- EWTA, pH 7.0 (23*(Z), and concentrated to >20 y/d1 by ultrafiltration. The concentrated, purified solutions may be stored in liquid nitrogen for several nunths without significant increase in the amount of metHb present. Preparation of Emulsions. An emulsion-forming liquid was prepared by mixing 95% SlOON, a high molecular weight paraffin oil, 3% ENJ-3029, a high molecular weight amine, and 2% Span 85, a sorbitan monooleate surfactant (9). The solubility of oxygen in SlOON was found by gas chromatography to be roughly 4 times greater than its solubility in water at room temperature and atmospheric pressure. Four parts of cold emulsion-forming liquid were added to one part of hemoglobin solution and the mixture emulsified by milling in a chilled tissue homogenizer for ca. 2 min. The resulting emulsion is stable for at least several weeks. However, it was found that oxyHbS autooxidizes more rapidly than usual when emulsified, so the emulsions were generally prepared the evening before their use, deoxygenated immediately, and stored under nitrogen until used. Measurement of Oxygen Equilibrium. Aliquots of deoxygenated emulsion were equilibrated with N 7:0, mixtures of known composition (Lif-0-Gen, Cambridge, MD.) by passing &e gas over the gently agitated emulsion. It was found that oxygen equilibrium was achieved within 15 minutes at room temperature (23+1OC). The sample to be measured was then transferred with a gastight syringe (previously flushed with the gas mixture being used) to a O.lmm optical cell (also flushed with the gas mixture being used). The cell was sealed with no gas space over the emulsion. Because of the high turbidity of even thin films of these emulsions, a novel spectrophotometric technique was used to measure fractional absorbance changes. The spectrophotometer employed (Aminco-Chance, American Instrument Co., with Beam-Scrambler accessory) has separate m3nochromators for the sample and reference beams, which pass alternately at 60 Hz through the optical cells. The reference beam is fixed at 548.9 run, an isosbestic point for oxy- and deoxyhenvglobin. The sample beam scans from 535-565 run, producing a difference spectrum: AA(A) [= A(X) - A(548.9)1. An apparent fractional oxygen saturation is defined as follows: app (x,p) I AA(A,p) - AA(X,O) Y d~(A,l atm) - AA(A,O)

981

Vol. 88, No. 3, 1979

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

982

Vol. 88, No. 3, 1979

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

where A~(h,p) refers to the differential absorbance recorded at wavelength at oxygen partial pressure p. Because of scattering, ~~&,~~umy~ay~y~;&~ x. However, the true fractional change In absorbance o globin, independent of scattering effects, should be given by lim = X+548.9

Y(P)

1

'app (X,P)

(A,p) was observed to vary essentially In our experimnts, yap with wavelength (with a sma P 1 slope) over the region 535-565 run. noise increased, as expected, in the regions immediately adjacent as AA(X,P) goes to zero for all p). The value of y(p) was (i.e., calculating the best least-squares straight line through the data for a single value of p, excluding a region 2.5 nm on either side bestic point, and interpolating to the isosbestic point.

linearly Experimental to 548.9 nm obtained by for y app(A,P) of the isos-

RESULTS AND DISCUSSION Figure

la shows

largest

microdroplets

parable

in size

lets

present

the bulk taining

a 1000x

observed

red

to

cells.

are much smaller

than

of the aqueous

Although these,

lb)

the

that

in a manner

in diameter,

larger

aa

In oxygenated almost

all

the deoxygenated

comparable

con-

observed

appeared

concentrated

the

slow

deformation

of

of weak birefringence. HbS is

to that

contain emulsions

microdroplets

appearance

com-

of the microdrop-

deoxygenated

we observed

The

roughly

microdroplets

of sufficiently

and concomitant

emulsion.

the majqrity

when an emulsion

indicate

the microdroplets

microns

was deoxygenated,

(Figure

observations

of an oxygenated

several

HbS solutions,

HbS solution

microdroplets These

phase.

However,

to be spherical.

g/al)

were

blood

HbA or dilute

(>20

magnification

observed

aggregating

within

in red blood

cells

(10). The oxygen from a number data

points

torr) for

equilibrium

normal

curve

hemoglobin

of these

results

in Figures

and emulsified

hemoglobin

(short

(n = 2.6)

calculated

summarized

2a fall,

of an Adair-type

coefficient of the

ficance

of homogeneous

fitting

the Hill

data

shown in Figure

same oxygen squares

equilibrium

curve solutions is

to within dashed

equation

under

line), to the

threefold:

983

which combined oxygen

good agreement similar

solutions.

experimental

and half-saturation are in

2a and 2b were

with

conditions

All

error,

of the

on the

was obtained data.

obtained

by least-

The values

pressure

(p50

the literature (11).

The signi-

of

= 11 values

Vol. 88, No. 3, 1979

IO

8lOCHEMlCAL

I

,

,

,

,

AND 8lOPHYSlCAL

p---+--’

a A,’

Cl8P 0.6 -

04

A’

-_-+----0.0 0.0a-a -4 ’ ’’ 1’ 0.0 0.0 0.4 0.4

6’

I’-

6”

/A-----’

’

’

i

’

1

0.8 0.8

1

1.2 1.2

’

’

1

1

1.6 1.6

’

I

2.0 2.0

08

log pop (torr)

12

:

1.6

J

log PO* (kvr)

a. Fractional saturation plotted vs. log p (oxygen) for the following samples: HbA solution, 4 g/d1 (n=6); WS solution, 5 g/d1 (n=7); HbS solution #2, 5 g/d1 (n=7); emulsified HbS solution, 13 g/d1 in aqueous phase (n=6) i emulsified HbS solution, 17 g/d1 in aqueous phase (n=5). T = 23.5kO.5, aqueous phase in -05 M PO4, .l M Cl-, .Ol M EDTA, pH 6.952.05. Dashed line is least-squares fit of Adair-type expression to the combined data (see text). b. Fractional saturation plotted vs. log p (oxygen) for the following samples: emulsified HbS solution, 27% in aqueous phase (n=9); emulsified HbS solution #2, 27% in aqueous phase (n=7). Conditions as above. Continuous and long dashed curves are simulations of 27 g/d1 data, and long-short dashed curve is simulation of 17 g/d1 data (see text). Short dashed curve same as in Fig.Za.

(1)

They confirm

the oxygen

equilibrium

solution.

At present

liably

the

of hemoglobin,

both

The results

support

is not

constants

to the analysis

a critical

HbA are

of the method

Adair

essential

(2)

utility

the method

the "interior"

are not

below

Y

:

-

a2 a2-

2.

0.6

4;’ P

Y

0.4

RESEARCH COMMUNICATIONS

However,

below

the oxygen

to estimate these

(8,12,13) binding

re-

quantities

and in reference

observations

measuring

and emulsified

precise

k2 and k . 3

previous

above for

in homogeneous

sufficiently

discussed

HbS concentration,

described

1.

indicating

isotherms

that,

of HbS and

identical.

(3) employed

They show that here

does not

the technique

of emulsification

itself

significantly

2b are data

points

affect

of HbS solutions the oxygen

equilibrium

of

hemoglobin. Shown in Figure at a concentration the other Nonetheless,

data

of 27 g/al. sets

certain

made and conclusions

and do not qualitative

These admit

from data

two emulsified sets

have a larger

of a thorough

and semi-quantitative

drawn.

984

solutions

of Hbs

scatter

quantitative observations

than

analysis. may be

BIOCHEMICAL

Vol. 88, No. 3, 1979

The oxygen

equilibrium

right-shifted

and steepened Gill

lutions.

--et al.

right

smaller

solution. data

curve

shift

recent1

in the

In Figure data

using

also

aggregating

of a model

solution

gated

model,

(polymeric)hemoglobins.

mer has a lower crease

in the overall increasing

However, hemoglobin

monomer,

as polymer. bility

for fractional

oxygen

the partial

pressure

thereby

of monomeric

hemoglobin

ration,and

at pressures

brium

the monomer.

with

of the oxygen In order and the data

equilibrium to obtain points

value will

above the One would curve

saturation

of the oxygen

critical expect

at this

critical

semiquantitative shown it

was necessary

and aggrepoly-

to a deis

present.

the solubility

of hemoglobin

partial

of present

pressure,

the solu-

hemglobin

concent-

no polymer

can exist

in equili-

a discontinuity

in the slope

of the oxygen

between

ii

y, of an

leading

increases

value

agreement

(Method

as the mass

when polymer

to observe

the

the hemoglobin

to the total value

artifacts.

taken

the mass fraction

of

samples,

saturation,

model,

HbS

two sets

to simulate

does the monomer,

be equal

the

(monomeric)

of oxygen

decreasing

At some critical

p, is

to this

than

g/d1

33

by Minton

oxygen

of non-aggregating

oxygen

a

experinmntal

proposed

pressure,

According

affinity

between

attempts

the equilibrium

saturations

in

but

of the hemoglobin

and/or

previously

similar, curve

representing

of HbS at oxygen

of the oxygen

average

conditions,

substantially

HbS and HbA so-

a qualitatively

composition

curves

is

of more dilute

discrepancy

in the

shown

In this

HbS solutions

of the oxygenation

experimental

2b are

1).

y reported

of the quantitative

modifications

of reference

27 g/d1

to those

and steepening

may be due to differences

differences

of the

relative

(8)

The origin

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

the

to make several

pressure.

calculated

additional

curves assump-

tions: (1)

Aggregated

as P approaches-Pcritical

3.

HbS becomes

significantly

saturated

with

oxygen

(yp+0.3)

3

Hofrichter (14) found that polymeric HbS In contrast to this result, binds little or no carbon monoxide over the range of CO concentrations in which monomeric and polymeric HbS coexist in equilibrium.

98.5

Vol. 88, No. 3, 1979

(2) than

The mass fraction

that

calculated

of mononmric globin

hemoglobin

unpublished

that

results,

but

similar We also

obtained

is

in

fair

substantially

only

is

substantially

greater

1 when the concentration less

examination

than

the

total

of the derivation

for

small

agreement

values

hemo-

of the

of f p (Minton,

and emulsions

Hem-O-Scan

with

direct

Instrument

torr).

data

of the

concentrated

is

not well

solubility

equilibrium

Co., curves

of HbA and HbS, which

reproducible

value

14 g/d1 defined

by

determinations

could

of HbS emulsions

the measurement

equilibrium

n = 2.6 and psO z 12-13

Of p50 and the shapes

with

for

American oxygen

This

the oxygen

apparatus

(Hem-O-Scan,

to be approximately

4

to measure

available

or emulsions,

HbS was assumed

experiments.

(15).

reproducible

solutions

not

Silver for

with

well

"equilibrium"

(i.e.,

aggregating)

MD).

with

We

the above data HbS solutions

We suggest

curves

oxygena-

(<12 g/dl)

concentrated

be obtained.

oxygen

blood

Spring,

"dilute"

agreed

However,

of whole

using

obtained

HbS solutions

that

values

using cannot

the be

upon. It

between tional

is

may be valid

of these

attempted

curves

relied

of reference

ce,

of deoxy

conditions

a commercially

(Hill

fp,

(2)

A detailed

it

The solubility

the conditions

under

tion

present,

polymer,

RESEARCH COMMUNICATIONS

observations).

(3) under

equation

cT.

indicates

AND BIOPHYSICAL

of hemoglobin

using

concentration,

equation

our

BIOCHEMICAL

is

clear

oxygenation oxygenation

that

a comprehensive and aggregation

data

obtained

quantitative of HbS will

over

a wide

analysis require,

range

of the relations at a minimum,

of HbS concentrations

addiex-

of the present communicaceeding the solubility of deoxy HbS. The objectives 4. It may be recalled that one of the "dilute" HbS emulsions exhibiting an oxygenation curve indistinguishable from that of non-aggregating hemglobin had a concentration of 17 g/dl. If the solubility of deoxy HbS is really as low as 14 g/al, some fraction of the HbS in a deoxygenated 17 g/d1 solution should be aggregated, resulting in a decrease in the average oxygen affinity. In Figure 2b is plotted a simulated oxygenation curve for a 17 g/d1 HbS solution using the same parameter values that were used to generate one of the simulated curves for the 27 g/d1 solutions. It may be seen that while the oxygen affinity is in fact decreased at low oxygen pressures, the decrease is so small in absolute terms as to be within the scatter of measurements of the oxygenation curve for non-aggregating HbS solutions.

986

BIOCHEMICAL

Vol. 88, No. 3, 1979

tion

have been

HbS provide oxygen

fied

to demonstrate

a realistic

binding

describe

(a)

yet

properties

a technique

for

AND BIOPHYSICAL

that

emulsified

analytically

tractable

of hemoglobin measuring

the

in sickle oxygenation

RESEARCH COMMUNICATIONS

concentrated physical erythrocytes, equilibria

solutions model

for and

of the (b)

to

of the emulsi-

solutions.

ACKNOWLEDGEMENTS We thank J. Kurlansik of the Naval Medical Research Institute, Bethesda, MD, for assistance in the gas chromatographic determination of the solubility of oxygen in emulsifying oil, N.M. Rumen of the Uniformed Services University of the Health Sciences, Bethesda, MD, for making a Hem-O-Scan available to us, N. Li and J. Frankenfeld of Exxon Research and Engineering Co. for generous gifts of SlOON and ENJ-3029, and ICI America, for the gift of the Span 85.

PEFEPENCES 1.

2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15.

Minton, A.P. (1976). In Proceedings of the Symposium on Molecular and Cellular Aspects of Sickle Cell Disease, Hercules, J.I., Cottam, G.L., Waterman, M-R., and Schechter, A-N., eds. (DHEW-NIH, Bethesda, Md.) ~~-257-273. Minton, A.P. (1976) J.Mol.Biol. 100, 519-542. Minton, A.P. (1977) J.Mol.Biol. 110, 89-103. Seakins, M., Gibbs, W.N., Milner, P-F., and Bertles, J.F. (1973) J.Clin.Invest. 52, 422-432. 30, 317-335. May, A. and Huehns, E.R. (1976) Brit.J.Haemotol. Rossi-Bemardi, L., Luzzana, M., Saaja, M., Rossi, F., Perrella, M., and Berger, R. (1975) FEBS Letters 59, 15-19. Winslow, R.M. (1976). In Proceedings of the Symposium on Molecular and Cellular Aspects of Sickle Cell Disease, Hercules, J-I., Cottam, G.L., Waterman, MR., and Schechter, A.N., eds. (DHEW-NIH, Bethesda, Md.) ~~-235-251. Gill, S.J., Sksld, R., Fall, L., Shaeffer, T., Spokane, R. and Wyman, J. (1978) Science 201, 362-364. 47, 1179-1185. May, S.W., and Li, N.N. (1972) Biochem.Biophys.Res.Comm. Harris, J.W. (1950) Proc.Soc.Exp.Biol.Med. 75, 197-201. Antonini, E. , and Brunori, M. (1971) Hemoglobin and Myoglobin in Their Reactions with Ligands, (North-Holland, Amsterdam). Wyman, J., and Allen, D.W. (1954) Rev.Hemat. 9, 155-157. Noble, R.W. (1978).In Clinical and Biochemical Aspects of Abnormal Hemoglobins, Caughey, W., ed. (Academic Press, N-Y.). Hofrichter, J. (1978).In Clinical and Biochemical Aspects of Abnormal Hemoglobins, Caughey, W., Ed. (Academic Press, N.Y.). Ross, P-D., Hofrichter, J., and Eaton, W.A. (1977) J.Mol.Biol. 115, 111-134.

987