Vol. 111, No. 2, 1983
March
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
16, 1983
Pages
675-682
Polymorphic
Phase Preferences
of
Phosphatidic
Acid:
A "P and 2H NMR Study
S.B.
M.J.
Farren,
Hope and P.R.Cullis
Department
of Biochemistry
University
of British
Columbia
Vancouver
V6T lW5 Canada
Received
January
7, 1983
3'P NMR, 'H NMR and freeze fracture
techniques
have been employed
to investigate
the structural
preferences
of dioleoyl
phosphatidic
acid
(DOPA) under various
conditions
of pH and divalent
cation
content.
It is
shown that DOPA increasingly
prefers
the HII organization
below pH 5, and
that low levels
of Ca2+ and Mg2+ (Me2+/DOPA = 0.5) induce HII phase
Higher Mg2+ and Ca2+ levels
(Me2+/DOPA > 1.0)
structure
for pH < 6.0.
induce more complex structures
at pH > 5.0 which may correspond
to interThis work illustrates
the
mediates
between lamellar
and HII organization.
utility
of 2H NMR techniques
in conjunction
with suitable
'H labelled
lipids
to provide
structural
information
on lipid-water
systems, and
for determinations
of lipid
organization
suggests
that "P NMR techniques
can be applied
to advantage
in PA containing
membranes.
Phosphatidic
is
a minority
the
acid
lipid
(PA),
species
more interesting.
(PI)
[2,3,4]
processes
[4].
standing
most
cal
for
hexagonal
intriguing
may play
(HII)
given
dynamic
Such information
is
roles
in membrane
transport
the
PA (DOPA) under
is
structural
various
preferences
conditions
of PA as a Ca2+
role
influence
not
of factors
currently
indicating
in biologithat
unsatu-
at pH 6 adopts
Such behaviour
"non-bilayer"
and fusion
the
is
lipid
structure
processes.
of aqueous
dispersions
of pH and monovalent
of
and
0006-291X183
675
such
available,
obtained
(PC))
that
so called
an under-
of Mg 2+ [6].
[7,8]
one of
of PA requires
a report
suggestions
however,
fusion
egg phosphatidylcholine
presence
is,
of PA in the
of PA such as are
to this
synthesis,
of PA in membrane
roles
recent
Here we examine
dioleoyl
in the
role
of PA and the
species
from
It
and the possible
of these
The exception
phase
[l]
properties
unsaturated
egg PA (derived
membranes.
as a potential
and pH.
in phospholipid
due to the production
An understanding
membranes.
rated
as well
cations
notably
is
effect
of the physical
as divalent
in biological
This
phosphatidylinositol
ionophore
a key intermediary
$1.50
Copyright 0 1983 by Academic Press, Inc.
,411rights of reproduction in a~> form reserved.
BIOCHEMICAL
Vol. 111, No. 2, 1983
divalent
cation
deuterium
It
is
(*Ii)
shown
3.7.
results
is
in the
above
The presence
of Mg
to a particular
Higher
of DOPA which
of the acyl
absence
cations
2+
at lower
the
only
which
cannot
(Cll-'H
HII
DOPA).
the bilayer
(HII)
present
to DOPA ratios
(R = 1.0)
is
the hexagonal
organization
on pH and significant
of cations
chains
DOPA adopts
pH values
or Ca2+ at cation
dependence
levels
of divalent
and is
and *H NMR spectra
phase
a version
RESEARCH COMMUNICATIONS
at the Cl1 position
preferred
in a sharper
in "P
We employ
pH 6, whereas
increasingly
at pH = 6.0.
results
labelled
that
organization
phase
concentration.
AND BIOPHYSICAL
at pH =
R of 0.5
phase
at pH values
be unambiguously
formation
of 5 and higher
assigned
structure.
MATERIALS
AND METHODS
1,2 - (11,ll
dideutero)
dioleoyl-sn-glycero-3-phosphorylcholine
(Cll-'H
DOPC) was synthesized
employing
oleic
acid 'H labelled
at the
Cl1 position
as described
elsewhere
(9).
Dioleoyl
phosphatidic
acid
(Cll-*H
DOPA) was prepared
from the Cll-'H
DOPC employing
standard
procedures
utilizing
phospholipase
D [lo] and was purified
by carboxymethylcellulose
column chromatography
[lo].
All
lipids
were more than 99% pure as
indicated
by two dimensional
thin layer chromatography.
Samples for "P and *H NMR studies
were prepared
from 65 umol
phospholipid.
The lipid
was hydrated
in a buffer
made up from *H depleted
H20 containing
20 mM HEPES (pH = 8.0),
100 mM NaCl, 2 mM EDTA by vortex
mixing.
Divalent
cations
were added in appropriate
amounts from a 1 M stock
solution,
whereas
the pH of the samples was adjusted
by adding aliquots
of
100 mM HCl or NaOH. Following
addition
of cations,
or adjustment
of pH, the
sample was subjected
to an exhaustive
freeze-thaw
protocol
(3 freeze-thaw
cycles employing
liquid
N2) and the pH was subsequently
checked and further
This procedure
was found to be necessary
as some
ad justed
if necessary.
variability
in the results
was observed
in the absence of such a protocol,
which likely
arises
from a non-equilibrium
pH or cation
distribution
in the
lipid
sample.
"P and *H NMR studies
were performed
on a Bruker WP 200 Fourier
Transform
NMR spectrometer.
"P NMR experiments
were performed
at an
observation
frequency
of 81.0 MHz and accumulated
free induction
decays were
obtained
from up to 500 transients
employing
a 15 psec 90" radio frequency
pulse,
0.8 set interpulse
time, gated proton
decoupling
and a 20 KHz sweep
width.
An exponential
multiplication
corresponding
to a 50 Hz line broadening was applied
prior
to Fourier
transformation.
The 'H NMR experiments
were oerformed
at 30.7 MHz. Accumulated
free induction
decays were obtained
from up to 20,000 transients
employing
a 16 usec 90' pulse,
a 0.01 set
interoulse
time and a 31.25 KHz sweep width.
An exponential
multiplication
corresponding
to a 100 Hz line broadening
was applied
prior
to Fourier
transformation.
It may be noted that due to "dead time" limitations,
the broader
'H
In particular
the classic
"solid
state"
NMR spectra
obtained
are distorted.
'H NMR pattern
is not observed
for broader
bilayer
*H NMR spectra.
In
agreement
with other workers
(11) subject
to similar
instrumental
limita676
Vol. 111, No. 2, 1983
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
tions,
two peaks corresponding
to the two major peaks of the solid
state
pattern
are then observed,
but other detail
is lost.
Such difficulties
can
be overcome by the use of the quadrupolar
echo pulse sequence
[12], however
our spectrometer
does not have the required
pulse programming
capabilities.
Freeze-fracture
studies
were performed
on a Balzers
BAF 301 apparatus,
and the samples contained
25% (vol/vol)
glycerol
as a cryoprotectant.
Replicas
were viewed employing
a Phillips
400 electron
microscope.
RESULTS AND DISCUSSION
In previous
techniques
work
to determine
and biological
extended
the
(for
the
diesterified
value
and markedly
or sign
different
structural
such lipids
is
transitions
[13-151.
ease of synthesis
[9]
exhibited
label
by this
The "P
various
lineshapes
organization,
is
are
exhibit
amorphous
indicates
KHZ.
This
would
be most consistent
analogy
with
DOPA, as the
convenient
quadrupolar
in liquid
type
crystalline
of 8.7 KHz,
component,
previous
with
results
obtained
from
1.
of asymmetry
phase
observed,
is
somewhat
are
less
8 KHz.
with
the
the occurrence
This
677
(HII)
Cl1 position
splitting
due to
(~8 KHz)
phospholipid
systems.
DOPA at
with
a lamellar
(Fig.
sole
narrower
At pH 5 the
- one exhibiting
lineshape
for
other
observed
phase
is
*H NMR
a quadru-
splitting
component
conclusion
than
At pH 5
At pH 8 and 6 the
a quadrupolar
of HII
4(a)).
at pH 3.7 the
ambiguous.
of two components
[15].
of
whereas
but
is
'H NMR spectrum
results
approximately
which
the
associated
and another
(CSA)
At pH 8 and 6, the 31P NMR
freeze-fracture
is
anisotropy
Cll-'H
to
in a
investigated
bilayer
in Fig.
with
result
of bilayer-hexagonal
and the
the presence
splitting
latter
We therefore
can be
in contrast
could
shift
DOPA at the
AC is
polar
This
to label
lineshape
splitting"
a monoester
We chose
The 'H NMR spectra
"quadrupolar
is
chemical
to the occurrence
NMR
in model
such techniques
phospholipids.
phosphorus
'lP
of phospholipids
that
group
NMR spectra.
of the HI1
phospholipids.
spectrum
of the
illustrated
the
consistent
characteristic
obvious
phosphate
NMR and 'H NMR spectra
pH values
a somewhat
not
of 'H labelled
sensitive
7) we have employed
preferences
of other
"P
preferences
is
as the
phosphate
see ref.
structural
It
membranes.
to PA systems,
different
review
structure
reinforced
of 2.9
at pH 3.7
by
by
Vol. 111, No. 2, 1983
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
C,,-*H
C,,-?H
DOPA
DOPA
1
3’P
PH
+ Mg’+
: 0.5
PH
__n7A &A
01
40
-40
0
@pm)
4
that
is
02
-4
40
0
(PPW
-40
4
(k&J
Fig.
1.
81.0 MHz "P NMR and 30.7 MHz 'H NMR spectra at 25°C obtained
from aqueous dispersions
of dioleoyl
phosphatidic
acid (DOPA) 'H
labelled
at the Cl1 position
of the acyl chains (Cll-'H
DOPA).
The "P NMR and 'II NMR spectra were obtained sequentially
from
the same sample.
The pH was lowered by addition
of appropriate
aliquots
of 0.1 M HCl, and the sample was then subjected to a
rigorous
freeze-thaw
protocol
(see Methods) and the pH determined
subsequently.
For other details
of sample preparation
and data
manipulation
see Methods.
Fig.
2.
81.0 MHz "P NMR and 30.7 MHz 'H NMR spectra obtained at 25'C
for Cll-'H labelled
dioleoyl
phosphatidic
acid (cl~-~H D;JA>
in3u+ated at various pH values and in the presence of Mg at a
to Cll-*H DOPA molar ratio of 0.5.
The sample was
W
initially
hydrated in the N'MR buffer (see Methods) at pH 8, the
Mg2+ added (chloride
salt, 100 ml4 stock solution)
and the sample
was then subjected to a rigorous
freeze thaw protocol
(see Methods).
The pH was subsequently
checked and, if necessary, readjusted.
Any
such adjustments were again followed by the freeze-thaw
procedure.
The "P and 'H N'MRspectra were obtained sequentially
on the
same sample at a given pH and the pH was then lowered (or raised)
by addition
of aliquots
of .l M HCl (.l M NaOH). For other details
of sample preparation
and data accumulation
see Methods.
freeze-fracture
are
0
(kH4
in the
data
region
(unsaturated)
a net
negative
(see
Fig.
of 3.5 and 8 [16]
PA can adopt
charge
The pK's
4(b)).
and thus
the HII
on the headgroup.
678
of the PA phosphate
the
phase
results
under
This
of Pig.
conditions
contrasts
group
1 suggest
where
with
-4
there
the behaviour
BIOCHEMICAL
Vol. 111, No. 2, 1983
C,,-2H
AND BIOPHYSICAL
DOPA
RESEARCH COMMUNICATIONS
+ Cd
-?.A.5.3
;
40
Fig.
3.
so small
phosphatidylserine
that
be accommodated
some degree
within
The influence
istics
of the
observed
and implies
illustrated
structure
in Fig.
sample.
clearly
resolved.
The pH 6 spectrum
appears
bilayer
and HII
organization.
These
by the
corresponding
that
quadrupolar
splitting
PA headgroup
repulsion
can
and 'H NMR character-
for
shown in Fig.
As
give
a clear
At pH 8 a bilayer
an H II
supported
"P
2 and 3.
at pH 5.5 and below
strongly
the
electrostatic
the 31 P NMR spectra
of the
that
organization.
whereas
the
(kYiZ)-4
of inter-headgroup
the HII
of Mg2+ at R = 0.5
phase
[17]
of Mg2+ and Ca2+ on the
of DOPA is
presence
4
-40
81.0 MHz "P NMR and 30.7 MHz 'H NMR spectra obtained at 25°C
for dioleoyl
phosphatidic
acid 'H labelled
at the Cl1 position
(C11-'H DOPA) in the presence of Ca2+ (Ca*+/DOPA = 0.5, mol/mol)
The Ca2+ was added to the hydrated DOPA at
at various pH values.
pH 8 and the sample was then subjected to the same procedure as
indicated
in the legend to Fig. 2.
of (unsaturated)
is
If&)
structural
phase
"P
679
indication
lineshape
is
NMR lineshape
to represent
a mixture
identifications
'H NMR spectra.
the hilayer
2 in the
component
is
of
are
It
may be noted
at pH 8 and 6 is
Vol. 111, No. 2, 1983
AQ = 11.1
acyl
KHz,
chain
indicating
region
the major
phase
BIOCHEMICAL
for
effect
is
although
structure.
phase
5.3,
organization.
but
pH values
presumably
The origin
of the central
indicates
intermediate
signal
Mg
becomes
a shape which
(results
not
2+
that
of the
state"
structure
samples
cylinder
packed
type
HII
fracture
correspond
diaries
face
the
to ascertain
freeze-fracture
studies
in general,
lamellar
diameter
and thus
lamellar
highly
flattened
and HII
a broad
featureless
were
and give
phase
structure
2+
Mg and
both
performed
4(c).
This
cylinders
rise
of
4(b))
DOPA
to R
(short)
of closely
and the
may
as interme-
Such structures
to cylinders
'H NMR spectrum.
for
structure
proposed
[18].
nature
The results
regions
to Fig.
'H
the
corresponding
regions
disorganized.
lipid
and assumes
more directly
(compare
680
NMR
show the classical
however
organizations
characteristics,
motion
the "P
for
it
phases.
or HII
No extended
are apparent
a
however
allowing
Mg2+ and Ca2' concentrations
2+
are presented
in Fig.
for Ca
to the extended,
basically
bilayer
at pH
with
at pH 6 and above,
In order
cylinders
phase
a AQ of e8 KHz.
(R = 1.0)
visible.
of HII
coexists
and HII
but do not
are
bilayer
to ascertain,
'H NMR spectra
of structure
defined,
the HII
exhibits
bilayer
with
unequivocally,
between
reflect
intensity
to interpret
is,
HII
indicative
component
concentrations
become narrower
formed,
phase
that
more compli-
well
in a structure
in the
in
not
clearly
is difficult
be identified
pattern.
= 1.0 and the results
appears
predominantly
which
of lipid
Similarly,
at pH = 7.0 for
are difficult
2+
also
origin,
obtained
and Ca
shown).
1 it
somewhat
are
is
central
component
not
in the
the pH at which
on DOPA is
spectrum
a broad
of bilayer
Ca2+ at pH 6 and above
NMR "solid
to raise
NMR spectra
the
very much reduced
could
order
to Fig.
as indicating
the presence
between
At higher
"P
however
component,
clearly
is
The *Ii NMR spectra
at higher
appreciable
In comparison
R = 0.5)
be interpreted
At pH = 5.3
induces
RESEARCH COMMUNICATIONS
to pH = 6.
Above pH 6 the
could
2+
DOPA.
of Ca2+ (at
3).
they
bilayer
induced
The influence
(Fig.
the Mg
of Mg2+ at R = 0.5
structure
cated
that
AND BIOPHYSICAL
have
of varying
Vol. 111, No. 2, 1983
Fig.
4.
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Freeze-fracture
micrographs
of dioleoyl
phosphatidic
acid
(DOPA)
under
varying
conditions
of pH and divalent
cation
content:
(a)
DOPA hydrated
in buffer
at pH = 8.0 (lamellar
structure);
(b) DOPA
hydrated
at pH =
;;~;;,‘;‘;,‘$,;;~
yrated
8t pH = 8, to ,,,&“,$;f:f~~‘,f$
Ca*+/DOPA
ratio
of 1.0 (mol/mol),
and the pH adjusted
to pH = 7.0,
indicating
the presence
of short
tube-like
structures
oriented
in a
random
fashion.
The direction
of shadowing
is indicated
by the
arrowhead
in the lower
left
hand corner.
Magnification
spprox.
115,500.
681
Vol. 111, No. 2, 1983
In summary,
reduced
below
BIOCHEMICAL
DOPA progressively
5.3,
and the
levels
increases
Higher
Ca2+ and Mg2'
an intermediate
the presence
a recent
AND BIOPHYSICAL
adopts
levels
between
HII
result
bilayer
freeze-fracture
study
organization
as the pH is
is observed
in a disorganized
and HII
of Mg2+ and Ca2+ is
phase
of Mg 2+ and Ca2+ at lower
presence
the pH at which
the HII
RESEARCH COMMUNICATIONS
phases.
in broad
[19]
structure,
The behaviour
agreement
which
appeared
(R = 0.5)
to 6.0.
possibly
of DOPA in
with
the
results
of
while
this
work
was in
preparation.
ACKNOWLEDGEMENTS
P.R.C.
This
work
is a Scholar
was supported
of the Medical
by operating
Research
and major
Council
equipment
(MRC) of Canada.
grants
from
the MRC.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Michell,
R.H. (1975) Biochim.
Biophys.
Acta 415, 81-147.
Salmon, D.M. and Honeyman, T.W. (1980) Nature 284, 344-345.
Putney,
J.W., Weiss, S.J.,
van de Walle,
C.M. and Haddas, R.A. (1980)
Nature 284, 345-347.
Chem.
Tyson, C.A., Zande, H.V. and Green, D.E. (1976) J. Biol.
251, 1326-1332.
Koter,
M., de Kruijff,
B. and van Deenen, L.L.M.
(1978) Biochim.
Biophys.
Acta 514, 155-163.
Papahadjopoulos,
D., Vail,
W.J., Pangborn,
W.A. and Poste, G. (1976)
Biochim.
Biophys.
Acta 5,
265-283.
Cullis,
P.R. and de Kruijff,
B. (1979) Biochim.
Biophys.
Acta
2,
399-420.
Cullis,
P.R., de Kruijff,
B., Hope, M.J.,
Nayar, R. and Schmid, S.L.
(1980)
Can. J. Biochem. 58, 1091-1100.
Farren,
S.B. and Cullis,
P.R. (1982) Chem. Phys. Lipids
(submitted).
Comfurius,
P. and Zwaal, R.F.A.
(1977)
Biochim.
Biophys.
Acta
488, 36-42.
Brown, M.F. and Seelig,
J. (1977) Nature 269, 721-723.
Davis, J.H.,
Jeffrey,
K.R., Bloom, M., Valic,
M.I.
and Higgs, T.P.
(1976) Chem. Phys. Letters.
42, 390-394.
Gally,
H.U., Pluschke,
G., Overath,
P. and Seelig,
J. (1980)
Biochemistry
19, 1638-1643.
Cullis,
P.R., Farren,
S.B. and Hope, M.J. (1981) Can J. Spectros.
26, 89-97.
Tilcock,
C.P.S.,
Bally,
M.B., Farren,
S.B. and Cullis,
P.R. (1982)
Biochemistry
21, 4596-4601.
Trauble,
H. and Eibl,
H. (1974) Proc. Nat. Acad. Sci. USA 2,
214-219.
Hope, M.J. and Cullis,
P.R. (1980) Biochem.
Biophys.
Res. Comm.
92, 846-852.
R. and Verkleij,
A.J. (1981) Biochim.
Biophys.
Acta
van Venetie,
645, 262-269.
Verkleij,
A.J.,
de Maagd, R., Leunissen-Bijvelt,
J. and de Kruijff,
B.
(1982) Biochim.
Biophys.
Acta 684, 255-262.
682