NNNNNNNNNNNNNNNNNNNNN

Willi/NM HIWif/l/llii/Wlm/lfll

0583 8126

TNEBEI fhfllum_

W

LIBRA R Y
Michigan State
University

_._ ______ 1

This is to certify that the

thesis entitled

AN ELECTRON SPIN RESONANCE STUDY OF THE PLASMA
MEMBRANE OF THE HALOTOLERANT ALGA,

DUNALIELLA PRIMOLECTA
presented by

 

Donna Rae Fontana

has been accepted towards fulfillment
of the requirements for

M.S- degree in_Bioph¥.s.i_cs

 

1/4: Milly

Mfim:

Date 017 Du“f. [7480

 

07839

 

 
  

a... 1%
(l’fl-‘k\\ 'A

 

"=1“;

 

  
  

 

w:
25¢ per day per item

RETUMIN LIBRARY MATERIALS:

Place in book return to remove
charge from circulation records

 

 

AN ELECTRON SPIN RESONANCE STUDY OF THE PLASMA
MEMBRANE OF THE HALOTOLERANT ALGA,
DUNALIELLA PRIMOLECTA

 

by

Donna Rae Fontana

A THESIS

Submitted to
Michigan State University
in Partial Fulfillment of the Requirements

for the Degree of

MASTER OF SCIENCE

Department of Biophysics
June 1980

ABSTRACT

AN ELECTRON SPIN RESONANCE STUDY OF THE PLASMA
MEMBRANE OF THE HALOTOLERANT ALGA,
DUNALIELLA PRIMOLECTA

 

by

Donna Rae Fontana

Dunaliella primolecta was studied as a model for

 

salt tolerance in higher plants. Electron spin resonance
was used to monitor the in vivo fluidity of the plasma
membrane of this alga and the fluidity of alga lipid
extracts.

It was found that algae adapted to and suspended in
higher salt concentrations had more rigid membranes. Both
biochemical modification and a physical interaction between
the Na+ and lipids were implicated.

When the fluidity of the plasma membrane and lipid
extracts were determined as a function of temperature, two
or three events were observed. There was always an event
at 9-14°C and 39-43°C, and these were interpreted as the
onset and completion of the lipid phase transition. This

implies that 2; primolecta exists with its membrane.in the

 

mixed state. With higher NaCl concentrations, a third
event occurred around 20-22°C. This event was a lipid-lipid

interaction and not related to adaptation.

ACKNOWLEDGMENTS

I would like to thank Ken Leonards, Steve Briggs,
and Charles Caldwell for their helpful discussion and
advice. I would also like to thank the above and Jack Jen,
Kathy Strong, and Richard Allison for making my daily
experience in the laboratory a more pleasant one.

I am grateful to Dr. A. Haug for the opportunity to
work in his laboratory, for introducing me to Dunaliella,
and for his guidance.

The other members of my guidance and thesis
committees were helpful and I would also like to thank them.
.These members include Dr. A. El-Bayoumi, Dr. D. T. A.
Lamport, Dr. E. J. McGroarty, and Dr. H. Ti Tien.

Lastly, I am grateful to my husband Tom for his
stimulating conversation and his help with my figures.

This work was supported by DOE contract DB-ACOZ-

76ER01338.

ii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . .
LIST OF FIGURES o o o o o o o o o o o o 0 0 0 0 0

INTRODUCTION

REVIEW OF THE LITERATURE

Permeability of the Dunaliella Plasma
Membrane to Sodium Chloride. . . . . .
Volume Regulation in Dunaliella . . . . .
ATP-Driven Cation Exchange in Dunaliella.
Summary of Dunaliella-Salt Relations
Literature . . . . . . .
Fatty Acid Composition of Dunaliella
primolecta . . . . . . . .
Effect of Membrane State on Membrane
Permeability . . . . . . . .
Effect of Salt on Membrane Fluidity . . .

 

MATERIALS, METHODS, AND RESULTS

Organism, Medium, and Growth. . . .
ESR Measurements and Analysis . . .
Whole Cell Spin—labelling Procedure
Cellular Location of the Spin Label
ESR Experiments with Adapted Algae.
ESR Experiments with Non-Adapted Algae. .
ESR Experiments with Lipid Extracts . . .

DISCUSSION . . . . . . . . . . . . . . . . . . .
CONCLUSION . . . . . . . . . . . . . . . . . . .

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . .

iii

Page

iv

15
17

18
19

20
28

33
34
40
43
SO
58
7O

84
91

95

LIST OF TABLES

Fatty Acids of Dunaliella primolecta .

 

Percent Increase in Low Field Peak Height
upon Illumination as a Function of
Temperature .

Results from ESR Experiments on Non-Adapted
Q; primolecta . . . . . . . . . . .

 

Break Points Determined from ESR Experiments
on Lipid Extracts of D; primolecta.

 

Hyperfine Splitting (2T11) at Growth
Temperature Determined from ESR Experi-
ments on Whole-Alga Lipid Extracts of
D. primolecta . . . . . .

 

iv

Page
21

47

69

78

82

1.

LIST OF FIGURES

A representation of Dunaliella primolecta
traced from a drawing found in reference 11

Generation time of Q; primolecta ase:function
of the sodium chloride concentration in the
growth medium to which the algae were
adapted for at least thirty generations . . .

 

Representatives of the ESR spectra obtained
with in vivo labelling of the plasma
membrane of Q; primolecta with S-nitroxy-

 

stearate C O O O O O O O O O O O O O C O O O 0

Structures of S-nitroxy-stearate(2-(3-carboxy-
prOpyl)-4, 4-dimethyl-2-tridecyl-3-oxazolid-
inyloxyl) and TEMPO (2,2,6,6-tetramethyl-
piperidino-oxyl). . . . . . . . . . . . . . .

Spectra 0f 5"nitJ-‘CDXY-stearate labelled D.
primolecta before and after illumination. . .

 

Plot of the hyperfine splitting (_2T ) and
calculated order parameter as a unction of
temperature obtained with in vivo labelling
of the plasma membrane of 2; primolecta
adapted to and studied in growth medium
containing 2% NaCl (w/v)- - - - - . - . . - .

 

Plot of the hyperfine splitting (2T11) and
calculated order parameter as a function of
temperature obtained with in vivo labelling
of the plasma membrane of Q; primolecta
adapted to and studied in growth medium
containing 24% NaCl (w/v) - - - . - - - . . .

Plasma membrane events detected as break points
as a function of the sodium chloride concen-
tration in the growth medium to which the
Q; primolecta were adapted for at least 30
generations . . . . . . . . . . . . . . . .

 

Page

36

39

42

46

52

54

57

10.

ll.

12.

13.

14.

15.

Page
Plasma membrane fluidity, as determined by 2T11,
as a function of the sodium chloride con—
centration in the growth medium to which the
‘2; primolecta were adapted for at least 30
generations . . . . . . . . . . . . . . . . . 6O

 

Plot of the hyperfine splitting (2T11)and
calculated order parameter as a function of
temperature obtained with in vivo labelling
of the plasma membrane of Q; primolecta
adapted to growth medium containing 4% NaCl
(w/v) and studied in medium containing 8%
NaCl (w/v). . . . . . . . . . . . . . . . . . 63

 

Plot of the hyperfine splitting (2T11)and
calculated order parameter as a function of
temperature obtained with in vivo labelling
of the plasma membrane of D. primolecta
adapted to growth medium containing 8% NaCl
(w/v) and studied in medium containing 4%
NaCl (w/v). . . . . . . . . . . . . . . . . . 65

 

Plot of the hyperfine splitting (ZTifl and
calculated order parameter as a function of
temperature obtained with in vivo labelling
of the plasma membrane of D. primolecta
adapted to growth medium containing 8% NaCl
(w/v), washed four times in a sucrose: NaCl
solution, and studied in medium containing
4% NaCl (w/v) . . . . . . . . . . . . . . . . 68

Developed thin layer chromatography plate which
has been spotted with the neutral component
of the whole alga lipid extract; Fraction I
is the neutral lipids minus Fraction II
which was the green band eluted from the
silicic acid column. The shaded spots
react positively with sterol-sensitive
reagents, . . . . . . . . . . . . . . . . . . 72

Plot of the hyperfine splitting (2Tlp and
calculated order parameter as a function
of temperature obtained with lipids,
extracted from D. primolecta adapted to
medium containing 2% NaCl (w/v), and
suspended in distilled water- - - - - - - - 75

 

Plot of hyperfine splitting (2T1 )as a function
of temperature obtained with ipids,
extracted from D. primolecta adapted to
medium containing 2% NaCl (w/v), and
suspended in medium containing 12% NaCl
(w/v)............... .... 77

 

Vi

Page
16. Plot of the hyperfine splitting (2Tll) as a
function of temperature obtained with lipids,
extracted from Q; primolecta adapted to medium
containing 2% NaCl (w/v), and suspended in
medium containing 12% NaCl (w/v). . . . . . . . 81

vii

INTRODUCTION

Approximately 70% of the earth's surface is ocean.
Of the remaining area, one third is arid or semi-arid and
half of that has high saline soils (1). This salinity
becomes an agricultural problem when there is insufficient
transport of salt away from the area occupied by plant
roots. This is common in irrigated areas but can also
occur in non-irrigated crOplands and rangelands (2). With
the growing world population, the need to increase food
production in these areas has risen. Often the cost of
reclaiming these lands is too high or quality water for
leaching and irrigation is unavailable. As a result, there
has been increased interest in the mechanisms of salt
tolerance in plants.

High salinity poses a complex problem. It lowers
the water potential of solutions which can affect the
hydration of proteins and lipids suspended in them (3).
There is also an abundance of various ions which can exert
their own specific effects. As expected, this complex
challenge induces a wide variety of responses at all levels
of plant organization (4,5). A few of these responses are:
1. an increase in succulence; 2. inhibition of differ-

entiation; 3. changes in number and size of stomata; and

2

4. changes in the diameter of the stem. The effects of
high salinity differ with each type of plant. Poljakoff—
Mayber and Gale (6) attributed the variety of responses of
different plant species and varieties to salinity to
differences in: 1. their ability to exclude or compart-
mentalize salt; 2. their ability to adjust osmotically;
3. the stability of their membranes and macromolecules to
a high ionic concentration; 4. the ability to generate
compatible solutes, such as kinetin and glycerol, which
are capable of stabilizing the macromolecules and finally;
5. "the ability to carry out other adaptive modifications;
at the lowest possible cost to overall plant growth".

Studying this complex phenomenon, salt tolerance,
has proven to be difficult. The high level of organization
within a plant makes it almost impossible to separate
direct responses. Many of these responses to salt stress
intimately involve the plasma membrane. The ability to
exclude salt, the stability of the membrane when subjected
to osmotic shock, and the possible role of the plasma
membrane as a transducer of an environmental stress are all
related to the physical properties of the plasma membrane.
The study of the physical properties of higher plant
membranes, in vivo, is very difficult because of the
presence of the cell wall. Removal of the wall may alter
these properties (7). A model system is needed. It
should be a simple plant which is salt tolerant and yet is

amenable to study.

3

Dunaliella is a unicellular, wall-less green alga
which possesses a remarkable ability to tolerate a wide
range of salt concentrations. All species tested seem to
require a small amount of sodium, at least 30 mM (8,9),
but some are able to grow in saturated brine (10). These
characteristics make Dunaliella a possible model for salt
tolerance in higher plants.

The wall-less nature of Dunaliella suggested that it
be used to examine the relationship between the physical
properties of the plasma membrane and salt tolerance. The
following study examines the physical state of the plasma
membrane of Dunaliella primolecta with electron spin
resonance. The results are discussed in terms of salt

tolerance.

REVIEW OF LITERATURE

Dunaliella is a wall-less unicellular alga belonging
to the phylum ChlorOphyta and the order to Volvocales
(Figure 1). Its remarkable ability to tolerate a wide
range of NaCl concentrations has attracted the attention
of many researchers whose investigations have often
produced contradicting results.

Permeability of the Dunaliella Plasma Membrane to Sodium
Chloride

When studying salt tolerance, a basic question which
is often asked is whether the plasma membrane is permeable
to the sodium and the chloride in the medium. This
question was addressed by Marre‘ and Servattaz in 1959

during their studies on Dunaliella salina (12). Using

 

cryoscopic, conductivity, and radioactive tracer measure-
ments, they concluded that the plasma membrane of Q; salina
was highly permeable, especially to Na+ and Cl-. In 1965,
Trezzi et al. showed, with electron microscopy, the rapid
swelling or shrinking which occurs when 2; salina is
subjected to osmotic stress (13). These changes were
reversible. From this study, they concluded that the
membranes of Dunaliella are semipermeable to water and ions.

Three years after the study of Trezzi et al.,

FIGURE 1

A Representation of Dunaliella primolecta

 

Traced from a Drawing Found in Reference 11

STIGMA/

 

 

KFLAGELLA

 
 
 

-\GRANULES

C)
0% Y3

 

     

7
Johnson et al. reported that in vitro CO2 fixation and
enzymes extracted from D;_viridis were inhibited by NaCl
concentrations far less than those of the growth medium
(14). Some of the enzymes tested were phosphoribulose
isomerase, glucose-6-phosphate dehydrogenase and phospho-
glucose isomerase. They found that salt sensitivity was
not dependent on the NaCl concentration of the growth
medium and that ions such as Cs+, K+, Li+, and Ca2+ had
similar inhibitory effects. Preparing the extracts in NaCl
did not affect the inhibition, but dialysis showed that

glucose-G-phosphate dehydrogenase did have a sodium

requirement. The following year Jokela reported that, in

 

2; tertiolecta extracts, glucose-6-phosphate dehydrogenase
and perphosphatase were also sensitive to concentrations

of NaCl equivalent to those in the growth medium (15).

Again the concentration of NaCl in the medium during growth
did not affect the level of sodium sensitivity. Ben-Amotz
and Avron, 1972, demonstrated inhibition of lactate dehydro-
genase and glucose-6-phosphate dehydrogenase by sodium
chloride in cell-free preparations of 21.22523 (16). In
1974, Borowitzka and Brown showed that glucose-6-phosphate
dehydrogenase and glycerol dehydrogenase extracted from Q;

tertiolecta and Q; parva are similarly inhibited by an

 

equimolar concentration of K+ and Na+ which had an ionic
strength far less than that of the growth medium (17).
Balnokin et al. reported in 1979 that 02 evolution in

'isolated chloroplasts' of Q; salina and D; maritima showed

8
maximum activity at 0.1 - 0.2 M NaCl (18). This value was
independent of and less than the NaCl concentrations in
the growth media. The above results describing the sodium
sensitivity of isolated Dunaliella enzymes suggest that
the Na+ is excluded from the alga.

In 1969, M. Ginzburg published the results of a
series of experiments with D; 23533 which demonstrated that
sucrose and inulin (mol. wt. 5 000-5 500) could easily
permeate the plasma membrane (19). Both of these molecules
could equilibrate with half of the cell volume in less than
one minute. Since the chlorOplast in Dunaliella occupies
half of the cell volume, it was postulated that the plasma
membrane contains large pores which are not found in the
chlorOplast membrane. Later that year, Ben-Amotz and B.
Ginzburg showed that 2; 22512 is permeable to Tris and
phosphate buffers as well as DCMU (3(3,4-dichlorophenyl)-l,
l—dimethylurea) (20).

With permeability studies suggesting that the
membrane is leaky to sodium and activity studies suggesting
that enzymes and photosynthetic processes could not function
under conditions of high sodium concentration, the next
step was to actually measure the intracellular sodium
concentration.

The attempts to measure the internal Naf concentra-
tion of Dunaliella were perhaps best summed up by Wegmann
when he stated, "I also have some measurements from my lab

on the cell Na+ concentration. Different methods give

9
concentrations ranging from 10 mM to 2 M inside the cell"
(21). The following is a summary of the published attempts
to determine intracellular sodium and in some cases
potassium. In most of the attempts, the ions were measured
with atomic absorption. The summary below describes how
intracellular ions were separated from the extracellular
ones, as well as giving the concentration of the sodium in
the medium.

1969 by A. Jokela with D; tertiolecta (15)

ions separated by centrifugation of the
algae using blue dextran to measure the
remaining intercellular volume

 

medium concentration: NaCl 1.5% to 15%
(w/v) and KCl 0.38% (w/v)

measured intracellular concentrations:
NaCl 2.45% to 4.35% and KCl 0.57% to 1.0%

intracellular salinities increase with
increasing extracellular Na+

1973 Latorella and Vadas with 2; tertiolecta (22)

spun cells with no Na+ present in the wash;
then broke the cells and measured released
Na

medium concentration of Na+: 0.02 M to 3.0 M

measured intracellular Na+: 1.75 meg/108 cells
to 2.75 meg/108 cells

intracellular salinities almost constant at
extracellular salinities varying between
0-5 M and 2.5 M

1974 Borowitzka and Brown with D; tertiolecta and
D; viridis (l7)
centrifugation of algae using blue dextran to
measure intercellular volume

unable to measure intracellular Na+ because
of exchange during centrifugation and large
errors associated with corrections-for
extracellular Na+

measured 20-40 fold uptake of K+

10
1978 Gimmler and Schirling with Q; parva (23)

rapid centrifugation of the algae through an
inert layer of silicon oil with tritiated
water and l4C-sorbitol used to calculate
intracellular space

medium concentration of Na+: 1.4 M to 0.75 M
intracellular Na+: 0.59-0.8 N and 0.26-0.3 N,
respectively
1978 B. Ginzburg with D; parva (24)
technique not described

medium concentration of Na+: 2.0 M and of
K+: 0.02 M

intracellular concentration of Na+: 0.7-l.4
M and of K+: 0.1—0.3 M

1980 Balnokin et a1. with Q; salina and Q; maritima
(18)
measured medium Na+ before and after cell
rupture

medium concentration of Na+: 1 M to 3 M for
2; salina and 0.35 M to 1.3 M for Q;
maritima

intracellular Na+: for both 0.15 to 0.2 g
eq/liter

With external NaCl concentrations greater than
3 M for D; salina and 1.3 M for D; maritima,
the internal sodium concentration increases.
The most tolerant alga was able to maintain
its intracellular Na+ over a wider range.
Some of the above studies suggest that the internal Na+
level is nearly constant with varying external concentra-
tions of sodium (22, 18). Others suggest that the intra-
cellular level of Na+ increases as the alga is exposed to
more extracellular sodium (15,23,24). With the demonstra~
tion of in vitro sensitivity of the enzymes of Dunaliella
to NaCl and the permeability of the plasma membrane to

large molecules such as inulin, the question of the effec-

tiveness of the plasma membrane as a barrier to sodium still

11

remains open.
Glycerol as a Compatible Solute

Organisms which are able to grow under conditions
of low water activity usually contain what has been termed
a 'compatible solute'. A compatible solute is a substance
which is tolerated in high concentration and can act as an
osmoregulator, protector of enzyme activity and occasionally
as a food source. Examples of compatible solutes are K+,
C1-, and KCl in halophilic bacteria, sucrose in yeast, and

betaine in the salt tolerant grass, Chloris gayana. For

 

recent reviews see (9,25,26).
In 1964, Craigie and McLachlan reported that glycerol

was the major product of photosynthesis in 2; tertiolecta

 

and that its production increased with an increase in the
salinity of the medium (27). In 1971, Wegmann showed that
glycerol production could be stimulated by the addition of
sucrose or 2-deoxy-D-g1ucose as well as NaCl (28). It was
later shown that the modulation of glycerol production was
reflected in the glycerol concentration within the alga
(8,15,23,29,30).

The effect of high glycerol concentrations on enzyme
activity and salt inhibition was tested by Borowitzka and

Brown on Q; tertiolecta and Q;_viridis (l7). Glycerol, at

 

concentrations up to 3 M, slightly enhanced the inhibitory
action of the salts. The activity of glycerol dehydrogenase

was increased by glycerol concentrations up to 6 M, but

12
above that the glycerol had an inhibitory effect. Glycerol
also diminished the salt inhibition of this enzyme. There
were no detectable differences when comparing the enzymes

of 2; tertiolecta and 2; viridis. Ben-Amotz showed that

 

the RuDP carboxylase, lactate dehydrogenase, and PEP
carboxylase of D; EEEXE could tolerate up to 4 M glycerol
with not more than 50% inhibition (30). Balnokin et al.
reported that glycerol stimulated 02 evolution in destroyed
cells of Q; salina and D;_maritima (18). The high
tolerance of Dunaliella enzymes for glycerol, its ability
to stimulate chloroplasts, and the inverse relation which
exists between its concentration and the water activity

of the medium indicates that glycerol is acting as a
compatible solute in Dunaliella..

It should be noted however, that Schobert believes
that glycerol does not act as an osmoregulator in Dunaliella
(31). She thinks that glycerol, like other polyols,
replaces the hydration water which can be lost during
osmotic stress. The replacement of this water would
stabilize cellular macromolecules thereby protecting them.
These macromolecules would include nucleic acids, proteins,
and lipids.

Although it is generally agreed that glycerol is
involved in the salt tolerance of Dunaliella, the path by
which it is synthesized and degraded is not yet known.
Glycerol was first described by Craigie and McLachlan as a

product of photosynthesis (27). This was later confirmed

13
by Wegmann (28). In 1974, Frank and Wegmann made the

observation that Dunaliella sp. was unable to produce

 

glycerol in the dark (32). Ben-Amotz and Avron later
reported that 2; EEEZE was able to synthesize glycerol in
the dark (29). Borowitzka et a1. studied the problem
further with Q; viridis. They found that when the algae
were placed in 2.5 M NaCl, there was no effect of light on
glycerol production (33). At higher salt concentrations
glycerol production was lower when the algae were in the
dark, as compared to the light, but it still occurred. Both
groups hypothesized that, in the dark, starch was broken
down to produce glycerol (29,33). Kaplan et a1. later
calculated that if all newly produced glycerol was formed
directly from CO2 fixation, the 02:CO2 exchange ratio should
be 1.17 (34). With 2; salina, their measured ratio was
1.81. Therefore they concluded that glycerol was synthe~
sized largely at the expense of reserve carbohydrate, at
least in Q; salina. Whether the pathway of glycerol
production involves photosynthesis and/or breakdown of
reserve carbohydrate, it has been generally agreed to
include dihydroxyacetone phosphate which is hydrolyzed to
dihydroxyacetone (33,35). The dihydroxyacetone is then
reduced to glycerol. -

Ben-Amotz along with Borowitzka et a1. concluded
that the dihydroxyacetone reductase was reversible and
initiated the degradation of glycerol when the alga was

exposed to a solution of lower osmotic strength (33,35).

14
This conclusion resulted from the inability of Ben-Amotz
to detect glycerol leakage from alga in NaCl concentrations
greater than 0.6 M (30) and actual measurements on the in
vitro enzyme activity by Borowitzka et a1. (33). However,
Frank and Wegmann could detect only trace amounts of

radioactive metabolites when 2; tertiolecta containing
14

 

C—glycerol was suspended in a hypotonic solution of NaCl
(32). They were able to measure cellular synthesis, uptake,
and excretion of glycerol. Hellebust found that non-

stressed D; tertiolecta, in full sunlight, excreted 16%

 

of its assimilated carbon and most of it in the form of
glycerol (36). Enhuber and Gimmler also found the plasma
membrane of D; EEEXE permeable to glycerol (37). They
measured efflux rates of between 0.1 and 2 umoles of
glycerol/mg chlor0phyll/hr depending on the external NaCl
concentration. On the basis of the above results, it
appears that the plasma membrane of Dunaliella is probably
permeable to glycerol. It should be noted that a salt

tolerant yeast, Debaryomyces hansenii, also produces

 

glycerol as an osmoticum and that the plasma membrane of
this organism has been shown to be permeable to glycerol
(38).

The results discussed in this section suggest that
glycerol can act as a compatible solute in Dunaliella. Its
concentration is altered as a result of osmotic stress.

The glycerol can probably be synthesized in the dark from

reserve carbohydrate and in the light as a product of

15

photosynthesis. The membrane of Dunaliella seems to be
permeable to glycerol, which as a result would have to be

continually synthesized.

Volume Regulation in Dunaliella

 

In a wall-less organism, volume regulation is a
reflection of osmotic regulation. It was for this reason,
that Rabinowitch, Grover, and B. Ginzburg examined volume
regulation in D; EEEXE (39). A particle size analyzer
(PSA) was used to determine cell volume and it was reported
that these measurements agreed with those from photomicro-
graphs. If cell volume is linearly related to the inverse
of the external osmotic pressure, the cell is acting as an
osmometer with the volume at zero osmotic pressure
representing the portion of the cell which isrmflzosmotically
active. Rabinowitch et al. reported that 2;.EEEZE behaves
as an osmometer, but 60—80% of the cell is osmotically
inactive (39). This conclusion is difficult to accept
because 40% of the cell is dry weight (39). When examining
the plot of the changing volume as a function of time after
immersion in a hypo-curhypertonic medium, they found that
the curve fit that of a two component system. The
calculated flow rates of these components differed by an
order of magnitude. They declared these results consistent
with the idea that the water had to flow through two
membranes, each with a different permeability.

Gimmler et al. also did experiments of this type

16
using 2; pgrya and a PSA. They concluded that 20-40% of
the cell was osmotically inactive (40). They divided the
kinetics of the volume changes into three phases. The
first phase lasted a matter of seconds and was attributed
to water movement. After two minutes (phase two), the
osmotic pressure reached equilibrium. In the next phase
which lasted 150-180 minutes, the cells returned to their
original size. The glycerol concentration, however,
became fully equilibrated after 90 minutes. This indicates
that there is a factor involved in volume recovery other
than the glycerol synthesis.

In 1978, B. Ginzburg again reported on volume
regulation experiments which he had performed with D; EEEXE
(24). He suggested that the PSA was only capable of
measuring a part of the cell. He compared his PSA measure-
ments with those of microphotography and those gotten by
calculating pellet volume minus trapped volume divided by
the number of cells. The PSA measurements were approxi-
mately half of the others. He concluded that the small
PSA measurements were the result of a very permeable,
therefore conductive, Dunaliella plasma membrane. He then
assumed that the concentration of sodium inside the cell
equalled that outside, and calculated that the volume
containing the sodium was approximately half of the cell.
The potassium volume, calculated the same way, is of equal
space. It was then suggested that Dunaliella is divided

into two compartments. One compartment would be in rapid

l7

equilibrium with the external medium and the other would

be rich in glycerol and more tightly bound.
ATP-Driven Cation Exchange in Dunaliella

In most systems where salt tolerance has been
studied, a cation exchange driven by an ATPase has either
been shown or hypothesized. For a review see (3).
Dunaliella is no exception.

In 1969, Ben-Amotz and B. Ginzburg detected a light
induced proton uptake of 2-6 uequiv H+/min/mg chlorophyll
in Q; EEEZE (20). That same year, Jokela found ATPase
activity in an enriched plasma membrane fraction of D;

tertiolecta (15). This enzyme could be stimulated 22% by

 

25 mM Na+ and 51% by 25 mM K+. In 1975, Latorella and
Vadas noted that when the salinity of unbuffered medium

containing 2; tertiolecta was raised from 0.5 M to 1.0 M,

 

the medium pH rose from 7.0 to 7.65 (22). They saw no pH
change with a decrease in medium tonicity. Kaplan and
Schreiber used the quenching of 9-aminoacridine to reflect
the uptake of H+ in D; salina (41). Dark and DCMU caused
a decrease in quenching and electron transport uncouplers
put quenching at its dark level. The ApH was also very
sensitive to PCMPS (p-chloromercuribenzene sulfonic acid),
a poorly penetrating sulfhydryl specific reagent. When
Kaplan and Schreiber increased the sodium concentration of
the medium, the ApH increased. When the sodium concentra—

tion of the medium was decreased, the ApH decreased. The

18
ApH was also sensitive to ATPase inhibitors such as
quercetin, rutin, and phloridzin. Their results plus the
previous results suggest a plasma membrane bound Na+/H+
exchanger which is coupled to an ATPase.

The increasing K+. /K+

ratio 't ' '
in out w1 h increa51ng

sodium chloride in the medium led Gimmler et al. to suggest
that an ATPase-drived Na+/K+ antiport may be present in
Dunaliella (23). The enzyme described by Jokela (15) would
be consistent with this idea. It is possible that both
of these cation-exchange systems may be present and play
a role in maintaining the internal ion concentrations at

tolerable levels.

Summary of Dunaliella-Salt Relations Literature

 

There seem to be three distinct hypotheses which
attempt to describe the relationship between Dunaliella
and the salt which it tolerates. The first hypothesis is
the one shared by Ben-Amotz and Brown. Based on the
susceptibility of enzymes to Na+, they believe that the
membrane is effective in excluding the vast majority of
the sodium ions (16,17). They are also convinced that the
membrane is impermeable to glycerol if the NaCl concentra-
tion of the medium is above 0.6 M (29,30,33) and that the
glycerol is approximately equiosmotic with the sodium
chloride of the medium (30).

Gimmler espouses the second hypothesis. He believes

that the plasma membrane of Dunaliella is semipermeable to

19
sodium and that ATP-driven cation exchanges may help
maintain the internal ion concentrations (23). Although
he thinks that the plasma membrane is permeable to glycerol,
he believes that glycerol plays a major role in osmo-
regulation. He has also come to the conclusion that ions
such as Na+ and K+ are also significant in osmoregulation
(23). To explain the sensitivity of the Dunaliella enzymes
to Na+, Gimmler suggests that these enzymes may be
compartmentalized within the alga and therefore protected.
He also notes that the enzymes may not be Na+ sensitive
in vivo, but become so due to a change in some part of the
system during extraction (23). This would not be unprec-
edented. See (42) for more information.

The third hypothesis which described the relation-
ship between Dunaliella and its salt environment was put
forth by B. Ginzburg (24). It was described in the section
on volume regulation and basically states that Dunaliella
is divided into two compartments. One of these compartments
communicates rapidly with the external medium, while the

other does not.
Fatty Acid Composition of Dunaliella Primolecta

The composition of a membrane is a major factor in
determining its physical state. Unfortunately, the
composition of the plasma membrane of Dunaliella primolecta
is not known. Cuecas and Riley have used gas chromatography

and determined the fatty acid composition of lipids

20

extracted from whole cells of Q; primolecta (43). Their

 

results are shown in Table 1.

Effect of Membrane State on Membrane Permeability

 

When studying salt tolerance with Dunaliella, the
ability of the plasma membrane to serve as a barrier to
permeation is often questioned. The experiments which
suggest that the enzymes of Dunaliella are sensitive to
Na+ (14,15,16,l7,18) seem to require that the membrane
excludes sodium ions. If much of the glycerol produced by
the Dunaliella leaked through the plasma membrane, a
considerable carbon drain would be placed on the alga
because of the need for increased glycerol synthesis.
Therefore if Dunaliella's ability to tolerate salt is to be
understood, the factors which affect membrane permeability
should be investigated.

A molecule's rate of permeation through a membrane
is dependent upon the molecule, its size and charge, and
the membrane through which it is penetrating. Model
systems have been used quite successfully to investigate
the role of the membrane in permeation. The results of
these studies suggest that the size and charge of the polar
headgroup, the length and degree of unsaturation of the
lipid fatty acyl chains (the longer, saturated chains
decrease permeation), and the interaction between membrane
sterols and phospholipids influence permeability. For a

recent review see (44). All of these factors influence the

21

TABLE 1

Fatty Acids of Dunaliella primolecta

 

(43)

 

 

 

Fatty Acid Double Bond Position Percent (w/w)
12:0 --- 0.8
14:0 --— 4.7
14:1 ? 0.1
15:0 --- 1.7
16:0 --- 11.3
16:1 9* 9.8
16:2 9,12 8.3
16:3 6,9,12 7.4
16:4 6,9,12,15 5.9
17:0 --- 0.3
18:0 --- 0.1
18:1 9* 5.8
18:2 9,12 5.8
18:3 6,9,12 2.1
18:3 9,12,15 10.4
18:4 6,9,12,15 6.7
20:0 —-- 0.1
20:1 11* 0.2
20:2 8,11 0.8
20:3 8,11,14 1.3
20:4 5,8,11,14 0.4
20:4 8,11,14,17 1.7
20:5 5,8,11,14,17 9.7
22:0 --- 0.3
22:4 7,10,13,16 0.2
22:5 7,10,13,16,l9 3.9
24:0 --- 0.2

* There may be other isomers

22

physical state of the plasma membrane.

A plasma membrane is composed of proteins and lipids.
The lipids usually are in a bilayer configuration and can
exist in two physical states or phases. In the gel state,
the lipids are highly ordered and the membrane is very
rigid. As the temperature of the system is raised, the
lipids can undergo a phase transition and enter the liquid-
crystalline or fluid state. In a biological membrane,
this phase transition occurs over a wide temperature range
and in this range clusters of gel and fluid lipids coexist.
The existence of these clusters is often termed lateral
phase separation. The size and charge of the polar head-
groups, the length and degree of unsaturation of the fatty
acyl chains and the interaction between sterols and
phospholipids dictate the temperature at which the phase
transition begins and the width of the transition.

The phase transition of the plasma membrane plays
an important role in determining membrane permeability.
In 1968, de Gier et al. found a strong temperature depen-
dence when measuring the permeability of distearoyl-,
dipalmitoyl-, dimyristoyl-, l-stearoyl-Z-myristoyl-, and
l-stearoyl-Z-decanoyl phosphatidylcholine liposomes to
glycerol and erythritol (45). A rapid increase in the
permeation of these molecules corresponded with the phase
transition temperature (TT) of these lipids. Oku et al.,
using liposomes composed of dipalmitoylglycer0phosphocholine

and dicetyl phosphate (molar ratio 10/1), saw an increase

23
in permeability to glucose, sucrose, and inulin at the

phase transition temperature (46). At their respective TT,

Inoue found a drastic increase in the permeability of
dipalmitoylphosphatidylcholine (DPPC) and dimyristoylphos-
phatidylcholine (DMPC) to glucose (47). When these two
lipids were mixed, the temperature range for increased
permeability was broader and proportional to the molar

ratio of the lecithins in the liposome (47). PapahadjOpoulos

et a1. studied the diffusion of 22Na+ through sonicated

vesicles of DPPC and dipalmitoylphosphatidylglycerol (DPPG).
They found complex temperature dependences, but in both

cases the maximum permeation was at TT(48). They also used

22Na+

DPPG to study the relative penetration of and

14

C-sucrose. Below T the penetration rates were equal.

T’
At TT the vesicles seemed to discriminate in favor of the
22Na+, though both diffusion rates increased. Above TT
the difference in permeation rates decreased.

In 1975, when further investigating this dramatic

increase in permeability at T Blok et al. found that the

T’
magnitude of the increase depended on the chain length of

the fatty acids, the longer being less permeable (49,50).

They also measured the dependence of the permeation on

the size of the penetrating molecule (49). They hypothe-

sized that the ion or molecule was diffusing through pores
which developed at the edge of lipid clusters (49,50).

These lipid clusters, domains, are present in the greatest

number in the middle of the lipid phase transition. The

24
following year, Marsh et a1. measured the ability of the
spin label, TEMPOCHOLINE, to penetrate DMPC vesicles (51).
They also performed a statistical mechanical calculation
to determine the fraction of lipids located at domain
boundaries. Because of the excellent agreement between
the temperature where the measured maximum of penetration
occurred and the temperature at which they calculated the
maximum number of domain edges should be present, they also
concluded that the permeation occurred at domain boundaries.
All of the experimental results are consistent with the
idea that molecules diffuse through the pores which result
from molecular mismatch at domain boundaries. These domains
and their boundaries can be found during a phase transition
where cluster of gel and fluid lipids coexist.

When Mandersloot et a1. were studying K+ penetration
through vesicles made from egg lecithin/lysolecithin
mixtures, they found a correlation between membrane thick-
ness and permeation (32). They suggested that decreased
interaction between the lipids in the membrane may be
responsible for the increased permeation of TT (52).
Marcelja and Wolfe, based on their own theoretical calcu-
lations, have concluded that in an uncharged, single-
component bilayer domain formation can not account for the
increased permeability (53). They concluded that the
increase in permeability is due to a decrease in molecular
interactions. It is doubtful whether the calculations of

Marcelja and Wolfe are valid for a biological system.

25
Perhaps in a bioIogical membrane, both the decrease in
molecular interaction and the increase in the number of
domain borders are important factors in determining
membrane permeability.

Early in the studies of the effect of lipid phase
transitions on permeation through membranes, it was
discovered that adding cholesterol to the system decreased
all penetration (45,48,54). If enough cholesterol was
added, the temperature dependence of the membrane perme-
ability could be removed (47,48). Addition of a sterol
which could condense (decrease area per molecule ratio)

a phospholipid monolayer could result in a bilayer
permeability independent of the temperature if the bilayer
was composed of the same lipids as the monolayer (55,56).
In order to condense a monolayer, the sterol must contain
the planar sterol nucleus, 3B-hydroxy group and an intact
side chain (56). The work of Shah and Schulman, which
deals with the condensation of phospholipid monolayers by
cholesterol, suggests that cholesterol acts by occupying
cavities created by the thermal motion of the fatty acyl
chains (57). Because cholesterol in this location would
occupy no additional space, the area per molecule ratio
would decrease i.e., condensation. If cholesterol is in
this position, it would tend to fluidize lipids in the gel
state by disrupting their packing and rigidify the fluid
state lipids by inhibiting the motion of the fatty acyl

chains (58). If enough cholesterol is added, it can destroy

26

c00peration between lipid molecules and effectively abolish
the lipid transition (58). The weakening or abolition of
the lipid phase transition is then reflected in the
temperature dependence of permeation.

The above results seen with model systems are
consistent with observations on biological membranes.
Murata and Fork studied the temperature dependence of the

light-induced spectral shift of carotenoids in Cyanidium

 

caldarium and higher plant leaves (59). The shift seems

 

to result from the generation of a membrane potential
across the thylakoid membrane. See (59) for more details.
Using dark recovery as an indicator of ionic diffusion
through the thylakoid membrane, they concluded that ion
penetration is greatly affected by the physical state of
the lipids. The membrane was most permeable to ions during
the phase transition, when lipid domains were present.
This agrees with results from model systems.

Mc Elhaney et al. were able to alter the composition

of the plasma membrane of Acholeplasma laidlawii by adding

 

various fatty acids and sterols to the growth medium (60).
As in model systems, if cholesterol was not added, glycerol
permeation was strongly temperature dependent. The rate of
permeation increased with increasing unsaturation of the
enriching fatty acyl chains. In these cells, the addition
of cholesterol also decreased glycerol penetration. De
Kruyff et a1. studied the effect of various sterols on

glycerol and erythritol permeation in the same organism

27
(61,62). They again found that cholesterol was able to
reduce the permeability of the plasma membrane. The
lowering of the permeability could be correlated with the
condensing effect of cholesterol on monolayers derived from

A; laidlawii lipids (61). All 3B-hydroxy sterols with a

 

flat steroid nucleus were able to reduce erythritol
permeation (62). These structural requirements are
consistent with the requirements for monolayer condensation.
Epicholesterol, which does not contain a 38-OH group and
has a very limited ability to condense a monolayer, was not
able to reduce the permeability of the plasma membrane of

A; laidlawii to glycerol and erythritol (61). Othersterols

 

which, because of their structures, should not be able to
condense a monolayer were also not effective in reducing
permeability (62). Cholesterol also reduced the energy
content of the phase transition, as determined by differ-
ential scanning calorimetry (61). All of the above results
could have been predicted on the basis of the monolayer
studies.

Brukdorfer et al. were able to modify the membranes
of erythrocytes through a lipid exchange process (63). They
then studied the ability of glycerol to penetrate the
modified erythrocytes and found that the replacement of
38-hydroxy sterols with 3-ketosterols increased membrane
permeability. Cholesterol depletion also increased the
permeability. Keinhaus et al. studied N12+ penetration in

yeast and yeast sterol mutants (64). The nickel ions could

28
not penetrate the membranes of wild type yeAst, but could
easily enter the yeast sterol mutants. The above results

are again consistent with the results from model systems.

Effect of Salt on Membrane Fluidity

 

Since the physical state of the plasma membrane is a
crucial factor in determining membrane permeability, which
may be critical for salt tolerance, it is necessary to take
a closer look at the effects of ions on the physical
prOperties of membranes. The property which will be
examined is membrane fluidity and the discussion will be
limited to salts which dissociate into monovalent ions.

Ehrstrdm et a1. studied the effect of ions on the

fluidity of the cytOplasmic membrane of Bacillus subtilis.

 

They used electron spin resonance with a stearic acid spin
label and its methyl ester as probes (65). They found that
Na+, K+, and Li+ (0.1-l 000 mM) had a disorganizing effect
on the membrane, i.e., they increased membrane fluidity.
When Butler et a1. studied the problem with a cholestane
spin label and the white matter of beef brain, they found
Na+, K+, and Li+ (0.1-100 mM) increased the order, i.e.,
decreased the fluidity, of membrane lipids (66).

Since these early conflicting results on biological
membranes, most of the work has been done on model systems.
Trauble and Eibl found that these same monovalent ions

lowered the phase transition temperature of phosphatidic

acid bilayers by as much as 13°C (67). A lowering of the

29

phase transition temperature is indicative of membrane
fluidization. MacDonald et al. found that monovalent
cations lowered the phase transition temperature of
dipalmitoylphosphatidylserine (68). The pH and resultant
ionization state of the phospholipid headgroup appeared to
be a critical factor in the interaction between the ions
and the lipid (67,68). Trauble and Eibl attributed the
action of these cations to an effect of ionic strength on
the apparent pK of the lipid headgroup (67). The high
concentration of ions (approximately 1 M) necessary to
lower the transition temperature supports the idea that the
mode of action of these monovalent cations is through an
ionic strength effect not a specific adsorption (67).

Since the surface of most membranes is negatively
charged, Tréuble and Eibl assumed that the association

constant for the lipid headgroups can be defined as

K = Ko exp (eWO/kT) (1)

where K0 is the intrinsic association constant, e equals the
elementary charge, To is the surface potential, k is
Boltzmann's constant and T is the absolute temperature (67).
Using the Gouy-Chapman theory and a high surface potential

assumption, they calculated the surface potential to be

ZkT o

‘y = __

o n (2)
e VekT721T /fi

 

where o is the surface charge density,€:is the dielectric

30
constant, and n is the electrolyte concentration (67). By
examining this last equation, it can be seen thatincreasing
the ionic strength of the medium would decrease the surface
potential. This decrease in To can result in further
dissociation of the headgroup (Equation 1). Earlier
experimental work supports this conclusion (69,70).

A result of further dissociation would be the
increased electrostatic repulsion between lipid headgroups.
Trauble and Eibl concluded that it is this increased
repulsion which fluidizes the lipid bilayer and lowers TT
(71,72). Jacobson and Papahadjopoulos (73) and MacDonald
et a1. (68) challenged this conclusion because electrostatic
repulsion could not account for the observed large changes
in the entrOpy and enthalpy of the transition induced by
changes in surface charge. Both of these groups believe
that the change in surface potential also affects inter-
molecular interactions such as hydrogen bonding and van der
Waal interactions in the hydrocarbon region (68,73). A
later paper by Jahnig et a1. supports this conclusion (74).
These altered interactions could be important contributors
to the increased membrane fluidity (68,73).

The decrease in surface potential can change the
fluidity of a membrane by triggering the release of divalent
cations (68,72). These divalent cations are known to
rigidify a membrane (67,68,73), and their release would
result in a fluidization of the membrane. I

+
Jacobson and Papahadjopoulos (73) reported that Na

31

can increase the transition temperature of dipalmitoylphos-
phatidylglycerol. This indicates a rigidification upon the
addition of monovalent cations and supports the work of
Butler et a1. (66). Trauble et a1. studied this further
using methylphosphatidic acid dispersions (71). They
found that if the membrane's lipid headgroups were fully
ionized, the monovalent cations could act to rigidify the
membrane. They presumed that the cations act as screens
for the negative surface charges, thereby reducing
electrostatic repulsion. The rigidification which resulted
was expressed as an increase in the transition temperature,
though this effect saturated at about 0.6 M Na+.

The influence of ions on the surface potential and
the possible screening action of these ions can play a
role in lateral phase separation (71,72). Depending on the
lipids involved and their ionization state, the addition of
monovalents can support the mixing of lipid types or their
lateral separation. Tokutomi et a1. studied this with mixed
phosphatidylserine/phosphatidylcholine membranes (75). They
found that protonation of the serine carboxyl group caused
the solidification of the phosphatidylserine and its
separation from the phosphatidylcholine. If they used KCl
to increase the ionic strength of their medium, the pH
required to induce separation was much lower. Theyconcluded
that electrostatic interaction affects the reorganization,
but that the reorganization was stabilized by van der Waal

interactions.

32

The above discussion suggests that the effect of

monovalent ions on membrane fluidity is varied. These ions

can rigidify or fluidize the membrane, depending on the
ionization state of the lipid headgroups. They can also
interact with a single component of the membrane thereby

influencing lateral phase separation.

33

MATERIALS, METHODS, AND RESULTS

Organism, Medium, and Growth

 

Dunaliella primolecta LB 1000 was obtained from the

 

Starr Algal Collection (University of Texas, Austin). The
culture was rid of contamination by selecting colonies of
algae grown on agar plates containing growth medium plus
300 U/ml polymyxin B and 150 U/ml penicillin G. The final
purification step included the addition of 1000 U/ml
streptomycin sulfate to the liquid culture medium. When
the cultures were used, there was.no visible contamination
on agar plates supplemented with 0.25% (w/v) dextrose and
0.05% (w/v) sucrose and no contamination could be seen when
the sample was examined at 400x magnification when phase
Optics.

The algae were grown in three liter wide form culture
flasks containing one liter of modified Hochsteins medium
(15). The following changes in the medium were made: CaCl2
final concentration of 0.14 mM, KZHPO4 final concentration
of 0.12 mM. Also, vanadium dichloride was replaced with
vanadyl dichloride. Varying amounts of NaCl were added.
Moist air, enriched with 1.4% C02, was bubbled through the
continually stirred cultures. The flasks were kept at 28°C

under constant illumination of 1.3 mW/cmz. For the growth

34
experiments, 100 ml of culture was grown in a 250 ml flask
under the above conditions.

To determine the generation time, periodically an
aliquont of algae was removed from the culture and counted
in a Petroff-Hausser chamber with a Standard Zeiss micro-
sc0pe. The log of cell number was plotted vs. time and
the slope of the linear region was determined by least
squares analysis. Generation time was calculated from the
fitted lepe.

The results of a study which investigated the effect
of NaCl on generation time are shown in Figure 2. The
algae used for these measurements were adapted to the
respective NaCl concentrations for at least thirty genera-
tions before use. Each generation time given is the average
of two to four determinations. The results indicate that,

although the effect is saturating, increasing amounts of

NaCl lengthen the generation time of Q; primolecta. If the

 

generation time is taken as a measure of adaptation, the

Q; primolecta is better able to adapt to lower concentra-

 

tions of sodium chloride. However, if no NaCl is added to
the medium, the algae are not able to divide (results not
shown). This is in agreement with the minimum sodium

requirement referred to earlier.

ESR Measurements and Analysis

 

Electron spin resonance (ESR) was chosen as the

method to study the physical properties of the plasma

35

FIGURE 2

 

Generation time of 2; primolecta as a function
of the sodium chloride concentration in the
growth medium to which the algae were adapted

for at least 30 generations

36

33.3. .32
emumom m. m. e. N. o. m b e N o

 

 

«

 

d la a - d _ q a q -

 

 

m N _ O

as: .002

fl :2 2
(mp) awn. Nonlveawae

'0.

37

membrane of this halotolerant alga. The spectrometer used
in these experiments was a Varian X-band (model E—112) with
an attached temperature controller. The temperature of

the sample was continually monitored by means of external
thermocouple connected to a digital meter (Omega model 250).
The ESR runs were always made with the temperature increased
between spectra recording, allowing five minute equilibra-
tions. The power and modulation were set below levels

which resulted in saturation or line broadening. The sample
cavity could be illuminated with a Hanovia high pressure
xenon-mercury lamp. With a water filter in place, the
resultant light intensity was 27 mW/cm2 and upon illumina-
tion, the increase in sample temperature was not more than
0.3°C.

The hyperfine splitting (2Tll)was determined as shown
in Figure 3. This splitting is used as a measure of
membrane fluidity. When possible, the order parameter (S)
was calculated. This parameter varies between 0 and l, and
is related to the deviation of the fatty acid chains from
their position perpendicular (S=l) to the plane of the

bilayer. To calculate S, ZTL (see Figure 3) must be

measurable and this was not possible at low temperatures.
When 2T1 could be evaluated, the S calculation was performed
according to method number two of Griffith and Jost (76)
using a Varian 620/L100 computer. The plots of ZTllor S vs.

temperature were analyzed in terms of linear components,

using least squares analysis to fit lines to apprOpriate

38

FIGURE 3

Representatives of the ESR spectra obtained withixlvivo
labelling of the plasma membrane of D; primolecta with

 

S-nitroxy-stearate

39

mm3<o

 

 

1
o!

a
On.

ON.

0.9

O-(

 

 

0.-

ON:

On-

0?:

 

 

 

 

40
sections. The intersection of the regression lines
determined the 'break points'. This method of analyzing
ESR data has been shown to give results which agree with
those obtained with other techniques (77,78). In some
cases, the break points were found with the aid of a
computer program for transition point analysis (79). Both

methods gave the same results.

Whole Cell Spin-Labelling Procedure

 

The Dunaliella primolecta used in the ESR experi-

 

ments were again adapted to a particular sodium chloride
concentration for at least 30 generations before use. The
algae were harvested, while in their logarithmic stage of
growth, by centrifugation at 15009 for 10 minutes. They
were then washed twice with fresh growth medium before
labelling. After washing, they were resuspended in their
growth medium to a final concentration of 108 cells/ml. To
0.5 m1 of algae 12ul of 15 mM 5-nitroxy—stearate(5NS)
(Synvar Corp., Palo Alto, CA) in ethanol was added. The
structure of this spin label is shown in Figure 4. The
sample was then placed near the side of a 125 watt sonic
bath and sonicated for 10 minutes. After the above
treatment, the cells were examined microscopically and
found to be swimming. If then placed in culture medium,
the algae were able to grow and divide. Sodium ferricyanide
(1 mM) (K&K, Plainview, NY) was added prior to placing the

sample into the cuvette. Control experiments showed that

41

FIGURE 4
Structures of S-nitroxy-sterate (2(3-carboxypropylr4,
4-dimethyl-2-tridecyl-3-oxazolidinyloxyl)
and TEMPO (2,2,6,6-tetramethylpiperidino-oxyl)

42

IO

meamamemlxomtzb

43
the Na3Fe(CN)6 had no effect on the hyperfine splitting,

order parameter, or position of break points.

Cellular Location of the Spin Label

 

After 2; primolecta was labelled according to the
described procedure, it was important to establish the
position of the spin label within the cell. It had been
shown with various types of cells that cytoplasmic
. components are able to reduce nitroxide spin labels. When
a spin label is reduced, it can no longer generate an ESR
signal. Kaplan et al. reported that human lymphocytes and
mouse L-cells were able to reduce three different types of
nitroxide spin labels (80). Part of the signal could be
recovered by adding non-penetrating potassium ferricyanide.
If the ferricyanide was removed, the signal again decayed.
Rousselet et al. showed that sonicated red blood cell
preparations were able to rapidly reduce spin labels and
that intact erythrocytes were able to reduce the permeant
spin label, TEMPO (Figure 4) (81). Ross and McConnell used
spin label reduction as a measure of permeation into red
blook cells (82). Maruyama and Ohnishi studied the cyto-
plasmic reduction of nitroxides with Rhodospirillum rubrum
(83). They concluded that the reduction which occurs is
catalyzed by a protein and used NADH or NAD+ plus malate as
the electron donor. Furthermore, the depth of the spin
label in a membrane was not a factor in its reduction.

Chlamydomonas reinhardtii, which is in the same order

 

44
(Volvocales) as Dunaliella, is also able to reducenitroxide
spin labels (84).

To test whether 2; primolecta is able to reduce a

 

nitroxide spin label, the permeant spin label TEMPO was
added. The labelling procedure for TEMPO was as described
for SNS, but sonication and Na3Fe(CN)6 addition were
omitted. Upon TEMPO addition, the resulting signal rapidly
disappeared and was often totally gone before the sample
was placed in the spectrometer and a scan performed. The
TEMPO signal could be restored by illumination of the sample.
For details of illumination see section on ESR measurement
and analysis. This ability to oxidize reduced nitroxides
upon illumination seems likely to be common in chlorophyll-
containing plants. For more details see (85).

If 2; primolecta was labelled with 5N8 which had been

 

previously reduced by ascorbate and no Na3Fe(CN)6 added, no
signal was detectable. A signal could be generated by
illumination. If the algae were labelled with non-reduced
SNS, but again no ferricyanide added, the signal intensity
was enhanced upon illumination. This enhancement was
accompanied by a change in spectral shape and an increase
in rigidity (Figure 5). The increase in rigidity was
detected as an approximately one gauss increase in 2T11' The
intensity increase was temperature dependent as shown in
Table 2. These values were obtained with cells adapted to
medium containing eight percent NaCl.

This temperature dependence might suggest that as

45

FIGURE 5

Spectra of 5-nitroxy-stearate labelled 2; primolecta

 

 

before ( ) and after (---F-) illumination

46

 

 

 

 

 

 

47

TABLE 2

Percent Increase in Low Field Peak Height
upon Illumination as a Function of Temperature1

 

 

Temperature (°C) Percent Increase
6 0—2
15 8
26 32
35 58
42 0

 

1Average of two determinations

48

the temperature of the sample was increased, more SNS was
diffusing away from the plasma membrane to the internal
membranes. It could also be a property of the light
driven oxidizing system. To test these ideas, the sample
was illuminated at these temperatures in an order which
was not ascending. The percent increase was found to be
independent of the past sample temperature. There was one
notable exception. If the sample was equilibrated at 42°C,
it was no longer able to oxidize spin label at any
temperature. The dependence of the percent increase on
temperature was therefore a function of the oxidizing
system, and this oxidizing system denatures at 42°C.

After establishing that Dunaliella primolecta is

 

able and does reduce nitroxide spin labels, it was impor-
tant maestablish whether the rate of label exchange between
membranes favored the reduction of the SNS in internal
membranes. Sodium ferricyanide, which is present in the
sample, is able to reoxidize reduced nitroxides. Because
ferricyanide can not penetrate the plasma membrane (80), it
can only oxidize label exposed to the exterior. If the
exchange of label is more rapid than internal reduction,
the 5NS in internal membranes could be contributing to the
detectable signal. To test this, ESR temperature runs were
performed with and without ferricyanide. The analysis of
these runs showed that ferricyanide did not affect the
results. It did, however, prevent a slow reduction of the

signal intensity. To further study the problem, the

49
percent increase upon illumination was studied withaisample
containing Na3Fe(CN)6. The increase was found to be
independent of the presence of this external oxidant. These

results suggest that the observable signal originates from

spin label in the plasma membrane of Q; primolecta.

 

Another method of determining the position of a

spin label is by spectral line broadening. Addition of a
non-penetrating paramagnetic molecule such as nickel (see
(86) and section on permeation in literature review).will
cause line broadening, due to spin exchange, if the spin
label is in the plasma membrane (86). Unfortunately, the
spin label and the added paramagnetic molecule must collide
for this to occur. The distance in the membrane at which
SNS sits would necessitate adding high concentrations of

the paramagnetic molecule.

 

When NiCl2 was added to labelled D; primolecta,
concentrations in the range of 100-600 mM were not able to
produce consistent broadening as measured by the peak-to-
peak distance of the mid-field peak. These concentrations
of nickel chloride had a deleterious effect on the algae
as indicated by their inability to swimi As a result, the
broadening tests were not conclusive.

In a paper from Keith's laboratory, which deals with
the problem of label position, the authors mention that if
the spin label is in the plasma membrane about half of it
can be removed by washing the cells (87) . This is the case

with Q; primolecta labelled with SNS. The average of two

 

50
trials showed that washing removed about 50% of the detect-
able label.
All the results in this section suggest that SNS is
distributed throughout the algae. This distribution pro-
bably is established during the sonication step. Because

2; primolecta is able to reduce the label which is present

 

in internal membranes and the exchange between the internal
and plasma membrane is slower than this reduction, it
appears that the ESR signal is being generated from 5N5
within the plasma membrane. Washing studies support this

conclusion.
ESR Experiments with Adapted Algae

After establishing that most of the observable spin

label is located in the plasma membrane, 2; primolecta

 

adapted to various salt concentrations were examined.
Representative spectra are shown in Figure 3. The spectra
suggest that there are two populations present (88,89). The
rigid component, which was used in all calculations, varied
with temperature. The more fluid component, which appears
as a shoulder on the low field peak of the rigid component,
did not vary with temperature. The presence of both
components in the spectra obtained with spin labelled lipid

extract of D; primolecta rules out the possibility that

 

either component is due to protein boundary lipid.
Plots obtained from two typical ESR experiments are

shown in Figures 6 and 7. The break points determined from

51

FIGURE 6

Plot of the hyperfine splitting (2Tll) (mo-o-)
and calculated order parameter (-A-A-) as a function
of temperature obtained with in vivo labellingcfifthe

plasma membrane of Q; primolecta adapted to and

 

studied in growth medium containing 2% NaCl (w/v)

52

ORDER PARAMETER

mm.

on.

vm.

mm.

NV.

0?.

8

5i

mm.

mm.

Aoov MEDHQEMQEMP
be we me S. 9. mm mm mm em om m.

N.

 

 

-

_

d

q

q

q

d

-

_

 

 

 

O?

N?

mm
Om
No

vm

(ssnnb) ”12

53

FIGURE 7

Plot of the hyperfine splitting (ZTll) (-o-o-)
and calculated order parameter (-A-A-) as a function
of temperature obtained with in vivo labelling of the

plasma membrane of P__ primolecta adapted to and

 

studied in growth medium containing 24% NaCl (w/v)

54

ORDER PARAMETER

mm.

NW

wv.

on.

go.

mm...

NO.

00.

Nm mg .1» CV On No mN VN ON @—

Gov meb‘mma—zm...

N.

m

 

_

 

_

q

 

_

_

_

 

 

vv
we
ow
on
No
vn
mm
mm
Om
Nm
vm

mm

(ssnob) " 12

55
these experiments are plotted as a function of the NaCl
concentration to which the algae were adapted for at least
30 generations (Figure 8). Each data point in Figure 8 is
the result of three to six experiments at each salt
concentration, and the bars express standard error. The

ESR experiments of all the Q; primolecta studied show break

 

points around 9-14°C and 39-43°C. These temperatures seem
to correspond with the temperature range in which the
algae have the ability to oxidize reduced SNS upon
illumination. If the Dunaliella were adapted to a sodium
chloride concentration of eight percent or greater, three
break points can be seen. The additional break is always
around 20-22°C.

The breaks in slope which occur at 9-14°C and 39-43ML
independent of salt concentration, are indicative of the
onset and completion of the membrane lipid phase transition.

When examining a plot of 2T vs. temperature, it is evident

11
that there is a slight change in fluidity as a function of
temperature less than 9-14°C. After this break point,
fluidity changes more rapidly as the temperatures is
increased. At 39-43°C the rate of change of fluidity as a
function of temperature decreases. This plot shape

suggests that at temperatures less than 9—14°C, the membrane
lipids are in the gel or ordered state. Between 9-14°C and
39-43°C, the lipids appear to be in the mixed or separated
state. The transition would then be completed at 30-43°C

and at temperatures above this the lipids would be in the

56

FIGURE 8

Plasma membrane events detected as break points
as a function of the sodium chloride concentration
in the growth medium to which the 2; primolecta

 

were adapted for at least 30 generations

BREAK POINTS (°C)

45

4O

35

30

25

20

I 5

IO

 

 

 

P M '1 .
_ iiii‘
pi‘ i I

.. Eli

 

O 4 8 I2 IS 20

NoCl (%,w/v)

24 28

58

fluid state.

Because sodium can rigidify or fluidize a membrane
(see Review of literature) and the fluidity of a membrane
can be biochemically controlled by a cell (90), it was
interesting to compare the values of 2Tllat the growth
temperature (28°C). An increase in 2Tllindicates that the
membrane is more rigid and a decrease signifies that it is
more fluid. Figure 9 shows the values of ZTilplotted as a
function of the NaCl concentration to which the algae are
adapted. The Dunaliella are in their growth media at the
time of the measurement. These ZTilvalues are taken from
the line which was fitted to the segment of the ESR plots
which contained 28°C. The data points plotted are the
average of four to ten determinations and again the bars
show the standard error. It can be seen that increasing
the NaCl concentration in the culture medium, rigidifies

the plasma membrane of E;_primolecta. This rigidification

 

varies linearly with the external NaCl concentration
(correlation coefficient = 0.97). This effect does not

saturate, even up to 24% NaCl (w/v).

ESR Experiments with Non-Adapted Algae

 

In order to investigate the possible causes of the
break which occurred around 21°C in the plots of ESR data,
it was important to study the Dunaliella in a salt
concentration to which it was not adapted. If thismiddle

break was the result of an electrostatic interaction or

59

FIGURE 9

Plasma membrane fluidity, as determined by ZTll’
as a function of the sodium chloride concentration
in the growth medium to which the Q; primolecta

 

were adapted for at least 30 generations

60

A339; 502
N NN ON 2 b. e. N. o.

 

 

_ _ _ _ _ _ _ _

 

 

mv

Om

_m

Nm

mm

go

no

(ssnnb) ”12

61
disruption of the water layer, ESR plots from algae adapted
to four percent salt and tested in eight percent salt would
show three break points. If this middle break point was
the result of some biochemical adaptation, it should be
present in eight percent adapted algae independent of the
external NaCl concentration. These two particular salt
concentrations were chosen for study so as to minimize the
osmotic stress.

When 2; primolecta adapted to a particular concen-

 

tration of NaCl were studied in a medium which contained

a different NaCl concentration, they were washed four times

with the new medium before use. They were then labelled

as before, but in the new medium not their growth medium.
When Dunaliella adapted to four percent NaCl were

studied in medium containing eight percent NaCl, the plot

of ZTllvs. temperature revealed three breaks (Figure 10).

This indicated that the middle break point was due to a

physical interaction, not a biochemical adaptation. If

Dthrimolecta adapted to eight percent salt were washed as

 

described and then studied in four percent salt, three
breaks in slope were obtained (Figure 11). This result
seems to suggest some kind of biochemical adaptation, but
it could also indicate that enough NaCl remained even after
four washes to cause the break.

To distinguish between these two possibilities,
algae adapted to eight percent NaCl medium were washed four

times in a solution of 9 : l equiosmotic sucrose : 8% NaCl

62

FIGURE 10

Plot of the hyperfine splitting (ZTLQ (-o-o-)
and calculated order parameter (-‘—‘-) as a function
of temperature obtained with in vivo labelling
of the plasma membrane of Q; primolecta adaptedtx>

 

growth medium containing 4% NaCl (w/v) and studied
in medium containing 8% NaCl (w/v)

63

ORDER PARAMETER

Nv.
Ov.
Om.
vm.
on.
NO.
OO.

Ob.

Aoov meH<mmE§m.—.
Nb we S. 9. on N» 8 N 8 b.

N.

O

¢ 0

VI

 

 

—

q

_

_

q

_

fl

q

 

 

Ov

On

No

gm

Om

on

GO

NO

«O

OO

(ssnnb) "13

64

FIGURE ll

Plot of the hyperfine splitting (ZTiB (-o-o-)

and calculated order parameter (-A-A-) as a function
of temperature obtained with in vivo labelling

of the plasma membrane of D; primolecta adapted to

 

growth medium containing 4% NaCl (w/v) and studied

in medium containing 8% NaCl (w/v)

80v wm3H<mMQ2mF
Nb meteoebm NmmNeN 82 N. o v o e-

 

65

ORDER PARAMETER

 

 

 

(ssnnb) "12

 

66
(w/v). The salt was included in this special wash because
it was found earlier that if Dunaliella is washed in a

solution with no sodium, even though the solution is

equiosmotic, the cells will lyse (Fontana, D., unpublished
observation). After this wash, the algae were then washed
three times in 4% NaCl medium and spin-labelled as
previously described. The ESR data plots obtained with
these cells (Figure 12) had only two breaks; the break
around 21°C was absent. All the results obtained with non-
adapted algae are shown in Table 3. These last experiments

seem to suggest that D; primolecta is capable of tightly

 

binding some of the sodium and/or chloride ions. All of
the studies on non-adapted algae reveal that the break
around 21°C is due to some kind Of physical interaction
between the ions in the medium and the plasma membrane of

Q; primolecta.

 

When the ZTilat growth temperature of these non-
adapted algae were compared, it was obvious that the plasma
membranes of all were more rigid than the algae adapted
to the same external salt concentration to which they were
exposed during the experiment. This could be the result of
the stress suffered during the more extensive washing or
a result of their recent need for osmotic adjustment. It
is also apparent, that the algae adapted to a NaCl medium
concentration of 4% are more rigid when placed in 8% medium
than algae adapted to this medium and placed in 4% NaCl

medium. It can also be seen that, within experimental

67

FIGURE 12

Plot of the hyperfine splitting (ZTLB (-o-o-)
and calculated order parameter (-A-A-) as a function
of temperature obtained with in vivo labelling oftbe
plasma membrane of D_. primolecta adapted to growth
medium containing 8% NaCl (w/v), washed four times
in a sucrose:NaCl solution (see text for details),
and studied in medium containing 4% NaCl (w/v)

68

ORDER PARAMETER

on
en.
an.
N...
be.
ob.
en.
mm.
Nb.

OO.

NO O¢ cw 0v Om NO ON vN ON O.

.oev mm3H<mwn=2mH

N.

O

 

 

_

q

q

.

.

.

q

.

—

 

 

NV

vv

O?

O?

CO

NO

vm

OO

Om

OO

NO

(ssnob) "12

69

mafimump MOO uxma mom “amps chommm

 

 

N
mcoflum:NEuou0p 03» mo omnuo>m~
one
>.o N m.mm ova mm» we wm
one
cam
n.o N o.~m boa 02 we NO
one
ova
H.H N m.mm ooa 02 mm mm
.mmsmmv .Oov .>\3V .>\3V
ousumumasme ucmENummxm Omummpfi
530.5 unfleN Bios xnmbm ants 389$ 5:86: Honz a 53mm: Unz N

N

 

 

Hmuooaoefium 4m Umummo<lcoz so mucwEHHmmxm mmm ECHO muasmmm

m mAm<B

70
error, it is not possible to distinguish between the algae
placed in the 4% medium. The special wash, which was
necessary to prevent the event which caused the break point
at 20—22°C, does not cause a noticeable change in the
membrane fluidity. It is possible that the measurement is

not sensitive enough to detect a difference.

ESR Experiments with Lipid Extracts

 

The lipids of D; primolecta were extracted overnight

 

in CHCl3/CH20H (2:1, v/v). The proteins were removed as
described by (91). Hydrolysis at 110°C in 6N HCl for 18
hours, with subsequent paper electrOphoresis pH 1.9 (92)
and ninhydrin dip, failed to detect the presence of proteins.
In order to rid the sample of some of the contaminating
chlorophyll, the sample was layered on a silicic acid
column (loo-200 mesh). The column was first eluted with
CHCl3 and the green band which resulted was removed from
the rest of the lipids. The glycolipids and phospholipids
were eluted with acetone and methanol respectively. Thin
layer chromatography on a silica gel G plate, developed in
petroleum ether/diethyl ether/acidic acid (80:20:1, v/v/v)
with spots detected with iodine vapor, showed that no
component of the neutral lipid fraction was totally removed
with the green band (Figure 13). Fraction I is the neutral
lipid fraction minus Fraction II, the removed green band.
Spraying the neutral lipid TLC plate with acidic

acid/sulfuric acid (1:1, v/v) and SbCl3 in CHCl3 darkened

71

FIGURE 13

Developed thin layer chromatography plate which had
been spotted with the neutral component of the whole
alga lipid extract; Fraction I is the neutral
lipids minus Fraction II which was the green band
eluted from the silicic acid column

The shaded spots react positively with

sterol-sensitive reagents

 

72

 

F

00 - O OORVI

_ _

O

 

009 n

 

 

b

 

 

|.O

8.
O

6
O

4
_ O

N3<> em

2
O

0.0

 

FRACTION

73
two spots, those shaded in Figure 13. This darkening, plus
an RF value of 0.2 in this developing solvent is indicative
of sterols (93,94). The identity of the second darkened

component, RF = 0.03, was not determined. The presence of

sterols in Q; primolecta lipid extracts suggests that

 

sterols may be acting to reduce the permeability of its
membranes.

The lipids, minus the green band, were dried under
nitrogen and then under vacuum for 10 minutes. Double
distilled water was added so that the final concentration
was at least 10 mg/ml. Addition of a glass bead plus
careful vortexing produced an aqueous dispersion which was
sonicated in the bath sonicator for 10 minutes. The label,
5N8 in ethanol, was added so that the concentration of
spin label to lipid was less than 0.2% (w/w). The sample
was again sonicated for 10 minutes. Control experiments
showed that the presence of the ethanol had no effect on
the resulting spectra.

Typical plots of ESR data obtained with lipids from
algae adapted to 2% NaCl and 24% NaCl medium are shown in
Figures 14 and 15, respectively. The lipids extracted from
algae adapted to 2% NaCl medium were also suspended in 2%
medium for experiments. The procedure was as described but
the medium replaced the double distilled water. The results
of these experiments are shown in Table 4. All the ZTllvs.
temperature plots with these lipids showed only two breaks

and these were similar to those determined from plots

74

FIGURE 14

Plot of the hyperfine splitting (2TB) (-o-o-)
and calculated order parameter i-A‘A‘)
as a function of temperature obtained with
lipids, extracted from D; primolecta adapted to

 

medium containing 2% NaCl (w/v), and suspended
in distilled water

75

ORDER PARAMETER

Gov mm:...<mmm.2m...
mbNb meeeovon Nn mNeNoN b. N. m e o

 

- _ q q a . — — q . — _ _ . q

NN.

 

 

 

 

(ssnnb) ' '12

76

FIGURE 15

Plot of the hyperfine splitting (2TH) (-o-o-)
and calculated order parameter (-A-A-)
as a function of temperature obtained with
lipids, extracted from Q; primolecta adapted to

 

medium containing 24% NaCl (w/v), and suspended

in distilled water

77

ZT'I (gauss)

Ob

ON

O0

mm

mm

as

mm

mo

AO

Am

.3

AM

no

 

 

—
-

_

_

_ _ _ r _ _ _ _ _ _

 

 

O A

O

.m

a No Nb me am No no .3 so am
ngvmmbficmm Ace

um

NA

NO

.OO

.ON

.mO

.mA

.00

.AO

bN

.uO

.ub

.uO

HEIEWVHVd 83080

78

OOGNEHODOO #0: ".U.:

nouns ooaennneo ombsoo .on on

anequCHEHoumO 03H mo ommuo>m

m
N
H

 

 

.Nm .o.s com

omN .Uofi Ill

04H m.o.s obs 4N

.mm .mm .em

.eN -u- nu-

.oH .HN .HH N

AUov AUoV

enemas agent: .oov couomuuxm mums
.>\3o Honz NNH .>\3v Honz NN onmso mosses boss: gone

ca popcommsm

cw Ooccmmmsm

ca copcmmmsm

N

mamas bounnoa Honz N

 

muomuuxm

H

 

MDUOHOENHQ 4m mo

ONQNA so mucmENHmmxm mmm ECHO conflfiuoumo mucflom xmmum

v SHONE

79
obtained from whole cells adapted to salt concentrations
less than eight percent.

To determine the effect of sodium chloride on these
lipids, these experiments were repeated with the lipids
suspended in 12% NaCl medium. To do this, the lipid
sample was first placed in one half the usual volume of
double distilled water. After sonication and just prior
to placing the sample in the ESR cuvette, an equal volume
of double strength growth medium with 24% NaCl was added.
The plots of ESR data that resulted contained three breaks
(Figure 16 is typical). The results of these experiments
are also summarized in Table 4.

If most of the salt was then removed using a Folch
extraction (95) and an ESR temperature experiment performed
with the lipids suspended in double distilled water, only
two breaks in slope could be seen. These experiments with
lipid extracts demonstrate that all the break points seen

with intact Q; primolecta are independent of protein and

 

that the middle break is caused by a lipid phenomenon which
is modulated by sodium chloride. The presence of the low
and high temperature break points in the data plots of all
lipid extract experiments supports the interpretation that
these breaks do indeed represent the onset and completion
of the membrane lipid phase transition.

The values of ZTELat 28°C obtained from these ESR
experiments with lipid extracts are shown in Table 5. If

lipids from both 2% and 24% NaCl adapted algae were placed

80

FIGURE 16

Plot of the hyperfine splitting (ZTif
as a function of temperature obtained with lipids,
extracted from D; primolecta adapted to

 

medium containing 2% NaCl (w/v), and suspended in
medium containing 12% NaCl (w/v)

81

2‘1"l (gauss)

Om

O0

OO

mm

we

ON

GO

AO

am

2

AN

 

 

_

_

_ p _ — _ _ _ _ _ _

 

A

O

_N a No Nb No NN am so .3 so
AWZUWEDHCEW Aoov

 

2
8

pmsflfiumuwp you u.O.:

 

 

m
been: ooeaeunmo baboon "oNr ooN
mcoflumcfleuoumv o3u mo mmmum>mH
o.H N m.vm m.c.: . c.H N O.Nm em
o.N N m.mm m.o N m.Nm m.H N N.mv N
.mmsmmv Ammsmmv Ammzmmv Omuomuuxm oqu
3}: Hoez NNH 3}: 3oz N Non no 383 goes: some
ca popcommsm ca noncmmmnm GM Oopcommsm womafl coummpm Homz w

 

 

HmuomHOEwum 4m mo muomuuxm

OHQNA mmadlmaonz co mucmeflummxm mmm ECHO

UOGHEHOHOD musumummEme :u3ouo um mfiemv mcfluufiamm ocflmuomhm

m mqmdfi

83
in double distilled water, the lipid from the 24% adapted

Q; primolecta appeared more rigid. This suggests that at

 

least part of the increased rigidity as a function of
increasing salt in the growth medium that was seen with
the intact algae (Figure 9) is the result of biochemical
adaptation; Addition of sodium chloride containing medium
rigidified the lipids extracted from 2% NaCl adapted algae.
The experiments with lipid extracted from 24% NaCl adapted

Q; primolecta also suggest that salt rigidifies the lipids,

 

but the extent of the rigidification is less and with such
a small change there are problems with experimental
precision. With lipids suspended in 12% NaCl growth medium,
there appeared to be little or no difference in fluidity
between lipids from Q; primolecta adapted to 2% and 24%
NaCl medium. These 2T-

11
amount of NaCl in the growth medium brings about a

values suggest that increasing the

biochemical change which results in increased membrane
. . . + . .
rlgldlty. The presence of the Na 1n the medium also

rigidifies the membrane by means of a physical interaction.

DISCUSSION

Theories on salt tolerance suggest the involvement
of membranes. Whether it be as a membrane-bound enzyme, as
a turgor sensor, or as a barrier to ions, membranes
appear to play a critical role in salt tolerance. In
higher plants, the wall makes the in vivo study of

membranes very difficult. Therefore Dunaliella primolecta,

 

a wall-less unicellular green alga, was suggested as a
model for salt tolerant plants. The preference of Q;
primolecta for lower salt concentrations, though it can
grow and reproduce in 29% NaCl (w/v) (4.1 M), confirms
that this Dunaliella species like others is salt tolerant
not halophilic. Because, as this study shows, one can
specifically spin-label its plasma membrane in vivo it is
even more attractive as a model system for salt tolerance.

When performing ESR studies with 2; primolecta that

 

had been adapted to various salt concentrations, it was
found that the NaCl in the medium did affect the physical
state of the plasma membrane. The fluidity of the membrane
decreased as the external NaCl concentration was increased.
Salt concentrations greater than 8% (w/v) resulted in a
membrane event which was detected as a break point in

ESR data plots. This event always occurred at 20-22°C.

84

85

Both of these effects of NaCl were upon the lipids in the
membrane and neither required biochemical modifications

for their expression. However, results obtained with lipid
extracts suggest that part of the rigidification of the

plasma membrane seen with intact Q; primolecta with

 

increasing salt is the result of a biochemical modification.
The event at 20-22°C could be a lateral phase separation

as a result of a salt interaction with the membrane lipid
headgroups. The rigidification of the membrane, with
increasing NaCl, suggests that the Na+ ions are acting as
screens between the negatively charged lipid headgroups.

Two membrane events, which were also detected with
electron spin resonance, did not vary with external salt
concentration. These break points (events), one at 9-14°C
and one at 39—43°C, have been interpreted as the onset and
completion of the lipid phase transition. ESR studies with
whole cell lipid extracts support this conclusion. The
ability of the alga to oxidize reduced spin label is related
to photosynthetic capacity and since this alga appears to
be an obligate phototrOph (96), these temperatures repre-
sent growth limits. These results suggest that D;

primolecta must grow with its lipids in the mixed physical

 

state.

There is a large body of work which measures the
ability of the membrane to act as a permeability barrier
as a function of its physical state. Membranes are most

permeable to molecules, charged and uncharged, in their

86
mixed state. In vitro studies with enzymes extracted from
Dunaliella suggest that these enzymes are not salt tolerant.
Dunaliella produces glycerol as an osmoticum and for it to
be effective, it must remain within the alga or be
continually synthesized. (See the Review of Literature
for references and more details concerning the previous
three statements.) At first these facts seem to suggest
that Dunaliella primolecta should not grow with its lipids
in the mixed state, but this does not agree with the
results of this study.

There are three physical states in which the plasma
membrane of Q; primolecta might exist. These states are
the gel, mixed, and liquid crystalline states. In the gel
state the membrane is least permeable to all molecules (44).
However, in this physical state the membrane is least able
to withstand osmotic or mechanical stress. This has been

shown with Acholeplasma laidlawii (97) and Escherichia coli

 

(98). Therefore, the gel state would not be suitable for
an alga such as Q; primolecta which must be able to with—
stand osmotic stresses.

Of the other two physical states, the mixed and
fluid, 2; primolecta seems to require the mixed. Perhaps
the possible advantage of this state can be seen if volume
regulation in other organisms is examined. The giant alga,
Valonia, has been used to study volume regulation. An
advantage of this organism is that a known turgor pressure

can be applied across its plasma membrane. It was found

87
that increasing the turgor pressure reduced K+ influx; K+
is the ion that this alga uses for osmotic regulation. As
the turgor pressure of the cell was increased, the
elasticity of the Valonia cell wall also decreased. This
decrease in wall elasticity was pr0portional to the
decrease in active K+ uptake. It was concluded that the
cell wall stretch was mirrored by a stretch in the plasma-
lemma and this latter stretch inhibited active K+ transport.
It was proposed that the stretching of the plasmalemma is
the turgor sensor and it coordinates all the cellular
responses to osmotic stress. For a review of these experi-
ments see (99).

A similar conclusion, i.e. that the plasma membrane
acts as a turgor sensor, was reached as a result of studies
with erythrocytes. Potassium and chloride are the ions
involved in the volume regulation of this cell (100). When
duck erythrocytes were placed in hypotonic medium, the cells
shrank and there was a selective but transient increase in
K+ efflux which lasted until the cells returned to their
original volume (100). When human and duck erythrocytes
were placed in hypertonic medium, the cells swelled and
there was a net increase in K+ influx (101,102) until the
original cell volume was regained (101). This volume
regulation was found to be independent of the ouabain-
inhibitable Na+-K+ exchange (101). These responses to
osmotic stress were specific because K+ alone was involved;

they were not general ion responses. Like in Valonia,

88
volume regulation in red blood cells appears to be related
to the stretch and shrinking of the plasma membrane.

It was hypothesized that the stretching of the
plasma membrane may mechanically alter the conformation of
a critical membrane protein, probably indirectly via the
lipid matrix, thereby altering its activity (99,102). This
is not the only way in which a membrane stretch could be
reflected in enzyme activity. It is also known that the
activity of important membrane-bound proteins are modulated
by the physical state of the lipids which surround them.
For a recent review see (103). Borochov and Borochov have
shown that the swelling of rose protoplasts and phospholipid
vesicles is accompanied by an increase in the fluidity of
their membranes (104). This change in membrane fluidity
could result in a change in the activity of a membrane-
bound enzyme. A change in enzyme conformation may not be
necessary.

With both of these transducing mechanisms, it is the
degree of membrane fluidity change upon an alteration in
turgor pressure that determines the sensitivity of the
sensor. It is in the mixed lipid state that the membrane
fluidity is most affected by slight changes in pressure.

In the gel phase, an increase in turgor pressure would
rupture the cell. In the fluid phase, an increase in
turgor pressure would cause a slight increaSe in membrane
fluidity because of membrane stretching. In the mixed

state, an increase in turgor pressure would have the same

89
effect as raising the temperature, luau they would both
drive the phase transition toward completion. A decrease
in turgor pressure would have the opposite effect. This
can be readily seen with monolayer experiments where, near
the phase transition temperature, the phase change can be
induced by a small alteration in surface pressure (105,106).
This shift in the phase equilibrium would be reflected as
a large change in membrane fluidity. If the problem of
salt tolerance is looked at from a teleological perspective,

it is possible that D; primolecta chose to live with its

 

plasma membrane in the mixed state because in that state
its sensor was most sensitive to changes in turgor pressure.
If the enzymes of Dunaliella are indeed sensitive
to high NaCl concentrations in vivo as they are in vitro,
then the enhanced permeability of the plasma membrane in
the mixed lipid state would still be a problem for the alga.
The enhanced glycerol penetration would create a large
carbon drain for Dunaliella because it would increase the
demand for glycerol synthesis. Studies on model systems
have shown that some sterols reduce membrane permeability
in all physical states and are able to prevent the dramatic
increase in permeability at the phase transition. They do
this by fluidizing the gel state by their inhibition of
lipid packing and by rigidifying the fluid state by their
inhibition of fatty acyl chain movement. In essence, they
are able to abolish the abrupt phase transition, thereby

avoiding the mixed state. (See the Review of Literature

90
for more information.) The presence of sterols, as detected
in whole alga lipid extracts, may act to decrease membrane
permeability and the dramatic effect of the phase transition
on this permeability. However, the phase transition is

still observable in Q; primolecta so there would be some

 

increase in membrane permeability when the lipids are in the
mixed state.

It is also possible that even though the plasma
membrane is in the mixed lipid state, it has an altered
composition so that it is more permeable to less harmful ions
such as K+. This selective permeability has been demon-
strated in model systems (107, 108).

The ability of Dunaliella primolecta to detect and

 

react to a change in external salt concentration is crucial

if it is to survive changes in the salt concentration of

its environment. With its plasma membrane in the mixed state,
it is possible that it is then best able to react to these

changes.

CONCLUSION

Dunaliella primolecta was studied as a model system

 

for salt tolerance in higher plants. It was found that

Q; primolecta could grow and reproduce in a medium contain-
ing 0.25 to 29% NaCl (w/v). The generation time of this
alga was shorter at the lower salt concentrations, thereby
demonstrating that it is a salt tolerant not halophilic
alga.

Since the plasma membrane is thought to play a
crucial role in salt tolerance, it was important to study
this membrane in vivo. A technique was developed which
would allow the observation of this specific membrane with
electron spin resonance. The spin label, 5-nitroxy-stearate

was incorporated into all of the membranes of 2; primolecta.

 

The label which was located in internal membranes was then
reduced by the alga. It was shown that the exchange of
spin label between internal membranes and the plasma
membrane was slow compared to the rate of label reduction.
The addition of non-penetrating sodium ferricyanide
resulted in the oxidation of all reduced spin label which
was present in the plasma membrane. Because only non-
reduced spin label is detectable with the electron spin

resonance technique, the physical properties of the plasma

91

92
membrane could then be studied in vivo.

The ESR spectra which resulted from spin labelling
whole cell lipid extracts and the plasma membrane of the
intact alga were indicative of two lipid spin label
populations. There was a fluid population which appeared
not to vary with temperature. The rigid population changed
with increasing temperature and was the subject of all
the calculations.

Using the labelling technique described above, the
fluidity of the plasma membrane was monitored in vivo as
a function of the external salt concentration. It was
found that increasing the external salt concentration
rigidified the membrane. Based on model system studies,
this is suggestive of a fully ionized membrane where the
sodium ions act to screen the negative surface charges (71).
There was also evidence of a biochemical modification which
resulted in rigidification.

When the fluidity and the order of the plasma
membrane were examined as a function of temperature, three
events could be seen. These events were detected as a
change in lepe when the ESR parameters which correspond
to order and fluidity were plotted as a function of
temperature. One of these events or breaks occurred at
20-22°C, but only with external NaCl concentrations of
eight percent (w/v) or greater. The salt concentration at
which the alga had been growing for at least 30 generations

was not important in determining whether this event would

93
occur; it was the NaCl concentration in the medium at the
time of the measurement that was critical. If lipid
extracts were suspended in growth medium containing 12%
NaCl (w/v), this break could again be detected. All of
these results suggest that the event at 20-22°C is a lipid-
lipid interaction modulated by the NaCl in the medium.

The other two events detected by these ESR tempera-
ture studies always occurred at 9-14°C and 39-43°C and
were independent of the salt concentration in the medium.
These events or breaks have been interpreted as the onset
and the completion of the lipid phase transition. This
interpretation is supported by their occurrence in lipid
extracts.

Since the 2; primolecta used in these experiments

 

were grown at 28°C, these results suggest that the algae
exist with their plasma membrane in the mixed state. Model
systems show that membranes in this state are usually most
permeable to ions and other non—charged molecules (44).
Sterols act to reduce permeability (45,48,54) and their
presence in whole alga lipid extracts suggest that they may

be playing a similar role in Q; primolecta.

 

There has been speculation concerning the plasma
membrane as a sensor of osmotic stress in walled and wall-
less organisms (99,100,101,102). If the fluidity change
which occurs in the plasma membrane as a result of
osmotically induced shrinking or expanding is indeed a

sensor, then there is a good reason for 2; primolecta to

 

94
exist with its plasma membrane in the mixed lipid state.
It would be in that physical state that its sensor would be

most sensitive to slight changes in the environment.

BIBLIOGRAPHY

Epstein, E., 1977, Genetic Potentials for Solving
Problems of Soil Mineral Stress: Adaptation of
Crops to Salinity, in Plant Adaptation to Mineral
Stress, ed. by M.J. Wright, Cornell University,
Ithaca, p. 73-82.

Carter, D.L., 1975, Problems of Salinity in Agriculture,
.in Plants in Saline Environments, ed. by A.
Poljakoff-Mayber and J. Gale, Springer-Verlag,
Berlin, p. 25-35.

Kylin, A., and R.S. Quatrano, 1975, Metabolic and
Biochemical Aspects of Salt Tolerance, in Plants
in Saline Environments, ed. by A. Poljakoff-Mayber
and J. Gale, Springer-Verlag, Berlin, p. 147-167.

Poljakoff-Mayber, A., 1975, Morphological and Anatomical
Changes in Plants as a Response to Salinity Stress,
in Plants in Saline Environments, ed. by A.
Poljakoff-Mayber and J. Gale, Springer-Verlag,
Berlin, p. 97-117.

Jennings, D.H., 1976, Biol. Rev. 51:453-486.

 

Poljakoff-Mayber, A. and J. Gale, 1975, General
Discussion, in Plants in Saline Environments, ed.
by A. Poljakoff-Mayber and J. Gale, Springer-Verlag,
Berlin, p. 193-199.

Racusen, R.H., Kennersley, A.M., and A.W. Galston,
1977, Science 198:405-407.

McLachlan, J., 1960, Can. J. Microbiol. 6:367-379.

 

Brown, A.D., 1976, Bacteriol. Reviews 40:803-846.

 

Brock, T.D., 1975, J. Gen. Microbiol. 89:285-292.

 

Butcher, R.W., 1959, An Introductory Account of the
Smaller Algae of British Coastal Waters, Part I:
Introduction and Chlorophyceae, Fishery Investi-
gations Series IV, Her Majesty's Stationery Office,
London.

95

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

96

Marré, E., and O. Servattaz, 1959, Atti Accad. Nazl.
Lincei., Rend. Classe Sci. Fis. Mat. e Nat. 25:56 -
574.

 

Trezzi, F., Galli, M.G., and E. Bellini, 1965, Giorn.
Bot. Ital. 72:255-263.

 

Johnson, M.K., Johnson, E.J., Mac Elroy, R.D., Speer,
H.L., and B.B. Bruff, 1968, J. Bacteriol. 95:1461-
1468.

 

Jokela, A., 1969, Outer Membrane of Dunaliella
tertiolecta; Isolation and Properties. University
of California, San Diego, Ph.D. Dissertation.

Ben-Amotz, A. and M. Avron, 1972, Plant Physiol.
49:240-243.

 

Borowitzka, L.J. and A.D. Brown, 1974, Arch. Microbiol.
96:37-52.

 

Balnokin, Yu. V., Strogonov, B.P., Kukaeva, E.A., and
A.V. Medvedev, 1979, Soviet Plant Physiol. 26:441-
447.

 

 

Ginzburg, M., 1969, Biochim. Biophys. Acta 173:370-
376. .

Ben-Amotz, A. and B.Z. Ginzburg, 1969, Biochim.Bi0phys.
Acta 183:144-151.

 

Wegmann, K., 1978, in Energetics and Structure of
Halophilic Micro-Organisms, ed. by S.R. Caplan and
M. Ginzburg, Elsevier/North Holland Biomedical
Press, Amsterdam, p. 559.

Latorella,A.H. and R.L. Vadas, 1973, J. Phycol. 9:272-
277.

 

Gimmler, H. and R. Schirling, 1978, Z. PflanzenphysiOl.
87:435-444.

 

Ginzburg, B.Z., 1978, Regulation of Cell Volume and
Osmotic Pressure in Dunaliella, in Energetics and
Structure of Halophilic Microorganisms, ed. by S.R.
Caplan and M. Ginzburg, Elsevier/North Holland
Biomedical Press, Amsterdam, p. 543-560.

Hellebust, J.A., 1976, Ann. Rev. Plant Physiol. 27:485-
505.

26.

27.

28.

29.

30.
31.
32.

33.

34.

35.

36.
37.

38.

39.

40.

41.

42.

43.

97

Kauss, H., 1977, Biochemistry of Osmotic Regulation in
Plant Biochemistry II 13:119-140, ed. by D.H.
Northcote, University Park Press, Baltimore.

Craigie, J.S. and J. McLachlan, 1964, Can. J. Bot.
42:777-778.

 

Wegmann, K., 1971, Biochim. BiOphys. Acta 234:317-323.

 

Ben-Amotz, A. and M. Avron, 1973, Plant Physiol.
51:875-878.

 

Ben-Amotz, A., 1975, J. Phycol. 11:50-54.

 

Schobert, B., 1977, J. Theor. Biol. 68:17-26.

 

Frank, G. and K. Wegmann, 1974, Biol. Zbl. 93:707-723.

 

Borowitzka, L.J., Kessly, D.S., and A.D. Brown, 1977,
Arch. Microbiol. 113:131-138.

 

Kaplan, A., Schreiber, U., and M. Avron, 1976, Carnegie
Yearbook, p. 316-319.

Ben-Amotz, A. and M. Avron, 1973, FEBS Letters 29:153-
155.

 

Hellebust, J.A., 1965, Limnol. Ocean. 10:192-206.

 

Enhuber, G., and H. Gimmler, 1979, Plant Physiol. Suppl.

 

 

63:139.

Gustaffson, L., and B. Norkrans, 1976, Arch. Microbiol.
110:177-183.

 

Rabinowitch, S., Grover, N.B., and B.Z. Ginzburg, 1975,
J. Memb. Biol. 22:211-230.

Gimmler, H., Schirling, R. and U. Tobler, 1977, g;
Pflanzenphysiol. 83:145-158.

 

Kaplan, A. and U. Schreiber, 1976, Carnegie Yearbook,
p. 320-323.

Kushner, D.J., 1978, Life in High Salt and Solute
Concentrations: HalOphilic Bacteria, in Microbial
Life in Extreme Environments, ed. by D.J. Kushner,
Academic Press, London, p. 318-380.

Chuecas, L. and J.P. Rilery, 1969, J. Mar. Biol. Assoc.,
U.K. 49:97-116.

 

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

98

Blok, M.C., Van Deenen, L.L.M., De Gier, J., Op Den
Kamp, J.A.F., and A.J. Verkleiff, 1977, Some Aspects
of Lipid-Phase Transition on Membrane Permeability
and Lipid-Protein Association, in Biochemistry of
Membrane Transport, ed. by G. Semenza and E. Carpoli,
Springer-Verlag, Berlin, p. 38-46.

De Gier, J., Mandersloot, J.G., and L.L.M. Van Deenen,
1968, Biochim. Biophys. Acta 150:666-675.

 

Oku, N., Jojima, S., and K. Inoue, 1980, Biochim.
BiOphys. Acta 595:277-290.

Inoue, K., 1974, Biochim. Biophys. Acta 339:390-402.

 

Papahadjopoulas, D., Jacobson, K., Nir, S., and T. Isac,
1973, Biochim. Biophys. Acta 311:330-348.

 

Blok, M.C., Van Der Neut-Kok, E.C.M., Van Deenen, L.L.M.,
and J. De Gier, 1975, Biochim. BiOphys. Acta 406:187-
196.

 

Blok, M.C., Van Deenen, L.L.M., and J. De Gier, 1976,
Biochim. Biophys. Acta 433:1-12.

Marsh, D., Watts, A., and P.F. Knowles, 1976, Bio-
chemistry 15:3570-3578.

 

Mandersloot, J.G., Reman, F.C., Van Deenen, L.L.M., and
J. De Gier, 1975, Biochim. Biophys. Acta 382:22-26.

 

Maréelja, S., and J. Wolfe, 1979, Biochim. Biophys.
ACta 557:24-31.

PapahadjOpoulos, D., Nir, S. Ohki, 1971, Biochim.
Biophys. Acta 266:561-583.

 

Demel, R.A., Kinsky, S.C., Kinsky, C.B., and L.L.M. Van
Deenen, 1968, Biochim. Biophys. Acta 150:655-665.

Demel, R.A., Brukdorfer, K.R., and L.L.M. Van Deenen,
1972, Biochim. Biophys. Acta 255:321-330.

Shah, 0.0., and J.H. Schulman, 1967, J. Lipid Res.
8:215-225.

Ladbrooke, B.D., Williams, R.M., and D. Chapman, 1968,
Biochim. Biophys. Acta 150:333-340.

 

Murata, N., and D.C. Fork, 1967, Biochim. Biophys. Acta
461:365-378.

McElhaney, R.M., De Gier, J., and L.L.M. Van Deenen,
1970, Biochim. Biophys. Acta 219:245-247.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

99

De Kruyff, B., Demel, R.A., and L.L.M. Van Deenen, 1972,
Biochim. Bigphys. Acta 255:331-347.

 

De Kruyff, B., De Greef, W.J., Van Eyk, R.V.W., Demel,
R.A., and L.L.M. Van Deenen, 1973, Biochim. Biophys.
Acta 298:479-499.

 

Brukdorfer, K.R., Demel, R.A., De Grier, J., and L.L.M.
Van Deenen, 1969, Biochim. BiOphys. Acta 183:334-345.

 

Kleinhans, F.W., Lees, N.D.,.Bard, M., Haak, R.A.,
and R.A. Woods, 1979, Chem.'Phys. Lipids 23:143-154.

 

Ehrstrdm, M., Gdran Eriksson, L.E., Israelachvili, J.,
and A. Ehrenberg, 1973, Biochem. Biophys. Res. Comm.
55:396-402.

 

Butler, K.W., Dugas, H., Smith, I.C.P., and H.
Schneider, 1970, Biochem. Biophys.‘Res. COmm. 40:
770-776.

 

Trauble, H., and H. Eibl, 1974, Proc. Natl. Acad., USA
71:214-219.

 

MacDonald, R.C., Simon, S.A. and E. Baer, 1976,
Biochemistry 15:885-891.

 

Abramson, M.B., Katzman, R., and H.P. Gregor, 1964,

 

Abramson, M.B., Katzman, R., Wilson, C.E., and H.P.
Gregor, 1964, J. Biol. Chem. 239:4066-4072.

 

Tréuble, H., Teubner, M., Woolley, P., and H. Eibl,
1976, Biophys. Chem. 4:319-342.

 

Trduble, H., 1977, Membrane Electrostatics, in Structure
of Biological Membranes, ed. by S. Abramhamsson and
I. Pasher, Plenum Press, New York, p. 509-550.

Jacobson, K., and D. Papahadjopoulas, 1975, Biochemis-
try 14:152-161.

 

Jfihnig, F., Harlos, K., Vogel, H., and H. Eibl, 1979,
Biochemistry 18:1459-1468.

 

Tokutomi, S., Ohki, S., and S. Ohnishi, 1980, Biochim.
Biophys. Acta 596:192-200.

 

Griffith, O.H., and P.C. Jost, 1976, in Spin
Labeling: Theory and Applications, ed. by L.J.
Berliner, Academic Press, Inc., New York, p. 483.

 

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

100

Morrisett, J.D., Pownall, H.J., Plumlee, R.T., Smith,
L.C., Zehner, Z.E., Esfahani, M., and S.J. Wakil,
1975, J. Biol. Chem. 250:6969-6976.

 

Stuhne-Sekalec, L., and N.Z. Stanacev,. 1977, Can. J.
Biochem. 55:173-185.

Brunder, D.G., Coughlin, R.T., and E.J. McGroarty,
1980, Computers in Biol. and Med., in press.

 

Kaplan, J., Canonica, P.G., and W;J. Caspary, 1973,
Proc. Natl. Acad., USA 70:66-70.

Rousselet, A., Guthmann, C., Matricon, J., Brenvenue,
A., and P.F. Devaux, 1976, Biochim. Biophys. Acta
426:357-371.

 

Ross, A.H. and H.M. McConnell, 1975, Biochemistry
14:2793-2798.

 

Maruyama, K. and S. Ohnishi, 1974, J. Biochem. 75:1153-
1164.

Weaver, E.C., and H.P. Chon, 1966, Science 153:301-303.

Briggs, S.P., 1980, Location of Spin Labels in Oat Leaf
Protoplasts. Michigan State University, M.S. Thesis.

Hammerstedt, R.H., Keith, A.D., Bolts, Jr., R.C., and
P.W. Todd, 1979, Arch. Biochem. Biophys. 194:565-
580.

Keith, A.D., Snipes, W., and R.J. Mehlhorn, 1977,
Biophys. J. 19:205-218.

Takeuchi, Y., Ohnishi, S., Ishinaga, M., and M. Kito,
1978, Biochim. BiOphys. Acta 506:54-63.

Brotherus, J.R., Jost, P.C., Griffith, O.H., Keana,
J.P.W., and L.E. Hokin, 1980, Proc. Natl. Acad; Sci.
USA 77:272-276.

 

Huang, L., and A. Haug, 1974, Biochim. Biophys. Acta
352:361-370.

 

Rouser, G., and S. Fleischer, 1965, Isolation, Chara-
terization, and Determination of Polar Lipids of
Mitochondria, in Methods of Enzymology, ed. by R.W.
Eastabrook and M.F. Pullman, Academic Press, Inc.,
New York, Vol. 10, p. 392-394.

Offord, R.E., 1966, Nature 211:591-593.

93.

94.

95.

96.

97.

98.

99.

100.
101.

102.

103.

104.

105.

106.

107.

108.

101

Kates, M., 1972, Techniques of Lipidology: Isolation,
Analysis, and Identification of Lipids, American
Elsevier Publishing Co. Inc., New York.

Krebs, K.G., Heusser, D., and H. Wimmer, 1969, Z. Spray
Reagents, in Thin Layer Chromatography, ed. by E.
Stahl, Springer-Verlag, Berlin, p. 854-909.

Folch, J., Lees, M., and G.H. Sloane Stanley, 1957,
J. Biol. Chem. 226:497-509.

 

Droop, M.R., 1974, Heterotrophy of Carbon, in Algal
Physiology and Biochemistry, ed. by W.D.P. Stewart,
University of California Press, Berkeley, p. 530-559.

Van Zoelen, E.J.J., Van Der Neut-Kok E.C.M., De Gier,
J., and L.L.M. Van Deenen, 1975, Biochim. BiOphys.
Acta 394:463-469.

Haest, C.W.M., De Gier, J., Van Es, G.A., Verkleij,
A.J., and L.L.M. Van Deenen, 1972, Biochim. BiOphys.
Acta, 288.

Kauss, H., 1978, Osmotic Regulation in Algae in Progress
in Phytochemistry, Vol. 5, ed. by L. Reinhold, J.B.
Harborne and T. Swain, Pergamon Press, Oxford, p.
1-27. '

 

Kregenow, F.M., 1971, J. Gen. Physiol. 58:372-395.
Kregenow, F.M., 1971, J. Gen. Physiol. 58:396-412.

Poznansky, M. and A.K. Solomon, 1972, Biochim. Biophys.
Acta 274:111-118.

Sandermann, Jr., H., 1978, Biochim. Biophys. Acta 515:
209-237.

Borochov, A., and Borochov, H., 1979, Biochim. Biophys.
Acta 550:546-549.

Trauble, H., Eibl, H., and H. Sawada, 1974,
Naturwissenschaften 61:344-354.

 

Blume, A., 1979, Biochim. Biophys. Acta 557:32-44.

Scarpa, A. and J. De Gier, 1971, Biochim. BiOphys. Acta
241:789-797.

Scibona, G., Scuppa, B., Fabiani, C., and M. Pizzichini,
1978, Biochim. BiOphys. Acta 512:41-53.

 

 

nrcnlcnu STATE UN
lllll”WNWWNW"!!!
31293105

IV.

E33

LIBRARIES
IWIWIWIWI
8126