J . gen. Microbiol. (1968), 54, 21-32
Printed in Great Britain
21
The Membrane of the Streptobacillus moniliformis L-phase
By S . RAZIN A N D CHASIA BOSCHWITZ
Department of Clinical Microbiology, The Hebrew University-Hadassah
Medical School, Jerusalem, Israel
(Accepted for publication 23 May 1968)
SUMMARY
Membranes of the stable L-phase of Streptobacillus moniliformis were
isolated after lysis of the organisms by osmotic shock combined with ultrasonic treatment. The membranes were composed of about 53 % protein and
40 % lipid. About two-thirds of the lipids were polar; the rest consisted
almost entirely of cholesterol. Radioactive oleic and palmitic acids were
incorporated most efficiently from the growth medium into the membrane
lipids. The L-phase membranes were solubilized by sodium dodecyl sulphate
and sodium deoxycholate. Most of the solubilized membrane proteins could
be separated from membrane lipid by ammonium sulphate precipitation. The
precipitated proteins were very hydrophobic. The solubilized membrane proteins reassociated with membrane lipids after removal of the detergent by dialysis. The lipid :protein ratio in the resulting membrane re-aggregates
depended on the magnesium concentration in the dialysis buffer. Polyacrylamide-gelelectrophoresis showed that most of the protein species of the
original membrane were incorporated into membrane re-aggregates. The
protein composition of the re-aggregate was similar to, but not identicalwith,
that of the hydrophobic protein fraction. The data obtained indicate that
membranes of the stable L-phase of S. moniliformis resemble Mycoplasma
membranes.
INTRODUCTION
The stable L-phase variants are distinguished from their parent bacteria by the
absence of the rigid cell wall (Panos, 1967a). The loss of the cell wall might be spontaneous or induced, and is irreversible. The first stable L-phase was recovered from
Streptobacillus moniliformis cultures by Klieneberger in 1935 and designated as L-I
(Klieneberger, 1935). Since then stable L-phase variants have been obtained from
other bacteria. Most of the biochemical studies made so far relate to L-phase variants
of Proteus species and of Streptococcus pyogenes (Smith, 1964; Panos, 1967a) and
information on the biochemical properties of the S. moniliformis L-phase is scanty.
Devoid of cell walls and osmotically fragile, the stable L-phase variants of bacteria
are eminently suited for the investigation of the bacterial plasma membrane (Salton,
1967), although this membrane may have undergone certain modilkations in composition and structure upon transition to the L-phase to compensate for the loss of the
cell wall. Thus changes have been recorded in the composition of the plasma membrane lipids of Streptococcus pyogenes upon its transition to the L-phase (Panos,
I967b).
The stable L-phase variants morphologically resemble the organisms of the order
Mycoplasmatales. Both types of organisms are bounded by a single lipoprotein
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
22
S. R A Z I N A N D C. BOSCHWITZ
membrane, and are able to grow and multiply in the absence of a cell wall (Dienes &
Bullivant, 1967). Since the major subject of our studies during recent years has been
the biochemistry of the Mycoplasma membrane (Razin, I 967), it seemed worthwhile
to use the techniques developed to compare some of the properties of the L-phase and
the Mycoplasma membranes.
METHODS
Organisms, The stable L-phase of Streptobacillus moniliformis (L- I) was obtained
from Dr E. Kleineberger-Nobel(The Lister Institute for PreventiveMedicine, London).
Mycoplasma laidlawii (oral strain), serologically related to M . laidlawii strain A, was
isolated in our laboratory from the human oral cavity.
Growth conditions. The organisms were grown in 2 1. volumes, dispensed in 4 1.
flasks, of a modified Edward medium (Razin, 1963) containing 3 % (v/v) of horse
serum in place of Difco-PPLO serum fraction. For labelling membrane lipids, the
medium was supplemented with I pc./l. of each [I -14C]-oleic acid (specific activity
32.5 mc./mmole), [~-l~C]-palmitic
acid (specific activity 36.6 mc./mmole) and uniformly labelled [14C]-sodium acetate (specific activity I 5.7 mc./mmole) purchased
from the Radiochemical Centre, Amersham, England. The organisms were harvested
after 16-20 hr of static incubation at 37" by centrifugation at 12,000 g for 20 min.,
washed once and resuspended in 0.5 M-NaCl.
Examination of lysis of organisms. Lysis of washed organisms by osmotic shock, or
by alternate freezing and thawing was measured by the decrease in turbidity of
suspensions at 600mp, increase in the amount of materials absorbing at 260mp
released from the organisms, and by colony counts (Razin & Argaman, 1963; Razin,
1964). Lysis by ultrasonic treatment was done with a M.S.E. ultrasonic disintegrator
(60 W, 20 kc.) at 1-5 amp. Suspensions of organisms prepared in 0.5 M-NaCl were
treated in ice for periods of I to 60 sec. The degree of lysis was estimated by the same
criteria as used for osmotic lysis.
Isolation of cell membranes. L-phase organisms harvested from 2 to 4 1. medium
were resuspended in 10 to 20 ml. 0.5 M-NaC1. The black sulphide precipitate which
accompanied the centrifuged deposit was dissolved by adding dilute HCl to pH 2.5.
The organisms were then collected by centrifugation, washed twice in 500 volumes of
0.5 M-NaCI and resuspended in 20 ml. de-ionized water. The suspension was treated
in the ultrasonic disintegrator for 30 sec. and centrifuged at 6000g for 5 min. The
supernatant fluid was collected and the sediment resuspended in de-ionized water and
treated in the disintegrator for another 30 sec. The treated suspension was centrifuged
as above and the supernatant fluid combined with the first one. Phase-contrast microscopy showed the pooled supernatant fluids to consist of membraneous material.
Pancreatic deoxyribonuclease (I o pg./ml.) was added to the membrane suspension,
and after incubation at room temperature for 15 min. the membranes were collected
by centrifugation at 34,000g for 40 min. The membranes were washed 8 times alternately with 0.05 M-NaCl and 0.05 M-phosphate buffer pH 7.0 (Panos, Cohen & Fagan,
1966) and finally with de-ionized water. For chemical analysis the membranes were
resuspended in a small volume of de-ionized water and freeze-dried. For solubilization and re-aggregation experiments, the membranes were resuspended in ,%buffer
diluted 1/20 in de-ionized water (Razin, Morowitz & Terry, 1965) and kept at -20'
until used.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
L-phase membrane
23
Solubilization and reaggregation of membrane material. Solubilization of membranes
by detergents was tested by adding 0.6 ml. membrane suspension (equiv. 20 mg. protein/ml.) in 1/20p-buffer to 3 ml. solutions containing various concentrations of
sodium dodecyl sulphate (SDS) or sodium deoxycholate (DOC) in the same dilute
buffer. Changes in the extinction of the suspensions at 500 mp were measured after
15 min. incubation at 37". For re-aggregation experiments the membranes were
solubilized in SDS (0.5 mg. detergentlmg. membrane protein). The non-solubilized
material was removed by centrifugation at 34,000g for 40 min. and the clear solution
dialysed in the cold for 2 to 3 days against 1/20,!?-buffercontaining various magnesium
concentrations (Razin et al. 1965).The re-aggregated membrane material formed in the
dialysis bag was collected by centrifugation at 34,000g for 40 min.
Isolation of the hydrophobic membrane proteins. The method used was that of
Criddle, Bock, Green & Tisdale (1962)as modified by Rodwell, Razin, Rottem &
Argaman (1967).Membranes were dissolved in a mixture of SDS (I mg. detergent/
mg. membrane protein) and sodium deoxycholate (DOC, 2 mg. detergent/mg.
membrane protein). Ammonium sulphate was added to 12% saturation. The heavy
precipitate formed was collected by centrifugation at 10,000g for 10min. and dissolved in the detergent mixture. The procedure of protein precipitation and resolution
was repeated twice more and the final protein precipitate washed twice and resuspended in 1/20,&buffer. The protein suspension was kept at 4" until used, with
I/IO,OOO
thiomersalate to prevent bacterial growth.
Polyacrylamide-gel electrophoresis of membrane proteins. The procedure described
by Rottem & Razin (1967)was used. The gels contained 7-5% (wlv) acrylamide, 35 %
(v/v) acetic acid and 5 M-urea; the electrophoresis buffer was 10% (v/v) acetic acid.
Analytical methodr. Protein, carbohydrate and nucleic acids were determined in
defatted organisms or membranes. Protein was determined according to Lowry et al.
(1951)with crystalline bovine plasma albumin as standard. Nucleic acids were extracted according to Schneider (1945).RNA was determined by the method of Drury
(1948)and DNA according to Burton (1956). Total carbohydrate was estimated
according to Dubois et al. (1956)after removal of the nucleic acids and lipids. Lipids
were extracted with chloroform+methanol (2+ I, by vol.) as described by Razin,
Argaman & Avigan (1963). Polar lipids were separated from the non-polar lipids by
chromatography on silicic acid columns (Ansell & Hawthorne, I 964). The non-polar
lipids were eluted from the column with chloroform and the polar lipids with methanol
(Razin et al. 1966). The two fractions were weighed after evaporation of the solvents
under a stream of nitrogen. Cholesterol in the non-polar lipid fraction was separated
by thin-layer chromatography and estimated by the FeCl, reaction (Argaman &
Razin, 1965). Radioactivity in lipid fractions was determined in a Packard Tri-Carb
liquid scintillation spectrometer. The scintillation mixture consisted of 800 ml. dioxane, 150 d. toluene, 50 g. naphthalene, 10 g. 2,5-diphenyloxazole (PPO) and
I 50 mg. I,~-bis-2-(~-phenyl-oxazolyl)-benzene
(POPOP).
Membranes or membrane protein solutions in 1/20/3-buffer containing 0.02 M-SDS
(7 mg. proteinlml.) were centrifuged at 59,780rev./min. in a Spinco model E analytical
ultracentrifuge at 20'. Sedimentation coefficients were calculated without correction
for medium viscosity or density.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
S. RAZIN A N D C. BOSCHWITZ
24
RESULTS
Susceptibility of the L-phase organisms to lysis
Osmotic lysis. The L-phase organisms lysed in de-ionized water and dilute NaCl
solutions. Lysis manifested itself in a decrease in colony counts, in the extinction of the
suspensions, and a greater amount of material absorbing at 260 mp released from the
organisms (Fig. I).
10
0.25
0.12
0
0.37
0.50 0
0.12
0.25
0.37,
0.50
5
NaCl (M)
Fig. I. Osmotic lysis of the Gphase of Streptobacillus moniliformis andMycoplasmalaidlawii.
The extent of lysis in the various NaCl solutions was measured after 15 min. incubation at
37".Extinction readings at 600 mp of suspensions ( 0 ) ;extinction readings at 260 mp of the
supernatant fluids of the corresponding suspensions (A) ; number of viable particleslml.
suspension (0).
Table
I.
Killing by ultrasonic treatment of Streptobacillus moniliformis
L-phase and Mycoplasma laidlawii
Number of viable particles/ml.
Time of ultrasonic treatment (sec.)
0
3
5
I0
30
I80
r
A
l
L-phase
8.ox I 0 8
3'5 x 108
5-9x 107
3 . 6 ~10'
Mycoplasma
8.2 x 109
2-1x 109
I-4x109
1.1 x 109
2 - 2 x 107
g-ox 1 0 8
1.8 x 107
3'8 x I06
MycopZasma laidlawii organisms were included in these experiments for purposes of
comparison. The extinction readings at 600mp of the L-phase suspension in deionized water were not as low as those of M. laidlawii. This was largely due to the
presence of sulphide crystals in the L-phase suspensions. Large quantities of these
crystals formed during growth, sedimented with the organisms at harvest and could
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
L-phase membrane
25
not be completely separated from them by differential centrifugation. Phase-contrast
microscopy of the L-phase culture in Edward medium showed the spherical organisms
to differ very much in diameter: some had a diameter of several microns while the
diameter of others was near the limit of the microscope resolution. The larger forms
were the first to lyse when the tonicity of the medium was decreased, leaving behind
the smaller forms, some of which resisted lysis even in de-ionized water (Fig. I).
Lysis by ultrasonic treatment. The L-phase and Mycoplasma organisms rapidly
lysed by ultrasonic treatment. Most organisms lysed within the first 10 sec. of treatment (Fig. 2 ; Table I). Again, the first to lyse were the larger forms. The sulphide
crystals in the L-phase suspensions were responsible for most of the residual turbidity
after prolonged ultrasonic treatment.
Lysis by alternate freezing and thawing. The Streptobacillus moniliformis L-phase
organisms and Mycoplasma laidlawii organims were relatively resistant to lysis by
alternate freezing and thawing when suspended in 0.5 M-NaCl. Alternate freezing and
thawing of the L-phase suspension in de-ionized water added little to the lytic effectof
the de-ionized water.
0.5
0.4
a
8
Y
(d
80
a
0.3
ca
!
i
9 60
0.2,
za
3
0
g
=
.CI
Y
3
8
40
bl
0
."fi
5
0.1.
20
&
n
n
-0
10
20
30
40
50
60
Seconds of ultrasonic treatment
Fig. 2
v 7
0
2.5
7.5
5.0
MgCl, in dialysis buffer ( x
10.0
0
10-*M)
Fig. 3
Fig. 2. Lysis of the Lphase of Streptobacillus moniliformis and Mycoplasma Iaidhwii by
ultrasonic treatment. Extinction readings at 600 m p of suspensions ( 0 ) ;extinctionreadings
Lphase;
at 260 mp of the supernatant fluids of the corresponding suspensions (A).
--- mycoplasma.
Fig. 3. Effect of magnesium on the re-aggregation of protein and lipid of L-phase membranes solubilized by sodium dodecyl sulphate. Percentage of membrane protein reaggregated (0);percentage of membrane lipid reaggregated ( 0 ) ;ratio of lipid (expressed as
radioactivity) to protein in re-aggregates (A).
-y
Isolation and chemical composition of the L-phase membranes
The results of the lysis experimentsindicated that osmotic lysis could be used for the
isolation of the L-phase membranes. However, several problems had to be solved
before this method could be used for the isolation of large quantities of membranes,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
26
S. RAZIN A N D C. BOSCHWITZ
First, the accompanying sulphide crystals had to be removed. The addition of dilute
HC1 to pH 2.5 to the suspension dissolved the crystals, producing the characteristic
Has odour. The acid treatment did not cause lysis, not did it affect the gross chemical
composition of the cell membranes, as determined by comparative chemical analysis
of membranes isolated without acid treatment. Since the lysis experiments indicated
that some of the minute L-phase forms did not lyse by osmotic shock, ultrasonic
treatment was used with osmotic shock to assure maximal lysis. The yield of membranes obtained by the above method from I 1. of culture was about 80-100 mg. dry
weight.
Table 2. Chemical composition of organisms and membranes of
Streptobacillus moniliformis L-phase
Organisms
Membranes
A
r
Mean*
\
Range
A
I
Mean?
>
Range
% dry weight
Protein
Lipid
Carbohydrate
RNA
DNA
48.7
31.2
2'1
I 1.6
5'1
39'4-52'0
28-7-3
6.4
1.2-2-7
9-2-12-2
46-60
52-6
40.2
0.9
2.0
-
50'2-55' 2
37'0-44'0
0.8-0-9
I -4-28
0.8s
*
The mean of determinations performed on five different batches of cells.
The mean of determinations performed on eight different batches of membranes.
2 Determined on a single batch of membranes, all other batches were treated with pancreatic
deoxyribonuclease and were essentially free of DNA.
t
Table 3. Incorporation of radioactive fatty acids into lipids of
Streptobacillus moniliformis L-phase
Organisms were grown in 100 ml. Edward medium containing 3 % (v/v) horse serum and
of radioactive fatty acid.
Percentage of
Radioactivity (cornts/min.)
radioactivity
A
3
incorporated
r
Radioactive fatty
Whole
of total
organisms
Cell lipids
radioactivity
acid in medium
0.1pc
[I- 1 4 q -Palmitic acid
65,000
[I -14C]-Oleicacid
53,000
2,I 80
[14C]-Sodiumacetate
59'0
44'2
2.2
The chemical composition of organisms and membranes of the L-phase organism is
compared in Table 2. The membranes were essentially composed of protein and lipid,
and their low RNA content indicated that they were relatively free from cytoplasmic
contaminants. Silicic acid chromatography showed that polar lipids consisted about
two-thirds (range 56.7 to 75.2 %) of the total membrane lipids. The non-polar lipid
fraction consisted almost entirely of non-esterified cholesterol.
Membrane lipids were selectively labelled by adding radioactive oleic or palmitic
acids to the growth medium of the L-phase organism (Table 3). The long-chain fatty
acids were most efficiently incorporated into cell lipids, as shown by the high percentage of the total radioactivity in the growth medium which was incorporated. Sodium
acetate, the common precursor for fatty acid synthesis, was poorly incorporated by the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
L-phase membrane
27
L-phase organisms (Table 3). Since the long-chain fatty acids were selectively incorporated into membrane lipids, radioativity determinations could be used for the
estimation of the relative lipid content of various membrane fractions, as described
in the following sections.
Solubilization and re-aggregation of membrane material
The Lphase membranes were solubilized by detergents. Sodium dodecyl sulphate
(SDS) was more effective in membrane solubilization than sodium deoxycholate
(DOC). The clear solution obtained by solubilization of the membranes with SDS
became turbid after removal of the detergent by dialysis against dilute /3-buffer
containing magnesium ions. The re-aggregated membrane material was collected by
centrifugation, and chemical analysis showed it to consist of protein and lipid. The
Mg2+concentration of the dialysis buffer determined both the degree of re-aggregation
of membrane material and its composition (Fig. 3). Since membrane lipids re-aggregated at a lower Mg2+concentration than membrane proteins, re-aggregates obtained
at low Mg2+ concentration contained a higher percentage of lipid. The ratio of
polar to non-polar lipids in re-aggregates obtained at the various Mg2+concentrations
was about the same.
Table 4. Streptobacillus moniliformis L-phase :separation of membrane
protein from lipid by ammonium sulphate fractionation
The Lphase membranes containing labelled lipid were solubilized by SDS+DOC, and
membrane proteins were precipitated by ammonium sulphate at 12% saturation as described under Methods.
Protein
Radioactivity
-
Preparation
Solubilizedmembranes
Supernatant fluid obtained after
first (NH&$04precipitation
Combined supernatant fluids obtained
after second and third (NH&S04
precipitations
Combined washing fluids of
precipitate
Washed precipitate
A
r
7
% Total
mg.
% Total
Counts/min.
26.1
100.0
23.0
490,000
330,750
100'0
6.0
3.7
I 1.7
148,250
29.8
2.8
I 0.7
6,100
1'2
I 2-9
49'5
4,100
0.8
68.2
Characterization of membrane proteins of Streptobacillus moniliformis L-phase
A large portion of L-phase membrane protein could be separated from membrane
lipid by a method based on the procedure of Criddle et al. (1962) for the isolation of
the mictochondrial structural protein. The method, described in detail under Methods,
consisted of thc solubilization of the membranes by a mixture of SDS+DOC and
precipitation of the protein by ammonium sulphate. Table 4 summarizes the results
of a typical experiment and shows that the major part of the solubilized L-phase
membrane protein precipitated at 12 % saturation of ammonium sulphate, while
most of the membrane lipid remained in the supernatant fluid. The small amount of
lipid that precipitated with the protein could be almost completely removed by repeated dissolution of the precipitate in the detergent mixture and by reprecipitation
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
28
S. R A Z I N A N D C. BOSCHWITZ
with ammonium sulphate. Since the precipitated protein was highly hydrophobic it
could be washed in dilute p-buffer. The hydrophobic protein could be dissolved in 67 %
acetic acid, dilute NaOH or 0.02 M-SDS.The solution of the hydrophobic membrane
protein in SDS exhibited a single symmetrical schlieren peak in the analytical ultracentrifuge. This peak had an uncorrected sedimentation coefficient of 2-4 s, similar to
the single peak shown by whole membranes solubilized in SDS.
Fig. 4. Electrophoretic patterns of the Streptobaci//us moni/$ormis L-phase membrane
proteins. (A) Proteins of whole membranes; (B) proteins of membrane re-aggregate formed
at 0.01 M-MgCl,; (C) hydrophobic membrane proteins precipitated at 12 % saturation of
(NH&304;(D) membrane proteins in the supernatant fluid obtained after removal of the
re-aggregate formed at 0.01 M-MgCI,.
Electrophoretic analysis of the L-phase membranes in polyacrylamide gels containing 35% acetic acid and 5 M-urea revealed about 20 different protein bands
(Fig. 4). Membrane re-aggregates formed in the presence of 0-01M-MgCl, contained
most of the protein bands characterizing the original membrane. However, several
proteins were not incorporated into the re-aggregate and were detected in the supernatant fluid obtained after its sedimentation. The electrophoretic pattern of the hydrophobic protein fraction was similar to, but not identical with, that of the re-aggregate.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
L-phase membrane
29
DISCUSSION
Over 99 % of the Streptobacillus moniliformis L-phase organisms were lysed by
osmotic shock, or by ultrasonic treatment. However, in all experiments a small
portion of the forms did not lyse and remained viable. Moreover, the L-phase organisms
resisted lysis by alternate freezing and thawing in 0.5 M-NaCl. In these respects the
L-phase organisms resembled the Mycoplasma organisms rather than the bacterial
protoplasts which are extremely sensitive to lysis by these methods (Razin & Argaman,
1963; Razin, 1964). Phase-contrast microscopy revealed that the smaller forms in the
L-phase culture were the most resistant to lysis. Apparently, the larger surface area of
the minute forms allows a rapid efflux of intracellular solutes during the osmotic
shock so that the stretchingand tearing effects on the cell membrane are less. Thorsson
& Weibull(1958) also noticed that the small bodies in a Proteus L-phase culture were
relatively resistant to osmotic lysis. The higher resistance to osmotic lysis of the
Lphase than of bacterial protoplasts may also be due to variations in the chemical
composition of the plasma membrane. That the L-phase membrane is highly elastic is
evident from the fact that in hypotonic solutions the forms are able to swell markedly
without bursting, as can be seen in the phase-contrast microscope.
The composition of the membrane lipids which, according to the model of Danielli
& Davison (1935), form the backbone of the biological membrane, may have a marked
influence on its physiologicalproperties, as was recently shown in Mycoplasma (Razin
et al. 1966). Cohen & Panos (1966) found the membranes of a Streptococcuspyogenes
Lphase to contain 35-6% lipid, while membranes of protoplasts prepared from the
same bacterium contained only 15.3 % lipid. Similar results were obtained by
Ward & Perkins (1968) on comparing the lipid content of membranes of the L-phase
with protoplasts of Staphylococcus aureus. In both cases the percentage of glycolipids
was much higher in the L-phase membranes than in the corresponding protoplast
membranes. While oleic.acid was the predominant CISmonoenoic acid in the L-phase
membrane of S. pyogenes, cis-vaccenic acid predominated in the S. pyogenes protoplast membrane (Panos et al. 1966). Smith & Rothblat (1962) showed that the lipid
content of the L-phase of Streptococcus moniliformis was about 5 times that of the
bacterial form; L-phase variants of other bacteria contained 2-3 times as much lipid
as their parent bacteria. The high lipid content (about 40 %) found in the present
study in membranes of the S. moniliformis L-phase is in line with these observations.
The data obtained so far seem to indicate that when a bacterium is transformed into
its stable L-phase the percentage of lipid in the plasma membrane increases, and
changes occur in its composition. The high percentage of lipid in the L-phase membrane
resembles that found in Mycoplasma membranes (Razin, I 967).
The Streptobacillus moniliformis L-phase membrane also resembles Mycoplasma
membranes, and differs from protoplast membranes, in its high cholesterol content.
Almost all the non-polar lipid fraction of the L-phase membranes (about one-third
of the total lipid) consisted of cholesterol in the non-esterified form. Cholesterol is
apparently incorporated from the growth medium, as shown by Rebel, Bader-Hirsch
& Mandel (1963) for the L-phase of Proteus sp. strain ~ 1 8 The
. presence of large
amounts of cholesterol in the L-phase membranes may be associated with the fairly
high osmotic resistance of the organisms, since cholesterol was found to increase the
osmotic stability of parasitic mycoplasmas (Razin, I 967).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
30
S. R A Z I N A N D C. B O S C H W I T Z
The question whether the bacterial L-phase variants require cholesterol for growth
is of great importance for their differentiation from mycoplasmas which do require it.
The few stable L-phase variants of Proteus species (Medill & O’Kane, 1954; Abrams,
1955) and of Streptococcus pyogenes (van Boven, Kastelein & Hijmans, 1967) so far
grown in chemically defined media did not show a cholesterol requirement. Edward
(1953) when using a complex medium was unable to demonstrate that the Streptobacillus moniliformis L-phase required cholesterol. Since no defined medium has so
far been devised for the growth of this L-phase, the question of its cholesterol requirement does not seem to be definitely settled.
Solubilization of the L-phase membranes by detergents enabled the major part of
membrane protein to be separated from lipid. The protein fraction isolated was very
hydrophobic, resembling the ‘ structural ’ protein fractions isolated by detergent
action from a variety of biological membranes including those of Mycoplasma (Green
& Tzagoloff, I 966 ; Rodwell et al. 1967). Polyacrylamide-gel electrophoresis showed
this fraction to consist of most of the protein species found in the membrane. Polyacrylamide-gel electrophoresis has recently shown that the mitochondria1 structural
protein, formerly regarded as a homogeneous protein species also consists of several
different proteins (Beattie, Basford & Koritz, I 967).
Most of the membrane proteins which were separated from membrane lipids by the
detergent action re-associated with membrane lipid on removal of the detergent by
dialysis against a magnesium solution. The membrane re-aggregates consisted mostly
of typical triple-layered membranes ( S . Rottem, 0. Stein & S . Razin, unpublished).
As with Mycoplasma and bacterial protoplast membranes (Razin et al. 1965; Terry,
Engelman & Morowitz, 1967; Butler, Smith & Grula, 1967; Rottem, Stein & Razin,
1968) a divalent cation, e.g. Mg2+,was essential for the re-aggregation of the solubilized L-phase membrane material. The ratio of lipid to protein in the re-aggregate
was greater at a low Mg2+concentration, in accord with the results obtained with
Mycoplasma membranes. The varying lipid :protein ratio in membrane re-aggregates
was regarded as a contra-indication for the presence of homogeneous lipoprotein
subunits in the solubilized membrane material (Rottem et al. 1968).
Comparison of the proteins incorporated into membrane re-aggregates with those
found in the hydrophobic protein fraction and in the original membranes is of interest
(Fig. 4). The incorporation of solubilized membrane proteins into the re-aggregates was
selective. While the proportion of the majority of the membrane proteins was the
same in the re-aggregate as in the original membranes, some proteins did not reassociate with the membrane lipids and remained in the soluble supernatant fluid
fraction. Comparison of the electrophoretic patterns of the hydrophobic protein
fraction and of the re-aggregate shows the marked similarity of the two. But again,
some proteins found in the re-aggregate were not present in the hydrophobic protein
fraction, and vice versa. It seems, therefore, that membrane re-aggregates are not
formed only of the hydrophobic membrane proteins. This fits with the discovery of
the protein(s) responsible for NADH oxidation in membrane re-aggregates of Mycoplasma laidlawii, which was absent from the hydrophobic protein fraction (Razin et al.
1965; Rodwell et al. 1967). Polyacrylamide-gel electrophoresis seems to be a most
efficient tool for further studies on membrane reasssembly.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
L-phase membrane
31
This research was financed in part by grant no. FG-Is-174 made by the United
States Department of Agriculture under P.L. 480.
REFERENCES
ABRAMS,R. Y. (1955). A method for cultivation of L-forms in liquid media. J. Bact. 70, 251.
ANSELL,
G. B. & HAWTHORNE,
J. N. (1964). Phospholipids, p. 46. New York: Elsevier Publishing Co.
ARGAMAN,M. & RAZIN, S. (1965). Cholesteroland cholesterolesters in Mycoplasma. J.gen. Microbiol.
38, 153.
BJXITIE,D. S., BASFORD,R. E. & KORITZ,S. B. (1967). The inner membrane as the site of the in
vitro incorporation of L[14C]-leucineinto mitochondria1protein. Biochemistry 6, 3099.
BURTON,
K. (1956). A study of the conditions and mechanism of the diphenylaminereaction for the
colorimetric estimation of deoxyribonucleicacid. Biochem. J. 62, 3 I 5.
BUTLER,
T. F., SMITH,G. L. & GRULA,E. A. (1967). Bacterial cell membranes. I. Reaggregation of
membrane subunits from Micrococcus lysodeikticus. Can. J. Microbiol. 13, 1471.
COHEN, M. & PANOS,C. (1966). Membrane lipid composition of Streptococcuspyogenes and derived
L form. Biochemistry 5, 2385.
CRIDDLE,
R. S., BOCK,R. M., GREEN,
D. E. & TISDALE,H. (1962). Physical characteristics of proteins
of the electron transfer system and interpretation of the structure of the mitochondrion. Biochemistry I, 827.
DANIELLI, J. F. & DAVSON,H. (1935). A contribution to the theory of permeability of thin films.
J. cell. comp. Physiol. 5, 495.
DIENIB,L. & BULLWANT,
S. (1967). Comparison of the morphology of PPLO and Lforms of bacteria
with light and electron microscopy. Ann. N. Y. Acud. Sci. 143,719.
DRURY,H. F. (1948). Identification and estimation of pentoses in the presence of glucose. Arch.
Biochem. 19,455.
DUBOIS,M., GILLES,
K.,HAMILTON,
K., REBES,
P. & SMITH,
E. (1956). Determination of carbohydrates
by phenol. Analyt. Chem. 28,350.
EDWARD,D. G. ff. (1953). A difference in growth requirements between bacteria in the L-phase and
organisms of the pleuropneumonia group. J. gen. Microbiol. 8, 256.
GREEN,D. E. & TZAGOLOFF,
A. (1966). Role of lipid in the structure and function of biological
membranes. J. Lipid Res. 7 , 587.
-ERG=,
E. (1935). The natural occurrence of pleuropneumonia-like organisms in apparent
symbiosis with Streptobacillus moniliformis and other bacteria. J. Path. Bact. 40, 93.
LOWRY,0. H., ROSEBROUGH,
N. J., FARR,
A. L. &RANDALL,
R. J. (1951).Protehmeasurement with
the folin phenol reagent. J. biol. Chem. 193, 265.
MEDILL,M. A. & O’KANE,D. J. (1954). A synthetic medium for the L type colonies of Proteus.
J. Bact. 68, 530.
PANOS,C. (1967a). Biochemistry of bacterial L-form growth and structure. In A Microbial Enigma
Mycoplasma and Bacterial Lforms, Ed. by C . Panos. P. 167. Cleveland: The World Publishing
Company.
PANOS,C. (19676). L-forms, lipid alterations and inhibition of cell wall biosynthesis. Ann. N. Y.
Acad. Sci. 143, 152.
PANOS,C., COHEN, M. & FAGAN,
G. (1966). Lipid alterations after cell wall inhibition. Fatty acid
content of Streptococcuspyogenes and derived L-form. Biochemistry 5 , 1461.
RAZIN,
S. (1963). Osmotic lysis of Mycoplasma. J. gen. Microbiol. 33, 471.
RAZIN, S. (1964). Factors influencing osmotic fragility of Mycoplasma. J. gen. Microbiol. 36, 451.
RAZIN,S. (1967). The cell membrane of Mycoplasma. Ann. N. Y. Acad. Sci. 143, I 15.
RAZIN,S. & ARGAMAN,
M. (1963). Lysis of Mycoplasma, bacterial protoplasts, spheroplasts and
L-forms by various agents. J. gen. Microbiol. 30, 155.
RAZIN, S., ARGAMAN,M. & AVIGAN,J. (1963). Chemical composition of Mycoplasma cells and
membranes. J. gen. Microbiol. 33,477.
RAZIN,S., MOROWITZ,
H. J. & TERRY,
T. M. (1965). Membrane subunits of Mycoplasmu laidluwii
and their assembly to membranelike structures. Proc. natn. Acud. Sci., U.S.A. 54, 219.
RAZIN, S., TOURTELLO-~~E,
M. E., MCELHANEY,
R. N. & POLLACK,
J. D. (1966). Influence of lipid
components of Mycoplmma Iaidlawii membranes on osmotic fragility of cells. J. Bact. gr, 6og.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52
32
S. R A Z I N A N D
C. BOSCHWITZ
KEBEL,G., BADER-HIRSCH,
A. M. & MANDEL,P. (1963). Recherches sur les lipides des formes L
dkrivbes du Proteus P 18. I. Presenceet absorption du cholesterol. Bull. Soc. Chim. Biol. 45, 1327.
RODWELL,
A.W., RAZIN,S., ROTTEM,S. & ARGAMAN,
M. (1967). The association of protein and
lipid in Mycoplasma laidlawii membranes disaggregated by detergents. Arch. Biochem. Biophys.
122, 621.
ROTTEM,S. & RAZIN,S. (1967). Electrophoretic patterns of membrane proteins of Mycoplasma.J.
Bact. 94,359.
ROTTEM,S., STEIN,0. & RAZIN,S. (1968). Reassembly of Mycoplasma membranes disaggregated by
detergents. Arch. Biochem. Biophys. 125,46.
SALTON,
M. R. J. (1967). Structure and function of bacterial cell membranes. A . Rev. Microbiol. 21,
417.
SCHNEIDER,
W. C. (1945). Phosphorus compounds in animal tissues. I. Extraction and estimation of
deoxypentose nucleic acid and of pentose nucleic acid. J. biol. Chem. 161, 293.
SMITH,P. F. (1964). Comparative physiology of pleuropneumonia-like and L-type organisms. Bact.
Rev. 2 8 , ~ .
SMITH,P. F. & ROTHBLAT,
G. H. (1962). Comparison of lipid composition of pleuropneumonia-like
and L-type organisms. J. Bact. 83, 500.
TERRY,
T. M., ENGELMAN,
D. M. & MOROWITZ,
H. J. (1967). Characterization of the plasma membrane of Mycoplasm Zaidlawii. 11. Modes of aggregation of solubilized membrane components.
Biochim. Biophys. Acta 135, 391.
THORSSON,
K. G . & WEIBULL,C. (1958). Studies on the structure of bacterial L forms, protoplasts
and protoplast-like bodies. J. Ultrastr. Res. I,412.
VANBOVEN,C. P. A., KASTELEIN,
M. J. W.& HIJMANS,
W. (1967). A chemically defined medium for
the L-phase of group A streptococci. Ann. N . Y. Acad. Sci. 143,749.
WARD,
J. B. & PERKINS,
H. R. (1968). The chemical composition of the membranes of protoplasts
and L-forms of Staphylococcus aureus. Biochem. J . 106, 391.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:33:52