Journal of General Microbiology (1984), 130, 27-37.
Printed in Great Britain
27
The Lipids of the Rumen Fungus Piromonas communis
By P A T R I C K K E M P , * D A V I D J . L A N D E R A N D C O L I N G . O R P I N
Biochemistry Department, ARC Institute of Animal Physiology, Babraham,
Cambridge CB2 4AT, UK
(Received 31 May 1983; revised 6 September 1983)
~
~~~
The major phospholipids of the anaerobic rumen phycomycete Piromonas communis were phosphatidylethanolamine (38 %), phosphatidylcholine (26%) and phosphatidylinositol (1 3 %); no
sphingolipids, glycolipids, plasmalogens or phosphonyl lipids were detected. Free fatty acids,
triacylglycerols, 1 :2 diacylglycerolsand a variable amount of 1 :3 diacylglycerol were identified,
as were minor amounts of squalene and a triterpenol which is probably tetrahymanol. Approximately half the fatty acids were straight chain, even 12 to 24 carbon, saturated acids, the
remainder being even 16 to 24 carbon, mono-unsaturated fatty acids. The double bonds in all
except the 16 carbon acid were in the 0 9 position. The unsaturation is introduced by a A9
desaturase which uses stearic acid as substrate and which does not use oxygen as a terminal
electron acceptor.
from acetate and glucose was incorporated into the fatty acids of all
complex lipids, as were lauric, myristic, palmitic, stearic and oleic acids. [ 14C]Choline was
incorporated into phosphatidylcholine and [ 14C]ethanolamine into phosphatidylethanolamine
and phosphatidylcholine. Label from [ ITIserine was recovered in phosphatidylserine and phosphatidylet hanolamine, but was not detected in p hosphatidylcholine.
INTRODUCTION
The anaerobic rumen phycomycetes Piromonas communis, Neocallimastix frontalis and
Sphaeromonas communis, are found in sheep in highest numbers when the animals are fed highroughage diets. Under these conditions rumen fungi could contribute up to 8% of the microbial
mass (Orpin, 1981). Piromonas communis can degrade plant structural carbohydrates including
cellulose and xylans and use the products as sources of energy and carbon (Orpin, 1981). It is
known that glycogen accumulates in the vegetative phase and zoospores (Munn etal., 1981). The
metabolism of these fungi may be a significant part of the rumen economy and, since they have
chitin in their cell walls (Orpin, 1977b), they may be somewhat resistant to microbial attack and
could reach the abomasum and lower digestive tract to provide the host animal with a source of
nutrients unmodified by bacterial activity.
Nearly all the choline in hay diets is in the free form, which is rapidly degraded to
trimethylamine in the rumen (Neil1 et al., 1979) unless it is taken up by protozoa and incorporated into protozoal phosphatidylcholine. Though many of the protozoa are selectively retained
in the rumen (Leng, 1982; Weller & Pilgrim, 1974), those which pass to the lower gut may
contribute a small amount of choline to the host animal. High-roughage diets do not encourage
high protozoal populations, so it may be that the rumen fungi can replace protozoa in decreasing
ruminal choline degradation.
Orpin & Letcher (1979) showed that cr-linolenic acid stimulated the in vitro growth of N .
frontalis. a-Linolenic and linoleic acids are very low in poor hay diets (Dawson & Kemp, 1970)
and uptake by fungi provides a possible means for some to escape hydrogenation in the rumen
(Dawson & Kemp, 1970; Kemp et al., 1975)and so increase the amount available to satisfy the
host's small requirement for these essential fatty acids (Lindsay & Leat, 1977).
To advance our knowledge of the metabolism of rumen fungi, their contribution to rumen
(W)2?-I2K7i8410001-1383$02.00 0 1984 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
28
P. K E M P , D . J . LANDER A N D C. G. ORPIN
metabolism and to the nutrition of the host animal, we have determined the lipid composition of
P.communis and examined the uptake of compounds which would be expected to contribute to
fatty acid and complex lipid synthesis.
METHODS
Materials. Unlabelled substrates were from Sigma and tetrahymanol was purified by TLC from the alkali-stable
lipids of Tetrahymena pyrformk grown in a non-sterol-containing medium. 1-I4C-Labelled long-chain fatty acids,
[2-I4C]acetateand ethanolamine, [Me-14C]cholineand [U-I4C]serineand glucose were obtained from Amersham.
Culture of organism. Firomonus communis was grown in a medium containing (g I - ] ) : centrifuged rumen fluid
(Orpin, 1977aj, 200; bactocasitone (Difco), 10; yeast extract (Difco), 2.5; NaHCO,, 6; KH2P04,0.45; K2HP04,
0.45; NaCI, 0.9; (NH4)2S04,0.9; MgS04. 7H20,0.09; CaCI,. 2H20, 0.12; cellobiose, 0.25; L-cysteine, 0.25; and
resazurin, 0.01. This solution was equilibrated with oxygen-free C 0 2and autoclaved in sealed bottles for 20min at
10 Ibf/in2 (69 kPa; 115 "C). After cooling, a sterile solution of ampicillin (66 mg 1- * ) was added followed by a
filter-sterilized vitamin solution (10 ml I - ] ) made up as follows (mg ] - I ) : thiamin, 10; riboflavin, 200; calcium
pantothenate, 600; nicotinic acid, 10oO; nicotinamide, 1OOO; folk acid, 50; cyanocobalamin, 200; D-biotin, 200;
pyridoxine hydrochloride, 100; and 4-aminobenzoic acid, 50.
Maintenance cultures were subcultured daily using a 10% inoculum from the previous day's culture.
Experimental cultures were prepared in the same way and the vegetative growth and zoospores harvested 20-30 h
later. The vegetative growth was harvested by filtration using four layers of muslin and pressed dry with filter
paper; zoospores were recovered by centrifugation of the filtrate at 500 g for 5 min. Samples were either extracted
at once or, after storing at - 70 "C, frozen samples were plunged directly into the extraction solvent. No difference
in lipid composition was noted between these treatments.
Media with 14C-labelled substrates. Water soluble 14C-labelledsubstrates were added to the medium as solutions.
I4C-Labelledfatty acids, dissolved in methanol, were added to 5 ml media and the methanol was boiled off in a?
autoclave. The suspension was sonicated (FS 100 ultrasonic bath; Decon Ultrasonics Ltd, Hove, UK) for 5 min
experiments
r
were carried out with
at 50 "C before adding to the bulk medium for sterilization. All 1 4 C - p r e ~ ~ r so
50 ml medium containing 5 pCi carrier-free precursor, except for some experiments with linoleic and a-linolenic
acids.
Extraction of lipids. The cell pellets were first dispersed in 15 vol. (w/v) acetone using an Ultra-Turrax
homogenizer, allowed to stand at room temperature for 3Omin, then filtered using glass fibre paper, or
centrifuged. The extraction was repeated with the same volume of CHCI,/MeOH (2 : 1, v/v), followed by
CHCI,/MeOH (2 : 1, v/v) + 0.5% (v/v) conc. HCI and finally again with CHCI,/MeOH (2 :1 v/v). Refluxing the
pellet for 2 h with 2% (w/v) anhydrous methanolic HCI (Gray, 1976) indicated an extremely low plasmalogen
content and hence that extraction with acidic solvents could be used.
The acetone extracts were dried on a rotary evaporator ( < 40 "C) and dissolved in the combined CHCI,/MeOH
extracts and washed with 0.25 vol. 0.9% (w/v) NaCl (Folch et al., 1957). The lower phase was washed once with
upper phase in which the NaCI solution was replaced with water. The lower phase was evaporated to dryness and
the lipids stored at - 16 "C in CHCI3/MeOH (2 :I v/v). To check the efficiency of the extraction the pellets were
refluxed with 6% (w/v) KOH in aqueous methanol 95% (v/v) or with 2.5% (v/v) conc. H,S04 in methanol. The
fatty acids and methyl esters were quantified by GLC.
Separation and analysis of lipids. The washed lipid extract was separated on a silicic acid column (Mallinckrodt
CC4 Special); 1 g silicic acid was used for each 25 mg total fatty acid in the sample and a column was chosen to
give a bed height at least 10 times its diameter. The column was poured in CHCI,/MeOH (1 : 1, v/v) and washed
with acetone then chloroform. The lipid extract was applied in chloroform and the colum eluted successively with
four column volumes of chloroform, acetone and CHCl,/MeOH (1 : I , v/v). The neutral and phospholipid
fractions were subjected to TLC on silica-gel plates (Merck, Kieselgel 60F254).The neutral lipids were separated
in one dimension using hexane/diethyl ether/acetic acid (70:30:1, by vol.) and the phospholipids in two
dimensions using a saturation chamber (Parker & Peterson, 1965). Chloroform/methanol/acetone/water/ethyl
formate (50 :40 :10 :4 :1.5, by vol.) was used as solvent in the first dimension and chloroform/methanol/ammonium
hydroxide (sp. gr. 0*88)/diethylamine (1 10: 50: 15: 10, by vol.) in the second dimension. Spraying with 2',7'dichlorofluorescein followed by examination in UV light was used to detect all lipids. Ninhydrin (0.2%, w/v) in
acetone was used to locate phospholipids containing amino groups and the spray of Vaskovsky & Kostetsky (1968)
to locate phospholipids. For choline-containing lipids, the Dragendorff reagent was used (Beiss, 1964). Individual
lipids were eluted with chloroform (neutrals) or chloroform/methanol (1 :1, v/v) and quantified by GLC analysis of
their fatty acid methyl esters using an internal standard of methyl-heptadecanoate (Christie, 1973). Phospholipids
were analysed for phosphate after digestion with perchloric acid (Dawson, 1976) and by the electrophoretic
examination of their water-soluble products after 04N transacylation using methylamine (Clarke & Dawson,
I98 1). Plasmalogens were analysed by HgCl,/trichloroacetic acid hydrolysis (Dawson et al., 1962). After
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
Piromonas communis lipids
29
separation by ionophoresis, the hydrolysis products were detected using ninhydrin and a phosphate-specific spray
(Dawson et al., 1962). Lipids stable to the above treatments were subjected to the aqueous, methanolic HCI
hydrolysis procedure of Gaver & Sweeley (1965) and the fatty acid esters and long chain bases were examined by
GLC on SE 30 (see below) at 200 "C. The water-soluble products were examined by ionophoresis (Dawson et al.,
1962). The aminoethylphosphonate content of the phosphatidylethanolamine fraction was examined by
hydrolysis with 5 M-HCI at 120 "C for 48 h (Dawson & Kemp, 1967).
Preparation and analysis ofjiatry acid methyl esters. Ester-bound fatty acids were freed by alkaline hydrolysis
(Kemp et al., 1975) and methylated with diazomethane (Schlenk & Gellerman, 1960; Roper & Ma, 1957). To
determine amide-bound fatty acids, samples were refluxed with 2.5% (v/v) conc. H2S04in methanol, or heated at
90 "C overnight in a sealed tube with 1 M-HCI in aqueous methanol (Gaver & Sweeley, 1965).
A Pye 104 gas-chromatograph with a flame-ionization detector was used for analysis of the fatty acid methyl
esters with glass columns packed with either polyethylene glycol adipate 10% (PEG A) or 3% SE 30 on 100-120
mesh Chromosorb W (Phase Separations Ltd, Queensferry, Clwyd, UK). Argon was used as carrier gas and
analyses were carried out at either 184 "C or 197 "C, respectively. Peak areas were measured with a Hewlett
Packard 338044 integrator and methyl heptadecanoate was used as an internal standard. Radio-gas
chromatography and determinations of the geometry and positions of double bonds were according to the methods
of Kemp et al. (1975). The cis and trans monoenoic fatty acid methyl esters were first separated on thin layers of
silver nitrate-impregnated silicic acid, followed by oxidation by the periodate-permanganate method of von
Rudloff (1 956). The resulting dicarboxylic acid mono-methyl esters were methylated and examined by GLC.
Physical measurements. Infrared spectra were obtained in CC1, solution using a Perkin-Elmer model 157G
spectrophotometer. Mass spectra were recorded on an AEI MS 50 instrument using both a direct insertion probe
and a gas chromatography inlet. The beam energy was 70 eV. Proton NMR spectra were recorded at 400 MHz in
deutero-chloroform using trimethylsilane for the lock signal.
RESULTS A N D DISCUSSION
Lipid extraction
The yields of vegetative growth of P . communis varied between 0.36 g and 1.I g dry weight per
litre of culture. Yields of zoospores varied from nil to 50 mg dry weight 1- l . Phospholipid phosphorus in the washed chlorofonn/methanol extract (Folch lower phase, Folch et al., 1957) was
2.05 mg per g dry weight of vegetative growth, GLC analysis showed that 92-97% of the total
fatty acids recoverable from the cell pellet were extracted by the solvents used. There were about
4mol fatty acids per mol phospholipid phosphorus when the fungi were grown on
unsupplemented medium. The pellet remaining after the solvent extraction of the vegetative
growth was refluxed with either KOH or H2S04in methanol and by heating at 37 "C with 0.2 MNaOH in methanol (see Methods) to quantify any residual lipid. Small amounts of fatty acid
methyl esters, representing 3-7% of the total fatty acids, were the only lipid components
recovered by both alkaline and acid methanolysis. Comparison of the fatty acid methyl esters
released by these procedures revealed only minor differences in quantity and composition. Since
amide-bound fatty acids are stable to mild alkaline treatment (37 "C)the residual fatty acids in
the pellet are most probably bound as esters, not amides.
The dehydrated aqueous phase (Folch et al., 1957) was also refluxed with KOH in methanol or
H2S04 in methanol. No more than 0.3% of the acids already removed from the pellet were
recovered, showing that glycolipids make an insignificant contribution to the total lipids.
Treatment of the whole cells or any of the extracts by the anhydrous methanolic HC1 procedure
of Gray (1976) yielded very small amounts ( < 0.5 % total fatty acids) of dimethyl acetals of longchain aldehydes. The aqueous methanolic HCl hydrolysis of Gaver & Sweeley (1965) gave
similar low recoveries of long-chain bases and glycerol ethers.
Silicic acid column chromatography
Neutral lipids, including free fatty acids, were eluted with chloroform, glycolipids with
acetone and phospholipids with CHClJMeOH (1 :1, v/v). Elution with more polar solvents did
not remove any more lipid material. No phosphorus was found in either the chloroform or
acetone eluates and the latter contained only 1% of the total fatty acids applied to the column,
confirming the absence of glycolipids.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
30
P . KEMP, D . J . LANDER AND C . G . ORPIN
Table 1. Composition (%) of phospholipid fraction obtained by silicic acid chromatography
Method of analysis
r
Phosphatidylethanolamine
Phosphatidylc holine
Phosphatidylinositol
Phosphatidylserine
Phosphatidic acid
Cardiolipin
Phosphatidylglycerol
Unidentifiedt
Not recovered
ND,
1
Methylamine hydrolysis
(Clarke & Dawson, 1981)
Phosphate content of lipids
from 2-dimensional TLC
38-0
25.7
12.3
5.6
4.8
3.1
32.0*
23.3
11-7
10.0
5.8
6.7
0.5
9.0
1 -0
ND
5.8
4.7
Not detected.
* Phosphatidylethanolamine frequently ran as two overlapping crescents, both of which contained exclusively
phosphatidylethanolamine and no aminoethylphosphonate, but which differed in their fatty acid composition (see
text).
t Not identified as aldehyde, sphingolipid or glycerolether material.
Phospholipids
The CHClJMeOH eluate from the silicic acid column had a phosphorus to fatty ester ratio of
1 : 1-98,which suggests that the phospholipids are exclusively of the diacyl-form which is unusual
for a rumen organism. No sphingolipids or plasmalogens were detected in significant amounts.
Phospholipid analysis by the hydrolytic method of Clarke & Dawson (1981) using methylamine,
gave results (Table 1) similar to those obtained by determining the phosphate content of the
individual phospholipids separated by two-dimensional TLC on silicic acid plates (Table 1).
A slightly better separation was obtained by ionophoresis of the water-soluble alkaline
degradation products of the phospholipids (Clarke & Dawson, 1981) than by two-dimensional
TLC of the parent phospholipids. Phosphatidylethanolamine often ran as two overlapping
crescent-shaped spots. We found no evidence for aminoethylphosphonate (which occurs in the
lipids of the rumen ciliate protozoa) in either component of this composite spot; after treatment
with strong acid (Dawson & Kemp, 1967), only inorganic phosphate and ethanolamine were
recovered. Fatty acid analysis of the two phosphatidylethanolamine components showed that
the faster running component in the second TLC solvent had 62% of its fatty acids as monoenoic
acids, whilst the slower had only 34% unsaturated. The difference resulted from almost a
doubling of the proportion of each unsaturated fatty acid except that of palmitoleic acid.
The fatty acid composition of the major phospholipids is shown in Table 2. There was a slight
excess of saturated over unsaturated acids in the unfractionated phospholipids and this was
exaggerated in the fatty acids of phosphatidylcholine which had the bulk of the lauric (12:0),
myristic (14 :0) and palmitoleic (16: 1) acids but proportionally less stearic (18 :O), oleic (18: 1)
and nervonic acids (24 : 1) compared with the other phospholipids. Phosphatidylethanolamine
and phosphatidylinositol have similar compositions. Phosphatidylserine had a low proportion
of 16 : O and 16 : 1 but the highest proportion of 18 : O and 18 : 1 and 12.9% of the combined Czs
acids.
Neutral lipids
The neutral lipid fractions eluted from silicic acid columns with CHCl, contained free fatty
acids, whose contribution was equal to that of the phospholipid fatty acids, except when the
fungi were grown in media fortified with fatty acids. Raised levels of the added fatty acids were
then found in the free fatty acid fractions.
Separation of the neutral lipids by TLC revealed three major and three minor components.
Each was eluted and characterized by TLC and GLC using known standards, and then by
alkaline hydrolysis followed by identification of the products. The three major components were
triacylglycerol (RF0.85), free fatty acid ( R F0.4) and 1 :2 diacylglycerol ( R F0.22). One of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
31
Piromonas communis lipids
Table 2. Fatty acid composition (percentage weight) of the major phospholipids separated by
TLC
Fatty acid*
12 :o
14:O
16:O
16: 1
18:O
18: 1
20:o
20: 1
22 :o
22: 1
24:O
24: 1
A:Saturated
...
Total
phospholipid
Phosphatidylcholine
14.2
10-5
20.3
1.5
11.0
35.5
0.3
2.0
0-4
0.6
0.4
3.3
42.9:57.1
12.7
14.7
28.0
5.0
7.1
30.0
0.1
0-5
0.1
0.4
0.3
1'1
37:63
Phosphatidylethanolamine
Phosphatidylinosito1
Phosphatidylserine
1.4
4.8
22.7
0.9
3.7
8.3
0.3
19-5
48.8
1.7
3-1
0.4
0.4
3.7
9.2
6 1 -8 : 38.2
<O-1
4.9
25.7
0.5
14.9
43.2
0.8
2.6
0.9
1.3
0.7
4.5
52.1 :47.9
1.1
17.8
41-2
1.6
3.3
0.2
1 *6
0.3
4.0
51.2:48-8
* The fatty acids represent 98% of those eluted from the column and the total percentage weight of those acids
listed has been adjusted to 100%. The esters were chromatographed on 10% PEG A at 184 "C and checked on
SE 30 at 197 "C.
Table 3. Analysis of the methyl esters of the fatty acids derived from the neutral lipid fraction of
one sample
Analyses are given as percentage weight of the recovery of the sum of the acids indicated. Straight chain
acids containing even numbers of carbon atoms account for 95-99% of all acids eluted from the column.
The esters were chromatographed at 184 "C on a 10% PEG A column and checked at 197 "C on a 3%
SE 30 column.
Fatty acid
12:o
14:O
16:O
16: I
18 : O
18: 1
20:o
20: 1
22:o
22 : 1
24 : O
24: 1
A :Saturated
Triacylglycerol
(9.6%)
Free fatty acid
(48.2%)
1 : 2 Diacyl glycerol
1 :3 Diacylglycerol*
3.2
4.2
15.4
2.9
10.4
58.2
<O.I
2.2
5.7
14.7
16.9
7.5
21-6
25.2
1.2
2.4
0-2
0.6
0.5
3.5
39.2 :60.8
4.2
10.9
24.3
0.5
13-3
41.3
1.4
4.3
1943
8.7
14.7
<0.1
...
0.6
1.I
1.8
65.6 : 34.3
(34%)
0.4
2.3
0.1
0.3
0.3
2.1
46.5 :53.5
(8.2 %I
44.8
0.6
0-4
0.2
0.7
14
3-5
58.1 :41.9
* 1 : 3 Diacylglycerol varies from sample to sample being from about 10-25% of the amount of 1 :2
diacylglycerol.
minor components was identified as 1 :3 diacylglycerol ( R , 0-35)and was only separated with
difficulty from a component stable to both acid and alkaline hydrolysis as was a third minor
component ( R F0.91). The composition of the fatty acids and acylglycerols is given in Table 3.
The fatty acid analysis of the 1 : 3 diacylglycerol was slightly different from that of the 1 : 2
diacylglycerol and the amount of 1 :3 diacylglycerol varied from sample to sample.
The fatty acids of the triacylglycerols were more unsaturated than those of either
diacylglycerol and the free fatty acids had the lowest content of unsaturated acids (Table 3)
unless unsaturated acids were added to the medium. The data from the sample analysed show
these differences well and also show a high content of 1 :3 diacylglycerol, though this was usually
3-5 %.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
32
P . KEMP, D . I . LANDER AND C . G . ORPIN
The two unknown minor components were recovered at least 95% pure from the plates and
examined by GLC on SE 30 at 240 "C. Their retention times relative to cholesterol were 0-54( R F
by TLC 0.91) and 2-66 (RFby TLC 0.35). Both samples were examined by mass spectrometry
using a gas chromatograph inlet. The faster running component gave a molecular ion (M +)at an
integral mass of m/e 410 suggesting a molecular formula of C30H50. We identified this
component as squalene, since the fragmentation pattern of squalene and the unknown were
indistinguishable and co-chromatography was demonstrated by TLC and GLC.
The second unknown component gave a molecular ion at m/e 428-4098, which is consistent
with a formula of C30H51 0 H(M+ calculated = 428.4018). Ions present at integral masses of m/e
410 (M-18), 413 (M-15) and at 395 [M-(18 + l5)] suggest the loss of both H 2 0 and a methyl
group. Two very prominent ions at 191 (base peak) and 207 (40% of base peak), the latter
retaining the oxygen atom, suggested a saturated pentacyclic triterpenol (see Budzikiewicz et al.,
1964). The fragmentation pattern was very similar to that of tetrahymanol run on the same day.
The unknown and tetrahymanol co-chromatographed on TLC and GLC. The 400 MHz lH
NMR spectrum of the unknown and tetrahymanol were identical both in the shifts of the eight
methyl singlets and in the appearance of the ill-resolved 26 proton methylene region. While the
spectra cannot be interpreted in detail, it seems safe from this 'fingerprint' test to conclude that
the unknown is indeed tetrahymanol.
It is probable that squalene and the triterpenol replace sterols in the membranes of these
fungi. Some other fungi do not have sterols, but the aerobic phycomycetes do (Ourisson &
Rohmer, 1982) and, though Saccharomyces cerevisae can grow anaerobically, it requires sterol
synthesized during earlier aerobic growth (Goodwin, 1973; Brennan et al., 1974). Triterpenols
are typical prokaryotic constituents and rare in eukaryotes (Ourisson & Rohmer, 1982).
Fatty acids
In all analyses of fatty acid methyl esters by GLC, 95-97% of esters eluted from either SE 30
or PEG A, were in the series containing even numbers of carbon atoms, the straight-chain series
with from 12 to 24 carbons, or the mono-unsaturated series with from 16 to 24 carbons (see
Tables 2 and 3). Minor components were tentatively identified as branched chain, odd-carbon
straight-chain, hydroxy- and 0x0-acids and occasionally an octadecadienoic acid. All probably
have their origins in the rumen fluid used in the medium. Analyses revealed an approximate 1:1
relationship between saturated and unsaturated fatty acids in cultures not supplemented with
fatty acids.
Geometry and position of double bonds
When the fatty acid methyl esters were separated on silica gel plates impregnated with silver
nitrate, three well-separated bands were detected. By comparison with standards, the three
bands should have contained saturated, trans and cis fatty acid-methyl esters, respectively. The
fastest running band contained only saturated esters, and the slowest band contained the esters
of the 16:1, 18:l and 20:l fatty acids. The middle band, expected to contain the transunsaturated esters, contained a small amount of 18 :1 and the unsaturated esters 22 :1 and 24 :1.
Since rechromatography with authentic standards confirmed that cis 24 : 1 (nervonate) cochromatographed with methyl elaidate, we concluded that all the monoenes were of cis
configuration except the small trans 18 :1 component. The greater RFof the 22 :1 and 24 :1 esters
was because of their lower polarity conferred by the longer paraffin chain. This conclusion was
confirmed by the low infra-red absorption in the region of 960 cm- (trans double bond) which is
accounted for by the 18:l in the faster monoene band.
Gas-chromatographic analysis of the products obtained by oxidizing the unsaturated acids
with periodate/permanganate (von Rudloff, 1956) indicated that most of the unsaturated acids
were of the 0 9 series (Table 4). However, no 7-dicarboxylic acid was recovered, suggesting that
the palmitoleic is 0 7 and may arise from the yeast extract in the medium. This probably
accounts for the high recovery of C9 dicarboxylic acid (83.4%) (Table 4) as only 79% was
expected from the cis and trans 18 :1. The yield of the CI3-dicarboxylicacid is low if the parent
C22:1is all 09, but this is the only discrepancy unexplained by the results.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
Piromonas communis lipids
33
Table 4. Double bond positions in the total unsaturated fatty acids of P. communis
Total fatty acids, obtained by KOH hydrolysis, were methylated and separated by argentation TLC into
saturated and two other bands designated cis and ‘trans’.These were analysed by GLC, the unsaturated
esters were oxidized (von Rudloff, 1956) and the dicarboxylic acids methylated and analysed by GLC.
The only true trans acid was trans 9; 22: 1 and 24: 1 run faster than other cis acids because of their
reduced polarity. Values are given as percentage weight of material recovered from GLC.
Monoenoic
fatty acid
16: I
18:l
20: 1
22 : 1
24: 1
cis
‘trans’
band
A position
expected if 0 9
Dicarboxylic
acid isolated
A position
found
3.2
76.4
5.0
-
7
9
11
13
15
C, None
C, 82.8 + 0-6
c,1 4.9
C I 30.4
C I S11.3
9
9
11
13
15
band
-
0.6
-
2.6
12.2
Since these results suggested desaturation of stearic acid followed by chain elongation, we
checked one culture for octadecenoic acid content before and after 20 h growth. There was an
increase of 8.6 times the concentration of octadecenoic acid originally present, of which some
may have been trans. This indicated that synthesis of stearate and desaturation of stearate to
oleate was a possibility. To check that oxygen was not involved in the desaturation, we used
culture media which were fully reduced (using resazurin as indicator) and incubated the cultures
in an anaerobic chamber. Transfer through five sub-cultures followed by extraction under
anaerobic conditions, revealed no change in the unsaturated fatty acid content. Cyanide at
M had no effect on the proportion of unsaturated fatty acids or on growth. Neither rotenone
(
M) nor 2-n-heptyl-4-hydroxyquinoline-N-oxide
(
M) were inhibitory (Mahler &
Cordes, 1966) and we could find no spectroscopic evidence for cytochromes. We conclude that
oxygen, nitrate and sulphate are not used as electron acceptors. These findings suggest the
presence of an anaerobic A9 desaturase similar to that suggested by Shapiro & Wertheimer
(1943), but unsubstantiated by other workers (Bloomfield & Bloch, 1960). We are pursuing these
findings in an effort to establish the identity of the terminal electron acceptor. Oleic acid could
also be formed by elongating the cis-dodec-3-enoic acid derived from the 3-hydroxy-dodecenoic
acid of the fatty acid synthetase, but this pathway has only been described in anaerobic bacteria
(Gurr, 1974).
Labelling of fatty acids and complex lipids with l4C-Iabelled precursors
Orpin & Letcher (1979) showed that growth in vitro of Neocallimastix frontalis could be
stimulated by some unsaturated fatty acids. Oleic acid, present in rumen fluid in small quantities
compared to its 11-trans isomer is the major unsaturated fatty acid found in all lipids of P.
communis. This finding prompted us to examine the source of the fatty acids using 1-14C-labelled
precursors, and precursors of choline [a limiting factor in sheep nutrition (Neil1 et al., 1979)]
since phosphatidylcholine represents 25 % of the total phospholipids.
Analyses of cells grown with 1-14C-labelled precursors showed that all long-chain fatty acids
were well taken up and incorporated into cell lipids in the range 53-81 % of the added substrate
(Table 5). There were different yields of cells and some increased uptake occurred with greater
cell weights. However, since it was not possible to ensure that the distributions of the substrates,
which are insoluble, were identical, the meaning of the finding is uncertain. The uptake of
labelled linoleic acid was halved by diluting the [ l-14C]linoleicacid with unlabelled linoleic acid
(20 pg ml- l) and a similar result was found with a-linolenic acid. Acetate and glucose were
poorly incorporated into lipids, but both these substrates are effectively diluted ; acetate is
diluted by acetate released by the fungi into the medium and glucose, by glucose derived from
the cellobiose in the medium.
[U-14C]Glucose(Table 5 ) gave the most uniform labelling of fatty acids. All fatty acids except
palmitoleic acid were labelled and the proportion of total activity in each acid was very close to
that contributed by the mass of each acid to the total fatty acids. Label from [U-14C]glucosewas
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
0.6
Glucose
-
14.0
9.8
15.0
6.6
5.0
96.0
-
-
93.6
18.6
25.0
-
1.0
4.0
57.6
-
9-3
11-0
4.4
4-4
-
-
1.2
0.9
1.1
tr
-
-
29.0
64.0
-
c,t,c conj.
-
0.4
tr
tr
-
3.8
0.5
-
-
-
-
1.28
2.5
3.2
3.9
tr
-
3.7
1.5
c,c,c 18:3
1.1
38.2
41.8
36.0
81.0
-
-
-
N o labelling of fatty acids
t,t conj.
0.73
c,t conj.
21.9
40.0
0
1-6 12.8
-
-
-
-
-
0.3
tr
3.1
-
-
2.6
0.5
1.3
tr
-
1.7
-
-
1.1
6.2
6.6
13.2
tr
-
-
-
c,t,t/t,t,c conj.
1.23
1-3
1.6
1.8
tr
-
-
-
-
3
24:O 24:l
+ FA of P and G
+ FA of P and G
+ FA of P and G
PS 30%, PE 30%, unidentified 40%
PE 60%, PC 30%, unidentified 10%
PC only (by TLC)
FFA
FFA
FFA
FFA + FA of P and G
FFA + FA of P and G
FFA + FA of P and G
FFA + FA of P and G
FFA + FA of P and G
FFA + FA of P and G, traces of
squalene and triterpenol
FFA + FA of P and G (86%)
squalene ( 5 7 3 , triterpenol (9%)
Lipids labelled?
* All the long chain fatty acids used were I-'sC-labelled, glucose and serine were U-14C-labelled,acetate and ethanolamine 2-14C-labelledand choline was MeI4C-labelled. I4C-Labelledprecursor (5 pCi) was added to the medium (50 ml) and in the case of linoleic and a-linolenic acids, bottles with 1 mg unlabelled carrier
were examined.
t FFA, fatty acids; FA, 0-acyl-bound fatty acids; P, phospholipids; G, glycerides; PS, phosphatidylserine; PE, phosphatidylethanolamine; PC,
phosphatidylcholine.
$ Configuration of double bonds: c, cis; t , trans; conj., one pair of double bonds conjugated.
Configuration of double bonds 1 . . . c,c 18:2
73.7
Linoleic acid (all cis)
79.9
55.6
+ 1 mg carrier
34.2
a-Linolenic acid (all cis)
+ 1 mg carrier
42-0
Serine
Ethanolamine
Choline
0.1
Mass <% weight) . . . 8.4 10.2 16.8
Relative specific activity
(% activity/% mass) . . . 0.79 0.96 1.1
60.7
53.4
55.5
31-0
81-5
2-0
0.1
tr
30.4
98.0
Fatty acids labelled (% of incorporation)
1
4
c
L
Incorporated
incell lipids 12:O 14:O 16:O 16:l 18:O 18:l 20:O 20:l 22:O 22:l
Lauric acid
Myristic acid
Palmitic acid
Stearic acid
Oleic acid
Acetate
I4C-Labelled
precursor*
Fatty acid methyl esters of complex lipids were prepared by hydrolysing with 6% KOH in 95% aqueous MeOH (w/v/v). The fatty acids were extracted
with hexane after acidification and methylated using fresh diazo-methane. Methyl esters were separated by GLC and the effluent was split in the ratio
9 : 1. One part was analysed for mass using a flame ionization detector and nine parts were combusted to C 0 2 and HzO by passing over heated copper
oxide; the H 2 0was reduced to Hz over heated iron filings and the 14C02was counted in a gas-flow counter coupled to a dual-pen chart-recorder and a
scaler. tr, Detectable but too small to quantify, ~ 0 . 1 % .
Table 5 . Uptake and incorporation of I4Cfrom l4C-labe1ledprecursors into lipids of P . communis
P
w
Piromonas communis lipids
35
also detected in squalene (5%) and the triterpenol (9%). 14C from [l-14C]acetate was
incorporated only in trace amounts into acids with more than 18 carbons. Squalene and the
triterpenol were also labelled. Chain elongation of saturated fatty acid precursors and oleic acid
was the most noticeable feature when cultures were grown with these substrates. When
incubated with [ 1-l4C]palmitate, 3.1 % of the total activity in the fatty acids was recovered in
oleate, presumably produced by the desaturation of stearate. Added stearic acid was not
desaturated but was converted to longer chain saturated homologues. Palmitoleic acid, though
detectable in the mass traces, was never found labelled and we assume it is taken up from the
medium. It may be that in the rumen the fungi utilize many available exogenous fatty acids. If
this source becomes inadequate, the deficiency may be met by synthesis. In situations such as a
very high roughage diet exogenous lipid would be in short supply.
The exceptions to chain elongation were linoleic and a-linolenic acids, which were extensively
incorporated into all fractions. Incorporations into free fatty acids, acylglycerols and
phospholipids, were 18,45 and 37 %, respectively, for carrier-free [ 1-14C]linoleicacid and 63, 18
and 20% with 20pg carrier ml-l; for [l-14C]linolenic acid with 20 pg carrier ml-'
incorporations were 49,37 and 18%, respectively. These polyunsaturated acids were extensively
conjugated and some cisltrans isomerization was detected by TLC and GLC. The free fatty acid
fraction was enriched with conjugated isomers relative to the complex lipids; with carrier-free
linoleic acid as substrate the ratio of non-conjugated diene to conjugated diene in the free fatty
acids was 36 :64 and in the complex lipids, 82 :18. When carrier was used, the values were 43 :57
and 77 :23, respectively. For a-linolenic acid with added carrier the values for non-conjugated to
conjugated were 53 :47 in the free acids and 75 :25 in the complex lipids.
Since conjugation of both linoleic and a-linolenic acids is the first step in the bacterial
hydrogenation of these acids to stearic acid (Dawson & Kemp, 1970), we looked carefully for
evidence of hydrogenation. Freezing and thawing destroyed conjugation activity in P .
communis, but not with bacterial hydrogenators. No increased activity was found in fungal
cultures grown without antibiotics. The conjugation of linoleic and a-linolenic acids may
marginally reduce the potential damaging effects that polyunsaturated acids have on
membranes.
[Me-14C]Choline was rather poorly (0.1 %) incorporated by cells during growth and only
phosphatidylcholine was labelled. Compared with this, 13% of the [2-l4C]ethano1amine was
incorporated into complex lipids, 60% into phosphatidylethanolamine and 30% into
phosphatidylcholine. The phosphatidylcholine was presumably generated by the sequential
methylation of phosphatidylethanolamine (Bremer & Greenberg, 1961). [U-14C]Serine was
poorly incorporated (0.45 % of added label); phosphatidylserine (30%) and phosphatidylethanol
amine (30%) were the only identified labelled lipids. The label found in phosphatidylcholine
was no more than twice background. This was unexpected, since, if the methylation pathway
from phosphatidylethanolamine to phosphatidylcholine (Bremer & Greenberg, 1961) operates
when ethanolamine labels the phosphatidylethanolamine, then phosphatidylethanolamine
derived by the decarboxylation of phosphatidylserine should also undergo methylation unless
there are two pools of phosphatidylethanolamine.No label from the three bases was found in
fatty acids, but unidentified minor lipids were labelled from serine and ethanolamine. The
ability to synthesize choline may help the fungi to survive in an environment where choline is
usually in short supply even on choline-rich diets (Neill et al., 1979). Choline not taken up by
protozoa is rapidly converted to trimethylamine and the methyl groups are eventually lost as
methane (Neill et al., 1979). Choline is only freely available just after feeding when the fungal
zoospores germinate, but they may not take up sufficient choline for their growth requirements.
Ethanolamine and serine would be present at low levels in the rumen (Mangan, 1972) and
available for the synthesis of phosphatidylserine and phosphatidylethanolamine,which could be
converted to phosphatidylcholine to provide for the requirements of the fungal membrane
synthesis.
The rumen fungi examined have very similar fatty acid, squalene and triterpenol content.
Experiments comparing N. frontalis with P . communis suggest that differences in fatty acid
and complex lipid composition are within the experimental variability of the methods.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
36
P . K E M P , D. J . L A N D E R AND C . G . O R P I N
Zoospores have also been examined and these are very similar in lipid composition, if not
identical, to the parent organism.
Both N.frontalis and P . communis contain the cis,24 : 1 (A 5 , fatty acid and the triterpenol. We
have investigated the possibility of correlating concentration of the total rumen cis, 24 :1 (A15,
fatty acid and the triterpenol with the numbers or mass of fungi in the rumen. We have found
some twofold variability of the C24:1content of the fungi even when grown under laboratory
conditions, and there are dietary sources of long-chain fatty acids (Body & Grace, 1983). On
PEG A, but not SE 30 GLC columns, methyl hydroxystearate may interfere with estimations.
The triterpenol has a long retention time on SE 30 and is eluted after much of the dietary
sterol, but it is a relatively small component in the rumen sterol fraction. There may be dietary
sources of this compound, but, if these could be eliminated, the use of specific ion monitoring by
mass-spectrometry might provide a method for assessing rumen fungal biomass.
We thank Dr J . K. M. Sanders of the Chemistry Department, University of Cambridge for the interpretation of
the NMR spectra and also John Eagles of the Mass Spectrometry Group, Food Research Institute, Norwich, UK.
REFERENCES
BEISS, U. ( 1964). Zur papierchromatographischen
determining individual phospholipids in a complex
Auftrennung von Pflanzenlipiden. Journal of Chromixture by successive chemical hydrolyses. Biomatography 13, 104-1 10.
chemical Journal 84, 497-501.
BLOOMFIELD,
D. K. & BLOCH,K. (1960). The formation
FOLCH,J ., LEES,M. & SLOANE
STANLEY,
G. H. (1 957).
of A9-unsaturated fatty acids. Journal of’ Biological
A simple method for the isolation and purification of
Chemistry 235, 337-345.
total lipids from animal tissues. Journal of Biological
BODY,D. R. & GRACE,N. D. (1983). Changes in the
Chemistry 226, 497-509.
GAVER,
R. C. & SWEELEY,
C. C. (1965). Methods for
composition and state of unesterified ClJ-C32 fatty
acids during their passage through the ovine digesmethanolysis of sphingolipids and direct determination of long-chain bases by gas chromatography.
tive tract. Journal of the Science of Food and
Journal of the American Oil Chemists Society 42,294Agriculture 34, 8- 12.
298.
D. M. (1961). Methyl
BREMER,J. & GREENBERG,
GOODWIN,
T. W. (1973). Comparative biochemistry of
transferring enzyme system of microsomes in the
sterols in eukaryotic microorganisms. In Lipids and
biosynthesis of lecithin (phosphatidylcholine). BioBiomembranes of Eukaryotic Microorganisms, pp. 1 chimica et biophysica acta 46, 205-2 16.
36. Edited by J. A. Erwin. London : Academic Press.
D. M. &
BRENNAN,
P. J., GRIFFIN,P. F. S., L~SEL,
GRAY,G. M. (1976). Gas chromatography of the longTYRELL,
D. (1 974). The lipids of fungi. Progress in the
chain aldehydes. In Lipid ChromatographicAnalysis,
Chemistry of’ Fats and Other Lipids 14, 51-89.
2nd edn, vol. 3, pp. 897-923. Edited by G. V.
D. H.
BUDZIKIEWICZ,
H., DJERASSI,C. & WILLIAMS,
Marinetti. New York: Marcel Dekker.
( 1964). Structural Elucidation of Natural Products
GURR,M. 1. (1974). The biosynthesis of unsaturated
by Mass Spectrometry, vol. 1, pp. 121 - 1 39. San Franfatty acids. In Biochemistry of Lipids, pp. 181-235.
cisco: Holden-Day.
Edited by T. W. Goodwin. London: Butterworth.
CHRISTIE,W. W. ( I 973). Lipid Analysis. Oxford :
Pergamon.
D. (1975). The
KEMP,P., WHITE,R. W. & LANDER,
CLARKE,
N. G. & DAWSON,
hydrogenation of unsaturated fatty acids by five
R. M. C. (1981). Alkaline
O-N-transacylation. A new method for the quanbacterial isolates from the sheep rumen, including a
titative deacylation of phospholipids. Biochemical
new species. Journalof General Microbiology 90,100Journal 195, 301-306.
114.
R. M. C. (1976). Analysis of phosphatides
DAWSON,
LENG, R. A. (1982). Dynamics of protozoa in the
and glycolipids by chromatography of their partial
rumen of sheep. British Journal of Nutrition 48, 339hydrolysis or alcoholysis products. In Lipid Chroma415.
tographic Analysis, 2nd edn, vol. 1, pp. 149-172.
LINDSAY,
D. B. & LEAT,W. M. F. (1977). Oxidation
Edited by G. V. Marinetti. New York: Marcel
and metabolism of linoleic acid in fed and fasted
sheep. Journal of Agricultural Science 89, 2 15-221.
Dekker.
MAHLER,
H. R. & CORDES,E. H. (1966). Biological
DAWSON,
R. M. C. & KEMP,P. (1967). The aminoChemistry, 2nd edn, p. 686. New York: Harper &
ethylphosphonate-containinglipids of rumen protoRow.
zoa. Biochemical Journal 105, 837-842.
MANGAN,J . L. (1972).Quantitative studies on nitrogen
DAWSON,
R. M. C. & KEMP,P. (1970). Biohydrogenametabolism in the bovine rumen. British Journal of
tion and dietary fats in ruminants. In Physiology of
Nutrition 27, 261-283.
Digestion and Metabolism in the Ruminant, pp. 504MU”, E. A., ORPIN,C. G. & HALL,F. J. H. (1981).
5 18. Edited by A. T. Phillipson. Newcastle-uponUltrastructural studies of the free zoospore of the
Tyne: Oriel Press.
DAWSON,
R. M. C., HEMINGTON,
N . & DAVENPORT, rumen phycomycete NeocaI1imasti.ufrontalis. Journal
of’ General Microbiology 125, 3 1 1-323.
J . B. (1962). Improvements in the method of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02
Piromonas communis lipids
NEILL,A. R., GRIME,D. W., SNOSWELL,
A. M.,
NORTHROP,
A. J., LINDSAY,
D. B. & DAWSON,
R. M.
C. (1979). The low availability of dietary choline for
the nutrition of sheep. Biochemical Journal 180,559565.
ORPIN,C. G. (1977~).The rumen flagellate Piromonas
comrnunis: its life-history and invasion of plant
material in the rumen. Journal of General Microbiology 99, 107-1 17.
ORPIN,C. G. (1977b). The Occurrence of chitin in the
cell walls of the rumen organisms Neocallimastix
frontalis, Piromonas communis and Sphaeromonascommunis. Journal of General Microbiology 99, 2 15-2 18.
ORPIN,C. G. (1981). Fungi in ruminant degradation.
In Agricultural Science Seminar :Degradation of Plant
Cell-wall Material, pp. 36-47. London : Agricultural
Research Council.
ORPIN,C. G . & LETCHER,
A. J. (1979). Utilization of
cellulose, starch, xylan and other hemicelluloses for
growth by the rumen phycomycete Neocallimastix
frontalis. Current Microbiology 3, I2 1- 1 24.
OURISSON,
G. & ROHMER,M. (1982). Prokaryotic
polyterpenes: phylogenetic precursors of sterols. In
Current Topics in Membrane Transport 17, 153- 182.
Edited by F. Bonner & A. Kleinzeller. New York:
Academic Press.
37
PARKER,
F. & PETERSON,
N. F. (1965). Quantitative
analysis of phospholipids and phospholipid fatty
acids from silica gel thin-layer chromatograms.
Journal of’ Lipid Research 6, 455-460.
ROPER,R. & MA,T. S. (1957). Diazomethane as a
reagent for microsynthesis. Microchemical Journal 1,
245-260.
VON RUDLOFF,E. ( 1 956). Periodate-permanganate
oxidations. V. Oxidation of lipids in media containing organic solvents. Canadian Journal of Chemistry
34, 1413-1418.
SCHLENK,
H.& GELLERMAN,
J. L. (1960). Esterification of fatty acids with diazomethane on a small
scale. Analytical Chemistry 32, 1412-1414.
SHAPIRO,B. & WERTHEIMER,
E. (1943). Fatty acid
dehydrogenase in adipose tissue. Biochemical Journal
37, 102-104.
VASKOVSKY,
V. E. & KOSTETSKY,E. Y. (1968).
Modified spray for the detection of phospholipids on
thin-layer chromatograms. Journal of Lipid Research
9, 396.
A. F. (1974). Passage of
WELLER,R. A. & PILGRIM,
protozoa and volatile fatty acids from the rumen of
sheep and from a continuous in ritro fermentation
system. British Journal of Nutrition 32, 341-35 1.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 18:04:02