—— —-~—v

 

 

THE {SOLATION AND? IDENTEFiCAfiON OF
TYPE-SPECIFRC UPEDS FROM MYCOBACTERM

Thesis for the beam of Ph. D.

MICHIGAN STATE UNIVERSiTY

Robert Wyman Walker
€9.63

THESYS

This is to certify that the

thesis entitled
The Isolation and Identification of Typeuspecific Lipids
from Mycobacteria

presented by

Robert Hyman Walker

has been accepted towards fulfillment
of the requirements for

Doctor of Philosoph¥_______dqpeein________ujcrobioiogy

1 Y
101' professor

M

Date September 3, 1963

 

LIBRARY

Michigan State
University

 

ABSTRACT

THE ISOLATION AND IDENTIFICATION OF TYPE-
SPECIFIC LIPIDS FROM MYCOBACTERIA

by Robert Wyman Walker

The isolation and identification of type-specific
lipids from mycobacteria. 1963. - Ethanol-ether extracted
lipids from 25 strains of mycobacteria were fractionated
by adsorption chromatography. Infrared spectra of the frac-
tions were recorded. Type-specific lipid compounds were
found in the extracts of human, bovine, avian and atypical
strains.

Dimycoceronate of phthiocerol was found in the lipids
of two human strains and one bovine strain. Mycoside B was
also isolated from the bovine strain. Other type-specific
lipids isolated were: mycoside A from photochromogenic
strains, mycoside F from an E, fortuitum strain, mycoside C
from g, gyigm_strains, mycoside D from a strain isolated
from a bovine mesenteric lymph node and mycoside CM from a
nonphotochromogenic laboratory strain and from 12 organisms

isolated from swine mesenteric lymph nodes, bovine body

lymph nodes and bovine Peyer's patches.

THE ISOLATION AND IDENTIFICATION OF TYPE-

SPECIFIC LIPIDS FROM MYCOBACTERIA

BY

Robert Wyman walker

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Microbiology and Public Health

1963

K:- d I I t/ ‘4
..., . / .
I/// VC/ 'J 4

Acknowledgements

The author would like to express his thanks to the many
who have freely contributed advice and a helping hand facil-
itating the successful completion of this work: Drs. W. L.
and Virginia Mallmann, Dr. James A. Ray, Dr. J. R. Brunner,
Dr. R. S. Lipe and the technical help of the Michigan State

University Bovine Tuberculosis Project.

ii

Table of Contents

Page
Introduction . . . . . . . . . . . . . . . . . . . . 1
Historical . . . . . . . . . . . . . . . . . . . . . 3
Adsorption chromatography . . . . . . . . . . . 4
Atypical or anonymous mycobacteria . . . . . . . 6
Infrared spectroscopy . . . . . . . . . . . . . 7
Infrared spectroscopy of mycobacterial lipid
fractions and lipid compounds as a means of
strain identification . . . . . . . . . . . . . 10
Types of lipids present in mycobacteria . . . . l4
Odor producing compounds 20
Pigments 20
Glycerides 21
Fatty acids 22
Glycolipids 24
Type-specific glycolipids 26
Gas chromatography of mycobacterial fatty
acids 28
Materials and Methods . . . . . . . . . . . . . . . 30
Gas chromatography . . . . . . . . . . . . . . . 35
Organisms used in study . . . . . . . . . . . . 38
Results . . . . . . . . . . . . . . . . . . . . . . 42
Discussion . . . . . . . . . . . . . . . . . . . . . 99

iii

Page
Gas chromatography . . . . . . . . . . . . . . . 103

summary 0 O O O O O O O O O O O O O O O O O O O O O 105

Literature Cited 0 O C O O O O O O O C O O O O O O O 106

iv

List of Tables

Page

TABLE 1. Weight of lipid eluted from chromato-

graphic column . . . . . . . . . . . . . . 57
TABLE 2. Order of elution of mycobacterial lipid

components from a Florisil-packed column . 72
TABLE 3. Approximate percentages of fatty acids

(C8 to C19) present in the ether-acetic

acid r% eluted lipid fraction of five

strains . . . . . . . . . . . . . . . . . 75
TABLE 4. Approximate percentages of total lipid

recovered eluted by each solvent series

for 1]- Strains O O O O O O O I O O O O O O 76

FIGURE
1.
2.

FIGURE

0‘ U1 wk on N
0

FIGURE

uh b) N H
e o 0

List of Figures

 

Page
1. Infrared spectra . . . . . . . . . . . . 78
P-8 ether-ethanol extract minus phosphatides

Ravenel ether-ethanol extract minus
phosphatides

P-8 diglycerides-saturated fatty acids,
Smith's "C"

B.C.G. monoglyceride, Smith's "D"

Ravenel long chain unsaturated hydroxy
esters (Noll, 1957)

2. Infrared spectra . . . . . . . . . . . . 79
M, bovis - Ravenel ethyl esters of fatty

acids

380$ ethyl esters of fatty acids

M, bovis - Ravenel Kx napthoquinone

IS

. tuberculosis - H37Rv Kx napthoquinone

. bovis - B.C.G. triglyceride

l3 HZ

. tuberculosis - H37RV diglyceride

3. Infrared spectra . . . . . . . . . . . . 80

P-39 glycerol monomycolate
P-39 long chain fatty acids
l86C—1 shorter chain fatty acids

M. tuberculosis - H37R dimycoceronate
_ . A
of phthiocerol

P-8 Mycoside A

jg. bovis - Ravenel Mycoside B

vi

FIGURE

vBllOH-l Mycoside C

4. Infrared spectra

172C1-l JABS

93C-0 Mycoside CM contaminated with fatty

acids

3808 Mycoside CM

_M, avium - Mycoside C

158C-O Mycoside C
7lC-O Mycoside D

H37Rv phosphoglycolipid

5. Infrared spectra

172C -1 amide-like compound

1

l7lC-1 amide-like compound
186C-l amide-like compound
35 foruitum probably Mycoside F

P-4 phospholipid
186C-l Mycoside CM

6. Infrared spectra

2441-1 Mycoside CM
62D-O MycoSide CM
152 C-1 Mycoside CM
193C2-1 Mycoside CM

7. Infrared Spectra

M
17lC-l Mycoside CM
68C-0 Mycoside CM
186C-l Mycoside CM
P-39 Mycoside CM
172C1-1 Mycoside CM

Vii

Page

81

82

83

84

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

M, avium elution curve . . . . . . .

2441-1 GIUtion curve 0 o o o o o o o

68C-0 elution curve . . . . . . . .
172C1-1 elution curve . . . . . . .
93C-O elution curve . . . . . . . .
62D-O elution curve . . . . . . . .
186C-l elution curve . . . . . . . .
fl, foruitum elution curve . . . . .
193C2-1 elution curve . . . . . . .
193C1-1 elution curve . . . . . . .
M, tuberculosis H37RV elution curve
P-39 elution curve . . . . . . . . .
M, Qggi§_B.C.G. elution curve . . .
.M. tuberculosis H37RA elution curve
P-4 elution curve . . . . . . . . .
3805 elution curve . . . . . . . . .

P-8 6111121011 cun’e o o o o o o o o o

P-lS elution curve . . . . . . . . .

viii

Page

85

85

86

86

87

87

88

88

89

89

90

9O

91

91

92

92

93

93

Page

FIGURE 26. 228C-1 elution curve . . . . . . . . . . 94
FIGURE 27. 7lC-0 elution curve . . . . . . . . . . 94
FIGURE 28. BllOH-l elution curve . . . . . . . . . 95
FIGURE 29. 152C-1 elution curve . . . . . . . . . . 95
FIGURE 30. 17lC-l elution curve . . . . . . . . . . 96
FIGURE 31. 158C-0 elution curve . . . . . . . . . . 96
FIGURE 32. M, boyi§_Ravenel elution curve . . . . . 97

FIGURE 33. Tracings of gas chromatographic
experiments . . . . . . . . . . . . . . 98

ix

Introduction

Mycobacterial lipids have attracted much attention in
the past. Immunologists, medical bacteriologists, pathol-
ogists, and biochemists have examined the abundant lipid
materials of tubercle bacilli in efforts to explain the
elusive pathogenic nature of these organisms. It has, there-
fore, been only a matter of time when systematists would
also study the lipids in an effort to exploit differences
facilitating classification of the mycobacteria.

The advent of the newer techniques of lipid chemistry,
adsorption chromatography and infrared spectroscopy has been
very careful for the examination of lipids. The first of
these techniques, adsorption chromatography, has made pos-
sible the separation of lipids into relatively pure com-
pounds which by previous techniques was impossible, or at
best, extremely difficult. The second technique, infra-red
spectroscopy, has simplified characterization and identifi-
cation of lipid compounds.

With the aid of infrared spectroscopy and adsorption
chromatography the author has attempted to isolate and
identify lipid compounds from twenty-five strains of myco—

bacteria. Particular emphasis has been placed upon

identification of so-called specific lipids, compounds of
possible utility in classifying the atypical organisms
studied.

Among the strains studied were included virulent and
avirulent human and bovine strains. In addition, avian and

atypical organisms from human, bovine, and porcine sources

were studied.

Historical

Tubercle bacillary lipids have been the object of extensive
research. Since Hammerschlag (1889) reported that the
tubercle bacillus contained 28% lipid, these componenets
have attracted much attention. Additional impetus was af-
forded when Sternberg (1902) demonstrated that the lipids
play some part in pathological responses. Defatted bacilli
were less pathogenic.

Subsequent investigations were concerned primarily with
determining the crude lipid content of cells of various
strains of mycobacteria. Hampered by lack of lipid material
and a lack of adequate techniques of lipid chemistry, de-
tailed chemical studies were few.

In the 1930's and 1940's knowledge of the chemical con-
stituents of mycobacterial lipids was advanced by the clas-
sical research of Dr. R. J. Anderson and associates at Yale
University. Under the sponsorship of the Medical Research
Committee of the National Tuberculosis Association these
workers obtained large volumes of lipid materials for their
studies (Anderson, 1940).

Using involved, tedious techniques of classical lipid

chemistry these workers were able to identify and characterize

many new compounds. A new series of liquid saturated fatty
acids, a number of higher alcohols, a pigment and optically
active hydroxy acids of high molecular weight were reported
(Anderson, 1940).

Anderson (1940) suggested that the neutral fats were
not glycerides but fatty acid esters of trehalose*. The
phosphatides were combinations of fatty acids, organic phos-
phorus acids, and a phosphate containing glycoside. The
glycoside yielded inositol and mannose on hydrolysis. The
waxes were a complex mixture unique for each type of bacil-
lus. Some waxes were compounds of hydroxy acids and phos-
phate containing polysaccharides, others complex glycerides,
while others were esters of hydroxy acids with trehalose.

With the advent of the newer techniques of lipid chem-
istry such as adsorption and gas chromatography and infra-
red spectroscopy, tremendous advances have been possible
in attempting to elucidate the complex chemical structure

of mycobacterial lipids.

Adsorption Chromatography
The technique of adsorption chromatography has greatly

facilitated the study of lipids (Asselineau, 1952; Smith

 

*Asselineau (1952) and others (Kubica et al., 1956;
Asselineau and Moron, 1958) have shown mycobacteria to have
a high glyceride content in the neutral fat fraction.

et al., 1960a). Separations of complex lipid mixtures pre-
viously difficult or impossible are now performed with ease.
This technique relies upon the differing affinities of
lipid classes (and lipid compounds) for the surface of a
finely divided insoluble material of large surface area

(the adsorbent) (Lederer and Lederer, 1958). Percolation
of organic solvents (the eluants) of increasing polarity
sequentially through the adsorbent causes displacement of
the less tightly held lipids with subsequent elution from
the adsorbent.

Collection of sequential small fractions of solvent and
eluted lipid serve to separate the lipid components as they
are eluted from the column. The presence of lipid in the
eluate may be determined by weighing fractions from which
solvent has been removed. Other quantitating techniques
such as phosphate determination, sterol determination, and
so forth may also be used in special cases.

Depending upon the adsorbent and types of eluants used,
complex lipid mixtures may be fractionated into classes or,
in certain cases, individual compounds. Silicic acid or
Florisil (a synthetic aluminum silicate) will separate a
lipid mixture into the following classes, in order of

elution: hydrocarbons, fatty acid esters, saturated fatty

acid—triglycerides, unsaturated triglycerides, diglycerides,
monoglycerides, fatty acids, glycolipids, glycolipid pep-
tides, free fatty acids and phosphatides (Carrol, 1961;

Wren 1961).

AtypicalTpr Anonymous Mycobacteria

The isolation of atypical or anonymous mycobacteria
from human tuberculosis patients is not new (Alvarez and
Tavel, 1885). Until recently, however, these organisms
were frequently regarded as saprophytes and were discarded.
In some instances the anonymous group of organisms may be
the causative agents of human disease (Runyon, 1960; Wood

et 31.. 1956: Tarshis and Frisch, 1952a’b'c

). As high as
10 to 15% of mycobacterioses may be caused by these
organisms.

Runyon (1959, 1960) has grouped the anonymous acid-fast
organisms into four groups. Group I, the photochromogens,
possess the ability with blue light inducement to form a
yellow pigment. Group II, the skotochromogens, are yellow
or orange pigmented in the light or dark. Group III, the
non-photochromogens, possess no pigmentation. All of these

three groups, in contradistinction to the fourth group, are

moderately slow growing organisms, requiring one to two

weeks for visible colony growth. Group IV, the rapid
growing organisms, are generally non-pigmented and require
three to four days for visible colony formation.

Atypical mycobacteria similar to the Runyon types of
human disease have been isolated frequently by members of
the Michigan State University Bovine Tuberculosis Project
(Mallmann, Mallmann and Ray, 1962). Various morphological
and cytochemical criteria, as well as animal infectivity a
and allergenicity, have been used to study the organisms.
The atypical mycobacteria from bovine and porcine sources
are heterogeneous and should be regarded as highly variable
and adaptable. In agreement with Tarshis and Frisch (1952c)
they also suggest some of the atypicals may occupy a posi-
tion midway between the classical pathogenic and sapro-

phytic acid fast organisms.

Infrared Spectroscopy

Infrared spectroscopy is a useful tool for identifying
or characterizing an organic compound. The infrared spec—
trum of a molecule is definitive and unique, each spectrum
being a "fingerprint," unlike that of any other compound

(Cole, 1956).

Organic molecules and chemical groups composing these
molecules vibrate at frequencies within the infrared range.
The frequency of these vibrations is dependent upon the
strength of the bonding forces and the mass of the atoms
involved. As a beam of infrared light is passed through
a chemical sample, it absorbs energy from those wave lengths
(or frequencies) of light corresponding to the vibrating
frequencies of the molecule. Continuous variation of the
frequency of incident light and synchronous recording of
the intensity of emerging light gives a spectrum showing
absorption bands (Cole, 1956).

Experience has shown that certain atomic groupings (OH,
NH, CO, and so forth) are associated with narrow absorption
regions of the infrared spectrum. The proper interpreta—
tion of such spectra, therefore, gives immediate informa-
tion concerning the presence or absence of constituent
atomic groups (Noll, 1956).

Infrared spectroscopy has been used in an effort to de-
termine chemical differences in both complex and simple
systems. Blout and Mellors (1949) showed differences in
spectra of various types of tissues studied. Blout and
Fields (1948 and 1949) suggested infrared spectroscopy had

possible application in differentiating and identifying

nucleic acids, nucleotides, and nucleosides. Other inves-
tigators have attempted to use infrared spectroscopy to
use infrared spectroscopy to study the amide linkage of
globular proteins (Klotz and Griswold, 1949) and the lipo-
proteins (Freeman et al., 1953).

Many investigators have attempted to use infrared
spectroscopy as a rapid means of identifying bacteria,
yeasts, and viruses. Minor changes in media were found to
alter spectra radically (Stevenson and Bolduan, 1952;
Levine et al., 1953a). Riddle et a1. (1956) and Kenner et
a1. (1958) studied 650 strains from 201 species of 3 genera.
Conditions of cultivation were found of vast importance in
spectra reproducibility. Incubation temperature, time of
growth, medium composition, pH, and methods of seeding the
medium must be rigidly controlled. Sample preparation for
presentation to the instrument was also found to be very
critical.

Other investigators have succeeded in showing spectral
differences with yeasts (Simon and Hedrick, 1955), glycogen
from enteric bacteria (Levine et al., 1953b), and acetic
acxid bacteria (Scopes, 1962). Benedict (1955 studied the

e-‘L=':L""ects of nucleic acid, polysaccharide, and lipid removal

fIVZMm several virus preparations.

10

Infrared spectra of bacteria, viruses, and yeasts are
the cumulative results of spectra of each chemical entity
from the organism involved. Substances present in higher
concentration have a greater effect on the spectrum than
those of lesser concentration. As a consequence, the pres-
ence of minute amounts of a substance is masked by non-
specific background substances (protein, polysaccharides,
and so forth), and detection of slight chemical differences
necessary for species differentiation is extremely difficult.

Separation and study of fractions isolated from organ—
isms may have more utility in differentiation of species.
Such fractions as lipid, polysaccharide, phospholipid, and
similar compounds, devoid of interfering non-specific back-
ground substances, would seem to be of use.

Infrared Spectroscopy of
Mycobacterial Lipid Fractions
and Lipid Compounds as a Means
of Strain Identification

Randall, Smith, Colm, and Nungester (1951) detailed
growth and fractionation procedures which facilitated the
use of infrared spectroscopy in production control of
-R. C. G., (Bacillus of Calmette and Guerin, an avirulent

Whacterium bovis) , other avirulent M. bovis and M. 3;};-

lELEZMMggyiyg cultures. They found that the chloroform soluble

11

fraction and acetone soluble and insoluble portions of this
fraction gave similar spectra within strains and differing
spectra between strains between strains studied.

The authors suggested that with carefully controlled
cultural conditions, reproducibility between production
lots might be possible. Further, they added, the use of
infrared spectroscopy might be useful for the preparation
of reproducible antigenic fractions, useful for differen-
tiation between strains, and useful for the subsequent
elucidation of chemical differences between drug resistant
and sensitive strains or virulent and avirulent strains of
mycobacteria.

Randall, Smith, and Nungester (1952) were able to dif-
ferentiate between avirulent and virulent strains of M,
Egggrcurgsis. Ethanol killed cells were extracted with
methanol-chloroform (1:1). A portion of this extract was
held at 5 C. After removal of the precipitate the extract
was subjected to precipitation at -30 C. Both the soluble
and insoluble fractions exhibited a high degree of repro-
ducibility within strains. A noticeable difference was
observed between strains. One fraction, the -30 C insoluble
fraction, showed the greatest promise for the differentiation

between virulent and avirulent strains of M, tuberculosis.

12

Differentiation between M, boyi§_and M, tuberculosis was
possible by means of infrared spectra of specific lipid
fractions (Smith, Harrell, and Randall, 1954; Harrell,
1954). Fractions studied included the 5 and -30 C chloro—
form-methanol (1:1) insoluble fractions as well as those
obtained by adsorption chromatography and batch elution
techniques. The 5 C chloroform-methanol insoluble fraction
showed no significant difference between the two species
studied. The - 30 C insoluble fraction served to differen-
tiate the bovine from the human strains. An absorption
band at 6.63 u was present in lipids from the former and
not the latter.

As a logical extension of previous work, Kubica, Randall,
and Smith (1956) attempted to isolate the chemical consti—
tuents that were the cause of differences in infrared
spectra of the crude extracts. Fourteen human strains were
studied. From the ether—ethanol (1:1) extracts 9 different
chemical constituents were isolated by adsorption
chromatography.

Lipids from 78 strains of multifarious sources were
fractionated and examined by infrared spectroscopy for the
presence of type specific lipids (Smith et al., 1960a).

The presence of certain specific lipids in the majority of

13

strains studied was confirmed. Further, certain of the
lipids were identified chemically. GB and GA as well as

cord factor are complex glycolipids. JAV and JAT are
glycolipid peptides.

Included in this report were studies to determine en—
vironmental effects on the development of specific lipids.
Culture age, medium composition, and drug resistance or
senstitivity were found to have no effect upon the elabora-
tion of specific lipids.

Smith, Randall, MacLennan, and Lederer (1960b) sug-
gested renaming the 4 type specific glycolipids mycosides.

As a consequence, G became mycoside B, and GA J

B ' AV and JA

T
became mycosides A, C, and D, respectively.

A correlation existed between colony morphology, strain,
and presence of a specific mycoside (Fregnan, Smith and
Randall, 1961). Using oleic acid albumin agar (Difco) for
cultivation, they observed a constancy of colony type and
strain (indicated by the presence of a specific mycoside)
for M, fortuitum and M, kansasii.

Fourteen of the 15 strains of M, kansasii (a group I
atypical organism) showed a K type colony morphology during

the later stages of colony development (after 18 days).

The K colony type was a smooth or wrinkled flat colony

14

possessing a regular or slightly undulated margin with a
central dark spot. The colony was dry in appearance and
was easily suspended in water.

All of the 9 strains of M, fortuitum were of an F type

 

of colony morphology, which was a smooth colony with
raised surface and round, smooth, regular margins. The
colony on gross appearance appeared moist and glistening;
it also was easily suspended in water.

The authors suggest that a close study of colony mor-
phology would be helpful in the early identification of
mycobacteria.

Another type specific glycolipid, mycoside F, was found
in all 8 strains of M, fortuitum studied (Randall, Rao,
Fregnan, and Smith, 1961). Mycoside F, composed 4 fatty
acids and trehalose. had an infrared spectrum similar to
cord factor (trehalose 6-6' dimycolate) but appears markedly
less toxic.

Types of Lipids Present
in Mycobacteria

Most investigators have utilized an extraction procedure
devised by Anderson (1940) and modified by Asselineau (1952)
for initial preparation of the lipid fractions to be

studied (see below).

15

 

 

 

 

 

Cells
alcohol
ether 1:1
extract
ether
F l
soluble insoluble
FATS (1)* crude phosphatide
boiling
acetone
F U
soluble insoluble
WAX A (2) phosphatide (3)

*Components of each fraction are discussed below.

16

Cell Residue I

chloroform

Crude wax

ether-methanol

 

 

 

 

 

 

 

 

i |
soluble insoluble
soft wax (4) purified wax (6)
hot
methanol acetone
' ' I
a
n
I
soluble soluble (7) | insoluble
Wax B (5) Wax C : Wax D (8)
I
I
' saponified
l
I
______._ _._ _.__ _._ _n._ __.
I 1
l . .
. U
Water soluble fraction (9) nsaponifiable
fraction

l
Ealcohol-ether

I i

soluble (10) insoluble (ll)

 

17

Cell residue II

ether-ethanol 1:1
1% H Cl

 

firmly bound lipids

ether-
chloroform

 

 

V j

filtrable (12) unfiltrable (13)

Anderson (1940), Seibert (1950), and Seibert and Soto-
Figueroa (1958) listed compounds and saponification products
found in each of the above fractions isolated from human
strains of mycobacteria.

1. Fats (saponified)

The fats are composed of trehalose, glycerol, a or B
leprosol, cerotic, linoleic, linolenic, oleic, palmitic,
phthioic, hexacosanoic, stearic, and tuberculostearic acids,
and the pigment phthiocol.

2. Wax A (saponified)

Mycolic acid, phthiocerol, phthiodiolone and phthiotriol,

carbonydrate esters of fatty acids, glycerides, and some

18

true waxes make up Wax A. The saturated fatty acids present
in this fraction are palmitic, stearic, hexacosanoic, tu-
berculostearic, phthioic, and a C31 acids. The unsaturated

acids are of the C26 series. Phthiocerol and various poly-
saccharides are also present.

3. Phosphatides (saponified)

The fatty acid components are mostly palmitic. Small
amounts of oleic, palmitoleic and phthioic acids are also
present. The water soluble components are phosphoric acid,
glycerophosphoric acid (or similar compounds) and the
sugars mannose, inositol, and other unidentified hexoses.

4. Soft Wax (saponified)

The soft wax consists of glycerol, phthiocerol, palmitic,
phthioic, mycolic, and stearic acids.

5. Wax B (saponified)

Phthiocerol, phthiotriol, phthiodiolone, mycolic acids,
and glycerides are present in Wax B.

6. Purified Wax

The purified wax is a mixture of glycerides, true waxes,
and fatty acid esters of polysaccharides.

7. Wax C

Wax C like Wax B also contains phthiocerol, phthiotriol,

phthiodiolone, and mycolic acid.

19

8. Wax D

Wax D consists chiefly of lipopolysaccharides containing
amino acids.

9. Water soluble fraction

Mannose, D-arabinose, glucosamine, inositol, and galactose
compose the water soluble portion of the saponified purified
wax.

10. Ether soluble fraction

D-eicosanol-Z, D-octadeconol-Z, phthiocerol, mycoserosic
acid, mycolic, oleic, palmitic, phthioic, stearic, tetra—
cosonoic, and tuberculostearic acids have been identified
in the ether soluble unsaponifiable fractions.

11. Ether insoluble

Mycolic, pentacosanoic, phthoic and tetracosanoic acids
are present in the ether insoluble fraction.

12. Filtrable firmly-bound lipids

Saponified filtrable firmly-bound lipids yield glycerol,
D—eicosanol—Z, D-octadecanol-Z, arabinose, galactose,
inositol, mannose,mycolic and tetracosanoic acids.

13. Unfiltrable firmly-bound lipids

D-eicosanol-Z, arabinose, galactose, inositol, mannose,
and glucosamine have been identified in the saponified un-

filtrable firmly—bound lipids.

20

ngr Producing Compounds

Mycobacterial lipids have a characteristic odor, var-
iously described as sweet, fruity, or perfume-like. Several
aromatic compounds have been isolated. Anisic, phenylacetic,
crotonic, phthalic and salicylic acids in free or ester
form (Asselineau, 1952; Asselineau and Lederer, 1953) are
probably the odor producing compounds. Aldomedone, also
present, may, in addition, also contribute to the odor

(Anderson, 1940).

Pigments

Several pigments give mycobacterial lipids a dark red-
dish color. Anderson (1940) and Ball (1934) isolated a
pigment, red in alkaline solution, yellow in acid solution
which was identified as 2—methyl-3—hydroxy 1, 4-napthoquinone,
phthiocol. Subsequent investigators have shown phthiocol
may be a degradation product (Asselineau, 1952).

N011 (1958) isolated 2 napthoquinones from the acetone-
soluble fat of a human strain. The first fraction, eluted
from silicic acid by a 1:1 petroleum ether-benzene mixture
was called Kx, the second, eluted by ether-methanol 5%, Ky.
Chemical studies revealed that Kx' was a polyisoprenoid

derivative of 1, 4-napthoquinone homologous to vitamin K2

21

but with a longer side chain. Later work (Noll et al.,
1960) has shown Kx to be 2-methyl-3-solanesy1-1, 4 naptho-
quinone. The author postulated that KX may be the parent
compound of Anderson's phthiocol.

Compound Ky (Knoll, 1958), because of greater solu-
bility properties in polar solvents, is probably more polar
than Kg, lacking the long hydrocarbon side chain. From
infrared spectra the author deduced that K.y is a precusor
or degradation product of Kx differing, however, from
phthiocol in the point of attachment of a hydroxy group.

The carotenes o, B, andfivas well as leprotene have also
been found in mycobacteria (Takeda, 1944; Asselineau and
Lederer, 1953). Ebina et al. (1962) have shown the presence
of a and B carotene in a photochromogenic organism. Light
inducement was shown to cause an increase in the production

of B carotene.

Glycerides

Very little has been done to determine the location and
type of fatty acids present in mycobacterial glycerides.
The predominance of palmitic acid in the saponification
products of the acetone soluble fats would suggest that the

glycerides are constructed chiefly of palmitic acid.

22

Asselineau and Moron (1958), however, have found that the
triglycerides from human strains are principally composed
of mycocerosic acid. he di- and mono-glycerides contained
phthienoic acid (C27) and a C18 acid. From a streptomycin
resistant strain they also isolated a C60 hydroxy acid pre-
viously unreported.

Demarteau-Ginsburg and Miquel (1962) isolated ethylene
glycol (OH-CHz-CHZ-OH) from saponification products of a
neutral fraction from M, ayigm, The fraction was eluted by
ether from a chromatographic column. In addition, glycerol
and mycolic acid were also isolated. This fraction, ap-

parently, was a mixture of glycerides of mycolic acid and

mycolic acid esters of ethylene glycol.

Fatty Acids

Lipid composition, especially that of fatty acids, is
affected both quantitatively and qualitatively by the growth
conditions and medium composition (O'Leary, 1962; Asselineau
and Lederer, 1960). Apparently, unawareness of these facts
has resulted in many inconsistencies found in the litera—
ture. A brief outline of mycobacterial fatty acids will be
attempted, keeping in mind that qualitative differences in

fatty acid composition from strain to strain probably occurs.

23

Commencing with the saturated fatty acids, those that
have been identified are caproic, an unnamed C6 acid,
myristic, and palmitic acids. Palmitic acid is the most
prevalent fatty acid in mycobacteria (Anderson, 1940) and
in most bacteria in general (O'Leary, 1962). Other satu-
rated fatty acids are stearic, dihydroxystearic, arachidic,
behenic, lignoceric, and hexacosanoic.

Unsaturated fatty acids that have been identified are
palmitoleic, the C18 series (oleic, linoleic, and linolenic),
and some C26 series acids as yet unnamed.

Completing the list of mycobacterial fatty acids are
the branched chain acids tuberculostearic, a C17 or 18
(Cason and Tavs, 1959), a C19 acid, an unnamed C26 acid,
mycolipenic acid, an unsaturated C27 acid, and mycocerosic
acid. Phthioic acid, according to Anderson (1940) a branched
C26 acid, has been shown to be a mixture of at least 12
branched fatty acids (Asselineau, 1952) of C23 to 31. The
most important acids of this group are a C27 and a C29 acid
called phthienoic acids.

Included in the branched chain saturated fatty acid
group are the mycolic acids. These unique acids, more wax-

1ike than fatty acid-like, are hydroxy acids constructed of

24

long branched carbon chains of approximately 88 carbon atoms
(Anderson, 1940).

Asselineau and Lederer (1955) stated that the mycolic
acids are a group of closely related compounds. Mycolonic
acid (C87) is the parent acid, differences occurring with
the addition of hydroxy, methoxy, or oxo groups of differing
loci. Some strains may produce as many as 7 or 8 different
mycolic acids. Most human strains, however, contain 2
acids (a and 8) comprising as much as 8.3% of the organism's
total dry weight (Asselineau and Gendre, 1954; Asselineau,
1960).

In order to classify the 40 or more mycolic acid types
thus far isolated, Asselineau and Lederer (1955) have sug-
gested that strain and order of elution from silicic acid
chromatographic columns be used in the nomenclature. Thus,
alpha mycolic acid from strain test was designated 3-hydroxy—
x-methoxy-mycolonic acid (l-Test).

Mycolic acid may be attached to phosphatides, polysac-
charides, or glycerol in the mycobacterial cell (N011 and

Jackim, 1958; Tsumita, 1956).

Glycolipids
Lederer (1961) recently reviewed the chemistry and bio-

logical activity of glycolipids from mycobacteria.

25

Knowledge of glycolipids has been very recent. The advent
of adsorption chromatography and infrared spectroscopy has
made possible separation and characterization of these com—
pounds. Certain glycolipids are specific for strains of
mycobacteria (Smith et al., 1960b).

Glycolipids, as isolated from mycobacteria, are glycosides
or esters. The esters contain an ester linkage between the
sugar and lipid moieties as in cord factor. The glycosides
are either glycosides of phenolic alcohols (mycoside B) or
of inositol (phosphoglycolipids) or they are attached to the
carboxyl group of D-alanine (mycoside C).

Bloch, Sorkin, and Erlenmeyer (1953) demonstrated a
crude lipid of high toxicity for mice which could be iso-
lated from living organisms by neutral solvent extraction.
Because this lipid was thought responsible for the corded
growth pattern of certain virulent mycobacteria it was named
cord factor. After several years of investigation (Noll
and Bloch, 1953; Noll et al., 1956) the chemical structure
of the toxic lipid was determined to be trehalose 6-6'
dimycolate.

Wax D was found to consist principally of esters of
polysaccharides (Lederer, 1961). The waxes from bovine,

avian, and saprophytic mycobacteria are composed of nitrogen-

26

free glycolipids, those from the human strains contain
alanine, glutamic acid, and aezdiamino pimelic acid.

Bovine wax D contains 50-80%»mycolic acid, the carbo-
hydrate moiety has been found to contain arabinose, mannose,
and galactose. Wax D from avain and saprophytic strains
was found to be also mycolic acid esters devoid of amino
acids.

Human mycobacterial wax D, comprising 6—8%»of the dry
weight of virulent organisms and 2% of the avirulent strains,
had a molecular weight of approximately 30-50,000. It was
composed of mycolic acid and peptido-polysaccharides. The
sugars isolated from the polysaccharides were D-arabinose,
Demannose, D-galactose, D-glucosamine, and D-galactosamine.
Lederer (1961) suggested that wax D might be a "monomer" of

the cell wall, esterified heavily with mycolic acid.

Type Specific Glycolipids

Studies of mycosides A, B, and C have shown that all
contain the rare O-methylated 6-deoxy—hexose in glycosidic
linkages. Mycoside A contains 2—0-methyl-fucose, 2-0-methyl-
rhamnose, and 2, 4 di-O—methyl-rhamnose. The lipid moiety
is a di-or tri-mycocerosate of an aromatic alcohol (Mac-

Lennan, Randall, and Smith, 1961).

27

Mycoside B contains 2-0-methyl-rhamnose. The lipid
moiety has not been identified but consists of 79 carbon
atoms, the rhamnose molecule being linked glycosidically
to one phenolic hydroxy group (MacLennan et al., 1961).

Mycoside c (Smith et al., 1960b) has been found to be
a mixture of closely related glycolipids separable by silicic

acid chromatography. Mycoside C contains 6-deoxy-talose,

l
6-deoxy-3-0-methy1-talose, and 3-4—di—0—methyl-rhamnose

as well as the amino acids D-phenylalanine, D—allothreonine,
and D-alanine linked in a penta-peptide. The lipid portion,

though not purified, appears to be a mixture of C hydroxy

20
acids (Ikawa, Snell, and Lederer, 1960; Jolles et al.,
1961; Ikawa and Snell, 1962). MacLennan (1962) has shown

that the monosaccharide units from mycoside C are glucose,

1
arabinose, rhamnose, 3-0-methy1rhamnose, 2, 3 and 3, 4-di-
O-methyl-rhamnose, 6-deoxy-talose, and 3-0-methy1-6-de-
oxytalose.

Other mycosides have been isolated. Chaput, Michel,

and Lederer (1962) reported the isolation and characteriza-

tion of a glycolipid-peptide from M, marianum. This compound

 

N O .

(named myc081de CM) had a chemical formula C108 H185 7 28

A mycoside has also been isolated from M, paratuberculosis

(Laneelle and Asselineau, 1962). This compound, also a

28

glyco-lipid peptide contains the amino acids phenylalanine,
alanine, isoleucine, and leucine with traces of valine.

The fatty acid portion appears to be normal chain acids of
16, 18, 20, 22, and 24 carbon atoms.

Gas Chromatography of Myco-
bacterjal Fatty Acids

 

Mycobacterial fatty acids have been found to be a com-
plex mixture of compounds. With the advent of the sensitive
analytical instrument - gas chromatography, many previously
unrecognized fatty acids have been identified. Cason and
coaworkers (whose findings are summarized below) have greatly
increased knowledge in this area.

Alcohol-ether extracts from virulent human strains were
saponified, methylated and fractionated by distillation of
esters (Cason et al., 1953). The three major components were
found to be palmitic acid (28-34%), C18 and C19 acids (32-
39%) and higher molecular weight acids (16-21%). About
one-third of the higher molecular weight acids were found
to be a methyl—o, B-unsaturated acids. The C18 and 19
fraction was found to consist mostly of n-octadecenoic,
stearic and tuberculostearic acids.

Essentially the same fatty acid content was found for

the avirulent human strains (Cason et al., 1956). Again

29

using fractional distillation, palmitic acid was found to
comprise 17-22%.of the total fatty acids. The C18 and 19
fraction comprised 25-40% of the total and the over C20
fraction 32-33%. Contrary to the results found with viru-
lent strains, however, no C27 acid fraction could be found.
Using the newer technique of gas-liquid chromatography
Cason and Tavs (1959) examined the C18-l9 fatty acid frac-
tion previously isolated by fractional distillation. This
fraction (from virulent strains) was found to be composed

of 42.5% tuberculostearic acid (C19 branched) and 49% C18

normal C and

components. Small amounts of branched C17, 17

branched C18 acids were found. The C18 fraction was found
to be oleic acid.

The C15 to C17 fractions were examined more closely
by Cason and Miller (1963). One per cent of the n-Cl6 fatty
acids, isolated by preparative gas chromatography was found
to consist of an unsaturated acid lO-hexadecenoic acid.
The branched C17 fraction contained 20% of the unsaturated
acid lO-methyl-9-hexadecenoic acid. The remainder of the

branched C17 fraction consisted of equal parts of 8- and 10-

methyl-hexadecanoic acid.

Materials and Methods

Mass cultures of organisms were initiated by removing
organisms from stock culture slants, inoculating a few
milliliters of Dubos (Difco) or Middlebrook broth (Difco),
and incubating 2-4 weeks. A few drops of the resulting
culture were placed in 10-20 tubes containing 5-10 ml. of
Dubos broth with Tween 80 and 1% glucose. These tubes were
incubated 2-4 weeks. Five or 10 of these tubes were then
used to inoculate each of 3 or 4 large bottles (1 gallon
syrup bottles) containing either 500 ml. of a synthetic
medium (Wong and Weinzirl, 1936; Fregnan, Smith, and Randall,
1961; Patterson, 1961) or 500 m1. of Dubos broth, 1% glu-
cose without Tween 80.

After a l to 3 month incubation period these cultures
were divided into 50 to 100 m1. portions and used as inocu-
lum for 35 to 40 Other gallon bottles containing 500 ml. of
synthetic medium each. After a 2-3 months growth period,
depending upon the strain used, the mass cultures were
harvested by filtration on a Buchner funnel with filter
paper or glass wool serving as the filtering substance.

The moist cells were then extracted for 2 days with diethyl

Ether-ethanol (1:). Two or 3 additional extractions of 3

30

31

hours each were also performed with fresh extractant. The
extracts were combined, filtered through a Seitz filter, and
evaporated to dryness at 40 C under vacuum.

Non-lipid components were initially removed by par-
titioning the lipid between ether and water. Subsequently,
the technidque of Folch, Lees, and Stanley (1957) was used.
In this procedure lipid is partitioned between a chloroform
or chloroform-methanol (2:1) mixture and a .6% sodium
chloride solution.

Phosphatides were removed from the crude extract by re-
peated washings with boiling acetone. The insoluble residue
constituted the crude phosphatide fraction which was not
chromatographed. The lipids were then dried and stored at
-30 C.

Prior to chromatographic separation the lipids were dis-
solved in chloroform and contacted with anhydrous magnesium
sulfate for 1 to 2 days at room temperature to remove
moisture.

The solvents used were prepared in the following man-
ner. Ether, anhydrous purified (Baker) was redistilled
over sodium metal; or ether, anhydrous-reagent grade (Baker)
was stored over sodium metal prior to use. Benzene, reagent

grade (Merck) was redistilled and stored over sodium.

32

Normal hexane (Skellysolve B, Phillips Petroleum Company,
Bartlesville, Oklahoma) was washed with concentrated sulfuric
acid, 10% KOH and distilled water. It was then redistilled
and stored over sodium. Other solvents (chloroform, methanol,
acetone, and ethanol) were reagent grade and were used with-
out prior treatment.

Chromatographic columns (45 to 50 mm. x 150 mm.) were
prepared by packing under approximately 5 psi. air pressure
a slurry of 100 to 200 mesh Florisil* (Floridin Company,
Tallahassee, Florida (suspended in hexane-benzene (1:1).
After washing the column with 5 or 6 column volumes of the
hexane-benzene mixture the column was charged.

A one to three g. weighted sample was dissolved in a
small volume of hexane-benzene (1:1) and applied to the
column. The column was then developed with 1800 ml. amounts
of the following solvent series: hexane-benzene (50%), ben-
zene, benzene-ether (5%), benzene-ether (50%), ether-
methanol (1%), ether-methanol (5%), ether-methanol (20%),
and ether-acetic acid (1%). Sequential fractions of 250

ml. each were collected. Solvents were removed by warming

 

*Florisil is a synthetic adsorbent composed of 15.5%
magnesium oxide, 84% silicon dioxide, and 0.5% sodium sul-
fate activated at 650 C (Carroll, 1961).

33

at 40—45 C each fraction in a stream of filtered, dried air.
The weight of lipid present was determined either by removal
of an aliquot and evaporated to dryness and weighed in a
tared vial or by weighing the entire fraction after trans-
ferral to a tared vial.

Lipid fractions were then analyzed with a recording
double beam spectrophotometer (Beckman IR-5, Beckman Instru-
ments, Fullerton, California). Prior to analysis lipid
samples were spread on sodium chloride optical plates and
pressed to the proper thickness by another plate. Wax-like
components were dissolved in a small amount of benzene and
smeared on the plates, the benzene being removed by evapora-
tion before recording a spectrum.

In certain instances fractions were subjected to another
chromatographic separation. Using a smaller column (20 x
100 mm.) and gradient elution, 100 to 400 mg. of a fraction
or pool of fractions was rechromatographed on silicic acid
or alumina. Silicic acid was prepared using a modified
Hirsch and Ahrens (1958) technique. Two hundred grams of
Mallinckrodt silicic acid (Mallinckrodt Chemical Works,

St. Louis, Missouri, analytical reagent in powder form) was
placed in a 1 liter glass cylinder. One liter of absolute

methanol was added to the cylinder and the silicic acid was

34

suspended by rapid agitation. After 30 minutes of settling,
the methanol and suspended silicic acid was discarded.

This procedure was repeated once more with methanol and
twice with diethyl ether. The silicic acid was then dried
to remove ether and activated by heating at 110 C for 24
hours.

Alumina (Aluminum Company of America, East St. Louis,
Illinois, Activated Alumina, grade F-20, of ungraded particle
size) was prepared by washing 3 times with methanol and
twice with chloroform. It was then activated by heating at
110 C for 24 hours. Both the silicic acid and alumina were
stored over activated alumina prior to use.

A solvent gradient of increasing polarity was obtained
by flowing, dropwise, a more polar solvent from one 500 ml.
separatory funnel into another funnel of the same size con-
taining a magnetic stirring device. Solvent from the lower
funnel was applied to the chromatographic column at approx—
imately the same rate as the incoming polar solvent.

Occasionally, as a test of homogeneity, lipid samples
were chromatographed on glass fibre filter paper impregnated
with silicic acid. Fibre glass paper (Whatman GF-A, W. and
R. Balston Limited, London, England) was impregnated ac-

cording to the method of Dieckert and Reiser (1955), Brown

35

Yeadon Goldblatt and Dieckert (1957) modified by Cormier,
Jonan, and Girre (1959). Twenty cm. squares of paper were
immersed in a sodium silicate solution (specific gravity
1.28), blotted between sheets of Whatman #2 filter paper and
then immersed in 5N HCl for several minutes. The impreg—
nated paper was then washed in running tap water for 1/2
hour, rinsed in 3 changes of distilled water, and dried.

It was then washed in benzene, ether, and methanol, dried,
and stored over activated alumina prior to use.

Narrow strips of impregnated paper out to fit 1 X 8
inch test tubes are spotted 1 inch from one end with 10 ul.
of a chloroform solution containing 50-200 ug. of lipid.
Several strips were so prepared for each lipid sample and
developed in a series of solvents of differing polarity.
One hour was required for the solvent front to ascend to
the end of the paper. The strips were then dried, sprayed
with aqueous 50% (vol./vol.) sulfuric acid and charred
over a hot plate. Organic material was located by charred

areas.

Gas Chromatography
Free fatty acid samples from five organisms (71C-0,

M. bovis Ravenel, M. Tuberculosis H37Rv, M. avium and

36

172C1-l) were analyzed by gas chromatography. The fatty
acids were methylated according to the method of Gehrke and
Goerlitz (1963). From 200 to 500 mg. of fatty acid were
placed in a 100 ml. flask with 10 to 20 m1. of distilled
water. Phenolphthalein indicator was added. The sample
was then titrated to a phenolphthalein end point with potas-
sium hydroxide, a slight excess was added and the flask was
then heated on a steam bath to convert the fatty acids to
potassium salts. A few ml. of ethanol was added and then a
two—fold excess of silver nitrate was added to convert the
fatty acids to silver salts.

Water was removed from each sample using a rotary vacuum
film evaporator. The samples were then stored under vacuum
in a desiccator over Drierite (W. A. Hammond Drierite
Company, Xenia, Ohio) for several days to remove the last
traces of water. Each sample was then placed in a 2 dram
vial and pulverized with a stirring rod. Two to 4 ml. of
10%.(vol./vol.) methyl iodide in n-pentane solution was
added to each vial. The vials were stoppered and stored 8—
12 hours at room temperature prior to gas chromatographic
analysis.

Supernatant solvent containing the fatty acid methyl

esters was decanted from the silver iodide residue and

37

transferred to a separate container. Excess n-pentane and
methyl iodide was removed with a stream of nitrogen. The
methyl esters were then dissolved in an equal volume of n-
pentane.

Five to ten ul. of ester-n-pentane mixture was injected
into the gas chromatograph (Aerograph model A-905 with
thermal conductivity detector, column composed of 60—80
mesh firebrick solid support coated with 20% diethylene glycol
succinate liquid phase, manufactured by Wilkins Instruments
and Research Inc., Walnut Creek, California).

A gas hase of helium flowing at 42 ml. per minute was
used. Column temperature was maintained at 185 C. Reten-
tion time was correlated with carbon number of the methylated
fatty acid by running a standard mixture of fatty acid
esters under the same conditions as the unknown samples.

The laboratory strains used in this study were obtained
from the following sources: photochromogens P-8 and P—4,
skotochromogen P-15, and nonphotochromogen P-39 Dr. E. H.
Runyon, Veterans Administration Hospital, Salt Lake City,
Utah; M, ayiMM, M, tuberculosis strains H37Rv and H37RA’ M,

bovis strains B.C.G., and Ravenel and M, fortuitum from

 

Communicable Disease Center, U. S. Public Health Service,

38

 

 

 

 

 

 

coflnma mcfl3m
III III awaken I coflmoa HHH HIU mmm
Immune humusmmofi
msowmma GMWMMH sonoH msfl3m
upon one was up III III HmEhop I coflmoa HHH HINU mma
Hmuoa In D H o Imuucw muoucmmme
0>Hummmc
coflmma ocfi3m
III III Awaken I seemed H HIU mmH
Immune humucomoe
> so
mcoflmoa Godwoa soflmma madam
>©on one deepen III III assume I :onoH HHH HIO mma
HmooH Immune Immune huoucmmoe
w :0
msoamoa H cofimoa mcfi3m
coflmoa H
soon one Hashes III III Hmsuoe I cosmos HHH HI omea
HmooH Imuuca Imuucw mumpsommE
mmam mnemoouw
mcfi3m mm>amu mooXUwav muwnnmm meadow cflmflno somssm Honfioz

How mufiuwcmmonumm

 

(NWSHm cw meD mEmacmmho

39

wmmnnmm “penance an oucoasnfl> omnmonocws

 

 

 

 

 

 

 

ocfl>on
III III III I monoumo >H HImoHHm
m.uo>mm
cofimoa ocw>on .IIIII
Esfl>m
+ II... HMEHOU OUOG QQEH .l OIU mmH
2
loans“ muoucomme
Godmo mcw>on
woos noama anewow I woos HHH o a No
new case I Immune enema
.¥ >003
ncofimwa sodmo mcfl>on
woos £QE>H III III mfiuma Impoc
unaommommum a Hmnucw £QE>H HHH OIU am
one QAXm .s mumucmmme
encamoa soamo
woos £QE>H . H
can saxm so III IIIs HmEpr mcfl>on HHH OIO mo
ommnmnmow3 Imuusfis
sownma mcfl3m
III III Hmsuoo I coamoa HHH HIH «em
Imuuafl muwucmmmE
mmam mGHQDOHw
on m > n . m .
fl3m m Hmo cmxoflnu nufinnmm museum ca HMO somcsm Honesz

40

 

 

 

 

coHnoH ocflzm
III III amEHoo I soamoa HHH HIO Aha
Immune mhwucmmofi
msw3m
III III M I cowmma HH HID mma
wuoucomoa
soflmoa woos cofimma ocwsm
seems one III III Hmenme I eoemma HHH oIo mm
Hmeuoplmuusfi Immune muoucmnos
mmflm mcflmsouo
msw3m mo>amu mcoxowzo nuannmm meadow CHoHHO sowssm HmQEsz

41

Atlanta, Georgia. The strains isolated by the Michigan
State University Bovine Tuberculosis Project are described

above.

Results

Table 2 lists most of the lipid compounds present in
the ether-ethanol 50%»extracts of mycobacteria, their order
of elution from a Florisilepacked chromatographic column
and the appearance of these lipids.

The first lipids eluted from the column are esters of
fatty acids. They appear as a clear oil in the first few
fractions and then as a yellow or reddish orange oil.

The red pigment is a napthoquinone, Kx, 2-methyl-3-
solanesyl-l, 4-napthoquinone (Noll et al., 1960).

Next in order of elution are the saturated trigly-
cerides. In human and some bovine strains the wax dimy-
coceronate of phthiocerol is mixed with the glycerides.

Benzene elutes from the column more saturated fatty
acid triglycerides white or light yellow in appearance.
Unsaturated fatty acid triglycerides appear in the benzene-
ether 5% eluates. Diglycerides mixed with some triglycerides
appear in the benzene-ether 50%Ieluates. Mycoside B appears
in either the late benzene-ether 50% or early ether-methanol
1% eluates of bovine lipids.

Monoglycerides, glycerol monomycolate, long chain fatty

acids and the napthoquinone Ky appear in the ether-methanol

42

43

1% fractions. The lipopeptide J frequently appears in

ABS
these fractions from nonphotochromogenic organisms. The
early ether-methanol 1% eluates of photochromogenic organ-
isms are composed chiefly of mycoside A, a hard transparent
wax-like material.

The ether~methanol 5%.e1uates are composed chiefly of
fatty acids. Mycoside F, a glycolipid from M, fortuitum
and skotochromogens, may appear in the late ether-methanol
5% or early ether-methanol 20% eluates.

Mycoside CM from nonphotochromogens, mycoside C from
M, ayrgm, and mycoside D from skotochromogens appear in the
early ether-methanol 20% eluates. All of these mycosides
are light brown hard transparent wax-like compounds. Fatty
acids appear in low concentration in the ether-methanol
20% fractions, most, however, are eluted in the ether-
acetic acid 1% fractions.

Table 4 is comprised of data obtained from Table 1.
Weights of lipid eluted from chromatographic columns by
each solvent series have been summed and expressed as per-
centages of the total lipid eluted from the columns. Cal-

culations from eleven representative strains have been

presented.

44

Fifty-five per cent of the total recovered lipid from
strain H37RA was eluted by the hexane-benzene 50% solvent
series. Of this fraction approximately one-half was liq-
uid at room temperature and eluted in the first two eluates.
The infrared spectra of these eluates were identical to
those of fatty acid ethyl esters (Smith, 1962). Eighty-
three per cent of the recovered lipid was eluted by the
solvents hexane—benzene 50%, benzene, benzene-ether 5%
and benzene-ether 50%. These fractions consist chiefly
of fatty acid esters, dimycoceronate of phthiocerol, tri-
and diglycerides. The ether-methanol series consists
mostly of monoglycerides and long chain fatty acids and
accounted for 7% of the recovered lipid. The ether-acetic
acid 1% fraction (mostly fatty acids) comprised about 9%
of the recovered lipid.

Lipids extracted from strain M, tubergglgsis H37Rv con-
tained very little fatty acid esters. The hexane-benzene
50% eluate comprised only 3% of the total recovered lipids.
The hexane-benzene 50% to benzene—ether 50% eluates con-
tained only 42% of the eluted lipids. The ether-methanol
5%Iand 20% eluates contained 25%.of the lipid eluted.
Several of the ether-methanol 20% fractions were chromato-

graphed on silicic acid using gradient elution techniques.

45

Most of these fractions consisted of fatty acids (probably
of long carbon chains) as indicated by their infrared
spectra. Some of the material was glycerol and about 5%
was composed of a material having an infrared spectrum very
similar to the phosphoglycolipid of Noll and Jackim (1958).
Thirty per cent of the recovered lipid consisted of ether-
acetic acid 1% eluates composed mostly of fatty acids.

Nearly one—half of the hexane-benzene 50% eluates of
_M; boyig Ravenel consisted of a reddish oily napthoquinone--
fatty acid ester mixture. The remainder of the eluates
consisted of a light yellow or white fat. Chromatography
of some of these fractions on alumina and silicic acid
served to separate them into triglycerides and dimycocero—
nate of phthiocerol. Eighty-seven per cent of the recovered
lipid was eluted by the hexane-benzene 50%, benzene—ether
50%.series eluents. Very little lipid (3%) was eluted
by the ether-methanol series. The majority of the ether-
methanol 1% fractions, however, consisted of the "type-
specific" glyco-lipid mycoside B*, comprising a little less
than 1% of the recovered lipid.

Lipids from M, bgy;§_B.C.G. were also studied. Dimyco-

ceronate of phthiocerol could not be found in late

 

*Identity verified by Smith (1963)

46

hexane-benzene 50% and early benzene eluates of lipids from
this organism. This compound could not be found in frac-
tions separated by silicic acid chromatography from the
original fractions. Mycoside B was not found in the lipids
from this organism. Rechromatography on silicic acid did
not serve to separate this lipid from the other compounds
present. Mycoside B, if present, may have been insuffi-
cient quantity to be detected. A large portion of the
ether-methanol 1% fraction consisted of a compound Whose
spectrum was identical with that Of a monoglyceride of
stearic or palmitic acid (Kubica et al., 1956). This frac-
tion comprised 5-6% of the total lipid reovered.

The early hexane-benzene 50%Ieluates from lipids of
strain P-8 (a photochromogen) were composed mostly of a
mixture of fatty acid esters and the napthoquinone Kx.

They comprised about 15% of the total recovered lipid.
Sixty-six per cent of the lipid was eluted by the hexane-
benzene 50%.to benzene-ether 50% solvent series. The greater
portion of the ether-methanol 1% eluted fraction consisted

of a hard transparent wax identified as the glycolipid

mycoside A*. The ether-methanol 20% eluate comprised 10%

 

*Identity verified by Smith (1963)

47

of the recovered lipid. Chromatography of the ether-
methanol 20%.fractions on silicic acid served to separate
them into fatty acids and a phospholglycolipid-like material.
The majority of this fraction was not eluted from silicic
acid by chloroform-methanol 10%Iand may have consisted of
phospholipid and/or glycerol.

Seventy-six per cent of the recovered lipid was eluted
by the hexane—benzene 50%»to benzene-ether 50% solvent
series from strain 71C-0. Two per cent of the lipids were
eluted in each of the solvent mixtures, ether-methanol 1%
and ether-methanol 5%. These fractions consisted mostly
of a napthoquinone, monoglycerides, fatty acids and free
glycerol. The ether-methanol 20% eluates were a dark
reddish brown hard wax. These fractions were combined and
separated by chromatography on silicic acid. A concave
gradient elution pattern chloroform-benzene 50%Ito chloro-
form-methanol 5%.was used. The early eluates consisted of
fatty acids, then fatty acid-napthoquinone mixtures. The
early eluates comprised about 20% of the ether-methanol
20% eluates. A hard light brown wax was eluted by l to 5%
methanol in chloroform. This wax, comprising about 80% of
the ether-methanol 20% fractions (8% of the recovered lipid)

was identified as mycoside D by infrared spectroscopy.

48

Lipids eluted by the solvent series hexane-benzene 50%
to benzene=ether 50% from extracts of M, fortuitum com—
prised 24% of the recovered lipid. Fifteen per cent of
the lipid was eluted by ether-methanol 5% and 34% by ether-
methanol 20%, These fractions were chromatographed on
alumina and silicic acid using gradient elution techniques.
A compound having a spectrum very similar to that of myco-
side F (Randall et al., 1961) was isolated from both the
ether-methanol 5% and ether-methanol 20% fractions. This
compound, probably mixed with fatty acids and traces of
napthoquinone, was a light brown transparent fat-like ma-
terial. It was eluted from silicic acid by chloroform-
methanol 5%.and comprised about one quarter of the ether-
methanol 20% eluates.

Sixty-two per cent of the recovered lipids of M, gyigm_
was eluted by the hexane-benzene 50%Pbenzene-ether 50%
series. Twenty-seven per cent of the lipid was eluted by
ether-methanol 20%. Infrared spectra of the early ether-
methanol 20% fractions were identical with those of myco-
side C*. Chromotography of the ether-methanol 20% frac-

tions on alumina served to separate the lipids into mycoside

 

*Identity verified by Smith (1963)

49

C and fatty acids. Mycoside C comprised 10-12% of the total
recovered lipid.

The lipid composition of strain 158C-0 (an M, gying
like organism) was very similar to that of the laboratory
strain of M, aylgg,

Strain P-39's recovered lipid fraction consisted of
64% eluted by the hexane-benzene 50% to benzene-ether 50%
eluent series. Most of the ether-methanol 1% fraction con-
sisted of a yellow fat-like material having an infrared
spectrum very similar to that of glycerol monomycolate. A
small portion of the ether-methanol 5%Ieluate consisted of

material having an infrared spectrum identical to J (a

ABS

lipopeptide)*. The early ether-methanol 20% eluates were
composed of a hard light brown transparent wax-like material

having an infrared spectrum identical with mycoside CM*.

The lipid composition of strains 62D-0 and 2441-1 were

very similar to that of P-39. Both contained mycoside CM.

The ether-methanol 20% eluates from 62D-0 were separated

on silicic acid into fatty acids and mycoside C Mycoside

M.
CM comprised nearly 15% of the recovered lipid of this or-

ganism. The lipopeptide JABS was identified in the

 

*Identity verified by Smith (1963)

50

ether-methanol 5% eluates from strain 2441-1 and not those
from strain 62D—0.

Mycoside C comprising 7-12% of the recovered lipid,

Ml

2-1, 193C1-1,

172C1-1, 68C—0, 171C-1, 186C-1, 228C-1, 380$, BllOH-l and

93C-0. Compound JABS was identified in eluates from 172C1-l,

was also isolated from strain 152C-l, 193C

l7lC-l and 186C-l in addition to strains 2441-1, 62D-0 and
P-39 mentioned above.

Lipid from laboratory strain P-15 (a group II skotochrom-
ogen) was also studied. Mycoside D could not be isolated
from the lipids of this organism. Repeated chromatography
of the various lipid fractions did not allow isolation of
this lipid. A compound having a spectrum suggestive of myco-
side F was eluted from a florisil-packed column by 1 to 2%
methanol in ether. No other specific lipid was observed in
eluates from P-15 lipids.

Figure 1 consists of infrared spectra. The first 2
spectra are of ether-ethanol extracts of strains P—8 and
M, ngi§_Ravenel from which phosphatides have been removed
by boiling acetone fractionation. Spectra of whole ex-
tracts, in general, are very similar to those of glycerides,
the major components of these extracts. Spectra 3 and 4

are of a di-and a mono-glyceride identical with Kubica,

51

Randall and Smith's (1956) spectra for compounds C and D.
Compound D, isolated from eluate #5 of the ether-methanol
1% series of M, bgyi§_B.C.G., is a monoglyceride of stearic
or palmitic acid. Compound C, isolated from early benzene-
ether 50%.eluates of strain P-8 is a diglyceride, possibly
of stearic and palmitic acids. The last spectrum is of a
brown gummy material isolated from early ether-methanol
eluates of M, bgyi§_Ravenel. Its spectrum is similar to
that of the unsaturated long chain hydroxy esters of Noll
(1957).

Spectra l and 2 (Figure 2) are of a clear oily material
isolated from the early hexane-benzene 50%»eluates of M,
ng;§_Ravenel and 3808. They are identical to the spectra
of ethyl esters of fatty acids supplied by Smith (1962).
Spectra 2 and 3 (of a dark reddish material) are identical
to those of Noll's (1958) compound K.x a napthoquinone later
identified as 2-methyl-3-solanesyl l, 4-napthoquinone (Noll
et al., 1960).

Spectrum 1 (Figure 3) is of a white fat-like material
isolated from the second ether-methanol 1% eluate of strain
P—39. The spectrum is identical to that of a glycerol mono-
mycolate (Noll, 1957). Smith (1963) suggested this sample
might be glycerol monomycolate or a monoglyceride of other

fatty acids.

52

Spectra 2 and 3 are similar to the fatty acid spectra
of Kubica et a1. (1956). The first compound was isolated
from the late ether-methanol 5% eluates of strain P-39,
the second from the early ether—acetic acid 1% eluates of
strain 186C-1. Spectrum 4 is similar to that of dimyco-
ceronate of phthiocerol. This compound, isolated from
rechromatographed late hexane-benzene 50% fractions of M,
tuberculosis H37RA' was a white soft fat-like material.
Smith (1963) has indicated this sample was relatively im-
pure but suggestive of dimycoceronate of phthiocerol.
Spectra 5 and 6 are of mycosides A* and B.* The first com-
pound was isolated from early ether-methanol 1% eluates of
strain P-8, and the second compound from the second ether-
methanol 1% eluate of strain M, bgy;§_Ravenel.

Spectrum 1 (Figure 4) is the spectrum of a light yellow
hard wax-like material isolated from eluates of silicic
acid chromatographed of ether-methanol 5% fractions from
strain 172C -1. The spectrum is identical with that of the

1

lipopeptide J * (Smith et al., 1960a). Spectrum 2 is

ABS

identical with the spectrum of mycoside CM* of strain P-39.

This sample was isolated from the first eluate of the

 

*Sample identity verified by Smith (1963)

53

ether-methanol 20% fractions. Spectrum 3 is also mycoside

C This sample was isolated from eluates of silicic acid

M'
chromatography of original ether-methanol 20% fractions
from strain 380s. Spectra 4 and 5 and of identical com-
pounds. The first isolated from eluates of alumina chromato-
graphy of ether—methanol 20% eluates of M. a_V_i_1_1Ml_ and the
second from the first eluate of ether-methanol 20% frac—
tions from strain 158C-0. The compounds from M, gyigm_and
158C-0 are both mycoside C. Spectrum 6 was recorded of a
hard light brown clear wax-like material isolated from
silicic acid chromatographic eluates of early ether—methanol
20% fractions from strain 7lC-0. This spectrum is identical
with that of mycoside D of Smith et al. (1960a). Spectrum
7 was taken of a brown fat-like material eluted during
silicic acid chromatography of the ether-methanol 20%Ifrac-
tions of M, tubercu10§;§_H37Rv. This compound, eluted from
the column by chloroform-methanol 5-10% has a spectrum sim-
ilar to that of a phosphoglycolipid isolated from a human
strain by Noll and Jackim (1958).

Spectra l, 2, and 3 (Figure 5) were taken of a light
yellow or light brown hard wax-like material isolated from
strains l7lC-l, 172C -1 and 186C-l. This compound (or mix-

1

ture of compounds) appeared in the first fraction of the

54

ether-methanol 1% eluates of strain l7lC-1 and the early
eluates of silicic acid chromatographic separations of
ether-methanol 20% fractions from strains 172C1—1 and 186C-l.
These compounds appear to be amide-like compounds or mixtures
of compounds. Spectrum 4 was recorded of a light brown
clear fat-like material isolated from eluates of silicic
acid chromatographic separation of ether-methanol 20% frac-
tions of M, fortuitum. This compound, which comprised
about one-fifth of the ether-methanol 20% fractions had a
spectrum similar to that of mycoside F (Fregnan et al.,
1961).

Spectrum 5 was recorded of a light yellow hard wax-
like material isolated from eluates of silicic acid chroma-
tographic separation of ether—methanol 20%»fractions of
strain P-4. This compound or compounds has a spectrum
similar to the phospholipid compounds described by Noll and
Jackim (1958).

Spectra l to 4 (Figure 6) and l to 6 (Figure 7) were
taken of hard light brown transparent wax-like compounds
isolated from the early ether-methanol 20% fractions of

strains 2441—1, 62D-0, 152C-l, 193C -1, BllOH-l, 171C-1,

2

68-0, 186C-l, P-39 and 172C1-1. All are identical with the

spectrum of mycoside C (Smith et al., 1960a).

M

55

Figures 8-32 are eluation curves plotted from data pre-
sented in Table l for the 25 organisms studied. Solvent
series are plotted against milligrams of lipid eluted. Each
solvent or solvent pair consists of 6 to 9 fractions of
250 ml. each. The solvents used, in order of increasing
polarity, were hexane-benzene 50%»(HB50), benzene (B),

benzene—ether 5% (BES)' benzene-ether 50%(BE5 ), ether-

0
methanol 1% (EM1)' ether—methanol 5%»(EM5), ether-methanol
20%»(EM20) and ether-acetic acid 1% (EA1)'

Most of the fatty acids of carbon chain length 8 to
19 in the ether-acetic r% eluted fractions were of 16 to
18 carbon atoms (see Figure 33). Approximately 60%»of the
fatty acids (C8-19) from M, tuberculosis H37RV were C16

unsaturated. M, avium, 172C -1 and 71C-0 fractions were

1
composed of 30-40% saturated C16 and 5—10%.unsaturated C16.
M, M22;§.Ravene1 fractions contain very little C16 to C19
fatty acids, most of the fatty acids (C8-l9) were C14
(30%), C13 (which may be unsaturated C12 or branched C14-
20%) and C8-12 (30%).

C18 fatty acids comprised 35 to 51% of the C8-l9 fatty
acids of 172C1—l, M, avium, M; tuberculosis H37Rv and 7lC-0.
With the exception of strain 71C-0 over 80% of the C18 acids

 

56

were unsaturated. About one-half of the C18 acids from
71C-0 were found to be saturated fatty acids.

Several strains (M, avium, H37Rv and 71C-0) had slight
amounts of a fatty acid with a retention time similar to
that of a C19 or C19 branched acid. Further identification
of this fraction was not possible. The column and condi-
tions used for analysis were not adequate for better resolu-

tion of the higher fatty acids.

57

TABLE 1

WEIGHT OF LIPID ELUTED FROM

CHROMATOGRAPHIC COLUMN

 

 

 

 

H37RV 9-39 ESE. H37RA 9-15 68C-0

H850 1 14* 56 86 140 237 8
2 20 88 97 396 203 29

3 20 75 24 249 303 15

4 6 53 8 170 202 25

5 2 54 3 77 104 25

6 5 50 4 53 57 23

7 '0 38 0 46 49 26

8 0 10 0 72 41 w 53

9 4 36 5 -- 60 __

B 1 15 91 63 89 135 ' 104
2 74 147 123 65 0 37

3 146 73 62 20 0 o

4 46 38 36 14 4 3

5 33 24 21 41 5 --

6 13 14 13 13 o --

7 12 1 10 25 4 --

8 -- 5 -- -- -- --

*mg. of lipid per 250 ml. of eluate

 

58

TABLE l-—Continued.

BE

EM

50

 

H37RV P-39 BCG H37RA P—15 68C-0
25 39 56 129 50 25
45 26 34 127 36 21
48 4 33 26 16 7
38 0 34 19 6 3
24 6 31 5 3 --
18 4 0 6 o __

9 7 23 3 0 __

143 99 '87 29 20 3
42 13 13 9 4 0
13 16 5 3 0 0

7 2 5 0 0 0
6 3 0 0 o __
0 O 0 0 0 —-
3 0 0 o o -_
12 14 7 3 0 8
39 20 11 15 10 10
14 5 13 9 0 0
7 5 11 0 0 0
8 9 14 4 3 _.
5 7 7 o o _-
3 6 -- 3 o --

 

 

 

 

EM 1

EM20 1

EA 1

Weight

of lipid
applied to
column (mg.)

 

47

57

30

28

20

22

15

128

53

49

31

21

17

14

467

71

2020

*accidentally lost

P-39

 

15

10

12

9

31

10

8

46

76

35

26

18

16

13

201

2242

59

TABLE l--Continued.

H37RV

BCG

26

20

‘16
26
20
15

16

1660

H37RA

 

13

16

28

19

2787

P-15

 

12

8

19

47

34

22

17

3054

68C-0

 

19

15

18

31

280

98

54

24

17

12

48

24

1136

Weight of
lipid
recovered
(mg.)

Per cent
recovery

HB50

60

TABLE l-—Continued.

 

 

 

 

H37RV 9-39 BCG
1998 1670 1287
98 75 78
19302-1 fliuiffig 9-4
36 10 255
82 100 . 43
18 10 40
29 0 65
47 10 65
55 10 55
59 15 43
147 23 33
182 68 100
124 43 57
69 223 10
35 7 3
l6 2 o
15 o 0

o 0 0

0 -- 18

 

H37RA 9-15

1808 --
65 --

380$ 186C-l
32 53
45 125
20 107
50 48
53 28
33 15
28 15
33 10
150 50
95 175
45 120
23 65
20 35
15 20
10 15

 

68C-0

 

1065

94

93C-0

 

35

44

44

15

11

11

13

101

227

112

73

50

30

27

BE

EM

50

61

TABLE 1--Continued.

193c2-1 Miufffi; 3:4-
101 58 55
53 143 28
35 38 10
18 7 2
12 3 0
5 2 0
3 0 0
48 20 ‘ 25
11 38 5
5 20 0
3 15 0
3 7 0
3 5 0
2 0 0
18 2 135
19 58 43
10 68 8
5 38 2
3 23 5
3 10 0

380$
30
48
35
25
23
15

2

63

28

10

10

20

28

23

18

12

12

12

186C—1 93C-0

 

10

63

25

15

23

18

10

13

10

10

13

115

80

33

20

15

15

11

36

32

24

16

11

8

10

43

62

TABLE l--Continued.

column (mg.)

 

 

193c2-1 fliufffi; P—4 3325. 186C-1 930-0
EMS 1 10 7 23 33 200 21
2 18 105 13 18 125 12
3 13 85 8 8 55 7
4 13 65 0 8 33 6
5 15 53 25 8 23 8
6 16 18 0 10 22 7
7 14 43 0 13 25 8
EMZO 1 174 385 43 200 100 184
2 125 240 65 245 218 98
3 68 140 55 43 115 49
4 52 85 33 33 85 33
5 43 60 30 25 70 28
6 31 20 25 20 50 22
7 28 43 12 23 45 26
-EA1 1 19 50 15 183 45 15
2 374 525 122 23 725 712
3 22 13 7 15 63 11
Weight
:fipiizgdto 4611 3954 3688 2101 3278 2288

63

TABLE 1--Continued.

 

 

 

 

£0 for-
193c2-1 tuIEEh P-4 3803 186C-1 93c-0
Weight
°f 1191a 2328 2859 1582 2008 3385 2302
recovered
(mg.)
Percent 51 72 43 96 103 101
recovery
62D-0 172C -1 5’ 2441-1 193C -1 228c-1
l av1um 1
2
8850 1 210 34 1 33 2 2
2 222 81 . 18 151 27 2
3 52 40 32 32 56 3
4 32 57 15 7 54 8
5 30 52 9 3 39 8
6 24 50 13 2 18 18
7 43 51 25 1 10 20
8 -- -- 36 -- -- --
B 1 136 155 234 1 16 30
2 115 218 107 2 9 24
3 70 149 68 9 3 13
4 38 93 128 32 2 6
5 28 60' 39 35 2 3
6 22 46 27 23 1 2

64

TABLE 1--Continued.

532:2. 17201-1 QZEEE. 2441-1 19301-1 2280-1
17 116 144 54 73 7
83 47 80 18 31 6
34 46 19 7 11 16
31 41 20 8 6 15
21 29 26 17 4 12
16' 19 16 17 4 7
11 20 9 11 19 5

8 68 34 19 126 44
40 13 12 14 18 34
13 4 3 2 4 9

8 2 1 1 2 3

5 1 1 1 0 2

4 1 0 1 0 1

3 0 0 0 0 0

5 19 12 4 37 28
52 34 17 11 16 5
33 21 8 13 9 2
20 11 3 8 7 2
14 7 2 5 6 2
11 6 2 3 7 2

65

TABLE 1--Continued.

332:2. 17201-1 82588. 2441-1 19301-1 2280-1
EMS 1 48 34 1s 8 23 1
2 50 36 7 32 32 20
3 34 21 3 14 14 13
4 28 15 3 10 13 4
5 26 11 3 7 15 4
6 24 10 2 7 15 3
7 23 10 0 7 11 2
EM20 1 408 76 247 17 76 52
2 158 176 125 126 80 64
3 90 122 57 40 55 20
4 59 79 31 30 39 10
5 49 63 16 26 30 6
6 43 47 12 24 27 5
7 34 45 8 26 19 4
EAl 1 744 34 124 21 442 83
2 32 449 5 424 58 245
3 7 23 2 10 13 6
4 2 8 0 3 7 3
Weight
:Ep1::;dto 3696 3765 2265 1630 1953 1110

column (mg.)

66

TABLE l--Continued.

 

 

 

62D-O 1720 -1 $4 2441-1 1930 -1 2280-1
1 av1um 1

Weight of

llpld 3335 2892 1865 1400 1599 893

recovered

(mg.)

Per cent 90 77 82 86 82 80

recovery

710-0 BllOH-l 1520-1 1710-1 1580-0
7

HB50 1 56 31 7 7 163
2 116 8 28 318 311
3 48 14 62 76 125
4 70 10 40 119 65
5 62 2 21 98 65
6 50 3 21 76 44
7 83 3 8 -- 50

B 1 311 136 3 203 121
2 149 106 2 114 28
3 94 72 2 42 11
4 75 55 1 19 10
5 43 44 2 9 26
6 33 42 3 7 45

BE

BM

50

67

TABLE l--Continued.

71C-O
259

26
20
20
16

9
29

39

B110H-1

88

16

9

11

11

22

10

11

11

41

152C-1

36

9

16

15

11

171C-1

85

85

21

10

58

47

14

27

33

12

158C-0
90

22

68

TABLE l--Continued.

110:g_ B110H-l 1520-1 1710-1 1580-0
EMS 1 16 30 28 50 24
2 3 4 16 51 26
3 3 4 14 23 10
4 9 5 14 14 8
5 2 5 13 15 9
6 4 6 12 15 11
7 4 128 76 19 125
EMZO 1 92 59 ' 193 137 300
2 24 3 64 49 161
3 17 8 45 41 57
4 30 2 31 38 30
5 27 1 28 34 23
6 22 2 24 30 15
7 17 3 20 28 13
EAl 1 161 152 318 242 172
2 21 9 152 388 37
3 10 6 10 12 5
4 4 5 1 8 4
Weight of
applied 2393 1583 2278 3223 2927

to column
(mg.)

Weight of
lipid
recovered
(mg.)

Per cent
recovery

HB

50

69

TABLE l--Continued.

110:0 BllOH-l
2166 1326
91 84
p_-§_
1 95
2 92
3 93
4 181
5 77
6 39
7 19
8 18
9 19
1 95
2 109
3 20
4 1
5 0
6 1

WWW

1500 2717 2364

66 84

Ravenel

17

70

36

34

128

126

88

42

27

111

159

141

59

23

EM

EM

50

70

TABLE l-Continued.

fl
1 168
2 3
3 0
4 1
5 0
6 0
7 0
1 16
2 117 '
3 31
4 10
5 2
6 4
7 16
1 18
2 10
3 5
4 3
5 0
6 0
7 0

Ravenel

129

0

71

TABLE l-Continued.

P-8 Ravenel

EM20 l 43 8
2 44 1
3 28 5
4 l4 0
5 16 0
6 10 0
7 8 4

EA1 l 123. 122
2 12 2
3 2 1
4 0 0

Weight of

lipid

applied to 2070 1948

column (mg.)

Weight of

11916 1560 1377

recovered

(mg.)

Per cent 75 71

recovery

72

xm3
cums ucmummmcmua

x03
cum; ucmosHmcmna

umm umom 30-
law» proud no 00803

06m umom 30a
law» unmfia no 00803
06m umom 30a

law» usmfla Ho muH£3
umm umom 00803

x63 0006 00803

H80 flmflwpwm

H80 860-0

 

mocmumwmm<

mcmmOEounoouogm

mcH>om

0002

quZ

“mo:

“no:

CMESQ USN 0Gfi>0m

quZ
umoz

 

 

coflusnwuumwn awmnum

ZZDAOU QfiMUflmldeHmOAh 1 20mm mBZMZOQZOU
QHmHA A<HMMHBU¢mOUME m0 ZOHBDAm m0 mfiflmo

< anamoohz

m mnamoohz

mmpaumomamfln

mpfiom
wuumm pmumuoummcs
I mmpanoomamflue

mpfiom muumm pmumnsumm
I mmoflnmohamfiue

mpfiom huumm pmumusumm
mo mmpwumumamflna

HOHmUOfinpnm
mo mumcoumooomawn

Iv .H HammcmHOm
umuflanumsnmv xx

mumumm Odom mumm
#cmcomeou

N WAM<B

8H Hocmnuosuumrum

Rom ngumlmcmucmm

Xom Hmnumlmcwucwm

Rm ngumlwfiwNCQm

mcmucwm

xom mcmucmalocmxmm

Rom meNchIwcmxmm

Rom mcmncmnwmcmxmm

Rom mcmncmnwmcmxmm

 

ucm>aomxmcflusam

73

x03 cum: ucmummmcmue

06m 6860 008::

pan
mcflaamumzuo ouflnz

x03 umom muH£3

mmpflnoomam
Ho 0800 muumm ca
pm>aommflp ooh xnmn

umm
mcflaamummuo muflnz

umm
umom zed-08 uamfiq

x03 umom «#833
mocmummmmfl

mammoeouno
Iouozmcoc 0cm

ESGMflHmE .2

mammoeouguouoxm

pcm Esufisuuom .2

 

umoz

55.." >0 .fi cam
mammOEOHzoouoxm
mamOOEounoouonmcoz

umoz

“moi

umoz

quS

 

008020880680 cflmuum

20 mpfimoowz

AmCOfluumum RON Hocmnume

IHmSum cw Hmmmmm
OmHm hmEV m mpfimouzz

mpfiom
muumm mango mcoq

mpflummmomwa m I mm¢b

mcwsvonummc m I mm

mpflom muumm cacao mcoq

mmpwnmowamocoz

Amcoauomnm

Km Hocmnuwfiluwsum

c8 ucmmmnm Omamv
wumaoomeoaoa HOHmohao

ucwcomeoo

.UmDGHUSOUIIN HAQ<B

8cm Hocmnumennmsum

8m Hocmnumeuumsum

8m Hocmsuwerumnum

*- Hocmrumaunmnum

8H Hocmauwauumsum

88 Hocmnumernmnum

8H Hocmnumaunonum

x; Hocmnuoeuuonum

 

ucm>Hom ucfiusam

74

pan
mcflaamumhuo mufin3

x03
puma ucmummmcmua

x03
who: ucmummmcmna

 

wonmumoaa<

 

umoz 00806 muumm

mammoeounuouoxw Q mcfimoohz

Esfl>m am 0 wpfimoohz
coflusnwuumfln cannum newcomaou

.0mscflucoonum mqm<e

RH 0800 uaumomlumaum

Rom HocmanEIumcum

Rom doomsumEIHmsum

 

ucm>aom ccwwmflm

75

mm
NN
Hm

0m
0m

OIUHB

wao Uo£ocmun 80

 

 

mZHdmem m hO ZOHBU<Mh QHmHA

II N 0
II mm m0
II m o
0 mm m0
II II m
0 am mm
0 Ho 00
om H II
om II II
0m 0 v 0

mac 00008500085 09 >020

Gmflocmun 0H0

0m 00008000000 000
0 000080000 000
m0 000 00008
II 00000080 000
00 00008000000 000
um 000080000 000
00 080 80008
I: 000
II 0Amv mHU
00 maoumo
Huaomna

QMBDAm RH QHU< UHBHUflImmmfim HEB ZH BZMmflMm
AmHU OB mUv mQHU< Mfiafih m0 mmw<Bszmmm MB<SHxUmmm¢

m mflm<9

76

HH

on

ha
mm
mm

OIUHB

 

 

m CH om m RH 0800 Ufluwomlnmnum
OH a ma m RON HocmsumEIHmnum
m 0 3 m .80 00000000780008
NH H m m “RH Hocmflumelumnum
s 0 0 0 ”Mmmwummoww
NH 0a in N mcmucmm
I I 0H 0H Rm 80A00Imcmucmm
«A on 5H NH mcmncmm
00 av m mm Rom mcmucwnImamxwm
.l. and >050 0020

mIm a .fl mfimoawmumnsu mfimoawmumnsu

mZH4MBm HH mom mmHMmm BZN>QOm EQQm Mm QHBDQM
Qmmm>00mm QHmHA A<BOB m0 mmU<BZm0mmm MB<ZHNOMQA¢

¢ mflmdfi

77

mm
mm

OH

HIH¢¢N

vN
mm

00
N
0
ma
ma

Olnmm

NH
ma

#0
m
m
0N
hm

GMIm

NH
mm

OIUmmH

mm
M
NH
mm
m

 

Ena>0 .2

ON
0m
ma

\Omo‘d‘

x8 0800 080000I8000m
808 8000;008:80000
80 80000008I8000m

88 80000008I8000m

Aommm 08 ommmv

wmmucmoumm 80009

Rom Hmzumlmcmucwm
Rm ngumImcmNGmm

wcmncwm

Rom wcmuchIwcmxmm

 

Esufisuuom

.000080000II0 00008

.2

   

P-8 ether—ethanol extract
minus phosphatides

 

   
   
   

 

 

Ravenel ether-ethanol extract
minus phosphatides

 

 

 

 

 

P—8 diglyceride-saturated
fatty acids, Smith's "C"

   
   

 

 

 

 

 

 

BCG monoglyceride.
Smith's "D"

 

 

Ravenel long chain unsaturated
OHS esters (Noll, 1957)

 

 

1'1 12 1'3 14 15

00 ‘F
4:.
U1
0‘
\l .L
(1)
KO
l—'
0

Figure 1. Infrared spectra

 

EL bovis-Ravenel ethyl esters of
fatty acids

 

J

    

380$ ethyl esters of fatty acids

 

g, bovis-Ravenel KX napthoquinone

 

{/1

M. tuberculosis-H37R
— . V
, K.X napthoquinone

 

    
      
   

fly bovis-B.C.G.
triglyceride

 

 

 

*—

 

g, tuberculosis-H37R
. . v
diglyceride

 

V
1.

7 8 9 10 1'1 1'2 13 14 15
0

Figure 2. Infrared spectra

‘1

L-
.0.
U1.
m

    
  
    
  

P-39 glycerol monomycolate

 

 

P-39 long chain fatty acids
/’

l86C-l shorter chain fatty acids

8. tuberculosis H37RA
dimycoceronate of phthiocerol

’é. .. an“.

 

r

 

 

 

P-8

r Mycoside A

 

 

     

fl, bovis Ravenel
Mycoside B

 

 

l A l I
U V V t W

3 4 5 6 7 8 9 H 10 11 12 13 14 - 15

Figure 3. Infrared spectra

q
4-
qr
JL

 

 

 

93C-O
Mycoside CM
contaminated with fatty acids

3805

M co 'de C
y 81 M

V g, avium
1 ’ Mycoside C

 

 

lSBC-O
Mycoside C

7lC—O
Mycoside D

  
 

  

H37RV
phosphoglycolipid

 

 

 

 

 

I L . . A l L A
V v r v v y

r

3456789u101112131415

Figure 4. Infrared spectra

r—fl

/--\

l72Cl-l

amide-like compound

 

17lC-l
amide-like
compound

 

 

  
  
  

  

186C-l
amide—like compound

 

    
 
 
 
  
 

  

ghyfortuitum
probably mycoside F

 

 

 

  

P-4
phospholipid?

 

 
 

 

l86C-l
mycoside CM

1 A
7’ 1

7 8 9u1'0 11 12 1'3 14 15

l
j

w-
p.
m
0‘

Figure 5. Infrared spectra

244I-l
mycoside CM

 
 
  

 
 

62D-O
mycoside CM

 
 

 

152C-1
mycoside C

M

193C2-l
mycoside CM

A A L
1 1 T

6 7 8 9 LL10 11 12 13 14 15

Figure 6. Infrared spectra

700"

600‘* Figure 8. g: avium elution curve

500..
mg. 400 o
300 n
200‘»

100 4

 

 

 

HB50 B BE5 BESO EMl EM5 EM20 EAl

Eluent

700 “
600 “ Figure 9. 2441-1 elution curve
500 “
mg. 400 0
300 w

200 0

100 0

 

 

 

//A\JN\~F—T*\ 8”\=;f/\“‘r , L5

1 U

HB50 B BE5 BE50 EMl EM5 EMZO EAi

Eluent

mg.

mg.

700

600

500

400

300

200

100

700 ‘

600

500 «

400

300

200

100 ..

GI

 

Figure 10. 68C-0 elution curve

I I I T Ffi

 

 

HB ' B B EMT
50 E5 BF501”le 5 EM20 EAl

Eluent

Figure 11. 172Cl-l elution curve

 

 

 

I I t I I 1 1
E

HB B BE BE EM EM EM

//\\_1 . k3
50 5 50 1 5 20 A1
Eluent

mg.

mg.

700

600"

500

400-

300“

200‘

100-

700 “

600 ‘

500

400

300

200

100

1

 

 

 

 

 

 

 

 

 

 

 

Figure 12. 93C-O elution curve
L
I I I (A {\r T L
50 B BE5 BE50 EMl EM5 EM20 EAl
Eluent
4- Figure 13. 620—0 elution curve
.1.
I I I I/A\\\#/fi\\-L—" ‘TEA '
H350 B BE5 BE50 EMl EM5 EMZO l

Eluent

mg.

mg.

700‘

6001'

500‘

400'

300"

200»

100'-

 

Figure 14. 186C-l elution curve

 

 

 

 

700”

600"

500“

400L

300"

200L

100"

 

 

Eluent

Figure 15. fl, fortuitum elution curve

 

 

 

 

 

 

 

T | T T

EM20 EAl

HB50 B BES BESO EMl EM5

Eluent

mg.

mg.

700 "

600 n

500

400 3-

300

200

100

700

600

500

400

300 '

200

100 "

Figure 16. 193C2-l elution curve

 

 

 

MA L

I I V ' V ' '

HB50 B BE5 BE50 EMl EM5 EM20 EAl

Eluent

Jp

Figure 17. 193Cl-l elution curve

 

 

 

1

EA

B BE5 BESO EMl EM5 EMZO

50 1

Eluent

mg.

mg.

700

600

500

400

300

200 -

100

700

600

500

400

300

200

100

 

 

 

 

 

 

 

 

.P Figure 18. M, tuberculosis H37Rv
elution curve
'1.
EM5 EM20 EAl
Eluent
1-
Figure 19. P-39 elution curve
1b
9
+5
+ 1 A I/\‘\ T T j
H350 B BE5 BE50 EMl EM5 EM20 EAl

Eluent

700 «-
600 «-
500 ‘1

Figure 20. y, bovis B.C.G. elution curve

400‘-
mg.
300"

200'-

100‘-

 

F 3 I I fir I I fi

HB50 H BE5 BE50 EMl EM5 EM20 EA1

Eluent

700‘L

I

600‘

500*
Figure 21. M, tuberculosis H37RA elution curve

/\\ _/\_ 1{*\1 /\\_ /\ifi
BE5 EAl

 

fir

O EMl EM5 EM20

Eluent

 

mg.

mg.

700‘

600‘

5001*

400‘-

300-

2001

 

100‘ 4

Figure 20. M, bovis B.C.G. elution curve

 

700"

600‘

500"

400‘

3000

2001.

100‘

 

V

 

L
HB50

V

M J
I I I I I
EMl EMS EMZO EAl

Eluent

H BE5 BE50

 

Figure 21. M, tuberculosis H37RA elution curve

BE5 EAl

fir

O EMl EM5 EM20

Eluent

I! $.11 '..

mg.

mg.

700"

600 ‘1

500

400

300

200

100

700

600-

500

400

300

200 .

100

Figure 22. P—4 elution curve

1.

+

1b

 

«I
1

 

HBSO B BE5 BESO EMl EM5 E

 

 

 

M20 EAl
Eluent
Figure 23. 380s elution curve
0
{I
HB50 B BE5 BESO EMl EM5 EM2O EAl

Eluent

mg.

mg.

' 200 J

700 0
600 -
Figure 24.
500 v
400
300 v

200 0

100 J.

 

700 v
600
Figure 25.
500 v

400 “

300 0

 

100

HB50 B

P-8 elution curve

BE50 EMl EM5 EM

Eluent

P—15 elution curve

20

EA

BE5 BE50 EMl EM

Eluent

mg.

mg.

700 -

600

500 8

400

300

200

100

700

600

500 ‘

400

300

200 u

100

Figure 26. 228C—l elution curve

 

_f/\/\ /\ 1/\

HB B BE'BEjEM EMTEM EA

 

50 5 50 l 5 20 1
Eluent
Figure 27. 71C-0 elution curve

 

AM

f I I I I I I I
HB50 B BE5 BE50 EMl EM5 EM50 EAl

Eluent

 

mg.

mg.

700“

6001'

500‘

400~

300‘

200‘-

100'-

T

V

 

Figure 28. BllOH-l elution curve

 

7001

600‘

500‘

400 *-

300‘

2001-

100 J-

T

 

H350 B BE 5 BE - EM EM EM EA

Eluent

Figure 29. 152C-1 elution curve

 

HB

50 EM EM EM EA

50 l 5 20 l
Eluent

MA K
‘ 1 T 1 T I 1
B BES BE

700”

600‘

T

Figure 30. l7lC-l elution curve

V

500‘
400"

mg.
300‘»

200 4-
100-L M
HB ' 'BE 'BE %‘ L
50 B 5 50 1

 

 

 

 

I I
EM5 EMZO EAl

Eluent

700‘r

600--
' Figure 31. 158C-0 elution curve

500~>

400‘
mg.

I

300--

200-

 

100

{\a {\g. /r\_¥

I
BE50 EMl EM5 EM20 EAl

1

 

Eluent

v

700'

 

 

600'I
Figure 32. M. bovis Ravenel elution curve
500 "
400 v
300 “
200 n
100 0
AA A AM
I l I T I I W
81350 B ‘BESO EMl 8M5 EMZO 'EAl

Eluent

71C-O

 

 

 

 

 

 

 

 

 

172Cl—1

0? Ruin

 

 

L
StaYt EH3 POlhf

 

M, bovis Ravenel

 

 

 

 

 

4 ITIME

 

Figure 33. Tracings of gas chromatographic experiments.

Discussion

The distribution of type-specific lipids among the 25
strains of mycobactiera was nearly as expected for the types
of organisms studied. With the exception of strains P-lS,

M, bovis BCG, 193C -1, B110H-1, 152C-1, and 71C-0 each or-

1
ganism possessed the specific lipids expected for the type

of strain studied (as determined by morphological and cyto-
chemical methods). Lipids from human strains contained di-
mycoceronate of phthiocerol, the virulent bovine strains--
dimycoceronate of phthiocerol and mycoside B, M, fortuitum--
mycoside F, photochromogens (Runyon group I)-—mycoside A,
nonphotochromogens (group III)--mycoside CM and avian strains--
mycoside C.

Lipids from strain P-15, a skotochromogenic strain ob-
tained from Dr. E. H. Runyon, did not contain mycoside D.
Occasionally cultures of group II organisms will change
colony form (Fregnan et al., 1962). This change from a
smooth colony form to a rough form results in the loss of
ability to produce mycoside D. It was possible that during
cultivation of this organism the rough surface growing var-
iant was selected during the numerous transfers. As a con-
sequence, lipids extracted from the variant did not contain

mycoside D.

99

100

Mycoside B was not isolated from the lipids of M, boyi§_
B.C.G. According to Smith et al. (1960a) this lipid con-
stitutes from 1-5% of the total ethanol-ether extracted
lipid. This lipid could not be detected initially in el-
uates from the benzene-ether 50% and ether-methanol 1%
fractions. Chromatoqraphic separation of these fractions
on silicic acid did not separate mycoside B from these frac-
tions. Conceivably, this lipid might have been present in
a low and undetectable concentration.

Strain 7lC-0, typed by cytochemical means as a group
III organism, was found to elaborate mycoside D, a type-
specific lipid of group II organisms. This organism was
found to have a high rate of mutation from the white colony
form to the yellow form. The yellow mutant apparently was
selected by the cultivation conditions employed and became
the predominating organism. This organism grew on the bot-
tom of the culture bottles as a slimy yellow growth exhib-
iting no photochromic reaction. Navalkar et al. (1963)
have observed that some scotochromogenic organisms of a
smooth colony type may produce either mycoside CM or mycoside
D.

Strain 193C -1 is also a strain which has been observed

1

to from mutant colony forms with high frequency. This

101

organism and 193C2-l was isolated from the same mesenteric
lesion of a swine. Their simultaneous presence in the
same lesion, originally thought to be a case of multiple
infection may, in fact, have been a matter of colony varia-
tion. The conditions of cultivation may have served to
select the type III mutant. It grew as a whitish slimy mass
on the bottom of the flasks and contained mycoside CM'
Strain BllOH-l, classified as a group IV organisms by
morphological and cytochemical tests, was found to produce

mycoside C Growth rate characteristics are the only cri-

M'
teria used to separate some organisms into group III or
group IV. Strain BllOH-l was an intermediate organism
which could have been classed as a group III or a group IV.

Strain 152C-l apparently is another case of a high rate
mutant of differing colony form being selected by the con-
ditions of cultivation. This organism grew on the surface
in a rough form, somewhat different from the smooth slimy
growth expected of the original isolant.

Growth and collection of lipids for this study had
been nearly completed when Smith and fellow workers at the
University of Wisconsin published the first of their series

of papers correlating colony morphology and presence of

specific-lipids (Fregnan et al., 1961). It was, therefore,

102

impossible to use Smith's procedures to maintain and cul-
tivate one colony form. Application of these procedures

to such strains as 152C-l, 193C2-l, and 7lC-0 might have
provided some interesting observations concerning distribu-
tion of specific lipids and colony variants.

Minor difference may be observed in the infrared spectra
of mycoside CM isolated from the different strains. These
differences are probably due to the presence of other lipids,
such as fatty acids and napthoquinone derivatives to a
greater or lesser extent. Some of the fractions were not
rechromatographed and undoubtedly are in a more impure form
than the other fractions.

Not all of the lipid applied to the chromatographic
columns was recovered. As much as 50% of the material in
some cases was not eluted. The unrecovered material prob-
ably consisted mostly of glycerol, phospholipids and mycolic
acids. Glycerol may have been present in farily high con-
centration in lipid samples which had been partitioned be-
tween ether and water for non-lipid removal. This method
is known to be inefficient in the removal of glycerol
(Smith, 1963).

Phosphatide removal also my have been incomplete. Sol-

vent fractionation in certain cases usually does not effect

103

a 100% separation of lipids, some of one lipid or the other
will be partially soluble in the other solvent (Asselineau,
1952). In this case some of the phosphatides may have dis-
solved in the boiling acetone and been carried over to the
fraction to be chromatographed. Most of the phospholipids
would be tightly held to the column and not be eluted with
the solvents used in this study (Carroll, 1962). The my—
colic acids also are not eluted from florisil by the sol-
vents used in this study (Smith, 1963). Volatile compounds
lost during manipulations may also account for some of the
lipid not recovered.

Compound J a lipopeptide found in the lipids of

ABS '
172C1-l, l7lC-l, 244I-1, 62D-0, l86-l and P-39, appears to
be closely related to mycosides C and D (Smith et al., 1960a).
JABS resembles these lipids after their sugar moities have

been removed.

Gas Chromatography

The fatty acid fraction, eluted by ether-acetic acid 1%
appears to be a complex mixture of compounds, particularly
in the case of strain 7lC-0. Fatty acids isolated from this
strain (identified by retention time) are C12, C13, C14,

branched C15 (or unsaturated C14), 015, traces of branched

104

C16, C16, unsaturated C16, branched C17, C17, C18, unsatu-
rated C18 and branched C19 fatty acids.

The presence of many of these fatty acids has been shown
for virulent human strains by Cason and Tavs (1959) and
Cason and Miller (1963). These investigators have found
C15, branched C16, C16, unsaturated C16 (lo-hexadecanoic),
branched C17 saturated (8- and lO-methyl hexadecanoic) and
branched C17 unsaturated (10methyl-9-hexadecenoic) fatty
acids.

The branched C19 fraction may be tuberculostearic acid
(lO-methylstearic acid) (Cason and Tavs, 1959).

Nearly one-third of the fatty acids isolated from vir-
ulent and avirulent M. tuberculosis have carbon chain
lengths over 20 (Cason et al., 1953; 1958). In this study
concentrations of these acids were not determined, and may
have exceeded one-third of the total fattyanids. A large
portion of the fatty acids from g: boyi§_Ravenel was not
eluted from the gas chromatographic column and may have
been composed of fatty acids of carbon chain length ex-

ceeding 20.

Summary

Lipids extracted by ether-ethanol from 25 strains of
mycobacteria were fractionated by adsorption chromato-
graphy. Infrared spectra of the fractions were recorded.
Type-specific lipid compounds were found in the extracts
of human, bovine, avian and atypical strains.

Dimycoceronate of phthiocerol was found in the lipids
of two human strains (H37RA and H37RV) and one bovine strain
(M, boyi§.Ravene1), mycoside B was isolated from M, bgyi§_
Ravenel but not from M, 291i§_B.C.G. Other type-specific
lipids found were: mycoside A isolated from two photo—
chromogens (P-4 and P-8), mycoside F from M, fortuitum,
mycoside C from M, ayigm_and 158C-0 (isolated from a bovine
mesenteric lymph node), mycoside D from 71C-0 (also from
a bovine mesenteric lymph node) and mycoside CM.from strain
P-39 and 12 organisms isolated from swine mesenteric lymph
nodes and bovine body lymph nodes and Peyer's patches
(strains 172C -1, 186C-l, 193C

1 1 2
68C-0, BllOH-l, 93C-0, 152C-l and 171C—1).

-1, 193C -1, 228C-1, 2441-1,

105

Literature Cited

Alvarez and Tavel. 1885. Recherches sur le bacille de
lustgasten. Arch. Physiol. Pathol. 6,303-308 (cited
Mallmann, Mallmann and Ray, 1962).

Anderson, R. J. 1940. The chemistry of the lipids of
tubercle bacilli. Harvey Lect. 271-313.

Asselineau, J. 1952. Lipides du bacille tuberculeux.
Adv. Tuberc. Res. 551-55. S. Karger, Basel/New York.

Asselineau, J. 1960. Sur la structure d'un acide mycolique
isole' de la souche humaine Canetti de bacille tuber-
culeux. Bull. Soc. Chim. 135-141.

Asselineau, J. and T. Gendre. 1954. Sur les acides myco-
lieques de trois souches humaines de M, tuberculosis:
H37RA, H37RV, et un mutant H37RV streptomycino-resist-
ant. Bull. Soc. Chim. 1226-1233.

Asselineau, J. and E. Lederer. 1953. Chimie des lipides
bacteriens. Fortsch. der Chem. Orgon. Naturst. 195
172-273.

Asselineau, J. and E. Lederer. 1955. Chemical structure
and biological activity of mycolic acids. CIBA Symp.
Exper. Tuberc. p. 14-40.

Asselineau, J. and E. Lederer. 1960. Chemistry and
metabolism of bacterial lipides. in Lipide Metab-
olism, K. Kloch ed. J. Wiley and Son Inc. New York,
p. 337-406.

Asselineau, J. and J. Moron. 1958. Sur la composition
des grasses acetonosolubles de souches humaines de
bacille tuberculeux. Bull. Soc. Chim. Biol. 495899-
911.

Ball, E. G. 1934. Studies on oxidation reduction. XXI.

Phthiocol--the pigment of the human tubercle bacillus.
J. Biol. Chem. 106:515-524.

106

107

Benedict, A. A. 1955. Group Classification of virus
preparations by infrared spectroscopy. J. Bacteriol.
625264-269.

Bloch, H., E. Sorkin and H. Erlenmeyer. 1953. A toxic
lipid component of the tubercle bacillus. I. Isola-
tion from petroleum ether extracts of young bacterial
cells. Am. Rev. Tuberc. 615629-643.

Blout, E. R. and M. Fields. 1948. On the infrared spectra
of nucleic acids and certain of their components.
Science 107:252-257.

Blout, E. R. and M. Fields. 1949. Absorption spectra.
VII. The infrared spectra of some nucleic acids,
nucleotides and nucleosides. J. Biol. Chem. 178:
335-343.

Blout, E. R. and R. C. Mellors. 1949. Infrared spectra
of tissues. Science 110:137-138.

Brown, M., D. A. Yeadon, L. A. Goldblatt and J. W. Dieckert.
1957. Chromatography of phospholipids and related com-
pounds on glass paper impregnated with silicic acid.
Anal. Chem. 22,30-31.

Carroll, K. K. 1961. Separation of lipid classes by
chromatography on Florisil. J. Lip. Res. 2,135-141.

Cason J., C. F. Allen, W. DeAcetis and G. J. Fonken. 1956.
Fatty acids from the lipides of non-virulent strains
of of the tubercle bacillus. J. Biol. Chem. 220:893-904.

Cason J. and W. T. Miller. 1963. Complexity of the mix-
ture of fatty acids from tubercle bacillus. J. Biol.
Chem. 238:883-887.

Cason J., G. Sumrell, C. F. Allen, G. A. Gillies and S.
Elberg. 1953. Certain characteristics of the fatty
acids from the lipides of the tubercle bacillus. J.
Biol. Chem. 295,435-447.

108

Cason, J. and P. Tavs. 1959. Separation of fatty acids
from tubercle bacilli by gas chromatography: identifi-
cation of oleic acid. J. Biol. Chem. 234:1401-1405.

Chaput, M., G. Michel and E. Lederer. 1962. Structure du
mycoside CM, peptido-glycolipide de Mycobacterium
marianum. Biochim. et Biophys. Acta 6;,310-326.

Cole, A. R. H. 1956. Infrared spectra of natural products.
Fortsch. der Chem. Organ. Naturst. 13,1-69.

Cormier, M., P. Jonan and L. Girre. 1959. Spearation
des lipides par chromatographie sur papier imprégne'
d'acide silicique. II. Possibilités et limites de
la methode. Bull. Soc. Chim. Biol. 41,1037-1046.

Demarteau-Ginsburg, H. and A. M. Miquel. 1962. Sur la
presence d'ethylene-glycol dans les lipides de myco-
bacteries. Bull. Soc. Chim. Biol. 445678-682.

Dieckert, J. W. and R. Reiser. 1955. Separation of
phospholipids on silicic acid impregnated glass filter
paper. Fed. Proceed. 14:202-203.

Ebina, T., M. Motomiya, K. Munakata and O. Satake. 1962.
Pigments of unclassified mycobacteria. Am. Rev. Resp.
Diseases. 86:740-743.

Folch, J., M. Lees and G. H. S. Stanley. 1957. A simple
method for the isolation and purification of total
lipids from animal tissues. J. Biol. Chem. 226:497-
509.

Freeman, N. K., F. T. Lindegren, Y. C. Ng and A. V. Nichols.
1953. Infrared spectra of some lipoproteins and related
lipids. J. Biol. Chem. 203:293-304.

Fregnan, G. B., D. W. Smith and H. M. Randall. 1961. Bio-
logical and chemical studies on mycobacteria. J. Bac-
teriol. 82:517—527.

Fregnan G., D. W. Smith and H. M. Randall. 1962. A mutant
of a scotochromogenic mycobacterium detected by colony
morphology and lipid studies. J. Bacteriol. 8;;828-836.

109

Gastambide-Odier, M. M., D. W. Smith, H. M. Randall and A.
O. Koevoet. 1957. Comparison of the native lipids of
atypical acid-fast bacilli with the lipids of known
mycobacterial types. Am. Rev. Tuberc. 153843-845.

Gehrke, C. W. and D. F. Goerlitz. 1963. Quantitative
preparation of methyl esters of fatty acids for gas
chromatography. Anal. Chem. 55376-80.

Hammerschlag, A. 1889. Ueber bakteriologisch-chemische
untersuchung der tuberkelbacillen. Cent. fur Bakt.
55702.

Harrell, W. K. 1954. Fractionation and differentiation
of human, bovine and avian strains of Mycobacterium
tubeggulosis by means of their infrared spectrums.
Thesis, University of Michigan, Ann Arbor, Michigan.

Hirsch, J. and E. H. Ahrens, Jr. 1958. The separation of
complex lipide mixtures by the use of silicic acid
chromatography. J. Biol. Chem. 233:311-320.

Ikawa, M. and E. E. Snell. 1962. The occurrence of D-
alloisoleucine and D-leucine in mycosides and peptido-
lipids of bacterial origin. Biochim. et Biophys. Acta
593186-188.

Ikawa, M., E. E. Snell and E. Lederer. 1960. Occurrence
of D-phenylalanine, D-allo—threonine and other D—amino
acids in peptido-lipids of bacterial origin. Nature
5555558-560.

Jollés, P., F. Bigler, T. Gendre and E. Lederer. 1961.
Sur la structure chimique du "mycoside C," peptido-
glycolipide de Mycobacterium avium. Bull. Soc. Chim.
Biol. 453177-192.

Kenner, B. A., J. W. Riddle, S. W. Rockwood and R. H.
Bordner. 1958. Bacterial identification by infrared
spectroscopy. II. Effect of instrumental and environ-
ment variables. J. Bacteriol. Z5;l6-20.

110

Klotz, I. M. and P. Griswold. 1949. Infrared spectra
and the amide linkage in a native globular protein.
Science 109:309-310.

Kubica, G. P., H. M. Randall and D. W. Smith. 1956.
Correlation of biologic properties of strains of myco-
bacterium with their infrared spectrums. IV. Frac-
tionation and comparison of the lipids of human strains
of Mycobacterium tuberculosis by means of their in-
frared spectrums. Am. Rev. Tuberc. 153529-538.

Laneelle, G. and J. Asselineau. 1962. Isolement de pep-
tidolipides a partir de Mycobacterium paratuberculosis.
J. Biochim. et Biophys. Acta 55:731-732.

Lederer, E. 1961. Glycolipids of acid-fast bacteria.
Adv. Carbohyd. Chem. 55;207-238. M. L. Wolfrom ed.,
Academic Press, New York and London.

Lederer, E. and M. Lederer. 1957. Chromatography; a re-
view of principle and applications. 2nd Edn. Elsevier
Pub. Co. Amsterdam, New York.

Levine, S., H. J. R. Stevenson and R. H. Bordner. 1953a.
Identification of glycogen in whole bacterial cells by
infrared spectroscopy. Science 118:141-142.

Levine, S., H. J. R. Stevenson, L. A. Chambers and B. A.
Kenner. 1953b. Infrared spectrophotometry of enteric
bacteria. J. Bacteriol. 55:10-15.

MacLennan, A. P. 1962. The monosaccharide units in spe-
cific glycolipids of Mycobacterium avium. Biochem. J.
55:394-400.

MacLennan, A. P., H. M. Randall and D. W. Smith. 1961.
The occurrence of methyl ethers of rhamnose and fucose
in specific glycolipids of certain mycobacteria. Bio-
chem. J. 595309-318.

Mallmann, V. H., W. L. Mallmann and J. A. Ray. 1962. Cas-
sical and atypical pathogenic mycobacteria of human and
animal origin. 8th Annual International Congress for
Microbiology, Montreal, p. 121.

111

Navalkar R., E. Wiegeshaus, D. W. Smith and G. Fregnan.
1963. Colony form and mycosides of scotochromogenic
mycobacteria. Ann. Meeting of Nat. Tuberc. Assn.
Denver, Colo. Medical Abstracts 3C/25.

Noll, H. 1956. The chemistry of cord factor, a toxic
glycolipid of M, tuberculosis. Adv. Tuberc. Res.
1:149-182, S. Karger, Basel/New York.

Noll, H. 1958. The chemistry of the native constituents
of the acetone-soluble fat of M, tuberculosis (Brevan—
nes). II. Isolation and properties of a new crystal-
line napthoquinone derivative related to vitamin K2.
J. Biol. Chem. Q5259l9-929.

Noll, H. and H. Bloch. 1953. A toxic lipid component of
the tubercle bacillus ("cord factor"). Amer. Rev.
Tuberc. 513828-852.

Noll, H., H. Bloch, J. Asselineau and E. Lederer. 1956.
The chemical structure of the cord factor of Mycobac-
terium tuberculosig, Biochim. et Biophys. Acta 295
299-309.

Noll, H., R. Ruegg, V. Gloor, G. Ryser and O. Isler. 1960.
Die struktur einer vitamin-KZ-verbindung aus tuberkel-
bazillen und die synthese hoherer isoprenologe der vit-
amin-KZ-reihe. Helv. Chim. Acta 555433-438.

Noll, H. and E. Jackim. 1958. The chemistry of the native
constituents of the acetone soluble fats of Mycobac-
terium tuberculosis (Brevannes). I. Glycerides and
phospholipids. J. Biol. Chem. 255,903-917.‘

O'Leary, W. M. 1962. The fatty acids of bacteria. Bac-
teriol. Rev. 55:421-427.

Patterson, D. S. P. 1960. The effect of trace metals on
the growth and metabolism of Mycobacterium tubercuigsgg
avium. Tubercle 455191-202.

Randall, H. M., D. W. Smith, A. C. Cohn and W. J. Nungester.
1951. Correlation of biologic properties of strains of
mycobacteria with infrared spectrums. Amer. Rev. Tuberc.
55:372-380.

112

Randall, H. M., D. W. Smith and W. J. Nungester. 1952.
Correlation of biologic properties of strains of myco-
bacteria by means of their infrared spectrums. II.

The differentiation of 2 strains H37RV and H37RA
of M, tubercglgsi§_by means of their infrared spectrums.
Am. Rev. Tuberc. 555477-480.

Randall, H. M., S. V. Rao, G. B. Fregnan and D. W. Smith.
1961. Preliminary observations on the relationship
between glycolipid composition and colony morphology
of different strains of mycobacteria. Amer. Rev.
Resp. Disease 53,117.

Riddle, J. W., P. W. Kabler, B. A. Kenner, R. H. Bordner.
S. W. Rockwood and H. J. R. Stevenson. 1956. Bac-
terial identification by infrared spectrophotometry.
J. Bacteriol. 225593-603.

Runyon, E. H. 1959. Anonymous mycobacteria in pulmonary
disease. Med. Clin. H. A. 55:273-290 (cited by Mall-
mann, Mallmann and Ray, 1962).

Runyon, E. H. 1960. The anonymous mycobacteria in human
disease. 6th Annual Ivan H. Mattson I Memorial Con-
ference. J. S. Chapman ed., Charles C. Thomas,
Springfield, Ill.

Scopes, A. W. 1962. The infrared spectra of some acetic
acid bacteria. J. Gen. Microbiol. 25569-79.

Seibert, F. B. 1950. Constituents of mycobacteria. Ann.
Rev. Microbiol. 5:35-52.

Seibert, F. B. and E. Soto-Figueroa. 1958. Constituents
of mycobacteria in the chemistry and chemotherapy of
tuberculosis. E. R. Long, 3rd edn., The Williams and
Wilkins Co., Baltimore, Md.

Simon, S. and L. R. Hedrick. 1955. Infrared spectrophoto—
metry of Hansenula and Saccharomyces whole yeast cells
and yeast cellulose. J. Bacteriol. 5534-8.

 

Smith D. W. 1962. Personal communication.

113

Smith, D. W. 1963. Personal communication.

Smith, D. W., H. K. Harrell and H. M. Randall. 1954. Cor-
relation of biologic properties of strains of mycobac-
teria with their infrared spectrums. Amer. Rev. Tuberc.
52:505-510.

Smith, D. W., M. M. Gastambide-Odier, A. L. Koevoet and
H. M. Randal. 1957a. Specificity of lipid content of
members of the genus Mycobacterium. Bacteriol. Proc.
p. 83.

Smith, D. W., H. M. Randall, M. M. Gastambide-Odier and A.
L. Koevoet. 1957b. The characterization of mycobac-
terial strains by the composition of their lipid ex-
tracts. Ann. N. Y. Acad. Sci. 55:145-157.

Smith, D. W., H. M. Randall, A. P. MacLennan, R. K. Putney
and S. V. Rao. 1960a. Detection of specific lipids
in mycobacteria by infrared spectroscopy. J. Bacteriol.
15:217-227.

Smith, D. W., H. M. Randall, A. P. MacLennan and E. Lederer.
1960b. Mycosides: A new class of type-specific glyco-
lipids of mycobacteria. Nature 196:887-888.

Sternberg, K. 1903. Ueber die folgen der einverleibung
toter tuberkelbacillen. Cent. fur Bakt. 525645.

Stevenson, H. J. R. and O. E. A. Boldwan. 1952. Infrared
spectrophotometry as a means for identification of bac-
teria. Science 116:111-112.

Takeda, Y. and T. Ohta. 1944. Carotenoid in human tubercle
bacillus. J. Pharm. Soc. Jap. 5g;67-68, Chem. Abst.
55,1198 (1951).

Tarshis, M. S. and A. W. Frisch. 1952a. Chromogenic acid-
fast bacilli from human sources. I. Cultured studies.
Amer. Rev. Tuberc. 555278-288.

114

Tarshis, M. S. and A. W. Frisch. 1952b. Chromogenic acid-
fast bacilli from human sources. II. Pathologic studies.
Amer. Rev. Tuberc. 555289-301.

Tarshis, M. S. and A. W. Frisch. 1952c. Chromogenic acid-
fast bacilli from human sources. III. Hypersensitivity
studies. Amer. Rev. Tuberc. 555301-315.

Tsumita, T. 1956. Studies on the lipid of B. C. G. I.
Glyceryl mono-mycolate in wax C fraction of the lipid of
B. C. G. Jap. Jour. Med. Sci. Biol. 55205-216, Biolog.
Abst. 5556443 (1957).

Wong, S. and J. Weinzirl. 1936. An inexpensive synthetic
medium for growing Mycobacterium tuberculosis. Amer.
Rev. Tuberc. 555577:583.

Wood, L. E., V. B. Buhler and A. Pollak. 1956. Human infec-
tion with the "yellow" acid fast bacillus. Amer. Rev.
Tuberc. 155917-929.

Wren, J. J. 1961. Chromatography of lipids on silicic
acid. Chromatog. Revs. 55111-133.

 

111111111101IIIIIIIIIIIIIIIII;

1 7 9519