THE ULTRASTRUCTURE 0F
TREPONEMA REFRINGENS AND THE
ISOLATION AND CHARACTERIZATION

OF CYTUPLASIIIC FIBRIIS

A Disserioflon
Io: Hm“ U99 roe oI pIx. D.
MICHIGAN STATE UNIVERSITY
Susan Radtke Eipert

I976

:Hrwfi

This is to certify that the

thesis entitled
THE ULTRASTRUCTURE OF TREPONEMA REFRINGENS
AND THE ISOLATION AND CHARACTERIZATION
OF CYTOPLASMIC FIBRILS
presented by

Susan Radtke Eipert

has been accepted towards fulfillment
of the requirements for

_Eh...D.._ degree in W937

We L

 

 

 

/ Major professox

Date 1976

 

0-7639

 

I

.-u--.—-—- —-——-

-u—vvvww—rrmn. . . >1 ~.-—. ~o~ w~-'»....._. —.—-.~ a...» .'DA .w

. I... n... 4 .. .m.

 

 

 

 

ABSTRACT
THE ULTRASTRUCTURE OF TREPONEMA REFRINGENS

AND THE ISOLATION AND CHARACTERIZATION
OF CYTOPLASMIC FIBRILS

BY

Susan Radtke Eipert

The structural relationship of treponemal cytoplasmic fibrils
to other cell components was determined and the fibrils were isolated
and characterized ultrastructurally and chemically. Scanning and
transmission electron microscopy of whole and disrupted cells of
Treponema refringens (formerly the Nichols nonpathogenic strain of
T. pallidum) confirmed that the cells were spiral and had a cell
envelope resembling that of gram negative cells. Axial filaments,
originating subterminally near each end of the cell, remained
enclosed in the outer envelope through their entire length. A band
of parallel fibrils extended along the cell on the inside of the
cytoplasmic membrane. The band was oriented along the inner curve
of the cell's helix. The nuclear area of the cell was always
located adjacent to the fibrils. In micrographs of spher0plast-
like forms an actual connection between fibrils and DNA was evident.
Experiments to demonstrate further this connection by cell frac-
tionation or electron microscopy of the DNA from partially lysed

cells were inconclusive.

Susan Radtke Eipert

After various lytic agents were tested, a procedure was
developed to lyse completely the cells and release the cytoplasmic
fibrils. Cells were treated with guinea pig serum, lysozyme and
Brij 58 overnight and then sonicated until the suspension cleared.
A fibril-containing fraction was isolated by equilibrium ultracen-
trifugation of the lysate on a 0 to 57% or O to 76% linear
Renografin gradient. A band which contained most of the fibrils,
along with some membrane and axial filament hooks and basal bodies,
was found on the gradient. After treatment with DNAse and/or
Triton x-1oo, the fibril fraction was further purified by two or
more Renografin gradients. A protein with a molecular weight of
97,000 was shown to be the major component of the partially puri-
fied fibril preparation by SDS polyacrylamide gel electrophoresis.
Cross sectioned fibrils were polygonal, with visible substructure.
Electron micrographs of negatively stained isolated fibrils showed
two views or types of fibrils: wide and narrow. Each wide fibril
had a 3.0 nm.periodicity. A model of fibril structure is proposed
to account for the ultrastructural observations. Each fibril is
definitely a complex arrangement of several strands rather than a
simple tubule. Much of the electron microscopic evidence can be
accounted for by a model in which four strands are arranged so that
each is at a corner of a rectangular cross section. Fibrils in situ
are viewed from the narrow side, whereas both the wide and the
narrow sides are viewed in preparations of isolated fibrils.

Treponemal fibrils bear little ultrastructural or chemical
similarity to other prokaryotic or eukaryotic intracytoplasmic

structures. The function of the fibrils is unknown, but speculations

Susan Radtke Eipert
can be made based on their location within the cell. Their position
on the insides of the curves of the cell suggests that they may,
along with the peptidoglycan, play a role in maintaining the shape
of the organism. A likely possibility is that fibrils play a
role in DNA replication, segregation or transcription since the DNA
of these treponemes was always located next to the fibrils.

The ultrastructure of Treponema hgodysenteriae was also studied.
The morphology of these cells was similar to that of T. refringens
with the most obvious difference being the presence of 9 to 11 axial
filaments on each end of the cell as opposed to one to 3. Attempts
to demonstrate intracytoplasmic fibrils by electron microscopy of
negatively stained whole cells and cell fractions were inconclusive.
The axial filaments displayed a polarity sometimes seen in bacterial
flagella--the distal end of each fragment of a broken filament was

indented while the other end was pointed.

THE ULTRASTRUCTURE OF TREPONEMA REFRINGENS AND THE ISOLATION

AND CHARACTERIZATION OF CYTOPLASMIC FIBRILS

BY

Susan Radtke Eipert

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Microbiology and Public Health

1976

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to Dr. Sam
Black for his interest, guidance and encouragement through the
years. Stuart Pankratz has provided invaluable instruction and
advice. I would like to thank Dr. Tom Corner for acting as substi-
tute major professor and for his helpful comments on this manuscript.

Appreciation is also extended to my committee members, Dr. L.
Snyder and Dr. E. Kuntz, and to those whose equipment I used: Dr.
L. Snyder, Dr. R. Patterson, Dr. T. Beaman, Dr. L. Velicer, Dr. R.
Costilow and Dr. G. Hooper.

Cultures were provided by Dr. S. Jackson (T. refringens) and
J. Kinyon (T. hyodgsenteriae). Special thanks to Leigh Martin, who
cultured T. hyodysenteriae and provided comments and discussion.

These investigations were supported by the Department of
Microbiology and Public Health and by a National Institutes of

Health training grant through this department.

ii

 

 

M}

 

 

TABLE OF

INTRODUCTION. . . . . . . . . . . .

REVIEW

OF THE LITERATURE. . . . . .

Historical Perspective on Spirochetes and

Treponemes . . . . . . . .

CONTENTS

Shape and Morphology of Spirochetes. .
Treponemal Cytoplasmic Fibrils . . . .
Spirochetal Cell Envelope Fibrils. . .

Isolation of Structural Components

MATERIALS AND METHODS . . . . . . .

RESULTS

Organism and Cultivation . .
Nephelometry . . . . . . . .
CLBS Lysis Procedure . . . .
Fibril Isolation Procedure .
Chemical Characterization of
Light Microscopy . . . . . .

Phase Microscopy. . .

Darkfield Microscopy.

O O O O O
O O O O O
O O O O O

Fibrils .

Fluorescence Microscopy . . . .

Electron Microscopy. . . . .

Scanning Electron Microscopy. .

Negative Stains . . .
DNA Spreading . . . .
Thin Sections . . . .
Shadowing . . . . . .

Growth Curves. . . . . . . .
General Ultrastructure . . .

Determination of Optimal Microscopy
Shape and Size of Typical Cells

Cell Envelope . . . .

Spherical Atypical Cells. . . .

Axial Filaments . . .
Cytoplasmic Fibrils .
DNA and Cytoplasm . .
Induction of Spherical Forms

Treatment with Penicillin . . .
Treatment with Lysozyme . . . .

iii

from Treponemes . .

Techniques.

Page

l—‘l—‘dU‘lw

H H

l3
13
14
14
15
15
15
l6
16
17
l7
l7
17
18
18

19

19
21
21
24
30
37
43
52
56
62
62
62

(3‘1
5
l

. I>II. .
.5 r

 

 

LI'

Cell Lysis .

Treatment with Guinea
Treatment with Guinea

and Sonication.

Fibril Isolation .
Sucrose Gradients .
Renografin Gradients.

Chemical Characterization of

DNA—Fibril Connection. .
Aqueous Spreading Technique for
Microscopy. . .
3H-Thymidine Labeling . . . . . . .
Ultrastructure of Fibrils.

Thin Sectioned and Shadowed Fibrils

Fibrils

Serum .
Serum, Detergents

Negatively Stained Fibrils. .
Effect of Chemical, Physical or
Treatment of Fibrils.

DISCUSSION. . . .

O 0

Cell Envelope and Axial Filaments. .
Fibril Purification. . .
Fibril Location in Cell.
Fibril Ultrastructure. .
Comparison of Fibrils with Other
Eukaryotic Structures.
Prokaryotic Structures.
Eukaryotic Structures .
Function of Fibrils. . .
Maintenance of Shape.

Cell Division . .

APPENDIX. . . . .

LITERATURE CITED.

iv

Prokaryotic

Electron

Enzymat c

O O O

Page

63
63

75
79
79
80
88
91

92
92
93

100
114
118

118
120
122
123

129
129
131
132
133
134

137

144

Table

LIST OF TABLES
Page

Terminology used for treponemal cytoplasmic fibrils
and axial filaments by various authors . . . . . . . . . 9

Generation times of cultures inoculated with popu—
lations of varying ages. . . . . . . . . . . . . . . . . 19

Mean diameter of normal cells as measured from

micrographs produced by transmission electron

microscopy (TEM) and scanning electron microscopy

(SEM). . . . . . . . . . . . . . . . . . . . . . . . . . 3O

Inductions of convoluted forms by suspension of
treponemes in lysozyme solutions and buffers . . . . . . 64

Disruption of cells after treatment with guinea pig
serum (GPS) plus detergent or sonication or both . . . . 78

Treatments which did not disaggregate clumps of
fibrils, as determined by electron microscopy. . . . . . 88

Means of various fibril dimensions measured from
fibrils in thin sectioned cells. . . . . . . . . . . . . 98

Mean measurements of various dimensions of nega—
tively stained fibrils . . . . . . . . . . . . . . . . . 99

Figure

10

11

12

13

14

15

16

LIST OF FIGURES

Growth curve showing growth of cultures using
inocula of various ages. . . . . . . . . . . . . . .

Growth curve showing slow growth of culture inocu-
lated with 20-month-old culture and its recovery
When transferred 0 O O O O O O O O O O O I I O O O 0

Phase micrograph of T. refringens. . . . . . . . . .

Darkfield micrograph of T. refringens from a 6-day
culture 0 O O O O O O O O O O O O O O O O O O O O O 0

Longitudinal section through cell from 3-day culture

Scanning electron micrograph of helical treponeme
mug straighter cells C O O O C O O O O O O O O O 0

Scanning electron micrograph of helical treponeme
among straighter cells . . . . . . . . . . . . . . .

PTA-stained cells which have recently divided. . . .
Thin section of cells from a 5-day culture . . . . .

PTA-stained divided cells (from a 3-day culture)
still bridged with outer envelope (OE) . . . . . . .

Thin sections of cells from a 5-day culture. . . . .
Outer envelope vesicles and striated tubules (ST)
with knobbed ends (K) from population of cells dis-

rupted with guinea pig serum in Tris . . . . . . . .

PTA-stained cell with a striated tubule (ST) extend-
ing from its outer envelope (0E) . . . . . . . . . .

Thin sectioned striated tubules. . . . . . . . . . .

Thin sectioned cells, from a 9—day culture, which
have internal membrane structures (arrows) . . . . .

PTA-stained cell, from a 4-day culture, with numerous

membrane structures (arrows) . . . . . . . . . . . .

vi

Page

20

25

25

26

27

28
29

31

33

34

35

36

38

38

Figure Page
17 PTA-stained cells from a 4-day culture . . . . . . . . . 4O

18 Three PTA-stained cells, two wide and one narrow,
bridged with outer envelope (OE) . . . . . . . . . . . . 41

19 Phase micrograph of convoluted cell. . . . . . . . . . . 42
20 Phase micrograph of a spheroplast-like cell. . . . . . . 42
21 Thin sectioned convoluted form . . . . . . . . . . . . . 42

22 Thin sectioned spheroplast-like cell surrounded with

OE . O C O O O O O O O O O O O O O C O O I O O O O O O O 42

23 Two partially convoluted cells and a completely
convoluted cell, stained with PTA. . . . . . . . . . . . 44

24 Ammonium molybdate-stained cell which was disrupted
with guinea pig serum and lysozyme . . . . . . . . . . . 45

25 Negative stain of band A of Renografin gradient. . . . . 47

26 Uranyl oxalate-stained basal body (BB) and hook (H)
with visible substructure. . . . . . . . . . . . . . . . 48

27 Uranyl oxalate-stained basal body (88) which has
stain deposited around the point at which the axial
filament hook was attached (arrow) . . . . . . . . . . . 48

28 Thin sectioned cell with axial filament hook (H)
and basal body (88) apparently embedded in pocket
of cytoplasmic membrane (CM) . . . . . . . . . . . . . . 48

29 Longitudinal sections of axial filaments in peri-
plasmic space (AF) and within a tube of membrane (T) . . 49

30 Cross sections of cells with axial filaments (AF) in
periplasmic space. . . . . . . . . . . . . . . . . . . . 50

31 Thin sectioned axial filament (AF) in space between
outer envelope and cytoplasm of spheroplast-like

fom . C O O O O O O O O O O O C O O O O O O O O O O O 0 SO

32 Thin sections of cells and tubes of membrane (T)
containing axial filaments . . . . . . . . . . . . . . . 51

33 Negative stain of cells from a 2-month culture . . . . . 53

34 Distribution of the number of fibrils per cross
section of cell. . . . . . . . . . . . . . . . . . . . . 54

35 Longitudinal section of cell with fibrils (F) adjacent
to inner edge of cytoplasmic membrane (CM) . . . . . . .

m
Ln

vii

Figure Page

36 PTA-stained cell disrupted with guinea pig serum and
lysozyme O O O O O O O O O O O O O O Q C I O O O O O O O 5 7

37 Thin sectioned cell with fibrils (F) and nuclear
area (N) ViSible O O O O O O O O O C O O O O I O O O O O 58

38 Thin sectioned spheroplast-like cell . . . . . . . . . . 5q
39 Thin sectioned spheroplast-like cell . . . . . . . . . . 59
40 Two wet-mounted cells stained with euchrysine. . . . . . 61
41 Wet-mounted cells stained with euchrysine. . . . . . . . 61

42 Two recently divided cells vitally stained with
euchrysine . . . . . . . . . . . . . . . . . . . . . . . 61

43 wet-mounted spheroplast-like cell stained with
euchrysine . . . . . . . . . . . . . . . . . . . . . . . 61

44 Phase micrograph of 5-day culture suspended in 0.067
M phosphate buffer, pH 7.8, with 2% n-butanol and 100
g/ml lysozyme O O O O O O O O O O O I O O O O O O O O O O 6 5

45 Phase micrograph of a 5-day culture suspended in
0.067 M phosphate buffer, pH 7.8, guinea pig serum . . . 65

46 Cells disrupted with guinea pig serum and negatively
stained with uranyl acetate. . . . . . . . . . . . . . . 67

47 Cell disrupted with guinea pig serum and negatively
stained with ammonium molybdate. . . . . . . . . . . . . 68

48 Cells disrupted with guinea pig serum and lysozyme
and negatively stained with PTA. . . . . . . . . . . . . 69

49 Cell disrupted with guinea pig serum and lysozyme
and negatively stained with 0.5% PTA, pH 6.7 . . . . . . 7O

50 Cell disrupted with guinea pig serum and lysozyme
and negatively stained with ammonium molybdate . . . . . 71

51 Cell disrupted with guinea pig serum and lysozyme
and negatively stained with PTA. . . . . . . . . . . . . 72

52 PTA-stained spiral band of fibrils from cell disrupted
with guinea pig serum and lysozyme . . . . . . . . . . . 73

53 Thin sections of disrupted cells and membrane vesicles

from culture treated with guinea pig serum and
1y sozyme O O O O O O O O O O O O O O O O O O O O O O O 0 7 4

viii

Figure Page

54 Cells disrupted with guinea pig serum and Brij 58
and negatively stained with PTA. . . . . . . . . . . . . 76

55 Ammonium molybdate stain of cell sonicated after
exposure to guinea pig serum and lysozyme. . . . . . . . 77

56 Silicon tungstate-stained membrane from Renografin
band A O O O O O O O O O O O O O O O O O C O O O O O O O 77

57 Average position of bands A and C when cell lysate,
with or without DNase, was centrifuged to equili-
brium on 0 to 76% Renografin gradients . . . . . . . . . 82

58 Ammonium molybdate-stained material from Renografin
barld A O O O O O O C O O O O O O O O O O O O O O O O O O 8 3

S9 PTA-stained material from Renografin band D. . . . . . . 84

60 PTA stain of large clump of fibrils from Renografin
band C O O O O O O O O O O O O O O O O O O O O O O O O O 86

61 Flow chart of procedure for isolation of fibrils
from treponemes. . . . . . . . . . . . . . . . . . . . . 87

62 Uranyl oxalate—stained band C from third Renografin
gradient 0 O O O O O I O O I O O I I O O O O O O O O O O 89

63 Absorbance readings of SDS polyacrylamide electro-
phoresis gels stained with Coomassie blue. . . . . . . . 90

64 DNA from T4 bacteriophage. . . . . . . . . . . . . . . . 93

65 Cell disrupted with guinea pig serum, prepared for
electron microscopy by the Kleinschmidt spreading
technique and stained with PTA and uranyl acetate. . . . 93

66 Cells disrupted with CLBS, prepared for electron
microscopy by the Kleinschmidt spreading technique,
stained with uranyl acetate and rotary shadowed
with platinum. . . . . . . . . . . . . . . . . . . . . . 94

67 Cell disrupted with CLBS, prepared for electron
microscopy by the Kleinschmidt technique and stained
with uranyl acetate. . . . . . . . . . . . . . . . . . . 94

68 Direct cross sections of fibrils in normal cells
(a,b,c), a convoluted cell (d) and disrupted cells

(e,f). . . . . . . . . . . . . . . . . . . . . . . . . . 96

69 Drawings of sectioned fibrils drawn approximately
to scale and positioned on the cytoplasmic membrane. . . 97

ix

rt

Figure Page
70 Sectioned disrupted cell containing fibrils (F)

which have been freed from the cytoplasmic membrane
101

71 Longitudinally sectioned fibrils (F) . . . . . . . . . . 101

72 Isolated fibrils which were shadowed with platinum-
carbon in the direction of the arrow . . . . . . . . . . 102

73 PTA-stained narrow (N) and wide (W) two-stranded
fibrils. o o o o o o o o o o o o o o o o o o o o o o o o 103

74 PTA-stained wide two-stranded fibrils (W) from pre-
paration which had been kept frozen. . . . . . . . . . . 104

75 Silicon tungstate-stained wide fibrils with ribbon
(R) of stained material lying on each side . . . . . . . 104

76 Drawings of fibril types and substructure seen in
preparations of isolated fibrils stained with PTA

or 001‘ o o o o o o o o o o o o o o o o o o o o o o o o o 106

77 UOx-stained wide fibrils with periodic cross stria-
tions (P) I o 0 I o o o o o o o o o o o o o o o o o o o o 107

78 UOx-stained wide fibrils with ribbons (R) alongside. . . 107
79 UOx-stained wide fibrils . . . . . . . . . . . . . . . . 108
80 UOx-stained wide (W) and narrow (N) fibrils. . . . . . . 109
81 UOx-stained wide and narrow fibrils. . . . . . . . . . . 109
82 Distribution of widths of narrow fibrils . . . . . . . . 111
83 Distribution of widths of wide fibrils . . . . . . . . . 112

84 Each histogram shows the distribution of wide fibril
widths from micrographs taken of one microscope grid . . 113

85 PTA-stained spiral band of narrow fibrils from fibril
preparation which had been dried and resuspended . . . . 115

86 Negative stain of narrow fibrils in bands (arrows) . . . 116

87 PTA-stained fibrils (F) disrupted with exposure to
4 M urea overnight . . . . . . . . . . . . . . . . . . . 116

88 Relationship between components of cross and longi-
tudinally sectioned fibrils. . . . . . . . . . . . . . . 125

89 Model of ultrastructure of treponemal fibril as con-
structed from micrographs of negatively stained fibrils. 128

X

Figure Page
90 T. hyodysenteriae negatively stained with PTA. . . . . . 139
91 Enlargement of Figure 90 . . . . . . . . . . . . . . . . 140

92 PTA-stained bundles of fibers from CLBS—disrupted
T. hyodysenteriae. . . . . . . . . . . . . . . . . . . . 141

93 PTA-stained axial filament fragments from CLBS-
disrupted T. hyodysenteriae. . . . . . . . . . . . . . . 141

94 PTA-stained axial filament fragments from CLBS-
disrupted T. hgodysenteriae. . . . . . . . . . . . . . . 142

xi

suf

PIE

tier
eve:

dis:

— — ___..__-_---__-_ ’—-—

 

I NTRODUCT ION

Since the first observation of spirochetes, their unique
morphology has fascinated many researchers. Light microscopy was
sufficient to show the helical nature of the cells and even the
presence of axial filaments, but the use of electron microscopy
resolved many aspects of spirochetal ultrastructure. It also
allowed the discovery in treponemes of a previously unknown
structure--cytoplasmic fibrils. Although discovered in 1968, the
fibrils have not been well characterized. The only clearly estab-
lished fact is that several parallel fibrils are located on the
inside of the cytoplasmic membrane of treponemes. Little informa-
tion has been available on their chemistry or ultrastructure and
even the question of whether there are one or more bands has been
disputed.

This study was undertaken to determine the function of the
cytoplasmic fibrils in treponemes. This objective was pursued by
clarifying the structural relationship of treponemal fibrils to
other cell components and by isolating the fibrils to examine their
ultrastructural and chemical nature.

The fibrils were found to be adjacent to the nuclear area of
the cell and to be located on the inner side of the cell's curves.

Based on this evidence, two possible functions for fibrils were

 

 

I'li‘

L

2

proposed--shape maintenance and association with DNA in the process
of cell division.

Treponemal cytoplasmic fibrils had not been isolated, nor had
a lysis procedure been developed which efficiently released fibrils
from the cells. In the present study several methods were tried
before a satisfactory procedure was developed. The use of guinea
pig serum, Brij 58 and sonication disrupted the cells to a degree
which permitted the isolation of the fibrils from all other cell
components visible in the electron microscope. Examination of the
ultrastructure of isolated fibrils and fibrils within cells pro-
vided evidence on which each fibril was prOposed to be a complex
arrangement of strands. Study of the chemical and ultrastructural
nature of the fibrils demonstrated that they are not similar to any

other known prokaryotic or eukaryotic structure.

I.

64

H

eel

H
u

at.

 

REVIEW OF THE LITERATURE

Historical Perspective on Spirochetes and Treponemes

 

Antony van Leewenhoek reported to Robert Hooke in 1681 that he
had seen "a sort of animalcules that had the figure of our river-
eels" in fecal material. These animalcules had "a wavy nimble motion"
and bent their bodies serpent-wise, and shot through the stuff as
quick as a pike through the water" (27). Van Leewenhoek later
observed similar spiral organisms in material from between teeth.
Dobell believed the descriptions to be those of spirochetes.

In 1838 Ehrenberg classified the bacteria within the animal
kingdom and defined the genus of large free-living flexible helical
filaments as Spirochaeta. It has only one species, S. plicatilis
(17). His drawings and terminology suggest that he considered the
spirochetes to be helical chains of bacteria rather than single
helical cells. Cohn classified the bacteria with the plants rather
than the animals and in his 1875 classification system even sug-
gested that S. plicatilis was nothing more than a colorless form of
the alga Spirulina. The genus Spirochaete was described as long
spiral flexuous filaments of cells without phycochrome (16).

Winter's 1881 classification system (112) is one example of
the many classifications of that time. In it, Spirochaeta was
distinguished from Bacillus and Beggiatoa only by being composed of
a long flexible spiral filament of cells rather than a straight chain

3

4
of cells. In an English translation of this classification, bacteria
belonging to the genus Spirochaeta were described as

...cells united in long slender threads, mostly showing

narrow spiral windings. The filaments have the liveliest

of movements and clearly prOpel themselves forward and

backward, but are also able to bend in various ways. (19)

The practice of describing spirochetes as spiral filaments of
cells rather than single spiral cells was followed in most, but not
all, classification systems in the 19th century. This practice
changed near the turn of the century, so that by 1917, when the
Committee of the Society of American Bacteriologists on Characteri-
zation and Classification of Bacterial Types made its preliminary
report (111), spirochetes were clearly considered individual spiral
organisms.

Spirocheta pallidum and other spirochetes were given the
generic name of Treponema in 1905 by Schaudinn (93,94) shortly
after he and Hoffmann (95) discovered spirochetes in syphilitic
chancres. At the same time that they found spirochetes connected
with syphilis, they found another type of spirochete, T. refringens,
in condylomata of a patient without syphilis (93,95). Swellengrebel
initiated the term Spirochaetaceae, which contained the genera
Spirochaeta and Treponema as well as a new genus, Borrelia (16).
Through the years, other genera have been included with the spiro-
chetes or removed from that group. The most recent classification
(99) lists 5 genera in the order Spirochaetales and the class

Spirochaetaceae: Spirocheta, Cristispira, Leptospira, Borrelia

and Treponema.

The t
blotype Iii
T. pallidu
from a rat
pallidum.
successful
Pathogenic
SO~called c
Other specj
inunologic
paI-‘L’O‘Jenic:

saPmphytic

Spirocl
”ganisms (C

darkfield

For examPle,

ChdnCres an:

is a right-1

 

pallidum Ce;

Cell,

 

Beside:
spiroch e tes

   
 

5

The treponeme used in the present studies is Treponema refringens
biotype Nichols, formerly called the Nichols nonpathogenic strain of
T. pallidum. It has been cultured since 1933, when it was isolated
from a rabbit infected with the pathogenic Nichols strain of T.
pallidum. At the time, Kast and Kolmer (61) considered it to be a
successful cultivation of T. pallidum, even though it was no longer
pathogenic for rabbits. Recently this organism, along with other
so-called cultivable strains of T. pallidum, was reclassified in
other species of treponemes (99). Since these organisms differ
immunologically and morphologically from T. pallidum and are not
pathogenic (108), they probably were originally cultured from

saprophytic genital treponemes rather than T. pallidum (98).

Shape and Morphology of Spirochetes

 

Spirochetes have usually been described as spiral or helical
organisms (or chains of organisms). Occasional reports based on
darkfield Observation of the cells disputed this, claiming that some
treponemes are not helical, but rather are in planar wave forms.

For example, Sequira (96) reported that T. pallidum from syphilitic
chancres and T. pertenue are flat wave forms, whereas T. refringens
is a right-handed spiral. More recently, Cox (22) found that T.
pallidum cells are made up of flat wave forms, in l to 5 planes per
cell.

Besides being helical, or perhaps planar wave forms, and flexible,
spirochetes are defined by the presence of axial filaments inter-
twining with the protoplasmic cylinder and enclosed by the outer

envelope (99). Bradfield and Cater (14) hypothesized that the axial

 

filaments a
filaments a
which pulls
(72) showed
from one er.
subtermina‘.
a variable I
ends overla.
responsible
(59) demons‘
ing cells wI'
the organisr
Spirocheta 5
cOiled COM;
Vere mthinta;
QIYCan laYez
PeptidoglyCa
my Shape.

The sire

 

6
filaments are the cause of the spiral shape of spirochetes. Axial
filaments attached at each end of a cell would provide the tension
which pulls the cell into a spiral shape. Listgarten and Socransky
(72) showed, however, that the axial filaments are not continuous
from one end of the cell to the other. Each filament is inserted
subterminally at one end only and extends towards the other end for
a variable distance. Usually filaments originating from opposite
ends overlap. As further evidence that the axial filaments are not
responsible for the shape of spirochetes, Joseph and Canale-Parole
(59) demonstrated that the removal of the axial filaments by treat-
ing cells with hydrochloric acid did not alter the coiled shape of
the organisms. Furthermore, the peptidoglycan layer purified from
Spirocheta stenostrepta or S. litoralis frequently retained its
coiled configuration (59,60). Thus, they concluded that spirochetes
were maintained in a permanently coiled shape by the helical peptido-
glycan layer, and not by tension on the cell by axial filaments.
Peptidoglycan isolated from leptospires (102) also retained its
wavy shape.

The similarity of axial filaments to bacterial flagella in
ultrastructural and chemical characteristics (59) has been the basis
for assuming that axial filaments play a role in spirochetal motility,
but that role has not been proven. It has not been possible to
demonstrate loss of motility after removal of axial filaments because
axial filaments lie inside the outer envelope and cannot be removed
by nonlethal treatments. Mutant spirochetes lacking axial filaments

or axial filament components have not yet been obtained (59). The

 

use of ant
informatic

Surrc
Plasnic me:
1

(‘1). The

gram‘negat.

 

ference be
membrane (
lecan (33
0f Spiroch
in Contras
gap hen-lee
transparex
cytoplaSm

(48'62'88

7
use of antisera against axial filaments yielded no conclusive
information (9).

Surrounding the cytoplasm of gram-negative cells is the cyto-
plasmic membrane, the peptidoglycan layer and the outer membrane
(21). The cell envelope of spirochetes is similar to that of
gramrnegative bacteria, with the most obvious morphological dif—
ference being the presence of the axial filaments between the outer
membrane (termed outer envelope in spirochetes) and the peptido-
glycan (53,70,87,88). In electron micrographs the peptidoglycan
of spirochetes is located directly next to the cytoplasmic membrane,
in contrast to most gram-negative organisms, in which there is a
gap between the membrane and the peptidoglycan (70,75). An electron
transparent area which contains tiny strands is present inside the
cytoplasm of spirochetes and is regarded as the nuclear area

(48,62,88,91,107).

Treponemal Cytoplasmic Fibrils

 

In 1968 Oinnnikov and Delektorskij observed filaments in the
cytoplasm of sectioned treponemes (83). They also mentioned that
there were several thin, parallel bands which stretched along the
body of negatively stained treponemes. These they regarded as the
pattern of the treponeme surface. At a meeting of the Italian
Society of Parasitology in 1968, two reports at a round table dis-
cussion on the fine morphology of spirochaetes mentioned unusual
structures: Swain and Anderson (101) reported wedge— or comma-shaped
masses in the cytoplasm next to the cytoplasmic membrane of sections

of T. pallidum, and Babudieri (4) observed small bundles of slender

 

fibrils in
not realize
structures.
of "micrctj
the cytopla
Since
one or more
negatively
Cross Sect;
pathOgenic
1" PEItenuE.
nemes such
(4’48'51.SE
(45) and T.
is “0 unive
confused vi
The di.
in the rang
a diameter <
Since they <
the axial f:
“m (63). T)
Hougen to be
and 5 to 20
the end of t
tents (47,5j

0n
9 band 0;

 

8
fibrils in negatively stained Reiter treponemes. It was apparently
not realized that these were probably different views of the same
structures. Also in 1968, Klingmuller et a1. (68) reported a band
of "microtubule filaments" lying next to the cytoplasmic membrane in
the cytoplasm of T. pallidum.

Since that time, other researchers have repeatedly observed
one or more bands of parallel fibrils situated along the length of
negatively stained treponemes or under the cytoplasmic membrane of
cross sectioned treponemes. These fibrils have been reported in
pathogenic treponemes such as T. pallidum (47,68,80,82,83,10l,107),
T. pertenue (81), and T. cuniculi (49) and in nonpathogenic trepo-
nemes such as the Reiter, Kazan, Stavropol and Nichols strains
(4,48,51,58,80,82,83), T. microdentium, T. minutem, and T. calligyrum
(45) and T. phagedenis, T. refringens and T. vincentii (44). There
is no universal name for these fibrils and often the terminology is
confused with that for axial filaments (Table l).

The diameters of the fibrils have usually been reported to be
in the range of 6 nm (58) to 9 nm (51). Klingmfiller et al. reported
a diameter of 4 nm (68), but this measurement is open to question
since they gave 10 nm as the diameter of axial filaments, whereas
the axial filament diameter probably ranges from 12 nm (58) to 30
nm.(63). The number of fibrils per band was reported by Hovind
Hougen to be 6 to 8 (44,45,47,49),lmn:wider ranges of 4 to 14 (58)
and 5 to 20 (51) have also been noted. The fibrils originate near
the end of the cell, close to the insertion points of the axial fila—
ments (47,51). Most researchers apparently assumed that there is

one band of fibrils running the length of the cell (51,58,68,107),

 

Table 1.

Authors

 

Klingmilllei

v _ .
Ovcmnikov

Jones et a1

JaCkSOn and

Hovind How
Andersen

 

9

Table 1. Terminology used for treponemal cytoplasmic fibrils and
axial filaments by various authors

 

 

Axial Cytoplasmic
Authors Date filaments fibrils
Klingmuller et a1. 1968 Fibrillen Mikrotubuli-
filamente
Ovcinnikov and Delektorskij 1969 superficial deep fibrils
fibrils
Jones et a1. 1970 axial fibers body fibrils
Jackson and Black 1971 axial filaments fibrils
Hovind Hougen and Birch- 1971 endoflagella microtubules
Andersen
Hovind Hougen 1972 flagella microtubules
Wiegand et a1. 1972 axial filaments deep filaments
Hovind Hougen 1974 flagella tubules

 

but it has been reported that there are 2 or 3 bands of fibrils
rather than one (4,66,79) or that bands originate at either end of
the cell and interdigitate in the middle (44,45,57). Many investi-
gators have stated that the fibrils are located on the same side of
the cell as, and directly under, the axial filaments (44,47,48,49,68,
81,101).

Fibrils were at least partially released from autolyzed cells
(51) or from cells treated with Myxobacter AL-l protease (44,45,47,48),
sonication and detergent (51), trypsin, or sodium dodecyl sulfate
and homogenization (50). Fibrils released from cells often remained

parallel and spiral (50,51) and were sometimes broken (44). The

 

fibrils of
T. phageder

The ul
tubular bec
(44,47,495
Possibly st
Outer stran
Units. Cro
coma-shape
°b1°n9 (107
centers (49
fibrils has
prObabll' ar

HOVind
c01.1ch be 115

Treponema S

haVe been f

 

MCIOSCOPY

been SUCCes:

10
fibrils of T. refringens appeared to be less fragile than those of
T. phagedenis or T. Vincentii (44).

The ultrastructure of the fibrils has usually been described as
tubular because negative stain appeared to penetrate the lumen
(44,47,49,51). Jones (58) noted that the fibrils have a beaded or
possibly striated appearance and Jackson (50) hypothesized that the
outer strands of the fibrils are the outer edges of spherical sub-
units. Cross sectioned fibrils are not round, but wedge- or
comma-shaped (101), slightly elongated (81), triangular (50), or
oblong (107). Fibril cross sections have slightly electron dense
centers (49) or a punctate structure (79). The chemical nature of
fibrils has not been mentioned, with the sole exception that they
probably are protein because they are sensitive to trypsin (51).

Hovind Hougen (47) has suggested that the presence of fibrils
could be used as a criterion for classification of the genus
Treponema since they appear to be unique to that genus. Fibrils
have been found in every treponeme species observed by electron
microscopy with the single exception of T. genitalis (46). Attempts
to find fibrils in leptospires (26,47) or borreliae (75) have not
been successful. Fibrils with a diameter of 5 to 8 nm were seen
in preparations of intact and disrupted cells of Spirocheta stenc—
strepta (59), but no mention was made of their number, location or
ultrastructure.

Hypotheses as to the function of treponemal cytoplasmic fibrils
have included motility (58,80,81,83), skeletal support (48,81,82)
and maintenance of the spiral shape (51,58,83). It has also been

suggested that fibrils may be the site for enzymatic processes

 

(82,101) , i

protein (48

Other
envelOpe of
the presenc
the Proton
trePOIlemes
0r hYaluron
found to be
had a mOlec:
tion) ' Sim
spirocheteS
iC'terobaemc
leptospireS
(110). in d
of tho 0m;
and in the
v“? 30111th

as3°Ciated

11
(82,101), intracytoplasmic transport (48), or synthesis of flagellar

protein (48).

Spirochetal Cell Envelope Fibrils

 

Other types of fibrils have also been reported in the cell
envelope of spirochetes. Yanigihara and Mifuchi (115,116) reported
the presence of branched microfibers 3 to 5 nm in diameter between
the protoplasmic cylinder and the outer envelope of Borreliae
treponemes and leptospires that had been treated with dilute alkali
or hyaluronidase. The leptospiral microfibrils were isolated and
found to be carbohydrate and protein in a 1:1 ratio. The protein
had a molecular weight of 72,000 (Yanigihara, personal communica-
tion). Similar fibrils have been observed in the envelope of large
spirochetes (12,72), in sodium deoxycholate-treated Leptospira
icterohaemorrhagiae (114), in the cell wall of thin sectioned
leptospires (98), in sodium deoxycholate-treated sex-ratio organisms
(110), in disrupted Borrelia Vincentii (11), in the outer envelope
of two out of seven species of treponemes observed by Hougen (44,46),
and in the outer envelope of T. refringens (50). These descriptions
vary somewhat and probably portray more than one type of fibril
associated with the cell envelope of various species. The location,
ultrastructure and generally smaller size of these spirochetal peri-
mural fibrils indicate that they are clearly different from treponemal

cytoplasmic fibrils.

Isolation of Structural Components from Treponemes

 

Many antigens and other cell components have been chemically

extracted from treponemes (53,86), but isolation of intact structural

 

component
outer env
was solut
dialysis

had been

axial fil
A cell we.
was isola1
cell fract
CYtoplasmj

been devel

 

12
components from spirochetes has been limited to axial filaments,
outer envelope, and a cell wall preparation. The outer envelope
was solubilized by sodium dodecyl sulfate and then reaggregated by

dialysis against a MgCl solution (36,57). After the outer envelope

2
had been solubilized with sodium dodecyl sulfate or butanol, the
axial filaments were sheared from the cells in a blender (8,34).

A cell wall preparation including peptidoglycan and polysaccharide
was isolated by centrifugation after cells were disrupted with a
cell fractionator and proteolytic enzymes (3). No internal or

cytoplasmic components have been isolated, nor has a lysis procedure

been developed which would allow isolation of such components.

 

The CL
NiChols nor
5- Jackson,
C°nsiSted T
(BEL). 1.4=
0'025: and
LaboramriE

m" A 91

from 10 to
InCUbatiOn
Vacuum jar
r0“finely e

at 7000 to

To “01-

 

MATERIALS AND METHODS

Organism and Cultivation

 

The culture of Treponema refringens, formerly called the
Nichols nonpathogenic strain of T. pallidum, was obtained from Dr.
S. Jackson, Baylor College of Medicine, Houston. The growth medium
consisted of (9/100 m1 glass distilled water):spirolate broth base
(BBL), 1.45; brain heart infusion (BBL), 1.85; tryptone (DIFCO),
(0.025; and asparagine (DIFCO), 0.05. Filtered rabbit serum (Flow
Laboratories) was heated for 30 min at 56 C to inactivate complement
and added to the autoclaved medium at a concentration of 9% (vol/
vol). A 9% (vol/vol) inoculum.was added and cultures were grown in
from 10 to 400 mu lots in screw-capped tubes or prescription bottles.
Incubation was at 35 C inside an anaerobic jar (Torbal) or glass
vacuum jar under a nitrogen atmosphere. Cultures were transferred
routinely every 4 to 7 days. Cells were harvested by centrifugation

at 7000 to 10,000 x g for 10 min.

Nephelometry

 

To monitor growth, 1 ml of culture was diluted with 9 ml of
phosphate buffer (pH 7.0) and read in a nephelometer (Coleman
Nepho-Colorimeter). According to Jackson (50), a suspension of

8
2 x 10 cells per ml had a reading of about 40 nephelos units.

13

 

The
in 1 all C
serum (re
Company 0
aPproxima
solution 4
temperatui
watt Ultra
15 to 45 S

3.10405 (S

35 c.

z?
:9;
(D

—

 

14

CLBS Lysis Procedure

 

The pellet from 200 ml of centrifuged culture was resuspended
in 1 ml of Tris buffer (0.04 M Tris pH 7.6) and 1.5 ml guinea pig
serum (reCOnstituted lyOphilized serum from Grand Island Biological
Company or fresh frozen serum from Suburban Serum Labs), after which
approximately 5 mg of lysozyme (Sigma) and 0.6 ml of a 5% (wt/vol)
solution of Brij 58 were added. The suspension was left at room
temperature from 2 h to overnight and then sonicated on an MSE 100
watt Ultrasonic Disintegrator until the suspension cleared (usually
15 to 45 sec). Three milligrams deoxyribonuclease I, E.C. No.
3.1.4.5 (Sigma) was added and the lysate was shaken for an hour at

35 C.

Fibril Isolation Procedure

 

Linear 0 to 57% or 0 to 76% Renografin (Squibb Renografin-76;
66% meglumine diatrizoate and 10% sodium diatrizoate) gradients were
made with 0.04 M Tris buffer (pH 7.4). Four milliliter gradients
were made in 5 ml tubes and 1 ml of CLBS lysate was layered on top.
Gradients were centrifuged in a SW 50.1 rotor in a Beckman Model
L3—50 ultracentrifuge at 130,000 x g for 2 h. The band containing
fibrils was collected with a Pasteur pipet and dialyzed overnight
against water or buffer. The dialyzed fibrils were incubated with
DNase, Triton X-100, or both and centrifuged again on a Renografin
gradient. The incubation and centrifugation of the fibrils was

repeated one more time.

 

Sodi
gels used
Fairbanks
glass tub!
allowed t(
and allows
Polyacryla
amperage o
Pretein We
0n the 981:
were Scanm
ph°t0meter
5“! “fiber ar

protei

 

 

 

15

Chemical Characterization of Fibrils

 

Sodium dodecyl sulfate (SDS) polyacrylamide electrophoresis
gels used to analyze fibril preparations were made according to
Fairbanks (29) with the following modifications. Acid cleaned
glass tubes were coated with dimethyl dichlorosilane (Sigma),
allowed to dry and then rinsed with hot water, 95% ethanol, water
and allowed to dry again. Gels (8 to 11 cm high) containing 7.5%
polyacrylamide and 1% SDS were pre-electrOphoresed at a constant
amperage of 2 mA per tube. Protein samples containing 20 to 50 pg
protein were boiled 5 min with 1% SDS and 2% mercaptoethanol, layered
on the gels and run at 4 mA per tube. Coomassie blue-stained gels
were scanned at a wavelength of 540 nm on a Gilford 2400-8 Spectro-
photometer. The molecular weights of standards had been determined
by Weber and Osborn (106).

Protein was analyzed by Lowry's method (73), hexose by the
anthrone procedure (2), DNA by the diphenylamine procedure (2), and

pentose by the orcinol test (109).

Light Microscopy,

 

All light microscopy and photomicrography was performed with
a Zeiss photomicroscope. The automatic exposure control was used

for phase and darkfield photography.

Phase Microscopy,
Wet mounts of cells were observed with phase optics using a
100 X oil-immersion objective and photographed using Kodak Plus-X

film.

 

Darkfield

When
the light
included a

lens with

For f
a high We
necessary
devflopmen
staining
and tilted
diethl’l ed
and some C
5 x Crysrs
stained f d
After ring

slide an d

in the prr

 

 

l6

Darkfield Microscgpyfi

 

When mounts of cells were observed by darkfield microscopy,
the light source was a high—pressure mercury lamp and the optics
included a cardioid condenser and a 100 X oil immersion objective

lens with an iris diaphragm.

Fluorescence Microscopy

 

For fluorescence microscopy, the microscope was equipped with
a high pressure mercury lamp. Exposure times of 2 to 15 min were
necessary with Kodak High Speed Ektachrome film pushed in its
development to an ASA of 640. In the procedure used for fluorescent
staining of fixed cells (5), smears on coverslips were air dried
and fixed overnight in Carnoy's solution and in 95% ethanol:95%
diethyl ether (1:1). After fixation, the smears were rehydrated
and some of them were treated with bovine pancreas ribonuclease
5 X crystallized (Nutritional Biochemicals). All smears were
stained for 20 min in acridine orange McIlvaine's buffer (pH 6.0) .
After rinsing, the wet coverslip was placed, cells down, on a clean
slide and the edges were sealed with nail polish. Many variations
in the procedure were tried. The major ones included the substi-
tution of euchrysine (in potassium-sodium phosphate buffer, pH
7.4) for acridine orange and the inclusion of deoxyribonuclease I,
E.C. No. 3.1.4.5 (Sigma) after the RNase incubation.

For vital staining, a wet mount was prepared containing culture
and one drop of a 0.02% solution of acridine orange in McIlvaine's
buffer, pH 2.4 to 6.0, or euchrysine in phosphate buffer, pH 7.4.
Agar coated slides helped to slow or stop the motion of the cells

for photography.

17

Electron Microscopy,

 

Scannipg Electron Microscopy,

 

Cells were collected on a 0.2 pm Nucleopore filter and fixed
in 2% glutaraldehyde or in 2% glutaraldehyde plus 1% osmium tetroxide.
The fixatives were dissolved in 0.067 M phosphate buffer (pH 7.2),
with or without 0.05 M or 0.15 M NaCl. After dehydration in an
alcohol series (5 min each in 50%, 70%, 95% and 100% ethanol), the
cells were dried in a Sorvall critical-point dryer and then sputter-
coated with gold. Cells were observed and photographed with an ISI

Super Mini-SEM scanning electron microscope.

Negative Stains

 

A11 transmission electron micrographs were taken with a Philips
EM-300 electron microscope operating at 80 Kv. For negative stain—
ing, cells were put on a carbon-Formvar grid and stained with a drop
of stain. Isolated fibrils were stained on a thin carbon or Parlodian
coated grid by floating the grid on the stain. The concentrations
of stains used, unless otherwise noted, were: 0.5% phosphotungstate,
pH 7.5 (PTA): 0.5% silicon tungstate, pH 7.0; 2% ammonium molybdate,
pH 7.5; 0.4% uranyl acetate in 0.05 M HCl, and 0.5% uranyl oxalate,

pH 6.8 (00x).

DNA Spreading

 

A variation of the Kleinschmidt spreading technique was used
for visualization of treponemal DNA (23). Various cell preparations
in 0.5 M ammonium acetate, 0.1 mg/ml cytochrome c (Nutritional Bio-

chemicals) were dropped onto the surface of 0.25 M ammonium acetate

 

in a parafi
was picked
ethanol an?
ethanol. 9

shadowed w 1

evaporatori

Cells
0.22 m Ki.
according ‘1
in 3‘ gluta
embedded if
113 Ultrotc
“Ian“ ace:

solutio“ ( J

 

18
in a paraffin coated glass petri plate. The cytochrome c monolayer
was picked up on a carbon-Parlodian coated grid, dehydrated in 50%
ethanol and stained for 30 sec with 0.002% uranyl acetate in 90%
ethanol. Occasional grids were also stained with PTA or rotary
shadowed with platinum in a Denton Vacuum DV—502 high vacuum

evaporator.

Thin Sections

 

Cells to be embedded for thin sectioning were collected on a
0.22 um Millipore filter, embedded in agar and fixed with osmium
according to the Ryter-Kellenberger procedure (65). Pre-fixation
in 3% glutaraldehyde was used only occasionally. The cells were
embedded in Epon and sectioned with a DuPont diamond knife on a
LKB Ultrotome III ultramicrotome. Sections were stained with 2%
uranyl acetate for 30 min and 0.5% lead citrate in a sodium hydroxide

solution (about pH 12) for 5 min.

Shadowing

Isolated fibrils on carbon-Parlodian grids were shadowed with
platinum-carbon at an angle of 15° in a Denton Vacuum DV-502 high

vacuum evaporator .

H —
A"
.~ “nu .‘rt—r

 

Nephe
Cultures (
existed be
hr PortiOn

 

RESULTS

Growth Curves

 

Nephelometry was used to record the growth of T. refringens
cultures (Figure 1). It was assumed that a direct relationship
existed between bacterial numbers and nephelos units (28). A 48
hr portion of each growth curve representing the exponential growth

phase was used to calculate the generation time (Table 2). The

Table 2. Generation times of cultures inoculated with p0pulations
of varying ages

 

Time period used

 

Age of inoculum in calculation Generation time
3 days Day 1 to day 3 23 hours
9 days Day 2 to day 4 35 hours
21 days Day 2 to day 4 40 hours
8 months Day 6 to day 8 49 hours

 

generation time was longer when older cultures were used as inoculum.
Growth did occur when a 20-month-old culture was used as inoculum,
although the stationary phase was quickly reached. The culture
inoculated with the 20-month-old population appeared to have returned
to normal since when it, in turn, was used as inoculum, the culture

19

 

'/\("'LII1rv

klr“r1 LJ r‘

0f iDOCu

20

 

100.4. I '
904
80 4 ’/

 

NEPHELOS UNITS
is
9

 

 

10612 35667 8§1b

TIME (DAYS)

 

Figure 1. Growth curve showing growth of cultures using
inocula of various ages. The growth was measured at one—day inter-
vals by nephelometry of a 1 to 10 dilution of each culture. Age
of inocula denoted by:

'V 3 days
El 9 days
0 21 days
a 8 months
V'20 months

grew a

regula

genic
appare
which
day 1
result
abOVe.

21

grew at a rate comparable to that of a culture inoculated with
regularly transferred stock (Figure 2).

Jackson (50) reported the generation time of Nichols nonpatho-
genic T. pallidum (T. refringens) as 45 hr. However, this was
apparently calculated from the earliest part of the growth curve,
which probably included the lag phase. A calculation based on
day 1 to day 3 of her graph, which is part of the exponential phase,
resulted in a generation time of 24 hr, similar to the value reported
above. L. Martin (personal communication) found that the generation

time of this organism is 20 hr.

General Ultrastructure

 

Determination of Optimal Microscopy Techniques

The diameter of Treponema refringens approaches the limit of
resolution of the light microscope, making observation with bright—
field microscopy difficult. Phase contrast and darkfield microscopy,
because of the increased contrast they provide, were found to be the
best modes for light microscopy studies. Fluorescent microscopy of
acridine orange- or euchrysine-stained cells proved useful for the
observation of RNA, but was of little value in the study of DNA
(see General Ultrastructure: DNA and Cytoplasm).

Various procedures were tested in order to gain adequate preser-
vation of the cells for scanning electron microscopy (SEM). Gluta-
raldehyde, in some cases followed by osmium tetroxide, was used in
phosphate buffer, with or without NaCl added to produce the approxi-
mate osmotic conditions (0.3 osmolal) determined by Hardy and Nell

(35) to preserve best the typical shape of treponemes. The cells

22

 

1 .4
88. ,
80‘ l
701

m 60‘

+—

32 50-

:3
40:

m

0

..J 30‘

Lb

I

o.

e 20*
10 , . .

 

 

 

 

o i 2 3 21 5 6 i 53 9'9 1'0
TIME(DAYS)

Figure 2. Growth curve showing slow growth of culture inoculated
with 20-month-old culture and its recovery when transferred. Minimal
growth occurred when a 20-month population was used as inoculum ( ‘ ).
When the culture which had been inoculated with the 20-month popula-
tion was itself used as inoculum after 16 days (I ) it grew as well

as a culture inoculated with a regularly transferred 21-day pOpula-
tion ( 0 ).

23

were then dehydrated in an alcohol series and subjected to critical-
point drying. The majority of cells in all preparations were rela-
tively straight, but some cells in each preparation were helical.
Those specimens fixed with both glutaraldehyde and osmium had the
highest percentage of well preserved cells, regardless of the
osmotic conditions.

Phosphotungstic acid (PTA), silicon tungstate and ammonium
molybdate, at varying pH values, were all satisfactory negative
stains for study of the cells by transmission electron microscopy
(TEM). Various fixation procedures were tested in the preparation
of thin sections for TEM. When the Ryter-Kellenberger fixation
procedure was preceded with glutaraldehyde fixation, the membranes
were well fixed, but the cytoplasm was somewhat clumped. Neither
the DNA nor the cytoplasmic fibrils was well defined. Much better
fixation of the cell contents resulted when the Ryter-Kellenberger
procedure was repeated using a shorter time for osmium fixation
and deleting the glutaraldehyde prefixation. The cytoplasm was
evenly distributed, the fibrils were clearly visible, and fine
strands were seen in an electron lucent space typical of the nuclear
area in Ryter-Kellenberger fixed bacteria. When lanthanum nitrate
was substituted for the Ryter-Kellenberger uranyl acetate post-
fixation, the DNA clumped inside the nuclear area. One drawback
to Ryter-Kellenberger fixation was that the outermost membrane, known
in treponemes as the outer envelope, was farther away from the cell
than when glutaraldehyde was used prior to the Ryter-Kellenberger

Procedure.

24

Shape and Size of Typical Cells

 

Typical living treponemal cells viewed by darkfield or phase
microscopy were undulate (Figures 3 and 4). This appearance was
determined by electron and light microscopy to be analogous to the
projection of a three-dimensional helix onto a two-dimensional
screen. Under the light microscope the lower portion of each turn
of the helix appeared out of focus when the upper part was in focus
(Figure 3). Furthermore, the wavy appearance was maintained on all
sides as the cells rotated (due to water currents or cell motility).

Cells thin sectioned for TEM were occasionally cut in such a
plane that their helical nature was apparent. The cell in Figure
5 winds back and forth and also passes out of and re-enters the
plane of the figure. SEM micrographs were even more conclusive
as to the helical nature of the cells and also demonstrated that
the helix is right-handed (Figures 6 and 7). Considerable variation
of wavelength and amplitude was noted from cell to cell and often
even within the same cell. The mean wavelength was about 880 nm
and the mean amplitude was about 170 nm. Negatively stained cells
were dried and flattened onto the surface of the supporting film
and consequently were not suitable as evidence for the helical nature
of the cells.

SEM, light microscopy, and TEM of negative stains disclosed
various other information about the shape of treponemes. The terminal
portions of cells were often straight (Figures 6 and 7; see also
Figures 3 and 4) with tapered ends (Figure 8). Occasional parts of
an otherwise helical cell were also straight (Figure 7), often near

division points.

Partia
of Ce?
cells
Visib;

I...

 

 

25

 

 

Figure 3. Phase micrograph of T. refringens. Note
partially out of focus cells which indicate spiral nature
of cells (long arrows), and the straight ends on some
cells (short arrows). Occasional convoluted cells are
visible. Bar r 10 um.

 

-

Figure 4. Darkfield micrograph of T. refringens
from a 6 day culture. Bar = 10 um.

26

 

Figure 5. longitudinal section through cell from 3 day
culture. Bar = 0.5 pm.

Figure 6. Scanning electron micrograph of helical treponeme
among straighter cells. Bar = 0.5 pm.

 

In

t ‘E
o "f )I ‘

Figure 7. Scanning electron micrograph of helical treponeme
among straighter cells. Note straight areas of cell (arrows).
Bar = 0.5 um.

 

29

 

Figure 8. PTA-stained cells which have recently divided.
Note tapered ends. Bar = 0.5 um.

 

Tabl
sectioned

scanning 1

Table 3.

W

Preparatio

Thin sect i
Negatively

CritiCal_p<

encrmously
Cell 19mm
tip) from (
2 to 7 (la)
cells from

from 8 to I

 

I
(Cu

V) are ti
Lain Se Stir
Hhich are

adarkl
(:on 7.3 +

30

Table 3 shows the diameter of the cell as measured from thin
sectioned cells, negatively stained cells and cells viewed with the

scanning electron microscope. The length of the cells varies

Table 3. Mean diameter of normal cells as measured from micrographs
produced by transmission electron microscopy (TEM) and
scanning electron microscopy (SEM)

 

Diameter Number of cells

 

Preparation of cells Microscope (nm) measured
Thin sectioned TEM 263 i, 4 66
Negatively stained TEM 226 :_ 9 ll
Critical-point dried SEM 196 + 17 ll

 

enormously, especially in the exponential growth phase cultures.
Cell lengths (measured on phase micrographs from cell tip to cell
tip) from 6 to 20 um were more common in actively growing cultures
(2 to 7 days) and occasionally much longer cells were seen. The
cells from a 15-day culture were found to be more uniform, varying

from 8 to 16 pm, with a mean length of about 12 um.

Cell Envelope

 

Outer envelope (0E), peptidoglycan, and cytoplasmic membrane
(CM) are the components of the treponemal cell envelope visible in
thin sections (Figure 9). Both OE and CM are trilaminar membranes
which are composed of an electron light layer bounded on each side
by a dark layer. The mean width of the OE and CM were calculated

to be 7.3 :_0.2 nm and 7.0 :_0.4 nm, respectively. The CM was

Fig;
IaYers Cf
' PE;
the fibr

 

 

 

 

31

 

 

Figure 9. Thin section of cells fran a 5 day culture. The
layers of the cell envelope are clearly visible: outer envelope
(OE), peptidoglycan (P) and cytoplasmic membrane (CM). Note also
the fibrils (F), nuclear area (N), and ribosomes (R). Bar = 0.5 um.

32

difficult to measure accurately because the adjacent cytoplasm
obscured its inner boundary. The empirical ratio of the width of
dark:light:dark components was about 2:3:2 for both membranes. The
layer assumed to be peptidogylcan was a uniformly dense band (6.0
1 0.4 nm wide) lying in direct contact with the outer layer of the
CM.

Prior to complete separation of dividing cells, the OE was the
only structure connecting the two cells. After division, a tail of
OE often remained (Figure 10). The identification of the membrane
tail as OE was confirmed by examination of a thin section of a
recently divided cell (Figure 11). OE could sometimes be identi-
fied in negative stains of cells or membrane vesicles by its unique
substructure. This consisted of electron dense particles about 8
nm in diameter which were hexagonally packed and separated by approxi-
mately 2 nm of electron light space (Figures 25 and 54). Vesicles
of OE from disrupted cells were also occasionally found joined in
large masses (Figure 12).

These masses of OE vesicles usually were connected to branched
or unbranched tubules with distinctive striations and terminal knobs
(Figure 12). Striated tubules were occasionally associated with the
OE surrounding normal cells (Figure 13), and more often with the OE
surrounding disrupted cells (Figure 46). Striated tubules had a
fairly uniform diameter of 35 i'2 nm in both negative stains and
thin sections. In negative stains, dark bands with a periodicity of
4.0 to 5.5 nm extended across the central dark area (lumen) of the
tubule and into part of the lighter area on each edge (Figure 13).

Striated tubule substructure was more apparent in thin sections,

33

 

- .H—u _
’. #' C.
t .0” 1-“ -

x\~()E

Figure 10. PTA—stained divided cells (from a 3 day culture)
still bridged with outer envelope (OE). Both daughter cells have
membranous structures (arrows) and one has a tail of OE remaining
from previous division. Bar = 0.5 pm.

\|(|

 

34

 

 

Figure 11. Thin sections of cells from a 5 day culture. OE
extends past end of one cell. Cross sections show fibrils (F) and
nuclear area (N). The internalized membrane is continuous with
the cytoplasmic membrane (CM). Bar = 0.5 pm.

35

 

Figure 12. Outer envelope vesicles and striated tubules (ST)
with knobbed ends (K) from population of cells disrupted with
guinea pig serum in Tris. Outer envelope substructure (S) is
visible. Negatively stained with ammonium molybdate. Bar = 0.5 pm.

36

 

Figure 13. PTA-stained cell with a striated tubule (ST)
extending from its outer envelope (OE). Dark bands (B) cross
the tubule but do not extend completely across the outer white
area. Bar - 0.5 um.

37

showing membrane (presumably the structure which is continuous with
the OE) surrounding the inner banded region of the tubule which in
turn forms the lumen (Figure 14).

The CM, sometimes with peptidoglycan fragments attached,
occasionally invaginated into the cell. These internalized membranes
were common in thin sections of cells from 5- and 9-day cultures
(Figure 15), but not in 3-day cultures (Figure 5). Occasionally
the opening to the periplasmic space was visible (Figure 11).
Membranous structures were also apparent (at an average of about
760 nm apart) in negative stains of cultures more than 3 days old
(Figure 16) and occasionally in younger cultures (Figure 10). Cells
with these structures had more negative stain deposited along their
sides and were narrower than cells lacking such structures (in
negative stains, 153 :_8 nm as opposed to 226 :_9 nm; in thin sec-
tions 245 :,7 nm as opposed to 263 :_4 nm). The narrow negatively
stained cells with membrane structures were generally less wavy than
the wider cells (Figure 17), as were many cells in old cultures
observed under the light microscope. Divided but still connected
cells of both types could be found in negative stains (Figures 10
and 18) and occasionally a narrow cell and wide cell were still

connected with OE (Figures 18 and 23).

Spherical Atypical Cells

 

In aging cultures and under certain conditions described below,
normal T. refringens cells converted to two types of spherical
atypical cells 2 to 6 um in diameter: a convoluted form and a

spheroplast-like form (Figures 19 and 20).

envelope.
1..
p ‘3-"
’, .’ -x
‘ ’I' a a y

38

 

 

Figure 14. Thin sectioned striated tubules. The banded
region (B) is covered with a membrane (M), presumably outer

Bar = 0.5 um.

   

Figure 15. Thin sectioned cells, from a 9 day culture, which
have internal membrane structures (arrows). Note also the fibrin

(F) and nuclear area (N). Bar = 0.5 um.

39

 

Figure 16. PTA-stained cell, from a 4 day culture, with
numerous membrane structures (arrows). Bar = 0.5 pm.

4O

 

Figure 17. PTA-stained cells from a 4 day culture. One is
wide, the other is narrow with membrane structures (arrow).
Bar = 0.5 pm.

41

O
A
-_ . 1 ‘
1rg;m_ 1§E¢FJIEE:“
..rlr' “w— '
. _- M 11"."

.
I,
‘ 0E
“ I
4" ,

Figure 18. Three PTA-stained cells, two wide and one
narrow, bridged with outer envelope (OE). The narrow cell has
membrane structures (arrows). Bar = 0.5 um.

.E: m.o n mom .on mzo mo mocmuum pom Amy hogan Hansen wuoz .Azuv oamunaoa oaammamouho an ouoason
ma EmnHmou>u owumsumwa .mo gums oooodouw9m Hams oxfialummamouwsmm ooqofiuowm owns .NN madman
.E: m.o u Hum .Amu asumflmOpwumom one Azov meanness uafimnflmouxu sue: monsoon moonwaau uwsmmam
Iouwu mo msofiuuom on» monsouusm Amos omoaw>cm House .suou oousao>coo owcowuoom owns .HN shaman
.E: 0H n ucm .EmmHmoumo
ooumsumflo monsouusm Amos smeao>so nouso .HHou uxqalumuamouonmm mo nomumouuwfi manna .om owsmfim
.5: OH H hum .HHoo pouodo>coo mo smoumowoflfi omega .mH ousmah

 

43

The convoluted spheres formed when the protoplasmic cylinder,
bounded by the CM and peptidoglycan, curled up inside the outer
envelope. A negative stain of two partially convoluted cells and
one completely convoluted cell is shown in Figure 23. Thin sections
through a convoluted form revealed the expanded outer envelope
surrounding the curled-up cytoplasmic cylinder which is bounded
by CM and peptidoglycan (Figure 21). The cytoplasm of spheroplast-
1ike cells was disrupted, in the form of one or more small spheres
within the ballooned—out sphere of OE (Figure 20). From the outer
layer inward, thin sections of these cells revealed an expanded OE,
CM with traces of peptidoglycan attached and, finally, disrupted
cytoplasm (Figure 22).

Cultures incubated three to four weeks or longer were found to
contain large numbers of spheroplasts. It should be noted that
attempts to induce these forms in younger populations with lysozyme
or penicillin failed. A few convoluted forms were present in normal
cultures of any age (Figure 3) and were readily induced, especially
in cultures beyond the exponential growth phase (see Induction of

Spherical Forms).

Axial Filaments

 

The axial filament was composed of a basal body, a hook and a
filament. The filament portion was wavy much like bacterial flagella
and consisted of a core and a sheath (Figure 24) which together had
a diameter of approximately 20 nm. The core alone measured about
13 nm in diameter. The hook connecting the basal body and filament

was not sheathed and had the same diameter as the core as well as a

44

   

 

Figure 23. Two partially convoluted cells and a completely
convoluted cell, stained with PTA. The convoluted portion of
each cell is completely surrounded with outer envelope (OE).

One cell has membrane structures (arrows) along its length.
Bar = 0.5 pm.

45

 

Figure 24. Ammonium molybdate-stained cell which was dis-
rupted with guinea pig serum and lysozyme. Fibrils (F) are visible
within the cell. Axial filaments (AF) with basal bodies (BB) and
hooks (H) emerge through break in OE. The sheath (Sh) is not
present on the distal part of either axial filament, thus revealing
the core (C). Bar = 0.5 um.

46
distinctive substructure (Figure 26). The basal body was mushroom—
shaped when viewed laterally (Figure 24) and appeared to be composed
of a round cap, with a diameter of approximately 50 nm, situated on
the end of the hook (Figure 25). Basal bodies no longer attached to
hooks are present in Figure 27.

The basal body, or insertion apparatus, of each axial filament
was rooted in the cytoplasmic cylinder near the end of the cell.
Usually one, two or three axial filaments were seen at each end of
a cell. Figure 24 shows a disrupted cell in which the axial fila-
ments appear released from the OE but still attached to the cytoplasmic
cylinder. It was difficult to determine on the basis of negative
stains whether the basal body was actually in the cytoplasm, with
the axial filament extending through the cytoplasmic membrane, or
whether the basal body was only embedded in a pocket in the CM,
while remaining outside the membrane (Figure 58). The basal body
appears to lie outside the CM in a micrograph of a thin sectioned
cell in Figure 28.

In thin sections, one, two or three axial filaments were some-
times seen just inside the OE within the periplasmic space of typical
cells (Figures 28, 29 and 30) or between the OE and the CM of either
type of atypical spherical cell (Figure 31). However, many cross
sections of normal cells did not contain any axial filaments (Figures
15 and 9) and axial filaments not in the cell, but simply enclosed
in trilaminar membrane were also seen in longitudinal or cross
section (Figure 32; see also Figures 29 and 30). Occasionally a
connection between this membrane tube and cellular OE was visible

(Figure 32; see also Figure 29) indicating that the axial filament

47

 

Figure 25. Negative stain of band A of Renografin gradient.
Note outer envelope vesicles with substructure (S), axial filament
hooks with basal bodies (BB) and a fibril (F). Bar = 0.5 um.

48

Figure 26. Uranyl oxalate-stained basal body (BB) and hook
(H) with visible substructure. Bar = 0.1 um.

 

Figure 27. Uranyl oxalate—stained basal body (BB) which has
stain deposited around the point at which the axial filament hook
was attached (arrow). A fragment of axial filament (AF) is also
present. Bar = 0.1 um.

 

Figure 28. Thin sectioned cell with axial filament hook
(H) and basal body (BB) apparently embedded in pocket of cyto—
plasmic membrane (CM). Bar = 0.1 um.

49

 

 

Figure 29. Longitudinal sections of axial filaments in peri—
plasmic space (AF) and within a tube of membrane (T). Bar =
0.5 um.

ll‘iilh I... ul\

SO

 

Figure 30. Cross sections of cell with axial filaments (AF)
in periplasmic space. Also present is a cross sectioned tube (T)
of membrane enclosing an axial filament. Bar = 0.1 pm.

 

Figure 31. Thin sectioned axial filament (AF) in space
between outer envelope and cytoplasm of spheroplast-like form.
Bar = 0.1 pm.

 

51

 
  

\
llh ‘ ’ .
‘1 t ‘
‘ 1.;
Figure 32. Thin sections of cells and tubes of membrane (T)
containing axial filaments. The membrane of the tube is continuous
with cellular OE at the arrows. Bar = 0.5 um.

In

’00

't)

Pu

(I)
to

was protruding away from its normal position next to the cytoplasmic
cylinder while still enclosed in OE. Such tubes of membrane sur-
rounding axial filaments were commonly seen in thin sections but
were seen in negative stains of only old (Figure 33) or disrupted
(Figure 48) cells. They were never seen in cells subjected to
critical-point drying for SEM. Striated tubules were occasionally
seen extending from the OE of normal negatively stained or thin
sectioned cells (Figure 13). Their substructure was clearly more
complex and their diameter more uniform than the simple tubes of OE
which enclosed axial filaments. Compare Figure 13 with Figure 48

and Figure 14 with Figure 32.

Cytoplasmic Fibrils

A band of several cytoplasmic fibrils ran through the cell
adjacent to the inner side of the CM: the location of these fibrils
was most evident in cross sections of cells (Figure 15; see also
Figures 9 and 11). The number of fibrils per cell varied from as
few as 5 to as many as 21. The most predominant numbers fell in
the range of from 6 to 10 (Figure 34). Since the cytoplasm of con-
voluted cells was undisturbed, the fibrils were in their normal
[position in these cells (Figure 21). More than one cross section
of the band was usually present on the inside of the CM of sphero—
‘plast-like cells (Figure 22). Details concerning the size and
ultrastructure of fibrils are reported in a later section.

In longitudinal sections, the fibrils lay adjacent to the CM,
crossing from one side of the section to the other (Figure 35).

Fibrils invariably lay along the inside of the curves in those

 

Figure 33. Negative stain of cells from a 2 month culture.
Axial filaments (AF) arising from basal bodies (BB) near the end of
each cell remain within the periplasmic space or extend away from
the cell within a tube (T) of outer envelope (OE). Fibrils (F) can be
seen within each cell and extending away from the cells. Stained
with 0.5% PTA pH 6.7. Bar = 0.5 um.

54

 

uumasn or 03055 szcnous
3. ?

 

 

 

 

 

 

 

 

 

 

 

 

1* r1 r1
56789muuuummummmm

NUMBER OF FIBRILS PER CROSS SECTION

 

Figure 34. Distribution of the number of fibrils per cross
section of cell. The number of fibrils was counted in 98 treponemal
cross sections in which the fibril band was clearly visible.

lh

Figure 35.

Longitudinal section of cell with fibrils (F)
adjacent to inner edge of cytoplasmic membrane (CM).

area (N) is found next to the fibrils at several points along the

Nuclear

cell. At arrow, fibrils cross from the inside of one curve of
the section to the inside of the next curve. Ribosomes are visible
within the cell. Bar = 0.5 um.

 

56

sections which demonstrated the helical nature of the cell (Figures
5 and 35).

Fibrils were not ordinarily visible in negatively stained cells
but some very old (Figure 33) or disrupted (Figures 36 and 24) cells
as well as an occasional cell from younger cultures were transparent
enough that the fibrils were visible. The apparent crossing of
fibrils at the edge of the cell (Figure 36) demonstrated that they
were spirally arranged within the cytoplasmic cylinder. The ends
of the fibrils were never clearly seen; thus their exact location
was unclear. Specialized structures were apparently not present on

the fibril ends.

DNA and Cytoplasm

The cytoplasm in thin sections of T. refringens was evenly
distributed and often included irregularly shaped dense particles
approximately 14.5 nm wide which were presumably ribosomes (Figure
35). The area next to the band of fibrils was fibrillar and less
dense--having the appearance of the typical nuclear area in Ryter-
Kellenberger fixed cells (Figure 9; see also Figures 15 and 21).
'The nuclear area most probably adjoined the fibrils along the
entire length of the cell (Figure 37; see also Figures 29 and 32).
(The DNA strands in spheroplast-like forms and other cells with
«disrupted cytoplasm appeared to be directly attached to fibrils
(Figures 38, 39 and 22).

Fluorescence microscopy was employed as another means of study-
ing the location of DNA and RNA in the cell. Smears of treponemes

were made on slides, fixed, and stained with acridine orange or

57

‘

 

 

 

Figure 36. PTA-stained cell disrupted with guinea pig
serum and lysozyme. The cytoplasm is unevenly distributed and
fibrils (F) are visible within the cell. Note apparent crossing
of fibrils at edge of cell (arrows). Bar = 0.5 pm.

58

 

 

Figure 37. Thin sectioned cell with fibrils (F) and nuclear
area (N) visible. The longitudinally sectioned fibril (arrow)
adjacent to the cytoplasmic membrane resembles a trilaminar
membrane. Bar = 0.5 pm.

 

59

‘ II. a 93‘.» '

 

L‘-

'd.’

 

Figure 38. Thin sectioned spheroplast-like cell. Note
strands of DNA (D) attached to fibrils (F). Bar = 0.5 um.

 

Figure 39. Thin sectioned spheroplast-like cell. Note
strands of DNA (D) attached to fibrils (F). Bar = 0.5 um.

6O

euchrysine. The alternate method was to stain unfixed cells in a
wet mount. Similar results were obtained with either method except
that the fluorescence of fixed cells was less intense and faded
much more rapidly. Vitally stained cells were used for photography
and observation except when it was necessary to fix the cells for
experiments with nucleases. Euchrysine was usually used instead of
acridine orange because euchrysine-stained cells fluoresced more
intensely, if less red, than acridine orange—stained cells, while
the mechanism of the action of the two stains is the same.

Cells stained with euchrysine fluoresced green, various shades
of orange (Figure 40), or orange alternating along the cell length
with green (Figure 41). Cells treated with RNase and stained with
euchrysine were green rather than yellow or orange, indicating
that when present RNA stained orange as expected. Cultures in the
exponential growth phase contained a higher percentage of orange
cells or cells with orange areas and a lower percentage of all-green
cells than older populations, presumably because of the increased
metabolic activity in younger cells. Dividing cells in which one
daughter cell contained more orange than the other were not
uncommon (Figure 42). This is difficult to explain since it is
unusual for a cell to divide into two cells, only one of which is
actively metabolising.

Cells treated with DNase subsequent to RNase did not lose the
green coloring as should have happened if the green fluorescence
was due to DNA. Both before and after DNase treatment, the outer
envelope extending between divided cells, as well as the membrane

and cytoplasm of spheroplast-like cells, stained green (Figure 43).

61

 

Figure 40. Two wet-mounted cells stained with euchrysine. One
cell is orange, the other yellow-green. Bar = 10;im in Figures 40
through 43.

Figure 41. Wet-mounted cells stained with euchrysine. Cells
are orange and green alternately along their lengths.

Figure 42. Two recently divided cells stained with euchrysine.
Cell on left has alternating orange and green areas whereas cell on
right is mostly orange. They are connected with green outer
envelope (OE).

Figure 43. Wet-mounted spheroplast-like cell stained with
euchrysine.

62

Since the green fluorescence did not prove to be specific for DNA,
acridine orange- or euchrysine-staining was not appropriate for

determining whether DNA extended the entire length of the cell.

Induction of Spherical Forms
As a first step in cell lysis or for visualization of DNA by
the Kleinschmidt technique it was desirable to produce a population

of osmotically sensitive spheroplast-like forms.

Treatment with Penicillin

Penicillin was added in various amounts to 3- or 4-day cultures
grown in the regular medium, with or without 10% (w/v) sucrose,
25‘ (w/v) sucrose or 1 M ammonium acetate. When 0.1 U penicillin
per ml was used, the culture turbidity increased at the normal rate,
but many of the cells were extremely long (up to 10 times the mean
length) and without visible division points. Little or no growth
was evident in penicillin concentrations of l U per ml or 10 U per
ml. The population included a high percentage of convoluted forms,

with some emaciated and normal cells, but few spheroplasts.

Treatment with Lysozjgne

Attempts to induce spheroplasts with lysozyme resulted instead
in convoluted forms. The result was not specific to lysozyme treat-
ments, as seen below. A culture~was centrifuged and the pellet
resuspended in buffer or water, with or without an osmotic stabilizer,
EDTA or lysozyme. Phase microscopy was used to examine the cells
after one to two hours as well as after 24 hours. Convoluted forms

developed more readily in cultures that were beyond the exponential

63
phase of growth, even when cultures were only centrifuged and resus-
pended in buffer or water. The results are summarized in Table 4.
Lysozyme-EDTA in Tris and sucrose appeared to induce all cells

of any age culture to convolute. Lesser but quite severe effects
were seen in any other lysozyme solution (Figure 44) or whenever
Tris was present, with or without lysozyme. These effects were
more pronounced after one day than after one hour. The number of
spheroplast forms did not increase under any of these conditions;

nor did severe cell disruption or lysis occur.

Cell Lysis

 

The attempt to isolate cytoplasmic fibrils from treponemes
required that they be lysed. Guinea pig serum was tried as a lytic
agent because Nevin and Guest (77) had reported that the complement
and lysozyme present in guinea pig serum had a lytic effect on

several treponemal strains.

Treatment with Guinea Pig Serum

The addition of guinea pig serum (GPS), with or without added
lysozyme, to cell suspensions produced mangled or emaciated cell
forms. Blebs were often evident at points along the cell (Figure
45). Electron microscopy confirmed that this treatment disrupted
the cells more than any other lysozyme solution although most cells
were still enclosed in outer envelope and not fragmented into smaller
'pieces. In the least disrupted cells, the cytoplasm appeared
unevenly distributed in a striped pattern and the cells were more
transparent so that fibrils became visible within them (Figure 36).

In some cases, the outer envelope was affected, with striated tubules

.oeusHo>coe eue3 mHHeo eunumu> Ham ++++ .oeusao>

Icoo eue3 mHHeo ecu mo mom menu euoe +++ .oeusflo>coo eyes mHHeo emu mo «om aaeuesaxoummu ++ .oeusao>coe euez
mHHeu emu mo wow menu mmeH + .emcmno o: I "mzoaaom me oeouooeu mm3 ecowmcemwsm Hmuceefiuemxe cw mHHeo oeusao>
Icoe uo HenEsc ens .zmoomouowa emmcm an meusuaso Heeuoc muuz oeummfiou eue3 msoaesemmsm Hmucefiuuemxmm

 

 

 

 

 

 

+++ ++ +++ ++ H.m mm .muue z Ho.o as mmouosm A>\3o wmm .meom z Ho.o
+++ + +++ + H.m mm .muua z Ho.o cu emouosm amm
I u m.> mm .uemmsn euenmmosm z noo.o cu emouusm wmm
+++ +++ + H.m mm .muue z Ho.o
: u m.» an .ueuesn eumammonm z seo.o
+ 0.» mm .ueeeoa mumnmmoem 2 500.0
+ u ueumz
chuusHOm Houucoe
++++ ++ H.m mm .muua z Ho.o cu emouosm A>\3V woe .490m 2 Ho.o
++++ ++ H.m mm .muue z Ho.o as emouosm x>\3s «mm .aeom z Ho.o
+++ H.m mm .muue s Ho.o as .Hocuusn «m
+++ + H.m mm .muue z Ho.o
4 m.h mm .uemusn
6 +++ ++ manganese z seo.o as .mmouosm fi>\3v emu .Hocmusn am
+++ ++ m.s mm .uouesn oumemmonm z seo.o as .Hocousn an
++ o.m mm .ueues cu emouose .>\&V «mm .meom z H.o
++ + . o.m mm .ueumz as eumueom saucosem z m .asom s H.o
”ecouusHOm msuzoaaom cu eEMNoe w HS\m: ooa
.mmmlmm a: mflm 8:
Am>eo HNINHV unease sumo meusuaso ecowusHOm scumcemwsmem
meusuasu neoao cu mHHeo emcee Huuusecomxe cw mHHeu
so ecuusaom mo uoeuum co souusaom mo uoemmm

 

.eun on ueuue one u: H ueuue mason oeusao>eoo mo Memes:

emu ecflsueueo ou news we; hmoumouewfi enema .meom no Hossusn .uenwauneum ouuoaeo as usosuus no

:uu3 .Heuusn no :Ouusaom eahuommH m as oeocemwseeu ueHHem emu use oemsuwuuceo eu3 eusuaso e59
.mueumsn one accuusHOm es>N0m>H cu mesesomeuu no soumcemmsm an meson oeusao>soo mo esouuosocH .v edges

65

/c

Figure 44. Phase micrograph of 5 day culture sus-
pended in 0.067 M phosphate buffer, pH 7.8, with 2%
n-butanol and 100 g/ml lysozyme. After one day, most
of the cells had convoluted (C). Compare with normal
culture in Figure 3. Bar = 10 um.

 

Figure 45. Phase micrograph of a 5 day culture sus—
pended in 0.067 M phosphate buffer, pH 7.8, and guinea
pig serum. After one day, many of the cells had blebs (B).
Bar = 10 um.

66

or vesicles of OE extending from the cell (Figures 46 and 47).
Masses of outer envelope and striated tubules were also present
apart from the cells (Figure 12). The blebs visible with the phase
microscope were seen also with the electron microscope (Figure 48).

Some cells were more severely disrupted than those just
described, as they were no longer cylindrical (Figure 49), or had
holes in the cytoplasmic membrane and outer envelope through which
fibrils or axial filaments extruded (Figures 50 and 24). A few cells
were so disrupted that nothing was visible but a band of fibrils
and membrane vesicles, some of which had the substructure of outer
envelope (Figure 51). The fibril bands often retained a spiral
configuration (Figure 52).

Cells treated with guinea pig serum and lysozyme were embedded
for thin sectioning. Many spheroplast-like forms were seen and the
cytoplasm of most cells was diffuse. Scattered among the cells were
a large number of membrane vesicles, many of which were discontinuous
at one or more points. Many membrane vesicles were not trilaminar
(Figure 53). It was impossible to determine from the pictures if
this isolated one-layered membrane was originally outer envelope or
cytoplasmic membrane which lost its three-layered appearance or if
it was the peptidoglycan layer.

Similar results were observed in cells resuspended in either
reconstituted lyophilized GPS or in fresh-frozen GPS and also in
amounts of GPS varying from 0.03 ml to 1.0 ml per ml of suspension.
This covered a range of 0.01 to 1.7 ml GPS per ml of original culture.
The best disruption was seen at pH 8.0, whereas at pH 6.5 most cells

were unaffected. The amount of lysozyme added (from 0.02 to 1.5 mg

67

 

 

Figure 46. Cells disrupted with guinea pig serum and nega—
tively stained with uranyl acetate. Striated tubules extend from
cells. Bar = 0.5 um.

68

 

Figure 47. Cell disrupted with guinea pig serum and negatively
stained with ammonium molybdate. Note fibrils visible within cell
(F) and vesicles of outer envelope. Bar = 0.5 um.

69

g ‘3}.

 

Figure 48. Cells disrupted with guinea pig serum and lyso-
zyme and negatively stained with PTA. Several blebs (B) have
formed on one cell. Axial filaments in a tube of outer envelope
(AF) are visible in another cell. Bar = 0.5 um.

70

 

Figure 49. Cell disrupted with guinea pig serum and lyso-
zyme and negatively stained with 0.5% PTA pH 6.7. Outer envelope
(OE) and fibrils (F) are visible. Bar = 0.5 pm.

71

 

Figure 50. Cell disrupted with guinea pig serum and lyso-
zyme and negatively stained with PTA. Fibrils (F) emerge from the
end of the cell through holes in the cytoplasmic membrane (CM) and
outer envelope (OE). Bar = 0.5 um.

Figure 51. Cell disrupted with guinea pig serum and lysozyme
and negatively stained with PTA. Membrane (M) and fibrils (F)
remain. Bar = 0.5 um.

 

73

 

Figure 52. PTA-stained spiral band of fibrils from cell
disrupted with guinea pig serum and lysozyme. Bar = 0.5 um.

74

' V" J

c

“ M's-ef- :91 30

l .
, -e‘;zq':ms'd's- . 1

 

Figure 53. Thin sections of disrupted cells and membrane
vesicles from culture treated with guinea pig sermn and lysozyme.
Membrane vesicles are discontinuous at points (arrow) and some
have lost their trilaminar appearance (M). Bar = 0.5 um.

75
per ml of suspension) did not appear to affect the amount of dis-
ruption. In fact, it was not necessary to add any lysozyme in
addition to that naturally occurring in GPS to get a high degree of
cell disruption.
Treatment with Guinea Pig Serum,
Detergents and Sonication

Guinea pig serum treatment resulted in disrupted cells with the
concomitant release of some fibrils; however, most fibril bands
remained in the cell. Most cells were still membrane bound and
incompletely lysed as was also evident from the still turbid suspen-
sion. In an attempt to completely lyse the cells, either Brij 58
(Figure 54), sodium deoxycholate, sodium dodecyl sulfate, or Triton
X-100 was added to GPS-treated cells. These detergents did not
result in greater disruption or more efficient fibril release and
left suspensions still turbid. Sonication before or after exposure
to GPS resulted in shorter pieces of typical GPS-disrupted cells
(Figure 55) and a still turbid suspension.

A combination of treatment with GPS (containing complement and
lysozyme), Brij 58 and sonication, hereafter called the CLBS pro-
cedure, did effect complete lysis (Table 5). A cell suspension of
1.6 ml required 15 s sonication while a volume of 2.5 to 6.0 m1
required 20 to 35 s. Vortex mixing in place of sonication did not
clear the suspension and sodium deoxycholate could not replace
Brij t8. CLBS-treated cells viewed under the electron microscope
revealed free fibril bands of varying lengths, clumps of individual

fibrils, and axial filaments.

76

 

Figure 54. Cells disrupted with guinea pig serum and Brij 58
and negatively stained with PTA. Fibrils (F) are partially
released from cell and outer envelope substructure (S) is visible.
Bar = 0.5 um.

77

 

Figure 55. Ammonium molybdate stain of cell sonicated after
exposure to guinea pig serum and lysozyme. Some fibrils have been
released from cell, but others are still enclosed (F). Bar =
0.5 pm.

 

Figure 56. Silicon tungstate—stained material from Renografin
gradient band A. Bar = 0.5 um.

78

Table 5. Disruption of cells after treatment with guinea pig serum
(GPS) plus detergent or sonication or both

 

 

Condition of Clearing of
majority of turbidity of
Treatment cellsa cell suspensionb
GPS with or without added lysozyme + -
GPS with detergent (Brij 58, sodium + -
deoxycholate, sodium dodecyl
sulfate, or Triton X-100)
GPS with sonication ++ -
GPS with sodium deoxycholate and ++ -
sonication
GPS with Brij 58 and sonication +++ +
(01.88)

 

aCells were observed by electron microscopy and rated accord-
ing to the following criteria:

- No disruption
+ Mildly disrupted cells still surrounded
with outer envelope
++ Short pieces of mildly disrupted cells
+++ Cells completely lysed; fibrils freed
bCell suspension was visually observed for loss of turbidity.

- Suspension remained turbid
+ Suspension became clear

The CLBS lysis procedure was determined to be the best treatment
for further use in the isolation of the fibrils. Routine preparation
involved treatment of 440 ml of culture. Under identical conditions,
*when three times this volume (1320 ml) was treated, lysis was not
effected. Sonication of small portions of this GPS-Brij 58-treated

suspension failed to clear the suspension.

7‘)

Fibril Isolation

 

Sucrose Gradients

 

Treponeme preparations treated with GPS and lysozyme were
placed onto a gradient of from 5 to 40% (w/v) sucrose and centri-
fuged for 10 min at 48,000 x g in an attempt to isolate fibrils.
Fractions collected by puncturing the bottom of the tube were
dialyzed and negatively stained for electron microscopic observa-
tion. Although most membrane was found near the top of the gradient,
membrane fragments or vesicles as well as fibrils were scattered
throughout the upper half of the gradient. The lower half contained
axial filament and membrane fragments, with whole treponemes and
large protoplasmic cylinder fragments near the bottom. When similar
cell preparations, with or without DNase treatment, were centrifuged
under the same conditions on a 5 to 25% (w/v) sucrose gradient,
whole cells and large fragments pelleted while the distribution of
other cell components resembled that described above. Visible bands
were sometimes formed in these gradients, dependent upon the degree
of cell disruption and the mass of cells put onto the gradient.
These bands were not homogeneous accumulations of any particular
cell component.

Cells treated with GPS, lysozyme and DNase were further dis—
rupted by detergent treatment (sodium deoxycholate or Triton X-lOO),
sonication or both. Distribution of the cell components on a 5 to
25% sucrose gradient was similar to that found in earlier gradients
with one difference: no pellet of whole cells formed when cells

were disrupted with GPS, lysozyme and sonication, with or without

80

detergent. Treatment with GPS, lysozyme and detergent resulted in
a small pellet.

When the CLBS lysis procedure was used to disrupt the cells,
on a S to 25% gradient a pellet of fibrils and membrane was formed
with no whole cells. When centrifuged on a 5 to 67% sucrose
gradient from 15 min to 2 hr at 106,000 x g, the cell lysate was
not distributed into distinct hands. This gradient was collected
on a fraction collector while the absorption at 280 nm was simul-
taneously read. The top 1/8 of the gradient was dominated by con—
taminating proteins including DNase, lysozyme, and serum components.
One distinct protein absorbance peak was detected near the bottom
of the gradient in the portion containing large cell fragments and

clumps of fibrils.

Renografin Gradients

When cells were disrupted with CLBS and DNase and centrifuged
on Renografin density gradients, the cell components were more
clearly resolved than on comparable sucrose gradients. A low speed
centrifugation step was not used on the lysate prior to gradient
[centrifugation because separation of fibrils from unlysed cells did
not occur. Fibril clumps pelleted with whole cells at low speeds
while many individual fibrils remained in suspension.

CLBS-treated cells centrifuged on gradients of 0-76% Renografin
at 130,000 x 9 formed distinct bands. Since the position of the
bands changed with further centrifugation after 1 1/2 hr, but not
after 2 hr, it was determined that equilibrium was reached at 2 hr.

{fine average position and size of the two main bands (A and C) can be

81

seen in Figure 57. Band C was flocculent, while band A was more
homogeneous. Smaller bands, often with diffuse edges, were seen
between band A and band C as well as directly below band C. Their
presence, size, and location in relation to the major bands varied
with minor changes in the procedure (e.g., length of sonication,
volume of disrupted cells placed on the gradient).

Band A consisted largely of membrane vesicles (Figure 56;
see also Figure 25), much of which had the substructure typical of
outer envelope. Occasional fibrils, axial filaments and striated
tubules were present. Light bands and diffuse material between
band A and band C did not contain major amounts of any particular
cell component other than membrane. Band C was composed of fibrils
with varying numbers of small membrane vesicles as well as occasional
axial filament hooks and basal bodies (Figure 58). Although a few
fibrils were seen in other parts of the gradient, most were concen-
trated in band C. There was considerable variation of fibril length
with most fibrils shorter than those found in situ.

One or two narrow, fairly distinct bands were usually seen
immediately below band C. These could be distinguished from band
C only if the gradient covered a sufficiently narrow density range.
The content of these bands was similar to that of band C, with more
membrane and a higher percentage of long fibrils (Figure 59). A
gradient of 0-57% Renografin allowed adequate separation of bands,
while still retaining the whole spectrum of bands.

The location of bands A and C in 34 0—76% Renografin linear
gradients from 10 experiments employing almost identical disruption

procedures (CLBS and DNase) were recorded. Since the mean density

82

 

 

 

 

 

 

 

 

 

wnu onus:
. A 1.14 g4.“
) c _ 1.22 gm
I
20. mm 15.0mm
wnuour 0111151:

A 114 SA»:
1: — 1.19 g/ml

20.] mm 16.8mm

 

 

 

Figure 57. Average position of bands A and C when cell lysate,
with or without DNase, was centrifuged to equilibrium on 0 to 76%
Renografin gradients. Average height of lower edge of each band and
average buoyant density is recorded.

83

 

 

Figure 58. Ammonium molybdate-stained material from Reno-
grafin band C. Membrane (M) . fibrils (F), and an axial filament
hook with basal body (BB) are visible. Bar = 0.5 um.

84

 

Figure 59. PTA-stained material from Renografin band D.
Membrane and long fibrils are present. Bar = 0.5 um.

85
of 76‘ Renografin had been measured (1.435 :_.009 g/ml), the buoyant
density of a particular band could be calculated from its position
in the linear gradient. By this method the average buoyant density
of the lower edge of band A was found to be 1.14 :_0.02, while that
of band C was 1.22 :_0.01 g/ml (Figure 57).

The use of DNase in the disruption procedure directly affected
the position of band C, but not that of band A. The buoyant density
of band C on gradients when DNase had not been used was 1.193 :_ V
0.003 g/ml, as compared to 1.22 :_0.01, the average buoyant density
of band C with DNase (Figure 57).

Even after dialysis, band C was very flocculent, especially
when DNase had not been used in the disruption procedure. The
fibrils and membrane settled to the bottom of a test tube from
suspension in only a few minutes. The resuspended fibrils, when
viewed with the electron microscope, were in large clumps (Figure
60). In order to determine a method of disrupting the clumps, the
fibrils were treated by various chemical and physical procedures
before electron microscopy. Aggregation was still apparent after
the treatments listed in Table 6.

For further purification of the fibrils, band C was dialyzed
to remove the Renografin and treated with DNase. Triton x-100 was
also added because it solubilizes both the outer and cytoplasmic
membranes of Escherichia coli (31) . After the suspension was centri-
fuged a second.time on a Renografin gradient (Figure 61), the band
C which formed contained fibrils with less membrane contamination
than band C on the previous gradient. When band C was treated again

*with.DNase and Triton x-100, and centrifuged on a Renografin gradient

 

Figure 60. PTA stain of large clump of fibrils from
Renografin band C. Bar = 0.5 pm.

87

Cells (treated with CLBS and DNase)

Renografin gradient (0 to 76% or 0 to 57%)
130,000 x g, 2 hr

 

band C other bands discarded

dialyzed overnight
treated with DNase and Triton X-100

Renografin gradient
130,?00 x g, 2 hr

 

r

band C other bands discarded
dialyzed overnight
treated with DNase and Triton X-100

Renografin gradient
130,000 x g, 2 hr

 

band C 1
other bands discarded

dialyzed overnight

Figure 61. Flow chart of procedure for isolation of fibrils
from treponemes. Cells lysed with guinea pig serum, Brij 58 and
sonication were centrifuged on a Renografin gradient. Band C was
collected, dialyzed and treated with DNase and Triton x-1oo and
centrifuged on another Renografin gradient. Band C was collected
and the process was repeated. After the third centrifugation band
C was again dialyzed.

88

Table 6. Treatments which did not disaggregate clumps of fibrils, as
determined by electron microscopy

 

Sonication

2% Triton X-100

0.01 N HCl (pH 2)

0.025 M ammonium acetate buffer at pH 5.0
5 to 1 M ammonium acetate at pH 7.5
S to l M ammonium acetate at pH 8.6
5 to l M ammonium acetate at pH 9.5
.1 to 2 M NaCl
1 to 2 M NaCl and sonication
M CaC12

50% chloroform-methanol

33% ethanol

33% ethanol and sonication

2 M urea

 

a third time, the resulting band C was a fairly purified preparation
of fibrils with few or no visible contaminants (Figure 62). This

type of preparation was used for chemical studies on fibrils.

Chemical Characterization of Fibrils

 

Several fibril preparations which had been purified three times
on Renografin gradients were put onto SDS polyacrylamide electrophoresis
gels. One major band of molecular weight 97,000 (96,960 :_1180) and
several minor bands were resolved after staining the gels with
Coomassie blue (Figure 63a). On less purified preparations, the
97,000 molecular weight peak was proportionately smaller in comparison
to the other peaks (Figure 63b). Renografin-prepared band A, con-
taining mostly membrane with occasional fibrils, had only a small peak

at the molecular weight of about 97,000 (Figure 63c).

89

 

Figure 62.
Renografin gradient.

Bar

 

Uranyl oxalate-stained band C from third

0.5 um.

9O

 

98,000

91,000

\.

 

 

 

 

1_ 1 1 A 1 1 L

130“ 8,000

fl

 

ABSORBANCE (540 nm)

46,000

when”:

01 0‘72 0‘? 0:4 05 0:6 0:7 011 if 03 0.4 0.5 0.0 0.7“
maumvz mommy

 

 

 

 

 

 

 

 

 

Figure 63. Absorbance readings of SDS polyacrylamide electro-
pfluaresis gels stained with Coomassie blue. Peaks are labeled with
Imalecular weights obtained by comparison with standards.

a. Fibril preparation purified on three Reongrafin gradients.
b. Fibril preparation purified on two Renografin gradients.
c. Band A from a Renografin gradient.

d. Molecular weight standards:

B-galactosidase (130,000)
bovine serum albumin (68,000)
ovalbumin (46,000)

pepsin (35,000)

91

The yield was estimated by Lowry's protein analysis of dialyzed
fibril preparations. Approximately 125 to 320 ug protein were found
in band C after one Renografin centrifugation of 120 ml of CLBS-
treated culture. After two more Renografin centrifugations there
were only about 30 ug protein. Using Jackson's calculation of the
dry weight of these treponemes (50), it was calculated that the
protein in the purest fibril preparation represented about 0.2% of
the total dry weight of the cells.

Fibrils purified by three centrifugations on Renografin gradients
were subjected to further chemical analyses. The preparation was
first dialyzed overnight with several changes of water. Analysis
by Lawry's method indicated about 50 ug/ml protein in this particu-
lar preparation. Hexose was not detected by the anthrone method,
but it must be noted that the smallest amount detectable in this
test is 25 ug/ml. The diphenylamine test revealed approximately
10 ug DNA/m1, while the orcinol test indicated pentose at a concen-
tration of 34 ug/ml. The determination of the ratio of absorbance
at 280 nm to the absorbance at 260 nm gave inconclusive results
because Renografin absorbs light very strongly at both wavelengths

and is difficult to dialyze away completely.

DNA-Fibril Connection

 

Further evidence for the apparent relationship between DNA and
fibrils noted in thin sectioned cells was pursued by the aqueous
spreading technique for electron microscopy of disrupted cells and
by gradient centrifugation of cells labeled with 3H-thymidine before

disruption.

92

Aqueous Spreading Technique for Electron Microscopy

 

The Kleinschmidt spreading technique for electron microscopy
was used on various cell preparations to visualize a direct connec-
tion between fibrils and DNA. DNA strands on the electron micro-
graphs were identified by comparison with control preparations of
T4 DNA (Figure 64) and by the absence of these strands in cell
preparations treated with DNase.

Strands of material (probably DNA) were seen emerging from
the ends of cells disrupted with GPS or the CLBS procedure (Figure
65). In such cases the cytoplasmic fibrils were still inside the
cell, making it impossible to note any connection between fibrils
and DNA. The DNA was spread out from bands of fibrils (or from the
cellular material surrounding fibrils) of more disrupted cells

(Figure 66). The DNA strands often extended directly from the

broken end of a fibril band (Figure 67).

3H-Thymidine Labeling

An alternative method of determining the attachment of DNA to
fibrils is to demonstrate that DNA segregates in gradients with
fibrils rather than with membrane. Detection of DNA would then be
possible by growing cells in a medium containing a radioactive DNA
precursor. Incorporation of 3H-thymidine by T. refringens was
demonstrated by adding 1, 5, 10 or 50 uCi of 3H-thymidine to 12 ml
of a 2-day-old culture. The cells contained the maximum radioactivity
after 3 more days of growth in 50 uCi 3H-thymidine and thereafter

lost radioactivity until the 6-day level was equal to that at 6 hr.

93

 

Q I '
v n
0, ‘0 ‘ '
, 0 . '._ . -
= . 'a‘ . f : . g ‘
~ { ‘ ’ ‘ o a ‘ ‘1
i k ‘ .' ‘ ‘ . £‘ N .‘_o n .
-. ‘ . ' A .‘I 2‘”
.1. 3 . " . w -. p, ‘ ‘ ‘
e ' _ ' . . . ..
. . ". . 1' '- '. O ' - ‘ _ ‘0’," e ' t . ~
0 . ’. ' . a. 0.. .
. 0 f X. "I. ‘ ‘3 ‘
’ .1 . .- . ~ :‘. _’3‘ i' ' 9
01 o . ,.' '{KT ‘~\ 2 . . \J’:*,'" “<- ,.,‘ o J.” ) .L
‘ J .I o 0“ .a ’ 0 (I a. . ‘ 1‘ .. t _ l— . (‘1
. ~, ' . .. . ..
' fi o ,fi'k' .( .v"""v- . ‘9, w } .9
, s x l “s -. .. ,
C " 1 $ ' I 1"? " ‘3‘ 0 y ‘1‘ a. ' . i
. \ 1 ‘.‘ '5 z p... ‘ s " 8‘ _.
0 #5 {P ,I 1‘ ‘I \ '\~’
- ’ 4‘s 94$va (" i 6 2 HA \‘x‘ \I-‘.. "
‘ I .., ‘ pi: _‘ "‘ ‘ ‘ (\J.
' L , | r s D $ '0‘ 4 i. -..u"
’ ' . f\. ' ‘ fl“) ’ i “Y- , .' ‘
’ M‘ Hf; _ / ‘ 'f -.‘L
l ’ Ag ~ 0 t a ‘ “’\; ( ’1’ I \J( ,. ‘s
. I. . - ‘— ‘ , ‘ 1‘... . .X" :5.) . ‘;Ls.)
. -‘ ' ' ' .‘i' o ‘—‘ I." r N‘v "L -‘ '
. . .. . .’ ' 1. . 9 .7 ‘VAzr‘ ' ’._> ’. ‘ ‘ . ‘ ' . ’3‘ .
g . ' - . C. ‘u‘. o- -4’.-...‘-‘--.~'~ _.l ‘- . ..'I ‘ f A ...-
DNA from T4 bacteriophage. Isolated DNA was
- 1 pm.

Figure 64.

prepared for electron microscopy by the Kleinschmidt spreading
technique and stained with uranyl acetate. Bar —

 

, fl .
Cell disrupted with guinea pig serum, prepared

for electron microscopy by the Kleinschmidt spreading technique
Strands of DNA (D)

Figure 65.
and stained with PTA and uranyl acetate.
emerge from the cell. Bar = 1 fim.

lilies-535i
.a. .

94

 

Figure 66. Cells disrupted with CLBS, prepared for
electron microscopy by the Kleinschmidt spreading technique,
stained with uranyl acetate and rotary shadowed with platinum.
DNA (D) emerges from near fibrils (F). Bar = 1 um.

   

Figure 67. Cell disrupted with CLBS, prepared for electron
microscopy by the Kleinschmidt technique and stained with
uranyl acetate. DNA (D) emerges from near fibrils (F). Bar
= 1 pm.

95

Labeled cells were disrupted by CLBS (with varying sonication
times), either with or without DNase, and centrifuged on Renografin
gradients. The results were inconclusive since radioactivity peaks
were found in the fractions containing band A and band C as well as

in other areas of the gradient.

Ultrastructure of Fibrils

 

Thin Sectioned and Shadowed Fibrils

 

The fibrils in thin sectioned cells were studied to gain addi-
tional information about their shape, substructure and location in
the cell. It was apparent in any slightly oblique or cross sectioned
cell that the fibrils were directly adjacent to the inner boundary
of the cytoplasmic membrane, but for information on the shape and
substructure of the fibrils it was necessary that the section be
cut perpendicular to the fibrils. These sections were not common
and could be identified by the trilaminar appearance of the CM
adjoining the fibrils (Figure 68). The substructure and shape of
cross sectioned fibrils varied, but there were certain features which
all or most sections had in common. The width of the fibril section
changed from its base (next to the CM) to its tip (the part extend-
ing furthest into the nuclear area of the cell), forming a shape
resembling a pentagon (Figures 68 and 69a). Almost all fibrils had
a cap, or narrower area, at their tip. The widest part of the
fibrils was usually just below the cap and the base was sometimes,
but not always, narrower than this. The base of the fibril averaged
10.8 nm and the base-to-base interval between fibrils about 3.5 nm.

The height of the fibril section from the CM to the tip averaged

96

 

Figure 68. Direct cross sections of fibrils in normal cells
(a,b,c), a convoluted cell (d) and disrupted cells (e,f). The sub-
structure of the fibrils is visible, as is the nuclear area into
which they project. The trilaminar appearance of the adjoining
cytoplasmic membrane indicated that the fibrils were cut approxi-

mately perpendicular to their length. Bar = 0.1 um.

 

 

97

 

 

 

    
   
 

 

 

  
 

a
b
CAP
wmssr POINT
110m smr
BASE
c
DENSE AREA moan THAN INNER ‘
LAYER or M ALONE) DENSE MATERIAL
ms: m0
110m BAND
0 m

  

Figure 69. Drawings of sectioned fibrils drawn approximately to
scale and positioned on the cytoplasmic membrane.

a. Generalized shape of sectioned fibril. The terms used in the
text and Table 7 to describe the parts of the sections are defined:

1. width of base
2. width of widest point
3. height
4. height, excluding cap
5. tip

hatched area; cap

b. Substructure of various cross sectioned fibrils.

c. Cross sectioned fibril which includes the elements common to
most sectioned fibrils in which substructure is visible.

d. Substructure of various longitudinally sectioned fibrils.

98
14.5 nm but was difficult to determine accurately since DNA or other
cellular material often obscured the tip and there was no clear
distinction between the base of the fibril and the inner dark layer
of the CM. Measurements of various dimensions of sectioned fibrils
as well as fibrils prepared by other electron microscopic techniques

are summarized in Tables 7 and 8.

Table 7. Means of various fibril dimensions measured from fibrils
in thin sectioned cells (see Figure 69 for clarification

 

 

 

 

 

 

of terms)
Mean measure- Number of fibrils
Dimension measured ment (nm) measured
Cross sections of fibrils
Width at base of fibril 10.8 :_0.6 37
Width of fibril at widest point 14.0 i_l.5 16
(if wider than base)
Height of fibril from cytoplasmic 14.5 :_0.9 48
membrane (CM)
Height of fibril from CM, 9.6 :_0.4 15
excluding cap
Fibril-to-fibril interval 3.5 i 0.5 29
between bases
Longitudinal sections of fibrils
Height of fibril including 7.3 :_0.2 28
dark band
Height of fibril including dark 16.2 :_l.7 21
band and dense material
outside band
Cross sections of loose fibrils
inside disrupted cells
Widest dimension 12.6 :_ .8 6
Dimension perpendicular to 12.4 i 4.0 6

widest dimension

 

99

Table 8. Mean measurements of various dimensions of negatively
stained fibrils. Fibrils were stained with phosphotungstate
(PTA) or uranyl oxalate (UOx) (see Figure 76 for clarifi-
cation of terms)

 

 

 

PTA-stained fibrils UOx-stained fibrils

Mean Number of Mean Number of

measure- fibrils measure- fibrils
Dimension measured ment (nm) measured ment (nm) measured

 

Negative stains of isolated fibrils

Width of wide fibril 12.7 :_ 138 11.9 :_0.3 239

Width of narrow 8.4 i 0.3 121 7.7 :_0.3 97
fibril

Width of strand of 3.4 :_0.3 41 2 2 i 0.2 39
wide fibril

Width of strand of 3.1 :_0.7 37 3.8 :_0.8 30
narrow fibril

Width of ribbon 5.4 :_0.4 30 4.3 :_0.5 31
alongside wide
fibril

Width of fibril in 8.4 :_0.4 35
band of fibrils

Interval between 4.3 :_0.6 30
fibrils in band

Periodicity of stria- 3.0 + 0.3 12

tions on wide fibrils

Fibrils within negatively_stained cells

Width of fibrils 7.6 i 0.6 34
Interval between 3.5 t 0.9 10
fibrils

 

The electron density varied with some of the sectioned fibrils.
The pattern of this substructure varied but some fairly common elements
included: a base made up of three (or two) dark areas and an electron

light strip just below the widest point (Figure 69b, 69c). Fibrils

100
were sometimes free from the CM in disrupted cells (Figure 70).
Dense areas were visible, but the fibrils' dimensions varied (Table
7) and they offered no clarification of fibrillar structure.

Longitudinal sections of fibrils were rare and even more diffi-
cult to measure accurately than cross sections. Adjacent to the
inner dark layer of the CM, which was perhaps thicker than usual
and included part of the CM, was sometimes seen a light band and
then a dense band, resembling another trilaminar membrane (Figures
69d and 37). Often outside the dense band more dark material was
found (Figures 37 and 35). Cross striations were seen only
occasionally in longitudinally sectioned fibrils (Figure 71), but
their periodicity (about 3 nm) was similar to the periodicity of
negatively stained fibrils reported later.

Isolated fibrils were shadowed with platinum-carbon. The
resolution attained did not allow a distinction to be made between
different fibril orientations and did not reveal any substructure
(Figure 72). The average height of the shadowed fibrils from the

grid was 6.6 :_0.4 nm.

Negatively_Stained Fibrils

 

Negative stains of fibrils revealed the complexity of their
substructure. The initial impression of PTA, silicon tungstate,
or ammonium molybdate-stained fibrils was that they were tubular
with stain penetrating into the lumen or lying in the center groove
created by a collapsed tube (Figure 73; see also Figure 58). More
careful inspection of negative stains revealed that some of the two-

stranded fibrils were wider than the others (Figures 73 and 74).

101

 

 

Figure 70. Sectioned disrupted cell containing fibrils (F)
which have been freed from the cytoplasmic membrane (CM).
outer envelope (OE) still surrounds the cell.

 

Figure 71. Longitudinally sectioned fibrils (F).
periodicity is evident in one of the fibrils (FP).

  

7”.-

mg

as

.1

. ‘7‘! 4" .
. 2;?- . . .
‘ztyv 3:1 .1

*1: "‘. hkl'.“ , ".
a. ‘ W ‘4;me :
\ .

L

15.1‘.
“a ~§§fl;.
" "33 (z.
1. “RT; ‘1'.“

5

      
   

         
   

      
  

    

 

  

  
     

Fm. ‘ g
2“ .. ,4 3‘s“ A" “ “We. :11. ..‘a‘!
.' a. ski ‘1' .. 15:
‘ xvii).

 
  

    
 
   

I‘ 1‘ . ,1“. ,' Rk ‘ _ ‘ gait“ .
mi ‘1 +7. «$“h‘fi ' -. - . ‘81}. xziisifift.“ j
2’“ fiéh.) “‘q§\‘hfi\fitfifiuflq 3%“ flfi a: is" i 1‘ 3&3?“ £595“ 31:9le ‘31,}: EL?” ‘ f ,
, 3. . 2, '1Qg.§9§h:n:‘-\;:‘LL ‘ . \ _ - 2,1“. ‘kd: ‘ 'h.’ ”:14. «‘6‘; .9;‘|"
. 1. "v 551%“): a: - .» 7 , - «a: a . . wens.
. _ . .1

. l\.“.‘~¢) .2 ,
i-c-V.‘ L .-'§-T~ '
from»; -_ ~. 1 9

1.
5v»,
r,

       
 

_ "1:3”;

 
   

      
     
   
  
 
 

         

~ 5
l

  

 
  
  

 

.- 1. J .. »
.vl- I . . _.._ \

‘ V". . W" “. v‘ -, ‘1 \J \;‘ f

":1; What. .‘Y “V :AJ\Q&I\ 1“- .

‘

Figure 72. Isolated fibrils which were shadowed with
platinum-carbon in the direction of the arrow. Bar = 0.5 um.

103

 

Figure 73. PTA-stained narrow (N) and wide (W) two-stranded
fibrils. Bar = 0.5 pm.

 

Figure 74. PTA-stained wide two-stranded fibrils (W) from
preparation which had been kept frozen. Bar = 0.1 um.

 

Figure 75. Silicon tungstate—stained wide fibrils with
ribbon (R) of stained material lying on each side. One strand
of some of the fibrils is broken (arrows). Bar = 0.1 um.

105

Sometimes these wide fibrils were edged on both sides with a faint
ribbon of material whose width varied from fibril to fibril (Figures
75 and 76a). One strand of the fibril was often broken away from
its companion (Figure 75). This made the fibrils appear to be
composed of two parallel strands rather than a tube.

In an attempt to reveal finer details of substructure, fibrils
were stained with uranyl oxalate (UOx), a negative stain which
provides higher contrast and greater penetrability into intramoleculr
crevices than PTA (39). Preparations stained with UOx had the same
fibril types or appearances as with PTA:‘ narrow two-stranded forms,
and wide two-stranded forms, with and without ribbons (Figures 77 and
78). In addition, very definite cross striations could be seen on
the wide fibrils (Figures 77 and 76b). Small extensions with a
periodicity of about 3 nm projected inward from each strand of the
wide fibrils, meeting at the center of the fibril to create a ladder
effect. When part of one strand was broken away, the other strand
retained its periodic extensions (Figures 78, 79 and 76c), and when
fibrils split, the break appeared in the center, leaving the small
projections of each strand (Figures 78 and 76c). The cross stria-
tions usually appeared to be perpendicular to the strands, but in
some places were at a slight angle.

Sometimes when one strand of a wide fibril was broken away,
the other strand apparently twisted to reveal a narrow fibril (Figures
77, 80 and 76d). The inner strand of this narrow fibril was con-
tinuous with a strand from the wide fibril. The outer strand of the

narrow fibril appeared to come from underneath the wide fibril strand

106

 

 

 

   

 

a ==
L Irmnmummuummmmnmmmwum 1.111(1))!1'1)!(Jill’gfliflll'ilfilll’l'lI)'l'l'jlllfl‘lj'iiifllfl’la
m
0 WW“... 1 . 1 .

 

Figure 76. Drawings of fibril types and substructure seen in
preparations of isolated fibrils stained with PTA or UOx. Fibrils
are drawn approximately to scale. See Table 8 for measurements.

a. PTA-stained narrow fibril, wide fibril without ribbon and
wide fibril with ribbon.

b. UOx-stained narrow fibril, wide fibril without ribbon and
wide fibril with ribbon. Wide fibrils have cross striations.

c. UOx-stained wide fibril with one broken strand and wide
fibril with strands split apart. Each strand retains its extensions.

d. UOx-stained wide fibrils (without and with ribbons) when
one or both strands have twisted into what appears to be narrow
fibrils.

107

 

Figure 77. UOx-stained wide fibrils with periodic cross
striations (P). A strand of one wide fibril appears to twist
into narrow fibril (N) at point where other strand is broken

away (arrow). Bar — 0.1 pm.

 

 

Figure 78. UOx-stained wide fibrils with ribbons (R)
alongside. Each strand of wide fibrils retains its periodic
projections when fibril splits (long arrow) or when one strand
is broken away (short arrows). Bar = 0.1 um.

108

 

Figure 79. UOx—stained wide fibrils. Each 'strand of a
fibril retains its periodic projections when the other strand is
broken away (arrows). Bar = 0.1 um.

 

Figure 80. UOx-stained wide (W) and narrow (N) fibrils. One
strand of a wide fibril splits away from the other strand and
twists (arrow) into a narrow fibril. Bar = 0.1 um.

5

 

Figure 81. UOx-stained wide and narrow fibrils. Both
strands of one fibril twist (arrow), resulting in two narrow
fibrils (N). Bar = 0.1 um.

110
or to extend from the dark ribbon alongside the wide fibril (Figure
76d). In Figure 81, both strands of a wide fibril twisted, after
splitting apart, resulting in two narrow fibrils.

The widths of many fibrils, both the side and the narrow
types, were measured after being stained with PTA or UOx. The mean
widths of UOx-stained narrow and wide fibrils were significantly
less than the respective mean widths of fibrils stained with PTA
(or ammonium molybdate or silicon tungstate). The averages of the
widths of several hundred wide and narrow fibrils are reported in
Table 8.

The widths of both wide and narrow fibrils also varied widely
within each stain group (Figures 82 and 83), but there was a normal
distribution of widths in micrographs taken from each grid, and
within each grid the ratio of the mean width of narrow fibrils to
the mean width of wide fibrils remained close to 0.6. The overall
variation in fibril widths was due to the normal distribution in
widths measured from each grid as well as significant differences
among means calculated from each grid (Figure 84). The average
width of fibrils from even the same preparation would often be
different when calculated from micrographs taken at different
times. It could not be determined whether the differences among
various grids were due to variations in stain penetration or actual
variations in the fibrils, perhaps because of different degrees of
degradation. Varying stain penetration was the less likely reason
because when the same fibril preparation was stained on separate
grids on the same day the fibril widths calculated from each grid

were the same. No consistent correlation was found between any of

111

 

Q)

 

 

25‘
20-
15-
10‘

 

 

wvruvu1trvrti'1‘1'1'17'1‘"

 

 

 

 

 

 

 

 

 

 

NUMBER OF FIBRILS
bu
9

 

l
VIIIIII‘I IITIIIIIIY VIII

Ln

 

 

 

 

 

 

 

 

 

 

'r—-* ii)
4.5 55 6.5 7.5 85 95 10.5
to to to to to to to

5.4 64 7.4 8.4 9410:4114

WIDTH OF NARROW FIBRILS (nm)

Figure 82. Distribution of widths of narrow fibrils.

a. Distribution of widths of 97 UOx-stained narrow fibrils
from 42 micrographs.

b. Distribution of widths of 121 PTA-stained narrow fibrils
from 36 micrographs.

112

 

60

ID

 

50-

 

4o;

 

30«

 

 

20‘
105

 

 

1111111|Y111TIIIIIIIIII

 

 

 

 

 

 

 

 

 

ttrtT I '1

4o-

 

S’IltiSL-J :JO 8 3 SWON

 

30f

 

2C)f

 

IV I

10"

 

 

 

 

 

 

 

 

 

25 6.5 95105 11.5 12.5 135 14.5 15.5 165
to to to to to to to to to to

8.4 9.4 ’04 11.4 124134 14.4 15.4 16.4 17.4
WIDTH OF WIDE FlBRlLS(nm)

Figure 83. Distribution of widths of wide fibrils.

a. Distribution of widths of 239 UOx-stained wide fibrils
from 67 micrographs.

b. Distribution of widths of 138 PTA-stained wide fibrils
from 34 micrographs.

113

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

 

15 -a . 154-6
‘2 '7 I
10: 103

to : .1
93 .E E
E 5_ 5_
g; . .
LL : r-1 :
5 .c ..d
o: ' I
Lu .1 1 w
CD15 5_
E : l
I) . .
210. 10-
5-2 F7 555

L . - I'— _[|;L16"J
91011121314

 

101112131415 .
tomwmmmmm mmmmmwmm

9.9 109119129139149‘59169 9.91 091191 29139149159 169

WIDT H(nm) W'DTH(nm)

Figure 84. Each histogram shows the distribution of wide fibril
~widths from micrographs taken of one microscope grid. These examples

were chosen because they are near the extremes of the range of means
for each strain.

UOx-stained; mean width

a. = 10.8 i_0.5 nm
b. PTA-stained; mean width = 11.4 :_0.4 nm
c. UOx-stained; mean width = 13.5 :_l.5 nm
d. PTA-stained; mean width = 13.3 i 0.3 nm

114

several factors and the variations in means. The factors considered
were: age of culture, length of time from preparation of fibrils

to electron microscopy, type or pH of buffer (water), variations in
electron microscope magnifications, errors in print magnification,
and magnification of prints used for measurement.

The strands composing the UOx-stained narrow fibrils were
wider than those in the wide fibrils. PTA-stained strands of narrow
fibrils, however, were not significantly different from those of
wide fibrils (Table 8). The ribbons of material sometimes seen
alongside the wide fibrils varied in width from 1.2 to 8.5 nm.

In micrographs of newly disrupted cells, the fibrils sometimes
remained in bands (Figures 85 and 86). These bands were usually
but not always still attached to membrane. The fibrils in bands
were definitely of the narrow variety, since comparisons could be
made with individual narrow and wide fibrils on the same micrograph
(Figure 86). Also, the mean width of fibrils in bands was similar
to that of individual narrow fibrils, as was the width of fibrils
occasionally visible inside negatively stained cells (Table 8). The
interval between fibrils in negatively stained bands compared
favorably with that measured within negatively stained and cross
sectioned cells (Tables 7 and 8).

Effect of Chemical, Physical or
Enzymatic Treatment of Fibrils

 

One drop of a suspension of fibrils was mixed with one drop of
any of various solutions and allowed to stand at room temperature for
10 to 45 minutes, after which the fibrils were negatively stained

and viewed with the electron microscope. No effect was observed on

Figure 85. PTA-stained spiral band of narrow fibrils from
fibril preparation which had been dried and resuspended. Note
possible wide fibril (W). Bar = 0.1 um.

 

116

 

Figure 86. Negative stain of narrow fibrils in bands
(arrows). Compare with wide (W) and narrow (N) fibrils. Stain
was 2% PTA pH 6.7. Bar = 0.1 um.

 

 

I}

Figure 87. PTA—stained fibrils (F) disrupted with exposure
to 4 M urea overnight. Bar = 0.1 um.

117

fibrils in a final concentration of 33% ethanol, 50% chloroform-
methanol, 2% Triton X-100, 2% sodium deoxycholate, 0.005 N HCl or
2 M urea. Fibrils were not visible after treatment with 1 N HCl,
0.005 N NaOH, 1.3% sodium dodecyl sulfate or 6 M urea. Fibrils
exposed overnight to 4 M urea were still two-stranded but were
wrinkled (Figure 87). Fibrils which had been frozen and thawed
(Figure 74) or dried and resuspended (Figure 85) looked normal,
while those which were repeatedly sonicated broke into shorter
pieces. No change was observed a few minutes after fibrils were
treated with 1% trypsin at pH 7.7, but after 1 hr at RT, no fibrils
remained. One percent pronase at pH 7.7 destroyed the fibrils

immediately.

DISCUSSION

Cell Envelope and Axial Filaments

 

The Ryter-Kellenberger fixation used in this study is not the
ideal procedure for electron microscopic studies of bacterial cell
envelopes, but nevertheless some comments can be made about the
structure of the treponemal cell envelope based on these micro-
graphs. The findings here confirm earlier reports that the trep0nemal
cell envelope is similar to that of gram-negative cells (53,70,87,
88). The outer envelope was often expanded away from the cytoplasmic
cylinder, but the expansion and separation of the outer membrane
from the rest of the cell is not uncommon in osmiumefixed gram-
negative bacteria. The preservation of the outer membrane can be
improved when glutaraldehyde pre-fixation is used (75,84). Although
this technique was tried on T. refringens, it proved undesirable for
the observation of the intracytoplasmic structures that were the
main targets of this research.

Membrane invaginations were observed only in old, degenerating
cells. Ultrastructural studies to determine the conditions for
the occurrence of these membrane structures, their possible function
or their similarity to bacterial mesosomes would have to be done on
cells fixed by various procedures, including glutaraldehyde pre-

fixation.

118

119

Axial filaments have usually been reported to lie between the
outer envelope and the cytoplasmic cylinder, but they have been
seen in striated tubules (7,10,11,52,59,76) or in simple tubes of
membrane (11,72). Listgarten and Socransky (72) reported that
axial filaments attached to one end of the cell extended beyond
the other end of the cell, within a sleeve of outer envelope,
after cell division. In the present work, however, axial filaments
were more often seen in tubes of membrane directed away from the
cell than in their commonly reported position--wound around the
cytoplasmic cylinder inside the outer envelope. Little signifi-
cance can be ascribed to the position of these filaments because
the fixation used was not optimum for preservation of the outer
envelope. Negatively stained cells did not show these structures
unless they had been previously disrupted.

Striated tubules may only form when the outer envelope is
disrupted since they are often seen without axial filaments or in
association with outer envelope vesicles. The preparation for
electron microscopy may be important to their formation since they
are more often seen in negatively stained cells than in thin
sectioned cells. The functional significance of striated tubules
or membrane sleeves and their relationships to axial filaments in
normal cells is not known.

Researchers have reported that the basal body of an axial
filament is inserted in the cytoplasm, with the filament protruding
through a hole in the cytoplasmic membrane into the periplasmic
space (43,48,52). This conclusion has been based mostly on the

results of negative stains, in which the relationship between axial

120

filaments and cytoplasmic membrane cannot be exactly determined.
The few micrographs available of thin sectioned basal bodies (see
Figure 28 above and Figures 7, 8 and 9 in Jackson and Black [52])
suggest that the basal body remains outside the cytoplasmic membrane,
and that it is embedded in a pocket of cytoplasmic membrane rather
than positioned inside the membrane. The innermost ring of the
attachment apparatus of the bacterial flagellum is associated with
the cytoplasmic membrane, but is not actually inside the membrane (26).
Further evidence about the position of the axial filament basal
body may increase understanding of the role of axial filaments in
spirochetal motility.

It has been reported that fibrils are found on the same side
of the cell as the axial filament (44,47,48,49,68,81,101). This
relationship would be expected since the fibrils are found con-
sistently on the insides of the curves of the cell where the "axial"
filaments apparently are usually positioned (7,14,56,107). In this
study no correlation was found between the positions of fibrils and
those of the axial filaments in thin sectioned cells in which axial
filaments were present. This is probably due to the displacement of
the axial filament rather than the fibrils. The axial filaments
could have been displaced due to inadequate fixation of the outer
envelope or because the axial filaments of this strain of T.

refringens are defective in some way.

Fibril Purification

 

For technical reasons it was practical to use no more than

440 ml culture at a time for fibril isolation. Thus only about 120

121
ug of fibrils were available from each preparation for use in
chemical analysis, ultrastructural studies, etc. This small amount
of fibrils available precluded getting more definitive results from
the chemical analyses. The orcinol and diphenylamine tests did
show that the RNA and DNA were not completely removed and, thus,
the preparation was not chemically pure. A few protein peaks
besides the 97,000 molecular weight peak were present on the
tracing of the SDS acrylamide gel indicating that contaminating
proteins were present, although little cellular contamination was
visible with the electron microscope. It was not resolved whether
fibrils are composed of only one large protein subunit (97,000
molecular weight) or more than one subunit, possibly including
the 84,000 molecular weight peak. The possibility of carbohydrate
or lipid components has not been excluded.

Isolated fibrils have a tendency to clump together, as evi-
denced by the flocculent appearance of a suspension of fibrils and
the large aggregates visible under the electron microscope. Elec-
tron microscopy was used to gauge the disaggregation effect of
various chemicals because, with the small amount of fibrils avail-
able, this could not be done by visual determination. It was
noticed, however, that DNase reduced the rapid settling of fibril
clumps. With electron microscopy, no chemicals were found to dis—
aggregate the fibrils (Table 4), but it is possible that the clumps
were dispersed only to reassemble as the dispersing agent dialyzed
into the negative stain or as the suspension dried during preparation
for electron microscopy. Thus, even though Triton X-100 did not

appear to disaggregate fibril clumps, its success in the purification

122

procedure may have been due to disaggregation of the clumps as well

as to solubilization of membrane contaminants.

Fibril Location in Cell

 

Treponemal fibrils are found in a specific location in the
cell--next to the cytoplasmic membrane on the inside of the curves
of the cell. Fibrils in the cell follow the shortest route from
one end of the cell to the other, staying on the inside of the
cell's windings. Thus, the amplitude of the fibrils' helix would
be less than that of the cell and the length of the fibrils which
extend from end to end of a cell would be shorter than the cell's
length. Long fibrils released from a cell often partially retain
their spiral or curved shape, whether or not they are still organized
in a band (Figures 59, 52 and 85). In no instances do the fibrils
appear to be in a rigid helical form.

It is assumed that axial filaments are the agents of spiro-
chetal locomotion but it is not known how they effect this motion.
Many of the theories of the mechanism of translational movement
in spirochetes are based on the phenomenon of waves traveling down
the length of the spirochete (55,105). The apparently fixed
position of fibrils on the inside of the treponemal waves argues
against the traveling wave theory. Unless fibrils moved around
the inside circumference of the cell membrane while waves traveled
the cell, it would be expected that fibrils be seen also on the

outer side of curves, as well as at all other positions.

No data were found to suggest that each cell contains two

overlapping bands of fibrils, as was reported by Hovind Hougen

123

(44,45,47), rather than one band running the full length of the cell.
Areas in which two bands of fibrils interdigitate were never seen

in those negatively stained cells in which fibrils were visible.

Such an area remains a possibility because fibrils in these cells
are difficult to count accurately (Figure 36). All cross sections
of cells contained only one band, but those with 13, 14 or more
fibrils (Figure 34) could possibly have been cut through areas

where two smaller bands interdigitated. It is possible that
interdigitation of the fibrils occurs only at division points.

The observations here did not support the claim by some observers
(4,66,79) that the fibrils are in two or three bands, all of which
run the full length of the cell at different positions around the
circumference. Some cite the visible crossing of fibrils in nega-
tively stained cells as evidence that there are two separate bands
(4,79), but this crossing is actually due to the flattening of one
spiral band of cells (Figure 36).

Khorvat (66) apparently based his diagram showing three separate
fibril bands on a micrograph of a band of isolated fibrils which had
separated into three groups. This is flimsy evidence for such a
theory, especially since the micrograph bears a remarkable simi-
larity to one of Oinnnikov and Delektorskij's pictures of axial

filaments (79).

Fibril Ultrastructure

 

Some proposals can be made for the shape and structure of
treponemal fibrils from the evidence produced by electron microscopy.
The fibrils are more complex than simple tubules in both thin

sectioned and negatively stained preparations.

124

In thin sectioned fibrils there are several common structures
which can be found in both .cross and longitudinally sectioned cells.
It is likely that the dark band in longitudinal sections corresponds
to the widest part of the cross sectioned fibril, whereas the light
bands in longitudinal sections correspond to the light strips in
cross sections (Figure 88). The base of the longitudinally sec-
tioned fibril is indistinguishable from the edge of the inner layer
of the cytoplasmic membrane. The material sometimes seen beyond
the dark band in longitudinal sections could then be analogous to
the cap. Measurements of these dimensions do not correspond
exactly (height of cross sectioned cell excluding cap is 9.6 nm,
height of longitudinally sectioned fibril from the cytoplasmic
membrane through the dark band is 7.3 nm; height of cross sectioned
fibril including cap is 14.5, height of longitudinally sectioned
fibril including material above dark band is 16.2 nm) (Table 7).
but because of the difficulties in knowing the exact orientation of
the slice through the fibrils, especially longitudinally sectioned
ones, these measurements compare closely enough for the above
relationship to be feasible.

Both PTA-stained and UOx-stained fibril preparations contain
two distinct fibril types or orientations: wide and narrow. Since
the distribution of widths of narrow fibrils slightly overlaps that
of the wide fibrils (Figures 80 and 81), it might appear that wide
fibrils are simply fibrils on one end of a range of widths. How-
ever, the periodic striations seen on wide fibrils in UOx-stained
preparations clearly distinguish wide fibrils from narrow ones.

Thus there are definitely two separate fibril appearances or types.

H

1)...
/

 

Figure 88. Relationship between components of cross and
longitudinally sectioned fibrils.

1. cap
2. dark band
3. light area
4. base

126
It can be assumed that wide fibrils are either degraded forms of the
narrow types seen in situ or that the wide and narrow fibrils are
merely two different orientations of the same fibrils.

The narrow fibrils are seen in isolated fibrils preparations
as well as within cells and in bands from disrupted cells while
wide fibrils are seen only in isolated fibril preparations. The
average fibril width reported by other researchers, who usually
measured fibrils in situ, ranged from 6 to 9 nm, which corresponds
to the narrow width. The fact that wide fibrils are seen only in
preparations of isolated fibrils seems to support the hypothesis
that they are degraded fibrils. If it is assumed that fibrils are
tubules, despite the evidence from cross sections, the wide fibrils
could be flattened tubules since the width of wide fibrils is
approximately half the circumference of narrow fibrils. Other ways
of explaining the wide fibrils as degraded narrow fibrils are pos-
sible, but the observation of wide fibril strands which twist and
then appear as narrow fibrils is difficult to explain if wide
fibrils are merely degraded narrow fibrils, and is a serious objec-
tion to this hypothesis.

The second hypothesis, that wide and narrow fibrils are two
different views of the same type of fibril, is supported by more
evidence and a model can be proposed on the basis of this evidence.
Fibrils are probably not simple two-stranded structures or cylinders
since they do not appear flat or circular in cross sections. The
twisting of wide fibril strands is the most helpful observation when
trying to construct a three-dimensional model of fibrils. Once a

strand of a wide fibril is broken off (or the two strands are split

127
apart), the other appears to be free to twist on its side, revealing
that it was composed of two strands on top of one another. It is
proposed, then, that each fibril is composed of four strands, each
at a corner of a rectangular cross section. When the fibril is
viewed from the narrow side, it has a narrow two-stranded appearance,
and when viewed from the wide side, it has a wide two-stranded
appearance (Figure 89a). The periodic projections visible on each
strand of the wide pair apparently join together all four strands.
The two strands of each narrow pair are held tightly together by
means of the periodic components and, when the fibril splits in half,
it does so between the two narrow pairs, each of which retains its
periodic projections (Figure 89b). The periodicity can usually be
seen only from the wide side, where the strands are far enough apart
to allow stain penetration.

When fibrils inside a cell or in a band are viewed, they are
seen on their narrow side because that is the side which is attached
to the cytoplasmic membrane in this model. When fibrils are freed
from the band, they are preferentially oriented on the grid support
film on their wide side and only occasionally are they positioned on
their narrow side. The height of shadowed fibrils is closer to the
narrow than to the wide fibril width, giving further evidence that
fibrils more often lie on their widest side. When one narrow pair
of strands is broken away from the fibril, the remaining pair twists
to lie flat (Figure 89c). The ribbons sometimes seen on wide fibrils
are a little more difficult to explain, but these fibrils could be
partially disorganized. If the pair of strands in contact with the

grid were spread apart, without splitting completely apart, the lower

 

128

 

 

 

 

 

 

 

 

 

 

a 1
i 1 MINIMIMMIMWW -.
N {—2
2 _
b 3 4
l 1
N 4 3 4
“ ’ ’ " "th1? ,,
0
5 s
i 1 5 .
)4 - T11 ,, , 11 r f
d
7
.541.

   

 

 

Figure 89. Model of ultrastructure of treponemal fibril as
constructed from micrographs of negatively stained fibrils. Left-
hand column cross sections through model. Right—hand column is
what is seenwhen UOx-stained fibril is viewed in direction of arrow.

a.
b.
c.

d.

complete fibril

complete fibril, one strand of which has broken away
wide fibril in which one strand has twisted to reveal
narrow fibril

wide fibril with ribbons on each side

129
strands would be covered with more negative stain than the upper
pair (Figure 89d), resulting in a dark area of varying width along
each side of the fibrils.

Certain difficulties arise in trying to correlate the above
model with the appearance of thin sectioned fibrils. The narrowest
side of the cross sectioned fibrils faces the cytoplasmic membrane,
both in these micrographs and in Wiegand's (107), as it does in the
model, and the space between the strands on the negatively stained
wide fibril (containing the periodic substructure) would correspond
with the light strip seen in cross and longitudinally sectioned
fibrils. However, there is nothing in the negatively stained
material suggestive of the asymmetry one would expect due to the
cap seen in cross sections. Furthermore, the base of the cross
sectioned fibrils often appears to be three-stranded rather than two.
The differences seen between thin sectioned fibrils and negatively
stained fibrils could be due to the different fixations and stains
used in their preparation for electron microscopy.

The model proposed here thus accounts for much of the evidence
from thin sectioned and negatively stained fibrils. Certain con-
clusions are inescapable. Fibrils are not simple tubules or solid
rods but rather are formed of an arrangement of strands. Moreover,
a definite periodicity is present.

Comparison of Fibrils with Other Prokaryotic
and Eukaryotic Structures

 

 

Prokaryotic Structures

 

Prokaryotes contain a variety of nonmembrane-enclosed inclusion

bodies, some of which are proteinaceous (97). There are, however,

130
only a few which bear any similarity to treponemal cytoplasmic
fibrils. Ten nanometer’ fibrillar structures have been found in
Cytophaga columnaris (85) and Thiovulum majus (24).

The fibrils found in the gliding bacterium, Cytophaga columnaris,
are associated with the inner layer of the outer membrane (85). The
fibrils run parallel to each other in wide bands which sometimes
cross each other. The fibrils are about 10 nm wide and their center—
to—center distance is about 16 nm. Triangular cross sections of
these fibrils attached to the outer membrane look remarkably like
the treponemal fibrils attached to the cytoplasmic membrane. In
freeze-etched cells, these fibrils appear as linear arrays of par-
ticles (18), whereas treponemal fibrils are still fibrillar in
freeze-etched cells (51). It has been suggested that these fibrils
are the agents of gliding motility in Cytophaga (85). However, such
fibrils have not been observed in Myxococcus xanthus, another gliding
bacterium (18).

Thiovulum majus is a very large bacterium containing an
"organite antapical" which was suspected by its discoverers to be
the organ of motility (30). It is a polar body located between the
inner and outer membranes and composed of a clump of about a hundred
fibrils (10 nm wide) oriented perpendicularly to the cell surface.
Since Thiovulum was later shown definitely to have typical bacterial
flagella (24), the function of the fibrils remains obscure.

The fibrils found in Cytophaga and Thiovulum are similar to
treponemal cytoplasmic fibrils in size but are located in the peri-
plasmic space rather than in the cytoplasm. The only intracytoplasmic

structures similar to treponemal fibrils in size are the components

131
of a cytoplasmic membrane-associated organelle in Halobacterium
halobium (20). The organelle is composed of a series of wavy
striated fibrils, from 7.5 to 10 nm wide, which are quite different
morphologically from treponemal fibrils.

Treponemal fibrils have been called microtubules (48), but are
not the same as other bacterial "microtubules" (90,104), which are
morphologically similar to the rhapidosomes found in various species
of bacteria (113), including Spirocheta litoralis (59). Rhapido-
somes are large tubules (20 to 30 nm in diameter) and normally have
a definite length. They often appear only in degenerating cells and

have no known biological activity (25,69).

Eukaryotic Structures

 

A11 eukaryotic movement, muscular, ciliar, flagellar or cyto-
plasmic, is apparently realized through the ATP-dependent sliding
interaction of fibrillar proteins (1). Treponemal fibrils are not
at all similar to eukaryotic microtubules. Even rhapidosomes, which
are nearly the same size as microtubules, have only superficial
structural similarities to them (25).

Actin and myosin are present in nonmuscle cells of widely sepa—
rated eukaryotic phyla, including Protozoa, but have not been found
in prokaryotes, with the possible exception of an actin-like protein
in E. coli (74). Actins are essentially the same in all cells. The
globular monomers, with a molecular weight of about 45,000, polymerize
into double helical filaments about 6 nm wide. Myosins vary more
than actins but are all made of heavy chains (180,000 to 240,000

molecular weight) and light chains (12,000 to 19,000 molecular

132
weight). Most myosins have a globular head and a rod—like tail.
All myosins have similar properties of actin binding and ATP
hydrolysis (89). Treponemal fibrils do not resemble actin, myosin
or tubulin either in molecular weight or ultrastructure. It would
be interesting, however, to know if fibrils are involved in a
sliding filament motility interaction or have any other related
properties, e.g., ATPase activity or ability to stimulate ATPase

activity.

Function of Fibrils

 

Cytoplasmic fibrils appear to be unique to the genus Treponeme.
The question may be asked why treponemes should have fibrils while
other spirochetes appear not to have them. Even though all spiro-
chetes are obviously related and have peculiar characteristics in
common, there are many differences among the species. Spirochetes
are free-living, pathogenic or nonpathogenic host-associated and
range in size from 0.1 to 0.75 pm thick and 2 to 250 um long (15).
Among other chemical and serological differences is the variation
in peptidoglycan composition. Treponeme and Spirocheta have peptido—
glycan similar to that of gram-negative cells with the exception
that ornithine is present rather than meso-diaminopimelic acid.
Leptospires, on the other hand, do have meso-diaminopimelic acid (3).

The function of treponemal cytoplasmic fibrils is unknown, but
it seems unlikely that they are nonfunctional. They are in specific
relationships with other cell components (DNA, membrane) and are in
a specific location relative to the shape of the cell. Fibrils are

not transitory but are found in cells of all growth stages. Based

133

on the location of the fibrils, two suggestions can be made as to
their possible function: 1) shape maintenance and 2) DNA replica—

tion, segregation, or transcription.

Maintenance of Shape

 

Because isolated peptidoglycan often retains the shape of the
cell, it is commonly cited as the shape maintaining component of the
cell. But Henning has pointed out that it is not clear whether
peptidoglycan is the only device for maintaining shape. Other cell
components which may be involved in various bacteria are outside
protein layers, the outer membrane or the cytoplasmic membrane (41).
The shape-maintaining function must also be separated from the shape-
specifying role since the information required to produce peptido-
glycan of a specific shape could only involve either enzymes or a
matrix onto which the peptidoglycan is laid down (42).

Treponemal fibrils are in a position to pull a rod into a
helical shape, as was once hypothesized for axial filaments. Brad-
field and Cater (l4) wound an elastic band spirally around a piece
of rubber tubing which had a metal rod inside and fastened the
elastic to the tube very few inches. When the rod was removed, the
tube was in the form of a spirochete. Tubing can also be held in a
helical shape by an internal string (representing cytoplasmic fibrils).
The string need only be passed through the tube and tied to the tube
at both ends after pulling so that the string is shorter than the
tube. If the tube is transparent it can be readily demonstrated
that the string follows the inside of the curves and is helical, but

with a smaller amplitude than that of the tubing.

134

Axial filaments were assumed not to be responsible for the
helical shape once it was discovered that each filament is attached
to the cell at only one point. There are also difficulties in
ascribing the role of shape maintenance to the fibrils. For this
model to be correct, it is necessary that there be only one band of
fibrils attached at each end of the cell or that, if two bands are
present and interdigitate, the fibrils be securely attached to the
membrane. The nature of treponemal fibril attachment to the cyto-
plasmic membrane is not known and there appears not to be a specialized
attachment apparatus at the end of the fibrils.

In any case, fibrils could not be the sole mechanism for shape
maintenance since they do not completely surround the cell. They
could only provide the tension to pull an already-formed cylinder

into a helical shape.

Cell Division

 

The location of the nuclear area of treponemes next to the
fibrils was consistently observed in the electron micrographs of
thin sectioned T. refringens. Except for two vague references to
a lighter area near the fibrils (48,83), this has not been previously
reported. Most investigators have not used the Ryter-Kellenberger
fixation which provides for the best preservation of the bacterial
nuclear area (64). Even so, a few published pictures do allow the
observation of the relationship between the nuclear area and the
fibrils to be extended to other species of treponemes (44,57,101).
When cells were grown in 3H-thymidine and then lysed and centri-

fuged on a gradient, the labeled DNA was distributed into several

135

areas of the gradient. This procedure failed to demonstrate the
attachment of DNA to either fibrils or membrane components. The
major difficulty in this approach is that the disruption procedure
must release the fibrils from the membrane completely while allowing
the DNA to remain attached to either fibrils or membrane. The only
way in which fibril and membrane could be completely separated from
one another in fibril isolation procedures was with the use of DNase.

The Kleinschmidt spreading procedure demonstrated DNA released
from the cell while still attached to cell components. The resolu-
tion was not good enough, however, to demonstrate the attachment
of DNA to fibrils rather than membrane. The lack of resolution in
this procedure resulted from the necessity of using cytochrome c
as a spreading agent and uranyl acetate as a stain in order to
visualize the DNA. These are not good conditions for clear visuali-
zation of the fibrils.

The data, then, which best demonstrate the attachment of DNA to
fibrils is from micrographs of thin sectioned cells. DNA was located
near the fibrils in all cases and no evidence was seen for the
attachment of DNA to the membrane. In thin sections of spheroplast-
like cells, whose cytoplasm was disrupted, the strands of DNA could
be seen directly attached to the cytoplasmic fibrils.

The attachment of DNA to the cell membrane of several bacteria
has been established (71). Cytological studies demonstrated this
attachment in micrographs of thin sectioned gram-negative and gram-
positive cells (92) and of disrupted cells prepared by the
Kleinschmidt spreading technique (32,40,103). Biochemical frac-

tionation studies have also demonstrated this association (71).

136

This connection has been hypothesized to be necessary for DNA repli-
cation, segregation or transcription (54,71). Since the DNA in
treponemes appears to be attached to fibrils rather than membrane,
one or more of the roles usually ascribed to the membrane may
actually be a function of cytoplasmic fibrils.

There is recent evidence that E. coli DNA may be attached to
the membrane at a region where the cell wall closely adheres to
the membrane. It has been suggested that association with the wall
may be necessary if the site is to have sufficient physical strength
for segregation of the chromosomes at cell division (78). In trepo-
nemes, perhaps fibrils are responsible for the segregation of DNA
which is taken care of by the mitotic apparatus in eukaryotes and

the membrane or membrane-wall in at least some of the bacteria.

APPENDIX

APPENDIX

ULTRASTRUCTURE OF TREPONEMA HYODYSENTERIAE

Introduction

 

Harris et a1. (88) isolated two types of spirochetes from the
colon of pigs infected with swine dysentery. Further work demon-
strated that one of the spirochetes could produce the disease in
healthy pigs and the name T. hyodysenteriae was proposed for the
organism. Its inclusion in the genus Treponema was based on its
requirement for anaerobiosis, structure, and location in the host
(37). Treponema hyodysenteriae is a large, loosely coiled spiro-
chete, measuring 0.3 to 0.4 um in diameter and S to 8 pm in length.
Seven to nine axial filaments originate from each end of the cell
(33,37).

Ultrastructural studies on this organism were initiated in an
attempt to determine if cytoplasmic fibrils were present and could

be isolated with the same procedure used for T. refringens.

Materials and Methods

 

Treponeme hyodgsenteriae, obtained from J. M. Kinyon, Iowa
State University, Ames, was grown in a prereduced anaerobically
sterilized broth at 37 C as described by Kinyon and Harris (67).

Cells were disrupted with the CLBS procedure and centrifuged

on a Renografin gradient as described earlier for T. refringens.

137

138
Several of the resulting bands were dialyzed against distilled
water and used for electron microscopy.
Whole cells or cell fractions were negatively stained with
0.5% PTA, pH 7.5, on a carbon-Formvar coated grid and observed with

a Philips 300 electron microscope.

Results

The outer envelope loosely enclosed the cytoplasmic cylinder
and the axial filaments (Figure 90). Nine to 11 axial filaments
were inserted subterminally at each end of the cell. The insertion
points were at various distances from the end of the cell (Figure
91).

When T. hyodysenteriae was treated with the CLBS procedure
used on T. refringens, the cells lysed. Electron microscopy
revealed membrane, occasional striated tubules and many fragments
of axial filaments, some with hooks. No fibrils were found.

The bands which formed after centrifugation in a Renografin
gradient were not in the same pattern as that for T. refringens.
Most of the membrane appeared in a band near the top of the gradient.
In a band of approximately the same density as that of fibrils of
T. refringens, a few bundles of 3 nm fibrous structures were seen
(Figure 92).

Axial filaments, in pieces of various lengths, were found in
several bands. The filaments were sheathed whereas their hooks,
if present, were not sheathed (Figures 93 and 94). An electron
dense line ran down the middle of many of the filaments. The fila-

ments were broken jaggedly at many points and often the segments

139

 

Figure 90. T. hyodysen'teriae negatively stained with PTA.
The cell has partially curled up inside the outer envelope (0E).
Bar = 0.5 um.

 

Figure 91. Enlargement of Figure 90. Points where axial
filament basal bodies insert into the cell are visible (arrows).
Bar = 0.5 um.

141

 

 

Figure 92. PTA-stained bundles of fibers from CLBS-disrupted
T. hyodysenteriae. Bar = 0.5 pm.

 

Figure 93. PTA-stained axial filament fragments from CLBS-
disrupted T. hyodysenteriae. Hooks (H) are unsheathed. Bar =
0.5 um.

142

 

 

Figure 94. PTA—stained axial filament fragments from CLBS-
disrupted T. hyodysenteriae. Sheath (Sh) is visible, as well as
electron dense line running dOWn the center of the filament (L).
Axial filaments are broken (B) and still attached, apparently with
the sheath. Bar = 0.5 pm.

143

were still attached to each other, presumably by the sheath (Figure
94). The breaks were often V-shaped and the filaments appeared to
have polarity since filament on one side of each break was pointed
and was indented on the other (Figure 94). It was determined that
the distal end of the filament had the indentation by looking at

filament segments which still had hooks attached (Figure 93).

Discussion

 

It is still unclear whether T. hgodysenteriae contains cyto-
plasmic fibrils of the type found in other treponemes. Fibrils were
not seen in guinea pig serum-disrupted cells under the same condi-
tions in which they are easily found in T. refringens. The fibrous
structures found after centrifugation are narrower than cytoplasmic
fibrils but are about the same size as the strands of which cyto-
plasmic fibrils are composed. Perhaps fibrils from various trepo-
nemal species differ in their fragility. The demonstration of
fibrils next to the membrane of thin-sectioned T. hyodysenteriae
cells would be required to determine if cytoplasmic fibrils were
present.

Short fragments of bacterial flagella reportedly possess dif-
ferently formed ends (13), but this has not yet been reported for
spirochetal axial filaments. The indented, or incised, end of a
bacterial flagellum is on the distal end as was reported above for

T. hyodysenteriae axial filaments.

LITERATURE CITED

10.

ll.

LITERATURE CITED

Alicea, H. A. and F. L. Renaud. 1975. Actin-tubulin homology
revisited. Nature (London) 257:601-602.

Ashwell, G. 1957. Colorimetric analysis of sugars, p. 73-
105. £2_S. P. Colowick and N. 0. Kaplan (eds.), Methods in
Enzymology, Vol. III. Academic Press, New York.

Azuma, I., T. Taniyama, Y. Yamamura, Y. Yanigihara, Y. Hattori,
S. Yasuda, and I. Mifuchi. 1975. Chemical studies on the
cell walls of Leptospira biflexa strain Urawa and Treponema
pallidum strain Reiter. Jap. J. Microbiol. 19:45-51.

Babudieri, B. 1972. Discussions, p. 157-159. 1§_B. Babudieri
(ed.), The Fine MOrphology of Spirochaetas. Leonarda Edizioni
Scientificha, Rome.

Bertalanffy, L., F. Masin and M. Masin. 1956. Use of acridine-
orange fluorescent technique in exfoliative cytology. Science
124:1024-1025.

Bharier, M. and D. Allis. 1974. Purification and characteri-
zation of axial filaments from Treponema phagadenis Biotype
reiterii (the Reiter treponeme). J. Bacteriol. 120:1434-1442.

Bharier. M. A., F. A. Eiserling and s. c. Rittenberg. 1971.
Electronmicroscopic observations on the structure of Treponema
zuelzerae and its axial filaments. J. Bacteriol. 105:413-421.

Bharier, M. A. and S. C. Rittenberg. 1971. Chemistry of
axial filaments of Treponema zuelzerae. J. Bacteriol. 105:
422-429.

Bharier, M. A. and S. C. Rittenberg. 1971. Immobilization
effects of anticell and antiaxial filament sera on Treponema
zuelzerae. J. Bacteriol. 105:430-437.

Birch-Andersen, A., K. Hovind Hougen, and C. Borg-Petersen.
1973. Electron microscopy of Leptospira. I. Leptospira
strain Pomona. Acta Path. Microbiol. Scand. sect. B 81:
665-676.

Bladen, H. A. and E. G. Hampp. 1964. Ultrastructure of
Treponema microdentium and Borrelia Vincentii. J. Bacteriol.
87:1180-1191.

144

12.

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

145

Blakemore, R. P. and E. Canale-Parola. 1973. Morphological
and ecological characteristics of Spirochaeta plicatilis.
Arch. Mikrobiol. 89:273-289.

Bode, W. 1973. The bacterial flagella and the flagellar
protein flagellin. Angew. Chem. Int. Ed. 12:683-693.

Bradfield, J. R. G. and D. B. Cater. 1952— Electron—microscopiC

evidence on the structure of spirochaetes. Nature (London)
169:944-946.

Breznak, J. A. 1973. Biology of nonpathogenic, host-
associated spirochetes. Crit. Rev. Microbiol. 2:457-489.

Buchanan, R. E. 1925. Classification of the genera and
higher groups of bacteria. Historical, 1773-1922, p. 15-108.
$2_General Systematic Bacteriology. Williams and Wilkins,
Baltimore.

Bulloch, W. 1936. Classification of bacteria, p. 171-203.
12_The History of Bacteriology. Oxford University Press,
London.

Burchard, R. P. and D. T. Brown. 1973. Surface structure of
gliding bacteria after freeze-etching. J. Bacteriol. 114:
1351-1355.

Burrill, T. J. 1882. Classification of bacteria, p. 32-65.
In The Bacteria: An Account of Their Nature and Effects,
Together With a Systematic Description of the Species. H. W.
Rokka, State Printer and Binder, Springfield, Illinois.

Cho, K. Y., C. H. Doy and E. H. Mercer. 1967. Ultrastructure
of the obligate halophilic bacterium Halobacterium halobium.
J. Bacteriol. 94:196-201.

Costerton, J. W., J. M. Ingram and K.-J. Chang. 1974.
Structure and function of the cell envelope of gram-negative
bacteria. Bacteriol. Rev. 38:87-110.

Cox, C. D. 1972. Shape of Treponema pallidum. J. Bacteriol.
109:943-944.

Davis, R. W., M. Simon and N. Davidson. 1971. Electronmicro-
scopic heteroduplex methods for mapping regions of base
sequence homology in nucleic acids, p. 413-428. £2_L.
Grossman and K. Modaves (eds.), Methods in Enzymology, Vol.
21. Academic Press, New York.

de Boer, W. E., J. W. M. LaRiviére and A. L. Houwink. 1961.
Observations on the morphology of Thiovulum majus Hinze.
Antonie van Leeuwenhoek J. of Microbiol. and Serol. 27:447-456.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

146

Delk, A. S. and C. A. Dekker. 1972. Characterization of
rhapidosomes of Saprospira grandis. J. Mol. Biol. 64:287-295.

De Pamphilh5.M.I~ and J. Adler. 1971. Attachment of flagellar
basal bodies to the cell envelope: Specific attachment to the
outer, lipopolysaccharide membrane and the cytoplasmic membrane.
J. Bacteriol. 105:396-407.

Dobell, C. 1932. The first observations on entozoic protozoa
and bacteria, p. 217-255. £2_Antony van Leeuwenhoek and His
"Little Animals". Harcourt, Brace and Co., New York.

Ellinghausen, H. C. 1959. Nephelometry and a nephelo-culture
flask used in measuring growth of leptospires. Am. J. Vet.
Res. 20:1072-1076.

Fairbanks, G., T. L. Steck, and D. F. H. Wallach. 1971.
Electrophoretic analysis of the major polypeptides of the
human erythrocyte membrane. Biochem. 10:2606-2617.

Faure-Fremiet, E. and Rouiller, C. 1958. Etude au microscope
électronique d'une bacterie sulfureusa, Thiovulum majus Hinze.
Exptl. Cell Res. 14:29-46.

Filip, C., G. Fletcher, J. L. Wulff and C. F. Earhart. 1973.
Solubilization of the cytoplasmic membrane of Escherichia coli

by the ionic detergent sodium-lauryl sarcosinate. J. Bacteriol.
115:717-722.

Ganesan, A. T. 1967. Particulate fractions in macromolecular
synthesis and genetic transformation, p. 19-47. £2_Voge1,

H. J. et al. (eds.), Organizational Biosynthesis. Academic
Press, Inc., New York. '

Glook, R. D., D. L. Harris, and J. P. Kluge. 1974. Localiza-
tion of spirochetes with the structural characteristics of
Treponema hyodysenteriae in the lesions of swine dysentery.
Inf. Imm. 9:167-178.

Hardy, P. H., W. R. Fredericks and E. E. Nell. 1975. Isola-
tion and antigenic characteristics of axial filaments from the
Reiter treponeme. Inf. Imm. 11:380-386.

Hardy, P. H. and E. E. Nell. 1961. Influence of osmotic
pressure on the morphology of the Reiter treponeme. J. Bac-
teriol. 82:967-978.

Hardy, P. H., E. E. Nell and W. R. Fredericks. 1975. Isola-
tion and immunologic study of treponemal outer envelopes, p.
72. Am. Soc. of Microbiol. Proc.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

147

Harris, D. L., Glock, R. D., C. R. Christensen, and J. M.
Kinyon. 1972. Inoculation of pigs with Treponema hyodysen-
teriae (new species) and reproduction of the disease. Vet.
Med./Small An. Clin. January, p. 61-64.

Harris, D. L., J. M. Kinyon, M. T. Mullin and R. D. Glock.
1972. Isolation and propagation of spirochetes from the
colon of swine dysentery affected pigs. Can. J. Comp. Med.
36:74-76.

Haschemeyer, R.Ii. and R. J. Myers. 1972. Negative staining,
p. 101-147. IE_M. A. Hayat (ed.), Principles and Techniques
of Electron Microscopy: Biological Applications, Vol. 2.

Van Nostrand Reinhold Co., New York.

Heidrich, H.-G. and W. L. Olsen. 1975. DNA-envelope complexes
from Escherichia coli: A complex-specific protein and its
possible function for the stability of the complex. J. Cell
Biol. 67:444-460.

Henning, U. 1975. Determination of cell shape in bacteria.
Annu. Rev. Microbiol. 29:45-60.

Henning, U. and U. Schwarz. 1973. Determinants of cell
shape, p. 413-440. IE_L. Leive (ed.), Bacterial Membranes
and walls. Marcel Dekker, New York.

Holt, S. C. and E. Canale-Parola. 1968. Fine structure of
Spirochaeta stenostrepta, a free-living anaerobic spirochete.
J. Bacteriol. 96:822-835.

Hovind Hougen, K. 1974. The ultrastructure of cultivable

treponemes. I. Treponema phagadenis, Treponema vincentii,
Treponema refringens. Acta Path. Microbiol. Scand. sect. B
82:329-344.

Hovind Hougen, K. 1974. The ultrastructure of cultivable
treponemes. II. Treponema calligyrum, Treponema minutem,
Treponema microdentiium. Acta Path. Microbiol. Scand. sect.
B 82:495-507.

Hovind Hougen, K. 1975. The ultrastructure of cultivable
treponemes. III. Treponema genitalis. Acta Path. Microbiol.
Scand. sect. B 83:91-99.

Hovind Hougen, K. 1972. Further observations on the ultra-
structure of Treponema pallidum Nichols. Acta Path. Micro-
biol. Scand. sect. B 80:297-304.

Hovind Hougen, K. and A. Birch-Andersen. 1971. Electron
microscopy of endoflagella and microtubules in Treponema
Reiter. Acta Path. Microbiol. Scand. sect. B 79:37-50.

 

4!

In

In

In

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

148

Hovind Hougen, K., A. Birch-Andersen, and H.-J. S. Jensen.
1973. Electron microscopy of Treponema cuniculi. Acta
Path. Microbiol. Scand. sect. B. 81:15-26.

Jackson, S. W. 1969. The fine structure of Treponema
pallidum Nichols strain. Thesis. Baylor University,
Houston, Texas.

Jackson, S. and S. H. Black. 1971. Ultrastructure of
Treponema pallidum Nichols following lysis by physical and
chemical means. I. Envelope, wall, membrane and fibrils.
Arch. Mikrobiol. 76:308-324.

Jackson, S. and S. H. Black. 1971. Ultrastructure of
Treponema pallidum Nichols following lysis by physical and
chemical means. II. Axial filaments. Arch. Mikrobiol.
76:325-340.

Jackson, S. W. and P. N. Zey. 1973. Ultrastructure of
lipopolysaccharide isolated from Treponema pallidum. J.
Bacteriol. 114:838-844.

Jacob, F., S. Brenner and F. Cuzin. 1963. On the regulation
of DNA replication in bacteria. Cold Spring Harbor Symp.
Quant. Biol. 28:329-348.

Jahn, T. L. and E. C. Bovee. 1965. Movement and locomotion
of micro-organisms. Annu. Rev. Microbiol. 19:21-58.

Jepsen, O. B., K. Hovind Hougen and A. Birch-Andersen. 1968.
Electron microscopy of Treponema pallidum Nichols. Acta
Path. Microbiol. Scand. 74:241-258.

Johnson, R. C., M. S. Wachter and D. M. Ritzi. 1973. Trepo-
neme outer cell envelope: Solubilization and reaggregation.
Inf. and Imm. 7:249-258.

Jones, R. H., Z. Yoshii and O. J. Carver. 1970. Some morpho-
logic features of axial fibers and body fibrils of treponemata,
p. 110-111. _I_n_C. J. Arcenaux (ed.), Proc. 28th Annu. Mtg.
Elec. Micros. Soc. Am. Claitor's Publishing Division, Baton
Rouge.

Joseph, R. and E. Canale-Parola. 1972. Axial fibrils of
anaerobic spirochetes: Ultrastructure and chemical charac-
teristics. Arch. Mikrobiol. 81:146-168.

Joseph, R., S. C. Holt, and E. Canale-Parola. 1973. Peptido-
glycan of free-living anaerobic spirochetes. J. Bacteriol.
115:426-435.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

149

Kast, C. C. and J. A. Kolmer. 1933. One successful cultiva-
tion of Spirocheta pallida from syphilitic chancre of the
rabbit. Amer. J. Syph. 17:533-538.

Kawata, T. 1961. Electron microsc0py of fine structure of
Borrelia duttonii. Jap. J. Microbiol. 5:203-209.

Kawata, T. and T. Inoue. 1964. Fine structure of the Reiter
treponeme as revealed by electron microscopy using thin
sectioning and negative staining techniques. Jap. J. Micro-
biol. 8:49-65.

Kellenberger, E. 1962. The study of natural and artificial
DNA-plasms by thin sections, p. 233-249. In_R. J. C. Harris
(ed.), The Interpretation of Ultrastructure, Vol. 1. Academic
Press, New York.

Kellenberger, E., A. Ryter and J. Séchaud. 1958. Electron
microscope study of DNA-containing plasms. II. Vegetative
and mature phage DNA as compared with normal bacterial
nucleoids in different physiological states. J. Biophys.
Biochem. Cytol. 4:671-676.

Khorvat, I. K. ‘1972. Structure and movement of a treponeme.
Acta Microbiol. Acad. Sci. Hung. 19:121-124.

Kinyon, J. A. and D. L. Harris. 1974. Growth of Treponema
hyodysenteriae in liquid medium. Vet. Rec. 95:219-220.

Klingmfiller, G., Y. Ishibashi, K. Radke. 1968. Der elek-
tronenmikroskopische Aufbau des Treponema pallidum. Arch.
ffir Klin. u. Exp.Dermatol. 233:197-205.

Kuhrt, M. and J. L. Pate. 1973. Isolation and characteriza-
tion of tubules and plasma membranes from Cytophaga columnaris.
J. Bacteriol. 114:1309-1318.

Lataste-Dorolle, C. 1972. Structure and nature of the
membrane of treponemas, p. 59-62. IB_B. Babudieri (ed.),
The Fine Morphology of Spirochaetas. Leonardo Edizioni
Scientifiche, Rome.

Leibowitz, R. J. and M. Schaechter. 1975. The attachment of
the bacterial chromosome to the cell membrane. Int. Rev.
Cytol. 41:1-28.

Listgarten, M. A. and S. S. Socransky. 1964. Electron micros-
copy of axial fibrils, outer envelope, and cell division of
certain oral spirochetes. J. Bacteriol. 88:1087-1103.

Lowry, O. H., N. J. Rosebrough, A. L. Farr and R. J. Randall.
1951. Protein measurement with the folin phenol reagent.
J. Biol. Chem. 193:265—275.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

150

Minkoff, L. and R. Damadian. 1975. Actin-like proteins from
Escherichia coli: Concept of cytotonus as the missing link
between cell metabolism and the biological ion-exchange resin.
J. Bacteriol. 125:353-365.

Murray, R. G. E. 1968. Bacterial cell wall anatomy in rela-
tion to the formation of spheroplasts and protoplasts, p. 1-18.
In_L. B. Guze (ed.), Microbial Protoplasts and L-Forms.
Williams and Wilkins, Baltimore.

Nauman, R. K., S. C. Holt and C. D. Cox. 1969. Purification,
ultrastructure, and composition of axial filaments from
Leptospira. J. Bacteriol. 98:264-280.

Nevin, T. A. and W. J. Guest. 1967. Complement and lysozyme
requirements for spirochetolysis in guinea pig serum. J.
Bacteriol. 94:1388-1393.

Olsen, W. L., H.-G. Heidrich, K. Hannig and P. H. Hofschneider.
1974. DNA complexes isolated from Escherichia coli by free-
flow electrophoresis: Biochemical and electron microscope
characterization. J. Bacteriol. 118:646-653.

Oinnnikov, N. M. and V. V. Delektorskij. 1972. The fibrils
of Treponema pallidum (an electron—microscope investigation).
WHO/VDT/RES/72.28l. World Health Organization, Geneva.

Oinnnikov, N. M. and V. V. Delektorskij. 1971. Current
concepts of the morphology and biology of Treponema pallidum
based on electron microscopy. Brit. J. Vener. Dis. 47:315-
328.

Oinnnikov, N. M. and V. V. Delektorskij. 1970. Treponema
pertenue under the electron microscope. Brit. J. Vener. Dis.
46:349-379.

Ovcinnikov, N. M. and V. V. Delektorskij. 1969. Further
material on the morphology of Treponema pallidum under the
electron microscope. Brit. J. Vener. Dis. 45:87-116.

Oinnnikov, N. M. and V. V. Delektorskij. 1968. Further
study of ultrathin sections of Treponema pallidum under the
electron microscope. Brit. J. Vener. Dis. 44:1-34.

Pate, J. L. and E. J. Ordal. 1967. The fine structure of
Chondrococcus columnaris. I. Structure and formation of
mesosomes. J. Cell Biol. 35:1-13.

Pate, J. L. and E. J. Ordal. 1967. The fine structure of
Chondrococcus columnaris. III. The surface layers of
Chondrococcus columnaris. J. Cell. Biol. 35:37-51.

 

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

Pillot, J. 1969. Analytical serology of Treponemataceae,
p. 123-183. £2_J. B. G. Kwapinski (ed.), Analytical Serology
of Microorganisms, Vol. 1. Interscience Publishers, New York.

Pillot, J. 1965. Contribution a 1'étude du genre Treponema:
Structures anatomique et antigenique. Thesis, L'Universite
de Paris, Paris.

Pillot, J. and A. Ryter. 1965. Structure des spirochetes.
I. Etude des genera Treponema, Borrelia et Leptospira au
microsc0pe électronique. Ann. Inst. Past. 108:791-804.

Pollard, T. D. and R. R. Weihing. 1974. Actin and myosin
and cell movement. Crit. Rev. Biochem. 2:1—65.

Pope, L. M. and P. Jurtshuk. 1967. Microtubule in Azoto-
bacter vinelandii strain 0. J. Bacteriol. 94:2062-2064.

Ritchie, A. E. and H. C. Ellinghausen. 1965. Electron
microsc0py of leptospires. I. Anatomical features of
Leptospira pomona. J. Bacteriol. 89:223-233.

Ryter, A. 1968. Association of the nucleus and the membrane
of bacteria: A morphological study. Bacteriol. Rev. 32:
39-54.

Schaudinn, F. 1905. Zur Kenntis der Spirochaeta pallida.
Deutsche Med. Wochensch. 31:1665-1667.

Schaudinn, F. 1905. Korrespondenzen. Deutsche Med. Wochensch.
31:1728.

Schaudinn, F. and E. Hoffmann. 1905. Vorlaufiger Bericht
fiber das Vorkommen von Spirochaeten in syphilitischen
Krankheitsprodukten u. bei Papillomen. Arb. Kaiserl.
Gesundh. 22:528-534.

Sequira, P. J. L. 1956. The morphology of Treponema pallidum.
Lancet 2:749.

Shively, J. M. 1974. Inclusion bodies of prokaryotes. Annu.
Rev. Microbiol. 28:167-187.

Simpson, C. F. and F. H. White. 1961. Electron microscope
studies and staining reactions of leptospires. J. Inf. Dis.
109:243-250.

Smibert, R. M. 1974. The spirochetes, p. 167-195. £2.R- E.
Buchanan and N. E. Gibbons (eds.), Bergey's Manual of Determina-
tive Bacteriology. The Williams and Wilkins Co., Baltimore.

Smibert, R. M. 1973. Spirochaetales, a review. Crit. Rev.
Microbiol. 2:491-552.

 

101

102

103

104

105

106

10'.

10:

10

11

11

ll

101.

102.

103.

104 O

105.

106.

107.

108.

109.

110.

111.

112.

113.

152

Swain, R. H. A. and N. Anderson. 1972. The ultra-structure
of species of treponemas and borreliae, p. 115-139. I2_B.
Babudieri (ed.), The Fine Merphology of Spirochaetas.
Leonardo Edizioni Scientifische, Rome.

Takashima, I and R. Yanagawa. 1975. Immunizing effects of
structural components of Leptospira icterohaemorrhagiae.
Zentrallbl. Bakt. Par. Inf. Hyg. 233:93-98.

Tremblay, G. Y., M. J. Daniels and M. Schaechter. 1969.
Isolation of a cell membrane-DNA-nascent RNA complex from
bacteria. J. Mol. Biol. 40:65-76.

van Iterson, W., J. F. M. Hoeniger and E. N. van Zanten.
1967. A "microtubule" in a bacterium. J. Cell Biol. 32:1-10.

Wang, C.-Y. and T. L. Jahn. 1972. A theory for the locomo-
tion of spirochetes. J. Theor. Biol. 36:53-60.

Weber, K. and M. Osborn. 1969. The reliability of molecular
weight determinations by dodecyl sulfate-polyacrylamide gel
electrOphoresis. J. Biol. Chem. 244:4406-4412.

Wiegand, S. E., P. L. Strobel and L. H. Glassman. 1972.
Electron microscopic anatomy of pathogenic Treponema pallidum.
J. Invest. Dermatol. 58:186-204.

Willcox, R. R. and T. Guthe. 1966. Culture of T. pallidum in
vitro: II. The cultured strains, p. 54-60. In Treponema
pallidum--A Bibliographical Review of the Merphology, Culture
and Survival of T. pallidum and Associated Organisms. World
Health Organization, Geneva.

Williams, C. A. and M. W. Chase. 1968. Carbohydrate analysis,
p. 282-305. £2.C- A. Williams and M. W. Chase (eds.), Methods
in Immunology and Immunochemistry, Vol. II. Academic Press,
New York.

Williamson, D. L. 1974. Unusual fibrils from the spirochete-
like sex ratio organism. J. Bacteriol. 117:904-906.

Winslow, C.-E.A., J. Bradhurst, R. E. Buchanan, C. Krumwede,
L. A. Rogers and G. H. Smith. 1917. The families and genera
of bacteria. J. Bacteriol. 2:205-566.

Winter, G. 1884. Classe, Schizomycetes, p. 33-67. In_
Rabenhorst's Kryptogamen-Flora von Deutschland, Oesterreich
u. der Schweiz. Eduard Kummer, Leipzig.

Yamamoto, T. 1967. Presence of rhapidosomes in various species
of bacteria and their morphological characteristics. J.
Bacteriol. 94:1746-1756.

114.

115.

116.

153

Yanagawa, R. and S. Faine. 1966. Morphological and serological
analysis of leptospiral structure. Nature (London) 211:823-
826 O

Yanigihara, Y., Y. Hattori and I. Mifuchi. 1975. Purifica-
tion of microfibers of spirochetes, p. 4. Abstracts First
Symposium on the Biology of the Parasitic Spirochetes.
Minneapolis.

Yanigihara, Y. and I. Mifuchi. 1968. Microfibers present in
surface structures of Leptospira. J. Bacteriol. 95:2403-2406.