ago-L-
p“

1" . ’3

1345556!
- ~er

.3:

‘ 1
r .

l

”E.“

m
m. a,“

u»

‘n ..

‘ n
H.
3

I”...

.; q
r:~

r c. '
,.. '2'.
- N,
‘tfi'z

 

-—-..m‘\

Wears

3003

37/ 7 7 M

DR
”\

 

U B RARY '7 “336mm”
Michigan State 3
University

Physiological Ecology of Termite Gut Spirochetes

presented by

Joseph Rex Graber

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Microbiologl

 

  

 

   

jor Profe s Signature

/2?9wgflm3r
a

Date

MSU is an Manama AcdonlEqud Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/Dateouepes-sz

 

 

PHYSIOLOGICAL ECOLOGY OF TERMITE GUT SPIROCHETES
By

Joseph Rex Graber

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHEOSOPHY
Department of Microbiology and Molecular Genetics

2003

ABSTRACT
PHYSIOLOGICAL ECOLOGY OF TERMITB GUT SPIROCHETES
By

Joseph Rex Graber

T reponema Sp. strains ZAS-l and ZAS-2 were the first spirochetes to be isolated
frOm termite hindguts and are also the first known spirochetal H2/COz-acetogens. The
ZAS strains were examined for nutritional and physiological properties relevant to in situ
growth and survival. In addition to using H2+C02 as growth substrates, both strains grew
on a variety of organic compounds and were capable of mixotrophic growth (i.e. the
simultaneous use of both H2 and organic substrates). Enzyme activities of the
Wood/Ljungdahl pathway of acetogenesis were detected in both strains, whose H2
thresholds were within the range typical of acetogens (650 and 490 ppmv, respectively).
Both strains were able to maintain growth in the presence of small amounts of O; (0.5%,
vol/vol) and possessed enzyme activities which could mediate protection from 02. These
results demonstrate that T reponema strains ZAS-1 and ZAS-2 are nutritionally versatile,
perform acetogenesis by the Wood/Ljungdahl pathway, and are likely able to tolerate
oxidative stress in the partially hypoxic hindgut environment.

T reponema strain ZAS-9, an additional spirochete from Z. angusticollis hindguts,
was not a homoacetogen, and in fact produced H; as a major product during sugar
fermentation. This strain also differed from ZAS-1 and ZAS-2 in a variety of other
properties, suggesting that strains ZAS-1 and ZAS-2 be assigned to a single new species

of T reponema, whereas ZAS-9 should be considered a separate new Treponema species.

Strains ZAS-1 and ZAS-2 had requirements for folate, and important cofactor in
acetogenesis. On the notion that other termite gut microbes must supply folate in situ,
heterotrophic organisms were isolated from the guts of the termite Zootermopsis
angusticollis and screened for folate secretion. Two folate-secreting isolates (Serratia
strain ZFX-1 and Lactococcus strain ZFX-2) were identified; both strains produced a
compound that could replace the folate requirements of ZAS-1 and ZAS-2. The folate
produced by both ZF X strains was identified as folinate. These results suggest that
strains ZFX-1 and ZFX-2 are capable of supplying folinate to acetogenic spirochetes in
the termite gut, and may also be important in providing folates to other members of the
gut microbiota, as well as to the termite host.

Despite the fact that methanogenesis is the more energetically favorable process,
H2/COz-homoacetogens are the primary Hz-consumers in the guts of wood-feeding
termites. To explain this observation, I hypothesized that gut methanogens are inhibited
by pteridine compounds. The pteridine compound lumazine is an inhibitor of
methanogenesis, and pteridines are known to be unusually prevalent in insect physiology.
A variety of pteridines, as well as termite gut extracts, were tested for inhibition of
cultures of the termite methanogen Methanobrevibacterfiliformis; only lumazine showed
significant inhibitory activity. No lumazine was detected in termite hindguts by TLC and
HPLC analysis. These results indicate that gut methanogens are not inhibited by
pteridine compounds. The marginalization of methanogenesis in wood-feeding termites

must be explained by other factors.

ACKNOWLEDGEMENTS

I would never have been able to complete the research presented in this
dissertation without the support of my fiiends, family, and coworkers. First and
foremost, my appreciation goes to my advisor, Dr. John Breznak, for the guidance and
encouragement he has provided over the past five years and for teaching me what it
means to truly be a scientist. I would also like to thank my fellow graduate students,
especially Bradley Stevenson, Dan Buckley, Joel Klappenbach, Deborah Hogan, Joel
Hashimoto, Keri Byzek, Merry Riley, Stephanie Eichorst, Kristin Huizinga, John Wertz,
Dawn Greensides, Nikki Lebrasscur, and Kirsten Fertuck. Thanks to Kwi Kim for being
a super tech and for looking out for all of the "kids" in the lab (myself included). I would
also like to thank the members of my committee for their consistently solid advice: Dr.
Mike Klug, Dr. C. A. Reddy, Dr. Mark Scriber, and especially Dr. Tom Schmidt who,
owing to the "symbiotic" nature of the Schmidt and Breznak labs, has been like a second
advisor to me. I am very grateful to the Department of Microbiology & Molecular
Genetics for providing a superb educational environment and to the Center for Microbial
Ecology for providing financial support that has allowed me to present my research at a
number of scientific meetings and conferences. Finally, I would like to thank my mother,
Faye, my father, Doug, and my brother, Jerome; whatever measure of success I've

achieved personally and professionally is largely owed to them.

iv

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... vii
LIST OF FIGURES ....................................................................................................... viii
CHAPTER 1: INTRODUCTION .................................................................................... 1
Termite Gut Spirochetes: A Historical Perspective ................................................ 1
Nature & Function of the Termite Hindgut Community ........................................ 6
Physiochemical Gradients in the Termite Hindgut ............................................... 10
Acetogenesis vs. Methanogenesis in Wood-feeding Termites ............................. 12
The Spatial Resource Partitioning Hypothesis ..................................................... 17
Isolation of Termite Gut Spirochetes .................................................................... 20
Dissertation Research ........................................................................................... 25
References ............................................................................................................. 27

CHAPTER 2: PHYSIOLOGY AND NUTRITION OF T REPONEMA STRAINS
ZAS-l AND ZAS-2, Hz/COz-ACETOGENIC SPIROCHETES FROM TERMITE

HINDGUTS ..................................................................................................................... 37
Introduction ........................................................................................................... 37
Materials & Methods ............................................................................................ 38
Results ................................................................................................................... 42

Nutrition and growth of T reponema strains ZAS-1 and ZAS-2 ...................... 42
Mixotrophic growth ......................................................................................... 46
Hydrogen thresholds ........................................................................................ 46
Hz/Coz-acetogenic enzyme activity ................................................................ 49
Oxygen tolerance and oxidative stress enzymes ............................................. 49
Discussion ............................................................................................................. 53
References ............................................................................................................. 60

CHAPTER 3: TAXONOMY OF T REPONEMA STRAINS ZAS-l, ZAS-2, AND

ZAS-9 ............................................................................................................................... 66
Introduction ........................................................................................................... 66
Materials & Methods ........................................................................................... 67
Results ................................................................................................................... 68

Morphology of T reponema strains ZAS-1, ZAS-2, and ZAS-9 ................ 68
Nutrition, growth, and fermentation products of strain ZAS-9 ................ 70
G + C content of DNA .............................................................................. 70
Discussion ............................................................................................................. 73
References ............................................................................................................. 76

CHAPTER 4: INTERSPECIES COFACTOR TRANSFER SUPPORTS THE
FOLATE REQUIREMENT OF ACETOGENIC TERMITE GUT

SPIROCHETES .............................................................................................................. 77
Introduction ........................................................................................................... 77
Materials & Methods ............................................................................................ 79
Results ................................................................................................................... 86

F olate requirements of T reponema strains ZAS-1 and ZAS-2 ................. 86
Isolation of folate-secreting bacteria from termite hindguts ..................... 88
Characterization of folate-secreting strains .............................................. 88
F olate production by Serratia str. ZFX-l and Lactococcus str. ZFX- ..... 94
Discussion ............................................................................................................. 97
References ........................................................................................................... 102

CHAPTER 5: ACETOGENESIS VS. METHANOGENESIS IN THE TERMITE

HINDGUT: THE LUMAZINE HY POTHESIS ........................................................ 107
Introduction ........................................................................................................ 107
Materials & Methods .......................................................................................... 111
Results ................................................................................................................. 1 15

Methanogen inhibition assays ................................................................. 115

TLC of R. flavipes gut extracts ............................................................... 118

HPLC analysis of R. flavipes gut extracts .............................................. 118

Discussion ........................................................................................................... 120
CHAPTER 6: SUMMARY .......................................................................................... 129

vi

LIST OF TABLES

Table 1.1. Redox potentials, Gibbs free energy changes, and H2 thresholds for various

anaerobic hydrogen consuming processes ........................................................................ 14
Table 2.1. Substrate utilization by Treponema strains ZAS-1 and ZAS-2 ...................... 44
Table 2.2. Cell yields and acetate production by Treponema strain ZAS-2 growing under
unitrophic and mixotrophic conditions ............................................................................. 48
Table 2.3. Enzyme activities relevant to Hz/COz-acetogenesis in T reponema str. ZAS-l
and ZAS-2 ......................................................................................................................... 51
Table 2.4. Oxygen and peroxide detoxifying enzyme activities in Treponema str. ZAS-1
and ZAS-2 ......................................................................................................................... 54
Table 3.1. Substrates Used as energy sources by T reponema strain ZAS-9 ................... 71
Table 3.2. Fermentation products of T reponema strain ZAS-9 ....................................... 72
Table 4.1. Substrates utilization by Serratia strain ZFX-l and Lactococcus strain

ZFX-2 ................................................................................................................................ 93
Table 5.1. Thin layer chromatography of pteridines and R. flavipes gut extract ........... 119

vii

LIST OF FIGURES

Figure 1.1. Reticulitermes flavipes, a wood-feeding lower termite, pictured with a gut
extracted from a separate individual. The volume of the hindgut is approximately 1 ul.
Adapted from reference 18 ................................................................................................. 2

Figure 1.2. Phase contrast micrograph of diluted hindgut contents collected from the
termite Reticulitermes flavipes. Spirochetes, including some attached to protozoa (P), are
indicated by white arrows. Non-spirochetal prokaryotes are indicated by black arrows.
Also labeled is an undigested wood particle (W). Adapted from reference 18 ................. 4

Figure 1.3. Carbon flow during cellulose digestion in wood-feeding termites. The
thickness of arrows indicates the relative significance of the indicated microorganims in
the hindgut .......................................................................................................................... 7

Figure 1.4. Radial profiles of oxygen (0) and hydrogen (0) in an agarose-embedded
hindgut from Reticulitermes flavipes. The circular inset indicates the relative sizes of the
oxic and anoxic zones in the gut (lefi half) and the hydrogen concentration gradient (right
half). Adapted from reference 21 ..................................................................................... 11

Figure 1.5. In situ Gibb's free energies for methanogenesis (El) and acetogenesis (0)
under termite hindgut conditions. Values were calculated at 25°C for pH 7.2 and the
following substrate and product concentrations: acetate, 80 mM; methane, 1 mM;
bicarbonate, 60 mM .......................................................................................................... 16

Figure 1.6. Diagram of the termite hindgut in radial cross-section, showing the spatial
distributions of various microbial groups and gradients of 02 (blue) and H2 (red) .......... 19

Figure 1.7. Morphology of termite gut Treponema strain ZAS-l by phase contrast (a)
and transmission electron microscopy of intact (c) and transverse sectioned (b) cells. The
morphology of strain ZAS-2 is virtually identical to that of ZAS-1. Arrows indicate
periplasmic flagella, whose insertion points are subterminal (c). Scale bars = 10 pm (a)
and 0.1 pm (b, c) ............................................................................................................... 21

Figure 1.8. Phylogenetic tree inferred from 168 rDNA sequences of termite gut
treponemes (strains ZAS-1, ZAS-2, and ZAS-9), representative known spirochetes, and
spirochetal 16S rDNA clones generated directly from termite gut contents (T gut clones).
A maximum likelihood technique (fastDNAml) was used to generate the tree. The
vertical line on the right delimits a distinct subgroup (the "termite cluster”) within the
genus Treponema. Scale bar represents units of evolutionary distance and is based on
sequence divergence ......................................................................................................... 22

Figure 1.9. The Wood/Ljungdahl pathway of acetogenesis, highlighting enzymes of the
methyl and carbonyl branches. THF=Tetrahydrofolate. Adapted from reference 36 ..... 24

viii

Figure 2.1. H2 stimulation of chemoorganotrophic growth of Treponema strain ZAS-2.
Substrates added to 2YACo medium under N2/CO2 head spaces (80/20, v/v) were: 10
mM maltose (O), 16 mM H2 (A) 10 mM maltose plus 16 mM H2 (Cl), and no added
substrates (V) .................................................................................................................... 45

Figure 2.2. Mixotrophic growth of Treponema strain ZAS-2 on H2+CO2 and maltose.
Cells were grown in bottles containing 100 ml of maltose supplemented 2YACo medium
and 645 ml gas phase ........................................................................................................ 47

Figure 2.3. Determination of H2 threshold for replicate cultures (0 and Cl) of Treponema
strain ZAS-2, growing under mixotrophic conditions (H2 + CO2 plus
trimethoxybenzoate). H2 additions are indicated by arrows. The threshold value is
indicated by a dashed line .............................................................................. 50

 

Figure 2.4. Effect of 02 on the growth of Treponema strain ZAS-2. Cultures were
initially grown under 80% H2:20% CO2, with O2 additions made at 125 hours (indicated
by arrow). The resulting concentration of O2 in the culture tube headspace (percentage
by volume) was: 0 (CI), 0.5 (O), l (A), 2.5 (V), and 5 (O) .............................................. 52

Figure 3.1. Morphology of T reponema strains ZAS-2 (A & B) and ZAS-9 (C and D) by
phase contrast (A & C) and transmission electron microscopy (B & D). Insets in panels
A and C show cells of each strain, with wavelengths indicated by arrowed bars.
Periplasmic flagella are indicated by arrows in panels B & D. Strain ZAS-1 is
morphologically indistinguishable from strain ZAS-2. Bars, 5 pm (A and C), 2.5 pm (A
and C insets), and 0.1 mm (B & D). Panel B is adapted from reference 4, and is
presented here for comparison .......................................................................................... 69

Figure 4.1. Generalized structure of folate compounds (A) and pteridine portion of the
reduced form of folate compounds, tetrahydrofolate (B). Individual derivatives vary in
reduction state, C1 substitution (at N5 or N”), and number of glutamyl residues
(generally from 1 to 9) ...................................................................................................... 78

Figure 4.2. Folate utilization by Treponema strains ZAS-l and ZAS-2. All folate
compounds were provided at a final concentration of 500 ng/ml. The specific compounds
were: folic acid (El), dihydrofolate (O), tetrahydrofolate (A), S-formyltetrahydrofolate
(i.e. folinic acid) (V), and no addition (0) ....................................................................... 87

Figure 4.3. Folate secretion assay of strains ZFX-1 and ZFX-2. Putative folate secretion
is indicated by the formation of satellite colonies of the folate-requiring bioassay
organism, Enterococcus hirae, surrounding surface colonies of ZFX-l (inset A) and
ZFX-2 (inset B) ................................................................................................................. 89

ix

Figure 4.4. Phylogenetic trees inferred from 16S rDNA sequences (~1500 nt) of ZFX-1
(A), ZFX-2 (B), and related organisms. A maximum likelihood technique (fastDNAml)
was used to generate the trees. The homologous sequence from Desulfovibrio senezii
was used as an outgroup in tree A (not shown). The scale bars represent units of
evolutionary distance and are based on sequence divergence .......................................... 91

Figure 4.5. Phase contrast micrographs of Serratia strain ZFX-1 (A) and Lactococcus
strain ZFX-2 (B) ............................................................................................................... 92

Figure 4.6. HPLC chromatograms showing fluorescent (280/359 nm) compounds
present in culture filtrates of strains ZFX-1 (A) and ZFX-2 (B) and for a standard mixture
(C) of THF (peak 1, 1 ng), 5-CH3-TI-IF (peak 2, 1 ng). and S-HCO-THF (peak 3, 5 ng). In
panels A & B, dashed and solid lines indicate samples taken at the time of inoculation
and at the end of logarithmic growth, respectively. The peak eluting at 31 minutes in
panel A could not be identified, but did not correlate to any folate standard ................... 95

Figure 4.7. Growth of Treponema strains ZAS-1 and ZAS-2 in folate-free medium 2YA
supplemented with folinate (O) or culture filtrates of Serratia strain ZFX-l (O) or
Lactococcus strain ZFX-2 (Cl). Culture filtrates were added at 10% (v/v) final
concentration. F olinate was provided at 500 ng/ml final concentration. Negative controls
(A) were supplemented with 10% uninoculated GM3 medium ....................................... 96

Figure 5.1. Generalized chemical structure of pterin compounds (A) and production of
lumazine from pterin via enzymatic deamination (B) .................................................... 109

Figure 5.2. Methane production by growing cells of M. filr'formis following addition of
lumazine (O), R. flavipes gut extract (A), and anoxic water (Cl). Lumazine was provided
at a final concentration of 1 mM, and gut extract was added at 5% (vol/vol). The time of
additions is indicated by the arrow ................................................................................. 116

Figure 5.3. Methane production by growing cells of M. filiformis following addition of
biolumazine (<>), xantholumazine (A), and isoxantholumazine (V), all provided at 0.5
mM final concentrations. Additions of anoxic water (Cl) and 0.1 mM lumazine (0) were
used as controls. The time of additions is indicated by the arrow .................................. 117

Chapter 1

Introduction

Termite Gut Spirochetes: A Historical Perspective

In 1877, Joseph Leidy published his first observations on the strikingly complex
community of microbes inhabiting the intestines of the wood-feeding termite
Reticulitermesflavipes (56) (Figure 1.1). Leidy initially sought to discover the food
source of the insects by examining their gut contents. However, upon microscopic
examination of fluid collected from hindguts of worker termites, he reports:

"The brownish matter proved to be the semi—liquid food; but my

astonishment was great to find it swarming with myriads of parasites,

which indeed actually predominated over the real food in quantity.

Repeated examination showed that all individuals harbored the same

world of parasites wonderful in number, variety, and form. "
Leidy reported detailed observations on a variety of protozoa inhabiting the gut fluid and
also observed an abundant and diverse population of somewhat smaller organisms
possessing a distinctive undulate morphology and rapid motility (Figure 1.2). Initially
referred to by Leidy as Vibrio termitis, these organisms would subsequently be
recognized as spirochetes (35).

Spirochetes belong to an independent division (Spirochaetes) within the domain
Bacteria. Nearly all representatives share a set of distinctive morphological features,
including helical or undulate protoplasmic cylinder and periplasmic flagella located

within the outer cell membrane (23, 25, 43). Spirochetes are generally

1 mm

Hindgut \litlgut

 

Figure 1.1. Reticulitermes flavipes, a wood-feeding lower termite,
pictured with a gut extracted from a separate individual. The volume
of the hindgut is approximately 1 ul. Adapted from reference 18.

chemoheterotrophs, consuming carbohydrates, amino acids, long chain fatty acids, or
long chain fatty alcohols as energy substrates (24). Despite these points of similarity, the
various spirochetal species are physiologically diverse and capable of colonizing a wide
range of habitats. This is illustrated by the differences between the free-living members
of the genus Spirochaeta (5 7), most frequently found in aquatic or sedimentary
environments, and the generally host-associated genus Treponema, which form
relationships (ranging fi'om commensal to pathogenic) with a wide range of vertebrate
and invertebrate animals (67).

Termite hindguts harbor the most diverse and abundant population of spirochetes
found in any known environment. Spirochetes can account for up to 50% of the
prokaryotic cells observed in the small but densely packed termite hindgut (109-10ll total
prokaryotic cells/ml) (74) (Figure 1.2). Termite gut spirochetes are morphologically
diverse with cells ranging in length from 3 to 100 um and displaying a variety of
wavelengths or body pitches (10); individual temrite species typically harbor from 12 to
15 visibly distinguishable morphotypes (94). These cells were readily identified as true
spirochetes by the in situ observation of characteristic periplasmic flagella in tranmission
electron micrographs (3, 9, 10). Spirochetes appear to colonize the central lumen of the
hindgut almost exclusively and are only rarely observed among the dense biofilrn of
organisms on the hindgut epithelial surface (9). In the lumen, spirochetes are observed
both as free-swimming individuals and as epibionts attached to the surfaces of various
gut protozoa (26, 83-85). In the case of the protozoa Mixotricha paradoxa, regularly
arranged rows of spirochetes cover most of the cell surface. The coordinated motion of

these epibionts propels the host cell in a unique motility symbiosis (26).

 

Figure 1.2. Phase contrast micrograph of diluted hindgut contents collected from
the termite Reticulitermes flavipes. Spirochetes, including some attached to
protozoa (P), are indicated by white arrows. Non-spirochetal prokaryotes are
indicated by black arrows. Also labeled is an undigested wood particle (W).
Adapted fiom reference 18.

Despite the conspicuous abundance of spirochetes in termite guts, attempts to
isolate these organisms in pure culture met with repeated failure. To circumvent this
problem, many researchers turned to cultivation independent molecular methods to
survey the composition of the spirochetal community (1, 2, 70, 72, 74). By cloning and
analyzing spirochetal 16S rDNA genes in community DNA collected from termite
hindguts, these studies revealed that spirochetal morphological diversity in termite guts
was paralleled by a high degree of phylogenetic diversity. Lilbum et a1. (59) undertook a
comprehensive phylogenetic survey of spirochetal 16S rDNA sequences collected from
seven termite species spanning five of the seven extant termite families. All spirochetal
16s rDNA sequences collected grouped within the genus Treponema. However, nearly
all of these sequences formed a distinct cluster (the “termite cluster”) within the genus
and had only limited similarity to the 16S sequences other treponemcs (S 91%).
Curiously, the termite spirochetal sequences were most closely affiliated with
Spirochaeta stenostrepta and Spirochaeta caldaria, two fi’ee-living spirochetes that
nevertheless group with the neponemes on the basis of 16S rDNA sequences. In a single
termite species (R. flavipes), it was estimated that 26 distinct spirochetal phylotypes were
present, corresponding to at least 21 new species. FISH analysis of spirochetes in situ
with a series of specific rRNA targeted probes resulted in labeling of both free-swimming
spirochetes and protozoan epibionts, but individual phylotypes appeared to be restricted
to one of the two life-styles (66).

While these studies provided new information on spirochetal phylogeny,
definitive information on the physiological properties of the termite gut spirochetes

remained elusive. However, the observation that spirochetes were consistently present in

seemingly healthy termites suggested a non-pathogenic or even beneficial relationship.
This hypothesis was tested by Eutick et al. (40), using antibiotic treatments to selectively
eliminate spirochetes from the guts of termites. While these treatments led to a reduction
in termite vitality, it was impossible to determine if this was the direct result of spirochete
elimination (as opposed to the loss of some other non-spirochetal member of the
community) and if so, what specific benefits the host had derived from the spirochetal
population. Until a representative could be isolated in pure culture, the role of

spirochetes in the termite hindgut would remain a matter of speculation.

Nature & Function of the Termite Hindgut Community

Studies of the termite hindgut symbiotic system have most frequently focused on
two problems posed by the utilization of wood as a primary food source by termites:
First, how is this relatively refractory biopolymer converted to a metabolically useful
source of carbon and energy? Second, how do termites thrive on a nutrient source
containing low (0.05%) nitrogen? Accordingly, the roles of the hindgut microbial
community in the digestion of lignocellulose and provision of carbon, energy, and
nitrogen to their termite hosts evolved as central areas of research interest.

Carbon flow in the hindgut of lower (i.e. evolutionarily basal termites comprising
six of the seven termite families) wood-feeding termites is illustrated in Figure 1.3.
Cellulose hydrolysis occurs by the combined actions of termite cellulases secreted fi'om
the salivary glands (45, 96, 97) and cellulases of anaerobic protozoa (69, 98-100), the
latter of which ferment the liberated glycosyl units within phagosomes by anaerobic

protozoa residing within the hindgut:

 

Cellulose

Termite cellulases & Protozoa

i

Soluble Glycosides
l

Lactic Acid Bacteria (?)

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Protozoa & Protozoa &
Other Gut Microbes Lactate Other Gut Microbes
Unknown
Acetate < H2 + 002
I Acetogenic Bacteria

Termite Respiration Methanogenic Archaea

 

 

 

C02 + H20 CH4

 

 

 

 

 

 

Figure 1.3. Carbon flow during cellulose digestion in wood-feeding termites.
The thickness of arrows indicates the relative significance of the indicated
microorganims in the hindgut.

l.) [C6H1005]n + 3n H2O —-> 2n CH3COOH + 2n CO2 + 4n H2
The H2 and CO2 produced can serve as substrates for homoacetogenic bacteria or
methanogenic archaea, which produce acetate or methane, respectively:
2.) 4 H2 + 2 CO2 —) CH3COOH + 2 H20
3.) 4H2+CO2—)CH4+2H2O

Acetate is absorbed through the gut epithelium and oxidized to fuel the insect’s energy
metabolism (42), while methane and unused H2 passively diffuses out of the gut. In the
context of the symbiosis, methanogens could be viewed as parasites that consume
reducing equivalents that could otherwise be directed towards acetate production.

Acetate dominates the hindgut pool of volatile fatty acids, accounting for 94-98
mol% of all VFAs at concentrations ranging 58 to 81 mM (68). Oxidation of microbially
produced acetate was estimated to satisfy between 71 and 100% of the host termite’s
energy requirements (68). Given this observation, and the fact that methanogenesis is
severely limited in the termite hindgut (see below), carbon flow in the hindgut was
thought to be dominated by reactions 1 and 2, with bacterial acetogenesis accounting for
up to 33% of total acetate production (15). However, recent studies using microinjection
of radiolabelled substrates into intact guts have revealed that up to one-third of the total
carbon flux passes through a previously unrecognized pool of lactate. The low standing
concentration of lactate measured in the gut reflects the rapid turnover of this substrate to
acetate (93). This observation has revised the estimated carbon flux accounted for by
H2/CO2—acetogenesis to 10.5% of the total.

Although the polysaccharide components of lignocellulose are very efficiently

digested during passage though the hindgut (74-99% of cellulose, 65-87% hemicellulose)

(38), high-molecular-weight core lignin does not appear to be significantly degraded in
wood-feeding termites (16, 32, 38, 44). There is, however, evidence for some
modification of lignin during gut passage via demethylation of the aromatic moieties of
lignin sidechains (38), and a number of organisms capable of degrading aromatic lignin
monomers have been isolated from termite hindguts (20, 51, 92). The contribution of
these lignin-consuming activities to termite nutrition is currently unknown, but is likely
to be relatively minor in comparison to cellulolytic activities.

Hindgut microbes also play important roles in the nitrogen economy of termites
(90). Reduction of N2 to ammonia (NH3) via the enzyme nitrogenase is a property unique
to prokaryotes, and any N2 fixation measured in termites (either by the acetylene
reduction assay or incorporation of l5N2 isotope) is thus attributable to bacteria or
archaea. N2 fixation has been detected and quantified in a large number of different
termite species (91), and a number of N2-fixing bacteria have been isolated from termites
(41, 52, 77). A highly diverse set of nifH genes (encoding dinitrogenase reductase) have
been detected in the hindguts a number of termite species (71, 73), and direct
amplification of nif genes from community mRN A allowed an elegant assessment of
which phylogenetic classes of nifl-I homologues were being expressed in the gut of the
termite Neotermes koshunensis (65). In addition to providing newly fixed nitrogen to
termites, members of the hindgut community act to prevent the loss of fixed nitrogen by
recycling the nitrogen present in the excretory product uric acid. Potrikus and Breznak
(78-81) demonstrated that uric acid is transported to the hindgut via the Malpighian
tubules and isolated a variety of bacteria capable of anaerobically degrading uric acid

with the liberation of ammonia (which is presumably reabsorbed by the termite).

Physiochemical Gradients in the Termite Hindgut

Early studies of redox conditions existing within termite hindguts indicated an
effectively anoxic environment (5, 6). While this result was in keeping with the presence
of strict anaerobes in the hindgut, such as cellulolytic protozoa and methanoarchaea, a
number of other observations suggested a more complex oxygen status. In cultivation
studies of gut bacteria, a large number of the isolates were observed to be aerotolerant,
facultatively aerobic (39, 87), or even strictly aerobic (92). Moreover, the ready
mineralization of lignin monomers fed to termites was shown to be dependent on hindgut
bacteria and the presence of O2, and a variety of bacteria isolated fiom hindguts were
shown to be capable of O2-dependent in vitro degradation of aromatic monomers of
lignin (20, 51, 92).

An elegant series of studies by Brune and coworkers (19, 21, 37) addressed the
issue of O2 in termite guts by making fine scale microelectrode measurements of the in
situ physiochemical characteristics of termite guts. These studies revealed a highly
structured environment with distinct radial gradients of H2 and 02 concentration (Figure
1.4). 02 partial pressures were highest near the gut epithelium and rapidly decreased to
non-detectable levels within 20-200 pm of the periphery. Conversely, hydrogen partial
pressure was very high (approx. 50 mbar, or 50,000 ppmv) in the central luminal portion
of the gut, and decreased steeply towards the gut periphery. Two main zones of H2
consumption were observed; one in the center of the lumen, and a second directly at the
gut epithelium. The hindguts of both lower and higher termites had pH values of around

7. These results provided convincing new evidence that as much as 60% of the

10

Partial pressure of gases (m bar)

60

l

80
i

l J

 

 

Depth (um)

 

1200

1500

 

 

? Agarose

 

Gut

Figure 1.4. Radial profiles of oxygen (0) and hydrogen (0) in an agarose-
embedded hindgut from Reticulitermes flavipes. The circular inset indicates
the relative sizes of the oxic and anoxic zones in the gut (left half) and the
hydrogen concentration gradient (right half). Adapted from reference 21.

11

gut volume is in fact hypoxic in character and also demonstrated that the luminal region
of termite hindguts is among the most H2-rich environments observed in nature (62).
Anaerobiosis in the gut lumen appears to be maintained largely by respiratory

activity of gut bacteria, and elimination of these organisms with antibiotic treatments
renders the gut fully oxic (95). This observation fits well with the epithelial zone of O2
consumption observed by Brune et al. (19) (Figure 1.4). High rates of O2-dependent
acetate oxidation have been observed in the gut (93), suggesting that acetate produced in
the central lumen may be one of the major substrates of O2-consuming epithelial biofilrn.
While this acetate consumption would represent a loss of oxidizable substrate for the
host, acetate oxidizers (and other O2-consuming bacteria) could be viewed as playing an
essential symbiotic role in maintaining conditions of anoxia in the lumen, which favors

the continued homoacetic fermentation of cellulose.

Acetogenesis vs. Methanogenesis in Wood-feeding Termites

A significant role for H2-consuming, CO2-reducing homoacetogens in termite
hindguts was first suggested by the dominance of acetate in the hindgut VFA pool and
the observation of relatively low rates of H2 and CH4 emission from termites (68). It was
also observed that H2 and CH4 emission rates were significantly increased in termites fed
antibacterial drugs (68). These observations suggested that H2 consumption in the
hindgut was dominated by bacterial homoacetogens. This hypothesis was confirmed by
the demonstration of H2-dependent fixation of 14CO2 to acetate by termite gut
homogenates and the inhibition of this activity by antibacterial drugs (11). Brauman et

al. (7) found that acetogens out-process methanogens as H2-consumers in the majority of

12

wood-feeding termite species examined (although the opposite was true in termites from
other feeding guilds, such as the soil-feeders). This situation, while unusual, is highly
beneficial to the termite due to the significant contribution of acetogens to termite
nutrition (15, 93).

The dominance of acetogenesis over methanogenesis in the termite hindgut is a
puzzling situation. In anoxic habitats, various metabolic classes of organisms (sulfate-
reducing bacteria, methanogens, H2/CO2-acetogens, etc) compete for reducing potential,
usually in the form of H2 produced by ferrnentative organisms. For a given electron
donor (in this case, H2) thermodynamics suggests that the group of organisms using the
most exergonic terminal electron-accepting metabolic process (i.e. the process with the
most positive redox potential, E°') should dominate other competitors, as implied by the
equation: AG°' = -n F AE°' [n = number of e' transferred, F = Faraday constant, AE°' = E°'
redox compound - E°' (H+/H2) ]. This correlates to progressively lower minimum
thresholds values of H2 (33). The minimum H2 threshold is the H2 concentration at which
an organism's metabolic reaction is still sufficiently exergonic to produce a useful amount
of energy. This amount, referred to as the critical Gibbs free energy (AG), is the amount
of energy required to translocate 1 mol of protons through a full energized cell
membrane, which in turn correlates to the minimum energy quanta sufficient to drive the
synthesis of ATP (approximately -23 kJ/mol substrate, or 1/3 of an ATP) (30, 31, 33).
Thus, sulfate-reducing bacteria (average E°’= -217 mV) are able to lower the H2
concentration to a point at which methanogens (average E°’= -238) can no longer derive

useful energy from their metabolism. Table 1.1 provides a list of Gibbs fiee energy

13

Table 1.1. Redox potentials, Gibbs free energy changes, and H2 thresholds for various

anaerobic hydrogen consuming processes”.

 

 

E°' b AG°'/reaction H2 Thresholdc

Metabolic Process: (mV) (kJ) (ppmv)
Acetogenesis:

4 H2 + 2 HCO;' + H+ -> CH3COO' + 4 H20 -279 -104.4 260 - 950
Methanogenesis:

4 H2 + HCO3' + H+ - CH4 + 3 H20 -238 -135.6 25 - 100
Sulfate Reduction:

4H2+SO.2'+H*-> HS'+4H2O -217 452.0 8-16

 

‘Adapted from values presented in reference (33).

l’E°' of the couples accepting electrons fiom H2 calculated relative to the redox potential

0sz (414 mV)

cRange of H2 threshold values observed for representatives of each of the metabolic types

grown in pure culture.

14

values, redox potentials, and H2 thresholds of some typical groups of hydrogen
consuming anaerobes. This model of competition is referred to as the threshold model
(29, 31, 86), and is generally in good agreement with actual H2 concentrations measured
in natural habitats in which the dominant H2-consuming process was known (27, 49, 62,
63). Methanogens, as would be predicted, dominate as the terminal electron sink
organisms in habitats in which CO2/I-ICO3' is the main terminal electron acceptor,
including gastrointestinal systems such as bovine rumen (4, 13, 34, 64, 82).

In the termite hindgut, CO2 appears to be the primary electron acceptor. Under these
conditions, methanogens should out-process acetogens due to the more positive E°’ value
of the CO2/CH4 redox couple, which equates to a more exergonic reaction and lower H2
threshold (see Table 1). This is certainly the case for most other gastrointestinal systems,
including the bovine rumen (4, 13, 34, 64, 82). Figure 1.5 shows free energy values of
acetogenesis and methanogenesis calculated for pH levels and concentrations of acetate,
bicarbonate, and methane typical of the termite gut environment and are calculated across
the range of radial hydrogen concentrations measured in the termite hindgut. While
methanogenesis is clearly the more favorable reaction at all H2 concentrations, acetogens
appear to effectively dominate H2 consumption in the termite hindgut.

A variety of homoacetogenic bacteria have been isolated from several species of
termites, including members of the genera Sporomusa, Acetonema, and Clostridium (12,
47, 48). These organisms, however, did not seem to colonize the hindgut at population
densities sufficient to support measured rates of H2/CO2 acetogenesis (15) and molecular
. phylogenetic studies of hindgut microbial communities of other termite species have

never detected phylotypes closely related to the isolated termite acetogens (70, 72).

15

-120-

-110-
E ..
g -100-
o
m ..
9
2 -90-
3
2 '80“
(>6 .
(D -70-
<1 .
E
17, -60-
s. 1

-50e

-40-

'30 l I I I l I I I I I

 

 

1o 20 30 4o 50
Hydrogen partial pressure (mbar)

Figure 1.5. In situ Gibb's free energies for methanogenesis (El) and
acetogenesis (0) under termite hindgut conditions. Values were
calculated at 25°C for pH 7.2 and the following substrate and product
concentrations: acetate, 80 mM; methane, 1 mM; bicarbonate, 60

mM.

16

Furthermore, the H2 thresholds of the isolated acetogens (251-380 ppmv) were
significantly higher than those of the cultured termite methanogens (36-45 ppmv) (17).
In coculture competitions for limited H2, the termite acetogen Sporomusa termitida was
consistently dominated by the termite methanogen Methanobrevibacter cuticularis (17).
Given these observations, it was considered doubtful that the cultured termite acetogens
represented the dominant acetogens in the termite hindgut.

It is unknown what factors in the termite gut favor acetogenesis over the more
energetically favorable process of methanogenesis. There are other habitats in which
acetogens dominate methanogens, including mildly acidic (76) or carbon limited (46)
freshwater sediments. Aceto genesis is also favored over methanogenesis at low
temperatures (28, 88, 89). None of these conditions, however, would appear to be
relevant to the termite hindgut habitat. Mixotrophic growth by the simultaneous
utilization of organic substrates and H2 has also been proposed to increase energy yields
and competitive ability of acetogens (50, 60, 61), and mixotrophic growth has been
demonstrated in the termite acetogen Sporomusa termitida (14). Hydrogen thresholds of
acetogens growing mixotrophically are indeed lower than those measured under
unitrophic growth, but are still significantly higher than those characteristic of
methanogenesis (75), making the significance of mixotrophy in acetogen/methanogen

competitions somewhat debatable.

The Spatial Resource Partitioning Hypothesis

The puzzling dominance of acetogenesis over methanogenesis in the termite hindgut

has led to speculation that direct competition for H2 does not actually occur between the

17

two groups. This was first suggested by the localization of methanogens in the hindgut of
the termite R. flavipes: F420 autofluorescent cells (a diagnostic characteristic of
methanogens) were rarely observed in the luminal gut fluid and instead appeared to
colonize the gut epithelium almost exclusively (53, 54). Considered in light of the
physiochemical gradient data of Ebert and Brune (37) and Brune et al. (19), this
observation suggests an alternative model of H2 consumption in termite guts referred to

as the spatial resource partitioning hypothesis (22, 37).

In this model, methanogens and acetogens occupy distinct physiochemical niches
within the hindgut, precluding any direct competition for H2 (see Figure 1.6).
Methanogens appear to be restricted to the region of lowest hydrogen concentration on
the gut epithelium, where they may be additionally inhibited by inwardly diffirsing 02. If
acetogens were to inhabit the central lumen, they would have access to hydrogen
produced by gut protozoa at levels 100 to 1000 fold higher than their minimum
thresholds (37). This would be consistent with the two regions of H2 consumption (one
in central lumen, and one near the gut epithelium) observed in the lumen by Ebert and
Brrme (37) (Figure 1.4).

As promising as this hypothesis may seem, it is not without shortcomings. First,
it offers no explanation for the counterintuitive restriction of methanogens to the gut wall,
where substrate H2 is scarce and inhibitory O2 is abundant. Also, in freshly prepared gut
homogenates (in which all microbial spatial associations have been effectively disrupted),
H2-dependent acetogenesis from CO2 still consistently dominates methanogenesis (11).

This would imply that other, as yet unknown, factors also contribute to the dominance of

18

 

Bacteria

Cellulolytic Protozoans

Figure 1.6. Diagram of the termite hindgut in radial cross-section, showing
the spatial distributions of various microbial groups and gradients of 02
and H2.

homoacetogens over methanogens as H2-consumers in the hindguts of wood-feeding

termites.

Isolation of Termite Gut Spirochetes

In 1999, years of periodic but persistent cultivation efforts were finally rewarded
with the first successful enrichment and isolation of spirochetes fi'orn the hindgut of the
dampwood termite Zootermopsis angusticollis (55). A number of factors contributed to
the success of these enrichments: the medium was low in fermentable carbon sources,
was supplemented with rumen fluid, and was incubated under an anoxic H2/CO2
atmosphere. The enrichment medium was also supplemented with rifamycin and
phosphomycin (antibiotics to which spirochetes are resistant) and bromoethanosulfonate
(a selective inhibitor of methanogens). The development of this enrichment medium was
the result of patient microscopic observation of differential spirochetal growth in a
variety of exploratory media of varying constitutions over 2-3 months (owing to the slow
growth of the termite gut spirochetes) (17, 18).

Of the dozen initially isolated spirochete strains, two (strains ZAS-1 and ZAS-2)
grew to sufficient densities to permit physiological studies (55). The two strains are
similar in size (0.2 pm by 3 to 7 urns) and display the characteristic morphological
features of spirochetes (i.e. undulate cell shape and periplasmic flagella) (Figure 1.7).
The 168 rDNA sequences of ZAS-1 and ZAS-2 are 98% similar to each other and, like
other termite gut spirochetes, group within the so-called "termite cluster" of the genus
T reponema (Figure 1.8). The most similar l6S rDNA sequences among cultivated

spirochetes were those of S. stenostrepta and S. caldaria (92-93% similarity). Both

20

 

Figure 1.7. Morphology of termite gut Treponema strain ZAS-1 by phase
contrast (a) and transmission electron microscopy of intact (c) and transverse
sectioned (b) cells. The morphology of strain ZAS-2 is virtually identical to that
of ZAS-1. Arrows indicate periplasmic flagella, whose insertion points are
subterminal (c). Scale bars = 10 pm (a) and 0.1 pm (b, c).

21

T Gut Clone RFS3 '

   
 
   
  
  
  
   
  

T Gut Clone RF825
ZAS-1
ZAS-2
T Gut Clone ZASBQ 'Tennite
ZAS-9 Cluster"

T Gut Clone NL-1
Spirochaeta stenostrepta
Spirochaeta caldan'a
Treponema pallidum
Treponema denticola
Spirochaeta zuelzerae
I I Treponema succinifaciens
Treponema bryantii
—-I Treponema pectinovorum
—I Spirochaeta halophila
Spirochaeta aurantia
Bomelia burgdorfen’
I Cn'stispira pectinis
I I Leptonema illini
, Leptospira biflexa
I Brachyspira hyodysenten’ae
I Brevinema andersonii

 

 

 

 

 

 

 

 

 

 

I Escherichia coli
0.1 0

 

Figure 1.8. Phylogenetic tree inferred from 168 rDNA sequences of termite
gut treponemes (strains ZAS-1, ZAS-2, and ZAS-9), representative known
spirochetes, and spirochetal 16$ rDNA clones generated directly from termite
gut contents (T gut clones). A maximum likelihood technique (fastDNAml)
was used to generate the tree. The vertical line on the right delimits a distinct
subgroup (the "termite cluster“) within the genus Treponema. Scale bar
represents units of evolutionary distance and is based on sequence
divergence.

22

strains required anoxic conditions and provision of yeast autolysate and an ll-cofactor
mixture in the growth media (55).

The most interesting property of ZAS-1 and ZAS-2, however, was their ability to
carry out H2/CO2 acetogenesis. This capability was first suggested by the consumption of
headspace H2 and CO2 in culture tubes with the coincident production of stoichiometric
amounts of acetate as the sole end product. Subsequently, it was shown that pure cultures
of strains ZAS-1 and ZAS-2 incorporated l4CO2 into both C atoms of acetate. To further
test for the presence of the characteristic Wood/Ljungdahl pathway of acetogenesis
(Figure 1.9) (36), both strains were successfully assayed for the presence of three
characteristic enzyme activities (carbon monoxide dehydrogenase, formate
dehydrogenase, and hydrogenase) of the pathway. While the possibility that termite
hindgut spirochetes were acetogens had been hypothesized many years ago (8), the
confirmation of this metabolic activity in ZAS-1 and ZAS-2 was still something of a
surprise, as no other spirochete had ever been shown to be capable of H2/CO2
acetogenesis.

Given their abundance in the hindgut microbial community, the observation of
acetogenesis in the ZAS strains suggests that spirochetes may be the dominant acetogens
in the termite hindgut. This hypothesis is supported by the primarily luminal localization
of spirochetes, which places them among (or attached to) H2-producing protozoa and
within the luminal zone of H2-consumption observed by Ebert et al. (37). This
interpretation, however, should be viewed with a note of caution. It is presently unknown
what percentage of spirochetes in the termite hindgut are actually acetogens and whether

acetogenic spirochetes exist in termites other than Z. angusticollis. An

23

Methyl Branch Carbonyl Branch
C02 002

 

Formate Dehydrogenase 21H]
HCOOH
Formyl-THF Synthetase THF’ ATP 2 [H] \I
ADP
HCO-THF CO Dehydrogenase
Methenyl-THF Cyclohydrolase 4 [H]
Methylene-THF Dehydrogenase
Methylene-THF Reductase
CH3-THF
Methyltranserase CO Dehydrogenase v
CH3-corrinoid [CO]
CoA
CH3-CO-S-CoA - - +cell carbon
CoA Pi
CH3-CO-P
ADP
ATP
CH3COOH

Figure 1.9. The Wood/Ljungdahl pathway of acetogenesis, highlighting
enzymes of the methyl and carbonyl branches. THF=Tetrahydrofolate.
Adapted from reference 36.

24

additional spirochete subsequently isolated from Z. angusticollis, T reponema strain ZAS-
9 (more thoroughly described in Chapter 3), is not capable of H2/CO2-acetogenesis. This
observation suggests that the morphological and phylogenetic diversity observed in
termite-associated spirochetes is likely paralleled by a similarly high degree of
physiological diversity. In addition to their role in the provision of acetate to their hosts,
termite spirochetes are likely to perform a variety of other functions in the termite
hindgut.

In keeping with this hypothesis, strain ZAS-9 was recently determined to be
capable of N2 fixation (58), another property not previously observed in spirochetes.
Strains ZAS-1, ZAS-2, and ZAS-9 each possess two distinct nrfl-I homologs, several of
which were nearly identical to nifH sequences known to be expressed in termite hindguts
(58, 65). Nitrogen fixation was unambiguously demonstrated in ZAS-9 by the acetylene
reduction assay, by the fixation of 15N2 into cellular 15N, and by growth with N2 as the
primary nitrogen source. ZAS-1 and ZAS-2, however, showed only low levels of
nitrogenase activity and no enhancement of nitrogen limited growth by the provision of
N2. These results strongly suggest that distinct populations of spirochetes make
significant contributions to host nutrition both in terms of carbon (via acetogenesis) and

nitrogen provision.

Dissertation Research
The research presented in this dissertation focuses on the physiological ecology of termite

gut spirochetes. Chapter 2 presents a more thorough physiological characterization of

25

strains ZAS-l and ZAS-2, with special attention focused on those properties relevant to
survival within the termite hindgut and their role as symbiotic homoacetogens. Chapter 3
provides further description of strain ZAS-9, explores the taxonomic and genomic
properties of all three ZAS strains, and proposes Latinate species epithets. Chapter 4
examines the requirement of strains ZAS-1 and ZAS-2 for the vitamin folate (a critical
cofactor in acetogenesis) and the role of other members of the hindgut microbiota in
providing this factor in situ. Chapter 5 revisits the issue of acetogenesis vs.
methanogenesis in termite hindguts, and tests an alternative to the spatial resource
partitioning hypothesis. Finally, chapter 6 summarizes the main conclusions of these

studies.

26

References

Berchtold, M., W. Ludwig, and H. Konig. 1994. 16S rDNA sequence and
phylogenetic position of an uncultivated spirochete fi'om the hindgut of the

termite Mastotermes darwiniensis Froggatt. FEMS Microbiology Letters
123:269-273.

Berchtold, M., and H. Konig. 1996. Phylogenetic analysis and in situ
identification of uncultivated spirochetes from the hindgut of the termite
Mastotermes darwiniensis. Syst Appl Microbiol 19:66-73.

Bermudes, D., D. Chase, and L. Margulis. 1988. Morphology as a basis for
taxonomy of large spirochetes symbiotic in wood-eating cockroaches and
termites. Int J Syst Bacteriol 38:291-302.

Bernalier, A., M. Lelait, V. Rochet, J.-P. Grivet, G. R. Gibson, and M.
Durand. 1996. Acetogenesis from H2 and CO2 by methane and non-methane-
producing human colonic bacterial communities. F ES Microbial. Ecol. 19:193-
202.

Bignell, D. E. 1980. Determination of pH and oxygen status in the guts of lower
and higher termites. J Insect Physiol 26: 183-188.

Bignell, D. E. 1984. Direct potentiometric determination of redox potentials of
the gut contents in the termites Zootermopsis nevadensis and Cubitermes severus
and in three other arthropods. J Insect Physiol 30: 169-174.

Brauman, A., M. D. Kane, M. Labat, and J. A. Breznak. 1992. Genesis of
acetate and methane by gut bacteria of nutritionally diverse termites. Nature
257:1384-1387.

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

Breznak, J. A., and H. S. Pankratz. 1977. In situ morphology of gut microbiota
of wood-eating termites [Reticulitermesflavipes (Kollar) and Coptotermes
formosanus Shiraki). Appl Environ Microbiol 33:406-426.

27

10.

ll.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Breznak, J. A. 1984. Hindgut spirochetes of termites and Cryptocercus
punculatus, p. 67-70. In N. R. Krieg and J. G. Holt (ed.), Bergey's Manual of
Systematic Bacteriology, vol. 1. Williams & Witkins, Baltimore.

Breznak, J. A., and J. M. Switzer. 1986. Acetate synthesis fi'om H2 plus CO2 by
termite gut microbes. Appl Environ Microbi0152:623-630.

Breznak, J. A., J. M. Switzer. and H. J. Seitz. 1988. Sporomusa termitida sp.
nov., an H2/CO2-utilizing acetogen isolated from termites. Arch Microbiol
150:282-288.

Breznak, J. A., and M. D. Kane. 1990. Microbial H2/CO2 acetogenesis in animal
guts: nature and nutritional significance. FEMS Microbiol. Rev.:309-314.

Breznak, J. A., and J. S. Blum. 1991. Mixotrophy in the termite gut acetogen,
Sporomusa termitida. Arch Microbiol 156: 105-1 10.

Breznak, J. A. 1994. Acetogenesis fiom carbon dioxide in termite guts. In H. L.
Drake (ed.), Acetogenesis. Chapman & Hall, New York.

Bro-alt, J. A., and A. Brune. 1994. Role of microorganisms in the digestion of
lignocellulose by termites. Annu Rev Entomol 39:453-487.

Breznak, J. A. 2000. Ecology of prokaryotic microbes in guts of wood- and
litter- feeding termites, p. 209-231. In T. Abe, D. E. Bignell, and M. Higashi (ed.),
Termites: Evolution, Sociality, Symbioses, Ecology. Kluwer Academic,
Dordrecht/Norvvell, MA.

Breznak, J. A., and J. R. Leadbetter. 2002. Termite Gut Spirochetes. In M.
Dworkin (ed.), The Prokaryotes: an evolving electronic resource for the
microbiological community. Springer-Verlag, New York.

Brune, A., D. Emerson, and J. A. Bro-alt. 1995. The termite gut microflora as
an oxygen sink: rrricroelectrode determination of oxygen and pH gradients in guts
of lower and higher termites. Appl Environ Microbiol 61 :2681-2687.

Brune, A., E. Miambi, and J. A. Breznak. 1995. Roles of oxygen and the
intestinal microflora in the metabolism of lignin-derived phenylpropanoids and

28

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

other monoaromatic compounds by termites. Appl Environ Microbiol 61 :2688-
2695.

Brune, A. 1998. Termite guts: the world's smallest bioreactors. Trends
Biotechnol 16: 16-21 .

Brune, A., and M. Friedrich. 2000. Microecology of the termite gut: structure
and function on a microscale. Curr Opin Microbiol 3:263-269.

Canale-Parola, E. 197 7. Physiology and evolution of Spirochetes. Bacteriol Rev
41:181-204.

Canale-Parola, E. 1984. The Spirochetes, p. 38-70. In N. R. Krieg and J. G. Holt
(ed.), Bergey's Manual of Systematic Bacteriology, 9th ed, vol. 1. Williams &
Wilkins, Baltimore.

Charon, N. W., E. P. Greenberg, M. B. H. Koopman, and R. J. Limberger.
1992. Spirochete Chemotaxis, Motility, and the Structure of the Spirochetal
Periplasmic Flagella. Res Microbiol 143:597-603.

Cleveland, L. R., and A. V. Grimstone. 1964. The fine structure of the flagellate
Mixotricha paradoxa and its associated microorganisms. Procedings of the Royal
Society of London Series B 159:668-86.

Conrad, R., B. Schink, and T. J. Phelps. 1986. Thermodynamics of H2-
consuming and H2 producing metabolic reactions in diverse methanogenic
environments under in situ conditions. FEMS Microbial. Ecol. 38:353-360.

Conrad, R., and B. Wetter. 1990. Influence of temperature on energetics of
hydrogen metabolism in homoacetogenic, methanogenic, and other anaerobic
bacteria. Arch Microbiol 155:94-98.

Conrad, R. 1995. Soil microbial processes involved in production and
consumption of atmospheric trace gases, p. 207-250, Advances in Microbial
Ecology, Vol 14, vol. 14.

Conrad, R. 1996. Soil microorganisms as controllers of atmospheric trace gases
(H2, CO, CH4, OCS, N20, and NO). Microbiol Rev 60:609-640.

29

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

Conrad, R. 1999. Contribution of hydrogen to methane production and control of
hydrogen concentrations in methanogenic soils and sediments. FEMS Microbiol
Ecol 28:193-202.

Cookson, L. J. 1987. I“C-Lignin degradation by three Australian termite species.
Wood Sci.Technol. 21:11-25.

Cord-Ruwisch, R., H. Seitz, and R. Conrad. 1988. The capacity of
hydrogenotrophic anaerobic bacteria to compete for traces of hydrogen depends
on the redox potential of the terminal electron acceptor. Arch. Microbiol.
149:350—7.

De Graeve, K. G., J. P. Grivet, M. Durand, P. Beaumatin, C. Cordelet, G.
Hannequart, and D. Demeyer. 1994. Competition between reductive

acetogenesis and methanogenesis in the pig large-intestinal flora. J. Appl.
Biotechnol. 76:55-61.

Dobell, C. 1912. Research on the spirochetes and related organisms. Archiv firr
protistenkunde 26:119-230.

Drake, H. L. 1994. Acetogenesis, acetogenic bacteria, and the acetyl-CoA
"Wood/Ljundahl" pathway: Past and current perspectives, p. 1-60. In H. L. Drake
(ed.), Acetogenesis. Chapman and Hall, New York.

Ebert, A., and A. Brune. 1997. Hydrogen concentration profiles at the oxic-
anoxic interface: a microsensor study of the hingut of the wood-feeding lower
termite Reticulitermesflavipes. Appl Environ Microbiol 63:4039-4046.

Esenther, G. R., and T. K. Kirk. 1974. Catabolism of aspen sapwood in
Reticulitermesflavipes. Ann Entomol Soc Am 67:989-991.

Eutick, M. L., R. W. O'Brien, and M. Slaytor. 1978. Bacteria from the gus of
Australian termites. Appl Environ Microbiol 35:823-828.

Eutick, M. L., P. Veivers, R. W. O'Brien, and M. Slaytor. 1978. Dependence
of the higher termite, Nasutitermes exitiosus and the lower termite, Coptotermes
lacteus on their gut flora. J Insect Physiol 24:363-368.

30

41.

42.

43.

45.

46.

47.

48.

49.

50.

French, J. R. J., G. L. Turner, and J. F. Bradbury. 1976. Nitrogen fixation by
bacteria from the hindguts of termites. J Gen Microbiol 95:202-206.

Hogan, M. E., M. Slaytor, and R. W. Obrien. 1985. Transport of volatile fatty
acids across the hindgut of the cockroach Panesthia cribrata Saussure and the
termite, Mastotermes darwiniensis Froggatt. J Insect Physiol 31:587-591.

Holt, S. C. 1978. Anatomy and chemistry of spirochetes. Microbiology Reviews
42:114-160.

Hyodo, F., J. Azuma, and T. Abe. 1999. Estimation of the effect of passage
through the gut of a lower termite, Coptotermesformosanus Shiraki, on lignin by
solid-state CP/MAS 13C NMR. Holzforschung 53:244-246.

Inoue, T., K. Murashima, J. I. Azuma, A. Sugimoto, and M. Slaytor. 1997.
Cellulose and xylan utilisation in the lower termite Reticulitermes speratus. J
Insect Physiol 43:235-242.

Jones, J. G., and B. M. Simon. 1985. Interaction of acetogens and methanogens
in anaerobic freshwater sediments. AEM 49:944-948.

Kane, M. D., A. Brauman, and J. A. Bro-alt. 1991. Clostridium mayombei Sp-
Nov, an H2/CO2 acetogenic bacterium fi’om the gut of the African soil-feeding
termite, Cubitermes speciosus. Arch Microbiol 156:99-104.

Kane, M. D., and J. A. Breznak. 1991. Acetonema longum Gen-Nov Sp-Nov, an
H2/CO2 acetogenic bacterium from the termite, Pterotermes occidentis. Arch
Microbiol 156:91-98.

Kramer, H., and R. Conrad. 1993. Measurement of dissolved H2 concentrations
in methanogenic environments with a gas-diffusion probe. FEMS Microbiol Ecol
12:149-158.

Krumholz, L. R., S. H. Harris, S. T. Tay, and J. M. Sulflita. 1999.
Characterization of two subsurface H2-utilizing bacteria, Desulfomicrobium
hypogeium sp. nov. and Acetobacterium psammolithicum sp. nov., and their
ecological roles. Appl Environ Microbiol 65:2300-2306.

31

51.

52.

53.

S4.

55.

56.

57.

58.

59.

60.

Kuhnigk, T., E.-M. Borst, A. Ritter, P. Kampfer, A. Graf, H. Hertel, and H.
Konig. 1994. Degradation of lignin monomers by the hingut flora of xylophagous
termites. Syst Appl Microbiol 17:76-85.

Kuhnigk, T., J. Branke, D. Krekeler, H. Cypionka, and H. Konig. 1996. A
feasible role of sulfate-reducing bacteria in the termite gut. Syst Appl Microbiol
19:139-149.

Leadbetter, J. R., and J. A. Breznak. 1996. Physiological ecology of
Methanobrevibacter cuticularis sp. nov. and Methanobrevibacter curvatus sp.
nov., isolated from the hindgut of the termite Reticulitermesflavipes. Appl
Environ Microbiol 62:3620-3631.

Leadbetter, J. R., L. D. Crosby, and J. A. Breznak. 1998. Methanobrevibacter
filiformis sp. nov., a filamentous methanogen fi'om termite hindguts. Arch
Microbiol 169:287-292.

Leadbetter, J. R., T. M. Schmidt, J. R. Graber, and J. A. Bro-alt. 1999.
Acetogenesis fi'om H2 plus CO2 by spirochetes from termite guts. Science
283:686-689.

Leidy, J. 1874-1881. The parasites of the termites. J. Acad. Nat. Sci. (Phil)
8:425-447.

Leschine, S., B. J. Faster, and E. Canale-Parola 2002, posting date. Free-
Living Saccharolytic Spirochetes: The Genus Spirochaeta. The prokaryotes: an
evolving electronic resource for the microbiological community. Dworkin, M.
(ed.). [Online.]

Lilburn, T. C., K. S. Kim, N. E. Ostrom, K. R. Byzek, J. R. Leadbetter, and
J. A. Breznak. 2001. Nitrogen fixation by symbiotic and fi'ee-living spirochetes.
Science 292:2495-2498.

Lilburn, T. G., T. M. Schmidt, and J. A. Breznak. 1999. The phylogenetic
diversity of termite gut spirochetes. Environ. Microbiol. 1:331-345.

Liu, S., and J. M. Sulflita. 1993. H2-CO2 dependent anaerobic O-demethylation
activity in subsurface sediments and by an isolated bacterium. Appl Environ
Microbiol 59:1325-1331.

32

61.

62.

63.

64.

65.

66.

67.

68.

69.

Liu, S., and J. M. Sulflita. 1995. H2 as an energy source for mixotrophic
acetogenesis by the reduction of CO2 and Syringate by Acetobacterium woodii
and Eubacterium limosum. Curr Microbiol 31 :245-250.

Lovely, D. R., and S. Goodwin. 1988. Hydrogen concentration as an indicator of
the predominant terminal electron-accepting process in aquatic sediments.
Geochimicaet Cosmochimica Acta 52:2993-3003.

Lovley, D. R., D. F. Dwyer, and M. J. King. 1982. Kinetic analysis of
competition between sulfate reducers and methanogens for hydrogen in
sediments. Appl Environ Microbiol 43:1373-1379.

Morvan, B., F. Bonnemoy, G. Fonty, and P. Gouet. 1996. Quantitative
determination of H2-utilizing acetogenic and sulfate reducing bacteria and

methanogenic archaea fiom the digestive tract of different mammals. Curr.
Microbiol. 32:129-133.

Noda, S., M. Ohkuma, R. Usami, K. Horikoshi, and T. Kudo. 1999. Culture-
independent characterization of a gene responsible for nitrogen fixation in the
symbiotic microbial community in the gut of the termite Neotermes koshunensis.
Appl Environ Microbiol 65:4935-4942.

Noda, S., M. Ohkuma, A. Yamada, Y. Hongoh, and T. Kudo. 2003.
Phylogenetic position and in situ identification of ectosymbiotic spirochetes on
protists in the termite gut. Appl Environ Microbiol 69:625-633.

Norris, S. J., B. J. Paster, A. Moter, and U. B. Gobe12002, posting date. The
Genus Treponema. The prokaryotes: an evolving electronic resource for the
microbiological community. Dworkin, M (ed.). [Online.]

Odelson, D. A., and J. A. Breznak. 1983. Volatile fatty acid production by the
hindgut microbiota of xylophagous termites. Appl Environ Microbiol 45: 1602-
161 3.

Odelson, D. A., and J. A. Breznak. 1985. Nutrition and growth characteristics of
Trichomitopsis termopsidis, a cellulolytic protozoan from termites. Appl Environ
Microbiol 49:614-621.

33

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

Ohknma, M., and T. Kudo. 1996. Phylogenetic diversity of the intestinal
bacterial community in the termite Reticulitermes speratus. Appl Environ
Microbiol 62:461-468.

Ohkuma, M., S. Noda, R. Usami, K. Horikoshi, and T. Kudo. 1996. Diversity
of nitrogen fixation genes in the symbiotic intestinal microflora of the termite
Reticulitermes speratus. Appl Environ Microbiol 62:2747-2752.

Ohkuma, M., and T. Kudo. 1998. Phylogenetic analysis of the symbiotic
intestinal microflora of the termite Cryptotermes domesticus. F ems Microbiology
Letters 164:389-395.

Ohkuma, M., S. Noda, and T. Kudo. 1999. Phylogenetic diversity in the
nitrogen fixation genes in the symbiotic microbial community in the guts of
diverse termites. Appl Environ Microbiol 65:4926—4934.

Paster, B. J., F. E. Dewhirst, S. M. Cooke, V. Fussing, L. K. Poulsen, and J.
A. Breznak. 1996. Phylogeny of not-yet-cultured spirochetes from termite guts.
Appl Environ Microbiol 62:347-352.

Peters, V., P. H. J anssen, and R. Conrad. 1998. Efficiency of hydrogen
utilization during unitrophic and mixotrophic growth of Acetobacterium woodii
on hydrogen and lactate in the chemostat. FEMS Microbiol Ecol 26:317-324.

Phelps, T. J., and J. G. Zeikus. 1984. Influence of pH on terminal carbon
metabolism in anoxic sediments from a mildly acidic lake. Appl Environ
Microbiol 48: 1088-1095.

Potrikus, C. J., and J. A. Breznak. 1977. Nitrogen-fixing Enterobacter
aglomerans isolated fi‘om guts of wood-eating termites. Appl Environ Microbiol
33:392-399.

Potrikus, C. J., and J. A. Breznak. 1980. Uric acid degrading bacteria in guts of
termites [Reticulitermesflavipes (Kollar)]. Appl Environ Microbiol 40:117-124.

Potrikus, C. J., and J. A. Breznak. 1980. Anaerobic degradation of uric acid by
gut bacteria of termites. Appl Environ Microbiol 40: 125-132.

34

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

Potrikus, C. J., and J. A. Breznak. 1980. Uric acid in wood-eating termites.
Insect Biochemistry 10:19-&.

Potrikus, C. J., and J. A. Breznak. 1981. Gut bacteria recycle uric acid nitrogen
in termites - a strategy for nutrient conservation. Proc Natl Acad Sci U S A
78:4601-4605.

Prins, R. A., and A. Lankhorst. 1977. Sythesis of acetate from CO2 in the
cecum of some rodents. FEMS Microbiol. Letters 1:255-258.

Radek, R., K. Hausmann, and A. Breunig. 1992. Ectobiotic and endocytobiotic
bacteria associated with the termite flagellate Joenia annectens. Acta
Protozoologica 31 :93-107.

Radek, R., J. Rosel, and K. Hausmann. 1996. Light and electron microscopic
study of the bacterial adhesion to termite flagellates applying lectin
cytochemistry. Protoplasma 193:105-122.

Rosel, J., R. Radek, and K. Hausmann. 1996. Ultrastructure of the trichomonad
flagellate Stephanonympha nelumbium. J Eukaryot Microbiol 43:505-511.

Schink, B. 1997 . Energetics of syntrophic cooperation in methanogenic
degradation. Microbiol. Molec. Bio. Rev. 61:262-280.

Schultz, J. E., and J. A. Breznak. 1978. Heterotrophic bacteria present in
hindguts of wood-eating termites [Reticulitermesflavipes (Kollar)]. Appl Environ
Microbiol 35:930-936.

Schulz, S., and R. Conrad. 1996. Influence of temperature on pathways to
methane production in the permanently cold profundal sediment of Lake
Constance. FEMS Microbiol Ecol 20: 1-14.

Schulz, S., H. Matsuyama, and R. Conrad. 1997. Temperature dependence of
methane production from different precursors in a profundal sediment (Lake
Constance). FEMS Microbiol Ecol 22:207-213.

Slaytor, M., and D. J. Chappell. 1994. Nitrogen metabolism in termites. Comp
Biochem Physiol 107:1-10.

35

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

Tayasu, I., A. Sugimoto, E. Wada, and T. Abe. 1994. Xylophagous termites
depending on atmospheric nitrogen. Naturwissenschafien 81:229-231.

Tholen, A., B. Schink, and A. Brune. 1997. The gut microflora of Reticulitermes
flavipes, its relation to oxygen, and evidence for oxygen-dependent acetogenesis
by the most abundant Enterococcus sp. FEMS Microbiol Ecol 24: 137-149.

Tholen, A., and A. Brune. 2000. Impact of oxygen on metabolic fluxes and in
situ rates of reductive acetogenesis in the hindgut of the wood-feeding termite
Reticulitermesflavipes. Environ. Microbiol. 2:436-449.

To, L. P., L. Margulis, D. Chase, and W. L. Nutting. 1980. The symbiotic
community of the Sonoran desert termite: Pterotermes occidentis. Biosystems
13:109-137.

Veivers, P. C., R. W. Obrien, and M. Slaytor. 1982. Role of bacteria in
maintaining the redox potential in the hindgut of termites and preventing entry of
foreign bacteria. J Insect Physiol 28:947-951.

Watanabe, H., H. N akamura, G. Tokuda, I. Yamaoka, A. M. Scrivener, and
H. Noda. 1997. Site of secretion and properties of the endogenous endobeta-1,4-
glucanase components from Reticulitermes speratus (Kolbe), a Japanese
subterranean termite. Insect Biochem Mol Biol 27 :305-313.

Watanabe, H., H. Noda, G. Tokuda, and N. Lo. 1998. A cellulase gene of
termite origin. Nature 394:330-331.

Yamin, M. A. 1978. Axenic cultivation of the cellulolytic flagellate
T richomitopsis termopsidis (Cleveland) from the termite Zootermopsis. J
Protozool 25:535-538.

Yamin, M. A. 1980. Cellulose metabolism by the termite flagellate
T richomitopsis termopsidis. Appl Environ Microbiol 39:859-863.

Yamin, M. A. 1981. Cellulose metabolism by the flagellate Trichonympha from a
termite is independent of endosymbiotic bacteria. Science 21 1:58-59.

36

Chapter 2

Physiology and Nutrition of Treponema strains ZAS-1 and

ZAS-2, H2/CO2-Acetogenic Spirochetes from Termite Hindguts

Introduction

Spirochetes are among the most abundant microbial groups in termite hindguts,
accounting for up to one-half of the prokaryotic community (43). For more than a
century, however, our knowledge of these organisms was largely limited to sporadic
reports of their presence in various termite species, their morphological diversity, and
their physical association with termite gut protozoa (8). Although elimination of
spirochetes from the termite gut led to a decrease in termite survivorship (21), specific
roles of spirochetes and factors contributing to their abundance in the hindgut have
remained obscure.

Over the past ten years, our understanding of termite hindgut spirochetes has
advanced dramatically. Cultivation-independent molecular approaches revealed that they
group within the genus T reponema and that the large majority of 16S rDNA clones form
a phylogenetically discrete cluster [the “termite c1uster”] within this genus (35). These
studies also revealed a striking degree of phylogenetic diversity amongst the termite gut
treponemes, with as many as 21 distinct species occurring within a single termite host
species (35). A few years ago, the first pure cultures of these organisms were isolated in
our laboratory and were found to possess metabolic capabilities hitherto unknown in the

Spirochaetes division of the Bacteria, including acetogenesis fi'om H2 + CO2 (32) and N2

37

fixation (34). Both of these processes are unique to prokaryotes, and have been
demonstrated to be important in the provision of carbon, nitrogen, and energy to termites
(6, 47). Acetogenesis plays a particularly prominent role in termite metabolism: 71-
100% of the host’s energy requirements can be met by oxidation of acetate produced by
hindgut microbes, and it has been estimated that from 10.5% to 33% of this acetate
production is attributable to H2/CO2 acetogenesis (6, 42, 52).

The availability of pure cultures of termite gut spirochetes has allowed the
exploration of properties relevant to their growth and survival in situ. In this paper, we
report on the nutritional, physiological and biochemical properties of Treponema strains
ZAS-1 and ZAS-2, H2/CO2-acetogenic spirochetes isolated from hindguts of the
California dampwood termite, Zootermopsis angusticollis (Hagen) (32). Additional
information regarding the taxonomy, nomenclature, and genomic properties of these

strains is included in chapter 3.

Material & Methods

Media and cultivation methods

Routine cultivation of strains ZAS-1 and ZAS-2 was in butyl rubber stoppered
tubes or bottles containing ca. one-filth their volume of medium 2YACo (32) under an
atmosphere of 80% H2 / 20% CO2 (v/v). Medium 2YACo consisted of a mixture of
inorganic salts, vitamins, cofactors, and 2% (v/v) yeast autolysate (equiv. to ca. 2.2 mg
dry solids/m1). The medium was buffered by inclusion of 70 mM NaHCO3 and 10mM 3-
N-[morpholino] propanesulfonic acid (MOPS) and reduced with dithiothreitol (DTT) (1

mM final conc.). The pH of the medium prior to inoculation was 7 .2. Unless otherwise

38

noted, cultures were grown at 30°C on a reciprocal shaker (50 oscillations per min.) with

vessels held in a horizontal position.

Nutritional and growth studies

The ability of commercial yeast extracts to replace yeast autolysate in 2YACo
medium was tested by using the following products at 6 mg/ml final concentration:
Tastone nos. 900, 154, and 310; Amberex nos.1003 and 695; and Amberferm nos. 5925,
5902, and 5021 (Red Star Bioproducts, Juneau, Wisconsin). Cofactor requirements were
evaluated by testing for an attenuation of growth in 2YACo medium lacking one of the
eleven cofactors present in the cofactor stock solution (32). Candidate compounds were
then tested in reciprocal experiments in which they alone were incorporated into medium
2YACo instead of the cofactor mixture.

Substrate utilization studies were performed with cells growing under an
atmosphere of N2/CO2 (80/20, v/v). For ZAS-1, medium 2YACo was modified to
' contain Amberferm 5902 (4 mg/ml) in place of yeast autolysate. Unmodified medium
2YACo was used for ZAS-2, with a small amount H2 (16 mM) added to the N2/CO2
headspace to aid in the initiation of growth (see below). H2 was not added to cultures
grown on methoxylated aromatics. An increase in cell yield (>20%) in the presence of a
substrate, as compared to its absence, was taken to indicate utilization of the substrate as
an energy source. Cell growth was determined by measuring the optical density (OD) of
cultures at 600 nm with a Milton Roy Spectronic 20 colorimeter. OD readings were
converted to cell number by reference to a standard curve relating these quantities.

Substrate carbon balances were determined under conditions of substrate-limited growth.

39

Carbon recoveries for methoxylated aromatic compounds were calculated based on
acetate production expected fiom demethylation of the aromatic (R) substrate according
to the following equation:

R(-OCH3)n + 0.5n C02 -> R (-OH)n + 0.75n CH3COOH + 0.5n H20
Organic acid production was determined by using a high performance liquid
chromatograph (HPLC) with refractive index detection (4). Aromatic compounds were
analyzed using a Beckrnan model 127 HPLC equipped with a model 168 photodiode
array detector and an Alltech Lichrosorb RP-18 column (250 x 4.6 mm, 10 um particle
size). The mobile phase was 0.1% phosphoric acid with a methanol gradient, linearly
increasing from 48 to 55% in 30 minutes. The flow rate was 1.5 mein.

To test for mixotrophic growth, strain ZAS-2 was grown in 750 ml bottles
containing 100 ml 2YACo media with 2 mM maltose and a 650 ml headspace composed
of 20% H2, 20% CO2, 60% N2 (vol/vol). Consumption of H2 was followed by gas
chromatography (5). Maltose consumption and organic acid production were followed

by using the anthrone assay (1) and HPLC analysis (above), respectively.

Determination of hydrogen thresholds

Hydrogen thresholds were determined as described by Lovley (37). In brief,
cultures were grown under H2/C02 (80/20, v/v) in medium 2YACo unmodified (ZAS-2)
or modified to contain 4 mg/ml Amberferm 5902 in place of yeast autolysate (ZAS-1).
When cultures reached mid-log phase (at which point further growth of both strains was
strictly dependent on the presence of H2), the headspace was replaced with N2/CO2

(80:20, v/v), followed by the introduction of a small amount of H2 (ca. 6000 ppmv). The

40

basal level to which this H2 was consumed (i.e. the H2 threshold) was determined through
three cycles of H2 addition and consumption for each culture. H2 was monitored by using
a Trace Analytical RGA2 gas chromatograph equipped with a RGD2 trace gas detection

unit.

Enzyme Assays

Cells from mid-log phase cultures (ODooo values of 0.15 to 0.3) were harvested
by centrifugation (16,000 x g for 10 min) and resuspended at 10x their original
concentration in appropriate assay buffer (as cited below) containing 10 mM
ethylenediaminetetraacetic acid (EDTA), 1 mM phenylrnethanesulfonyl fluoride (PMSF),
and leupeptin (10 rig/ml) to inhibit proteases. While held at 4°C, cells were disrupted by
sonication 3 times for 30 s each with a Branson Model 450 sonifier (power setting of 5;
50% duty cycle) equipped with a stepped micro-tip. The resulting crude cell extracts
were assayed for enzyme activities.

Formyltetrahydrofolate (formyl-THF) synthetase, methenyl-THF cyclohydrolase
and methylene-THF dehydrogenase (41) and methylene-THF reductase (39) were
assayed as described. THF and THF derivatives used in the assays were obtained from
Schircks Laboratories (J ona, Switzerland). Catalase was assayed by measuring the rate
of decrease in the A240 of H202 (2). Oxidase and peroxidase activities coupled to the
oxidation of NADH or NADPH were assayed as described by Stanton (50). Superoxide
dismutase was assayed by the xanthine/xanthine oxidase-cytochrome c reduction method

(22). Protein content of cell extracts was measured by the Lowry assay (38).

41

Absorbance measurements were made using a Perkin-Elmer Lambda 14 UVN IS

spectrophotometer.

Oxygen Tolerance

Cells growing under anoxic conditions were tested for the ability to maintain
growth after the addition of various concentrations of O2 to the headspace. Replicate
cultures (n = 3) were grown under H2/CO2 with oscillation (above) in 18 mm anaerobe
tubes containing Sml of medium 2YACo modified by the inclusion of 10 mM maltose,
but containing no D'I'I‘ reducing agent. When cells reached mid-log phase, the headspace
was balanced to atmospheric pressure with H2/CO2, and sterile 02 was injected to a final
headspace concentration of 0.5%, 1%, 2.5%, or 5% (v/v). Cultures were then

reincubated, and firrther growth was monitored as described above.

Results

Nutrition and growth of Treponema strains ZAS-1 and ZAS-2

Cells of both strains grew in medium 2YACo under H2+CO2 within an initial pH
range of 6.5 to 7.8, with an optimum at pH 7.2. No growth was observed in media with
an initial pH 56.0 or 28.0. Both strains grew within a temperature range of 23°C to
32°C, with an optimum at 30°C. No growth occurred at 4°C, or 34°C. Under optimum
conditions (H2/CO2 + organic substrates), the shortest doubling time of cells was 24 hrs

(ZAS-1) and 29 hrs (ZAS-2).

42

A variety of commercial yeast extracts (6 mg/ml final concentration) were tested
for their ability to replace yeast autolysate in medium 2YACo. Tastone 900, Amberferm
5925, and Amberferm 5902 could replace yeast autolysate for growth of strain ZAS-1,
whereas Tastone 154, Tastone 310, Amberex 1003, and Amberex 695 were suitable
replacements for growth of strain ZAS-2. No single product was effective for both
organisms. Growth rates and cell yields of ZAS-1 and ZAS-2 in media containing
commercial yeast extracts were similar to those measured in media 2YACo.

Besides H2 + C02, a variety of hexoses, pentoses, and disaccharides were also
used as energy sources by both strains and were fermented homoacetogenically (Table 1).
Curiously, however, when grown on organic substrates under N2/CO2, strain ZAS-2
displayed prolonged lag phases (272 h) prior to the initiation of growth. The provision
of small amounts of H2 (16 mM) or the use of larger inocula (> 5% v/v) eliminated such
lags (Figure 2.1). Strain ZAS-2 was additionally able to utilize methoxylated aromatic
compounds (syringate, ferulate, vanillate, and trimethoxybenzoate) as energy sources
when supplied at 52.5 mM (higher concentrations inhibited growth). Cell doubling
times were typically greater than 100 hours on these substrates, and acetate production
was consistent with demethylation of the compounds (i.e. the aromatic ring did not
appear to be cleaved) (Table 1). This was confirmed for trimethoxybenzoate, which was
quantitatively converted to gallic acid and acetate. Neither strain was capable of growth
on other C1 compounds or methyl group donors tested (methanol, formats, C0, betaine,
or choline). Relatively low concentrations of CO were inhibitory to both strains; addition
of 1% CO (v/v) to the headspace of actively growing cultures resulted in the immediate

cessation of growth.

43

5.80508 005 28555 050880 05 .8 850350805 805 508095 850388 058000 8 50009.
.NOQNE 05 88000 00 b08000 88020 50>C0T£ 8 .2308 00 505800.. .858 88555 .8 0w8=00.~0mo

.2833 505 05 wet—00— 080508 00$ 8:508 oo<>m 505508 8 850805
08500 85 538m .8.“ 50500.80 05 85 538w 0088358580 .8 085580 .588 5088.555 053 0050800.. 858 85 020$...

8888800 .8.“ 05: 5080008 8 85 G3

00:05.80 8 503838 .3888 00? «Comm 8 0885 53 .8 530..» 8w 8.5 .8850 .555 .3 508:8 8: 0.53 088:0 580 .0883
.080bw .05>:.Sa .0880— .00 .0885 85508 .08an 8.: .3 508058 0.8 508:8 8: :5 50505 008835 .948 on: 5 85
G48 8 00 .983 5880800 050805 50538508 05 05 8085 28m .8 85880080 85 a 5 50288 0.53 0008038 :5

 

 

m2 - v.80 - .20.? .8 228;
m _ m - .38 - .20.: .8 2238
8 - exam - .20.? .3. 85:80
5 - .88 - .2: ._ .8 3880885085
- R - $8 .3. .20.? 828:8
8 R :3 $3 .20.: .03.... 8932
2: 3 saw as .25; .052 as?
- 0N - 50 .3. .2me 08:58...
2 mm as $3 .20.: .250 3882
K mm :3 $3 .20.? .2085 88:8
we 3 .80 :8 page .208 .09.:
«-08 705 ”-98 708 ~08 708 .8825
35%: mun—uh. Mam—£509 Sewn—€250”— 83oo< .— na\m=oo “:3? ~02

 

 

 

N-m<N 5:0 Tm<N 0:850 efioeofioé .3 8050853 ofibmnsm A." 0.55.

 

 

o '160'260 '360 '460 '560'660'760
Time (hrs)

Figure 2.1. Hz stimulation of chemoorganotrophic growth of Treponema
strain ZAS-2. Substrates added to 2YACo medium under N2/C02 head
spaces (80/20, v/v) were: 10 mM maltose (O), 16 mM H2 (A) 10 mM
maltose plus 16 mM H2 (Cl), and no added substrates (V).

45

In order to determine which cofactors in 2YACo medium were required for growth, ZAS-
1 was grown through 3 successive transfers in 2YACo media lacking one of the eleven
cofactors. Cultures deficient in folinate (formyltetrahydrofolate) displayed successively
decreasing growth yields with each transfer. Further testing demonstrated that folinate
alone (500 ng/ml, final concentration) could replace the eleven cofactor mixture for both

strains.

Mixotrophic growth

Strain ZAS-2 was capable of mixotrophic growth (i.e. the simultaneous utilization
organic substrates and H2 for energy). Growing in the presence of both maltose and H2 +
CO2, strain ZAS-2 displayed a significantly shorter doubling time (29.2 d: 1.2 hrs.) than
when growing on maltose (65.9 i 2.4 hrs.) or H2 + CO2 (50.3 :t 2.1 hrs.) alone.
Moreover, maltose and H2 were consumed simultaneously during growth (Figure 2.2),
and cell yields and acetate production from maltose plus H2 + CO2 were close to the sum
of cell yields and acetate production when grown on either substrate alone (Table 2.2).
Strain ZAS-2 also appeared to be capable of mixotrophy with H2 + C02 plus other
organic substrates: doubling times during growth on H2 + CO2 plus trimethoxybenzoate
(38.1 :t 2.5 hrs) or xylose (35.2 i 1.5 hrs) were significantly shorter than during

unitTOphic growth on either organic substrate alone (Table 1).

Hydrogen thresholds

The lowest level to which H2 could be consumed (i.e. the H2 threshold) was

determined for mid-log phase cultures of H2/CO2 grown cells, after replacing the

46

~
1

L l A .l A

cg: 11 10.1
e j 8
V \O
:I:N ‘ 1 8
<1 ‘ . ¢

 

5: -D-Acetate or O- Maltose (mmol)
l

n
a a
O

1 1

O

 

 

 

 

. , . r. , . r . , . . . , 0.01
50 100 150 200 250 300 350
Time(hrs)

c—

Figure 2.2. Mixotrophic growth of Treponema strain ZAS-2 on
H2+C02 and maltose.

47

Table 2.2. Cell yields and acetate production by Treponema strain ZAS-2 growing under
unitrophic and mixotrophic conditions.

 

 

Substrate Acetate Cell Yield‘
Substrate Consumeda (mmol) Produced' Recoveryb (cells/ml)
(mmol)

Maltose 0.15 :l: 0.04 0.84 a: 0.19 93% 2.4 x 108
H2+CO2 5.23 d: 0.35 (H2) 1.23 i 0.13 94% 1.7 x 108
Maltose 0.15 i 0.01 (maltose)

plus plus 1.95 a: 0.03 89% 4.3 x 108
H2+CO2 5.20 i 0.19 (H2)

 

“Averages derived from replicate cultures (11 = 3).

l’Acetate recovery as a percentage of substrate carbon (maltose) or electron (H2)
consumption.

48

headspace with N2/CO2 (80/20, v/v) and measuring the extent of consumption of three
successive additions of small amounts (ca. 6000 ppmv) of H2. ZAS-1 and ZAS-2
displayed H2 thresholds of 490 ppmv (s.d. 40 ppmv) and 650 ppmv (s.d. 50 ppmv)
respectively. The H2 threshold of ZAS-2 decreased moderately to 350 ppmv (s.d. 30

ppmv) when growing mixotrophically on H2 + CO2 plus trimethoxybenzoate (Figure 2.3).

H2/CO2-acetogenic enzyme activities

Previous work in our laboratory demonstrated that strains ZAS-1 and ZAS-2
possessed hydrogenase, forrnate dehydrogenase, and CO dehydrogenase, the latter a
characteristic enzyme of C02-reductive acetogenesis via the Wood/Ljungdahl pathway
(32). To further confirm the presence of this pathway in ZAS-1 & 2, enzymes associated
with the conversion of forrnate to the methyl-group of acetate were assayed. Cell extracts
of ZAS-l and ZAS-2 displayed activities for forrnyl-THF synthetase, methenyl-THF
cyclohydrolase, methylene-THF dehydrogenase, and methylene-THF reductase, the four
characteristic enzymes of the methyl group-forming branch of the Wood/Ljungdahl
pathway (Table 2.3). The forrnyl-THF-synthetase activity of both strains was dependent
on the presence of ATP, formate, and THF. The activity of methylene-THF-
dehydrogenase was NADP dependent in both strains; no activity was observed with
NAD. In all cases, enzyme activity was abolished in cell extracts that had been boiled for

10 minutes.

49

2000 -

1000 J

 

 

 

 

 

Time (h)

Figure 2.3. Determination of H2 threshold for replicate
cultures (El and O) of Treponema strain ZAS-2, growing
under mixotrophic conditions (H2 + C02 plus
trimethoxybenzoate). H2 additions are indicated by arrows.
The threshold value is indicated by a dashed line.

50

Table 2.3. Enzyme activities relevant to H2/CO2-acetogenesis in Treponema str. ZAS-1

 

 

and ZAS-2
Specific Activii

Enzyme ZAS-1 ZAS-2
Hydrogenaseb 0.47 0.45
CO dehydrogenaseb 1.28 0.64
F ormate dehydrogenaseb 0.13 0.20
F ormyl-THF-synthetase 0.76 0.71
Methenyl-THF-cyclohydrolase 0.12 0.1 1
Methylene-THF-dehydrogenase 0.39 0.33
Methylene-THF-reductase 0.18 0.22

 

aSpecific activities are expressed as micromoles of substrate used (or product formed) per
minute per milligram of protein.

bActivities for these enzymes were previously published in reference (32), and are
included here for comparison.

51

 

 

0.2 -
o/8
0
0.] " a
. A\
. 73\ A\A
a/.o N
c: /:’. . e \V
O
50 /O O
O . 0
l
,
. /
0
0.01- v
' I ' I I ' I ' I
0 50 100 150 200 250
Time (hrs)

Figure 2.4. Effect of 02 on the grth of Treponema strain ZAS-
2. Cultures were initially grown under 80% H2:20% C02, with
02 additions made at 125 hours (indicated by arrow). The
resulting concentration of 02 in the culture tube headspace
(percentage by volume) was: 0(13), 0.5 (O), l (A), 2.5 (V), and

5 (0).

52

Oxygen tolerance and oxidative stress enzymes

Mid-log phase cultures of strains ZAS-l and ZAS-2 were able to maintain growth
after the addition of 0.5% 02 to the headspace atmosphere, albeit with a slight and
transient decrease in growth rate (Figure 2.4). Addition of 21% 02, however, led to the
rapid cessation of growth. The tolerance of cells to limited 02 exposure prompted assays
for enzymes associated with oxidative stress protection. Cell extracts of both strains
showed low levels of NADH- and NADPH peroxidase, whereas neither exhibited
activities of catalase or superoxide dismutase (SOD). Both strains had low levels of
NADPH oxidase, but relatively high levels of NADH oxidase activity was seen only in
ZAS-2 (Table 2.4). Exposure of cultures to 0.5% 02 12 hours prior to enzyme assays did
not significantly alter the levels of any of the tested enzyme activities in either of the

strains, and activity of all enzymes tested was abolished in boiled cell extracts.

Discussion

Treponema strains ZAS-1 and ZAS-2 possess a variety of properties that are
typical of traditional “acetogens”, including the use of the acetyl-CoA (Wood/Ljungdahl)
pathway for reductive synthesis of acetyl-CoA/acetate from H2 + CO2 for energy
generation and carbon fixation (16). Operation of this pathway, first suggested by the
presence of CO dehydrogenase in cells (32), was further supported in this study by the
demonstration of the corresponding enzymes catalyzing conversion of CO2 to the methyl
group of acetate (Table 3). Strains ZAS-1 and ZAS-2 are also similar to other

homoacetogens in being nutritionally versatile (17) and capable of homoacetogenic

53

Table 2.4. Oxygen and peroxide detoxifying enzyme activities in Treponema str. ZAS-1

 

 

and ZAS-2.

Specific Activii
Enzyme ZAS-1 ZAS-2
NADH Oxidase n.d. 560
NADPH Oxidase 7 10
NADH Peroxidase 10 40
NADPH Peroxidase 3 5
Catalase n.d n.d
Superoxide Dismutase n.d. n.d.

 

‘Specific activities are expressed as nanomoles pyridine nucleotide oxidized per minute

per milligram protein. n.d., activity not detected.

54

growth on a variety of organic substrates likely to be present in the termite gut, including
hexoses, pentoses, disaccharides, and (in the case of ZAS-2) the methyl group of
methoxylated aromatic compounds. Although the high-molecular-weight core lignin of
ligrrocellulose does not appear to be degraded significantly during passage through the
guts of wood-feeding termites (7, 12, 20, 25), there is evidence for the demethylation of
aromatic lignin moieties (20). As such, it seems likely that acetogenic spirochetes
contribute to termite nutrition via aceto genesis from fermentation and demethylation of
organic substrates in situ, as well as fiom H2 and CO2 produced by other members of the
gut microbiota.

Strain ZAS-2 [and apparently also ZAS-1 (32)] was capable of mixotrophy, i.e.
simultaneous use of H2 + CO2 and organic substrates as energy sources (Figure 2; Table
2). This capability has been demonstrated in other acetogens, including those isolated
from termite guts (5). Given the seemingly constant and relatively high standing pool of
H2 in hindguts (l 8), mixotrophy may well be a frequent, if not constant, mode of growth
for the ZAS strains in situ. Indeed, some strains may have become so adapted to the
relatively high concentration of H2 in termite hindguts that they now require it for optimal
growth. This was suggested by the stimulatory effect of H2 on utilization of organic
substrates by strain ZAS-2 (Figure 1), which may be indicative of a requirement for H2 as
a low potential electron donor for some step(s) in biosynthesis. Although ZAS-2
possesses a hydrogenase (Table 3) and presumably forms some H2 during fermentation of
organic substrates, at low cell densities such free H2 may diffuse away fiom cells rapidly
and thereby limit the rate of growth. Only as cell density (and hence H2-forming capacity

per ml) increases and H2 accumulates would the specific growth rate increase

55

incrementally. This interpretation is supported by the broadly concave shape of the
growth curve of ZAS-2 on maltose in the absence of added H2 (Figure 1).

The H2 thresholds of ZAS-1 and ZAS-2 (490 :1: 40 ppmv and 650 :l: 50 ppmv,
respectively) were also well within the range reported for other H2-utlilizing acetogens
(13). For ZAS-2, the H2 threshold was somewhat lower during mixotrophic growth on
H2 plus trimethoxybenzoate (350 :l: 30 ppmv), but was not lowered to a level which
would make ZAS-2 competitive with H2-utilizing methanogens under H2-limited growth
conditions. In any event, direct competition with methanogens for H2 may be largely
irrelevant; given the high H2 concentrations measured in the central region of termite
hindguts where spirochetes are most abundant [ca. 50,000 ppmv (18)], it is difficult to
imagine that spirochetes are ever seriously limited for H2 in vivo.

In contrast to other homoacetogens, the amount of energy conserved by strains
ZAS-1 and ZAS-2 during H2/CO2 acetogenesis is rather low. Assuming that the protein
content of cells is 55% of the dry mass (40), it can be calculated fiom the information in
Table 1 that the cell yields of ZAS-1 and ZAS-2 (mg dry mass/mmol substrate) are
respectively 0.1 and 0.2 during growth on H2 + CO2, and 7.9 and 6.3 for glucose. This is
substantially lower than that of other acetogens, whose cell yields range fiorn 0.5 to 4.2
with H2 and 32.7 to 53.0 with hexoses (14, 19, 46, 53). However, this seemingly
inefficient coupling of acetogenesis to growth may contribute to the effectiveness of the
spirochetes as syrnbionts. Acetate produced by gut microbes is the primary energy
source for termites (6), and ZAS-1 and ZAS-2 will generate acetate as their sole end
product without a concomitantly large increase in biomass and nutrient demand that

could be detrimental to the host.

56

Previous studies have shown that the peripheral region of termite hindguts is
microoxic (9, 18). 02 concentrations near the gut epithelium are 50-100 14M, and
decrease steeply to anoxia within about 200 pm of the gut wall. It seems reasonable to
assume that spirochetes inhabiting the hindgut lumen are likely to encounter this
microoxic zone periodically. Although ZAS-l and ZAS-2 are “strict anaerobes” in the
traditional sense (i.e. they are incapable of growth in air), their ability to tolerate low
levels of oxygen exposure could be of adaptive value in their natural habitat. Both ZAS-
1 and ZAS-2 were capable of maintaining growth in the presence of 0.5% (vol/vol)
oxygen (equivalent to a dissolved 02 concentration of 6 11M). This ability suggests that
the spirochetes could survive transient exposure to substantially higher concentrations of
02 (a situation that is perhaps more analogous to that which they would encounter in
vivo). Both strains possessed activities for NADH and NADPH peroxidase activities that
could afford protection against H202. ZAS-2 also exhibited activity of the O2-consuming
enzyme NADH oxidase. The activities of peroxidase and oxidase were similar to those
of Brachyspira hyodysenteriae, an aerotolerant anaerobic spirochete that colonizes the
gastrointestinal tract of pigs (50). Studies have shown that free-living acetogens display
varying degrees of aerotolerance and can grow under atmospheres ranging fi'om 0.3%
(Acetobacterium woodii) to 6% (Clostridium glycolicum RD-l) oxygen (28, 30). A
recent study revealed a similar degree of aerotolerance, as well as active reduction of 02
using H2 or organic compounds as reductants, in a variety of non-spirochetal acetogens
isolated fi'om termite hindguts (3). Like the acetogens examined in these studies, both
strains lack superoxide dismutase (SOD) activity. The ability to tolerate low levels of 02

has also been observed in the termite hindgut-associated methano gens

57

Methanobrevibacter cuticularis and M. curvatus (31), suggesting that aerotolerance may
be a common trait among other anaerobes in the termite hindgut.

Studies of the nutritional requirements of ZAS-l and ZAS-2 revealed a strict
requirement for folate compounds. This observation was surprising, given the fact that
tetrahydrofolate plays a critical role as a one-carbon carrier in the methyl-group-forming
branch of the Wood/Ljungdahl pathway (36). Most bacteria are capable of de novo folate
biosynthesis, and numerous homoacetogens grown in defined media have no apparent
folate requirements (23, 33, 45). By contrast, many host-associated bacteria require an
exogenous source of folate, including ruminal [Treponema bryantii (49)] and genital-
associated [T. phagedenis (51)] spirochetes and numerous nonspirochetal members of the
bovine rumen microbial community (24, 48). The source of folate compounds for the
ZAS strains in vivo is currently unknown. Although provision of folates by the termite
host is a possibility (folate synthesis has been demonstrated in the mosquito, Aedes
aegypti (27, 54, 55)), most insects require a dietary source of pre-forrned folate. A more
likely source of folate compounds are other members of the complex hindgut community.
A variety of bacteria have been shown to secrete folate compounds in vitro (15, 26), as
well as in gastrointestinal tracts (10, 11, 29, 44). This hypothesis is further explored in
chapter 4.

In summary, Treponema strains ZAS-1 and ZAS-2 possess an assortment of
nutritional and physiological properties that would appear to make them well-adapted to
life as termite hindgut symbionts. In light of the phylogenetic diversity of termite gut
spirochetes revealed by cultivation-independent molecular analyses (35), it seems certain

that the few strains currently in culture (32, 34) offer only an introductory glimpse into

58

the physiological diversity of termite gut spirochetes and their importance to termite

nutrition.

Acknowledgements

This research was funded by grants from the National Science Foundation (IBN97-09000
and IBN01-14505). HPLC analysis of aromatic compounds was performed with
equipment provided by the laboratory of Dr. J arnes Tiedje, with the aid of Benjamin
Griffin. I am also grateful to Dr. Tom Schmidt and members of his laboratory for helpful

discussions and criticisms.

59

References

1.

10.

ll.

Ashwell, G. 1963. Colorimetric analysis of sugars, p. 73-105. In S. P. a. N. O. K.
Colowick (ed.), Methods in Enzymology, vol. 3. Academic Press, New York.

Beers, R. F., and I. W. Sizer. 1952. Spectrophotometric method for measuring
the breakdown of hydrogen peroxide by catalase. J Biol Chem 195:133-140.

Boga, H. 1., and A. Brune. 2003. Hydrogen-dependent oxygen reduction by
homoacetogenic bacteria isolated from termite guts. Appl Environ Microbiol
69:779-786.

Breznak, J. A., and J. M. Switzer. 1986. Acetate synthesis fiom H2 plus CO2 by
termite gut microbes. Appl Environ Microbiol 52:623-630.

Breznak, J. A., and J. S. Blum. 1991. Mixotrophy in the termite gut acetogen,
Sporomusa termitida. Arch Microbiol 156:105-110.

Breznak, J. A. 1994. Acetogenesis from carbon dioxide in termite guts. In H. L.
Drake (ed.), Acetogenesis. Chapman & Hall, New York.

Breznak, J. A., and A. Brune. 1994. Role of microorganisms in the digestion of
lignocellulose by termites. Annu Rev Entomol 39:453-487.

Breznak, J. A., and J. R. Leadbetter. 2002. Termite Gut Spirochetes. In M.
Dworkin (ed.), The Prokaryotes: an evolving electronic resource for the
microbiological community. Springer-Verlag, New York.

Brune, A., D. Emerson, and J. A. Bro-ak. 1995. The termite gut microflora as
an oxygen sink: microelectrode determination of oxygen and pH gradients in guts
of lower and higher termites. Appl Environ Microbiol 61 :2681-2687.

Camilo, E., J. Zimmerman, J. B. Mason, B. Golner, R. Russell, J. Selhub, and
I. H. Rosenberg. 1996. Folate synthesized by bacteria in the human upper small
intestine is assimilated by the host. Gastroenterol 110:991-998.

Camilo, M. E., J. Zimmerman, M. Nadeau, B. Golner, J. Selhub, J. Mason,
R. Russell, and I. H. Rosenberg. 1994. Folate Synthesized by the Intestinal
Microflora Are Assimilated by the Human Host. Gastroenterol 106:A599-A599.

60

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Cooksan, L. J. 1987. ”C-Lignin degradation by three Australian termite species.
Wood Sci.Technol. 21:11-25.

Cord-Ruwisch, R., H. Seitz, and R. Conrad. 1988. The capacity of
hydrogenotrophic anaerobic bacteria to compete for traces of hydrogen depends
on the redox potential of the terminal electron acceptor. Arch. Microbial.
149:350-7.

Daniel, S. L., T. Hsu, S. 1. Dean and H. L. Drake. 1990. Characterization of the
H2- and CO-dependent chemolithotrophic potentials of the acetogens Clostridium
thermoaceticum and Acetogenium kivui. J Bacterial 172:4464-4471.

Deguchi, Y., T. Morishita, and M. Mutai. 1985. Comparative studies an
synthesis of water-soluble vitamins among human species of Bifidabacteria.
Agric. Biol. Chem. 49:13-19.

Drake, H. L. 1992. Acetogenesis and acetogenic bacteria, p. 1-15. In J. Lederberg
(ed.), Encyclopedia of Microbiology, vol. 1. Academic Press, New York.

Drake, H. L., S. L. Daniel, K. Kusel, C. Matthies, C. Kuhner, and S. Braus-
Strameyer. 1997. Acetogenic bacteria: what are the in situ consequences of their
diverse metabolic versatilities? Biofactors 6:13-24.

Ebert, A., and A. Brune. 1997. Hydrogen concentration profiles at the oxic-
anoxic interface: a microsensor study of the hingut of the wood-feeding lower
termite Reticulitermesflavipes. Appl Environ Microbial 63:4039-4046.

Eichler, B., and B. Schink. 1984. Oxidation of primary aliphatic alcohols by
Acetobacterium carbinolicum sp. nov., a homoacetogenic anaerobe. Arch
Microbiol 140: 147-152.

Esenther, G. R., and T. K. Kirk. 1974. Catabolism of aspen sapwood in
Reticulitermesflavipes. Ann Entomol Soc Am 67:989-991.

Eutick, M. L., P. Veivers, R. W. O'Brien, and M. Slaytor. 1978. Dependence
of the higher termite, Nasutitermes exitiosus and the lower termite, Coptotermes
lacteus on their gut flora. J Insect Physiol 24:363-368.

61

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

Flohe, L., R. Becker, R. Brigelius, E. Lengfelder, and F. Otting. 1988.
Convenient assays for superoxide dismutase, p. 287-293. In J. Miquel (ed.), CRC
handbook of fi'ee radicals and antioxidants in biamedicine, vol. 3. CRC Press,
Boca Ratan.

Genthner, B. R. S., C. L. Davis, and M. P. Bryant. 1981. Features of rumen and
sewage sludge strains of Eubacterium limosum, a methanol-utilizing and H2-CO2-
utilizing species. Appl Environ Microbiol 42:12-19.

Herbeck, J. L., and M. P. Bryant. 1974. Nutritional requirements of the
intestinal anaerobe Ruminococcus bromii. Appl.Micrabiol. 28:1018-1022.

Hyada, F., J. Azuma, and T. Abe. 1999. Estimation of the effect of passage
through the gut of a lower termite, Coptotermesformosanus Shiraki, on lignin by
solid-state CP/MAS 13C NMR. Holzforschung 53:244-246.

Iwai, K., M. Kobashi, and H. Fujisawa. 1970. Occurence of Crithidia factors
and folic acid in various bacteria. J Bacterial 104:197-201.

Jaffe, J. J., J. J. McCormack, E. Meymarian, and H. M. Daremus. 1977.
Comparative activity and properties of lactate dehydrogenase, xanthine
dehydrogenase, and dihydrofolate reductase in normal and Brugia pahangi
infected Aedes aegypti. J. Parasitol. 63:547-553.

Karnholz, A., K. Kusel, A. Gossner, A. Schramm, and H. L. Drake. 2002.
Tolerance and metabolic response of acetogenic bacteria toward oxygen. Applied
and Environmental Microbiology 68:1005-1009.

Krause, L. J., C. W. Forsberg, and D. L. O'Connor. 1996. Feeding human milk
to rats increases Bifidobacterium in the cecum and colon which correlates with
enhanced folate status. J Nutr 126:1505-1511.

Kusel, K., A. Karnholz, T. Trinkwalter, R. Devereux, G. Acker, and H. L.
Drake. 2001. Physiological ecology of Clostridium glycolicum RD-l, an
aerotolerant acetogen isolated fiom sea grass roots. Appl Environ Microbiol
67 :4734-4741 .

Leadbetter, J. R., and J. A. Breznak. 1996. Physiological ecology of
Methanobrevibacter cuticularis sp. nov. and Methanobrevibacter curvatus sp.

62

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

nov., isolated fiom the hindgut of the termite Reticulitermesflavipes. Appl
Environ Microbial 62:3620-3631.

Leadbetter, J. R., T. M. Schmidt, J. R. Graber, and J. A. Breznak. 1999.
Acetogenesis from H2 plus CO2 by spirochetes from termite guts. Science
283:686-689.

Leigh, J. A., F. Mayer, and R. S. Wolfe. 1981. Acetogenium kivui, a new
thermophilic hydrogen-oxidizing, acetogenic bacterium. Arch Microbial 129:275-
280.

Lilburn, T. C., K. S. Kim, N. E. Ostrom, K. R. Byzek, J. R. Leadbetter, and
J. A. Breznak. 2001 . Nitrogen fixation by symbiotic and fi‘ee-living spirochetes.
Science 292:2495-2498.

Lilburn, T. G., T. M. Schmidt, and J. A. Breznak. 1999. The phylogenetic
diversity of temrite gut spirochetes. Environ. Microbiol. 1:331-345.

Ljungdahl, L. G. 1984. Other functions of folates, p. 555-579. In R. L. B. a. S. J.
Benkovic (ed.), Folates and Pterins, vol. 1. John Wiley and Sons, New York.

Lovely, D. R. 1985. Minimum threshold for hydrogen metabolism in
methanogenic bacteria. Appl Environ Microbiol 49: 1530--31.

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

MeBmer, M., G. Wahlfarth, and G. Diekert. 1993. Methyl chloride metabolism
of the strictly anaerobic, methyl chloride-utilizing homoacetogen strain MC. Arch
Microbiol 160:383-387.

Neidhardt, F. C., J. L. Ingraham, and M. Schaechter. 1990. Physiology of the
Bacterial Cell. Sinauer Assoc., Sunderland, Massachusetts.

O'Brien, W. E., and L. G. Ljungdahl. 1972. Fermentation of fructose and
synthesis of acetate fiom carbon dioxide by Clostridium formicoaceticum. J
Bacteriol 109:626-632.

63

42.

43.

45.

46.

47.

48.

49.

50.

51.

Odelson, D. A., and J. A. Breznak. 1983. Volatile fatty acid production by the
hindgut nricrabiota of xylophagous termites. Appl Environ Microbiol 45:1602-
1 613 .

Paster, B. J., F. E. Dewhirst, S. M. Cooke, V. Fussing, L. K. Poulsen, and J.
A. Breznak. 1996. Phylogeny of not-yet-cultured spirochetes from termite guts.
Appl Environ Microbiol 62:347-352.

Rang, N., J. Selhub, B. R. Goldin, and I. H. Rosenberg. 1991. Bacterially
synthesized folate in rat large intestine is incorporated into host tissue falyl
polyglutarnates. J Nutr 121 : 1955-1959.

Schink, B. 1984. Clostridium magnum sp. nov., a non-autotrophic
homoacetogenic bacterium. Arch Microbial 137 :250-255.

Sharak-Genthner, B. R., and M. P. Bryant. 1987. Additional characteristics of
ane-carban-campaund utilization by Eubacterium limosum and Acetobacterium
woodii. Appl Environ Microbiol 53:471-476.

Slaytor, M., and D. J. Chappell. 1994. Nitrogen metabolism in termites. Comp
Biochem Physiol 107:1-10.

Slyter, L. L., and J. M. Weaver. 1977. Tetrahydrofolate and other growth
requirements of certain strains of Ruminococcusflavefaciens. Appl Environ
Microbiol 33:363-369.

Stanton, T. B., and E. Canale-Parola. 1980. Treponema bryantii sp nov, a
rumen spirochete that interacts with cellulolytic bacteria. Arch Microbiol
127:145-156.

Stanton, T. B. 1989. Glucose metabolism and NADH recycling by T reponema
hyodysenteriae, the agent of swine dysentery. Appl Environ Microbial 55:2365-
2371.

Steinman, H. G., V. I. Oyama, and H. O. Schulze. 1954. Carbon dioxide,
cocarboxylase, citrovorum factor, and caenzyme A as essential growth factors for
a saccharalytic T reponeme. J Biol Chem 211 :327-335.

52.

53.

54.

55.

Tholen, A., and A. Brune. 2000. Impact of oxygen on metabolic fluxes and in
situ rates of reductive acetogenesis in the hindgut of the wood-feeding termite
Reticulitermesflavipes. Environ. Microbial. 2:436-449.

Tschech, A., and N. Pfennig. 1984. Growth yield increase linked to caffeate
reduction in Acetobacterium woodii. Arch Microbiol 137 :163-167.

Venters, D. 1971. The mode of systemic insecticidal action of sulphaquinaxaline
against mosquitoes. Trans. R. Soc. Trap. Med. Hyg. 65:24-25.

Venters, D. 1972. Folate metabolism in the muscles of Aedes aegypti. Insect
Biochemistry 2:153-160.

65

Chapter 3

Taxonomy of Treponema stains ZAS-1, ZAS-2, and ZAS-9

Introduction

Treponema strains ZAS-l, ZAS-2, and ZAS-9 are currently the only termite gut
spirochetes isolated and studied in pure culture (4, 5). The results reported by Leadbetter
et al. (4) and in Chapter 2 of this dissertation established that strains ZAS-1 and ZAS-2
are highly similar to each other in terms of morphology, 16S rRNA similarity (98%), and
in exhibiting a homoacetogenic metabolism. By contrast, strain ZAS-9 is readily
distinguishable from ZAS-1 and ZAS-2 morphologically (ZAS-9 cells are longer and
have Shorter wavelengths) and ZAS-9 is not a homoacetogen. Moreover, the 16S rRNA
sequence of ZAS-9 is only 93% and 92% similar to that ZAS-1 and ZAS-2, respectively
(5). ZAS-9 also shows 10-fold higher nitrogenase activities than either of the other
strains in vitro and is capable of unambiguous N2-dcpendent growth, suggesting that
ZAS-9 may be a more significant provider of fixed nitrogen to termites than ZAS-1 or
ZAS-2 (5).

This chapter presents a more thorough characterization of the nutritional and
physiological properties of strain ZAS-9 and compares a number of taxonomically
relevant properties between the three ZAS strains. Overall, the results suggest that ZAS-
l and ZAS-2 should be regarded as two strains of single new species in the genus
Treponema, whereas strain ZAS-9 should be considered a separate new species of

Treponema distinct hour that accommodating ZAS-1 and ZAS-2.

66

Material and Methods

Media and Cultivation Conditions

Routine cultivation of Treponema strains ZAS-1 and ZAS-2 was as described in
Chapter 2 with one exception: in medium 2YACo, the cofactor solution was replaced
with folinate (500 ng/ml). Cultivation of strain ZAS-9 was identical to the other strains,

except that medium 2YACo supplemented with fermentable sugars was used.

Substrate Utilization Studies

Substrate utilization by strain ZAS-9 was determined with cells growing under an
atmosphere of 80% N2/20% CO2 (vol/vol) except during testing of H2 as a growth
substrate (80% H2/20% CO2). Cultures were grown at 30°C. An increase in cell yield
(>20%) in the presence of a substrate, as compared to its absence, was taken to indicate
utilization of the substrate as an energy source. Cell growth was determined by
measuring the optical density (OD) of cultures at 600 nm on a Milton Roy Spectronic 20
calorimeter. OD readings were converted to cell number by reference to a standard curve
relating these quantities. Substrate-product carbon balances were determined under
conditions of substrate-limited growth. Organic acid production was determined by using
high performance liquid chromatograph (HPLC) with refractive index detection (2).
Ethanol production was enzymatically measured (ethanol assay kit 332-A; Sigrna-

Aldrich, St. Louis, Missouri). Production of H2 was determined by gas chromatography
(3).

67

Genomic G-l-C Content Determination

The guanosine plus cytosine (G+C) content of genomic DNA was calculated fi'om
the apparent ratios of deoxyguanosine and deoxythymidine (dGuo/dThd), which were
determined by using the HPLC method of Mesbah et al. (6), with the following
modifications: genomic DNA was harvested from mid-log phase cultures by using a
Qiagen genomic DNA isolation kit with a 500/G purification column (Qiagen Inc.,
Chatsworth, California). A Shimadzu HPLC equipped with a model SPD-lOA UV/V IS
detector (Shimadzu Corp., Kyoto, Japan) and an Alltech Alltirna C18 column (250 x 4.6
mm, 5 pm particle size) (Alltech Assoc., Deerfield, Illinois) were used. Data was

analyzed by using the Shimadzu 82 Start (v. 7.1) software package.

Results
Morphology of Treponema strains ZAS-1, ZAS-2, and ZAS-9
The morphology of strains ZAS-9 and ZAS-2 is depicted in Figure 3.1. Cells of
ZAS-2 were 3 to 7 pm in length and had a wavelength of 2.3 :l: 0.2 um (11 = 15). Strain
ZAS-1 was virtually indistinguishable fi'am ZAS-2 on a morphological basis.
In contrast, cells of ZAS-9 were longer (10 to 12 pm in length by 0.2 to 0.3 pm in width)
than cells of ZAS-1 and ZAS-2 and displayed a shorter wavelength [1.1 :l: 0.1 pm (n =

15)], giving the cells a more tightly coiled appearance.

68

 

 

 

 

 

 

 

 

 

Figure 3.1. Morphology of Treponema strains ZAS-2 (A & B) and ZAS-9 (C
and D) by phase contrast (A & C) and transmission electron microscopy (B &
D). Insets in panels A and C show cells of each strain, with wavelengths
indicated by arrowed bars. Periplasmic flagella are indicated by arrows in
panels B & D. Strain ZAS-1 is morphologically indistinguishable from strain
ZAS-2. Bars. 5 um (A and C), 2.5 pm (A and C insets), and 0.1 pm (B & D).
Panel B is adapted from reference 4. and is presented here for comparison.

69

Nutrition, Growth, and Fermentation Products of strain ZAS-9

Energy sources used by ZAS-9 for growth are presented in Table 3.1.
Carbohydrates were the only growth-supportive substrates; no growth occurred in the
presence of organic acids or the amino acid glycine. ZAS-9 was also incapable of using
H2 (+ CO2) as an energy sources, and no enhancement of growth (in terms of rate or cell
yield) was observed when H2 was provided along with a fermentable carbohydrate (data
not shown). ZAS-9 did not grow in medium 2YACo in the absence of added growth
substrates. Major products of maltose fermentation by ZAS-9 were acetate, adrenal, H2,
and CO2 (Table 3.2). The carbon and electron recoveries (90% and 88%, respectively),
and the lack of any other detectable products in HPLC profiles, suggested that these were

the only fermentation products of strain ZAS-9.

G+C Content of DNA

Genomic G+C content of the three ZAS strains were similar: ZAS-1, 51.04 :1:
0.09 mol%; ZAS-2, 50.87 :t 0.17 mol%; ZAS-9, 50.02 i 0.11 mol% (n=6 for each strain).
G+C content was also measured for the DNA of the free-living spirochete Spirochaeta
aurantia J 1 (65.60 :t 0.05 mol%) for comparison. As a control, the G+ C content was
determined for Myxococcus xanthus DK1622, a strain previously assayed by the same
method (6); the value determined here (67.41 :t 0.07 mol%) was in good agreement with
that reported in the previous study (67.55 i 0.02 mol %). For all strains examined, peaks
corresponding to the four unmodified deoxynucleosides (dGua, dThy, dAde, and dCyt)

were the only peaks observed in HPLC chromatograms.

7O

Table 3.1. Substrates Used as Energy Sources by Treponema strain ZAS-9

 

 

Substratea Doubling Time (h) Yield (cells/ml)
Glucose 35 3.1x10“
Fructose 37 2.6x108
Ribose 42 1.4x108
Xylose 35 2.8x108
Maltose 47 5.6x108
Cellobiase 45 4.8x108

 

“All substrates were provided at a final concentration of 5 mM except for H2 (see
methods). H2, mannitol, arabinose, sucrose, trehalose, glycine, lactate, pyruvate, and uric
acid were not utilized. No growth was observed in medium 2YACo in the absence of

added substrates.

71

Table 3.2. Fermentation Products of T reponema strain ZAS-9

 

 

Product mmol per 100 mmol
maltose“

Acetate 280

Ethanol 80

H2 520

C02 360”

 

“Carbon recovery = 90%; electron recovery = 88%.

bAssumed to be equal to the sum of acetate plus ethanol

72

Discussion

As reported by Leadbetter et al. (4) and in Chapter 2, ZAS-1 and ZAS-2 are both
homoacetogens capable of growth on H2 + CO2 as well as a variety of organic
compounds. The only significant differences observed between the two strains have been
the ability of ZAS-2 to grow on methoxylated aromatic compounds (albeit very slowly)
and its apparent reliance on H2 as a growth stimulant. ZAS-1 and ZAS-2 are otherwise
highly similar to each other in terms of morphology, physiology, 16S rRNA sequence,
and genomic G+C content. Given these results, it seems reasonable to conclude that
ZAS-1 and ZAS-2 represent two strains of a single species.

By contrast, strain ZAS-9 differs substantially from ZAS-1 and ZAS-2 in a
number of taxonomically relevant properties, including cell morphology and its capacity
for nitrogen fixation and N2-dependant growth. Unlike ZAS-1 and ZAS-2, ZAS-9 is not
a homoacetogen, and does not consume H2 during growth. In fact, H2 is a major product
of fermentation by ZAS-9. ZAS-9’s non-homoacetogenic nature of ZAS-9 is also
supported by the fact that cells possess a hydrogenase activity (1.15 U/mg protein), but
no formate dehydrogenase or CO dehydrogenase (J. A. Breznak, personal
communication). The demonstration of H2 evolution by ZAS-9 and H2/C02-acetogenesis
by ZAS-1 and ZAS-2 suggests that interspecies H2 transfer between spirochetes may be
an important component of H2 turnover in the termite hindgut.

In addition to the genomic G+C content data reported here, other genomic
properties of the ZAS strains have been the subject of studies by other members of our
laboratory. As with G+C content, genome sizes determined by pulsed field gel

electrophoresis of macrorestricted genomic DNA were fairly similar between the three

73

strains: ZAS-1, 3.46 Mb; ZAS-2, 3.84 Mb; ZAS-9, 3.9 Mb (Brendan Keough & Kwi
Kim, personal communication). Each strain also possessed two copies of the small
subunit rRNA-encoding gene (Kwi Kim, personal communication).

The S. aurantia J1 genomic G + C content determined here (65.6 mol%, as
measured by direct HPLC analysis of deoxynucleosides) was in relatively good
agreement with value previously reported by Breznak and Canale-Parola (1) for this
organism as measured by the equilibrium density centrifugation (pCsCl) method (64.5
mol%). Both values, however, were significantly higher than the G+C content inferred
by the thermal denaturation (Tm) method (60.4 mol%) in the same study, in which a
discrepancy of approximately 4 to 6 mol% G+C content between the pCsCl and Tm
methods was also reported for the spirochetes S. stenostrepta and S. Iitoralis (1). While
the basis for this discrepancy is unlmown, the results of this study suggest that genomic
G+C contents for free-living spirochetes are more accurately measured by direct HPLC
quantification of deoxynucleosides or density centrifugation than by thermal
denaturation.

In summary, T reponema strains ZAS-1 and ZAS-2 are sufficiently similar to be
considered two strains of a single new species, whereas the morphological and
physiological differences displayed by strain ZAS-9 support the assignment of this strain
to a separate new species. Latinate species epithets for these organisms are currently
under consideration. These results also further our knowledge of diversity in terrnite-
associated spirochetes; the previously reported morphological and phylogenetic diversity
of these organisms appears to be paralleled by physiological diversity (of an as yet

unknown extent). Our current understanding of the ZAS strains suggests that the primary

74

contribution of ZAS-l and ZAS-2 to termite nutrition is through H2/CO2-acetogenesis,
while ZAS-9 is likely to be of greater importance in the provision of fixed nitrogen to

their hosts.

Acknowledgements
Thanks to Kristina Stredwick for providing cultures of Myxococcus xanthus DK1622 and
to Kwi Kim, Brendan Keough, and Dr. John Breznak for sharing results of their own

ongoing studies.

75

References

Breznak, J. A., and E. Canale-Parola. 1975. Morphology and physiology of
Spirochaeta aurantia strains isolated from aquatic habitats. Arch Microbial
105:1-12.

Breznak, J. A., and J. M. Switzer. 1986. Acetate synthesis from H2 plus CO2 by
termite gut microbes. Appl Environ Microbiol 52:623-630.

Breznak, J. A., and J. S. Blum. 1991. Mixotrophy in the termite gut acetogen,
Sporomusa termitida. Arch Microbiol 156:105-110.

Leadbetter, J. R., T. M. Schmidt, J. R. Graber, and J. A. Breznak. 1999.
Acetogenesis fiom H2 plus CO2 by spirochetes from termite guts. Science
283:686-689.

Lilburn, T. C., K. S. Kim, N. E. Ostrom, K. R. Byzek, J. R. Leadbetter, and
J. A. Breznak. 2001. Nitrogen fixation by symbiotic and flee-living spirochetes.
Science 292:2495-2498.

Mesbah, M., U. Premachandran, and W. B. Whitman. 1989. Precise
measurement of the G + C content of deoxyribonucleic acid by high performance
liquid chromatography. Int J Syst Bacterial 39:159-167.

76

Chapter 4

Interspecies Cofactor Transfer Supports the Folate

Requirement of Homoacetogenic Termite Gut Spirochetes

Introduction

Folate is found in all living cells, playing a central metabolic role as a carrier of
one-carbon (C1) groups. In particular, the synthesis of purines, pyrimidines, and several
amino acids require folate cofactors (9). “Folate” is actually a generic term referring to
folic acid (i.e. pteroylpolyglutarnate) compounds, all of which are composed of a
pteridine ring, a p-aminobenzoate moiety, and one or more glutamate residues (Figure
4.1). Individual folate derivatives vary in reduction state, nature and position of C1
substitution, and degree of polyglutamation. Intracellular folates generally occur at the
biologically active tetrahydrofolate (TI-IF) level of reduction, but are otherwise diverse in
form and highly variable between organisms (8). While the majority of animals require a
dietary source of folates, plants and bacteria are generally capable of de novo folate
synthesis.

In Chapter 2, T reponema strains ZAS-1 and ZAS-2 were shown to have a strict
requirement for folate. This was particularly surprising for homoacetogenic organisms,
since three enzymes on the methyl-forming branch of the Wood/Ljungdahl pathway of
acetogenesis require THF as a cofactor (27). Other homoacetogens appear to be capable

of synthesizing folate, and a number of acetogens grown in defined media have no

77

A

0H ?
—NH
k.“ N/

NH2

Pterin Ring p-Aminobenzoate

B. on

)\ I H
\ H
NH2 N m H

COOH

Glutamate

Figure 4.1. Generalized structure of folate compounds (A) and the
pteridine portion of the reduced form of folate compounds,
tetrahydrofolate (B). Individual derivatives vary in reduction state, C1
substitution (at N5 or N10), and number of glutamyl residues (generally

from 1 to 9).

78

apparent folate requirements (12, 26, 39). The absence of the ability to synthesize folate,
however, is not uncommon in host-associated organisms, including the spirochetes
T reponema bryantii (42) and T. phagedenis (43).

In considering potential sources of folate compounds for the ZAS strains in the
termite hindgut, it was hypothesized that the most likely candidates would be other
members of the gut microbial community. Folate secretion has been observed in
members of diverse bacterial genera, including Bacillus, Pseudomonas, Aeromonas,
Serratia (l 8), Propionibacterium (l6), and Bifidobacterium (10). Moreover, production
of folate by bacteria has long been thought to play an important role animal nutrition (37)
and has been clearly demonstrated in a number of gastrointestinal systems (6, 7, 22, 29,
37,38)

Research presented in this chapter examines the folate requirements of
Treponema strains ZAS-1 and ZAS-2 in greater detail and identifies folate-secreting
bacteria from the hindgut of the termite Zootermopsis angusticollis (the original source of

the ZAS strains) that are likely to provide folate to the ZAS strains in vivo.

Material and Methods
Termites
Specimens of Zootermopsis angusticollis were collected in the San Gabriel
Mountains of southern California and maintained in the laboratory on Pinus ponderosa
stump material in polypropylene containers. Specimens were used within one month of

their collection.

79

Folate Compounds

Folic acid, dihydrofolate (DHF), tetrahydrofolate (THF), 5-methyltetrahydrofolate (5-
CH3-THF), 5-10-methylenetetrahydrofolate (5-10-CH2-THF), 5-formyltetrahydrofolate
(5-HCO-THF; folinate), lO-formylfolate (IO-HCO-folate), and di- and triglutamate forms
of folic acid were obtained fi'om Schirck’s Laboratories (Jana, Switzerland). Standard
solutions of folate compounds were prepared as described in Kdnings et al. (21) and
contained 10 mM 2-mercaptoethanol (final concentration) and 2% Na - ascorbate (w/v) to
prevent the oxidation of reduced folate compounds. All preparative and analytical
procedures involving folate compounds or culture filtrates were carried out under reduced

lighting conditions to minimize photochemical degradation.

Media and Cultivation Methods

Routine growth of Treponema strains ZAS-1 and ZAS-2 was as described in
Chapter 2, with the following exception: medium 2YACo was modified to medium
2YAFo by replacing the ll-cofactor mixture (25) with 500 ng/ml folinate (final
concentration). Medium 2YA was identical to 2YAFo, but lacked folinate. Culture
media for folate-excreting strains (ZFX strains) varied according to experimental
conditions. Medium GMl was adapted from a bifidobacterial enrichment medium TPY
(2) and consisted of (in grams per liter): glucose, 15; tryptone (Difco), 10; phytone
(BBL), 5; yeast extract (Difco), 2.5; cysteine - HCl, 0.5; K2HP04, 2; MgCl2 - 6 H20, 0.5;
ZnSO4 - 7 H2O, 0.25; CaCl2, 0.15; FCC13, 0.001; Tween 80 (1 ml/L). Plates for anoxic
incubations were supplemented with sterile PdCl2 (30 mg/L final concentration) after

autoclaving. Medium GM2 contained (g/l): NaCl, 1.0; KCl, 0.5; MgCl2-6H2O, 0.4;

80

CaCl2-2H2O, 0.1;NH4C1, 0.3; KH2PO4, 0.2; Na2SO4, 0.15; NaHCO3, 5.8; 3-N-
[morpholino] propanesulfonic acid (MOPS) (10 mM final concentration) and 0.5% (v/v)
yeast autolysate. Trace element solution SL1 1, selenite-tungstate solution, 7-vitamin
solution, and vitamin-B12 solution were added as described by Widdel and Bak (50).
Complex medium GM3 contained 0.2% yeast extract, 0.2% peptone, and 20 mM glucose.
Semi-defined medium GM4 contained (g/l): NaCl, 1.0; KCl, 0.5; MgC12-6H20, 0.4;
CaCl2-2H2O, 0.1; NH4C1, 0.3; KH2PO4, 0.2; Na2SO4, 0.15; NaI-ICO3, 5.8; casamino acids
(Difco), 5; MOPS (10 mM). Afier autoclaving, medium GM4 was supplemented with
(g/l): glutathione, 0.01; asparagine, 0.04; glutarnine, 0.04; uracil, 0.02; adenine, 0.01;
guanine, 0.01; glucose (20 mM final concentration) and trace element and vitamin
solutions described in Van Neil et al. (47). Routine cultivation of ZF X strains was on
plates of Reinforced Clostridial Medium (Difco) containing 2% agar at 30°C. All

described were at an initial pH of 7.2-7.4.

Enrichment, Enumeration, and Isolation of Termite Hindgut Heterotrophs

Worker larvae of Z. angusticollis were degutted with forceps, removing any
attached midgut segments fiom hindguts. Two extracted hindguts were placed in 5 ml of
dithiothreitol (DTT)-reduced buffered salt solution (24) and homogenized while held in
an anoxic glove box (3). Enumerations were performed by preparing a serial 10-fold
dilution series of gut homogenate in tubes of anoxic ZFX medium. Aliquots (0.1 ml) of
the resulting dilutions were spread in triplicate on plates of GMl medium containing 2%
agar and incubated under anoxic (95% N2: 5% H2), hypoxic (98.5% N2: 1.5% 02), or oxic

(air) conditions at 25°C. Colonies on the high dilution plates were enumerated, then

81

grouped by colony and cell morphology. and representatives were selected for folate-

secretion bioassays.

Folate Secretion Biaassays

The folate secretion bioassay was a modification of the folate diffusion bioassay
described by Hewitt and Vincent (15). The folate-requiring bioassay organism,
Enterococcus hirae was obtained from the American Type Culture Collection (ATCC
8043). Liquid cultures of E. hirae were grown overnight in brain heart infusion (Difco)
at 37°C, and 0.25 ml of the culture was transferred to 25 ml of 50% strength AOAC
folate assay medium (AOAC-F A) (Difco) and was again incubated at 37°C overnight.
Twenty ml of the resultant culture was added to 250 ml AOAC-F A medium containing
2% agar, which had been autoclaved and cooled to 50°C. This suspension was dispensed
into Petri plates and used for bioassays. To test for folate secretion, colonies of interest
were picked and patched onto the assay plates and incubated overnight under oxic,
hypoxia, or anoxic conditions. Formation of satellite E. hirae colonies around colonies of
the test organism was taken to indicate secretion of a putative folate compound(s).
Bifidobacterium infantis strain S12 (ATCC 15697), a known folate secretar (10), was
used as a positive control. Strains yielding a positive test result were designated ZFX

strains, streaked for isolation, and subjected to further characterization.

Nucleotide Sequence Analysis of 16s rDNA
The 16s rDNA gene was PCR amplified from putative folate-secreting strains

using primers 8f(5'-AGAGTTTGATCCTGGCTCAG-3') and 1492r (5'-

82

GGTTACCTTGTTACGACT‘T-3'). Each 100-ul PCR reaction mixture contained the
primers (30 pM each), deoxynucleoside triphosphates (Boehringer-Mannheim; 50 11M
each), MgCl2 (2 mM), 2.5 U of T aq polymerase (Gibco), 10 pl of PCR buffer (supplied
with enzyme) and a small amount of colony material. The reactions were performed with
a Gene Amp model 9600 thermocycler (Perkin-Elmer). PCR amplifications consisted of
a 3-min hold at 95°C, followed by 30 cycles consisting of 30 s at 95°C, 30 s at 55°C, and
45 s at 72°C. After thermocycling, an additional extension was performed for 10 min at
72°C. Positive controls contained E. coli genomic DNA and negative controls contained
no template. After amplification, PCR products were purified using a QIAquick PCR
purification kit (Qiagen, Valencia, CA). and subjected to restriction fragment length
polymorphism (RF LP) analysis to eliminate redundancy (23). Each amplirner was
digested with RsaI and HpaII (New England Biolabs) and the resulting fiagments were
resolved on 2.5% NuSieve gel (FMC BioProducts) to yield a restriction fiagment length
polymorphism (RF LP) pattern. One representative of each RFLP banding pattern was
selected for sequencing. Nearly full length (~1500 nt) 16S rDNA sequences were
determined using an ABI PRISM 3100 Genetic Analyzer and overlapping eubacterial
sequencing primers: 8f, 339f, 515f, 700f, 776f, 9341', 1100f, 337r, 531r, 6851‘, 1100r, and
1492r. Phylogenetic analysis was performed using the Ribosomal Database Project
sequence database (28) and the ARB software package (www.biol.chemie.tu-
muenchen.de). Sequences were aligned using the ARB automatic aligner, followed by
manual correction of ambiguous regions. A maximum likelihood method (fastDNAml)

was used to generate phylogenetic trees (34).

83

Nutritional and growth studies

Substrate utilization was assessed by using medium GM2 as a basal medium to
which test substrates were added. Media were held in rubber stoppered 18 mm anaerobe
tubes (Bellco Glass; Vineland, NJ) under atmospheres of 100% N2 or in foam-stoppered
nephlometry flasks under air. Incubations were at 30°C. An increase in cell yield
(>20%) in the presence of a substrate, as compared to its absence, was taken to indicate
utilization of the substrate as an energy source. Cell yield was determined by measuring
the optical density (OD) of cultures at 600 nm with a Milton Roy Spectronic 20
calorimeter. Soluble metabolic products were detected by high performance liquid
chromatography (HPLC) with refractive index detection (3), and H2 production was
measured by gas chromatogaphy (4).

To prepare culture filtrates of ZF X strains, sub-samples of stationary phase
cultures gown in medium GM2 were supplemented with 10 mM 2-mercaptoethanol
(final concentration) and 2% Na - ascorbate (w/v) and adjusted to pH 7. Samples were
centrifuged at 10,000 g for 10 minutes, and supematants were degassed, flushed with N2,
passed through a 0.2 pm syringe filter, and stored under N2 at -80°C until analysis.
Production of folate compounds supporting the gowth of Treponema strains ZAS-1 and
ZAS-2 was assessed by replacing folinate in medium 2YAFo with 10% (v/v) ZFX culture
filtrates. Cultures of ZAS-1 and ZAS-2 that served as inocula in these experiments were
gown in folinate-free 2YA medium through two transfers to reduce residual folinate

carried over with the inoculum.

84

Identification of Folate Compounds

Identification of unknown folate compounds was done by comparing HPLC
retention times with those of folate standards. The HPLC protocol used was a
modification of the procedure described by Pheiffer et al. (35). Folates were separated by
using a Shimadzu HPLC system (Shimadzu Corp., Kyoto, Japan) equipped with an
Alltech Alltima C13 column (250 x 4.6 mm, 5 pm particle size) (Alltech Assoc.,
Deerfield, Illinois). Gradient elution was performed with acetonitrile and 33 mM
phosphoric acid (pH 2.3) was performed as follows: acetonitrile was held at 5% for the
first 9 minutes, then was raised linearly to 7% over the next 13 min, then raised linearly
to 16% for the next 9 min, held at 16% for 14 nrin, then raised to 25% over two min, held
at 25% for 5 min, and finally decreased to 5% acetonitrile within 3 min. The flow rate
was 0.8 mllmin, and sample injection voltune was 20 pl. Separated folate compounds
were monitored by using a Shimadzu model SPD-lOA UVN IS detector set at 280 nm
and a model RF-lOA fluorescence detector set 280 nm excitation and 356 emission
wavelengths for reduced folate compounds and 360/460nm for 10-HCO-folate (20). Data
was analyzed with the Shimadzu EZ Start (v. 7.1) software package. F olate
concentrations reported for culture filtrates are corrected for folates carried over in
culture inocula.

Rat plasma conjugase (RPC) was used to remove excess glutamate residues (i.e.
more than a single glutamate) from folate compounds, as these interfere with HPLC
analysis. RPC was prepared as described in Pfeiffer et al. (35) using rat plasma obtained

from Pel-Freez Biologicals (Rogers, AR). Activity of the crude RPC preparation was

85

confirmed as described previously (35). ZFX strains were gown in medium GM4, and
filtrates were collected as described above. RPC preparation was added to filtrates at 5%
(v/v), and the mixture was incubated at 37°C for one hour, heated to 100°C for 5 min,
and cooled an ice for five minutes. Samples were centrifuged at 10,000 x g for 10

minutes at 5°C. Supematants were collected and stored at -80°C until analysis.

Results

Folate Requirements of Treponema strains ZAS-1 and ZAS-2

Treponema strains ZAS-1 and ZAS-2 were tested for their ability to use folates in
the three major states of reduction (folic acid, DHF, and THF) and folinate (5-HCO-THF)
(Figure 4.2). ZAS-2 was able to use all of the folates tested, whereas ZAS-1 gew only
when provided with THF or folinate. Growth of ZAS-1 appeared to be inhibited in the
presence of folate and dihydrofolate compared to the limited gowth seen in the negative
control cultures. Addition of 500 ng/ml folinate to these “inhibited” cultures eliminates
this effect (data not shown), suggesting that the relatively large pools of folate and
dihydrofolate may bind to (and thus rendered inaccessible) the small amounts of gowth
supportive folinate carried over with the inoculum (36). Provision of p-aminobenzoate,
glutamate, various pteridines (pterin, 6-hydroxymethlypterin, pteroic acid), and the pterin
precursor guanosine 5’-triphosphate (GTP) did not alleviate the requirement of either

strain for preformed folates.

86

0.8 : ZAS-1

 

 

 

 

fit

0 '5'0'100'150 200'250'

 

Time (hrs)

Figure 4.2. Folate utilization by 'li’eponema strains ZAS-1 and ZAS-2. All
folate compounds were provided at a final concentration of 500 ng/ml. The
specific compounds were: folic acid (El), dihydrofolate (O), tetrahydrofolate
(A), 5-formyltetrahydrofolate (i.e. folinic acid) (V), and no addition (0).

87

Isolation of Folate-Secreting Bacteria from Termite Hindguts

Colonies derived from the highest dilutions of Z. angusticollis gut homogenates
were sorted on the basis of morphological characteristics (both colonial and microscopic)
and incubation condition (oxic, anoxic, or hypoxic). A total of 32 colonies were then
bioassayed for folate secretion by using Enterococcus hirae, an organism with a strict
folate requirement. Eighteen of the screened colonies supported the gowth of satellite
colonies of E. hirae in the folate-free bioassay medium, implying secretion of a putative
folate into the surrounding medium (Figure 4.3). These colonies were designated “ZFX”
strains. Bifidobacterium infantis str. S12, a known folate secretar (10), was used as a
positive control organism in bioassays. The diameter of the zone of E. hirae satellite
colonies surrounding ZF X colonies was similar to that observed surrounding B. infantis
812 colonies. Of the colonies yielding negative results, thirteen failed to gow on the
assay plates (perhaps indicating a requirement for folate in these isolates) and one gew
but did not support the gowth of E. hirae. Colonies yielding positive results were

streaked for isolation and subjected to further characterization.

Characterization of Folate-Secreting Strains

16S rDNA genes were PCR amplified from genomic DNA of the eighteen
putatively folate-secreting isolates using bacteria specific primers. Analysis of the
eighteen clones revealed two distinct RF LP banding patterns. Nearly complete 16S
rDNA sequences were obtained for one isolate representing each RFLP pattern,
desigrated strains ZFX-1 and ZFX-2. Phylogenetic analysis of these sequences indicated

that strain ZFX-1 gouped with the genus Serratia, whereas ZFX-2 gouped within the

88

('35-- W

"i\ /

Bioassay Organism

 

Figure 4.3. Folate secretion assay of strains ZFX-1 and ZFX-2.
Putative folate secretion is indicated by the formation of satellite
colonies of the folate-requiring bioassay organism, Enterococcus hirae,
surrounding surface colonies of ZFX-1 (inset A) and ZFX-2 (inset B).

89

genus Lactococcus (Figure 4.4). ZFX-1 had 99.7% 16S rRNA sequence similarity to
Serratia grimesii and 99.1% similarity of Serratia liquefaciens. Strain ZFX-2 was 99.9%
similar in 168 rRNA sequence to Lactococcus lactis subsp. lactis.

All ZFX-1 and ZFX-2 type isolates gew well on plates incubated under oxic,
anoxic, and hypoxic conditions. Cells of all strains corresponding to the ZFX-1 RFLP
pattern were motile gam-negative rods approximately 0.6-0.8 x 1.5-2 pm in size, and all
ZFX-2 type isolates were non-motile coccoid cells approximately 0.5-0.8 x 0.9-1.1 pm in
size (Figure 4.5). Taken together with the RF LP data, these results suggest that all folate-
secreting isolates obtained under the three enrichment conditions were similar, if not
identical, to representative strains ZFX-1 or ZFX-2. Mean numbers of strains ZFX-1 and
ZFX-2 occurring per Z. angusticollis gut were estimated to be 5.1 i 0.6 x 105 and 5.4 i
0.7 x 105 (n = 3) respectively.

A variety of hexoses, pentoses, and disaccharides were used as gowth substrates
by both strains during gowth under aerobic conditions; ZFX-1 was additionally capable
of using organic acids (Table 4.1). During anaerobic gowth on glucose, strain ZFX-1
produced a mixture of succinate, acetate, ethanol, 2-3 butanediol, and H2. No organic
products were detected during aerobic gowth of ZFX-l on glucose. Under both oxic and
anoxic conditions, ZFX-2 produced lactate as the sole end product fi'om glucose. Lactate
production was stoichiometrically consistent with glucose consumption, and the

fermentation pattern was not altered during gowth in a complex medium (GM3).

90

Serratia plymuthica

Aranicola proteolyticus
Serratia proteomaculans
Serratia fonticola

 
  
  
  

Serratia grimesii
Serratia quuefaciens
ZFX-1

r——[7Aphid S-Symbiants

Serratia fican'a
Serratia entomophila

_l -— Serratia odorifera
Serratia marcesens

0.10

 

 

 

B ZFX-2
Lactococcus lactis subsp. lactis
Lactococcus lactis subsp. cremorum
Lactococcus lactis TmL05
Lactococcus garvieae
Lactococcus plantarum
l— Streptococcus infantan’us

Streptococcus bovis
Streptococcus equi

Streptococcus sanguinis
l l— Streptococcus pyogenes
Streptococcus canis
ll

,, Escherichia coli

 

 

 

 

 

 

 

 

 

 

 

 

0.10

 

Figure 4.4. Phylogenetic trees inferred from 16S rDNA sequences (~1500 nt) of
ZFX-1 (A), ZFX-2 (B), and related organisms. A maximum likelihood technique
(fastDNAml) was used to generate the trees. The homologous sequence from
Desulfovibrio senezii was used as an outgroup in tree A (not shown). The scale
bars represent units of evolutionary distance and are based on sequence
divergence.

91

. I
A o
-.-
\ 1 -~
‘ \
’ ’ 5pm
B I
t
a! 0 °
r. '° r-
1.
0.
’Q
3 r. 5pm

 

Figure 4.5. Phase contrast micrographs of Serratia
strain ZFX-1 (A) and Lactococcus strain ZFX-2 (B).

92

Table 4.1. Substrate Utilization by Serratia strain ZFX-1 and Lactococcus strain ZFX-2

 

 

Strain ZFX-l Strain ZFX-2
Substratea Doubling Time (h) Doubling Time (h)
Glucose 1 .30 0.73
Mannitol 1.45 0.73
Ribose 2.65 0.81
Xylose 1.92 0.82
Maltose 1 .22 0.75
Cellobiose n.u. 0.85
Trehalose 1.11 0.81
Acetate 1.85 n.u.
Pyruvate 1 .58 n.u.
Lactate 1.67 n.u.
Fumarate 1.59 n.u.
Malate 1.50 n.u.
Succinate l . 1 8 n.u.

 

'All substrates were provided at a final concentration of 10 mM except for methoxylated
aromatic compounds (2mM). Substrates tested but not utilized are indicated by n.u.
Arabinose, glycine, formate, methanol, ferulate, vanillate, syringate, and
trimethoxybenzoate were not utilized by either strain.

93

Folate Production by Serratia strain ZFX-2 and Lactococcus strain ZFX-1

To characterize the putative folate compound(s) secreted by the ZF X strains,
cultures were grown in anoxic semi-defined medium GM4. Culture filtrates were
collected at the time of inoculation (i.e. 0 hrs.) and at the end of logarithmic growth (9
hrs. for ZFX-1, 7 hrs. for ZFX-2). Reducing agents (2-mercaptoethanol and
Na-ascorbate) were immediately added to protect folates from oxidative damage, and
filtrates were then treated with rat plasma conjugase (RPC) to remove excess glutamate
residues prior to analysis.

Good resolution of folate standards was observed in HPLC analysis using
fluorescence detection for reduced folates and UV absorbance for unreduced forms (see
methods and Figure 4.6). A distinct peak with a retention time identical to that of
authentic folinate (i.e. 5-HCO-THF) was observed in culture filtrates of both ZFX strains
(Figure 4.6). Concentrations of folinate in culture fluids of ZFX-l and ZFX-2 at the end
of growth were 146 ng/ml and 117 ng/ml, respectively (corrected for folinate carried over
in inocula). An additional fluorescent compound with a retention time of 31 minutes was
produced during growth of ZFX-1; the identity of this compound is unknown, but it did
not correspond to any standard folate compound tested under the three detection methods.
No other peaks corresponding to folates were observed using any of the three detection
methods, suggesting that folinate is the sole folate compound secreted by strains ZFX-1
and ZFX-2 at significant levels.

Provision of 10% culture filtrates (v/v) of strains ZFX-l and ZFX-2 supported the
growth of Treponema strains ZAS-1 and ZAS-2 in the absence of added folate

compounds (Figure 4.7). The filtrates used in these experiments were not treated with

94

 

 

 

 

 

 

 

 

 

 

15‘ A
’>‘
é
a) 10-
O
C
0)
O
(I)
9
.3 5‘
Ll.
O‘N‘ .7
20 2'5 3'0 35 4o
15- B
9
E,
8 10
C
Q)
0
(D
9
s 5*
LI.
o~”"“'
20 25 30 35 4o
15* C 2
S‘
E
8 10-
c 1
8 3
(I)
§
2 5‘
LL
0:
20 25 3'0 35 40
Time (min)

Figure 4.6. HPLC chromatograms showing fluorescent (280/359 nm)
compounds present in culture filtrates of strains ZFX-1 (A) and ZFX-2 (B)
and for a standard mixture (C) of THF (peak 1, 1 ng), 5-CH3-THF (peak 2,
1 ng), and 5-HCO-THF (peak 3, 5 ng). In panels A & B, dashed and solid
lines indicate samples taken at the time of inoculation and at the end of
logarithmic growth, respectively. The peak eluting at 31 minutes in panel
A could not be identified, but did not correlate to any folate standard.

95

0.8 3

ZAS-1 If.

095

 

0.01-
' ifrfl ' r '

'I'I I'I'I'I
0 50 100 150 200 250300350400 450 500

0.8 -

ZAS-2

A L44

 

0,01. . .. . .
o ' 5'0 '160'1éo'zoo'zéo'aoo'356466456560
Time (hrs)

 

 

Figure 4.7. Growth of Treponema strains ZAS-1 and ZAS-2 in folate free
media supplemented with folinate (O), cultures filtrates of Serratia strain
ZFX-1 (O) and Lactococcus strain ZFX-2 (El). Culture filtrates were
added at 10% (v/v) final concentration. Folinate was provided at 500
ng/ml final concentration. Negative controls (A) were supplemented with
10% uninoculated GM3 medium.

96

RPC, indicating that ZAS-1 and ZAS-2 are capable of using the folinate compounds
secreted by the ZF X strains in their native states of polyglutamation. Provision of ZFX
culture filtrates did not replace the requirement of either spirochete for yeast autolysate in
medium 2YACo, and addition of ZAS culture filtrates did not result in any discernible

enhancement of folate production by either ZF X strain (date not shown).

Discussion

The observed folate requirement of Treponema stains ZAS-1 and ZAS-2 in vitro
suggested that a source of folate compounds must exist within the termite hindgut. Given
the critical nature of this cofactor in the homoacetogenic metabolism of these organisms
[(27); Chapter 2] and the importance of C02-reductive acetogenesis in termite nutrition
(5, 45), determination of the source of folates in situ was deemed important to a better
understanding of the physiological ecology of the ZAS strains and the termite hindgut
symbiotic system as a whole.

There are several potential sources of folate in termite hindguts. Since plants are
capable of folate synthesis, wood ingested by termites is one possibility. However, the
primarily structural role of lignocellulose in plants suggests that this material is likely be
relatively low in folate, since this cofactor is normally involved in active biosynthetic
processes. The termite host itself is also potential source, although the ability of termites
to synthesize folate compounds is unknown. De novo synthesis of folate has been
demonstrated in mosquito muscle tissue (19, 48, 49), but the majority of insects require a

dietary source of folates. In any event, it seems unlikely that the gut microbial

97

community would have significant access to folate pools existing within host cells. As
such, the most likely source of folates would be other microbes inhabiting the hindgut.

F olate synthesis is a common capability in bacteria, and secretion of folate
compounds has been demonstrated in a variety of bacterial species, including members of
the genera Bacillus, Pseudomonas, Aeromonas, Serratia (l8), Propionibacterium (16),
and Bifidobacterium (10). Bacterial folate production has been observed in a variety of
gastrointestinal systems (7, 29, 37), and host uptake of bacterially produced folates has
been unambiguously demonstrated in rat and human gastrointestinal systems (6, 38). The
presence of folate-requiring bacteria in the bovine rumen (14, 41, 42) and observation of
THF production in rumen enrichment cultures (41) suggests that extensive folate cross-
feeding also occurs in the bovine rumen.

Folate secretion by Z. angusticollis gut isolates Serratia strain ZFX-1 and
Lactococcus strain ZFX-2 was demonstrated by both microbiological assay with E. hirae
and by direct detection of folate compounds in ZF X culture filtrates. Folate
concentrations in ZF X culture fluids were within the range observed for other folate
secreting bacteria, which secrete folates to concentrations 20—160 ng/ml in culture fluids
(10, 16, 18). Levels of folate production similar to those of ZFX-2 (100 ng/ml) have
been reported in Lactococcus lactis strain M61363. In contrast to ZFX-2 though, 90% of
the folate produced by strain M61363 is accumulated intracellularly rather than being
secreted into the environment (17). Levels of folate production and secretion also vary
widely between individual species and strains of propionibacteria and bifidobacteria (10,

16). It has been hypothesized that strains showing the highest levels of folate secretion

98

lack feed-back inhibition mechanisms controlling folate release, a trait which may be
selected for in densely colonized gastrointestinal environments (10).

Although no Serratia strains have been identified in recent culture collections (11,
40, 44) or molecular studies (32, 33) of Reticulitermes sp. termite gut bacteria, several
strains identified as pathogenic Serratia marcescens have been isolated from Z.
angusticollis (13). However, the termites from which strain ZFX-1 was isolated appeared
to be in good health, and no subsequent signs of disease were observed in other members
of the same colony. These observations, and the presence of a relatively large population
of strain ZFX-1 in the gut (~5 x105 cells per gut), suggests that ZFX-1 is more likely to
be a beneficial member of the hindgut microbiota. Strain ZFX-1 showed 96% sequence
similarity to the 16S rRNAs of a number of secondary symbionts (S-symbionts) of aphids
(46) (Figure 4.4). The role of S-symbionts in aphids is currently unknown, but the
primary aphid symbiont Buchnera has been implicated in the provision of the vitamin
riboflavin to their hosts (30). The observation of folate secretion by Serratia strain ZFX-
1 raises the possibility that the related aphid S-symbionts could also be involved in the
provision of vitamins to their hosts.

Lactic acid bacteria (LAB), including Lactococcus species similar to strain ZFX-
2, are frequently cultured from termite hindguts (1 , 11, 40, 44). The population levels
determined for strain ZFX-2 in the Z. angmticollis hindgut are consistent with numbers
of LAB observed in other termite species. LAB are estimated to represent approximately
3% of the total gut bacterial count in the termite Reticulitennesflavipes (44). Lactococci
isolated from the guts of other termites exhibit a heterolactic fermentation under anoxic

conditions (producing lactate plus smaller amounts of ethanol, acetate, and formate) and

99

shift their fermentation pattern to favor mostly acetate production under oxic conditions
(1, 44). In contrast, strain ZFX-2 performs a homolactic fermentation of glucose under
both oxic and anoxic conditions. The lack of some medium component required for
acetate formation seems unlikely, since medium GM2 is similar in composition to the
culture medium used in previous studies of lactococci (l, 44), and growth on a more
complex media (GM3) did not result in any change in the fermentation pattern of ZFX-2.
The inability of strain ZFX-2 to shifi to the more energetically favorable fermentation
pattern favoring acetate production could indicate a lesion in a component of the pyruvate
dehydrogenase complex. Given the wide distribution of LAB among diverse termite
species (1), it is possible that these organisms are significant folate providers in many
hindgut systems. Considering to high degree of variability in folate secretion observed
among closely related strains of other folate-secreting bacteria though, it would be
necessary to evaluate folate secretion in strains isolated from diverse termites to
determine if this is a common property of termite-associated Lactococcus species.

The relatively high levels of folinate secretion observed in strains ZFX-1 and
ZFX-2 suggests that these organisms represent the major source of folate for Treponema
strains ZAS-1 and ZAS-2 in the Z. angusticollis hindgut. In situ confirmation of this
hypothesis, however, presents a considerable technical challenge due to the very small
volume of extracellular fluid in termite hindguts (0.27-3.2 pl) (31) and the likelihood that
secreted folates are efficiently absorbed by folate-requiring members of the dense gut
microbial community. In mammalian gastrointestinal systems, investigators have tracked
the incorporation of radiolabelled p-aminobenzoate into folate compounds, demonstrating

in situ bacterial production and host-uptake of folates (6, 38). This method could prove

100

useful in confirming bacterial folate synthesis in termite hindguts and determining the in
situ role of folate-secretors in providing this vitamin to their termite hosts, as well as to

other members of termite hindgut microbial communities.

Acknowledgments

I wish to thank to Dr. Jesse Gregory (University of Florida) and Dr. Tsunenobu Tarnura
(University of Alabama) for helpful comments on folate assay techniques. Thanks also to
Dr. Jared Leadbetter (California Technical Institute) for providing the Z. angusticollis

termites used in this study.

101

10.

References

Bauer, S., A. Tholen, J. Overmann, and A. Brune. 2000. Characterization of
abundance and diversity of lactic acid bacteria in the hindgut of wood- and soil-
feeding termites by molecular and culture-dependent techniques. Arch Microbiol
173:126-137.

Biavati, B., and P. Mattarelli. 2001. The Family Bifidobacteriaceae. In M.
Dworkin (ed.), The Prokaryotes: an Evolving Electronic Resource. Springer-
Verlag, New York.

Breznak, J. A., and J. M. Switzer. 1986. Acetate synthesis from H2 plus C02 by
termite gut microbes. Appl Environ Microbi0152:623-630.

Breznak, J. A., and J. S. Blum. 1991. Mixotrophy in the termite gut acetogen,
Sporomusa termitida. Arch Microbiol 156:105-110.

Breznak, J. A. 1994. Acetogenesis from carbon dioxide in termite guts. In H. L.
Drake (ed.), Acetogenesis. Chapman & Hall, New York.

Camila, E., J. Zimmerman, J. B. Mason, B. Golner, R. Russell, J. Selhub, and
I. H. Rosenberg. 1996. Folate synthesized by bacteria in the human upper small
intestine is assimilated by the host. Gastroenterol 110:991-998.

Coates, M. E., J. E. Ford, and G. F. Harrison. 1968. Intestinal synthesis of
vitamins of the B complex in chicks. Br J Nutr 22:493-500.

Cossins, E. A. 1984. Folates in biological materials, p. 1-60. In R. L. Blakely and
S. J. Benkovic (ed.), Folates and Pterins, vol. 1. John Wiley & Sons, New York.

Cossins, E. A. 2000. The fascinating world of folate and one-carbon metabolism.
Can. J. Bot-Rev. Can. Bot. 78:691-708.

Deguchi, Y., T. Morishita, and M. Mutai. 1985. Comparative studies on
synthesis of water-soluble vitamins among human species of Bifidobacteria.
Agric. Biol. Chem. 49:13-19.

102

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Eutick, M. L., R. W. O'Brien, and M. Slaytor. 1978. Bacteria from the gus of
Australian termites. Appl Environ Microbiol 35:823-828.

Gentbner, B. R. S., C. L. Davis, and M. P. Bryant. 1981. Features of rumen and
sewage sludge strains of Eubacterium Iimosum, a methanol-utilizing and Hz-Coz-
utilizing species. Appl Environ Microbiol 42:12-19.

Grimont, P. A. D., and F. Grimont. 1978. The genus Serratia. Annu Rev
Microbiol 32:221-248.

Herbeck, J. L., and M. P. Bryant. 1974. Nutritional requirements of the
intestinal anaerobe Ruminococcus bromii. App1.Microbiol. 28:1018-1022.

Hewitt, W., and S. Vincent. 1989. Theory and Application of Microbiological
Assay. Academic Press, San Diego, CA.

Hugenholtz, J., J. Hunik, H. Santos, and E. Smid. 2002. Nutraceutical
production by propionibacteria. Lait 82:103-112.

Hugenholtz, J., W. Sybesma, M. N. Groot, W. Wisselink, V. Ladero, K.
Burgess, D. van Sinderen, J. C. Piard, G. Eggink, E. J. Smld, G. Savoy, F.
Senna, T. Jansen, P. Hols, and M. Kleerebezem. 2002. Metabolic engineering
of lactic acid bacteria for the production of nutraceuticals. Antonie Van
Leeuwenhoek International Journal of General and Molecular Microbiology
82:217-235.

Iwai, K., M. Kobasbi, and H. Fujisawa. 1970. Occurence of Crithidia factors
and folic acid in various bacteria. J Bacteriol 104:197-201.

Jaffe, J. J., J. J. McCormack, E. Meymarian, and H. M. Doremus. 1977.
Comparative activity and properties of lactate dehydrogenase, xanthine
dehydrogenase, and dihydrofolate reductase in normal and Brugia pahangi
infected Aedes aegypti. J. Parasitol. 63:547-553.

Kariluoto, M. S., L. T. Vahteristo, and V. I. Piironen. 2001. Applicability of
microbiological assay and affinity chromatography purification followed by high-
perfonnance liquid chromatography (HPLC) in studying folate contents in rye. J
Sci Food Agric 81:938-942.

103

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Konings, E. J. M. 1999. A validated liquid chromatographic method for
determining folates in vegetables, milk powder, liver, and flour. J AOAC Int
82:119-127.

Krause, L. J., C. W. Forsberg, and D. L. O'Connor. 1996. Feeding human milk
to rats increases Bifidobacterium in the cecum and colon which correlates with
enhanced folate status. J Nutr 126:1505-1511.

Laguerre, G., M. Allard, F. Revoy, and N. Amarger. 1994. Rapid identification
of rhizobia by restriction fragment length polymorphism analysis of PCR-
arnplified 16S rRNA genes. Appl Environ Microbiol 60:56-63.

Leadbetter, J. R., and J. A. Breznak. 1996. Physiological ecology of
Methanobrevibacter cuticularis sp. nov. and Methanobrevibacter curvams sp.
nov., isolated from the hindgut of the termite Reticulitennesflavipes. Appl
Environ Microbiol 62:3620-3631.

Leadbetter, J. R., T. M. Schmidt, J. R. Graber, and J. A. Breznak. 1999.
Acetogenesis from H2 plus C02 by spirochetes from termite guts. Science
283:686-689.

Leigh, J. A., F. Mayer, and R. S. Wolfe. 1981. Acetogenium kivui, a new
thermophilic hydrogen-oxidizing, acetogenic bacterium. Arch Microbiol 129:275-
280.

Ljungdahl, L. G. 1984. Other firnctions of folates, p. 555-579. In R. L. B. a. S. J.
Benkovic (ed.), Folates and Pterins, vol. 1. John Wiley and Sons, New York.

Maidak, B. L., G. J. Olsen, N. Larsen, R. Overbeek, M. J. McCaughey, and
C. R. Woese. 1997. The RDP (Ribosomal Database Project). Nucleic Acids Res
25:109-110.

Miller, H. T., and T. D. Lackey. 1963. Intestinal synthesis of folic acid in
monoflora chicks. J Nutr 80:236-242.

Nakabachi, A., and H. Ishikawa. 1999. Provision of riboflavin to the host aphid,
Acyrthosiphon pisum, by endosymbiotic bacteria, Buchnera. J Insect Physiol
45:1-6.

104

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

Odelson, D. A., and J. A. Breznak. 1983. Volatile fatty acid production by the
hindgut microbiota of xylophagous termites. Appl Environ Microbiol 45:1602-
1613.

Obkuma, M., and T. Kudo. 1996. Phylogenetic diversity of the intestinal
bacterial community in the termite Reticulitermes speratus. Appl Environ
Microbiol 62:461-468.

Obkuma, M., and T. Kudo. 1998. Phylogenetic analysis of the symbiotic
intestinal microflora of the termite Cryptotermes domesticus. Ferns Microbiology
Letters 164:389-395.

Olsen, G. J., H. Matsuda, R. Hagstrom, and R. Overbeek. 1994. FasDNArnl: a
tool for construction of phylogenetic trees of DNA sequences using maximum
likelihood. Comput. Appl. Biosci. 10:41-48.

Pfeiffer, C. M., L. M. Rogers, and J. F. Gregory. 1997. Determination of folate
in cereal-grain food products using trienzyme extraction and combined affinity
and reversed-phase liquid chromatography. J Agric Food Chem 45:407-413.

Pfleiderer, W. 1985. Chemistry of naturally occurring pterins, p. 43-114. In R. L.
a. B. Blakely, S. J. (ed.), Folates and Pterins, vol. 2. John Wiley & Sons, New
York.

Rodriguez, M. S. 1978. A conspectus of research on folacin requirements of
man. J Nutr 108:1987-2075.

Rong, N., J. Selhub, B. R. Goldin, and I. H. Rosenberg. 1991. Bacterially
synthesized folate in rat large intestine is incorporated into host tissue folyl
polyglutarnates. J Nutr 121:1955-1959.

Schink, B. 1984. Clostridium magnum sp. nov., a non-autotrophic
homoacetogenic bacterium. Arch Microbiol 137 :250-255.

Schultz, J. E., and J. A. Breznak. 1978. Heterotrophic bacteria present in
hindguts of wood-eating termites [Reticulitermesflavipes (Kollar)]. Appl Environ
Microbiol 35:930-936.

105

41.

42.

43.

45.

46.

47.

48.

49.

50.

Slyter, L. L., and J. M. Weaver. 1977. Tetrahydrofolate and other growth
requirements of certain strains of Ruminococcusflavefaciens. Appl Environ
Microbiol 33:363-369.

Stanton, T. B., and E. Canale-Parola. 1980. Treponema bryantii sp nov, a
rumen spirochete that interacts with cellulolytic bacteria. Arch Microbiol
127:145-156.

Steinman, H. G., V. I. Oyama, and H. O. Schulze. 1954. Carbon dioxide,
cocarboxylase, citrovorum factor, and coenzyme A as essential growth factors for
a saccharolytic T reponeme. J Biol Chem 211:327-335.

Tholen, A., B. Schink, and A. Brune. 1997. The gut microflora of Reticulitermes
flavipes, its relation to oxygen. and evidence for oxygen-dependent acetogenesis
by the most abundant Enterococcus sp. FEMS Microbiol Ecol 24:137-149.

Tholen, A., and A. Brune. 2000. Impact of oxygen on metabolic fluxes and in
situ rates of reductive acetogenesis in the hindgut of the wood-feeding termite
Reticulitermesflavipes. Environ. Microbiol. 2:436-449.

Unterman, B. M., P. Baumann, and D. L. McClean. 1989. Pea aphid symbiont
relations established by analysis of 16S rRNAs. J Bacteriol 171 :2970-2974.

van Niel, E. W. J., and B. Hahn-Hagerdal. 1999. Nutrient requirements of
lactococci in defined growth media. Appl Microbiol Biotechnol 52:617-627.

Venters, D. 1971. The mode of systemic insecticidal action of sulphaquinoxaline
against mosquitoes. Trans. R. Soc. Trop. Med. Hyg. 65:24-25.

Venters, D. 1972. Folate metabolism in the muscles of Aedes aegypti. Insect
Biochemistry 2:153-160.

Widdel, F., and F. Bak. 1992. Gram-negative mesophilic sulfate-reducing
bacteria, p. 3358-3378. In A. Balows, H. G. Truper, M. Dworkin, W. Harder, and
K.-H. Schleifer (ed.), The Prokaryotes, 2nd ed, vol. 4. Springer-Verlag, New
York.

106

Chapter 5
Acetogenesis vs. Methanogenesis in the Termite Hindgut:

The Lumazine Hypothesis

Introduction

The puzzling dominance of acetogenesis over methanogenesis in guts of wood-
feeding termites was discussed in Chapter 1. To summarize, bacterial H2/C02-acetogens
and archaea] methanogens are both present in termite hindguts, but the former clearly
out-process the latter as Hz-consumers, as indicated by: i) the low rates of H2 or methane
emission measured in most wood-feeding termites (3, 35); ii) the increased rate of
methane emission observed after feeding termites antibacterial drugs (35); and iii) the
preferential Hz-dependent reduction of 14CO; to acetate, rather than methane, in termite
gut homogenates (5). The dominance of H2/COz-acetogenesis in termite hindguts is
surprising, since methanogens utilize a more energetically favorable form of metabolism
and are the dominant Hz-consumers in most anoxic habitats in which C02/HC03' is the
primary terminal e' acceptor (2, 7, 13, 33, 37). While unusual, the dominance of termite
C02-reductive acetogenesis as an electron sink is highly beneficial to termites: acetate
production in the hindgut fuels the majority of the insect’s energy metabolism (35),
whereas CH4 emission represents a loss of otherwise useful carbon and energy.
Explaining this peculiar situation is important in understanding the functional ecology of

the termite hindgut community.

107

The spatial resource partitioning (SRP) hypothesis (10, 15) discussed in Chapter 1
offers the best current explanation for the dominance of H2/CO2-acetogenesis in termite
hindguts. In this model, methanogens and acetogens occupy distinct physiocherrrical
niches within the hindgut and hence do not directly compete for H2. In most termites,
methanogens appear to be restricted to the region of lowest hydrogen concentration on
the gut epithelium (22, 23), where they may be additionally inhibited by inwardly
diffusing O2 (9). In contrast, homoacetogens such as Treponema strains ZAS-1 and
ZAS-2 colonize the anoxic lumen of the hindgut, where hydrogen production by gut
protozoa and other luminal microbes maintains H2 at levels 10 to 100 fold higher than the
threshold for H2/CO2-acetogenesis (15). Although the SRP hypothesis helps to reconcile
the prevalence of acetogenesis in termite hindguts, it offers no explanation for the
restriction of methanogens to the gut epithelium, nor does it explain why acetogens
continue to dominate H2-consumption in termite gut homogenates, wherein the spatial
distribution of microbes is homogenous and all members should have equal access to H2
(3, 5). While the spatial separation of acetogens and methanogens undoubtedly
contributes to the dominance of acetogenesis as the primary H2-sink, it seems likely that
some other factor(s) also affects this outcome. It is possible that methanogens are in
some way prevented from colonizing the H2 rich gut lumen, perhaps due to the presence
of some inhibitory compound.

Pteridines are a class of fluorescent compounds possessing a pyrimidine-pyrazine
ring structure. Most naturally occurring pteridine compounds are 2-amino-4-hydroxy
derivatives and are collectively referred to as pterins (Figure 5.1A). Pterins are involved

in a variety of physiological functions, and all living organisms appear to contain at least

108

 

R1
NH2 R2
N Pterin N
N / I \ Deaminase N / \
L / —- I
NH2 N N HO \ N N
Pterin Lumazine

Figure 5.1. Generalized chemical structure of pterin compounds (A) and
production of lumazine from pterin via enzymatic deamination (B).

109

trace amounts of them. Insects, however, contain higher concentrations of pterins than
most organisms and seem to have developed more varied physiological firnctions for
these compounds (49). Insects are capable of synthesizing a wide variety of pteridine
compounds (19, 46, 47), which play significant roles in pigmentation (1, 29, 31, 40),
nitrogen storage and excretion (11, 18), and development (20, 30).

Although there have been no studies of pteridines in termites, these compounds
could play a significant role in the competition between hindgut acetogens and
methanogens: the pteridine compound lumazine has been demonstrated to be a potent
inhibitor of methanogenesis (34). Lumazines are usually formed by the enzymatic
deamination of a pterin (Figure 5.13) (38). Lumazine compounds (1, 16, 17, 36) and
pterin deaminase activities (17, 43-45) have been detected in a wide range of insect
species, and various bacteria have also been shown to possess pterin deaminase activities
(26-28, 41). Nagar-Anthal et al. (34) demonstrated that relatively low concentrations of
lumazine (0.09-0.6 mM) completely inhibited growth and methanogenesis in
phylogenetically diverse methanogens, whereas much higher concentrations had no
apparent inhibitory effects on non-methanogenic organisms, including H2/CO2-
homoaceto gens. The compound tested in that study, lumazine (2, 4-pteridinedione), is
the dearninated derivative of unsubstituted pterin; other lumazine and pteridine
compounds were not examined for anti-methanogenic properties.

The observation that lumazine inhibited methanogenesis, in conjunction with the
known importance of pterin metabolism in insects, raises the possibility that lumazine (or
other pteridine compounds) function as inhibitors of methanogenesis in hindguts of

wood-feeding termites. In this model, the termite host would excrete lumazine (or related

110

pterins, which could be subsequently deaminated by gut bacteria) into the hindgut. The
presence of this compound in the extracellular gut fluid would act to inhibit methanogens
in the H2 rich gut lumen, leaving homoacetogens as the primary H2-consuming group.
This would also potentially explain the restriction of methanogens to the gut epithelium,
since organisms living in biofilm communities are typically less susceptible to the action
of inhibitory compounds (12, 48).

In this chapter, the lumazine hypothesis is investigated in three ways: i) the
susceptibility of the termite gut methanogen Methanobrevibacterfilrfannis to inhibition
by lumazine, other pteridine derivatives, and termite gut extracts is evaluated; ii) the
acetogenic termite gut spirochetes, Treponema strains ZAS-1 and ZAS-2 are screened for
excretion of pteridines and the ability to deaminate exogenous pterins; and iii) gut
extracts of the termite Reticulitermesflavipes are screened for the presence of lumazine

and other pteridine derivatives.

Material and Methods
Organisms, Media, and Cultivation Conditions
T reponema strains ZAS-1 and ZAS-2 and Methanobrevibacterfilzfonnis (DSM
11501) were grown in medium 2YACo (24) as described in Chapter 2, with the following
modification: the cofactor mixture was either omitted (M. filiformis) or replaced with 500
ng/ml folinate (strains ZAS-1 and ZAS-2). All strains were cultivated under anoxic

conditions with atmospheres containing 80% H2 and 20% C02.

111

Termite Gut Extract Preparation

Reticulitermesflavipes worker termites were collected in Dansville, Mich. as
described in Chapter 4 and used within 24 hours. Termites were chilled to 4°C and
degutted by using forceps (4). To estimate the concentration of R. flavipes gut extract
that would yield measurable amounts of pteridine compounds, termite biomass was
compared with that of Pyrrhocoris aptems, an insect in which pteridine concentrations
have been quantified (36). A total of 300 extracted guts were pooled in a glass tissue
homogenizer containing 1.5 ml of 50 mM Na2CO3/NaHC03 buffer (pH 10). The
extraction of pteridine compounds fiom gut homogenates was a modification of the
procedure described in Melber et a1. (30). Homogenates were heated to 100°C for one
minute and immediately cooled on ice, followed by the addition of 5 m1 of chloroform.
Suspensions were mixed by vortexing and centrifuged for 10 min at 3000 x g. The upper
aqueous phase (containing pterins) was removed and added to 0.2 ml of chloroform,
mixed, and centrifuged for 5 min at 10,000 x g. The upper aqueous phase was collected
and stored at -80°C until analysis. All preparations were carried out under reduced

lighting conditions.

Treponema strain ZAS-l and ZAS-2 Pterin Deaminase Assays

Pterin compounds (0.1 mM, final concentration) were added to cultures of ZAS-1
and ZAS-2 at the time of inoculation. At the onset of stationary phase, culture filtrates
were obtained by centrifuging cells for 10 min at 10,000 x g and collecting the

supematants, which were then neutralized, degassed, and passed through a 0.2 um pore

112

size syringe filter into a pre-sterilized sealed tube filled with N2. Pterin deaminase
activity was assessed by TLC analysis of filtrates (see below) to detect any shifts in Rf
value or coloration consistent with conversion of the pterin compound to its lumazine

derivative.

Methanogen inhibition assays

The effects of various pteridine compounds and R. flavipes gut extracts on M.
filiformis were tested in replicate (n=2-4) liquid cultures. Methane production was
assessed by gas chromatography as described previously (35). Pteridines, gut extracts, or
ZAS strain culture filtrates were added to cultures after the onset of growth (as
determined by optical density readings) and methane production. Termite gut extracts
were added to culture tubes at 5% (vol/vol) final concentration, a dose based on
comparison with potentially inhibitory pteridine concentrations measured in P. apterus
(see above). Pteridines were provided at final concentrations of 0.5 mM (lumazines) and
0.2 mM (pterins) unless otherwise noted. ZAS culture filtrates were provided at a final
concentration of 10% (vol/vol). Sterile anoxic water was added to negative control tubes.
At the time of additions, culture tubes were wrapped in aluminum foil to protect
compounds fi'om photochemical degradation. Subsequent assessments of culture growth
were based on methane production alone. All pteridine compounds were obtained from

Schirck’s Laboratories (Jona, Switzerland).

113

Thin Layer Chromatography

As an initial screen for the presence of pteridine compounds, R. flavipes gut
homogenates were analyzed by thin layer chromatography (TLC) (30). Gut homogenate
(10 pl) was applied to 20 x 20 cm TLC plates of CEL 400-10 0.1 mm microcrystalline
cellulose (Machery-Nagle; Diiren, Germany). Chromatograrns were developed by using
an ascending solvent [isopropanolz 2% (w/v) NH3-acetate (1:1)] in a sealed tank lined
with Whatrnan no. 1 filter paper. For two-dimensional chromatography, plates were
developed in the first dimension as described above, allowed to dry, and developed in the
second dimension using 3% NH4C1. Fluorescent compounds on TLC plates were
visualized under a long wavelength (366 nm) U.V. lamp (Black-Ray, Ultra-Violet
Products, San Gabriel, CA). The ratio of distance traveled by the solute to that of the
mobile phase (Rf value) and visual appearance of fluorescant spots in gut extracts were

compared to those of standard pteridine compounds (500 uM concentrations applied as

10 ul spots).

HPLC Analysis

Preparative one-dimensional TLC was performed prior to HPLC analysis of
fluorescent compounds in gut extract. A total of 50 ul of R. flavipes gut extract was
applied to a TLC plate and developed as described above. Each fluorescent spot was
scraped from the glass backing with a razor blade and resuspended in 1 ml of 10 mM
phosphoric acid buffer (pH 3.2) containing 4% methanol. The resulting suspension was
agitated for 5 min, centrifuged for 15 min at 12,000 x g to sediment cellulose particles,

and the supernatant was collected and used for HPLC analysis.

114

The HPLC system and coltunn used for analysis was described in Chapter 4. The
HPLC protocol was a modification of the method described by Porcar et al. (36). The
mobile phase consisted of 10 mM potassium phosphate buffer (pH 3.2) with 0.5%
methanol (v/v) pumped at a flow rate of 1 mein. The injection volume was 20 pl, and
each sample was analyzed by fluorescence detection at the following wavelengths

(excitation/emission): 335/440 nm, 350/410 nm, and 365/490 nm.

Results

Methanogen Inhibition Assays

Lumazine, other pteridine compounds, and gut extracts of Reticulitennesflavipes
were tested for the ability to inhibit methanogenesis by Methanobrevibacterfilifonnis, a
methanogen previously isolated from the gut of R. flavipes (23). The addition of
lumazine to actively growing of cultures of M. filiformis resulted in a substantial
reduction in the rate of methane emission; this effect was consistently observed at final
lumazine concentrations ranging from 0.1 to 1 mM (Figures 5.2 and 5.3). In contrast, no
inhibition of M. filzformis was observed following the addition of 5% (v/v) R. flavipes gut
extract (Figure 5.2) or extracts of whole R. flavipes worker termites (data not shown).
Addition of alternatively substituted lumazine derivatives also yielded no significant
reduction in CH. emission (Figure 5.3). A variety of pterin compounds commonly
observed in insects (pterin, biopterin, neopterin, leucopterin, xanthopterin, and
isoxanthopterin) added to M. filiformis cultures at near saturating concentrations (0.2
mM) also had no inhibitory effects on methane emission. Analogous experiments using

culture filtrates of T reponema strains ZAS-1 and ZAS-2 provided at 10% (vol/vol) final

115

0.063 '-
0.054-
0.045-
0.036"

o.027- l

Methane (mmol)

0.018“

0.009-

 

 

l ' l ' l ' 1 fl 1

u ‘ l
o 20 4o 60 80 100 120
Time (h)

Figure 5.2. Methane production by growing cells of M. filiformis following addition
of lumazine (0), Fl. flavipes gut extract (A), and anoxic water (D). Lumazine was
provided at a final concentration of 1 mM, and gut extract was added at 5%
(vol/vol). The time of additions is indicated by the arrow.

116

0.081 "

0.072 ‘-

0.063 ‘

0.0544

0.045 ‘

Methane (mmol)

0.036 -'

0.027 '-

0.018 "

0.009 *

 

 

 

V V I I I I I I l
20 4o 60 80 100 120 140
Time (h)

Figure 5.3. Methane production by growing cells of M. filiformis following addition
of biolumazine (O), xantholumazine (A), and isoxantholumazine (V), all
provided at 0.5 mM final concentrations. Additions of anoxic water (El) and 0.1

mM lumazine (0) were used as controls. The time of additions is indicated by the
arrow.

117

concentration also had no effect on growth or methanogenesis by M. filiformis (data not

shown).

Thin Layer Chromatography of R. flavipes Gut Extracts

TLC analysis of R. flavipes gut extracts revealed six fluorescent spots which were
designated A through F (Table 5.1). Two-dimensional TLC separations of gut extract
showed no additional spots, suggesting that the six spots represented the only fluorescent
compounds present at detectable levels. A variety of pteridine compounds (both pterins
and lumazine derivatives) commonly found in insects were also screened. Of the six
spots detected in R. flavipes gut extracts, five (A-E) were not similar in Rf value or
coloration to any screened pteridine. Spot F, however, was similar to leucopterin in both
properties (Table 5.1). To assay T reponema strains ZAS-l and ZAS-2 for pterin
deaminase activity, the pterins listed in table 5.1 were added to ZAS-1 and ZAS-2
cultures at the onset of growth and culture filtrates were harvested at the end of lag phase.
No detectable shifts in Rf value or coloration were observed in any of the compounds,
indicating a lack of pterin deaminase activity. TLC analysis of late log phase and
stationary phase culture filtrates of Treponema strains ZAS-l and ZAS-2 showed no

detectable production of fluorescent compounds during growth.

HPLC Analysis of R. flavipes Gut Extracts
To remove potentially interfering compounds present in R. flavipes gut extracts,
the extracts were pre-purified by TLC prior to HPLC analysis. This purification was first

tested by applying a total of 2.5 nmol of lumazine to TLC plates, developing the plate as

118

Table 5.1. Thin Layer Chromatography of Pteridines and R. flavipes Gut Extract

 

Compound Rf Value Color

 

Gut Extract Fluorescent Spots”:

Spot A 0.99 Whitish-Green
Spot B 0.85 Blue-Green
Spot C 0.66 Blue-Green
Spot D 0.56 Blue-White
Spot E 0.35 Light Green
Spot F 0.13 Green

Pteridine Comp_ound Standards :

Lumazine 0.52 Light Green
Xantholumazine 0.27 Blue-Green
Isoxantholumazine 0.31 Dark Purple
Pterin 0.50 Blue
Xanthopterin 0.26 Green
Isoxanthopterin 0.24 Purple
Biopterin 0.58 Blue-Purple
Neopterin 0.50 Blue-Purple
Leucopterin 0.14 Green

 

‘Gut extract and 0.5 mM pteridine standard solutions were applied to TLC plates as 10 ul
spots.

119

described above, and extracting the resultant spot from material excised from the plate.
Recovery of lumazine from this material (as determined by HPLC analysis) was greater
than 90%. HPLC analysis of pterin and lumazine standards (listed in table 5.1) resulted
in good linearity of response for micro- to nanomolar concentrations of all compounds
tested.

The six fluorescent compounds present in R. flavipes gut extracts were subjected
to HPLC analysis with fluorescence detection. Five of the compounds (A, B, C, D, and
E) showed no detectable peaks at any of the three pteridine-specific fluorescence
excitation/emission wavelength combinations tested (data not shown). Spots C and E
showed U.V. absorbance at 250 nm, suggesting that what had been interpreted as
fluorescence on TLC plates may have actually been U.V. absorbance. Compound F
yielded a single fluorescent peak with similar retention time to that of standard
leucopterin. The amount of leucopterin present in gut extract equates to a concentration
of 0.04 nmol per R. flavipes gut. If all of the detected leucopterin was present in gut fluid
and assuming an approximate volume of 0.27 ul fluid per gut (35), leucopterin

concentration in gut fluid would be 0.15 mM.

Discussion

The results of this study demonstrate that neither lumazine nor other pteridine
compounds are likely to play a significant role in the inhibition of termite gut
methanogens. Of the ten pteridines tested, only unsubstituted lumazine (2, 4-
pteridinedione) showed any significant inhibition of the termite methanogen M. filiformis.

While lumazine has been reported in the fruit fly Drosophila melanogaster (21, 39) and

120

the honeybee Apis mellifica (14), this compound was not present at detectable levels in
gut extracts of R. flavipes by either TLC or HPLC analysis. The only pteridine
compound observed in gut extracts was leucopterin, which was shown to be non-
inhibitory to M. filiformis at higher concentrations (200 uM) than could have been
present in R. flavipes gut fluid (150 uM at maximum). A more likely explanation for the
presence of leucopterin in the gut would be as a nitrogen excretory product, since this is
one of roles played by pterin compounds in other insects (49). It remains possible that
lumazine (or pteridines) are present in the temrite hindgut at concentrations below the
detection limits of the assays used in this study, but it is doubtful that such low levels of
these compounds would have any significant activity against methanogens in the gut.

In a series of independent experiments conducted by Dr. John Breznak, filter
paper containing the methanogen-inhibitor bromoethanosulfonate (BES) or lumazine (5
and 500 uM final concentrations, respectively), were fed to R. flavipes worker termites.
Methane emission rates measured in live termites fed BES were significantly reduced,
whereas lumazine-fed termites actually showed a substantial increase in rates of methane
emission (John Breznak, personal communication). The cause of this outcome is as yet
unknown, but could be explained by either the degradation of ingested lumazine by the
termite or gut microbiota (possibly to methanogenic precursors) or by the presence of a
population of gut methanogens more resistant to lumazine inhibition than M. fihformis.
Regardless of the reason, the failure of exogenous lumazine to inhibit methanogens in
situ further indicates that this compound is unlikely be significantly involved in

methanogen inhibition in the termite hindgut.

121

At this time, the failure of termite gut methanogens to dominate H2 consumption
in the termite hindgut remains somewhat enigmatic. While it is possible that some non-
pteridine inhibitor of methanogens is present in the gut lumen, the abolition of
methanogenesis observed in the BBS-feeding studies [(32) and those discussed above]
suggests that colonization of the gut epithelium affords methanogens no special
protection from inhibitory substances. Also, incubation of live termites (32) and agarose
embedded termite guts (42) under H2-enriched atmospheres results in an increase in
methane emission, suggesting that substrate limitation is the dominant limiting factor for
the population of epithelial methanogens. This inference is further supported by the
observation of significantly higher rates of methane emission in Zootermopsis
angusticollis than in other wood-feeding termites (3). Trichomonad flagellates containing
endosymbiotic methanogens colonize the gut of Z. angusticollis (25), and by living
within these protozoa, intracellular methanogens apparently escape the restriction of
others methanogens to the gut wall and gain increased access to H2 produced by cellulose
degradation.

Taken together, these observations suggest that the spatial resource partitioning
hypothesis articulated by Brune and coworkers (10, 15) remains the best current
explanation for the dominance of H2/CO2-acetogens as H2 consumers in the hindguts of
most wood- and grass-feeding termites. Since no termite gut H2/CO2-acetogen has yet
displayed any property which would make them unusually competitive with methanogens
[(6, 8), and Chapter 2], it would be predicted that successful colonization of the gut
lumen by methanogenic archaea would result in a methanogenically dominated H2-flow,

equating to a significant loss of energy for the termite host. This makes the

122

counterintuitive restriction of methanogens to the gut epithelium all the more puzzling.
This unusual situation may yet be explained by considering other factors specific to
growth and survival in termite hindguts, which could include avoidance of grazing by gut
protozoa, avoiding being washed out of the gut, or a reliance on maintaining a close
physical association with some other member of the epithelial microflora. Determining
which (if any) of these factors plays a role in influencing the localization of methanogens
in the hindgut remains a critical unresolved issue in understanding the functional ecology

of the termite hindgut symbiotic system.

123

References

Bel, Y., M. Porcar, R. Sacha, V. Nemec, and J. Ferre. 1997. Analysis of
pteridines in Pyrrhocoris apterus (L) (Heteroptera, Pyrrhocoridae) during
development and in body-color mutants. Arch Insect Biochem Physiol 34:83-98.

Bernalier, A., M. Lelait, V. Rocbet, J.-P. Grivet, G. R. Gibson, and M.
Durand. 1996. Acetogenesis fiom H2 and CO2 by methane and non-methane-
producing human colonic bacterial communities. FES Microbial. Ecol. 19:193-
202.

Brauman, A., M. D. Kane, M. Labat, and J. A. Breznak. 1992. Genesis of
acetate and methane by gut bacteria of nutritionally diverse termites. Nature
257:1384-1387.

Breznak, J. A., and H. S. Pankratz. 1977. In situ morphology of gut microbiota

of wood-eating termites [Reticulitermesflavipes (Kollar) and Coptotermes
formosanus Shiraki). Appl Environ Microbiol 33:406-426.

Breznak, J. A., and J. M. Switzer. 1986. Acetate synthesis from H2 plus CO2 by
termite gut microbes. Appl Environ Microbiol 52:623-630.

Breznak, J. A., J. M. Switzer, and H. J. Seitz. 1988. Sporomusa termitida sp.
nov., an H2/CO2-utilizing acetogen isolated from termites. Arch Microbiol
150:282-288.

Breznak, J. A., and M. D. Kane. 1990. Microbial H2/CO2 acetogenesis in animal
guts: nature and nutritional significance. FEMS Microbiol. Rev.:309-3l4.

Breznak, J. A. 2000. Ecology of prokaryotic microbes in guts of wood- and
litter- feeding termites, p. 209-231. In T. Abe, D. E. Bignell, and M. Higashi (ed.),
Termites: Evolution, Sociality, Symbioses, Ecology. Kluwer Academic,
Dordrecht/Norwell, MA.

Brune, A., D. Emerson, and J. A. Breznak. 1995. The termite gut microflora as
an oxygen sink: microelectrode determination of oxygen and pH gradients in guts
of lower and higher termites. Appl Environ Microbiol 61 :2681-2687.

124

10.

11.

12.

l3.

14.

15.

16.

17.

18.

19.

20.

Brune, A., and M. Friedrich. 2000. Microecology of the termite gut: structure
and function on a microscale. Curr Opin Microbial 3:263-269.

Cochran, D. G. 1975. Excretion in insects, p. 179-281. In D. J. C. a. B. A. Kelby
(ed.), Insect Biochemistry and Function. John Wiley and Sons, New York.

Costerton, J. W., Z. Lewandowski, D. E. Caldwell, D. R. Korber, and H. M.
Lappinscott. 1995. Microbial Biofihns. Annu Rev Microbiol 49:711-745.

De Graeve, K. G., J. P. Grivet, M. Durand, P. Beaumatin, C. Cordelet, G.
Hannequart, and D. Demeyer. 1994. Competition between reductive

acetogenesis and methanogenesis in the pig large-intestinal flora. J. Appl.
Biotechnol. 76:55-61.

Dustmann, J. H. 1971. Pteridine bei der honigbiene Apis met/{flea isolierung
neuer lumazine. Hoppe Seylers Z Physiol Chem 352:1599-1600.

Ebert, A., and A. Brune. 1997. Hydrogen concentration profiles at the oxic-
anoxic interface: a nricrosensor study of the hingut of the wood-feeding lower
termite Reticulitermesflavipes. Appl Environ Microbiol 63:4039-4046.

Goto, M., M. Konishi, K. Sugiura, and M. Tsusue. 1966. The structure of a
yellow pigment from the mutant lemon of Bombyx mori. Bull Chem Soc Jpn
39:929-932.

Gyure, W. L. 1974. Catabolism of isoxanthopterin during the development of the
silkworm, Bombyx mori. Insect Biochemistry 4:303-312.

Harmsen, R. 1966. The excretory role of pteridines in insects. Journal of
Experimental Biology 45:1-13.

Harmsen, R. 1969. The effect of atmospheric oxygen pressure on the
biosynthesis of simple pteridines in the pierid butterflys. J Insect Physiol 15:2239-
2244.

Hudson, B. W., A. H. Bartel, and R. Craig. 1959. Pteridines in milkweed bug,
Oncopeltus fasciatus (Dallas) H. Quantitative determination of pteridine content
of tissues during growth. J Insect Physiol 3:63-73.

125

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Katah, S., T. Sueaka, and S. Yamada. 1967. Neopterin in the sepia mutant of
adult Drosophila melanogaster. Insect Biochemistry 10:119-123.

Leadbetter, J. R., and J. A. Breznak. 1996. Physiological ecology of
Methanobrevibacter cuticularis sp. nov. and Methanobrevibacter curvatus sp.
nov., isolated from the hindgut of the termite Reticulitennesflavipes. Appl
Environ Microbiol 62:3620—363 1.

Leadbetter, J. R., L. D. Crosby, and J. A. Breznak. 1998. Methanobrevibacter
filiformis sp. nov., a filamentous methanogen from termite hindguts. Arch
Microbiol 169:287-292.

Leadbetter, J. R., T. M. Schmidt, J. R. Graber, and J. A. Bro-ak. 1999.
Acetogenesis from H2 plus CO2 by spirochetes from termite guts. Science
283:686-689.

Lee, M. J., P. J. Schreurs, A. C. Messer, and S. H. Zinder. 1987. Association
of methanogenic bacteria with flagellated protozoa fi'om a termite hindgut. Curr
Microbial 15:337-341.

Levenberg, B., and O. Hayaishi. 1959. A bacterial pterin deaminase. J Biol
Chem 234:955-961.

Levy, C., and W. S. McNutt. 1962. The biological transformation of
xanthopterin by a bacterium isolated from soil. Biochemistry 1:1161-1170.

McNutt, W. S. 1962. The metabolism of isoxanthopterin by Alcaligenesfaecalis.
J Biol Chem 238:1116-1121.

Melber, C., and G. H. Schmidt. 1992. Identification of fluorescent compounds
in certain species of Dysdercus and some of their mutants (Heteroptera,
Pyrrhocoridae). Comp Biochem Physiol 101:115-133.

Melber, C., and G. H. Schmidt. 1994. Quantitative variations in the pteridines
during the postembryonic development of Dysdercusspecies (Heteroptera,
Pyrrhocoridae). Comp Biochem Physiol 108:79-94.

126

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

Melber, C., and G. H. Schmidt. 1997. Body colouration related to the deposition
of pteridines in the epidermis and other organs of Dysdercus species (Insecta;
Heteroptera: Pyrrhocoridae). Comp Biochem Physiol 1 16:17-28.

Messer, A. C., and M. J. Lee. 1989. Effect of chemical treatments on methane
emission by the hindgut rrricrobiota in the termite Zootermopsis angusticollis.
Microb Ecol 18:275-284.

Marvan, B., F. Bonnemoy, G. Fonty, and P. Gouet. 1996. Quantitative
determination of H2-utilizing acetogenic and sulfate reducing bacteria and

methanogenic archaea from the digestive tract of different mammals. Curr.
Microbial. 32:129-133.

NagarAntbal, K. R., V. E. Warrell, R. Teal, and D. P. Nagle. 1996. The pterin
lumazine inhibits growth of methanogens and methane formation. Arch Microbiol
166:136-140.

Odelson, D. A., and J. A. Breznak. 1983. Volatile fatty acid production by the
hindgut microbiota of xylophagous termites. Appl Environ Microbiol 45:1602-
1 61 3.

Porcar, M., Y. Bel, R. Sacha, V. Nemec, and J. Ferre. 1996. Identification of
pteridines in the firebug, Pyrrhocoris apterus (L) (Heteroptera, Pyrrhocoridae) by
high-performance liquid chromatography. J Chromatogr A 724:193-197.

Prins, R. A., and A. Lankhorst. 1977. Sythesis of acetate from CO2 in the
cecum of some rodents. FEMS Microbiol. Letters 1:255-258.

Rembold, H. 1985. Catabolism of pterins, p. 155-177. In R. L. B. a. S. J.
Benkovic (ed.), Folates and Pterins, vol. 2. John Wiley and Sons, New York.

Sagiura, K., and M. Gata. 1967. Biosythesis of pteridines in Drosophila
melanogaster. Biochem Biophys Res Commun 28:687-691.

Tabata, H., T. Hasegawa, M. N akagasbi, S. Takikawa, and M. Tsuse. 1996.
Occurence of biopterin in the wings of Morpho butterflies. Experientia 52:85-87.

Takikawa, S., C. Kitayama-yokokawa, and M. Tsusue. 1979. Pterin deanrinase
from Bacillus megaterium. J Biochem (Tokyo) 85:785-790.

127

42.

43.

45.

46.

47.

48.

49.

Tholen, A., and A. Brune. 2000. Impact of oxygen on metabolic fluxes and in
situ rates of reductive acetogenesis in the hindgut of the wood-feeding termite
Reticulitermesflavipes. Environ. Microbiol. 2:436-449.

Tsusue, M. 1967. Occurence of sepiapterin deaminase in the silkworm, Bombyx
mori. Experientia 23:1 16-1 17.

Tsusue, M. 1971. Studies on sepiapterin deaminase fi'om the silkworm, Bombyx
mori. J Biochem (Tokyo) 69:781-788.

Tsusue, M., and T. Mazda. 1977. Occurrence of sepiapterin deaminase in the
normal type silkworm, Bombyx mori. Experientia 33:854-855.

Watt, W. B. 1967. Pteridine biosynthesis in the butterfly Colias eurytheme. J Biol
Chem 242:565-572.

Watt, W. B. 1972. Xanthine dehydrogenase and pteridine metabolism in Colias
butterflys. J Biol Chem 247:1445-1451.

Whiteley, M., M. G. Bangera, R. E. Bumgarner, M. R. Parsek, G. M. Teitzel,
S. Lory, and E. P. Greenberg. 2001. Gene expression in Pseudomonas
aeruginosa biofilrns. Nature 413:860-864.

Zeigler, 1., and R. Harmsen. 1969. The biology of pteridines in insects. Adv.
Insect. Physiol. 6:139-207.

128

Chapter 6

Summary

This dissertation has investigated of the physiological properties of the termite gut
spirochetes T reponema strains ZAS-l , ZAS-2 (3), and ZAS-9 (5) and their relationships
with other members of the termite hindgut microbial community. Although over a
hundred years of research had yielded a substantial amount of data on the abundance,
diversity, and distribution of termite gut spirochetes, their physiological properties and
firnctional roles in situ were poorly understood, largely due to a lack of any cultured
representatives. The availability of pure cultures of Treponema strains ZAS-l, ZAS-2,
and ZAS-9, the first spirochetes to be isolated fiom the guts of wood-feeding termites,
has finally made in vitro physiological studies possible. Gaining further understanding of
the physiology, nutrition, and growth requirements of these organisms in vitro has
resulted in the formulation of new hypotheses regarding their in situ roles and
relationships with other members of the termite gut microbial community.

The research presented in Chapter 2 focused on the physiological properties
relevant to the in situ growth and survival of Treponema strains ZAS-1 and ZAS-2,
H2/CO2-homoacetogenic termite gut spirochetes. Strains ZAS-1 and ZAS-2 proved to be
nutritionally versatile and were capable of mixotrophic utilization of H2 and organic
substrates, including various carbohydrates and (in the case of ZAS-2) methoxylated
aromatic compounds. These results suggest that the ZAS strains are likely to contribute

to termite nutrition via acetogenesis from fermentation and demethylation of organic

129

substrates in situ, as well as from H2 and CO2 produced by other members of the gut
microbiota. ZAS-l and ZAS-2 were shown to be similar to other homoacetogens in
possessing enzyme activities of the Wood/Ljungdahl pathway and displaying H2
thresholds within the range of typical of known acetogens. In comparison to other
homoacetogens however, ZAS-l and ZAS-2 were less efficient in terms of energy
conservation; this may represent an adaptation to a symbiotic lifestyle which prevents
excessive cell growth (which could be detrimental to the host) while still allowing
production of acetate to fuel termite energy metabolism. Both strains had requirements
for folate, a curious observation considering the importance of this cofactor in
acetogenesis. Although ZAS-l and ZAS-2 are strict anaerobes, both strains are capable
of some degree of O2 tolerance and detoxification, which is likely an adaptation to the
partially hypoxic conditions encountered by spirochetes in termite hindguts. Taken
together, these findings demonstrate that Treponema strains ZAS-1 and ZAS-l are well
adapted to life in the termite hindgut and suggests that their relationship to their termite
hosts is beneficial in nature.

Chapter 3 provided a more thorough description of Treponema strain ZAS-9,
which differed significantly fi'om ZAS-l and ZAS-2 in terms of morphology and
metabolism. ZAS-9 is not a H2/CO2-homoacetogen and in fact produced H2 as a product
of sugar fermentation. This raises the possibility that interspecies H2 transfer between
spirochetes may be an important component of H2 turnover in the termite hindgut. In
terms of taxonomy, the results suggest that ZAS-1 and ZAS-2 should be regarded as two
strains of single new species in the genus Treponema, whereas strain ZAS-9 should be

considered a separate new species of Treponema distinct fi'om that accommodating ZAS-

130

1 and ZAS-2 (Latinate species epithets are currently under consideration). The
physiological differences observed among the three strains of termite gut spirochetes
currently in pure culture likely offers only an introductory glimpse into the functional
diversity of these organisms. Future cultivation efforts should continue to yield new
insights into the physiological capabilities of termite gut spirochetes.

The observation of folate requirements in Treponema strains ZAS-l and ZAS-2 in
Chapter 2 resulted in an investigation of the in situ source of this cofactor, which was
presented in Chapter 4. Since folate-secreting bacteria had been observed in a variety of
other gastrointestinal systems(1, 2, 7), it was hypothesized that other members of the
termite gut microbial community were providers of folate. Consistent with this
hypothesis, two folate-secreting strains were isolated from guts of the termite
Zootermopsis angusticollis, Serratia strain ZFX-1 and Lactococcus strain ZFX-2.
Provision of culture filtrates of ZFX-1 and ZFX-2 supported the growth of Treponema
strains ZAS-l and ZAS-2 in the absence of added folates. The folate compound
produced by both ZF X strains was determined to be folinate (5-HCO-tetrahydrofolate),
which was demonstrated to be a growth supportive form of the cofactor for ZAS-l and
ZAS-2. These results suggest that folate secreting bacteria such as strains ZFX-1 and
ZFX-2 are the most likely providers of folates to the ZAS strains in situ. The ZF X strains
may also be important in providing folate to other members of the gut microbial
community to the termite host.

Finally, Chapter 5 tested the hypothesis that methanogenic archaea are inhibited
by the presence of pteridine compounds in the termite hindgut, allowing H2/CO2-

homoacetogens to act the primary H2-consumers in the guts of wood-feeding termites.

131

The pteridine compound lumazine has been shown to be inhibitory to methanogenesis
(6), and pteridine compounds are typically present in insects at relatively high levels (as
compared to other animals) (8). The presence of inhibitory pteridines in the lumen of the
termite hindgut would help to explain the counterintuitive restriction of methanogens to
the region of lowest H2 concentration in the gut, the epithelium. However, lumazine was
the only pteridine which had any inhibitory effect on the termite methanogen
Methanobrevibacterfilyformis, and this compound was not detected in gut extracts of the
termite Reticulitermesflavipes. It was therefore deemed doubtful that pteridines play a
significant role in the inhibition of methanogenesis in the R. flavipes hindgut, and the
localization of termite methanogens to the gut epithelium must be explained by other
factors.

In concluding this dissertation, it seems appropriate to revisit the pioneering work
of Dr. Joseph Leidy (4). In 1881, after years of microscopic observations describing the
diverse inhabitants of termite hindguts, Leidy makes the following comment:

"Termites... are so common, easily obtained and preserved alive, and their

parasites are so exceedingly numerous, constant in their occurrence, and

curious, that once the fact becomes sufi'iciently known, the insects will
become subjects to illustrate at once the infinity of life and the wonders

that are revealed by the microscope. "

Over one hundred years later, the termite hindgut continues to fascinate
microbiologists, and it seems safe to conclude that it will continue to do for at

least another hundred years to come.

132

 

References

Camila, E., J. Zimmerman, J. B. Mason, B. Golner, R. Russell, J. Selhub, and
I. H. Rosenberg. 1996. Folate synthesized by bacteria in the human upper small
intestine is assimilated by the host. Gastroenterol 110:991-998.

Krause, L. J., C. W. Forsberg, and D. L. O'Connor. 1996. Feeding human milk
to rats increases Bifidobacterium in the cecum and colon which correlates with
enhanced folate status. J Nutr 126:1505-1511.

Leadbetter, J. R., T. M. Schmidt, J. R. Graber, and J. A. Breznak. 1999.
Acetogenesis fiom H2 plus CO2 by spirochetes fi'om termite guts. Science
283:686-689.

Leidy, J. 1874-1881. The parasites of the termites. J. Acad. Nat. Sci. (Phi1.)
8:425-447.

Lilburn, T. C., K. S. Kim, N. E. Ostrom, K. R. Byzek, J. R. Leadbetter, and
J. A. Breznak. 2001. Nitrogen fixation by symbiotic and fi'ee-living spirochetes.
Science 292:2495-2498.

NagarAnthal, K. R., V. E. Warrell, R. Teal, and D. P. N agle. 1996. The pterin
lumazine inhibits growth of methanogens and methane formation. Arch Microbiol
166:136-140.

Slyter, L. L., and J. M. Weaver. 1977. Tetrahydrofolate and other growth
requirements of certain strains of Ruminococcusflavefaciens. Appl Environ
Microbiol 33:363-369.

Zeigler, 1., and R. Harmsen. 1969. The biology of pteridines in insects. Adv.
Insect. Physiol. 6:139-207.

133

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llélijliliflsliizilflil'i1111(1)1|

144