Download Influence of diet on the structure and function of the bacterial hindgut

Molecular Ecology (1998) 7, 761–767
Influence of diet on the structure and function of the
bacterial hindgut community of crickets
J . W. S A N T O D O M I N G O , * † ¶ M . G . K A U F M A N , ‡ M . J . K L U G , * † ‡ W. E . H O L B E N , † * *
D. HARRIS,†§ and J. M. TIEDJE*†§
*Department of Microbiology, Michigan State University, East Lansing, MI 48824, USA, †Center for Microbial Ecology, Michigan
State University, East Lansing, MI 48824, USA, ‡W. K. Kellogg Biological Station, Hickory Corners, MI 49060, USA, §Crop and
Soil Sciences Department, Michigan State University, East Lansing, MI 48824, USA
Abstract
The effect of diets varying in carbohydrate and protein content on the structure and function of the hindgut microbiota of crickets was evaluated by determining bacterial densities, fermentation activity, and guanine plus cytosine (G + C) profiles of the DNA
extracted from the microbial hindgut community. DNA isolated from the gut community
was fractionated and quantified according to G + C content as a comprehensive, coarselevel measure of the composition and structure of the community. The bacterial densities
measured by direct counts were not significantly different among the four diets. The
crickets were initially reared in the laboratory on cricket chow, which resulted in a
hindgut community dominated by bacteria with a G + C content between 32% and 57%.
Crickets shifted to an alfalfa diet showed a similar hindgut community G + C profile,
although microbial populations with DNA between 35% and 45% G + C were more abundant in alfalfa- than chow-fed crickets. The apparent complexity of the gut community
was reduced in crickets fed beet-pulp and protein-based diets compared to those fed
chow and alfalfa, and was dominated by populations with a low percentage G + C content. Hindgut communities in crickets fed pulp and protein diets also showed a decrease
in hydrogen and carbon dioxide production, suggesting that these diets affected the biochemical activity of the hindgut community. The protein-based diet resulted in a decrease
in the rate of evolution of volatile fatty acids, while the ratio of butyrate production to
acetate and propionate production was significantly higher in these crickets. Our results
show the emergence of a new microbial community structure concomitant with changes
in microbial biochemical activity due to shifts in the cricket’s dietary regime.
Keywords: gut fermentation, hindgut microbial community, insect gut, microbial community structure
Received 9 July 1997; revision received 18 November 1997; accepted 1 January 1998
Introduction
Understanding the relationship between community
structure and function is one of the central themes in
microbial ecology. The insect gut community is well
suited to such studies because microbial metabolism of
Correspondence: J. W. Santo Domingo.
¶Present address (for correspondence): Westinghouse Savannah
River Company, Savannah River Technology Center, Building
704-8T, Aiken, South Carolina, SC 29808, USA. Fax: +01-8035577223; E-mail: [email protected].
**Present address: Division of Biological Sciences, The University
of Montana, Missoula MT 59812, USA
© 1998 Blackwell Science Ltd
ingested food is often important to the animal’s nutrition
(Buchner 1965; Jones 1984; Campbell 1989; Breznak &
Brune 1994), it is a flow-through system that would be
expected to rapidly respond to change of substrates, it
often has a relatively dense and easily recoverable
biomass, and normally the insect system is relatively inexpensive and easily adapted to laboratory conditions and
experimental treatments (Cruden & Markovetz 1987;
Kaufman et al. 1989).
Although it is well documented that the gut of insects
provides excellent conditions for the prolific growth of
microorganisms (Bignell 1984), such microbial communities are rather poorly understood. Extrapolation from
762
J . W. S A N T O D O M I N G O E T A L .
studies of gut communities of higher organisms (Moore &
Holdeman 1974; Hungate 1975) suggests that the guts of
many insects should also harbour diverse microbial communities comprised of many types of bacteria, some with
unique physiological requirements. In fact, several reports
show that strict and facultative anaerobic Eubacteria as
well as Archaebacteria and Protozoa inhabit the gut of certain insects, e.g. termites and cockroaches (Ulrich et al.
1981; Breznak 1984; Cruden & Markovetz 1987; Breznak &
Brune 1994; Ohkuma & Kudo 1996; Paster et al. 1996). The
functional disruption of gut microbiota by feeding antibiotics or axenic-rearing techniques have supported the
hypothesis that microbial symbionts play key roles in
complementing the nutrition of their host (Bracke et al.
1978; Kaufman et al. 1989; Kaufman & Klug 1991). Thus,
perturbations that disrupt the community structure of gut
microbes could adversely affect the nutrition of the insect.
While some studies have reported on diets affecting
some gut-community functions of insects, e.g. methanogenesis, fewer studies have attempted to link functional
changes to population shifts. In one such study the investigators noted a decrease in streptococci and lactobacilli
inhabiting the foregut of cockroaches after a switch to low
protein–high-fibre diets, which was accompanied by a
decrease in the rate of lactate and acetate production
(Kane & Breznak 1991). In contrast, methanogenic activity
was found to increase in the hindgut of cockroaches fed
the same diet, although methanogen population levels
were not measured directly. Studies with crickets have
shown that total microbial densities do not significantly
change with changes in the host diet (Santo Domingo
1994), even though the profile of fermentation products
may change (Kaufman & Klug 1991). It has been shown
that microbial degradation of dietary soluble carbohydrates fluctuates with diet treatments in crickets
(Kaufman 1988), and that the microbial response to levels
and types of dietary carbohydrates serves to maintain a
relatively constant supply of fermentation products to the
insect (Kaufman & Klug 1990). However, it is unknown
whether the microbial response is a product of shifts in
bacterial populations or shifts in metabolic activity of the
existing populations. To address this question, we tested
the hypothesis that changes in insect diet result in
changes in the composition of the gut microflora, that in
turn lead to changes in fermentation processes. This was
accomplished by comparing the bacterial community
structure in the hindguts of crickets fed different diets by
measuring the distribution of guanine plus cytosine (%
G + C) content in DNA fragments extracted from the gut
community. This approach is comprehensive for all community members, whether culturable or not, and provides an approximate family- to genus-level resolution of
community composition (Holben & Harris 1995; Holben
et al. 1998). In addition, rates of production of fermenta-
tion products in the hindgut were measured as an index
of microbial community function.
Materials and methods
Cricket diets and gut preparation
House crickets (Acheta domesticus) were reared in the laboratory until adults on a diet of Purina cricket chow. The
crickets were incubated under a 12-h light, 12-h dark cycle
in an environmental chamber at 30 °C and 60% relative
humidity. Crickets were then shifted to one of the following three diets: ground alfalfa hay, ground pulp from
sugar beet roots, or an artificial protein diet (40% casein,
50% alphacel fibre) based upon that used by McFarlane &
Distler (1981) for rearing A. domesticus. Alfalfa and pulp
diets were amended with salts and vitamins as in the protein diet. Crickets maintained on chow were used as controls. Water-soluble carbohydrates (Dubois et al. 1965)
were 50%, 51%, 20%, and < 5% of the total dry weight for
chow, pulp, alfalfa, and protein diets, respectively. Soluble
carbohydrate/protein ratios were 5:1, 2:1, 1:1, and 1:5 for
pulp, chow, alfalfa, and protein diets, respectively. Overall,
the diets represented a range of nutritional levels likely to
be encountered by omnivorous insects. A separate batch of
crickets fed on chow were exposed for 5 days to metronidazole incorporated in the water supply (300 µg/mL)
(Cruden & Markovetz 1979). After 5 days of these diets,
crickets were sacrificed and the hindguts were surgically
removed under a dissecting microscope using fine forceps.
Bacterial densities
Hindguts were homogenized in phosphate-buffered
saline (PBS; pH 7.2) using a sterile tissue grinder. Samples
were transferred to microcentrifuge tubes and vortexed
for 30 s. Direct counts were performed using subsamples
of hindgut homogenates fixed in 3.7% formaldehyde–PBS
solution and stained with 0.01% acridine orange as suggested by Hobbie et al. (1977). Samples were filtered
through a 0.2-µm pore size Irgalan black prestained polycarbonate membrane (Costar, Nuclepore Filtration
Products, Cambridge, MA). Membranes were placed on a
microscope slide and stored at 4 °C in the dark until analysed. Fluorescing cells from 10 randomly selected fields
were counted using a 63× oil-immersion objective on a
Leitz Orthoplan 2 epifluorescence microscope. Differences
in direct counts were evaluated by a one-way A N O VA .
Levels of significance were P < 0.05.
Fermentation metabolites
Hindguts were quickly removed and placed individually
in preweighed 2 mL Vacutainers® under an oxygen-free
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 761–767
INFLUENCE OF DIET ON THE CRICKET HINDGUT BACTERIAL COMMUNITY
N2 headspace. Six replicate hindguts from each diet treatment were weighed, incubated for 3 h at 30 °C, and analysed for H2 and CO2 using gas chromatography (Lovley
& Klug 1982). Samples were frozen in the vials by immersion in liquid N2 immediately after headspace sampling,
and aqueous extracts were later analysed for short-chain
fatty acids using high-performance liquid chromatography (HPLC) (Lawson & Klug 1989). Three replicates from
each diet were treated as above but without incubation to
determine initial (time-zero) levels of fermentation products. Only data for those compounds which showed significant production (slope from regression analysis)
during the incubation period are presented.
Concentrations and percentages of fermentation products
were compared after suitable transformation with standard A N O VA techniques. Levels of significance were
defined as P < 0.05 unless otherwise indicated.
Microbial community DNA extraction
To separate bacteria from the eukaryotic tissue, 15 homogenized hindguts were quickly spun (1 s) using a microcentrifuge. The supernatant was then carefully removed and
used to extract the hindgut microbial community DNA.
Microscopic examination of the supernatant revealed that
the gut tissue was efficiently removed using this
approach, thus minimizing the potential contribution of
host DNA to the final pool of nucleic acid. Other potential
sources of eukaryotic DNA are unlikely as protozoa and
fungi do not normally inhabit the hindgut of most of the
species of crickets examined, including A. domesticus
(Ulrich et al. 1981). Direct counts indicated that 78% of the
bacterial cells in the original gut homogenate were recovered using this approach, indicating that only a minor
portion of the community was removed with gut-wall
debris (data not shown).
To concentrate the microbial cells, the supernatant was
collected and centrifuged for 5 min at 6800 × g. Genomic
DNA was extracted from the bacterial pellet following the
procedure described by Visuvanathan et al. (1989), but
modified as follows: no heat treatment was employed as
the bactericidal agent; instead, 2% sodium azide (final
concentration) was used. Incubation times for the subtilisin, lysozyme, and pronase treatments were reduced to 7,
3, and 7 h, respectively. Phenol/chloroform extractions
were performed following a hexadecyltrimethylammonium bromide (CTAB; Sigma, St Louis, MO) pretreatment
(Ausubel et al. 1987). The organic phase was treated with
RNase A (Sambrook et al. 1989), followed by an additional
phenol/chloroform extraction. To precipitate the DNA,
samples were mixed gently with isopropanol (0.6 vols)
and stored at – 20 °C for 16 h. Samples were centrifuged at
4 °C for 15 min at 8000 × g and the supernatant was carefully removed. The precipitated DNA was allowed to air
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 761–767
763
dry, dissolved in sterile deionized water, and stored at
20 °C until analysed.
Direct counts and viable counts were performed prior
to phenol/chloroform extractions to determine the celllysis efficiency of the DNA-extraction methods. Viable
counts were determined as colony-forming units using
Trypticase Soy Agar (Difco) plates and incubating at 24 °C
for 3 days.
Percentage G + C profiles
A modified version of the CsCl–bisbenzimide method
developed by Holben & Harris (1995) was used in this
study. Approximately 50 µg of bacterial community DNA
was added to a CsCl solution of refractive index (RI) of
1.4000 and adjusted to a final RI of 1.3990. Bisbenzimide
(Sigma) was added (10 µL from a 1 mg/mL bisbenzimide
stock solution) to the CsCl–DNA solution, mixed well, and
transferred to 6-mL ultracentrifuge tubes. Samples were
centrifuged at 110 000 × g for 72 h using a TFT65.6 fixedangle rotor on a Sorvall OTD55B ultracentrifuge (Sorvall
Instruments, Wilmington, DE). Two replicate CsCl–bisbenzimide gradients were analysed for the hindgut communities of crickets fed each of the different diets. The
density gradients were carefully removed from the rotor
and aliquoted to microcentrifuge tubes as 100 µL fractions
using a gradient fractionator (Model 640, ISCO, Lincoln,
NE). The refractive index of each fraction was determined
using a Bausch and Lomb refractometer (Model ABBE-3 L,
Milton Roy Co., Rochester, NY). The DNA concentration
of each fraction was determined by UV absorption
(OD260). The percentage G + C of each fraction was
determined from the RI using a standard curve of percentage G + C vs. RI made with genomic DNA of bacteria of
known percentage G + C (Clostridium perfringens, 31.0%;
Enterobacter aerogenes, 57.5%; and Corynebacterium flaccumfaciens 70.4%). The linear correlation (R2) was 0.987.
Results
Direct counts of the Acheta domesticus microbial community ranged from ≈ 1.0 × 109 to 3.0 × 109 cells per gut
(Fig. 1) and did not differ significantly between diets.
Bacterial densities also did not vary with diet as determined by aerobic (Santo Domingo 1994) and anaerobic
(M. G. Kaufman and M. J. Klug, unpublished data) viable
counts.
Production rates of different fermentation products
were measured as an index of the metabolic activity of the
hindgut community. Similar rates of hydrogen evolution
were observed in the hindguts of crickets fed the pulp and
protein diets. However, such rates were significantly
higher in the hindgut of crickets fed chow or alfalfa
(Fig. 2). Rates of CO2 evolution were significantly lower in
764
J . W. S A N T O D O M I N G O E T A L .
Fig. 2 H2, CO2, and VFA production from in vitro assays of
hindgut communities derived from crickets fed the indicated
diets. Values are concentrations per mg of gut fresh weight after
3-h incubations and are presented as mean ± 1 SE (n = 6).
Visuvanathan et al. (1989), which was reported to lyse
both Gram-positive and Gram-negative bacteria, resulted
in a 91% reduction of the direct counts and ≈ 99.9% of the
viable counts (data not shown), indicating highly efficient
lysis of the gut bacterial populations. The DNA yield
using this method was 6.29 µg of DNA per gut, which is
equivalent to a recovery of 35% of the theoretical DNA
yield estimated from direct counts and an average DNA
content for bacteria of 9 × 10–9 µg per cell (Holben 1997).
Considering the indicated lysis efficiency based on two
different analyses, it seems likely that the relatively low
recovery of DNA reflects nonspecific loss due to postlysis
processing steps (e.g. phenol/chloroform extractions and
isopropanol precipitations) rather than large numbers of
bacteria refractory to lysis in the starting sample.
The genomic DNA extracted was ≈ 25 kb in size as
determined by agarose gel electrophoresis. None of the
other three methods assessed lysed more than 45% of the
hindgut bacterial community (based on direct counts)
and yielded less than 2 µg DNA per cricket hindgut.
Subsequent work was performed using the modified
Visuvanathan method as the DNA recovered was considered to be representative of the entire community based
on lysis efficiency.
To investigate whether diet changed the structure and
composition of hindgut bacterial communities, we generated profiles of the gut microbial community DNA in terms
of DNA relative abundance vs.%G + C content (Fig. 4). The
bacterial community profiles of crickets on chow and
alfalfa diets exhibited a broad distribution of DNA ranging
from about 27% to 60% G + C content (Fig. 4a). These broad
peaks are indicative of the presence of a relatively rich
number of bacterial populations spanning this percentage
G + C range as individual populations typically generate a
single, coherent peak whose central point is in good agreement with literature values for that bacterial species
hindguts from crickets fed the protein-based diet compared
to crickets fed alfalfa or beet-pulp diets (Fig. 2). Similarly,
the total volatile fatty acids (VFA) evolution rate was lower
in the protein-fed crickets than in those fed on alfalfa or
beet-pulp diets. Protein-fed crickets exhibited the highest
ratio of butyrate to acetate and butyrate to propionate
(Fig. 3), although the rate of total VFA production for
crickets on the protein diet was less than for crickets on
alfalfa or pulp (Fig. 2).
We evaluated four different methods for bacterial lysis
and recovery of DNA from the cricket hindgut community. These methods ranged from large-scale protocols
(Ausubel et al. 1987; Holben et al. 1988) requiring purification of nucleic acids using CsCl–ethidium bromide gradients to small-scale DNA-extraction procedures (Ausubel
et al. 1987; Visuvanathan et al. 1989). The method of
Fig. 3 Percentage of the total VFA (see Fig. 2) as acetate, butyrate,
and propionate from crickets fed the indicated diets. Values are
mean ± 1 SE (n = 6).
Fig. 1 Bacterial densities of cricket hindguts from animals fed the
indicated diets and determined by acridine orange direct counts
(AODC). Triplicates were used to determine each value. Bars represent standard errors.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 761–767
INFLUENCE OF DIET ON THE CRICKET HINDGUT BACTERIAL COMMUNITY
765
and 53% G + C content. This community profile is also
consistent with a less-complex bacterial community than
was indicated in the alfalfa- and chow-fed crickets and is
also different in composition from the community
observed in beet-pulp-fed crickets.
When the antibiotic metronidazole was added to
crickets fed chow, the community profile changed to
two predominant DNA peaks at 47.5 and 55% G + C
(Fig. 4c). The sharp nature of these peaks and the clear
differences between this community and those not
treated with metronidazole indicates that this drug has
had a profound effect on the gut community, probably
by reducing the number of anaerobic bacteria (Bracke
et al. 1978).
Discussion
Fig. 4 Percentage G + C profiles of hindgut community DNA
from crickets fed the indicated diets. The crickets subjected to
metronidazole were also fed chow. Standards are DNA from
Clostridium perfringens (31.0% G + C), Enterobacter aerogenes
(57.5% G + C), and Corynebacterium flaccumfaciens (70.4% G + C).
The error bars represent standard errors of two gradients using
DNA homogenized from one batch (15 individuals) of crickets.
(Holben & Harris 1995; Holben et al. 1997; Fig. 4c). Thus, the
generalized distribution of DNA across this broad range of
percentage G + C content is consistent with the presence of
a diverse bacterial community representing multiple phylogenetic groups as the majority of known members of bacterial genera typically have percentage G + C contents
spanning a range of less than 10% (Holben & Harris 1995).
Crickets fed cricket chow had a rather equitable distribution of bacteria across the 27–60% G + C content range,
while in crickets fed alfalfa, bacteria of 35–45% G + C content were relatively more abundant.
Two distinct peaks in the bacterial community profile,
one at 35% G + C content and the other at ≈ 43% G + C
content were obtained from crickets fed beet-pulp
(Fig. 4b). No DNA from bacterial populations with a
G + C content higher than 47% was obtained from crickets
on this diet. This G + C profile with its narrower, sharper,
bimodal appearance is consistent with a less-complex
community with no component populations of high percentage G + C content. The bacterial community from
crickets on the protein diet also was bimodal in appearance with peak DNA concentrations centred at about 39%
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 761–767
Diet affected both the structure and function of the cricket
hindgut microbial community, although the microbial
densities remained constant. The protein diet caused the
most dramatic change as the rate of H2, CO2, and total
VFA production was less than for all other diets. The community profile for this gut community indicated
decreased complexity compared to the communities
observed in alfalfa- and chow-fed crickets; particularly
noticeable was the virtual absence of bacterial populations having 45–50% G + C content. The pulp diet also
appeared to reduce the complexity compared to that of
chow- and alfalfa-fed animals but to a lesser extent than
for the protein-based diet. One possible explanation is
that the variety of soluble carbohydrates in beet-pulp is
reduced compared to alfalfa as the pulp diet is based on
root material after sugars have been extracted while the
alfalfa diet is composed of ground but otherwise unmodified leaf and shoot material. While no direct evidence was
provided by this study, this suggests a possible link
between substrate diversity and the predominant populations in the cricket gut community. Overall gut fermentation, however, was not significantly reduced by the pulp
diet as CO2 and VFA production rates were approximately equivalent to those observed for chow- and
alfalfa-fed crickets.
The results indicated that the composition of the
cricket hindgut microbial community is mostly dominated by bacteria whose DNA has a G + C content of less
than 55%. No DNA fragments with a G + C content of
more than 60% were detected for any diet. Bacteria with a
G + C content greater than 60% tend to be oxidative in
their metabolism and dominate soil environments and
the plant phyllosphere, the normal habitat and food
sources of these crickets (Martoja 1966). In fact, typical
soil bacterial communities are dominated by bacteria in
the 55–75% G + C range (Holben & Harris 1995). Hence,
the cricket hindgut community seems to be one that is
766
J . W. S A N T O D O M I N G O E T A L .
selected by the gut conditions and not simply reflective
of the ingested microbiota.
It is interesting that overall microbial densities, as
determined by direct counts (this study) and viable
counts (M. G. Kaufman and M. J. Klug unpublished
data), did not change after altering the crickets' dietary
regime, while changes in the community structure and
metabolic profile were indicated. One hypothesis is that
the changes observed in the VFA profiles are a result of
the de novo induction of different metabolic pathways
without changes in the relative abundance of the
hindgut microbial populations. However, the results in
this study showed that the hindgut community percentage G + C profiles changed after changes in diet, implying that shifts in the numerically dominant populations
occurred due to the dietary perturbations. It is perhaps
surprising that these diets did have as much impact in
the community structure as was observed because they
are thought to contain rather common substrates which
most gut-inhabiting genera should be able to metabolize. Because of the difficulty of culturing many of the
members of this community, these results underscore the
value of studies based on directly extracted community
DNA to study the structural dynamics of insect gut communities. Further studies are needed to demonstrate
whether changes in the numerically dominant populations are truly responsible for the changes in fermentation profiles observed in Acheta domesticus.
The percentage G + C profile method indicates community composition with resolution at approximately
the family to genus level, and thus much diversity at the
species level and ecotype level could still be present,
even in communities with apparently low complexity.
While this is admittedly a coarse level of resolution, differences in the overall structure of the microbial communities in the guts of crickets on different diet
regimens were clearly demonstrated. As percentage
G + C content is indicative of bacterial taxonomic
groups (Holt & Krieg 1984), such information can also
be useful in devising improved culturing strategies.
This approach thus represents an appropriate tool for
comparative studies of total bacterial community structure in a single, relatively simple analysis which was the
objective of this work. It is, however, not well suited to
demonstrating the presence or absence of specific
groups, genera or populations without being coupled to
additional analyses with finer resolving power, such as
hybridization analyses with highly specific functional or
phylogenetic probes. Moreover, this method is not sensitive to minor members of the community. However, it is
reasonable to assume that many microbial processes are
largely driven by the most-abundant group of microorganisms inhabiting flow-through ecosystems such as
the insect gut.
Acknowledgements
We thank Mary Ann Bruns for technical advice on the use of
CsCl–bisbenzimide gradients, Sandy Marsh for technical assistance with fermentation metabolite analyses, and James Kastner
for critically reviewing the manuscript. This study was supported
by the National Science Foundation Grant No. DEB-9120006.
References
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG,
Smith JA, Struhl K (1987) Current Protocols in Molecular
Biology. Greene Publishing Associates and Wiley-Interscience.
New York, NY.
Bignell DE (1984) The arthropod gut as an environment for
microorganisms. In: Invertebrate–Microbial Interactions (eds
Anderson JM, Rayner ADM, Walter DEH), pp. 205–228.
Cambridge Press, New York, NY.
Bracke JW, Cruden DL, Markovetz AJ (1978) Effect of metronidazole on the intestinal microflora of the American cockroach,
Periplaneta americana L. Antimicrobiological Agents in
Chemotherapy, 13, 115–120.
Breznak JA (1984) Biochemical aspects of symbiosis between termites and their intestinal microbiota. In: Invertebrate–Microbial
Interactions (Anderson JM, Rayner ADM, Walter DEH), pp.
173–203. Cambridge Press, New York, NY.
Breznak JA, Brune A (1994) Role of microorganisms in the digestion of lignocellulose by termites. Annual Review of
Entomology, 39, 453–487.
Buchner P (1965) Endosymbiosis of Animals with Plant
Microorganisms. Wiley Interscience, New York NY.
Campbell BC (1989) On the role of microbial symbionts in herbivorous insects. In: Insect–Plant Interactions (ed. Bernays EA),
pp. 1–44, CRC Press, Boca Raton, FL.
Cruden DL, Markovetz AJ (1979) Carboxymethyl cellulose
decomposition by intestinal bacteria of cockroaches. Applied
and Environmental Microbiology, 38, 369–372.
Cruden DL, Markovetz AJ (1987) Microbial ecology of the cockroach gut. Annual Review of Microbiology, 41, 617–643.
Dubois M, Giles KA, Hamilton JK, Rebers PA, Smith F (1965)
Colorimetric method for the determination of sugars and
related substances. Analytical Chemistry, 28, 350–356.
Hobbie JE, Daley RJ, Jasper S (1977) Use of Nuclepore filters for
counting bacteria by fluorescent microscopy. Applied and
Environmental Microbiology, 33, 1225–1228.
Holben WE (1997) Isolation and purification of bacterial
community DNA from environmental samples. In: ASM
Manual of Environmental Microbiology (eds Hurst CJ, Knudsen
GR, McInerney MJ, Stetzenbach LD, Walter MV), pp.
431–436. American Society for Microbiology Press,
Washington DC.
Holben WE, Harris D (1995) DNA-based determination of total
bacterial community structure in environmental samples.
Molecular Ecology, 4, 627–631.
Holben WE, Jansson JK, Chelm BK, Tiedje JM (1988) DNA probe
method for the detection of specific microorganisms in the soil
bacterial community. Applied and Environmental Microbiology,
54, 703–711.
Holben WE, Noto K, Sumino T, Suwa Y (1998) Molecular analysis
of bacterial communities in a three-compartment granular
activated sludge system indicates community-level control by
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 761–767
INFLUENCE OF DIET ON THE CRICKET HINDGUT BACTERIAL COMMUNITY
incompatible nitrification processes. Applied and Environmental
Microbiology, in press.
Holt J, Krieg NR (1984) Bergey’s Manual of Systematic Bacteriology,
Vol. 1, pp. 602–662. Williams and Wilkins, Baltimore, MD.
Hungate RE (1975) The rumen microbial ecosystem. Annual
Review of Ecology and Systematics, 6, 39–66.
Jones CG (1984) Microorganisms as mediators of plant resource
exploitation by insect herbivores. In: A New Ecology: Novel
Approaches to Interactive Systems (eds Price PW,
Slobodochikoff CN, Gaud WS), pp. 53–99. Wiley Interscience,
New York, NY.
Kane MD, Breznak JA (1991) Effect of host diet on production of
organic acids and methane by cockroach gut bacteria. Applied
and Environmental Microbiology, 57, 2628–2634.
Kaufman MG (1988) The role of anaerobic bacterial metabolism
in the nutrition of crickets (Orthoptera: Gryllidae). PhD
Thesis, Michigan State University, MI.
Kaufman MG, Klug MJ (1990) Microbial community metabolism
in the digestive tract of crickets (Orthoptera: Gryllidae):
implications for omnivorous insects. In: Microbiology in
Poecilotherms (ed. Lesel R), pp. 69–72. Elsevier Science
Publishers.
Kaufman MG, Klug MJ (1991) The contribution of hindgut bacteria to dietary carbohydrate utilization by crickets
(Orthoptera: Gryllidae). Comparative Biochemistry and
Physiology, 98A, 117–123.
Kaufman MG, Klug MJ, Merritt RW (1989) Growth and food utilization parameters of germ-free house crickets, Acheta domesticus. Journal of Insect Physiology, 35, 957–967.
Lawson DL, Klug MJ (1989) Microbial fermentation in the
hindguts of two stream detritivores. Journal of North America
Benthological Society, 8, 85–91.
Lovley DR, Klug MJ (1982) Intermediary metabolism of organic
matter in the sediments of a eutrophic lake. Applied and
Environmental Microbiology, 43, 552–560.
Martoja R (1966) Sur quelques aspects de la biologie des orthoptires in relation avec la presence des concentrations microbiennes (bacteria intestinals, rickettsies). Annals of Entomology
(France), 2, 753–840.
McFarlane JE, Distler MHW (1981) The effect of rutin on growth,
fecundity, and food utilization in Acheta domesticus (L.).
Journal of Insect Physiology, 28, 85–88.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 761–767
767
Moore WEC, Holdeman LV (1974) Human fecal flora: the normal
flora of 20 Japanese–Hawaiians. Applied Microbiology, 27,
961–968.
Ohkuma M, Kudo T (1996) Phylogenetic diversity of the intestinal bacterial community in the termite Reticulitermes speratus.
Applied and Environmental Microbiology, 62, 461–468.
Paster BJ, Dewhirst FE, Cooke SM, Fussing V, Poulsen LK,
Breznak JA (1996) Phylogeny of not-yet-cultured spirochetes
from termite guts. Applied and Environmental Microbiology, 62,
347–352.
Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory Press, NY.
Santo Domingo JW (1994) Use of nucleic acid methods to study
the bacterial community of the cricket hindgut. PhD Thesis,
Michigan State University, MI.
Ulrich RG, Buthala DA, Klug MJ (1981) Microbiota associated
with the gastrointestinal tract of the common house cricket,
Acheta domesticus. Applied and Environmental Microbiology, 41,
246–254.
Visuvanathan S, Moss MT, Stanford JL, Hermo-Taylor J,
McFadden JJ (1989) Simple enzymic method for isolation of
DNA from diverse bacteria. Journal of Microbiological Methods,
10, 59–64.
The CsCl gradients were performed in the laboratory of Jim
Tiedje while the volatile fatty acids analyses were performed in
the laboratory of Michael Klug. J. W. Santo Domingo, William
Holben, and Dave Harris participated in the molecular aspects of
the project. Michael Kaufman provided guidance on all aspects
regarding cricket microbiology and analysis of fermentation
products. Previous studies regarding the microbial ecology of
crickets have relied on culturable techniques. Hence, our current
understanding of the composition and structure of the cricket
hindgut microbial community is mostly based on culturable
populations. This study represents one of the first attempts to
examine the hindgut microbial ecology of omnivorous insects
using a DNA-based approach and was possible due to our
common interest in applying molecular techniques to study the
relationship between structure and function in natural microbial
communities.