THE ROLE OF MACROINVERTEBRATES IN BURULI ULCER DISEASE IN
GHANA, WEST AFRICA
By
Ryan K. Kimbirauskas

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Entomology
2011

ABSTRACT
THE ROLE OF MACROINVERTEBRATES IN BURULI ULCER DISEASE IN
GHANA, WEST AFRICA
By
Ryan K. Kimbirauskas
Buruli ulcer (BU) is an emerging, neglected, infectious disease most often
associated with poor, rural communities within developing nations. To date, the
disease has been reported from at least 32 countries, with the highest frequency
of new cases being reported from the West African nations of Cote D’ Ivoire,
Benin and Ghana. It is known that Mycobacterium ulcerans is the pathogen
responsible for causing BU disease; however, researchers have yet to
conclusively identify the extent of the pathogen’s distribution in the environment,
the reservoir(s) of the pathogen in nature, or the mode(s) of transmission to
humans. It is widely accepted that BU disease is in some way related to
exposure to freshwater environments, and furthermore, it has been suggested
that human activities leading to environmental disturbance increase risk of BU
infection. Aquatic macroinvertebrates have been implicated as both potential
reservoirs and vectors M. ulcerans infection to humans; however, field-based
ecological studies to investigate the role of macroinvertebrates in BU disease
have not been conducted. The purpose of this study was to: 1) characterize and
compare overall macroinvertebrate communities from aquatic environments in
Ghana, West Africa: 2) identify macroinvertebrate community associations with
the presence and absence of M. ulcerans in aquatic environments: and 3)
identify potential relationships between specific macroinvertebrates and M.

ulcerans. Results from this large survey of aquatic environments in Ghana
suggest that macroinvertebrate communities and individual taxa may be useful
sentinels for initial identification of pathogen presence or habitat conditions
associated with disease agent transmission; however, further studies are needed
to elucidate the exact role of macroinvertebrates as reservoirs of M. ulcerans and
potential vectors of BU.

This dissertation is dedicated to my parents and sister: Paul, Cinda and Paula
Kimbirauskas, respectively.

iv

ACKNOWLEDGEMENTS

I would first like to thank my advisor, Dr. Richard Merritt. Throughout my
graduate career he provided endless support, encouragement, friendship, and
new opportunities that have made me a better scientist, person and future
mentor. I must also thank Dr. Michael Kaufman, Dr. Eric Benbow, and Dr. Mollie
McIntosh, who not only served on my committee, but also helped with field
logistics, data management and analyses for this project. I would like to
acknowledge collaborators on this project: Charles Quaye, Charles Yeboah,
Lydia Mosi, and Daniel Boyake from the University of Ghana; and Dr. Heather
Williamson and Dr. Pamela Small from the University of Tennessee. There are
numerous individuals at MSU I would also like to thank who played a significant
role in my development during my graduate years. The administrative and
support staff for the Department of Entomology is the best around and not only
made my work easier, but were always pleasant to visit. I would like to recognize
and thank the many people who have helped either in the field or laboratory
processing samples: Todd White, Rebecca Kolar, Tom Alwin, Bethany Coggins,
Alvin Makhon-Moore, and Sanjeev Mahabir. Gary Parsons should also be
recognized for his help and guidance in identifying insect specimens. Finally I
would like to thank the members of the Merritt Lab, including: Jaree Johnson,
Kristi Zurwaski, Osvaldo Hernandez, Mollie McIntosh, Eric Benbow, Christian
Lesage, Todd White, Emily Campbell, and Matt Wessner; for helping make this a
lot of fun and more than just the process of obtaining a degree.

v

TABLE OF CONTENTS
LIST OF TABLES ......................................................................................... viii
LIST OF FIGURES ....................................................................................... x
CHAPTER 1
INTRODUCTION TO THE ASSOCIATIONS BETWEEN AQUATIC INSECTS
AND THE ECOLOGY OF BURULI ULCER DISEASE ................................. 1
Figures ................................................................................................... 9
Literature Cited....................................................................................... 11
CHAPTER 2
ASSOCIATIONS BETWEEN MYCOBACTERIUM ULCERANS AND
MACROINVERTEBRATE ASSEMBLAGES IN AQUATIC ENVIRONMENTS
OF GHANA, WEST AFRICA ........................................................................ 19
Introduction ............................................................................................ 19
Materials and Methods........................................................................... 22
Study Location and Scale ....................................................... 22
Macroinvertebrate Sample Collections ................................... 22
Detection of Mycobacterium Ulcerans .................................... 23
Primers, PCR conditions and sequencing .............................. 24
Site Classification.................................................................... 24
Data Analyses ......................................................................... 25
Results ................................................................................................... 26
Discussion.............................................................................................. 29
Tables .................................................................................................... 37
Figures ................................................................................................... 41
Literature Cited....................................................................................... 44
CHAPTER 3
SEASONAL DIFFERENCES IN BENTHIC MACROINVERTEBRATE
ASSEMBLAGES IN RELATION TO THE PRESENCE OF M. ULCERANS IN
AQUATIC ENVIRONMENTS OF GHANA, WEST AFRICA ......................... 56
Introduction ............................................................................................ 56
Materials and Methods........................................................................... 59
Study Location and Site Selection .......................................... 59
Macroinvertebrate Sample Collections ................................... 60
Detection of Mycobacterium Ulcerans .................................... 60
Primers, PCR conditions and sequencing .............................. 61
Data Analyses ......................................................................... 62
Results ................................................................................................... 63
Discussion.............................................................................................. 75
Tables .................................................................................................... 84
Figures ................................................................................................... 95
Literature Cited....................................................................................... 97

vi

CHAPTER 4
DETECTION OF NATURAL PREY IN THE GUTS OF AFRICAN CREEPING
WATER BUGS (HEMPTERA: NAUCORIDAE) USING SEQUENCED CLONES
OF PCR-AMPLIFIED GUT CONTENTS....................................................... 105
Introduction ............................................................................................ 105
Materials and Methods........................................................................... 108
Sample Collection and Preparation ........................................ 108
DNA Extraction and PCR ........................................................ 108
Molecular Cloning ................................................................... 109
Data Analysis .......................................................................... 109
Results ................................................................................................... 110
Discussion.............................................................................................. 111
Tables .................................................................................................... 114
Figures ................................................................................................... 115
Literature Cited....................................................................................... 118

vii

LIST OF TABLES
Table 2.1: Number of sites macroinvertebrates were observed and the
total macroinvertebrate specimens in lotic and lentic habitats,
Ghana, W. Africa. A total of 73,892 invertebrates from 77 unique
taxa were identified. ER+ represents waterbodies where the
pathogen was detected and ER- represents waterbodies where the
pathogen was not detected................................................................ 37
Table 3.1: Locations and general information for each of the six waterbodies
used in the seasonal analysis of macroinvertebrate associations with BU
and M. ulcerans. MW = modified wetland; MP = modified pond; BU- = no
cases of BU had been reported from the village; and BU+ = at least 3
cases of BU had previously been reported from the village………….84
Table 3.2: Macroinvertebrate taxa and total specimens collected June
2007 to July 2008 .............................................................................. 85
Table 3.3: Model summary of the multiple regression analysis of
macroinvertebrate metric relationships with presence and absence
of BU cases, within the six seasonal study sites ............................... 88
Table 3.4: Descriptive statistics for regression analysis of macroinvertebrate
metrics and the presence and absence of BU cases. BU- = no cases of
BU reported prior to the beginning of this study and BU+ = at least 3
cases of BU reported prior to the beginning of this study…………… 89
Table 3.5: Model summary of the multiple regression analysis of
macroinvertebrate metric relationships with presence and absence
of BU cases, within the six seasonal study sites ............................... 90
Table 3.6: Descriptive statistics for regression analysis of macroinvertebrate
metrics and the presence and absence of ER within each waterbody
during the entire study period…………………………………………... 91
Table 3.7: Summary table indicating that two of four hypotheses reached
statistically significant differences for the repeated measures profile
analyses. The dependent variables were total macroinvertebrate
abundance, total taxa, Shannon-Weiner Diversity, Simpson’s
Heterogeneity, Margalef’s Richness, and Pielou’s Eveness.............. 92
Table 3.8: Descriptive statistics for the repeated measures profile analysis of
macroinvertebrate metrics. Results presented here are for both season
and waterbody……………………………………………………………. 93

viii

Table 3.9: Significant positive and negative Pearson correlations (P <
0.05) between macroinvertebrate taxa and both cases of Buruli
ulcer and ER detection within waterbodies ........................................ 94
Table 4.1: Results of comparisons to the NCBI nucleotide database using
blastn queries of rrnL and cox1sequences that were PCRamplified from Naucoridae guts and cloned (see methods). Taxa
listed were, in each case, first on the hit table. Only full-length
inserts (450 – 478 bp rrnL, 658 bp cox1) were considered. Most
full-length inserts were identical to our sequences obtained from
direct sequencing of sampled Naucoridae and resulted in a top
blastn hit of Macrocoris sp. (rrnL) or Hemiptera sp. (cox1) (data not
shown) ............................................................................................... 114

ix

LIST OF FIGURES
Figure 1.1: Global map showing distribution and concentration of Buruli
ulcer cases. These data were obtained from the World Health
Organization and represent confirmed cases as of 2009. Map
provided by the World Health Organization (2009) and modified to
fit formatting requirements for this dissertation .................................. 9
Figure 1.2: Conceptual model illustrating potential reservoirs and
movement of Mycobacterium ulcerans within and among aquatic
environments. Dark arrows indicate potential movement within a
waterbody and; dashed lines and arrows represent potential
dissemination pathways to other water bodies. This diagram was
published in Merritt et al. (2005) and modified to fit formatting
requirements. All drawings made by RA MacKarrall ......................... 10
Figure 2.1: Map illustrating the locations of 98 water bodies sampled for
macroinvertebrates from five geographic regions in southern
Ghana, Africa. Villages were randomly selected within each region
and water bodies were selected within each village, based on
location (<100-200m from community housing structures) and
human use (daily domestic activities) to reflect aquatic
environments with potential human exposure to Buruli ulcer.
Community discussions on water body selection were conducted in
each village as described by Benbow et al. (2005). .......................... 41
Figure 2.2: A three-axis NMDS solution explained 70% of the total variation in the
macroinvertebrate community (stress: 16.9, p=0.004), with 1% on axis 1,
42% on axis 2, and 27% on axis 3. No significant differences in the
overall macroinvertebrate community structure between sites that were
ER+ and ER- (MRPP: A=0.001, p = 0.23) were observed. The NMDS
ordination did identify differences in macroinvertebrate community
structure based on water body flow (MRPP: A=0.046, p < 0.000)….. 42
Figure 2.3: A non-metric multi-diminsional scaling (NMDS) ordination of
macroinvertebrate communities collected from 98 waterbodies in
Ghana, Africa. Each circle or triangle represents the overall
macroinvertebrate community at each site and symbols closer
together have more similar community structure while symbols
further apart were more dissimilar communities. Closed circles
represent ER+ and open triangles represent ER- habitats. A threeaxis NMDS solution explained 80% of the total variation in the
macroinvertebrate community (stress: 15.4, p=0.004), with 38% on
axis 1, 14% on axis 2, and 28% on axis 3. We found significant
differences between lotic sites that were ER+ and ER- (MRPP:
A=0.01, p= 0.02) ................................................................................ 43

x

Figure 3.1: Estimated marginal means for six macroinvertebrate metrics across
five sampling seasons. The macroinvertebrate metrics were total
specimens, total taxa, Shannon-Weiner Diversity, Simpson’s
Heterogeneity, Margalef’s Richness, and Pielou’s Eveness. The sampling
seasons are listed as Season 1 (June 2007), Season 2 (November 2007),
Season 3 (February 2008), Season 4 (April 2008), and Season 5 (July
2008).…….…………………………………………………………………95
Figure 3.2: Estimated marginal means for six macroinvertebrate metrics across
six waterbodies. The macroinvertebrate metrics were total specimens,
total taxa, Shannon-Weiner Diversity, Simpson’s Heterogeneity,
Margalef’s Richness, and Pielou’s Eveness. The were from the following
villages: Site 1= Otinibi, Site 2= Danfa, Site 3= Teimen, Site 4= Afieman,
Site 5= Kotoku, Site 6= Nsakima………………………………………...96
Figure 4.1: Maximum likelihood phylogenetic tree of rrnL using a GTR model of
evolution, including newly sequenced Naucoris sp. and potential prey
(blue terminals), cloned PCR products from Naucoris sp. guts and
mouthparts (red terminals), and highly ranked sequences according to
blastn queries (black, see text for criteria)……………………………. 115

xi

CHAPTER 1
INTRODUCTION TO THE ASSOCIATIONS BETWEEN AQUATIC INSECTS
AND THE ECOLOGY OF BURULI ULCER DISEASE
Buruli ulcer (BU) is an emerging neglected disease caused by infection of
Mycobacterium ulcerans (Walsh et al. 2008, Duker et al. 2006, WansbroughJones and Philips 2006, van der Werf et al. 2005). The disease can inflict people
of any age or gender, although nearly 70% of the cases occur in children under
the age of 15 years and in some communities more females are infected than
males (WHO 2000, 2008; Duker et al. 2004). Confirmed cases of BU have been
reported from 32 countries mainly in Africa, Australia, southeast Asia, China,
Central and South America, and the Western Pacific (Johnson et al. 1999; WHO
2000, 2008, Guerra et al. 2008, Walsh et al. 2009) (Fig. 1.1). Endemism is
primarily confined to tropical and subtropical climates (WHO 2000; Duker 2006);
however, outbreaks have occurred in a few isolated regions in temperate
Australia (Hayman 1991; Veitch et al. 1997; Johnson et al. 2009). Infection rates
are more severe in rural and remote areas of developing nations (WHO 2008),
and the highest numbers of new cases come from the west African nations of
Cote d’Ivoire, Ghana, and Benin, where BU is now the second most frequent
mycobacterial disease in humans after tuberculosis (Debecker et al. 2004,
Amofah et al. 2002, Sopoh et al. 2007). Cases of BU appear to be on the rise
throughout endemic regions, however, true incidence is difficult to determine due
to poor case confirmation and surveillance measures (WHO 2008).
The genus Mycobacterium comprises more than 50 species, most of

1

which are nonpathogenic environmental bacteria closely related to the soil
bacteria Streptomyces and Actinomyces (Cosma et al. 2003). M. ulcerans,
however, is a facultative environmental pathogen belonging to the M. marinum
complex and is closely related to M. tuberculosis and M. leprae, the causative
agents of tuberculosis and leprosy, respectively (Chemlal et al. 2002; Kaser et al.
2009; Stinear et al. 2000a; Stinear et al. 2004; Yip et al. 2007). The major
virulence determinant of M. ulcerans is an immunosuppressant toxin called
mycolactone, which is a polyketide-derived macrolide secreted by M. ulcerans
causing cell necroses and tissue damage in infected individuals (George et al.
1999; Gunawardana et al. 1999; Demangel, et al, 2009). Mycobacterium
ulcerans is characterized as a slow-growing mycobacteria, sensitive to UV light
(Stinear et al. 2004), and has optimal growth under a narrow range of
temperatures (WHO 2000; Yeboah-Manu et al. 2004; Boisvert and Schroder
1977; Garrity et al. 2001) and in oxygen deprived environments (Palomino et al.
1998). The combination of these characteristics suggests M. ulcerans has
adapted to a specific niche and does not live freely in the environment (Stinear et
al. 2000a, Stinear et al. 2004, Stinear et al. 2007). Improved PCR techniques
have allowed for more accurate testing for M. ulcerans in environmental samples
(Ross et al. 1997; Stinear et al. 1999; Stinear et al. 2000b,c, 2004, Johnson et al.
2005, Fyfe et al. 2007, Lavender et al. 2008; Williamson et al. 2008), and have
contributed to the detection of M. ulcerans DNA from soil and mud, detritus,
biofilms, filtered water, fish, frogs, snails, spiders, several insect groups and other
invertebrates (Williamson et al. 2008, Benbow et al. 2008, Marsollier et al. 2002b,

2

2004a, b, Eddyani et al. 2004, Johnson et al. 2007, Portaels et al. 1999, 2001,
2008, Stinear et al. 2000b, Kotlowski et al. 2004, Fyfe et al. 2007, Trott et al.
2004). These findings have provided more insight into the distribution of M.
ulcerans throughout aquatic environments; however, a thorough understanding
of the ecology of M. ulcerans is lacking and remains understudied.
Buruli ulcer has been referred to as the “mystery disease”, in part,
because researchers still do not know how the disease is spread or where the
primary source of M. ulcerans is in the environment. Most epidemiological
studies have associated cases of BU with proximity and prolonged exposure to
freshwater environments (Lunn et al. 1965, Revill and Barker 1972, Barker and
Carswell 1973, Duker et al. 2006, Marston et al. 1995, Walsh et al. 2008,
Portaels 1995, Debacker et al. 2006, Noeske et al. 2004, Johnson et al. 2007,
Wagner et al. 2008a, WHO 2000). It has further been suggested that people
living in areas prone to flooding are at higher risk of infection (Barker and
Carswell 1973, Wagner et al. 2008a; Radford 1974b, Barker 1972 Meyers et al.
1996, Portaels 1995, Hayman 1991). Anthropogenic disturbances to
waterbodies and adjacent landscapes have also been linked to higher disease
incidence. In particular, the damming of streams and rivers, modification of
wetlands, deforestation practices, increased agriculture development, and sand
mining operations are believed to promote proliferation of M. ulcerans in the
environment and therefore increase risk of becoming infected (Hayman 1991b;
Marston et al. 1995; Meyers et al. 1996; Johnson et al. 1999; Portaels et al.
2001, Wagner et al. 2008a, Merritt et al. 2005, Duker et al. 2006, Kibadi et al.

3

2008). Although nearly all the epidemiological studies on BU have associated
disease outbreaks with communities in close proximity to disturbed aquatic
environments, the source of infection and mode of transmission still remains a
mystery (Merritt et al. 2010).
Two hypotheses explaining potential pathways for M. ulcerans infection
have been proposed. The first hypothesis suggested M. ulcerans could be
inhaled or ingested as an aerosol (Connor and Lunn, 1965; Hayman, 1991;
Veitch et al. 1997; Johnson et al., 1999); however, this hypothesis has since
been considered unlikely as a primary mode of transmission. The second
hypothesis, which is more widely accepted, is mechanical transmission where M.
ulcerans enters an individual through direct contact with the pathogen from
contaminated soils, water, plant biofilms and aquatic insects (Barker 1971;
Radford, 1974; Hayman, 1991; Johnson et al., 1999; Portaels, 1995; Portaels,
1999; Portaels, 2001; Merritt et al. 2005). Portaels et al. (1999) first
hypothesized that predacious aquatic insects infected with M. ulcerans
mechanically transmit the bacteria to humans through bites and offered a model
describing the movement of M. ulcerans through trophic pathways. Merritt et al.
(2005) elaborated on the role of aquatic invertebrates in maintaining M. ulcerans
in aquatic food webs and expanded on this model with the addition of potential
pathways for the dissemination of M. ulcerans between waterbodies (Fig. 1.2).
The work by Portaels and colleagues prompted researchers to more closely
investigate the role of aquatic insects, particularly aquatic hemipterans, as
potential vectors and environmental reservoirs of M. ulcerans. Most of the

4

subsequent field research initiatives to investigate these relationships have been
conducted in Africa and Australia, where BU cases are most prevalent. The
following section presents a brief overview of published studies examining the
associations between BU and aquatic invertebrates on these two continents.
Most of the research investigating associations between aquatic insects
and BU in Africa has taken place in west Africa where disease incidence is
greatest. Portaels et al. (1999) first suspected that aquatic insects might be
reservoirs of M. ulcerans following detection of the pathogen in water bugs
(Hemiptera: Gerridae) collected from wetlands in endemic areas of Benin. Their
results led to the development of the first transmission model involving an aquatic
insect and initiated a series of studies placing biting aquatic hemipterans,
particularly Belsotomatidae and Naucoridae, as potential vectors and reservoirs
of M. ulcerans. More recently, Portaels et al. (2008) cultured M. ulcerans from a
water strider (Gerris sp.) and became the first to successfully culture M. ulcerans
from the environment. Benbow et al. (2008) were the first to conduct a largescale survey of aquatic invertebrates associated with waterbodies in both BU
endemic and non-endemic regions. From their research in Ghana they
concluded that aquatic insects were unlikely vectors, in part due to relatively low
numbers of biting hemipterans (Belostomatidae, Naucoridae) compared to
reported disease occurrence (Benbow et al. 2008). In addition, they found that M.
ulcerans was more widespread in the environment than previously believed and
reported several aquatic insect taxa that tested positive for M. ulcerans
(Williamson et al. 2008). A number of studies have provided similar results and

5

identified M. ulcerans in association with aquatic insects, as well as snails,
tadpoles, and fish (Kotlowski et al. 2004; Marrion et al. 2010; Morsolier et al.
2004a; Portaels et al. 2001, 2008).
A series of laboratory experiments by Marsollier and colleagues
demonstrated that naucorid water bugs (Naucoris cimicoides sp.) could become
infected by feeding on inoculated prey and then transmit M. ulcerans to
uninfected mice (2002b). They also reported that M. ulcerans could survive and
multiply within the salivary glands of N. cimicoides sp. (2002a, 2003, 2004a,
2005). Results from these studies reinforced the hypothesis of an insect vector
and environmental reservoir, however, these results have been scrutinized
because African naucorids were not used, the amount of innocula present in the
naucorids was much higher than what would be found in nature, and the results
only showed indirect transmission (Benbow et al. 2008). Mosi et al. (2008)
investigated the trophic movement of M. ulcerans experimentally using
belostomatids (Appasus sp.) collected from Ghana and found that water bugs
can become infected after feeding on inoculated prey; however, they concluded
that replication of M. ulcerans did not occur in the salivary complex and was most
likely restricted to the exoskeleton. Wallace et al. (2010) found that M. ulcerans
could become concentrated in filter-feeding mosquito larvae and then acquired
by predaceous mosquito larvae up a food chain. However, the bacteria was
found to not pass through all instars nor survive metamorphosis to the adult
stage. Together, these results provide evidence that M. ulcerans can become
concentrated and passed trophically up an aquatic food chain and support the

6

hypothesis of an aquatic invertebrate reservoir; however, the role of an aquatic
insect as a vector involved in actual transmission of M. ulcerans requires further
investigation (Merritt et al. 2010).
Outbreaks of BU in Australia have occurred in confined areas and most of
the research efforts have been directed towards associating mosquitoes with
disease incidence. Johnson et al. (2007) sampled salt marsh mosquitoes
following an outbreak of BU and found M. ulcerans positive adult mosquitoes in
pooled samples collected from the outbreak area. Further epidemiological work
has supported the mosquito vector hypothesis (Fyfe et al. 2007; Lavender et al.
2008); however, conclusive evidence demonstrating transmission by adult
mosquitoes is lacking. Tobias et al. (2009) conducted feeding experiments in an
attempt to connect a potential environmental source of M. ulcerans with adult
mosquitoes and found that mosquito larvae could consume and concentrate M.
ulcerans, but they were unable to demonstrate that the bacteria can persist
beyond the fourth instar. These results were consistent with findings by Wallace
et al. (2010), which collectively are significant in that they showed that M.
ulcerans can be maintained in aquatic food webs. It should be further noted that
despite the literature from both Australia and Africa showing an association of
aquatic insects in the transmission of this disease, major scientific criteria are
lacking for implicating the roles of living agents as biologically significant
reservoirs and/or vectors of pathogens (Merritt et al. 2010).


In order to better understand associations between aquatic
macroinvertebrates and BU disease, more quantitative studies evaluating the

7

ecology of M. ulcerans are needed. The major objective of this research was to
systematically assess and characterize the macroinvertebrate communities
within aquatic habitats of disease endemic and non-endemic areas in Ghana,
West Africa. The specific objectives of this study were divided into the following
three chapters: 1) Associations between M. ulcerans and benthic
macroinvertebrate assemblages in aquatic environments of Ghana, West Arica,
2) Seasonal differences in aquatic macroinvertebrate assemblages in relation to
the presence of M. ulcerans in waterbodies of Ghana, West Africa, and 3) Gut
content analysis of Naucoridae and trophic relationships of benthic
macroinvertebrates in waterbodies of Ghana, West Africa.

8




Number of reported BU cases, 2009

1000 and above
500-1000
100-500
Less than100
Previously reported cases







No cases reported



Figure 1.1. Global map showing countries where Buruli ulcer disease has been
confirmed and the number of cases reported in 2009 for each country. Map
provided by World Health Organization (2009) and modified to fit formatting
requirements.
“For interpretation of the references to color in this and all other figures, the
reader is referred to the electronic version of this thesis (or dissertation).”




9








Figure 1.2. Conceptual model illustrating potential reservoirs and movement of
Mycobacterium ulcerans within and among aquatic environments. Dark arrows
indicate potential movement within a water body; dashed lines and arrows
represent potential dissemination pathways to other water bodies. This diagram
was published in Merritt et al. (2005) and modified to fit formatting requirements.
All drawings made by RA MacKarrall.




10


LITERATURE CITED




11


Literature Cited

Aiga H, Amano T, Cairncross S, Domako JA, Nanas OK, et al. (2004)
Assessing water-related risk factors for Buruli ulcer: A case-control study in
Ghana. Am J Trop Med Hyg 71: 387–392.
Amofah GK, Bonsu F, Tetteh C, Okrah J, Asamoa K, et al. (2002) Buruli ulcer
in Ghana: results of a national case search. Emerging Infectious Diseases 8:
167–170.
Asiedu K, Etuaful S (1998) Socioeconmoic implications of Buruli ulcer in
Ghana: a three-year review. Trans R Soc Trop Med & Hyg 59: 1015–1022.
18. Portaels F, Chemlal K, Elsen P, Johnson PDR, Hayman JA, et al. (2001)
Mycobacterium ulcerans in wild animals. Rev sci tech Off int Epiz 20: 252–264.
Barker, DJP (1971) Buruli disease in a district of Uganda. J. Trop. Med. Hyg. 74:
260-264.
Barker DJP (1972) The distribution of Buruli disease in Uganda. Trans R Soc
Trop Med Hyg. 66:867-74.
Barker DJP, Carswell JW (1973) Mycobacterium ulcerans infection among
Tsetse control workers in Uganda. International Journal of Epidemiology 2: 161–
165.
Benbow M, Williamson H, Kimbirauskus R, McIntosh M, Kolar R, et al. (2008)
Aquatic invertebrates as unlikely vectors of Buruli ulcer disease. Emerg Infect Dis
14: 1247–1254.
Boisvert H, Schroder KH (1977) Skin ulcer caused by Mycobacterium ulcerans in
Cameroon II: Bacteriologic study. Bull Soc Pathol Exot Filiales 70: 125–131.
Chemlal, K, Huys, G, Laval, F, et al. (2002) Characterization of an unusual
Mycobacterium: a possible missing link between Mycobacterium marinum and
Mycobacterium ulcerans. Jornal of Clinical Microbiology 40:2370.
Connor DH, and Lunn F (1965) Mycobacterium ulcerans infection. Int. J. Leprosy
33:698.
Cosma CL, Sherman DR, and Ramakrishnan L (2003) The Secret Lives of the
Pathogenic Mycobacteria. Annu. Rev. Microbiol. 57:641–76.
Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, et al. (2004)
Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin,
1997-2001. Emerg Infect Dis 10: 1391–1398.



12


Demangel C, Stinear TP, Cole ST (2009) Buruli ulcer: reductive evolution
enhances pathogenicity of Mycobacterium ulcerans. Nat Rev Microbiol 7: 50–60.
Duker AA, Carranza JM, Hale M (2004) Spatial dependency of Buruli ulcer
prevalence on arsenic-enriched domains in Amansie West District, Ghana:
implications for arsenic mediation in Mycobacterium ulcerans infection. Int J
Health Geogr. 3:19.
Duker AA, Portaels F, Hale M (2006) Pathways of Mycobacterium ulcerans
infection: A review. Environment International 32: 567–573.
Eddyani M, Ofori-Adjei D, Teugels G, De Weirdt D, Boakye D, et al. (2004)
Potential role for fish in transmission of Mycobacterium ulcerans disease (Buruli
Ulcer): an environmental study. Appl Environ Microbiol 70: 5679–5681.
Fyfe JAM, Lavender CJ, Johnson P, Globan M, Sievers A, et al. (2007)
Development and Application of Two Multiplex Real-Time PCR Assays for
the Detection of Mycobacterium ulcerans in Clinical and Environmental Samples.
Appl Environ Microbiol 73: 4733–4740.
Garrity GM (2001) Bergey’s Manual of Systematic Bacteriology; In: Garrity GM,
ed. New York: Springer-Verlag.
George KM, Chatterjee D, Gunawardana G, Welty D, Lee T, et al. (1999)
Mycolactone: a polyketide toxin from Mycobacterium ulcerans required for
virulence. Science 283: 854–857.
Gunawardana G, Chatterjee D, George KM, Brennan P, Whittern D, et al. (1999)
Mycolactone A and B: toxins of Mycobacterium ulcerans. J Amer Chem Soc 121:
6092–6093.
Guerra H, Palomino JC, Falconi E, Bravo F, Donaires N, et al. (2008)
Mycobacterium ulcerans Disease, Peru. Emerging Infectious Diseases 14: 373–
377.
Hayman J. (1991) Postulated epidemiology of Mycobacterium ulcerans infection.
International Journal of Epidemiology. 20(4): 1093-8.
Hayman J (1991) Mycobacterium ulcerans infection. The Lancet 337: 124.
Johnson PDR, Stinear TP, Hayman JA (1999) Mycobacterium ulcerans — a
minireview. J Med Microbiol 48: 511–513.
Johnson PDR, Stinear TP, Small PLC, Pluschke G, Merritt RW, et al. (2005)
Buruli ulcer (M. ulcerans Infection): new insights, new hope for disease



13


control.PLoS Med 2(4): e108.
Johnson PDR, Hayman JA, Quek TY, Fyfe JAM, Jenkin GA, et al. (2007)
Consensus recommendations for the diagnosis, treatment and control of
Mycobacterium ulcerans infection (Bairnsdale or Buruli ulcer) in Victoria,
Australia. Medical Journal of Australia 186: 64–68.
Johnson PDR, Lavender CJ (2009) Correlation between Buruli ulcer and vectorborne notifiable diseases, Victoria, Australia. Emerging Infectious Diseases 15:
614–615.
Kaser, M., Hauser, J., Small, P., Pluschke, G. (2009). Large Sequence
Polymorphisms Unveil the Phylogenetic Relationship of Environmental and
Pathogenic Mycobacteria Related to Mycobacterium ulcerans. Appl. Environ.
Microbiol. 75: 5667-5675.
Kibadi K, Panda M, Tamfum JM, Fraga AG, Filho AL, et al. (2008) New foci of
Buruli ulcer, Angola and Democratic Republic of Congo. Emerging Infectious
Diseases 14: 1790–1792.
Kotlowski R, Martin A, Ablordey A, Chemlal K, Fonteyne P, et al. (2004) Onetube
cell lysis and DNA extraction procedure for PCR-based detection of
Mycobacterium ulcerans in aquatic insects, molluscs and fish. Journal of Medical
Microbiology 53: 927–933.
Lavender CJ, Stinear TP, Johnson PDR, Azuolas J, Benbow ME, et al. (2008)
Evaluation of VNTR typing for the identification of Mycobacterium ulcerans in
environmental samples from Victoria, Australia. FEMS Microbiology Letters
287: 250–255.
Lunn HF, Connor DH, Wilks NE, Barnley GR, Kamunvi F, et al. (1965) Buruli
(Mycobacterial) ulceration in Uganda. East African Medical Journal 42: 275–288.
Marion E, Eyangoh S, Yeramian E, Doannio J, Landier J Aubry J et al. (2010)
Seasonal and regional dynamics of M. ulcerans transmission in environmental
context: deciphering the role of water bugs as hosts and vectors. PLoS Negl Trop
Dis. 6:4(7):e731.
Marsollier L, Robert R, Aubry J, Andre JS, Kouakou H, et al. (2002) Aquatic
insects as a vector for Mycobacterium ulcerans. Applied and Environmental
Microbiology 68: 4623–4628.
Marsollier L, Legras P, Manceau AL, Saint-Andre´ JP, Aubry J, et al. (2002)
Role des punaises d’eau dans la transmission de M. ulcerans. BULL ALLF or
Bulletin de l’ALLF 10: 23–25.
Marsollier L, Aubry J, Saint-Andre JP, Robert R, Legras P, et al. (2003)



14


Ecology and transmission of Mycobacterium ulcerans. Pathologie Biologie 51:
490–495.
Marsollier L, Severin T, Aubry J, Merritt RW, Saint Andre JP, et al. (2004)
Aquatic snails, passive hosts of Mycobacterium ulcerans. Appl Environ Microbiol
70: 6296–6298.
Marsollier L, Stinear TP, Aubry J, Saint-Andre J-P, Robert R, et al. (2004)
Aquatic plants stimulate the growth of and biofilm formation by Mycobacterium
ulcerans in axenic culture and harbor these bacteria in the environment. Applied
and Environmental Microbiology 70: 1097–1103.
Marsollier L, Aubry J, Coutanceau E, Andre JPS, Small PL, et al. (2005)
Colonization of the salivary glands of Naucoris cimicoides by Mycobacterium
ulcerans requires host plasmatocytes and a macrolide toxin, mycolactone.
Cellular Microbiology 7: 935–943.
Marsollier L, Andre J, Frigui W, Reysset G, Milon G, et al. (2006) Early
trafficking events of Mycobacterium ulcerans within Naucoris cimicoides. Cellular
Microbiology.
Marsollier L, Brodin P, Jackson M, Kordulakova J, Tafelmeyer P, et al. (2007)
Impact of Mycobacterium ulcerans biofilm on transmissibility to ecological niches
and Buruli ulcer pathogenesis. PLoS Pathogens 3: e62.
Marston BJ, Diallo MO, Horsburgh CR Jr., Diomande I, Saki MZ, et al.
(1995) Emergence of Buruli ulcer disease in the Daloa region of Cote D’ivoire.
Am J Trop Med Hyg 52: 219–224.
Merritt RW, Benbow ME, Small PLC (2005) Unraveling an Emerging Disease
Associated with Disturbed Aquatic Environments: The Case of Buruli Ulcer.
Frontiers in Ecology and the Environment 3: 323–331.
Merritt RW, Walker ED, Small PLC, Wallace JR, Johnson PDR, et al. (2010)
Ecology and Transmission of Buruli Ulcer Disease: A Systematic Review. PLoS
Negl Trop Dis 4(12): 1-15.
Meyers WM, Tignokpa WM, Priuli GB, Portaels F (1996) Mycobacterium ulcerans
infection (Buruli ulcer): first reported patients in Togo. British Journal of
Dermatology 134: 1116–1121.
Mosi L, Williamson H, Wallace JR, Merritt RW, Small PLC (2008) Persistent
association of Mycobacterium ulcerans with West African predaceous insects of
the family Belostomatidae. Applied and Environmental Microbiology 74:
7036–7042.




15


Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, et al. (2004) Buruli ulcer
disease in Cameroon rediscovered. Am J Trop Med Hyg 70: 520–526.
Palomino JC, Obiang AM, Realini L, Meyers WM, and Portaels F (1998) Effect of
Oxygen on Growth of Mycobacterium ulcerans in the BACTEC System. Journal
of Clinical Microbiology. Vol. 36, No. 11, p. 3420-3422,
Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, et al. (2006)
Mycobacterium ulcerans disease (Buruli Ulcer) in a rural hospital in Bas-Congo,
Democratic Republic of Congo, 2002-2004. American J Tropical Medicine
and Hygiene 75: 311–314.
Portaels, F. (1995) Epidemiology of mycobacterial diseases. Clin. Dermatol.
13:207-222.
Portaels F, Elsen P, Guimaraes-Peres A, Fonteyne P, Meyers WM (1999)
Insects in the transmission of Mycobacterium ulcerans infection. The Lancet 353:
986.
Portaels F, Chemlal K, Elsen P, Johnson PDR, Hayman JA, et al. (2001)
Mycobacterium ulcerans in wild animals. Rev sci tech Off int Epiz 20: 252–264.
Portaels F, Manuel S, Meyers W (2009) Buruli ulcer. Clinics in dermatology 27:
291–305.
Portaels F, Meyers WM, Ablordey A, Castro AG, Chemlal K, et al. (2008) First
Cultivation and Characterization of Mycobacterium ulcerans from the
Environment. PLoS Neglected Tropical Diseases 2: e178.
Radford AJ (1974) Mycobacterium ulcerans: a review, I: Epidemiology. Papua
New Guinea Medical Journal 17: 129–133.
Revill WDL, Barker DJP (1972) Seasonal distribution of mycobacterial skin
ulcers. Brit J prev soc Med 26: 23–27.
Ross BC, Marino L, Oppedisano F, Edwards R, Robins-Browne RM, et al. (1997)
Development of a PCR assay for rapid diagnosis of Mycobacterium ulcerans
infection. Journal of Clinical Microbiology 35: 1696–1700.
Silva MT, Portaels F, Pedrosa J (2007) Aquatic Insects and Mycobacterium
ulcerans: an association relevant to Buruli ulcer control? PLoS Medicine 4: e63.
Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, et al. (2007) Buruli
ulcer surveillance, Benin, 2003-2005. Emerg Infect Dis 13: 1374–1376.
Stinear T, Ross, BC, Davies JK, Marino L, Robins-Browne RM, Oppedisano F,



16


Grant A, Jenkin GA, Potaels, F, Ross, BC, Oppedisano F, Sievers A, Johnson
PDR (1999) Identification and Characterization of IS2404 and IS2606: Two
Distinct Repeated Sequences for Detection of Mycobacterium ulcerans by PCR.
American Society for Microbiology vol. 37, No. 4.
Stinear T, Jenkin GA, Johnson PDR, Davies JK (2000) Comparative Genetic
Analysis of Mycobacterium ulcerans and Mycobacterium marinum Reveals
Evidence of Recent Divergence. Journal of Clinical Microbiology vol. 182, No. 22.
pp-6322-6330.
Stinear T, Davies JK, Jenkin GA, Hayman JA, Oppedisano F, et al. (2000)
Identification of Mycobacterium ulcerans in the environment from regions in
Southeast Australia in which it is endemic with sequence Capture-PCR. Appl
Environ Microbiol 66: 3206–3213.
Stinear T, Davies JK, Grant A, Jenkin GA, Potaels, F, Ross, BC, Oppedisano F,
Purcell M, Hayman JA, Johnson PDR (2000) A Simple PCR Method for Rapid
Genotype Analysis of Mycobacterium ulcerans. Journal of Clinical Microbiology
vol. 38, No. 4. pp-1482-1487.
Stinear TP, Mve Obiang A, Small PL, Frigui W, Pryor MJ, et al. (2004) Giant
plasmid-encoded polyketide synthases produce the macrolide toxin of
Mycobacterium ulcerans. Proc Natl Acad Sci U S A 101: 1345–1349.
Stinear T, Johnson PDR (2007) From marinum to ulcerans: a mycobacterial
human pathogen emerges. Microbe 2: 187–194.
Tobias N, Seemann T, Pidot S, Porter J, Marsollier L, et al. (2009) Mycolactone
gene expression is controlled by strong SigA-like promoters with
utility in studies of Mycobacterium ulcerans and buruli ulcer. PLoS Negl Trop Dis
3: e553.
Trott KA, Stacy BA, Lifland BD, Diggs HE, Harland RM, et al. (2004)
Characterization of a Mycobacterium ulcerans-like infection in a colony of African
tropical clawed frogs (Xenopus tropicalis). Comp Med 54: 309–317.
van der Werf TS, Stienstra Y, Johnson C, Phillips R, Adjei O, et al. (2005)
Mycobacterium ulcerans disease. Bulletin of the World Health Organization 83:
785–791.
Veitch MGK, Johnson PDR, Flood PE, Leslie D, Street AC, et al. (1997) A large
localized outbreak of Mycobacterium ulcerans infection on a temperate southern
Australian island. Epidemiol Infect 119: 313–318.
Wagner T, Benbow ME, Burns M, Johnson RC, Merritt R, et al. (2008) A
Landscape-based Model for Predicting Mycobacterium ulcerans Infection (Buruli



17


Ulcer Disease) Presence in Benin, West Africa. EcoHealth 5: 69–79.
Wagner T, Benbow ME, Brenden T, Qi J, Johnson RC (2008) Buruli ulcer
disease prevalence in Benin, West Africa: associations with land use/cover and
the identification of disease clusters. International Journal of Health Geographics
7: 25.
Wallace J, Gordon M, Hartsell L, Mosi L, Benbow M, et al. (2010) Interaction
of Mycobacterium ulcerans with mosquito species: Implications for transmission
and trophic relationships. Applied Environ Microbiol 76: 6215–6222.
Walsh D, Portaels F, Meyers W (2008) Buruli ulcer (Mycobacterium ulcerans
infection). Transactions of the Royal Society of Tropical Medicine and Hygiene
102: 969–978.
Walsh D, Eyase F, Onyango D, Odindo A, Waitumbi JN et al. (2010) Short
Report: Clinical and Molecular Evidence for a Case of Buruli Ulcer
(Mycobacterium ulcerans Infection) in Kenya. Am. J. Trop. Med. Hyg., 81(6), pp.
1110–1113
Wansbrough-Jones M, Phillips R (2006) Buruli ulcer: emerging from obscurity.
Lancet 367: 1849–1858.
Williamson HR, Benbow ME, Nguyen KD, Beachboard DC, Kimbirauskas RK, et
al. (2008) Distribution of Mycobacterium ulcerans in Buruli Ulcer Endemic and
Non-Endemic Aquatic Sites in Ghana. PLoS Neglected Tropical Diseases 2:
e205.
WHO, ed (2000) Buruli ulcer. Mycobacterium ulcerans infection. Geneva,
Switzerland: WHO. 118 p.
WHO (2008) Buruli ulcer: progress report, 2004–2008. In: WHO, ed. Weekly
epidemiological record. Geneva, Switzerland: World Health Organization 83:
145–156.
Yeboah-Manu D, Bodmer T, Mensah-Quainoo E, Owusu S, Ofori-Adjei D, et al.
(2004) Evaluation of decontamination methods and growth media for primary
isolation of Mycobacterium ulcerans from surgical specimens. J Clin Microbiol 42:
5875–5876.
Yip MJ, Porter JL, Fyfe JA, Lavender CJ, Portaels F, et al. (2007) Evolution of
Mycobacterium ulcerans and other mycolactone-producing mycobacteria from a
common Mycobacterium marinum progenitor. J Bacteriol 189: 2021–9.




18


CHAPTER 2
ASSOCIATIONS BETWEEN MYCOBACTERIUM ULCERANS AND
MACROINVERTEBRATE ASSEMBLAGES IN AQUATIC ENVIRONMENTS OF
GHANA, WEST AFRICA
INTRODUCTION
Buruli ulcer (BU) is an emerging skin disease caused by an infection of
Mycobacterium ulcerans (Walsh et al. 2008, Duker et al. 2006, WansbroughJones and Philips 2006, van der Werf et al. 2005). This disease is generally
considered non-fatal; however, infections often result in cell necroses, which may
lead to severe ulcerations, disfigurement, and disability in humans (Asiedu and
Etuaful 1998, Amofah et al. 2002, Johnson et al. 2005, van der Werf et al. 1999,
2005, Wansbrough-Jones and Philips 2006). Johnson et al. (1996) reported
incidence of BU in isolated temperate regions of Australia, but BU is most
prevalent in tropical and subtropical climates, with the highest number of new
cases being reported from sub-Saharan West Africa (WHO 2008)(Fig. 2.1). It is
widely accepted that BU incidence is associated with exposure and proximity to
freshwater habitats (Barker & Carswell 1973, Radford 1975, Hayman and
McQueen 1985, Hayman 1991, WHO 2003, Aiga et al. 2004, Porteals et al.
1999, Merritt et al. 2005, Debacker et al. 2006; Thangaraj et al. 1999), yet many
questions regarding the ecology of M. ulcerans remain unanswered, including the
pathogen’s natural reservoir(s), environmental distribution, and method of
transmission to humans (Merritt et al. 2005, 2010).
Epidemiological studies have linked cases of BU with aquatic
environments that are both lentic (i.e. ponds, lakes) and lotic (i.e. streams,

19

rivers)(Merritt et al. 2010). Field studies have also associated BU endemicity
with disturbance to waterbodies through modification of freshwater habitats and
their adjacent landscapes (Hayman 1991b; Marston et al. 1995; Meyers et al.
1996; Johnson et al. 1999; Portaels et al. 2001, Wagner et al. 2008a, Merritt et
al. 2005, Duker et al. 2006, Kibadi et al. 2008). Merritt et al. (2010) provided a
thorough review of published work on the transmission and ecology of BU
disease and reported that anthropogenic influences; such as, mining activity,
damming of waterbodies, deforestation practices, and agricultural development
were among the most commonly cited factors attributed to environmental
disturbance and BU disease incidence. Natural disturbances, such as flooding
events, also have been proposed to be factors that could lead to potential
increased risk of MU infection (Barker and Carswell 1973, Wagner et al. 2008a;
Radford 1974b, Barker 1972 Meyers et al. 1996, Portaels 1995, Hayman 1991).
Merritt et al. (2005) proposed a model describing how anthropogenic and natural
disturbances may lead to the proliferation of MU in aquatic environments;
however, this model has not yet been field-tested and the source of M. ulcerans
in the environment remains unknown.
Aquatic macroinvertebrates have received the most attention as potential
reservoirs and biological vectors of M. ulcerans. The isolation and successful
culturing of M. ulcerans from a water strider (Gerris sp.) in Benin, West Africa
(Portaels et al. 2008) and laboratory experiments demonstrating water bugs
(Naucoris sp.) can transmit M. ulcerans to a mammal model (Marsollier et al.
2004) have supported the role of aquatic insects as reservoirs and vectors. While

20

these findings demonstrate a direct association of macroinvertebrates with M.
ulcerans and BU disease, aquatic invertebrates may also provide valuable
information into the ecology of this disease through indirect associations with M.
ulcerans in the environment. Aquatic macroinvertebrates are often used as
biological indicators of water quality and analysis of macroinvertebrate
communities can identify short and long term disturbances to aquatic
environments (Merritt and Cummins 1996). If M. ulcerans proliferation is
associated with disturbed aquatic environments, then aquatic insects could
possibly be useful indicators of environmental conditions suitable for the
establishment of M. ulcerans and an increased risk of BU infection.
The role of aquatic macroinvertebrates in the transmission of BU has been
proposed by several authors (Johnson et al. 2005, Portaels et al., 1999,
Marsollier 2004, Merritt et al. 2005, Wansbrough-Jones and Phillips 2006);
however, field based ecological studies to specifically address the association
between macroinvertebrate communities and M. ulcerans are limited (Benbow et
al. 2007). An initial step in understanding the potential role of macroinvertebrates
in the ecology of BU, whether direct or indirect, is identifying the relative
abundances and composition of the aquatic macroinvertebrate communities in
relation to the disease pathogen. As part of a large-scale systematic study, I
surveyed 98 water bodies from BU endemic and non-endemic regions, in Ghana,
West Africa to: 1) characterize and compare overall macroinvertebrate
communities from aquatic environments in Ghana: 2) identify macroinvertebrate
community metrics associated with the presence and absence of M. ulcerans in

21

aquatic environments: and 3) identify potential relationships between specific
macroinvertebrates and pathogen presence.
MATERIALS AND METHODS
Study Location and Scale. A large-scale, standardized assessment of
aquatic habitats was conducted to characterize benthic macroinvertebrate
communities in Ghana, West Africa. In this study water bodies (n=98) were
selected from individual villages located within three distinct regions of southern
Ghana: the Greater Accra (n=29) and Ashanti regions (n=39), which are endemic
for the disease, and the Volta region (n = 30), which is non-endemic. Villages
were randomly selected within each region and water bodies were selected
within each village, based on location (<100-200m from community housing
structures) and human use (daily domestic activities) to reflect aquatic
environments with potential human exposure to Buruli ulcer. Community
discussions on water body selection were conducted in each village as described
by Benbow et al. (2005). Various types of water bodies were selected from all
regions, including streams, rivers, wetlands, ponds, fetches and reservoirs.
Water bodies and thus aquatic macroinvertebrate communities, were surveyed
on a single sampling date in 2005 (6 July to 15 August), 2006 (7 July to 15
August), or 2007 (15 August to 7 September).
Macroinvertebrate sample collections. All samples were collected from
the littoral margins of water bodies. Within each water body, two 10–20-m
transects were measured parallel to the shoreline and positioned through the
dominant macrophyte community. Along each transect, two floating 1-m2

22

polyvinyl chloride (PVC) quadrats were randomly placed and invertebrates were
collected by sweeping within the quadrat with a 500-µm mesh dip net. The
quadrats floated on top of the water and delineated 1 m2 of area to be sampled.
Three sweeps of the dip net were performed from the water surface to the bottom
substrate for comprehensive sampling of specimens in the water column, and all
samples were collected using the same technique. Contents within each net
were washed through a 500-µm sieve, preserved in 99% ethanol, and
transported to the laboratory for identification. All specimens were enumerated
and identified to lowest possible taxon under dissecting microscope using African
regional keys (Durand and Leveque 1981; Invertebrates of South Africa Identification Keys, vols. 2-10, 1999-2007), and keys from elsewhere (Merritt et
al. 2008).
Detection of Mycobacterium ulcerans. To identify potential
relationships between macroinvertebrate communities and M. ulcerans, biofilms
and water samples were collected at each waterbody. Biofilms were collected
from the surfaces of dominant macrophytes and detritus (n=3), and a composite
water sample was collected from open water areas within the water body at the
mid-water column depth. From the composite, five 100-200ml sub-samples were
filtered through a 1.6 micron fiberglass filter followed by a 0.2 micron
nitrocellulose filter (Whatman Inc). Filters were sealed inside aluminum foil
packets for later laboratory analysis. In addition, a 500 µl of sample liquid was
used for DNA extraction. Extracted DNA was also collected from M. ulcerans
agy99, Mycobacterium marinum 1218, or water for use as positive and negative

23

controls. All samples were processed at the University of Tennessee, Knoxville,
Tennessee and follow methods as previously described by Williamson et al.
(2008).
Primers, PCR conditions and sequencing. A tiered PCR detection
method was used for the identification of M. ulcerans in which DNA was
subjected to amplification of the enoyl reductase (ER) domain and variable
tandem repeat (VNTR) sequences. The enoyl reductase domain is partially
responsible for production of the toxin mycolactone, and was used as a
presumptive identification for mycolactone producing mycobacteria (MPMs)
including M. ulcerans (George et al. 1999; Sizaire et al. 2002). ER-PCR positive
samples were then subjected to VNTR analysis to identify M. ulcerans (MU) from
other MPMs, and to match sample profiles to known VNTR profiles obtained from
patients (Ablordey et al. 2005; George et al. 1999; Johnson et al. 2007; Portaels
et al. 2001; Sizaire et al. 2002). Primers and PCR conditions for amplification of
the enoyl reductase domain as well as VNTR loci, including BNTR MIRU 1, locus
6, ST 1 and locus 19, were used as described by Williamson et al. (2008).
Site Classification. Sites were classified into pre-defined groups based
on the overall presence or absence of M. ulcerans. Due to the complexity of M.
ulcerans detection in environmental samples, the presumptive presence of M.
ulcerans was detected using PCR for the ER domain and thus represents the
presence of M. ulcerans and any additional mycolactone producing
microorganisms (MPMs). Although this provides an overestimate of M. ulcerans,
it provides a maximum estimate for toxin-producing mycobacteria. Thus, sites

24

were classified into the pre-defined groups ER+ or ER- to identify potential
relationships between macroinvertebrate communities and M. ulcerans.
Data Analyses. Multivariate tests were performed to characterize overall
macroinvertebrate community structure among sites. For these procedures, a
total of 98 sites were analyzed and all macroinvertebrate data (relative
abundance as a proportion) were transformed using the arcsin square-root
calculation (Mielke 1991). Rare taxa were determined as those taxa occurring in
fewer than 5% of all sites, and were eliminated from analyses to improve the
detection of potential relationships (McCune and Grace 2002). Nonmetric multidimensional scaling (NMDS) and multi-response permutation procedure (MRPP)
were used to analyze differences in the overall macroinvertebrate community
structure among pre-defined groups (ER+/-). Since multiple MRPP tests were
completed, it was necessary to calculate a Bonferroni adjusted α (and
corresponding p value) of 0.008 to assist in interpreting statistically significant
differences. Indicator species analyses (ISA) were performed to identify specific
macroinvertebrate taxa that best characterize or represent our predefined
groups. Monte Carlo randomization tests were used to assess indicator
significance (McCune and Grace 2002). Multivariate analyses were repeated for
subsets of data based on water body flow, with 50 lotic sites (e.g., rivers,
streams) and 48 lentic sites (e.g., ponds, wetlands, reservoirs). Bonferroni
adjusted α (and corresponding p value) of 0.025 were used to analyze subset
data. All multivariate statistical analyses were conducted in PC-ORD (Version 5).
Paired sample t-tests were used to compare macroinvertebrate

25

community metrics between the pre-defined groups (ER+/-). For all sites
community diversity and similarity indices (Shannon-Weiner Diversity, Simpson’s
Heterogeneity, Margalef’s Richness, Pielou’s Eveness, % Dominant taxa, %
Diptera taxa, and Total taxa), percent functional-feeding group abundances
(Filter-Collector, Gather-Collector, Engulfing-Predator, Piercing-Predator,
Scraper-Grazer, Shredder, and ratio of Scraper to Filter-Collector) (Merritt and
Cummins 2006), and certain relative taxa abundances (Belostomatidae,
Naucoridae, Nepidae, and Culicidae) were calculated and compared between
ER+ and ER- water bodies. For lotic sites, EPT taxa richness (EphemeropteraPlecoptera-Trichoptera taxa) and the ratio EPT taxa to total organisms were
calculated (Rosenberg and Resh 1993). For lentic sites, EOT taxa richness
(Ephemeroptera-Odonata-Trichoptera taxa), ESTD taxa richness
(Ephemeroptera-Sphaeriidae-Trichoptera-Odonata taxa), and percent Corixidae
abundances were calculated and compared between ER+ and ER- water bodies
(Radar 2001, USEPA 2001, Helgen and Gernes 2002). To meet the assumptions
of normality and equal variances data were log + 1 transformed. For percentage
composition differences, data were arc-sine square root transformed. The
nonparametric Wilcoxon/Kruskal-Wallis rank sum test was used when
appropriate. The t-tests were analyzed using SAS software (8.2 2001).
RESULTS
Aquatic macroinvertebrates were analyzed from 98 sites in southern
Ghana to assess overall community differences based on the presence of
environmental M. ulcerans. A total of 73,892 invertebrates from 77 unique taxa

26

were identified in this study (Table 2.1). A three-axis NMDS solution explained
70% of the total variation in the macroinvertebrate community (stress: 16.9,
p=0.004), with 1% on axis 1, 42% on axis 2, and 27% on axis 3. No significant
differences in the overall macroinvertebrate community structure between sites
that were ER+ and ER- (MRPP: A=0.001, p = 0.23) were observed. The NMDS
ordination did identify differences in macroinvertebrate community structure
based on water body flow (MRPP: A=0.046, p < 0.000; Fig. 2.2), indicating that
sites with flowing water were characterized by a different macroinvertebrate
community compared to sites with standing water. To eliminate variation in the
macroinvertebrate community due to flow, we classified all sites into two
separate flow groups (lotic and lentic) and repeated all multivariate and univariate
analyses; this allowed for comparisons of macroinvertebrate and bacterial
communities controlling for variation due to flow regime.
Benthic macroinvertebrates were surveyed from 48 lentic water bodies in
southern Ghana to assess differences in macroinvertebrate community structure
between sites based on the presence or absence of environmental M. ulcerans.
Among the 48 lentic sites, 34 were identified as ER+ and 14 ER-. A total of
42,498 invertebrates from 69 unique taxa were identified and used for final
analysis of lentic water bodies. A three-axis NMDS solution explained 71% of
the total variation in the macroinvertebrate community (stress: 17.7, p=0.004),
with 21% on axis 1, 30% on axis 2, and 20% on axis 3. No significant differences
were detected in the overall macroinvertebrate community between sites that
were ER+ and ER- (MRPP: A=0.000, p = 0.358), and there were also no

27

significant differences in macroinvertebrate community metrics between lentic
sites that were ER+ and ER- (a=0.05).
Aquatic macroinvertebrates were surveyed from 50 lotic water bodies in
southern Ghana to assess differences in macroinvertebrate community structure
between sites based on the presence or absence of environmental M. ulcerans.
Among the 50 lotic sites, 38 were identified as ER+ and 12 ER-. A total of
31,394 invertebrates from 76 unique taxa were identified and used for final
analysis of lotic water bodies. A three-axis NMDS solution explained 80% of the
total variation in the macroinvertebrate community (stress: 15.4, p=0.004), with
38% on axis 1, 14% on axis 2, and 28% on axis 3. We found significant
differences between lotic sites that were ER+ and ER- (MRPP: A=0.01, p = 0.02;
Fig. 2.3), and 7 macroinvertebrate taxa were identified as significant indicators of
ER+ or ER- lotic waterbodies. Indicators of ER+ were Pleidae (ISA: p < 0.010),
Gerridae (ISA: p<0.019) (Hemiptera), Hydroacari (ISA: p < 0.021) (Acarina), and
Libellulidae (ISA: p < 0.042) (Odonata). Indicators of ER- were Elmidae (ISA: p <
0.026) (Coleoptera). Simuliidae (ISA: p < 0.035) (Diptera), and Calopterygidae
(ISA: p < 0.044) (Odonata).
There were also significant differences in commonly used
macroinvertebrate community metrics between lotic sites that were ER+ and ER(A=0.05). Total taxa counts were higher in ER+ water bodies (29.14, std 6.5)
than in ER- water bodies (24.0, std 6.1)(p0.02), as was mean taxa richness
(Margalef’s) (4.49, std 0.81)(3.91, std 0.71)(p= 0.0323). Percent dominance of
the top three taxa was lower in ER+ sites (60.2, std 10.1) than in ER- sites (68.8,

28

std 10.1)(p0.0137). The functional-feeding group consisting of piercing-predators
had a higher mean percent in ER+ sites (0.066, std 0.061) compared to ER- sites
(0.024, std 0.0237, p0.038), and the ratio of scrapers to collector-filterers was
lower in ER+ sites (6.13, std 16.22) compared to ER- sites (6.32, std 8.10)(p=
0.041).
DISCUSSION
The role of aquatic macroinvertebrates in the transmission of BU has been
proposed by several authors (Johnson et al. 2005, Portaels et al., 1999,
Marsollier 2004, Merritt et al. 2005, Wansbrough-Jones and Phillips 2006);
however, field based ecological studies to specifically address the association
between macroinvertebrate communities and M. ulcerans are limited (Benbow et
al. 2007). An initial step in understanding the potential role of macroinvertebrates
in the ecology of BU is identifying the relative abundances and composition of the
aquatic macroinvertebrate communities in relation to the disease pathogen. As
part of a large-scale systematic study, 98 water bodies were surveyed from BU
endemic and non-endemic regions, in Ghana, West Africa to: 1) characterize and
compare overall macroinvertebrate communities from aquatic environments in
Ghana; 2) identify macroinvertebrate community metrics associated with the
presence and absence of M. ulcerans in aquatic environments; and 3) identify
potential relationships between specific macroinvertebrates and pathogen
presence. When water bodies were separated by flow regime, I found
differences in macroinvertebrate community structure and function in relation to

29

the presence of M. ulcerans, and also identified specific taxa that may potentially
be used as biological indicators of M. ulcerans in aquatic environments.
Several studies have been conducted on aquatic invertebrates in West
Africa; however, most of these have focused on specific taxa of medical
importance (i.e. Anopheles sp., Simulium sp., Bulinus sp.) and rarely have
provided data on entire macroinvertebrate communities (Resh et al. 2004, Hynes
1975a,b, Thorne et al. 1997, Thorne et al. 2000). Hynes et al. (1975a,b)
conducted studies in rivers of Ghana and sampled the riffle habitat community to
examine annual life-cycles and drift behavior of benthic macroinvertebrates.
Thorne et al. (1997, 2000) anchored artificial substrates to streambeds in
southern Ghana to sample benthic communities in an examination of the
responses of macroinvertebrates to gradients of pollution. They found similar
community responses to pollution observed in studies from temperate areas, and
concluded that established macroinvertebrate community metrics can be used to
characterize sites of differing water qualities in the tropics. In 1974, an
independent ecological oversight committee initiated a long-term monitoring
program in West Africa to evaluate the effects of insecticides used to control
black flies and Onchocerciasis. In these studies, riverine benthic communities
that occupy the same habitats as black flies were sampled to evaluate changes
in aquatic fauna studies in relation to insecticide treatments. Resh et al. (2004)
summarized results from this 29 year study and concluded that permanent
damage to non-target invertebrate communities due to insecticides was unlikely,
but also reported that clear associations with macroinvertebrate communities and

30

treatment effect were difficult to analyze due to seasonal variations and lack of
pre-treatment community data. Direct comparisons of my work in Ghana to
these studies, and others in West Africa, are difficult due to differences in specific
research objectives and sampling strategies. Therefore, information used in the
interpretation of my results was largely drawn from research on
macroinvertebrate communities conducted elsewhere.
Rapid bioassessment techniques that use macroinvertebrates to assess
water quality incorporate the use of metrics to assess environmental degradation
by measuring changes in the macroinvertebrate community and comparing them
to a predicted response in relation to increased disturbance (Metcalfe-Smith
1994, Resh and Jackson 1993, USEPA 1996, Barbour et al. 1995, 1999). My
analysis of lotic water bodies revealed that macroinvertebrate total taxa and taxa
richness (Margalef’s) were significantly greater when M. ulcerans was detected. I
also found that percent taxa dominance was higher in waterbodies with M.
ulcerans. The predicated responses of these measurements are that total taxa
and taxa richness decrease with increased environmental perturbation, and
percent dominance increases with increased environmental perturbation (Plafkin
et al 1989). Proliferation of M. ulcerans in the aquatic environment is believed to
be associated with natural and anthropogenic disturbances (Hayman 1991b;
Marston et al. 1995; Meyers et al. 1996; Johnson et al. 1999; Portaels et al.
2001, Wagner et al. 2008a, Merritt et al. 2005, Duker et al. 2006, Kibadi et al.
2008). My data indicated that the macroinvertebrate communities of waterbodies
without M. ulcerans were more characteristic of disturbed habitats. While these

31

community metrics allow for general statements to be made regarding water
quality and overall community health within and between waterbodies, direct
associations between my results and the ecology of M. ulcerans should be made
with caution.
Functional-feeding groups were investigated to better understand potential
ecological associations between macroinvertebrate communities and M.
ulcerans. Where other metrics rely strictly on taxonomic groupings, functionalfeeding group classifications are based on morpho-behavioral mechanisms of
food acquisition and provide insight into the balance between food resource
availability and the predictable response of aquatic insect assemblages
(Cummins and Klug 1979, Merritt and Cummins 1984, Merritt and Cummins
2006). My analysis revealed that the ratio of scraper-grazers (i.e. snails) to
collector-filterers (i.e. mosquitoes, black flies) was greater in waterbodies when
M. ulcerans was detected. A shift in the dominance of the scraper-grazer
community can be an indication of increased periphyton (i.e. attached algae,
diatoms)(Merritt and Cummins 2006), and periphyton assemblages have been
linked to the presence and absence of M. ulcerans in waterbodies of Ghana
(Miller et al. unpublished data). Periphyton enrichment in aquatic habitats has
also been associated with eutrophication (Davis 1994, McCormick and
Stevenson 1998, Gaiser et al. 2005), which has been proposed by several
authors to play a significant role in the establishment of M. ulcerans in the
environment (Hayman 1991b; Marston et al. 1995; Meyers et al. 1996; Palomino
et al. 1998, Johnson et al. 1999; Portaels et al. 2001, Merritt et al. 2005, Duker et

32

al. 2006, Kibadi et al. 2008). The observed increase of the scraper-grazer group
in relation to M. ulcerans in my study supports an association of the pathogen
with nutrient enrichment and eutrophication of aquatic habitats, and suggests that
macroinvertebrate feeding-group analyses may be a viable way to identify
environmental conditions favorable for the establishment of M. ulcerans.
The feeding group comprised of piercing-predators, which includes
families of biting hemipterans implicated as vectors of BU (i.e. Belostomatidae,
Naucoridae), also was greater in waterbodies when M. ulcerans was detected.
Upon further examination of this community, it was an abundance of the pygmy
backswimmer (Hemiptera: Pleidae) that was responsible for the significant
difference among these waterbodies. Pleidae are predators of micro-crustaceans
(i.e. Cladocera, Copepoda, Ostracoda) and to date have not been formerly
associated with the ecology of M. ulcerans or BU disease. Williamson et al.
(2008) processed more than 100 pleid specimens collected from field samples
and found no direct associations with M. ulcerans and this group, but they did
identify M. ulcerans associated with micro-crustaceans. Considering the feeding
behavior of Pleidae and the growing body of work indicating that M. ulcerans is
transferred trophically (Eddyani et al. 2004, Marsollier et al. 2004a,b, Duker et al.
2006, Mosi et al. 2008, Wallace et al. 2010), a closer look into the potential role
of this group as an environmental reservoir of M. ulcerans may be warranted. It
should be noted that although the piercing-predator community was greater in
waterbodies with M. ulcerans, an investigation of Belostomatidae (giant water
bugs) and Naucoridae (creeping water bugs) yielded no identifiable differences

33

between waterbodies. In addition, the overall relative abundances of these two
families were low among all waterbodies sampled in this survey. These data
were consistent with reports by Benbow et al. (2008), whom suggested a
possible role as reservoirs of M. ulcerans for biting Hemiptera, but that caution
should be taken when describing the role of biting hemipterans in BU
transmission.
An indicator species analysis was used to detect relationships between
specific macroinvertebrates and M. ulcerans. As a result, I found 4 taxa that
were identified as indicators of the pathogen in the environment. These taxa, all
of which are predators, included: Hemiptera (Pleidae, Gerridae), Odonata
(Libellulidae), and Hydroacrines (water mites). Overall, these taxa are considered
to be tolerant of moderate levels of pollution and physical disturbance and, as a
result, not strong indicators of a particular environmental condition (Bode et al.
1996; Hauer and Lamberti 1996; Hilsenhoff 1988; Plafkin et al. 1989). In lotic
systems, an increase in the abundance of predators can indicate a healthy
biological community (Karr et al. 1986; Morley 2000); however, these particular
macroinvertebrates are more characteristic of lentic habitats and this
generalization does not relate due to my sampling strategy which focused on the
marginal zones. The association of these taxa with M. ulcerans does suggest an
ecological connection between pathogen and waterbodies with riparian margins
subject to prolonged periods of inundation. This is consistent with field research
and epidemiological data that have associated BU disease with areas prone to
flooding (Lunn et al. 1965, Revill and Barker 1972, Barker and Carswell 1973,

34

Duker et al. 2006, Marston et al. 1995, Walsh et al. 2008, Portaels 1995,
Debacker et al. 2006, Noeske et al. 2004, Johnson et al. 2007, Wagner et al.
2008a, WHO 2000). Additional field collections, particularly to address seasonal
variation, will provide more insight as to whether these taxa have a more specific
connection with the ecology of M. ulcerans.
In conclusion, my investigation of macroinvertebrate community
associations with M. ulcerans did produce results that suggest potential use for
these communities to be useful as indicators of environmental conditions
preferable to the proliferation of the pathogen in the environment; however these
data should be treated with caution. First, while there were a few metrics and
taxa associated with the presence of M. ulcerans, there were several more that
were not. This could be the result of sampling strategy, taxonomic resolution
used for macroinvertebrate identifications, inaccurate estimate of the presence or
absence of M. ulcerans within sites, or simply that there aren’t true associations
between this bacteria and the macroinvertebrate community that can be
measured with the standard biomontoring techniques. Second, the sampling
strategy used in this study was aimed to standardize collections among all sites
by targeting the marginal habitats where it is believed M. ulcerans is most likely
to flourish and where people most likely contact the pathogen in the environment.
While sampling this habitat allowed for comparison to be made between sites,
the true macroinvertebrate community profile within individual waterbodies may
have been missed by not sampling additional habitats. For example, analyses of
lotic waterbodies are most often based on collections made from riffles, pools

35

and stream runs, whereas our collections were based on marginal habitats
typically out of the current. One suggestion for future studies would be to sample
lotic habitats in a way that incorporates collections of benthic communities from
habitats apart from the vegetative and marginal zone. This might produce a
different community assemblage more characteristic of stream and river systems
and would allow for more accurate comparisons of data with common
biomontoring practices and previous studies conducted in West Africa. Third, the
use of ER positivity as an indication of the presence and absence of M. ulcerans
may overestimate the true distribution of the pathogen in the environment. There
also was the potential for the pathogen to have been present at a waterbody but
not collected or possibly not enough DNA to have been collected to generate an
adequate positive confirmation in the laboratory. If either of these conditions
were true, then comparisons of macroinvertebrate communities between and
among waterbodies with and without the pathogen would likely produce different
results than what were observed. The methods used to confirm positive
detection of M. ulcerans were up to date at the time of this study, however
increasing the number of samples collected per site to determine pathogen
presence or absence might have led to more positive waterbodies than what
were observed during this study. This study did identify potential ecological
relationships between macroinvertebrates and M. ulcerans in the environment,
but further field-based studies are needed to more completely understand the
specific role M. ulcerans may play on benthic macroinvertebrate communities.

36

Table 2.1.

Total specimens and number of sites observed in lotic and lentic habitats, Ghana, W. Africa
LOTIC
ER+ (N= 38)
ER- (N= 12)

Taxon (Higher, Lowest)

LENTIC
ER+ (N= 34)
ER- (N= 14)

#Sites Total #Sites Total #Sites Total #Sites Total
Obsrvd. Spec. Obsrvd. Spec. Obsrvd. Spec. Obsrvd. Spec.

Annelida
Clitellata, Hirudinea
Clitellata, Oligochaeta

14
36

68
572

3
10

9
189

18
32

129
1764

7
12

43
333

Arthropoda
Arachnida, Araneae
Arachnida, Hydracarina

33
26

280
775

6
4

20
6

31
27

344
903

12
13

89
176

Crustacea, Branchiopoda Cladocera
Lynceidae
Crustacea, Decapoda
Atyidae
Potamonautidae
Crustacea, Maxillopoda
Copepoda
Crustacea, Ostracoda
Ostracoda

9
2
23
4
10
15

544
29
550
7
470
1669

1
5
4
1
7

22
99
6
4
41

13
5
5
20
26

694
282
365
2123
4082

6
2
3
1
5
11

32
24
7
3
271
497

Insecta, Coleoptera

4
26
22
17
27
3
24

18
273
880
36
435
4
227

6
11
5
8
8

56
285
17
48
13

6
24
3
6
21
2
30

24
263
3
49
578
33
673

1
14
2
8
13

2
269
5
294
220

Curculionidae
Dytiscidae
Elmidae
Gyrinidae
Hydraenidae
Hydrobiidae
Hydrophilidae

37

Table 2.1 (cont'd). Total specimens and number of sites observed in lotic and lentic habitats, Ghana, W. Africa
Insecta, Coleoptera

Lampyridae
Noteridae
Scirtidae

10
8
17

21
290
125

1
1
3

1
1
4

8
27
13

30
498
144

4
10
6

15
40
73

Insecta, Collembola

Entomobryiidae
Isotomidae

17
4

98
42

5
-

12
-

13
3

90
16

5
-

20
-

Insecta, Diptera

Athericidae
Ceratopogonidae
Chaoboridae
Chironomidae
Culicidae
Dixidae
Empididae
Ephydridae
Psychodidae
Sciomyzidae
Simuliidae
Stratiomyiidae
Syrphidae
Tipulidae

4
30
5
37
24
4
4
8
10
11
3
3
13

8
364
9
4911
941
15
11
15
42
122
7
4
25

1
11
12
4
3
6
7
2
-

3
109
1714
10
5
29
134
6
-

30
10
34
26
1
5
2
4
2
10
3
9

392
42
9121
1256
1
7
8
13
2
25
3
31

10
5
14
12
1
1
1
1
1
5
4

62
9
1597
392
13
1
1
1
1
31
21

Insecta, Ephemeroptera

Baetidae
Caenidae
Heptageniidae
Leptophlebiidae
Polymitarcyidae

38
34
20
17
1

3714
2529
356
254
1

12
11
6
6
-

615
454
93
27
-

34
24
2
1
7

3666
565
38
2
16

14
10
4

1623
117
19

38

Table 2.1 (cont'd). Total specimens and number of sites observed in lotic and lentic habitats, Ghana, W. Africa
Insecta, Ephemeroptera

Tricorythidae

5

36

2

16

-

-

-

-

Insecta, Hemiptera

Belostomatidae
Corixidae
Gerridae
Hebridae
Hydrometridae
Mesoveliidae
Naucoridae
Nepidae
Notonectidae
Pleidae
Saldidae
Veliidae

20
7
25
6
5
28
14
6
19
21
1
27

145
28
114
9
6
112
30
10
147
563
1
138

2
1
2
2
7
2
1
3
1
9

3
3
4
8
20
3
1
17
2
41

17
7
19
1
5
26
12
10
28
22
2
16

166
111
112
1
7
210
118
14
651
675
2
93

10
5
10
4
7
12
5
7
12
7
2
8

58
34
48
4
11
121
19
16
218
98
3
75

Insecta, Lepidoptera

Pyralidae

10

21

2

4

7

25

2

2

Insecta, Odonata

Calopterygidae
Chlorocyphidae
Coenagrionidae
Corduliidae
Gomphidae
Libellulidae
Protoneuridae

2
2
25
13
11
24
31

2
9
401
77
31
394
529

3
1
4
7
4
4
10

12
1
12
74
10
6
99

24
9
29
28

242
141
449
942

8
4
9
10

79
61
189
279

Insecta, Plecoptera

Perlidae

3

9

2

3

-

-

-

-

39

Table 2.1 (cont'd). Total specimens and number of sites observed in lotic and lentic habitats, Ghana, W. Africa
Insecta, Trichoptera

Ecnomidae
Hydropsychidae
Hydroptilidae

2
6
3

2
26
12

1
3
2

1
5
3

2
2
1

2
31
4

1
-

14
-

Insecta, Trichoptera

Leptoceridae
Polycentropodidae

26
5

160
10

4
1

16
1

4
-

11
-

2
-

8
-

Sphaeriidae
Ancylidae
Bithyniidae
Lymnaeidae
Physidae
Pilidae
Planorbidae
Thiaridae

5
11
3
3
9
8
34
20

60
87
9
20
20
22
1603
393

4
1
1
5
9
6

18
2
4
8
152
619

1
8
1
8
5
6
32
9

10
100
1
36
44
11
1577
136

4
1
1
8
2

82
24
4
384
33

Mollusca
Bivalvia, Veneroida
Gastrpoda

40

Figure 2.1. Map illustrating the locations of 98 water bodies sampled for
macroinvertebrates from five geographic regions in southern Ghana, Africa.
Villages were randomly selected within each region and water bodies were
selected within each village, based on location (<100-200m from community
housing structures) and human use (daily domestic activities) to reflect aquatic
environments with potential human exposure to Buruli ulcer. Community
discussions on water body selection were conducted in each village as described
by Benbow et al. (2005).

41












Figure 2.2. A non-metric multi-dimensional scaling (NMDS) ordination of
macroinvertebrate communities collected from 98 waterbodies in Ghana, Africa.
Each circle or triangle represents the overall macroinvertebrate community at
each site and symbols closer together have more similar community structure
while symbols further apart were more dissimilar communities. Closed circles
represent lotic habitats and open triangles represent lentic habitats. A three-axis
NMDS solution explained 70% of the total variation in the macroinvertebrate
community (stress: 16.9, p=0.004), with 1% on axis 1, 42% on axis 2, and 27%
on axis 3. No significant differences in the overall macroinvertebrate community
structure between sites that were ER+ and ER- (MRPP: A=0.001, p = 0.23) were
observed. The NMDS ordination did identify differences in macroinvertebrate
community structure based on water body flow (MRPP: A=0.046, p < 0.000).








42







































Figure 2.3. A non-metric multi-diminsional scaling (NMDS) ordination of
macroinvertebrate communities collected from 98 waterbodies in Ghana, Africa.
Each circle or triangle represents the overall macroinvertebrate community at
each site and symbols closer together have more similar community structure
while symbols further apart were more dissimilar communities. Closed circles
represent ER+ and open triangles represent ER- habitats. A three-axis NMDS
solution explained 80% of the total variation in the macroinvertebrate community
(stress: 15.4, p=0.004), with 38% on axis 1, 14% on axis 2, and 28% on axis 3.
We found significant differences between lotic sites that were ER+ and ER(MRPP: A=0.01, p = 0.02).







43


LITERATURE CITED




44


Literature Cited

Ablordey, A, Swings, J, Hubans, C, Chemlal, K, Locht, C, Portaels, F, Supply, P
(2005) Multilocus variable-number tandem repeat typing of Mycobacterium
ulcerans. J Clin Microbiol 43, 1546–1551.
Aiga H, Amano T, Cairncross S, Domako JA, Nanas OK, et al. (2004)
Assessing water-related risk factors for Buruli ulcer: A case-control study in
Ghana. Am J Trop Med Hyg 71: 387–392.
Amofah GK, Bonsu F, Tetteh C, Okrah J, Asamoa K, et al. (2002) Buruli ulcer
in Ghana: results of a national case search. Emerging Infectious Diseases 8:
167–170.
Asiedu K, Etuaful S (1998) Socioeconmoic implications of Buruli ulcer in
Ghana: a three-year review. Trans R Soc Trop Med & Hyg 59: 1015–1022.
18. Portaels F, Chemlal K, Elsen P, Johnson PDR, Hayman JA, et al. (2001)
Mycobacterium ulcerans in wild animals. Rev sci tech Off int Epiz 20: 252–264.
Barbour, MT, JB Stribling, JR Karr (1995) Multimetric approach for establishing
biocriteria and measuring biological condition. Pages 63-77 in WS Davis and TP
Simon (editors). Biological assessment and criteria. Tools for water resource
planning and decision making. Lewis Publishers, Boca Raton, Florida.
Barbour, MT, JM Diamond, CO Yoder (1996) Biological assessment strategies:
Applications and Limitations. Pages 245-270 in DR Grothe, KL Dickson, and DK
Reed-Judkins (editors). Whole effluent toxicity testing: An evaluation of methods
and prediction of receiving system impacts, SETAC Press, Pensacola, Florida.
Barker, DJP (1971) Buruli disease in a district of Uganda. J. Trop. Med. Hyg. 74:
260-264.
Barker DJP (1972) The distribution of Buruli disease in Uganda. Trans R Soc
Trop Med Hyg. 66:867-74.
Barker DJP, Carswell JW (1973) Mycobacterium ulcerans infection among
Tsetse control workers in Uganda. International Journal of Epidemiology 2: 161–
165.
Benbow M, Williamson H, Kimbirauskus R, McIntosh M, Kolar R, et al. (2008)
Aquatic invertebrates as unlikely vectors of Buruli ulcer disease. Emerg Infect Dis
14: 1247–1254.
Bode, RW, MA Novak, LA Abele (1996) Quality assurance work plan for
biological stream monitoring in New York State. NYS Department of



45


Environmental Protection; Division of Water; Bureau of Monitoring and
Assessment; Stream Biomonitoring Unit; Albany, NY.
Boisvert H, Schroder KH (1977) Skin ulcer caused by Mycobacterium ulcerans in
Cameroon II: Bacteriologic study. Bull Soc Pathol Exot Filiales 70: 125–131.
Chemlal, K, Huys, G, Laval, F, et al. (2002) Characterization of an unusual
Mycobacterium: a possible missing link between Mycobacterium marinum and
Mycobacterium ulcerans. Journal of Clinical Microbiology 40:2370.
Chessman, BC (1995) Rapid assessment of rivers using macroinvertebrates: A
procedure based on habitat-specific sampling, family level identification and a
biotic index. Australian Journal of Ecology 20:122-129.
Connor DH, and Lunn F (1965) Mycobacterium ulcerans infection. Int. J. Leprosy
33:698.
Cosma CL, Sherman DR, and Ramakrishnan L (2003) The Secret Lives of the
Pathogenic Mycobacteria. Annu. Rev. Microbiol. 57:641–76.
Cummins, KW, MJ Klug (1979) Feeding ecology of stream invertebrates. Annual
Review of Ecology and Systematics 10: 147-172.
Cummins, KW, RW Merritt (1981) Application of invertebrate functional feeding
groups to wetland ecosystem functions and biomonitoring. Pp. 85-11 in RB
Rader, DP Batzer, SA Wissinger (editors). Bioassessment and Management of
North American Freshwater Wetlands. John Wiley and Sons, New York.
Davis, SM (1994) Phosphorus inputs and vegetation sensitivity in the
Everglades. In Davis, S. M. & J. C. Ogden (eds), 355 Everglades: The
Ecosystem and its Restoration. St. Lucie Press, Boca Raton, FL, USA, 357–378.
Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, et al. (2004)
Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin,
1997-2001. Emerg Infect Dis 10: 1391–1398.
Debacker M, Portaels F, Aguiar J, Steunou C, Zinsou C, et al. (2006) Risk factors
for Buruli ulcer, Benin. Emerg Infect Dis 12: 1325–1331.
Demangel C, Stinear TP, Cole ST (2009) Buruli ulcer: reductive evolution
enhances pathogenicity of Mycobacterium ulcerans. Nat Rev Microbiol 7: 50–60.
Duker AA, Carranza JM, Hale M (2004) Spatial dependency of Buruli ulcer
prevalence on arsenic-enriched domains in Amansie West District, Ghana:
implications for arsenic mediation in Mycobacterium ulcerans infection. Int J
Health Geogr. 3:19.



46


Duker AA, Portaels F, Hale M (2006) Pathways of Mycobacterium ulcerans
infection: A review. Environment International 32: 567–573.
Durand, JR, Leveque, C (1981) Flore et Faune Aquatiques de l’Afrique SaheloSoudaniene, 2 volumes. ORSTOM, Paris.
Eddyani M, Ofori-Adjei D, Teugels G, De Weirdt D, Boakye D, et al. (2004)
Potential role for fish in transmission of Mycobacterium ulcerans disease (Buruli
Ulcer): an environmental study. Appl Environ Microbiol 70: 5679–5681.
Fyfe JAM, Lavender CJ, Johnson P, Globan M, Sievers A, et al. (2007)
Development and Application of Two Multiplex Real-Time PCR Assays for
the Detection of Mycobacterium ulcerans in Clinical and Environmental Samples.
Appl Environ Microbiol 73: 4733–4740.
Gaiser, EE, DL Childers, RD Jones, JH Richards, LJ, Scinto, JC Trexler (2005)
Periphyton responses to eutrophication in the Florida Everglades: cross-system
patterns of structural and compositional change. Limnology and Oceanography
50: 342–355.
Garrity GM (2001) Bergey’s Manual of Systematic Bacteriology; In: Garrity GM,
ed. New York: Springer-Verlag.
George KM, Chatterjee D, Gunawardana G, Welty D, Lee T, et al. (1999)
Mycolactone: a polyketide toxin from Mycobacterium ulcerans required for
virulence. Science 283: 854–857.
Gunawardana G, Chatterjee D, George KM, Brennan P, Whittern D, et al. (1999)
Mycolactone A and B: toxins of Mycobacterium ulcerans. J Amer Chem Soc 121:
6092–6093.
Guerra H, Palomino JC, Falconi E, Bravo F, Donaires N, et al. (2008)
Mycobacterium ulcerans Disease, Peru. Emerging Infectious Diseases 14: 373–
377.
Hauer, FR, GA Lamberti (editors) (1996) Methods in Stream Ecology. Academic
Press. New York. p 674.
Hayman, J, A. McQueen (1985) The pathology of Mycobacterium ulcerans
infection. Pathology 17:594-600.
Hayman J (1991a) Postulated epidemiology of Mycobacterium ulcerans infection.
International Journal of Epidemiology. 20(4): 1093-8.
Hayman J (1991b) Mycobacterium ulcerans infection. The Lancet 337: 124.



47


Helgen, JC, MC Gernes (2001) Monitoring the condition of wetlands: indexes of
biological integrity using invertebrates and vegetation. Pp. 167-185 IN: RB
Rader, DP Batzer, and SA Wissinger (eds.), Bioassessment and Management of
North American Freshwater Wetlands, John Wiley & Sons, New York, NY.
Hilsenhoff, WL (1988) Rapid field assessment of organic pollution with a family
level biotic index. Journal of the North American Benthological Society 7(1):6568.
Hynes, JD (1975a) Annual cycles of macro-invertebrates of a river in southern
Ghana. Freshwater Biology (5): 71-83.
Hynes, JD (1975) Downstream Drift of Invertebrates in a River in Southern
Ghana. Freshwater Biology 5(6): 515-531.
Johnson PDR, Veitch MGK, Leslie D, Flood PE, Hayman JA (1996) The
emergence of Mycobacterium ulcerans infection near Melbourne. Medical
Journal of Australia 164: 76–78.
Johnson PDR, Stinear TP, Hayman JA (1999) Mycobacterium ulcerans — a
minireview. J Med Microbiol 48: 511–513.
Johnson PDR, Stinear TP, Small PLC, Pluschke G, Merritt RW, et al. (2005)
Buruli ulcer (M. ulcerans Infection): new insights, new hope for disease
control.PLoS Med 2(4): e108.
Johnson PDR, Hayman JA, Quek TY, Fyfe JAM, Jenkin GA, et al. (2007)
Consensus recommendations for the diagnosis, treatment and control of
Mycobacterium ulcerans infection (Bairnsdale or Buruli ulcer) in Victoria,
Australia. Medical Journal of Australia 186: 64–68.
Johnson PDR, Lavender CJ (2009) Correlation between Buruli ulcer and vectorborne notifiable diseases, Victoria, Australia. Emerging Infectious Diseases 15:
614–615.
Karr, JR, KD Fausch, PL Angermeier, PR Yant, lJ Schlosser (1986) Assessing
biological integrity in running waters: A method and its rationale. Special
publication 5. Illinois Natural History Survey.
Kaser, M., Hauser, J., Small, P., Pluschke, G. (2009). Large Sequence
Polymorphisms Unveil the Phylogenetic Relationship of Environmental and
Pathogenic Mycobacteria Related to Mycobacterium ulcerans. Appl. Environ.
Microbiol. 75: 5667-5675.
Kibadi K, Panda M, Tamfum JM, Fraga AG, Filho AL, et al. (2008) New foci of



48


Buruli ulcer, Angola and Democratic Republic of Congo. Emerging Infectious
Diseases 14: 1790–1792.
Kotlowski R, Martin A, Ablordey A, Chemlal K, Fonteyne P, et al. (2004) Onetube
cell lysis and DNA extraction procedure for PCR-based detection of
Mycobacterium ulcerans in aquatic insects, molluscs and fish. Journal of Medical
Microbiology 53: 927–933.
Lavender CJ, Stinear TP, Johnson PDR, Azuolas J, Benbow ME, et al. (2008)
Evaluation of VNTR typing for the identification of Mycobacterium ulcerans in
environmental samples from Victoria, Australia. FEMS Microbiology Letters
287: 250–255.
Lunn HF, Connor DH, Wilks NE, Barnley GR, Kamunvi F, et al. (1965) Buruli
(Mycobacterial) ulceration in Uganda. East African Medical Journal 42: 275–288.
Marion E, Eyangoh S, Yeramian E, Doannio J, Landier J Aubry J et al. (2010)
Seasonal and regional dynamics of M. ulcerans transmission in environmental
context: deciphering the role of water bugs as hosts and vectors. PLoS Negl Trop
Dis. 6:4(7):e731.
Marsollier L, Robert R, Aubry J, Andre JS, Kouakou H, et al. (2002) Aquatic
insects as a vector for Mycobacterium ulcerans. Applied and Environmental
Microbiology 68: 4623–4628.
Marsollier L, Aubry J, Saint-Andre JP, Robert R, Legras P, et al. (2003)
Ecology and transmission of Mycobacterium ulcerans. Pathologie Biologie 51:
490–495.
Marsollier L, Severin T, Aubry J, Merritt RW, Saint Andre JP, et al. (2004a)
Aquatic snails, passive hosts of Mycobacterium ulcerans. Appl Environ Microbiol
70: 6296–6298.
Marsollier L, Stinear TP, Aubry J, Saint-Andre J-P, Robert R, et al. (2004b)
Aquatic plants stimulate the growth of and biofilm formation by Mycobacterium
ulcerans in axenic culture and harbor these bacteria in the environment. Applied
and Environmental Microbiology 70: 1097–1103.
Marsollier L, Aubry J, Coutanceau E, Andre JPS, Small PL, et al. (2005)
Colonization of the salivary glands of Naucoris cimicoides by Mycobacterium
ulcerans requires host plasmatocytes and a macrolide toxin, mycolactone.
Cellular Microbiology 7: 935–943.
Marsollier L, Andre J, Frigui W, Reysset G, Milon G, et al. (2006) Early
trafficking events of Mycobacterium ulcerans within Naucoris cimicoides. Cellular
Microbiology.




49


Marsollier L, Brodin P, Jackson M, Kordulakova J, Tafelmeyer P, et al. (2007)
Impact of Mycobacterium ulcerans biofilm on transmissibility to ecological niches
and Buruli ulcer pathogenesis. PLoS Pathogens 3: e62.
Marston BJ, Diallo MO, Horsburgh CR Jr., Diomande I, Saki MZ, et al.
(1995) Emergence of Buruli ulcer disease in the Daloa region of Cote D’ivoire.
Am J Trop Med Hyg 52: 219–224.
McCormick, PV, RJ Stevenson (1998) Periphyton as a tool for ecological
assessment and management in the Florida Everglades. Journal of Phycology
34: 726–733.
McCune, B, JB Grace (2002) Analysis of Ecological Communities, vol. MJM
Software Designs, Glenden Beach, Oregon.
Merritt, RW, KW Cummins, TM Burton (1984) The role of aquatic insects in the
processing and cycling of nutrients. pp. 134-163. In: Resh, VH and DN
Rosenberg (eds.). The ecology of aquatic insects. Praeger Scient., N.Y. 438 p.
Merritt, RW, Cummins KW (1996) An introduction to the aquatic insects of North
America. 3 ed. Kendall / Hunt, Iowa.
Merritt RW, Benbow ME, Small PLC (2005) Unraveling an Emerging Disease
Associated with Disturbed Aquatic Environments: The Case of Buruli Ulcer.
Frontiers in Ecology and the Environment 3: 323–331.
Merritt, RW, KW Cummins (2006) Trophic relationships., pp. 585-610. In: Hauer,
F. R. and G. A. Lamberti (eds.). Methods in Stream Ecology (2nd ed.). Academic
Press, San Diego, CA.
Merritt, RW, Cummins, KW, MB Berg (2008) An introduction to the aquatic
insects of North America. Fourth Edition. Kendall/Hunt Publishing Co., Dubuque,
Iowa. 1158pp.
Merritt RW, Walker ED, Small PLC, Wallace JR, Johnson PDR, et al. (2010)
Ecology and Transmission of Buruli Ulcer Disease: A Systematic Review. PLoS
Negl Trop Dis 4(12): 1-15.
Meyers WM, Tignokpa WM, Priuli GB, Portaels F (1996) Mycobacterium ulcerans
infection (Buruli ulcer): first reported patients in Togo. British Journal of
Dermatology 134: 1116–1121.
Mielke, PW (1991) The application of multivariate permutation methods based on
distance functions in the earth sciences. Earth Science Review 31:55-71.




50


Morley, SA (2000) Effects of urbanization on the biological integrity of Puget
Sound lowland streams: restoration with a biological focus. Seattle, University of
Washington, M.S. Thesis.
Mosi L, Williamson H, Wallace JR, Merritt RW, Small PLC (2008) Persistent
association of Mycobacterium ulcerans with West African predaceous insects of
the family Belostomatidae. Applied and Environmental Microbiology 74: 7036–
7042.
Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, et al. (2004) Buruli ulcer
disease in Cameroon rediscovered. Am J Trop Med Hyg 70: 520–526.
Palomino JC, Obiang AM, Realini L, Meyers WM, and Portaels F (1998) Effect of
Oxygen on Growth of Mycobacterium ulcerans in the BACTEC System. Journal
of Clinical Microbiology. Vol. 36, No. 11, p. 3420-3422,
Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, et al. (2006)
Mycobacterium ulcerans disease (Buruli Ulcer) in a rural hospital in Bas-Congo,
Democratic Republic of Congo, 2002-2004. American J Tropical Medicine
and Hygiene 75: 311–314.
Plafkin, JL, MT Barbour, KD Porter, SK Gross, RM Hughes (1989) Rapid
bioassessment protocols for use in streams and rivers: Benthic
macroinvertebrates and fish. U.S. Environmental Protection Agency, Office of
Water Regulations and Standards, Washington, D.C. EPA 440-4-89-001.
Portaels, F. (1995) Epidemiology of mycobacterial diseases. Clin. Dermatol.
13:207-222.
Portaels F, Elsen P, Guimaraes-Peres A, Fonteyne P, Meyers WM (1999)
Insects in the transmission of Mycobacterium ulcerans infection. The Lancet 353:
986.
Portaels F, Chemlal K, Elsen P, Johnson PDR, Hayman JA, et al. (2001)
Mycobacterium ulcerans in wild animals. Rev sci tech Off int Epiz 20: 252–264.
Portaels F, Meyers WM, Ablordey A, Castro AG, Chemlal K, et al. (2008) First
Cultivation and Characterization of Mycobacterium ulcerans from the
Environment. PLoS Neglected Tropical Diseases 2: e178.
Portaels F, Manuel S, Meyers W (2009) Buruli ulcer. Clinics in dermatology 27:
291–305.
Rader, RB (2001) An introduction to wetland bioassessment and management.
Pages 1-9 in Bioassessment and Management of North American Freshwater
Wetlands. John Wiley and Sons, New York.




51


Radford AJ (1974) Mycobacterium ulcerans: a review, I: Epidemiology. Papua
New Guinea Medical Journal 17: 129–133.
Radford AJ (1975) Mycobacterium ulcerans in Australia. Aust NZ J Med 5: 162–
169.
Resh, VH, JK Jackson (1993) Rapid assessment approaches to biomonitoring
using benthic macroinvertebrates. Pages 195-233 in DM Rosenberg and VH
Resh (editors). Freshwater biomonitoring and benthic macroinvertebrates.
Chapman and Hall, New York.
Resh VH, Leveque C, Statzner B (2004) Long-Term, Large-Scale Biomonitoring
Of The Unknown: Assessing the Effects of Insecticides to Control River
Blindness (Onchocerciasis) in West Africa. Annu. Rev. Entomol. 2004. 49:115–
39
Revill WDL, Barker DJP (1972) Seasonal distribution of mycobacterial skin
ulcers. Brit J prev soc Med 26: 23–27.
Rosenberg, DM, VH Resh ( eds.) (1993) Freshwater biomonitoring and benthic
macroinvertebrates. Chapman and Hall, New York. 488 p.
Ross BC, Marino L, Oppedisano F, Edwards R, Robins-Browne RM, et al. (1997)
Development of a PCR assay for rapid diagnosis of Mycobacterium ulcerans
infection. Journal of Clinical Microbiology 35: 1696–1700.
Silva MT, Portaels F, Pedrosa J (2007) Aquatic Insects and Mycobacterium
ulcerans: an association relevant to Buruli ulcer control? PLoS Medicine 4: e63.
Sizaire, V, Nackers, F, Comte, E, Portaels, F (2006) Mycobacterium ulcerans
infection: control, diagnosis, and treatment. Lancet Infect Dis 6, 288–296.
Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, et al. (2007) Buruli
ulcer surveillance, Benin, 2003-2005. Emerg Infect Dis 13: 1374–1376.
Stinear T, Ross, BC, Davies JK, Marino L, Robins-Browne RM, Oppedisano F,
Grant A, Jenkin GA, Potaels, F, Ross, BC, Oppedisano F, Sievers A, Johnson
PDR (1999) Identification and Characterization of IS2404 and IS2606: Two
Distinct Repeated Sequences for Detection of Mycobacterium ulcerans by PCR.
American Society for Microbiology vol. 37, No. 4.
Stinear T, Jenkin GA, Johnson PDR, Davies JK (2000) Comparative Genetic
Analysis of Mycobacterium ulcerans and Mycobacterium marinum Reveals
Evidence of Recent Divergence. Journal of Clinical Microbiology vol. 182, No. 22.
pp-6322-6330.




52


Stinear T, Davies JK, Jenkin GA, Hayman JA, Oppedisano F, et al. (2000)
Identification of Mycobacterium ulcerans in the environment from regions in
Southeast Australia in which it is endemic with sequence Capture-PCR. Appl
Environ Microbiol 66: 3206–3213.
Stinear T, Davies JK, Grant A, Jenkin GA, Potaels, F, Ross, BC, Oppedisano F,
Purcell M, Hayman JA, Johnson PDR (2000) A Simple PCR Method for Rapid
Genotype Analysis of Mycobacterium ulcerans. Journal of Clinical Microbiology
vol. 38, No. 4. pp-1482-1487.
Stinear TP, Mve Obiang A, Small PL, Frigui W, Pryor MJ, et al. (2004) Giant
plasmid-encoded polyketide synthases produce the macrolide toxin of
Mycobacterium ulcerans. Proc Natl Acad Sci U S A 101: 1345–1349.
Stinear T, Johnson PDR (2007) From marinum to ulcerans: a mycobacterial
human pathogen emerges. Microbe 2: 187–194.
Thangaraj HS, Evans MRW, Wansbrough-Jones MH (1999) Mycobacterium
ulcerans; Buruli ulcer. Transactions of the Royal Society of Tropical Medicine and
Hygiene 93: 337–340.
Thorne, R, Williams WP (1997) The response of benthic macroinvertebrates to
pollution in developing countries: a multimetric system of bioassessment.
Freshwater Biology (1997) 37, 671–686
Thorne, R, Williams, WP, Gordon, C (2000) The macroinvertebrates of a polluted
stream in Ghana. Journal of Freshwater Ecology 15(2): 209-217.
Tobias N, Seemann T, Pidot S, Porter J, Marsollier L, et al. (2009) Mycolactone
gene expression is controlled by strong SigA-like promoters with
utility in studies of Mycobacterium ulcerans and buruli ulcer. PLoS Negl Trop Dis
3: e553.
Trott KA, Stacy BA, Lifland BD, Diggs HE, Harland RM, et al. (2004)
Characterization of a Mycobacterium ulcerans-like infection in a colony of African
tropical clawed frogs (Xenopus tropicalis). Comp Med 54: 309–317.
U.S. Environmental Protection Agency (1996) The volunteer monitor's guide to
quality assurance project plans. U.S. Environmental Protection Agency, Office of
Wetlands, Oceans, and Watersheds, Washington, D.C. EPA 841-B-96-003.
U.S. Environmental Protection Agency (2001) Indicators for Monitoring Bilogical
Integrity of Inland Freshwater Wetlands. U.S. Environmental Protection Agency,
Office of Water, Wetlands Divison, Washington, D.C. EPA 843-R-01.
van der Werf TS, Stienstra Y, Johnson C, Phillips R, Adjei O, et al. (2005)



53


Mycobacterium ulcerans disease. Bulletin of the World Health Organization 83:
785–791.
Veitch MGK, Johnson PDR, Flood PE, Leslie D, Street AC, et al. (1997) A large
localized outbreak of Mycobacterium ulcerans infection on a temperate southern
Australian island. Epidemiol Infect 119: 313–318.
Wagner T, Benbow ME, Burns M, Johnson RC, Merritt R, et al. (2008a) A
Landscape-based Model for Predicting Mycobacterium ulcerans Infection (Buruli
Ulcer Disease) Presence in Benin, West Africa. EcoHealth 5: 69–79.
Wagner T, Benbow ME, Brenden T, Qi J, Johnson RC (2008b) Buruli ulcer
disease prevalence in Benin, West Africa: associations with land use/cover and
the identification of disease clusters. International Journal of Health Geographics
7: 25.
Wallace J, Gordon M, Hartsell L, Mosi L, Benbow M, et al. (2010) Interaction
of Mycobacterium ulcerans with mosquito species: Implications for transmission
and trophic relationships. Applied Environ Microbiol 76: 6215–6222.
Walsh D, Portaels F, Meyers W (2008) Buruli ulcer (Mycobacterium ulcerans
infection). Transactions of the Royal Society of Tropical Medicine and Hygiene
102: 969–978.
Walsh D, Eyase F, Onyango D, Odindo A, Waitumbi JN et al. (2010) Short
Report: Clinical and Molecular Evidence for a Case of Buruli Ulcer
(Mycobacterium ulcerans Infection) in Kenya. Am. J. Trop. Med. Hyg., 81(6), pp.
1110–1113
Wansbrough-Jones M, Phillips R (2006) Buruli ulcer: emerging from obscurity.
Lancet 367: 1849–1858.
Williamson HR, Benbow ME, Nguyen KD, Beachboard DC, Kimbirauskas RK, et
al. (2008) Distribution of Mycobacterium ulcerans in Buruli Ulcer Endemic and
Non-Endemic Aquatic Sites in Ghana. PLoS Neglected Tropical Diseases 2:
e205.
WHO, ed (2000) Buruli ulcer. Mycobacterium ulcerans infection. Geneva,
Switzerland: WHO. 118 p.
WHO (2008) Buruli ulcer: progress report, 2004–2008. In: WHO, ed. Weekly
epidemiological record. Geneva, Switzerland: World Health Organization 83:
145–156.
Yeboah-Manu D, Bodmer T, Mensah-Quainoo E, Owusu S, Ofori-Adjei D, et al.
(2004) Evaluation of decontamination methods and growth media for primary



54


isolation of Mycobacterium ulcerans from surgical specimens. J Clin Microbiol 42:
5875–5876.
Yip MJ, Porter JL, Fyfe JA, Lavender CJ, Portaels F, et al. (2007) Evolution of
Mycobacterium ulcerans and other mycolactone-producing mycobacteria from a
common Mycobacterium marinum progenitor. J Bacteriol 189: 2021–9.




55


CHAPTER 3
SEASONAL ASSOCIATIONS BETWEEN MYCOBACTERIUM ULCERANS AND
MACROINVERTEBRATE ASSEMBLAGES IN AQUATIC ENVIRONMENTS OF
GHANA, WEST AFRICA
INTRODUCTION
Buruli ulcer (BU) is an emerging, neglected, infectious disease most often
associated with poor, rural communities within developing nations (WHO 2008).
To date, the disease has been reported from at least 32 countries, with the
highest frequency of new cases being reported from the West African nations of
Cote D’ Ivoire, Benin and Ghana (WHO 2008). It is known that Mycobacterium
ulcerans is the pathogen responsible for causing BU disease (WHO 2000);
however, researchers have yet to conclusively identify the extent of the
pathogen’s distribution in the environment, the reservoir(s) of the pathogen in
nature, or the mode(s) of transmission to humans (WHO 2008; Merritt et al.
2010). For these reasons, in part, BU is referred to as the mysterious disease
(WHO 2000).
It is widely accepted that BU disease is in some way related to exposure
to freshwater environments (Aiga et al. 2004; Marston et al. 2005; Raghunathan
et al. 2005; WHO 2008). Several epidemiological studies have associated
cases of BU with proximity and prolonged exposure to disturbed aquatic habitats
(Lunn et al. 1965; Revill and Barker 1972; Barker and Carswell 1973; Marston et
al. 1995; Portaels 1995; Noeske et al. 2004; Debacker et al. 2006; Duker et al.
2006; Johnson et al. 2007; Wagner et al. 2008a; Walsh et al. 2008), and
furthermore, it has been suggested that human activities, such as- surface
56

mining, damming of waterbodies, deforestation practices, and agricultural
development are contributing factors leading to environmental disturbance and
increased risk of BU infection (Hayman 1991b; Marston et al. 1995; Meyers et al.
1996; Johnson et al. 1999; Portaels et al. 2001; Merrit et al. 2005; Duker et al.
2006; Kibadi et al. 2008; Wagner et al. 2008a,b). These disturbances are
believed to provide environmental conditions favorable for the establishment and
proliferation of M. ulcerans in aquatic habitats (Hayman 1991b; Portaels 1999;
Merritt et al. 2005; Williamson et al. 2008; McIntosh et al. submitted 2010). BU
incidence also has been reported to increase during prolonged dry periods, after
flooding events and in areas prone to flooding (Portaels 1989; Daire et al. 1993;
Meyers et al. 1996; Dabacker et al. 2004; Merritt et al. 2005; Duker et al. 2006;
Walsh et al. 2010), suggesting BU infection may be related to season.
It is hypothesized that M. ulcerans is acquired from the environment either
through inoculation of the pathogen into skin lesions or from a biological vector
(WHO 2000, 2008). There is a growing body of work that suggests aquatic
macroinvertebrates may be vectors of M. ulcerans to humans (Portaels et al.
1999; Marsoilier et al. 2002b; Johnson et al. 2007; Marion et al. 2010), and also
environmental reservoirs of the pathogen (Portaels et al. 1999, 2001, 2008;
Marsoilier et al. 2002a; 2003, 2004a, 2005; Fyfe et al. 2007; Johnson et al. 2007;
Lavender et al. 2008; Williamson et al. 2008; Tobias et al. 2009; Marion et al.
2010; Wallace et al. 2010). In Australia, mosquitoes are believed to play a role in
BU transmission (Johnson et al. 1999, 2007; Fyfe et al. 2007; Lavender et al.
2008; WHO 2008) and disease outbreaks have been correlated with Ross River

57

virus and Barmah Forest virus, both of which are vectored by mosquitoes
(Johnson et al. 2009). In West Africa, aquatic biting Hemiptera populations have
been associated with BU infection and have been proposed to be both reservoirs
and vectors of M. ulcerans to humans (Portaels et al. 1999; Marsoilier et al.
2002b, 2007; Marion et al. 2010). Despite the literature showing an association
of aquatic insects in the transmission of this disease, major scientific criteria are
lacking for implicating the roles of living agents as biologically significant
reservoirs and/or vectors of pathogens (Merritt et al. 2010).
The role of aquatic macroinvertebrates as potential reservoirs or vectors of
BU is well documented as reviewed by Merritt et al. (2010); however, field based
ecological studies to specifically address these associations are few (Benbow et
al. 2007, Merritt et al. 2010). An initial step in understanding the potential role of
macroinvertebrates in the ecology of BU, whether direct or indirect, is identifying
the relative abundances and composition of the aquatic macroinvertebrate
communities in relation to the disease and disease pathogen. As part of a
standardized assessment of the temporal patterns of macroinvertebrate
communities, I surveyed 6 waterbodies selected from villages that were known to
have reported cases of BU (n= 3) and villages with no previous record of BU (n=
3), in Ghana, West Africa to 1) characterize and compare seasonal variation in
overall macroinvertebrate communities from aquatic environments in Ghana; 2)
identify macroinvertebrate community metrics associated with the presence and
absence of BU cases and M. ulcerans within these environments; and 3) identify
potential relationships between macroinvertebrates, BU cases and M. ulcerans.

58

MATERIALS AND METHODS
Study location and site selection. A standardized, seasonal
assessment of aquatic habitats was conducted to characterize seasonal variation
in benthic macroinvertebrate communities in Ghana, West Africa. In this study, a
total of 6 waterbodies were selected from villages located within the Greater
Accra Region of southern Ghana. Villages were selected based on reported BU
case data (Ghana Ministry of Health) and personal communication with local
researchers familiar with BU infected and uninfected communities. Three villages
from the Ga West District were identified as endemic for BU (Afieman, Kotoku,
Nsakima) and 3 villages from the Ga East District were identified as BU nonendemic (Otinibi, Danfa, Teiman). Within each village, one waterbody was
selected based on location (<100-200m from community housing structures),
human use (daily domestic activities) to reflect aquatic environments with
potential human exposure to BU, and community discussions as described by
Benbow et al. (2005). Waterbodies varied in size and macrophyte community
composition, but all were characterized by slow flowing water and identified as
either modified ponds (MP, n= 3) or modified wetlands (MW, n= 3). Location,
GPS coordinates and general description for each water body are provided in
Table 3.1.
Seasonal sampling strategy. The climate of southern Ghana, where this
study was conducted, is tropical and seasons are characterized by wet and dry
periods. The dominant wet season occurs between September and November,
followed by a dry season December to March. A wet season also occurs

59

between April and June, followed by another dry period between July and
August. Aquatic macroinvertebrate communities in this study were surveyed on
a single sampling date from each waterbody in June 2007 (wet), November 2007
(wet), February 2008 (dry), April 2008 (wet) and July 2008 (dry).
Macroinvertebrate sample collections. Aquatic macroinvertebrate
communities were collected from the littoral margins of waterbodies. Within each
waterbody, two 10–20-m transects were measured parallel to the shoreline and
positioned through the dominant macrophyte community. Along each transect,
two floating 1-m2 polyvinyl chloride (PVC) quadrats were randomly placed and
macroinvertebrates were collected by sweeping within the quadrat with a 500-µm
mesh aquatic dip net. The quadrats floated on top of the water and delineated 1
m2 of area to be sampled. Three sweeps of the dip net were performed from the
water surface to the bottom substrate for comprehensive sampling of specimens
in the water column and all samples were collected using the same technique.
Contents within each net were then washed through a 500-µm sieve, preserved
in 99% ethanol, and transported to the laboratory for identification. All specimens
were enumerated and identified to lowest possible taxon under a dissecting
microscope using African keys (Durand and Leveque 1981; Invertebrates of
South Africa - Keys, vols. 2-10, 1999-2007), and keys from elsewhere (Merritt et
al. 2008). Voucher specimens are maintained in the Entomological Collection
Museum, Department of Entomology at Michigan State University.
Detection of Mycobacterium ulcerans. To identify potential
relationships between macroinvertebrate communities and M. ulcerans in the

60

environment, biofilms and water samples were collected at each water body
during each sampling event. Biofilms were collected from artificial substrates
(glass slides), surfaces of dominant macrophytes, and detritus, and a composite
water sample was collected from open water areas within the water body at the
mid-water column depth. From the composite, five 100-200ml sub-samples were
filtered through a 1.6 micron fiberglass filter followed by a 0.2 micron
nitrocellulose filter (Whatman Inc.). Filters were sealed inside aluminum foil
packets for later laboratory analysis. In addition, a 500 µl biofilm sample was
used for DNA extraction. Extracted DNA was also collected from M. ulcerans
agy99, Mycobacterium marinum 1218, or water for use as positive and negative
controls. All samples were processed at the University of Tennessee, Knoxville,
Tennessee and follow methods as previously described by Williamson et al.
(2008).
Primers, PCR conditions and sequencing. A tiered PCR detection
method was used for the identification of M. ulcerans in which DNA was first
subjected to amplification of the enoyl reductase (ER) domain. The ER domain
is partially responsible for production of the toxin mycolactone and was used for
the presumptive identification of M. ulcerans (George et al. 1999; Sizaire et al.
2002). ER-PCR positive samples were then subjected to VNTR analysis to
identify M. ulcerans from other mycolactone-producing mycobacteria, and to
potentially match sample profiles to known VNTR profiles obtained from patients
(George et al. 1999; Portaels et al. 2001; Sizaire et al. 2002; Ablordey et al.
2005; Johnson et al. 2007). Primers and PCR conditions for amplification of the

61

ER domain as well as VNTR loci, including BNTR MIRU 1, locus 6, ST 1 and
locus 19, were used as described by Williamson et al. (2008). For this study, the
presumptive PCR test for the ER domain was used for the determination of M.
ulcerans from environmental samples. Although this presumptive identification
potentially gives an overestimate of M. ulcerans in the environment, it also
provides a maximum estimate for all mycolactone-producing mycobacteria.
Data Analyses. Descriptive and inferential statistics were used to test the
relationships between macroinvertebrate communities and selected independent
variables. The independent variables were: site, season, presence or absence of
reported cases of BU (BU+/BU-), and presence or absence of the pathogen
(ER+/ER-) within waterbodies. The macroinvertebrate community metrics
evaluated in this study comprised both diversity and similarity indices (Total
Abundance, Total Taxa, Shannon-Weiner Diversity, Simpson’s Heterogeneity,
Margalef’s Richness and Pielou’s Eveness). Prior to all analyses, data hygiene
and data screening were undertaken to ensure the variables of interest met
appropriate statistical assumptions. Thus, the following analyses follow a similar
analytic strategy in that the dependent variables were first evaluated for
normality, linearity and homoscedasticity. Subsequently, two methods were used
to analyze macroinvertebrate metrics in relation to the independent variables.
First, multiple linear regressions were run to investigate relationships between
macroinvertebrate communities and the independent variables BU cases
(BU+/BU-) and pathogen (ER+/ER-)(SPSS v 17.0). Second, repeated measures
profile analyses were run to detect amount of shared variance and strength of

62

relationship between the macroinvertebrate community metrics and the
independent variables: BU cases (BU+/BU-), pathogen (ER+/ER-), sampling
season (June 2007, November 2007, February 2008, April 2008, July 2008), and
between waterbodies (n= 6)(SPSS 17.0). Finally, Pearson correlation coefficient
was used to identify potential relationships between individual macroinvertebrate
taxa (family) and the independent variables BU cases (BU+/BU-) and pathogen
(ER+/ER-)(SPSS 17.0). Correlation coefficients were performed through
parametric tests using SPSS 17.0 with p > 0.05 and p > 0.01 as threshold for
significance.
RESULTS
Macroinvertebrate composition and distribution among sites. A total
of 25,104 invertebrates from 69 unique taxa were identified and used for
analyses in this study (Table 3.2). The greatest abundance of benthic
macroinvertebrates was recorded in February (n= 9456), despite the fact that the
waterbody at Teimen village was not sampled during this month, followed by
June (n= 5854), November (n= 5016), April (n= 2829) and July (n=2083). The
greatest number of different taxa collected was recorded in June (n= 54);
followed by November (n= 50), April (n= 49), February (n= 46) and July (n= 42).
The most abundant taxa throughout the entire study were mayflies
(Ephemeroptera: Baetidae, n= 4030, 16%); followed by midges (Diptera:
Chironomidae, n= 3725, 14.8%) and cladocerans (n= 2621, 10.4%). The
majority of the cladocerans were collected from one waterbody during February
(n= 2522), but were completely absent in all April collections and relatively few

63

were collected during June (n= 54), November (n= 5) and July (n= 40). Taxa
dominance shifted throughout the season. Baetid mayflies were the most
abundant taxa collected February, April and July (n= 983, 734, 418, respectively);
however, mosquitoes (Diptera: Culicidae) were the dominant taxa in June (n=
896) and midges (Chironomidae) were the dominant taxa in November (n=
1241). Several taxa were found only during one season. The phantom midge
(Diptera: Chaoboridae, n= 19) was only collected June 2007, but it occurred in
three of six study sites that season. The collembolan family Entomobryiidae was
collected during every season; however, only one specimen from the
collembolan family Sminthuridae was collected during this entire study and it was
collected during June, as were Sphaerid clams (n= 8) and polymitarcid mayflies
(n= 7). In November, three beetle taxa (Elmidae, Gyrinidae and Hydrobiidae)
were collected that were not found during the other sampling events.
There also were - taxa that were recorded from only one waterbody.
Three taxa (Hemiptera: Saldidae, Neuroptera: Sysiridae, and Coleoptera:
Elmidae) were only found in the waterbody at Nsakima village; two taxa
(Collembola: Sminthuridae and Veneroida: Sphaeriidae) were only found in the
waterbody at Kotoku village; and one taxa was only found in the waterbodies at
Danfa (Ephemeroptera: Oligoneuridae), Afieman (Ephemeroptera:
Polymitarcidae) and Teimen (Diptera: Simuliidae). Belostomatidae were
completely absent from one site (Kotoku, Ga West) and Naucoridae were absent
from two sites (Otinibi, Ga East and Afieman, Ga West). Together, these data
demonstrate the importance of sampling multiple sites over a period of time to

64

more completely understand the macroinvertebrate communities in these
waterbodies and to make more accurate associations between these
communities and the ecology of BU disease.
Results for multiple linear regression analysis based on BU cases.
Multiple linear regression analysis was run to investigate relationships between
macroinvertebrate communities within waterbodies identified as either BU+ or
BU- (SPSS 17.0). There was not a significant relationship between BU case
data based on the combined group of macroinvertebrate metrics (Total
Abundance, Total Taxa, Shannon-Weiner Diversity, Simpson’s Heterogeneity,
Margalef’s Richness and Pielou’s Eveness); R = .321, R2 = .103, F (6, 113) =
2.160, p = .052 (two-tailed). The R-squared value demonstrates that 10.3% of the
variance in the macroinvertebrate metrics can be explained the by the presence
or absence of BU cases. Table 3.3 presents a model summary of the multiple
regression analysis of macroinvertebrate metric relationships with presence and
absence of BU cases, and Table 3.4 shows the descriptive statistics from the
regression analysis.
Results for multiple linear regression analysis for pathogen (ER+/ER). Multiple linear regression analysis was run to investigate relationships
between macroinvertebrate communities within waterbodies identified as either
ER+ or ER- (SPSS 17.0). There was a significant relationship between ER
presence and absence based on the combined group of macroinvertebrate
metrics (Total Abundance, Total Taxa, Shannon-Weiner Diversity, Simpson’s
Heterogeneity, Margalef’s Richness and Pielou’s Eveness); R = .384, R2 = .148,

65

F (6, 113) = 3.26, p = .005 (two-tailed). The R-squared value demonstrates that
14.8% of the variance in the macroinvertebrate metrics can be explained the by
presence or absence ER within waterbodies. Table 3.5 is a model summary of
the multiple regression analysis of macroinvertebrate metric relationships with
presence and absence of ER in waterbodies. The contribution of each predictor
variable, when the others are controlled for, was evaluated using the
standardized Beta for each coefficient. None of the individual variables made a
statistically unique contribution to the model. Together, these results indicate that
the combined macroinvertebrate metrics are potential predictors of M. ulcerans,
but that not one individual metric is a predictor of the pathogen in the
environment. Table 3.6 shows the descriptive statistics from the regression
analysis.
Summary results for profile analyses. Repeated measures profile
analyses were run to detect amount of shared variance and strength of
relationship between macroinvertebrate community metrics and the independent
variables: BU cases (BU+/BU-), pathogen (ER+/ER-), sampling season (June
2007, November 2007, February 2008, April 2008, July 2008) and individual
waterbodies (n= 6)(SPSS 17.0). Results indicated there were significant
differences in the dependant variables (Total Abundance, Total Taxa, ShannonWeiner Diversity, Simpson’s Heterogeneity, Margalef’s Richness and Pielou’s
Eveness) depending on sampling season (n= 5) and individual waterbody (n= 6),
but that there were no significant differences in the six macroinvertebrate metrics

66

between neither BU cases (BU+/BU-) nor pathogen (ER+/ER-)(p= .542)(Table
3.7).
Missing data and univariate outliers. A test for univariate outliers was
conducted and none were found to exist within the distribution. Univariate outliers
were sought by converting observed scores to z-scores and then comparing case
values to the critical value of +/-3.29, p < .001. Case z-scores that exceed this
value are greater than three standard deviations from the normalized mean.
Missing data were investigated by running frequency counts (SPSS 17.0). No
cases were missing, thus, 114 responses from participants were received and
114 were entered into the multiple regression models (n = 114). Before analysis,
basic parametric assumptions were assessed. That is, for the criterion and
predictor variables, assumptions of normality, linearity, and homoscedasticity of
variance were evaluated. Results showed the variables to be normally distributed
and assumed to meet parametric assumptions.
To examine the assumption of homogeneity of variance Box’s M-Test of
Equality of Covariance Matrices was run (Seber 1984). This test was run to
determine if the dependent variable distributions were equal across the levels of
the independent variable (BU+/BU-). Results from the test found that the
distributions were not equal across groups for cases, F (21, 51212.449) = 7.846,
p < .001. These results suggest that the two distributions were not equally
distributed and therefore the homogeneity of variance assumption is not met.

67

Profile analysis for BU case data. Repeated measures profile analyses
were run to detect amount of shared variance and strength of relationship
between macroinvertebrate community metrics within waterbodies identified as
either BU+ or BU- (SPSS 17.0). Using Wilks’ criterion, the profiles did not
significantly deviate from parallelism, F (5, 114) = 1.001, p = .420, partial etasquared = .042. For the between-groups test, there were no statistically
significant differences found for the dependent variables when scores were
averaged over all BU case data; F (1,118) = 2.082, p = .152, partial-eta squared
= .017. The partial eta-squared statistic means that 1.7% of the reason why the
combined macroinvertebrate community metrics varied was due to the effect of
the independent variable (BU cases). The test of within-subjects effects reveals
that there are no significant differences in macroinvertebrate community metrics
between the independent variable’s (BU cases); F (5, 114) = .000, p = 1.00. In
other words, the dependent variable measures (Total Abundance, Total Taxa,
Shannon-Weiner Diversity, Simpson’s Heterogeneity, Margalef’s Richness and
Pielou’s Eveness) do not vary depending on the reported BU case data.
Missing data and univariate outliers. A test for univariate outliers was
conducted and none were found to exist within the distribution. Univariate outliers
were sought by converting observed scores to z-scores and then comparing case
values to the critical value of +/-3.29, p < .001. Case z-scores that exceed this
value are greater than three standard deviations from the normalized mean.
Missing data were investigated by running frequency counts in SPSS (version
17.0). No cases were missing, thus, 114 responses from participants were

68

received and 114 were entered into the multiple regression models (n = 114).
Before analysis, basic parametric assumptions were assessed. That is, for the
criterion and predictor variables, assumptions of normality, linearity, and
homoscedasticity of variance were evaluated. Results showed the variables were
normally distributed and assumed to meet parametric assumptions.
To examine the assumption of homogeneity of variance Box’s M-Test of
Equality of Covariance Matrices was run. This test was run to determine if the
dependent variable distributions were equal across the levels of the independent
variable (ER+/ER-). Results from the test found that the distributions were not
equal across groups for pathogen, F (df 21, 37359.566) = 10.630, p < .001.
These results suggest that the two distributions were not equally distributed and
therefore the homogeneity of variance assumption is not met.
Profile analysis for pathogen (ER+/ER-). Repeated measures profile
analyses were run to detect amount of shared variance and strength of
relationship between macroinvertebrate community metrics within waterbodies
identified as either ER+ or ER- (SPSS 17.0). Using Wilks’ criterion, the profiles
did not deviate significantly from parallelism, F (5, 114) = .140, p = .983, partial
eta-squared = .006. For the between-groups test, there were no statistically
significant differences found for the dependent variables when scores were
averaged for ER+ and ER- waterbodies; F (1,118) = .140, p = .339, partial-eta
squared = .008. The partial eta-squared statistic means that 0.8% of the reason
why the combined macroinvertebrate community metrics varied was due to the
effect of the independent variable (ER). The test of within-subjects effects

69

reveals that there are no significant differences between dependant variables
and independent variables; F (5, 114) = .140, p < .983. In other words, the
macroinvertebrate metrics (Total Abundance, Total Taxa, Shannon-Weiner
Diversity, Simpson’s Heterogeneity, Margalef’s Richness and Pielou’s Eveness)
did not vary between waterbodies that were either ER+ or ER-.
Profile analysis of macroinvertebrate associations with Season. All
cases were examined for accuracy and found to be correctly recorded. Further,
there were no cases with missing values. A test for univariate outliers was
conducted for each group and none were found to exist within the distributions;
thus, 120 responses were received and 120 were entered into the Profile
Analysis model; n = 120. There were a number of variables that were either
skewed, kurtotic, or both (Table 3.8). Normality of the distributions is assumed
when z-skew coefficients were less than the critical value of +/- 3.29 (Tabachnick
and Fidel 2008). In addition, the variables differed in ranges of scores. There are
some tests of profile analysis that evaluate difference of scores in the dependent
variables, making the scaling of the variables important. For this reason,
dependent variables were standardized into z-scores around the mean for this
analysis.
To examine the assumption of homogeneity of variance Box’s M-Test of
Equality of Covariance Matrices was run (Seber 1984). This test was run to
determine if the dependent variable distributions were equal across the levels of
the independent variable. Results from the test found that the distributions were
not equal across groups, F (df 84, 25266.805) = 4.7, p < .001. These results

70

suggest that the two distributions were not equally distributed and therefore the
homogeneity of variance assumption is not met.
Profile analysis based on season. Repeated measures profile analyses
were run to detect amount of shared variance and strength of relationship
between macroinvertebrate community metrics and sampling season (n=
5)(SPSS 17.0)(Figure 3.1). Using Wilks’ criterion, the profiles deviated
significantly from parallelism, F (20, 369.095) = 3.08, p < .001, partial etasquared = .120. For the between-groups test, no statistically significant
differences were found among groups (sampling season) when scores were
averaged over all seasons F (4, 115) = 2.14, p = .08, partial-eta squared = .069.
The partial eta-squared statistic means that 6.9% of the reason why the
combined macroinvertebrate community metrics varied was due to the
independent variable (sampling swason); F (5, 111) = .000, p = 1.0. In other
words, while the grouped macroinvertebrate metrics were useful as predictors of
sampling season, the macroinvertebrate metrics individually were not (Table 3.8).
Missing data and univariate outliers. All cases were examined for
accuracy and found to be correctly recorded. Further, there were no cases with
missing values. A test for univariate outliers was conducted for each group and
none were found to exist within the distributions; thus, 114 responses were
received and 114 were entered into the Profile Analysis model; n = 114. There
were a number of variables that were either skewed, kurtotic, or both (Table 3.8).
Normality of the distributions is assumed when z-skew coefficients were less
than the critical value of +/- 3.29 (Tabachnick & Fidel, 2008). In addition, the

71

variables differed in ranges of scores. There are some tests of profile analysis
that evaluate difference of scores in the dependent variable, making the scaling
of the variables important. For this reason, dependent variables were
standardized into z-scores around the mean for this analysis.
To examine the assumption of homogeneity of variance Box’s M-Test of
Equality of Covariance Matrices was run (Seber 1984). This test was run to
determine if the dependent variable distributions were equal across the levels of
the independent variables (waterbodies, n= 5). Results from the test found that
the distributions were not equal across groups, F (df 105, 20867.275) = 5.89, p <
.001. These results suggest that the two distributions were not equally distributed
and therefore the homogeneity of variance assumption is not met.
Profile analysis based on waterbody. Repeated measures profile
analyses were run to detect amount of shared variance and strength of
relationship between macroinvertebrate community metrics and individual
waterbodies (n= 6)(SPSS version 17.0)(Figure 3.2). Using Wilks’ criterion, the
profiles deviated significantly from parallelism, F (25, 410.134) = 2.60, p < .001,
partial eta-squared = .104. For the between-groups test, there were statistically
significant differences found among groups (waterbody) when scores were
averaged over all sites F (5, 114) = 4.963, p < .001, partial-eta squared = .179.
The partial eta-squared statistic means that 17.9% of the reason why the
combined macroinvertebrate community metrics varied was due to the effect of
waterbody. The test of within-subjects effects reveals that there are no
significant differences in the individual macroinvertebrate metrics between each

72

of the waterbodies; F (5, 110) = .000, p = 1.00, partial-eta squared = .000. In
other words, while the grouped macroinvertebrate metrics were useful as
predictors of different waterbodies, the macroinvertebrate metrics individually
were not.
Macroinvertebrate taxa correlations with BU cases. Significant
positive and negative Pearson correlations (P < 0.05) occurred between
macroinvertebrate taxa collected within waterbodies from villages with reported
cases of Buruli ulcer (Table 3.9). Significant positive correlations were observed
between Buruli ulcer cases and total abundance of Caenidae (r= .214, P < 0.01),
Chironomidae (r= .256, P < 0.01), Coenagrionidae (r= .280, P < 0.01),
Hydraenidae (r= .191, P < 0.05), Naucoridae (r= .202, P < 0.05), Noteridae (r=
.386, P < 0.01), Physidae (r= .246, P < 0.01), Planorbidae (r= .215, P < 0.05),
Pleidae (r= .368, P < 0.01) and Scirtidae (r= .232, P < 0.01). Significant negative
correlations were observed between Buruli ulcer cases and total abundance of
Atyidae (r= -.261, P < 0.01) and Gerridae (r= -.341, P < 0.01).
Macroinvertebrate taxa correlations with pathogen. Significant
positive and negative Pearson correlations (P < 0.05) occurred between
macroinvertebrate taxa and waterbodies that were ER+ and ER- (Table 3.9).
Significant positive correlations were observed between pathogen presence,
based on ER detection from the waterbody, and total abundance of Atyidae (r=
.213, P < 0.05) and Gerridae. Significant negative correlations were observed
between pathogen presence, based on ER detection from the waterbody, and
total abundance of Anclyidae (r= -.272, P < 0.01), Chironomidae (r= -.221, P <

73

0.05), Dytiscidae (r= -.234, P < 0.01), Hydraenidae (r= -.245, P < 0.01),
Hydrophilidae (r= -.241, P < 0.01), Naucoridae (r= -.259, P < 0.01), Noteridae (r=
-.300, P < 0.01), Planorbidae (r= -.184, P < 0.05) and Pleidae (r= -.195, P <
0.05).
Seasonal distributions and composition of Hemiptera taxa.
Hemiptera taxa abundances were relatively low throughout the entire study (n=
1571) and represented only 6.2% of the total macroinvertebrates collected. The
greatest abundances of Hemiptera were recorded during the February (n= 411)
collections; however, all other macroinvertebrate taxa were also collected in
greater proportions during this season and total Hemipterans only compromised
4.3% of the February collections. The months with the greatest concentrations
(relative abundance) of Hemipterans compared to all specimens collected were
April (12.3%) and July (10.7%), whereas the concentrations of Hemiptera in June
(5.5%), November (5.4%) and February (4.3%) were relatively low. Eleven
families of Hemiptera were recorded during this study (Table 3.2), eight of which
(Belostomatidae, Gerridae, Hydrometridae, Mesoveliidae, Naucoridae, Nepidae,
Notonectidae and Pleidae) were collected in every month throughout the study
period. Of the remaining three families, Veliidae and Corixidae were collected
every month with the exception of July, and Hebridae were only collected in June
and April. When only looking at the Hemiptera taxa composition, the most
abundant families were the Notonectidae (29.9%) and Pleidae (26.5%), followed
by Mesoveliidae (13.3%) and Gerridae (12.6%). Belostomatidae and Naucoridae
were less abundant at 7.2% and 3.7%, respectively.

74

There were marked differences in Hemiptera composition between
waterbodies sampled. The waterbody in Teimen village had the highest
proportion of water bugs over all seasons compared to all other taxa (25.8%),
whereas the proportion of water bugs collected from waterbodies in Otinibi
(3.2%) and Nsakima (3.5%) villages were much lower. Of the 6 waterbodies
sampled during this study, only the waterbody in Danfa village produced all
eleven Hemiptera taxa. Naucoridae were never collected at Afieman and
Nsakima village, and Belostomatidae were never collected at Kotoku village.
DISCUSSION
Aquatic macroinvertebrates have been proposed by several authors to be
possible reservoirs of M. ulcerans or vectors of the pathogen to humans (Merritt
et al. 2010). Despite these potential connections with both the ecology of M.
ulcerans and BU transmission, standardized ecological studies aimed to
investigate macroinvertebrate community associations with the pathogen and
disease are limited (Benbow et al. 2008, Merritt et al. 2010). To date, most field
studies investigating aquatic invertebrates have primarily targeted specific taxa,
and sampling strategies have been either qualitative or lack adequate replication.
An initial step in understanding the role aquatic macroinvertebrates might play in
the ecology of BU, whether direct or indirect, is identifying the distribution and
composition of the entire macroinvertebrate community in relation to the disease
and disease pathogen. As part of a standardized assessment of the temporal
patterns of macroinvertebrate communities, I surveyed 6 waterbodies selected
from villages that were known to have reported cases of BU (n= 3) and villages

75

with no previous record of BU (n= 3), in Ghana, West Africa to characterize and
compare overall macroinvertebrate seasonal variation, community metrics, with
the presence and absence of BU cases and M. ulcerans within these
environments.
Results generated from this study identified no significant relationship
between macroinvertebrate community measurements in relation to BU+ and
BU- villages. In other words, the aquatic macroinvertebrate communities in
waterbodies within villages reporting at least one case of BU, prior to the
beginning of this study, were similar to those in waterbodies from villages
reporting no cases of BU. This result is comparable to what was reported by
Benbow et al. (2008), who conducted survey studies of aquatic
macroinvertebrate communities in endemic and non-endemic areas of southern
Ghana. In central Cameroon, Marion et al. (2010) report higher abundances of
Hemiptera taxa (water bugs) from a BU endemic stretch of the Nyong River than
what were found in a stretch of the same river associated with a village identified
as BU non-endemic. Marion et al. (2010) and I both present data that indicate
greater concentrations of macroinvertebrates during dry seasons; however, while
I found higher abundance of total macroinvertebrates in waterbodies from
endemic areas, water bug total abundances and percent composition were
highest in the non-endemic areas. One possible explanation for the differences
in taxa abundances and composition between these two studies was that
macroinvertebrate communities were expected to differ in relation to water body
and habitat availability. My field sites were all lentic (slow-flowing) habitats,

76

whereas the study site sampled by Marion et al. (2010) was a large river
characterized by flowing water (lotic). Together, these variations suggest
additional collection sites should be included in order for a more comprehensive
evaluation of invertebrate communities in relation to BU endemicity.
My profile analysis of macroinvertebrate community metrics in association
with the presence and absence of M. ulcerans, based on presumptive ER testing,
also generated no statistically significant relationships across seasons. The
multiple linear regression model indicated a significant relationship between
combined macroinvertebrate metrics and M. ulcerans, but not one individual
metric could be used as a predictor of the presence of M. ulcerans in the
environment. The significant result of the combined community analyses
suggests a complex biological system, and points toward the importance for
future studies to include collections of the entire macroinvertebrate community to
elucidate associations between biological communities and BU disease.
The only other study that compared macroinvertebrate community
measurements with the presence of M. ulcerans in the environment was
conducted in a companion study to this and results were reported in Chapter 2 of
this dissertation. In that study, using multivariate analyses (NMDS), I found no
significant relationships between the macroinvertebrate community and presence
of environmental M. ulcerans in lentic habitats, as was the case in the current
study. There were, however, significant differences in the macroinvertebrate
community profile and bioassessment metrics in relation to M. ulcerans in lotic
habitats. This further underlines the importance of considering habitat for future

77

development of proper experimental designs and when analyzing data regarding
macroinvertebrate communities in relation to M. ulcerans and BU endemicity.
I identified significant relationships between macroinvertebrate community
metrics and both season and individual waterbodies during the current study. In
both cases, the profile analysis of grouped means was significant but no
individual macroinvertebrate metrics could be identified as good indicators of
either season or waterbody. Variation in macroinvertebrate distribution and taxa
composition between and among waterbodies can be due to several factors;
including, physical, chemical, and biological parameters (USEPA 1996, 2001;
Merritt et al. 2008). In Malawi, central Africa, McLachlan (1975) investigated the
role of aquatic macrophytes in the recovery of the benthic fauna after a dry
season and found associations between macroinvertebrate taxa and specific
plant communities. Prior to that, Petr (1968) reported differences in
macroinvertebrate taxa composition and abundance in relation to the habitats
provided by the floating Pistia stratiotes and submersed Ceratophyllum
demersum in Volta Lake (Ghana, West Africa). Aquatic plants have also been
shown in the laboratory to promote biofilm and M. ulcerans development
(Marsolier et al. 2004b), and associations between macrophyte communities and
the presence of environmental M. ulcerans were reported by McIntosh et al.
(submitted 2010). General macrophyte community measurements (dominant
taxa, percent surface coverage) were collected during my study; however, the
results reported here were based entirely on macroinvertebrate community
assemblages in relation to BU disease parameters. Additional analyses

78

incorporating the abiotic and biotic measurements recorded during this study are
under way.
A focal interest of this study was to investigate temporal variations in
macroinvertebrate community distributions and compositions in relation to M.
ulcerans and BU disease. While strong relationships were not found between
macroinvertebrate communities and pathogen or BU case data, there were
significant relationships between macroinvertebrates and season. In this study,
overall macroinvertebrate abundances were greatest in the February sampling
event, which for southern Ghana represents a dry season. Marion et al. (2010)
reported high densities of Hemiptera densities from Cameroon during collections
made in January, and studies on the life histories of Naucoridae in Costa Rica
showed a similar pattern (Stout 1981, 1982). The work by Stout on Naucoridae
indentified water bug distribution patterns that varied due to season and
environmental condition, but it was also concluded that seasonal patterns in total
abundance did not exist due to the presence of all life stages of Naucoridae that
were collected throughout the year (Stout 1981). Asynchronous life history
patterns are often a characteristic of aquatic macroinvertebrates in tropical
environments (Merchant and Yule 1996; Huryn and Wallace 2000).
I found variation in the frequency and composition of Hemiptera taxa
throughout the sampling period, but in general Hemipterans only represented a
small percent (6.2%) of the total macroinvertebrate community, and in February
the percent composition of hemipterans dropped to 4.3% of the total community.
When comparing the composition of the entire hemipteran community,

79

Belostomatidae and Naucoridae represented a small percentage of the total
Hemiptera taxa composition at 7.2% and 3.7%, respectively. Benbow et al.
(2008) found comparable abundance and composition patterns in Hemiptera taxa
during their investigation of macroinvertebrate communities in Ghana, and
concluded that these data do not rule out the possibility of biting Hemiptera or
other invertebrates as vectors or possible reservoirs for M. ulcerans, but that
caution should be used in describing their role in transmission based solely on
abundance patterns.
Belostomatidae and Naucoridae have received the most attention from
researchers in West Africa as potential vectors of M. ulcerans, and in my study
the abundances of these two taxa were low and in some cases these water bugs
were missing completely from entire waterbodies. Belostomatidae were
completely absent from one site (Kotoku, Ga West) and Naucoridae were absent
from two sites (Otinibi, Ga East and Afieman, Ga West), despite sampling each
waterbody 5 times over the period of one year. Future investigators of
macroinvertebrate associations with BU disease should consider the potential
spatial and temporal variations in the entire community composition, as well as
proper replication of study sites, in order to enhance the accuracy of statements
that can be drawn from their studies.
Pearson correlation’s identified several individual macroinvertebrate taxa
that were correlated with either reported cases of BU, the presence and absence
of environmental M. ulcerans, based on the presumptive ER test, or both. Of
particular interest, considering their proposed associations with BU disease, are

80

the correlations found with Belostomatidae and Naucoridae. Belostomatid water
bugs were found to be negatively correlated with BU case data, which differs
from what has been found elsewhere (Marsoilier et al. 2004; Marion et al. 2010)
and is not what would be expected if these insects are intimately involved with
BU disease transmission, as outlined by Benbow et al. (2008). Naucorid water
bugs on the other hand were positively correlated with BU cases data, but were
also negatively correlated with the presence of M. ulcerans. This pattern of
positively correlated with BU cases and at the same time negatively correlated
with M. ulcerans, or negatively correlated with BU and positively correlated with
M. ulcerans was found throughout the results from these analyses. There are
multiple rationales that could explain these relationships; however, the
Interpretation of these results should be made with caution.
First, the relatively low numbers of some of these taxa, particularly the
Belostomatidae and Naucoridae which were missing completely from sites, might
be contributing to decreased power and inability to detect a true effect (Kendall
and Gibbons 1990). Second, although a number of taxa were identified as
predictors of BU cases (n= 13) and M. ulcerans (n= 11), several more (n= 69)
were used for this analysis and there is potential for unwarranted artifacts to
emerge as a result (Kendall and Gibbons 1990). Finally, correlations between
two variables do not automatically mean that they are directly associated with
each other (correlation versus causation). What these results do is offer insight
into the potential associations between specific macroinvertebrate taxa and BU
disease ecology, which suggest a complex interaction between BU and biological

81

communities. Furthermore, these results provide a framework for further studies
to be directed toward elucidating these relationships.
Epidemiological studies have identified patterns in the emergence of BU
cases that can be attributed to seasonal variation (Daire et al. 1993, Portaels
1989, Meyers et al. 1996; Johnson et al. 2007), and empirical data suggest that
seasonal patterns in rainfall and subsequent flooding may provide environmental
conditions that are favorable for the establishment and proliferation of M.
ulcerans (Hayman 1991b; Portaels 1999; Merritt et al. 2005, Williamson et al.
2008; McIntosh et al. submitted). For these reasons, and the proposed
associations between aquatic macroinvertebrates and BU disease, an area of
interest should be to develop a more complete understanding of the seasonal
variations in macroinvertebrate communities within BU endemic and nonendemic areas. In this study, I identified variations in macroinvertebrate
distributions and compositions in relation to both waterbody and season, and
also identified individual taxa that show potential associations with both BU and
M. ulcerans. Results from this study should be used as a framework to aid in the
development of forthcoming studies with the goal of examining associations
between macroinvertebrate communities, M. ulcerans, and BU.
It should be noted that the case data used in this study was obtained
through passive surveillance practices, provided by the Ghana Ministry of Health,
and may not reflect the true disease incidence. Likewise, many of the
associations made between macroinvertebrates and M. ulcerans were based on
the presumptive ER test, which might over-estimate the true distribution of M.

82

ulcerans in the environment. These data were also interpreted under the
assumption that the waterbodies where the bacteria were not detected were in
fact M. ulcerans negative sites. While these factors should not be overlooked,
neither should the potential value that the information this study provides to
scientists involved with BU research.

83







84








85








86








87








88








89








90





91








92








93








94


Figure 3.1. Estimated marginal means for six macroinvertebrate metrics across five
sampling seasons. The macroinvertebrate metrics were total specimens, total taxa,
Shannon-Weiner Diversity, Simpson’s Heterogeneity, Margalef’s Richness, and Pielou’s
Eveness. The sampling seasons are listed as Season 1 (June 2007), Season 2 (November
2007), Season 3 (February 2008), Season 4 (April 2008), and Season 5 (July 2008).

95










Figure
3.2.
Estimated
marginal
means
for
six
macroinvertebrate
metrics
across
six

waterbodies.

The
macroinvertebrate
metrics
were
total
specimens,
total
taxa,

Shannon‐Weiner
Diversity,
Simpson’s
Heterogeneity,
Margalef’s
Richness,
and

Pielou’s
Eveness.

The
were
from
the
following
villages:
Site
1=Otinibi,
Site
2=Danfa,

Site
3=Teimen,
Site
4=
Afieman,
Site
5=Kotoku,
Site
6=Nsakima.








96


LITERATURE CITED




97


Literature Cited

Ablordey A, J Swings, C Hubans, K Chemlal, C Locht, F Portaels, P Supply
(2005) Multilocus variable-number tandem repeat typing of Mycobacterium
ulcerans. J Clin Microbiol 43, 1546–1551.
Aiga H, Amano T, Cairncross S, Domako JA, Nanas OK, et al. (2004)
Assessing water-related risk factors for Buruli ulcer: A case-control study in
Ghana. Am J Trop Med Hyg 71: 387–392.
Barker DJP (1973) Epidemiology
of
Mycobacterium
ulcerans
infection. Trans R
Soc Trop Med Hyg. 67:(1)43.
Barker DJP, Carswell JW (1973) Mycobacterium ulcerans infection among
Tsetse control workers in Uganda. International Journal of Epidemiology 2: 161–
165.
Batzer DP, SA Wissinger (1996) Ecology of insect communities in nontidal
wetlands. Annu Rev Entomol. 41:75–100.
Benbow M, Williamson H, Kimbirauskus R, McIntosh M, Kolar R, et al. (2008)
Aquatic invertebrates as unlikely vectors of Buruli ulcer disease. Emerg Infect Dis
14: 1247–1254.
Darie H, T Le Guyadec, JE Touze (1993) "Aspects épidémiologiques et cliniques
de l'ulcère de Buruli en Côte-d'Ivoire". Bull Soc Pathol Exot Filiales 86: 272–276.
Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, et al. (2004)
Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin,
1997-2001. Emerg Infect Dis 10: 1391–1398.
Debacker M, Portaels F, Aguiar J, Steunou C, Zinsou C, et al. (2006) Risk factors
for Buruli ulcer, Benin. Emerg Infect Dis 12: 1325–1331.
Dobson M, Magana AM, Lancaster J, Mathooko JM. (2007) Aseasonality in the
abundance and life history of an ecologically dominant freshwater crab in the Rift
Valley, Kenya. Freshw Biol. 52:215–25.
Donohue I, Irvine K. (2004) Seasonal patterns of sediment loading and benthic
invertebrate community dynamics in Lake Tanganyika, Africa. Freshw Biol.
49:320–31.
Duker AA, Carranza JM, Hale M (2004) Spatial dependency of Buruli ulcer
prevalence on arsenic-enriched domains in Amansie West District, Ghana:
implications for arsenic mediation in Mycobacterium ulcerans infection. Int J
Health Geogr. 3:19.



98


Duker AA, Portaels F, Hale M (2006) Pathways of Mycobacterium ulcerans
infection: A review. Environment International 32: 567–573.
Durand, JR, Leveque, C (1981) Flore et Faune Aquatiques de l’Afrique SaheloSoudaniene, 2 volumes. ORSTOM, Paris.
Fyfe JAM, Lavender CJ, Johnson P, Globan M, Sievers A, et al. (2007)
Development and Application of Two Multiplex Real-Time PCR Assays for
the Detection of Mycobacterium ulcerans in Clinical and Environmental Samples.
Appl Environ Microbiol 73: 4733–4740.
George KM, D Chatterjee, G Gunawardana, D Welty, T Lee, et al. (1999)
Mycolactone: a polyketide toxin from Mycobacterium ulcerans required for
virulence. Science 283: 854–857.
Hayman J (1991a) Postulated epidemiology of Mycobacterium ulcerans infection.
International Journal of Epidemiology. 20(4): 1093-8.
Hayman J (1991b) Mycobacterium ulcerans infection. The Lancet 337: 124.
Huryn AD, Wallace JB (2000) Life history and production of stream insects. Annu
Rev Entomol. 45:83–110.
Invertebrates of South Africa (1999-2007) Identification Keys, vols. 2-10. Rhodes
University, Department of Zoology and Entomology, Johannesburg, South Africa.
Johnson PDR, Stinear TP, Hayman JA (1999) Mycobacterium ulcerans — a
minireview. J Med Microbiol 48: 511–513.
Johnson PDR, Stinear TP, Small PLC, Pluschke G, Merritt RW, et al. (2005)
Buruli ulcer (M. ulcerans Infection): new insights, new hope for disease
control.PLoS Med 2(4): e108.
Johnson PDR, Hayman JA, Quek TY, Fyfe JAM, Jenkin GA, et al. (2007)
Consensus recommendations for the diagnosis, treatment and control of
Mycobacterium ulcerans infection (Bairnsdale or Buruli ulcer) in Victoria,
Australia. Medical Journal of Australia 186: 64–68.
Johnson PDR, Lavender CJ (2009) Correlation between Buruli ulcer and vectorborne notifiable diseases, Victoria, Australia. Emerging Infectious Diseases 15:
614–615.
Kendall MG, Gibbons JD (1990) Rank Correlation Methods (5th edition). London:
Arnold.




99


Kibadi K, Panda M, Tamfum JM, Fraga AG, Filho AL, et al. (2008) New foci of
Buruli ulcer, Angola and Democratic Republic of Congo. Emerging Infectious
Diseases 14: 1790–1792.
Lavender CJ, Stinear TP, Johnson PDR, Azuolas J, Benbow ME, et al. (2008)
Evaluation of VNTR typing for the identification of Mycobacterium ulcerans in
environmental samples from Victoria, Australia. FEMS Microbiology Letters
287: 250–255.
Lunn HF, Connor DH, Wilks NE, Barnley GR, Kamunvi F, et al. (1965) Buruli
(Mycobacterial) ulceration in Uganda. East African Medical Journal 42: 275–288.
Marion E, Eyangoh S, Yeramian E, Doannio J, Landier J Aubry J et al. (2010)
Seasonal and regional dynamics of M. ulcerans transmission in environmental
context: deciphering the role of water bugs as hosts and vectors. PLoS Negl Trop
Dis. 6:4(7):731.
Marsollier L, Robert R, Aubry J, Andre JS, Kouakou H, et al. (2002) Aquatic
insects as a vector for Mycobacterium ulcerans. Applied and Environmental
Microbiology 68: 4623–4628.
Marsollier L, Aubry J, Saint-Andre JP, Robert R, Legras P, et al. (2003)
Ecology and transmission of Mycobacterium ulcerans. Pathologie Biologie 51:
490–495.
Marsollier L, Severin T, Aubry J, Merritt RW, Saint Andre JP, et al. (2004a)
Aquatic snails, passive hosts of Mycobacterium ulcerans. Appl Environ Microbiol
70: 6296–6298.
Marsollier L, Stinear TP, Aubry J, Saint-Andre J-P, Robert R, et al. (2004b)
Aquatic plants stimulate the growth of and biofilm formation by Mycobacterium
ulcerans in axenic culture and harbor these bacteria in the environment. Applied
and Environmental Microbiology 70: 1097–1103.
Marsollier L, Aubry J, Coutanceau E, Andre JPS, Small PL, et al. (2005)
Colonization of the salivary glands of Naucoris cimicoides by Mycobacterium
ulcerans requires host plasmatocytes and a macrolide toxin, mycolactone.
Cellular Microbiology 7: 935–943.
Marsollier L, Andre J, Frigui W, Reysset G, Milon G, et al. (2006) Early
trafficking events of Mycobacterium ulcerans within Naucoris cimicoides. Cellular
Microbiology.
Marston BJ, Diallo MO, Horsburgh CR Jr., Diomande I, Saki MZ, et al.
(1995) Emergence of Buruli ulcer disease in the Daloa region of Cote D’ivoire.
Am J Trop Med Hyg 52: 219–224.
McIntosh MD, H Williamson, ME Benbow, RK Kimbirauskas, C Quaye, D



100


Boakye, PLC Small, RW Merritt (2010) Associations between Mycobacterium
ulcerans and Aquatic Plant Communities of West Africa: implications for Buruli
ulcer disease. (submitted 2010)
McLachlan AJ (1975) The role of aquatic macrophytes in the recovery of the
benthic fauna of a tropical lake after a dry phase. Limnology and Oceanography.
20(1) pp. 54-63.
Merchant R and CM Yule (1996) A method for estimating larval life spans of
aseasonal aquatic insects from streams on Bougainville Island, Papua New
Guinea. Freshwater Biology 35, 101-107.
Merritt RW, Benbow ME, Small PLC (2005) Unraveling an Emerging Disease
Associated with Disturbed Aquatic Environments: The Case of Buruli Ulcer.
Frontiers in Ecology and the Environment 3: 323–331.
Merritt, RW, Cummins, KW, MB Berg (2008) An introduction to the aquatic
insects of North America. Fourth Edition. Kendall/Hunt Publishing Co., Dubuque,
Iowa. 1158pp.
Merritt RW, Walker ED, Small PLC, Wallace JR, Johnson PDR, et al. (2010)
Ecology and Transmission of Buruli Ulcer Disease: A Systematic Review. PLoS
Negl Trop Dis 4(12): 1-15.
Meyers WM, Shelly TA, Conner DH, Meyers EK (1974) Human Mycobacterium
ulcerans infections developing at sites of trauma. American Journal Trop Med
Hyg 23: 919-923.
Meyers WM, Tignokpa WM, Priuli GB, Portaels F (1996) Mycobacterium ulcerans
infection (Buruli ulcer): first reported patients in Togo. British Journal of
Dermatology 134: 1116–1121.
Meyers WM, Tignokpa WM, Priuli GB, Portaels F (1996) Mycobacterium ulcerans
infection (Buruli ulcer): first reported patients in Togo. British Journal of
Dermatology 134: 1116–1121.
Mosi L, Williamson H, Wallace JR, Merritt RW, Small PLC (2008) Persistent
association of Mycobacterium ulcerans with West African predaceous insects of
the family Belostomatidae. Applied and Environmental Microbiology 74: 7036–
7042.
Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, et al. (2004) Buruli ulcer
disease in Cameroon rediscovered. Am J Trop Med Hyg 70: 520–526.
PASW Statistics v. 17.0 (SPSS: An IBM Company).
Petr T (1968) Population changes in aquatic invertebrates living on two water



101


plants in a tropical man-made lake. Hydrobiologia 32:449–85
Portaels, F (1989) Epidemiologie des ulcères à Mycobacterium ulcerans. Ann
Soc Belg Med Trop. 69: 91–103.
Portaels, F (1995) Epidemiology of mycobacterial diseases. Clin. Dermatol.
13:207-222.
Portaels F, Elsen P, Guimaraes-Peres A, Fonteyne P, Meyers WM (1999)
Insects in the transmission of Mycobacterium ulcerans infection. The Lancet 353:
986.
Portaels F, Chemlal K, Elsen P, Johnson PDR, Hayman JA, et al. (2001)
Mycobacterium ulcerans in wild animals. Rev sci tech Off int Epiz 20: 252–264.
Portaels F, Meyers WM, Ablordey A, Castro AG, Chemlal K, et al. (2008) First
Cultivation and Characterization of Mycobacterium ulcerans from the
Environment. PLoS Neglected Tropical Diseases 2: e178.
Raghunathan PL, Whitney EAS, Asamoa K, Stienstra Y, Taylor TH Jr., et al.
(2005) Risk factors for Buruli Ulcer disease (Mycobacterium ulcerans Infection):
results from a case-control study in Ghana. Clinical Infectious Diseases 40:1445–
1453.
Revill WDL, Barker DJP (1972) Seasonal distribution of mycobacterial skin
ulcers. Brit J prev soc Med 26: 23–27.
Seber GAF (1984) Multivariate observations. New York: John Wiley & Sons,
Inc. (Section 9.2.6).
Silva MT, Portaels F, Pedrosa J (2007) Aquatic Insects and Mycobacterium
ulcerans: an association relevant to Buruli ulcer control? PLoS Medicine 4: e63.
Sizaire V, F Nackers, E Comte, F Portaels (2006) Mycobacterium ulcerans
infection: control, diagnosis, and treatment. Lancet Infect Dis 6, 288–296.
Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, et al. (2007) Buruli
ulcer surveillance, Benin, 2003-2005. Emerg Infect Dis 13: 1374–1376.
Stout JR (1981) How abiotic factors affect the distribution of two species of
tropical predaceous aquatic bugs (Family: Naucoridae). Ecology. 62:1170–8.
Stout JR (1982) Effects of harsh environment on the life history patterns of two
species of tropical aquatic hemiptera (Family: Naucoridae). Ecology 63: 75–83.
Tabachnick BG, LS Fidel (2008) Using Multivariate Statistics (5th Edition). Allyn
and Bacon, Inc. Boston, MA USA.



102


Tobias N, Seemann T, Pidot S, Porter J, Marsollier L, et al. (2009) Mycolactone
gene expression is controlled by strong SigA-like promoters with
utility in studies of Mycobacterium ulcerans and buruli ulcer. PLoS Negl Trop Dis
3: e553.
U.S. Environmental Protection Agency (1996) The volunteer monitor's guide to
quality assurance project plans. U.S. Environmental Protection Agency, Office of
Wetlands, Oceans, and Watersheds, Washington, D.C. EPA 841-B-96-003.
U.S. Environmental Protection Agency (2001) Indicators for Monitoring Bilogical
Integrity of Inland Freshwater Wetlands. U.S. Environmental Protection Agency,
Office of Water, Wetlands Divison, Washington, D.C. EPA 843-R-01.
Wagner T, Benbow ME, Burns M, Johnson RC, Merritt R, et al. (2008a) A
Landscape-based Model for Predicting Mycobacterium ulcerans Infection (Buruli
Ulcer Disease) Presence in Benin, West Africa. EcoHealth 5: 69–79.
Wagner T, Benbow ME, Brenden T, Qi J, Johnson RC (2008b) Buruli ulcer
disease prevalence in Benin, West Africa: associations with land use/cover and
the identification of disease clusters. International Journal of Health Geographics
7: 25.
Wallace J, Gordon M, Hartsell L, Mosi L, Benbow M, et al. (2010) Interaction
of Mycobacterium ulcerans with mosquito species: Implications for transmission
and trophic relationships. Applied Environ Microbiol 76: 6215–6222.
Walsh D, Portaels F, Meyers W (2008) Buruli ulcer (Mycobacterium ulcerans
infection). Transactions of the Royal Society of Tropical Medicine and Hygiene
102: 969–978.
Walsh D, Eyase F, Onyango D, Odindo A, Waitumbi JN et al. (2010) Short
Report: Clinical and Molecular Evidence for a Case of Buruli Ulcer
(Mycobacterium ulcerans Infection) in Kenya. Am. J. Trop. Med. Hyg., 81(6), pp.
1110–1113.
Wansbrough-Jones M, Phillips R (2006) Buruli ulcer: emerging from obscurity.
Lancet 367: 1849–1858.
Williamson HR, Benbow ME, Nguyen KD, Beachboard DC, Kimbirauskas RK, et
al. (2008) Distribution of Mycobacterium ulcerans in Buruli Ulcer Endemic and
Non-Endemic Aquatic Sites in Ghana. PLoS Neglected Tropical Diseases 2:
e205.
WHO, ed (2000) Buruli ulcer. Mycobacterium ulcerans infection. Geneva,
Switzerland: WHO. 118 p.




103


WHO (2008) Buruli ulcer: progress report, 2004–2008. In: WHO, ed. Weekly
epidemiological record. Geneva, Switzerland: World Health Organization 83:
145–156.




104


CHAPTER 4

DETECTION OF NATURAL PREY IN THE GUTS OF AFRICAN CREEPING
WATER BUGS (HEMPTERA: NAUCORIDAE) USING SEQUENCED CLONES OF
PCR-AMPLIFIED GUT CONTENTS
Introduction
Buruli ulcer (BU) is a neglected emerging disease of skin and soft tissue that
leads to scarring and disability (Johnson et al. 2005, Merritt et al. 2005). It is
caused by Mycobacterium ulcerans, an environmental pathogen that produces a
destructive polyketide toxin (George et al. 1999). The disease has been reported in
humans from at least 32 countries, with a large number of cases reported from
West Africa (Duker et al. 2006; Walsh et al. 2008). While transmission of the
disease to human beings remains unclear, BU outbreaks have been associated
with freshwater habitats (Thangaraj et al. 1999), particularly in areas where the
landscape is disturbed by natural events such as flooding, or through deforestation,
dam construction, agricultural diversion, or mining (Thangaraj et al. 1999, Johnson
et al. 2005, Merritt et al. 2005, Duker et al. 2006, Wansbrough-Jones and Philips
2006).
A critical step in understanding BU transmission is elucidating the diet of
organisms that may potentially act as reservoirs and vectors of the Mycobacterium
pathogen in nature. Non-human mammals and reptiles have been tested in the
environment without positive findings for the pathogen (Radford 1974), and several
arthropod disease vectors (i.e., bedbugs, black flies, mosquitoes) tested negative
in early studies (Revill and Barker 1972, Portaels et al. 2001). However, only a few
organisms in each taxonomic group were tested in these early studies, and insect
105

sampling methods were neither systematically employed nor standardized as
discussed by Benbow et al. (2008). Portaels et al. (1999) were first to suggest that
aquatic bugs (Hemiptera) might be reservoirs of M. ulcerans in nature, and recently
they described the first isolation in pure culture of M. ulcerans from a water strider
(Hemiptera: Gerridae, Gerris sp.) from Benin, West Africa (Portaels et al. 2008). A
survey study (Portaels et al. 2001) based on detecting M. ulcerans DNA in aquatic
insects (Hemiptera, Odonata, Coleoptera) in African BU-endemic swamps
confirmed the earlier findings. More recent studies in Australia have suggested that
mosquitoes may be involved in transmission (Johnson et al. 2007).
Using experiments, Marsollier et al. (2002, 2003) demonstrated that M.
ulcerans could survive and multiply within the salivary glands of the aquatic bug
Naucoris cimicoides (Hemiptera, Naucoridae), and that N. cimicoides could
transmit the mycobacteria to mice (Marsollier et al. 2002). Naucoris spp. live in
freshwater ponds, lakes, and slow-flowing sections of streams and rivers.
Naucoridae are predacious in both the immature (nymph) and adult stages,
although little is known of the ecology and prey preferences of Naucoridae in
nature. This is particularly true in developing countries where Buruli ulcer is most
prevalent (WHO 2008). Most aquatic hemipterans are believed to be generalist
predators on other aquatic invertebrates (Merritt et al. 2008), although some,
including naucorid species, have mouthparts designed to aid in feeding on prey
larger then themselves (e.g., Cohen 1995) such as tadpoles (Polhemus and
Polhemus 1988) and larval fish (Louarn and Cloarec 1997).
Most Hemiptera feed by injecting digestive enzymes into prey and ingesting
the liquefied tissues through a tube-like proboscis (extra-oral digestion) (Cohen
106

1995). This feeding mode presents a challenge to the study of their diet, largely
eliminating the use of standard morphological identification of chitinous body parts
in the gut. As a result, studies of hemipteran diets have typically used
immunoassays employing prey-specific monoclonal antibodies (Greenstone 1996),
PCR tests (Sheppard and Harwood 2005), or both (e.g., Fournier et al. 2008). One
limitation of most antibody- and DNA-based applications is that some knowledge of
the potential prey is required. Antibodies target epitopes that are specific to
proteins from target prey species, and most DNA methods employ species- or
taxon-specific primers in PCR tests to determine the presence/absence of target
prey species or taxa (e.g., Agustí et al. 1999, 2000, 2003, Read 2002, Cuthbertson
et al. 2003, Jarman et al. 2004, de León et al. 2006). While these methods can be
powerful and have been verified for their accuracy using laboratory feeding
experiments (Chen et al. 2000, Foltan et al. 2005, Harper et al. 2005, 2006,
Sheppard et al. 2005, Harwood et al. 2007, McMillan et al. 2007), a major limitation
arises when there is little or no prior knowledge of the prey in their natural habitat.
Here I examined the diet of a common predator in freshwater ecosystems in
West Africa (Naucoris sp.) in a first attempt to understand its role in a tropical pond
food web and potential trophic connections in relation to the pathogen M. ulcerans.
In the absence of any a priori knowledge of their prey, I attempted to PCR-amplify
all DNA in the Naucoris sp. gut using universal primers, clone the PCR product,
and sequence a subset of clones. I then matched the resulting gut-content
sequences to sequences obtained from potential prey collected from the same
habitat, and to publically available sequence databases.

107

Materials and Methods
Sample collection and preparation. Naucoris sp. water bugs and
potential prey populations were sampled 9 August 2009 from one waterbody within
the village of Saduase, Ga East District, Ghana, Africa. All macroinvertebrates
were collected using a 500µm D-frame aquatic net. All Naucoris sp. were
transferred immediately to individual vials, while all other macroinvertebrates were
considered to be potential prey items and stored separately. All specimens were
preserved in 95% EtOH in the field. In the laboratory, Naucoris sp. were sexed
and guts were carefully removed under a dissecting microscope. To expose the
guts, heads were dissected and incisions were made laterally along the abdomen
to peel back the exoseleton. Guts were then removed with forceps and stored
separately in fresh 95% EtOH. Prior to handling each specimen, all instruments
were rinsed with distilled water, flame treated, and wiped with individual Kimwipes.
All samples were stored at 4°C prior to DNA extraction.
DNA extraction and PCR. Genomic DNA was extracted from Naucoris sp.
adults (n = 29) and nymphs (n = 14) as well as the potential prey sampled:
Ephemeroptera (mayflies, n = 4), Odonata (damselflies, Zygoptera) (n = 8),
Coleoptera (beetles, n = 4), Diptera (flies, Chironomidae, n = 3), and Arachnidae
(spiders, n = 3) using DNeasy tissue kits (Qiagen GmbH, Hilden, Germany).
Genomic DNA was extracted from Naucoris sp. guts (n = 60) using the QIAamp
DNA Stool Mini Kit (Qiagen GmbH, Hilden, Germany) (King et al. 2008). Field
samples were first centrifuged for 1 min at 2000 g. The ethanol was poured off and
the dry weight of the pellet was determined. All remaining steps followed the
manufacturer’s protocol, except that only half the recommended volume of
108

buffers/InhibitEX was used. Primers LR-N-13389 (alias 16ar, Simon et al. 1994)
and 16b2 (5’ - TTTAATCCAACATCGAGG - 3’) were used to amplify a ca. 440-bp
fragment of mitochondrial rrnL (16S) for all samples using standard methods. The
5’ (DNA barcode) region of cox1 (COI) was amplified for four Naucoris sp. adults
using primers LCO-1490 and HCO-2198 (Folmer et al. 2004) in order to potentially
match individuals with existing databases. PCR products were purified using the
QIAquick PCR Purification Kit (Qiagen GmbH, Hilden, Germany) and sequenced in
both directions using the PCR primers. Samples were analyzed on either a CEQ
8000 (Beckman Coulter) or a 3500xL (Applied Biosystems) automated sequencer.
Molecular cloning. Cloning was used to differentiate among multiple
possible PCR products obtained from Naucoris sp. guts. Both the rrnL and cox1
PCR primers (above) target a broad range of organisms including crustaceans,
insects, and vertebrates, thus could be useful for simultaneously amplifying
multiple taxa that may be present in the gut. PCR products were run out on a 2 %
agarose gel and purified using the MinElute gel extraction kit (Qiagen GmbH,
Hilden, Germany). The clone libraries were created with the pGEM-T-Easy-kit
(Promega GmbH, Mannheim, Germany) following the manufacturer’s protocol.
Insert size was examined using PCR of plasmids with the primers SP6 (5’ATTTAGGTACACTATAG) and T7 (5’-AATACGACTCACTATAGG). Large inserts
(n = 576) were cleaned with PEG and sequenced using primer SP6.
Data analysis. All sequences were assembled and edited using
CodonCode Aligner v 3.5 (Codon Code Corporation, Dedham, USA). For Naucoris
sp. and prey, forward and reverse sequences were assembled and edited for each
specimen. Naucoris sp. cox1 and rrnL sequences were first compared to the NCBI
109

nucleotide database using blastn queries (http://blast.ncbi.nlm.nih.gov). Clone
sequences were assembled and contigs were edited in order to generate
consensus sequences for each contig (contig sizes on Table 4.1). All full-length
sequences obtained from clone libraries (ca. 450-478 bp rrnL, 658 bp cox1) were
compared to the NCBI nucleotide database using blastn queries.
Phylogenetic analysis was conducted on all insect and arachnid rrnL
sequences that were newly generated by direct sequencing or cloning. I also
included 18 other Hemiptera rrnL sequences obtained from the list of blastn hits,
most of which came from a recent mtDNA phylogeny (Hua et al. 2009). For
comparison of potential prey sequences with the NCBI database, I downloaded all
blastn hits with > 90 % identity to each query genotype. All sequences were
aligned using clustalW (align.genome.jp) and a phylogenetic tree search was
conducted on the matrix using a maximum likelihood approach in PhyML v 3.0
(Guindon and Gascuel 2003) under a GTR model of evolution (as determined by
Modeltest v 3.7, Posada and Crandall 1998).
Results
PCR amplification of cox1 and rrnL was equally successful for Naucoris sp.
but rrnL was more consistently amplified for gut samples and potential prey. There
were no Naucoris sp. cox1 sequences available on the NCBI nucleotide database
and the top hit of the blastn query was an unclassified Hemiptera (AAG5301
voucher ENT-OUBS-156, HM381306), whereas the database contained 8 rrnL
sequences for Naucoridae collected from Madagascar, Europe, North and Central
America, and the Philippines (Hebsgaard et al. 2004). Combining the newly
generated rrnL sequences with blastn query results produced an aligned matrix of
110

62 taxa and 460 characters (sequence length 336 – 443 bp). In the maximum
likelihood rrnL gene tree (ln L = -8123.81514; Fig. 4.1) our Naucoris sp. sequences
were clearly nested within published data for the Hemiptera, the closest relative
being Ambrysus sp. collected from North America (Fig. 4.1). Ambrysus sp. was the
second-ranked blastn hit, with Macrocoris sp. from Madagascar (Hebsgaard et al.
2004) the top hit but phylogenetically more distant in our analysis of the same
sequences (Fig. 4.1).
Three of our cloned rrnL sequences were identical to a field-caught
Coleoptera species sampled at the study site as potential prey. The top database
match (using blastn) to this sequence was Spercheus (Spercheidae) (Table 4.1)
and the second-ranked match was Hydrobius sp. (Hydrophilidae)(not shown), but it
is clear from our phylogenetic search that it is distantly related to both (Fig. 4.1).
None of the other potential prey species that we collected from the sampling site
and sequenced were recovered from gut sequence clones (Fig. 4.1). Nonetheless,
a number of interesting non-insect species were recovered from guts and identified
with blastn queries (Table 4.1). These included Afrixalus sp. (Anura: Hyperoliidae),
a sub-Saharan genus of frog for which one rrnL sequence was recovered from the
clone library. Sequences from the cox1 clone library included Embata and
Floscularia (Rotifera), and Pythium (Oomycete fungi). All other full-length clone
sequences were identical to our Naucoridae sequences obtained using direct
sequencing of PCR products (rrnL shown in Fig. 4.1).
Discussion
Buruli ulcer (Mycobacterium ulcerans infection) causes severe morbidity in
human populations associated with degraded freshwater habitats, but neither the
111

reservoir nor the mode of transmission of M. ulcerans is known (Merritt et al. 2005).
Here I investigated an abundant aquatic predator from a BU-endemic area in
Ghana, Naucoris sp. water bugs (Hemiptera, Naucoridae). While Naucorids have
been implicated in the transmission of M. ulcerans in laboratory studies, a limited
knowledge of their place in aquatic food webs in nature makes it difficult assess the
potential source and sinks of the pathogen. This is the first investigation of which I
am aware to clone and sequence PCR products from universal primers to
determine a Hemipteran diet without any prior knowledge of potential prey.
Our approach led to a broader perspective of the role of Naucoris sp. in the
aquatic food web. Using a standard PCR-based method, prey-specific primers for
the 5 taxa of field-caught prey would have designed (e.g., Agusti et al. 2003). From
this, we would have probably generated positive tests for the Coleoptera. In
contrast, the universal primers, sequenced clones, and database queries used
here allowed us to identify DNA in the guts that came from taxa that were not fieldcollected. These included fungi, rotifers, and an anuran, although immature
anurans were collected from the field site and thus known to be present. A
limitation of the database queries is clearly the fact that the extent of the database
plays an important role. Using the newly generated Naucoris sp. sequences, none
of the blastn query results gave a close match. Even the barcode cox1 sequence
gave a fairly meaningless match (Hemiptera sp.) to the Barcode of Life database
(www.barcodinglife.org).
Combining public databases and our own newly generated prey sequences
was beneficial for confirming that the prey Coleoptera sequence cloned from the
gut matched the field-caught Coleoptera species at the same habitat. Although
112

neither the database nor our new sequence could provide identification, at least I
could conclude with confidence that Naucoris sp. prey on the Coleoptera resident
to the sample site in Ghana, West Africa. These sequences were identical and,
based on the phylogenetic gene tree, quite different from any species in GenBank.
Interestingly, the blastn query and phylogenetic analysis gave different results.
The second-ranked blastn result was phylogenetically closer to the query
sequence than the top-ranked blastn query result. The more detailed phylogenetic
analysis, using more of the available data and a GTR model of sequence evolution,
probably revealed the closer relative. Although both query hits were relatively
distant and neither is probably a good match, it does suggest that a phylogenetic
approach is more accurate than a blast result in the absence of a complete
database.
In conclusion, this approach provided the means to study an aquatic
hemipteran diet without any prior knowledge of potential prey and despite the
difficulties of extra-oral digestion. Naucoris sp. in this area of West Africa feed on a
wide range of prey and body sizes, including rotifers, insects, and anurans. Further
work on M. ulcerans transmission will be aided by this food web information. These
results also suggest the approach could be successfully used to study the complex
interactions within aquatic food webs, including even feeding on fungi. These
results corroborate previous suggestions that DNA-based approaches using
universal primers and cloning provide an important tool for studying the prey
spectrum of predators with unknown diets.





113

Table 4.1. Results of comparisons to the NCBI nucleotide database using blastn queries of rrnL and
cox1sequences that were PCR-amplified from Naucoridae guts and cloned (see methods). Taxa listed were,
in each case, first on the hit table. Only full-length inserts (450 – 478 bp rrnL, 658 bp cox1) were considered.
Most full-length inserts were identical to our sequences obtained from direct sequencing of sampled
Naucoridae and resulted in a top blastn hit of Macrocoris sp. (rrnL) or Hemiptera sp. (cox1) (data not
shown).
gene

Taxon

Accession

n

% identity

region
rrnL

query length /

e-value

Source

alignment length
Spercheus emarginatus

AM287063

3

85.59

450 / 444

4 E -135

(Coleoptera)

Bernhard et al.
(2006)

Afrixalus sp.(Anura)
cox1

AF215431

1

98.47

478 / 457

0

Vences (2000)

Embata parasitica

EF650597

3

88.87

658 / 602

0

Unpublished A

EU499896

2

75.99

658 / 633

1 E -124

Unpublished B

EU350529

4

80.18

658 / 661

0

Unpublished C



















(Rotifera)
Floscularia melicerta
(Rotifera)
Pythium acanthophoron
(Oomycetes)






A- Herniou, E.A. and Fontaneto, D. (unpubl.)
B- Fontaneto, D., Chen, K. and Castillo, K. (unpubl.)
C- Jackson, C. A. R., de Cock, A. W. A. M., Vijayan, P., Robideau, G. P. and Levesque, C. A. (unpubl.)




114











Figure 4.1. Maximum likelihood phylogenetic tree of rrnL using a GTR model of evolution, including newly sequenced
Naucoris sp. and potential prey (blue terminals), cloned PCR products from Naucoris sp. guts and mouthparts (red
terminals), and highly ranked sequences according to blastn queries (black, see text for criteria).




115


Figure 4.1 (cont’d).











116


Figure 4.1 (cont’d).


























































117


LITERATURE CITED

118

Literature Cited
Agustí N, Unruh TR, Welter SC (2003) Detecting Cacopsylla pyricola (Hemiptera:
Psyllidae) in predator guts using COI mitochondrial markers. Bulletin of
Entomological Research, 93, 179–185.
Agustí N, de Vicente MC, Gabarra R (1999) Development of sequence amplified
characterized region (SCAR) markers of Helicoverpa armigera: a new polymerase
chain reaction-based technique for predator gut analysis. Molecular ecology, 8,
1467-1474.
Agustí N, de Vicente MC, Gabarra R (2000) Developing scar markers to study
predation on Trialeurodes vaporariorum. Insect Molecular Biology, 9, 263–268.
Benbow M, Williamson H, Kimbirauskus R, McIntosh M, Kolar R, et al. (2008)
Aquatic invertebrates as unlikely vectors of Buruli ulcer disease. Emerg Infect Dis
14: 1247–1254.
Bernhard D, Schmidt C, Korte A, Fritzsch G, Beutel R G (2006) From terrestrial
to aquatic habitats and back again – molecular insights into the evolution and
phylogeny of Hydrophiloidea (Coleoptera) using multigene analyses. Zoologica
Scripta, 35, 597-606.
Chen Y, Giles KL, Payton ME, Greenstone MH (2000) Identifying key cereal
aphid predators by molecular gut analysis. Molecular Ecology, 9, 1887–1898.
Cohen AC (1995) Extra-oral digestion in predaceous terrestrial Arthropoda.
Annual Review of Entomology, 40, 85-103.
Cuthbertson AGS, Fleming CC, Murchie AK (2003) Detection of Rhopalosiphum
insertum (apple-grass aphid) predation by the predatory mite Anystis baccarum
using molecular gut analysis. Agricultural and Forest Entomology, 5, 219–225.
de León JH, Fournier V, Hagler JR, Daane KM (2006) Development of molecular
diagnostic markers for sharpshooters Homalodisca coagulata and Homalodisca
liturata for use in predator gut content examinations. Entomologia Experimentalis
et Applicata, 119, 109–119.
Duker AA, Portaels F, Hale M (2006) Pathways of Mycobacterium ulcerans
infection: A review. Environment International, 32, 567-573.
Foltan P, Sheppard S, Konvicka M, Symondson WOC (2005) The significance of
facultative scavenging in generalist predator nutrition: detecting decayed prey in
the guts of predators using PCR. Molecular Ecology, 14, 4147–4158.

119

Fournier V, Hagler J, Daane K, de León J, Groves R (2008) Identifying the
predator complex of Homalodisca vitripennis (Hemiptera: Cicadellidae): a
comparative study of the efficacy of an ELISA and PCR gut content assay.
Oecologia, 157, 629-40.
Greenstone MH (1996) Serological analysis of arthropod predation: past, present
and future. In: The Ecology of Agricultural Pests: Biochemical Approaches (eds
Symondson WOC, Liddell JE), pp. 265–300. Chapman & Hall, London.
George KM, Chatterjee D, Gunawardana G et al. (1999) Mycolactone: a
polyketide toxin from Mycobacterium ulcerans required for virulence. Science,
283, 854–7.
Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate
large phylogenies by maximum likelihood. Systematic Biology, 52(5), 696-704.
Harper GL, King RA, Dodd CS et al. (2005) Rapid screening of invertebrate
predators for multiple prey DNA targets. Molecular Ecology, 14, 819–827.
Harper GL, Sheppard SK, Harwood JD et al. (2006) Evaluation of temperature
gradient gel electrophoresis for the analysis of prey DNA within the guts of
invertebrate predators. Bulletin of Entomological Research, 96, 295–304.
Harwood JD, Desneux N, Yoo HJS et al. (2007) Tracking the role of alternative
prey in soybean aphid predation by Orius insidiosus: a molecular approach.
Molecular Ecology, 16, 4390–4400.
Hebsgaard MB, Andersen NM, Damgaard J (2004) Phylogeny of the true water
bugs (Nepomorpha: Hemiptera-Heteroptera) based on 16S and 28S rDNA and
morphology. Systematic Entomology, 29, 488-508.
Jarman SN, Deagle BE, Gales NJ (2004) Group-specific polymerase chain
reaction for DNA-based analysis of species diversity and identity in dietary
samples. Molecular Ecology, 13, 1313–1322.
Johnson PD, Stinear TP, Small PL et al. (2005) Buruli ulcer (Mycobacterium
ulcerans): new insights, new hope for disease control. PLoS Med, 2, e108.
Johnson PDR, Azuolas J, Lavender CJ et al. (2007) Mycobacterium ulcerans in
mosquitoes captured during outbreak of Buruli ulcer, Southeastern Australia.
Emerging Infectious Diseases, 13, 1653-1660.
King RA, Read DS, Traugott M, Symondson WOC (2008) Molecular analysis of
predation: a review of best practice for DNA-based approaches. Molecular
ecology, 17, 947-963.
120

Louarn H, Cloarec A (1997) Insect predation on pike fry. Journal of Fish Biology,
50(2), 366-370. doi: 10.1111/j.1095-8649.1997.tb01364.x.)
McMillan S, Kuusk AK, Cassel-Lundhagen A, Ekbom B (2007) The influence of
time and temperature on molecular gut content analysis: Adalia bipunctata fed
with Rhopalosiphum padi. Insect Science, 14, 353–358.
Marsollier L, Aubry J, Saint-Andre JP et al. (2003) Ecology and transmission of
Mycobacterium ulcerans. Pathologie Biologie, 51, 490-495.
Marsollier L, Robert R, Aubry J et al. (2002) Aquatic insects as a vector for
Mycobacterium ulcerans. Applied and Environmental Microbiology, 68, 46234628.
Merritt RW, Cummins KW, Berg MB (2008) An introduction to the aquatic insects
of North America. 4th ed. Kendal/Hunt Pub. Co., Duduque, IA. 1158 p.
Merritt R, Benbow M, Small P (2005) Unraveling an emerging disease associated
with disturbed aquatic environments: the case of Buruli ulcer. Frontiers in
Ecology and the Environment, 3, 323-331.
Polhemus DA, Polhemus JT (1988) The Aphelocheirinae of Tropical Asia
(Heteroptera: Naucoridae). Raffles Bulletin Zoology, 36, 167-300.
Posada D, Crandall KA (1998) Modeltest: testing the model of DNA substitution.
Bioinformatics 14: 817-818.
Portaels F, Chemlal K, Elsen P et al. (2001) Mycobacterium ulcerans in wild
animals. Revue Scientifique et Technique (International Office of Epizootics), 20,
252-264.
Portaels F, Elsen P, Guimaraes-Peres A, Fonteyne P, Meyers WM (1999)
Insects in the transmission of Mycobacterium ulcerans infection. The Lancet,
353, 986.
Portaels F, Meyers WM, Ablordey A, et al. (2008) First cultivation and
characterization of Mycobacterium ulcerans from the environment. PLoS
Neglected Tropical Diseases, 2, e178.
Radford AJ (1974) Mycobacterium ulcerans infection in Papua New Guinea.
Papua and New Guinea Medical Journal, 17, 145-149.
Read DS (2002) Sequencing of aphid DNA and primer design for the detection of
aphid remains in predators. Undergraduate Dissertation, Cardiff University.
Cardiff, UK.

121

Revill WDL, Barker DJP (1972) Seasonal distribution of mycobacterial skin
ulcers. British Journal of Preventive & Social Medicine, 26, 23-27
Sheppard SK, Harwood JD (2005) Advances in molecular ecology: tracking
trophic links through predator-prey food-webs. Functional ecology, 19, 751-762.
Sheppard SK, Bell J, Sunderland KD, Fenlon J, Skervin D, Symondoson WOC
(2005) Detection of secondary predation by PCR analyses of the gut contents of
invertebrate generalist predators. Molecular Ecology, 14, 4461–4468.
Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook PC (1994) Evolution
weighting and phylogenetic utility of mitochondrial gene sequences and
compilation of conserved polymerase chain reaction primers. Annals of the
Entomological Society of America, 87, 651- 700.
Swanstrom J, Chen K, Castillo K, Barraclough TG, Fontaneto D (in press)
Testing for evidence of inefficient selection in bdelloid rotifers: do sample size
and heterogeneity matter? Hydrobiologia
Thangaraj HS, Evans MRW, Wansbrough-Jones MH (1999) Mycobacterium
ulcerans: Buruli ulcer. Transactions of the Royal Society of Tropical Medicine and
Hygiene, 93, 337-340.
Vences M (2000) Phylogenetic studies of ranoid frogs (Amphibia:Anura), with a
discussion of the origin and evolution of the vertebrate clades of Madagascar.
PhD Thesis Universitaet Bonn.
Walsh D, Portaels F, Meyers W (2008) Buruli ulcer (Mycobacterium ulcerans
infection). Transactions of the Royal Society of Tropical Medicine and Hygiene,
102, 969-978.
Wansbrough-Jones M, Phillips R (2006) Buruli ulcer: emerging from obscurity.
Lancet, 367, 1849-1858.


WHO (2008) Buruli ulcer: progress report, 2004–2008. In: WHO, ed. Weekly
epidemiological record. Geneva, Switzerland: World Health Organization 83:
145–156.





122