murdoch research repository · page 2 of 43 accepted manuscript 2 20 abstract 21 trypanosoma...
TRANSCRIPT
MURDOCH RESEARCH REPOSITORY
This is the author’s final version of the work, as accepted for publication following peer review but without the publisher’s layout or pagination.
The definitive version is available at http://dx.doi.org/10.1016/j.vetpar.2011.12.010
McInnes, L.M., Dargantes, A.P., Ryan, U.M. and Reid, S.A. (2012) Microsatellite typing and population structuring of Trypanosoma
evansi in Mindanao, Philippines. Veterinary Parasitology, 187 (1-2). pp. 129-139.
http://researchrepository.murdoch.edu.au/8678/
Copyright: © 2012 Elsevier B.V..
It is posted here for your personal use. No further distribution is permitted.
MURDOCH RESEARCH REPOSITORY
This is the author’s final version of the work, as accepted for publication following peer review but without the publisher’s layout or pagination.
The definitive version is available at http://dx.doi.org/10.1016/j.vetpar.2012.01.007
McInnes, L.M., Dargantes, A.P., Ryan, U.M. and Reid, S.A. (2012) Microsatellite typing and population structuring of Trypanosoma
evansi in Mindanao, Philippines. Veterinary Parasitology, 187 (1-2). pp. 129-139.
http://researchrepository.murdoch.edu.au/8678/
Copyright: © 2012 Elsevier B.V..
It is posted here for your personal use. No further distribution is permitted.
Accepted Manuscript
Title: Microsatellite typing and population structuring ofTrypanosoma evansi in Mindanao, Philippines
Authors: L.M. McInnes, A.P. Dargantes, U.M. Ryan, S.A.Reid
PII: S0304-4017(11)00848-XDOI: doi:10.1016/j.vetpar.2011.12.010Reference: VETPAR 6183
To appear in: Veterinary Parasitology
Received date: 3-9-2011Revised date: 3-12-2011Accepted date: 13-12-2011
Please cite this article as: McInnes, L.M., Dargantes, A.P., Ryan, U.M., Reid, S.A.,Microsatellite typing and population structuring of Trypanosoma evansi in Mindanao,Philippines, Veterinary Parasitology (2010), doi:10.1016/j.vetpar.2011.12.010
This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.
Page 1 of 43
Accep
ted
Man
uscr
ipt
1
Microsatellite typing and population structuring of Trypanosoma evansi in Mindanao, 1
Philippines
2
3
McInnes, L.M.1*
, Dargantes, A.P.1,2
, Ryan, U.M. 1
and Reid, S.A.3
4
1Division of Health Sciences, School of Veterinary and Biomedical Sciences, Murdoch 5
University, Murdoch, Perth WA 6150 6
2College of Veterinary Medicine, Central Mindanao University, University Town, Musuan, 7
Bukidnon, Philippines 8710 8
3Public Health Division, Secretariat of the Pacific Community, B.P. D5, 98848 Noumea cedex, 9
New Caledonia 10
11
* Author for correspondence: 12
Linda M. McInnes 13
Division of Health Sciences, School of Veterinary and Biomedical Sciences, Murdoch 14
University, Murdoch, Perth WA 6150 15
Phone: +61893602495 16
FAX: +61893104144 17
Email: [email protected] 18
19
*Manuscript
Page 2 of 43
Accep
ted
Man
uscr
ipt
2
ABSTRACT 20
Trypanosoma evansi, a blood-borne protozoan parasite with an extensive geographical range is 21
the causative agent of the livestock disease known as surra. A total of 140 out of 179 T. evansi 22
isolates collected between 2006 and 2007 from 44 villages (comprising of 16 reported surra 23
outbreaks) in 3 provinces (Agusan del Sur (ADS), Surigao del Sur (SDS) and Agusan del Norte 24
(ADN)) in Mindanao, Philippines were each successfully genotyped using a suite of 7 25
polymorphic microsatellites. The study identified 16 multi locus genotypes (MLG) within the T. 26
evansi isolates and evidence of the spread of surra outbreaks from one village to another, most 27
likely due to the movement of infected animals. Genotyping provided evidence of population 28
sub-structuring with 3 populations (I, II and III (only 1 isolate)) identified. The most abundant 29
population was II, which was the predominant population in ADS and SDS (p=0.022). In 30
addition, buffalo mortality was statistically higher in outbreak areas associated with isolates from 31
population I (13.6%) than with isolates from population II (6.9%) (p=0.047). The present study 32
has highlighted the utility of microsatellite loci to improve understanding of the epidemiology of 33
T. evansi and in tracking surra outbreaks. 34
35
KEYWORDS 36
Trypanosoma evansi, surra, epidemic, microsatellites, DNA, Mindanao, Philippines. 37
38
Page 3 of 43
Accep
ted
Man
uscr
ipt
3
1. Introduction 39
Trypanosoma evansi is endemic in Africa, the Middle East, Central and South America, and 40
many parts of Asia including the Philippines (OIE, 2008). Trypanosoma evansi, although able to 41
infect most classes of mammals, is principally a livestock disease (surra), causing considerable 42
mortality and morbidity in bovines, equines and camelids worldwide. In the Philippines, surra 43
outbreaks have been reported in all 13 regions and resultant mortalities in horses, buffaloes and 44
cattle between 1989 and 1997 have been estimated at a cost of US$1.15 million per annum 45
(Manuel, 1998). There has recently been a large number of severe outbreaks in the Philippines 46
with an outbreak in the North Samar province reported to have caused buffalo mortalities in 47
excess of a thousand over a 6-month period (Cresencio et al., 1994; Manuel, 1998). 48
Surra was first introduced into the Philippines in the early 19th
century with the 49
importation of infected cavalry horses from China by the American forces during the American-50
Spanish war (Manuel, 1998). The initial and successive surra cases which occurred in the 51
Philippines were predominately reported as high mortalities of equines (Manresa, 1935; Randall 52
and Schwartz, 1936; Manuel, 1998). Mindanao, the second largest and southernmost island in 53
the Philippine archipelago appeared to be free of surra until 1989 when the first reported cases of 54
surra (38% equine and bovine mortality) were recorded in the southern Mindanao province of 55
Sarangani (Mercado, R. pers. comm., 2010). Subsequent to the Sarangani outbreak, further surra 56
outbreaks have occurred in a range of Mindanao provinces (Fig. 1). At present, the economic 57
losses due to surra in Mindanao have been estimated at US$1.58 million per year from every 10 58
villages where the disease is moderately to highly prevalent (Dobson et al., 2009). 59
Surra presents as a highly variable clinical disease with the variation in disease 60
manifestation (acute, sub-acute or chronic) associated with differences in host susceptibility 61
Page 4 of 43
Accep
ted
Man
uscr
ipt
4
and/or parasite strain virulence (OIE, 2010). An important feature of T. evansi infections is the 62
sporadic nature of surra outbreaks which are often localised. High mortalities are most often 63
attributed to the movement of infected animals into areas previously free from infection or the 64
movement of surra-free animals into endemic areas (Payne et al., 1988). The endemic nature of 65
T. evansi in the absence of high mortality outbreaks has led to speculation that surra may be 66
enzootically stable in some areas such as Indonesia (Payne et al., 1991a; Payne et al., 1991b). 67
Microsatellites, which are highly discriminatory genetic tools have been used to examine 68
T. cruzi phylogeography (Llewellyn et al., 2009a; Llewellyn et al., 2009b), differences in T. cruzi 69
clinical disease (D'Avila et al., 2009), T. brucei gambiense population structuring (Koffi et al., 70
2009; Simo et al., 2010), T. brucei mixed strain infection investigations (Balmer and Caccone, 71
2008), mating in T. congolense (Morrison et al., 2009) and T. evansi population structuring in 72
Kenya (Njiru and Constantine, 2007; Salim et al., 2011). As T. evansi is widely believed to have 73
evolved from T. brucei which was spread beyond the tsetse fly belt (Hoare, 1972). The genetic 74
similarity between T. brucei and T. evansi has led others to propose that T. evansi should be 75
considered a subspecies of T. brucei rather than a species (Lai et al., 2008). This genetic 76
similarity between T. brucei and T. evansi means the T. brucei genome is highly useful for 77
development of T. evansi genetic tools. In the present study, microsatellite markers developed 78
from T. brucei genetic information were used to characterise the isolates and parasite populations 79
from barangays (the Philippines‟ equivalent of a village) without surra outbreaks and those 80
isolates involved in the late 2005 to 2007 T. evansi outbreaks in 16 barangays in Mindanao, 81
Philippines and their possible association with clinical disease, patent parasitaemia and temporal 82
and spatial sub-structuring. 83
84
Page 5 of 43
Accep
ted
Man
uscr
ipt
5
2. Materials and methods 85
2.1. Sources of isolates and sample collection 86
A total of 179 T. evansi isolates were collected from infected livestock (mostly buffaloes) 87
from 44 barangays (with or without surra outbreaks) in the provinces of Agusan del Sur (ADS), 88
Surigao del Sur (SDS) and Agusan del Norte (ADN), as shown in Figures 2a and b, in Mindanao, 89
southern Philippines during surra epidemiological surveys in 2006–2007. The field surveys were 90
conducted in a total of 89 barangays from 36 municipalities in 5 provinces in Mindanao 91
involving more than 3,000 village livestock (Dargantes, 2010). From 16 surra outbreak 92
barangays (Table 1), 660 local swamp buffaloes (Bubalis bubalis), 24 cattle, 50 horses and 33 93
goats were randomly bled. Whole blood (6-9 mL) was aseptically collected from the jugular vein 94
of livestock into a labelled ethylenediamine-tetra acetic acid (EDTA) vacutainer vial (Vacuette®, 95
Greiner Bio-One, USA). Blood samples were placed inside a cooler box until examination. The 96
microhaematocrit centrifugation technique (MHCT) and mouse inoculation test (MIT) were used 97
to detect T. evansi in blood samples as described in Dargantes et al. (2009). Mice parasitaemic 98
with T. evansi were bled and the blood was stored in EDTA vials at -20ºC until required. 99
Information on each animal sampled including geographical origin (barangay, municipality and 100
province), MHCT parasite enumeration, card agglutination test for Trypanosoma evansi (CATT), 101
host species, blood packed cell volume (PCV), age and body condition was recorded. Data on the 102
percentage mortality of buffaloes, cattle, horses and goats in conjunction with the outbreaks of 103
surra were obtained from the municipal veterinary personnel or local barangay officials. All field 104
and laboratory activities were conducted with approval from the Murdoch University Animal and 105
Human Ethics Committees (Permit numbers R881/01 and 2007/034, respectively). 106
107
Page 6 of 43
Accep
ted
Man
uscr
ipt
6
2.2. DNA extraction 108
Whole genomic DNA was extracted from blood collected from mice inoculated with the 109
blood of a T. evansi-infected original host (M) and from the whole blood from the corresponding 110
original host (OH) using a modified protocol with a MasterPure™ DNA Purification Kit 111
(EPICENTRE® Biotechnologies, Madison, Wisconsin, U.S.A.) as described by McInnes et al. 112
(2011). DNA was stored at -20ºC until use. 113
114
2.3. Microsatellite typing tool development 115
Microsatellites were identified from T. brucei genome sequences sourced from GenBank 116
by screening for repeating DNA motifs utilising the software program Tandem Repeats Finder 117
(Benson, 1999). Primers were designed in flanking regions of selected microsatellites chosen on 118
the basis of perfect repeating motifs, originating from different contigs and where possible from 119
different chromosomes to minimise the chance of linkage. An initial 15 microsatellites were 120
chosen for analysis (11 dinucleotide microsatellites and 4 trinucleotide microsatellites) and 121
primers were designed with the aid of software program Amplify v3.1. Primers were tested on 3 122
different T. evansi isolates (collected in different years from livestock in Central, and East Java 123
and Sumbawa, Indonesia) previously shown to be genetically different (C. Constantine, pers. 124
comm., 2007) and the resultant amplicons electrophoresed on 4% High resolution agarose 125
(Fisher Biotec, Australia) to ascertain the best candidates for further analysis. Promising 126
microsatellites, which appeared to display some degree of heterogeneity amongst the three 127
isolates, were chosen for fluorescent labelling and multiplexing. Of the 15 microsatellites, only 4 128
were deemed useful after evaluation was completed on 20 T. evansi isolates (11 loci were 129
discarded due to monomorphism, unscoreable results and poor PCR amplification). Chosen loci 130
Page 7 of 43
Accep
ted
Man
uscr
ipt
7
Tbb1 and Tbb10 were combined in one multiplex (TEM1) and Tbb5 and Tbb9 in a second 131
multiplex (TEM2) with fluorescent labels as indicated in Table 2. 132
All designed primers were tested on the DNA extracted from the blood of a mouse 133
experimentally infected with a T. evansi isolate using the following PCR conditions: 94ºC for 5 134
min followed by 35 cycles of 94ºC for 30 sec, 56ºC for 20 sec and 72ºC for 30 sec, finishing with 135
a final extension of 72ºC for 45 min. All multiplex PCRs were carried out in 25 µL reactions 136
comprising of 1 x PCR buffer, 1.5 mM MgCl2, 0.1 mM of each dNTPs, 0.4 µM of each primer, 137
0.04 units of Tth+ DNA polymerase (Fisher Biotec, Australia) and 1 µL DNA template. All PCR 138
amplicons generated from this T. evansi isolate were purified using a MO BIO UltraCleanTM 15 139
DNA Purification Kit (MO BIO Laboratories Inc. West Carlsbad, California, USA) and 140
sequenced using an ABI PrismTM
Terminator Cycle Sequencing kit (Applied Biosystems, Foster 141
City, California, USA) on an Applied Biosystem 3730 DNA Analyzer to confirm PCR 142
amplification of correct loci. 143
144
2.4. Microsatellite typing of samples 145
DNA extracted from the Philippine T. evansi samples (both M and OH samples) were 146
amplified using the two developed multiplexes (TEM1 and TEM2) following conditions 147
described above as well as with microsatellite primers designed by Biteau et al. (2000) for the 148
following loci: MORF-CA, MT3033-AC/TC, M6C8-CA, MT3033-TA and MEST19-AT/GT. 149
The five loci identified by Biteau et al. (2000) were multiplexed to conserve resources. Loci 150
MORF-CA and MT3033-AC/TC were combined in one multiplex (Bite1) with MORF-CA 151
fluorescently labelled with 6-FAM (Applied Biosystems, Life Technologies, Carlsbad, 152
California) and MT3033-AC/TC with Vic (Applied Biosystems). Loci M6C8-CA, MT3033-TA 153
Page 8 of 43
Accep
ted
Man
uscr
ipt
8
and MEST19-AT/GT were combined in a second multiplex (Bite2) with the forward primers for 154
M6C8-CA labelled with NED (Applied Biosystems), MT3033-TA with Vic (Applied 155
Biosystems) and MEST19-AT/GT with 6-FAM (Applied Biosystems). Conditions for the Bite1 156
and Bite2 multiplexes were as follows: 94ºC for 5 min followed by 30 cycles of 94ºC for 30 sec, 157
53ºC for Bite1 and 50ºC for Bite2 for 30 sec and 72ºC for 30 sec, and a final extension of 72ºC 158
for 45 min. The PCR reaction reagents for both Bite1 and Bite2 were the same as previously 159
described for the TEM1 and TEM2 multiplexes. All fluorescently labelled PCR amplicons were 160
analysed on an ABI 3730 DNA Analyzer (Applied Biosystems Inc.) in conjunction with a 161
GeneScan™ 600 LIZ® Size Standard (Applied Biosystems). Allele identification was performed 162
manually using the ABI program Genemapper v3.5 (Applied Biosystems). The term multilocus 163
genotype (MLG) is applied in this paper to isolates with identical genotype established from 164
several genetic loci. 165
166
2.5. Population analysis 167
Population structuring analyses of datasets using the 4 (developed as part of the present 168
study) and 7 loci (the 4 developed loci and 3 loci published by Biteau et al. (2000) were 169
conducted using the software program Structure 2.3.3, which is a Bayesian model-based 170
clustering method for inferring population structure and assign individuals to populations from 171
multilocus genotype data (Pritchard et al., 2000). Simulations using the admixture and correlated 172
frequencies model were run for K=1-7 with a burn in period of 20,000 and 100,000 MCMC 173
repetitions and the average log probability of data clustering (L(K)) from 15 replicates 174
calculated. 175
Page 9 of 43
Accep
ted
Man
uscr
ipt
9
The construction of Neighbour-Joining (NJ) phylogenetic trees of microsatellite data 176
from T. evansi isolates were carried out using Powermarker 3.25 (Liu and Muse, 2005), and the 177
trees visualised using MEGA 4.1 (Tamura et al., 2007). The NJ genetic distance trees were 178
constructed with the shared allele model (DAS) (Jin and Chakraborty, 1994) and 1,000 bootstrap 179
replicates for all isolates typed at the 4 loci (TEM1 and TEM2) (159 isolates) and with all 7 loci 180
(140 isolates). 181
Microsatellite allelic frequencies, Nei‟s unbiased measures of genetic distance and 182
inbreeding coefficient (FIS), Wright‟s fixation index (FST) values and departure from Hardy–183
Weinberg equilibrium (heterozygote deficiency or excess) were calculated with the genetic 184
software GENEPOP v.3.4 (http://genepop.curtin.edu.au) originally developed by Raymond and 185
Rousset (1995). In addition, the standardised index of association (IAS), a measure of the 186
association between alleles at different loci, was calculated using the software program LIAN - 187
LInkage ANalysis, Version 3.5 (http://adenine.biz.fh-weihenstephan.de/cgi-bin/lian/lian.cgi.pl) 188
developed by Haubold and Hudson (2000). The null hypothesis of linkage equilibrium for the T. 189
evansi dataset was tested by Monte Carlo simulation (1,000 iterations) and a parametric test 190
(Haubold et al., 1998). The multi locus genotype (MLG) index (G/N), a measure of clonality 191
expressed as a fraction derived from the number of MLGs (G) compared with the number of 192
isolates (N), was calculated. Genic differentiation with calculation of FST values between 193
populations (determined by Structure 2.3.3), provinces, year of sampling and provinces together 194
with years of sampling were calculated using GenAlEx (Peakall and Smouse, 2006). 195
196
2.6. Statistical analysis 197
Page 10 of 43
Accep
ted
Man
uscr
ipt
10
The populations identified amongst the isolates were recorded and coded together with 198
other information gathered about each host in a dataset using the Statistical Package for Social 199
Sciences (SPSS) 17.0.1 (SPSS Inc., Chicago, IL, USA). The statistical significance of any 200
differences in values for categorical variables (population, geographical origin, MHCT, CATT 201
and body condition) was determined using Pearson‟s Chi-square (χ2) (p<0.05) and for continuous 202
variables (host animal age, mortality and PCV) using Student‟s T-test (p<0.05). 203
Analysis of the predominant T. evansi population (as determined by Structure 2.3.3), and 204
buffalo mortality required the removal of any sampling locations that could not be assigned a 205
predominant population (i.e. barangays with 2 isolates, each assigned by Structure 2.3.3 to 206
different populations). This reduced the dataset analysed from 44 to 41 barangays. 207
208
2.7. Geographical representation of T. evansi microsatellite information 209
The spatial relationships of the T. evansi isolate genotypes were visualised by plotting 210
populations, epidemics and buffalo mortality figures together with isolate genotypes on a map of 211
Mindanao using a geographical information system (GIS) program Quantum GIS 1.5.0-Tethys 212
(http://qgis.org/). 213
214
3. Results 215
3.1. Microsatellite selection procedure 216
The loci M6C8 and MT3033TA sourced from Biteau et al. (2000) which were initially 217
combined in two multiplexes, were monomorphic in 120 Mindanao T. evansi isolates and these 218
loci were therefore discarded. 219
Page 11 of 43
Accep
ted
Man
uscr
ipt
11
DNA sequencing of amplicons of each of the loci (Tbb1, Tbb5, Tbb9, Tbb10) developed 220
in the present study confirmed that the loci amplified from T. evansi isolates were the same loci 221
identified from the T. brucei genome sequence data (data not shown). Therefore a total of 7 loci 222
(4 developed during the present study and 3 sourced from Biteau et al. (2000)) were used for 223
further analysis. 224
225
3.2. Microsatellite typing of original host (OH) and mouse derived (M) DNA extracts 226
Microsatellite typing with the multiplexes TEM1, TEM2, Bite1 and Bite2 were attempted 227
on 262 DNA samples from 195 isolates: 179 T. evansi mouse derived (M) isolates and 99 T. 228
evansi original host derived (OH) samples. Of the 179 T. evansi M isolates, 83 (46%) had a 229
corresponding OH sample, i.e. DNA extractions from the blood of an infected buffalo and the 230
blood of a mouse infected with a sample from the same buffalo. Trypanosoma evansi M samples 231
were more successfully amplified than the OH samples; 7 loci were amplified from 70.9% 232
(127/179) M isolates but only 13.1% (13/99) OH isolates were amplified at all 7 loci. 233
Of the 195 isolates tested, 6 isolates typed had 3-4 alleles at each microsatellite locus 234
typed. Two OH isolates (A7B-387OH and A73-807OH) and 4 M isolates (A7E-788M, A7E-235
859M, A7E-1079M and SN7-151M) demonstrated mixed loci. The complementary M isolates 236
(A7B-387M and A73-807M) to the two mixed OH isolates (A7B-387OH and A73-807OH) did 237
not demonstrate the same mixture of alleles but contained alleles present in the OH isolate 238
mixtures. 239
240
3.3. Population structuring 241
Page 12 of 43
Accep
ted
Man
uscr
ipt
12
An unrooted genetic distance tree with NJ 1000 bootstrap was constructed with the 242
shared allele evolutionary model from 140 isolates typed at 7 loci (Fig. 3) and 179 isolates at 4 243
loci (not shown). Population structuring analyses conducted on 4 and 7 loci datasets using the 244
software program Structure 2.3.3 (Pritchard et al., 2000) were congruent (i.e. T. evansi isolates 245
which were present in both datasets, were assigned to the same populations in both analyses). 246
Furthermore, the highest average log probability of data clustering (L(K)) from 15 replicates 247
calculated for the 7 loci dataset was determined for K=3 (-435.867). The three populations 248
identified were designated I, II and III and are represented in the constructed NJ genetic distance 249
tree (Fig. 3). The predominant population in the 7 loci dataset was population II which 250
comprised 67.1% (94/140) of isolates followed by population I with 32.1% (45/140) of isolates 251
and population III with 0.7% (1/140) of isolates. The analysis also identified 16 MLGs (sets of 252
isolates that appeared to be identical when typed with a given set of genetic markers in a 253
basically clonal species) (Tibayrenc and Ayala, 1991) within the three populations (I, II and III). 254
The largest MLG groups were A and C; MLG A comprised 21 identical isolates and MLG C had 255
55 isolates. 256
257
3.4. Population analysis 258
Analysis of the 140 isolates typed at 7 loci revealed complete heterozygosity at all of the 259
7 microsatellite loci analysed (no homozygotes present in any loci in any isolate genotyped) with 260
significant deviations from Hardy-Weinberg equilibrium (p<0.001). The mean inbreeding 261
coefficient (FIS) of the 7 loci was -0.3458 and negative FIS values were also generated for each 262
individual locus. Levels of polymorphism varied between loci, with the total number of alleles 263
per locus ranging from 3 (MORFCA) to 12 (Tbb5). The MLG index (G/N) for 7 loci was 0.14 (6 264
Page 13 of 43
Accep
ted
Man
uscr
ipt
13
unique and 14 repeated MLGs). The standardised index of association (IAS) for the complete 265
dataset was 0.6956 with significant linkage disequilibrium amongst all combinations of loci 266
(p<0.001). 267
The expected and observed heterozygosity, number of alleles and FIS values of each locus 268
within populations I, II and III are presented in Table 3. Nei‟s genetic distance, FST and geneflow 269
(NM) of isolates divided by population (I and II; III excluded due being only 1 isolate), province 270
(ADS and SDS), year of collection (2006 and 2007), or combination of province and year of 271
collection (ADS-2006, ADS-2007, SDS-2006 and SDS-2007) were calculated (Table 4). There 272
was little evidence in population structuring according to province or year of collection (0.022 273
and 0.021 respectively). There were significantly different FST values (0.118-0.154) generated for 274
isolates collected from ADS in 2006 compared with those collected from ADS in 2007 or from 275
SDS in either 2006 or 2007. The FST value between populations I and II was 0.117. There was a 276
significant linkage of all loci for isolates divided on the basis of Structure 2.3.3 population, 277
province, year of collection or province and year of collection based on calculated IAS values 278
(p<0.001). 279
280
3.5. Statistical and geographical analysis 281
No significant differences were seen between population I and II isolates (III was omitted 282
due to insufficient data) and province in regard to any categories except for province and buffalo 283
mortalities (p<0.05). Population II was detected in a significantly higher proportion of samples 284
from ADS (58.7%, 95% C.I. 48.9-68.1%) and SDS (81.3% 95% C.I. 67.4-91.1%) (p=0.022) 285
(Fig. 4). The mean buffalo mortality in locations with a predominant population I of T. evansi 286
Page 14 of 43
Accep
ted
Man
uscr
ipt
14
isolates was significantly higher (13.6%) compared to locations with a predominant II population 287
of T. evansi isolates (6.9%) (p=0.047). 288
Analysis of MHCT, CATT, buffalo mortality and PCV with T. evansi MLGs could only 289
be conducted for two MLGs; A and C (sufficient numbers for analysis), which belonged to the 290
same population (II). MLG A which comprised 21 isolates was primarily isolated in Surigao del 291
Sur, whilst MLG C which comprised 55 isolates was predominantly isolated in Agusan del Sur 292
(Fig. 6). There were no significant differences on the MHCT and CATT results and PCV values 293
of hosts infected with either MLG A or C. Buffalo mortality in areas with a reported outbreak 294
had significantly higher mortality when MLG A T. evansi isolates were present than in areas 295
where MLG C isolates were present (p<0.05). The mean buffalo mortality involving MLG A 296
isolates was 17.1% and 5.9% for MLG C isolates. There was a more visible association between 297
geographical location of MLG A isolates and areas of recorded buffalo mortality (4 barangays) 298
compared with MLG C and areas of buffalo mortality (1 barangay) (Fig. 5). 299
300
3.6. Correlation between epidemic and T. evansi MLG 301
Samples were collected from livestock within 3 months of first outbreak report in 9 of the 302
17 reported outbreaks of surra. The T. evansi MLG most likely to be the source of the outbreak 303
could not always be inferred because many of the epidemic sites sampled yielded too few 304
isolates to be conclusive. Several isolates of the same MLG were typed in different outbreaks 305
and isolates collected from some outbreaks comprised more than one T. evansi genotype as 306
presented in Table 5. 307
308
4. Discussion 309
Page 15 of 43
Accep
ted
Man
uscr
ipt
15
The population structure of T. evansi in Mindanao exhibited evidence of clonal 310
expansion, in particular within epidemics with the observation of 14 MLGs in the dataset as well 311
as significant linkage, negative FIS values and 100% heterozygosity at each locus. High 312
heterozygosity occurs in clonal organisms due to accumulations of random mutation events 313
(Halkett et al., 2005; De Meeus et al., 2006). Due to the absence of the tsetse fly, vector in which 314
T. brucei is known to undergo sexual recombination (Peacock et al., 2011), T. evansi within the 315
Philippines and other non tsetse fly areas is inferred to be asexually reproducing. The complete 316
heterozygosity observed in the T. evansi isolates from the Philippines is incompatible with 317
meiotic segregation (Tibayrenc et al., 1990) and supportive of T. evansi reproduction being 318
asexual in the Philippines. However, heterozygosity may also be due to the typing of mixed 319
strain infections. The presence of 14 T. evansi MLGs in Mindanao isolates, however, negates the 320
notion that the heterozygous microsatellite profiles produced are the result of mixed infections 321
(the combination of two homozygote strains giving rise to two alleles at each locus). The finding 322
of several MLGs amongst the Philippine isolates is in contrast to a recent study conducted by 323
Salim et al. (2011) on T. evansi in camels within Sudan. Salim et al. (2011) noted that the 324
absence of any MLGs in their dataset was unusual for what is believed to be a clonal organism. 325
They argued that their dataset may have been subject to “frequent allelic dropouts, Wahlund 326
effects or both”. 327
Mixed strain infections of T. evansi were evident in 6 isolates (the presence of 4 alleles at 328
each locus), 2 of which originated from OH isolates. Mixed infections have been reported in 329
cattle (T. congolense) (Morrison et al., 2009), humans (T. brucei) (Truc et al., 2002) and in pigs 330
(T. brucei) (Jamonneau et al., 2003). Intra-host variation has been observed in the ribosomal 331
internal transcribed spacer 1 of T. evansi from buffalo and cattle from Thailand, which could be 332
Page 16 of 43
Accep
ted
Man
uscr
ipt
16
due to natural mixed strain infections (Khuchareontaworn et al., 2007; Areekit et al., 2008). 333
Mixed trypanosome species and strain infections are also known to occur in vectors such as the 334
tsetse fly (MacLeod et al., 1999). Tabanids are considered opportunistic feeders (Magnarelli and 335
Anderson, 1980; Gouteux et al., 1989; Muzari et al., 2010) and recent studies have shown that 336
mixed host blood meals are frequent in some tabanid species (Muzari et al., 2010). 337
It is important to recognise that in vivo systems such as the MIT may exert „selective 338
pressures‟ (Jamonneau et al., 2003) that may lead to a reduced recognition of naturally occurring 339
mixed infections. Concentration of trypanosomes by methods such as MIT has however, been 340
very important for trypanosome analysis due to the characteristic low parasitaemia that is 341
observed in most naturally T. evansi-infected hosts. Direct amplification of microsatellites 342
without the aid of any concentration methodologies or PCR sensitivity enhancing strategies is 343
frequently difficult. Duffy et al. (2009) were only able to amplify a suite of 8 microsatellites 344
from 31 of 304 (10%) T. vivax-positive DNA samples. Nested PCRs may have increased 345
sensitivity without the need for a potentially selective concentration process such as MIT, 346
however, the reported increases in sensitivity achieved by nested microsatellite PCR design vary. 347
Valadares et al. (2008) reported a 100-1,000-fold increase in the sensitivity with typing T. cruzi 348
from infected tissue samples, whilst Morrison et al. (2007) only reported a 5-fold increase in 349
sensitivity with a nested PCR using dried blood on FTA filter cards. PCR detection on FTA 350
cards can however be problematic as reported by Cox et al ,(Cox et al., 2010) who observed that 351
single punch nested PCR analyses for trypanosomes can lead to considerable underestimations of 352
prevalence. 353
The three populations (I, II and III) identified amongst the T. evansi isolates genotyped 354
from Mindanao may be evidence of the existence of only three T. evansi lineages in the 355
Page 17 of 43
Accep
ted
Man
uscr
ipt
17
northeastern Mindanao provinces (ADS and SDS). The population analysis indicates that the 356
outbreaks investigated in ADS and SDS were primarily caused by two T. evansi lineages which 357
have differentiated over time into two populations (I and II). The isolate AS-76 from Mapaga, 358
Prosperidad, ADS was the only representative of T. evansi population III in Mindanao. 359
Identification of only one isolate for this population suggests that this isolate may be a recent 360
introduction to the northeastern Mindanao provinces (ADS and SDS). The genetic similarity of 361
AS-76 with an isolate LZ-1 (typed at 4 loci) from the Philippine island Luzon suggests that AS-362
76 isolate may have originated from Luzon. Mindanao is, however, an island geographically 363
removed from Luzon with restrictions of animal movement due to Mindanao being considered 364
foot and mouth disease (FMD) free without vaccination by the Office Internationale des 365
Epizooties (OIE) since 2001 (OIE, 2001). It is therefore more likely that AS-76 is an isolate from 366
a lineage of T. evansi present in some of the Mindanao provinces not examined in this study. A 367
comprehensive survey to genotype T. evansi isolates from all Mindanao provinces as well as 368
other Philippine provinces would be required to elucidate the origins of the 3 Mindanao T. evansi 369
populations. The present study did not test any isolates from the southern Mindanao provinces 370
which were the first to report surra outbreaks. 371
The provinces of ADS and SDS in Mindanao, although situated adjacent to one another, 372
are separated by a mountain range with few connecting roads. The mountain range, lack of 373
interconnecting roads and the political insurgency in the region may have restricted animal 374
movement between the provinces and thereby reduced the spread of T. evansi strains between 375
provinces. However, comparison of isolates from these two provinces did not support the 376
hypothesis that there is a significant genetic divide in isolates (low FST values of 0.022) with 377
considerable gene flow (11.3) measured between the provinces. This was also evident with T. 378
Page 18 of 43
Accep
ted
Man
uscr
ipt
18
evansi population II isolates present in both provinces and significantly more prominent in both 379
ADS and SDS provinces than population I. The transmission of T. evansi MLGs between 380
provinces was exemplified by MLGs A and C with isolates of these MLGs present in both 381
provinces on either side of the mountain range. 382
Significant variations on the FST values of Mindanao T. evansi based on population sub-383
structuring analysis may suggest that the isolates collected from ADS in 2006 were different 384
from the isolates collected from the same province in the following year as well as from isolates 385
collected from SDS in either 2006 or 2007. However, this could be an anomaly attributed to the 386
small sample size (n= 9) of isolates collected from ADS in 2006. 387
The susceptibility to T. evansi infection varies depending on the host species. However, it 388
is not clear whether host-specific strains exist in the areas sampled. It is interesting to note that 389
the outbreak recorded in Ebro, San Francisco identified a T. evansi MLG (K) from 4 horses 390
which was not present in any of the other surra outbreaks in nearby barangays during the same 391
year. This may have been due to a strain more pathogenic to horses than to buffalo or simply due 392
to successful containment of the Ebro outbreak. In the present study there were no villages 393
sampled which contained both horse and buffalo surra strains. This was usually the result of high 394
surra mortality in horses leading to the elimination of most horses from surra epidemic areas 395
prior to sampling. The high susceptibility or possible high pathogenicity of T. evansi to horses 396
was evident in the outbreaks in Ladgadan, Del Rosario, Mapaga and Causwagan, ADS which 397
recorded equine mortalities in excess of 30%. In Banahao, Lianga, SDS, T. evansi isolates were 398
however recovered from 2 horses which were identified as MLG C, the most likely causative 399
agent of the large outbreak in buffalo in Guadalupe and Concordia. As the Guadalupe/Concordia 400
Page 19 of 43
Accep
ted
Man
uscr
ipt
19
outbreak occurred immediately before the Banahao outbreak, it may be due to translocation of T. 401
evansi-infected animals. 402
The evidence that some MLGs were localised to specific barangays is probably a 403
reflection of the transmission dynamics of T. evansi that require close proximity between hosts 404
(Barros and Foil, 2007). In Mindanao barangays, the mechanical transmission of T. evansi by 405
tabanids is likely to be enhanced by the common practice of communal tethering of livestock in 406
pastures often adjacent to or in close proximity to irrigation canals, swamps and rivers or by 407
using a common wallowing pool as in the case of buffaloes (Dargantes, A. pers. comm., 2010). 408
Areas with abundant water are likely to have high population of tabanids, as most tabanids 409
require water for breeding and development (Mitzmain, 1913; Service, 1986; McElligott and 410
Lewis, 1996; Butt et al., 2008). 411
Transmission of T. evansi by animal translocation is a logical explanation for the 412
outbreaks reported in Guadalupe and Concordia, Esperanza, ADS. These outbreaks occurred in 413
villages geographically close to one another, around the same time of the year (Jan-Mar 2007) 414
and were both strongly associated with T. evansi MLG C. The MLG finding and temporal and 415
spatial proximity of the outbreaks suggest this was in fact a single connected event rather than 416
two separate outbreaks. The analysis revealed that the MLG responsible for these two outbreaks 417
was the same, which then caused an outbreak in Ladgadan, San Francisco, ADS in the following 418
year. This may be due to the movement of an infected host from Guadalupe or Concordia to 419
Ladgadan and seasonal fluctuations in tabanid abundance culminating in an outbreak the 420
following year in a new location. 421
In the present study, increased buffalo mortality was more associated with T. evansi 422
population I isolates than with population II isolates, and with MLG C isolates than with MLG A 423
Page 20 of 43
Accep
ted
Man
uscr
ipt
20
isolates from population II. However, it is important to note that the variation in pathogenicity 424
could be due to the many other potentially influencing factors such as immune status and other 425
contributing health issues of the hosts. Microsatellite typing is able to assist in elucidating this 426
issue but influencing factors would need to be controlled or a more extensive survey over a 427
considerably longer period of time would be required. Data from the application of microsatellite 428
typing of T. evansi isolates from Mindanao shows that outbreaks were caused by the same T. 429
evansi strain (MLG C), and that the T. evansi strains in the northeastern Mindanao provinces 430
represent three lineages. Furthermore, the present study has demonstrated the value of 431
microsatellites for the study of complex infectious agents such as T. evansi. However, larger and 432
more representative datasets are required to provide a more accurate description of the origin of 433
outbreaks, the transmission factors promoting spread of outbreaks and the importance of strain 434
variation in pathogenicity. 435
436
437
Acknowledgements 438
Field surveys and collection of isolates were carried out in coordination with the Mindanao 439
Unified Surra Control Approach (MUSCA) group headed by Dr. R. Mercado and the help from 440
veterinarians and staff of collaborating government agencies in Agusan del Norte, Agusan del 441
Sur and Surigao del Sur, and the members of the CMU Surra Team. The authors are also grateful 442
for the significant assistance of Drs. J.R. Dargantes, P. Calo, M. Samar, J.A. Abella, R. 443
Sumagang, E. Igsoc, M. Pagobo and of D. Miguel and K. Quero. The financial support from the 444
Australian Centre for International Agricultural Research (ACIAR), Central Mindanao 445
University (CMU), International Foundation of Science (IFS), Murdoch University and the 446
Page 21 of 43
Accep
ted
Man
uscr
ipt
21
Australian Biosecurity Cooperative Research Centre (AB-CRC) which contributed to the 447
research presented in the article is acknowledged. 448
449
Conflict of interest statement 450
The authors of this article declare that they have no conflicts of interest with regard to the work 451
presented herein. 452
453
References 454
Areekit, S., Singhaphan, P., Kanjanavas, P., Khuchareontaworn, S., Sriyapai, T., Pakpitcharoen, 455
A., Chansiri, K., 2008. Genetic diversity of Trypanosoma evansi in beef cattle based on 456
internal transcribed spacer region. Infect. Genet. Evol. 8, 484-488. 457
Balmer, O., Caccone, A., 2008. Multiple-strain infections of Trypanosoma brucei across Africa. 458
Acta Trop. 107, 275-279. 459
Barros, A.T., Foil, L.D., 2007. The influence of distance on movement of tabanids (Diptera: 460
Tabanidae) between horses. Vet. Parasitol. 144, 380-384. 461
Benson, G., 1999. Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids 462
Res. 27, 573-580. 463
Biteau, N., Bringaud, F., Gibson, W., Truc, P., Baltz, T., 2000. Characterization of Trypanozoon 464
isolates using a repeated coding sequence and microsatellite markers. Mol. Biochem. 465
Parasitol. 105, 187-202. 466
Butt, C., Hicks, B., Campbell, C., 2008. The diversity and abundance of Tabanidae in black 467
spruce forests and sphagnum bogs in Gros Morne National Park, Newfoundland, Canada. 468
. J. Acad. Entomol. Soc. 4, 7-13. 469
Page 22 of 43
Accep
ted
Man
uscr
ipt
22
Cox, A.P., Tosas, O., Tilley, A., Picozzi, K., Coleman, P., Hide, G., Welburn, S.C., 2010. 470
Constraints to estimating the prevalence of trypanosome infections in East African zebu 471
cattle. Parasites and Vectors 3, 82. 472
Cresencio, R.O., Roeder, P.L., Tongson, M.T., Gonzales, A.T., Jacobo, L.V., De la Peña, R.C., 473
Abuso, O.T., 1994. Incident of disease in carabaos and cattle with Trypanosoma evansi 474
involvement. In, Annual Convention of the Veterinary Practitioners Association of the 475
Philippines, Manila, Philippines. 476
D'Avila, D.A., Macedo, A.M., Valadares, H.M., Gontijo, E.D., de Castro, A.M., Machado, C.R., 477
Chiari, E., Galvao, L.M., 2009. Probing population dynamics of Trypanosoma cruzi 478
during progression of the chronic phase in chagasic patients. J. Clin. Microbiol. 47, 1718-479
1725. 480
Dargantes, A.P., Mercado, R.T., Dobson, R.J., Reid, S.A., 2009. Estimating the impact of 481
Trypanosoma evansi infection (surra) on buffalo population dynamics in southern 482
Philippines using data from cross-sectional surveys. Int. J. Parasitol. 39, 1109-1114. 483
Dargantes, A.P., 2010. Epidemiology, control and potential insect vectors of Trypanosoma 484
evansi (surra) in village livestock in southern Philippines. Murdoch University, Perth. 485
De Meeus, T., Lehmann, L., Balloux, F., 2006. Molecular epidemiology of clonal diploids: a 486
quick overview and a short DIY (do it yourself) notice. Infect. Genet. Evol. 6, 163-170. 487
Dobson, R.J., Dargantes, A.P., Mercado, R.T., Reid, S.A., 2009. Models for Trypanosoma evansi 488
(surra), its control and economic impact on small-hold livestock owners in the 489
Philippines. Int. J. Parasitol. 39, 1115-1123. 490
Page 23 of 43
Accep
ted
Man
uscr
ipt
23
Duffy, C.W., Morrison, L.J., Black, A., Pinchbeck, G.L., Christley, R.M., Schoenefeld, A., Tait, 491
A., Turner, C.M., MacLeod, A., 2009. Trypanosoma vivax displays a clonal population 492
structure. Int. J. Parasitol. 39, 1475-1483. 493
Gouteux, J.P., Noireau, F., Staak, C., 1989. The host preferences of Chrysops silacea and C. 494
dimidiata (Diptera: Tabanidae) in an endemic area of Loa loa in the Congo. Ann. Trop. 495
Med. Parasitol. 83, 167-172. 496
Halkett, F., Simon, J.C., Balloux, F., 2005. Tackling the population genetics of clonal and 497
partially clonal organisms. Trends Ecol. Evol. 20, 194-201. 498
Haubold, B., Travisano, M., Rainey, P.B., Hudson, R.R., 1998. Detecting linkage disequilibrium 499
in bacterial populations. Genetics 150, 1341-1348. 500
Haubold, B., Hudson, R.R., 2000. LIAN 3.0: detecting linkage disequilibrium in multilocus data. 501
Linkage Analysis. Bioinformatics 16, 847-848. 502
Hoare, C.A., 1972. The trypanosomes of mammals A zoological monograph. Blackwell 503
Scientific Publications, Oxford. 504
Jamonneau, V., Barnabe, C., Koffi, M., Sane, B., Cuny, G., Solano, P., 2003. Identification of 505
Trypanosoma brucei circulating in a sleeping sickness focus in Cote d'Ivoire: assessment 506
of genotype selection by the isolation method. Infect. Genet. Evol. 3, 143-149. 507
Jin, L., Chakraborty, R., 1994. Estimation of genetic distance and coefficient of gene diversity 508
from single-probe multilocus DNA fingerprinting data. Mol. Biol. Evol. 11, 120-127. 509
Khuchareontaworn, S., Singhaphan, P., Viseshakul, N., Chansiri, K., 2007. Genetic diversity of 510
Trypanosoma evansi in buffalo based on internal transcribed spacer (ITS) regions. J. Vet. 511
Med. Sci. 69, 487-493. 512
Page 24 of 43
Accep
ted
Man
uscr
ipt
24
Koffi, M., De Meeus, T., Bucheton, B., Solano, P., Camara, M., Kaba, D., Cuny, G., Ayala, F.J., 513
Jamonneau, V., 2009. Population genetics of Trypanosoma brucei gambiense, the agent 514
of sleeping sickness in Western Africa. Proc. Natl. Acad. Sci. U.S.A. 106, 209-214. 515
Lai, D.H., Hashimi, H., Lun, Z.R., Ayala, F.J., Lukes, J., 2008. Adaptations of Trypanosoma 516
brucei to gradual loss of kinetoplast DNA: Trypanosoma equiperdum and Trypanosoma 517
evansi are petite mutants of T. brucei. Proc Natl Acad Sci U S A 105, 1999-2004. 518
Liu, K., Muse, S.V., 2005. PowerMarker: an integrated analysis environment for genetic marker 519
analysis. Bioinformatics 21, 2128-2129. 520
Llewellyn, M.S., Lewis, M.D., Acosta, N., Yeo, M., Carrasco, H.J., Segovia, M., Vargas, J., 521
Torrico, F., Miles, M.A., Gaunt, M.W., 2009a. Trypanosoma cruzi IIc: phylogenetic and 522
phylogeographic insights from sequence and microsatellite analysis and potential impact 523
on emergent Chagas disease. PLoS Negl. Trop. Dis. 3, e510. 524
Llewellyn, M.S., Miles, M.A., Carrasco, H.J., Lewis, M.D., Yeo, M., Vargas, J., Torrico, F., 525
Diosque, P., Valente, V., Valente, S.A., Gaunt, M.W., 2009b. Genome-scale multilocus 526
microsatellite typing of Trypanosoma cruzi discrete typing unit I reveals phylogeographic 527
structure and specific genotypes linked to human infection. PLoS Pathog. 5, e1000410. 528
MacLeod, A., Turner, C.M., Tait, A., 1999. A high level of mixed Trypanosoma brucei 529
infections in tsetse flies detected by three hypervariable minisatellites. Mol. Biochem. 530
Parasitol. 102, 237-248. 531
Magnarelli, L.A., Anderson, J.F., 1980. Feeding behaviour of tabanidae (Diptera) on cattle and 532
serological analyses of partial blood meals. Environ. Entomol. 9, 664-667. 533
Manresa, M., 1935. Studies on surra: I. The outbreak of surra in 1933 in the College of 534
Agriculture. The Philippine Agriculturist 23, 749-757. 535
Page 25 of 43
Accep
ted
Man
uscr
ipt
25
Manuel, M.F., 1998. Sporadic outbreaks of surra in the Philippines and its economic impact. J. 536
Protozool. Res. 8, 131-138. 537
McElligott, P., Lewis, D., 1996. Distribution and abundance of immature tabanidae (Diptera) in a 538
subarctic Labrador peatland. Can. J. Zool. 74, 1364-1369. 539
McInnes, L., Hanger, J., Simmons, G., Reid, S., Ryan, U., 2011. Novel trypanosome 540
Trypanosoma gilletti sp. (Sarcomastigophora: Trypanosomatidae) and the extension of 541
the host range of Trypanosoma copemani to include the koala (Phascolarctos cinereus). 542
Parasitology 138, 59-70. 543
Mitzmain, M.B., 1913. The surra conveying fly of the Philippines and some factors concerned in 544
control measures. Philippine Agricultural Review 6, 371-379. 545
Morrison, L.J., McCormack, G., Sweeney, L., Likeufack, A.C., Truc, P., Turner, C.M., Tait, A., 546
MacLeod, A., 2007. Use of multiple displacement amplification to increase the detection 547
and genotyping of Trypanosoma species samples immobilized on FTA filters. Am. J. 548
Trop. Med. Hyg. 76, 1132-1137. 549
Morrison, L.J., Tweedie, A., Black, A., Pinchbeck, G.L., Christley, R.M., Schoenefeld, A., 550
Hertz-Fowler, C., MacLeod, A., Turner, C.M., Tait, A., 2009. Discovery of mating in the 551
major African livestock pathogen Trypanosoma congolense. PLoS One 4, e5564. 552
Muzari, M.O., Burgess, G.W., Skerratt, L.F., Jones, R.E., Duran, T.L., 2010. Host preferences of 553
tabanid flies based on identification of blood meals by ELISA. Vet. Parasitol. 174, 191-554
198. 555
Njiru, Z.K., Constantine, C.C., 2007. Population sub-structuring among Trypanosoma evansi 556
stocks. Parasitol. Res. 101, 1215-1224. 557
Page 26 of 43
Accep
ted
Man
uscr
ipt
26
OIE, 2001. Resolutions adopted by the International Committee of the OIE. In: Epizooties, O.I.d. 558
(Ed). Recognition of the Foot and Mouth Disease Status of Member Countries, Vol. 17, 559
International Committee of the Office Internationale des Epizooties Paris, pp. 1-52. 560
OIE, 2008. Trypanosoma evansi infections (including surra): Aetiology Epidemiology Diagnosis 561
Prevention and Control References. In: Epizooties, O.I.d. (Ed). Technical disease cards, 562
Office International des Epizooties. 563
OIE, 2010. Trypanosoma evansi infections (including surra), Manual of Diagnostic Tests and 564
Vaccines for Terrestrial Animals Vol. 1, Office International des Epizooties, pp. 1-14. 565
Payne, R.C., Ward, D.E., Usman, M., Rusli, A., D., D., Husein, A., 1988. Prevalence of bovine 566
haemoparasites in Aceh Province of Northern Sumatra: implications for imported cattle. 567
Prev. Vet. Med., 275-283. 568
Payne, R.C., Sukanto, I.P., Djauhari, D., Jones, T.W., 1991a. Trypanosoma evansi infection in 569
bovine and buffalo calves in Indonesia. Vet. Parasitol. 38, 253-256. 570
Payne, R.C., Sukanto, I.P., Djauhari, D., Partoutomo, S., Wilson, A.J., Jones, T.W., Boid, R., 571
Luckins, A.G., 1991b. Trypanosoma evansi infection in cattle, buffaloes and horses in 572
Indonesia. Vet. Parasitol. 38, 109-119. 573
Peacock, L., Ferris, V., Sharma, R., Sunter, J., Bailey, M., Carrington, M., Gibson, W., 2011. 574
Identification of the meiotic life cycle stage of Trypanosoma brucei in the tsetse fly. Proc 575
Natl Acad Sci U S A 108, 3671-3676. 576
Peakall, R., Smouse, P.E., 2006. GenALex 6: genetic analysis in Excel. Population genetic 577
software for teaching and research. Mol. Ecol. Notes 6, 288-295. 578
Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure using 579
multilocus genotype data. Genetics 155, 945-959. 580
Page 27 of 43
Accep
ted
Man
uscr
ipt
27
Randall, R., Schwartz, S., 1936. A survey for the incidence of surra in the Philippine Island. 581
Veterinary Bulletin (US Army) 33, 99-108. 582
Raymond, M., Rousset, F., 1995. GENEPOP (version 1.2): population genetics software for 583
exact tests and ecumenicism. J. Hered. 86, 248-249. 584
Salim, B., de Meeus, T., Bakheit, M.A., Kamau, J., Nakamura, I., Sugimoto, C., 2011. 585
Population genetics of Trypanosoma evansi from camel in the Sudan. PLoS Negl Trop 586
Dis 5, e1196. 587
Service, M.W., 1986. Lecture notes on medical entomology. Blackwell Scientific Publications, 588
Oxford. 589
Simo, G., Njiokou, F., Tume, C., Lueong, S., De Meeus, T., Cuny, G., Asonganyi, T., 2010. 590
Population genetic structure of Central African Trypanosoma brucei gambiense isolates 591
using microsatellite DNA markers. Infect. Genet. Evol. 10, 68-76. 592
Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: Molecular Evolutionary Genetics 593
Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, 1596-1599. 594
Tibayrenc, M., Kjellberg, F., Ayala, F.J., 1990. A clonal theory of parasitic protozoa: The 595
population structures of Entamoeba, Giardia, Leishmania, Naegleria, Plasmodium, 596
Trichomonas, and Trypanosoma and their medical and taxonomical consequences. Proc. 597
Natl. Acad. Sci. U. S. A. 87, 2414-2418. 598
Tibayrenc, M., Ayala, F.J., 1991. Towards a population genetics of microorganisms: The clonal 599
theory of parasitic protozoa. Parasitol. Today 7, 228-232. 600
Truc, P., Ravel, S., Jamonneau, V., N'Guessan, P., Cuny, G., 2002. Genetic variability within 601
Trypanosoma brucei gambiense: evidence for the circulation of different genotypes in 602
Page 28 of 43
Accep
ted
Man
uscr
ipt
28
human African trypanosomiasis patients in Cote d'Ivoire. Trans. R. Soc. Trop. Med. Hyg. 603
96, 52-55. 604
Valadares, H.M., Pimenta, J.R., de Freitas, J.M., Duffy, T., Bartholomeu, D.C., Oliveira Rde, P., 605
Chiari, E., Moreira Mda, C., Filho, G.B., Schijman, A.G., Franco, G.R., Machado, C.R., 606
Pena, S.D., Macedo, A.M., 2008. Genetic profiling of Trypanosoma cruzi directly in 607
infected tissues using nested PCR of polymorphic microsatellites. Int. J. Parasitol. 38, 608
839-850. 609
610
611
Page 29 of 43
Accep
ted
Man
uscr
ipt
Table 1
Details of the locations of putative Trypanosoma evansi outbreaks in 16 barangays in Mindanao,
Philippines between late 2005 and 2007 and the associated livestock mortality percentages.
Barangay Municipality Province Epidemic
Year
Epidemic
Months
Mortality %
Buffalo Cattle Goat Horse
Parang Cantilan SDS 2005 Dec NR NR NR NR
San Isidro San Francisco ADS 2006 Jan-Mar 17.7 12.8 0.0 0.0
Ebro San Francisco ADS 2006 Jan-Mar NR NR NR NR
Libas Gua San Miguel SDS 2006 Jan-Mar NR NR NR NR
Ladgadan San Francisco ADS 2006 Jul-Sep 5.9 5.4 0.0 40.0
Mamis Barobo SDS 2006 Jun-Aug 17.1 6.0 9.1 0.0
Del Rosario Sibagat ADS 2006 Mar-Apr 14.6 0.0 0.0 35.0
Mapaga Prosperidad ADS 2006 Mar-Apr 8.3 0.0 0.0 35.7
Bayan Marihatag SDS 2006 Oct-Nov 7.9 0.0 100.0 0.0
San Ignacio Trento ADS 2007 Apr-Jun 16.5 0.0 0.0 0.0
Dimasalang San Luis ADS 2007 Dec-Mar 6.0 17.9 0.0 0.0
Concordia Esperanza ADS 2007 Jan-Mar 5.6 8.3 0.0 0.0
Guadalupe Esperanza ADS 2007 Jan-Mar 6.0 0.0 0.0 0.0
Mambalili Bunawan ADS 2007 Jan-Mar 5.1 4.9 0.0 17.4
Banahao Lianga SDS 2007 Mar-May NR NR NR NR
Causwagan Talacogon ADS 2007 Nov-Feb 14.0 0.0 0.0 41.7
NR: Not recorded
SDS: Surigao del Sur
ADS: Agusan del Sur
Table
Page 30 of 43
Accep
ted
Man
uscr
ipt
Table 2
Primer sequences and modifications, amplicon sizes, and repeat motifs of microsatellite
multiplexes (TEM1 and TEM2) designed on the basis of T. brucei GenBank accession data.
Locus
name
Primer name - sequence 5´-3´
(modifications: PIG tail, fluorescent label)
T. brucei
Amplicon size
bp
T. brucei
Acc. no. T. brucei motif
TEM1
Tbb1 Tbb1F - TTTGAAGAATTTGTACCTCGGTG`3
Tbb1R- gttcttCTTTCAACCGCCTTATCAGC’3 192 AL92960 GT49
Tbb10 Tbb10F- TGTGGCTCCTACTTTCCGTG`3
Tbb10R- gttcttCTCCCTTTTGTCAGTGTGGC`3 262 NW001076899 CA48
TEM2
Tbb5 Tbb5F- CGCAACAACTTTCACATACGC`3
Tbb5R- gttcttTGAGGGTTCTTCGCTTACAC`3 240 AC159443 CA50
Tbb9 Tbb9F- CGTTTGTCGCTGCCTTTGTG`3
Tbb9R- gttcttTGTAGTGCCGATTAGGGTTG`3 239 NW001076895 TG51
Page 31 of 43
Accep
ted
Man
uscr
ipt
Table 3
Characterisation of the 7 microsatellite loci used in the population analysis of 3 populations (I, II
and III as determined by Structure 2.3.3 analysis) of T. evansi isolates from Mindanao, the
Philippines.
Population Locus Number of
isolates
Number of
Alleles Ho He FIS
I
Tbb1 45 3 1.000 0.511 -0.957
Tbb10 45 3 1.000 0.511 -0.957
Tbb5 45 5 1.000 0.579 -0.726
Tbb9 45 3 1.000 0.540 -0.850
MORFCA 45 2 1.000 0.500 -1.000
MT3033 45 3 1.000 0.566 -0.768
MEST19 45 3 1.000 0.511 -0.957
II
Tbb1 94 8 1.000 0.789 -0.268
Tbb10 94 7 1.000 0.790 -0.267
Tbb5 94 11 1.000 0.871 -0.149
Tbb9 94 4 1.000 0.750 -0.334
MORFCA 94 3 1.000 0.619 -0.615
MT3033 94 9 1.000 0.826 -0.210
MEST19 94 8 1.000 0.788 -0.268
III
Tbb1 1 2 1.000 0.500 -1.000
Tbb10 1 2 1.000 0.500 -1.000
Tbb5 1 2 1.000 0.500 -1.000
Tbb9 1 2 1.000 0.500 -1.000
MORFCA 1 2 1.000 0.500 -1.000
MT3033 1 2 1.000 0.500 -1.000
MEST19 1 2 1.000 0.500 -1.000
Ho: observed heterozygosity, He: expected heterozygosity, FIS: inbreeding coefficient.
Page 32 of 43
Accep
ted
Man
uscr
ipt
Table 4
Nei’s genetic distance, Wright’s Fixation Index (FST) and gene flow (Nm) for the different
divisions in T. evansi isolates collected in Mindanao, the Philippines and typed with 7
microsatellite loci. Isolates were grouped by province of origin (Agusan del Sur (ADS) or
Surigao del Sur (SDS)) and/or collection year (2006 or 2007) or population structure as
determined by software program Structure 2.3.3 (populations I and II).
Isolates’ population divisions Nei’s genetic distance FST Nm
ADS / SDS 0.122 0.022 11.276
2006 / 2007 0.142 0.021 11.48
ADS-2006 / ADS-2007 0.893 0.118 1.87
SDS-2006 / SDS-2007 0.049 0.011 22.9
ADS-2006 / SDS-2006 1.204 0.154 1.37
ADS-2007 / SDS-2007 0.066 0.012 20.54
ADS-2006 / SDS-2007 0.903 0.124 1.77
ADS-2007 / SDS-2006 0.171 0.032 7.68
I / II (Structure populations) 0.259 0.067 3.476
Page 33 of 43
Accep
ted
Man
uscr
ipt
Table 5
The 7 microsatellite multi locus genotypes (MLGs) and number of isolates from putative
Trypanosoma evansi outbreaks occurring in 16 barangays in Mindanao, Philippines between late
2005 and 2007.
Barangay Municipality Province Epidemic
Year Months
MLG typed at 7 loci
and number of isolates (n)
Parang Cantilan SDS 2005 Dec MLG I-2
San Isidro San Francisco ADS 2006 Jan-Mar NT
Ebro San Francisco ADS 2006 Jan-Mar MLG K-4
Libas Gua San Miguel ADS 2006 Jan-Mar MLG I-1 & MLG M-1
Ladgadan San Francisco ADS 2006 Jul-Sep MLG E-1
Mamis Barobo SDS 2006 Jun-Aug MLGA-10, MLG M-4 & unique-2
Del Rosario Sibagat ADS 2006 Mar-Apr NT
Mapaga Prosperidad ADS 2006 Mar-Apr MLG J-1 & unique-1
Bayan Marihatag SDS 2006 Oct-Nov MLG E-2
San Ignacio Trento ADS 2007 Apr-Jun MLG G-6, MLG H-5, MLG M-3 &
unique-1
Dimasalang San Luis ADS 2007 Dec-Mar MLG C-3
Concordia Esperanza ADS 2007 Jan-Mar MLG C-12, MLG L-1 & unique-2
Guadalupe Esperanza ADS 2007 Jan-Mar MLG C-20 & MLG L-1
Mambalili Bunawan ADS 2007 Jan-Mar NT
Banahao Lianga SDS 2007 Mar-May MLG C-2
Causwagan Talacogon ADS 2007 Nov-Feb MLG E-2 & MLG L-1
NT: Not typed at all 7 loci
MLG: multi locus genotype
Unique: an isolate with a unique multi locus genotype
Page 34 of 43
Accep
ted
Man
uscr
ipt
Fig. 1. A map of Mindanao in the Philippines showing the provinces where surra outbreaks have been reported in 1989-2009 (Dargantes, J.R., 1
Mercado, R. and Alforque, J. pers. comm., 2010).2
Figure
Page 35 of 43
Accep
ted
Man
uscr
ipt
Fig. 2 a and b. The locations of Trypanosoma evansi isolate sampling and surra epidemics (late 3
2005-2007) in Mindanao, Philippines. a. Depicts the Philippines with an inlay box over an area 4
of Mindanao which is enlarged in b. and contains marked T. evansi sampling locations () and 5
epidemic locations (). 6
7
Page 36 of 43
Accep
ted
Man
uscr
ipt
Fig. 3. Neighbour-joining tree inferred from the Share allele (DAS) genetic distances calculated 8
for the microsatellite data of 140 Trypanosoma evansi isolates from Mindanao, Philippines 9
genotyped at 7 loci. Unique isolates are identified by the individual code (i.e. A7E-912M) and 10
MLGs (groups of identical isolates) by a letter and a number signifying the number of clones of 11
that type (i.e. A21, is a MLG comprising 21 identical isolates). Percentage bootstrap values were 12
generated from 1,000 replicates. Scale bar represents branch lengths. The three populations (I, II 13
and III) inferred by Bayesian model-based analysis with program Structure 2.3.3 are circled and 14
marked accordingly. 15
16
17
18
19
Page 37 of 43
Accep
ted
Man
uscr
ipt
Fig. 4. Trypanosoma evansi isolates from Mindanao buffaloes classified into three populations I 20
(blue), II (pink) and III (green) based on analysis of population structuring by the software 21
program Structure 2.3.3. The size of the symbol reflects the number of isolates of that type 22
collected from that location. The three provinces from which samples were collected and 23
analysed were Agusan del Sur (ADS) and Surigao del Sur (SDS) and Agusan del Norte (ADN) 24
(no buffalo isolates collected). 25
26
27
28
Page 38 of 43
Accep
ted
Man
uscr
ipt
Fig. 5. Geographical representation of the occurrences of Trypanosoma evansi MLG A () and 29
MLG C () isolates from buffalo in Mindanao, Philippines in relation to buffalo mortality 30
figures (). The size of the MLG symbols reflects number of isolates and the size of buffalo 31
mortality symbol the buffalo mortality percentage. The three provinces from which samples were 32
collected and analysed were Agusan del Sur (ADS) and Surigao del Sur (SDS) and Agusan del 33
Norte (ADN) (no buffalo isolates collected). Major roads in provinces are marked. Arrows 34
identify the location of MLG isolates which occurred in another province to all the other isolates 35
of the respective MLG. 36
37
Page 39 of 43
Accep
ted
Man
uscr
ipt
Figure
Page 40 of 43
Accep
ted
Man
uscr
ipt
Figure
Page 41 of 43
Accep
ted
Man
uscr
ipt
Figure
Page 42 of 43
Accep
ted
Man
uscr
ipt
Figure
Page 43 of 43
Accep
ted
Man
uscr
ipt
Figure