To investigate the foot-and-mouth disease virus (FMDV) serotypes circulating in Uganda’s cattle population, both serological and virological analyses of samples from outbreaks that occurred during 2012–2013 were performed. Altogether, 79 sera and 60 oropharyngeal fluid (OP)/tissue/oral swab samples were collected from herds with reported FMD outbreaks in seven different Ugandan districts. Overall, 61/79 (77%) of the cattle sera were positive for antibodies against FMDV by PrioCHECK FMDV NS ELISA and solid phase blocking ELISA detected titres ≥ 80 for serotypes O, SAT 1, SAT 2 and SAT 3 in 41, 45, 30 and 45 of these 61 seropositive samples, respectively. Virus neutralisation tests detected the highest levels of neutralising antibodies (titres ≥ 45) against serotype O in the herds from Kween and Rakai districts, against SAT 1 in the herd from Nwoya district and against SAT 2 in the herds from Kiruhura, Isingiro and Ntungamo districts. The isolation of a SAT 2 FMDV from Isingiro was consistent with the detection of high levels of neutralising antibodies against SAT 2; sequencing (for the VP1 coding region) indicated that this virus belonged to lineage I within this serotype, like the currently used vaccine strain. From the Wakiso district 11 tissue/swab samples were collected; serotype A FMDV, genotype Africa (G-I), was isolated from the epithelial samples. This study shows that within a period of less than one year, FMD outbreaks in Uganda were caused by four different serotypes namely O, A, SAT 1 and SAT 2. Therefore, to enhance the control of FMD in Uganda, there is need for efficient and timely determination of outbreak virus strains/serotypes and vaccine matching. The value of incorporating serotype A antigen into the imported vaccines along with the current serotype O, SAT 1 and SAT 2 strains should be considered.
Citation: Namatovu A, Tjørnehøj K, Belsham GJ, Dhikusooka MT, Wekesa SN, Muwanika VB, et al. (2015) Characterization of Foot-And-Mouth Disease Viruses (FMDVs) from Ugandan Cattle Outbreaks during 2012-2013: Evidence for Circulation of Multiple Serotypes. PLoS ONE 10(2): e0114811. https://doi.org/10.1371/journal.pone.0114811
Received: September 2, 2014; Accepted: October 19, 2014; Published: February 9, 2015
Copyright: © 2015 Namatovu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: Accession numbers for the new nucleotide sequences can be found on Genbank, as follows: Serotype A sequences U74/13 (KP089984) U75/13 (KP089985) Serotype SAT 2 sequences U30/13 (KP089986) U35/13 (KP089987) U45/13 (KP089988) U69/13 (KP089989) U73/13 (KP089990)
Funding: This work was funded by the Danish International Development Agency (DANIDA) under the Transboundary Animal Diseases in East Africa project, DFC no. 10-006KU (HS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Foot-and-mouth disease (FMD) is a highly infectious disease of cloven hoofed animals characterized by the formation of vesicles in, and around, the mouth and on the feet [1–3]. The disease is caused by infection with FMDV (genus Aphthovirus, family Picornaviridae) which exists in seven antigenically diverse serotypes (O, A, C, Asia 1, SAT 1, SAT 2 and SAT 3) [4, 5] that cause indistinguishable clinical disease . FMD is endemic in Uganda with outbreaks occurring frequently ; cattle can show overt clinical signs, while it is generally subclinical in small ruminants [1, 8, 9]. Although mortality is generally low, this disease causes significant economic losses through reduction in milk production, loss of draught power and loss of access to profitable international livestock and livestock product markets [10–12]. Thus, control of this disease holds the potential to enhance food security, poverty alleviation and national development [11, 13].
In Uganda, control strategies for FMD outbreaks include quarantine and ring vaccination of cattle using imported trivalent vaccines (O, SAT 1 and SAT 2) . However, the success of these efforts is hampered by uncontrolled animal movements, inadequate surveillance and delayed reporting of FMD outbreaks. Since vaccination against one of the seven FMDV serotypes does not protect against other serotypes , it is important to know which serotypes are circulating. Moreover, variation between FMDV strains within a given serotype may result in poor coverage and may necessitate matching of one or more vaccine strains against the circulating FMDVs , which is still a challenge in East Africa .
According to Vosloo et al., all FMDV serotypes, other than Asia 1, have been detected in East Africa, however, serotype C has not been isolated since 2004 [18, 19]. In Uganda, the first FMD outbreak in cattle was reported in 1953 , with serotype O being responsible for the majority of the subsequent outbreaks. According to recent studies on Ugandan outbreaks from 2006 to 2011, topotype EA-2 serotype O FMDVs have been isolated, while the current O serotype vaccine strain incorporated in the imported trivalent vaccines belongs to the EA-1 topotype [20–22]. Other than serotype O FMDV, serotype A and SAT 2 viruses have been identified in cattle in 2002 and 2004, respectively [14, 23], while serotypes SAT 1, SAT 2 and SAT 3 FMDVs have been reported in Ugandan African buffalo (Syncerus caffer) [24, 25]. Very recent characterization of Ugandan and Kenyan FMDV outbreak strains disclosed simultaneous outbreaks with different strains of serotype O [22, 26] and separately SAT 2 viruses , emphasizing the necessity for prompt and accurate diagnosis, including regular typing of circulating strains, for effective control measures to be implemented .
Uganda is currently at stage 1 of the FAO/OIE defined Progressive Control Pathway for FMD (PCP-FMD) [17, 28] and to progress along this pathway, towards improved control of FMD, it is important to generate more knowledge about the epidemiology of FMD in the country. The aim of this study was to characterize the FMDVs responsible for seven of the Ugandan 2012–2013 FMD outbreaks in cattle.
Materials and Methods
Study area, sampling strategy and sample collection
This study was performed using outbreak samples collected from Ugandan districts with reported FMD outbreaks in 2012–2013. In 2012, 33 cattle sera were submitted to the National Animal Disease Diagnostic and Epidemiological Centre (NADDEC) from three districts (Kiruhura, Kween and Nwoya) (Fig. 1). In 2013, oral epithelial tissues, oral swabs and/or oropharyngeal fluids (OPs) were collected from one outbreak herd in each of Isingiro, Ntungamo, Rakai and Wakiso districts and in addition sera were collected from these herds in Isingiro, Ntungamo and Rakai, totaling 46 sera, 30 OPs, 16 epithelial tissues and 14 oral swabs. These samples were collected following requests by the respective District Veterinary Officers (DVOs) and subsequent approval by the Commissioner Animal Health, Ministry of Agriculture Animal Industry and Fisheries. The DVOs sought the consent of each herd owner before sampling, and samples were collected from, Isingiro District (GPS: 36M 0248868, UTM 9918082), Ntungamo District (GPS: 36M 019262, UTM 989192), Rakai District (GPS: 36M 0315178, UTM 9894397) and Wakiso District (GPS: 36N 0447204, UTM 0041487). No GPS coordinates are available for the samples submitted in 2012.The farmers were interviewed about the suspected source of the outbreak, previous outbreaks, last vaccination and type of vaccine used, while the DVOs were interviewed about the history of outbreaks and vaccination, the vaccine used and the instituted control measures.
The map shows the seven districts of Uganda that reported FMD outbreaks during 2012–2013. These districts were: Kiruhura, Kween and Nwoya during the 2012 outbreaks plus Isingiro, Ntungamo, Rakai and Wakiso during the 2013 outbreaks.
The epithelial tissues were preserved in 50% phosphate buffered saline (PBS) and 50% glycerol while the oral swabs and OPs were preserved using PBS as recommended by OIE . Sera were extracted from blood samples and stored at -20°C at NADDEC, while epithelial tissues, OPs and oral swabs were kept in liquid nitrogen during transportation to NADDEC and then stored at -80°C.
Detection of antibodies against FMDV non-structural proteins (NSPs). Sera were screened for antibodies against FMDV NSPs using the PrioCHECK FMDV NS kit (Prionics AG, Zurich Switzerland) following the manufacturer’s instructions except that optical density (OD) values were measured at dual wavelengths (450 nm and 620 nm) using a Multiskan Ascent spectrophotometer (Thermo Labsystems Oy, Helsinki, Finland). Results were expressed as percentage inhibition (PI) relative to the mean of the negative controls; PI = 100—((OD of test sample (OD450—OD620)/OD of mean negative control (OD450—OD620)) ×100), and sera with PI ≥ 50% were considered positive.
Detection of serotype-specific antibodies against FMDV using solid phase blocking ELISAs (SPBE). All samples that were positive in the Priocheck FMDV NS ELISA were tested for antibodies against each of the seven FMDV serotypes using SPBEs as described . Screening was done using a sample dilution of 1:10 and the results expressed as optical density percentage (ODP) using the formula; ODP = (OD of test sample (OD450nm–620nm) / mean OD of negative controls (OD450nm–620nm)) × 100. Samples were regarded as positive if ODP values were equal to or lower than: 50% for serotypes O, SAT 1, SAT 2 and SAT 3, 45% for serotype A and 35% for serotypes C and Asia 1 .
Positive reactions were titrated in the appropriate SPBEs (using two-fold dilution series from 1:10 to 1:1280). Titres were expressed as the reciprocal of the highest positive dilution and titres ≥ 80 were considered positive for each serotype .
Detection of serotype-specific neutralising antibodies against FMDV using virus neutralization tests (VNTs). For each district, selected, representative samples with titres ≥ 80 in the SPBEs were tested in the corresponding VNTs for the presence of neutralising antibodies. The protocol was adapted from the procedure described in the OIE terrestrial manual  and was described previously . Briefly, two-fold serum dilutions were reacted for 1 hour with 100 TCID50 of virus suspension (O Manisa, A Iraq 24/64, C Noville, Asia 1 IND 8/79, SAT 1 BOT/68, SAT 2 ZIM 5/81 or SAT 3 ZIM 4/81) in equal volumes, in 96-well plates. Subsequently, swine kidney cell suspension was added to the wells and the plates were incubated at 37°C for three days. The plates were examined for cytopathic effect (CPE) under the microscope and the 50% end-point titres were calculated according to Reed and Muench, . A titre ≥ 45 was considered positive, 16 – 44 was considered doubtful and < 16 negative.
Detection of FMDV by virus isolation (VI). Epithelial tissue preparation and VI procedures were adapted from the description in the OIE terrestrial manual  as described . Briefly, samples (OPs or oral swab suspension or 10% suspension of epithelial tissues in Eagles minimal essential medium including 2 million I.U. benzylpenicillin, 1g dihydro-streptomycin sulphate, 0.5g neomycin sulphate and 8.5¼g amphotericin per litre (MEM+)) were inoculated onto monolayers of primary bovine thyroid (BTY) cells in five wells using a dilution of 1:2 and in five wells using a dilution of 1:4 in a 96 well tissue culture plate. Epithelial samples were cultured on separate plates, while for OPs/oral swabs, samples from the same herd were cultured on the same plate but separated by wells with uninfected cells. The plates were incubated for 2 hours at 37°C (with 5% CO2), followed by a wash with MEM+ including 2% fetal calf serum (FCS) and addition of 150μl MEM+ with 2% FCS per well. The plates were incubated for 1–2 days, as above, and observed for CPE every 24 hours. CPE-positive samples were frozen for harvesting, while samples without CPE after 48 hours were harvested by freeze-thawing and passaged again on BTY monolayers as described above. All samples that did not show CPE following the second passage were considered negative.
Confirmation of FMDV in CPE-positive samples by antigen ELISA (Ag ELISA). CPE-positive harvests from all districts were tested in the Lindholm in-house antigen ELISA following a protocol based on the description in the OIE terrestrial manual  as described . Samples giving ODs > 0.2 higher than the negative control were considered positive, while those with ODs 0.1—0.2 above the negative control were inconclusive and those with ODs < 0.1 above the negative control were considered negative.
FMDV RNA extraction and detection using quantitative RT-PCR (RT-qPCR). RNA was extracted from FMDV antigen positive cell culture harvests using the Blood kit Viral RNA Mini Spin Protocol in the QIAamp Viral RNA Mini Kit (Qiagen, Hamburg, Germany). cDNA was synthesized using the Perkin Elmer (PE) Biosystem TaqMan kit (Perkin Elmer, Foster city, California, USA) and RT-qPCR was performed using the TaqMan PCR kit (PE Applied Biosystems, Foster city, California, USA) with FMDV 3D-F (ACTGGGTTTTACAAACCTGTGA) and FMDV 3D-R (GCG AGTCCTGCG ACG GA) primers and the 3D-P probe (TCCTTTGCACGCCGTGGGAC) labelled with 6-carboxyfluorescein as a reporter dye and tetramethylrhodamine as a quencher at the 5′ and 3′ ends respectively .
cDNA synthesis and sequencing of FMDV VP1 coding region. The cDNA for sequencing was synthesized using Ready-To-Go You-Prime First-strand Beads (GE Healthcare Life Sciences, Uppsala, Sweden) according to the manufacturer’s instructions using 28μl template and 7μl of a four primer mix (85 ¼l of pdN6 random hexamer (50 ng/¼l), 5 ¼l of T27V (TTTTTTTTTTTTTTTTTTTTTTTTTTTV) (25 pmol/¼l), 5 ¼l of KDS 1R (CCAGTCCCCTTCTCAGATC) (25 pmol/¼l) and 5 ¼l of 9-G PN 55 (GGGGGGGGGGGGGGH) (25 pmol/¼l)). The PCR products corresponding to the VP1-coding region were prepared for sequencing as previously described  except that the SAT 2 primers used were the forward 13-K PN 100 (5′–3′ GGGTGGBBGTSTWMCAGRTSACMGACAC) and reverse 1-O PN 15 (5′–3′ GAAGGGCCCAGGGTTGGACTC) (where B is C or G or T; S is G or C, W is A or T, M is A or C, R is A or G). For serotype A samples, the primers used were forward 13-K PN 101 (5′–3′ CACTGCTAYCAYKCNGARTGGGA) and reverse 10-P PN 34 (5′–3′ CAGGGTTGGACTCMACGTCTCC) (where M is A or C; Y is T or C; K is G or T; R is G or A and N is G or A or T or C). All primers were used at a concentration of 5 pmol/μl.
Gel purification of the PCR amplicons (795–942 bp) was performed using the GeneJET PCR purification kit (Thermo Fisher Scientific). Cycle sequencing was performed with 10–15 ng of PCR product (as determined using a Nanodrop spectrophotometer (Thermo Scientific, Wilmington, DE, USA)) and the BigDye terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City California, USA) on an ABI 3700 automated DNA Sequencer (Applied Biosystems).
Descriptive serological data were entered in Microsoft Excel and imported into Stata 9 (Statacorp, Lakeway Drive College Station, Texas, USA) where the analysis was performed using the median and interquartile range as a measure of central tendency and spread respectively. For molecular data, base calling of nucleotide sequences was performed using seqMan Pro software (Lasergene package, DNAstar, Inc., WI, USA). Published sequences relevant to this study were obtained using BLAST searches from NCBI (http://www.ncbi.nlm.nih.gov/). The sequences analyzed correspond to the VP1 coding region and details of the new and previously published sequences are summarised in Table 1. Multiple sequences were aligned using MUSCLE (codons)  incorporated in MEGA 5 . A phylogenetic tree of these sequences was estimated using the Neighbor-Joining method . The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (2000 replicates) is shown next to the branches . The tree is drawn to scale with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Tamura 3-parameter method  and are in the units of the number of base substitutions per site. The rate variation among sites was modelled with a gamma distribution (shape parameter = 1). The analysis for serotype A and SAT 2 involved 24 and 17 different nucleotide (nt) sequences, respectively, and codon positions included were 1st+2nd+3rd. There were a total of 638 and 648 nt positions in the final dataset for serotype A and SAT 2, respectively. Evolutionary analyses were conducted in MEGA 5 .
Detection of antibodies against FMDV NSPs
Altogether 61 of the 79 (77%) cattle sera collected from herds with reported FMD outbreaks had antibodies against the FMDV NSPs with herd prevalences ranging from 33% to 93% (Table 2).
Detection of serotype-specific antibodies against FMDV using SPBE
Antibody titres ≥ 80 against one or more of all serotypes except Asia 1 were found using SPBEs in 59 of the 61 samples that had tested positive for antibodies against FMDV NSPs. Thus, 41, 7, 1, 45, 30 and 45 samples had SPBE titres ≥ 80 against serotypes O, A, C, SAT 1, SAT 2 and SAT 3, respectively (Table 2). One sample with titre 80 in SPBE for antibodies against C FMDV (from Isingiro) and four of the seven samples with titres 80–160 in SPBE for antibodies against A FMDV (from Kween) had much higher titres against two or more of serotypes O, SAT 1, SAT 2 and SAT 3. A fifth sample from Kween and two samples from Ntungamo had SPBE A titre 160 as the highest titre. Seven sera had titres ≥ 80 against only one serotype (A: 1; SAT 1: 1; SAT 2: 3; SAT 3: 2), while 52 sera had titres ≥ 80 against multiple serotypes (data for individual animals can be found in the S1 Table). An overall evaluation of SPBE titres pointed towards the outbreaks in Nwoya and Ntungamo being caused by SAT 1 and SAT 2 FMDV, respectively, while the one in Kween was most likely caused by O FMDV and the one in Rakai by either O or SAT 1 FMDV. The SPBE results were very complicated in Isingiro and Kiruhura, where very high titres against three or more of serotypes O, SAT 1, SAT 2 and SAT 3 in most sera impeded a serotype-specific diagnosis based on these assays.
Detection of neutralising antibodies using VNTs
Serum samples with antibody titres ≥ 80 in a serotype-specific SPBE were tested in VNTs for antibodies against the same serotype if sufficient serum was available. None of the SPBE antibody titres of 80–160 against serotype C and A were confirmed in the VNTs. Furthermore, all but three of 42 samples with SPBE antibody titres ≥ 80 against serotype SAT 3 were negative or inconclusive in the corresponding VNT. However, the number of SPBE positive samples with neutralising antibodies against serotypes O, SAT 1 and SAT 2 (titres ≥ 45) were 24/41, 13/38 and 25/25, respectively (Table 2). The VNTs determined high antibody titres against serotype O FMDV in all tested sera from Kween and Rakai with lower titres against SAT 1 in some sera from Rakai, while the detection of anti-SAT 1 antibodies in the SPBE was confirmed for samples from Nwoya (Table 2). All SPBE positive sera from Ntungamo, Kiruhura and Isingiro tested in VNT had high neutralizing titres against SAT 2 FMDV with lower titres against other serotypes in some sera (Table 2). The three sera with neutralising antibodies against SAT 3 came from three different herds/districts and had higher neutralising antibody titres against at least one other serotype (O, SAT 1, SAT 2) (see S1 Table), indicating that these apparent anti-SAT 3 reactions were likely cross reactivity generated by exposure to other serotypes.
Virus characterization by VI, Ag ELISA, RT-qPCR and sequencing
Eight of 30 OPs, six of 16 epithelial samples and two of 14 oral swabs produced CPE during one or two passages on primary BTY cells. These samples were from the districts of Isingiro (eight), Ntungamo (two), Rakai (three) and Wakiso (three). The FMDV antigen ELISA identified serotype SAT 2 FMDV antigen in cell culture harvests from one OP, three epithelia and one oral swab from Isingiro and serotype A FMDV antigen in those from three epithelia from Wakiso (data not shown). Moreover, these eight virus harvests had Ct-values < 25 in RT-qPCR assays. From these harvests, amplicons corresponding to the complete VP1 coding region were generated by RT-PCR and sequenced. The VP1 coding sequences, between 633 and 642 nucleotides in length, were determined and compared with other sequences from NCBI, Genbank, using BLAST. The Isingiro and Wakiso sequences were most closely matched to other SAT 2 and A FMDV sequences, respectively (Table 1), consistent with the Ag ELISA data on these virus isolates. Identification of SAT 2 as the serotype responsible for the outbreak in the Isingiro district was consistent with the high titres of neutralising antibodies against SAT 2 detected in the sera from this district (Table 2).
Phylogenetic analysis of serotype A FMDVs
Assessment of phylogenetic relationships using selected, genotype-defined serotype A FMDV isolates (Table 1) showed that the VP1 sequences of the Wakiso strains (U74/2013 and U75/2013) clustered with other strains belonging to the Africa genotype (G-I) (Fig. 2). The two Wakiso strains had 100% nt identity with each other within the VP1 coding region and so had the same predicted amino acid sequences (Fig. 3). The Wakiso viruses had pairwise nt identity of 93% with both K3/2013 and K148/12 Kenyan isolates and amino acid (aa) identity of 93% and 92% for K3/2013 and K148/12 respectively. Thus they belonged to the same lineage but were in different sub-lineages [37–39]. The Wakiso sequences had pairwise nt identity of 82% (521/633) and 82% (517/633), and pairwise aa identity of 87% and 89% (Fig. 3), with the K5/1980 (Africa (G-I)) and K35/1980 (Africa (G-VII)) vaccine strains currently produced in Kenya, respectively . Hence, based on a cut off of > 15% nt difference in the VP1 coding region of FMDV (non-SAT serotypes) for separating genotypes [39, 41], these viruses belonged to different genotypes than both the current Kenyan vaccine strains.
The two 2013 Ugandan outbreak FMDV sequences are marked with asterisks (*), and K35/1980** and K5/1980** are the current serotype A virus vaccine strains produced in Kenya and last used in Uganda in 2002. Bootstrap values greater than 50% are marked on the tree. Two of the three global topotypes within serotype A (Africa and Asia) are indicated, as are some of the African genotypes found in Eastern Africa (G-I, G-IV and G- VII). The topotype Asia serotype A FMD viruses were used to out root the tree.
The VP1 amino acid sequences were inferred from one of the Ugandan 2013 serotype A FMDV nucleotide sequences (marked with an asterisk (*)) and selected African serotype A FMDV strains. The dots indicate identity with the vaccine strain sequence (K5/1980**) that is the reference sequence.
One hundred and twelve (18%) variable sites between the Wakiso serotype A VP1 sequences and the vaccine strain (K5/1980) were determined across the 633 nucleotides, which encode substitutions at 27/211 amino acids (13%) in the deduced protein sequence (Fig. 3). The integrin receptor binding motif, RGD, (residues 144–146) within the Wakiso viruses was conserved but there were a number of amino acid substitutions in the region flanking the RGD cell attachment site (up to residues +3 and from -1 to -9). These amino acid variations occurred within known antigenic regions; for example in antigenic site 1 [42, 43] methionine (M) was replaced by leucine (L) at residue 147 while alanine (A) was replaced by glycine (G) at position 148.
Phylogenetic assessment of SAT 2 FMDVs
Except for one nucleotide substitution at position 339 (C for T) in the sequence of the virus isolate U35/13, the five Isingiro virus sequences from 2013 were identical within the VP1 coding sequence. Assessment of phylogenetic relationships with selected SAT 2 FMDV isolates (Table 1) showed that the five Isingiro viruses clustered with known strains belonging to the SAT 2 lineage I (Fig. 4). Comparison of these recent viruses with the 2004 Ugandan SAT 2 cattle isolate (UGA/11/2004), a partial VP1 sequence (345 nt), showed that they belonged to the same genotype/lineage (pairwise nt identity of 90% (309/345 nt) and aa identity of 94% (108/115 aa)) [4, 41]. However, based on the complete VP1 sequence, the Isingiro viruses were most closely related to a recent Kenyan isolate (K125/12) (Wekesa S., unpublished data) having a pairwise nt identity of 99% and aa identity of 100%, and were also closely related to a Tanzanian isolate from 2009 (TAN/43/2009) (pairwise nt identity of 97% and aa identity of 98%).
The relationships between the Ugandan 2013 SAT 2 FMD viruses and selected East African SAT 2 viruses are shown. The five 2013 Ugandan and one 2012 Kenyan outbreak FMDV (K125/12) sequences are marked with asterisks (*), while the strain K52/84** is the SAT 2 FMD virus strain incorporated in the trivalent vaccine (O, SAT 1 and SAT 2) as used in Uganda.
The Ugandan 2013 SAT 2 sequences had 87% nt identity (557/642 nt) and pairwise aa identity of 94% with the SAT 2 vaccine strain (K52/84) which is incorporated in the trivalent vaccines (O, SAT 1 and SAT 2) currently used in Uganda. Therefore, these viruses belonged to the same lineage as the vaccine strain.
Eighty five (13%) variable sites were determined across the 642 nt between the Isingiro 2013 SAT 2 outbreak sequences and the vaccine strain (K52/84) which encode substitutions of 12/214 (6%) amino acids in the deduced protein sequence (Fig. 5). The RGD motif (residues 144–146) within these viruses was conserved and so was the flanking region of the RGD motif as far as the -5 and +10 positions (Fig. 5). Conservative changes occurred at residue 138 where Asp (D) was replaced by Glu (E) and at position 157 where Ser (S) was replaced by Thr (T), both residues are located in the flanking region of the RGD motif cell attachment site [44, 45] which also constitutes an important antigenic site.
The VP1 amino acid sequences were inferred from one of the 2013 outbreak nucleotide sequences (marked with an asterisk (*)) and earlier SAT 2 FMDV strains. Identities with the reference vaccine strain sequence (K52/84**) are indicated by dots. Similar amino acid variation exists between the reference sequence and the Ugandan 2013 isolates and a Kenyan outbreak isolate (K125/12*) sequence (unpublished data).
The aim of this study was to characterize the serotypes of FMDV responsible for Uganda’s 2012–2013 outbreaks. The detection of antibodies against FMDV NSPs in sera from herds with recent reports of clinical signs of FMD strongly supported the clinical diagnosis. Moreover, the higher seroprevalences (60–92%) of antibodies against FMDV NSPs in herds from Isingiro, Kiruhura, Kween, Ntungamo and Rakai districts probably indicate that sampling was done after considerable spread of the infection within these herds, while the lower seroprevalence in the herd in Nwoya (33%) likely indicates sampling at an earlier phase of the outbreak .
Determining the serotype of the causal FMDV based on the SPBEs was not straightforward, since 87% of the samples had antibody titres ≥ 80 against more than one serotype and with most herds having high titres against serotypes O, SAT 1, SAT 2 and SAT 3. In addition, the VNT also detected neutralising antibodies against more than one serotype in Ntungamo, Kiruhura and Isingiro. The cross-reactions observed between the SPBEs, and to a lesser extent between the VNTs, are in accordance with previous reports [9, 46–49], and have for the SPBEs been observed to be more prominent for the first 1–4 weeks after experimental infection of calves (Tjørnehøj, K., unpublished data). The detection of antibodies against multiple serotypes may indicate some antigenic relationship between the virus serotypes , or could be due to a complicated serological picture with heterotypic responses caused by priming effects from past exposures from one or more infections and/or vaccinations [9, 46, 49]. Nevertheless, overall evaluation of the SPBE titres limited the likely serotype responsible for the six outbreaks that were investigated by serology to only one serotype for three outbreaks and to just two serotypes in another outbreak. Moreover, the VNT, performed on selected SPBE positive sera, clearly identified the responsible FMDV serotype in all six outbreaks (as O, SAT 1 or SAT 2).
The serological identification of SAT 2 as the responsible virus in Isingiro was confirmed by the isolation of SAT 2 FMDV from clinical material. No serum samples were available from the Wakiso district but serotype A FMDV was successfully isolated and characterized by antigen ELISA and VP1 sequencing. Unfortunately, it was not possible to isolate viruses from the samples collected from the outbreaks in Isingiro, Ntungamo and Rakai. However, the Isingiro district borders the Kiruhura and Ntungamo districts, and since sera from these districts had the highest antibody titres against SAT 2, it is not unlikely that the outbreaks in these districts were all part of the same SAT 2 FMDV outbreak.
The 2013 outbreak of serotype A virus in the Wakiso district was close to the capital city (Kampala); this serotype has not been reported in Uganda since 2002 . The population of cattle in this area is low and thus this FMD outbreak was suspected to have originated from another country, possibly via feed or fomites or newly introduced animals. The two identical (within VP1) serotype A FMDVs belonged to a different sub-lineage from those recently found in neighbouring Kenya (K148/2012 and K3/2013) (7% nt difference), hence, the Ugandan serotype A FMDV outbreak was not related to the Kenyan 2012–2013 outbreak. The VP1 sequence of the isolates had 18% diversity from both the K5/1980 (G-I) and K35/1980 (G-VII) vaccine strains. In fact, the Wakiso FMDVs cluster with viruses within a new genotype that has arisen from within the Africa topotype (G-I) as discussed recently for Kenyan FMD viruses .
The five SAT 2 isolates from Isingiro were collected from a single epidemiological unit and the VP1 coding sequences were identical except for one nt change in one of the five isolates and belonged to lineage (I) within the SAT 2 strains. Moreover, these isolates were only distantly related to the most recent Ugandan SAT 2 cattle isolate from 2004 (UGA/11/2004) (10% nt difference) , but were closely related to a recent Kenyan isolate (K125/12) (1% nt difference) and a Tanzanian isolate from 2009 (TAN43/2009) (3% nt difference). This provides further evidence for the transboundary mobility of FMDV in the East African region [20, 22], probably due to the increased, uncontrolled, cross border movement given the growth of cross border livestock trade . The 2013 Ugandan SAT 2 VP1 sequences had 13% nt difference from the vaccine strain (K52/84) which showed that they were rather different although belonging to the same genotype .
The current results indicate resurgence of serotypes A and SAT 2 FMDV in Uganda. The significant diversity between both virus serotypes and their respective current vaccine strains included multiple amino acid variations (Figs. 3 and 5), especially in the critical antigenic regions. This indicates a need for vaccine matching studies to establish the level of protection conferred against the currently circulating viruses by the available vaccine strains.
In conclusion, the presence of antibodies against FMDV in sera collected from six unvaccinated herds following reports of FMD outbreaks supported the FMD diagnosis, and the FMDV serotypes responsible for the outbreaks were determined as O, SAT 1 and SAT 2 by the use of SPBEs and VNTs. The serology based serotyping of the outbreaks was made more complicated due to the cross-reactions identified with the SPBEs and also, to a lesser degree, with VNTs. The FMDV responsible for one of these outbreaks was confirmed by isolation of serotype SAT 2 FMDV and the FMDV involved in a seventh outbreak was identified by isolation of serotype A FMDV. The serotype A isolates were rather distantly related to the serotype A FMDV vaccine strains incorporated in the vaccines imported to Uganda in 2002 . In contrast, the Ugandan 2013 SAT 2 viruses belonged to the same lineage, (I), as the FMDV vaccine strain incorporated in the trivalent vaccines (O, SAT 1 and SAT 2) imported by Uganda but have amino acid substitutions within antigenic sites. Therefore, to enhance the control of FMD in Uganda, there is need for increased surveillance, including monitoring with efficient and timely characterization of outbreak virus strains, as well as vaccine matching. The value of incorporating an appropriate serotype A FMDV vaccine strain into the imported vaccines used in Uganda should be assessed.
The authors extend their special thanks to: Martin Essau, Eugene Arinaitwe and Simon Peter Ruhweza at NADDEC for their assistance during sample collection; Jani Christiansen, Jane Borch, Inge Nielsen and Preben Normann at the National Veterinary Institute, Technical University of Denmark (DTU), Lindholm for their technical assistance and Mr. Bernard Mwesigwa of FAO, Uganda, for assistance with drawing the map.
Conceived and designed the experiments: AN CA KT GB HS VM. Performed the experiments: AN CA MD SW. Analyzed the data: AN KT GB HS CA VM. Wrote the paper: MD SW KT GB HS CA VM.
- 1. Alexandersen S, Zhang Z, Donaldson AI, Garland AJ (2003) The pathogenesis and diagnosis of foot-and-mouth disease. J Comp Pathol 129: 1–36. pmid:12859905
- 2. Grubman MJ, Baxt B (2004) Foot-and-mouth disease. Clin Microbiol Rev 17: 465–493. pmid:15084510
- 3. Alexandersen S, Mowat N (2005) Foot-and-mouth disease: host range and pathogenesis. Curr Top Microbiol Immunol 288: 9–42. pmid:15648173
- 4. Knowles NJ, Samuel AR (2003) Molecular epidemiology of foot-and-mouth disease virus. Virus Res 91: 65–80. pmid:12527438
- 5. Stanway G, Brown F, Christian P, Hovi T, Hyypia T, et al. (2005) FamilyPicornaviridae. In Fauquet CM, Mayo MA, Maniloff J, Desselberger U, Ball LA,editors. Virus Taxonomy: classification and nomenclature of viruses. Eighth Report of the International Committee on Taxonomy of Viruses. Elsevier/Academic Press, London. pp. 757–778.
- 6. Vosloo W, Bastos AD, Sangare O, Hargreaves SK, and Thomson GR (2002) Review of the status and control of foot-and-mouth disease in sub-Saharan Africa. Rev Sci Tech 21: 437–449. pmid:12523685
- 7. Ayebazibwe C, Tjornehoj K, Mwiine FN, Muwanika VB, Okurut AR, et al. (2010) Patterns, risk factors and characteristics of reported and perceived foot-and-mouth disease (FMD) in Uganda. Trop Anim Health Prod 42: 1547–1559. pmid:20526861
- 8. Balinda SN, Tjornehoj K, Muwanika VB, Sangula AK, Mwiine FN, et al. (2009) Prevalence estimates of antibodies towards foot-and-mouth disease virus in small ruminants in Uganda. Transbound Emerg Dis 56: 362–371. pmid:19909475
- 9. Mwiine FN, Ayebazibwe C, Olaho-Mukani W, Alexandersen S, Balinda SN, et al. (2010) Serotype specificity of antibodies against foot-and-mouth disease virus in cattle in selected districts in Uganda. Transbound Emerg Dis 57: 365–374. pmid:20696028
- 10. Kitching RP (2002) Clinical variation in foot-and-mouth disease: cattle. Rev Sci Tech 21: 499–504. pmid:12523690
- 11. Rutagwenda T (2003) The implications and consequences of foot-and-mouth disease outbreak on food security and household incomes in selected districts in Uganda; A draft consultancy report (Study report 2): FAO, Rome (Italy).Animal Production and Health Div. TCP/UGA/ 0168 (E) 1.
- 12. Rweyemamu MM, Leforban Y (1999) Foot-and-mouth disease and international developments. Adv Virus Res 53: 111. pmid:10582097
- 13. Perry BD, Rich KM (2007) Poverty impacts of foot-and-mouth disease and the poverty reduction implications of its control. Vet Rec 160: 238–241. pmid:17308024
- 14. Muleme M, Barigye R, Khaitsa ML, Berry E, Wamono AW, et al. (2012) Effectiveness of vaccines and vaccination programs for the control of foot-and-mouth disease in Uganda, 2001–2010. Trop Anim Health Prod 45: 35–43. pmid:22956440
- 15. Domingo E, Baranowski E, Escarmis C, Sobrino F (2002) Foot-and-mouth disease virus. Comp. Immunol. Microbiol. Infect. Dis. 25: 297–308.
- 16. OIE (2012) Foot-and-mouth disease [Chapter 2.1.5]. In: Manual of diagnostic tests and vaccines for terrestrial animals 2012. OIE, Paris, France http://www.oie.int/fileadmin/Home/eng/Health_standards/tahm/2.01.05_FMD.pdf. Accessed on 31/08/2013.
- 17. Namatovu A, Wekesa SN, Tjornehoj K, Dhikusooka MT, Muwanika VB, et al. (2013) Laboratory capacity for diagnosis of foot-and-mouth disease in Eastern Africa: implications for the progressive control pathway. BMC Vet Res 9: 19. pmid:23347795
- 18. Sangula AK, Siegismund HR, Belsham GJ, Balinda SN, Masembe C, et al. (2010) Low diversity of foot-and-mouth disease serotype C virus in Kenya: evidence for probable vaccine strain re-introductions in the field. Epidemiol Infect.
- 19. Roeder PL, Knowles NJ, (2009) Foot-and-mouth disease virus type C situation: the first target for eradication. In the global control of FMD—Tools, Ideas and Ideals. FAO, Erice, Italy, Rome.
- 20. Balinda SN, Sangula AK, Heller R, Muwanika VB, Belsham GJ, et al. (2010) Diversity and transboundary mobility of serotype O foot-and-mouth disease virus in East Africa: implications for vaccination policies. Infect Genet Evol 10: 1058–1065. pmid:20619358
- 21. Kasambula L, Belsham GJ, Siegismund HR, Muwanika VB, Ademun-Okurut AR, et al. (2012) Serotype identification and VP1 coding sequence analysis of foot-and-mouth disease viruses from outbreaks in eastern and northern Uganda in 2008/9. Transbound Emerg Dis 59: 323–330. pmid:22117844
- 22. Namatovu A, Belsham GJ, Ayebazibwe C, Dhikusooka MT, Wekesa SN, et al. (2013) Challenges for serology-based characterization of foot-and-mouth disease outbreaks in endemic areas; identification of two separate lineages of serotype O FMDV in Uganda in 2011. Transbound Emerg Dis.
- 23. Balinda SN, Belsham GJ, Masembe C, Sangula AK, Siegismund HR, et al. (2010) Molecular characterization of SAT 2 foot-and-mouth disease virus from post-outbreak slaughtered animals: implications for disease control in Uganda. Epidemiol Infect 138: 1204–1210. pmid:20003615
- 24. Kalema-Zikusooka G, Bengis RG, Michel AL, and Woodford MH (2005) A preliminary investigation of tuberculosis and other diseases in African buffalo (Syncerus caffer) in Queen Elizabeth National Park, Uganda. Onderstepoort J Vet Res 72: 145–151.
- 25. Ayebazibwe C, Mwiine FN, Tjørnehøj K, Balinda SN, Muwanika VB, et al. (2010) The role of African buffalos (Syncerus caffer) in the maintenance of foot-and-mouth disease in Uganda. BMC Vet Res 6: 1–13.
- 26. Wekesa SN, Muwanika VB, Siegismund HR, Sangula AK, Namatovu A, et al. (2013) Analysis of recent serotype O foot-and-mouth disease viruses from livestock in Kenya: evidence of four independently evolving lineages. Transbound Emerg Dis.
- 27. Sangula AK, Belsham GJ, Muwanika VB, Heller R, Balinda SN, et al. (2010) Co-circulation of two extremely divergent serotype SAT 2 lineages in Kenya highlights challenges to foot-and-mouth disease control. Arch Virol 155: 1625–1630. pmid:20614146
- 28. FAO/OIE (2011) Principles, stage description and standards. In the Progressive control pathway for FMD control (PCP-FMD). http://www.faoorg/ag/againfo/commissions/docs/PCP/PCP-26012011pdf 2011. Accessed on 02/02/2014.
- 29. Reed LJ, Muench H (1938) A simple method of estimating fifty per cent endpoints. Am J Hyg 27: 493–497.
- 30. Callahan JD, Brown F, Csorio FA, Sur JH, Kramer E, et al. (2002) Use of a portable real-time reverse transcriptase-polymerase chain reaction assay for rapid detection of foot-and-mouth disease virus. J Am Vet Med Assoc 220: 1636–1642. pmid:12051502
- 31. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215: 403–410. pmid:2231712
- 32. Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 32: 1792–1797. pmid:15034147
- 33. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739. pmid:21546353
- 34. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425. pmid:3447015
- 35. Felsenstein J (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783–791.
- 36. Tamura K (1992) Estimation of the number of nucleotide substitutions when there are strong transition-transversion and G + C-content biases. Mol Biol Evol 9: 678–687. pmid:1630306
- 37. Samuel A, Knowles N, Mackay D (1999) Genetic analysis of type O viruses responsible for epidemics of foot-and-mouth disease in north Africa. Epidemiol Infect 122: 529–538. pmid:10459658
- 38. Mohapatra JK, Subramaniam S, Pandey LK, Pawar SS, De A, et al. (2011) Phylogenetic structure of serotype A foot-and-mouth disease virus: global diversity and the Indian perspective. J Gen Virol 92: 873–879. pmid:21228130
- 39. Samuel A, Knowles N (2001) Foot-and-mouth disease type O viruses exhibit genetically and geographically distinct evolutionary lineages (topotypes). J Gen Virol 82: 609–621. pmid:11172103
- 40. Wekesa SN, Sangula AK, Belsham GJ, Muwanika VB, Heller R, et al. (2014) Genetic diversity of serotype A foot-and-mouth disease viruses in Kenya from 1964 to 2013; implications for control strategies in eastern Africa. Infect Genet Evol 21: 408–417. pmid:24368254
- 41. Vosloo W, Knowles N J, Thomson GR (1992) Genetic relationships between southern African SAT-2 isolates of foot-and-mouth disease virus. Epidemiol Infect 109: 547–558. pmid:1334842
- 42. Baxt B, Vakharia V, Moore D, Franke A, Morgan D (1989) Analysis of neutralizing antigenic sites on the surface of type A12 foot-and-mouth disease virus. J Virol 63: 2143–2151. pmid:2467993
- 43. Kitson JD, McCahon D, Belsham GJ (1990) Sequence analysis of monoclonal antibody resistant mutants of type O foot and mouth disease virus: evidence for the involvement of the three surface exposed capsid proteins in four antigenic sites. Virology 179: 26–34. pmid:1699353
- 44. Crowther J, Rowe C, Butcher R (1993) Characterization of monoclonal antibodies against a type SAT 2 foot-and-mouth disease virus. Epidemiol Infect 111: 391–406. pmid:7691630
- 45. Sangare O, Bastos A, Venter E, Vosloo W (2004) A first molecular epidemiological study of SAT-2 type foot-and-mouth disease viruses in West Africa. Epidemiol Infect 132: 525–532. pmid:15188721
- 46. Hedger RS, Barnett ITR, Gradwell DV, Travassos Dias P (1982) Serological tests for foot-and-mouth disease in bovine serum samples. Problems of interpretation. Rev Sci Tech 1: 387–393.
- 47. Hamblin C, Barnett ITR, Crowther JR (1986) A new enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against foot-and-mouth disease virus. II. Application. J Immunol Methods 93: 123–129.
- 48. Hamblin C, Barnett ITR, Hedger RS (1986) A new enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against foot-and-mouth disease virus. I. Development and method of ELISA. J Immunol Methods 93: 115–121.
- 49. Mwiine FN, Ayebazibwe C, Alexandersen S, Olaho- Mukani W, Okurut Ademun A R. et al. (2010) Seroepidemiological investigation of foot-and-mouth disease virus serotypes in cattle around Lake Mburo National Park in south-western Uganda. J Vet Med Anim Health 2: 46–54.
- 50. COMESA (2009) Hidden value on a hoof: cross-border livestock trade in eastern Africa. Common market for eastern and southern Africa comprehesive agriculture development programme. Policy brief No. 2 http://www.caadp.net/pdf/COMESA%20CAADP%20Policy%20Brief%202%20Cross%20Border%20Livestock%20Trade%20(2).pdf.
- 51. Kasanga CJ, Wadsworth J, Mpelumbe‐Ngeleja CAR, Sallu R, Kivaria F, et al. (2014) Molecular characterization of foot‐and‐mouth disease viruses collected in Tanzania between 1967 and 2009. Transbound Emerg Dis.
- 52. Knowles NJ, Nazem Shirazi MH, Wadsworth J, Swabey KG, Stirling JM, et al. (2009) Recent spread of a new strain (A‐Iran‐05) of foot‐and‐mouth disease virus type A in the Middle East. Transbound Emerg Dis 56: 157–169. pmid:19432637
- 53. Ayelet G, Mahapatra M, Gelaye E, Egziabher BG, Rufeal T, et al. (2009) Genetic characterization of foot-and-mouth disease viruses, Ethiopia, 1981–2007. Emerg Infect Dis 15: 1409–1417. pmid:19788808
- 54. Habiela M, Ferris NP, Hutchings GH, Wadsworth J, Reid SM, et al. (2010) Molecular characterization of foot‐and‐mouth disease viruses collected from Sudan. Transbound Emerg Dis, 57: 305–314. pmid:20626708