This cross-sectional study was carried out from January 2016 to June 2018 across 13 counties in Kenya where camels are reared (Table 1 and Fig. 1). The four camel breed types in Kenya sampled in this study were from historical and recent camel rearing tribes (Mburu et al. 2003). Therefore, based on the geographical division of breed types, the 13 counties were grouped into five groups. For human sampling, high-risk groups such as camel herders and their immediate families were targeted. Sampling mainly focused on high-risk locations such as around watering points and common browsing locations that attract several herds of camels from different regions.
Table 1. Univariate analysis of factors associated with ELISA positive camels for MERS-CoV in Kenya.
For camel sampling, camel owners were informed about the study in a language that they understood. The camels were physically restrained, and 10 mL of blood from the jugular vein was drawn into anti-coagulant vacuum blood collection tubes. Additionally, 1163 nasal swabs samples were collected in RNAlater® (Ambion, Foster City, CA, USA) and virus transport medium.
The research was conducted in accordance with the Helsinki Declaration for sampling of human subjects. Informed consent was obtained from camel owners, handlers, and family members or their guardians (in case of underage children) from whom blood was collected. Five milliliters of blood from consenting study participants were collected by a clinician or phlebotomist from a peripheral vein into an anti-coagulant vacuum blood collection tube. Sociodemographic data were also collected from the participants (Supplemental Table S2). Other data such as age and sex were also collected.
Blood samples were centrifuged, plasma was aliquoted and stored in liquid nitrogen while in the field, and the samples were transported to Nairobi for storage at -80 ℃ A total of 1, 163 dromedary camels were sampled from 13 counties and 486 human subjects from 10 counties. Camel and human samples were exported to Wuhan Institute of Virology, Chinese Academy of Science in Hubei, Peoples Republic of China. All samples were transported according to IATA international regulations for transporting viable samples.
An in-house anti-MERS-CoV IgG ELISA kit was developed based on the purified spike protein receptor binding domain. This highly sensitive and specific ELISA was previously validated for use with samples from camel and human (Zohaib et al. 2018). This anti-MERS-CoV IgG ELISA was used with minor modifications. Camel samples were tested at 1:20 dilution and goat anti-camel IgG-horseradish peroxidase conjugate (Alpha Diagnostic International, San Antonio, TX, USA) was used as the secondary antibody at 1:3000. Based on the microneutralization test, a cut-off value of 0.35 was determined. For human samples, plasma was tested at a dilution of 1:20 and anti-human IgG-horseradish peroxidase conjugated monoclonal antibody (Kyab Biotech Co., Ltd, Wuhan, China) was used as the secondary antibody at 1:15000.
A microneutralization assay was performed as described previously (Perera et al. 2013). Briefly, Vero B4 cells were seeded into 96-well plates. Plasma samples were incubated at 56 ℃ for 1 h. MERS-CoV (EMC strain) was diluted with DMEM to 100 TCID50/50 μL. The plasma samples were diluted by twofold in DMEM and incubated with MERS-CoV at 37 ℃ for 30 min. The medium was removed from the cells and 50 μL virus-plasma mixture was added. The virus-plasma mixture was removed after 1 h and 100 μL DMEM plus 2% fetal bovine serum (FBS) and 1% penicillin/streptomycin was added. The cells were incubated at 37 ℃ with 5% CO2, and the cytopathic effect (CPE) was observed and recorded at 4 days post-infection. Samples that inhibited CPE at a dilution of 1:20 were considered as positive.
Viral RNA was extracted from camel nasal swabs stored in RNAlater® (Ambion) using a viral RNA extraction kit (Roche, Basel, Switzerland) according to the manufacturer's instructions. All camel nasal swabs were screened for MERS-CoV using two independent TaqMan quantitative reverse transcription PCR assays for the nucleocapsid gene (N) according to the WHO testing algorithm as described previously (Lu et al. 2014). Additionally, camel samples positive according to RT-qPCR were also screened by MERS-CoV-specific RT-PCR targeting the N gene to confirm the presence of MERS-CoV in the samples as described previously (Corman et al. 2012).
Virus isolation from two positive nasal swabs with high viral loads was attempted using Vero cells. Vero cell monolayers were maintained in DMEM supplemented with 10% FBS. Nasal swab specimens in viral transport medium were diluted in DMEM before being added to the Vero cells. After incubation at 37 ℃ for 1.5 h, the inoculum was removed and replaced with fresh DMEM containing 2% FBS and antibiotics. The cells were incubated at 37 ℃ for 5 days and observed daily for CPEs. The culture supernatant and cells were examined for the presence of virus by the RT-qPCR N2 assay targeting the MERS-CoV N gene (Corman et al. 2012) and immunofluorescence assay using a rabbit antibody (prepared in house) against the MERS-CoV N protein.
Libraries for next-generation sequencing were prepared using an Illumina Truseq mRNA kit (TruSeq Stranded mRNA Library Prep Kit, Cat #RS-122-2101, Illumina, San Diego, CA, USA) following the manufacturer's instructions. The sequencing was performed on a HiSeq 3000 sequencer. The data obtained was analyzed in Metavisitor (a suite of galaxy tools) as described previously (Carissimo et al. 2017). The BLAST-guided scaffold was then used to reference align the reads in Geneious R11. Phylogenetic analysis was performed in MEGA7. For MERS-CoV samples that were not selected for full-genome sequencing, specific RT-PCRs were set-up to amplify the partial S gene and fragment covering the gene regions of orf3, orf4a, and orf4b as described previously (Smits et al. 2015; Chu et al. 2018).
ELISA-positive samples were statistically analyzed (Chi Square test) to detect associations between location, age, and sex. Univariable analysis was performed and odds ratios along with their 95% confidence intervals (CIs) were calculated. A P value < 0.05 was considered as significant in all analyses. Statistical analysis was performed in R (v3.5.1) with epicalc (v18.104.22.168) and the DescTools (v0.99.25) packages.
Study Area and Design
Molecular Detection of MERS-CoV in Camel Nasal Swabs
Full Genome Sequencing of MERS-CoV and Analysis
A total of 1163 plasma samples was collected from camels in 13 counties in Kenya between January 2016 to June 2018. Age and gender data for the 14 camels were missing. From the remaining 1149 samples, 801 (69.71%) were female and 348 (30.29%) were male. Most plasma samples (611; 52.54%) were collected from the northeastern part of Kenya (region C), which also has the largest herd of camels in Kenya and borders the Republic of Somalia. The demographic distribution of plasma in different camel breed regions and administrative counties is presented in Table 1 and Fig. 1. A total of 792 of the 1163 (68.10%) camel plasma samples tested positive by ELISA. Seroprevalence varied significantly (P < 0.001) among regions in the country, ranging from the highest in region B (79.86, 95% CI 74.90-84.06) followed by region C (75.29, 95% CI 71.72-78.54), region A (48.72, 95% CI 41.00-56.50), region E (42.11%, 95% CI 23.14-63.72), and region D (16.67%, 95% CI 10.20-26.05). The significantly (P < 0.001) highest prevalence was observed in Marsabit county (87.34%, 95% CI 78.24-92.98) (Supplemental Table S1). The seroprevalence of MERS-CoV increased with age and was significantly higher (P < 0.001) in adult camels > 7 years (82.37%, 95% CI 79.50-84.91) compared to sub-adults > 4 years < 7 years (58.57%, 95% CI 46.88-69.37) and juvenile camels < 4 years (36.05%, 95% CI 30.98-41.46). Significantly higher (P < 0.001) seroprevalence was observed in female (74.28%, 95% CI 71.14-77.19) than in male camels (53.74%, 95% CI 48.48-58.90). The geographical location and age of the camels were the main factors affecting the MERS-CoV seroprevalence in Kenya (Table 1).
A total of 486 human plasma samples were collected from 10 counties in Kenya from March 2017 to June 2018, among which 231 (48.63%) subjects were male and 244 (51.37%) were female; gender data for 11 human samples were missing. Among the humans sampled, 95.27% stated that they have had close contact with livestock including camels. Three hundred people in this group stated they have regular contact with camels as herdsmen.
Twenty of the 486 human plasma samples showed positive results by ELISA. Of these, eight were from West Pokot, five were from Tana River, four were from Garissa, two were from Wajir, and one was from Isiolo Counties. Ten seropositive samples were male and nine were female, gender data of one ELISA reactive sample was missing. Twelve of the 20 ELISA-positive individuals were in frequent contact with camels. A micro-neutralization test was performed on all ELISA-reactive samples and none were positive in the neutralization assay.
Eleven camel nasal swabs (Table 2) were positive for both N2 and N3 by RT-qPCR and nested PCR for the N gene. Among them, five were from adult camels, four were from juvenile camels, and one was a sub-adult camel. Age data for one positive camel was not available. High viral loads were observed in juvenile camels compared to in adults. Sequencing of the PCR-positive N gene samples revealed 100% identical sequences. Contamination was ruled out by repeating the experiments in independent laboratories and targeted PCR amplification and sequencing of the N gene. We partially sequenced the S gene of eight samples as described previously (Table 2) (Smits et al. 2015). Of these eight, six samples were 100% identical, whereas the other two samples showed one nucleotide difference compared to the other six positive samples. All MERS-CoV isolates from Kenya clusters within sub-clade C2, which is associated with the African clade (data not shown).
Table 2. Detection of MERSCoV genomes in eleven dromedary camels from Kenya.
Of the 11 samples detected in this study, two specimens with a high viral load were selected for virus isolation and full-length sequencing. Cytopathic changes in Vero cells were observed sequentially for 5 days post-infection (Fig. 2A). Camel MERS-CoV isolates in Vero cells were confirmed by immunofluorescence assay (Fig. 2B) and RT-qPCR (data not shown). Genetic nucleotide identity was 100% between two Kenyan viruses and 99.63%-99.77% for viruses from sub-clade C2 (Fig. 3), > 99.14% within camel and human MERS-CoV from the Middle East, and 97.82%-98.10% for viruses from sub-clade C1. Full-length sequencing was also performed on RNAlater® preserved nasal swab samples. No nucleic acid differences were detected between isolated viruses and RNAlater® preserved samples. The results of phylogenetic analysis of Kenyan MERS-CoV sequences along with other relevant viruses are shown in Fig. 3. The tree was rooted against a MERS-CoV-related bat coronavirus from South Africa (KC869678.4). Viruses from Kenya clustered together with sequences from Ethiopia, Egypt, Burkina Faso, Nigeria, and Morocco in clade C. Within clade C, viruses from Kenya clustered with viruses from Ethiopia and Egypt in sub-clade C2. Characteristics signature deletions in orf4b have been observed in viruses from Africa in sub-clade C1 but not in those from Egypt and sub-clade C2. Notably, Ethiopian and Egyptian MERS-CoV encode full-length orf4b (246 amino acids). Two viruses detected in this study were unique because they had a truncated orf4b (244aa) in sub-clade C2 (Fig. 4), whereas the other three sequences encoded full-length orf4b of 246 amino acids. Previously, a deletion pattern in the orf3 region of African MERS-CoV has also been reported (Chu et al. 2018). However, MERS-CoV from Kenya encodes full-length orf3. The amino acid residues of Kenyan sample C1215 differed from the EMC strain throughout the virus genome. The amino acid residues in the spike protein receptor binding domain of C1215 contained the amino acid substitutions S10F and S148P in the receptor binding motif. Although these two mutations were observed in the spike receptor binding domain region of MERS-CoV, the camel plasma and human plasma were effectively neutralized by the EMC strain from Kenya as described previously; thus these mutations likely do not affect the affinity of the protein for the host receptor (Corman et al. 2014; Liljander et al. 2016), although further studies are needed to confirm this.
Figure 2. Isolation of camel MERS-CoV C1215 and C1272. A Induction of cytopathic effect on Vero cells. The images were taken by NIS Elements F (ECLIPSE TS100, Nikon). Original magnification: 100 ×. B Successful isolation of camel MERS-CoV was confirmed by immunofluorescent antibody staining using rabbit antibody against the MERS-CoV N protein. The columns (from left to right) show staining of nuclei (blue), virus replication (red), and both nuclei and virus replication (merged double-stain images). The images were taken by a confocal microscope. Scale bar = 100 μm.
Figure 3. Phylogenetic analysis of MERS-CoV full genomes using neighbor-joining method in MEGA7. Bootstrap values of nodes are shown. Bootstrap values along branches are for 1, 000 replicates. The tree was rooted against a MERS-CoV related bat coronavirus Neoromicia/PML-PHE1/RSA/2011 (KC869678) from South Africa. To allow for greater resolution of the viruses of interest, the long branch of KC869678 was removed. Detected MERS-CoV viruses in this study are red colored and identified with circle node markers (●). Scale bar indicates nucleotide substitutions per site.
Figure 4. Schematic diagram showing the alignment of orf4b from Kenya and other previously reported MERS-CoV strains from Africa compared to HCoV-EMC. Putative ORFs are represented in black and proteins are shown as gray bars. Stop codon are indicated by asterisks and amino acid lengths are indicated.
MERS-CoV Seroprevalence among Camel Populations in Kenya
MERS-CoV Seroprevalence in Humans in Kenya
Molecular Detection of MERS-CoV
Table Table S1. Detection of MERS-CoV from thirteen counties of Kenya, 2016-2018
Table Table S2. Demographic characteristics of human samples tested in this study