Detection of Rift Valley Fever Virus Interepidemic Activity in Lower Moshi area of Kilimanjaro Region, North Eastern Tanzania: A Community Survey

Background: Rift Valley fever virus (RVFV) is a zoonotic arbovirus of public health impact infecting livestock, wildlife, and humans mainly in Africa and other parts of the world. Despite its public health importance, mechanisms of RVFV maintenance during inter-epidemic (IEPS) periods and potentially spread to new areas remain unclear.We aimed to comparatively examine exposure to RVFV and RVFV infection among humans, goats and mosquitoes in an agro-pastoral community in Lower Moshi area of Moshi rural district. Results:Results show that the male gender was related to RVFV seropositivity (χ2 = 5.351; p=0.030). Being 50 years and above was related to seropositivity (χ2=14.430; p=0.006) whereas bed net use, larger numbers of persons living in the same house (>7 persons) and RVFV seropositivity in goats were related to higher seropositivity to RVFV among humans (χ2=6.003; p=0.021, χ2=23.213; p=0.000 and 27.053; p=0.000), respectively.RVFV antibody concentrations were only marginally higher in humans without statistically signicant difference [t (112) =0.526; p=0.60)]. By the use of RT-qPCR, goats exhibited the highest RVFV infection rate of 4.1%, followed by humans (2.6%), Aedes spp(2.3%), and Culex spp(1.5%). Conclusions: In the absence of RVFV infection data in areas nearby the study site, our ndings suggest Lower Moshi area as a potential hotspot for RVF, posing the danger of being a source of RVFV spread to other areas. Goats had the highest infection rate, suggesting goats as important hosts in the virus maintenance during IEPs. We recommend the design and implementation of strategies that will warrant effective active surveillance of RVF through the identication of RVF hotspots for targeted control of RVF.


Background
Rift Valley fever virus (RVFV) is a zoonotic arbovirus affecting livestock and humans mainly in Africa and the Arabian Peninsula [1][2][3][4] although recent reports indicate the presence of RVF in other parts of the world [5]. According to the World Health Organization (WHO), Rift Valley fever (RVF) is a priority disease due to its considerable public health impact in areas where it occurs and the inadequate interventions to control it [6]. It is also considered an important threat to agriculture in African countries including Tanzania [7][8][9]. Transmission of RVFV to animals is mainly through bites by infected Aedes and Culex mosquitoes, whereas human transmission largely through direct contact with tissues of RVFV-infected animals [10].
It has been previously suggested that maintenance of the virus in animals during inter-epidemic (IEPS) periods and potentially spread to new areas through animal movement. Disease pathology and endemic maintenance within mammalian hosts have been reviewed [11,12]. Although the transmission of RVFV by mosquito vectors to animals and humans has been described, less is known about the role of animals, humans, and vector mosquitoes in maintaining the virus during IEPS. The maintenance mechanisms during IEPS become interesting due to the absence of a clear understanding of where the virus hides during the "silent" periods. Some explanations have been made regarding the possible mechanisms by which the virus is maintained during IEPS. Previous work has documented low-levels of RVFV exposure in northern, central, eastern, and southern Tanzania as a key mechanism of virus maintenance [13,14]. Most people infected by RVFV remain asymptomatic although a small percentage present with clinical disease. Other reports have hypothesized critical mechanisms for survival of RVFV during long inter-epizootic periods as vertical transmission through mosquito eggs to mosquito offspring [15][16][17].
Maintenance of RVFV depend on the presence of competent vectors and hosts but must coincide with multiple factors such as su cient livestock density, rainfall providing vector breeding sites, and temperatures that support vector development and pathogen replication [18], but differential exposure of RVFV in high-risk agropastoral communities in Northern Tanzania has not been examined. We aimed to comparatively examine exposure to RVFV and RVFV infection among humans, goats and mosquitoes in an agropastoral community in Lower Moshi area of Moshi rural district.

Study Design and Site
A community-based, cross-sectional survey was conducted in three villages of lower Moshi in Moshi district, Kilimanjaro region of Tanzania. Data were collected between March and June 2020 involving 3 villages, namely Mikocheni, Chemchem, and Arusha Chini. Lower Moshi is located on the southern foothills of Mount Kilimanjaro ( Figure 1). On the west, Lower Moshi is bordered by the Kikuletwa River, Hai District, and Manyara Region. To the east Lower Moshi borders Mwanga district. Lower Moshi elevation ranges between 700 and 800 m above sea level. The main Rift Valley Fever vectors in this area are Culex spp, Mansoniaspp, Anopheles spp, and Aedes spp [19]. Numerous water streams cross the area and they form the irrigation channels for rice and sugar cane. The rice irrigation schemes have structured and unstructured canal networks; covering an area of about 1,100 hectares. During the rainy season, temporary pools that serve as malaria vector breeding sites are formed. Their persistence beyond the rains contributes to further malaria transmission. The area has two rainy seasons; the long rains which run from March to May and the short rainy season from November to December. The average annual rainfall is about 900 mm per year [20].

Participants and sample collection
Participants in this study were males and females aged between 10 and 70 years, who were either smallholder crop farmers or livestock keepers and willing to participate in this study. Animal sampling was carried out by animal health experts from the Tanzania Veterinary Laboratory Agency (TVLA). Up to 15 goats were selected from each herd by systematic sampling technique where every 3 rd and 5 th animal was included depending on the size of the herd. Selected animals were manually restrained and 3 ml of blood collected through jugular venipuncture using a vacutainer needle. Human blood sampling was done by expert phlebotomists from the Kilimanjaro Christian Medical Center (KCMC). Three milliliters of blood were collected from the median cubital vein by venipuncture. Each sample from both animals and humans was divided into two aliquots of 1.5 ml each and placed into plain and EDTA vacutainer tubes, respectively. To each sample in an EDTA tube, 4.5ml of Tri Reagent (Zymo Research, Irvine, CA, U.S.A.) were added. The mixture was gently mixed by shaking for 1 minute and immediately shipped to the KCRI biotechnology laboratory at 4°C, for analyses. Demographic data from participants were collected using electronic forms designed using Open Data Kit (ODK)tools (https://opendatakit.org/) deployed in Android tablets.
Mosquito trapping BG Sentinel trap (BGS) (Biogents AG, Regensburg, Germany) to target outdoor host-seeking adult mosquitoesparticularlyAedes spp, Ochlerotatusspp, Culex spp, Mansoniaspp, and Anopheles spp [21]. BGS traps were used in combination with the BGS-Lure, a dispenser that releases emanations such as those found on human skin (lactic acid, ammonia, and caproic acid) [22]. The BGS-Trap, developed by BioGents GmbH (Regensburg, Germany), consists of an easy to transport, collapsible white bucket with white gauze covering its opening. In the middle of the gauze cover, there is a black tube through which a down ow is created by 12V DC fan that causes any mosquito in the vicinity of the opening tube sucked into a catch bag [22]. Mosquitoes were immediately morphologically identi ed in the eld and consequently sorted according to species, sex, and whether fed or unfed.
Laboratory procedures RVFV competitive ELISA (cELISA) Serumwas extracted from the plain vacutainer tubes at the end of eachday by centrifugation of clotted blood at3000 rpm for 5 min. An extracted serum sample was then transferred into 2 ml sterile cryovials using a sterile Pasteur pipette. All serum samples were tested for the presence of antibodies against RVFV using a competitive ELISA (cELISA)using the ID Screen RVFCompetition Multi-Species kit (ID-vet, Grables, France), which detects both IgG and IgM antibodies directed against the RVFV nucleoprotein (NP). Validation tests for the test kit have shown a sensitivity of between 91 and 100% and a speci city of 100%. The cELISA was performed according to the instructions of the manufacturer and as described previously [23,24]. To control the validity of each plate, the mean value of the two negative controls (OD NC ) was computed whereby a plate was considered valid if the OD NC was>0.7. For a valid plate, the mean value of the two positive controls divided by OD NC had to be <0.3. For each sample, the competition percentage was calculated by dividing OD sample /OD NC ) × 100. If the value was ≤0.4, the sample was considered positive while a value > 0.5 was considered negative. Only samples that tested positive for cELISA were subjected to RT-qPCR for RVFV detection.
For detection of RVFV RNA in humans and goats, RNA was extracted from Trizol archived blood in EDTA tubes using DirectZol miniprep kit (Zymo Research, Irvine, CA, U.S.A.) by using the Boom method. To isolate RVFV RNA from mosquitoes, pools of 10-50 unfed monospeci c female mosquitoes were placed in cryovials and transferred into Lysing Matrix, impact-resistant tubes containing 1.4 mm ceramic beads (MP Biomedicals, CA, USA). Samples were disrupted by bead beating at 10,000 x g for 1min, spun at 1000 g for 10 min at 4°C. The supernatant was transferred into labeled RNase-free tubes. Puri cation procedures were done using Direct-zol™ RNA miniprep kit (Irvine, CA, U.S.A) following the manufacturer's instructions.
For both human/goat and mosquito samples, RNA concentration and quality check were performed using NanoDrop™ 2000 Spectrophotometer (Thermo Scienti c, NY, USA) before storage at -80°C. RVFV RNA was detected using TaqMan probe-based one-step RT-qPCR targeting the RVFV Gn gene as described by Gudo and colleagues [2] using Applied Biosystems ViiA7 PCR platform, Thermo Scienti c, NY, USA).

Nature of data and data Analysis
Data analysis was performed using IBM SPSS v.26 (IBM® Corp., Armonk, NY, USA). Descriptive data were presented as frequencies and percentages, means, and medians. Categorical data were reported as a tabulation of proportions and compared between humans and goats. Chi-squared statistic (χ 2 ) was used to examine associations between seropositivity to RVFV and RVFV infection in both humans and goats. Mean IgM and IgG concentrations were compared between humans and goats by paired t-test. Percent positivity to RVFV infection in goats, humans, and mosquitoes were reported as histograms.

Ethical issues
This study obtained approval by the Kilimanjaro Christian Medical University College (KCMUCo) Research and Ethics Committee (CRERC) with approval certi cate #2419. This study was also approved by the Kilimanjaro Regional and District Administrative Secretaries, District Medical and Veterinary O cers, and local village and ward executive o cers of respective villages. Before commencement of sample collection, written informed consent was obtained from all study participants aged 18 years and above by signing consent forms whereas parents and/or legal guardians of participants under 18 years and participants who could not read or write consented on behalf. All authors hereby con rm that all procedures in this study were approved by CRERC and were performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki. Authors also con rm that all procedures that involved animals in this study were conducted in compliance with the ARRIVE guidelines.

Demographic Characteristics of Human Participants
A total of 266 human participants were enrolled in the study. Of the participants, more than half (56.4%) were females. The median age (interquartile range) of participants was 45 . The majority of participants (74.4%) came from households with more than 4 persons in the same house. With regards to the participant's education, 63.2% of participants had attained primary school education. Most participants (72.9%) kept livestock (cattle, sheep, goats, and/or chicken). Three quarters (75.2%) of the participants reported having used an insect side treated bed-net the night before the interview. (Table 1)  results are presented in Table 2. Results show that the male gender was signi cantly more related to RVFV seropositivity (χ 2 = 5.351; p = 0.030). Likewise, participants aged 50 years and above were more seropositive as compared to their younger counterparts (χ 2 = 14.430; p = 0.006). Bed net use, larger numbers of persons living in the same house (> 7 persons), and RVFV seropositivity in goats were related to higher seropositivity to RVFV (χ 2 = 6.003; p = 0.021, χ 2 = 23.213; p = 0.000 and 27.053; p = 0.000), respectively (Table 2). Among the selected factors analyzed for possible association with IgM/IgG RVFV seropositivity in goats, only IgM/IgG RVFV seropositivity in humans had a signi cant relationship (χ 2 = 27.053; p = 0.000) ( Table 3).

Comparison of mean IgM/IgG concentrations in humans and goats
When mean concentrations of antibodies to the RVF virus were compared between goats and humans, it was observed that RVFV antibody concentrations were only marginally higher in humans without statistically signi cant difference [t (112) = 0.526; p = 0.60)] ( Table 4). Percentages of RVFV seropositive humans and goats as well as PCR results for viral infections were determined (Fig. 2). Compared to humans, goats were more seropositive to RVFV (23.3% seropositive goats against 13.2% seropositive humans).  Findings from this study indicate that13.2% and 23.3% of tested humans and goats had circulating antibodies to RVFV, respectively. Our ndings emphasize an active exposure to RVFV during IEPs as previously reported by some studies across geo-ecological zones of Tanzania [19,[25][26][27][28].
In this study, goats had higher exposure rates to RVFV compared to humans. Aedes spp, the major vector for RVFV, is known to have bimodal daily feeding behavior with both exophagic and exophilic behaviors [29], feeding on a wide range of mammalian hosts. Consequently, this behavior can be implicated as a key behavior in its role as a vector for many zoonotic infections. Despite its preference for human hosts [29], we report higher seropositivity in goats. The transmission of RVFV is not absolutely dependent on the presence of vector mosquitoes. Direct human contact with infected animal tissues has been reported as a signi cant factor for its transmission from animals to humans [30][31][32]. Not all of the human participants in this study were directly involved in activities that bring them into direct contact with infected tissues such as infected aborted fetuses and those working in slaughterhouses, which could partly explain the lower seropositivity to RFVFV in humans compared to goats.
In the current study, RVFV RNA was detected in humans, goats, and mosquitoes. Goats exhibited the highest infection rate of 4.1%, followed by humans (2.6%). Viral RNA was also detected in 2.3% and 1.5% of tested Aedes spp and Culex spp mosquito pools. This study was conducted to shed light on the maintenance mechanisms of RVFV by investigating both exposure and infection rates in mammalian and arthropod vectors. To our knowledge, this is the rst study conducted in Tanzania to concomitantly report on RVFV diagnosis in humans, animals, and mosquitoes. Many of the previous studies that sought to understand the epidemiology of RVFV in Tanzania, either focused on sero-epidemiology or could not detect RVFV RNA in mammalian and arthropod vectors. Although the interactions of arboviruses and their vectors are complex and their epidemiology is poorly understood, our ndings support the hypothesis that during IEPs, RVFV is likely maintained by localized low-level transmission between mosquito vectors and mammalian hosts without any noticeable clinical symptoms [15,[32][33][34]. Evidence for RVFV transmission during IEPs has previously been reported among humans, livestock, and wild animals in Tanzania and elsewhere [1-4, 6, 19, 24-28, 33, 35-38]. Although goats were more seropositive to RVFV compared to humans, paired comparison of mean anti-RVFV IgG/IgM concentrations revealed no difference that existed between humans and goats.
Some factors were signi cantly associated with seropositivity to RVFV in humans including male gender, living more than 4 persons in a household, being older than 50 years, not using an insecticide-treated bednet, and higher RVFV seropositivity in goats. RVFV seropositivity in Humans was consequently associated with seropositivity in goats. Males, especially in agropastoral communities seem to be more active outdoors for various subsistence activities including farming and grazing which bring them into frequent contact with RVFV susceptible or infected animals. This nding stresses the need for continued distribution, access, and usage of LLINs, especially among rural and agro-pastoral communities that are more prone to zoonotic diseases.
The study site is characterized by features that are supportive of vector mosquito breeding and intimate human-animal interaction. In the absence of reports on RVFV infection in areas nearby the study area, [19] the detection of antibodies to RVFV in humans and goats and detection of RVFV in humans, goats, and mosquitoes in the study area suggests the site to be a potential RVF hotspot.

Conclusion
Here, we present data that reveals the detection of anti-RVFV antibodies in humans and goats and isolation of RVFV from humans, goats, and mosquitoes in an area with the necessary features for mosquito breeding. Authors' Contributions JOC, RSM, BTM conceived the project, overall study implementation and wrote the manuscript.
MSK, RMB, PGH, SIM participated in data collection and made signi cant inputs in writing the manuscript.
MSK, JOC and JMV analyzed and interpreted data