key: cord-315972-5g2hnk1x
authors: Tong, Suxiang; Conrardy, Christina; Ruone, Susan; Kuzmin, Ivan V.; Guo, Xiling; Tao, Ying; Niezgoda, Michael; Haynes, Lia; Agwanda, Bernard; Breiman, Robert F.; Anderson, Larry J.; Rupprecht, Charles E.
title: Detection of Novel SARS-like and Other Coronaviruses in Bats from Kenya
date: 2009-03-17
journal: Emerg Infect Dis
DOI: 10.3201/eid1503.081013
sha: 
doc_id: 315972
cord_uid: 5g2hnk1x

Diverse coronaviruses have been identified in bats from several continents but not from Africa. We identified group 1 and 2 coronaviruses in bats in Kenya, including SARS-related coronaviruses. The sequence diversity suggests that bats are well-established reservoirs for and likely sources of coronaviruses for many species, including humans.

T he 2003 outbreak of severe acute respiratory syndrome (SARS) generated renewed interest in coronaviruses (CoV) and the source for the SARS CoV that caused the outbreak in humans (1) . Serologic studies demonstrated that the virus had not previously circulated in human populations to any large extent and suggested a source of zoonotic origin (2) (3) (4) . A likely natural viral reservoir for the virus was not identifi ed until horseshoe bats (Rhinolophus spp.) in several regions in the People's Republic of China were demonstrated to harbor SARS-like CoVs (5) . Subsequently, a number of other SARS-like CoVs, as well as CoVs from antigenic groups I and II, were identifi ed from bats in Asia, Europe, and North America, and coronavirus antibodies were detected in African bat species (6) (7) (8) (9) (10) (11) . It is not surprising that a growing number of CoVs have been detected in bats. To date, >60 viral species have been detected in bats because their biodiversity (second only to rodents), high population densities, wide distribution, and ability to fl y over long distances allow them to harbor and easily spread multiple infectious agents. Bats have long been known as natural hosts for lyssaviruses and more recently have been recognized as potential reservoirs for emerging human pathogens, including Ebola, Marburg, Nipah, and Hendra viruses in addition to SARS-CoV (12, 13) .

Given the association of bats with emerging infectious diseases, fi eld surveys were performed during July-August 2006 in the southern portion of Kenya ( Figure 1 ). The selection of sites was based on preliminary data regarding bat roost locations and observations of bats in the fi eld during the survey. Attempts were made to collect specimens from 10-20 animals of each species present in each location. Bats were captured manually and by using mist nets and hand nets; adults and subadults of both sexes were captured. Each bat was measured, sexed, and identifi ed to the genus or species level when possible. Blood samples and oral and fecal swabs were collected; the animals were then euthanized in compliance with fi eld protocol. Blood, fecal swabs, and selected tissue samples were transported on dry ice from the fi eld and stored at -80°C.

Fecal swabs (n = 221; Table) were screened for the presence of CoV RNA using 2 semi-nested reverse transcription-PCR (RT-PCR) assays. For the pan-coronavirus RT-PCR, conserved primers were designed from highly conserved regions of the RNA-dependent RNA polymerase (RdRp) gene 1b based on available CoV sequences (1st and 2nd round forward 5′-ATGGGITGGGAY TATCCWAARTGTG-3′; 1st round reverse 5′-AATTAT ARCAIACAACISYRTCRTCA-3′; 2nd round reverse 5′-CTAGTICCACCIGGYTTWANRTA-3′). For the pan-bat coronavirus RT-PCR, conserved primers were designed from the same highly conserved regions based on available bat CoV sequences and presumed to be more specifi c to bat coronaviruses (1st and 2nd round forward 5′-ATGGGITGGGAYTATCCWAARTGTG-3′; 1st round reverse 5′-TATTATARCAIACIACRCCATCRTC-3′; 2nd round reverse 5′-CTGGTICCACCI GGYTTNACRTA-3′). Total nucleic acids were extracted from 200 μL of a phosphate buffered saline suspension of each swab by using the QIAamp Mini Viral Elute kit (QIAGEN, Santa Clarita, CA, USA), according to the manufacturer's instructions. The seminested RT-PCR was performed by using the SuperScript III One-Step RT-PCR kit and Platinum Tag Kit (Invitrogen, San Diego, CA, USA). The positive PCR products were purifi ed by gel extraction by using the QIAquick Gel Extraction kit (QIAGEN) according to the manufacturer's instructions; they were then sequenced on an ABI Prism 3130 automated sequencer (Applied Biosystems, Foster City, CA, USA), according to the manufacturer's instructions.

Of 221 bat fecal swabs examined, 41 (19%) were positive by at least 1 of the 2 seminested RT-PCR assays (Table) . One specimen had 2 distinct CoV sequences, each amplifi ed by 1 of the 2 PCR assays, giving a total of 42 distinct CoV sequences. To characterize the overall diversity of CoV sequences, in this study a phylogenetic tree (Figure 2 ) of the 121-bp fragment of RdRp was generated from 39 coronaviruses from bats in Kenya and 47 selected human and animal coronaviruses from the National Center for Biotechnology Information database based on the Bayesian Monte Carlo Markov Chain method (14) . Three of the 42 sequences were not of suffi ciently high quality to include in this tree. Some nodes had low Bayesian posterior probabilities ( Figure 2 ). Longer sequences from these viruses are needed to refi ne their phylogenetic relationships.

Among the 39 sequences in the tree, 23 belonged to previously defi ned group 1 and were mapped into 5 different sequence clusters. The 121-bp sequences in these 5 clusters had an average nucleic acid (NA) sequence identity of 88%, 85%, 81%, 77%, and 80% when compared with the next closest previously characterized CoVs (i.e., BtCoV1A, BtHKU8, BtHKU7, HCoV229E, and BtCoVA970, respectively). The remaining 16 sequences would likely be placed into group 2. Two sequences from Chaerophon spp. bats (location 17) were closely related to a SARS-like CoV cluster, including 1 sequence shown in Figure 2 (BtKY15) and another (BtKY16) that was 1 of the 3 low-quality sequences excluded from the tree. These 2 NA sequences show ≈89% identity with the nearest previously characterized bat: SARS-like CoV, BtCoVRF1, shows ≈80% NA sequence identity to SARS CoV (Urbani strain) and ≈63% NA sequence identity to the human group 2 CoV HCoVOC43. The 15 remaining NA sequences were grouped into 2 clusters. One cluster contains the recently described BtHKU9 with >95% NA sequence identity, and the other cluster (BtKY18-like cluster) contains no other previously known CoVs, with <75% NA sequence identity to BtHKU9. The pattern of CoV detections by bat species and location demonstrates several features concerning coronaviruses in bats. A given bat species in the same location can harbor several distinct CoVs as noted for Chaerophon spp. (location 17), Miniopterus infl atus (location 5), and Rousettus aegyptiacus (location 2 and 16); similar CoVs can also been seen in the same type of bat in different locations, as noted for BtCoV1A-like cluster CoVs being detected in Miniopterus spp. bats of 4 species from different locations. One M. infl atus bat from location 5 harbored 2 different, but closely related, CoVs, 1 (BtCoV 36) from the BtCo-V1A-like cluster and 1 (BtCoV 35) from the BtHKU8-like cluster ( Figure 2 ). CoVs of these 2 closely related clusters were detected in Miniopterus spp. bats, but not detected in other bat genera, including those that shared roosts with Miniopterus spp. bats. This fi nding is consistent with studies from China in which BtCoV1A-like and BtHKU8-like CoVs were frequently identifi ed but only in Miniopterus spp. bats (15) . This may suggest that viruses of the BtCo-V1A-like cluster and the BtHKU8-like cluster are specifically adapted to Miniopterus spp. bats and not easily transmitted to other bat species.

In contrast, other genetically similar CoVs were detected in several different bat species. For example, CoVs from th BtHKU7-like cluster were detected in both Chaerophon spp. and Otomops martinsseni bats; CoVs from the BtCo-VA970-like cluster were detected in Cardioderma cor and Rousettus aegyptiacus bats; CoVs from the BtKY18-like cluster were detected in Chaerophon spp., Eidolon helvum, and R. aegyptiacus bats; and CoVs from the BtHKU9-like cluster were detected in Hipposidereos commersoni and R. aegyptiacus bats.

These data demonstrate that the CoV diversity in bats previously detected in Asia, Europe, and North America is also present, possibly to a greater extent, in Africa. The extent of this diversity among CoVs may be shown more clearly through additional studies in bats, and increased demonstration of CoV diversity in bats may require a reconsideration of how they should be grouped. The frequency and diversity of CoV detections in bats, now in multiple continents, demonstrate that bats are likely an important source for introduction into other species globally. Understanding the extent and diversity of CoV infection in bats provides a foundation for detecting new disease introductions that may, like SARS, present a public health threat.

A novel coronavirus associated with severe acute respiratory syndrome

Antibody response of patients with severe acute respiratory syndrome (SARS) to nucleocapsid antigen of SARS-associated coronavirus

Infectious diseases emerging from Chinese wet-markets: zoonotic origins of severe respiratory viral infections

Isolation and characterization of viruses related to the SARS coronavirus from animals in southern China

Severe acute respiratory syndrome coronavirus-like virus in Chinese horseshoe bats

Identifi cation of a novel coronavirus in bats

Coronavirus antibodies in African bat species

Detection and prevalence patterns of group I coronaviruses in bats, northern Germany

Detection of group 1 coronaviruses in bats in North America

Prevalence and genetic diversity of coronaviruses in bats from China

Molecular diversity of coronaviruses in bats

Bats as a continuing source of emerging infections in humans

Fruit bats as reservoirs of Ebola virus

BEAST: Bayesian evolutionary analysis by sampling trees

Coronaviruses in bent-winged bats (Miniopterus spp

We thank Janet C. Beagley, Evelyne Mulama, Heather Burke, Dorine Bonyo, Edwin Danga, Leonard Nderitu, Solomon Gikundi, Lydia Kigo, M. Kariuki Njenga, and Olga Urazova for excellent technical and logistical assistance. We thank Yu Li, Jane Iber, and Cara Burns for their discussion and support in the phylogenetic analysis.The study was supported in part by the Global Disease Detection program (Centers for Disease Control and Prevention, Atlanta, Georgia).Dr Tong is a virologist in the Gastroenteritis and Respiratory Virus Laboratory Branch of the Division of Viral Diseases, Centers for Disease Control and Prevention. Her interests are in novel emerging CoVs.