Identification of feline-like rotavirus strains in Sabah State of Malaysian Borneo Authors: Asgini Binti Mohd Jelfree, Daisuke Mori, Yuki Akari, Jecelyn Leaslie John, Lia Natasha Amit, Syed Sharizman bin Syed Abdul Rahim, Satoshi Komoto, Julia Joseph, Kamruddin Ahmed Published: 2026 ABSTRACT AbstractChildren in low- and middle-income countries experience a greater burden of rotavirus infection relative to developed countries. Malaysia is a middle-income country; however, like other South East Asian countries, studies on rotavirus are scarce. In an effort to understand the strain diversity, we performed a study in children under five years old who attended 4 health care centers in the Papar district of Sabah state of Malaysian Borneo and collected samples from 2020 to 2021. This study provides robust molecular evidence that strains detected in Sabah, Malaysia, were generated by intergenogroup reassortment. Reassortment is one of the evolutionary mechanisms that generate diversity in rotavirus strains. We characterized a rotavirus which possessed a long RNA pattern and an unusual combination of G6P[9]. These strains have a unique constellation of G6-P[9]-I2-R3-C3-M3-A3-N2-T6-E3-H6 with an AU1-like backbone and components from DS-1, indicating an intergenogroup reassortant. Of the 11 gene segments, five were closest to feline and 2 were closest to human strains, respectively; the rest were related to human–animal strains. Novel rotaviruses generated by reassortment might influence the disease outcome and vaccination effort.Keywords: Rotavirus, Reassortant, Feline, G6P[9], Children, Diarrhea Introduction Rotavirus is a nonenveloped, double-stranded RNA (dsRNA) virus from the Sedoreoviridae family. The glycoprotein VP7 defines the G genotypes, and the protease-sensitive protein VP4 defines the P genotypes. The distribution of G and P genotypes varies between and within countries and years. The rotavirus genome is made up of 11 segments of double-stranded RNA. These segments encode for six structural proteins (VP1, VP2, VP3, VP4, VP6, and VP7) and six non-structural proteins (NSP1–NSP6). The most common G/P genotype combinations causing infections in humans are G1P[8], G2P[4], G3P[8], G4P[8], G9P[8] and G12P[8]. Among them, G3P[8], G9P[8], and G12P[6] are the most prevalent genotypes in Malaysia; other genotypes include G9P[4], G1P[8] and G8P[8] [1–3]. Most human RVA strains can also be classified into two major and one minor genotype constellations, the Wa-like (G1/3/4/9/12-P[8]-I1-R1-C1-M1-A1-N1-T1-E1-H1), DS-1-like (G2-P[4]-I2-R2-C2-M2-A2-N2-T2-E2-H2), and AU-1-like (G3-P[9]-I3-R3-C3-M3-A3-N3-T3-E3-H3) genotype constellations [4]. In the present communication, we report the detection of the G6 rotavirus genotype, an occasional pathogen typically transmitted sporadically from animal sources [5]. In an effort to contribute to the understanding of the evolution of these strains, genetic characterization was done using whole-genome sequencing. The present study reports naturally occurring intergenogroup reassortant strains between animal and human rotaviruses. Materials and methods From July 2020 to August 2021, watery stool samples were collected from children under 5 years of age with diarrhea attending Kinarut Health Clinic, Bongawan Health Clinic, and Ibu Anak Papar Health Clinic in the Papar district of Sabah state of Malaysia. A case of diarrhea was defined as the passing of three loose stools during a 24-h period. Before sample collection, written informed consent was taken from the guardians of the children. The age, gender, and race of the children were recorded. The collected stool samples were sent under cold chain conditions to the laboratory of Universiti Malaysia Sabah, Kota Kinabalu, Sabah, Malaysia. The samples were stored at −80 °C until used. All samples were tested for Salmonella and Shigella by stool culture using blood agar, and MacConkey’s agar. Additionally, Vibrio cholerae infection was also tested by culturing rectal swab and/or stool samples on thiosulfate citrate bile salt agar (TCBS). Stool samples were diluted tenfold in phosphate-buffered solution. Rotavirus was identified using a commercial solid phase sandwich immunochromatography kit according to the manufacturer’s instructions (SD BIOLINE Rotavirus RTK.). Rotavirus genomic RNA was extracted from rotavirus-positive samples using a QIAamp Viral RNA Mini Kit according to the manufacturer’s instructions (Qiagen GmbH, Hilden, Germany). Whole-genome sequencing of rotavirus was done by Illumina MiSeq next-generation sequencing as described previously [6]. Genotypes were determined by submitting nucleotide sequences to the rotavirus genotyping tool available on the Rotavirus A Genotyping Tool Version 0.1 (https://mpf.rivm.nl/mpf/typingtool/rotavirusa/). Phylogenetic analyses were done using the maximum likelihood method in MEGA X software (https://www.megasoftware.net) following nucleotide sequence alignment with CLUSTAL W. The branching patterns were evaluated using a bootstrap analysis of 1000 replicates. Nucleotide identities among gene segments from different lineages were calculated using online tools (www.bioinformatics.org). The sequence data of the viruses were deposited in the GenBank database with accession numbers LC830266 to LC830309. Results A total of 90 samples were collected during the study period. Their median age was 26 months (range 6–59 months) months. The male:female ratio was 1.7:1. Eighty-one (90%) samples were Malaysian and nine (10%) were non-Malaysian; four of them were Filipinos and five of them were Indonesians. The predominant symptom was diarrhea (n = 90) followed by fever (n = 81), vomiting (n = 30), and dehydration (n = 22). Stool culture and rectal swab culture for V. cholerae were done on 80 (98.8%) and 66 (81.5%) children, respectively. All samples were negative for Salmonella, Shigella and V. cholerae. Four (4.5%) samples were positive for rotavirus. None of these children were vaccinated for rotavirus. The ages of the patients ranged from 23–30 months, two were males and two were females. Details are given in Table 1. Table 1Characteristics of human cases infected with feline origin reassortant rotavirus strains in Sabah, MalaysiaSample no.GenderAge (month)ResidenceNationalityClinical informationBody temperature (oC)VomitingDiarrheaDay of onset (diarrhea)DehydrationPPR 39M30Pengalat Besar, PaparMalaysian38NoYesD4NoPPR 40F30PPRT Kinarut/UNHCRNon-Malaysian38YesYesD4YesPPR 41F23PPRT Kinarut/UNHCRNon-Malaysian39YesYesD4YesPPR 42M27PPRT Kinarut/UNHCRNon-Malaysian38YesYesD3Yes Characteristics of human cases infected with feline origin reassortant rotavirus strains in Sabah, Malaysia Illumina MiSeq next-generation sequencing produced 9.3 × 105 reads (~ 138 bp average length), 1.4 × 106 reads (~ 138 bp average length), 1.3 × 106 reads (~ 139 bp average length), 1.2 × 106 reads (~ 141 bp average length) for strains P39, P40, P41, and P42, respectively. The complete nucleotide sequences of all 11 genes of these study strains were obtained. The lengths of the nucleotide and deduced amino acid sequences of the 11 genes of these two study strains, with related sequence read data, are summarized in Table 2. Whole-genome sequencing showed that all four strains possessed a unique genotype constellation of G6-P[9]-I2-R3-C3-M3-A3-N2-T6-E3-H6 with an AU1-like backbone (Table 3). Nucleotide sequence identities as well as closest strains are summarized in Table 4. Of the four strains, three were from patients PPR 40, PPR 41, and PPR 42 who attended Kinarut Health Clinic. They lived in the same neighborhood of PPRT Kinarut/United Nations High Commissioner for Refugees (UNHCR) camp. PPR 40 and PPR 41 stayed together and notably, did not have any pets. However, there were many stray cats and dogs in the area. Patient PPR 39 attended Papar Health Clinic and lived in Pengalat Besar which is 7 km from UNHCR camp. We were unable to retrieve neighborhood information because the patient’s guardian could not be contacted using the number provided at the clinic. Table 2Sequence data for the 11 gene segments of four Malaysian G6P [9] strains, P39, P40, P41, and P42Study strainTotal readsa (average length)GeneVP7VP4VP6VP1VP2VP3NSP1NSP2NSP3NSP4NSP5P391,010,276 (138.6)Nucleotides; bp (% coverage of the full-length)1046 (98.5%)2359 (100%)1296 (95.6%)3268 (99.0%)2701 (99.7%)2568 (99.1%)1565 (99.2%)1042 (98.4%)1056 (98.3%)627 (83.6%)638 (95.7%)Deduced amino acids; aa (% coverage of the full-length)326 (100%)775 (100%)422 (100%)1088 (100%)887 (100%)835 (100%)491 (100%)317 (100%)313 (100%)175 (100%)198 (100%)Reads mapped to gene segment437172877461404389421314011512185087746Maximum depth of reads10731718335437821719323728522910Average coverage57.9101.582.8260.4225.2227.4124.1201.1242.8193.910.0P401,547,350 (137.5)Nucleotide; bp (% coverage of the full-length)1051 (99.0%)2350 (99.9%)1300 (95.9%)3292 (99.7%)2695 (99.5%)2571 (99.2%)1552 (98.4%)1037 (97.9%)1059 (98.6%)715 (95.3%)632 (94.8%)Deduced amino acids; aa (% coverage of the full-length)326 (100%)775 (100%)422 (100%)1088 (100%)887 (100%)835 (100%)491 (100%)317 (100%)313 (100%)175 (100%)198 (100%)Reads mapped to gene segment19127313307522,98416,28415,4555998533361193116208Maximum depth of reads37533887351164132579970479884681545Average coverage250.1427.9325.2960.0830.8826.6531.4707.1794.5599.245.3P411,483,432 (139.6)Nucleotide; bp (% coverage of the full-length)1047 (98.6%)2349 (99.6%)1322 (97.5%)3291 (99.7%)2699 (99.7%)2581 (99.6%)1553 (99.4%)1042 (98.4%)1057 (98.4%)725 (96.7%)629 (94.3%)Deduced amino acids; aa (% coverage of the full-length)326 (100%)775 (100%)422 (100%)1088 (100%)887 (100%)835 (100%)491 (100%)317 (100%)313 (100%)175 (100%)198 (100%)Reads mapped to gene segment6551031196613,170726214,364300927775215111376Maximum depth of reads15460954893582777739465275138413Average coverage87.361.3207.6558.7375.6776.9270.5372.0688.8214.316.9P421,386,020 (140.3)Nucleotide; bp (% coverage of the full-length)1051 (99.0%)2349 (99.6%)1324 (97.6%)3291 (99.7%)2703 (99.8%)2571 (99.2%)1553 (99.4%)1040 (98.2%)1057 (98.4%)728 (97.1%)648 (97.2%)Deduced amino acids; aa (% coverage of the full-length)326 (100%)775 (100%)422 (100%)1088 (100%)887 (100%)835 (100%)491 (100%)317 (100%)313 (100%)175 (100%)198 (100%)Reads mapped to gene segment10924422217115,81210,20911,7483322386347802221150Maximum depth of reads194894543912113460351758366560731Average coverage145.8264.1230.1674.1529.9641.1300.1521.1634.5428.032.5aSequence reads remaining after adapter trimming.Average coverage = (N × L)/GN: reads mapped to gene segmentL: average lengthG: gene length Sequence data for the 11 gene segments of four Malaysian G6P [9] strains, P39, P40, P41, and P42 aSequence reads remaining after adapter trimming. Average coverage = (N × L)/G N: reads mapped to gene segment L: average length G: gene length Table 3Comparison of genomic constellation of G6P[9] strains from Sabah, Malaysia with other related strainsConstellationDigital object identifier of publicationCountryStrainG6-P[9]-I2-R3-C3-M3-A3-N2-T6-E3-H6Present studyMalaysiaP39, P40, P41, and P42G6-P[9]-I2-R2-C2-M2-A3-N2-T3-E3-H310.1007/s11262-011-0624-6JapanKF17G6-P[9]-I2-R2-C2-M2-A3-N1-T6-E2-H310.3201/eid1904.121383Tunisia17237G6-P[9]-I2-R2-C2-M2-A3-N2-T3-E2-H310.1016/j.meegid.2018.05.030GermanyGer29-14G6-P[9]-I2-Rx-Cx-Mx-A3-N2-T3-E3-H310.1002/jmv.23644ItalyPG05G3-P[3]-I3-R3-C2-M3-A9-N2-T3-E3-H610.1007/s00705-018-3835-5JapanFRV72G3-P[9]-I3-R3-C3-M3-A3-N3-T3-E3-H610.1099/jgv.0.001548JapanTo16-11G3-P[9]-I2-R2-C2-M2-A3-N2-T3-E2-H310.1007/s00705-023-05874-8ArgentinaArg17139G8-P[14]-I2-R5-C2-M2-A3-N2-T6-E12-H310.1128/JVI.02246-08ArgentinaArg/Chubut/99G8-P[1]-I2-R2-C2-M2-A3-N2-T6-E2-H310.1002/jmv.21974KenyaB12G3-P[14]-I2-R3-C3-M3-A9-N2-T6-E2-H310.1016/j.meegid.2014.04.009AustraliaRCH272G6-P[6]-I2-R2-C2-M2-A2-N2-T2-E2-H210.1099/vir.0.2008/003780-0Belgium/MaliB1711G3-P[9]-I3-R3-C3-M3-A3-N3-T3-E3-H610.1128/JVI.02257-07JapanAU-1G3-P[9]-I3-R3-C3-M3-A3-N3-T3-E3-H310.1099/vir.0.070771-0JapanFRV-1G3-P[3]-I3-R3-C2-M3-A9-N2-T3-E3-H610.1016/j.virol.2008.07.041AustraliaCat97G3-P[9]-I2-R2-C2-M2-A3-N1-T3-E2-H310.1099/vir.0.027425-0ItalyBA222G6-P[6]-I2-R2-C2-M2-A2-N2-T2-E2-H210.1099/vir.0.000174GhanaPML1965G6-P[14]-I2-R2-C2-M2-A11-N2-T6-E2-H310.1128/mra.00008-23USA3000015004G6-P[14]-I2-R2-C2-M2-A11-N2-T6-E2-H310.1016/j.meegid.2011.05.012EgyptEGY3399G6-P[8]-I2-R2-C2-M2-A2-N2-T2-E2-H210.1016/j.virol.2022.10.010BrazilLVCA30390G6-P[1]-I2-R2-C2-M2-A3-N2-T6-E2-H110.1002/jmv.25625PakistanPAK99 Comparison of genomic constellation of G6P[9] strains from Sabah, Malaysia with other related strains Table 4Nucleotide sequence identity and closest strains for each geneGeneStrainAccession NoNucleotide sequence identity (%)P39P40P41P42VP4RVA/Human-wt/CHN/BJ-F2086/2019/G3P[9]PP811619.197.297.1197.1197.11VP7RVA/Human-wt/BEL/B1711/2002/G6P[6]EF554087.198.9598.9598.9598.95NSP1RVA/Cat-wt/ITA/BA222/2005/G3P[9]GU827412.196.0496.0196.0196.01NSP2RVA/Human-wt/USA/6235/2003/G3P[3]JF804996.197.3697.3497.3697.36NSP3RVA/Guanaco-wt/ARG/Chubut/1999/G8P[14]FJ347108.198.0198.0298.0198.01NSP4RVA/Cat-tc/JPN/FRV72/1990/G3P[3]LC328259.194.4395.1195.1895.20NSP5RVA/Dog-wt/BRA/IAL-M202/BRA/2017/G3P[3]ON322741.198.5998.5898.5798.61VP1RVA/Dog-wt/USA/A79-10/1979/G3P[3]EU708934.196.5496.5496.5496.54VP2RVA/Human-wt/KEN/KLF2230/2013/G3P[9]MZ097275.198.9998.9998.9998.99VP3RVA/Human-wt/AUS/RCH272/2012/G3P[14]KF690127.197.6297.6397.6097.63VP6RVA/Dog-wt/BRA/IAL-M202/BRA/2017/G3P[3]ON454215.195.7695.6995.6895.61 Nucleotide sequence identity and closest strains for each gene Phylogenetic analyses of individual gene segments are shown in (Figs. 1 and 2). Phylogenetic analyses (Figs. 3, 4, 5, 6 and 7) showed that the 5 gene segments clustered significantly with 72–100% bootstrap support with feline strains (VP4, VP1, NSP2, NSP4, NSP5). VP2 and VP7 clustered with human strains (Figs. 2, 8) with 98–100% bootstrap support. Gene segments NSP1, and VP3 related to feline–human strains (Figs. 1, 9); NSP3 clustered with guanaco–human strains (Fig. 10) and the VP6 clustered with human and different animal strains (Fig. 11). This indicates that the strains identified in this study may have evolved via multiple reassortment events involving feline, human and other animal rotaviruses. Phylogenetic tree constructed using the nucleotide sequences of the NSP1 gene of samples collected (labeled with a filled circle) and global strains using the Maximum Likelihood method, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the VP7 gene of samples collected (labeled with a filled circle), and of global strains using the maximum likelihood method, mid-point rooted, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the NSP2 gene of samples collected (labeled with a filled circle) and global strains using the maximum likelihood method, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the NSP4 gene of samples collected (labeled with a filled circle) and global strains using the maximum likelihood method, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the NSP5 gene of samples collected (labeled with a filled circle) and global strains using the maximum likelihood method, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the VP1 gene of samples collected (labeled with a filled circle) and global strains using the maximum likelihood method, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the VP4 gene of samples collected (labeled with a filled circle) and of global strains using the maximum-likelihood method, midpoint-rooted, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed from the nucleotide sequences of the VP2 gene of samples collected (labeled with a filled circle) and of global strains, using the maximum-likelihood method, midpoint-rooted, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the VP3 gene of samples collected (labeled with a filled circle) and global strains using the maximum likelihood method, mid-point rooted, with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the NSP3 gene of samples collected (labeled with a filled circle) and of global strains, using the maximum likelihood method with bootstrap analysis of 1000 replicates Phylogenetic tree constructed using the nucleotide sequences of the VP6 gene of samples collected (labeled with a filled circle) and global strains using the maximum likelihood method, mid-point rooted, with bootstrap analysis of 1000 replicates Discussion The detection of the unusual G6P[9] strain in Sabah is noteworthy as it is the first such strain reported from Malaysia. As this constellation has not been reported in other countries to date, we assume local emergence through multiple reassortment events. Nevertheless, parental feline or human strains may already be circulating globally or regionally, remaining undetected due to limited surveillance. The co-infection of multiple strains of rotaviruses is attributed to the emergence of intergenogroup reassortant strains [7]. However, the parental strains of this constellation have not been identified in Malaysia and are likely the source of the genes present in our intergenogroup reassortant strains. This is not unexpected, as co-circulation of parental strains with such reassortants in the same location and period is rarely documented. It is also important to note that studies on animal rotaviruses are scarce in Malaysia, even though they likely play a key role in the emergence of these reassortant strains. Future investigations of feline and bovine rotaviruses characterized from Sabah may shed light on the evolution of these strains. The reasons for detecting these strains in the Papar district remain unclear. However, compared with other districts in Sabah, Papar has a higher proportion of marginalized populations, undocumented migrants, and seasonal workers. Rapid industrial development and economic opportunities have attracted immigrants from neighboring Indonesia and the Philippines [8]. As a result, 23.2% (31,500/135,637) of Papar’s population is non-Malaysian (https://www.mycensus.gov.my/index.php/ms), and many refugees and undocumented immigrants reside in UNHCR camps [9]. The factors that triggered the evolution and facilitated the spread of G6P[9] in this context are unknown. It may be due to the pattern of adaptation to selective constraints of the camp situation, such as overcrowding, malnutrition, and close interaction with infected animals. However, this would be further validated using environmental or animal sampling. Notably, G6 is the most prevalent genotype in cattle worldwide [10] whereas P[9] is commonly found in cats [11], which correlates with findings from the phylogenetic trees for NSP3, which clustered with guanaco–human strains and VP6 which clustered with various animal origins. However, the scarcity of data, especially sequence data from Malaysia, constrains the ability to place these strains within a local context. The first report of a human infection with a G6 strain was reported in Belgium in a child who returned from Mali [12]. G6 has been reported in combination with P[8] [13], P[6] [10], G6P[6] in Ghana [14] in Mali, Italy, and Burkina Faso [15–17], G6P[1] in Pakistan [18], and G6P[9] in Japan [19], all belonging to a DS-1-like genetic constellation. Increasing evidence suggests that intergenogroup reassortant strains with novel genome constellations might influence the effectiveness of rotavirus vaccines [20]. Therefore, such strains should be considered in future vaccine development efforts. It is crucial to explore the spread of atypical strains and investigate any fitness gains as a result of intergenogroup reassortment events. The detection of an intergenogroup reassortant in Sabah further emphasizes the ubiquitous nature and diversity of rotavirus strains in circulation. The four naturally occurring reassortant strains identified in this study share a common genome constellation, suggesting derivation from a common ancestor and successful propagation among human populations in Sabah. Continuous rotavirus surveillance, incorporating whole-genome sequencing, is essential to monitor the circulating rotavirus strains in Malaysia. Acknowledgements We thank the Director General of the Ministry of Health for giving permission to publish this study. Author contributions A. B. M. J., J. L. J., S.S. B. S. A. R., K.A. conceptualization. A. B. M. J., and K.A.: writing. A. B. M. J., D. M., Y. A., J. L. J., L. N. A., S. S. B. S. A. R., S. K., J. J., and K. A.: editing. A. B. M. J., D. M., Y. A., J. L. J., and L. N. A.: methodology and investigation. S. S. B. S. A. R., S. K., and K. A.: resources, funding acquisition and supervision. All authors contributed to the article and approved the submitted version. Funding This study was funded by the Universiti Malaysia Sabah fund (AM18006). Data availability The sequence data of viruses were deposited in the GenBank database with accession numbers LC830266 to LC830309. Declarations Ethics approval and consent to participate Ethics approval was obtained from the Medical Research Ethics Committee (NMRR-19-3850-51720) and registered with the National Medical Research Registrar, Ministry of Health Malaysia. Written informed consent was obtained from prospectively included patients before study participation. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The sequence data of viruses were deposited in the GenBank database with accession numbers LC830266 to LC830309.