Globally, Zika virus (ZIKV) is a public health concern. Although about 80% of ZIKV infection is asymptomatic, the development of microcephaly in fetuses born to mothers infected with ZIKV during pregnancy makes the infection alarming. Furthermore, the lacks of ZIKV-specific antiviral agents or vaccine and of a convenient rapid test make the situation worse. Recently, ZIKV has been implicated as a cause of microcephaly, Guillain–Barré syndrome, and encephalitis1. In addition to mosquito-dependent transmission, ZIKV can be transmitted by the transplacental route, blood transfusion, and sexual intercourse2.

ZIKV is a single-stranded positive-sense RNA virus with a 10.7 kb genome encoding a single polyprotein that is cleaved into three structural proteins (C, prM/M, and E) and seven nonstructural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5)3. ZIKV is a member of the family Flaviviridae, genus Flavivirus, and is transmitted between humans by Aedes mosquito species. ZIKV causes a dengue fever-like syndrome in humans which is known as Zika fever or ZIKV disease.

The virus was first isolated in 1947 from a febrile sentinel rhesus monkey in the Zika forest research station of the East African Virus Research Institute in Entebbe, Uganda4. The virus was first detected outside Africa in Aedes aegypti from Bentong, Pahang, Malaysia in 19515 and in 1977/78, the first human cases were detected in neighboring Indonesia6. However, as far as we know human ZIKV infections had not been reported from Malaysia before the recent outbreak in Singapore in 2016, although in 2015, ZIKV infection was reported in a German tourist who traveled to the Sabah state of Malaysian Borneo in 20147. Previous seroepidemiological studies have detected the presence of antibodies against ZIKV in humans and in orangutans in Sabah8^, 9^, 10. These results indicate the possible presence of a sylvatic cycle in nonhuman primates and the risk of human exposure to ZIKV.

On September 1, 2016, the first autochthonous case of ZIKV infection in Malaysia was detected in a man in Kota Kinabalu, the capital of Sabah state. Subsequently, a second autochthonous case was also detected in Kota Kinabalu. These raised concerns about whether the strain had invaded recently or was an indigenous type that had been circulating for a long time. Therefore, this study was undertaken to analyze the outbreak cases and to determine their relationship with the ecologic, entomologic, and host determinants of ZIKV in Sabah.

Ethical Statement

The study was registered and ethical clearance obtained from the Medical Research and Ethics Committee, Minisitry of Health, Malaysia (NMRR-17-795-35053). Informed verbal consent was obtained from the patients, their household members, and volunteers.

Background, Location, and Population

Malaysia is a federation of 13 states and three federal territories. These are located in two regions: 11 states and two federal territories on Peninsular Malaysia, and the other two states (Sarawak and Sabah) and one federal territory (Labuan) in Borneo.

Sabah is situated in the northern part of Borneo, bordering Sarawak to the southwest and Kalimantan (Indonesia) to the south. Sabah has an equatorial climate with tropical rainforests and abundant animal and plant species. Kota Kinabalu is the capital city as well the economic center. According to the 2015 census, the population of Sabah is 3,543,500 (https://web.archive.org/web/20160212125740/http://pmr.penerangan.gov.my/index.php/info-terkini/19463-unjuran-populasi-penduduk-2015.html).

Study and Surveillance Period

A Zika alert was issued by the Ministry of Health of Malaysia on February 4, 2016 (https://kpkesihatan.com/2016/08/30/zika-alert-dan-arahan-pentadbiran-untuk-pemantauan-dan-pengurusan-jangkitan-virus-zika-surat-edaran-kpk-4-februari-2016/) and was updated on September 11, 2016 (https://kpkesihatan.com/2016/09/11/updated-zika-alert-dan-arahan-pentadbiran-untuk-pemantauan-dan-pengurusan-jangkitan-virus-zika/). The present study includes the cases of ZIKV infection that occurred in Sabah and were reported by hospitals, health-care personnel or health authorities from other countries to the Sabah State Health Department up to December 2016. Serum and urine samples were collected from the local patients and their household members for the detection of ZIKV.

Following health education activities at Tanjung Lipat and Damai, Kota Kinabalu, on October 10 and 16, 2016, respectively, ZIKV testing was offered to those residing within a 400m radius of the first and second cases. Serum and urine samples were collected for the detection of ZIKV from these volunteers.

Mosquitoes were collected from 30 locations in Kota Kinabalu where the Zika cases resided or had visited. The insects were transported to the laboratory and their species was identified morphologically. The mosquitoes were stored at –80 ^oC until further testing for ZIKV. To each mosquito-containing tube, 150 µl of Eagle’s minimal medium was added and the mosquitoes were crushed with an electric pellet mixer. After centrifugation at 2500 rpm for 10 min at 4 ^oC, 100 µl of supernatant was collected from each tube and placed in a new tube. TriZol (Invitrogen, Carlsbad, CA, USA) was added to 50 µl of this supernatant and nucleic acids extracted using QIAamp Viral RNA Mini Kits (QIAGEN GmbH, Hilden, Germany) according to the manufacturer’s instructions.

To determine whether nonhuman primates are involved in the sylvatic cycle of ZIKV in Sabah, blood samples were collected on October 12 and 19, 2016 from wild Macaca fascicularis (M. fascicularis) by a veterinarian and serum was separated. These wild monkeys were captured in the Kampung Lopak of Beaufort district which is about 104 km from Kota Kinabalu and transported to Kota Kinabalu for preparation to release them in the wild.

Definition and Laboratory Confirmation

The definition of a case of ZIKV infection was according to the WHO guideline. Patients and their household members were tested for ZIKV by reverse transcription-quantitative polymerase chain reaction (RT-qPCR) performed at the Institute of Medical Research, Kuala Lumpur.

Samples from the first and second cases, volunteers and M. fascicularis were tested for ZIKV infection at the laboratory of the Faculty of Medicine and Health Sciences, Universiti Malaysia Sabah. ZIKV genomic RNA was extracted from samples using a QIAamp Viral RNA Mini Kit (QIAGEN GmbH, Hilden, Germany) according to the manufacturer’s instructions. ZIKV was detected by amplifying the NS511 and envelope protein genes 12 by reverse transcription–polymerase chain reaction (RT–PCR). Samples were also subjected to PCR using universal flavivirus primers that amplify the NS5 gene13.

Determination of Nucleotide Sequences

The nucleotide sequence of the amplicons was determined using a BigDye terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions and the product was run on an ABI Prism 3100 Genetic Analyzer (Applied Biosystems).

Phylogenetic Analysis

Nucleotide sequence identity was searched using BLAST, multiple sequence alignment was performed using ClustalW and phylogenetic trees were constructed using MEGA 5.05 and the neighbor-joining method14, based on the nucleotide sequences. Bootstrap analysis of 1000 replicates was performed to determine the significance of the branching of the constructed tree.

Data Analysis

Demographical and epidemiological results were recorded for analysis using Microsoft Excel software.

The first case of ZIKV infection was a 61-year-old man from the Likas area of Kota Kinabalu. He went to Kuala Lumpur on August 25 and then to Johor Bahru on August 27, during which time he had fever. On August 29, he returned to Kota Kinabalu, developed watery diarrhea three times per day, and was weak. On August 30, he went to Luyang Health Clinic, where a blood smear for malaria and an NS1 test for dengue virus were negative. He was prescribed paracetamol and went home. On August 31, he went to the Emergency Department of Queen Elizabeth Hospital 2. He was again suspected of having dengue, however again the NS1 test was negative. On September 1, the patient was suspected of having ZIKV infection because of his history of visiting Johor Bahru, which borders Singapore. Samples were sent for the detection of ZIKV and infection was confirmed by RT–qPCR. On September 2, he was transferred to the High Dependency Unit, but died on September 3, because of septic shock with multiorgan failure. The urine sample from September 1 was also positive for NS5 (192 bp and 214 bp) and envelope protein (365 bp) genes by RT–PCR. We designated this strain SZ1-2016. All amplicons were successfully sequenced.

In search and destroy activities within a 400-m radius of the patient's home, 477 of 556 (85.7%) houses were inspected. Mosquito breeding places around the locality showed that 11 of 1946 containers examined were positive for larvae. The Aedes Index (AI) was 0%, the Breateau Index (BI) was 0% and the Container Index (CI) was 1.49%.

The second case was a 60-year-old woman; a resident of Damai, Kota Kinabalu. On September 24, she noticed rashes on her neck and back (Figure 1).

Maculopapular rash of Zika virus infection Maculopapular rash of Zika virus infection Maculopapular rash on the back of the second case of Zika virus infection in Sabah.

She had no fever, conjunctivitis, arthralgia, or arthritis, and no recent history of traveling to ZIKV-endemic countries or history of contact with confirmed ZIKV cases. From September 13 to 18, she visited several places and parks to take photos of wild animals and birds, such as Kawang Forest Centre in Papar district which is 30 km from Kota Kinabalu, Perdana Park Tanjung at Kota Kinabalu, Kinarut Outward Bound School Papar, Sewerage plant in Kota Kinabalu, and Shopping malls in Kota Kinabalu. She acknowledged that on September 23, she was bitten by mosquitoes while taking photographs in Sabah Museum, Kota Kinabalu. Both serum and urine samples from September 28 were positive for ZIKV, as were urine sample from September 30 (6 days after onset) and October 7 (13 days after onset). The urine sample from October 14 (20 days after onset) was negative for ZIKV. The urine sample but not serum sample from September 30 was positive for NS5 and envelope protein genes by RT–PCR. We designated this strain SZ2-2016. All amplicons were successfully sequenced.

During search and destroy activities within a 400-m radius of the patient’s residence, 236 of 269 (87.7%) houses were inspected. Mosquito breeding places around the locality showed that 2 of 1202 containers examined were positive for larvae. The AI was 0 %, BI 0%, and CI 0.3 %.

The third case was a 57-year-old female Taiwanese tourist. Her tour group visited Sabah from September 23 to 27, 2016. On September 24 the group visited Kinabalu park in Mount Kinabalu where they did canopy walk. On September 25 visited Borneo Kellybays in Tuaran district for water and land activities in mangrove park. On September 26 they went to Klias, wetland in Beaufort, 110 km from Kota Kinabalu. The main attractions at this wetland are proboscis monkeys and fireflies. The group continued visiting other places around Kota Kinabalu on September 27. On September 28, the patient returned to Taiwan. On October 5, 2016 she developed headache, skin rash, muscular pain, and fever, and was hospitalized from October 6–8. Her urine sample from October 8, was positive for ZIKV by RT–PCR. The search and destroy activities in 15 places she visited showed that only 2 of 479 containers examined were positive for Aedes larvae. The AI was 6.6 %, BI 6.6, and CI 0.4%.

All 10 household members of the first ZIKV-infected patient and one member of the second were negative for ZIKV by RT–qPCR. From Tanjung Lipat, samples were collected from 18 volunteers out of about 200 people who attended the health education activities. Among them, only one had a skin rash and another had a travel history to Singapore in May 2016. From Damai, samples were obtained from eight volunteers of about 50 attendees. All were asymptomatic and had no recent history of traveling to Singapore. All urine samples were negative for ZIKV RT–PCR.

There were 30 pooled samples of mosquitoes. The number of mosquitoes in each pooled sample varied from 3 to 11 per tube with a median value of 10/tube. Two pools contained 10 and 8 Culex fascocephala mosquitoes and one pool contained 10 Culex spp.; otherwise all samples (255) were A. albopictus. All mosquitoes were female and were negative for ZIKV by RT–PCR. A total of four serum samples were collected from wild M. fascicularis: three were male and one female. All were negative for ZIKV by RT–PCR.

Phylogenetic tree of NS5 gene

Phylogenetic tree constructed based on the nucleotide sequences of the NS5 gene of ZIKV strains. All strain names begin with the DNA Data Bank of Japan/European Molecular Biology Laboratory/GenBank accession numbers. The strains detected in the first (SZ1-2016) and second (SZ2-2016) cases in Sabah are indicated by black squares, which are followed by the strain numbers. The numbers adjacent to nodes represent the bootstrap values; values less than 70% are not shown. The scale bar shows the genetic distance, which is expressed as nucleotide substitutions per site.

Phylogenetic analyses of the partial nucleotide sequences of NS5 (232 nt) and envelope protein (365 nt) gene showed that the ZIKV from Sabah belongs to the Asian lineage. In the NS5 tree, SZ1-2016 and SZ2-2016 were in a big cluster containing recent outbreak strains from the Americas, Singapore, Thailand, Haiti, and French Polynesia (Figure 2), indicating that the Sabah ZIKV shares a common ancestor with these strains. However, SZ1-2016 and SZ2-2016 were not close to each other. The nucleotide and amino acid identities between SZ1-2016 and SZ2-2916 were 98% and 100%, respectively.

The envelope protein genes from all three ZIKV strains from Sabah formed an independent cluster with strains detected in 2013 and 2010 from Thailand and Cambodia, respectively (Figure 3).

Phylogenetic tree of envelope protein genes of ZIKV strains.

Phylogenetic tree constructed based on the nucleotide sequences of the envelope protein gene of ZIKV strains. All strain names begin with the DNA Data Bank of Japan/European Molecular Biology Laboratory/GenBank accession numbers. The strains detected in the first (SZ1-2016) and second (SZ2-2016) cases in Sabah are indicated by black squares, which are followed by the strain numbers. KY126348.1 Malaysia/2016/1609Tw was detected in a Taiwanese patient visited Sabah. The numbers adjacent to nodes represent the bootstrap values; values less than 70% are not shown. The scale bar shows the genetic distance, which is expressed as nucleotide substitutions per site.

The Sabah strains were very close to each other and shared 97–99% and 100% nucleotide and amino acid identities, respectively. Our strains shared 97–99% nucleotide and 100% amino acid identity with 2016 outbreak strains from other parts of Malaysia (MyH318_JB, MyH326PL_JB, and MyH334PL_Sar) and Singapore (SG-108, and SG-117). These strains shared 97–100% and 100% nucleotide and amino acid identity among themselves.

Surveillance, risk assessment, and intervention were strengthened throughout Malaysia in response to the 2016 outbreak of ZIKV in neighboring Singapore. The Malaysian Ministry of Health undertook regular surveillance from June 2015 during the South American outbreak, but no ZIKV was detected in 784 samples tested up to August 2016 (https://kpkesihatan.com/2016/08/28/kenyataan-akhbar-kpk-28-ogos-2016-situasi-terkini-virus-zika-di-malaysia/). However, during the peak of the 2016 Singapore ZIKV outbreak from September through December 2016, eight out of 849 samples tested were positive for ZIKV (https://kpkesihatan.com/2016/12/18/kenyataan-akhbar-kpk-18-disember-2016-situasi-terkini-zika-di-malaysia-kes-zika-ke-8/). Except for the cases detected in Sabah, all ZIKV infections identified in Malaysia were directly or epidemiologically linked to the outbreak in Singapore. Therefore, the epidemiology of ZIKV infection in Sabah deserves special attention.

Because of a case report of ZIKV infection in a German traveler before the recent worldwide outbreak of ZIKV7 and serological evidence of ZIKV human infection in Sabah and nearby Labuan island more than six decades ago9, the report of three autochthonous cases of ZIKV infection rang an alarm bell that the ZIKV in Sabah might be phylogenetically distinct from the recent worldwide outbreak strains. Phylogenetic analysis of the envelope protein gene confirmed that the ZIKVs circulating in Sabah is different from the recent outbreak strains but related to strains from Cambodia and Thailand. The nucleotide sequences of the envelope protein gene of four recent outbreak strains of ZIKVs from Malaysia available in the GenBank are shorter in length therefore we could not use those in our phylogenetic analysis. Except strain MyH319_Sbh other three Malaysian strains are epidemiologically linked to Singapore outbreak strains therefore Singapore strains used in our phylogenetic analyses can represent them. Strain MyH319_Sbh and our strain SZ1-2016 were same because they were detected in the samples from the same patient in two different laboratories. Phylogenetic analyses of the NS5 gene supported the difference of Sabah strains by showing longer branch length, indicating that several substitutions have occurred compared with other strains lineage. In the phylogenetic tree of envelope genes, SZ1-2016 and SZ2-2016 clustered together, but are relatively divergent in NS5 tree. This is not unexpected for ZIKV as has been reported previously15. The nucleotide sequence of the NS5 gene of the strain detected in the Taiwanese patient who traveled Sabah and strains detected in other parts of Malaysia were not available in GenBank. Therefore we could not include those in the phylogenetic analysis. A full genome sequence of ZIKV is needed to conclusively determine the evolutionary history of the Sabah strain. Sabah is a popular tourist destination: the most recent five-year data showed that each year an average of 97,038 (845,910–1,089,320) international visitors arrived (http://www.sabahtourism.com/business/statistic). Therefore, it is possible that the movement of this large number of people brings strains from other parts of the world that then evolve locally.

The previous report of the presence of ZIKV antibodies in orangutans8^, 10 prompted us initially to believe that ZIKV has a sylvatic transmission cycle in Sabah. Although all serum samples from nonhuman primates in this study were negative for ZIKV, because of the small sample size we cannot rule out the existence of a sylvatic cycle. In Africa, ZIKV exists in a sylvatic transmission cycle involving nonhuman primates and forest-dwelling Aedes mosquitoes, but such a cycle has not been identified in Asia16^, 17.

Our unpublished data showed that A. albopictus is the predominant Aedes mosquito in Kota Kinabalu, as exemplified also by the species of mosquitoes collected during the present study. It is not known why all A. albopictus collected in the present study were negative for ZIKV. In urban and suburban environments, ZIKV is transmitted via a human–mosquito–human transmission cycle involving A. aegypti and to a lesser extent, A. albopictus16. It should be noted that the first ZIKV isolated in Malaysia in 1966 was also from A. aegypti5 . Even in A. aegypti, the competence to transmit ZIKV depends on the mosquito strain17. During the recent Pan-American epidemics, A. aegypti was reported as the primary and the only confirmed vector for ZIKV18. However, a study from Singapore showed that A. albopictus is able to transmit ZIKV, although the study used Ugandan not Asian strain19. Whereas, using different mosquitoes and viral strains it was concluded that A. albopictus is capable of transmitting ZIKV, the competence is potentially depended on geographic origin of both the mosquito population and viral strain20. Therefore more study is needed to find out whether local A. albopictus is competent to transmit ZIKV or the prevalence of A. albopictus infected with ZIKV is low in Sabah. Furthermore, our results showing no ZIKV in Culex fascocephala and Culex spp. may support the findings that although ZIKV has been isolated in 20 species of the genera Aedes, Anopheles, Eretmpodites, and Mansonia21, C. quinquefasciatus, C. pipiens and C. torrentium are refractory to virus transmission22^, 23 .

Because no ZIKV infection was detected in other parts of the state, suggesting that the infection was possibly limited to Kota Kinabalu. Further tests on volunteers confirmed that ZIKV transmission occurred on a limited scale. The educational activities for dengue vector control are an ongoing campaign by the Sabah State Health Department. From July through December 2016, educational activities about dengue and ZIKV infections were conducted in 26 dengue hotspots. These efforts might have a considerable impact on the vector population, resulting in limited ZIKV transmission in Sabah. This is supported by the fact that ZIKV was not detected in any mosquitoes tested and that the AI, BI, and CI indices were all considerably lower than the standard set by the Ministry of Health Malaysia in which the standard AI, BI, and CI indices are 1%, 5%, and 10%, respectively.

In conclusion, we consider that the ZIKV currently circulating in Sabah belongs to the Asian lineage and different from the recent outbreak strains in other parts of the world. We consider that the rate of ZIKV infection in Sabah is low and sporadic, possibly because of limited transmission of ZIKV by the presence of mainly A. albopictus. It remains to be determined whether these findings represent the results of ongoing vector control measures and public education by the Sabah State Health Department or a unique epidemiology of ZIKV in Sabah. More surveillance is needed to determine whether any sylvatic cycle exists in nonhuman primates in Sabah. Therefore, a large prospective study collecting serum samples from humans, animals, and mosquitoes from different areas is needed to determine the extent of ZIKV distribution in Sabah.

The authors have declared that no competing interests exist.

Jiloris Julian Frederick Dony is the corresponding author. He can be contacted at (jiloris@moh.gov.my). The authors also select Kamruddin Ahmed as another corresponding author. His email address: ahmed@ums.edu.my.

Access to data is restricted to protect the confidentiality of individuals and premises involved in this outbreak. Researchers interested in accessing anonymized minimum data should write to Dr Julaidah Binti Sharip (pkk.kotakinabalu@moh.gov.my), Research Documentation Committee,Kota Kinabalu Area Health Office, who will assess the request.