The study was approved by the Institutional Human Ethics Committee at the (IHEC Number-NIV/IEC/Mar/2021/D-9 dated 9th April 2021) and the Institutional Biosafety Committee at ICMR-NIV, Pune.

The retrospective serum, urine, plasma, and whole blood samples from May to October 2021 were tested for ZIKV as per the following inclusion criteria: (a) collected in the acute phase of infection (within 7 days of onset of illness); (b) screened for DENV and CHIKV by IgM ELISA; and (c) stored at –20°C at the respective centers.

A total of 1520 clinical samples, serum (1253), plasma (99), whole blood (120), and urine (48) were collected from 1,475 patients across 16 VRDLs from 14 States/Union Territories (UTs) in India [Delhi, Kerala, Punjab, Rajasthan, Uttar Pradesh (UP), Uttarakhand, Madhya Pradesh (MP), West Bengal, Bihar, Odisha, Telangana, Assam, Jharkhand, and Bihar]. Samples fulfilling the inclusion criteria were packed in dry ice in triple-layer packaging, as per the International Air Transport Association (IATA) guidelines. The samples were subsequently transferred to the apex laboratory at ICMR-NIV, Pune, for molecular diagnosis, serology, and genomic analysis.

All samples were accompanied by a case report form (CRF) of the patient, among which most of the forms were incomplete with respect to symptoms, comorbidities, clinical management, complications, and outcomes. Therefore, all cases found positive for ZIKV by qRT-PCR (n = 67) were interviewed telephonically in English, Hindi, and Malayalam language and the above details were elicited. The telephonic interviews also helped in validating the information captured in the CRF forms. The patients were surveyed for their demographic details, duration of illness, and clinical symptoms including details of hospitalization. Each call lasted for at least 10 min.

All 1,520 clinical samples received from the VRDLs at the apex laboratory at ICMR-NIV, Pune were screened for the presence of ZIKV, DENV, and CHIKV by CDC trioplex real-time RT-PCR assay (Santiago et al., 2018). Further, the Zika positive samples were screened with monoplex ZIKV real-time reverse transcriptase polymerase chain reaction (rRTPCR) (Lanciotti et al., 2008) and the quantification of the RNA copy numbers was performed by a standard curve generated using ZIKV in vitro transcribed RNA.

The anti-Zika IgM antibodies were determined using ZIKV Detect IgM Capture Enzyme-linked Immunosorbent Assay (ELISA) (InBios International Inc., Seattle, WA, United States) as per the manufacturer’s instructions (In-Bios, 2017). For each sample, an immune status ratio (ISR) was calculated by dividing the ZIKV recombinant envelope (E) glycoprotein antigen absorbance (OD₄₅₀) value by the cross-reactive control antigen OD₄₅₀ value. ISR value > 1.90 indicates presumptive ZIKV positive.

The anti-dengue IgM ELISA was performed by using ICMR-NIV dengue IgM capture ELISA kit in a subset of positive samples. Briefly, in prewashed anti-human IgM coated wells, 50 μl diluted serum samples (1:100) were added. The plate was incubated at 37°C for 1 h. In the subsequent step Dengue antigen (50 μl) was added into each well and incubated again at 37°C for 1 h. At the end of the incubation, 50 μl of avidin-HRP was added and incubated at 37°C for 1 h. Plates were washed five times after each incubation. The TMB substrate (100 μl) was added and the plate was incubated at room temperature for 10 min. After adding the stop solution, the optical density (OD) was measured at 450 nm. The OD value of less than 2.0 was considered as negative, and more than 3.0 as positive while OD values between upper and lower cut-off (2.0–3.0) were considered as equivocal. Respective positive and negative controls were used in the assay (Cecilia et al., 2011).

The positive samples were further analyzed for the neutralizing antibody (NAb) titers by the plaque reduction neutralization test (PRNT90) as described earlier (Russell et al., 1967; Ward et al., 2018; Ribeiro et al., 2020). The presence of neutralizing antibodies was determined as described by Russell et al. with some modifications (Ribeiro et al., 2020). Vero CCL-81 cell suspension (1.0 × 10⁵/mL/well) was added into 24-well tissue culture plates and was incubated for 16–24 h at 37°C with 5% CO₂ to obtain confluent monolayer. The serum samples were inactivated at a 56°C-water bath for 30 min. A ten-fold serial dilution of serum samples were mixed with an equal volume of virus suspension containing 50–60 plaque-forming units (PFU/0.1 ml) and incubated at 37°C for 1 h. The virus–serum mixtures (0.1 ml) were added onto the Vero CCL-81 cell monolayers and incubated at 37°C with 5% CO₂ with intermittent shaking for uniform absorption of the virus. After 1 h, the suspension was aspirated, and the cell monolayer was gently overlaid with an overlay medium containing 2% carboxymethylcellulose (CMC) and 2X MEM containing 2% FBS (1:1). The plates were further incubated at 37°C with 5% CO₂ for 4 days. The plates were terminated by decanting overlay medium followed by 0.1% saline wash and were stained with 1% amido black for 1 h at room temperature. The plates were washed in running tap water and air-dried. Plaque number was counted, and antibody titers are defined as a reciprocal of the highest dilution of tested serum that resulted in a reduction of viral infectivity by 90% when compared to the non-neutralization control (Russell et al., 1967; Ward et al., 2018).

Zika virus positive samples were sequenced using next generation sequencing (NGS) and the standard sanger sequencing method as described earlier (Yadav et al., 2018, 2020). Genomic characterization of the viral specimens was carried out using NGS with the quantified RNA. Briefly, RNA was extracted using Magmax RNA extraction kit (Thermo Fisher Scientific, United States), quantified by Qubit™ RNA high sensitivity (HS) kit (Thermo Fisher Scientific, United States) and the RNA libraries were prepared by using TruSeq stranded total RNA library preparation kit. Quantification of libraries was performed using Qubit™ dsDNA HS kit (Thermo Fisher Scientific, United States; kit and Qibitvrsion 2.0). Quantified libraries were pooled, normalized, and loaded on the Illumina Nextseq500/550 sequencing platform. Reference-based mapping was performed to retrieve the complete genome sequence of the virus isolates using the CLC Genome Workbench v20. Samples, where the complete genome could not be retrieved, were further tried for partial envelop [E] gene sequencing using the Sanger sequencing method. As described above, DENV positive samples having a cyclic threshold (Ct) value of < 30 were also sequenced using the TruSeq stranded total RNA library preparation kits and Miniseq sequencing platform. The complete genome sequences were retrieved using reference-based mapping with the DENV-I, II, III, and IV serotypes. A neighbor-joining (NJ) phylogenetic tree was constructed from the coding region of the ZIKV complete genome, partial E-gene, DENV1, DENV2, DENV3, and DENV4 separately using the Tamura 3-parameter model along with gamma distribution as the rate variation parameter. A bootstrap replication of 1,000 cycles was performed to assess the statistical robustness of the generated tree. The amino acid variation for each gene, as well as the net nucleotide and amino acid divergence were identified using the MEGA software version 7.0. The pairwise comparison was used to find the percentage identity for nucleotide and amino acid using CLC genomic workbench version 2.0.

For the partial ZIKV E gene amplification, the primer sets used were: Zika_F1555:AGGACAGGCCTTGACTTTTCAGAT and Zika_R_2350:GTGAGAACCAGGACATTCCTCC generating 795bp product. This was followed by a heminested PCR using Zika_F1555: AGGACAGGCCTTGACTTTTCAGAT and Zika_R_1950: TACTGTACCTCCACTGTGACTGT resulting in a 400bp product. The RT-PCR conditions followed for amplification were: reverse transcription at 50°C for 10 min, initial PCR activation at 98°C for 2 min, 35 amplification cycles of denaturation at 98°C for 10 s, annealing at 52°C for 10 s, extension at 72°C for 1 min. The PCR conditions for heminested PCR were: initial denaturation at 98°C for 2 min, 35 cycles of denaturation at 98°C for 10 s, annealing at 52°C for 30 s extension at 72°C for 30 s, and a final extension at 72°C for 5 min. The partial E gene was further sequenced by the sanger sequencing method and used for analysis as described earlier (Yadav et al., 2020).

After rRTPCR testing of 1,520 clinical samples, 75 samples from 67 patients were found to be positive for ZIKV, three of which were cases of co-infection. ZIKV and DENV co-infection was found in two cases while ZIKV, CHIKV, and DENV co-infection was observed in one case. Apart from this, samples from 121 patients were DENV positive while 10 were CHIKV positive. Dual positivity of DENV and CHIKV was observed in five cases. Serotyping of dengue was done and circulation of all four serotypes was seen. DENV-II was found to a larger extent in 26 patients followed by 7 cases of DENV-IV, 2 cases of DENV-I, and one case of DENV-III. Diagnosis of the DENV and CHIKV positive samples had been missed in initial screening at VRDLs as the primary test conducted there was NS1 and IgM ELISA for DENV and IgM ELISA for CHIKV. The ZIKV RNA log copy number ranged from 4.1 to 7.1 for plasma (n = 28) and urine (n = 10) samples, while 5.0–8.4 for serum samples (n = 29).

The ZIKV positive samples detected during retrospective testing at ICMR-NIV, Pune revealed the presence of ZIKV in Amritsar, Punjab (1/120); New Delhi (1/64); Aligarh, UP (2/288); Jodhpur, Rajasthan (1/120); Ranchi, Jharkhand (1/120); Hyderabad, Telangana (1/60), and Thiruvananthapuram, Kerala (60/120), during 2021 (Table 1). This positivity was additional to the earlier reported positivity in the same year from Thiruvananthapuram, Kerala (166 cases) (unpublished data, Govt. of Kerala); Kanpur, UP (151 cases) (unpublished data, VRDL at King George Medical University, Lucknow); and Pune, Maharashtra (1 case). The retrospective analysis undertaken by us shows the spread of ZIKV to newer states and cities in India such as Delhi, Jharkhand, Punjab, Telangana, Jodhpur, and Aligarh. The ZIKV positivity so far reported in India is depicted in Figure 1.

Among the 67 ZIKV positive cases, 13.43% (Rocha et al., 2019) were hospitalized while 86.56% (58) of the cases were managed on an outpatient basis. The median age and the interquartile range of the patients included in the study were 30 years (22−43 years). A total of 34 (50.7%) of ZIKV-positive cases were males. All 67 patients were symptomatic with fever 56 (83.6%) and rash 52 (77.6%) being the most prominent. Followed by arthralgia 30 (44.9%) and conjunctivitis 23 (34.3%). Microcephaly was reported in one patient. The median duration of illness amongst all cases was 6 days (range 2 to 10 days) (Figure 2 and Supplementary Table 4).

Three cases of co-infection of ZIKV with other flaviviruses were confirmed by qRT-PCR. One case each from Aligarh, UP, and Thiruvananthapuram, Kerala were confirmed for DENV and ZIKV positivity by qRT-PCR assay while one case (51 years/M) from Amritsar, Punjab was confirmed to have co-infection with all three, namely, DENV, ZIKV, and CHIKV. There was no significant difference in the profile of clinical symptoms and duration of illness between the patients suffering from standalone ZIKV and those with co-infection (CHIKV and DENV).

Six cases presented with atypical clinical presentation as compared to the rest. This included respiratory distress (n = 4 cases), seizure (n = 1 case), and suspected microcephaly at birth (n = 1 baby) related to ZIKV infection and of them three cases were hospitalized. Details of the cases are as follows:

(1) Four cases with respiratory distress—Four ZIKV positive cases from Rajasthan (24/M; case 1), Telangana (21/F; case 2), and Kerala (88/M; case 3 and 33/M; case 4) states presented with the complaints of respiratory distress screened negative for SARS-CoV-2 and Influenza. Their fever lasted for an average duration of 4–5 days. Only case 3 was hospitalized in August 2021 with complaints of fever and loss of appetite. He was known hypertensive and asthmatic. The illness resolved uneventfully in 4 days and the patient was discharged. Cases 2, 3, and 4 presented with respiratory distress, high-grade fever, and arthralgia. Cases 1 and 2 additionally reported vomiting and rash. None of the patients required supplemental oxygen or ventilation. The respiratory distress resolved within 4–5 days.

(2) A Case with seizure—A 45-year-old female presented with acute onset fever, arthralgia, conjunctival congestion, maculopapular rash, and convulsions in a tertiary care hospital in Thiruvananthapuram, Kerala. She had a previous history of seizures in 2002 after sustaining a traumatic head injury in a road accident. She was not on any anti-epileptic medication for the past several years. At the time of admission, her MRI brain was normal while EEG was abnormal with an intermittent sharp activity which was non-generalizable and focused in the left temporal region. As a management for seizures, she was given the anti-epileptic drug lacosamide (100 mg BD). She tested positive for ZIKV and was discharged post-recovery after 5 days of admission. She was asked to continue the anti-epileptic medications and no sequel has been observed during the follow-up visit in February 2022.

(3) A case of suspected microcephaly at birth—From Thiruvananthapuram, Kerala, a preterm (35 weeks + 5 days) baby girl weighing 1.68 kg (small for gestational age) was delivered in July 2021 by cesarean section due to placenta praevia. The baby was a suspect case of microcephaly as her head circumference at birth was 28 cm. According to Indian Standards for Fetal biometry head circumference at 35 weeks is 30 cm (Warrier and Ashokan, 2016). Her APGAR score was normal and there was no neurological deficit, seizure, or limb weakness. No visual or hearing impairment or any other signs of congenital zika syndrome were recorded (Freitas et al., 2020). Within 2 days of birth, she was diagnosed with congenital broncho-pneumonia and neonatal jaundice. The baby was intubated and started on empirical treatment with broad-spectrum antibiotics. Her serum samples were positive for ZIKV RNA by qRT-PCR and negative for TORCH pathogens (Toxoplasma, Rubella, Cytomegalovirus, Herpes Simplex). Serum samples of her mother were ZIKV negative by qRT-PCR, suggestive of probable infection resolution after initial transient viremia. The baby recovered after 12 days of birth and was discharged with advice to further follow up. She was regularly followed up for the gross motor, fine motor reflexes, head and chest circumference, length, weight gain, and neurological development. She gained normal head circumference as per the Indian Academy of Pediatrics growth chart (37 cm/40.5 cm) post 3/6 months of birth. During her last follow-up at 6 months of age, she had shown good weight gain, normal head circumference (40.5 cm), no neurological, ocular, auditory problems, or any developmental milestone delays.

Out of 29 ZIKV positive samples from different patients, four samples showed the presence of anti-Zika IgM antibodies [Thiruvananthapuram (n = 3) and Aligarh (n = 1)]. The anti-Zika IgM positive samples were also checked for flavivirus cross-reactivity and one sample was found to be positive for anti-dengue antibodies while the remaining three were negative by ELISA. In addition, the Zika PRNT90 assay confirmed the presence of Zika-specific antibodies in all four samples.

Next generation sequencing was performed on all the ZIKV, DENV, and CHIKV positive samples, for identifying the currently circulating ZIKV, DENV, and CHIKV strains in various parts of the Indian population. The analysis involved generating a consensus sequence from all sequence reads for each sample using CLC genomic software version 21.0.4. Out of the 67 ZIKV positive cases [Ct value range—29–33], the partial fragment of the E gene was sequenced for 32 ZIKV positive cases while one complete genome was retrieved (Figures 3A,B). To better understand the relationship between these sequences, we used the envelope coding region to construct the phylogenetic analysis between the ZIKV strains (Figure 3B). The phylogenetic tree shown in Figure 3B indicated that ZIKV sequences obtained in the current study belonged to the Asian lineage and were closely related to the strain previously reported from India (Yadav et al., 2019; Malhotra et al., 2020). The percent identities were calculated in CLC Genomics workbench version 22.0. The identities observed were 89%, 95.53%, and 99.67% at the nucleotide level and 96.75%, 96.99%, and 99.397% at amino acid (AA) level (Supplementary Table 1) between the complete genome of ZIKV obtained in this study and the ZIKV sequences from Uganda (NC_012532), Kerala (OM666892), and Rajasthan (MK238036), respectively (Figure 3A). The comparison at the nucleotide and AA level was performed for the partial envelope gene with the sequences from Kerala and Rajasthan and it indicated 97%–100% and 95%–99%, and 96%–100% and 92%–100% identity, respectively (Figure 3B and Supplementary Table 1). All the ZIKV sequences from the current study belong to Asian lineage (Figures 3A,B). The ZIKV sequences in the current study sequences were matched with the reference genome of ZIKV (NC_012532) from Uganda. Whereas the 32 ZIKV partial sequences showed an average of 84% nucleotide and 91% amino acid identities with the reference genome of ZIKV (NC_012532) from Uganda (Supplementary Table 1). The amino acid changes observed are depicted in Supplementary Figure 1. The point mutation in prM-S139N reported amongst the children with microcephaly was not observed in the retrieved ZIKV sequences. The NS1-A188V mutation reported to enhance ZIKV infectivity in the laboratory Aedes aegypti mosquitoes was also not observed in the retrieved sequence (Liu et al., 2017). The genomic data retrieved in the current study are provided in Supplementary Table 3.

Out of 129 DENV positive samples, sequences from 36 serum samples were retrieved and analyzed with neighbor-joining phylogenetic method using Tamura-3 as substitution model and a bootstrap replication of 1,000 cycles. The analysis indicated the circulation of DENV-1 (n = 2), DENV-2 (n = 26), DENV-3 (n = 1), and DENV-4 (n = 7) serotypes in different geographical locations of the country. The genomic data retrieved for the DENV (1-4) (Dick, 1952; Moore et al., 1975; Rothan et al., 2018; Pielnaa et al., 2020) is given in Supplementary Table 2. The presence of the DENV-2 serotype was confirmed in Haldwani, Aligarh, Jodhpur, Jabalpur, and Secunderabad districts. The DENV-4 serotype was seen in Aligarh and Punjab while the DENV-III serotype was confirmed from Thiruvananthapuram. The co-circulation of all the three DENV-1, DENV-2, and DENV-4 serotypes was seen in Aligarh. The phylogenetic analysis of the two sequences of DENV-1 serotype showed alignment with genotype V of the Indian sequence. Furthermore, 26 sequences of the DENV-2 formed two distinct genotypes [Cosmopolitan I (n = 12) and Cosmopolitan III (n = 14)] having similarities with the sequences from Pakistan, Sri Lanka, India, and China. One sequence of DENV-3 clustered with sequences from India, Pakistan, Brazil, Singapore, and Thailand which belonged to genotype III. While seven sequences of the DENV-4 clustered with sequences from Tamil Nadu and belonged to genotype I (Figures 4A–D). Of the CHIKV positive samples, the Ct range was higher (Asadi-Pooya, 2016; Warrier and Ashokan, 2016; Landry and St George, 2017; Lina Villa et al., 2017; Liu et al., 2017; Ward et al., 2018; Yadav et al., 2018, 2020; Freitas et al., 2020; Malhotra et al., 2020; Silva et al., 2020; Low et al., 2021; Stiasny et al., 2021) and NGS could not lead to giving useful sequences to be used for analysis. The amino acid changes in the NS5 protein of DENV-1-DENV-4 observed with respect to its reference are depicted in Supplementary Figures 2, 5 respectively. The NS5 protein of DENV is nearly 900 peptide residue with multifunctional activity. The 1/3rd region toward the N terminal encoding for the methyltransferase and the rest toward the C terminal encodes for the RNA-dependent RNA Polymerase (RdRp). It is observed that the mutations are observed in both the segments of the protein. CHIKV genomic sequences were not retrieved from the real-time RT-PCR samples in the study; the reason being low volume and higher CT values of the samples.