Abstract
Rotaviruses are among the major causes of viral acute gastroenteritis in newborns and children younger than 5 years worldwide. The ability of rotaviruses to remain infectious in harsh environments as well as in the wastewater treatment process makes them one of the most prevalent enteric viruses. The current study aimed to determine the presence of rotavirus genomes and to analyze them phylogenetically in secondary treated wastewater (TW) samples. In total, 13 TW samples were collected from September 2017 to August 2018. Viral concentration was carried out using the absorption-elution method, and after RNA extraction and cDNA synthesis, real-time and conventional polymerase chain reaction (PCR) were performed. A phylogenetic tree was drawn using Maximum Likelihood and Tamura 3-parameter using MEGA v.6 software. Rotavirus genomes were detected in 7/13 (53.8%) and 3/13 (23.07%) samples using reverse transcription (RT)-PCR and conventional PCR, respectively. Accordingly, phylogenetic analysis revealed G4P[8], G9P[4], and G9P[8] genotypes among the samples. The presence of rotavirus in secondary TW samples discharged into surface water emphasizes the importance of monitoring and assessing viral contamination in the water sources used for agricultural and recreational purposes.
HIGHLIGHTS
Rotaviruses are one of the most important enteric viruses that can transmit via fecal contaminated water.
Treated wastewater was monitored for rotaviruses for 1 year.
Rotaviruses genome were detected in samples using RT-PCR by conventional PCR.
Phylogenetic analysis revealed three G4P[8], G9P[4], and G9P[8] genotypes among samples.
INTRODUCTION
Numerous documented waterborne outbreaks have dramatically increased the importance of water resources sanitation (Wu et al. 2011). In spite of strict control on drinking and authorized services of wastewater treatment processes, it seems that waterborne diseases are still one of the most critical public health issues throughout the world (Okoh et al. 2007).
Enteric viruses are significant causes of waterborne diseases which may lead to great economic and social impact, particularly in developing countries. Viral gastroenteritis usually happens due to contamination of drinking/recreational water with sewage. On the other hand, since treated wastewater (TW) is widely used in agriculture, unsuccessful treatment processes may bring infectious virus particles to farmland and raise the risk of viral contamination of crops (Rodríguez-Díaz et al. 2009). Enteric viruses are host-specific and since fecal-oral transmission is the main route of infection, their presence in water resources could be evidence of fecal contamination, and monitoring for these viruses may be a useful tool to monitor for fecal pollution in water systems (Fong et al. 2005).
Rotaviruses, as one of the major waterborne viruses, are classified into the Reoviridae family. They are characterized by segmented double-stranded RNA with a three-layer capsid (Nick et al. 2019).
As with other enteric viruses, the fecal-oral route is the predominant route of rotaviruses transmission. They cause gastroenteritis with the symptoms of vomiting, acute diarrhea, and dehydration (Shrestha et al. 2019).
Although many studies have been performed to monitor the occurrence of rotaviruses in raw and TW in the world, there is sparse data on molecular detection and characterization of the viruses in water samples in Iran (Rodríguez-Díaz et al. 2009; Kargar et al. 2013; Ruggeri et al. 2015).
The aim of the current study was to assess the secondary TW samples of a wastewater treatment plant (WWTP) in Tehran City for contamination by rotavirus which could give rise to environmental and public health concerns. Virus molecular characterization of positive samples was also used for genotyping and demonstration of the common genotypes disseminated in the study area.
MATERIALS AND METHODS
TW samples
The capacity of the studied wastewater plant was 4.5 × 108 m3 located in the south of Tehran, Iran. The final TW is fed to an uncovered canal and used for irrigating the farmlands nearby. The sampling was conducted once a month (two for October) for 1 year, from September 2017 to August 2018, from secondary TW. A total of 13 samples were collected in a sterile bottle and immediately transferred to the laboratory on ice for further analysis. Sample specifications including sampling date, turbidity, pH, chemical oxygen demand (COD), and biological oxygen demand (BOD) were recorded.
Concentrating the samples
The sample concentration method utilized in the present study was described previously by the United States Environmental Protection Agency (USEPA) (Fout et al. 2010). Briefly, 5 L of secondary TW samples was aseptically passed through positively charged membranes (Zeta Plus 1MDS, 3M, USA, 0.45 μm pore size, 47-mm diameters) via the filtration apparatus with a vacuum pump 24 L/s (Sartorious, Goettingen, Germany).
The turbidity and pH of the collected TW were measured; if the turbidity was greater than 50 NTU (nephelometric turbidity unit), then the samples were pre-filtrated with sterile 47-mm membrane filters (MF-Millipore Membrane Filter, 0.8 μm pore size, USA). Also, for the samples with pH values of greater than 8.0, it was adjusted to 6.5–7.5 before starting the concentration process. The elution procedure was conducted as mentioned before (Fout et al. 2010; Tavakoli Nick et al. 2019). Briefly, 7.5 mL (per-filter) of 1.5% beef extract (wt/vol) with 0.05 M glycine (pH 9.5) solution was gently passed through a filter twice. Afterwards, the eluate was acidified and flocculated by adjusting the pH to 3.5 ± 0.1, stirred gently at room temperature for 30 min, and centrifuged at 2,500 g for 15 min at 4 °C. The precipitate was suspended in 30 mL of sterile 0.15 M sodium phosphate solution (pH: 9.0–9.5) and centrifuged for 10 min at 4 °C at 6,000 g. The supernatant was then collected, and pH was adjusted to 7.0–7.5. Further, 10 mL of the eluate was re-concentrated using a centrifugal concentrator 30,000 MWCO (Millipore Amicon ultra-15 centrifugal filter unit, USA) and centrifuged at 4,000 g for 10–15 min at 4 °C to obtain a final volume of 500 μL (Fout et al. 2010).
Viral RNA extraction
Viral RNAs were extracted from 140 μL of the final concentrated elute using QIAamp RNA mini kit (QIAGEN, Germany) according to the manufacturer's instructions. The extracted RNAs were frozen at −70 °C.
Reverse transcription reaction
cDNA synthesis was carried out using Revertaid Reverse Transcriptase kit (Thermos Fisher Scientific, USA) according to the manufacturer's protocol.
Real-time and conventional polymerase chain reaction
Real-time polymerase chain reaction (PCR) was conducted at a final volume of 30 μL using rotavirus probe base kit (Altona, Hamburg) and Rotor Gene Q 5plex Platform system (QIAGEN, Germany). Conventional PCR was performed using two sets of primers to amplify the target fragments of VP7 and VP4 genes to confirm the results of real-time PCR. Conventional PCR with the first set of primers (VP7 primers) (Table 1) was performed using 3 μL of synthesized cDNA and 10 pmol of the first set primers at the final volume of 20 μL (Gómara et al. 2001). Conventional PCR with the second set of primers (VP4 primers) (Table 1) was performed using 3 μL of synthesized cDNA and 10 pmol of the second set primers at the final volume of 20 μL (Gentsch et al. 1992). The products of conventional PCR were then separated on a 1.5% agarose gel. The expected products were visualized by safe stain and UV image capture system (Vivilber E-BOX CX5) (the sizes of VP7 and VP4 PCR products were 881 and 663 bp, respectively). Finally, the PCR products were sent for bi-directional sequencing to Codon Genetics Group Company, Iran.
Conventional PCR primers used for rotaviruses detection
Primers . | Sequence (5-3) . | Product size . | References . | |
---|---|---|---|---|
First set primers | VP7-Forward | 5-ATGTATGGTATTGAATATACCAC-3 | 881 bp | Gómara et al. (2001) |
VP7-Reverse | 5-AACTTGCCACCATTTTTTCC-3 | |||
Second set primers | VP4-Forward | 5-TATGCTCCAGTNAATTGG-3 | 663 bp | Gentsch et al. (1992) |
VP4-Reverse | 5-ATTGCATTTCTTTCCATAATG-3 |
Primers . | Sequence (5-3) . | Product size . | References . | |
---|---|---|---|---|
First set primers | VP7-Forward | 5-ATGTATGGTATTGAATATACCAC-3 | 881 bp | Gómara et al. (2001) |
VP7-Reverse | 5-AACTTGCCACCATTTTTTCC-3 | |||
Second set primers | VP4-Forward | 5-TATGCTCCAGTNAATTGG-3 | 663 bp | Gentsch et al. (1992) |
VP4-Reverse | 5-ATTGCATTTCTTTCCATAATG-3 |
Sequence analysis and phylogenetic tree construction
Sequencing data were edited by BioEdit program (Hall 1999), aligned with the reference strains using the CLUSTAL W method (Thompson et al. 1994) and further compared with the NCBI GenBank database (http://www.ncbi.nlm.nih.gov). Phylogenetic trees were constructed using the maximum-likelihood method with 1,000 bootstrap replicates. MEGA 6 software (Tamura et al. 2013) was utilized to analyze the sequences and construct the phylogenetic tree. For each gene, the dataset was analyzed to determine the best-fit nucleotide substitution model with the lowest Bayesian Information Criterion score. Considering the best-fit models chosen, the Tamura 3-parameter (T92) nucleotide substitution model with the discrete Gamma distribution (G) and invariant sites (I) was used for VP7 sequences, and T92 + G was applied for VP4 sequences.
RESULTS
Biological oxygen demand (BOD) and COD of samples had a range of 5.6–180 mg/L and 23–448 mg/L, respectively. Also, the pH and turbidity were within the range of 7.22–7.87 and 6–10 NTU, respectively.
According to the real-time results, rotaviruses were detected in 53.8% (7 out of 13 samples) and conventional PCR in 23.07% (3 out of 11 samples) of secondary TW samples (Table 2). Out of seven positive samples detected by the real-time PCR method, only three samples (3 of 7, 42.85%) were amplified by conventional PCR for both VP7 and VP4 genes. To determine the G type and P type of the Iranian isolates, suitable PCR products were sent for sequencing. The successfully sequenced samples were aligned and compared with relevant reference sequences retrieved from GenBank.
The results of real-time PCR and conventional PCR for detecting rotaviruses in secondary TW samples
Sample period . | Real-time PCR results . | Conventional PCR results . | |
---|---|---|---|
2017 | September | Negative | Negative |
October | Negative | Negative | |
October | Negative | Negative | |
November | Positive | Negative | |
December | Positive | Negative | |
2018 | January | Negative | Negative |
February | Positive | Positive | |
March | Positive | Positive | |
April | Positive | Negative | |
May | Negative | Negative | |
June | Positive | Positive | |
July | Negative | Negative | |
August | Positive | Negative | |
Total (% positive) | 53.8 | 23.07 |
Sample period . | Real-time PCR results . | Conventional PCR results . | |
---|---|---|---|
2017 | September | Negative | Negative |
October | Negative | Negative | |
October | Negative | Negative | |
November | Positive | Negative | |
December | Positive | Negative | |
2018 | January | Negative | Negative |
February | Positive | Positive | |
March | Positive | Positive | |
April | Positive | Negative | |
May | Negative | Negative | |
June | Positive | Positive | |
July | Negative | Negative | |
August | Positive | Negative | |
Total (% positive) | 53.8 | 23.07 |
One isolate sampled in June (code IR-RIGLD-SHTN-VP7-WRTGA-0303) was classified as G9P[8], another isolate collected in March (code IR-RIGLD-SHTN-VP7-WRTGA-1208) belonged to G4P[8] and finally, the collected sample in February (code IR-RIGLD-SHTN-VP7-WRTGA-1118) was associated with G9P[4] genotype (Figures 1 and 2).
Phylogenetic tree of partial nucleotide sequences of the genes coding for viral structural proteins VP7 (564 nucleotides, nt 216-780) of the Iranian rotavirus isolated from waste water (the studied isolates were indicated by black triangles); the Tamura 3-parameter model with discrete Gamma distribution and invariant sites was used followed by the maximum-likelihood method. Bootstrap values (with 1,000 repetitions) lower than 70% have not been shown. The scale bars on the bottom left corners illustrate substitutions per nucleotide site.
Phylogenetic tree of partial nucleotide sequences of the genes coding for viral structural proteins VP7 (564 nucleotides, nt 216-780) of the Iranian rotavirus isolated from waste water (the studied isolates were indicated by black triangles); the Tamura 3-parameter model with discrete Gamma distribution and invariant sites was used followed by the maximum-likelihood method. Bootstrap values (with 1,000 repetitions) lower than 70% have not been shown. The scale bars on the bottom left corners illustrate substitutions per nucleotide site.
Phylogenetic tree of partial nucleotide sequences of the genes coding for viral structural proteins VP4 (534 nucleotides, nt 187-721) of the Iranian rotavirus isolated from waste water (the studied isolates were indicated by black triangles); the Tamura 3-parameter model using a discrete Gamma distribution was used followed by the maximum-likelihood method. Bootstrap values (with 1,000 repetitions) lower than 70% have not been shown. The scale bars on the bottom left corners display substitutions per nucleotide site.
Phylogenetic tree of partial nucleotide sequences of the genes coding for viral structural proteins VP4 (534 nucleotides, nt 187-721) of the Iranian rotavirus isolated from waste water (the studied isolates were indicated by black triangles); the Tamura 3-parameter model using a discrete Gamma distribution was used followed by the maximum-likelihood method. Bootstrap values (with 1,000 repetitions) lower than 70% have not been shown. The scale bars on the bottom left corners display substitutions per nucleotide site.
DISCUSSION
Rotavirus is a member of the enteric viruses, and as with other enteric viruses, it is remarkably resistant to wastewater treatment processes (Petrinca et al. 2009). In many countries, rotaviruses are responsible for most viral waterborne gastroenteritis outbreaks (Redwan & Attar 2012). Since water and wastewater systems are important avenues for the transmission of human waterborne diseases, monitoring viruses in urban TW could be helpful to determine the potential risk of viruses circulating throughout the community (Ruggeri et al. 2015).
Indeed, in order to monitor a low number of viruses in water resources, reliable, efficacious, and sensitive methods are needed. In the present study, the frequency of rotaviruses in secondary effluent samples was evaluated for 1 year. In this study, rotavirus was detected in 53.8% of samples using real-time PCR and in 23.07% of samples using conventional PCR. The higher rate of positivity with real-time in comparison to conventional PCR confirms the higher sensitivity of real-time PCR for molecular detection of rotaviruses. According to the sequencing and phylogenetic analyses in the present study, three genotypes of rotavirus including G9P[4] (collected in February), G4P[8] (collected in March), and G9P[8] (collected in June) were identified. In line with the results of our study, Betancourt et al. in Venezuela detected rotavirus in 66.7% of sewage samples and in 83% of sewage-polluted river samples and reported two P types including P[4] and P[8] and two G types including G1 and G9. The predominant genotypes in their study were G1P[4] followed by G1P[8] (Rodríguez-Díaz et al. 2009); the presence of G9, P[4], and P[8] types of rotavirus was in agreement with the present study. The discrepancies among the genotypes detected in the present study and in the Betancourt study can be due to the differences in their geographic location and detection methods.
Nezar and Ruba in Saudi Arabia employed the same method and detected rotavirus in 65% of TW samples (Redwan & Attar 2012). In 2006, Myrmel et al. determined the presence of rotavirus in inlet and outlet samples of three WWTPs in Norway. In total, rotaviruses were detected in 64.3% of inlet and in 53.3% of the outlet samples using nested PCR. Note that despite the differences in the concentration methods and sample sizes, the results of this study are similar to ours (Myrmel et al. 2006). A similar prevalence of the virus was reported by Arraj et al. in France who identified rotaviruses in 44.8% of treated effluent samples and in 37.9% of raw effluent samples using reverse transcription (RT)-PCR (Arraj et al. 2008).
A higher incidence of rotavirus was detected in China in 100% of influent and effluent as well as 90% reclaimed effluents of samples (He et al. 2008). It seems that the higher rates of rotavirus in some studies compared with the present study could be attributed to the concentration method and virus recovery.
Interestingly, the only report of rotaviruses in wastewater in Iran was conducted by Kargar et al., who detected rotaviruses in 26.67% of effluent samples. In that study, they did not employ a standard concentration method to recover rotaviruses from the samples. Note that the reported genotypes in the mentioned study were G1 and G4 (Kargar et al. 2013). The prevalence rate of rotaviruses in the current study together with different incidence rates of this virus among the other reports (Smith & Gerba 1982; Bosch et al. 1988; Mehnert et al. 1997) most probably arise from different volumes of samples, concentration method, sensitivity virus recovery, detection methods, and the geographic area in which the study was conducted.
Since the prevalence of virus circulating in the wastewater samples of a region can reflect the frequency of the virus in a community, the presence of viruses in wastewater could be an appropriate method for monitoring the virus across a community (Carter 2005). In a study conducted in Iran, rotavirus was identified at a high rate of 59% among children's stool samples with the identified genotypes being classified as G4P[8], G2[P4], P[8] with G nontypeable, G1[P8], G4 with P nontypeable. These results confirm the circulation of G4 and P[8] among community and urban wastewater of Tehran city (Eesteghamati et al. 2009).
The epidemiological studies among children in Iran suggested a high prevalence of rotavirus, while some genotypes were similar to ours. In 2016, Makvandi et al. detected rotavirus in 32% of children's stool specimens in Ahvaz/Iran. The results of G/P typing revealed that G9P[8] and G2P[4] were predominant followed by G1P[8], G12P[8], G4P[8], G2G9P[4], G9P[4] P[8], G3P[8], G9P[4], G2P[8], and G9P [nontypeable] (Azaran et al. 2018). In another study conducted by Kargar et al. in 2012, 46.0% of children's stool samples were positive for rotavirus in Jahrom city in which G1, G2, G3, G4, and G9 were the G types. In another study by the same researcher in Shiraz city, 34.78% of children's stool samples were positive for rotavirus. Also, the G typing analysis implied that the most common genotypes were mixed genotypes, nontypeable genotypes followed by G4. Notably, although Jahrom and Shiraz are located in same area in the south of Iran, G9 was not found in samples from Shiraz (Kargar & Akbarizadeh 2012; Kargar et al. 2012).
CONCLUSIONS
The present study revealed that rotavirus may remain detectable in wastewater treatment processes. The high prevalence of rotaviruses in the secondary effluent samples highlights the public health concern of these viruses. Indeed, regarding our results, it seems that the same genotypes are circulating across the community and environment. Notably, in spite of the water crisis as well as fertilizing feature of TW, these types of resources are usually used for irrigation or recreational purposes, and thus, a healthy TW is an interesting source. Further, regarding the importance of transmission of rotaviruses through water, periodical monitoring of wastewater treatment processes is required to control these waterborne viruses together with other microbial indicators. Taken together with population-based studies in Iran, it appears that G9P[8], G4P[8], and G9P[4] genotypes of rotaviruses might be suggested for national vaccination programs.
ACKNOWLEDGEMENTS
The present study was supported by the Research Institute for Gastroenterology and Liver Diseases (RIGLD), Shahid Beheshti University of Medical Sciences, Tehran, Iran (Grant No. 1022). The authors would like to thank Mrs R Rostami and Dr E Jalilzadeh, Department of Water and Wastewater Quality Control Laboratory, Water and Wastewater Company, Tehran, Iran and all of the RIGLD staffs, especially Mrs Sh Kazemian.
CONFLICT OF INTEREST
The authors declare that there is not any conflict of interest.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.