Low occurrence of Hepatitis A virus in water samples from an urban area of Southern Brazil

  • Fernanda Gil de Souza Universidade Feevale
  • Francini Pereira da Silva Universidade Feevale
  • Rodrigo Staggemeier Universidade Feevale
  • Caroline Rigotto Universidade Feevale
  • Fernando Rosado Spilki Universidade Feevale
Keywords: HAV, Stream, ICC-PCR, RT-PCR, Hepatitis, Nested-RT-PCR, Enteric virus


Hepatitis A virus (HAV), a member of Picornaviridae family, is the main causative agent of acute viral hepatitis in the world, mainly in developing countries. HAV may be present in contaminated water and food and its presence is often associated to a lesser extent with socioeconomic factors and environmental quality. The main goals in the present study were to standardize a cell culture combined to a polymerase chain reaction protocol for the detection and quantification of viral viability and analyze whether the virus could be found in water samples collected in four urban streams of Sinos River watershed. Virus recovery was assayed from known virus concentrations measured in experimentally contaminated raw and ultrapure water (MilliQ®). Recovery rates ranged from 270% in raw water to 15,000% in ultrapure water. In a second step, a qPCR coupled to a previous passage in cells, demonstrated more analytical sensitivity when compared to samples assayed without a previous passage in cell cultures. HAV genome was detected in only 1 of 84 samples analyzed, pointing to a very low occurrence of HAV in water samples in the studied region. These findings are remarkable, since no more than 5% of the domestic sewage in this area is treated pointing to a low occurrence of HAV in the population living nearby during the study period.


Download data is not yet available.
How to Cite
Souza, F., Silva, F., Staggemeier, R., Rigotto, C., & Spilki, F. (2018). Low occurrence of Hepatitis A virus in water samples from an urban area of Southern Brazil. Revista Do Instituto De Medicina Tropical De São Paulo, 60, e69. Retrieved from https://www.revistas.usp.br/rimtsp/article/view/151628
Original Articles


Hepatitis A virus (HAV) belongs to the genus Hepatovirus , within the Picornaviridae family, order Picornavirales. The genome is composed of a positive single-stranded RNA, with approximately 7.5 Kb of length 1 . Two noncoding regions are found in the genome flanking regions nominated P1 (comprising the ORF encoding the structural proteins VP1, VP2, VP3 and VP4) and P2 and P3, which encodes non-structural proteins related to viral replication 1 . HAV capsids are icosahedral and the viral particle is non-enveloped, thus being resistant under environmental conditions, although the virus is transitionally enveloped during traffic from the liver to the intestinal epithelium 1,2 .Three of the 6 HAV genotypes are associated with infections in humans (I-III) 3 .

HAV particles are highly stable in the environment and may be detected for long periods in various matrices such as surface and groundwater, sea waters and sediments 4 . Through the years, HAV outbreaks have been reported with increased frequency in large urban areas 1 . HAV outbreaks are related mostly to low socioeconomic levels and low environmental sanitation 1 . The spread of the virus through contaminated water and food increased the number of individuals who became not naturally immunized in so called developed regions. Actually, the absence of children vaccination in many places poses a challenge for the control of HAV infections nowadays 5 . As water and food brought up the nature of the infection, it is a priori advisable that studies on the environmental monitoring of HAV should be done, aiming to fulfill the knowledge about the epidemiology, whenever it is possible.

In Brazil, hepatitis A is a great matter of concern to public health, especially in the Northern part of the country. Children’s immunization has been routinely implemented. Since 2014, the vaccine was included in the governmental calendar, increasing the access to immunization. In Rio Grande do Sul, the Southernmost Brazilian State, HAV cases decayed in the last 10 years. In 2014, there was a very low prevalence of the disease, with no more than 100 cases reported 6 . Many studies in other parts of the country showed a decrease of less than half that would be related to the viral prevalence in environmental samples 7,8 .

Molecular-based methods are useful for the detection of HAV genomes in water samples 9 . Adoption of protocols integrating molecular detection with cell culture, namely ICC-PCR, may allow inferring about the viral viability, providing information on the infectivity and health risks 10 . Additional ICC-PCR benefits may also include removal of inhibitors and increasing of viral particles counting through a cell passage, thus enhancing the analytical sensitivity of the technique when compared to molecular or cell-based detections alone 11 .

The main goal of the present study was to standardize and apply molecular techniques coupled to a cell culture, the polymerase chain reaction providing quantification and evaluation of HAV viral viability in water samples from streams in urban areas of the Sinos River watershed. This watershed is highly impacted by discharges of untreated domestic sewer in water bodies, and has been used as the main source of drinking water for nearly 1.500.000 inhabitants, and the waters were previously found to be contaminated by other enteric viruses 12,13 .



From 2012 to 2014, 0.5 L of water samples were collected, aseptically bimonthly, in four streams crossing 4 cities of Sinos River watershed: Estancia Velha/Portao (cities of Estancia Velha and Portao, 73,494 habitants), Schmidt (Campo Bom city, 60,074 habitants), Pampa and Luiz Rau (both located in Novo Hamburgo city, 238,940 habitants, Figure 1 ); two samples per stream, near the source and the mouth, totaling 84 samples. Water samples were transported to the laboratory in freezer blocks. Afterwards, samples were kept at 4 °C until processing. As mentioned before, this period was characterized by a very low prevalence of HAV-induced hepatitis in this region according to official reports.

Description of the counties Estancia Velha, Novo Hamburgo and Campo Bom.

Figure 1: Description of the counties Estancia Velha, Novo Hamburgo and Campo Bom.

Viral concentration

Samples were concentrated by an adapted adsorption-elution method 14 described before 15 . Briefly, before filtration, pH was measured and adjusted to 4.5-5.5. Afterwards, 12.5 mL of MgCl2 1M were added to samples. Five hundred millilitres of samples were filtered with the aid of a vacuum pump (760 mm Hg), using a 0.45 µm nitrocellulose negatively charged membrane (HA 47 mm, Millipore), for the adsorption of viral particles present in the sample; 87.5 mL of 0.5 mM H2SO4 was passed through the membrane to rinse out the cation. Then, 2.5 mL of solution 1 mM of NaOH were added to the membrane for elution of viral particles. In 2.5 mL of the final eluate, 12.5 μL of 50 mM H2SO4 solution and 12.5 μL TE 100X buffer were added for neutralization, then samples were aliquoted and stored at -80 °C until further processing.

Viral nucleic acids extraction

A volume of 400 uL of the final viral eluate was submitted to nucleic acid extraction using RTP® DNA/RNA Virus Mini Kit (Invitek®, Germany), following the manufacturer’s instructions. The final volume of 60 uL containing putatively viral RNAs was stored at -80 °C.

Recovery of the viral concentration method

To assay the viral recovery rates of the filtration-elution method described, aliquots of 500 mL of ultrapure RNAse/DNAse free water (Direct 8®, MilliQ®) and of raw water of Pampa stream, previously tested negative for HAV, were artificially inoculated by spiking 1 mL of a viral suspension of 5.2 x 107 HAV (HM 175) CG/ 5 -µL. Before the viral inoculation, 1 mL of each sample vial was used as the negative control. Borosilicate bottles containing 500 mL of water were left closed at room temperature (21 oC) for 30 min. Afterwards, 1 mL of each sample was aliquoted before the viral concentration by the method described above. Final eluates of 2.5 mL were stored at -80 ºC until further analysis. Viral genome extraction was performed directly from samples or after inoculating them into cell monolayers (ICC-RT-qPCR) as described above and the following types of samples were analyzed: 1) water before virus inoculation as the negative control (NC); 2) HAV inoculated in water before concentration (BC); 3) virus inoculated in water after concentration, (AC); 4) virus that has a passageway by cell culture and analysis by ICC-RT-qPCR and Positive Control (PC).

Recovery rate calculation

Recovery rates were calculated using the values observed in the qPCR before the concentration process, defining as 100% of recovery, not evaluating losses in the processes of extraction and cDNA synthesis. Values obtained in 5 µL after the process of concentration were divided by the values obtained before the process and multiplied by 100, X = b/a × 100, where X = recovery rate (%), a = quantification result for 5 μL by qPCR after concentration, and b = result for 5 μL by qPCR before concentration.

HAV genome detection

Complementary DNA synthesis (cDNA) using the RNA extracted from samples and controls was performed using random primers of the High-Capacity cDNA Reverse Transcription kit™ (AppliedBiosystems®, USA), following the manufacturer’s instructions.

For quantitative PCR (qPCR), primers and a TaqMan® HAV-specific probe were used aiming at amplifying the conserved noncoding region 5’-UTR of the HAV genome, described before 16 (sense 5’-GGTAGGCTACGGGTGAAAC-3’; reverse 5’-AACAACTCACCAATATCCGC-3’; probe FAM5’-CTTAGGCTAATACTTCTATGAAGAGATGC-3’BHQ). All qPCR assays were performed using TaqMan® Universal PCR Master Mix (Applied Biosystems®). For a final reaction volume of 20 μL, the reagents mixture contained 100 nM of each primer and the Taqman® probe, and 5 μL of cDNA. Each run was composed of 1 cycle of 60 min, denaturation at 95 ºC for 15 min, followed by 40 cycles of 95 ºC for 10 s, 55 ºC for 20 s and 72 ºC for 15 s. All reactions performed in 48 wells microplates (MicroAmp Applied Biosystems) in a StepOne real-time PCR system (AppliedBiosystems®, USA).

Positive (plasmids) and negative (RNAse/DNAse free water) controls were included, and standard curves were made using decimal serial dilutions from 10-1 to 10-10 genome copies of a plasmid containing the amplicon, kindly provided by Dr. Celia Barardi (Federal University of Santa Catarina, Brazil). Following the addition of 15 μL of the reaction mixture and 5 μL of cDNA, the microplate was sealed. All the samples and controls were tested in duplicate, to determine the efficiency and the limit of analytical sensitivity, values were calculated from a standard curve with dilution series of controls containing known concentrations of the target. Typical values obtained for efficiency were 103% and R2 =0.99.

Viral viability assay by ICC-RT-qPCR

Viral viability evaluation was performed using RT-PCR real time assay coupled to cell culture (ICC-RT-qPCR - integrated cell culture qPCR ). This assay aims to detect possible infectious viral particles in environmental or biological samples. For the ICC-RT-qPCR, FRKH-4 cells ( Rhesus Monkey Kidney Fibroblasts) were cultivated in 24-well plates (2.5x105 cells/well) at 37 ºC and 5% CO2 atmosphere until cellular confluence. Afterwards, 500 μL of the following were added to each well: concentrated water samples diluted 1:2 (non-cytotoxic dilution) and sterilized in 0.22 µm membranes positive control (HAV HM175 5.2 x 107 CG/ 5 µL) and negative control (Eagle’s minimum essential medium – E-MEM). After 1 h of viral adsorption and uniform shaking, every 15 min, the initial inoculum was removed. In each plate cavity, 1 mL of E-MEM containing 1% of antibiotics (penicillin and streptomycin) and 2% of fetal bovine serum (FBS, Cultilab®) was added. Cell monolayers were incubated for 72 h at 37 ºC in atmosphere with 5% of CO2. Afterwards, the supernatant of each cavity was removed and the lysis buffer was added directly to the monolayer cell. Cell lysates were transferred to 1.5 mL microtubes and the nucleic acids extraction, cDNA synthesis and qPCRs were performed previously described protocols.


The analytical sensitivity of qPCR fell under 100 genomic copies in serial dilution curves performed with plasmid dilutions. During viral recovery assays using HAV-spiked in ultrapure water, the minimum viral detection limit was 3 genomic copies, when PCR was coupled to cell culture ( Table 1 ).

Table 1:
Analytical sensitivity limit of ICC-qPCR reaction to Hepatitis A virus artificially inoculated in ultrapure water before and after the viral concentration process by the adsorption-elution method.
Sample dilutions Quantification before concentration a Quantification after concentration a
Non-diluted 1.1x103 1.3x105
10-1 1.1x102 1.6x105
10-2 10 1.8x104
10-3 3 1.5x103
10-4 3 3.1x102

a values in genomic copies/5 µL.

The viral concentration protocol was efficient, as inferred from the results of recovery assays, because quantification was 150 X higher in ultrapure concentrated water with 6.3 x 104 GC/ 5 µL (genomic copies/ 5 µL) than in non-concentrated water samples with 4.2 x 102 GC /5 µL ( Table 2 ). In assays performed with ultrapure water samples spiked with HAV after the described cell culture ( Table 1 ), it was possible to observe an increase of approximately 1 log in the quantification when compared to samples assayed without a cell culture step ( Table 2 ). Before concentration, in samples assayed without a cell culture step, the value was 4.2 x 102 GC/ 5 µL and after cell culture, the value was 1.1 x 103 GC/ 5 µL. After concentration, in samples assayed without a cell culture step, the value was 6.3 x 104 GC/ 5 µL and after cell culture, the value was 1.3 x 105. From the environmental samples analyzed, only one was HAV-positive (1.5 x 104 cg/L) coming from point 4 of the Pampa stream, which crosses the larger city included in this study, namely Novo Hamburgo (July/2013).

Table 2:
Recovery rates of Hepatitis A (HAV) virus before and after the adsorption-elution concentration method in ultrapure and raw experimentally contaminated water.
Samples Quantification (CG/5uL) Recovery rate (%)
Ultrapure water before inoculation Neg Neg
Ultrapure water + HAV before concentration 4.2x102 100
Ultrapure water + HAV after concentration 6.3x104 15000
Raw water before inoculation Neg Neg
Raw water + HAV before concentration 1.04x103 100
Raw water + HAV after concentration 2.7x103 270


ICC-PCR assays have been used for detecting several other enteric viruses in environmental samples 10,17,18 . The results have shown that the qPCR using TaqMan® was adapted to the integrated use proposed in the cell culture protocol, increases the technique sensitivity possibly by viral growth that may enhance the number of genome copies and decrease amplification inhibitors.

The recovery percentage in raw water was lower than in ultrapure water as expected, probably due to the presence of inhibitors. An increase in the quantification of raw samples was also observed. In other studies, the best recovery was found in distilled water and treated wastewater effluents (100%), and the lowest recovery (10%) in seawater 19 . In addition, like in this study, no detection of HAV was found in coastal water 20 and for both, Norovirus (NoV) and Human astrovirus (HAstV), the recovery percentage from mineral and river water samples was between 18% and 64% 21 .

These results show that the use of integrated cell culture to ensure the detection of HAV in water may favor analytical sensitivity, as it was observed in the past for other enteric viruses 22 . Another advantage is the inference of measures to control infectivity and consequently the health risk posed by the presence of HAV particles in water 18 .

The Pampa stream is highly impacted and urbanized, receiving industrial garbage discharges and untreated sewer 23 . Other studies reported poor water quality conditions in this particular stream by monitoring of physical and chemical parameters 12 . Although the detection of a solely positive sample for HAV shows that no matter this stream has been hugely contaminated by domestic sewage, the detection of this virus under environmental matrices in often sporadic. No other samples were positive for HAV gene sequences, even after cell passages.

In another study conducted in the North region of Brazil, 23% of HAV positive samples were detected by nested -RT-PCR and 92% using RT-qPCR, but it may be speculated that these samples were collected during or after HAV outbreaks 7 . The same may have happened in another survey on recreational seawater in Florianopolis, in which 46% of the samples were found to be contaminated by HAV 24 . On the other hand, many studies in Brazil showed low occurrence or absence of HAV in water, as has also been reported for the area of the study, regarding acute viral hepatitis of official government reports throughout the sampling period. In 2009, HAV was detected in 8.3% of freshwater samples from Florianopolis 19 and no positive results for HAV were found in 108 samples analyzed in Rio de Janeiro 25 . Other studies have also shown a low incidence of HAV in environmental matrices, ranging from 20% to 32% positive samples, respectively, reinforcing the sporadic occurrence of HAV in surface waters in Brazil 26,27 .

However, since HAV is an endemic condition in Brazil, with the improvement of sanitation in recent decades, a decrease in the seroprevalence has been observed in several regions of the country. Unlike low endemic countries, the vaccine is only recommended depending on epidemiological studies of the region and risk groups involved, especially for children under five years old 28 .


Based on viral recovery assays and qPCR efficiency evaluation performed in this study and in the environmental samples analysis, we can conclude that there was a low virus occurrence in superficial waters in the region studied.



This work was supported by Feevale University, the Coordination for the Improvement of Higher Level Personnel (CAPES), and the National Council for Scientific and Technological Development (CNPq).


  1. , , (). Hepatitis A virus detection in food: current and future prospects. Lett Appl Microbiol 45, 1-5.
  2. , , , (). Naked viruses that aren’t always naked: quasi-enveloped agents of acute hepatitis. Annu Rev Microbiol 1, 539-560.
  3. , , , , , (). Evidence of hepatitis A virus person-to-person transmission in household outbreaks. PLoS One 9
  4. , , , , , (). Monitoring viruses in environmental samples. IJESER 3, 62-79.
  5. (). Hepatitis A virus infection: progress made, more work to be done. J Pediatr 87, 185-186.
  6. , , , , (). . Dia mundial de luta contra as hepatites virais: 2018. https://cevs.rs.gov.br/upload/arquivos/201807/25114428-panorama-das-hepatites-virais-b-e-c-no-rs.pdf
  7. , , , (). Detection of enteric viruses in sewage sludge and treated wastewater effluent. Water Sci Technol 61, 537-544.
  8. , , , , (). Spatial distribution of enteric viruses and somatic coli phages in a Lagoon used as drinking water source and recreation in Southern Brazil. Int J Hyg Environ Health 219, 617-625.
  9. , , , , , (). Hepatitis A virus in environmental water samples from the Amazon Basin. Water Res 41, 1169-1176.
  10. , , , , , (). Assessment of adenovirus, hepatitis A virus and rotavirus presence in environmental samples in Florianopolis, South Brazil. J Appl Microbiol 109, 1979-1987.
  11. , (). Development of an integrated cell culture: real-time RT-PCR assay for detection of reovirus in biosolids. J Virol Methods 139, 195-202.
  12. , , , , , (). Surface water quality in the Sinos River basin, in Southern Brazil: tracking microbiological contamination and correlation with physicochemical parameters. Environ Sci Pollut Res Int 22, 9899-9911.
  13. , , , , (). Quantitative vs. conventional PCR for detection of human adenoviruses in water and sediment samples. Rev Inst Med Trop Sao Paulo 57, 299-303.
  14. , , (). Development of a Virus Concentration method and its application to detection of enterovirus and norwalk virus from coastal seawater. Appl Environ Microbiol 68, 1033-1039.
  15. , , , , , (). First description of Adenovirus, Enterovirus, Rotavirus and Torque teno virus in water samples collected from the Arroio Dilúvio, Porto Alegre, Brazil. Braz J Biol 72, 323-329.
  16. , , , (). Development and evaluation of a broadly reactive TaqMan assay for rapid detection of hepatitis A virus. Appl Environ Microbiol 71, 3359-3363.
  17. , , , (). Detection of adenoviruses in shellfish by means of conventional-PCR, nested-PCR, and integrated cell culture PCR (ICC/PCR). Water Res 39, 297-304.
  18. , , , (). Detection of infectious rotavirus in naturally contaminated source waters for drinking water production. J Appl Microbiol 107, 97-105.
  19. , , , , (). Evaluation of HA negatively charged membranes in the recovery of human adenoviruses and hepatitis A virus in different water matrices. Mem Inst Oswaldo Cruz 104, 970-974.
  20. , , , , (). Evaluation of methods used to concentrate and detect hepatitis A virus in water samples. J Virol Methods 137, 169-176.
  21. , , , , , (). Evaluation of an adsorption-elution method for detection of astrovirus and norovirus in environmental waters. J Virol Methods 156, 73-76.
  22. , , (). Evaluation of methodology for detection of human adenoviruses in wastewater, drinking water, stream water and recreational waters. J Appl Microbiol 108, 800-809.
  23. , (). Monitoramento de ph, temperatura, OD, DBO e condições microbiológicas das águas do arroio Pampa em Novo Hamburgo (RS). Est Tecnol 5, 227-244.
  24. , , , , , (). Microbiological and physicochemical analysis of the coastal waters of southern Brazil. Mar Pollut Bull 64, 40-48.
  25. , , , , , (). Assessment of water quality in a border region between the Atlantic forest and an urbanized area in Rio de Janeiro, Brazil. Food Environ Virol 6, 110-115.
  26. , , , , , (). Nested multiplex PCR assay for detection of human enteric viruses in shellfish and sewage. J Virol Methods 125, 111-118.
  27. , , , , , (). Molecular detection of hepatitis A virus in urban sewage in Rio de Janeiro, Brazil. Lett Appl Microbiol 45, 168-173.
  28. , , , , , (). Multilevel analysis of hepatitis A infection in children and adolescents: a household survey in the Northeast and Central-west regions of Brazil. Int J Epidemiol 37, 852-861.