Science of the Total Environment 530–531 (2015) 314–322
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Behaviour and recovery of human adenovirus from tropical sediment under simulated conditions Hugo Delleon Silva a,b, Marco Aurélio Pessoa-de-Souza c, Gislaine Fongaro d, Carlos E. Anunciação e, Elisângela de P. Silveira-Lacerda f, Célia Regina Monte Barardi d, Marco Tulio Antonio Garcia-Zapata a,⁎ a
Núcleo de Pesquisas em Agente Emergentes e Re-emegentes, Instituto de Patologia e Saúde Pública, Universidade Federal de Goiás, Brazil Instituto Brasil de Ciência e Tecnologia, Anápolis, Brazil Departamento de Zootecnia, Pontifícia Universidade Católica de Goiás, Campus II, Goiânia, Brazil d Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Universidade Federal de Santa Catarina, Brazil e Laboratório de Diagnóstico Genético e Molecular, Departamento de Bioquímica e Biologia Molecular, Instituto de Ciências Biológicas II, Universidade Federal de Goiás, Brazil f Laboratório de Genética Molecular e Citogenética, Departamento de Biologia Geral, Instituto de Ciências Biológicas I, Universidade Federal de Goiás, Brazil b c
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Tropical solids decreased genome copy numbers and viral infectivity. • Organic matter did not influence genome copy numbers but decreased viral infectivity of HAdV-5. • Acidic pH hinders viral inactivation.
a r t i c l e
i n f o
Article history: Received 23 February 2015 Received in revised form 12 May 2015 Accepted 18 May 2015 Available online 3 June 2015 Keywords: Human adenovirus Infectivity Organic matter pH
a b s t r a c t This study assessed the contributions of pH and organic matter (OM) on the recovery of infectious human adenovirus 5 (HAdV-5) and genome copies (GCs) in waters that were artificially contaminated with tropical soil. The use of a mathematical equation was proposed based on the flocculation index of clay to assess the recovery of total GCs in these controlled assays. The results suggest that solids in the water reduced the viral genome copy loads per millilitre (GC·mL−1) and viral infectivity. OM did not influence the GC·mL−1 recovery rate (p N 0.05) but led to a 99% (2 log10) reduction in plaque-forming unit counts per millilitre (PFU/mL), which indicates that infectivity and gene integrity were non-related parameters. Our findings also suggest that acidic pH levels hinder viral inactivation and that clay is the main factor responsible for the interactions of HAdV-5 with soil. These findings may be useful for future eco-epidemiological investigations and studies of viral inactivation or even as parameters for future research into water quality analysis and water treatment. © 2015 Elsevier B.V. All rights reserved.
⁎ Corresponding author at: Núcleo de Pesquisas em Agente Emergentes e Re-Emegentes, Instituto de Patologia e Saúde Pública, Universidade Federal de Goiás, 74643-970 Goiânia, Goiás State, Brazil. E-mail address: [email protected] (M.T.A. Garcia-Zapata).
http://dx.doi.org/10.1016/j.scitotenv.2015.05.075 0048-9697/© 2015 Elsevier B.V. All rights reserved.
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
1. Introduction Human adenoviruses (HAdVs) belong to the family Adenoviridae, genus Mastadenovirus, which contains 57 different serotypes distributed across seven species (A–G) (ICTV, 2013). They are icosahedral, nonenveloped viruses containing a double-stranded linear DNA genome (Enriquez, 2002). These viruses are of substantial public health importance (Silva et al., 2010), as they are excreted in faeces, urine, and respiratory droplets (Metcalf et al., 1995; Jiang et al., 2001), and can cause a series of disease states in infected individuals by the respiratory and faecal–oral routes. Examples include upper respiratory tract infections (pharyngitis and tonsillitis), lower respiratory tract infections (bronchiolitis and pneumonia), conjunctivitis, cystitis, and gastroenteritis (Mena and Gerba, 2008). HAdVs are stable in the environment and resistant to water treatment methods (Thompson et al., 2003), particularly to UV irradiation (Liden et al., 2007). Furthermore, they are ubiquitous in the environment yearround (Katayama et al., 2008). These pathogens are prevalent in both treated and untreated water (Jiang et al., 2001; Silva et al., 2010; Fongaro et al., 2013) and are often detected in higher concentrations than other enteric viruses (Wong et al., 2010). Thus, HAdVs are indicated for use as viral biomarkers of environmental water and drinking water quality (Wyn-Jones et al., 2011; Silva et al., 2011). Detection of HAdVs in water destined for human consumption can be accomplished either by molecular techniques alone (Silva et al., 2011) or in combination of these techniques with cell cultures (Wyn-Jones et al., 2011; Garcia et al., 2012) that allows access infectivity (ability of the virus to replicate in permissive cells) (Herzog et al., 2008). Despite the great sensitivity of PCR, the main limitation is the lack of the correlation between the detected viral genome and viral infectivity, which limits conclusions about the significance for public health (Hamza et al., 2011). Despite advancements in the detection of HAdVs in different water sources, information is lacking on the relationship of these viruses with the solids, sediments, or suspended solids in the waters in which they are present.
315
Viral adsorption to solids is known to be an extremely complex process (Schijven and Hassanizadeh, 2000), and variations can be observed even among different serotypes of the same virus (Singh et al., 1986). Studies that report viral adsorption behaviour to solids usually employ phages as models (Schijven and Hassanizadeh, 2000). These studies are valid, but only as predictive models, and cannot infer the true permanence of infectious viruses, such as HAdVs. This is due to variations in the isoelectric point (pI), virion size, hydrophobicity, and capsid proteins (Schijven and Hassanizadeh, 2000), among other factors that have not yet been discovered. HAdVs are more prevalent than norovirus, enterovirus, hepatitis A virus, and human poliovirus in biosolids (Wong et al., 2010). Fong et al. (2010) found that 100% of sewage and effluent samples were infectious with HAdV. These findings suggest that matrix solids may play a crucial role in the gene integrity and infectivity of HAdVs. The interactions of HAdVs in soil solutions were recently analysed in two studies. In the first, Wong et al. (2012) investigated the influence of different concentrations of inorganic ions on the aggregation and deposition behaviours of HAdVs in sandy soil. In the second, Wong et al. (2013) evaluated the role of soil organic carbon (SOC) and solution-phase dissolved organic carbon (DOC) on sorption capacity and reversibility of organic carbon on adsorption of HAdVs. These two studies assessed adsorption behaviour using real-time quantitative PCR (qPCR) but did not evaluate the influence of solids and total organic matter on HAdV infectivity. qPCR is a highly sensitive technique, capable of detecting a small number of microorganisms (Botes et al., 2013), but when used alone, it is unable to distinguish between infectious and non-infectious viral particles. Furthermore, viral interactions with solid particles may lead to viral inactivation by loss of capsid integrity, with the consequent release of genetic material (Schijven and Hassanizadeh, 2000), which may be identified by qPCR either as an infectious or non-infectious particle. Therefore, the present study sought to (i) assess the recovery of infectious HAdV-5 and genome copies from simulated solutions containing tropical solids under controlled pH values in the presence or absence of organic matter; and (ii) define a mathematical equation to assess the recovery rate from clay soils under simulated conditions.
2. Material and methods The U.S. Environmental Protection Agency (USEPA, 2011) defines the acceptable level of total dissolved solids (TDS) in drinking water as 500 and 1000 mg·L−1 and at pH ranges of 6.0–9.5 (Brazil, 2005) and 6.5–8.5 (USEPA, 2011). Thus, the experiments were conducted to simulate maximum levels of solids in water contaminated with infectious HAdV-5, using the reference pH values of 6.0 and 8.0.
2.1. Soil characterisation and preparation Gleysoil (hydromorphic soil) is a typical soil of riverbanks in tropical environments (Rosolen et al., 2014). This type of soil is associated with poor land management (use and occupation) and intense precipitation runoff from the landscape. In these conditions, organic matter and minerals from these soils can reach the rivers (FAO, 2006; Reatto et al., 1998) and can be found during the water treatment process. Thus, gleysoil samples were collected at a depth of up to 15 cm from a native palm swamp area in the municipality of Bela Vista de Goiás, the central portion of the State of Goiás, Brazil in the Cerrado ecoregion (17° 00′S 48° 47′W). This type of soil is typically found at river and lake borders and wetlands (Rosolen et al., 2014) and is the most common sediment carried inside rivers and lakes (FAO, 2006; Reatto et al., 1998). The samples were air-dried, passed through a 2-mm mesh sieve, and any excess plant debris were removed with the aid of tweezers and a magnifier to yield the fine earth fraction described by Gee and Bauder (1986). The soil was subdivided into two portions: (i) soil with organic matter (WOM) and (ii) soil without organic matter (OM consumed by H2O2 – LOM – less organic matter/OM-free). The latter was treated with hydrogen peroxide (Whittig and Allardice, 1986) and autoclaved at 121 °C and 0.105 MPa for 3 h, three times, with 24-h intervals between each treatment (Zhao et al., 2008), to intentionally remove the soil organic matter. To ensure maximal homogeneity, the soil samples were crushed and stored at room temperature. The results of physicochemical and instrumental analyses of WOM and LOM soil samples are shown in Table 1. The clay fraction was determined by X-ray diffraction (XRD). The silt, clay (iron-free fraction and saturated with K, Mg, Mg + glycerol, K + 350 °C, and K + 550 °C), and sand fractions were also separated. Preparation for XRD and XRD itself was performed as described by Whittig and Allardice (1986) and Resende et al. (2005). Fig. 1 shows the X-ray diffractogram obtained.
316
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
Table 1 Physicochemical and instrumental analyses of soil samples containing organic matter (WOM) and without organic matter (LOM). Assay
Unit
WOM
LOM
Organic matter (OM)a Al (aluminium)b Si (silicon)b K (potassium)b Ti (titanium)b Fe (iron)b Iron — Fedc Iron — Feod Sande Silte Claye Degree of flocculationf Point of zero charge (PZC)g Densityh
dag·dm−3 % (wt) % (wt) % (wt) % (wt) % (wt) g kg−1 g kg−1 dag·dm−3 dag·dm−3 dag.dm−3 %
3.07 42.06 47.72 0.69 2.54 11.99 131.1 4.0 36.79 24.2 39 27 5.41 1.60
ND 41.72 46.70 0.47 1.97 9.14 20.2 3.0 ND ND ND ND 4.20 ND
i i
a
Measured by loss on ignition (Ball, 1964). Scanning electron microscopy/energy-dispersive spectroscopy (semiquantitative data). c Iron content of the clay fraction by the oxalate method (Whittig and Allardice, 1986). d Iron content of the clay fraction by the dithionite method (Whittig and Allardice, 1986). e Dispersion and pipette method (Embrapa, 1997). f Ratio of natural clay to total clay (Vettori, 1969, Embrapa, 1997). g PZC — potentiometric titration method (Embrapa, 1997). h Particle density (Embrapa, 1997). i Dimensionless quantity. b
Fig. 1. Overlay of X-ray diffractograms of original soil (clay with OM) and OM-free soil (clay without MO), showing the presence of 1:1 (kaolinite) clay as well as 2:1 (vermiculite and ilite) clays. This composition provided evidence of the degree of soil weathering, particularly by the presence of kaolinite and gibbsite, which are characteristic of highly weathered soils.
2.2. Preparation of viral inoculum A549 cells were cultured in 75-cm2 cell culture flasks containing Dulbecco's Modified Eagle's Medium (DMEM — High Glucose 1×) supplemented with 10% foetal bovine serum (FBS) and 1% PSA (penicillin G, streptomycin, and amphotericin B at final concentrations of 100 U·mL−1, 100 μg·mL−1, and 0.25 μg·mL−1, respectively). The cells were infected with human adenovirus 5 (species C, serotype 5 — HAdV-5) and cultured in a 5% CO2 incubator at 37 °C until cytopathic effects were observed under an inverted microscope. The cell monolayers were frozen and thawed three times, and any cellular debris was removed by centrifuging at 5000 rpm at 4 °C for 10 min. The viral inoculum consisted of 1-mL aliquots of the resulting supernatant.
2.3. HAdV inoculation Fig. 2 provides a scheme of the experimental study design. The method consisted of separately adding 5, 25, and 50 mg of original soil and OM-free soil to 50-mL sterile polypropylene tubes and adding 40 mL of ultrapure water, followed by brief manual agitation, to obtain suspended and settled solids. The pH of sediment-containing water solutions was adjusted to 6.0 and 8.0 (a different pH level for each soil type and concentration) with 0.1 M HCl. A 1-mL aliquot of viral inoculum was then added and ultrapure water was added to a final volume of 50 mL, to obtain TDS concentrations of 100, 500, and 1000 mg·L−1 for each test assay (TA). One control assay (CA — without addition of soil), which consisted solely of the viral inoculum in a 50-mL final volume of ultrapure water, was performed for each pH level. These control tubes were processed identically to the test tubes. The same experiment performed with original soil samples was run simultaneously with LOM soil. All concentrations were run in duplicate with repeats in triplicate. Tubes containing the virus and sediment solution were agitated at 150 rpm for 1 h at a temperature of 24 °C (Schijven and Hassanizadeh, 2000; Gantzer et al., 1994). After agitation, the tubes were rested for 1 h at 24 °C for precipitation of settleable solids. Two clear phases were observed: (i) the supernatant and (ii) the sediment. Briefly, 200-μL aliquots of the supernatant were collected from each tube and stored at −80 °C until processing. The remaining supernatant was removed, leaving only the pellet representative of settleable material, to which 2 mL of ASL buffer from the QIAamp DNA Stool Mini Kit (QIAGEN) was immediately added, following the manufacturer's instructions for nucleic acid extraction. Nucleic acids were also extracted from the 200-μL supernatant aliquots, using the same kit. All extracted samples were immediately stored at −80 °C.
2.4. Quantitative real-time PCR (qPCR) DNA samples were diluted 1:10 to avoid inhibition, and qPCR was performed as described by Hernroth et al. (2002). These specific primers anneal with the conserved region of the hexon gene of the HAdVs. All amplifications were run in duplicate in a StepOnePlus™ Real-Time PCR System (Applied Biosystems), using as standard a pBR322 plasmid containing part of the adenovirus hexon gene, to generate a standard curve (R N 0.98). The results were analysed to measure genomic copy numbers of HAdV-5 (GC·mL−1) in water and in sediment, and the mean GC·mL−1 values were compared to the mean results obtained in each CA.
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
317
Fig. 2. HAdV-5 copy numbers and infectious HAdV-5 counts in simulated aqueous solutions of solids obtained from tropical soil samples, under controlled pH and organic matter conditions.
2.4.1. Mathematical equation to infer the total copy number The GC·mL−1 values were used to construct a series of mathematical equations to infer the total copy number recovered in relation to the concentration of clay, both in the suspension (to simulate suspended solids in drinking water) and in the precipitate (to simulate solids that may settle as water is transported through pipes or containers). For the test assay (TA) of each soil treatment, the total HAdV genome copy numbers in the supernatant (Nsup) and in the precipitate (Nprec) were calculated according to the following equations, respectively: Nsup ¼
. h ðClay% V cm3 Þ
100
Nprec ¼ ðV cm3 Þ−
hh
i Floc GC mL−1
V cm3 Clay%
. 100
i i Floc GC mL−1
ð1Þ
ð2Þ
where V, Clay% and Floc are, respectively, the volume of sediment (cm3) in the tube, the clay content (%) of the matrix, and its degree of flocculation. Volume is the direct relationship between mass (g) and density (g·cm−3), where 1 cm3 is equal to 1 mL. The clay content is determined by soil texture analysis, and the degree of flocculation is the difference between total clay and natural clay after dispersion (Table 1). The total genome copy number for the TA (Nta) was obtained by compiling Eqs. (1) and (2) and deriving Eq. (3) below: Nta ¼ Nsup þ Nprec :
ð3Þ
The overall recovery rate (Rr), expressed in the number of copies, was calculated by Eq. (4):
Rr ð % Þ ¼
ðNca −Nta Þ 100 Nca
ð4Þ
318
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
where Nca is the total number of GCs in the CA. Eq. (4) may be rewritten as follows (Eq. (5)): Rr ð % Þ ¼
. hh 8 ðClay% V cm3 Þ
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Behaviour and recovery of human adenovirus from tropical sediment under simulated conditions Hugo Delleon Silva a,b, Marco Aurélio Pessoa-de-Souza c, Gislaine Fongaro d, Carlos E. Anunciação e, Elisângela de P. Silveira-Lacerda f, Célia Regina Monte Barardi d, Marco Tulio Antonio Garcia-Zapata a,⁎ a
Núcleo de Pesquisas em Agente Emergentes e Re-emegentes, Instituto de Patologia e Saúde Pública, Universidade Federal de Goiás, Brazil Instituto Brasil de Ciência e Tecnologia, Anápolis, Brazil Departamento de Zootecnia, Pontifícia Universidade Católica de Goiás, Campus II, Goiânia, Brazil d Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Universidade Federal de Santa Catarina, Brazil e Laboratório de Diagnóstico Genético e Molecular, Departamento de Bioquímica e Biologia Molecular, Instituto de Ciências Biológicas II, Universidade Federal de Goiás, Brazil f Laboratório de Genética Molecular e Citogenética, Departamento de Biologia Geral, Instituto de Ciências Biológicas I, Universidade Federal de Goiás, Brazil b c
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Tropical solids decreased genome copy numbers and viral infectivity. • Organic matter did not influence genome copy numbers but decreased viral infectivity of HAdV-5. • Acidic pH hinders viral inactivation.
a r t i c l e
i n f o
Article history: Received 23 February 2015 Received in revised form 12 May 2015 Accepted 18 May 2015 Available online 3 June 2015 Keywords: Human adenovirus Infectivity Organic matter pH
a b s t r a c t This study assessed the contributions of pH and organic matter (OM) on the recovery of infectious human adenovirus 5 (HAdV-5) and genome copies (GCs) in waters that were artificially contaminated with tropical soil. The use of a mathematical equation was proposed based on the flocculation index of clay to assess the recovery of total GCs in these controlled assays. The results suggest that solids in the water reduced the viral genome copy loads per millilitre (GC·mL−1) and viral infectivity. OM did not influence the GC·mL−1 recovery rate (p N 0.05) but led to a 99% (2 log10) reduction in plaque-forming unit counts per millilitre (PFU/mL), which indicates that infectivity and gene integrity were non-related parameters. Our findings also suggest that acidic pH levels hinder viral inactivation and that clay is the main factor responsible for the interactions of HAdV-5 with soil. These findings may be useful for future eco-epidemiological investigations and studies of viral inactivation or even as parameters for future research into water quality analysis and water treatment. © 2015 Elsevier B.V. All rights reserved.
⁎ Corresponding author at: Núcleo de Pesquisas em Agente Emergentes e Re-Emegentes, Instituto de Patologia e Saúde Pública, Universidade Federal de Goiás, 74643-970 Goiânia, Goiás State, Brazil. E-mail address: [email protected] (M.T.A. Garcia-Zapata).
http://dx.doi.org/10.1016/j.scitotenv.2015.05.075 0048-9697/© 2015 Elsevier B.V. All rights reserved.
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
1. Introduction Human adenoviruses (HAdVs) belong to the family Adenoviridae, genus Mastadenovirus, which contains 57 different serotypes distributed across seven species (A–G) (ICTV, 2013). They are icosahedral, nonenveloped viruses containing a double-stranded linear DNA genome (Enriquez, 2002). These viruses are of substantial public health importance (Silva et al., 2010), as they are excreted in faeces, urine, and respiratory droplets (Metcalf et al., 1995; Jiang et al., 2001), and can cause a series of disease states in infected individuals by the respiratory and faecal–oral routes. Examples include upper respiratory tract infections (pharyngitis and tonsillitis), lower respiratory tract infections (bronchiolitis and pneumonia), conjunctivitis, cystitis, and gastroenteritis (Mena and Gerba, 2008). HAdVs are stable in the environment and resistant to water treatment methods (Thompson et al., 2003), particularly to UV irradiation (Liden et al., 2007). Furthermore, they are ubiquitous in the environment yearround (Katayama et al., 2008). These pathogens are prevalent in both treated and untreated water (Jiang et al., 2001; Silva et al., 2010; Fongaro et al., 2013) and are often detected in higher concentrations than other enteric viruses (Wong et al., 2010). Thus, HAdVs are indicated for use as viral biomarkers of environmental water and drinking water quality (Wyn-Jones et al., 2011; Silva et al., 2011). Detection of HAdVs in water destined for human consumption can be accomplished either by molecular techniques alone (Silva et al., 2011) or in combination of these techniques with cell cultures (Wyn-Jones et al., 2011; Garcia et al., 2012) that allows access infectivity (ability of the virus to replicate in permissive cells) (Herzog et al., 2008). Despite the great sensitivity of PCR, the main limitation is the lack of the correlation between the detected viral genome and viral infectivity, which limits conclusions about the significance for public health (Hamza et al., 2011). Despite advancements in the detection of HAdVs in different water sources, information is lacking on the relationship of these viruses with the solids, sediments, or suspended solids in the waters in which they are present.
315
Viral adsorption to solids is known to be an extremely complex process (Schijven and Hassanizadeh, 2000), and variations can be observed even among different serotypes of the same virus (Singh et al., 1986). Studies that report viral adsorption behaviour to solids usually employ phages as models (Schijven and Hassanizadeh, 2000). These studies are valid, but only as predictive models, and cannot infer the true permanence of infectious viruses, such as HAdVs. This is due to variations in the isoelectric point (pI), virion size, hydrophobicity, and capsid proteins (Schijven and Hassanizadeh, 2000), among other factors that have not yet been discovered. HAdVs are more prevalent than norovirus, enterovirus, hepatitis A virus, and human poliovirus in biosolids (Wong et al., 2010). Fong et al. (2010) found that 100% of sewage and effluent samples were infectious with HAdV. These findings suggest that matrix solids may play a crucial role in the gene integrity and infectivity of HAdVs. The interactions of HAdVs in soil solutions were recently analysed in two studies. In the first, Wong et al. (2012) investigated the influence of different concentrations of inorganic ions on the aggregation and deposition behaviours of HAdVs in sandy soil. In the second, Wong et al. (2013) evaluated the role of soil organic carbon (SOC) and solution-phase dissolved organic carbon (DOC) on sorption capacity and reversibility of organic carbon on adsorption of HAdVs. These two studies assessed adsorption behaviour using real-time quantitative PCR (qPCR) but did not evaluate the influence of solids and total organic matter on HAdV infectivity. qPCR is a highly sensitive technique, capable of detecting a small number of microorganisms (Botes et al., 2013), but when used alone, it is unable to distinguish between infectious and non-infectious viral particles. Furthermore, viral interactions with solid particles may lead to viral inactivation by loss of capsid integrity, with the consequent release of genetic material (Schijven and Hassanizadeh, 2000), which may be identified by qPCR either as an infectious or non-infectious particle. Therefore, the present study sought to (i) assess the recovery of infectious HAdV-5 and genome copies from simulated solutions containing tropical solids under controlled pH values in the presence or absence of organic matter; and (ii) define a mathematical equation to assess the recovery rate from clay soils under simulated conditions.
2. Material and methods The U.S. Environmental Protection Agency (USEPA, 2011) defines the acceptable level of total dissolved solids (TDS) in drinking water as 500 and 1000 mg·L−1 and at pH ranges of 6.0–9.5 (Brazil, 2005) and 6.5–8.5 (USEPA, 2011). Thus, the experiments were conducted to simulate maximum levels of solids in water contaminated with infectious HAdV-5, using the reference pH values of 6.0 and 8.0.
2.1. Soil characterisation and preparation Gleysoil (hydromorphic soil) is a typical soil of riverbanks in tropical environments (Rosolen et al., 2014). This type of soil is associated with poor land management (use and occupation) and intense precipitation runoff from the landscape. In these conditions, organic matter and minerals from these soils can reach the rivers (FAO, 2006; Reatto et al., 1998) and can be found during the water treatment process. Thus, gleysoil samples were collected at a depth of up to 15 cm from a native palm swamp area in the municipality of Bela Vista de Goiás, the central portion of the State of Goiás, Brazil in the Cerrado ecoregion (17° 00′S 48° 47′W). This type of soil is typically found at river and lake borders and wetlands (Rosolen et al., 2014) and is the most common sediment carried inside rivers and lakes (FAO, 2006; Reatto et al., 1998). The samples were air-dried, passed through a 2-mm mesh sieve, and any excess plant debris were removed with the aid of tweezers and a magnifier to yield the fine earth fraction described by Gee and Bauder (1986). The soil was subdivided into two portions: (i) soil with organic matter (WOM) and (ii) soil without organic matter (OM consumed by H2O2 – LOM – less organic matter/OM-free). The latter was treated with hydrogen peroxide (Whittig and Allardice, 1986) and autoclaved at 121 °C and 0.105 MPa for 3 h, three times, with 24-h intervals between each treatment (Zhao et al., 2008), to intentionally remove the soil organic matter. To ensure maximal homogeneity, the soil samples were crushed and stored at room temperature. The results of physicochemical and instrumental analyses of WOM and LOM soil samples are shown in Table 1. The clay fraction was determined by X-ray diffraction (XRD). The silt, clay (iron-free fraction and saturated with K, Mg, Mg + glycerol, K + 350 °C, and K + 550 °C), and sand fractions were also separated. Preparation for XRD and XRD itself was performed as described by Whittig and Allardice (1986) and Resende et al. (2005). Fig. 1 shows the X-ray diffractogram obtained.
316
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
Table 1 Physicochemical and instrumental analyses of soil samples containing organic matter (WOM) and without organic matter (LOM). Assay
Unit
WOM
LOM
Organic matter (OM)a Al (aluminium)b Si (silicon)b K (potassium)b Ti (titanium)b Fe (iron)b Iron — Fedc Iron — Feod Sande Silte Claye Degree of flocculationf Point of zero charge (PZC)g Densityh
dag·dm−3 % (wt) % (wt) % (wt) % (wt) % (wt) g kg−1 g kg−1 dag·dm−3 dag·dm−3 dag.dm−3 %
3.07 42.06 47.72 0.69 2.54 11.99 131.1 4.0 36.79 24.2 39 27 5.41 1.60
ND 41.72 46.70 0.47 1.97 9.14 20.2 3.0 ND ND ND ND 4.20 ND
i i
a
Measured by loss on ignition (Ball, 1964). Scanning electron microscopy/energy-dispersive spectroscopy (semiquantitative data). c Iron content of the clay fraction by the oxalate method (Whittig and Allardice, 1986). d Iron content of the clay fraction by the dithionite method (Whittig and Allardice, 1986). e Dispersion and pipette method (Embrapa, 1997). f Ratio of natural clay to total clay (Vettori, 1969, Embrapa, 1997). g PZC — potentiometric titration method (Embrapa, 1997). h Particle density (Embrapa, 1997). i Dimensionless quantity. b
Fig. 1. Overlay of X-ray diffractograms of original soil (clay with OM) and OM-free soil (clay without MO), showing the presence of 1:1 (kaolinite) clay as well as 2:1 (vermiculite and ilite) clays. This composition provided evidence of the degree of soil weathering, particularly by the presence of kaolinite and gibbsite, which are characteristic of highly weathered soils.
2.2. Preparation of viral inoculum A549 cells were cultured in 75-cm2 cell culture flasks containing Dulbecco's Modified Eagle's Medium (DMEM — High Glucose 1×) supplemented with 10% foetal bovine serum (FBS) and 1% PSA (penicillin G, streptomycin, and amphotericin B at final concentrations of 100 U·mL−1, 100 μg·mL−1, and 0.25 μg·mL−1, respectively). The cells were infected with human adenovirus 5 (species C, serotype 5 — HAdV-5) and cultured in a 5% CO2 incubator at 37 °C until cytopathic effects were observed under an inverted microscope. The cell monolayers were frozen and thawed three times, and any cellular debris was removed by centrifuging at 5000 rpm at 4 °C for 10 min. The viral inoculum consisted of 1-mL aliquots of the resulting supernatant.
2.3. HAdV inoculation Fig. 2 provides a scheme of the experimental study design. The method consisted of separately adding 5, 25, and 50 mg of original soil and OM-free soil to 50-mL sterile polypropylene tubes and adding 40 mL of ultrapure water, followed by brief manual agitation, to obtain suspended and settled solids. The pH of sediment-containing water solutions was adjusted to 6.0 and 8.0 (a different pH level for each soil type and concentration) with 0.1 M HCl. A 1-mL aliquot of viral inoculum was then added and ultrapure water was added to a final volume of 50 mL, to obtain TDS concentrations of 100, 500, and 1000 mg·L−1 for each test assay (TA). One control assay (CA — without addition of soil), which consisted solely of the viral inoculum in a 50-mL final volume of ultrapure water, was performed for each pH level. These control tubes were processed identically to the test tubes. The same experiment performed with original soil samples was run simultaneously with LOM soil. All concentrations were run in duplicate with repeats in triplicate. Tubes containing the virus and sediment solution were agitated at 150 rpm for 1 h at a temperature of 24 °C (Schijven and Hassanizadeh, 2000; Gantzer et al., 1994). After agitation, the tubes were rested for 1 h at 24 °C for precipitation of settleable solids. Two clear phases were observed: (i) the supernatant and (ii) the sediment. Briefly, 200-μL aliquots of the supernatant were collected from each tube and stored at −80 °C until processing. The remaining supernatant was removed, leaving only the pellet representative of settleable material, to which 2 mL of ASL buffer from the QIAamp DNA Stool Mini Kit (QIAGEN) was immediately added, following the manufacturer's instructions for nucleic acid extraction. Nucleic acids were also extracted from the 200-μL supernatant aliquots, using the same kit. All extracted samples were immediately stored at −80 °C.
2.4. Quantitative real-time PCR (qPCR) DNA samples were diluted 1:10 to avoid inhibition, and qPCR was performed as described by Hernroth et al. (2002). These specific primers anneal with the conserved region of the hexon gene of the HAdVs. All amplifications were run in duplicate in a StepOnePlus™ Real-Time PCR System (Applied Biosystems), using as standard a pBR322 plasmid containing part of the adenovirus hexon gene, to generate a standard curve (R N 0.98). The results were analysed to measure genomic copy numbers of HAdV-5 (GC·mL−1) in water and in sediment, and the mean GC·mL−1 values were compared to the mean results obtained in each CA.
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
317
Fig. 2. HAdV-5 copy numbers and infectious HAdV-5 counts in simulated aqueous solutions of solids obtained from tropical soil samples, under controlled pH and organic matter conditions.
2.4.1. Mathematical equation to infer the total copy number The GC·mL−1 values were used to construct a series of mathematical equations to infer the total copy number recovered in relation to the concentration of clay, both in the suspension (to simulate suspended solids in drinking water) and in the precipitate (to simulate solids that may settle as water is transported through pipes or containers). For the test assay (TA) of each soil treatment, the total HAdV genome copy numbers in the supernatant (Nsup) and in the precipitate (Nprec) were calculated according to the following equations, respectively: Nsup ¼
. h ðClay% V cm3 Þ
100
Nprec ¼ ðV cm3 Þ−
hh
i Floc GC mL−1
V cm3 Clay%
. 100
i i Floc GC mL−1
ð1Þ
ð2Þ
where V, Clay% and Floc are, respectively, the volume of sediment (cm3) in the tube, the clay content (%) of the matrix, and its degree of flocculation. Volume is the direct relationship between mass (g) and density (g·cm−3), where 1 cm3 is equal to 1 mL. The clay content is determined by soil texture analysis, and the degree of flocculation is the difference between total clay and natural clay after dispersion (Table 1). The total genome copy number for the TA (Nta) was obtained by compiling Eqs. (1) and (2) and deriving Eq. (3) below: Nta ¼ Nsup þ Nprec :
ð3Þ
The overall recovery rate (Rr), expressed in the number of copies, was calculated by Eq. (4):
Rr ð % Þ ¼
ðNca −Nta Þ 100 Nca
ð4Þ
318
H.D. Silva et al. / Science of the Total Environment 530–531 (2015) 314–322
where Nca is the total number of GCs in the CA. Eq. (4) may be rewritten as follows (Eq. (5)): Rr ð % Þ ¼
. hh 8 ðClay% V cm3 Þ
