Infectivity of Giardia duodenalis Cysts from UV Light-Disinfected Wastewater Effluent Using a Nude BALB/c Mouse Model

Giardia duodenalis is a protozoan of public health interest that causes gastroenteritis in humans and other animals. In the city of Campinas in southeast Brazil, giardiasis is endemic, and this pathogen is detected at high concentrations in wastewater effluents, which are potential reservoirs for transmission. The Samambaia wastewater treatment plant (WWTP) in the city of Campinas employs an activated sludge system for sewage treatment and ultraviolet (UV) light for disinfection of effluents. To evaluate this disinfection process with respect to inactivating G. duodenalis cysts, two sample types were investigated: (i) effluent without UV disinfection (EFL) and (ii) effluent with UV disinfection (EFL+UV). Nude immunodeficient BALB/c mice were intragastrically inoculated with a mean dose of 14 cysts of G. duodenalis recovered from effluent from this WWTP, EFL, or EFL+UV. All animals inoculated with G. duodenalis cysts developed the infection, but animals inoculated with UV-exposed cysts released a lower average concentration of cysts in their faeces than animals inoculated with cysts that were not UV disinfected. Trophozoites were also observed in both groups of animals. These findings suggest that G. duodenalis cysts exposed to UV light were damaged but were still able to cause infection.


Introduction
Giardiasis is an intestinal infection caused by the protozoan Giardia duodenalis, and it represents the most commonly reported protozoan infection in humans and other animals. Some of the assemblage A and B subtypes of G. duodenalis have the potential for zoonotic transmission [1][2][3]. Cysts are the environmental stage of this organism, and they remain viable for several months under a range of environmental conditions and are extremely resistant to chemical disinfection. e infectious dose for these cysts is low; 10-25 cysts can cause illness [4]. Acquisition of giardiasis occurs through a faecal-oral route, which can be person to person, foodborne, or through contaminated water (drinking water or during recreational activities), with the last being the most signi�cant means of transmission of this infection [5][6][7].
A large outbreak of giardiasis recorded in 2003 (Boston, MA, USA), exhibited two modes of transmission, which illustrated the capacity of G. duodenalis to spread through multiple means of transmission; these two modes included a common outbreak by exposure to contaminated recreational water and a subsequent prolonged propagation through interpersonal transmission between people in the community [4]. is pathogen was responsible for 132 waterborne outbreaks that occurred in North America and Europe and consequently resulted in a public health concern [8][9][10]. In Norway (2004), 1,500 people were diagnosed with giardiasis, which was described as the largest waterborne outbreak recorded worldwide. e water treatment plant involved in this outbreak used chlorination for water treatment [11,12], and it is well established that G. duodenalis cysts are resistant to the chlorination process [8,13].
Many studies on the removal of this parasite have been performed in wastewater treatment plants (WWTP) using different treatment processes, and removal efficiencies of approximately 80.0 to 98.4% have been reported for G. duodenalis cysts [11,14]. However, cysts have been detected in effluent samples at concentrations of 10 3 to 10 4 cysts L −1 throughout the world, and little is known about the infectivity of cysts in these samples [15][16][17][18].
Ultraviolet (UV) disinfection is of growing interest for use in WWTPs because it has been demonstrated that UV radiation is effective against protozoans and does not generate by-products, which does occur with chlorination and ozonation [13,19,20]. However, because the biocidal effect of UV light is caused by the absorption of UV photons by nucleic acids (DNA/RNA) in cells, causing damage to the nucleic acids, the characteristics of the liquid to be disinfected (turbidity for example) are considered a signi�cant parameter for disinfection [21,22].
Research on the effect of UV radiation on cyst inactivation using in vitro methods underestimates the loss of infectivity for G. duodenalis cysts following UV exposure. For example, in vitro excystation is not a reliable assay for assessing infectivity and does not correlate well with animal infectivity assays [23]. Additionally, vital dye viability assays (using DAPI/PI) signi�cantly underestimate cyst inactivation compared with infectivity, which indicates that vital dye viability assays should not be used to de�ne inactivation [24].
Infectivity analyses using animal models are the best method for this purpose because they require the parasite to be able to reproduce and complete its entire lifecycle to be considered infectious [23,24].
e BALB/c nude animal model was based on the fact that nude mice are natural athymic mutants that are susceptible to infection at low doses. is susceptibility permitted the use of a smaller number of individuals per group, which is in accord with ethical aims of animal experimentation [25].
e natural, recessive nude mutation, which was initially reported to cause a loss of hair in homozygous mice, results in a rudimentary thymus that causes a reduction in the number of lymphocytes [26,27].
In the present study, we evaluated the infectivity of G. duodenalis cysts recovered from effluent from the Samambaia WWTP disinfected with UV light. Evaluation of infectivity was determined by an animal infectivity assay using BALB/c nude mice. In all experiments, animals inoculated with cysts that were not disinfected with UV light released a higher concentration of cysts in their faeces compared to animals inoculated with UV-exposed cysts. Trophozoites were also observed on histological slides containing duodenal samples from animals in both groups. e UV light reactor used in our experiments was operated at an actual scale; the effluent quality was highly variable (a high concentration of solids in suspension and high turbidity), affecting the performance of UV radiation.

Wastewater Treatment Plant.
e Samambaia wastewater treatment plant (WWTP), southeast of Campinas (22 ∘ 56 ′ 00 ′′ S, 47 ∘ 00 ′ 00 ′′ W), treats approximately 70 L/s of sewage using an activated sludge system. In the experimental phase, operating at a real scale, an In-Line 250 Liquids Disinfection System reactor using high intensity UV radiation (Germetec UV & GO Technology/ http://www.germetec.com.br) was installed in the WWTP to disinfect the effluent using a 25 to 30 mJ cm −2 dose (medium pressure UV light).

Cyst
Recovery. is study was conducted in real scale and therefore subject to the variations of the operational conditions of a WWTP. During two years, two litres of effluent disinfected by UV radiation and of effluent without UV disinfection were collected once in two weeks at the Samambaia WWTP and concentrated following cellulose ester membrane �ltration (porosity of 3 m, diameter of 47 mm, Millipore) in accordance with the method by Franco et al. [28]. e samples were eluted from each membrane by alternately scraping, the membranes with a smoothedged plastic loop and rinsing the membrane with elution solution for 20 minutes. e resulting liquid was centrifuged at 1,050 ×g for 10 minutes, and the concentrated pellet was washed and centrifuged again. e density of Giardia cysts was determined with �uorescent monoclonal antibody tests (IFA) (Meri�uor kit, Meridian Bioscience, Cincinnati, Ohio) according to the manufacturer's instructions. e samples were then maintained in a 2% potassium dichromate solution.

Infectivity Assay.
Groups of 3-4-week-old female nude BALB/c mice (Pasteur lineage) were purchased from CEMIB (Multidisciplinary Centre for Biological Investigation-UNICAMP) with the assurance that the mice were free from all possible protozoan infections. e animals were divided into the following 4 treatment groups according to the inoculum tested: (i) effluent with UV disinfection: EFL+UV ( 3); (ii) effluent without UV disinfection: EFL ( 3); (iii) effluent �ltered through cellulose ester membranes (nominal porosity 3 m) and thus presumably T 1: Presence of cysts and trophozoites in faeces, scrapings, and histological preparations from the duodenum and ileum of mice inoculated with cysts of G. duodenalis obtained from treated wastewater disinfected (or not) with UV light.

Groups
Cysts (faeces) Cysts (scrapings) Trophozoites (scrapings) Trophozoites (histological slides) Effluent+UV a e presence of cysts in faeces or intestinal trophozoites was found to be infection = positive and, b if not detected = negative. free of G. duodenalis cysts: EFLF ( ); and (iv) sentinel animals, which were not inoculated: ST ( ) to check for intercage contamination. To prevent undesirable microbial infections in these animals, prior to inoculation, the samples were sanitised with 1.0% sodium hypochlorite (60 minutes, 4 ∘ C), and the bleach was then washed out. All mice, the EFL+UV and EFL groups were intragastrically inoculated with a mean dose of 14 cysts/10 L. e inoculum was obtained from a pool of cysts recovered from four samples chosen at random from the samples collected over 2 years. e number of cysts in the effluent samples was determined by IFA. Mice were caged individually and maintained in isolation. From day 5 to 15 aer infection; faeces from all animals were collected and stored in 2% potassium dichromate and assayed by zinc sulphate �otation (speci�c gravity = 1.2) to isolate cysts. At day 15 aer inoculation, all animals were necropsied. A piece of the small intestine of each mouse was removed and �xed for histological examination, and mucosal scrapings from the duodenum and ileum were stained with Giemsa to detect trophozoites (the whole �eld was observed). Mice were recorded as Giardia infected if cysts or trophozoites were found in faecal specimens, in Giemsa-stained intestinal scrapings, or in histological slides (six slides/mice). e protocol used in these experiments was submitted to the Commission of Ethics in Animal Experimentation/ CEEA-IB-Unicamp (protocol 909-1) and was approved as being in accordance with the Ethical Principles in Animal Experimentation adopted by the Brazilian College of Animal Experimentation (COBEA).
e two reactions were performed in a 25 L volume containing 1x PCR buffer, 2,5 mM MgCl 2 , 0,2 mM each dNTP, 25 pmol of each primer, 1,25 U of Taq polymerase platinum (Invitrogen), and 1 L of DNA template from cysts isolated from effluent samples. PCR ampli�cation was conducted in a thermocycler (MJ PTC 100-MJ Research INC). e samples were denatured at 94 ∘ C for 3 minutes, followed by 40 cycles of 94 ∘ C for 1 minute, 60 ∘ C for 1 minute, and 72 ∘ C for 1 minute. Final elongation was performed at 72 ∘ C for 7 minutes. e obtained products were con�rmed by 3% agarose gel electrophoresis with ethidium bromide staining.
e two reactions to amplify the GDH gene contained 1x PCR buffer, 2 mM MgCl 2 , 0,2 mM each dNTP, 12,5 pmol of each primer, 0,625 U of Taq polymerase platinum (Invitrogen), and 1 L of DNA of effluent sample. PCR ampli�cation was conducted in a thermocycler (MJ PTC 100-MJ Research INC). e samples were denatured at 94 ∘ C for 3 minutes, followed by 40 cycles of 94 ∘ C for 30 seconds, 59 ∘ C for 30 seconds and 72 ∘ C for 1 minute. Final elongation was performed at 72 ∘ C for 7 minutes. e �rst reaction was performed in a volume of 25 L, of which 12 L was used as a template for the second reaction, which was performed in a volume of 100 L. e obtained products were con�rmed by 3% agarose gel electrophoresis with ethidium bromide staining.
Before sequencing the obtained GDH gene fragment, the products of the second round of PCR were puri�ed using the QIAGEN QIAquick PCR kit. e sequencing reaction was performed using the ABI Prism Big Dye Terminator Cycle Sequencing kit ver. 3.1 (Applied Biosystems), and the samples were sequenced in both directions at least four times using GDHF3 or GDHB5 primers. SeqMan ver. 5.01 (DNASTAR) soware was used to edit the sequences. Reference GDH gene sequences from the seven major G. duodenalis assemblage (A, B, C, D, E, F, and G, GenBank accession numbers L40509.1, L40508.1, U60985.1, U60986.2, AY178740, AY178744, and AY178748.1, resp.) were aligned using Clustal X soware [32]. e homologous nucleotide sequence from Giardia ardeae (accession number AF069060.2) was used as an out group. Phylogenetic analyses were conducted in the soware MEGA ver. 5.05 [33] using neighbour-joining and maximum likelihood algorithms. e model of nucleotide substitution that best �t the data was determined aer analysis with jModelTest soware [34]. Tamura-Nei93 distance [35] with gamma correction was used and bootstrap phylogeny test was performed in both methods with 10,000 replicates.

Statistical
Analysis. Student's t-test with a P value, 0.005 was used to analyse the data.

Results
Ampli�cation of fragments of the -giardin gene and the 220bp fragment of the GDH gene demonstrated the presence of G. duodenalis in the effluent samples from Samambaia WWTP. e sequencing reaction performed for the ampli-�ed fragment from the GDH gene using DNA from a positive sample con�rmed the presence of G. duodenalis assemblage A (Figure 1). e sequence that was obtained has been deposited in GenBank under accession number JN116502.
G. duodenalis cysts were not detected in the faeces of animals inoculated with effluent �ltered through membranes with a porosity of 3 m (EFLF-negative control), indicating that the �ltration process used for this effluent sample was efficient. e sentinel animal (ST) was also negative for cysts in the faeces, indicating that no contamination occurred inside the isolator. e animals in the EFL+UV group eliminated cysts in their faeces from the 5th day to the 11th day aer inoculation. e animals in the EFL group eliminated cysts in their faeces beginning on the 5th day aer inoculation, and this persisted to the end of the experiment (Figure 2).
With respect to the number of G. duodenalis cysts in the faeces per day, we found that the total number was similar between groups EFL+UV and EFL. However, there was a signi�cant statistical difference between the groups (P > 0,005). e animals of group EFL eliminated an average of 36.7 cysts/day (varying from 8.3 to 41.6 cysts/day) during the 15 days of the experiment, with the highest number of cysts being eliminated on the 13th day. e animals in the EFL+UV group eliminated an average 16.6 cysts/day (varying from 8.3 to 33.3 cysts/day), with the maximum number of cysts being eliminated on the 11th day.
Neither the intestinal scrapings from the duodenum/ ileum nor the enteric histological slides from animals in the EFLF and ST groups were positive for trophozoites. In contrast, the groups inoculated with G. duodenalis cysts (EFL+UV and EFL) trophozoites were observed (Table 1 and Figure 3).
ere was a difference observed with respect to the number of cysts released by the animals that developed the infection and in the number of trophozoites found in the microscopic slides (Figure 4).
Cysts were also observed in histological slides for samples obtained from animals in the EFL+UV group.

Discussion
e cysts detected in effluent from Samambaia WWTP and used to infect mice following UV light disinfection were designated as belonging to assemblage A. Assemblage A infects most vertebrates. Additionally, assemblages A and B are the only two assemblages known to infect humans and are thus considered to have zoonotic potential. e effluents from this WWTP aer UV light disinfection are discharged into the streams of Pinheiros, which then �ow to the Atibaia River. Several cities use the waters of this river as a water supply source, including Campinas, which is the third largest city in São Paulo state. Given that this pathogenic agent has a major impact on public health, these results show the importance of protecting surface water sources from sewage discharges.
Intragastric inoculation of mice with G. duodenalis cysts resulted in infection, as assessed by cysts eliminated in their faeces and presence of trophozoites in their small intestines. Our observation of cysts on the 5th day aer inoculation is in accordance with kinetics of Giardia infection observed in mice and with the capacity of the BALB/c strain to eliminate the infection in a short period of time compared to the other strains [36,37].
Typically, peak cyst release was observed at some point during the �rst 10 days of infection, followed by a decline in the cyst output and eventual elimination of the parasite at the end of 2 weeks.
e results obtained in this study con�rm that acquisition of this infection may occur at low doses. All of the animals in the EFL and EFL+UV groups inoculated with approximately 14 cysts developed an infection, but the infection was associated with an absence of clinical signs and there was a difference in infection intensity between groups. It was observed that the intensity of infection was signi�cantly lower in animals of group EFL+UV than animals of group EFL.
A lower level of inactivation of the cysts by the UV reactors may be explained by the UV dose delivered to the cysts being lower than necessary, the potential repair of cysts following UV radiation exposure, and the protection of cysts from inactivation due to particle shielding effects [37]. e radiation dose evaluated in this study was 25-30 mJ cm −2 , and several studies on the inactivation of protozoa by UV radiation have shown inactivation at a UV dose of 20 mJ cm −2 . Disinfection of G. duodenalis cysts with a UV dose of 1 mJ cm −2 based on evaluation in an infectivity assay (Meriones unguiculatus) was reported by Shin et al. [20], suggesting that the UV radiation dose used by the Samambaia WWTP should have been appropriate for the inactivation of G. duodenalis cysts.
It is possible that repair of the cysts may have occurred, but the repair mechanism was not evaluated in this study. Rochelle et al. [38] showed that Cryptosporidium spp. protozoa contain genes necessary for repair. However, there is no evidence that UV light-exposed oocysts can be sufficiently repaired to regain their preexposure levels of infectivity. Nevertheless, other components of the cell may be damaged by the UV disinfection performed [8,13].
e effectiveness of UV radiation is directly related to the UV dose absorbed by microorganisms, and the characteristics of the liquid to be disinfected. For instance, turbidity and the concentration of solids in suspension are considered signi�cant parameters affecting the performance of UV radiation for disinfection. ere are 3 principal factors involved in this interference: (a) the number and size of particles, which could cause dispersion of radiation and occurrence of shading areas; (b) the nature of particles (organic or inorganic); and (c) the degree of association of microorganisms and particles, which protects the microorganisms from UV radiation [22]. us, the properties of wastewater effluent may not be favourable for disinfection of the effluent by exposure to UV light [13,19].
Based on faecal coliform inactivation, some studies indicate that suspended solid concentrations and turbidity are greatly reduced by sand �ltration of secondary effluent wastewater samples and that this increases the performance of UV disinfection signi�cantly [39]. In the case of this study, G. duodenalis cysts recovered from the effluent of Samambaia WWTP maintain the ability to cause infection. Samambaia WWTP was designed to produce secondary effluent wastewater through biological treatment. e effluent quality observed during this study showed an average turbidity of 13.37 NTU (range 1.75 to 25.00) and average concentration of solids in suspension of 18.00 mg L −1 (range 2.00 to 34.00). ese parameters correspond to samples analysed in the infectivity assay. erefore, the UV disinfection employed 6 ISRN Parasitology in this case was not able to inactivate G. duodenalis cysts, which was attributed to effluent quality [15]. Li et al. [40]. assessed the infectivity of cysts in wastewater following disinfection with UV radiation and also reported, based on an animal infectivity assay, that this disinfection process was not completely efficient for inactivation of Giardia spp.
Even with their inherent variability, infectivity assays provide a direct measure of the ability of cysts to cause infection and represent the best method for the purpose of evaluating the efficiency of various processes of disinfection in the inactivation of pathogens, such as G. duodenalis cysts. However, it is necessary for researchers to obtain standardised protocols to minimise the variables associated with these assays.
e results obtained in this study have great public health relevance, considering that conventional processes of sewage treatment using activated sludge do not result in total removal of protozoan pathogens, and disinfection by light UV is not able to achieve complete inactivation of these pathogens before discharge to the environment (into streams). Individuals infected with a few cysts or trophozoites have the same probability of disseminating the parasite as those receiving high doses [41,42]. Sewage effluent is an important source of environmental contamination, particularly when it is water supplied for drinking, recreation, or agricultural purposes that is contaminated. e possibility exists that upon drinking this water, individuals infected with a decreased dose of cysts will develop subclinical giardiasis, as shown from the results obtained in this study, in which the animals in our model developed the sub-clinical form of this illness.
Giardiasis is well recognised as causing chronic infections. us, G. duodenalis cysts present in sewage in�uent cannot only be derived from symptomatic individuals, but also from asymptomatic persons. It is therefore noteworthy that a lack of clinical symptoms may preclude a search for treatment, and hence, asymptomatic individuals act as an important source of dispersion of G. duodenalis cysts.