Synergism in sequential inactivation of Cryptosporidium parvum with trypsin and UV irradiation

Cryptosporidium, a protozoan parasite, in wastewater presents a major public health concern for water safety. However, bactericidal efficiencies of conventional disinfection methods towards Cryptosporidium oocysts are still hampered owing to the presence of their thick outer wall. In this study, we present a novel UV inactivation process where the efficiency has been significantly enhanced by addition of a trypsin pretreatment stage. Notably, inactivation (log-reduction) of oocysts was noted to be 73.75–294.72% higher than that obtained by UV irradiation alone, under identical conditions. Experimental observations and supporting mechanistic analyses suggest that trypsin led to cleavage of the protein layers on the oocyst wall, facilitating penetration of UV radiation into the oocysts leading to degradation of their genomic DNA (gDNA). The dissociative effect of trypsin on the oocyst wall was indicated by the fact that 64.50% of oocysts displayed early apoptosis after trypsinization. Imaging by scanning electron microscopy indicated that this combined treatment led to substantial disruption of the oocyst coat, deforming their shape. This resulted in the release of cellular proteins and gDNA, their concentrations in bulk solution increasing by 1.22–8.60 times. As UV irradiation time was prolonged, gDNA was degraded into smaller fragments with lower molecular masses. Both laddering and diffuse smear patterns in gel analysis indicated significantly detrimental effects on gDNA and viability of oocysts. Overall, this study demonstrated enhancement of UV inactivation of Cryptosporidium oocysts by trypsin and explored the underlying mechanisms for the process.


Introduction
Cryptosporidium, as one of the common waterborne protozoan parasites, presents a major public health concern for water safety (Hamilton et al. 2018). It has the potential to be spread through both zoonotic and anthroponotic transmission by the fecal-oral route (Gharpure et al. 2019). Infection with Cryptosporidium can cause severe and life-threatening gastrointestinal disease (i.e., cryptosporidiosis) in humans, from which there is neither a vaccine nor an effective treatment (Li et al. 2016). Cryptosporidium oocysts have been found in various types of water worldwide (river, drinking water, and wastewaters) (Abeledo-Lameiro et al. 2018). The global emission of Cryptosporidium oocysts to surface waters has been estimated at 3 × 10 17 oocysts per year (Hofstra et al. 2013), which has been associated with high-profile outbreaks. Cryptosporidium was estimated to be responsible for 69 million cases of illness, and 57,203 deaths in 2016 (Kooh et al. 2021). During 2009 and2017, 444 Cryptosporidiosis outbreaks occurred in the USA and resulted in 7465 cases, of which 56.7% were associated with contaminated river/reservoir and recreational waters (Gharpure et al. 2019).
The ever-present risk of contamination by Cryptosporidium has therefore required the implementation of removal or inactivation processes by water utilities. The Cryptosporidium oocysts are usually of small size (3.5-6 μm) and low density (approximately 1.045 g cm −3 ), and possess high electronegativity with an isoelectric point below pH 3 (Xiao et al. 2022). These properties allow oocyst suspension to be relatively stable, due to the unfavorable filtration properties, weak gravitational settling, and repulsive electrostatic force (Xiao et al. 2022). Therefore, removal of oocysts by common sedimentation and activated sludge units (Rizk et al. 2019;Medeiros et al. 2020) is difficult. In many wastewater treatment plants (WWTPs), removal of oocysts is enhanced by flocculation, settling and filtration processes, or dissolved air flotation mediated by coagulants and frothing agents (Hatam-Nahavandi et al. 2015;dos Santos and Danie, 2016;Xiao et al. 2022).
Moreover, disinfecting agents such as chlorine, ultraviolet (UV), and ozone are often considered the final barriers in the prevention of pathogen transmission by wastewater effluents. Cryptosporidium has been found to be extraordinarily resistant to all chlorine-based disinfection techniques (i.e., chlorine dioxide, chlorine, and monochloramine), even at high doses (Hamilton et al. 2018). The oocysts are almost all viable, even after long exposure times (120 min) in 0.5 ppm of chlorine (Adeyemo et al. 2019). Ozonation has stronger inactivating capabilities, compared with chlorination (Nasser, 2016); however, its application usually requires corrosion-resistant equipment, efficient contacting systems, and careful control of disinfection by-products (Wu et al. 2019). UV irradiation offers an alternative for the removal of Cryptosporidium, and the use of this technique has been growing extensively in water/wastewater treatments due to its ease of operation, absence of halogens, and high disinfection efficiencies (Li et al. 2018). Nevertheless, the removal of Cryptosporidium across real wastewater treatment trains is often not achieved thoroughly. Reported removal efficiencies in different WWTPs ranged from under 10 to over 90% for Cryptosporidium oocysts (Rizk et al. 2019;Medeiros et al. 2020), and a recent investigation found that the mean concentration of live oocysts present in effluent from four WWTPs was as high as 4.15 oocysts L −1 , even when these WWTPs were equipped with modified secondary clarifiers with coagulation and ozone/UV disinfection units (Xiao et al. 2022). The residual concentrations of Cryptosporidium oocysts still exceeded the WHO drinking water quality standard (2006) (this requires < 1 oocyst per 10 L of water). Thus, more effective disinfectant regimes are urgently needed for the inactivation of the recalcitrant oocysts.
The main reason for the resistance of Cryptosporidium oocysts is that they are protected by a thick outer wall. This shell structure limits the entry of potentially harmful watersoluble chemicals and allows the long-term persistence of oocysts in harsh water/wastewater systems (Xiao et al. 2022). The durable oocyst wall has been found to be a protective barrier consisting of a complex multi-layer proteinlipid-carbohydrate matrix (Jenkins et al. 2010;Samuelson et al. 2013). The main hypothesis of this study is that if the protective coat of oocyst can be initially disrupted, this may then benefit the permeation of disinfectants and potentiate damage to the interior structures. Proteolytic enzymes (e.g., trypsin), which are widely used to detach cells from substrata and to separate cells from living tissues (Iranzo et al. 2002), are desirable to break proteins layer on the oocyst wall. We speculate that this dissociation of an oocyst wall will make it easier for UV radiation to penetrate and attack the genomic DNA. To the best of our knowledge, the effects of proteolytic enzyme processes on the efficacy of UV irradiation for the inactivation of Cryptosporidium oocysts, and determination of the underpinning mechanisms, have not been studied previously.
Therefore, the present study assessed the effectiveness of inactivation and unveiled the underlying mechanism of the proposed UV irradiation process, coupled with trypsin pretreatment (Fig. 1). C. parvum oocysts were employed as model organisms in this study, as it is one of the most commonly found coccidian parasite species in water. The inactivation efficiency was investigated through the monitoring of the viability of oocysts detected using an immunofluorescence assay and flow cytometry. Disinfection mechanisms were confirmed by morphology destruction, release of cellular proteins, and damage to gDNA.

C. parvum oocysts
Bovine-derived C. parvum oocysts were obtained from the laboratory of Xichen Zhang in Jilin University, China. Oocysts were genotyped, as previously described by Xiao et al. (2022), by direct PCR analysis and sequencing of the 16S rRNA gene. C. parvum oocysts were propagated from infected male Holstein calves and purified using glucose flotation and cesium chlorine gradient ultracentrifugation. The purified oocysts were stored at 4 °C in sterile phosphatebuffered saline (PBS) supplemented with antibiotic solution according to methods described by King et al. (2008). In all cases, purified oocysts were shipped on ice and were used for disinfection trails within 20 days. The excystation efficiency for oocysts used for infections and UV in activation experiments was in the range 95-99.5%. Scanning electron microscopy (SEM) (Philips XL30-ESEM, USA) was employed to observe the morphology of oocysts in these experiments.

UV irradiation of oocysts
The UV inactivation experiments were carried out using a low-pressure UV lamp (X6500, Xujiang, China) at a wavelength of 254 nm with an optical power output of 14 W. Twelve recline quartz cuvettes (10 mm) were used as reactors and were placed on a rotating table (15 s/revolution) under the lamp so that all reactors were completely and uniformly illuminated (Fig. 1). Aliquots (2.5 mL) of purified oocyst suspension (1 × 10 6 oocysts per mL) were added to the cuvettes. The solution pH was 7.0 and the temperature was maintained at 25 (± 0.3) °C using an air conditioning unit (DEM12MCP5, Emerson, China). The cuvettes were taken out at specific time intervals (2 min) for targeted analysis of oocysts, corresponding to the UV doses of 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 mJ·cm −2 (0 ~ 20 min), respectively. A radiometer (LS126C UVC, Linshang, China) was used to ensure the correct UV dose was applied to each sample. All experiments were undertaken in triplicate.

Viability determination
The viability of the oocysts after initial pretreatment with trypsin followed by UV irradiation was assessed and determined using vital dye 4', 6-diamidino-2-phenylindole (DAPI) and propidium iodide (PI) according to a method previously described Xiao et al. (2022). The numbers of live Cryptosporidium oocysts (DAPI + /PI-and DAPI-/PI-) were counted by fluorescence microscopy (BX63, Olympus, Japan). All samples were analyzed in triplicate, and the final density of live oocysts (number of oocysts per liter) determined as an average value. The reduction (log 10 reduction) in oocyst viability was interpreted as the inactivation ratio, log (N 0 /N t ), where N t is the number of live oocysts observed in the suspension after UV exposure and N 0 is the number counted prior to UV exposure.

Flow cytometry analysis
The influence of trypsin on Cryptosporidium oocysts was examined using Annexin V apoptosis assays (Robinson et al. 2020). Briefly, oocysts before and after trypsin treatment (0.25%, 20 min) were harvested (more than 1 × 10 5 oocysts) and the suspensions centrifuged at 2500 rpm for 5 min at room temperature. The resultant supernatant was removed Fig. 1 a Traditional UV inactivation of Cryptosporidium oocysts is hampered by their thick outer wall. b Scheme of the sequential inactivation of Cryptosporidium parvum with trypsin and UV irradiation and oocysts were washed once in PBS. Oocysts pellets were then resuspended in 200 μL binding buffer containing 4 μL of 0.5 mg mL −1 PI and 2 μL of Annexin V-FITC. Samples were then incubated in the dark at room temperature for 10 min. Early-stage apoptotic cells (AnnexinV + /PI −), necrotic cells (AnnexinV − /PI +), and late-stage apoptotic cells (AnnexinV + /PI +) were analyzed by flow cytometry (DxFLEX, Beckman Coulter, USA). The data were processed using CytExpert software (v 2.3; Beckman Coulter, USA).

Detection of extracellular proteins and DNA, and gDNA of oocysts
Oocyst suspensions before (untreated) and after the three treatments (treatment time 10, 20, and 30 min) were centrifuged at 6000 g, 4 °C for 15 min. The resulting supernatants were collected for analysis of both extracellular proteins and DNA (eDNA). The proteins in the supernatant were measured using a BCA protein quantitative detection kit (No Ml024499, Mlbio Company, China), and identified as extracellular proteins. The residual pellets were subjected to a combined mechanical-chemical cell lysis using the PowerFood™-Microbial DNA Isolation Kit (Biogenro, Beijing, China) in order to obtain intracellular (genomic) DNA (gDNA). Both eDNA and gDNA fractions were also purified using the Inhibitor Removal Technology (IRT) technique provided by this DNA Isolation Kit. DNA in both fractions was quantified by a Qubit™ 4 fluorometer (Thermo Fisher, USA). Quantification of gDNA degradation was conducted using a yield gel, as follows. Extracted gDNA was normalized to 100 ng μL −1 and initially evaluated by gel electrophoresis by loading 100 ng of gDNA in a 1% agarose gel along with DNA quantification standards (250-2000 bp). The distribution of DNA fragments was visualized using a shortwave UV light transilluminator.

Pretreatment of Cryptosporidium oocysts using trypsin
In the pretreatment trials, the relative trypsin activity was indicated by the change in absorbance at 253 nm for BAEE reaction (Fig. 2a). As the reaction time was extended, the product revealed a corresponding increase in UV absorbance. This increase appeared to be linear over Fig. 2 a The activity of trypsin in the reaction mixture used in this study. b The inactivation of Cryptosporidium oocysts using trypsin exclusively. Flow cytometry histogram and dot plots illustrating viability of Cryptosporidium oocysts before (c) and after (d) trypsin treatment (0.25%, 20 min) the early part of the reaction (0-20 min), corresponding to constant enzyme activity. This implied that the BAEE was fully hydrolyzed within 20 min, and that afterwards no significant conversion to product was observed. The activity of 1.00% trypsin was almost 2.20 times as that of 0.10% trypsin, suggesting a significant increase in activity with increasing trypsin concentration. These results were consistent with results previously reported (Crowell et al. 2013). The trypsin employed herein displayed considerable operational stability in the reaction mixture of pretreatment used in this study.
In a parallel experiment to this BAEE spectroscopic assay, trials to examine the viability of Cryptosporidium oocysts in trypsin solution were conducted using the DAPI/PI method. The results are illustrated in Fig. 2b, and demonstrated that using only trypsin had little influence on the oocyst. The inactivate rates were only 0.33%, 1.68%, 3.72%, and 3.89%, respectively, at trypsin concentrations of 0.10%, 0.25%, 0.50%, and 1.00% correspondingly. Although the oocysts' exposure to trypsin induced certain damage to membrane proteins, the trypsinization did not completely alter the integrity and function of the oocyst wall. The fluorescent dyes were thus still excluded by the oocysts treated just by trypsin and did not exhibit a response for the DNA stain (Rousseau et al. 2018). Consequently, the oocyst viability indicated by dye exclusion methods illustrated in Fig. 2b showed minimal changes. This was in accordance with previous reports that cell viability was not significantly altered by trypsin digestion alone (Heng et al. 2009;Tsuji et al. 2017).
Nevertheless, the trypsin-denudation of oocysts walls was evidenced using the Annexin V apoptosis assay by flow cytometry. The oocyst wall has been found to exist as a multi-layer protein-lipid-carbohydrate matrix (Chatterjee et al. 2010;Jenkins et al. 2010) with phospholipid compositions therein to contain phosphatidylcholine, phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine (PS), sphingomyelin, and cardiolipin (Jenkins et al. 2010). In normal cells, PS is located on the inner surface of cell wall (Nagata et al. 2016), but if the enzymatic damage to the cells/oocysts causes apoptosis and disrupts their structures, the phosphatidylserine may translocate and become exposed on the outside surface (Mariño and Kroemer 2013). Annexin V, with high affinity for PS, can thus be conjugated to fluorochromes (Robinson et al. 2020) and then used to indicate the influence of trypsin on oocyst walls. As shown in Fig. 2c, untreated samples predominantly (99.56%) appear in the lower left quadrant, which represent viable cells, whereas the majority of oocysts (64.50%) presented in the lower right quadrant after trypsin treatment (0.25%, 20 min), indicating early apoptosis (Fig. 2d). This verified the results of the viability analysis in illustrated in Fig. 2b. Oocysts yielded consistently high viability when dissociated with trypsin.

Effect on viability of trypsin-UV sequential disinfection
The inactivation behaviors of oocysts, which were expressed on a log scale, were observed under different experimental conditions (Fig. 3). Main inactivation constants obtained by regression analysis for UV irradiation experiments are listed in Table 1. In the control test (black circles in Fig. 3a-d), when trypsin was not applied (UV only), 0.72-log inactivation of oocysts was achieved at 16 mJ cm −2 . This result was similar to those observed by previous studies. Adeyemo et al. (2019) reported values of 20.8 mJ cm 2 for 1.12-log inactivation of Cryptosporidium oocysts, and that the oocysts were more resistant than Escherichia coli to UV irradiation (14 mJ cm −2 for 1.44-log) (Koivunen and Heinonen-Tanski 2005).
When 0.10% trypsin was used for pretreatment, the final log reductions of live oocysts increased to 1.25 (treatment time, 10 min), 1.79 (20 min), 2.73 (30 min), and 2.83 (40 min). Prolonging treatment time with trypsin significantly enhanced the subsequent microbicidal effect. Increase in trypsin concentration from 0.10 to 1.0% was found to also improve the inactivation performance of UV irradiation. For the scenario involving 10 min of trypsin treatment, the final log reductions of live cells at 20 mJ cm −2 were 1.25 (0.10%), 1.58 (0.25%), 1.70 (0.50%), and 2.28 (1.0%). The higher the trypsin concentration and the longer the treatment time, the more pronounced was found to be the improvement of inactivation. The highest inactivation efficiency of live oocysts in the present work was 2.86-log reduction, achieving a high inactivation rate of 99.86%.
Note that tailing phases (no further increase of inactivation at higher fluence) were observed in all scenarios as displayed in Fig. 3. This may be attributed to oocyst aggregation in solution (Medema et al. 2006), as particle agglomeration and precipitation could shield each other, or adsorb or scatter UV light, thus reducing the dose received by oocysts and thus the inactivation efficiency. As expected, there was a synergistic effect in sequential inactivation of oocysts with trypsin followed by UV irradiation. The combination of trypsin pretreatment and UV irradiation outperformed singular UV exposure by 1.74-3.98 times. We speculate that the trypsin displayed proteolytic activity and cleaved surface proteins when in contact with the oocysts (He and Buck 2010). The removal of protein building blocks consequently allowed the penetration of UV light into the oocysts, causing damage to DNA. This trypsinization process to solubilize wall proteins has been investigated by Iranzo et al. (2002), further evidenced by this study. Gabriel et al. (2019) reported that extreme UV irradiation of trypsin caused minimal impact on trypsin function, and that the activity and structure of trypsin were metastable. On the contrary, the activities of trypsin under different conditions of initial trypsin concentration declined obviously in our study (Fig. S1). This difference may be attributed to the different source and purity of the tested trypsins. In the present study, the final percentage activities of trypsin at 20 mJ·cm −2 were 5.03% (0.10%), 8.63% (0.25%), 11.15% (0.50%), and 18.82% (1.00%) (Fig. S1a), while the levels of activity were 50.27, 215.83, 557.46, and 1881.89 BAAE U/mL correspondingly (Fig. S1b). The final inactivation rates of Cryptosporidium parvum oocysts at 0.25%, 0.50%, and 1.00% were more than 99.80%, but showed no significant differences (p > 0.05) among these three groups (Fig. S1c). The final inactivation rate at 0.10% was less than 98% (Fig. S1c). We proposed that the minimum of initial trypsin concentration was 0.25% in this study, while the retained trypsin activity was only 215.83 BAAE U/mL. However, to our best knowledge, there has been no previous report that focused on the behaviors of trypsin in natural water environment. There is not enough clinical research data available to prove trypsin's safety or hazard for aquatic habitats if the trypsin is used in wastewater treatment and discharged into river, lake or pond, etc. It remains well worthy to be further studied the dose effect of trypsin on sensitive receptor, which will help to understand the possible eco-environmental impact of residual prypsin. Fig. 3 Inactivation of Cryptosporidium oocysts during 20-min UV irradiation under different trypsin pretreatment scenarios. The duration of trypsin pretreatment varied from 0 to 40 min, and the concentration of trypsin increased from 0.1 to 1.0%

Inactivation mechanisms indicated by leakage of intracellular proteins and damage to gDNA
SEM images indicated the near-spherical Cryptosporidium oocyst possessed a diameter of approximately 741 nm (Fig. 4a, untreated) and exhibited a damage-free and wellpreserved coat. Treatment by trypsin did not produce obvious changes in morphology, but surface wrinkling and pitting corrosion could be observed on the surface (Fig. 4a, 0.25% Trypsin, 20 min). In contrast, UV irradiation induced significant corrosive delamination and contraction of oocysts (Fig. 4a, UV irradiation, 20 min). Sequential treatment with trypsin (0.25%, 20 min) and UV irradiation (20 min) almost totally dissociated the oocysts, their coats were destroyed, and their shape was seriously deformed. Oocysts suffering this damage failed to maintain cellular integrity and their viability consequently declined. We also tried to use immunofluorescence assay to reveal the possible damage of oocysts after the treatment, but it is difficult to distinguish the debris from destroyed oocyst and impurities presented in solution in the fluorescein isothiocyanate (FITC) graphic. Nevertheless, the FITC method still helps identify the Cryptosporidium parvum oocysts. The stained oocysts with integrity show brilliant green fluorescence in the FITC graphic, round to ovoid (Fig. S2).
The changes of extracellular proteins, cell-free DNA, and gDNA before and after different treatments further verified the SEM analyses. Oocyst dissociations were observed to result in proteins release (i.e., glycoproteins and plasma membrane protein) (Fig. 4b). The concentration of extracellular proteins in systems involving UV alone, trypsin alone, and trypsin + UV treatments increased by 1. 34-5.46 Fig. 4 SEM images of the oocyst morphology (a), content of extracellular proteins in solution (b), gel electrophoresis of gDNA (c), and content of cell-free DNA (Ex-DNA) and gDNA before and after different treatments (d). The concentration of trypsin was 0.25%. The duration of all treatments was 20 min in a. "Trypsin + UV" denotes trypsin pretreatment for 20 min followed by UV irradiation for 10-30 min times compared with that in untreated oocysts. The leakage of genomic DNA was identical to the release of proteins (Fig. 4d). Combinations of trypsin and UV treatment clearly induced more severe release of proteins and leakage of DNA. The concentrations of free DNA in bulk solutions increased from 0.092 mg mL −1 for untreated oocysts to 0.432 mg mL −1 after UV treatment, 0.185 mg mL −1 for trypsin treatment, and 0.774 mg mL −1 from the trypsin + UV treatment.
In addition, gDNA was also damaged by UV irradiation (Fig. 4c), the predominant types including base modifications, strand breaks, and photoproducts (Jiang et al. 2007;Lai and Wang 2021), while strand breaks usually occur due to oxygen radical-mediated reactions, and base loss by strand scission (Hall et al. 2014). The single-or double-strand breaks observed in this study were quantified by separating DNA by gel electrophoresis and obtaining a quantitative image of the resulting distribution of DNA in the gel (Fig. 4c). Highly intact gDNA appears as a sharp and compact band (gel lane 1 in Fig. 4c), whereas degraded gDNA extends into lower molecular weight range (Permenter et al. 2015). Similar to the viability analysis, trypsin treatment alone did not alter the genome integrity (gel lanes 2-3 in Fig. 4c). In contrast, UV irradiation was noted to degrade genomic DNA into small fragments, the resulting DNA fragments visible as different bands (gel lanes 4-5 in Fig. 4c). When oocysts were subjected to both trypsin treatment and UV irradiation, the DNA was observed to become more highly fragmented (gel lanes 6-7 in Fig. 4c). As reaction times were prolonged, gDNA degradation could be observed to change from a laddering pattern to be a diffuse smear pattern, the latter indicating more significant degradation.

Conclusions
This paper has proposed a method involving trypsin pretreatment following UV irradiation for the effective inactivation of Cryptosporidium oocysts. The UV inactivation efficiency of oocysts with trypsin present was significantly higher than that in the system without trypsin. In the combined system, the trypsin firstly cleaved surface proteins in the oocyst wall, removed protein-building blocks, and then consequently allowed the penetration of UV light into the oocysts. The trypsinization-facilitated UV irradiation further induced DNA strand breaks and interfered with genome integrity. The underpinning mechanisms for this synergistic effect have been elucidated in depth with the support of enzyme activity assays, SEM analyses, flow cytometry, viability examination, leakage detection of cell proteins, and degradation of genomic DNA. The mechanistic investigations put forth in this study thus offer a new strategy for inactivation of Cryptosporidium in wastewaters.
Author contribution Dan Xaio and Wei Fan conceptualized, designed the experiments and wrote the manuscript. Dan Xaio, Nan Wang, Shiheng Chen, and Siyue Wang conducted the experiments and analyzed the data. Xiangyi Yuan and Mingxin Huo discussed the results and improved the manuscript. All the authors listed have made a substantial contribution to the work, and approved it for publication. Data availability The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval The author agrees with all ethical conditions given for publication in Environmental Science and Pollution Research.
Consent to participate All the authors have given consent to their contribution.