An Integrated Cryptosporidium Assay to Determine Oocyst Density, Infectivity

12 Cryptosporidium continues to be problematic for the water industry, with risk assessments 13 often indicating that treatment barriers may fail under extreme conditions. However, risk 14 analyses have historically used oocyst densities and not considered either oocyst infectivity or 15 species/genotype, which can result in an overestimation of risk if the oocysts are not human 16 infective. We describe an integrated assay for measuring oocyst density, infectivity and 17 genotype from a single sample concentrate, an important advance which overcomes the need 18 for processing multiple grab samples or splitting sample concentrates for separate analyses. 19 The assay incorporates an oocyst recovery control and is compatible with standard primary 20 concentration techniques. Oocysts were purified from primary concentrates using immuno21 magnetic separation prior to processing by an infectivity assay. Plate-based cell culture was 22 used to detect infectious foci, with a monolayer washing protocol developed to allow 23 recovery and enumeration of oocysts. A simple DNA extraction protocol was developed to 24 allow typing of any wells containing infectious Cryptosporidium. Water samples from a 25 variety of source water and wastewater matrices, including a semirural catchment, 26 wastewater, an aquifer recharge site and stormwater, were analyzed using the assay. Results 27 demonstrate that the assay can reliably determine oocyst densities, infectivity and genotype 28 from single grab samples for a variety of waters matrices and emphasize the varying nature of 29 Cryptosporidium risk extant throughout source waters and wastewaters. This assay should 30 therefore enable a more comprehensive understanding of Cryptosporidium risk for different 31 water sources, assisting in the selection of appropriate risk mitigation measures. 32 33 34 on Jne 6, 2017 by gest ht://aem .sm .rg/ D ow nladed fom 35 INTRODUCTION 36 Cryptosporidium is ubiquitous in source waters and wastewaters, presenting a treatment 37 challenge on account of its small size, resistance to chlorine disinfection and the absence of 38 more easily measured surrogates to allow treatment performance validation (1, 2). These 39 characteristics make Cryptosporidium problematic for the water industry and the ever-present 40 threat from this pathogen requires sound characterization and management of risks, including 41 validation and monitoring of critical control points (3-5). Removal of Cryptosporidium by 42 treatment processes can be highly variable (6), so validation and monitoring of individual 43 processes is important to ensure appropriate performance. Even for validated systems 44 treatment failure is possible under highly adverse conditions, and management of this risk 45 may require significant capital expenditure to provide sufficient risk mitigation for extreme 46 events. However, risk assessments frequently utilize historical monitoring data based on total 47 oocyst numbers, not considering either oocyst infectivity or species/genotype, resulting in 48 possible overestimation of risk. 49 50 Of the greater than 26 species or genotypes of Cryptosporidium that might be detected in the 51 environment, only C. parvum and C. hominis commonly infect humans (7). Furthermore, not 52 all oocysts excreted by an infected host are infectious and those that are infectious oocysts 53 can be rapidly inactivated by environmental conditions (8). Cryptosporidium species are 54 commonly monitored in waters using standard detection methodologies (e.g., USEPA 1622 55 and 1623). However, these methods provide no information on either the infectivity or 56 identity of the detected oocyst, providing little information on the relative health risk posed to 57 humans (9, 10). 58 59 on Jne 6, 2017 by gest ht://aem .sm .rg/ D ow nladed fom Recently, intensive efforts have been made to examine source waters destined for potable 60 consumption, as well as wastewaters intended for re-use, using genotyping assays and, to a 61 lesser extent, infectivity assays (9-13). The information from these assays provide an 62 indication of the risk to of human health (13). However, to better estimate risk there is a 63 requirement to understand oocyst density (number), infectivity and species (13). While a 64 number of assays are available to obtain such data, none are in a single assay format, meaning 65 that multiple grab samples must be processed to conduct the different analyses, increasing 66 analytical costs, or sample concentrates need to be split, compromising detection limits. The 67 heterogeneous distribution and low numbers of oocysts in surface waters also complicate the 68 use of multiple grab samples, where oocysts can be present in one sample and absent in the 69 next. These issues translate into higher costs to obtain data from all analyses, as well as 70 potential variability in data quality when using multiple grab samples. 71 72 To address a number of these shortcomings, Lalancette and colleagues developed a dual 73 direct detection assay to provide information on the fraction of infectious oocysts from a 74 water sample (10, 13). To achieve this, they combined a cell culture immunofluorescence 75 infectivity assay with a direct count of oocysts on the host cell monolayer as well as from 76 washes off of the monolayer using a filtration technique to capture oocysts. This prompted us 77 to adapt an optimized infectivity assay (14, 15) to not only include a total oocyst count for 78 calculating the percent of infectious oocysts, but to also incorporate recovery controls to 79 account for oocyst losses at different process steps. Furthermore, we developed an extraction 80 step to genotype Cryptosporidium lifecycle stages present in infectious foci and we assessed 81 the feasibility of genotyping oocysts remaining on monolayers that did not contain any 82 infectious foci. 83 84 on Jne 6, 2017 by gest ht://aem .sm .rg/ D ow nladed fom To determine the performance of our single format assay for measuring oocyst densities, 85 infectivity and genotype we applied it to water samples collected from: a series of rain events 86 in an Adelaide Hills semirural / agricultural catchment; various sampling points at three 87 wastewater treatment plants; an influent aquifer recharge site and; inlets for two stormwater 88 collection schemes. Here we present results from these activities demonstrating that this 89 single format assay is capable of determining oocyst densities, infectivity and genotype from 90 single grab samples across a variety of waters matrices, thereby providing a more 91 comprehensive understanding of Cryptosporidium risk. 92 93 MATERIALS AND METHODS 94 Single Format Assay Overview 95 Figure 1 presents an overview of the work flow for the single format assay to generate 96 information on oocyst densities, infectivity and genotype from water samples. The assay is 97 described as two discrete stages, Front-End Processing and Back-End Processing. Front-End 98 Processing involves primary concentration of oocysts from a water sample, followed by 99 secondary concentration and purification of oocysts by Immuno-Magnetic Separation (IMS). 100 The Back-End Processing, which is the novel part of the assay and focus of this paper, refers 101 to the elution of oocysts from the IMS beads and the subsequent steps required to determine 102 oocyst density, infectivity and genotype from a single sample. 103 104 Source of control Cryptosporidium oocysts 105 The C. parvum cattle isolate (Iowa strain) used in these studies was purchased from BTF 106 (Sydney, Australia). Oocysts were enumerated by dilution in phosphate buffered saline, 107 followed by staining with EasyStain, filtration onto 13 mm black polycarbonate membranes 108 (0.8 μm pore size) and enumeration using fluorescence microscopy (described in detail 109 on Jne 6, 2017 by gest ht://aem .sm .rg/ D ow nladed fom below). On receipt, the infectivity of each oocyst batch was determined using the focus 110 detection method (14, 15). All lots were stored at 4°C and used within 16 weeks of receipt. 111 ColorSeed (BTF, Sydney, Australia), which contains 100 Texas Red-labelled gamma 112 inactivated C. parvum oocysts, was used as an internal oocyst recovery control. 113 114 In vitro culturing of the HCT-8 cell line 115 Cells from the HCT-8 cell line (ATCC CCL-244; human ileocecal colorectal 116 adenocarcinoma) were maintained in 25 cm flasks with regular sub-culturing (3 times per 117 week) in RPMI 1640 growth medium with L-glutamine and supplemented with 15 mM 118 HEPES buffer, 100,000 U/L penicillin G, 0.1 g/L streptomycin and 10% fetal calf serum 119 adjusted to pH 7.4. HCT-8 cells were incubated at 37°C in a humidified CO2 (5% (vol/vol)) 120 incubator until confluent. All cell culture reagents were purchased from Sigma-Aldrich, 121 Sydney, Australia. Confluent monolayers were used to inoculate 48-well plates (Nunclon ∆ 122 Surface) at approximately 4×10 cells/well and grown until confluent over a 48 h incubation 123 period (humidified at 37°C with 5% (vol/vol) CO2) before infection. 124 125 Concentration and isolation of environmental oocysts from source waters and 126 wastewaters 127 All water samples were lodged with the Australian Water Quality Centre (a laboratory 128 accredited for Cryptosporidium analysis with the Australian National Association of Testing 129 Authorities) for Front-End Processing. Raw sewage samples were sub-sampled (250 mL – 13