Comparison of four commercial immunomagnetic separation kits for the detection of Cryptosporidium

Enteric protozoa known for their infectivity, Cryptosporidium spp., are parasitic pathogens that can be transmitted in humans via the faecal–oral route including from contaminated water sources and food production (Ryan et al. 2021; Costa et al. 2022). First reported to cause gastrointestinal infection in humans in 1976 (Nime et al. 1976), Cryptosporidium has now been considered a leading cause of diarrhoeal illnesses (cryptosporidiosis) to both young and immunocompromised individuals globally (O'Leary et al. 2021). Approximately 90% of cryptosporidiosis cases in humans are caused by C. hominis and C. parvum (Hassan et al. 2021). Cryptosporidium has a low infectious dose, and the environmentally robust oocysts are resistant to chlorine disinfectants, including those used in water treatment plants (Temesgen et al. 2021). Additionally, it has been identified that Cryptosporidium oocysts are able to survive for months in soil and water and are able to pass through filters for water treatment due to their small size of ∼4 to 5 μm (Andreoli & Sabogal-Paz 2019). Monitoring water sources for the prevalence of Cryptosporidium and setting up surveillance systems are important to validate the efficiency of water treatment; however, there are challenges with the existing analytical methods.

Currently, the gold standard for the detection of both Cryptosporidium and Giardia spp. in water is a three-step procedure outlined by the U.S. Environmental Protection Agency in Method 1623.1 (Hassan et al. 2021). The three main steps are (1) the concentration of the sample using filtration and centrifugation, (2) the separation of the sample using immunomagnetic separation (IMS), and (3) the enumeration of oocysts using immunofluorescent staining and microscopic analysis.

The IMS procedure involves adding antibody-coated magnetic beads to the concentrated sample and incubating to allow the beads to bind to the oocysts. The process is carried out in borosilicate glass culture tubes called Leighton tubes with flat sides and adjoining magnetic clips. In these Leighton tubes, the IMS buffers are incubated with the magnetic beads and the water sample of interest. The bead–oocyst complexes are magnetically separated from the solution, then the attached oocysts are dissociated from the beads via heat or 0.1 M hydrochloric acid treatment (Andreoli & Sabogal-Paz 2021).

Several different anti-Cryptosporidium IMS bead kits are commercially available. These include Dynabeads™ Cryptosporidium Kit (IDEXX) (‘Dynabeads’), Crypto-Grab™ Cryptosporidium oocyst Isolation kit from Waterborne, Inc. (‘Waterborne beads’), Isolate^®Cryptosporidium IMS kit from TCS BioSciences (‘TCS beads’), and CryptoBeads Cryptosporidium IMS beads from BioPoint Pty. Ltd (Sydney, Australia) (‘BioPoint beads’).

The aim of this study is to assess and compare the performance of anti-Cryptosporidium IMS kits currently available in the market, which are tested against different sample matrices. These will be carried out by analysing the recovery efficiencies of spiked oocysts as well as assessing the purity and ease of reading the slides during microscopy.

Phosphate-buffered saline (PBS), bovine serum albumin (BSA), anti-Mouse IgG (whole molecule)-fluorescein isothiocyanate (FITC) antibody produced in goat (anti-mouse IgG FITC), and anti-mouse IgM (μ-chain-specific)-FITC antibody produced in goat (anti-mouse IgM FITC) were purchased from Sigma Aldrich. BD TruCount™ Absolute Counting Tubes In Vitro Diagnostic (IVD) were purchased from BD BioSciences. For the IMS kits, CryptoBead Cryptosporidium IMS beads were supplied by BioPoint Pty. Ltd, Crypto-Grab™ Cryptosporidium oocyst Isolation kits were purchased from Waterborne, Inc., Dynabeads™ Cryptosporidium kits were purchased from IDEXX, and Isolate^®Cryptosporidium IMS kits were purchased from TCS BioSciences.

The magnetic beads were briefly described by their respective manufacturers as follows: BioPoint beads are coated with a purified anti-Cryptosporidium oocyst monoclonal antibody; Dynabeads are coated with purified antibodies against Cryptosporidium oocysts covalently bound to the surface; TCS beads are conjugated with oocyst wall-specific anti-Cryptosporidium mouse monoclonal IgG; and Waterborne beads are coated with the manufacturer's own mouse monoclonal antibodies prepared specifically against antigens located on the outer surface of oocysts. The accuracy of Dynabeads was stated to be determined by seeding experiments and the recovery efficiency might be affected by factors of the water sample concentrate seeded, such as relative turbidity. Apart from these, no additional information was available regarding the specific properties of the beads as well as on cross-reactivity, specificity, and sensitivity of the IMS kits.

Different sample matrices were analysed in IMS. Reagent-grade water was obtained from Livingstone (Sydney, Australia), while Cockle Creek Sediment is a river water sediment collected in Ku-ring-gai National Park, Australia. The clay sample was prepared from Fuller's Earth Clay purchased from N-essentials Pty. Ltd, and the quality control (QC) mud was a filter backwash sample from a water treatment plant in Sydney, Australia. Samples from the Cockle Creek Sediment, Fuller's Earth Clay, and QC mud were washed and resuspended in distilled water prior to use. Preliminary screening for the presence of indigenous oocysts was conducted on these sample matrices using the Dynabeads GC-Combo kit and ColorSeed™.

For all recovery experiments conducted, ColorSeed^® flow cytometry dispensed control samples were used (BioPoint, Pty. Ltd). These were suspensions that contained precisely enumerated oocysts, i.e., 100 or 1,000, which have been labelled with Texas Red and can be confirmed with fluorescence microscopy (Warnecke et al. 2003). ColorSeed Tween Rinse (BioPoint Pty. Ltd) was used to rinse ColorSeed tubes to ensure all oocysts are transferred to the tubes and to prevent the aggregation of oocysts when seeding.

Immunofluorescence staining was carried out using EasyStain™ kits provided by BioPoint Pty. Ltd and the Aqua-Glo™ G/C Direct Comprehensive Kit purchased from Waterborne, Inc.

For each type of magnetic beads, a 2 μL aliquot was suspended in a 100 μL anti-mouse IgG whole molecule or IgM μ-chain-specific FITC prepared to 1:32 working concentration with 1% BSA in PBS, vortexed thoroughly, and incubated for 30 min at room temperature. After staining, 200 μL of 1% BSA in PBS was added and samples were kept on ice until analysis. Flow cytometric acquisition was performed on a BD FACSCalibur and data analyses with BD CellQuest™ (BD BioSciences).

For fluorescence detection, the samples were initially identified using the forward scatter (FSC) and side scatter (SSC) detectors and then FITC fluorescence was measured using the FL-1 detector. TruCount™ tubes were used to determine the number of magnetic particles per millilitre (mL) of bead suspension into each tube. About 995 μL of 1% BSA in PBS and 5 μL of the magnetic particles were added and vortexed thoroughly.

The efficiency of the IMS beads to capture oocysts was assessed in three replicates for each kit as follows. In 1.5 mL microcentrifuge tubes, 100 μL of Buffer A (10×) and 100 μL of Buffer B respective for each kit were added. One vial of customised ColorSeed, given at 0.9 mL volumes, with 1,000 Cryptosporidium oocysts was vortexed and added to the microcentrifuge tube, and the ColorSeed tube was rinsed with ColorSeed Tween Rinse and the washings were added to the microcentrifuge tube. To each tube, 10 μL of the IMS beads were added. The samples were incubated with gentle rotation for 5, 15, and 30 min. The tubes were then attached to Isolate Micro-clip (ZAICS5, TCS BioSciences) magnetic clips and incubated for another 5 min. With the bead pellets captured on the magnets, the supernatants were pipetted onto membrane filters and stained as indicated below.

IMS was initially performed on reagent-grade water samples following the manufacturer's instructions respective to each brand. Similar sample preparation steps were instructed for BioPoint beads, Dynabeads, and TCS beads where 1 mL each of Buffer A (10×) and Buffer B were dispensed into tubes. Waterborne beads required the addition of Grab Buffer A (2×) only. One vial of ColorSeed was added to the tube and rinsed with ColorSeed Tween Rinse. Reagent-grade water was added to the tube, followed by 100 μL of IMS beads. The tubes were incubated at room temperature with gentle rotation as prescribed (a minimum of 60 min for all three kits and a minimum of 15 min for BioPoint beads). The tubes were then attached to magnets for magnetic capture of the bead–oocyst complexes. After tipping off the supernatants, the beads were rinsed out of the tubes and transferred into 1.5 mL microcentrifuge tubes for at least one washing step with Grab Buffer B for Waterborne beads and with the diluted Buffer A (1×) for the other three kits. To dissociate the oocysts from the beads, 50 μL of 0.1 M HCl was added to the microcentrifuge tubes. The tubes were vortex-mixed and then incubated at room temperature for 10 min before magnetic separation. The supernatant was collected and pipetted onto membrane filters as indicated in the Post IMS Analyses section.

IMS was similarly performed on the three sample matrices: Cockle Creek Sediment sample, Fuller's Earth Clay sample, and QC mud sample, each tested at 0.1 mL packed pellet volume. All the experiments were done in three replicates. As a contamination test, negative control assays were performed where the relevant IMS procedure for each respective kit was tested against distilled water without any oocyst spike. All control assays showed negative results for the presence of Cryptosporidium oocysts.

Immunofluorescence analyses were performed using Aqua-Glo™ kits and EasyStain™ kits to stain the oocysts. Isopore polycarbonate (PC) membrane filters of 13 mm diameter (Merck Millipore) with 0.8 μm pore size were placed onto 13 mm Swinnex filter holders set up on a manifold to allow drainage by vacuum pressure. The samples were pipetted onto the membrane and drained by vacuum. The membranes were stained with one drop of the fluorescent antibody for 15 min (EasyStain™) or 25 min (Aqua-Glo™), rinsed with Wash Buffer, and then mounted on glass slides for microscopy analyses.

Membranes were analysed using epifluorescence microscopy with an Olympus^® BX51 microscope. FITC staining was detected using a FITC filter (excitation, 480 nm; emission, 520 nm), and a Texas Red filter (excitation, 560 nm; emission, 630 nm) was used to confirm the target oocysts.

Equation (1) represents the recovery efficiency formula.

The statistical analysis was performed using Microsoft^® Excel^® for Microsoft 365 MSO (Version 2306 Build 16.0.16529.20100). Significance was determined using one-way analysis of variance (ANOVA) where a value of p<0.05 was considered statistically significant. This was followed by a pairwise Tukey–Kramer post hoc comparison to determine which IMS beads have recoveries that are significantly different.

Flow cytometry analyses of the IMS beads based on the FSC and SSC parameters are indicative of the particle size and refractive properties (Figure 1). BioPoint and TCS beads have higher FSC and SSC values compared to the other two IMS beads, which indicates a larger bead size. The lowest FSC and SSC values are observed in the Waterborne beads, showing that these beads are the smallest in size. Dynabeads exhibit the tightest population among all samples, denoting that these particles are highly uniform in size (Šafařík & Šafaříková 1999). The TCS Isolate beads and the Waterborne beads showed a significant number of aggregates representing 51% (TCS) and 77% (Waterborne) of the total particle numbers detected.

There was a large degree of variation between the different bead types in reactivity with the IgG whole molecule antibody. The BioPoint beads appeared densely coated with IgG antibody. These beads are, in fact, coated with an IgG1 antibody (Weir et al. 2000). The Dynabeads appear to be coated with an IgM antibody at a low level, while the TCS beads appear to be coated with a medium level of IgG antibody. The Waterborne beads appear to be coated with a very low level of IgM antibody.

The BioPoint beads showed faster capture efficiency over the shorter time periods than the other beads. The BioPoint, Dynabeads, and TCS beads all achieved a high capture efficiency at the 30-min time point at >94%, while Waterborne beads at 41.4% failed to capture most of the oocysts.

The Dynabeads, TCS, and BioPoint beads all performed well against all sample types. Two sets of analyses with reagent-grade water were carried out to compare the differences between staining with Aqua-Glo™ and EasyStain™ where a similar trend in the beads' results was observed. The Dynabeads showed a lower recovery with the Fuller's Earth Clay sample. The BioPoint beads performed the best across all four sample matrices with the recovery efficiencies ranging from 75 to 93%. In contrast, Waterborne beads had the lowest recovery efficiencies ranging from 0 to 8%.

The efficiency of the IMS process is influenced by several factors, including the rate and specificity at which the beads bind to the oocysts, the number of contaminating particles present in the sample; the number of magnetic beads used; and the incubation time allowed for the capture of the target organism (Wang et al. 2020). Additional operation parameters that affect the recovery efficiency are the magnetic capture strength and the dissociation step, wherein magnetic beads are separated from the target oocysts with heat or acid treatment (Robertson & Gjerde 2001; Andreoli & Sabogal-Paz 2021).

The tendency of an antibody to form a complex with an antigen pertains to its avidity, which is defined as the equilibrium association constant by convention (Greenspan 2023). Reciprocal to this measure of avidity is the equilibrium dissociation constant, determined by the ratio of the antibody dissociation rate or off-rate to the association rate or on-rate, which measures the velocity of how quickly the antibody dissociates or binds to the target, respectively (Greenspan 2023). The specificity, avidity, and the on-rate of the antibody used will influence the performance of the IMS beads as well as the amount of antibody coated onto the beads. The most widely available Cryptosporidium-specific monoclonal antibodies are IgM or IgG3 antibodies (Smith & Nichols 2010). Both IgM and IgG3 monoclonal antibodies are often ‘sticky’ and have slower on-rates and lower avidity than IgG1 monoclonal antibodies, causing the non-specific binding of this antibody to algal and mineral particles when testing water samples (Vesey et al. 1997; Weir et al. 2000).

Flow cytometry analysis of the different beads after staining with secondary antibodies shows that Dynabeads and Waterborne beads are coated with IgM antibodies, whereas BioPoint beads and TCS Isolate Cryptosporidium both have IgG antibodies. These results are congruent with previously published studies where it was identified that the Dynabeads have an IgM antibody and the TCS Isolate beads have an IgG3 antibody (Smith & Nichols 2010). BioPoint beads have an IgG1 antibody coated onto the magnetic particles, producing beads that have a higher affinity to the surface antigens of Cryptosporidium oocysts in water samples. The BioPoint beads showed a high capture efficiency even with a short incubation time of 5 min. This is probably due to the combined effect of the superior IgG1 antibody and the high concentration of antibody on the beads.

As determined through flow cytometry analyses, Dynabeads, TCS, and BioPoint beads were relatively similar in size and were supplied at similar bead concentrations. The Waterborne beads were found to be much smaller and were supplied at almost a 10-fold higher concentration. The amount of antibodies on the beads varied significantly. There was 10-fold more antibody on the BioPoint beads than the Dynabeads and double the amount of the TCS beads. The efficiency of immunocapture and magnetic separation is also affected by the particle size, where the large surface area of smaller beads should provide more antibody binding sites and allow more particles to interact with target organisms (Wang et al. 2020). This was, however, not observed with the Waterborne beads, where despite the high particle concentration there were very low levels of antibody detected on the beads, possibly causing the poor performance of these beads.

The four IMS kits utilised in this comparative study each have their specific manufacturer's instructions for use. Simply put, the whole IMS process can be divided into four sections: (1) mixing the sample with buffers and beads, (2) incubation to allow magnetic beads to bind to oocysts, (3) washing the sample from the tube to a microcentrifuge tube, and (4) dissociation of the oocysts from the beads, prior to staining and microscopic analyses. The main difference in the methods among the IMS kits was during the incubation and capture of beads. Dynabeads, TCS, and Waterborne beads all recommend a minimum of 1 h incubation with rotation at 15–20 rpm at room temperature, while BioPoint beads require a minimum of 15 min. The BioPoint protocol recommends performing the IMS capture in a 15 mL plastic centrifuge tube, whereas the other beads kits all recommend using a glass Leighton tube. Additionally, both Dynabeads and Waterborne protocols recommend the use of the Dynal^® MPC^®-1/MPC^®-6 and Dynal^® MPC^®-S magnets (Dynal) for the separation and doing the captures manually by rocking the tube by hand end-to-end at ∼90° for 2 min with ∼1 tilt/s. BioPoint beads and TCS beads protocols, on the other hand, use an automated capture where magnetic tube holders are rotated at 20 rpm for 5 min and then the supernatant is immediately tipped off. All four kits have the same dissociation process wherein 50 μL of 0.1 M HCl was added to the bead–oocyst pellet and magnetic separation was carried out following vortex mixing and sample incubation. This procedure during the IMS process is a crucial step as having oocysts remaining adherent to the beads following dissociation has become the main cause of reduced oocyst recovery (de Oliveira Pinto et al. 2016). It is ideal that a complete dissociation of the bead–oocyst complex occurs to obtain optimal recovery efficiency.

Resulting recovery efficiencies are highly influenced by both the methodology for IMS and the particularities inherent to the sample matrix of interest. There is then a need to evaluate IMS in different matrices such as those from natural sources like rivers and lakes as the efficiency of the U.S. Environmental Protection Agency (EPA) Method 1623.1 has been optimised for drinking water samples but not for complex matrices (Giglio & Sabogal-Paz 2018; Andreoli & Sabogal-Paz 2021). The selection of river sediment, clay samples, and filter backwash samples as the matrices of interest for this study is based on the complexity of these samples and the likelihood of such being analysed in water testing laboratories. Out of the three matrices, the Cockle Creek Sediment appeared to contain some magnetic debris. It is possible that this matrix had dissolved iron, the most common type of interference from both surface water and groundwater, in the form of either ferrous or ferric aqueous ionic species (Yakub & Stadterman-Knauer 2000). In solution, insoluble iron oxides and hydroxides are formed and further influence the overall turbidity of the matrix. (Yakub & Stadterman-Knauer 2000).

Interference in the separation of oocysts can be influenced by the turbidity of both organic and inorganic debris present in the sample such as clays and coagulants arising from the source waters or from additional treatment processes prior to IMS (U.S. EPA 2012). The Dynabeads showed variable recoveries with the Fuller's Earth Clay matrix. Previous studies have reported problems using Dynabeads with clay samples (McElroy et al. 2001).

Additionally, it must be noted that there is still a research gap in detecting oocyst present in filter backwash samples from high turbidity water samples (Ogura & Sabogal-Paz 2022). In water treatment plants, although most Cryptosporidium oocysts are removed during the settling and filtration stages, the small size and compressibility of the oocysts enable some to pass through (Ogura & Sabogal-Paz 2022). Similarly, recirculating the filter backwash water reintroduces the oocysts in the treatment process. Relatively good recoveries at ∼80% from BioPoint, Dynabeads, and TCS Isolate beads were still achieved despite the presence of particles that may adhere to oocysts, as commonly seen in filter backwash water samples (Ogura & Sabogal-Paz 2022).

After IMS processing, the oocysts dissociated from immunomagnetic beads are visualised under fluorescence microscopy. Manual counting of cells under fluorescence microscopy requires trained personnel and can be time-consuming, labour-intensive, and prone to errors (Aldughayfiq et al. 2023). The use of immunofluorescent staining enables the marked distinction of Cryptosporidium oocysts from the non-fluorescent background and makes the microscopic investigation easier (O'Leary et al. 2021). However, debris from IMS processing is often auto-fluorescent and can obscure the target cells, as shown in the Fuller's Earth Clay samples. Out of the four IMS kits tested, the resulting slides from the BioPoint beads had the least background and debris. The quality of the processed sample is most likely influenced by the type of antibody present on the IMS beads, in which the identified IgG1-coated beads are less prone to non-specific binding, compared to IgM and IgG3 antibodies. If a large amount of debris is present, then it is difficult to visualise oocysts and can reduce recovery efficiencies.

The presence of various materials such as sediments, dissolved metals, and particulates in sample matrices can hinder achieving optimal recovery efficiency in IMS. Conducting IMS against the different sample matrices tested the robustness of the different IMS bead kits and exhibited how both the method and the sample matrix affect the resulting slides and oocyst recoveries post-IMS processing.

Two of the IMS bead kits, Dynabeads™ Cryptosporidium and Waterborne Crypto-Grab™, have both been found to be coated with IgM antibodies. BioPoint beads and TCS Isolate beads are both coated with IgG antibodies, resulting in beads with higher affinity to oocyst samples. Among the four IMS bead kits, the BioPoint beads were the best performing when tested across different matrices both in terms of recovery efficiency and the quality of membranes following IMS. Fluorescence microscopy visualisation and enumeration of the recovered oocysts are influenced by both the method and the sample matrix processed. As there is a lack of documented protocols on how complex matrices should be analysed, it is highly recommended that further research and extensive testing be carried out to standardise and tailor these methods for various environmental water samples.

All relevant data are included in the paper or its Supplementary Information.

J.C.A. was employed by BioPoint Pty. Ltd, which has provided their products for this study. The outcomes of this research may lead to the development of products that may be licenced to the same company.