Final Report: FDP -- Development of Detection and Viability Methods for Waterborne Microsporidia

EPA Grant Number: R828041
Title: FDP -- Development of Detection and Viability Methods for Waterborne Microsporidia
Investigators: Hoffman, Rebecca M. , Borchardt, Mark , Marshall, Marilyn M. , Sonzogni, William C.
Institution: University of Wisconsin - Madison , University of Arizona
EPA Project Officer: Nolt-Helms, Cynthia
Project Period: August 1, 2000 through August 3, 2001
Project Amount: $375,037
RFA: Drinking Water (1999) RFA Text |  Recipients Lists
Research Category: Drinking Water , Water

Objective:

Microsporidia are a group of obligate intracellular pathogens capable of initiating disease in a plethora of vertebrate and invertebrate hosts. The transmissible form of these protozoa is a 2 µm, environmentally resilient spore. The presence of these organisms in surface water has been documented. The overall objective of this research project was to develop a strategy for the recovery and identification of human pathogenic microsporidia from natural waters. The specific objectives of this research project were to: (1) generate purified spores for development of analytic methods; (2) develop/optimize an efficient sample collection method; (3) sample concentration/purification by flow cytometry; (4) conduct diagnostic assay/viability testing; and (5) validate a finished method in natural waters.

Summary/Accomplishments (Outputs/Outcomes):

Identification of Variation Within Cell Culture Propagated Spore Suspensions

Encephalitozoon cuniculi (ATCC 50502), E. hellem, (ATCC 50451), and a duodenal isolate of E. intestinalis (ATCC 50603) spores were propagated and purified using published methods. Briefly, these methods included spore propagation in a rabbit kidney cell line and purification using Percoll density gradient separation. Flow cytometric analyses of purified spore suspensions produced over a 1-year period showed that freshly harvested suspensions produced using these commonly cited spore propagation methods were comprised of two subpopulations differing in physical size, permeability to vital dyes, antibody staining, and infectivity in cell culture assays. These findings were published by the research team in the August 2003 issue of Applied and Environmental Microbiology. Flow cytometry with cell sorting was used to prepare suspensions containing the larger, infectious spore type. These spores were used for subsequent method development.

Development of Polyclonal Antibody to Encephalitozoon and Characterization of an Existing Antimicrosporidia Antibody

Several antibodies were evaluated for use in concentration/purification methods. Monoclonal antibody 7G7 (Ted Nash, National Institutes of Health) exhibited a variable staining pattern, brightly staining the small spore subpopulation but only dimly staining the larger, infectious spore subpopulation. Rabbit polyclonal antibody produced against a cocktail of formalin-treated Encephalitozoon cuniculi, hellem, and intestinalis spores stained only E. cuniculi. An additional rabbit polyclonal generated against ultraviolet-treated Encephalitozoon spores was highly reactive with all three Encephalitozoon species and was used for the methods development work below.

Evaluation of Flow Cytometry With Cell Sorting (FCCS) for Isolation of Microsporidia From Reagent Water and Water Concentrates Prepared Using National Institute of Standards and Technology (NIST) Turbidity Standards

Flow cytometer prepared standards containing either 100 spores prelabeled with fluorescent polyclonal antibody or 100 unlabeled spores were prepared and the microsporidia sorted from nontarget particles using FCCS. Recovery of prelabeled microsporidia averaged 93.1 percent (sd=4.2, n=10) and recoveries of spores labeled with fluorescein isothiocyanate (FITC)-conjugated polyclonal antibody following spore standard preparation averaged 82.2 percent (sd=3.3, n=10). Recovery of spores spiked into an artificial turbidity matrix ranged from 63.9 to 79.4 percent (mean 69.5 percent, sd=6.4, n=5).

Evaluation of Continuous Flow Centrifugation for Concentration of Microsporidia Spores From Water Samples

The continuous flow centrifuge evaluated in this project was a modified blood aphersis unit produced by the Baxter Corporation (Round Lake, IL). This instrument, the Amicus Separator, is a channel-type continuous flow centrifuge that concentrates particles in a flexible plastic belt. Standards of precisely enumerated, fluorescently labeled E. intestinalis spores were added to 10 L filtered tap water, natural water samples, or 2 Nephelometric Turbidity Unit (NTU) water artificially created using NIST Tennessee River sediment Standard Reference Materials (SRM) 8406. Following sample concentration, particulates were eluted from the belt and the eluate concentrated using multiple, high-speed centrifugations. Sample concentrates were transferred directly onto well slides and examined microscopically. Recovery efficiencies are listed in Table 1.

 

Table 1. Spore Recovery From 10 L Spiked Reagent Water Using Chlorofluorocarbon (CFC) Concentration and Well Slide Analysis

Spore Conc. (L-1)
n
Mean (%)
SD (%)
Range (%)
10
14
61.5
12.2
38.7-75.5
100
10
63.8
8.9
52.0-75.5

Determination of Parasite Recovery Using CFC Combined With FCCS

FCCS purification of spores from CFC concentrates was evaluated for use when the increased pellet volume associated with more turbid samples interfered with direct well slide analysis. Standards containing either 10, 100, or 1,000 E. intestinalis spores prelabeled with FITC-conjugated polyclonal antibody were prepared using flow cytometry. Spores were spiked into 10 L filtered tap water and concentrated as described above. Fluorescent particles in the CFC concentrates were flow-sorted onto well slides and confirmed as E. intestinalis spores using epifluorescence microscopy. Recovery efficiency using this method averaged 44 percent (sd=11.7, n=3) at the 1 spore/L level, 63 percent (sd=9.9, n=10) at 10 spores/L, and 45 percent (sd=12.2, n=10) at 100 spores/L (see Table 2). Although FCCS effectively isolated spores from water, microscopic confirmation of spores remained necessary. The research team believed the difficulty and uncertainty associated with microscopic confirmation of these 2 µm protozoa precluded the use of FCCS as a detection method.

 

Table 2. Spore Recovery From 10 L Spiked Reagent Water Using CFC Concentration and Flow Cytometry With Cell Sorting

Spore conc. (L-1)
n
Mean (%)
SD (%)
Range (%)
1
3
44
11.7
31-51
10
10
63
9.9
43-77
100
10
45
12.2
30-62

Development of Immunomagnetic Separation (IMS) Methods for Purification of E. Intestinalis Spores From Water

The spore capture efficiency of eight IMS constructs was evaluated. The highest recoveries were achieved using indirect IMS methods, where water concentrates were labeled with polyclonal anti-microsporidia antibody and the antibody-labeled spores were purified using IMS particles coated with anti-rabbit IgG. Two IMS products, Dynabeads® (2.5 µm) and Captivate™ ferrofluid (200 nm), successfully purified spores from reagent water. The Dynal product recovered an average of 90.2 percent (sd=7.3, n=5) of the 1,000 spores seeded into reagent water, while the Captivate™ product recovery of spores seeded into reagent grade water averaged 79 percent at both the 500 (n=5) and 1,000 (n=5) seed level (sd=3.3 and 6.2, respectively). The Captivate™ product also was tested using dilute pond water and recovered 67 percent (sd=4.1, n=5) of the 500 spores seeded into the sample. Retrieval of the smaller Captivate™ magnetic particles from samples with higher turbidity was not successful. The authors speculate that particulate matter from the sample concentrate coats the magnetic particles and reduces their attraction to the magnet. Consequently, further IMS studies were performed using the Dynal IMS product. Additional evaluation of the IMS method using quantitative, real-time polymerase chain reaction (PCR) to assess relative recovery of spores in natural water samples showed that IMS using polyclonal antibodies may not offer the specificity needed for reliable detection of these protozoa. In reagent grade water samples, relative recovery of microsporidia spores (defined as the quantitative PCR value of the pre-IMS seed sample divided by the value of the post-IMS seed sample multiplied by 100) averaged 46.1 percent (sd=9.1). The IMS method failed to identify microsporidial nucleic acid in seven natural water concentrates seeded with 500 spores; however, internal positive controls identified inhibitors in all samples and therefore the perceived loss may not be completely attributed to lost spores. The research team suggests that future work developing monoclonal antibodies to microsporidia for use with IMS methods is necessary for advancement of microsporial detection methods.

Development of PCR Detection of E. Intestinalis Seeded in Natural Water Samples

PCR detection of seeded E. intestinalis spores using 16S rRNA primers was performed. Although detection of 25 spores in reagent grade water was achieved in some experiments, detection of even 100 spores seeded into natural water samples was variable. Three different commercially available DNA extraction kits were evaluated; however, the test results appear to be influenced by both the extraction method and water matrix. Most DNA extraction kits have been designed for use with clinical samples that tend to be more homogeneous in nature. Matrix composition and pH must be considered carefully when selecting DNA extraction methods if molecular analyses are to be successful.

Evaluation of Reverse Transcriptase PCR Cell Culture (RT-PCRCC) Assay as a Means to Assess Spore Viability

RT-PCRCC assays were performed on E. intestinalis spores using primers designed to the 16S rRNA, hsp70, or beta-tubulin genes to determine if viable spores could be differentiated from nonviable spores. In addition, the assays were performed with E. cuniculi and E. hellem spores to test whether these primer pairs could be used to detect multiple Encephalitozoon species. Microsporidial spores were heat inactivated and viability was determined by both RT-PCRCC assays and standard in vitro cell culture coverslip assays. PCR assays using primers designed to 16S rRNA performed on E. intestinalis, E. cuniculi, and E. hellem spores were able to amplify all three species but did not differentiate viable and nonviable spores. Hsp 70 gene amplification compared favorably with standard cell culture infectivity data and was able to differentiate between live and dead E. cuniculi and E. hellem spores. These primers, however, were not able to differentiate live and heat-killed E. intestinalis spores. Assays performed using beta-tubulin primers correlated with standard cell culture infectivity data and were able to differentiate between live and dead E. intestinalis and E. hellem spores, but failed to differentiate live and heat-killed E. cuniculi spores. Thus, to use a RT-PCR assay for water samples or viability testing, it may be necessary to use a nested primer approach, a multiplex approach, or develop additional primer sets that would identify all three species with one assay.

Evaluation of Quantitative, Real-Time PCR for Detection of Microsporidia in Water Samples

During Year 3 of the project, through a partnership with Dr. Donna Wolk at the Southern Arizona Veterans Administration (VA) Health Care System in Tucson, AZ, real-time PCR detection methods were applied to this research project. Spore standards were prepared using FCCS in either reagent water or in post-IMS water concentrates. Nucleic acid was extracted using the MagNA Pure LC Extraction system and subjected to real-time PCR on the ABI 7900HT using primers directed to the 16S rRNA gene. Quantitative standards were prepared for each real-time PCR run using 10, 50, 100, and 500 spores/sample. Data obtained from plotting the PCR cycle thresholds versus the corresponding quantitative standard concentrations were used to calculate a standard curve from which the unknown samples could be quantified. The assay is a multiplex reaction that includes an extraneous target (an internal positive control), which is monitored to assess the presence or absence of PCR inhibitors in water matrices. Microsporidia were detected at low levels in both reagent water and in post-IMS water concentrates (see Table 3).

 

Table 3. Detection of Seeded Microsporidia Using Real-Time PCR

Water Type
Spore Level
# Detects/Total # Tested
Reagent water
1
4/10
 
3
7/10
 
5
10/10
Artificial turbidity1
1
2/3
 
10
2/3
 
50
3/3
 
100
3/3
Marston River2
10
10/10
 
50
10/10
 
100
10/10
Poudre River2
10
10/10
 
50
10/10
 
100
10/10
1 2 NTU samples created using Tennessee River sediment.
2 Marston and Poudre River samples not tested at the one spore level.

Recovery Determination of the CFC-IMS-Quantitative PCR Method Using Seeded Natural Water Samples

Ten source waters (20 L each) were collected and shipped on ice to the Marshfield Clinic Research Foundation. Samples were split into 2 to 10 L aliquots. One 10 L sample was spiked with E. intestinalis spores prior to concentration by CFC as described previously, while the remaining 10 L was processed as an unspiked background control. Concentrated samples were shipped to the Wisconsin State Laboratory of Hygiene, where IMS was performed using the Dynal product. IMS beads were transferred to screw cap vials and shipped to the Southern Arizona VA Health Care System laboratory in Tucson, AZ, for quantitative, real-time PCR analysis. Although relative recovery percentages were quite low (0.01to 0.11 percent), 6 of the 10 samples tested positive at 50 spores/L (see Table 4).

 

Table 4. Recovery of Seeded Microsporidia From Natural Water Samples Using the CFC-IMS-Quantitative PCR Method

Water Source1 Municipality
Turbidity (NTU)
pH
Relative % Recovery
Lake Superior Duluth, MN
0.1
7.70
0.01
Lake Champlain Burlington, VT
0.2
7.05
ND
Bay of Green Bay Marinette, WI
0.3
7.95
0.11
Lake Michigan Kenosha, WI
0.5
7.90
0.07
Lake Michigan Manitowoc, WI
0.5
7.94
0.03
St. Clair River Algonac, MI
2.5
7.83
ND
Lake Winnebago Oshkosh, WI
5.8
7.85
0.02
Lake Winnebago Menasha, WI
8.6
8.00
0.04
Mississippi River Fort Madison, IA
9.4
8.06
ND
Ohio River Cincinnati, OH
28.9
7.58
ND
1Inhibitors to PCR were detected in all samples.
ND = not detected.

Although CFC and real-time PCR have been shown to effectively concentrate and detect microsporidia, respectively, the reliability of the IMS portion of the method is questionable. The research team believes that IMS recovery could be significantly improved if a new, highly specific monoclonal antibody were developed and incorporated into the assay. Without the ability to purify microsporidia from sample concentrates, method development efforts will stall, making it impossible to study the fate, ecology, and distribution of these pathogens in the environment and in drinking water.

The automated, real-time PCR assay reported here successfully identified low levels of microsporidia in several natural water samples. Like most other environmental molecular-based assays, however, nucleic acid amplification in other natural water samples was subject to inhibition. Further research should focus on identification of PCR inhibitors and the development of strategies to overcome them.


Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other project views: All 14 publications 3 publications in selected types All 3 journal articles
Type Citation Project Document Sources
Journal Article Hoffman RM, Marshall MM, Polchert DM, Jost BH. Identification and characterization of two subpopulations of Encephalitozoon intestinalis. Applied and Environmental Microbiology 2003;69(8):4966-4970. R828041 (Final)
  • Abstract from PubMed
  • Full-text: AEM Full Text
    Exit
  • Other: AEM PDF
    Exit
  • Journal Article Hoffman RM, Wolk DM, Spencer SK, Borchardt MA. Development of a method for the detection of waterborne microsporidia. Journal of Microbiological Methods 2007;70(2):312-318. R828041 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct
    Exit
  • Other: Science Direct PDF
    Exit
  • Journal Article Lambertini E, Spencer SK, Bertz PD, Loge FJ, Kieke BA, Borchardt MA. Concentration of enteroviruses, adenoviruses, and noroviruses from drinking water by use of glass wool filters. Applied and Environmental Microbiology 2008;74(10):2990-2996. R828041 (Final)
    R831630 (2007)
    R831630 (2008)
  • Abstract from PubMed
  • Full-text: AEM
    Exit
  • Other: AEM PDF
    Exit
  • Supplemental Keywords:

    microsporidia, encephalitozoon, protozoa, flow cytometry, continuous flow centrifugation, water, emerging pathogens, enterocytozoon, analytical chemistry, drinking water, environmental chemistry, environmental microbiology, health risk assessment, complex mixtures, contaminant candidate list, CCL, drinking water system, drinking water contaminants, exposure, microbial risk management, microbiological organisms, molecular biotechnologies, monitoring, natural waters, oligoprobe, parasites, pathogens, treatment, viability methods, water treatment., RFA, Scientific Discipline, Water, Environmental Chemistry, Health Risk Assessment, Analytical Chemistry, Environmental Microbiology, Environmental Monitoring, Drinking Water, complex mixtures, natural waters, pathogens, molecular biotechnologies, monitoring, encephalitozoon, microbiological organisms, waterborne disease, exposure and effects, exposure, viability methods, treatment, microbial risk management, oligoprobe, parasites, drinking water contaminants, microsporidia, water treatment, contaminant candidate list

    Progress and Final Reports:

    Original Abstract
  • 2001 Progress Report