Final Report: Improved Rapid Detection of Viable Waterborne Pathogens

EPA Contract Number: EPD06034
Title: Improved Rapid Detection of Viable Waterborne Pathogens
Investigators: Montagna, Richard A
Small Business: Innovative Biotechnologies International, Inc.
EPA Contact: Manager, SBIR Program
Phase: I
Project Period: March 1, 2006 through August 31, 2006
Project Amount: $69,984
RFA: Small Business Innovation Research (SBIR) - Phase I (2006) RFA Text |  Recipients Lists
Research Category: Drinking Water , Small Business Innovation Research (SBIR) , SBIR - Water and Wastewater


Project Description:

Phase I of this research project, conducted by Innovative Biotechnologies International, Inc., was designed to determine if an isothermic gene amplification system (Nucleic Acid Sequence Based Amplification [NASBA]) can be coupled with a simple, lateral, flow-based biosensor that uses liposomes to detect Cryptosporidium parvum and Cryptosporidium hominis in a sensitive and specific manner. In addition to demonstrating that the resulting CryptoDetect™ Test System can detect these species of Cryptosporidium, these studies also demonstrated that the system can detect as few as five oocysts and distinguish viable from non-viable oocysts. Furthermore, this effort also served as a feasibility study for final optimization and testing in “real world” samples anticipated in a subsequent Phase II effort.

Summary/Accomplishments (Outputs/Outcomes):

The initial Phase I effort focused on evaluating the following aspects of the CryptoDetect™ Test System:

  • Specificity of primer pairs and probes against closely related microbes.
  • Ability to detect both C. parvum and C. hominis.
  • Analytical sensitivity of the test in the presence of contaminating microorganisms.
  • Analytical sensitivity of the test using low numbers of flow cytometry sorted oocysts.
  • Ability to distinguish viable from heat-killed and ultraviolet (UV)-irradiated oocysts.
  • A head-to-head comparison of the test against U.S. Environmental Protection Agency (EPA) Method 1622.

Briefly, the principal of the test is that primer pairs were designed to use NASBA to amplify the heat-shock mRNA (coded by the hsp70 gene) and detect the resulting single-stranded amplicons by two probes. One “capture” probe was immobilized on a lateral flow test strip while a second “reporter” probe was attached to a liposome encapsulating a visually detectable dye. Due to the transparent nature of liposomes, formation of a hybridization complex consisting of the amplicons and “reporter” probe-conjugated liposomes, immobilized on the test strip by the “capture” probe, results in an easily visualized band (Figure 1). Using a simple, handheld reflectometer, the signal resulting from the band can be digitally quantitated and reported, thus eliminating any subjective interpretation of the results. The Phase I data support that the CryptoDetect™ Test System can detect as few as five viable oocysts of C. parvum and C. hominis. The signals generated by the detection of as few as 5 or 10 oocysts are crisp and easy to distinguish from negative samples (Figures 2 and 3). In addition, the NASBA primer pairs and probes designed to detect C. parvum and hominis also were able to detect C. felis and C. meleagridis, other Cryptosporidium species known to infect humans (Figure 4). Moreover, a minimum of five oocysts of each of these Cryptosporidium species is detectable. These data are consistent with our review of the literature as well as the BLAST analysis of the target sequences. Therefore, if any of the Cryptosporidium species that are known to infect humans are present in drinking water, the CryptoDetect™ Test System can accurately detect them.

Figure 1. Assay Format

Figure 2. Detection of Five Oocysts

Figure 3. Detection of 10 Oocysts

Figure 4. Dose-Response Curves for Human Pathogenic Cryptosporidium Species

Innovative Biotechnologies International also evaluated the ability of the CryptoDetect™ Test System to detect low numbers (i.e., 10 oocysts) of C. parvum oocysts in the presence of high concentrations (i.e., 50,000 organisms) of contaminating microorganisms. In all cases, each of the multiple replicates of the 10 oocysts were scored as “positive” in the presence of either E. coli O157:H7, Giardia intestinalis, or Oocystis minuta (Table 1).

Contaminating Organism

Number of Positive Samples

E. coli O157:H7

4 out of 4

Giardia intestinalis

4 out of 4

Oocystis minuta

4 out of 4


3 out of 4

Table 1. Detection of C. Parvum in the Presence of Contaminating Organisms

Our original Phase I application proposed evaluating the ability of the CryptoDetect™ Test System to distinguish between viable oocysts and those killed by either UV irradiation or boiling. To evaluate oocysts subjected to UV irradiation, eight samples each of 5, 50, 500, and 1,000 oocysts of C. parvum were irradiated with a dose of 10 mJ/cm2, using a collimated beam of UV by collaborators at Clancy Environmental Consultants, Inc. The samples then were shipped with cold packs and stored refrigerated for 48 hours until tested. Each group of eight samples then was tested along with a positive control (i.e., RNA isolated from 100 lysed C. parvum oocysts) and a negative control consisting of nuclease-free water.

Figure 5. Evaluation of UV-Treated Oocysts

Approximately 75 percent of the samples containing five oocysts that were UV-treated scored negative (Figure 5). The percentage of non-viable oocysts decreased, however, at each increasing oocyst level by approximately 25 percent, until all samples tested positive at 1,000 oocysts. Such unexpected results might have been caused by one of several factors. Most likely, at high oocyst doses, not all organisms were rendered nonviable by the UV treatment. In addition, while the dose employed has previously been shown to render the oocysts non-infectious, there are reports in the literature that suggest that UV irradiation can induce “stress proteins.” Because the oocysts were stored for more than 48 hours, it is not likely that such induced mRNAs would have survived long enough to be detected in our system. Although Innovative Biotechnologies International previously demonstrated that oocysts killed by boiling do, indeed, produce some heat shock mRNA during the early stages of the heat ramp up, such mRNA degrades within 48 hours and no longer is detectable in our system. Because the fundamental scientific questions raised by this outcome are beyond the scope of this Phase I study, Innovative Biotechnologies International will propose additional efforts during Phase II.

To verify, however, that the CryptoDetect™ Test System can, indeed, differentiate between viable and non-viable oocysts, Innovative Biotechnologies International subjected oocysts to boiling. Because our Phase I data support that as few as five oocysts (quantitated by flow cytometry) can be detected in our test system, Innovative Biotechnologies International evaluated 10 times that amount in the next set of experiments to assure that our system was being adequately challenged. Innovative Biotechnologies International prepared eight individual samples, each containing 50 viable oocysts of C. parvum in 100 μL. Four of those samples were placed in a boiling water bath for 15 minutes with vortexing every 5 minutes, while the remaining samples were left untreated. All of the treated and untreated samples then were allowed to incubate at room temperature for approximately 72 hours to assure the degradation of any hsp70 mRNA that might have been induced as a result of the early temperature ramp-up prior to cell death. All eight of the samples then were processed through the CryptoDetect™ Test System. All samples that were boiled (i.e., heat-killed, non-viable oocysts) were scored as negative, while all untreated, viable oocyst samples were scored as positive (Table 2). Therefore, unlike EPA Method 1622, the CryptoDetect™ Test System can discriminate between viable and non-viable C. parvum oocysts.


Number Tested

Number Positive

50 viable oocysts



50 boiled oocysts



Table 2. Detection of Viable C. Parvum Oocysts Versus Heat-Killed Oocysts

Close examination of all the test strips (Figure 6) used in this set of experiments reveals that no discernable bands were found in the capture zone of the negative controls or any of the boiled samples, while all viable oocysts yielded the expected strong positive signals.

Figure 6. Detection of Viable C Parvum Oocysts Versus Heat-Killed Oocysts

The final aspect of the Phase I effort was focused on comparing the performance of the CryptoDetect™ Test System with EPA to Method 1622. The C. parvum oocysts used in this portion of the study were from the Iowa isolate maintained at Waterborne, Inc., in New Orleans, Louisiana, and originally obtained from Harley Moon (National Animal Disease Center, Ames, IA) and provided to Innovative Biotechnologies International by Clancy Environmental Consultants, Inc. (CEC). The oocysts have been maintained by passage through experimentally infected mice, with fecal material collected and oocysts purified from contaminating debris by Percoll and sucrose density gradient centrifugation. Purified oocysts were stored by CEC in deionized water (DI) with penicillin, streptomycin, gentamicin, Amphotericin B and 0.01 percent Tween 20 at 3-5°C for up to 1 month post-shedding.

CEC prepared a working suspension of oocysts by diluting an aliquot of the stock suspension in sterile DI resulting in an approximate concentration of 25 oocysts in 100 μL. The actual concentration of the suspension then was determined using the 10 well count as described in EPA Method 1622/1623. For the presumptive 10 oocyst samples, the range of determinations was 4-11 with an average of 8 ± 2.6, while for the 25 oocyst samples, the range of determinations was 17-32 with an average of 24.8 ± 4.9. The spike suspensions then were used within 24 hours of enumeration.

The working suspension then was used to prepare a total of 12 individual spiked samples containing either a total of 10 or 25 oocysts within 10 ml volumes of DI water. Each of the replicates then were subjected to immunomagnetic separation as prescribed by EPA Method 1622 and then one-half of each sample group either continued through the standard Method 1622 procedure or subjected to the successive steps of the CryptoDetect™ Test System. For the limited number of samples that were investigated, very good results were obtained. As noted in Table 3, for samples evaluated from the immunomagnetic separation (IMS) through the CryptoDetect™ Test System, all three replicates of the 10 oocyst-spiked samples were scored as “positive,” whereas two of the three 25 oocyst-spiked samples were scored as positive. Table 4 reports the actual reflectometer readings of the individual strips evaluated using the CryptoDetect™ Test System. The background signal obtained in the portion of the test strip just upstream of the “capture” zone has been subtracted from the specific signal obtained in the “capture” zone to yield the final “Net Reading” reported.

Due to the experimental variation possible in the above approach (i.e., the actual number of oocysts introduced into the spiked samples is unknown and the percent recovery via IMS is not 100%), Innovative Biotechnologies International will suggest in its Phase II proposal to significantly increase the number of samples evaluated (the cost on a per sample basis prevented doing this in the current Phase I project) to obtain more statistically relevant data. Innovative Biotechnologies International also will suggest using numbers of oocysts counted by means of flow cytometry prior to IMS to avoid this additional point of variation.

Table 3. Comparision of CryptoDetect™ Test System with EPA Method 1622

Table 4. Evaluation of Spiked Samples Compared in Table 2


The Phase I data support the initial claims that the CryptoDetect™ Test System can detect low numbers of human pathogenic forms of Cryptosporidium species (especially parvum and hominis) and provide test results in an objective easy-to-interpret manner. Furthermore, the ability to distinguish viable from non-viable organisms will provide a valuable means to test drinking water for safety, especially in the case where oocysts that may have been rendered non-viable by effective means have not been eliminated successfully from the water. Method 1622 and Method 1623 are not able to distinguish viable from non-viable oocysts; the CryptoDetect™ Test System can make this distinction.

Supplemental Keywords:

Cryptosporidium parvum, Cryptosporidium hominis, NASBA, biosensor, liposomes, mRNA, heat-shock mRNA, hsp 70, viable oocysts, oocysts, gene amplification, safe drinking water, onsite testing, rapid testing,, RFA, Scientific Discipline, INTERNATIONAL COOPERATION, Water, Environmental Chemistry, Environmental Monitoring, Drinking Water, Environmental Engineering, cryptosporidium parvum oocysts, aquatic organisms, nanotechnology, waterborne pathogens, early warning, drinking water monitoring, nanosome method

SBIR Phase II:

Improved Rapid Detection of Viable Waterborne Pathogens  | Final Report