Abstract
In this study we examined the recovery of Cryptosporidium parvum and Giardia duodenalis from matrices containing various concentrations of dissolved iron. The organisms were recovered by using the immunomagnetic separation-immunofluorescent assay method, and the levels of recovery were compared to the dissolved iron concentrations. The levels of recovery of C. parvum decreased sharply at dissolved iron concentrations greater than 4 mg/liter, while the levels of recovery of G. duodenalis decreased sharply at concentrations greater than 40 mg/liter.
On 16 December 1998, the Environmental Protection Agency (EPA) finalized the Interim Enhanced Surface Water Treatment Rule. One of the purposes of this rule is to improve the control of microbial pathogens, including Cryptosporidium spp., in drinking water (10). As watershed characterization and management become more important to source water protection policies, it must be realized that the ability to detect and enumerate pathogenic protozoans in increasingly complex matrices is a paramount concern of both regulators and stakeholders. The effects of common types of interference, as well as complex matrix remediation techniques, must be thoroughly investigated. Critical issues, such as point and nonpoint source loading, environmental fate and transport, and evaluation of agricultural best management practices, all rely on accurate detection and enumeration of the target organisms.
The currently accepted method for detection and enumeration of Cryptosporidium and Giardia species is the filtration-immunomagnetic separation (IMS)-immunofluorescence assay (FA) method outlined in EPA Method 1623 (11). While this method represents a significant improvement over the ICR technique that was used previously (6), it should be recognized that it was developed primarily for untreated surface water and finished drinking water. Other researchers have shown that as the sample matrix becomes more complex, the level of recovery of Cryptosporidium spp. begins to decline due to various types of interference (4, 9). Therefore, if EPA Method 1623 is to be used to enumerate Cryptosporidium and Giardia species in complex matrices, such as agriculturally contaminated surface water, acid mine drainage, and raw sewage, it will be necessary to document the effects of common types of interference on the process and to develop and refine the method and/or the sample matrix in order to overcome the effects of the interference.
This study was undertaken to determine what, if any, effect dissolved iron has on the IMS-FA procedure. Dissolved iron was selected as a possible primary type of interference based on the following conclusions.
First, iron is extremely common in the environment. It is the second most abundant metal found in the earth's crust and is present primarily as iron ores (8). Iron is also commonly found in surface water and groundwater, where it is present primarily as either the ferrous [Fe(II)] or ferric [Fe(III)] aqueous ionic species (8).
Second, iron is commonly found in association with sewage and in wastewater treatment plants. At one location, raw and conventional activated sludge treated wastewater has been found to contain average iron concentrations between 0.5 and 20 mg/liter, and the concentrations in residual materials (sludge, ash) are between 5,000 and 30,000 mg/kg (Allegheny County Sanitary Authority, unpublished data). Iron compounds are commonly used in wastewater treatment plants for clarifying processes and for precipitation of phosphates. Elevated iron levels would be expected in the effluents of such treatment plants (1, 5).
Third, the solution chemistry of iron is such that it causes direct difficulties with the IMS-FA method. Soluble Fe(III) yields an acidic reaction by hydrolysis: 4Fe3+ + 12H2O → 4Fe(OH)3 + 12H+ (http://www.dep.state.pa.us/dep/deputate /minres/bamr/amd/science_of_amd.htm). Since the immunomagnetic beads dissociate in an acid environment, we hypothesize that this reaction may interfere with the pH mechanisms of the IMS-FA procedure. In addition, iron in solution forms insoluble iron oxides and hydroxides that contribute to the overall turbidity of a matrix. It has been shown by other researchers that turbidity has an effect on the IMS-FA method (4).
Finally, other researchers have shown that there is a relationship between aqueous iron and biological surfaces (2, 12). Iron has been shown to interact with the amphoteric surface functional groups that are associated with the cell wall structural polymers of microorganisms (12). We hypothesize that this interaction may interfere with association of the immunomagnetic beads with the cyst surface or with the fluorescein isothiocyanate (FITC)-monoclonal antibody assay.
Matrix remediation through the use of EDTA (disodium salt) was also examined in this study. The chelating effects of EDTA for metals are well known, and many researchers have documented the chemistry of iron-EDTA complexes (3, 7, 13).
Experimental procedure.
IMS recovery was performed in triplicate in lab-prepared deionized (DI) water containing various concentrations of dissolved iron. Two separate trials were performed with the initial concentration range. Each matrix was evaluated to determine its turbidity before spiking. Slides were enumerated by an FA. The entire experiment was repeated with EDTA (disodium salt) added in order to examine the remediation effects of a known trace metal chelator. Finally, additional concentrations of dissolved iron were studied in an attempt to gather more data for dissolved iron concentrations ranging from 4 to 400 mg/liter.
Dissolved iron matrix.
A dissolved iron matrix was prepared in the laboratory by dissolving 1.45 g of anhydrous ferric chloride (catalog no. F-7134; Sigma) in 1 liter of lab-prepared pure DI water. The iron solution was passed through a sterile 0.45-μm-pore-size membrane filter (type HAWG047S1; Millipore). The filtrate was examined by using flame atomic absorption to determine the concentration of dissolved iron. The filtrate was also tested to determine its turbidity and pH. The stock dissolved iron matrix was serially diluted to provide the concentrations analyzed in this experiment.
IMS.
Ten milliliters of each matrix was added to four Leighton tubes. One tube was used to monitor the separation pH after buffer was added. The remaining three tubes of each matrix were spiked with ∼300 viable Cryptosporidium oocysts and ∼300 viable Giardia cysts. The tubes were then subjected to the IMS procedure described in EPA Method 1623 (11) by using a GC Combo kit (catalog no. 730-02; Dynal AS, Oslo, Norway).
Slide staining and enumeration.
Slides were prepared and stained by using the Merifluor Cryptosporidium-Giardia direct FA (catalog no. 250050; Meridan Diagnostics, Cincinnati, Ohio) and the general procedure outlined in EPA Method 1623 (11). Application of 4′,6-diamidino-2-phenylindole (DAPI) was omitted. All slides were enumerated within 30 h of preparation.
Oocysts and cyst preparation.
Viable Cryptosporidium parvum Iowa oocysts and viable Giardia duodenalis H3 cysts were obtained from Waterborne, Inc. (New Orleans, La.). The oocysts and cysts were shipped and stored in phosphate-buffered saline containing antibiotics at 2 to 8°C. The stock solutions were enumerated with a hemacytometer and were diluted to spike dose concentrations with sterile lab-prepared pure DI water. The oocysts and cysts that were utilized in the aged studies were stored at 2 to 8°C in phosphate-buffered saline containing antibiotics for approximately 90 days.
EDTA chelation.
Approximately 0.1 g of EDTA (disodium salt; catalog no. BP 120-1 Fisher) was added to each Leighton tube after the Cryptosporidium oocysts and Giardia cysts were added. The tubes were agitated occasionally and left to react for 0.5 h. Then SL buffers were added, and the IMS-FA was performed exactly as described above. The pH of each matrix was determined after buffer was added.
Table 1 shows the matrix characteristics that were examined in the initial part of this study.
TABLE 1.
Characteristics of lab-prepared high-iron-concentration matricies
| Dissolved Fe concn (mg/liter) | Turbidity (NTU)a | pH of matrix after SL buffer addition | pH of matrix and EDTA after SL buffer addition |
|---|---|---|---|
| 0 | <1.0 | 7 | 7 |
| 0.4 | 1 | 7 | 7 |
| 4 | 4 | 7 | 7 |
| 40 | 25 | 7 | 7 |
| 400 | 120 | 5 | 5 |
NTU, nephelometric turbidity units.
Tables 2 and 3 show the levels of recovery of C. parvum and G. duodenalis, respectively, based on the four trials completed in this study.
TABLE 2.
Recovery of Cryptosporidium sp. from high-iron-concentration matrices
| Matrix iron concn (mg/liter) | Trial 1 (broad range, iron)
|
Trial 2 (broad range, aged)
|
Trial 3 (EDTA remediation)
|
Trial 4 (narrow range, iron)
|
||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| % Recovery | SD (%) | n | % Recovery | SD (%) | n | % Recovery | SD (%) | n | % Recovery | SD (%) | n | |
| 0 | 58.0 | 6.2 | 3 | 58.1 | 2 | 32.7 | 25.6 | 3 | ||||
| 0.4 | 54.7 | 6.3 | 3 | 52.1 | 14.1 | 3 | NDa | ND | ND | |||
| 4 | 51.0 | 3.7 | 3 | 20.6 | 18.1 | 3 | 29.5 | 4.3 | 3 | |||
| 20 | 3.2 | 2.0 | 3 | |||||||||
| 40 | 7.5 | 4.5 | 3 | 5.0 | 5.4 | 3 | 19.1 | 2.9 | 3 | |||
| 50 | 4.3 | 4.6 | 3 | |||||||||
| 100 | 0.2 | 0.2 | 3 | |||||||||
| 200 | 0.0 | 0.0 | 3 | |||||||||
| 300 | 0.0 | 0.0 | 3 | |||||||||
| 400 | 0.0 | 0.0 | 3 | 0.0 | 0.0 | 3 | 0.1 | 1.7 | 3 | |||
ND, not determined.
TABLE 3.
Recovery of Giardia sp. from high-iron-concentration matrices
| Matrix iron concn (mg/liter) | Trial 1 (broad range, iron)
|
Trial 2 (broad range, aged)
|
Trial 3 (EDTA remediation)
|
Trial 4 (narrow range, iron)
|
||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| % Recovery | SD (%) | n | % Recovery | SD (%) | n | % Recovery | SD (%) | n | % Recovery | SD (%) | n | |
| 0 | 74.0 | 6.7 | 3 | 26.0 | 3.7 | 3 | 29.0 | 23.5 | 3 | |||
| 0.4 | 71.2 | 10.8 | 3 | 25.2 | 10.5 | 3 | NDa | ND | ND | |||
| 4 | 68.9 | 13.6 | 3 | 22.9 | 7.7 | 3 | 6.2 | 5.3 | 3 | |||
| 20 | 46.6 | 8.3 | 3 | |||||||||
| 40 | 72.9 | 7.5 | 3 | 12.7 | 7.0 | 3 | 38.1 | 5.9 | 3 | |||
| 50 | 54.0 | 11.7 | 3 | |||||||||
| 100 | 52.1 | 8.8 | 3 | |||||||||
| 200 | 8.7 | 3.0 | 3 | |||||||||
| 300 | 5.7 | 3.1 | 3 | |||||||||
| 400 | 0.1 | 0.2 | 3 | 0.0 | 0.0 | 3 | 4.3 | 4.6 | 3 | |||
ND, not determined.
Several other noteworthy observations were made. Since pH is an important controlling factor in the association and dissociation of the magnetic beads, the pH of each matrix was determined after the two SL buffers from the Dynal kit were added (Table 1). Remediation with EDTA had no effect on the initial pH established by the buffers.
As determined by microscopic examination, the fresh Giardia cysts that were recovered from the higher-iron-concentration matrices exhibited distinct internal staining that ranged from bright red to a dull brick red and incomplete and/or faint FITC staining of the cyst wall. This staining pattern was not observed with fresh Giardia cysts recovered from distilled water or the low-iron-concentration matrices, which exhibited very faint or no red internal staining and bright, complete, apple green staining of the cyst wall. The Giardia cysts recovered from the EDTA-remediated iron matrices exhibited bright, complete, apple green staining of the cyst wall at all iron concentrations. Fresh and aged Cryptosporidium oocysts and aged Giardia cysts exhibited complete, apple green staining of the oocyst or cyst wall under all conditions.
In this study, we attempted to quantify the effects of dissolved iron on the IMS-FA results in terms of reduced level of recovery. We also examined the remediation effects of EDTA (disodium salt) on the matrix in terms of improved level of recovery compared to the untreated matrix, as well as any effects attributable to organism age.
As Tables 2 and 3 show, there appeared to be a threshold iron concentration above which the IMS-FA method was not able to enumerate the target organisms. For Cryptosporidium sp., the average level of recovery decreased from 58 to 0% over the range of dissolved iron concentrations in both trial 1 (fresh organisms) and trial 2 (aged organisms). For Giardia sp., the average levels of recovery decreased from 74 to 0% in trial 1 (fresh organisms) and from 26 to 0% in trial 2 (aged organisms). Cryptosporidium sp. appeared to exhibit greatly reduced levels of recovery at iron concentrations lower than the concentrations to obtain reduced levels of recovery of Giardia sp. This tendency could have been due to Giardia's larger cysts or to differences in cell wall constituents. It is noteworthy that the level of recovery of Cryptosporidium sp. decreased sharply in both trials before the matrix pH effect became significant (Table 1). This finding supports the theory that interference due to dissolved iron is due to more than a simple pH effect. Our microscopic examination of the incomplete FITC staining of fresh Giardia cysts also suggested that activity between dissolved iron and the cyst or oocyst surface is a significant source of interference.
A comparison of the data obtained in trials 1 and 2 shows that the levels of recovery for aged Cryptosporidium oocysts, as determined by the IMS-FA method, were very similar to those for fresh Cryptosporidium oocysts. This supports the findings of previous researchers who showed that the levels of recovery of oocysts determined by the IMS-FA method do not depend on oocyst age (4). There was, however, a great difference between the levels of recovery of fresh Giardia cysts and aged Giardia cysts as determined by IMS, and the levels of recovery of fresh Giardia cysts were as much as 50% greater. It was noted during both stock enumeration and FA microscopy that aged Giardia cysts tended to clump together much more than fresh Giardia cysts clumped together. We hypothesized that this clumping could lead to difficulties in stock enumeration and spike dose solution preparation.
A comparison of the levels of recovery in trials 1 and 3, both of which were performed with fresh organisms, shows that addition of EDTA (disodium salt) prior to the IMS-FA did not improve recovery of the target organisms. In fact, EDTA tended to have an inhibitory effect at low iron concentrations, resulting in lower levels of recovery, especially for Giardia sp. We also noted that addition of EDTA resulted in data that were much more variable than the data obtained in the experiment performed without EDTA. Microscopic examination of the Giardia cysts recovered from the EDTA-remediated high-iron-concentration matrices did not reveal the characteristic incomplete staining pattern observed in the earlier experiment. This suggests that EDTA successfully inhibited the surface interactions between the dissolved iron and the target organisms.
The recovery experiments in trial 4 were conducted to provide more information for dissolved iron concentrations ranging from 4 to 400 mg/liter. The levels of Cryptosporidium recovery declined sharply at concentrations between 4 and 20 mg/liter, while the levels of Giardia recovery declined sharply at concentrations between 100 and 200 mg/liter.
In conclusion, the data presented above show that high concentrations of dissolved iron have an inhibitory effect on the IMS-FA portion of EPA Method 1623. Whether this effect is a result of pH, surface interactions, magnetic activity, or a combination of these factors needs to be investigated further. We also found that EDTA (disodium salt) is not an effective remediation agent for lab-prepared high-iron-concentration matrices and probably has pronounced inhibitory effects of its own in low-iron-concentration matrices. Future efforts will be undertaken to evaluate other conventional means of matrix remediation.
Finally, additional recovery efforts suggested that the interference threshold concentration of dissolved iron is between 4 and 20 mg/liter for Cryptosporidium sp. and between 100 and 200 mg/liter for Giardia sp.; these findings demonstrate that Cryptosporidium oocysts are about 10 times more sensitive to dissolved iron than Giardia cysts.
Acknowledgments
This study was supported by the Allegheny County Sanitary Authority.
We thank Lisa Williams for performing the atomic absorption analysis, Sandra Brandon for preparing the manuscript, and Stanley States of the Pittsburgh Water and Sewer Authority for editorial assistance.
REFERENCES
- 1.Baker M, Blowes D W, Ptacek C J. Laboratory development of permeable reactive mixtures for the removal of phosphorus from onsite wastewater disposal systems. Environ Sci Technol. 1998;32:2308–2316. [Google Scholar]
- 2.Brezonik P L, King S O, Mach C E. The influence of water chemistry on trace metal bioavailability and toxicity to aquatic organisms. In: Newman M C, McIntosh A W, editors. Metal ecotoxicology: concepts and applications. Chelsea, Mich: Lewis Publishing; 1991. pp. 1–31. [Google Scholar]
- 3.Bryce A L, Kornicker W A, Elzerman A W. Nickel adsorption to hydrous ferric oxide in the presence of EDTA: effects of component addition sequence. Environ Sci Technol. 1994;28:2353–2359. doi: 10.1021/es00062a020. [DOI] [PubMed] [Google Scholar]
- 4.Bukhari Z, McCuin R M, Fricker C R, Clancy J L. Immunomagnetic separation of Cryptosporidium parvum from source water samples of various turbidities. Appl Environ Microbiol. 1998;64:4495–4499. doi: 10.1128/aem.64.11.4495-4499.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Caccavo F, Jr, Frolund B, Kloeke F V O, Nielsen P H. Deflocculation of activated sludge by the dissimilatory Fe(III)-reducing bacterium Shewanella alga Br Y. Appl Environ Microbiol. 1996;62:1487–1490. doi: 10.1128/aem.62.4.1487-1490.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Clancy J L, Fricker C R, Telliard W. Proceedings of the AWWA Water Quality Technology Conference, Denver, Colo. Denver, Colo: AWWA; 1997. New USEPA standard method for Cryptosporidium analysis in water (CD-ROM) [Google Scholar]
- 7.MacPhail A P, Patel R C, Bothwell T H, Lamparelli R D. EDTA and the absorption of iron from food. Am J Clin Nutr. 1994;59:644–648. doi: 10.1093/ajcn/59.3.644. [DOI] [PubMed] [Google Scholar]
- 8.Nebergall W H, Holtzclaw H F, Jr, Robinson W R. General chemistry. 6th ed. Lexington, Mass: Heath Publishing; 1980. [Google Scholar]
- 9.Rochelle P A, Deleon R, Johnson A, Stewart M H, Wolfe R L. Evaluation of immunomagnetic separation for recovery of infectious Cryptosporidium parvum oocysts from environmental samples. Appl Environ Microbiol. 1999;65:841–845. doi: 10.1128/aem.65.2.841-845.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.United States Environmental Protection Agency. Overland migration of Cryptosporidium oocysts. Washington, D.C.: United States Environmental Protection Agency; 1999. [Google Scholar]
- 11.United States Environmental Protection Agency. Method 1623: Cryptosporidium and Giardia in water by filtration/IMS/FA. Washington, D.C.: United States Environmental Protection Agency; 1999. [Google Scholar]
- 12.Warren L A, Ferris F G. Continuum between sorption and precipitation of Fe(III) on microbial surfaces. Environ Sci Technol. 1998;32:2331–2337. [Google Scholar]
- 13.Xue H, Sigg L, Kari F G. Speciation of EDTA in natural waters: exchange kinetics of Fe-EDTA in river water. Environ Sci Technol. 1995;29:59–68. doi: 10.1021/es00001a007. [DOI] [PubMed] [Google Scholar]