New Method Using Sedimentation and Immunomagnetic Separation for Isolation and Enumeration of Cryptosporidium parvum Oocysts and Giardia lamblia cysts

A new method for the isolation of Cryptosporidium parvum oocystsand Giardia lamblia cysts from biosolid samples has been developedthat utilizes sedimentation and immunomagnetic separation. Themethod was used to recover stained cysts and oocysts (spikeorganisms) from primary settled sewage sludge, anaerobicallydigested sewage sludge, and bovine manure. Recovery efficienciesassociated with this method were approximately 40 to 60% andwere significantly greater than those associated with similarmethods based on sucrose flotation (P < 0.001). The recoveryefficiency of the sedimentation-based method showed no significantreduction as a result of sample storage for up to 21 days (P> 0.05). Recovery efficiencies were determined by spikingsamples with prestained cysts and oocysts, allowing them tobe differentiated from those naturally present in the biosolidsamples. The prestained cysts and oocysts had been fixed in5% formalin, and the recovery efficiencies associated with thismethod may be different from recovery efficiencies for freshcysts or oocysts.

INTRODUCTION

Top

Introduction

Cryptosporidium parvum and Giardia lamblia are protozoan parasitescapable of infecting a variety of mammalian species, includinghumans (6). Consequently, these organisms can be found in bothsewage and animal manures (2). As biosolids are recycled toland, there are potential transmission routes for these organismsvia runoff and the contamination of water or foods cropped fromland. Consequently, it is important to quantify the risks ofcontamination associated with this practice, which in turn requiresaccurate methods to quantify the levels of C. parvum and G.lamblia in biosolids.

Several methods have been described to enumerate C. parvum andG. lamblia in various biosolids. These methods include NaClflotation and differential density centrifugation with Percolland sucrose (1, 4). Recovery efficiencies associated with thesemethods have been determined by a number of researchers, butthe results exhibit a considerable degree of variation. In addition,most studies involve the addition of C. parvum oocysts and G.lamblia cysts to samples in quantities that are several ordersof magnitude higher than background levels. This step may beundertaken to facilitate counting of the organisms (with a hemocytometer)or to prevent the levels of indigenous cysts and oocysts frominfluencing counts. Such high numbers of organisms influenceestimates of recovery efficiency since the number of organismsrecovered has been shown to vary with the quantity of organismsadded to the sample (4). Adding prestained organisms (spikeorganisms) at a level equal to or less than that normally observedin a sample may provide a more realistic assessment of a recoverytechnique's potential.

In this paper a method is described for isolating and enumeratingC. parvum oocysts and G. lamblia cysts in bovine feces and inboth raw and digested sewage sludge. The method combines a sedimentationstep with immunomagnetic separation followed by detection withepifluorescence microscopy. The method is compared with sucroseflotation, which was used in a recent survey of pathogens inbiosolids by UK Water Industry Research, Ltd. (3). The recoveryefficiency of the method as a function of storage time is alsoinvestigated.

MATERIALS AND METHODS

Top

Materials and Methods

Preparation of stock spike solutions.
Two stock spike solutions of approximately 10³ organisms/mlwere prepared, one containing C. parvum and the other containingG. lamblia. Cryptosporidium oocysts and Giardia cysts were obtainedfrom the University of Arizona. C. parvum oocysts (Harley Moonstrain, National Animal Disease Center, Ames, Iowa) were suppliedat a concentration of 10⁷ organisms ml^-1. They were producedin neonatal Holstein calves and purified from feces by usingdiscontinuous sucrose gradients. The Giardia cysts were suppliedat a concentration of 10⁶ organisms ml^-1; they had been shedfrom gerbils (Meriones unguiculatur) and purified from fecesby using discontinuous sucrose gradients. The C. parvum oocystsand G. lamblia cysts were nonviable and were preserved in 5%formalin. A total of 300 µl of the spike organisms wasmixed in a microcentrifuge tube with 1 ml of 0.003 M sulforhodamine101 acid chloride (Sigma) in methanol and incubated at 90°Cfor 60 min. The stained organisms were then centrifuged at 9,000x g for 3 min, and the supernatant was aspirated, leaving apellet of stained (oo)cysts and 100 µl of supernatantat the bottom of the tube. The organisms were then resuspendedin 1 ml of deionized water and centrifuged again at 1,300 rpmfor 3 min, and the supernatant was aspirated as described above.This wash step was repeated two more times. The stained organismswere enumerated with a hemocytometer, and their concentrationwas adjusted to 1,000 (oo)cysts ml^-1 via serial dilutions withdeionized water. For each organism, five replicates were analyzedto determine the exact concentration of organisms (see belowfor spike dose enumeration).

Spike dose enumeration.
Aliquots of 100 µl of the stock spike suspension werefiltered through 2-µm track-etched polyester membranes(Chemunex, Maisons-Alfort, France). Each membrane was wettedwith Tween 20 (0.1%, wt/vol) (Sigma, Poole, United Kingdom),placed on an aliquot (100 µl) of staining solution (fluoresceinisothiocyanate-labeled antibody with specificity for eitherCryptosporidium or Giardia) (TCS Microbiology, Botolph Claydon,Buckingham, United Kingdom) and incubated in a humidity chamber(for 1 h at 37°C). After staining, the membrane was transferredto a sintered glass support (Millipore, Billerica, Mass.), andthe excess antibody was drawn through the membrane with a -10-kPavacuum source. The membranes were placed onto glass microscopeslides and a drop of immunofluorescence assay mounting fluid(TCS) was placed on each. The membranes were then covered withglass coverslips, and the numbers of (oo)cysts on each membranewere determined by epifluorescence microscopy; cysts and oocystsfluoresced apple green when viewed under light with a wavelengthof 340 to 380 nm.

Biosolids.
The biosolids used in this experiment were primary settled sewagesludge, anaerobically digested sewage sludge, and bovine manure.The sludges were collected from a sewage treatment facilitythat served a population of 276,000 and that also received bothagricultural and industrial inputs. One-gram samples of thebiosolids were deposited in 50-ml plastic centrifuge tubes (Corning).Each of these biosolid samples was spiked with approximately10² (oo)cysts of both G. lamblia and C. parvum by adding 100-µlaliquots of each of the stock spike suspensions to the centrifugetubes containing the biosolids and vortexing for 5 s.

Isolation by sedimentation.
A 25-ml volume of 0.1 M phosphate-buffered saline (Sigma) wasadded to the centrifuge tube containing the spiked biosolidsample and vortexed for 60 s. Another 25 ml of phosphate-bufferedsaline was then added to the sample, and the tube was invertedfive times. The sample was left to stand for 60 min at roomtemperature, during which time the heavier particles of debrissettled at the bottom of the tube.

At the end of the 60-min period, a 10-ml disposable pipettewas used to transfer the top 45 ml of liquid to a clean 50-mlcentrifuge tube. During this step, care was taken not to disturbthe solids that had accumulated at the bottom of the tube. The45 ml of liquid was then brought to a final volume of 50 mlwith filtered deionized water and centrifuged for 5 min at 1,050x g (with no braking during the deceleration phase). The samplewas then further purified by immunomagnetic separation as describedbelow.

Isolation by sucrose flotation.
A 10-ml volume of 0.01% Tween 20 was added to the tube containingthe spiked biosolid and vortexed for 60 s. A 30-ml solutionof sucrose dissolved in deionized water to a density of 1.18g/ml was then deposited under the sample. This process was carriedout by injecting the sucrose into the bottom of the tube witha 50-ml syringe and a 15-cm 19-gauge steel syringe to avoiddisturbing the biosolid layer. The sample was then centrifugedat 1,050 x g for 10 min (with no braking during the decelerationphase). After centrifugation, the top 10 ml of the sample, theinterface between the layers, and the top 10 ml of sucrose werecarefully transferred with a 10-ml pipette to a new 50-ml centrifugetube. The contents of this tube were brought to a final volumeof 50 ml with deionized water, and the sample was then centrifugedonce more at 1,050 x g for 10 min. The sample was then purifiedby immunomagnetic separation as described below.

Purification by immunomagnetic separation.
Immediately after centrifugation, the top 45 ml of liquid wasremoved with a 10-ml pipette and discarded. Care was taken toensure that the pellet at the bottom of the tube was not disturbed.The sample was resuspended by vortexing for 10 s and then transferredwith a disposable 10-ml pipette to a Leighton tube (Dynal, Oslo,Norway) and brought to a final volume of 10 ml with filtereddeionized water.

The sample was further purified with a commercial kit for theimmunomagnetic separation of Giardia and Cryptosporidium (oo)cysts(Dynal) in accordance with the manufacturer's instructions.In brief, the samples were incubated with paramagnetic beadswhich were coated with antibodies raised against Cryptosporidiumor Giardia. The resulting complex of (oo)cysts and beads wasthen isolated from the sample with magnets and washed; the washingstep described in the instructions was repeated to further reducethe amount of particulate matter in the sample. The (oo)cystswere then dissociated from the beads, and the beads were removedfrom the purified sample with magnets.

After purification, the sample was transferred to a 12-mm-wellslide (Dynal) containing 10 µl of 1 M NaOH and air driedat 37°C. The sample was then fixed by applying a drop ofmethanol and allowed to air dry at room temperature. The slidewas stained with 50 µl of fluorescein isothiocyanate-labeledCryptosporidium and Giardia monoclonal antibodies (Waterbourne,Inc., New Orleans, La.). Cysts and oocysts in the purified samplewere identified by epifluorescence microscopy; these fluorescedapple green under light with a wavelength of 340 to 380 nm.Spike (oo)cysts were differentiated by checking for the inclusionof the sulforhodamine stain, which glowed red under light witha wavelength of 546 nm.

Comparison of sedimentation and sucrose flotation.
In order to compare the recovery efficiencies associated withboth sedimentation and sucrose flotation, five replicates ofeach of the spiked biosolid samples were processed by the twodifferent isolation methods. Samples were then further purifiedby immunomagnetic separation and enumerated by using epifluorescencemicroscopy. The numbers of spike (oo)cysts recovered from eachreplicate were recorded, and the mean recovery efficienciesfrom each method were compared by an unpaired t test.

Effect of storage time on recovery efficiency.
The effect of storage time on the recovery efficiency associatedwith sedimentation was investigated. Three replicates of thespiked biosolid samples of raw and digested sewage sludge werestored at 2 to 8°C for 0, 7, 14, and 21 days. These sampleswere then analyzed by sedimentation and immunomagnetic separation.The numbers of spike organisms recovered from each sample wererecorded, and a simple rectilinear regression was employed tocharacterize changes in the mean recovery efficiency from eachbiosolid as a function of storage time.

RESULTS

Top

Results

Spike dose enumeration.
Table 1 shows the number of organisms in 5,100-µl aliquotsof each of the stock suspensions. The mean concentration ofoocysts in the stock suspension of C. parvum was 1,408 oocystsml^-1, and the mean concentration of cysts in the stock suspensionof G. lamblia was 1,070 cysts ml^-1.

View this table:
[in this window]
[in a new window]
TABLE 1. Numbers of spike organisms detected in 100-µl replicates of each stock suspension

Comparison of sedimentation and sucrose flotation.
Figure 1 shows the recovery efficiencies associated with sedimentationand sucrose flotation for each combination of the biosolidsand organisms investigated. In each case the recovery efficienciesfor both isolation methods were compared by an unpaired t test.The results of this analysis are shown in Table 2.

View larger version (32K):
[in this window]
[in a new window]
FIG. 1. Mean recoveries associated with each isolation method. Error bars indicate standard deviations of 5 replicate samples.

View this table:
[in this window]
[in a new window]
TABLE 2. Comparison of recovery efficiencies associated with sedimentation and sucrose flotation

Effect of storage time on recovery efficiency.
Figure 2 shows recovery efficiencies after each storage periodfor each combination of the biosolids and organisms investigated.In each case the change in recovery efficiency as a functionof storage time was characterized by a simple rectilinear regression.The results of this analysis are shown in Table 3.

View larger version (36K):
[in this window]
[in a new window]
FIG. 2. Recovery efficiencies associated with sedimentation after storage of the spiked biosolid samples ar 2 to 8°C. Error bars indicate standard deviations of 3 replicate samples.

View this table:
[in this window]
[in a new window]
TABLE 3. Results of a regression analysis characterizing the recovery efficiency associated with sedimentation as a function of storage time (days)

DISCUSSION

Top

Discussion

Use of prestained organisms.
Prestaining Cryptosporidium and Giardia organisms with sulforhodamineallowed the enumeration of (oo)cysts at low levels [100 to 140(oo)cysts], and indigenous organisms did not influence estimatesof recovery efficiencies. In practice, both prestained organismscould readily be distinguished from those naturally presentin the biosolids. However, the prestained Giardia cysts fluorescedslightly less brightly, increasing the time needed to enumeratethem.

Comparison between sedimentation and sucrose flotation.
The recovery efficiencies associated with sedimentation werehigher than those associated with sucrose flotation for everycombination of biosolids and organisms investigated. Recoveriesassociated with sedimentation ranged from 38.9 to 58.5%, whereasthose associated with sucrose flotation were between 1.3 and5.4%. Evaluation of these differences with an unpaired t testindicated that they were significant in every case (P < 0.001).

The recovery efficiencies for sucrose flotation were somewhatlower than expected. In previous work conducted by this laboratory,recoveries of approximately 30 to 40% were obtained with sucroseflotation. However, previous recovery efficiencies were estimatedby determining the mean numbers of background cysts and oocystsand then deducting these values from the total counts. Becauseof the heterogeneity of biosolid samples and the small samplesizes (5 replicates), this method may have led to an overestimationof recovery efficiencies. The recovery efficiencies for sedimentationcompares favorably with the recovery efficiencies associatedwith other isolation methods. Kuczynska and Shelton (4) evaluatedeight methods for the detection of Cryptosporidium and foundthat the greatest recovery efficiency was 18.7% ± 5.9%.The higher recovery efficiencies associated with the methodpresented are possibly due to the smaller number of manipulationsteps involved. It is important to note, however, that the recoveryefficiencies associated with live cysts and oocysts or withorganisms that have been environmentally stressed may be differentthan recovery efficiencies associated with fixed cysts and oocysts.

Recovery efficiency of sedimentation as a function of storage time.
The sedimentation method is based in part on the differencein sedimentation rates of cysts and oocysts from rates of otherparticulates in the biosolids. Therefore, if prolonged contactbetween cysts and oocysts and the biosolids resulted in adherenceto particulates, then the recovery efficiency of the methodcould be reduced. To test this hypothesis, a simple rectilinearregression was used to investigate the effect of storage timeon the recovery efficiency of the sedimentation method. No significantassociation between these two variables was detected for anyof the combinations of organisms and biosolids investigated(P > 0.1).

The absence of any significant effect of storage time on recoveryefficiency may be because Cryptosporidium and Giardia did notreadily adhere to particles of biosolids. Alternatively, adherencemay have occurred rapidly, and so the effect of adhesion wasmanifested equally at all time points. This hypothesis is consistentwith the observations of Medema et al. (5) that the majorityof cysts and oocysts become attached to particulates withinthe first 5 h. It is also possible that storage time has aneffect on recovery efficiency but that the effect is small incomparison with other sources of variation affecting the methodand cannot, therefore, be detected statistically.

In conclusion, sedimentation in conjunction with immunomagneticseparation resulted in good (38.9 to 58.5%) recovery of Cryptosporidiumand Giardia. The use of prestained cysts and oocysts allowedfor accurate estimates of recovery efficiencies at spike concentrationsequal to or less than background levels. Recovery efficienciesremained stable across a range of biosolids and were not affectedby storage times of up to 21 days.

ACKNOWLEDGMENTS

I thank Thames Water Utilities for permission to present thispaper.

The views expressed are those of the author and not necessarilythose of Thames Water Utilities.

FOOTNOTES

* Mailing address: Thames Water Utilities, Spencer House, Manor Farm Rd., Reading, Berkshire RG2 0JN, United Kingdom. Phone: 0118 923 623 1. Fax: 0118 923 631 1. E-mail: jaime.massanet-nicolau{at}thameswater.co.uk.

REFERENCES

Top