The Minimum Infective Dose of Cryptosporidium in Gnotobiotic Lambs.
Report No DWI0338

1991

SUMMARY

In the two gnotobiotic lamb trials, L1 and L2, undertaken, all the lambs fed on Cryptosporidium contaminated water, whether at 50 or 5 oocysts per litre, contracted cryptosporidiosis and exhibited typical clinical symptoms. Mean oocyst outputs for each group of lambs were approximately 10^10. These mean outputs, and the shedding profiles exhibited by the two groups, approximate closely to the outputs and shedding profiles one would associate with experimental lambs infections initiated with 10^6 Cryptosporidium oocysts (4).

The lambs fed with water containing 50 oocysts per litre or 5 oocysts per litre exhibited a slightly longer prepatent period (5-6 days) than has been observed in lambs dosed with 10^6 oocysts (3-4 days); this may be a reflection of the time required for an infection initiated with a very few oocysts to multiply to the level at which oocysts are detectable in the faeces.

Given the probable normal distribution of oocysts in the water, it is likely that some feeds may have contained less oocysts than the mean while others may have contained more oocysts than the mean.

The mean cumulative oocyst intakes estimated for the two groups of lambs were 64 and 10 (allowing for a 3 day prepatent period) yet the observed prepatent periods were very similar, 95% of the lambs becoming patently infected by day 6. The near synchronous onset of oocyst shedding in both groups suggests an infection established at the first feed, rather than one which has been acquired by accumulation of oocysts to a threshold level.

In trial L2, an estimated initial intake of 2-5 oocysts induced such a rapid acquisition of infection that it must be concluded that the minimum infective dose for gnotobiotic lambs is one oocyst.

The validation data for trials L1 and L2 have highlighted the problems of detecting Cryptosporidium oocysts in raw and potable waters. Nominal contamination of the 2,000 litre tank for trial T2 was 5 oocysts per litre. Membrane filtration and in situ staining with a monclonal antibody kit produced a mean recovery of 3.11 oocysts per litre.

Replicate filters taken and processed by the SCA "Blue Book" method gave results ranging from 0 - 0.24 oocysts per litre - this "water supply" had demonstrably infected 10 out of 10 gnotobiotic lambs.

These two experiments have validated the lamb infection model as the most sensitive controlled system available for detecting minimal levels of infectivity (as distinct from numbers of oocysts) in potable water. It is now possible to design studies to assess the efficacy of treatments intended to eliminate infectivity at the very low levels of contamination likely to be found in treated waters. Previously, experimental designs have required very high oocyst counts to permit detection of changes in oocyst numbers or viability. The accuracy and sensitivity of these detection techniques are not sufficient to monitor the persistence of infectivity from high initial levels down to levels of below one viable oocyst per litre.

Copies of this report may be available as an Acrobat pdf download under the 'Pre 2000 Reports' heading on the DWI website.