Anticipating that the titer of scrapie infectivity in excreted urine would be low, we measured concentration by using limiting dilution titration, a method with which we have extensive experience quantitating TSE infectivity in blood and blood components. In a limiting dilution titration, all animals in the bioassay are inoculated with the highest concentration of inoculum that is tolerated by the intracranial (most efficient) route. Infectivity assorts randomly into the inoculated animals; provided that at least some, but not all, of the animals are infected, the concentration can be calculated from the Poisson distribution of the infections (1
). The method is highly sensitive and far more precise than other methods of TSE titration. We considered concentrating the urine before bioassay, but to circumvent uncertainties about the recovery of endogenous infectivity, we decided to inject the urine as collected.
We found TSE infectivity in the urine of hamsters that had no evidence of kidney or bladder inflammation. In contrast, Seeger et al. did not detect infectivity in the urine of scrapie-infected mice (11
) unless the mice were also affected by nephritis, in which case they found low levels of infectivity. Whether the bioassay they used was capable of detecting infectivity at the concentration we observed for hamsters is not clear. If it was not capable, then detection of infectivity in mice with nephritis implies a higher concentration of infectivity in urine excreted by a nephritic kidney. In another study, urine and feces from deer with chronic wasting disease failed to demonstrate infectivity when orally given to the same susceptible species (17
). Although usually an inefficient route of inoculation, the oral route did successfully transmit chronic wasting disease infectivity in saliva. The authors identified several possible reasons for the unsuccessful transmission by excreta, including incubation time, genotype, or sample size.
In our experiments, cross-contamination by feces can not be excluded as a source of infectivity. Although the metabolism cage effectively separated urine and feces, some contact is possible because of the anatomy of the hamster.
Protein misfolding cyclic amplification uses sonication to generate PrPres
and infectivity in vitro. Although we routinely disperse all samples by ultrasonication before injection, our conditions are much harsher than those used to generate PrPres
de novo (18
) and do not support protein misfolding cyclic amplification of PrPres
, or presumably infectivity (L. Gregori and R.G. Rohwer, unpub. data).
The kidney and bladder titers were far greater than expected compared with findings of historical studies in which, with only rare exceptions (19
), most attempts at transmission have been unsuccessful. These titers cannot be explained by the infectivity in residual blood (10 ID/mL) (1
). In addition, we observed PrPd
in the kidneys of scrapie-infected animals that had no indications of tissue inflammation. Heikenwalder et al. found PrPd
staining within follicular infiltrates only in kidneys of mice affected by nephritis and not in control mice with noncomplicated scrapie (12
). These data together with those by Seeger et al. (11
) suggested that renal inflammation might be a prerequisite for TSE infectivity in renal tissue and its excretion in urine. In contrast, our results indicate that renal inflammation is not necessary for the deposition of PrPd
in kidneys or for excretion of infectivity. One interpretation is that nephritis enhances the accumulation of PrPd
at sites of inflammation, consistent with the excretion of higher levels of infectivity inferred above for this same condition (11
Two studies of scrapie in naturally and experimentally infected sheep reported PrPd
depositions in the renal papillae (22
) and in the intraepithelial cortex, medulla, and papillae (23
). Similar to our findings, both studies indicated that not all scrapie tissues examined were positive for PrPd
. In chronic wasting disease, PrPd
staining was uniquely localized in the ectopic lymphoid follicle of the kidney of a whitetail deer (24
). All studies indicated either no changes (22
) or mild to no inflammatory changes of the kidney (23
). Thus, our histologic and immunohistochemical results for scrapie-infected hamsters are consistent with results found for sheep and deer and suggest that under normal conditions TSE diseases do not have concomitant inflammatory changes in the kidney.
That urine titer is similar to that of plasma suggests that urine infectivity may originate from blood (25
), but how the infectivity would be excreted is not clear. In general, proteins >40 kDa are not excreted and smaller proteins crossing the glomeruli are reabsorbed in the renal tubule and returned to the blood. If TSE infectivity is particulate (>40 kDa), its presence in urine might indicate abnormalities in renal filtration, perhaps related to the accumulation of PrPd
in the collecting tubules of the medulla. The accumulation of immunoglobulins in the urine of TSE-infected hamsters and humans may also indicate malfunction of the urinary system (9
). Excretion of a small C-terminal fragment of the normal cellular form of the prion protein in urine of infected and noninfected animals has been reported (27
), but PrPres
forms can only be inferred from the presence of infectivity. Nevertheless, excretion of proteins similar to PrPres
forms has been documented. Follicle-stimulating hormone is a glycosylated protein of 203 amino acids organized mostly as a β-sheet, which bears some remarkable similarities to β-rich forms of the prion protein. Follicle-stimulating and several similar hormones are excreted in urine at great enough concentration to be extracted commercially. Alternatively, TSE infectivity may be excreted by processes analogous to those responsible for the low-level virurias that occur during infections of the nervous system by mumps, measles, and West Nile virus (28
To the extent that results from the hamster model can be generalized to other TSE infections (and it has so far proven highly predictive), then even the very low concentrations of infectivity measured here could result in substantial environmental contamination. Several liters of urine and several thousand doses of TSE infectivity may be excreted daily over the course of the illness; even higher titers might be excreted by an animal with nephritis. The high stability of TSE infectivity would account for its persistence in pasture years after infected animals are removed (31
). Recent studies have shown that infectivity that is adsorbed and immobilized by soil minerals (32
) can still infect hamsters by oral exposure 29 months later (33
). Our study also warns of a possible risk from TSE contamination to fertility hormones and other medicinal products extracted from human urine.