Prion in Saliva of Bovine Spongiform Encephalopathy–Infected Cattle

To the Editor: A defi nitive diagnosis of bovine spongiform encephalopathy (BSE) in cattle usually relies on Western blot and immunohistochemical testing of samples from the obex region of the brainstem. These conventional diagnostic tests can detect the presence of the abnormal (disease-associated) form of the prion protein (PrPSc) in brain samples several months before the onset of clinical signs; however, there is no appropriate, universal tool for early preclinical and antemortem diagnosis of BSE. Furthermore, confi rmation of the disease is currently only possible by postmortem examination of brain tissues. In this study, we used the serial protein misfolding cyclic amplifi cation (sPMCA) technique to determine the presence of PrPSc in saliva samples collected from BSE-infected cows before and after the onset of disease (1). In a previous study (2), we analyzed the tissue distribution of PrPSc in cattle up to 66 months after they were orally inoculated with a relatively low dose (5 g) of homogenized brainstem from animals with naturally occurring BSE in England. In 2011, after publication of that study and 83.3 months after the cows were inoculated, clinical signs of BSE developed in 1 cow (no. 5444); necropsy was performed 84.7 months after inoculation. In addition, we used saliva samples from 2 BSE-affected cows (nos. 5413 and 5437) (2) to determine the presence of PrPSc. We collected saliva samples from animals at 4 monthly intervals, beginning in 2009, 56 months after inoculation. Samples were stored at -80°C until analysis. Using the sodium phosphotungstic acid precipitation method, we concentrated (100-fold) individual 1-mL saliva samples from each time point. We then diluted the concentrated samples 1:10 with the normal isoform of prion protein substrate containing 0.5% potassium dextran sulfate. Using the sPMCA technique as described (1), we amplifi ed the samples in 3–8 tubes, and we used Western blot to analyze the proteinase K–treated sPMCA products (2). Using Western blot and immunohistochemical tests, we detected the accumulation of PrPSc in brains collected at necropsy from the 3 cows examined. In addition, using the sPMCA technique, we detected PrPSc signal in 1) saliva samples that were concentrated from samples collected from the same 3 cows at necropsy and in 2) concentrated saliva samples that were collected from 2 of the cows (nos. 5413 and 5444) at the early clinical stages of disease. After saliva samples underwent 3 rounds of amplifi cation, we detected PrPSc in a saliva sample that was collected from cow number 5437 two months before the clinical onset of clinical symptoms (Figure). For 2 of the cows (nos. 5413 and 5437), the positive ratio of salivary PrPSc at round 4 of amplifi cation increased as the disease progressed (Figure). Because PrPSc signal could be detected in BSE-infected brain homogenates diluted up to 10−10 after 2 rounds of amplifi cation (1), we estimated PrPSc levels in the nonconcentrated original saliva samples to be lower than those in BSE-infected brain homogenate diluted to 10−12. No PrPSc signal was detected in samples collected from the 3 cows 3–5 months before the onset of clinical symptoms or from agematched noninfected controls, even after 4 rounds of amplifi cation. We demonstrated the presence of PrPSc in saliva of BSE-affected cows