In 2002, three rhesus macaques were inoculated with BSE intraperitoneally (10 mL of a 10% homogenate of brain from BSE-diseased cattle). As controls, 2 rhesus macaques received saline (10 mL) and 1 was untreated. All procedures involving rhesus macaques were performed at the Institute of Neuropathology, University Medical Center Hamburg-Eppendorf (Hamburg, Germany), in accordance with the German Animal Welfare Act and the Council Directive 86/609/EEC (Permit 33.42502/08–08.02 LAVES, Lower Saxony, Germany). Animals were observed for clinical signs of prion disease and, when signs of terminal prion disease became evident, were euthanized and underwent autopsy. In all 3 BSE-challenged macaques and none of the controls a progressive neurologic disease developed 49, 59, and 61 months postinoculation. Examination of brain by using hematoxylin and eosin staining showed typical neuropathologic features of vCJD (data not shown) and abundant deposits of PrP^Sc in the cortex, basal ganglia, and cerebellum in paraffin-embedded tissue blots performed as described by using 12F10 monclonal antiprion antibody (6) (Figure 1, panel A). The mobility of the unglycosylated PrP^Sc band and the glycoform ratio of proteinase K–digested PrP^Sc were similar to those in BSE when assessed by Western blot analysis by using monoclonal POM-1 antiprion antibody as described (7) (Figure 1, panel B).

Besides lymphoreticular tissues, the muscular compartment is targeted by prions (7,8). Thus, we assessed presence of PrP^Sc in skeletal and heart muscle by Western blot analysis with sodium phosphotungstic acid precipitation for enrichment of PrP^Sc and protein misfolding cyclic amplification by using published protocols (3). We could not detect substantial amounts of PrP^Sc in skeletal muscle (Figure 2, panel A). One macaque showed abundant PrP^Sc (≈1/100 of PrP^Sc found in brain) in heart in Western blot and protein misfolding cyclic amplification (Figure 2, panels A, B). Paraffin-embedded tissue blot analysis of this heart showed PrP^Sc as amyloid, occupying considerable stretches of heart tissue, mainly in the septum (Figure 2, panel C), whereas no PrP^Sc could be seen in hearts of other macaques (data not shown). These findings were confirmed by strong Congo red–positive patch-like depositions in cardiomyocytes in the heart of this monkey (Figure 2, panel D). The primate with cardiac PrP^Sc showed the longest disease duration (4 months, compared with 4 weeks for other BSE-infected monkeys), signs of cardiac affection when assessed by relevant makers of cardiac hypertrophy and of cardiac distress–associated inflammation, and only this macaque showed clinical signs of fatigue and signs of cardiac distress (i.e., venous congestion) on autopsy (Technical Appendix, Table). Histologic examination of heart tissue with hematoxylin and eosin staining and immunohistochemical stainings against B and T cells (CD20 [not shown] and CD3) did not provide evidence for toxic cardiomyopathy (i.e., fibrosis or vacuolization), nor did we find signs of inflammatory reaction (Figure 2, panel D).

Although the vCJD epidemic is declining, considerable concern exists that clinical characterastics of vCJD will shift. The most important genetic risk factor for development of vCJD is homozygosity for methionine on PRNP codon 129, and all but 1 patient with clinical vCJD carry this polymorphism (5). Thus, future cases of vCJD with longer incubation times are likely to comprise more patients with alternative codon 129 polymorphisms than methionine homozygosity. Data from rodent experiments indicate that clinical features of vCJD may differ in these patients (9). Thus, the next decades may see a shift in vCJD phenotypes. Further uncertainty for atypical cases in humans results from the possibility of secondary transmission of vCJD through blood products from subclinical carriers, which may lead to development of nonclassical vCJD phenotypes (5).

We showed that BSE infection of primates may occur as prion-amyloid cardiomyopathy. Because prion-amyloid cardiomyopathy developed in only 1 of 3 macaques, host-encoded factors, such as genetic makeup, probably influence development of this cardiac phenotype. All macaques are homozygous for methionine on PRNP codon 129; thus, prion-amyloid cardiomyopathy cannot be related to polymorphic codon 129 in our study (10). Cardiac involvement has been observed in a patient with sporadic CJD and is prominent in prion-diseased mice expressing PrP^C lacking its membrane anchor (11,12). We considered the possibility that preexisting pathology, such as spontaneous cardiomyopathy or inflammation of the heart, might have contributed to cardiac PrP^Sc, and the fact that we did not find any evidence for toxic cardiomyopathy or inflammation in the primate does not exclude this possibility. Because the macaque with abundant PrP^Sc deposition in heart had longer disease duration, it is also possible that longer disease duration, which favors centrifugal spread of prions to peripheral tissues, contributed to cardiac affection in this primate (7). Peripheral deposition of PrP^Sc in vCJD is well studied (3). We were surprised by the amount and deposition type of PrP^Sc in heart, reaching 1/100 of the amount seen in brain and deposited as amyloid across large stretches of heart tissue. Skeletal muscle of prion-diseased patients and nonhuman primates routinely harbor minimal amounts of PrP^Sc (<1/1000 that found in brain), and PrP^Sc in muscle is virtually impossible to detect by in situ methods (6,8,13). To our knowledge, PrP^Sc has not been detected in heart of vCJD-diseased persons or in patients with systemic amyloidosis, although primates orally exposed to BSE show very low amounts of cardiac PrP^Sc (8,14,15). The lack of cardiac PrP^Sc in vCJD may result from small cohorts investigated. Because the spectrum of vCJD is likely to change, broad application of current clinical criteria for vCJD in clinical practice may lead to underreporting of vCJD, missing atypical cases of vCJD.

In conclusion, we showed that BSE-infection of primates may lead to prion-amyloid cardiomyopathy. These data should be considered when vCJD surveillance is conducted.

RNA isolation method and supplementary results.