All monkeys and mice were housed at the Rocky Mountain Laboratories (Hamilton, MT, USA). Experimentation followed protocols approved by the National Institutes of Health Rocky Mountain Laboratories Animal Care and Use Committee.

CWD-positive brain homogenates were provided by E.S.W. and M.W.M. Contents of each pool were as follows: MD-1, 6 free-ranging mule deer from Wyoming (18); MD-2, 4 captive mule deer from Colorado; MD-3, 28 captive mule deer from Wyoming and Colorado (2,19); WTD-1, 7 captive white-tailed deer from Wyoming and Colorado (18,20); WTD-2, 1 wild white-tailed deer from Wyoming; Elk-1, 2 free-ranging elk from Wyoming (18); Elk-2, 6 elk from a South Dakota game farm; and Elk-3, 10 captive elk from Wyoming and Colorado. Normal elk brain was a pool from 2 elk from Montana obtained from Lynn Creekmore of the US Department of Agriculture.

For intracerebral injections, squirrel monkeys received either 2 mg or 20 mg brain in a total volume of 200 μL, and cynomolgus macaques received 5 mg in a total volume of 500 μL. Oral doses of 200 mg brain/mL were given on 5 different days at 2–6 day intervals. Squirrel monkeys received 3-mL doses; most macaques received 4-mL doses. The inoculum was given to anesthetized animals through a rubber gastric tube.

Brain homogenates diluted in phosphate buffered balanced solution containing 2% fetal bovine serum were inoculated intracerebrally into young adult mice. Volumes were 50 μL.

Mice expressing human PrP (tgRM and tg66) were generated by using a transgene, cosSHa.HumPrP, which was created by ligating the human PrP open reading frame into the cosSHa.Tet vector (21). The transgene was inoculated into eggs of FVBn–mouse PrP null mice in the laboratories of R.R. (tg66) and L.C. (tgRM). Each line of mice overexpressed human PrP as tested by Western blot with monoclonal antibody 3F4.

Tissues were prepared by making a 20% (wt/vol) homogenate in 0.01 M Tris buffer, pH 7.4. Samples to be analyzed for protease-sensitive PrP (PrPsen) contained the following protease inhibitors: 10 μmol/L leupeptin, 1 μmol/L pepstatin A, and 1 μg/mL aprotinin. Samples were sonicated for 1 min and centrifuged at 5,000 rpm for 10 min. Supernatants were mixed 1:1 in 2× sample buffer and boiled for 3 min before electrophoresis.

Preparation of samples for PrPres analysis has been described (18). Removal of carbohydrate residues from PrPres was performed by digestion with peptide-N-glycosidase F (22).

After electrophoresis, proteins were transferred to Immobilon polyvinylidene difluoride–P membranes (Millipore, Billerica, MA, USA), and PrP bands were detected with antibodies 3F4 (residues 109–112) (23), D13 (residues 96–106) (24) (InPro Biotechnology, Inc., South San Francisco, CA, USA), or L42 (residues 145–163) (r-Biopharm, Darmstadt, Germany) (25). Bands were detected by using enhanced chemiluminescence substrate (GE Healthcare, Piscataway, NJ, USA).

Routine formalin fixation, embedding, and tissue-sectioning protocols were followed. Tissues were stained with hematoxylin and eosin and analyzed for pathologic changes. Immunohistochemical staining was performed by using an automated Nexus stainer (Ventana, Tucson, AZ, USA). Anti-PrP antibodies D13 and 3F4 were used for PrPres immunostaining as described (26,27).

Primate genomic DNA was purified from whole blood, and PCR products were amplified by using PuRe Taq Ready-To-Go PCR beads (GE Healthcare). Two primers from the extreme outer ends of the open reading frame, including the previously published forward primer HM-1 (28) with mPrP-780R (5′-TCCCACTATCAGGAAGATGAGG-3′) or a combination of outer primers with internal primers mPrP-397F (5′-CCTTGGTGGCTACATGCTG-3′) and mPrP-416R (5′-CCAGCATGTAGCCACCAAG-3′), were used. Assembly comparisons were made against human, elk, mule deer, cynomolgus macaque, and squirrel monkey by using Sequencher version 4.6 (Gene Codes, Ann Arbor, MI, USA).

When the 8 pools of CWD (representing both wild and captive deer and elk) used as inocula were analyzed by immunoblot, PrPres in the 8 pools showed similar electrophoretic mobilities and glycoform patterns (Figure 1, panel A), but PrPres levels differed when quantitatively compared (Figure 1, panel C). To measure the level of infectivity in these pools, we titered each pool in transgenic mice expressing deer PrP (line 33; tgDeerPrP) (18). A typical endpoint dilution titration is shown in Figure 1, panel B. The 8 pools had 50% infectious dose (ID₅₀) titers ranging from 6.3 × 10⁷ to 5.0 × 10⁸ ID₅₀/g of brain homogenate (Figure 1, panel C). Comparison of titers with PrPres levels showed a partial correlation (Figure 1, panel C). For example, the CWD pool with the lowest infectivity titer (MD-2) was also the pool with the lowest PrPres level. However, for some pools, these tests showed discrepant values.

To test susceptibility to CWD, we inoculated squirrel monkeys with each of the 8 CWD pools described above. Of 13 squirrel monkeys, 11 became symptomatic (33–53 mo postinfection [mpi]) (Table 1). The most consistent and reliable clinical finding was a severe wasting syndrome. Weight loss (average decrease of 33%) was most pronounced in the final few months of infection. Affected monkeys also had rough, poor-quality coats despite continuing to eat and drink. In the final 3–5 weeks, monkeys became weak and less active and spent most of their time hunched at the bottom of their cage. When the monkeys were encouraged to move, they did so slowly and deliberately. In the terminal stage of disease, a few had muscle tremors, excessive salivation, and mild ataxia. Fine, coordinated movement such as eating food was rarely affected. Monkeys were euthanized when terminal-stage weakness and wasting compromised their mobility and ability to eat and drink.

No clear correlation between incubation period and amount of agent inoculated was noted (Table 1). For example, 3 pairs of monkeys received the same inocula but in amounts that differed by 10-fold (Elk-1, Elk-3, and MD-1). Two pairs that received the lower dose became clinically sick first (Elk-1 and Elk-3). Both members of the third pair (MD-1) were euthanized after 36 months (Table 1). Two animals received the same dose of WTD-1 pool, yet to date, only 1 animal has become clinically sick. Animals that received the CWD pool with the lowest titer (MD-2) had incubation periods similar to those receiving much higher titered inocula (Table 1).

In all monkeys with clinical signs, CWD was confirmed by Western blot detection of PrPres in brain (Figure 2, panels A, B). The glycoform pattern of PrPres was similar for all affected monkeys inoculated with different CWD pools (Figure 2, panel B). Because PrPres deposition may also occur outside the central nervous system, we also tested peripheral lymphoid tissues. For 3 of 11 monkeys that had PrPres in brain, PrPres was also found in spleen and lymph nodes (Figure 2, panel C). In general, PrPres levels were much lower in lymphoid tissues than in brain and were often not detected by Western blot. All nonlymphatic tissues tested (cardiac muscle, skeletal muscle, duodenum, jejunum, ileum, colon, salivary gland, kidney, and lung) were negative for PrPres by immunoblot.

Tissues from squirrel monkeys euthanized after intracerebral injection with CWD (Table 1) were also examined by histopathologic analysis, including staining with hematoxylin and eosin and immunohistochemical detection of PrPres. All monkeys examined had spongiosis in the cerebral cortex, caudate, putamen, and thalamus (Figure 3, panel A). In addition, PrPres deposition was observed in many brain regions with large PrPres-positive plaques in the thalamus, cerebellum, and spinal cord (Figure 3, panels C, E, F) and in smaller plaques spread out in the gray matter of the internal capsule and white matter of the corpus callosum (Figure 3, panels G, H). The most abundant and consistent location for PrPres staining was found in the frontal cortex and in the fiber tracts of the claustrum (Figure 3, panel I). The adjacent caudate had severe spongiosis and astrocytosis but minimal PrPres (Figure 3, panel I). PrPres was also detected in lymph nodes and spleen, within follicles, in areas resembling follicular dendritic cells (Figure 3, panels K, M). Immunohistochemical analysis showed no PrPres in heart, kidney, adrenal gland, skeletal muscle, salivary gland, tongue, pancreas, white fat, and all regions of the gastrointestinal tract.

To test a more natural route of infection, we exposed squirrel monkeys orally to CWD. Of the 15 exposed squirrel monkeys, 1 (no. 345) was found dead in its cage at 69 mpi; it had shown no neurologic signs or weakness. Western blot results indicated PrPres in brain, spleen, and lymph nodes (Figure 2, panel D). The level of PrPres in the brain of monkey 345 was comparable with that in end-stage intracerebrally inoculated monkeys; body weight at necropsy indicated a 33% decrease over the final 10 months. The high levels of PrPres and the severe wasting indicate that CWD infection could have been the cause of death. A second monkey, 303, was euthanized at 69 mpi because of suspicion of TSE after 2 weeks of progressive weakness, wasting, and eventual anorexia. PrPres analysis confirmed PrPres in brain (Figure 2, panel D), spleen, and lymph nodes. For monkeys 303 and 345, levels of PrPres in the lymph nodes and spleens were 10–100-fold lower than those in brain.

Two other orally infected monkeys were euthanized during the first 69 mpi (Table 2). Monkey 301 was euthanized at 39 mpi, after rapid onset of lethargy and anorexia that led to severe dehydration. Results of Western blot analysis for PrPres were negative in brain (Figure 1, panel B), spleen, lymph nodes, heart, skeletal muscle, duodenum, jejunum, ileum, colon, salivary gland, kidney, lung, and tonsil. However, immunohistochemical analysis detected PrPres in the spleen and 1 mesenteric lymph node from this monkey, indicating a low level of infection (Figure 3, panels J,K). Monkey 614 was euthanized at 44 mpi because it did not recover from anesthesia related to routine tuberculosis screening. Neither Western blot nor immunohistochemical analysis detected PrPres in brain, spleen, or lymph nodes of this monkey.

We inoculated cynomolgus macaques both orally and intracerebrally with 3 CWD inocula representing elk, mule deer, and white-tailed deer (Table 3). Of the cynomolgus macaques, 1 (no. 609) was euthanized at 48 mpi after it became aggressive. Brain (Figure 2, panel B), spinal cord, spleen, and lymph nodes were negative for PrPres by Western blot and immunohistochemical analysis. All remaining CWD-inoculated cynomolgus monkeys are currently (at 70 mpi) neurologically asymptomatic and have stable or increased body weights.

Amino acid substitutions in PrP can alter susceptibility to TSE agents, including CWD (18,29,30). To determine whether the lack of susceptibility in several intracerebrally inoculated squirrel monkeys (Table 1) was caused by PrP gene polymorphisms, we sequenced the PrP genes from 23 squirrel monkeys. When compared with published squirrel monkey sequences (28,31), variation was seen at residue 164, in the number of octapeptide repeats, and at residue 19 of the signal peptide (Table 4). However, these genetic differences in PrP did not appear to account for the lack of susceptibility of monkey 310, which was genotype A, because this genotype was also found in 5 of the CWD-positive monkeys. Because we were not able to sequence PrP of monkey 628, we could not assess the role of PrP variation in the lack of disease.

To determine whether passage of CWD in squirrel monkeys altered the tropism of the infectious agent, we inoculated tgDeerPrP mice and tg mice expressing human PrP (lines 66 and RM) intracerebrally with tissue homogenates from 3 CWD-positive squirrel monkeys (nos. 322, 308, and 301) with PrPres and from an intracerebrally inoculated cynomolgus macaque (no. 609). Clinical disease did not develop in any tgDeerPrP, tg66, or tgRM mice during 600–700 days (Table 5). The lack of transmission to tgDeerPrP mice from the 3 squirrel monkeys with detectable CWD PrPres indicated that either the infectivity levels were low in these squirrel monkeys or that the original cervid species tropism was altered by the passage in squirrel monkeys. Similarly, the lack of transmission to tg mice expressing human PrP implied that passage through squirrel monkeys did not facilitate adaptation to an agent with increased tropism for humans.