Transmissible Spongiform Encephalopathy: The Glycosylation Status of PrPC Is a Key Factor in Determining Transmissible Spongiform Encephalopathy Transmission between Species

The Glycosylation Status of PrPC Is a Key Factor in Determining Transmissible Spongiform Encephalopathy Transmission between Species

Frances K. Wisemana*, Enrico Cancellottia, Pedro Piccardoa,c, Kayleigh Iremongera*, Aileen Boylea, Deborah Browna, James W. Ironsideb, Jean C. Mansona and Abigail B. Diacka aNeurobiology Division, The Roslin Institute and R(D)SVS, University of Edinburgh, Easter Bush, United Kingdom bThe National Creutzfeldt-Jakob Disease Research & Surveillance Unit, University of Edinburgh, Edinburgh, United Kingdom cFood and Drug Administration, Rockville, Maryland, USA

B. Caughey, Editor

+ Author Affiliations

ABSTRACT

The risk of transmission of transmissible spongiform encephalopathies (TSE) between different species has been notoriously unpredictable because the mechanisms of transmission are not fully understood. A transmission barrier between species often prevents infection of a new host with a TSE agent. Nonetheless, some TSE agents are able to cross this barrier and infect new species, with devastating consequences. The host PrPC misfolds during disease pathogenesis and has a major role in controlling the transmission of agents between species, but sequence compatibility between host and agent PrPC does not fully explain host susceptibility. PrPC is posttranslationally modified by the addition of glycan moieties which have an important role in the infectious process. Here, we show in vivo that glycosylation of the host PrPC has a significant impact on the transmission of TSE between different host species. We infected mice carrying different glycosylated forms of PrPC with two human agents (sCJDMM2 and vCJD) and one hamster strain (263K). The absence of glycosylation at both or the first PrPC glycosylation site in the host results in almost complete resistance to disease. The absence of the second site of N-glycan has a dramatic effect on the barrier to transmission between host species, facilitating the transmission of sCJDMM2 to a host normally resistant to this agent. These results highlight glycosylation of PrPC as a key factor in determining the transmission efficiency of TSEs between different species.

IMPORTANCE The risks of transmission of TSE between different species are difficult to predict due to a lack of knowledge over the mechanisms of disease transmission; some strains of TSE are able to cross a species barrier, while others do not. The host protein, PrPC, plays a major role in disease transmission. PrPC undergoes posttranslational glycosylation, and the addition of these glycans may play a role in disease transmission. We infected mice that express different forms of glycosylated PrPC with three different TSE agents. We demonstrate that changing the glycosylation status of the host can have profound effects on disease transmission, changing host susceptibility and incubation times. Our results show that PrPC glycosylation is a key factor in determining risks of TSE transmission between species.

snip...

Transmission of TSE between different species often is limited by a species barrier to infection (6, 7). In experimental models of disease, the species barrier is characterized by an inefficient primary infection with low susceptibility and long incubation times in the new host. Adaptation to the new host then usually occurs in subsequent passages with an increased attack rate and shorter incubation time (6, 8). In naturally occurring TSE, the species barrier prevents transmission of certain agents between different species. However, some agents have been shown to be able to cross this barrier and cause devastating epidemics in a new host. For example, BSE in cattle can be transmitted to humans via the oral route to cause variant CJD (vCJD) (9, 10). BSE also was able to naturally infect a number of different species, such as goats, nyala, kudu, and domestic or captive wild cats (11–13). Understanding how the species barrier is regulated is important, so that the zoonotic potential of a TSE in other animal populations transmitting to humans can be assessed. This is particularly important for newly emergent strains of TSE in both farmed and wild animals (8, 14).

snip...

DISCUSSION

Expression of PrPC is known to influence incubation times of a TSE disease, with reduced levels of the protein resulting in longer incubation periods (48). Earlier studies showed conflicting results over whether PrPC expression levels are altered within the glycosylation transgenic mice (22). This most likely was due to the epitope recognition of the antibodies used and detection of only a subset of isoforms. Our expanded studies here, using a range of monoclonal antibodies within the C-terminal and central region of PrPC, are able to detect all isoforms of PrPC, demonstrating that G1, G2, and G3 mice do have lower levels of PrPC expression than wild-type mice. However, while lower levels of PrPC in the G1 and G3 mice may contribute to longer incubation times, the levels observed in these mice are not likely to explain the resistance to TSE disease observed here. Studies have shown that mice heterozygous for PrPC expression and with a level of PrPC expression similar to that of the G1 mice are fully susceptible to TSE disease, albeit with incubation periods of almost twice that of wild-type mice (48–50). Our studies were maintained to approximately 700 dpi, almost twice the incubation period of sCJD in G2 mice, which also show 50% PrPC expression. Thus, factors other than a reduction in PrPC expression level are likely to contribute to the resistance of these mice to TSE disease. While the lower expression of PrPC in G2 mice may contribute to the longer incubation period observed in this model after challenge with vCJD, the G2 mice are more susceptible to infection with the sCJDMM2- and G2-passaged 263K TSE agents despite expressing lower levels of PrPC than wild-type controls. Therefore, this enhanced susceptibility can be directly attributed to the altered glycosylation status of the host.

The monoglycosylated sCJDMM2 agent was transmitted to a normally resistant host (51) by removal of the glycans at PrP residue 196 (as removed in G2 mice). Moreover, sCJDMM2 became adapted in the G2 host and produced very short incubation times on the second pass. The data suggest that the presence of glycans at PrP residue 196 (as present in G1 or wild-type mice) is responsible for the sCJDMM2 transmission barrier; removal of this site may facilitate the interaction between host monoglycosylated PrPC and the infective monoglycosylated PrPSc, allowing replication of the infective agent. This is the first time that glycosylation-deficient transgenic mice have shown an enhanced susceptibility to TSE infection compared to that of wild-type mice. This suggests that glycosylation at the second glycosylation site can protect against transmission both between and within species.

Experimental transmissions from wild-type or G2 mice infected with the 263K strain provide additional evidence that similar glycosylation statuses of host PrPC and the PrPSc in the inoculated strain can greatly accelerate TSE incubation periods. Indeed, the incubation period in G2 recipients was almost half that of wild-type mice after challenge with the G2-263K strain.

In both primary and secondary passages of vCJD, incubation periods were shorter in wild-type mice than in mice in which the second PrPC N-glycan attachment site was disrupted. The shorter incubation periods were observed irrespective of the glycosylation status of the second site in the infecting PrPSc. While these differences in incubation time can be explained on the basis of lower PrPC expression levels in the G2 mice, we cannot discount the possibility that it indicates a preference of this strain for a PrPC diglycosylated host irrespective of the passage history of the strain. This may explain the ability of this agent to infect a large number of host species and its transmissibility across many species barriers.

G1 and G3 mice showed little susceptibility to infection throughout this study. Indeed, these transgenic mice did not develop any pathologically confirmed clinical TSE disease after inoculation with any of the three agents used, although asymptomatic infection in the form of PrP deposition was detected in extremely low numbers. This may be linked to an inability of this particular host PrPC to propagate nonmurine strains; previous experiments performed with a number of mouse-adapted scrapie strains by several routes have highlighted an intrinsic resistance of both G1 and G3 mice to infection (23, 52). Therefore, it is more likely that the resistance observed in G1 and G3 mice in this study is linked to a more general mechanism rather than an effect of the species barrier. Why the absence of the first glycosylation site should lead to such a dramatic loss of host susceptibility may be related to the conversion efficiency of PrPC to PrPSc. Some in vitro conversion assays have previously suggested that glycosylation inhibits the conversion activity (30). However, such in vitro systems have not revealed the complexity of the glycosylation issue observed in these in vivo studies. The resistance observed in the G3 mice likely is related to the absence of the G1 glycosylation. However, G3 mice also show more C1-truncated PrPC upon biochemical analysis than G1, G2, and wild-type mice. Previous in vitro studies have shown that higher levels of C1 PrPC are associated with resistance to TSE infection (53). In addition, G3 mice show the lowest PrPC expression of the three glycosylation mutants and a different PrPC localization (22). All of these factors might contribute to the resistance to TSE infection of this specific line of mice.

The absence of glycans at the second site may alter the biology of PrPC or PrPSc interaction in a very different way than that of the first glycosylation site. A number of biochemical properties and the cellular localization of PrPC in the G2 mice resemble that observed in wild-type and G1 mice (22). However, the presence/absence of carbohydrates in a specific portion of PrPC may influence other characteristics, such as the ultrastructural localization of PrPC (e.g., localization in a different portion of the cell membrane) or its conformation, and this may dictate the different susceptibility to infection of the G2 mice compared to that of the G1 mice.

We have argued that altered glycosylation status of PrPC alters the host susceptibility. An alternative explanation is that the point mutations inserted in order to modify the N-linked glycosylation sites on PrP are the cause of this change (22). Previous transmission studies performed by us (23) and Neuendorf et al. (20) have shown similar results upon primary passage of both ME7 and mouse BSE strains with prolonged incubation periods in mice deficient at the first glycosylation site despite utilizing different amino acid substitutions and expressing different levels of PrP. In addition, Ikeda et al. (54) showed that substitution of Asn residues to abolish glycosylation sites does not prevent conversion of PrPC to PrPSc. In this study, the differences between the wild-type and G2 hosts in susceptibility to primary passage with two human agents, vCJD and sCJDMM2 (characterized by an identical PrPSc sequence and PK cleavage pattern but a different glycoprofile), further argues for the glycosylation status being the main determinant of host susceptibility rather than the change in amino acid sequence.

The deposition of PrP in the brains of G2 mice infected with vCJD differed from that observed in wild-type mice infected with the same agent. First, the total amount of PrP that accumulated by disease endpoint appeared to be lower in G2 mice. This could be due to less PrPC being available for replication, or it could mean that the rates of misfolding, clearance, and/or toxicity of PrP are changed in the absence of glycosylation at the second site. In addition, small PrP aggregates were observed in G2 mice infected with vCJD, in contrast to the diffuse PrP deposition observed in wild-type mice. Large aggregated deposits of PrP also were observed in G2 mice challenged with sCJDMM2. These data suggest that PrPC that lacks the second glycosylation site has altered misfolding or clearance kinetics, which also may have an important effect on host susceptibility.

In summary, we propose that the transmission of TSE agents across different species can be profoundly influenced by posttranslational events in both PrPC and PrPSc. In particular, we have demonstrated that the glycosylation status of host PrPC (55, 56) can dramatically alter cross-species transmission characteristics and likely is important for this protein to act as a receptor for the incoming TSE agent.

On the other hand, the prevalence of certain PrPSc glycotypes in an infectious inoculum may determine its conformation and the ability to interact with the host and cause a TSE infection. This combination may lead to the binding between PrPSc and PrPC occurring through direct interactions between the glycan residues and/or different PrP regions that have been recently suggested to be important for TSE transmission between different species (57) or by interactions with a number of conversion cofactors previously suggested, such as host proteins or nucleic acids (58–60).

The dramatic effects in altered host susceptibility, in particular the resistance of the G1 and G3 mice to infection, suggests this mechanism provides an important focus for blocking the disease process and protecting the infected individual from neurodegeneration.

Previous Section

Next Section

ACKNOWLEDGMENTS

We acknowledge the excellent technical assistance of Irene McConnell, Val Thomson, Sally Carpenter, Kris Hogan, Gillian Macgregor, Sandra Coupar, Dorothy Kisielewski, and Winggee Liu and the statistical analysis assistance of Jill Sales, BIOSS. We thank Robert Somerville, Wilfred Goldmann, Rona Barron, and Nadia Tuzi for valuable discussions. Antibodies 8H4 and 7A12 were a kind gift of M. S. Sy, University of Cleveland.

This work was supported by the BBSRC and MRC. The NCJDRSU Brain Bank is part of the Edinburgh Brain Bank, which is funded by MRC. F.W. was funded by a Wellcome Trust Ph.D. studentship (069283). K.I. was funded by a BBSRC studentship.

NCJDRSU is supported by the Scottish Government and the Department of Health, England. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health. The findings and conclusions in this article have not been formally disseminated by the Food and Drug Administration and should not be construed to represent any agency determination or policy.

Previous Section

Next Section

FOOTNOTES Received 11 August 2014. Accepted 4 February 2015. Accepted manuscript posted online 11 February 2015. Address correspondence to Abigail Diack, abigail.diack@roslin.ed.ac.uk.

↵* Present address: Frances K. Wiseman, The Department of Neurodegenerative Disease, Institute of Neurology, University College, London, London, United Kingdom; Kayleigh Iremonger, Sheffield Institute for Translational Neuroscience (SITraN), Department of Neuroscience, The University of Sheffield, Sheffield, United Kingdom.

Citation Wiseman FK, Cancellotti E, Piccardo P, Iremonger K, Boyle A, Brown D, Ironside JW, Manson JC, Diack AB. 2015. The glycosylation status of PrPC is a key factor in determining transmissible spongiform encephalopathy transmission between species. J Virol 89:4738–4747. doi:10.1128/JVI.02296-14.

Previous Section

REFERENCES

FOOTNOTES Received 11 August 2014. Accepted 4 February 2015. Accepted manuscript posted online 11 February 2015. Address correspondence to Abigail Diack, abigail.diack@roslin.ed.ac.uk.