How independent are TSE agents from their hosts?

Abstract

Central to understanding the nature TSE agents (or prions) is how their genetic information is distinguished from the host. Are TSEs truly infectious diseases with host-independent genomes, or are they aberrations of a host component derived from the host genome? Recent experiments tested whether glycosylation of host PrP affects TSE strain characteristics. Wild-type mice were infected with 3 TSE strains passaged through transgenic mice with PrP devoid of glycans at 1 or both N-glycosylation sites. Strain-specific characteristics of 1 TSE strain changed but did not change for 2 others. Changes resulted from the selection of mutant TSE strains in a novel replicative environment. In general the properties of established TSEs support the genetic independence of TSE agents from the host, and specifically the primary structure of PrP does not directly encode TSE agent properties. However sporadic TSEs, challenge this independency. The prion hypothesis explains emerging TSEs relatively successfully but poorly accounts for the diversity and mutability of established TSE strains, or how many different infectious conformations are sustained thermodynamically. Research on early changes in RNA expression and events at the ribosome may inform the debate on TSE agent properties and their interaction with host cell machinery.

Keywords: :

• TSE agents
• prion hypothesis
• genetic information
• infectious disease
• glycosylation
• mutability

This article refers to:

The molecular composition of the agents causing the transmissible spongiform encephalopathies (TSEs) or how their structures encode the genetic information they require to express their phenotypic properties are very incompletely understood. But more fundamental to an understanding of their biochemical nature is how their genetic information is defined and distinguished from that of the hosts which they infect. Recent experiments have tested whether the glycosylation status of host PrP affects TSE strain characteristics. Wild-type mice were infected with 3 TSE strains passaged through transgenic mice with PrP devoid of glycans at the first, second or both N-glycosylation sites. Strain-specific characteristics of one TSE strain, 79A, changed when PrP^Sc was devoid of 1 or both glycans, but did not change for 2 other TSE strains (ME7 and 301C). The simplest explanation for these findings was that change in the phenotype of 79A resulted from the selection of mutant TSE strains in the novel replicative environment where post-translational changes to PrP had been engineered.¹

The established natural TSEs, scrapie in sheep and CWD in deer, are clearly caused by infectious agents passed from one host to another. However the position is not so clear cut with sporadic CJD or atypical forms of scrapie and BSE. They may be caused by a pre-existing independent agent, but more likely they occur spontaneously; and in humans apparently spontaneous disease occurs much more readily in individuals with certain mutations of PrP where the mutation can appear causal. So are the TSEs truly infectious diseases with genomes completely independent of their hosts, or are they an aberration of a host component which can be sustained indefinitely but nevertheless whose properties derive directly from the host genome, via sequence modifications and/or post-translational modifications of PrP? Are they fundamentally reliant on the host for their properties?

This question has lain at the heart of TSE research from the earliest days. In 1960 Parry argued that “scrapie is primarily a hereditarily determined disease” and that “As this hereditary scrapie can be transmitted by inoculation… it is suggested in the homozygous recessive individual a substance, the transmissible scrapie agent, is formed which is closely associated with the development of degenerative changes in several organs.”² However Dickinson responded: “Analysis of the incidence of natural scrapie…leads to the firm conclusion that the disease is not caused by the action of a simple recessive or dominant gene in the sheep. The total evidence on scrapie is shown to be consistent with the interpretation that a biologically independent pathogenic organism causes the disease.³ Dickinson showed that both natural and experimental scrapie was caused by an independent infectious agent whose replication was under the genetic control of the host, primarily a single host gene which in mice was named Sinc and in sheep Sip.^4,5 The gene has subsequently been shown to encode the PrP protein⁶ and has been renamed Prnp.

These 2 paradigms subsequently affected the development of hypotheses about the molecular composition of TSE agents. On the one hand there was an emphasis on an informational molecule that was independent of the host. This was presumed most likely to be a nucleic acid, either of a virino: defined as an independent informational molecule, protected by host protein, PrP,^7-9 or as the genome of a virus.^10,11 On the other hand an emphasis on the apparent host origins of some human forms of these diseases on the biochemical properties of the TSE agents led to the development of protein only hypotheses,^12,13 which originally proposed self-replicating protein but subsequently proposed that the host glycoprotein PrP assumes abnormal conformations that are infectious, i.e., that they initiate or seed further conformational change to normal PrP.¹⁴ The debate continues because of an inability to isolate and characterize TSE genomic nucleic acid, or to define in sufficient detail the conformational changes to PrP that the diversity and mutability of TSE strains require.

TSE Agent Properties

Similarly to other infectious agents, many different strains of TSE agent have been identified.¹⁵ In the absence of molecular genotyping methods they continue to be characterized using their phenotypic properties, as was the case with conventional infectious organisms. TSE strains were first phenotypically characterized using the measurement of incubation periods in mice differing in PrP genotype and from the amount and distribution of vacuolar lesions in the brain.^16,17 These methods remain among the most powerful for distinguishing between TSE isolates in mice. Other phenotypic properties can also now be measured: the amount and distribution of abnormal PrP deposition in the brain,¹⁸ the biochemical properties of PrP, e.g., its migration on SDS gels after digestion with a protease, the degree to which PrP^Sc is glycosylated,¹⁹ and the thermostability of TSE agents.²⁰ In addition mutational change to TSE agents can be identified after serial passage.²¹ The changes are stereotypic, i.e., the same changes are usually found when the conditions remain the same. The range of properties and the large number of discrete TSE agent phenotypes argue for an information-rich TSE genome.

The information available about sheep scrapie, CWD and from many strains of experimental TSEs can clearly be interpreted to support the independence of TSE agent genetic information from the host. However the sporadic cases of TSE, apparently arising spontaneously, challenge this independency. They appear to occur very rarely: sporadic CJD being responsible for 1 in 10,000 deaths in humans. It is possible that TSE agents may be harboured in some individuals as independent infectious agents without causing disease but then either are triggered to replicate and cause disease in their host or to be passed on to a more susceptible individual. Alternatively, the disease may arise spontaneously from host-embedded components analogously to retrovirus genomes, perhaps because of mutational change in a precursor TSE agent structure. Usually the infections that arise are not naturally infectious, but can be transmitted experimentally or when routes are created by human intervention, through cannibalistic rites associated with the transmission of kuru, the use of growth hormone extractions from CJD infected pituitary glands and the incorporation of BSE infected tissue into meat and bone meal used in cattle feed. The experimental strains that are derived from these sources behave with similar independence from the host as those derived from scrapie and CWD.

The prion hypothesis postulates disease arising de novo through the conversion of the conformation of the PrP protein to an abnormal, infectious conformation. The abnormally folded PrP can then induce other PrP molecules to adopt the same abnormal conformation. Different abnormal conformations are postulated which encode the different strains. Such ideas have been proposed to explain the cell to cell transmission of pathology in other diseases via abnormally folded protein, Aβ in Alzheimer disease (AD) and α-synuclein in Parkinson disease (PD). The prion hypothesis has some of its origins in the challenge to explain sporadic and mutation-associated TSEs and it does so relatively successfully. However at the molecular level it still struggles to account for the diversity and mutability of established TSE strains whether natural or experimental. Nor have questions been addressed about how many different infectious conformations of PrP can be sustained thermodynamically, rather than reverting to one or a few of the most thermodynamically stable folded structures. It should also be borne in mind that AD and PD appear etiologically distinct from established TSEs, and by inference emerging TSEs too. Specifically no infectious etiology has been demonstrated. However there may be common underlying mechanisms causing the production of abnormally folded protein, initiated directly by abnormally folded protein interacting directly with normal protein, or initiated by other factors disrupting the folding of protein.

Host encoded information in the PrP sequence or in post-translational modifications would result in the host specifying TSE agent properties. In other words, are polymorphisms/mutations of host PrP directly responsible for change in agent genotype? If not, what effect do they have on agent properties? When TSE agents are passaged experimentally through a new host whose PrP genotype differs from the host in which it was previously replicating, in some cases (e.g., ME7 passaged in mice with different PrP genotypes no change in agent phenotype is detected, but in other cases (e.g., passage of sheep scrapie into mice, to give 87A, which then breaks down to ME7) change does occur.²¹ Sometimes the change appears to occur immediately but in other cases change is delayed. Change in agent properties is not directed but the new replicative environment that the new genotype of PrP provides allows selection of mutant forms of TSE agent that replicate more efficiently. Similar arguments apply to the question of whether the carbohydrate moieties of PrP directly contribute to the genotype of TSE agents. Again in some cases removal of one or both carbohydrate moieties has little effect on the agent phenotype when assessed back in wild type mice, but in other cases change in TSE strain properties (mutation) takes place.^1,22,23 Overall, the primary structure of PrP, including its post-translational modifications, does not directly encode the independent information required by established TSE agents.¹

Current and Future Research Opportunities

Little is known about how TSE agents initiate pathological change in their hosts. TSE infections are slow and progressive. They can be targeted to some neurons but not others. The infection appears to spread from cell to cell, mainly neuron to neuron in the CNS.²⁴ Analysis of infected whole brain therefore leads to regional analysis of the brain at different stages of disease pathogenesis. By focusing on a single small area of brain it has been possible recently to examine RNA expression level changes through pathogenesis and this has shown substantial early change in RNA levels, from about midway through infection. Specifically a major cluster of genes was altered early during pre-clinical disease. This cluster contained a signature indicative of synaptic N-methyl-d-aspartic acid (NMDA) receptor signaling and the activation of neuroprotective pathways. These findings suggest that TSE agent replication results in the persistent stimulation of a programmed response that is mediated, at least in part, by synaptic NMDA receptor activity, initially promoting cell survival and neurite remodelling.²⁵ The patterns of expression alter as the disease progresses. The earliest events remain to be analyzed since results from earlier than halfway through the incubation period were not reported, and at that stage substantial change is already demonstrable. miRNA levels were also altered from early on. Although profound and specific changes to cellular homeostasis had been caused by the infection from early on, nothing is known about how TSE agents cause these changes. Further investigation of these very early molecular events may throw light on the primary molecular interactions of TSE agents with host cellular machinery and processes fundamental to neuronal function and survival, and how targeting of infection and pathology to some neurons but not others is achieved.

Another potential insight into the molecular mechanisms of TSE pathogenesis and hence the molecular function of TSE agents comes from experiments in which two inhibitors of the production of abnormally folded PrP were identified.²⁶ They also lengthened TSE incubation periods.²⁷ These inhibitors probably operate through interactions with domain V of the RNA of the large ribosomal subunit, by inhibiting the rRNA-mediated protein folding activity of the ribosome. Their effect was to interfere with the abnormal folding of PrP which was probably being chaperoned by the ribosome.²⁸ This finding conflicts with the assumption that abnormal PrP is obtained from the conversion of normal, mature PrP^C into abnormal forms by direct interaction with existing abnormal forms of PrP. However it supports the view that the inefficient glycosylation of PrP^Sc that occurs in some TSE models occurs biosynthetically, due to the abnormal folding of PrP when transiting the endoplasmic reticulum and Golgi apparatus.²⁹ These results suggest that mechanisms for the production of abnormally folded PrP are controlled at the heart of the protein translational machinery—but how such control is exercised is not known. Although it is possible to speculate that existing abnormally folded PrP interacts with the ribosomal folding machinery, it is equally valid to suggest that other molecules may be operating to induce abnormal folding of PrP. Other proteins may be affected too.³⁰

Can Two Views of the Causality of TSE Diseases Identified Above be Reconciled?

A scenario is developing in which TSE agents may emerge sporadically from host components, which are altered (mutated) to forms that sustain discrete information and that ensures its self-replication, can induce pathological change in the cells in which it is localized and can spread from cell to cell. If a form of these novel agents emerges that acquires the ability to infect lymphoid and secretory organs, this may allow excretion and deposition of the TSE agent in the environment and hence opportunities to naturally infect a new host. Hence, an independent infectious agent will have evolved. The molecular structures of these agents must provide a sufficient platform for TSE agents to evolve and carry the information required. The interaction of TSE agents with host cell machinery at the molecular level may now be further explored by examining the earliest changes in mRNA and miRNA expression levels and by examining how molecular pathogenesis is controlled at the ribosome. In turn these topics may inform the debate on the structure and function of TSE agents.

Abbreviations:
TSEs	=	transmissible spongiform encephalopathies
BSE	=	bovine spongiform encephalopathy
CJD	=	Creutzfeldt-Jakob disease
CWD	=	chronic wasting disease
AD	=	Alzheimer’s Disease
PD	=	Parkinson’s Disease

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

The author is most grateful to Dr Abigail Diack and Prof. Jean Manson for reading the manuscript and providing helpful suggestions.

Submitted

05/21/2013

Accepted

06/14/2013