|
||||||||||||||||||
 |
From: TSS (216-119-138-171.ipset18.wt.net)
Nature 425, 673 - 674 (16 October 2003); Prion diseases: A nucleic-acid accomplice? BYRON CAUGHEY AND DAVID A. KOCISKO Byron Caughey and David A. Kocisko are at the NIAID/NIH Rocky Mountain Laboratories, Hamilton, Montana 59840, USA. Prion proteins that trigger a cascade of protein misfolding in the brain are suspected of being the sole transmissible cause of some brain-destroying diseases. But nucleic acids could be their partner in crime. The seductive notion that a modified host protein might be the sole infectious agent of transmissible spongiform encephalopathies (TSEs) has tantalized the scientific world since the idea was suggested more than 30 years ago1. The ensuing 'prion hypothesis' was developed following the discovery that a host-cell protein called prion protein (PrPC) can exist in TSEs in an abnormal form (PrPSc). Considerable circumstantial evidence is consistent with this Nobel-prizewinning hypothesis, which holds that prions are "transmissible particles that are devoid of nucleic acid"2; however, direct proof of the 'protein-only' nature of TSE agents has remained elusive. In this issue, Deleault et al.3 provide striking evidence that nucleic acids might, after all, be involved in propagating TSE infectivity. Writing on page 717, the authors suggest that vertebrate single-stranded RNA is required for the in vitro amplification of PrPSc. How might prions replicate if they carry no genetic material? The premise of most models of prion replication is that the abnormal form, PrPSc, propagates itself by inducing a conformational change in PrPC, and/or triggering its aggregation. Evidence for this view first came from cell-free reactions, in which PrPSc induced the conversion of PrPC into a PrPSc-like aggregated state4. Both this conversion product and PrPSc are partially resistant to destruction by protein-digesting enzymes (unlike PrPC), and so they are often called PrPres. Cell-free conversion reactions are highly specific in ways that reflect biological features of TSE diseases5. For example, different TSE agents show particular patterns of interspecies transmission — similar patterns of agent-dependent conversion are also seen in the in vitro reactions. But whether PrPC is converted to the pathological form in the same way in the brain is unknown. Indeed, in vitro conversion models suffer from two limitations: PrPres formed in cell-free reactions has never been shown to constitute new TSE infectivity in animals, and most in vitro reactions yield substoichiometric quantities of PrPres relative to the PrPSc initially added5. But much higher yields of PrPres are obtained when homogenates from normal brains and TSE-affected brains are mixed together, suggesting that cellular cofactors might be important for PrPSc propagation in vivo6. A number of potential cofactors have been proposed, including nucleic acids, which are known to interact with PrP molecules in vitro7-9. Deleault et al.3 have now looked at the effect of nucleic acids on the propagation of PrPres in brain homogenates. They found that PrPres was amplified sixfold after mixing infected and normal brain homogenates. But when the homogenates were depleted of single-stranded RNA, no amplification was detected. In contrast, adding high concentrations of exogenous RNA (extracted from uninfected hamster or mouse tissues) boosted PrPres amplification about 24-fold. So these results suggest that single-stranded RNA molecules are necessary for PrPres amplification. Interestingly, the authors found that the activity of the RNA shows some species specificity — RNA from invertebrates failed to support amplification. Further characterization showed that the most active RNA contained more than 300 nucleotides and co-purified with ribosomal RNA (a form of RNA found at the site of protein synthesis in cells). But beyond these general characteristics, the identity of the active RNA is unknown. Do the new findings refute the popular protein-only prion hypothesis? Not necessarily, because, as the authors point out, RNA might be a host-cell cofactor required for the conversion of PrPC, rather than an obligatory or informational component of the transmissible prion. Deleault et al. showed that active RNA is also present in extracts of normal mammalian tissues, so these molecules are clearly not specific to TSE infections. Furthermore, biochemical analyses of extensively purified PrPSc preparations have provided no evidence that a nucleic acid component of the size specified by Deleault et al. is an essential element of an infectious unit10. Nonetheless, the possibility remains that small, variable and/or fragmented host-derived RNA sequences stabilize PrPSc infectivity. How might RNA promote the conversion of PrPC? Nucleic acid molecules (both RNA and DNA) are known to bind to PrPC, and under certain circumstances DNA can promote conformational changes in PrPC and trigger its aggregation into fibrils7, 8. It is possible that RNA behaves similarly during PrPres amplification. Curiously, Deleault et al. saw no effect on PrPres propagation after adding DNA molecules to the reactions or depleting homogenates of endogenous DNA using DNA-digesting enzymes. However, the added DNA molecules were probably far too small to be active, and the effects of the endogenous DNA might have been blocked by bound proteins. Large, negatively charged nucleic acid molecules (whether RNA or DNA) could act as scaffolds or catalytic templates that promote PrPSc–PrPC interactions8, 11. But in intact cells, nucleic acids and PrP molecules are usually separated by cell membranes: RNA molecules are found in the nucleus and cytoplasm, whereas PrPC and PrPSc molecules are mainly found outside the cell, on the cell surface or in intracellular organelles. Given these apparent logistical difficulties, it is worth considering a different view — that the RNA molecules might act as surrogates for a different cofactor that is crucial for the conversion of PrPC in vivo, but which is in short supply in the in vitro reactions. For instance, sulphated glycosaminoglycans (which, like RNA, are large polyanions) are found in the same parts of the cell as PrPC and PrPSc, and these molecules can strongly modulate PrPres formation in vitro and in vivo (see refs 11, 12 and references therein). Deleault and colleagues found that various other polyanions, including sulphated glycans, had no effect on PrPres amplification. However, given the complexity of brain homogenates, it is possible that the effects of these added polymers were masked by interactions with vast excesses of molecules other than PrP. Of all the polyanions that Deleault et al. tested, large single-stranded RNA molecules were most efficient at promoting PrPres amplification. The authors' observations raise a number of questions. Can RNA from different species influence the patterns of interspecies transmission of TSE agents, and the propagation of distinct TSE 'strains' and PrPSc conformations? Does the presence of a specific type of RNA affect TSE infectivity? And does RNA influence the progression of other diseases involving abnormal protein aggregation, such as Alzheimer's, Parkinson's or Huntington's diseases? Whatever the answers to these questions, the ability of RNA to induce the amplification of PrPres might also have practical implications. It is important to be able to detect very small amounts of PrPres in order to protect the food supply and for early diagnosis of TSEs such as Creutzfeldt–Jakob disease and bovine spongiform encephalopathy (BSE). Using RNA to amplify PrPres in cell-free conversion reactions could improve the sensitivity and reliability of diagnostic tests. So Deseault and colleagues' findings should be of interest whether or not RNA molecules turn out to be important players in PrPres amplification in vivo. References © 2003 Nature Publishing Group http://www.nature.com/ TSS
|
