Opening ceremony

Mark Zabel and Glenn Telling welcomed the crowd and thanked all of the sponsors and organizers. Debbie Yobs, president of the CJD Foundation, announced that CJD Foundation will be awarding another round of $50,000 research grants in February 2016. Letters of intent are due August 1, 2015. Suzanne Solvyns welcomed Deanna Simpson as the new co-chair of the CJD International Support Alliance, taking Florence Kranitz’s place after her retirement. Jean-Philippe Deslys announced that Alliance Biosecure has extended the deadline for its Call for Projects 2015 to June 30, so you can still submit a new application. Applicants passing the first round of review will be invited to present their proposal in Paris on October 28, with successful grantees notified in November. Alan Rudolph, CSU Vice President for Research, welcomed us and expressed his enthuiasm for prion research and the potential implications for human health.

Byron Caughey — PrP conformational conversions and the detection of prions

Ed Hoover gave a nice introduction to Byron Caughey. I learned that Dr. Caughey is a native of Ft. Collins and a CSU alum. His father, Winslow Caughey, was a professor at CSU. Dr. Caughey opened by noting that he was delivering this keynote lecture in place of Stanley Prusiner. In his characteristically self-effacing manner, Dr. Caughey quipped that he had tried to grow more hair in order to be an effective sub for Dr. Prusiner, but that it had grown on his face instead of his head.

Dr. Caughey opened with a general introduction to prions, which he defines as a “proteinaceous infections agent that lacks a nucleic acid genome”. He offered three requirements for a protein to be classified as a prion, which I’ll paraphrase as:

  1. It acquires self-propagating state that rarely accumulates spontaneously
  2. It replicates by recruiting the non-prion state of the same protein
  3. It can spread to naïve hosts, finding new sources of substrate protein

This last point requires that prions be able to escape their host and find new hosts. Using the example of chronic wasting disease, he listed the many routes by which CWD prions have been demonstrated to be transmissible [reviewed in Kraus 2013]. He raised the possibility of a spectrum of transmissibility. CWD, at one end of the spectrum, is readily transmissible between individuals in nature. Along the spectrum there may be other neurodegenerative diseases that involve proteins that acquire prion states that can be transmitted experimentally but are rarely or never transmitted in nature. At the other end there may be misfolded protein states that exhibit limited or no transmissibility.

A longstanding question in the prion field has been how strain identity is encoded in protein conformation. Similarly, it is unclear why synthetic amyloids made of recombinant PrP exhibit little or no infectivity, whereas purified brain-derived PrPSc fibrils are highly infectious. He then reviewed proposed structures of PrPSc. Most of the early structures included alpha helices, but it is now almost universally agreed that the true structure of PrPSc includes no alpha helix structure. He noted that other groups have reported evidence that synthetic recombinant PrP amyloids adopt a parallel in-register intermolecular beta sheet (PIRIBS) conformation [Cobb 2008, Tycko 2010]. This inspired Dr. Caughey and his group, in collaboration with Reed Wickner, to use solid-state NMR to query the structure of recombinant PrP amyloids derived from RT-QuIC reactions seeded with brain-derived prions. They have generated a PIRIBS model for these amyloids that is consistent with the solid-state NMR data [Groveman 2014]. This model also fits certain structural constraints that have been proposed for PrPSc, for instance based on the introduction of non-native disulfide bonds [Hafner-Bratkovic 2011]. The PIRIBS model would imply that the dimensions of amyloid fibrils seen on electron microscopy must be comprised of two intertwined “celery stalks” or “half-pipes”.

Dr. Caughey noted that infectious prions have a PK-resistant core from residue ~90-231 and require anionic co-factors [Deleault 2007, Wang 2010, Deleault 2012a, Deleault 2012b], whereas non-infectious PrP amyloids have a much smaller PK-resistant core, from ~160-231, and do not require co-factors. He therefore hypothesized that the “central lysine cluster” or CLC for short (KPSKPKTNMK, residues 101-110 in humans) could represent the group of positive charges that are responsible for binding anionic co-factors, and that neutralization of these positive charges by mutating the lysines to alanines or asparagines (K→A or K→N) might allow misfolding to proceed more readily. His group has now reported that mutating all four of these lysines (sets of mutations dubbed K4A or K4N) indeed allows recombinant PrP to adopt an amyloid state with a larger PK-resistant core, spanning ~20 kDa instead of just ~10 kDa for wild-type recombinant PrP [Groveman & Kraus 2015]. The CLC is in the most highly conserved portion of PrP’s amino acid sequence, and Dr. Caughey speculates this could be either due to its importance for PrPC’s native function, and/or because mutations here might increase the propensity to form PrPSc in vivo.

Next, Dr. Caughey pivoted to diagnostics for prion diseases. He reviewed the efforts by his lab and Ryuichiro Atarashi’s lab to develop RT-QuIC as a plate-based method for fluorescence detection of prion seeding activity [Wilham 2010, Atarashi 2011, Orru 2011]. The sensitivity of RT-QuIC reaches to 1 attogram (10-18 g) of PrPSc. Early tests on human CSF samples indicated diagnostic sensitivity on the order of 90% [Atarashi 2011, McGuire 2012], but Christina Orru has recently developed a new set of reaction conditions that appears to give more like 95% sensitivity for CSF samples [Orru 2015a]. The Caughey lab also collaborated with Gianluigi Zanusso, who for over a decade has been interested in the detection of PrPSc in the olfactory mucosa of Creutzfeldt-Jakob disease patients [Zanusso 2003]. Together, they found that performing RT-QuIC on olfactory mucosa brushings gave even better diagnostic accuracy than RT-QuIC on CSF, reaching ≥97% sensitivity and 100% specificity [Orru 2014]. RT-QuIC seems to be the closest thing so far to a definitive antemortem diagnostic test for CJD.

He then presented unpublished data which I can’t share here, but you can check out Christina Orru’s poster #P.47 tomorrow.

Dr. Caughey concluded by expressing his hope that RT-QuIC and similar assays will allow earlier diagnosis and discrimination between different protein folding diseases, and that it might one day even be used to monitor efficacy in clinical trials.

Reed Wickner — Yeast prions: how proteins template their conformation, anti-prion systems, prion clouds

Eric Ross, the sole yeast prionologist in the CSU prion group, introduced Reed Wickner. He recalled that a little over twenty years ago, [URE3] and [PSI+] were both considered to be “non-chromosomal genetic elements” with an unknown molecular basis. It was Dr. Wickner who realized that yeast have prions.

Dr. Wickner divided his talk into three stories, relating to structural biology [Gorkovskiy 2014], prion “clouds” [Bateman & Wickner 2013], and anti-prion systems [Wickner 2014]. Before beginning those stories, he introduced the four major proteins that will make appearances in his talk:

  • Ure2p has a native function related to nitrogen utilization. Its incorporation into [URE3] amyloid results in a loss of function.
  • Sup35p has a native function of translation termination. Its incorporation into [PSI+] amyloid results in a loss of function.
  • Rnq1p’s native function is unknown, but in its prion state, [PIN+], it can seed the formation of [PSI+] prions.
  • HETs is a protein of Podospora (the other three proteins above are of yeast). It achieves its native function only in the amyloid state, known as [Het-s], in which it governs heterokaryon incompatibility. He likened it to the major histocompatibility complex in humans.

The first three of these proteins can each adopt multiple different prion strains, whereas HETs has only one known prion state.

When Eric Ross was a student in Dr. Wickner’s lab, he experimented with shuffling the amino acid sequences of the Q/N-rich prion domains of Sup35p and Ure2p. They found that the formation of prions was sequence independent, even though transmission of prions required compatible sequences [Ross 2005a, Ross 2005b]. They ventured that the best explanation for this behavior would be if these yeast prions adopted parallel in-register intermolecular beta sheet (PIRIBS) conformations. In a PIRIBS conformation, the corresponding residues of each molecule are aligned (this is what “in-register” means), so sequence identity between molecules is critical. Meanwhile, the side chains of residues in PIRIBS hydrogen bond with the side chains of the corresponding residues in the molecules adjacent to them, rather than with other residues, thus shuffling the sequence makes little difference in the propensity to form prions. They have since found additional evidence in favor of the PIRIBS structure for yeast prions [Kryndushkin 2011]. It appears that differences in amino acid sequence dictate the positions of the folds in the PIRIBS structure [Gorkovskiy 2014].

Another protege of Dr. Wickner’s, David Bateman, collected Sup35p sequences from a variety of wild yeasts. He identified a number of naturally occurring polymorphisms in these yeast that hinder the transmission of [PSI+] prions [Bateman & Wickner 2012]. These polymorphisms may be analogous to the PRNP M129V polymorphism in humans. They then grew and subcloned a population of [PSI+] yeast for 75 generations, and found that on subcloning they were able to obtain divergent prion strains with differing abilities to infect yeast expressing the polymorphic wild Sup35p sequences [Bateman & Wickner 2013]. This is believed to arise from spontaneous mutation of prion conformations in the absence of selection. This argues in favor of the prion “cloud” hypothesis. Moreover, because laboratory experiments do inevitably select for living cells, studies of prions will tend to select for non-lethal prion variants.

In another set of experiments, Wickner and colleagues found that overexpression of Btn2p cures yeast of [URE3] prions [Kryndushkin 2008]. They hypothesized that the mechanism might be that Btn2p brings all of the prions together in one big clump, reducing the probability that one prion seed will segregate into a daughter cell. Indeed, even normal levels of Btn2p appear to reduce the number of prion seeds, as Btn2pΔ yeast retain prion seeds over a larger number of generations when grown in guanidine levels sufficient to prevent new prion propagation [Kryndushkin 2008]. They’ve since discovered that normal levels of Btn2p and Cur1 proteins are sufficient to cure most newly formed [URE3] prions, which have low seed numbers [Wickner 2014]. The [URE3] prions that cannot be cured by normal levels of Btn2p and Cur1 are those with elevated seed numbers, which Jonathan Weissman refers to as “strong” prion strains.

Dr. Wickner noted that his lab’s most recent work on yeast prion structural biology will be presented by Anton Gorkovskiy on Day 4 of this conference (talk O.20).