16 December 2004. To the student of neurodegenerative diseases, the Society for Neuroscience conference can seem like an overloaded buffet where only morsels of the food tickle one’s taste buds. By comparison, the satellite meetings offer a carefully selected menu with ample time to consume and digest new findings in good company. This was true of the 5th Neurobiology of Aging Conference, held on October 21 and 22. Organizers Ralph Nixon and Paul Coleman chose protein misfolding in Alzheimer’s and other age-related neurodegenerative diseases as the topic, and if full attendance and lively debate at the end of two intense days is any indication, this small conference was a resounding success.
Nixon, of New York University’s Nathan Kline Institute in Orangeburg, opened the meeting with a challenge. A growing number of scientists believe that the accumulation of incorrectly folded proteins is a molecular basis of many diseases, including Alzheimer’s, Huntington’s, Parkinson’s, prion diseases, as well as peripheral conditions and even some forms of cancer. The study of both protein misfolding and of defects in the cell’s protein disposal systems has come a long way in the last 15 years. Even so, knowledge gaps remain wide and deep, Nixon said. For one, whatever the investigators' favored pathogenic hypothesis, their talks tend to leave obscure how a given misfolded protein kills the cell. “Something happens to the proteolytic system under study, then comes a black box, then comes cell death. We need to know who the killers are and how the cells are dying,” said Nixon. For another, it remains unclear whether misfolding effects precede or follow synaptic deficits in AD pathogenesis. Consensus disease pathways remain an elusive wish for all of the major neurodegenerative diseases. To try to bridge those gaps, Nixon rallied the audience to speculate beyond their data and shake up collective thinking.
Approaching the problem from many different angles, the talks brought to mind the adage of the blindfolded scientits, each of whom touches a different part of the elephant and proclaims it to be a tree, a snake, or something else. The discussions did not quite gel into a clear picture of the elusive elephant, either, but as the next best thing, here are the component themes that resonated throughout the event and drew some consensus at the end:
- Oligomers of pathogenic proteins are universally toxic (Glabe talk). The field needs to clarify in-vivo mechanisms.
- Large aggregates are tombstones. They are a consequence of the disease process, not the cause. They are toxic, but only after soluble aggregates have wrought critical damage. This applies to Aβ, huntingtin, perhaps α-synuclein, possibly tau.
- Axonal transport blockages contribute to neurodegenerative diseases (Mandelkow, McMurray talks).
- The endosomal-lysosomal pathway shows abnormalities early, before inclusions/aggregates of pathogenic protein forms. This occurs in AD and HD (Nixon, McMurray talks). Changes in vesicular trafficking could be triggered by oligomers, kinases, or other events upstream from the aggregates.
- Different cellular protein degradation systems (proteasome, lysosomes) should be studied together (Cuervo talk).
- Chaperones are key to folding, but also other functions. Much remains to be learned (Frydman, Muchowski talks).
- A misfolded protein can sabotage its own elimination by creating a defect in a proteolytic system, or prevent the normal degradation of other proteins. Examples include α-synuclein (Cuervo talk) and ubiquitin (van Leeuwen talk). What’s more toxic to the cell?
- The proteasome plays a role in neurodegenerative diseases, but its place in the pathway remains unclear (van Leeuwen, Dawson talks).
What therapeutic prospects emerge from recent advances on protein misfolding in neurodegenerative disease? Here, again, are common themes and caveats:
- If pathogenic oligomers share a common toxic mechanism, could broad-spectrum treatments be developed against them? Could they be safe?
- Approaches influencing the levels of heat shock proteins might prevent proteins from misfolding.
- Neurodegeneration/cancer is the yin/yang of proteasomal degradation. Can we stimulate the proteasome without causing cancer?
- Quality-control degradative pathways in the endoplasmic reticulum (ER) represent an unexplored area for drug discovery. These pathways are multiple and able to recognize and remove abnormal proteins before they exit the ER and do their damage (Hampton talk).
- Stimulate lysosomal degradation. Caloric restriction can do it; are there other ways to beef up the system?
- With aging, overall protein degradation dwindles. Waste proteins accumulate and aggregates may only represent the tip of the iceberg. How specific should therapies boosting protein degradation be?
- Approaches to prevent aggregation are under way in AD. Extend them to other diseases?
Below are summaries of selected talks at this conference, and more discussion of the therapeutic ideas.
Fred Van Leeuwen
Up first was Fred Van Leeuwen from the Netherlands Institute for Brain Research in Amsterdam, who updated his hypothesis about molecular misreading and the ubiquitin proteasome system in neurodegeneration. The hypothesis proposes that frameshift transcription errors that occur continuously at a low level generate altered forms of certain proteins, including APP and ubiquitin B, which then contribute to the pathogenic process (see ARF related news story and Van Leeuwen et al., 1998). In the case of ubiquitin B, the altered form, UBB+1, cannot perform its normal function of ubiquitinating target proteins for degradation in the proteasome; instead, it becomes ubiquitinated itself. The resulting UBB+1-ubiquitin complex resists deubiquitination and could inhibit the proteasome (Lam et al., 2000). UBB+1 proteins occur in a variety of neurodegenerative diseases involving Aβ, tau, and huntingtin, but not α-synuclein (see De Pril et al., 2004 and Fischer et al., 2003).
At high doses, this protein inhibits the proteasome and causes cell death in culture. Its relevance in vivo, especially at low doses, has remained uncertain. Van Leeuwen addressed this issue by presenting initial data on transgenic mice that express UBB+1 transgenes postnatally in cortex and hippocampus. Both brain regions readily degrade low concentrations of UBB+1 (see Fischer et al., Soc. for Neuroscience, 2004 Abstract 336.1). They do not degrade high concentrations, however. By nine months of age, the high-expressing line of transgenics perform poorly in the Morris water maze and show proteomic changes downstream of UBB+1, (Van Dijk et al., Soc. for Neuroscience, 2004 Abstract 336.2.) This strain effectively represents a model for chronic proteasome inhibition, van Leeuwen added. Interestingly, a rat model of proteasome inhibition has generated recent interest in the Parkinson’s field because it mimics the disease more faithfully than do older models (see ARF related news story). Yet the comparison goes only so far, as the rats develop Lewy bodies, but Van Leeuwen’s group does not see UBB+1 and accompanying proteasome dysfunction in α-synuclein diseases.
Discussion after the talk centered on the question of whether UBB+1 accumulation and ensuing proteasome dysfunction is the first pathogenic event in AD, or it marks an earlier, unknown insult. Van Leeuwen noted that high concentrations of UBB+1 are required for proteasome inhibition, suggesting that other upstream compounds such as the ubiquitin-conjugating enzyme E2-25K/Hip-2 may be involved (Song et al., 2003). Clearly, Alzheimer disease is multifactorial, Van Leeuwen noted. Nixon asked what happens in the cell between proteasome inhibition and death? Attendants shared a sense that the proteasome field has not yet seen enough convergence and confirmation of data to allow conclusions about whether it plays a primary or secondary role in the disease process (for a review on this question, see Ciechanover and Brundin, 2003).
Ted Dawson illustrated one example of how emerging data are contradicting earlier, simpler hypotheses about the role of the proteasome in neurodegeneration. In Parkinson’s research, the spotlight fell on proteasomal degradation when it turned out that the parkin protein has an E3 ligase function (Shimura et al., 2001 and Zhang et al., 2000). Soon after that, potential substrates and interactors for parkin became known, and now the list includes synphilin-1, α-synuclein, CDCrel-1, p38, Pael-R, cyclin E, α/β-tubulin, synaptotagmin 11, among others. Researchers in Dawson’s lab at Johns Hopkins University in Baltimore are systematically testing which of these are relevant substrates in vivo, and whether their parkin-mediated degradation in the proteasome has an effect on disease. As a first step in this effort, Dawson reported that his group bred their parkin knockout mice (Von Coelln et al., 2004) to a strain transgenic for mutant human α-synuclein (Lee et al., 2002). Dawson reasoned that if parkin detoxifies α-synuclein via the proteasome, then its absence should worsen the phenotype of the α-synuclein mice. Alas, it did not. “We can take α-synuclein off the list of important parkin substrates,” Dawson said.
Next, Dawson tested synphilin-1, a parkin substrate and α-synuclein interactor that his group discovered (Chung et al., 2001). He noted that parkin does interact with synphilin-1, but that, at physiological concentrations, parkin appears to form alternate ubiquitin chains on amino acids other than the most studied one, lysine 48. In addition to tagging onto synphilin-1 conventional ubiquitin chains that lead to degradation, parkin also forms chains on lysine 63. These chains serve an unknown physiological function and lead to Lewy body formation, Dawson said. Taken together, Dawson does not view parkin as acting strictly in proteasome degradation. Instead, he believes it is a broad-spectrum neuroprotective agent, whose functions are determined in part by poorly understood modifications such as s-nitrosylation. —Gabrielle Strobel.
To be continued Friday, 17 December 2004.