Past Webinar
Will Probing Flies and Worms Bear Fruit in the Search for New Genes Involved in Alzheimer's?
Quick Links
Introduction
Mel Feany led this live discussion on 29 October 2002. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.
Transcript:
Live discussion held 29 October 2002, featuring Mel Feany of Brigham and Women's Hospital, Boston
Participants: Gabrielle Strobel (ARF); June Kinoshita (ARF); Mel Feany (Harvard Medical School, Boston); Brian Kraemer (University of Washington, Seattle); Josh Shulman (Harvard Medical School, Boston); Anne Hart (Massachusetts General Hospital/Harvard Medical School); David Shepherd (University of Southampton); Detlef Schmicker (Germany); Chris Link (University of Colorado, Boulder)
Note: Transcript has been edited for clarity and accuracy.
Gabrielle Strobel
Hi, Mel, guest of honor, so nice to "see" you again!
Gabrielle Strobel
Hello everyone, let's begin. I am Gabrielle Strobel, managing editor of the Alzforum and will be moderating today. Would everyone please just introduce themselves quickly? And obviously, don't worry about typing, we'll fix all that later in the transcript.
Mel Feany
Hi, I'm Mel Feany. I have a lab at Harvard Medical School and we model neurodegenerative diseases, including Alzheimer's disease and Parkinson's disease.
Brian Kraemer
Hi my name is Brian Kraemer (not a very imaginative nickname I know). I am from Seattle Washington. My email is kraemerb@u.washington.edu.
Josh Shulman
Sorry, I'm a postdoc in Mel Feany's lab, and I'm a medical student at Harvard...
Anne Hart
Hi, I am Anne Hart from Mass. General Hospital (hart@helix.mgh.harvard.edu).
David Shepherd
Sorry just disturbed by a visitor, this is David Shepherd at the University of Southampton in England.
Detlef Schmicker
Sure, Detlef Schmicker, Germany, second time in ARF discussion, start becoming interested in ARF due to father having Alzheimer, special interested in genetics at the moment because test is running.
Mel Feany
Welcome Anne and Josh. Thanks for coming. We have a great group already!
Gabrielle Strobel
And hello Anne, Chris and Joshua, I really appreciate that you could make it today where everyone is trying to get ready for Orlando, I suppose...
I have read your recent paper, Chris, and am excited to have you join us.
Chris Link
Thanks. Props go to Gin Fonte in my lab, who did most of the protein work, and our collaborator, David Friedman, who did the mass spec.
Mel Feany
Hi David. I'm glad you could come.
Anne Hart
I have often wished I could be two places at one time... but not twice in the same place.
Gabrielle Strobel
Let me start the discussion with a simple question: why is there no Drosophila model of Alzheimer's disease? Has anyone tried?
Mel, assuming your tau model is more one of tauopathies than of AD...
Mel Feany
Gabrielle, you raise a very interesting (and possibly controversial!) question. Since abnormal tau in the form of neurofibrillary tangles is one of the two key features of Alzheimer's pathology, we would argue that models created by expression of normal and mutant forms of tau are relevant to Alzheimer's disease. Maybe Chris could comment since he approaches modeling from the Ab perspective.
Brian Kraemer
What exactly do you mean by model? One might say there is no mouse model for AD either.
Gabrielle Strobel
Well yes, there is no complete model, but there are many mouse models that model aspects of AD and they do show overlapping, similar results, for all their faults.
Josh Shulman
Why, Brian would you say that there is no mouse model of AD?
Brian Kraemer
I was being facetious. If we agree that all models have their strengths and weaknesses, then I think the question is what can we learn from any given model.
Chris Link
My group has been trying to develop transgenic C. elegans models for AD and other neurodegenerative disease for about 10 years now (I guess we should know more than we do!) linkc@colorado.edu.
Gabrielle Strobel
Chris, what are the major difficulties? Besides measuring cognitive decline in roundworms...
Chris Link
I guess the difficulty is deciding what questions you specifically want to answer, and what is the best way to make the little worms do your bidding. I always try to stress that we are trying to model ASPECTS of AD. I think two aspects of AD that we can get at with worm models are the basic cell biology of Ab metabolism and toxicity.
Gabrielle Strobel
And Mel, have you tried modeling the amyloid side as well, and then crossing those models?
Brian Kraemer
So Mel, What do you think we can learn from the fly models for tauopathy(and Parkinson's)?
Josh Shulman
If you believe that all or at least some of the toxic effects in AD are mediated by tau, then models based only on tau can offer important lessons.
Mel Feany
Yes, Gabrielle, we tried to follow Chris' important lead by expressing Ab in flies. So far, the pathology is not impressive in our animals as compared to Chris' worms.
Gabrielle Strobel
Mel, are you expressing it intracellularly? I mean by that, is it present inside neurons or is all of it secreted?
Mel Feany
We have tried to express Ab both intracellularly and extracellularly (secreted) with about the same results.
Gabrielle Strobel
I am asking that because Sue Lindquist just had two papers in Science reporting extraordinary toxicity of a misfolded cytoplasmic prion protein, and Chris's paper about the Ab-chaperone interaction brought up speculative echoes in Ad...
Anne Hart
Has Mel tried expression in muscles and Chris tried expression in neurons?
Mel Feany
Yes, Anne, we have tried to express Ab in muscle. We are still looking at those animals, but there is nothing obvious so far. Chris, any thoughts on why Ab (we used your constructs among others) is apparently less toxic and less prone to aggregate (we do not see Thioflavin or Congo red positive material) in fly muscle compared to worm muscle?
Chris Link
Mel—animals with induced expression of Ab in muscle show rapid paralysis, but actually do NOT have significant fibrillar deposits. (I do suspect that they have some toxic conformer/oligomer, however). The difference in the fly and worm muscle phenotypes might have to do with the specific chaperone responses. Mel—have you looked at any HSP proteins in flies with muscle Ab?
I think we may have lucked into a particularly pathological situation with muscle expression of Ab. Our lines with pan-neuronal expression of Ab do not show as extreme a pathology as the muscle expression lines. I suspect this is telling us something about IBM, and perhaps by extension, AD.
Mel Feany
Chris makes the important point that model organisms address specific questions in neurodegeneration. Of course the key utility of flies and worms (in my view, at least) is the ability to perform forward genetic analysis.
Gabrielle Strobel
Mel, you wrote in the backgrounder that you have found some modifiers with forward genetics. Can you tell us anything about them?
Nancy Bonini has an abstract in Orlando about genetic and pharmacologic modifiers that prevent dopaminergic neuron loss in her Drosophila PD model. I am sure we are all curious what those could be.
Josh Shulman
If neurons degenerate in a particular model, are "cognitive" phenotypes, or any behavioral phenotype for that matter, important?
Chris Link
Josh asks about cognitive phenotypes in the worm and fly models. My bias is that these models should be informative without having to look at subtle behavioral differences. AD is ultimately a SEVERE brain disease with significant neuronal loss—this is what we most want to understand, and I think the most informative models will reflect this.
Josh Shulman
I agree, Chris. With robust and easily-scorable neurodegeneration in Tau flies, I personally don't think it would be particularly useful to chase after behavioral outcomes.
David Shepherd
I agree with Chris here, cognitive loss is not an issue. It's about understanding why neurons die!
Mel Feany
I think that Chris brings up an important point regarding the nature of neurodegenerative disease—that the most important feature from the pathologic perspective is the loss of neurons. Can we have a real model of a neurodegenerative disease without cell death?
Gabrielle Strobel
Perhaps cell death is slow, and AD starts with synaptic dysfunction?
Brian Kraemer
Are we sure that neuronal loss is really the most important aspect of disease? Might it not be the case that pathological changes occur years or decades before neuronal loss? Sure, I would agree that it is the neuronal loss that kills, but what causes the cognitive impairment and memory loss?
Josh Shulman
Is it not clear that cell loss in the hippocampus tracks with cognitive changes and memory loss, even in early to moderate stages of AD?
Chris Link
Hopefully, the neuronal dysfunction and eventual neuronal death are part of the same continuum, so if we can understand one, we can understand the other. Perhaps a related point is whether apoptosis plays an important role in AD. Although I would not be surprised if many neurons in AD do eventually undergo apoptosis, I would bet this is actually pretty far downstream of the initial toxic insult, which might be critical to the early neuronal dysfunction.
Mel Feany
Chris, given your results, we should look at HSP regulation. Our approach has been to look for significant pathology and phenotypes that lend themselves to second site modifier analysis. However, creating a fly model of Ab toxicity would be very important, and manipulating HSPs might help us characterize the system.
Gabrielle Strobel
I am glad Chris mentions the chaperone connection. Have you come across chaperones in your polyQ huntingtin model, Anne?
Anne Hart
We have not directly tested chaperones—but they have not come up in our genetic screens. Heat shock of animals expressing the expanded huntingtin fragment does not change toxicity.
Mel Feany
Anne, could you comment on your negative results with HSPs given the extensive literature on HSPs and polyglutamines in flies?
Anne Hart
Our genetic screens to date have looked for genes whose loss of dramatically exacerbate polyQ toxicity. Heat shock proteins may [be] redundant in our system; loss of function in any one gene may not have dramatic consequences. Alternatively, our level of expression, while deleterious to the neurons, may not be high enough to induce a heat shock response.
Chris Link
Mel—in worm muscle, Ab turn over seems very slow (non-existant?). I wonder if inhibition proteasome function in your flies might exacerbate phenotypes.
Gabrielle Strobel
Chris and Mel, Sue Lindquist showed high toxicity of an otherwise rapidly degraded misfolded prion protein when she temporarily blocked the proteasome.
Is Elan doing any work on fly and worm models?
Paul Shapiro
Elan is looking at these models.
June Kinoshita
Paul, from the perspective of drug screening, what types of fly and C. elegans models are most interesting?
Gabrielle Strobel
Paul, is Elan screening compound libraries against such models?
Paul Shapiro
Gabrielle, No we are not currently screening compound libraries against such models.
Gabrielle Strobel
Mel, would it be interesting to cross the flies expressing normal human tau in cholinergic neurons to ones expressing Ab (assuming they express it in the same cells), even if the phenotype in the latter is not impressive? It would be so interesting to get at how those two pathways intersect in the absence of FAD mutations!
Mel Feany
Gabrielle, we have done some preliminary crosses with our Ab flies with tau flies. So far, nothing too dramatic. Has anyone else (maybe Chris or David) tried these experiments?
Chris Link
Gabrielle—we have been trying pretty hard to study animals co-expressing tau and Ab in the same cells, but don't have anything interesting yet. Wildtype expression of human tau in worm neurons is sufficient to cause an "uncoordinated" (neuronal) phenotype, so we are trying to look for something beyond this.
Gabrielle Strobel
Anne I see you have a presentation in Orlando about the role of Creb, CBP and histone deacetylase in polyQ huntingtin-mediated neurotoxicity. Can you tell us a little bit about it?
Anne Hart
Our preliminary results suggest that loss of CREB function does not change polyQ toxicity. We have only tested one HDAC loss of function mutation...and it also had no effect. The CBP story is not yet clear; we are still assessing various loss of function and dominant negative alleles.
Josh Shulman
While it is important to test hypothesis about pathways such as apoptosis, proteasome, and chaperones, I still think the greatest contribution of worm and fly models will be the completely unexpected and novel modifiers that come from forward genetic screens.
Gabrielle Strobel
We are sitting on the edge of our seats, Josh...!
Josh Shulman
Our screens have identified several well-conserved, novel proteins as strong enhancers or suppressors of Tau.
Gabrielle Strobel
That's what I mean, you are making us so curious.
David Shepherd
What are these enhancers and supressors?
Josh Shulman
We are curious too....the difficulty is in figuring out what such proteins do....of course this is all still preliminary.
Gabrielle Strobel
Could it be the case in AD, like some suggest in huntingtin, that formation of intracellular inclusions, i.e. tangles, sequesters away proteins needed for transcription, or other functions? Can this be studied in your flies, Mel?
Mel Feany
Actually, Gabrielle, we are pretty excited about the idea that large inclusions sequester smaller (monomers/oligomers/protofibrils) toxic species and are therefore protective rather than toxic. I think fly models are ideal to test the role of inclusions because we can combine insights from forward genetic screens with genetic reagents designed to alter ubiquitin/proteasome and heat shock protein expression.
June Kinoshita
Mel, re: screening for modifier genes, I'm fascinated by the toxicity seen when APP is overexpressed. Have you looked for modifiers of that toxicity? (Also, have you tried inducible overexpression to get around the developmental effects?)
Mel Feany
June, re APP toxicity, we and others see significant toxicity in the nervous system and in non-neuronal tissues when we express full length APP (normal or AD mutation-linked). Expression of fly APP (which lacks Ab sequences) is also toxic. I think the challenge is to determine the relevance of these phenotypes to disease and/or the normal function of APP. Maybe Chris can comment.
Gabrielle Strobel
A quick search of the Neuroscience abstracts of these diseases and Drosophila and C. elegans did not net any hits. It might be that the search engine is weak, but it made me wonder: are there really only a handful of labs working on fly and worm models of neurodegenerative diseases? Basically present company and the Bonini/Trojanowski lab?
Brian Kraemer
We are working on a worm model for tauopathy in collaboration with the Trojanowski lab.
Chris Link
Among the worm contingent, Ralf Baumeister has done some APP and poly Q work, and Rick Morimoto's group has also published a couple of polyQ papers. I think some people working with Jim Thomas are also looking at worm tau models.
Anne Hart
I suspect more fly/worm models will appear in the next few years given these precedents. It is sometimes difficult to get support to start such "high risk" research... the foundations have a dramatic impact in these cases!
Gabrielle Strobel
Chris, you mentioned APP processing and the role of other cleavage products as one of the areas worm models could address. What would help get more of such projects started?
Chris Link
Gabrielle—First, you have to believe the other APP products are important to AD! Technically, it would be easy to directly express C-terminal fragments in worms. I forgot to mention that Chris Li has been working on the worm APP homolog apl-1. I don't know if human APP can rescue apl-1 loss-of-function mutations.
Gabrielle Strobel
Mel, do you think your flies would have less neurodegeneration if they were able to make full-fledged tangles, then?
Mel Feany
Gabrielle, yes, we thought that our modifiers might affect tangle formation and therefore help dissect the role of tangles. Unfortunately, none of the modifiers we have examined so far have any effects on tangle formation, although they do alter phosphoepitopes related to neuronal death in our system.
Gabrielle Strobel
Mel, very interesting. So absent a role in tangle formation, the modifiers interfere perhaps with the direct mechanism/pathway of toxicity?
Mel Feany
Gabrielle, we do think that the relevant pathway of tau toxicity does not necessarily include tangles (because tau is toxic in our system without tangle formation). However, I think the key test for all the fly and worm models of neurodegenerative disease will be making the connection back to human disease. So far, we are trying to relate our modifiers back to human disease by determining if any of the human homologs of our fly modifiers are linked to familial forms of Alzheimer's or Parkinson's disease.
June Kinoshita
Mel and Chris, there's interest now in the APP intracellular domain (AID or AICD). What have flies and worms shown to date re: AID/AICD?
Chris Link
The identification of FAD mutations in the Ab sequence (e.g., the "Dutch" and "Arctic" mutations) seems to support the notion that Ab is central. I think what we really need is for the mouse people to make models that JUST express Ab—this is a good way to disentangle APP from Ab toxicity.
Gabrielle Strobel
Chris, are there not mice that express just Ab (Frank LaFerla's 1995 paper showing apoptotic neuronal death) and some by Lennart Mucke expressing APP minigenes?
Chris Link
Gabrielle—I don't recall the details of those papers, but my memory is that they did not use simple signal peptide/Ab constructs that would get around requirements for processing, etc. I will look them up!
Mel Feany
Chris, do experiments injecting Ab in mice adequately address the APP vs Ab issue?
Brian Kraemer
Can these models be used to determine the root similarities of all diseases with toxic protein aggregates (ie AD, PD, Huntington's, etc)? Will this naturally come out of the modifier screens?
Mel Feany
Brian, we are particularly excited about the potential to explore similarities and differences by comparing genetic modifiers among different disease models. We have to keep in mind the caveat raised by Anne, though, in regard to HSPs and her model. Failing to identify a pathway as a mediator of toxicity does not automatically mean that the pathway is not important. Issues of genetic redundancy are particularly problematic here.
Gabrielle Strobel
All, we will soon host another chat, with Gunnar Gouras, about intracellular Ab and connections to misfolding, aggregation, toxicity. Join us for that, as well!
June Kinoshita
And we will be posting an interview with Sue Lindquist. Much food for thought!
Brian Kraemer
Mel, I agree comparing the results of genetic screens between flies and worms would be interesting and one organism may give modifiers not detected in the other due to redundancy etc. We are not yet at the stage of doing genetic screens, although that is certainly on my mind.
Chris Link
Mel—I guess there is still no consensus how much AD involves "outside-in" Ab toxicity. I could imagine that intracellular (never-secreted) Ab could play an important role, and this might not be replicated in the injection experiments. Has anyone injected Ab into (young, pre-plaque) mice?
Gabrielle Strobel
Chris, that is why Sue Lindquist's papers are so interesting. Miniscule, barely detectable amounts of misfolded cytoplasmic intraneuronal prion protein killed her mice. This is pure speculation but if such a species were at work in human AD neurons, we would never know.
Brian Kraemer
So Chris, have you done any screens for suppressors of Ab toxicity?
Chris Link
Brian, if we hadn't, I would have lost my worm certification! Unfortunately, we have not identified the suppressors we have, so we really haven't learned anything yet. A full-genome feeding RNAi screen is under consideration.
Gabrielle Strobel
Mel, in that vein, have any genes related to cholesterol come up in your screens? There is a slide talk in Orlando about an enzyme in the chol pathway located under the chromosome 10q linkage peak [that]is associated with AD risk in one sample.
Mel Feany
Gabrielle, not really. We have tried some candidate testing in lipid pathways, but nothing striking has emerged.
Detlef Schmicker
Is there any genetic linkage on NMDA receptors for AD, some AD drugs work on NMDA?
Brian Kraemer
Has anyone done drug screening on flies or worms? It seems to me that a high throughput screen of a small molecule library for inhibitors of toxic protein aggregation might be feasible.
Chris Link
Mel makes a great point—the key proof will be in human genetics (just like the FAD cases made APP/Ab central). In reference to the previous discussion about chaperone proteins, there is an HSP40 at 10q23—could this be the AD locus thought to be on chromosome 10?
Gabrielle Strobel
Hmm, interesting. I do not know its precise address, unfortunately.
I have to drop out. Please carry on as long as you like and thanks again! Hope to see you in Orlando.
Chris Link
Brian—we have tried a few compounds, looking to block amyloid formation, with no luck. The dirty little worm secret is that MANY compounds do not efficiently get into worms, so negative results are hard to interpret.
Mel Feany
I want to thank Gabrielle and Nico for all their hard work putting the chat together! Thanks also to everyone for all their comments.
Background
Background Text
By Mel B. Feany
The progressive loss of specific neuronal populations is what is thought to cause the major symptoms and early death seen in human neurodegenerative diseases. The identification of causative single gene mutations in families with inherited forms of these disorders has facilitated modeling of these diseases in experimental organisms, including the fruit fly Drosophila melanogaster and the roundworm Caenorhabditis elegans.
Many neurodegenerative diseases have now been modeled in Drosophila and C. elegans, including Alzheimer's, Parkinson's, Huntington's, human tauopathies, and several of the spinocerebellar ataxias. In our lab, we have introduced normal and mutated human tau into fruit flies, which then develop progressive neurodegeneration (see news). We have also found normal and mutant forms of human α-synuclein induce degeneration of dopaminergic neurons (see news). Second-site genetic analysis, i.e. looking for additional sites that enhance or suppress the primary phenotype, is currently ongoing for many of these models. Our group has identified numerous modifier genes, some expected, some novel, which we are now analyzing. This is where the strength of Drosophila genetics becomes apparent, as genetic-modifier screens yield results faster and more cheaply than in vertebrate models. Fruit fly modifiers can then be checked for mammalian homologues. In this way, fly genetics represents one strategy to try to identify genes involved in Alzheimer's that are now implicated by linkage peak analysis only. The table (to be posted) summarizes the main features of the currently available fly and worm models.
The approach relies on the assumptions that fruit fly genetics can indeed define the molecular pathways underlying human neurodegenerative processes and that it provides new therapeutic targets and drug screens.
Despite extensive research in vertebrate systems, many questions regarding the pathogenesis of neurodegenerative diseases still remain. What are the cellular pathways mediating neurodegeneration? What underlies the remarkable tissue and cellular specificity seen in these diseases? What is the role of inclusion bodies?
In our discussion it may be useful to consider the role that Drosophila and C. elegans models have in addressing these key questions.
- What are the strengths of the models?
- What features of the human diseases should simple models recapitulate?
- What are the limitations of the models? For example, the extensive anatomic differences between the fly and vertebrate motor systems might limit this model's contribution to answering questions about how loss of specific populations of dopamine neurons causes the abnormal movements in Parkinson's disease. Are there other differences between vertebrates and invertebrates that will restrict the utility of fly models of neurodegenerative disease?
- What other questions could be addressed with these models? AβPP processing and the downstream fate of AβPP's cleavage products, or intracellular actions and binding partners of Aβ come to mind.
Editor's Recommended References:
Link CD. Expression of human beta-amyloid peptide in transgenic Caenorhabditis elegans. Proc Natl Acad Sci U S A. 1995 Sep 26;92(20):9368-72. Abstract
Fonte V, Kapulkin V, Taft A, Fluet A, Friedman D, Link CD. Interaction of intracellular beta amyloid peptide with chaperone proteins. Proc Natl Acad Sci U S A 2002 Jul 9;99(14):9439-44. Abstract
Faber PW, Alter JR, MacDonald ME, Hart AC. Polyglutamine-mediated dysfunction and apoptotic death of a Caenorhabditis elegans sensory neuron. Proc Natl Acad Sci U S A. 1999 Jan 5;96(1):179-84. Abstract
Jackson GR, Wiedau-Pazos M, Sang TK, Wagle N, Brown CA, Massachi S, Geschwind DH. Human wild-type tau interacts with wingless pathway components and produces neurofibrillary pathology in Drosophila. Neuron 2002 May 16;34(4):509-19. Abstract
Muqit MM, Feany MB. Modeling neurodegenerative diseases in Drosophila: a fruitful approach? Abstract
Link CD. Transgenic invertebrate models of age-associated neurodegenerative diseases. Mech Ageing Dev 2001 Sep 30;122(14):1639-49. Abstract
Chan HY, Bonini NM. Drosophila models of human neurodegenerative disease. Cell Death Differ 2000 Nov;7(11):1075-80. Abstract
Further Reading About Flies' and Worms' Contributions to AD Gene Discovery:
Tickoo S, Russell S. Drosophila melanogaster as a model system for drug discovery and pathway screening. Curr Opin Pharmacol 2002 Oct;2(5):555. Abstract
Li J, Pauley AM, Myers RL, Shuang R, Brashler JR, Yan R, Buhl AE, Ruble C, Gurney ME. SEL-10 interacts with presenilin 1, facilitates its ubiquitination, and alters A-beta peptide production. J Neurochem 2002 Sep;82(6):1540-8. Abstract
Zambrano N, Bimonte M, Arbucci S, Gianni D, Russo T, Bazzicalupo P. feh-1 and apl-1, the Caenorhabditis elegans orthologues of mammalian Fe65 and beta-amyloid precursor protein genes, are involved in the same pathway that controls nematode pharyngeal pumping. J Cell Sci 2002 Apr 1;115(Pt 7):1411-22. Abstract
Gunawardena S, Goldstein LS. Disruption of axonal transport and neuronal viability by amyloid precursor protein mutations in Drosophila. Neuron 2001 Nov 8;32(3):389-401. (Related news)
Chung HM, Struhl G. Nicastrin is required for Presenilin-mediated transmembrane cleavage in Drosophila. Nat Cell Biol 2001 Dec;3(12):1129-32. Abstract
Zhang SX, Guo Y, Boulianne GL. Identification of a novel family of putative methyltransferases that interact with human and Drosophila presenilins. Zhang SX, Guo Y, Boulianne GL. Gene 2001 Dec 12;280(1-2):135-44. Abstract
Anderton BH. Alzheimer's disease: clues from flies and worms. Curr Biol 1999 Feb 11;9(3):R106-9. Abstract
References
News Citations
- In Fly Model of Tauopathy, Neurons Degenerate Without Tangles
- "Parkinson's" in a Transgenic Drosophila
- Axonal Transport Suggested as Function for APP
Webinar Citations
Other Citations
External Citations
Further Reading
No Available Further Reading
Comments
No Available Comments
Make a Comment
To make a comment you must login or register.