Summary

Vikram Khurana and Ole Isacson led this live discussion on 29 June 2005. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.

Transcript:

Live Discussion led by Ole Isacson and Vikram Khurana on 29 June 2005.

Participants: Chee Yeun Chung, McLean Hospital/Harvard Medical School; Ole Isacson, Harvard Medical School; Vikram Khurana, Harvard Medical School; Xinkun Wang, University of Kansas; Mark Smith, Case Western University; George Martin, University of Washington, Seattle; Paul Salvaterra, Beckman Research Institute, City of Hope; Keith Crutcher, University of Cincinnati; Dewitt James Lowell, retired; Sangram Sisodia, University of Chicago; Richard Fine, Boston University School of Medicine; June Kinoshita, Alzheimer Research Forum; Tom Fagan, Alzheimer Research Forum.

Tom Fagan
Hi all, I'm Tom Fagan, from Alzforum.

George Martin
Greetings from Seattle.

Vik Khurana
Hi, Tom. Thanks to you, June, and Nico for arranging this chat.

June
Hi, everyone. Hi, Vik and Ole. Thank you so much for agreeing to lead today's discussion.

Vik Khurana
No problem. Thanks for inviting us, June.

June
Shall we get started?

Vik Khurana
 Sure.

June
Perhaps Ole can start things off by quickly summarizing what his group found with regard to Parkinson disease (PD) neuronal vulnerability.

Ole Isacson
We found that the neurons in substantia nigra (SN) that die normally have a different gene expression profile than those that survive.

June
Are these differences found literally between neurons that are right next to one another, or are they differences between distinct adjacent regions in the SN?

Ole Isacson
Yes, using laser capture microscopy we could define the cells even when they were adjacent to each other. This gave us a "phenotype" or "character" of the cells; this provided molecular clues for response to injury.

Vik Khurana
Interestingly, in the study Ole refers to (see background text and Chung et al.), innate differences between cells influenced survival to PD-relevant toxic insults in primary culture.

Paul Salvaterra
How can you tell if differential gene expression is a cause or an effect from dying?

Vik Khurana
Paul, the study was in normal mouse tissue.

Xinkun Wang
Paul, I guess the in vitro study partially answered your question, i.e., they are part of the cause(s). Chee Yeun, Ole, I like that you tried some identified genes in cell culture and they did affect neuronal vulnerability.

Vik Khurana
Paul, I think you also raise the important issue of validation. I think eliciting differences is a start, but validation in vivo is also critical, and GAL4 and Cre/Lox techniques may be useful in this regard in flies and mice, respectively.

Paul Salvaterra
I have always been confused about how to interpret genetic neurodegenerative diseases. The mutant gene expression does not seem to correlate with the types of cells that die. Is this true in general? If so, it could mean that there must be something else to generate the cell death phenotype. Is this the angle for trying to define differential gene expression patterns in different types of neurons? What are the other gene candidates? Are they restricted to Parkinson models?

Vik Khurana
Paul, I think that's a fascinating aspect of this…that a similar toxic insult has different effects in different neurons is a key finding. The fact that animal models—even flies!—recapitulate this makes them especially useful.

Paul Salvaterra
Vik, I'm not sure how the fly models recapitulate cell type specificity. What is the best evidence for cell type specificity in any animal model, in your opinion?

Ole Isacson
Paul, the fact is that most of the mutated proteins that eventually kill some cells are expressed in most cells of the body. For example, in Huntington disease, the mutated protein also disturbs function of skin cells in patients (according to Seo et al., 2004). We think that only a few cells in the body cannot cope with the protein—and they are the ones that we try to distinguish from the ones that can—by looking at gene and protein expression profiles.

Vik Khurana
Paul, in fly models it's a little trickier to relate back to human because the fly brain is wired completely differently. However, parallels can be drawn—for example, in some spinocerebellar ataxia (SCA) diseases in which cholinergic neurons die, fly cholinergic neurons in the model (in the lamina) are also particularly vulnerable.

George Martin
Ole and Chee Yeun, it is wonderful to see this initiative in trying to unravel the mystery of differential neuronal susceptibility to neurodegeneration. I have a couple of questions—one technical and the other conceptual. The technical question is prompted by some proteomics friends who say there may be marked discordances between levels of mRNA as seen in expression arrays and levels of protein.

Ole Isacson
Chee Yeun will describe the technical reasoning for this.

Chee Yeun Chung
To answer George's questions, I still think that the mRNA level differences mean something. But it would be interesting to know protein level differences. But proteomics using laser capture microscopy (LCM) is not very advanced at the moment. It requires the capture of a lot more neurons....

George Martin
Here is my conceptual question: What if we are successful in engineering pars compacta dopaminergic cells with gene actions similar to those of resistant neighbors? Would there not then be some functional trade-offs?

Chee Yeun Chung
To add more to George's technical question, we are doing immunohistochemistry to verify some of the candidates in terms of their protein levels. Within the candidates I tried, mRNA level differences go together with protein differences. We are doing functional study after validation of their protein levels by immunohistochemistry.

Mark Smith
Ole, Vik, and Chee Yeun, what age were the mice and do you think that things might change with age? PD is age-related, after all.

Ole Isacson
The mice were adult (7-8 months). Aging does change a number of things...I am particularly interested in the damage to mitochondria as we age. I think significant functional loss of some mitochondria may occur, on average, at 45-50 years of age in humans? This could precipitate the disease as the cells have a harder time to cope.

Mark Smith
Ole...I agree on the mitochondrial aspect...but a swine to put your finger on mechanistically speaking! Ole, is this the age of peak MPTP toxicity? [The chemical MPTP, or 1,2,3,6-tetrahydropyridine, is known to cause Parkinson disease in both humans and mice; see ARF related news story.]

Vik Khurana
Mark, you have studied GluR expression differences in AD...AD is generally considered a more diffuse type of neurodegeneration than some other diseases. Is this just an understudied area or is AD really different?

Mark Smith
Vik, AD, as a disease, is certainly more widespread than PD; however, both share selective neurodegeneration. Your study will go a long way in figuring this out; it seems to have been something forgotten since the cholinergic hypothesis.

June
To Vik and Dewitt, in AD it has long been noted (by Paul Coleman, in particular, see Buell and Coleman, 1979 /new/detail.asp?id=214) that you will see in human AD brains healthy looking neurons right alongside those that are filled with tangles. I agree with Mark that there is selective neuronal vulnerability, most obviously in early stages of the disease.

Dewitt James Lowell
Have you studied any people? To all participants.

Chee Yeun Chung
To Dewitt, we verified some of mouse results in human using laser capture and real-time polymerase chain reaction (PCR).

Tom Fagan
Ole, Vik, Chee Yeun, are these innate differences present in the absence of any pathology?

Vik Khurana
Yes, Tom. The point of the study was to show how normal differences between neurons may influence disease vulnerability in disease states.

Xinkun Wang
Why do different neuron populations die in different neurodegenerative diseases? To everybody.

Paul Salvaterra
Xinkun, I think there are two general answers. The neurons may be intrinsically different—and the gene expression studies could address this—or the cells that die may be interacting with a toxic microenvironment. Anyone else have other ideas?

Vik Khurana
Paul and Xinkun, I've classified these as cell intrinsic and surrounding factors in the background reading. However, surrounding factors should also be expanded to include potential developmental factors (or environmental exposures) that could prime certain neuronal populations to die.

Mark Smith
Xinkun, my guess is selective vulnerability (mediated by gene expression differences as shown so well in this paper) to different insults. Once we figure out differences between populations, we can figure out what kills them (since they should be selectively vulnerable/resistant).

George Martin
Paul and Xinkun, a third formal possibility is that there may be differential cell-cell interactions.

June
George, I agree, and I think these may be quite difficult to get a handle on. Any thoughts about how to tackle this question?

George Martin
June, a good question. I think we will require a great deal more research on characterizing the interacting cell types. There could, for example, be more classes of glial cells than we know about.

Xinkun Wang
Good points, everyone. The cell-cell interactions seem much understudied but very interesting.

Vik Khurana
George and June, to continue this point, a recent study (Barroso-Chinea et al., 2005) looked at glial-derive neurotrophic factor (GDNF) production in the striatum influencing differential vulnerability in midbrain, presumably due to different GDNF uptake in vulnerable and resistant cells.

Mark Smith
Are there knockouts of any of the "protective/susceptible" genes? Would you then expect more widespread neurodegeneration with, say, MPTP?

Paul Salvaterra
George and June, wouldn't the fact that neighboring (touching, interacting?) neurons have different neurodegenerative fates make cell-cell interactions less likely as a cause?

Vik Khurana
Paul, I think you also have to consider the long distances over which neurons project. So it could be the neurons they project on to are more important than neurons anatomically adjacent.

Paul Salvaterra
Vik, I agree; that is what I meant by "toxic" microenvironment.

Tom Fagan
On cell-cell interaction, Don Cleveland has produced ALS-like symptoms in mice by putting mutant superoxide dismutase in glia, as most of you probably know (see ARF related news story).

Vik Khurana
June, I think the cell-specific expression systems might be very useful. For example, one could envisage expressing mutant proteins in glial cells alone and determining the effect on neuronal vulnerability in surrounding populations....

Ole Isacson
George, yes, cell-cell interaction is likely very important over time. As neurons (and glia) interact minute by minute, the gene expression as well as synaptic relationships change. One experimental advantage (for once) in cell culture was that the dopamine neurons we tested (A9 and A10) actually kept their relative vulnerability also when isolated from their normal cell-cell contacts.

Chee Yeun Chung
There is a recent paper in Neuron (Gu et al., 2005) which described controlling the expression of mutant huntingtin protein to different cortical neurons. It claims cell death of the pyramidal neurons needs interaction with other cells.

Tom Fagan
Chee Yeun, we covered that in our news section; see ARF related news story.

Vik Khurana
I also wanted to raise the issue to all about how insights from human patients may help tackle this question. For example, I think it's fascinating that patients with the same mutation can have very different phenotypes. Presumably genomic differences and environmental exposures in these patients would be fascinating (and difficult!) to study....

Dewitt James Lowell
Can genes, as neurodegenerative diseases develop, change and reproduce enough distorted structural forms to both cause regional differences and adjacent neuron differences?

George Martin
Vik, one answer is the powerful impact of the background genome on the phenotypic expression of a single gene mutation.

Xinkun Wang
I think oxidative stress must play an important role here. For example, in the hippocampus, treatment of pyramidal neurons with superoxide-generating chemicals reproduced the selective death pattern that occurs in ischemia and neurodegenerative diseases, i.e., CA1 neurons are sensitive while CA3 neurons are resistant.

June
Vik, there are some studies now going on in large families with presenilin 1 (PS1) familial AD mutations to try to determine factors that affect age of onset.

Vik Khurana
June, I think age of onset is interesting, but with regard to this question, different patterns might be fascinating to look at, also. For example, how can patients with the same synuclein or Park 8 mutation manifest a PD versus a Lewy body dementia phenotype?

George Martin
An interesting and unexplored question is the possibility that different neuron types have variations in the extent of what might be called gene expression excursions-oscillations in gene action from moment to moment; those neuronal types with wide excursions might be particularly vulnerable.

Vik Khurana
George, that's a fascinating idea. Have people looked at this in cells responding to disease-relevant toxic insults?

George Martin
Vik, not to my knowledge. It is essentially unexplored territory.

Vik Khurana
George, do you have any references describing the study of excursion-oscillations?

Paul Salvaterra
Vik, all circadian genes have oscillations that are well-documented. Most genes in yeast also oscillate.

George Martin
Vik, not that I can come up with immediately. I imagine that chromatin is a "breathing," dynamic structure.

Vik Khurana
George, perhaps in this regard, differences in the "chromatin code" (e.g., the histone code described by David Allis at Rockefeller and others) might be interesting to examine in susceptible and resistant neuronal populations.

George Martin
I have to go to another meeting. I have enjoyed this exchange.

Paul Salvaterra
I have a question related to Xinkun's comment. Would the laser dissection-gene expression studies be "good" enough to see quantitative changes in gene expression or are they more suited to qualitative differences? The quantitative changes may be more related to oxidative phenomena.

Chee Yeun Chung
Paul, the mRNA differences we found using microarrays are quite consistent among species, and we validated with real-time PCR, which is very quantitative.

Paul Salvaterra
Chee Yeun, wouldn't the "validation" also be subject to the same amplification artifacts as the original samples used for the microarray?

Chee Yeun Chung
Paul, the real-time PCR was done in un-amplified samples.

Paul Salvaterra
Chee Yeun, I am amazed that amplification does not induce a bias. I am also amazed that any microarray study would provide anything but crude quantitative estimates.

Chee Yeun Chung
Paul, sorry if I confused you. I agree that microarray study with amplification materials only provides crude differences. That is why we tried to verify our microarray results with real-time PCR where we used "non-amplified" LCM samples, which exclude bias from amplification. To Xinkun, there is a study in which gene expression profiles of CA1 and CA3 neurons were compared using LCM and microarray. You may find that interesting.

Xinkun Wang
Chee Yeun, I think I have read that paper (see Torres-Munoz et al., 2004).

Paul Salvaterra BR> I think there are many correlative studies and LCM/microarray is an interesting and new approach. What we need is to move into studies that show necessity for neurodegenerative disease (ND) and sufficiency. We need to know how neurons die in ND and looking at selective vulnerability is one way to get new clues.

Vik Khurana BR> Paul, I think using genetic approaches in mice and flies is a good start to look at the issues of causality that you raise....

Dewitt James Lowell
My two queries were to all participants.

Tom Fagan
Chee Yeun, would you hazard a guess, based on your data, as to how many different "types" of A9/A10 neurons there are?

Xinkun Wang
Chee Yeun, have you found any genes related directly to oxidative stress such as SOD?

Chee Yeun Chung
Xinkun, we found glutathione transferase genes more expressed in A9 but the fold differences were very low.

Xinkun Wang
Chee Yeun, that's very interesting. We have found that some oxidative stress response genes are expressed higher in the hippocampal CA1 (sensitive region).

Chee Yeun Chung
That is very interesting, Xinkun! I was puzzled, too. There is a study by Kweon et al. that may be interesting to you .

Xinkun Wang
Chee Yeun, thanks for letting me know the paper. I will look into it.

Vik Khurana
Ole and Chee Yeun, having established intrinsic differences between A9 and A10 DA neurons, what strategy is your lab using to take this further?

Ole Isacson
Chee Yeun and I are using the data from the gene expression differences—the neuroprotection studies in vitro inspired findings—to now study these molecules in vivo. Using rodent models of PD, we overexpress or block molecules to try to make the neurons more protected. Right now we are starting a few such "gene therapy" studies.

Tom Fagan
Ole, what targets are you going to look at first?

Ole Isacson
We are looking at molecules related to synaptic vesicle trafficking and kinase/phosphatase balance.

Chee Yeun Chung
By the way, these were the candidates that were verified in humans.

Xinkun Wang
Chee Yeun, you said that some protein-level changes correlated well with the gene mRNA level. What method did you use for the protein-level analysis? I think you could not get enough protein for Western blot.

Chee Yeun Chung
Well, I used immunohistochemistry. I know this is not very quantitative but you can still see the difference.

Xinkun Wang
Chee Yeun, is this work published? I'd like to see the data.

Chee Yeun Chung
No, not yet... I used double labeling with tyrosine hydroxylase (TH) antibody. G-protein-coupled inward rectifying current potassium channel type 2 (GIRK2) distribution in human midbrain was published just recently (see Mendez et al., 2005).

Vik Khurana
Tom, before we log out, just wanted to thank you again for moderating....

Tom Fagan
Folks, we are nearing the end of our official hour, so before people start to leave I want to thank Vik for doing a lot of leg work in getting this chat going, and Ole and Chee Yeun for generating the great data and agreeing to co-lead this chat today. I think it will be a topic that will be revisited many times in the future.

Ole Isacson
Thank you!

Chee Yeun Chung
Thanks for participating.

Tom Fagan
Thanks all for attending and firing great questions.

Xinkun Wang
Thanks, Tom. Bye!

Tom Fagan
 Bye.

 

Background

Background Text
By Vikram Khurana, M.D.
HCNR Fellow, Program in Neuroscience
Harvard Medical School

Review of concept

A universal feature of neurodegenerative diseases is the relative vulnerability of particular neuronal subpopulations. Not only do neurons secreting particular neurotransmitters appear to exhibit increased vulnerability to specific neurodegenerative diseases but, perhaps more strikingly, even within groups of neurons secreting a single neurotransmitter some groups are vulnerable while others are resistant to neurotoxic stimuli. Importantly, differential vulnerability of neurons is a feature of both sporadic and inherited forms of neurodegeneration, and is also observed in animal models of these diseases.

The pattern of neuronal death differs amongst different diseases:

In Parkinson disease (PD), whether sporadic or inherited, dopaminergic neurons of the substantia nigra pars compacta (SNPc or A9) are vulnerable, whereas other dopaminergic populations are relatively spared, including the adjacent ventral tegmental area (VTA or A10) neurons. This pattern is conserved in toxic (6-OHDA, MPTP, rotenone, proteasome inhibition) animal models of PD. Even in Drosophila, in which PD is modeled by overexpressing wild-type or mutant α-synuclein, one dopaminergic neuron group selectively degenerates (6). In Huntington disease (HD), one of several polyglutamine-associated neurodegenerative diseases, the medium spiny GABAergic projection neurons of the striatum degenerate, whereas the aspiny interneurons are spared. Intriguingly, in polyglutamine-associated diseases, the clinical phenotype and pattern of cell death becomes less distinct with increasing length of the polyglutamine tract. In amyotrophic lateral sclerosis (ALS or Lou Gehrig's disease), motor neurons do not display universal vulnerability. Some motor nuclei (III, IV, VI, and Onuf's nucleus) remain relatively intact during terminal stages of the disease, while others (V, VII, XII, and most of the spinal nuclei) usually degenerate.

In Alzheimer disease, while neuronal and synaptic loss occurs diffusely across the brain, a stereotyped spreading pattern of degeneration (entorhinal cortex à hippocampus à temporal cortexà parietofrontalcortex à basal forebrain à visual cortex) implies differential neuronal vulnerability. Indeed, even within particular regions, some neurons are more vulnerable than others (for example, CA1 is the most affected region of the hippocampus).

Understanding the basis of relative vulnerability is important. An understanding of why some neurons die and others survive when exposed to the same toxic stimulus would greatly enhance our understanding of disease pathogenesis and thereby enable more specific therapeutic strategies. For example, one could envisage pharmacologic manipulations targeted to correcting a particular "vulnerability factor" in a susceptible neuronal population. Characterization of these cell types might also allow for more refined stem cell/regenerative approaches in which transplanted cells more closely reflect lost neuronal subtype or in which cells are engineered to be more resistant to toxicity.

Defining vulnerability factors

Broadly, vulnerability of a particular neuronal population could be due to innate differences between neurons or due to surrounding factors. Innate differences: Apart from the particular neurotransmitter released by the neuron group, immunohistochemical methods have defined other differences between vulnerable and spared neurons. For example, in PD, to take only two of several differences that have been defined, vulnerable neurons in the substantia nigra are calbindin-negative and GIRK2 (G protein-coupled inwardly rectifying potassium channel 2)-positive, whereas adjacent surviving neurons in the VTA are calbindin-positive and GIRK2-negative. Innate differences may also include physical differences between neurons. For example, increased cell size has been postulated to be a factor in the vulnerability of large motor neurons in ALS. Interestingly, studies have described differences in susceptibility of dopaminergic neurons to MPTP between mouse strains in toxic models of PD. Investigating strain differences may therefore also provide insights into differential neuronal vulnerability. Laser-capture microdissection (LCM) is a technique that has greatly advanced our ability to characterize differences between neuron populations. In this technique individual neurons can be isolated and collection of a large number of neurons from a particular subpopulation allows for gene expression profiling. The major advance here has been the ability to capture a pure population of neurons without "contamination" from surrounding glia and neurons. Recently, LCM has been used by several groups to characterize innate differences between susceptible and spared neuronal populations. For example, studies have compared SNpc and VTA dopaminergic populations in normal rodents (1-3), MSN and interneurons in a mouse model of HD, and hippocampal CA1 and CA3 neurons in humans (4).

Surrounding factors: The differential vulnerability of a neuronal population could also be attributable to differences in cell-cell interactions (afferent or efferent neuronal connections or interactions with adjacent glial cells) or to other anatomic features such as vascular supply.

The use of Cre/Lox-P in mice and UAS/GAL4 in flies, which allow for temporally and spatially specific genetic manipulations, might prove very useful to investigate surrounding factors. Indeed, a recent study utilizing Cre/Lox-P demonstrated that pathological cell-cell interactions were crucial for developing cortical pathology in an HD mouse model (5). In flies, differential neuronal vulnerability has been described in models of spinocerebellar ataxia (6) and Parkinson disease (7).

Live Chat Discussion

We envisage a broad discussion of the subject, and welcome all to present their perspectives. Most broadly, we plan to discuss how elucidating mechanisms underlying differential vulnerability may be translated into therapeutic benefits for patients. We are fortunate to have as our invited guest Professor Ole Isacson from McLean Hospital/Harvard Medical School. The Isacson lab has recently performed two studies that raise important issues in this area. First, Chung et al. (1) performed LCM with microarray analysis to determine differential gene expression in SNpc and VTA neurons and proceeded to show that several differentially expressed genes alter vulnerability of cultured primary dopaminergic neurons to MPP+. This study raises the issue of functional validation of candidate vulnerability factors, and we plan to discuss experimental tools available to address this question. Second, Seo et al. (8) have shown that proteasomal function is inhibited in both vulnerable striatal neurons, non-striatal neurons, and also in skin fibroblasts from patients with Huntington disease, despite only striatal neurons being vulnerable in the disease. This study raises the possibility that a combination of innate neuronal factors, such as proteasomal function and mitochondrial activity, together with surrounding factors, such as BDNF availability, may interact to determine a neuron's ability to cope with a toxic insult.

References

1. Chung CY, S.H., Sonntag KC, Brooks A, Lin L, Isacson O. (2005) Cell type specific gene expression of midbrain dopaminergic neurons reveals molecules involved in their vulnerability and protection. Human Molecular Genetics, Epub ahead of print. Abstract

2. Greene, J.G., Dingledine, R. and Greenamyre, J.T. (2005) Gene expression profiling of rat midbrain dopamine neurons: implications for selective vulnerability in parkinsonism. Neurobiol Dis, 18, 19-31. Abstract

3. Grimm, J., Mueller, A., Hefti, F. and Rosenthal, A. (2004) Molecular basis for catecholaminergic neuron diversity. Proc Natl Acad Sci U S A, 101, 13891-13896. Abstract

4. Zucker, B., Luthi-Carter, R., Kama, J.A., Dunah, A.W., Stern, E.A., Fox, J.H., Standaert, D.G., Young, A.B. and Augood, S.J. (2005) Transcriptional dysregulation in striatal projection- and interneurons in a mouse model of Huntington's disease: neuronal selectivity and potential neuroprotective role of HAP1. Hum Mol Genet, 14, 179-189. Abstract

5. Gu, X., Li, C., Wei, W., Lo, V., Gong, S., Li, S.H., Iwasato, T., Itohara, S., Li, X.J., Mody, I. et al. (2005) Pathological Cell-Cell Interactions Elicited by a Neuropathogenic Form of Mutant Huntingtin Contribute to Cortical Pathogenesis in HD Mice. Neuron, 46, 433-444. Abstract

6. Ghosh, S., Feany, M.B. (2004) Comparison of pathways controlling toxicity in the eye and brain in Drosophila models of human neurodegenerative diseases. Hum Mol Genet 13, 2011-2018. Abstract

7. Feany, M.B. and Bender, W.W. (2000). A Drosophila model of Parkinson's disease. Nature, 404, 394-398. Abstract

8. Seo, H., Sonntag, K.C., and Isacson, O. (2004) Generalized brain and skin proteasome inhibition in Huntington's disease. Ann Neurol 56: 319-328. Abstract. Also see online coverage .

Comments

Make a Comment

To make a comment you must login or register.

Comments on this content

  1. My thoughts are that this as a very interesting approach. It
    capitalizes on our understandings of the populations of neurons that
    are (and are not) vulnerable to degeneration in Parkinson's disease,
    and uses relatively recent but now well developed methods to
    investigate reasons for the neurons' differential vulnerability.

    If only we knew which cells were (and were not) selectively vulnerable
    in AD we might try a similar approach. The problem is, I fear we
    don't. Not enough attention has been paid to this problem as the field
    has been engrossed with the amyloid theory of AD pathogenesis or, more
    recently, with inflammatory damage, unfortunately paying much less
    attention to the
    "targets" of these pathogenic processes. Perhaps your discussion will
    help focus discussion on this important area.

  2. The theme of the differential susceptibility of different cell types to toxic amyloid aggregates is of outstanding importance. We have recently carried out research where we investigated the different impairment of cell viability in a wide panel of differing cultured cell lines exposed to toxic prefibrillar amyloid aggregates of a protein unrelated to any amyloid disease (the N-terminal domain of the prokaryotic HypF). Indeed, we found a wide range of susceptibilities, some cell lines being heavily affected and undergoing apoptosis, and others behaving apparently in a normal way. We found that such a different susceptibility might be traced back to different biochemical equipment in terms of molecular machineries aimed at counteracting derangements of the intracellular ion balance (particularly, free calcium and oxidative stress). In fact, we found significant correlations among resistance to the toxic effects of the aggregates and basal levels of calcium ATPase, intracellular ATP levels, and total antioxidant capacity.

    However, more intriguingly, we found that the more resistant cells were those with the highest content of cholesterol in their plasma membranes and vice versa. In particular, we found that the ability of the prefibrillar aggregates to interact with the plasma membranes inversely and significantly correlated with the content in cholesterol; we found also that the cells more heavily affected by the presence of the aggregates, those displaying the lowest level of membrane cholesterol, significantly improved their resistance upon enrichment of their membrane cholesterol content. It is currently widely accepted that toxic amyloid aggregates interact with cell components, notably cell membranes, and that such an interaction is a key event in favoring their translocation into the cell, out of the cell, or in their trafficking among cell compartments. It also appears fundamental in modifying the structural and biochemical properties of cell membranes leading to heavy modification of its permeability properties and, possibly, of its efficiency in signal transduction and other functions.

    I think that, on this aspect, the biochemical and structural properties of cell membranes are of fundamental importance, as is suggested by the above reported results and by several other previously reported data. For example, it has been shown that phosphatidylserine (mostly found in the inner membrane leaflet in normal cells) may be a preferential site of interaction for the assemblies sharing the toxic fold rich in β structure found in prefibrillar aggregates of several peptides and proteins; this could also explain the selective toxicity of folding variants of some proteins such as α lactalbumin, the ectodomain of CD44 and aprotinin to tumoral cells (known to expose phosphatidylserine molecules on the outer membrane surface). In conclusion, I think that the different susceptibility of differing cell types to the same toxic aggregates is a key topic in understanding the cell biology of aggregate toxicity; on this aspect, a central issue is a better knowledge of the role of the membrane lipid composition in favoring or disfavoring their interaction with the toxic assemblies triggering the chain of events eventually leading to cell death.

References

Webinar Citations

  1. Differential Neuronal Vulnerability

News Citations

  1. A New Link Between Pesticides and Parkinson's Disease
  2. ALS—Is It the Neurons or the Glia?
  3. Mutant Huntingtin—Trojan Horse or Trebuchet?

Paper Citations

  1. . Cell type-specific gene expression of midbrain dopaminergic neurons reveals molecules involved in their vulnerability and protection. Hum Mol Genet. 2005 Jul 1;14(13):1709-25. PubMed.
  2. . Gene expression profiling of rat midbrain dopamine neurons: implications for selective vulnerability in parkinsonism. Neurobiol Dis. 2005 Feb;18(1):19-31. PubMed.
  3. . Molecular basis for catecholaminergic neuron diversity. Proc Natl Acad Sci U S A. 2004 Sep 21;101(38):13891-6. PubMed.
  4. . Transcriptional dysregulation in striatal projection- and interneurons in a mouse model of Huntington's disease: neuronal selectivity and potential neuroprotective role of HAP1. Hum Mol Genet. 2005 Jan 15;14(2):179-89. PubMed.
  5. . Pathological cell-cell interactions elicited by a neuropathogenic form of mutant Huntingtin contribute to cortical pathogenesis in HD mice. Neuron. 2005 May 5;46(3):433-44. PubMed.
  6. . Comparison of pathways controlling toxicity in the eye and brain in Drosophila models of human neurodegenerative diseases. Hum Mol Genet. 2004 Sep 15;13(18):2011-8. PubMed.
  7. . A Drosophila model of Parkinson's disease. Nature. 2000 Mar 23;404(6776):394-8. PubMed.
  8. . Generalized brain and skin proteasome inhibition in Huntington's disease. Ann Neurol. 2004 Sep;56(3):319-28. PubMed.
  9. . Striatal expression of GDNF and differential vulnerability of midbrain dopaminergic cells. Eur J Neurosci. 2005 Apr;21(7):1815-27. PubMed.
  10. . Gene expression profiles in microdissected neurons from human hippocampal subregions. Brain Res Mol Brain Res. 2004 Aug 23;127(1-2):105-14. PubMed.
  11. . Distinct mechanisms of neurodegeneration induced by chronic complex I inhibition in dopaminergic and non-dopaminergic cells. J Biol Chem. 2004 Dec 10;279(50):51783-92. PubMed.
  12. . Cell type analysis of functional fetal dopamine cell suspension transplants in the striatum and substantia nigra of patients with Parkinson's disease. Brain. 2005 Jul;128(Pt 7):1498-510. PubMed.

External Citations

  1. online coverage

Further Reading

No Available Further Reading