Jie Shen led this live discussion on 12 July 2002. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.
Live discussion held 12 July 2002, 12 noon-1 p.m EST with Jie Shen.
Participants: Jie Shen, June Kinoshita, Kiminobu Sugaya, Richard Bowen, Keith Crutcher, Hiroko, Stavros Therianos, J. Wesson Ashford
Note: The transcript has been edited for clarity and accuracy.
June Kinoshita: I think a lot of people are on vacation today, so perhaps we should go ahead and begin. Jie, in your discussion text you mention that you are starting to look at the results of knocking in PS1 FAD mutations. Can you describe that?
Jie Shen: The experiment is still at a very early stage...
June Kinoshita: OK, Jie, although we are all anxious to learn what effect PS1 FAD mutation knockins have, since that data are not yet ready for discussion, perhaps we can review the work on the role of PS1 on neural stem cell proliferation.
Jie Shen: PS1 is required for maintaining the proliferative state of neural progenitor cells, perhaps neural stem cells, as well.
June Kinoshita: Yes, I should have said progenitor cells. Oops.
Kiminobu Sugaya: So, without PS1, the cell differentiates?
Kiminobu Sugaya: Maybe [this is] because of lack of Notch processing?
Jie Shen: Yes, without PS1, cells differentiate through the reduction of Notch signaling.
Kiminobu Sugaya: Then PS1-knockout mice have low progenitor cells, right?
June Kinoshita: I think some previous transgenic mice with PS1 mutations were reported not to have any obvious abnormality. Were those against a background of wildtype PS1?
Jie Shen: We hope that if we can understand how PS1 regulates cell cycle of neural progenitor cells during development, we might be able to figure out how PS1 regulates cell cycle and fate of neural stem cells in the adult, which may be relevant to AD pathogenesis... Although PS1 is required for the regulation of cell cycle, overexpression of PS1 may not have gain-of-function effects. Regarding neural stem cells in the adult, the problem is that there is no specific marker or promoter known to be selectively expressed in these adult neural stem cells. We are trying to get around the problem...
June Kinoshita: Are there conditional PS1-knockouts that could help answer this question?
June Kinoshita: Jie, are you referring to the overexpression of mutant PS1 in Tg mice?
Jie Shen: Yes, most of PS1 transgenic mice are made in the wildtype PS1 background, so in these mice, there is more PS1.
Keith Crutcher: This may be completely off the wall, but I wonder if this fascinating role of PS1 in neurogenesis might suggest that mutations actually result in a hypoproliferation such that AD patients are actually at risk due to fewer neurons to begin with?
Jie Shen: Keith, I think it is possible. That is the reason we are looking into the KI mice for this.
Keith Crutcher: Has anyone actually looked at PS1 brains prior to AD (I expect there are a few) to see if this might be the case? I have wondered about whether the conclusion that AD involves significant neuronal loss might actually be due to already fewer neurons in individuals at risk...a variant of the reserve hypothesis.
June Kinoshita: Keith, I've wondered the same thing myself.
June Kinoshita: Jie, are there other approaches to gene silencing that might be applicable?
Jie Shen: June, the best approach is to have an inducible system selectively turn on and off genes in neural stem cells, but the problem is that there is no inducible system that is working as we would like, and there is no specific promoter, so we have to use other methods to get around these problems.
June Kinoshita: Jie, if the brain from a person with an FAD PS1 mutation came to autopsy, what would one look for?
Jie Shen: Keith, it is difficult to address the question in humans. By the time patients have died, fewer neurons in their brains could be due to neurodegeneration.
June Kinoshita: I think Keith is suggesting that one look at a brain of a mutation carrier who has not yet developed AD.
Keith Crutcher: Yes, June, that is what I was trying to suggest.
Keith Crutcher: Jie, I agree, but the evidence for neuronal degeneration is indirect, as you know. In fact, there are fewer neurons, but whether they were ever there to begin with is an open question...I think.
Jie Shen: I don't know whether the existing technology allows accurate assessment of neuronal numbers in live human brain.
Keith Crutcher: Jie, I meant postmortem analysis [of carriers who died of other causes].
June Kinoshita: There may be families with PS1 mutations who have elected to therapeutically abort a fetus... But that's such a delicate issue.
Keith Crutcher: Jie, are pre-AD PS1 mutant carriers characterized along other neurological dimensions? I don't know this literature.
June Kinoshita: There might be someone doing volumetric measurements of such individuals, but again, I don't know how many families are being genotyped at this time.
Jie Shen: I am not aware of any literature on that. I don't think people have focused on the issue.
June Kinoshita: I wonder if Drosophila or C. elegans might be a useful model system for addressing some of these questions. What do you think, Jie?
Jie Shen: These model systems are great to address the normal function, but I am not sure whether they are good systems for FAD mutations, due to sequence divergence.
June Kinoshita: Jie, can you explain? Are there significant differences between mammalian presenilins and their homologues in fly and worm?
Jie Shen: June, yes. I don't remember the exact number—around 50 percent.
Kiminobu Sugaya: I think APP production by PS mutation may also be important for stem cell differentiation or migration. We found sAPP promotes glial differentiation of stem cells.
Kiminobu Sugaya: Has anybody checked on this?
Jie Shen: Kiminobu, maybe. Unfortunately, we know so little of in the vivo function of APP and its family members in neural development. I wish someone would do what we are doing with presenilins to make double and triple cell type-specific conditional KO to dissect out their specific function.
June Kinoshita: FYI, Gabriel Corfas at Children's Hospital is interested in the function of APP in neural development.
Jie Shen: Great, I should talk to him...
June Kinoshita: Regarding sequence divergence, that's intriguing in a gene that regulates progenitor cell proliferation. So has it evolved in mammals to govern a function that is specific to mammalian brains?
Kiminobu Sugaya: Down's syndrome patient stem cells also mainly differentiated into glia rather than neurons.
Richard Bowen: I am sure many of you are aware of the work that has been done on AD and cell cycle. Mark Smith, among others, believes that the neuronal death is due to aberrant entry of terminally differentiated neurons into the cell cycle. Could it be that PS mutations are turning on the cell cycle in these terminally differentiated cells, resulting in dysfunction or apoptosis?
Jie Shen: PS in Drosophila may regulate cell cycle, too...
Richard Bowen: Down's syndrome patients also have much higher gonadotropin levels than the general population.
Richard Bowen: Jie, we are looking at how gonadotropins may affect the cell cycle. Since human chorionic gonadotropin hCG is extremely high during pregnancy, we are wondering if it might be affecting the presenilins.
Jie Shen: Richard, interesting.
Richard Bowen: Jie, does the PS mutation affect hormone levels in these animals?
Jie Shen: I don't know.
June Kinoshita: Jie, apropos of Richard's remark, do you have any thoughts about how presenilin might affect postmitotic neurons?
Richard Bowen: The tissue with the highest expression of PS is the testes, obviously a high rate of cell cycle there.
Kiminobu Sugaya: We have not seen any increased apoptosis in previous PS1 mutant transgenic mice.
Jie Shen: Presenilin has distinct roles in postmitotic neurons.
June Kinoshita: From what Jie writes in the discussion text, I gather that the effects of PS could be quite cell-type specific, so what happens in testes may not apply to brain, despite what we think of the gonad-brain connection!
Richard Bowen: What about cell cycle abnormalities in PS1 mutant mice?
Kiminobu Sugaya: We did not check that time.
June Kinoshita: Kiminobu, what are you referring to?
Kiminobu Sugaya: When I was with Mayo, I checked the TUNEL signal in these mice, but we did not see any increase.
J. Wesson Ashford: Another point on the testes, still wondering what the critical role of PS would be. Could there be an involvement with cell motility? That could be relevant to neurite migration.
Jie Shen: Yes, the functions of PS and Notch signaling are very context-specific.
Jie Shen: Richard, what type of experiment did Mark Smith do to show that terminally differentiated neurons can re-enter the cell cycle?
Richard Bowen: He looked at the expression of cell cycle-related proteins in AD vs. control brains.
Jie Shen: Expression of proteins involved in the cell cycle?
Richard Bowen: Jie, yes. I have a list of the ones he has identified that I can send you. Let me know: email@example.com
Jie Shen: I am not sure about the relationship between PS and apoptosis, since we did not see a difference in TUNEL+ cells between PS1-KO and control mice. But it remains possible that at other time points, and other cell types, PS may regulate cell death.
June Kinoshita: Not to toot our own horn, but Inez Vincent put together a beautiful summary of the data on our site.
Jie Shen: I guess, then, the evidence is indirect, since up- or downregulation of proteins involved in cell cycle does not necessarily mean cell cycle is affected.
Keith Crutcher: Yes, actually Mark made this point at the debate conference last summer, as well.
Kiminobu Sugaya: So the argument that PS1 changes a terminally differentiated neuron's cell cycle and causes neuronal cell death could be questionable.
Richard Bowen: Hiroko, Marc is not the only one who has done work on this. I believe there are over 30 papers on the subject of AD and cell cycle abnormalities. I think that Karl Herrup has even shown that there is polyploidy in AD neurons, but not in control. June, wasn't there a recent live discussion on the subject of cell cycle and AD?
June Kinoshita: Richard, yes. See Inez Vincent reference.
Jie Shen: June Kinoshita, on the subject, does Alzforum have the information on how to genotype patients who may have PS mutations?
June Kinoshita: Jie, patients with suspected FAD can be genotyped by Athena Diagnostics.
Jie Shen: Thanks, June. Not so much of normal PS1 in sporadic AD, but I do think we have learned a lot from PS1 function and dysfunction to speculate on AD pathogenesis.
Keith Crutcher: Jie, do you have any speculation on whether normal PS1 might play a role in nonfamilial AD, which has a much higher incidence?
June Kinoshita: For example, I'm wondering if anyone has published data on PS1 gene expression or protein levels in AD brain.
June Kinoshita: Someone must have.
Keith Crutcher: I don't know this literature, but I always wonder to what extent the mechanisms postulated in FAD are going to be relevant to most of AD.
Keith Crutcher: Jie, I think you made the argument in your paper that most PS1 mutations are likely to represent gains of function.
Richard Bowen: In the same line as Keith, does anyone know if PS1 is upregulated or downregulated in typical late-onset AD?
Keith Crutcher: Jie, are you saying that the level of PS1 is reduced in sporadic AD?
Jie Shen: Keith, I am actually not sure whether FAD mutations in PS are gain-of-function. But I do think that the study of FAD mutations in PS1 will help us to understand the disease progression, in general.
Keith Crutcher: Yes, I agree. Just curious about how to relate the FAD data to sporadic AD.
Jie Shen: No, I am saying that the mechanism of neurodegeneration may be shared between PS FAD and sporadic patients.
Keith Crutcher: Yes, very possible.
Jie Shen: Actually, I think disrupted synaptic function may be the cause leading to neurodegeneration.
June Kinoshita: But what causes the disruption of synaptic function?
Keith Crutcher: Again, it would be helpful to know if PS1 mutant carriers show any differences in synaptic density.
Richard Bowen: Jie, very exciting data. After I read more about it, I want to ask you some further questions. Sorry, I have to leave now.
Kiminobu Sugaya: Isoe-Wada et al. reported in 1999 reduced PS1 expression in sporadic AD.
Jie Shen: Molecular pathways regulate synaptic function, just as PS regulates neuronal differentiation through the Notch signaling pathway.
June Kinoshita: Do you think the role of PS1 in adult brain and neurodegeneration is related to its role in progenitor cell proliferation, or some other pathway?
Jie Shen: June, maybe, it could be a contributing factor.
June Kinoshita: This suggests that there could be an impairment in the brain's repair/plasticity mechanisms.
Kiminobu Sugaya: Since we found APP function in stem cell biology, I would like to speculate on PS1 effect through APP processing, also.
Jie Shen: For example, Aβ disrupts synaptic function, but no one knows yet how.
June Kinoshita: What genes regulate PS1 expression? I suppose that would be one place to look in terms of understanding why PS1 expression would be reduced in sporadic AD.
Jie Shen: Kiminobu, PS may affect downstream effects, such as synaptic function, etc., partially through their regulation of APP processing.
Hiroko: Please, tell me, is synaptic dysfunction a key of AD, Jie?
Kiminobu Sugaya: In a damaged neuron, is PS1 expression increased or decreased?
Jie Shen: Hiroko, I think it may be. I meant synaptic dysfunction may be an important step leading to neurodegeneration.
Jie Shen: June, I want to address your question about brain repair, etc.
June Kinoshita: Please, do.
Jie Shen: If adult neurogenesis plays an important role in normal brain function, possibly by replenishing neuronal populations, and if FAD mutations in PS affect adult neurogenesis, then this could be a contributing mechanism for AD.
June Kinoshita: Jie, that's an interesting hypothesis, and seems as though it should be pretty testable, using some of the rodent paradigms that were used to establish the existence of adult neurogenesis.
Stavros: Concerning gene regulation, several developmental genes such as prospero in Drosophila could regulate the expression of some cell cycle-related genes. Could such regulation, using the same set of developmental "selector" genes, but in the adult, regulate cyclins as well as PS1?
Jie Shen: Possible.
Kiminobu Sugaya: How do PS mutations affect adult neurogenesis?
Kiminobu Sugaya: Decrease by increasing proliferation of the stem cells...
Jie Shen: A lot of experiments need to be done... [it could] affect proliferation or cell fate.
J. Wesson Ashford: Jie, I think that the issue in Alzheimer's does not relate to neurogenesis or apoptosis, but as asked by Hiroko, the problem is loss of synapses. The synapses are most likely lost because of disruptions of neurite function associated with neuroplasticity.
Jie Shen: Wes, I agree with you that disrupted synaptic function, loss of synapses, then loss of neurons is likely to be the main cause of AD, but logically impaired neurogenesis could be a contributing factor, as well.
Kiminobu Sugaya: I agree. I think AD is also related to adult neurogenesis.
J. Wesson Ashford: Jie, the only problem that I can see associated with neurogenesis would be the difficulty that new neurons would have establishing new connections.
Jie Shen: I think we need to keep an open mind as to the cause of AD, and look into all possibilities. This approach will more likely lead to the whole truth.
June Kinoshita: That's a good note on which to wrap up today.
Keith Crutcher: Very interesting work, Jie! Thank you for taking the time to discuss it today.
Jie Shen: Thanks, everyone.
June Kinoshita: We've run over our hour, so I would like to thank Jie for her participation today, and to all of the audience for attending.
By Jie Shen
More than 80 mutations in the presenilin genes (PS1 and PS2) and about ten mutations in the gene encoding the amyloid precursor protein (AβPP) have been linked to familial Alzheimer's disease. These proteins—particularly the presenilins—play essential roles in cortical development. This raises fascinating questions about the role of neurogenesis, both during development and in the adult, in normal and pathological aging.
Mice lacking both PS1 and PS2 die before embryonic day 9 (Donoviel et al., 1999). PS1-/- mice, which are a partial loss-of-function presenilin mutant, do survive until birth and so permit us to assess PS1 function in neural development (Shen et al., 1997). Further analysis of PS1-/- mice revealed that PS1 is involved in the maintenance of the progenitor cell population and the regulation of neuronal differentiation through the Notch signaling pathway (Handler et al., 2000). In the absence of PS1, neural progenitor cells prematurely exit the cell cycle to differentiate into postmitotic neurons, thus depleting the pool of self-renewing progenitor cells needed for further brain development. By contrast, inactivation of PS1 in epidermal cells promotes hyperproliferation and tumorigenesis via the Wnt signaling pathway (Xia et al., 2001). These intriguing results suggest that PS1 may be directly involved in cell cycle regulation in a context-dependent manner. Since the PS1-null mutation represents only a partial loss of function of presenilins, a full characterization of presenilin function in neural development and the cell cycle still awaits the generation and analysis of cell type-specific, conditional null mutants of presenilins.
AβPP has been implicated in neuronal differentiation based on its expression pattern in vivo and in cultured neurons (Hung et al., 1992). And yet, there has not been direct evidence demonstrating an involvement of AβPP in neural development. APP-/- mice are viable and exhibit impaired learning and reduced locomotor activity (Muller et al., 1994; Zheng et al., 1995). Mice lacking either the amyloid precursor-like protein 1 or 2, (APLPs are members of the AβPP family,) are viable. However, APP-/-; APLP2-/- and APLP1-/-; APLP2-/- exhibit early postnatal lethality, indicating functional redundancies among these proteins (von Koch et al., 1997; Heber et al., 2000). However, the fact that the YENPTY motif at the C-terminal region of AβPP interacts with the PTB domain of mDab1, which is a key molecule in neuronal migration and cortical lamination, suggests that AβPP might be involved in the regulation of cortical lamination during development. Functional redundancies among AβPP family members may explain the lack of phenotypes in neural development in the single and double knockout mice. Mice lacking all three members of the AβPP family will determine whether the AbPP family is indeed involved in cortical development.
The fact that presenilins regulate neurogenesis has raised the question of whether FAD-linked mutations in presenilins lead to an impairment of neurogenesis, and to the subsequent question of whether defective neurogenesis contributes to the pathogenesis of AD. Fortunately, these questions can be addressed directly using a PS1 knockin mouse, in which a FAD-linked mutation is introduced into the PS1 gene in such a way that heterozygous knockin mice genetically resemble FAD patients (Guo et al., 1999). We are in the process of studying such mice, and their complete analysis will help us understand why genes involved in cortical development play a critical role in Alzheimer's disease.
- Donoviel DB, Hadjantonakis AK, Ikeda M, Zheng H, Hyslop PS, Bernstein A. Mice lacking both presenilin genes exhibit early embryonic patterning defects. Genes Dev. 1999 Nov 1;13(21):2801-10. PubMed.
- Shen J, Bronson RT, Chen DF, Xia W, Selkoe DJ, Tonegawa S. Skeletal and CNS defects in Presenilin-1-deficient mice. Cell. 1997 May 16;89(4):629-39. PubMed.
- Handler M, Yang X, Shen J. Presenilin-1 regulates neuronal differentiation during neurogenesis. Development. 2000 Jun;127(12):2593-606. PubMed.
- Xia X, Qian S, Soriano S, Wu Y, Fletcher AM, Wang XJ, Koo EH, Wu X, Zheng H. Loss of presenilin 1 is associated with enhanced beta-catenin signaling and skin tumorigenesis. Proc Natl Acad Sci U S A. 2001 Sep 11;98(19):10863-8. PubMed.
- Hung AY, Koo EH, Haass C, Selkoe DJ. Increased expression of beta-amyloid precursor protein during neuronal differentiation is not accompanied by secretory cleavage. Proc Natl Acad Sci U S A. 1992 Oct 15;89(20):9439-43. PubMed.
- Müller U, Cristina N, Li ZW, Wolfer DP, Lipp HP, Rülicke T, Brandner S, Aguzzi A, Weissmann C. Behavioral and anatomical deficits in mice homozygous for a modified beta-amyloid precursor protein gene. Cell. 1994 Dec 2;79(5):755-65. PubMed.
- Zheng H, Jiang M, Trumbauer ME, Sirinathsinghji DJ, Hopkins R, Smith DW, Heavens RP, Dawson GR, Boyce S, Conner MW, Stevens KA, Slunt HH, Sisoda SS, Chen HY, van der Ploeg LH. beta-Amyloid precursor protein-deficient mice show reactive gliosis and decreased locomotor activity. Cell. 1995 May 19;81(4):525-31. PubMed.
- von Koch CS, Zheng H, Chen H, Trumbauer M, Thinakaran G, van der Ploeg LH, Price DL, Sisodia SS. Generation of APLP2 KO mice and early postnatal lethality in APLP2/APP double KO mice. Neurobiol Aging. 1997 Nov-Dec;18(6):661-9. PubMed.
- Heber S, Herms J, Gajic V, Hainfellner J, Aguzzi A, Rülicke T, von Kretzschmar H, von Koch C, Sisodia S, Tremml P, Lipp HP, Wolfer DP, Müller U. Mice with combined gene knock-outs reveal essential and partially redundant functions of amyloid precursor protein family members. J Neurosci. 2000 Nov 1;20(21):7951-63. PubMed.
- Guo Q, Fu W, Sopher BL, Miller MW, Ware CB, Martin GM, Mattson MP. Increased vulnerability of hippocampal neurons to excitotoxic necrosis in presenilin-1 mutant knock-in mice. Nat Med. 1999 Jan;5(1):101-6. PubMed.
No Available Further Reading