Kiminobu Sugaya led this live discussion on 9 July 2002. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.
Live discussion held 9 July 2002, 12 noon-1 p.m EST with Kiminobu Sugaya.
Participants: Kiminobu Sugaya, George C., Peter Soba, June Kinoshita, J. Wesson Ashford
Note: The transcript has been edited for clarity and accuracy.
Kiminobu Sugaya: Well, I guess we can start. Recent progress in stem cell technologies suggests the probability of using neuroreplacement strategies as an AD therapy, but we may have to clear several hurdles. For example,
- Can we replace long-projecting cholinergic neurons?
- Does AD pathology affect stem cell biology? If so, how can we control it?
- Is it a good tactic to make cholinergic cells in vitro then transplant them into the brain
just throw these few ideas out to get us started. Anyone got any thoughts.
George C.: It seems that a key question is whether the neuronal loss/damage is due to extracellular factors, which would be expected to damage the stem cell derived-neurons or whether intracellular metabolic effects are responsible for the dementia.
Kiminobu Sugaya: George, I do not think A toxicity is the only factor. We found a direct effect of AbPP on stem cell migration and differentiation.
George C.: How was this demonstrated?
Kiminobu Sugaya: When we apply sAbPP to the stem cells, they differentiate more and, interestingly, more differentiate into glia.
George C.: Is there an effect of Ab on stem cell differentiation in culture? In vivo?
Kiminobu Sugaya: We also found AbPP transfection to the cell causes glial differentiation rather than neural differentiation. In vivo, we are conducting transplantation into AbPP transgenic mice, and will present data in this coming Neuroscience. We have also transplanted the stem cell to AbPP-knockout mice. In this case, they did not migrate well. We have not done any particular study with Ab , though I do not expect cell toxicity.
Peter Soba: But still the question remains, even if neuroreplacement works in the normal brain, does it also work in Alzheimer's brains?
June Kinoshita: How about nontoxic effects, e.g., on migration or differentiation?
Kiminobu Sugaya: Ab may have some effect, which we have to look for in a future study.
June Kinoshita: I saw a poster--I think it was at the Neuroscience meeting last year--by Barbara Tate, showing that stem cells migrated to the sites of Ab injection. It wasn't clear whether this was a response to Ab-induced injury, or whether Ab was acting as a signaling molecule. Could you comment on that?
Kiminobu Sugaya: Yes, it even may be because of the injury caused by the injection.
June Kinoshita: I don't think this effect was seen in the controls, which received sham injections.
Kiminobu Sugaya: Any brain injury increases AbPP expression in the brain, though. Anyway, we are thinking AbPP, not Ab , is the signal molecule for migration of stem cells.
June Kinoshita: Yes, so an interesting question is whether AbPP or an AbPP derivative is part of the brain's normal injury response, and whether this response includes the recruitment of stem cells to the injury site.
Kiminobu Sugaya: I think AbPP is the molecule to initiate migration or differentiation of stem cells.
June Kinoshita: Kiminobu, could you summarize for us the strongest evidence you have to date to support your hypothesis?
Kiminobu Sugaya: Okay.
- sAPP treatment increases differentiation of stem cells in vitro.
- 22C11 suppresses this effect
- High doses of sAPP increase glial differentiation
- AbPP transgene increases glial differentiation
- Stem cells transplanted to AbPP-knockout mice did not migrate or differentiate.
George C.: I believe Mark Mattson and coworkers have evidence in AbPP-transgenic mice that endogenous stem cell activity is depressed. Any evidence for this with transplanted stem cells?
Peter Soba: But still you see an increase in neuron numbers in AbPP-Tg mice.
June Kinoshita: On the surface, the Mattson results and yours would seem to contradict each other. What was the difference between his experiment and yours?
Kiminobu Sugaya: Stem cell activity depressed in AbPP transgenic mice could be caused by the over differentiation to glial cells. Actually, in AbPP transgenic mice, the stem cell population may be decreased by the glial differentiation, and this could happen in the AD brain, too.
June Kinoshita: Just trying to make sure I understand: From what you are saying, AbPP overexpression doesn't decrease stem cell activity per se, but it pushes stem cells towards differentiating into glia rather than neurons. Is that right?
Kiminobu Sugaya: AbPP overexpression just pushes glial differentiation of stem cells, which results in decreased stem cell population. I am not sure if Mattson counted neural differentiation of transplanted stem cells in control and AbPP-transgenic mice, though.
George C.: I don't think Mattson assessed differentiation.
Kiminobu Sugaya: Some Japanese groups reported increased proliferation of stem cells by sAbPP.
George C.: Any ideas on a mechanism for the effect of AbPP on stem cell migration and differentiation? What about endogenous stem cell activity in AbPP null mice?
Kiminobu Sugaya: We have not tested endogenous stem cell migration or differentiation in AbPP-knockout mice yet. It may be interesting to do.
Kiminobu Sugaya: I think brain injury increases AbPP expression. AbPP expression then induces stem cell migration and differentiation to the site. First, the cells would differentiate into glia, such glial cells may induce neuronal differentiation later, or support the neuronally differentiating stem cells.
Peter Soba: But the proliferative effect is only seen when growth factors are present, e.g., EGF, FGF.
Kiminobu Sugaya: Peter, you are right. In our case, we do not use any of the factors in vitro.
Peter Soba: So the effect of sAbPP might depend on which factors are present, and these might work cooperatively.
June Kinoshita: Are there any obvious abnormalities in the brains of AbPP-knockout mice? I was under the impression that there aren't.
Kiminobu Sugaya: There is no obvious phenotype in AbPP-knockout mice. AbPP may be more important for adult neurogenesis rather than corticogenesis.
June Kinoshita: So you would argue for AbPP not playing an important role in differentiation of brain cells during development, or are other proteins, e.g., AbPP homologues, able to step into that role? For example, if one knocked out AbPP and its homologues, how does that affect stem cell differentiation in development? And is that relevant to understanding neurogenesis in an adult brain?
Kiminobu Sugaya: During development AbPP may also work, because we can see increased expression of AbPP at some point. But APLP might be able to compensate for this effect in AbPP knockouts.
Kiminobu Sugaya: In our case, we saw some migration of stem cells to the hippocampus. This may be because APLP is rich in this area. As a matter of fact, if we knock out both AbPP and APLP, we don't get offspring.
June Kinoshita: Kiminobu, in what system did you observe this?
Kiminobu Sugaya: After AbPP-knockout mice transplantation.
J. Wesson Ashford: It seems that some of this might be like saying that the frontal lobes are not important since we can't measure their functions. I would think that we need to determine the function of AbPP carefully. One recent thought is that the two pathways-- - versus -secretase--may be important, but there would still be some balance, though probably less adaptive, without AbPP.
George C.: In general, how reproducible is stem cell migration and differentiation, and how difficult is this to quantify?
Kiminobu Sugaya: Good question, since the stem cell forms spheres, we cannot count the cells inside of the sphere.
June Kinoshita: Is that in vitro or in vivo?
Kiminobu Sugaya: Sorry, this is in vitro.
Kiminobu Sugaya: To overcome this problem, we have to use rather small spheres.
George C.: So you are transplanting whole neurospheres rather than dissociated cells?
Kiminobu Sugaya: Yes, we are transplanting whole spheres.
J. Wesson Ashford: So, Kiminobu, is your inference that AbPP and APLP can cross-cover for each other? It seems that ApoE-knockouts may be relatively normal. Also, there is a related case where the absence of the protein is not clearly bad, probably because others can cross-cover.
Kiminobu Sugaya: Wes, thank you for your input. I think they can cross-cover each other.
June Kinoshita: We need a conditional double or triple APP/APLP1 and 2-knockout, perhaps.
Kiminobu Sugaya: That is a good idea. Anybody have one?
June Kinoshita: Don't know. I'll keep an eye out for one!
Kiminobu Sugaya: I think such a compensation of knockout genes is always a problem with knockout mice models.
George C.: If quantification is difficult, how many mice of each genotype need to be transplanted to demonstrate a difference between null and wildtype mice, for example?
Kiminobu Sugaya: In vivo quantification is rather easy, because we do not see any migration of stem cells to the cortex. We can see the cells in the needle track, though.
George C.: Very interesting.
Kiminobu Sugaya: For the hippocampus, we may be able to count the cells reaching the pyramidal cell line. Lacking migration of stem cells in AbPP knockout mice may also be related to the Ab immunization problem.
J. Wesson Ashford: Kiminobu, I am still not completely happy with the cross-coverage issue. There must be some basic function that the AbPP provides, without which there is some deficit. Otherwise, all we would have to do to solve Alzheimer's disease would be to turn off the AbPP altogether.
George C.: Good point, and tet-on tet-off or cre-LoxP mice should be able to answer this question.
Kiminobu Sugaya: If we turn off AbPP, the stem cell could not migrate. Meaning that there would be no adult neurogenesis.
J. Wesson Ashford: And migration is an active process, probably related to dendrites and axons searching for new connections--neuroplasticity.
June Kinoshita: Sorry, Wes, I'm missing a step here. If AbPP is involved in neural stem cell differentiation, how would shutting it down cure Alzheimer's?
Kiminobu Sugaya: Again, shutting down AbPP causes no neuroplasticity. It should not cure the disease. I guess that is why Ab immunization failed.
June Kinoshita: I think that had more to do with an inflammatory response. There aren't any public data to help us understand what happened there with the Elan trials.
J. Wesson Ashford: My point, June, was that if AbPP-knockout mice were normal, and APLP cross-covered for it adequately to produce an animal without deficits, then it really would not be needed. If shutting down AbPP led to a lack of neuroplasticity, that would be bad for memory but good for AD. But the issue is more complex, I am sure, as Kiminobu is suggesting, probably underlying the reasons that the immunizations failed--about which we need to learn much more.
Peter Soba: What I heard is that there was also loss of synaptic plasticity, June.
June Kinoshita: Wes, that seems paradoxical. Alzheimer's involves a loss of plasticity, so how would even less neuroplasticity help?
Kiminobu Sugaya: June, maybe so, but in the long run, suppressing stem cells could be a big problem.
J. Wesson Ashford: I need to run. Sorry I was late. Very interesting and just getting better. The neuroplasticity question is a longer answer; check our papers and I will talk with you about that answer later. Bye.
Kiminobu Sugaya: Wes's point would be, even if we knock out AbPP, APLP may help to migrate the stem cells. So, I guess it's about the time to close this.
June Kinoshita: Well, there's clearly a lot of exciting work to be done to sort this all out. Let me know if you have any afterthoughts to add to the chat.
June Kinoshita: Kiminobu, thank you very much. Thanks to the audience.
Kiminobu Sugaya: Thank you, everybody. Talk to you later.
By Kiminobu Sugaya
The discovery of multipotent neural stem cells (NSCs) in the adult brain (Gould, 1999) has brought revolutionary changes to the theory of neurogenesis. This theory now suggests that regeneration of neurons can occur throughout life. Recent advances in stem cell technologies are expanding our capability of eventually replacing many types of tissues throughout the body. In particular, the successful transplantation of human NSCs (HNSCs) into aged rats with subsequent improvement of cognitive function (Qu et al., 2001) reinforces the potential feasibility of HNSC transplantation therapy for neurodegenerative diseases such as Parkinson's, Alzheimer's, and Huntington's (Fricker, 1999; Snyder, 1997).
Parkinson's disease has been a good target for neuroreplacement therapy because of its specific loss of large projecting dopaminergic neurons in the substantia nigra. These neurons are sending dopaminergic projections to the striatum, hence conventional treatment for Parkinsons' augments the dopamine content in the striatum with L-dopa. Based on experience with L-dopa treatment, transplantation of fetal tissue producing dopamine to the striatum of PD has been tested for many years. Many of the clinical trials significantly ameliorated the behavioral deficit, though with significant side effects (see ARF news story). However, the use of human fetal neuronal tissue not only raises ethical concerns but also is impractical because neural tissue from multiple fetuses is required for each patient. Thus we must seek alternatives to fetal tissue.
Neural stem cells, capable of spontaneously differentiating into neurons and glia, are the most promising candidates. Stem cells are often defined as self-renewing and multipotent. They can be isolated from the embryonic and adult rodent central nervous system and propagated in vitro in a variety of culture systems (Brannen, 2000; Doetsch, 1999). Our inability to grow NSCs in serum-free culture media has long been a major obstacle in understanding their physiology. Now, however, NSCs can be maintained or expanded in serum-free medium containing basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF) (Brannen, 2000, Fricker 1999). HNSCs differentiate into cells immuno-positive for the neuronal marker bIII-tubulin and the astrocyte marker glial fibrillary acidic protein in an unsupplemented medium, suggesting the genesis of neurons or astrocytes without exogenous differentiation factors. This capability to expand multipotent HNSCs in vitro offers a well-characterized and efficient source of transplantable cell types.
Even so, many researchers are still trying to differentiate dopaminergic cells in vitro and then transplant them into the basal ganglia as dopamine-release vehicles in the target area (Clarkson, 2001). This approach seems natural, just as we were transplanting the fetal dopaminergic tissue to the basal ganglia to increase its dopamine content, however, we must consider several issues. First, fully differentiated neuronal cells do not migrate, meaning they sit at the injection site. Second, a "cure" for PD is conceivable only if we replace degenerating dopaminergic neurons in the substantia nigra. Dopaminergic neurons transplanted to the basal ganglia may not be functionally regulated by the host brain, which may be a cause of side effects (Freed, 2001). Third, if the injection is made into the brain, the concomitant tissue destruction will cause monocyte recruitment, and the ensuing immune response will eliminate the transplanted cells. To avoid these problems, I suggest injecting undifferentiated stem cells into the brain ventricle.
Our group has succeeded in recovering cognitive function in aged rats by intraventricular injection of neural stem cells. Yet this work did not take into account the effect of pathological changes of the diseased brain, which may prevent regular differentiation or migration of the stem cells. For example in thinking about neuroreplacement therapy for AD, we must consider the following issues:
1. Can we replace long-projecting cholinergic neurons?
2. Does AD pathology affect stem cell biology? How?
In AD, memory deterioration involves degeneration of basal forebrain (BF) cholinergic neurons, so this neuron population should be replaced. Their long projections and expression of nerve growth factor receptor make cholinergic neurons phenotypically different from dopaminergic neurons. This has led some researchers to declare BF cholinergic neurons irreplaceable. However, it has since been shown that BF cholinergic neurons transplanted into striatum and nucleus basalis of Meynert (NBM, which provides major cholinergic input to the neocortex) of adult rat brain survived and expressed the cholinergic phenotype (Martinez-Serrano, 1996). Furthermore, engrafted cholinergic-rich (but not non-cholinergic) cell suspensions reversed the deficits in radial-arm maze performance previously caused by NBM excitotoxic lesion (Sinden, 1995). Although these results hint that it may be possible to replace BF cholinergic neurons, we have to prove that degenerating BF cholinergic neurons in AD can be replaced by HNSC transplantation.
How Does AβPP Enter the Picture?
While the physiological functions of AβPP remain unclear (join the other chats in this series!), prior work has generated a host of possibilities. The ones relevant to this discussion include AβPP's possible involvement in neurite outgrowth (Roch, 1993; Salinero, 2000); NSC proliferation (Hayashi, 1994; Ohsawa, 1999); epidermal basal cell proliferation (Hoffmann, 2000); neuronal migration (De Strooper, 2000); and neuronal differentiation (Ando, 1999). AβPP expression increases after brain injury, neuronal loss (Murakami, 1998) and axonal injury (Koszyca, 1998). AβPP potentiates the effect of neurotrophin (Wallace, 1997; Wang, 2000), and treatment with a monoclonal antibody to AβPP induces neuronal apoptosis, indicating the involvement of AβPP in cell survival (Rohn, 2000).
In our recent study (Society for Neuroscience 2000, http://sfn.ScholarOne.com/itin2001/; search by Sugaya), unsupplemented, serum-free differentiation causes HNSC apoptosis associated with an increase in AβPP expression. A monoclonal antibody recognizing the N'-terminal of AβPP inhibited this HNSC differentiation, but the addition of an exogenous a-secreted AβPP (sAβPP) to the culture media accelerated it. Furthermore, exogenous sAβPP induces gliogenesis rather than neurogenesis of HNSCs. These results indicate that apoptotic cells release AβPP fragment(s) to increase HNSC differentiation into glial cells, possibly contributing to gliogenesis seen in AD. Overexpression of AβPP in HNSCs by transfection with wild-type AβPP also induced glial differentiation.
Recently, Bahn et. al., reported that stem cells from people with Down's syndrome differentiated into astrocytes rather than neurons (see ARF news story. Since Downs' patients have inherited three copies of AβPP, (which resides on chromosome 21,) this abnormal differentiation may result from an overdose of AβPP. Down's patients develop AD by age 40. Given all this, transplantation therapy of AD with HNSCs may not be effective in an environment where AβPP metabolism is altered, since it might lead to excessive gliogenesis.
What are the implications of these findings for AD? It is not clear whether adult neurogenesis is essential for normal cognitive function in age. Yet it is tempting to speculate that pathologically altered AβPP metabolism could impair NSC migration and differentiation into the proper ratio of neurons and glia. Aged transgenic AβPP mice exhibit neuronal loss and extensive gliogenesis in the neocortex (Bondolfi, 2002). Although the rate of endogenous neuroregeneration in the adult brain may be minimal, in the long run a defect in this process might significantly harm normal brain function.
Incidentally, this possibility raises the question whether Aβ immunization, which may also reduce AβPP fragments, is helpful for maintaining stem cell function in AD? Here is why I say no to this question: HNSCs transplanted into AβPP knockout mice did not migrate or effectively differentiate into neurons in the cerebral cortex, where we have seen beautiful neural differentiation of transplanted HNSCs in wild-type mice (Society for Neuroscience, 2000). HNSCs may play important roles in neuroregeneration, and if AβPP is, indeed, involved in the regulation of HNSCs, as we propose, destruction of the AβPP system may jeopardize the maintenance of brain function.
Many in the field are talking about clinical trials of NSC transplantation for AD, though none are underway yet. This approach will take time to establish itself as a therapy for AD. We need to know what good NSCs can do in the AD brain and how the AD brain environment affects NSC biology. I hope our effort in stem cell and AD research will cut down the time needed to develop a better treatment for AD.
Sinden, J., Hodges, H. and Gray, J., Neural transplantation and recovery of cognitive function, Behavioral and Brain Sciences, 18 (1995) 10-35.
- Long-Term Follow-up of Parkinson's Patients Who Received Embryonic Cell
- Down and In—the Location of Amyloid-β Does Matter
- Gould, 1999
- Qu et al., 2001
- Fricker, 1999
- Snyder, 1997
- Brannen, 2000
- Doetsch, 1999
- Fricker 1999
- Clarkson, 2001
- Freed, 2001
- Martinez-Serrano, 1996
- Roch, 1993
- Salinero, 2000
- Hayashi, 1994
- Ohsawa, 1999
- Hoffmann, 2000
- De Strooper, 2000
- Ando, 1999
- Murakami, 1998
- Koszyca, 1998
- Wallace, 1997
- Wang, 2000
- Rohn, 2000
- http://sfn.ScholarOne.com/itin2001/; search by Sugaya
- Bahn et. al.
- Bondolfi, 2002
No Available Further Reading