A pair of studies published in the July 6 Neuron show that in two different mouse models of Down syndrome, the cause of neuron death boils down to a lack of neurotrophic factor support. One study, from Ahmad Salehi, Jean-Dominque Delcroix, William Mobley and colleagues at Stanford University in California, puts the blame on the amyloid precursor protein (APP)—its increased expression disrupts nerve growth factor (NGF) retrograde transport and leads to degeneration of cholinergic neurons, they report. The second paper, from Susan Dorsey and Lino Tessarollo of the National Cancer Institute in Frederick, Maryland, concludes that it is the upregulation of a truncated TrkB receptor that spells the demise of hippocampal neurons in a mouse trisomy model—the dominant-negative receptor isoform inhibits the action of another neurotrophin, brain-derived neurotrophic factor (BDNF).

Down syndrome (DS) leads to early onset AD-like pathology that many researchers believe is caused by the increased dose of the APP gene, present on the triplicated chromosome 21. Though the findings of Salehi and colleagues support this view, Dorsey and colleagues’ work suggests that APP is only part of a bigger picture. But together the two papers show in concrete detail how interrupting neurotrophic factor action causes neurodegeneration in DS models, and make the case that similar perturbations could contribute to neuron loss in Alzheimer disease and other neurodegenerative disorders. It is already known, for example, that there are defects in axonal transport early in the neurodegenerative processes in AD (see ARF related news story) and the work from Salehi and colleagues suggests that this could be linked to APP. Coincidentally, a third report last week, this one in Nature Neuroscience from the lab of Moses Chao of New York University, shows that the basis for another action of BDNF—its ability to stimulate neurotransmitter release—is also linked to transport. The actin-dependent motor Myo6 and an adaptor protein, GIPC1, which complex with the TrkB receptor, are essential for BDNF-stimulated release of glutamate.

The study by Salehi, Delcroix and Mobley, along with collaborators from across the U.S., continues their work on the degeneration of basal forebrain cholinergic neurons (BFCNs) in Ts65DN Down syndrome mice. These same cells degenerate early in AD, and in the DS model can be saved by application of NGF. The Ts65DN mice are trisomic for a portion of chromosome 16, which is analogous to human chromosome 21. Work from this group so far suggests that increased levels of APP contribute to neurodegeneration by interfering with the retrograde transport of the neurotrophic factor NGF from nerve endings in the hippocampus back to BFCN cell bodies (see Cooper et al., 2001, and coverage of Delcroix’s presentation at the SfN meeting in 2004). With their supply of NGF cut off, the BFCNs atrophy and die.

The new report shows this data, as well as additional work on the mechanism by which APP inhibits NGF transport. Experiments show that while there is no elevation of Aβ peptides in the mice, APP and its C-terminal cleavage fragments are elevated, and localized in part to signaling endosomes that also contain NGF and its receptor—after endocytosis at nerve terminals, NGF is transported as a signaling complex with its receptor in these early endosomes. In the Ts56Dn mice, there was no indication that NGF binding or internalization was changed. By immunostaining, Salehi and coworkers showed that the levels of APP and APP-CTF were increased in early endocytic vesicles in cholinergic nerve terminals. They also detected NGF and its receptors in the same vesicles, and showed that cholinergic terminals in the Ts56Dn mice contained enlarged vesicles containing APP, APP-CTF, and NGF. Further work will be needed to understand how vesicle enlargement might be related to the failure of transport. Importantly, they showed that disruption of transport was selective for NGF, since there was no change in general retrograde transport as measured by movement of Fluoro-Gold.

Since triplication of the APP gene in the context of trisomy 16 in the Ts65DN mice caused a 90 percent reduction in NGF transport and loss of BFCNs, the researchers checked transport in mice overexpressing the human APP gene, or the APP gene with the AD-causing Swedish mutation. In this case, mice expressing either wild-type APP or the Swedish mutation experienced a slowing of NGF transport by 40-60 percent, but no neuron loss. These results suggest either that more severe deficits in transport are necessary to trigger neurodegeneration, or that other genes present on the triplicated region also play a part. Mice overexpressing APP along with presenilin1 (PS1) had decreased NGF transport, but mice overexpressing only PS1 instead experienced an increase in NGF transport, for reasons that are still mysterious.

One of those other genes could code for TrkB. In a different mouse model of DS, Dorsey, Tessarollo, and colleagues from the National Cancer Institute found that alterations not in transport but instead in TrkB-mediated receptor signaling of BDNF lead to the early death of hippocampal and cortical neurons. They had shown previously that hippocampal neurons from the trisomy 16 mouse model of DS died quickly in culture, and had a defective response to their normal trophic factor, BDNF. Now, the group shows that the reason for this is that the cells overexpress a dominant negative form of the TrkB receptor, TrkB.T1. By reducing the levels of this truncated receptor to normal, they prevented cell death in the trisomy 16 neurons. That prevention was associated with restored signaling in response to BDNF, including a normalization of AKT kinase activity and intracellular calcium levels, both of which are known to contribute to cell survival. When they measured apoptosis in vivo in developing mouse brain, they found high levels of cell death in the cortex of Ts16 mice, which were decreased to normal by lowering expression the TrkB.T1 isoform. “These data suggest that neurodegeneration may not only be the result of a diminished supply of neurotrophins and provide direct evidence that neurons must express the correct set of receptor isoforms to transduce a proper survival signal in response to neurotrophins,” the authors conclude.

“The common message of both of these papers is that the inhibition of signaling or transport of a single neurotrophic factor may be partially responsible for the neuronal pathology observed in DS mouse models, and the same mechanisms may be affected also in AD,” write Finnish researchers Eero Castren of the University of Helsinki and Heikki Tanila of the University of Kuopio, in their accompanying preview. “The papers…underline the notion that the primary aim of treatment of neurodegenerative disorders is not to keep the neurons alive, but to keep them connected.”

Castren and Tanila outline other important avenues that need to be explored, including elucidating the processing steps for APP, the role of α- versus β- secretase processing, and the importance of various APP products in NGF transport. Adults with Down syndrome invariably develop AD, and recently a duplication of APP gene alone has been shown to cause AD (see ARF related news story). While the mechanisms of neurodegeneration in DS and AD are not exactly the same (there was no elevation of Aβ peptides in the Ts56Dn mice, for one), the same cell types are involved, and some commonalities, such as failure of axonal transport, may turn up based on this work. In addition, overexpression of truncated TrkB.T1 has been reported in AD (Ferrer et al., 1999), and when added to observations of BDNF deficiency in AD brain (e.g., Peng et al., 2005), the current work should spur research into the role of TrkB isoforms in AD.

In addition to its functions in keeping neurons alive and connected, BDNF released in response to synaptic activity acts through the TrkB receptor to acutely regulate neurotransmitter release and synaptic plasticity. In the third paper, Chao and colleagues show that the presynaptic complex of the actin motor Myo6 and the GIPC1 adaptor protein links TrkB to neurotransmitter release. Writing in the July 2 Nature Neuroscience, first author Hiroko Yano and coworkers report that the TrkB receptor directly associates with Myo6-GIPC1. Either Myo6 or GIPC1 knockout mice showed defects in BDNF-induced LTP in very young mice, while BDNF-independent LTP in adults was normal. The complex was required for BDNF-induced glutamate release, and lack of either protein compromised vesicle recycling in response to BDNF. Their conclusion that BDNF can use GIPC1 and the Myo6 actin motor to modulate neurotransmitter release at presynaptic locations lends an important role to actin filaments in neurotransmitter release, and sheds light on another function of BDNF in brain health.—Pat McCaffrey


  1. I think this study by Salehi and colleagues complements our work. If anything, the two studies combined stress once again the relevance of neurotrophin supply/signaling in supporting neuronal survival and function. What I find intriguing is that the two papers describe different mechanisms by which alterations in neurotrophin signaling can cause neuronal cell death, depending on the specific brain cell type affected. For example, Salehi et al. report that disrupted retrograde transport of NGF to the basal forebrain cholinergic neurons (BFCNs) causes degeneration of these neurons (I would like to note that BDNF is not a major signaling molecule in this neuronal cell population, which is why Salehi et al. find that the retrograde transport of BDNF and NT3 is below the limits of detection with the methodology used in their study). We find that an impairment of TrkB signaling causes cell death in cortical and hippocampal neurons, two cell populations that are responsive to BDNF and in which TrkB receptor isoforms alterations have been already described in Alzheimer disease (AD). As you know, all cell populations described in the two studies (BFCNs, cortical and hippocampal neurons) are affected in AD. Thus, these papers suggest that different cell populations in the brain may be affected by different genetic insults, and alternative mechanisms should be taken into account when considering therapies.

  2. NGF has a potent effect on cholinergic neurons in the basal forebrain, which are prone to degeneration in AD. The idea that NGF dysfunction is involved in AD has been around for some time, but it has never been taken seriously because of the prominence of the “Aβ” hypothesis. Now Mobley and colleagues show that APP acts to reduce the retrograde transport of NGF in these cholinergic neurons, a process that might be important for their survival. The significance of the work by Mobley et al. is that they provide a mechanistic link between APP and NGF signaling in the basal forebrain neurons, therefore putting NGF back into the center stage of the AD field. The immediate task now is to test whether this works in an AD model.

    The functional role of TrkB.T1, which is highly expressed in the brain, has been puzzling for some time now. One idea is that T1 has no function by itself, but prevents locally secreted BDNF from diffusion to distant places, and therefore ensures its local action. Another idea is that T1 can actually signal in glial cells in an unconventional way, but this idea is largely based on cell culture work, and there is no evidence that this works in vivo. The work of Tessarollo et al. sheds new light on the function of T1, using in vivo genetic approaches. They show that restoration of the physiological level of T1 by gene targeting rescues neuronal death in trisomy 16 mouse. Interestingly, T1 appears to affect selectively the Akt pathway, which is critical for neuronal survival. Given that the main function of BDNF in the brain is for synaptic plasticity rather than neuronal survival, this work offers a unique opportunity to study differential functions of BDNF in the brain.

  3. Yano and colleagues managed to proceed one step further in elucidating synaptic actions of neurotrophins. Although it was well established for quite some time that BDNF exerts presynaptic effects on the availability of presynaptic glutamate vesicles for synaptic transmission, the molecular determinants of this action were far from being understood. This paper now highlights new downstream signaling partners in the presynaptic actions of BDNF.

    The observation, in the early 1990s, that BDNF can enhance presynaptic functions of excitatory synapses (Lohof et al., 1993; Lessmann et al., 1994) was followed shortly thereafter by the discovery of an essential role of BDNF in Schaffer collateral LTP (Korte et al., 1995; Patterson et al., 1996). Also, in 1996, Figurov and colleagues (1996) found that one of the important presynaptic actions of BDNF is to avoid transmitter vesicle depletion upon repetitive activity of juvenile synapses, although this presynaptic BDNF effect cannot account for the impaired LTP in adult animals. It took another four years to learn, from the data by Jovanovic et al. (2000), that the effect of BDNF on availability of glutamate vesicles is mediated via synapsin 1, which is kind of a “chassis” for the transport of glutamate vesicles along actin filaments, to facilitate their “in time” arrival at the presynaptic active zone.

    With their most recent paper, Yano and coworkers now provide evidence that the actin-dependent motor protein Myo6 is linked via the adapter protein GIPC-1 to the transport of glutamate vesicles into the terminal. Important as this finding is, it raises—as new data usually do—a number of new issues concerning the molecular players involved in the presynaptic actions of BDNF, such as the following:

    1. What is the molecular impact of BDNF/TrkB signaling on the functions of GIPC-1 and Myo6, and is there a direct link to synapsin 1 function in vesicle transport?

    2. Since the basal presynaptic phenotype of the Myo6-/- and the GIPC-1-/- mice seems to be rather robust, acute knockdown of these proteins via siRNA in normally developed hippocampal neurons would further strengthen a direct and specific functional link of these proteins to the presynaptic modulation by BDNF.

    3. Given the also very prominent postsynaptic expression of Myo6 and GIPC-1 (and TrkB can be postsynaptic, too), the routes of postsynaptic actions of these downstream signaling molecules would be exciting to investigate. This is especially true, given that CA1-LTP is prominently expressed at postsynaptic locations (Malinow, 2003) and that Myo6 is involved in postsynaptic AMPA receptor shuttling, which mediates this form of LTP.

    4. The absence of any effects of Myo6 or GIPC knockouts, respectively, on LTP in adult animals raises questions about whether the LTP protocol was sensitive for pre- and postsynaptic BDNF signaling, or whether compensatory mechanisms were at work in these animals and might be responsible for bypassing BDNF signaling in postsynaptic LTP in these animals in adulthood.

    5. Finally, given the modulation of dopamine release via BDNF signaling (Blochl et al., 1996), it is tempting to speculate that BDNF, via the Myo6-GIPC-1 signaling, could also participate in the pathophysiology of Huntington and Parkinson diseases, known to originate from low dopamine release in the striatum. And even more exciting, Myo6 and GIPC could also participate in the trafficking of BDNF vesicles, which are known to depend on kinesin- and especially dynein-dependent motors in axons and dendrites (Gauthier et al., 2004).

    Of course, asking all these questions is much easier than finding the answers, and it is inherent to the paper by the Chao lab that we are now able to ask even more precise new questions regarding these topics.


    . Neurotrophins stimulate the release of dopamine from rat mesencephalic neurons via Trk and p75Lntr receptors. J Biol Chem. 1996 Aug 30;271(35):21100-7. PubMed.

    . Regulation of synaptic responses to high-frequency stimulation and LTP by neurotrophins in the hippocampus. Nature. 1996 Jun 20;381(6584):706-9. PubMed.

    . Huntingtin controls neurotrophic support and survival of neurons by enhancing BDNF vesicular transport along microtubules. Cell. 2004 Jul 9;118(1):127-38. PubMed.

    . Synapsins as mediators of BDNF-enhanced neurotransmitter release. Nat Neurosci. 2000 Apr;3(4):323-9. PubMed.

    . Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc Natl Acad Sci U S A. 1995 Sep 12;92(19):8856-60. PubMed.

    . BDNF and NT-4/5 enhance glutamatergic synaptic transmission in cultured hippocampal neurones. Neuroreport. 1994 Dec 30;6(1):21-5. PubMed.

    . Potentiation of developing neuromuscular synapses by the neurotrophins NT-3 and BDNF. Nature. 1993 May 27;363(6427):350-3. PubMed.

    . AMPA receptor trafficking and long-term potentiation. Philos Trans R Soc Lond B Biol Sci. 2003 Apr 29;358(1432):707-14. PubMed.

    . Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron. 1996 Jun;16(6):1137-45. PubMed.

  4. The study by Salehi, Delcroix, Mobley, and colleagues reporting that increased APP expression results in basal forebrain cholinergic neuron loss due to the inhibition of retrograde transport of NGF is most interesting. This cholinergic loss is reported in several neurodegenerative disorders including amyotrophic lateral sclerosis (ALS) (1).

    It's of interest that Hendy and Bonyhady (2) find that retrogradely transported NGF increases ornithine decarboxylase activity in rat superior cervical ganglia. I'd like to propose that there may be a feedback loop whereby ornithine decarboxylase decreases retrograde transport of NGF.

    There are several studies that support this hypothesis, including those of Yatin and colleagues (3), finding that Aβ peptides increase ODC activity; Nilsson and colleagues’ (4) finding that APP induces expression of ODC; and Virgili and colleagues’ (5) finding ODC activity increased in the SOD1 G39A transgenic mice, an animal model for ALS.

    The fact that increased activity of this enzyme is reported in H. pylori infection and is induced by insulin may help explain the association with both hyperinsulinemia and H. pylori infection in AD.

    There may be many implications if the increased activity of ODC affects the activity of other enzymes involved in ornithine metabolism. Signs of Ornithine transcarbamylase (OTC) deficiency include hyperammonemia, failure to thrive, seizures, mental retardation, and Alzheimer type 2 astrocytosis. Deficiency of this enzyme is also associated with basal forebrain cholinergic loss (6). Those with OTC deficiency are sensitive to valproic acid and this would need to be taken into consideration for those suggesting VPA therapy in AD. If substrate availability was reduced for the ornithine aminotransferase (OAT) reaction, then we may be looking at a hypoglutamatergic state. Reduced glutamate levels have been reported in several brain regions in Down syndrome (DS) (7). Are these problems further compounded in DS due to the fact that we have the gene pyridoxal kinase on Chr 21? Pyridoxal 5'-phosphate is a cofactor for ODC?

    The fact that vitamin E and resveratrol have been shown to inhibit ODC activity may give us reason to suspect that the ODC inhibitors antizyme and difluoromethylornithine (DFMO) may be a useful treatment in those with AD, DS, and ALS. When taking this into account, I note in the study by Tournoy et al. (8) that loss of presenilin in mice is associated with an autoimmune disease which mimics systemic lupus erythematosis. Elevated polyamines are seen in lupus-prone mice and interestingly, treatment of these mice with DFMO resulted in a significant increase in lifespan (9). Do those with PS1 mutations have increased ODC activity? Is ODC a longevity gene?


    . Disease-related regressive alterations of forebrain cholinergic system in SOD1 mutant transgenic mice. Neurochem Int. 2005 Apr;46(5):357-68. PubMed.

    . Retrogradely transported nerve growth factor increases ornithine decarboxylase activity in rat superior cervical ganglia. Brain Res. 1980 Oct 27;200(1):39-45. PubMed.

    . Alzheimer's amyloid beta-peptide associated free radicals increase rat embryonic neuronal polyamine uptake and ornithine decarboxylase activity: protective effect of vitamin E. Neurosci Lett. 1999 Mar 19;263(1):17-20. PubMed.

    . Antibody-bound amyloid precursor protein upregulates ornithine decarboxylase expression. Biochem Biophys Res Commun. 2006 Mar 24;341(4):1294-9. PubMed.

    . Regional and temporal alterations of ODC/polyamine system during ALS-like neurodegenerative motor syndrome in G93A transgenic mice. Neurochem Int. 2006 Feb;48(3):201-7. PubMed.

    . Evidence for forebrain cholinergic neuronal loss in congenital ornithine transcarbamylase deficiency. Metab Brain Dis. 2000 Mar;15(1):83-91. PubMed.

    . Excitatory amino acids and monoamines in parahippocampal gyrus and frontal cortical pole of adults with Down syndrome. Life Sci. 1997;60(15):1231-7. PubMed.

    . Partial loss of presenilins causes seborrheic keratosis and autoimmune disease in mice. Hum Mol Genet. 2004 Jul 1;13(13):1321-31. PubMed.

    . Successful treatment of lupus nephritis in MRL-lpr/lpr mice by inhibiting ornithine decarboxylase. Kidney Int. 1991 May;39(5):882-90. PubMed.

    View all comments by Mary Reid
  5. The articles by Dorsey and colleagues and Mobley's group (see Salehi et al., 2006) are particularly relevant to recent findings that our group has generated using human autopsy material to investigate the neurobiological mechanisms underlying cholinotrophic basal forebrain dysfunction during the progression of AD (see review by Counts and Mufson, 2004). Findings derived from autopsy material harvested from people with a premortem clinical diagnosis of non-cognitive impairment, mild cognitive impairment, and AD revealed that subtle alterations or shifts in the balance between proNGF and the high-affinity TrkA receptor for NGF may play a crucial role in the survival of cholinergic neurons during the progression of AD. Shifts in the balance between NGF and its high-affinity receptor may dysregulate pro-survival and initiate apoptotic signaling leading to cholinergic basal forebrain neuronal death. The findings of Dorsey et al. suggest that relatively small imbalances in the physiological levels of TrkB receptor isoforms affect neuronal survival by altering BDNF-induced pro-survival signaling. Together, these findings suggest that the success of exogenous neurotrophin therapy for neurological diseases such as AD may depend upon a fine balance between neurotrophin and receptor interactions.

    Several years ago, it was demonstrated that there is a reduction in retrogradely transported NGF within cholinergic neurons of the nucleus basalis in AD (Mufson and Kordower, 1997). The article by Salehi et al. provides information that APP acts to reduce the retrograde transport of NGF in the cholinergic cortical projection neurons, a process important for their survival and relevant to their selective vulnerability in AD. Moreover, this study links the amyloid hypothesis to cholinergic basal forebrain degeneration and their ultimate demise in AD.

    Together, these observations provide evidence of the complex nature of neurotrophin/receptor as well as defects in neurotrophin retrograde transport as putative functional deficits which ultimately play a role in neuronal survival and death in various disease states. Interestingly, Costantini et al. (2005) showed that a TrkA-to-p75NTR molecular switch activates amyloid-β expression during aging. This novel information demonstrates that subtle shifts in neurotrophin receptor balance not only can affect neuron survival in neurological disease but also during the normal aging process.


    . A TrkA-to-p75NTR molecular switch activates amyloid beta-peptide generation during aging. Biochem J. 2005 Oct 1;391(Pt 1):59-67. PubMed.

    . The role of nerve growth factor receptors in cholinergic basal forebrain degeneration in prodromal Alzheimer disease. J Neuropathol Exp Neurol. 2005 Apr;64(4):263-72. PubMed.

    . In vivo restoration of physiological levels of truncated TrkB.T1 receptor rescues neuronal cell death in a trisomic mouse model. Neuron. 2006 Jul 6;51(1):21-8. PubMed.

    . Nerve growth factor in Alzheimer's disease: defective retrograde transport to nucleus basalis. Neuroreport. 1995 May 9;6(7):1063-6. PubMed.

    . Increased App expression in a mouse model of Down's syndrome disrupts NGF transport and causes cholinergic neuron degeneration. Neuron. 2006 Jul 6;51(1):29-42. PubMed.

    View all comments by Elliott Mufson

Make a Comment

To make a comment you must login or register.


News Citations

  1. Varicose Axons: Traffic Jams Precede AD Pathology in Mice, Men
  2. San Diego: Too Much APP Blocks Transport, Starves Down's Neurons
  3. APP Double Dose Causes Early Onset AD

Paper Citations

  1. . Failed retrograde transport of NGF in a mouse model of Down's syndrome: reversal of cholinergic neurodegenerative phenotypes following NGF infusion. Proc Natl Acad Sci U S A. 2001 Aug 28;98(18):10439-44. PubMed.
  2. . BDNF and full-length and truncated TrkB expression in Alzheimer disease. Implications in therapeutic strategies. J Neuropathol Exp Neurol. 1999 Jul;58(7):729-39. PubMed.
  3. . Precursor form of brain-derived neurotrophic factor and mature brain-derived neurotrophic factor are decreased in the pre-clinical stages of Alzheimer's disease. J Neurochem. 2005 Jun;93(6):1412-21. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . BDNF-mediated neurotransmission relies upon a myosin VI motor complex. Nat Neurosci. 2006 Aug;9(8):1009-18. PubMed.
  2. . Increased App expression in a mouse model of Down's syndrome disrupts NGF transport and causes cholinergic neuron degeneration. Neuron. 2006 Jul 6;51(1):29-42. PubMed.
  3. . In vivo restoration of physiological levels of truncated TrkB.T1 receptor rescues neuronal cell death in a trisomic mouse model. Neuron. 2006 Jul 6;51(1):21-8. PubMed.
  4. . Neurotrophins and dementia-keeping in touch. Neuron. 2006 Jul 6;51(1):1-3. PubMed.