The cell surface receptor p75 has earned something of a bad boy reputation among Alzheimer disease (AD) researchers. The receptor is pro-apoptotic and implicated in the death of cholinergic neurons in the AD brain. But take the receptor out of the peripheral nervous system and you’ll see its “sympathetic” side. In this week’s PNAS online, researchers led by Kuo-Fen Lee at The Salk Institute for Biological Studies, La Jolla, California, report that p75 protects peripheral sympathetic nerves against Aβ toxicity and that loss of the receptor leads to premature death in transgenic mice expressing human mutant amyloid-β (Aβ) precursor protein. The paper suggests that in some situations p75 may actually protect against Aβ. The finding could lead to new strategies for dealing with deficits in sympathetic function that occur in AD, suggest the authors.

The neurotrophin receptor p75 is a member of the tumor necrosis factor family. It is mostly expressed during development but also hangs around long after, particularly in basal forebrain neurons, which are vulnerable in AD, in the CNS. Work from Margaret Fahnestock, McMaster University, Ontario, Canada, and colleagues indicates that pro-NGF, an immature form of the nerve growth factor, might contribute to neurotoxicity in AD by activating p75 in forebrain neurons (see Peng et al., 2004). Studies in Elizabeth Coulson’s lab at the University of Queensland, Brisbane, Australia, suggest that Aβ also activates the receptor, causing the demise of neurons in the hippocampus, another vulnerable region in AD (see Sotthibundhu et al., 2008). And while there are indications that Aβ prevents the growth of sympathetic nerves as well (see Tolar et al., 1998), it is not clear if that is related to p75.

To address this possibility, Lee and colleagues tested the effect of Aβ on sympathetic neurons from the superior cervical ganglia. First author Tasha Bengoechea and colleagues found that neurite growth was the same in neurons from wild-type or p75 knockout mice, but the neurons reacted differently to Aβ42. Oligomers (made by incubating 100 μM solutions for two hours at 37 degrees) inhibited neurite growth in wild-type neurons by 40 percent; but rather than protecting from this harm, ablating p75 only made it worse. In p75-null neurons Aβ42 oligomers reduced neurite growth by a further 20 percent, suggesting that p75 actually protects sympathetic nerves against Aβ.

To see whether this matters in vivo, Bengoechea and colleagues crossed p75 knockout mice with hAPP751 transgenic mice. While the APP/p75-/- offspring appeared to develop normally in utero, about 70 percent of them died within three weeks of birth. They had profound deficits in sympathetic innervation in multiple organ systems, unlike p75 knockouts in a normal mouse APP background. The APP/p75-/- mice lost sympathetic innervation to the sweat glands of their footpads and to the ventricles of their hearts, for example. These severe defects “are consistent with the hypothesis that Aβ is able to inhibit neurite outgrowth from sympathetic neurons through p75-independent pathways, whereas p75-mediated signaling pathways are protective against Aβ-induced neurotoxicity,” write the authors. They suggest that those protective pathways may involve the p75 partner TrkA and that together they promote the survival of the neurons through NGF signaling. “This balance of Trk and p75 is critically important for the development of the nervous system, particularly the sympathetic nervous system,” said Coulson in an interview with ARF. She was not involved in this study.

That raises the question of what APP/Aβ is doing so early in the life of these animals, since these effects on the sympathetic nervous system occur well before any accumulation of Aβ is detectable. “This paper is interesting because it is a developmental system and because they can’t detect any Aβ. So they are looking perhaps at a normal function for Aβ—or APP,” said Coulson. She suggested that the absence of p75, together with the increase in APP/Aβ, might subtly change p75/TrkA signaling. Work from Dale Bredesen’s lab at the Buck Institute for Age Research, Novato, California, has recently shown that APP and p75 directly interact and alter each other’s processing, for example (see Fombonne et al., 2009).

Lee and colleagues’ contention that Aβ is the toxic factor in this system draws support from an experiment knocking out one copy of BACE—the β-secretase that is essential for amyloidogenic APP processing. “These results strongly support that sympathetic innervation deficits were due to Aβ production in the p75-deficient AD mice,” write the authors. They acknowledge that BACE has other substrates besides APP and that rescue may be partly due to other mechanisms.

While most people focus on the central nervous system when studying AD, this work is relevant to humans. As the authors point out, AD patients experience sympathetic deficits, such as a drop in blood pressure when they stand up, and it is not clear if these are due to deficits in central regulation or peripheral sympathetic innervation. “The fact that Aβ at low levels is having a physiological effect and that [Lee] can link that to human patients showing hypotension and sympathetic dysfunction is interesting. Everyone just thinks about the brain,” said Coulson.—Tom Fagan


  1. The paper reports that aggregated Aβ reduces neurite outgrowth from p75-deficient sympathetic neurons, and that p75-deficient J9 AD mice show severe sympathetic innervation defects which can be partially prevented by elimination of one allele of BACE1. Bengoechea et al. conclude that p75 can protect the sympathetic nervous system from adverse effects of Aβ, in agreement with earlier reports on a neuroprotective effect of p75 against Aβ (Zhang et al., 2003; Costantini et al., 2005).

    The p75 knockout mouse used in the study (Lee et al., 1992; p75 Exon 3-/- mouse) expresses a p75 variant by splicing out Exon 3, and the resulting protein s-p75 is also coexpressed with wild-type p75 (von Schack et al., 2001). It lacks the neurotrophin binding site and includes the cysteine-rich domain 1 and the stalk, transmembrane and intracellular domains of p75. Another p75 knockout mouse (von Schack et al., 2001) produces a truncated s-p75 protein (here designated ts-p75) that encompasses a small portion from the stalk and the entire transmembrane and intracellular domains of p75 (Paul et al., 2004); ts-p75 is membrane-associated and (when overexpressed) able to trigger apoptotic signaling. Paul et al. concluded that aspects of the phenotype of the knockout mouse with this p75 variant "may reflect a gain-of-function mutation rather than loss of p75NTR function." This may also be true for the phenotype of the p75 Exon 3-/- mouse. If defects in the sympathetic nervous system of the p75 knockout mice expressing s-p75 or ts-p75 should be partly due to signaling of the respective p75 variant, and if the reduction of sympathetic innervation observed in the p75 Exon 3-/- J9 mouse is partly caused by Aβ (as indicated by the study of Bengoechea et al.), then it should be taken into consideration that aggregated Aβ might activate the two p75 variants.

    There is indeed evidence (as I presented in the full text of the Aβ-crosslinker-hypothesis) that, in addition to the known ("upper") binding site of p75 for Aβ within the four cysteine-rich domains, the stalk region of p75 contains a second specific binding site for Aβ that is adjacent to and perhaps overlapping the transmembrane domain ("stalk binding site"). Stimulation of a p75 mutant lacking the four cysteine-rich domains of full-length p75 with aged Aβ induces Ras activation similar to stimulation of wild-type p75, in contrast with a control construct that consists only of the extracellular and transmembrane domains of p75; peptide fragments from the putative stalk binding site for Aβ prevent Aβ-induced apoptosis in rat cerebellar neurons expressing full-length p75. These observations indicate that aged Aβ can bind and activate this p75 mutant, and they also suggest that Aβ can activate the s-p75 and ts-p75 proteins which include the entire stalk binding site for Aβ or an essential part of it. Since s-p75 is expressed at substantially lower levels than full-length p75 (von Schack et al., 2001), s-p75 signaling might be too weak to trigger apoptosis or should require a longer period of Aβ stimulation for that than wild-type p75. This reflection might apply to the observation by Sotthibundhu et al. (2008) that stimulation with Aβ for 24 hours causes apoptosis in hippocampal neurons with full-length p75 but not in neurons from a p75 Exon 3-/- mouse. During development, however, continual Aβ-induced signaling of s-p75 in the absence of full-length p75 might lead or contribute to deleterious effects.

    If certain defects in the sympathetic nervous system of p75 knockout mice expressing s-p75 or ts-p75 are partly due to Aβ-induced signaling of these p75 variants, then wild-type p75 (in normal mice) obviously suppresses negative Aβ-induced effects of coexpressed s-p75 and thereby has a neuroprotective effect against Aβ. How full-length p75 could protect neurons against Aβ and how the s-p75 variant could inhibit neurite outgrowth might be answered simultaneously. The Aβ-crosslinker-hypothesis postulates that short (non-β-sheet) Aβ oligomers can crosslink the p75 stalk binding site for Aβ with proteins such as APP, prion protein, and α-synuclein and thereby mediate neurotrophic and neuroprotective cooperation between p75 and these proteins; these hypothetical cooperations are activated by neurotrophin or Aβ stimulation of p75 via the neurotrophin binding site and the upper Aβ binding site. β-sheet Aβ should compete with Aβ monomers and oligomers for binding to the stalk binding site, and at sufficient concentrations, inhibit the Aβ-mediated neurotrophic cooperations. Since s-p75 does not include the neurotrophin binding site and probably also not the upper Aβ binding site of full-length p75, the activation of pre-formed cooperation links of s-p75 would be prevented, while β-sheet Aβ would be able to stimulate s-p75 (and thereby ceramide production) via the stalk binding site, eventually causing cofactor-independent effects of p75 such as ceramide-induced Rho activation (inhibiting neurite outgrowth) or ceramide-dependent apoptosis.


    . The expression of p75 neurotrophin receptor protects against the neurotoxicity of soluble oligomers of beta-amyloid. Exp Cell Res. 2005 Nov 15;311(1):126-34. PubMed.

    . Targeted mutation of the gene encoding the low affinity NGF receptor p75 leads to deficits in the peripheral sensory nervous system. Cell. 1992 May 29;69(5):737-49. PubMed.

    . A pro-apoptotic fragment of the p75 neurotrophin receptor is expressed in p75NTRExonIV null mice. J Neurosci. 2004 Feb 25;24(8):1917-23. PubMed.

    . Beta-amyloid(1-42) induces neuronal death through the p75 neurotrophin receptor. J Neurosci. 2008 Apr 9;28(15):3941-6. PubMed.

    . Complete ablation of the neurotrophin receptor p75NTR causes defects both in the nervous and the vascular system. Nat Neurosci. 2001 Oct;4(10):977-8. PubMed.

    . p75 neurotrophin receptor protects primary cultures of human neurons against extracellular amyloid beta peptide cytotoxicity. J Neurosci. 2003 Aug 13;23(19):7385-94. PubMed.

  2. Dr. Bloechl suggests that the particularly adverse effects of Abeta in the p75 knockout mouse might involve Abeta binding to a truncated form of p75 that is expressed in the "exon 3" knockout mouse. However, unpublished studies of several several laboratories in the field, including my own, have been unable to detect either the putative alternatively spliced transcript lacking exon 3, or the putative protein that would be expressed from such a transcript. In my opinion, the "exon 3" knockout mouse is completely null for p75 protein. Thus, I accept the interpretation of Lee and coworkers that p75 is protective against Abeta for sympathetic neurons, rather than the alternative interpretation that the knockout mouse expresses an Abeta-binding truncated form of p75.

  3. The controversy about the truncated s-p75 isoform, which is reflected in Dr. Bothwell's comment, is caused by the use of different p75ExonIII-/- mouse strains. A careful study by Naumann et al. (2002) demonstrated that the level of s-p75 mRNA is strain-dependent. Analysis by reverse transcription-PCR revealed that s-p75 mRNA in P15 whole brain and medial septum (MS) accumulates at much higher levels in Sv129 mice than in C57BL/6 (B6) mice, in contrast with comparable levels of full-length p75 mRNA in both strains. Naumann et al. further observed that pure B6 animals at P15 have approximately 33 percent fewer cholinergic MS neurons than Sv129 animals, and that the number of these neurons is elevated by 6.5 percent in the p75ExonIII-/- mouse with the original mixed Sv129/BALB/c background and by 13 percent in a corresponding congenic B6 mouse (relative to controls with wild-type p75). The authors interpreted the latter results to be due to the prevention of p75-mediated developmental cell death or/and to an increased efficiency of TrkA signaling.

    The fact that B6 mice with the p75ExonIV mutation, which prevents the expression of wild-type p75 and s-p75, have a substantially higher increase in the number of cholinergic MS neurons than B6 mice with the p75ExonIII mutation (28 percent:13 percent; Naumann et al.) indicates a negative influence of s-p75 on the number of these neurons in B6 p75ExonIII-/- mice. Naumann et al. supposed that "s-p75 partially compensates for the lack" of full-length p75. That B6 p75ExonIV-/- mice also express an isoform of p75 (which is a truncated s-p75; Paul et al., 2004) should not devalue this inference. This is because this isoform is pro-apoptotic and should diminish the increase in cholinergic MS neurons rather than enlarge it. The pro-apoptotic property of the truncated s-p75 (Paul et al.) might imply a similar function of s-p75. It is a serious outcome of the study of Naumann et al. that experimental results from p75ExonIII-/- mice can significantly depend on the genetic background of the used mouse strains, especially when developmental or long-term effects of the p75ExonIII mutation are investigated.

    The background of the p75-deficient mouse strain used by Bengoechea et al. is not listed in their paper. But even with low or undetectable expression of s-p75 it cannot be excluded at present that s-p75 influences the developing nervous system, in this respect not unlike Aβ. Bengoechea et al. could not detect this latter protein in developing and adolescent mice of the used AD model but they considered it to be responsible for the more severe sympathetic innervation defects in the p75-deficient AD mouse (compared to the p75-deficient mouse without overexpression of hAPP751). There is evidence that Aβ has an additional specific binding site within the stalk region of p75 and s-p75 and also that aged Aβ can activate a p75 mutant comprising the stalk, transmembrane, and intracellular domains of p75. Therefore, the negative effect of Aβ on sympathetic innervation in p75-deficient AD mice might involve activation of s-p75. This would not call into question p75-mediated neuroprotection against Aβ as observed by Bengoechea et al. and other researchers.


    . Complete deletion of the neurotrophin receptor p75NTR leads to long-lasting increases in the number of basal forebrain cholinergic neurons. J Neurosci. 2002 Apr 1;22(7):2409-18. PubMed.

    . A pro-apoptotic fragment of the p75 neurotrophin receptor is expressed in p75NTRExonIV null mice. J Neurosci. 2004 Feb 25;24(8):1917-23. PubMed.

    View all comments by Rudolf Bloechl

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Increased proNGF levels in subjects with mild cognitive impairment and mild Alzheimer disease. J Neuropathol Exp Neurol. 2004 Jun;63(6):641-9. PubMed.
  2. . Beta-amyloid(1-42) induces neuronal death through the p75 neurotrophin receptor. J Neurosci. 2008 Apr 9;28(15):3941-6. PubMed.
  3. . Sympathetic neurite outgrowth is greater on plaque-poor vs. plaque-rich regions of Alzheimer's disease cryostat sections. Brain Res. 1998 Mar 16;787(1):49-58. PubMed.
  4. . Selective vulnerability in Alzheimer's disease: amyloid precursor protein and p75(NTR) interaction. Ann Neurol. 2009 Mar;65(3):294-303. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . p75 reduces beta-amyloid-induced sympathetic innervation deficits in an Alzheimer's disease mouse model. Proc Natl Acad Sci U S A. 2009 May 12;106(19):7870-5. PubMed.