Dementia strikes in many forms besides Alzheimer disease, and two recent reports take a look at the biochemical basis for dementias in other settings. One paper, from the lab of Christopher Power, University of Alberta, Edmonton, shows that proteolytic processing of a cytokine produces a peptide with retooled receptor binding specificity. This processing converts the cytokine from helpful to harmful, as the truncated form causes neurotoxicity via the G protein-coupled receptor (GPCR) CXCR3. In the other paper, Eugenia Gurevich and coworkers at Vanderbilt University Medical Center in Nashville, Tennessee, report changes in the proteins that regulate GPCRs are specific to Parkinson disease patients with dementia, and not those without. This is the same subset of PD patients who often show resistance to L-dopa treatment.

The cytokine story, which is reminiscent of the proteolysis processing of amyloid precursor to toxic Aβ, starts with the activation of HIV-infected immune cells in the brain. These cells produce inflammatory mediators, including cytokines. Previously, Power and colleague Christopher Overall, University of Vancouver, British Columbia, showed that cleavage of the chemokine stromal cell-derived factor (SDF)1α by matrix metalloprotease 2 in activated monocytic cells produces a truncated, neurotoxic peptide product, SDF(5-67). Their new work, published online in last week’s PNAS, shows that the neurodegenerative properties of SDF(5-67) stem from an altered receptor binding specificity.

Although an SDF(5-67) fragment has been implicated in dementia associated with HIV, no one had ever measured the peptide in the brain, so lead author David Vergote of the Power lab and colleagues checked the brain tissue of four HIV-infected patients, two with and two without dementia. They found elevated SDF1α and SDF(5-67) in all the HIV-positive tissue compared with negative controls, with the highest amounts in the two patients with dementia. Immunoreactivity to SDF(5-67) occurred in macrophage/microglial cells, which also expressed MMP-2.

Having established that SDF(5-67) is, in fact, present in brain of people infected with HIV, the researchers took a closer look at its effects on different cells in vitro. They found that the SDF(5-67) fragment induced inflammatory genes in monocytes or astrocytes. In neurons, the peptide depressed whole cell currents and triggered caspase-3 and p53-dependent cell death. The parent peptide had none of these effects.

Intact SDF-1α binds to the chemokine receptor CXCR4, but the truncated peptide does not. It does, however, have affinity for CXCR3, and the authors showed that the peptide bound to this receptor on cultured cells. Blocking the interaction on neurons with anti-receptor antibodies, an antagonist peptide, or siRNA to the receptor prevented SDF(5-67)-induced neurotoxicity. In vivo, they used a mouse model of AIDS dementia, involving injection of SDF(5-67) directly into the striatum, to show that blocking CXCR3 with a receptor antagonist peptide prevented astrogliosis, microgliosis, neuronal loss, and behavioral problems. The peptide may do its damage both by direct toxicity and by indirect incitement of neuroinflammation, and the results suggest both of these effects are mediated by the switch in binding specificity to CXCR3.

Parkinson disease can also feature dementia in about 30-40 percent of patients as the disease progresses. The paper by Gurevich and colleagues, published in the Neurobiology of Aging online on November 27, shows that PD patients with dementia have distinct biochemical changes in the striatum compared to healthy controls or PD patients with no dementia. Working with Jeff Joyce at the Sun Health Research Institute, Sun City, Arizona, first author Evgeny Bychkov and colleagues establish that postmortem tissue from only this subset of patients displays increases in the protein and messenger RNAs for regulators of GPCRs, namely the arrestins and G protein-coupled receptor kinases (GRKs).

Arrestins uncouple GPCRs from downstream signaling molecules and can enhance their degradation; GRKs also modulate the signaling activity of GPCRs. To look for changes in these pathways, the authors collected postmortem brain samples from 21 patients with PD and 16 normal controls. Western blotting and RNase protection assays showed that the protein and messenger RNAs for arrestins 2 and 3, and GRKs 3 and 5, were elevated up to twofold in the striatum in the subset of Parkinson patients with dementia, but not in those without, or in control brains. Because arrestins not only uncouple GPCRs from G proteins, but also stimulate specific signaling pathways when they associate with the receptors, the researchers checked the protein levels of the secondary signaling molecules Erk, Akt, and GSK3, and found each was increased in the Parkinson with dementia group. In contrast, the dopamine receptor regulator DAARP-32 was not.

Dementia in PD sometimes comes along with amyloid plaques and neurofibrillary tangles, but the changes in signaling proteins were not related to the presence of this kind of pathology. Thus, the changes seemed specific for dementia, though more studies with larger samples will be necessary to confirm this.

Increases in arrestins/GRKs dampen GPCR signaling, so one might predict they would be decreased in PD as compensation for lower dopamine levels. The researchers did see some trend toward downregulation of the proteins in PD, but this situation was clearly reversed in PD with dementia. Consistent with this, the same group has previously reported that the dopamine receptor is downregulated in PD with dementia (Joyce et al., 2002).

PD with dementia is associated with loss of response of Parkinson symptoms to L-dopa. Based on this data, and their previous studies with nonhuman primates, the authors speculate that in patients who are sensitive to L-dopa, the drug may act via dopamine receptors to keep arrestin/GRK levels down. However, if patients have persistently elevated levels of arrestins/GRKs, they could end up with drug resistance via downregulation of DA receptors, and dementia via perturbation of multiple signaling mechanisms.—Pat McCaffrey


  1. The paper is very fascinating and opens up new avenues in Parkinson research.

    Recently, Fountaine and his coworkers published a paper in J. Neurochem. Res. which is also very interesting. They developed a system in which levels of α-synuclein can be acutely suppressed by using RNA interference (RNAi) in a physiologically relevant human dopaminergic cellular model. By using small interfering RNA (siRNA) molecules targeted to endogenous α-synuclein, they achieved 80 percent protein knockdown. They also showed that α-synuclein knockdown has no effect on cellular survival either under normal growth conditions over 5 days or in the presence of the mitochondrial inhibitor rotenone.

    Knockdown does, however, confer resistance to the dopamine transporter (DAT)-dependent neurotoxin N-methyl-4-phenylpyridinium (MPP[+]). Collectively, the data infers that α-synuclein suppression decreases dopamine transport in human cells, reducing the maximal uptake velocity (Vmax) of dopamine and the surface density of its transporter by up to 50 percent. But my thinking is how one can correlate those results with this paper? Do GPCR expression and dopamine transporters have a coupling effect or diverse effects? Do they take the direct or indirect pathway of neural transmission?


    . RNA interference-mediated knockdown of alpha-synuclein protects human dopaminergic neuroblastoma cells from MPP(+) toxicity and reduces dopamine transport. J Neurosci Res. 2007 Feb 1;85(2):351-63. PubMed.

  2. Proteolytic processing of SDF-1 reveals a new receptor specificity mediating HIV-associated neurodegeneration
    HIV-1 associated dementia (HAD) is characterized by a constellation of cognitive, behavioral, and/or motor abnormalities affecting a significant portion of infected children and adults with human immunodeficiency virus (HIV) (Ellis et al., 2002; Epstein and Gelbard, 1999; McArthur et al., 1999). Although the incidence of HAD has dropped to about 10 percent of all infected subjects with the advent of highly active antiretroviral therapy (HAART) (Sacktor et al., 2001), HAD remains a persistent problem in infected individuals as resistance to therapy grows with viral strain mutations and because of the limited ability of drugs to penetrate the blood-brain barrier. Thus, HAD will continue to be a significant complication of advanced HIV-1 disease (Carpenter et al., 2000; McArthur et al., 1999).

    The pathological correlate of HAD, HIV-1 encephalitis (HIVE), is characterized by the presence of HIV-1-infected and immune-activated mononuclear phagocytes (MP, brain macrophages, and microglia). The association among HIVE, inflammation, and neuronal injury is substantial (Glass et al., 1993). HIV-1-infected MPs secrete cytotoxic factors and viral proteins that cause synaptic damage, neuronal degeneration and cellular dropout (Gendelman, 1997; Gonzalez-Scarano and Martin-Garcia, 2005).

    Dr. Richard Power’s group in Canada has been very active in elucidating the molecular mechanisms leading to neuronal cell injury and death in HAD and other neurodegenerative disorders. In 2003, they published a paper in Nature Neuroscience demonstrating that matrix metalloproteinase-2 (MMP-2), upregulated by HIV-1-infected macrophages, readily cleaves four amino acids from the N-terminal of stromal cell-derived factor 1 (SDF-1), the endogenous ligand for CXCR4, resulting in the formation of SDF-1 (5-67), which becomes a potent neurotoxin (Zhang et al., 2003). In the current publication they extended this previous finding and provided evidence demonstrating the presence of SDF-1 (5-67) in the brain of an HIV-1 dementia subject who died from this disease (Vergote et al., 2006). Further, they provided new evidence that SDF(5-67) changes the protein’s receptor specificity from the normal CXCR4 receptor to CXCR3. SDF(5-67)-mediated neurotoxicity is through CXCR3 instead of CXCR4, leading to neurodegeneration in HAD.

    It is known that HIV-1-infected and/or immune-activated macrophages could regulate astrocyte SDF-1 production during HAD (Peng et al., 2006). The current finding indicates that SDF-1 produced from activated astrocytes could be potentially cleaved by MMP-2 released from infected and activated macrophages and the resulting SDF(5-67) could lead to mediated neurodegeneration. Similarly, microglia activation is an important feature of Alzheimer disease (AD) and is likely to be a key participant in disease progression. Although the etiology and neuropathology of HAD and AD are clearly distinct, common features of both diseases are microglial activation, brain inflammation, and neuronal injury (Cotter et al., 1999). While neuronal cell injury and death is likely, the molecular mechanisms leading to this pathology remain poorly understood. Although further studies are needed, it is quite possible that similar phenomena happen in the AD brain regarding SDF-1 production, SDF-1 cleavage, and SDF-1 fragment-mediated neurodegeneration. This investigation could provide new therapeutic avenues for the treatment of HAD and possibly AD.

    Although this remains an important area of investigation, many questions remain to be answered. First: Is activated MMP-2 the only enzyme responsible for cleavage of SDF-1 to SDF(5-67) or other fragments? Second: What factors would cause the activation of MMP-2, initiating the cleavage of SDF-1 and the increase of SDF(5-67)? Third: Why does there appear to be more SDF(5-67) in HAD samples than in non-demented subjects? How is the increase in SDF(5-67) related to HAD? Is there cause and effect? And last and most importantly, how is SDF(5-67) involved in activation of caspase-3 and p53 pathways, which lead to neuronal apoptosis? What are the upstream signaling pathways for these activations? Further investigation of these questions will not only help to elucidate the mechanism in HAD, but also provide new avenues for the investigation of neuronal injury mechanisms in other neurodegenerative diseases such as AD.

    See also:

    Cotter, R., Zheng, J., and Gendelman, H. E. (1999). The role of mononuclear phagocytes in neurodegenerative disorders: Lessons from multiple sclerosis, Alzheimer's disease and HIV-1 dementia. In Advances in Neurodegenerative Disorders, J. Marwah, and H. Teitelbaum, eds. (Scottsdale, AZ, Prominent Press), pp. 203-241.

    Gendelman, H. E. (1997). The Neuropathogenesis of HIV-1-Dementia. In The neurology of AIDS, H. E. Gendelman, S. A. Lipton, L. G. Epstein, and S. Swindells, eds. (New York, Chapman and Hall), pp. 1-10.


    . Antiretroviral therapy in adults: updated recommendations of the International AIDS Society-USA Panel. JAMA. 2000 Jan 19;283(3):381-90. PubMed.

    . Progression to neuropsychological impairment in human immunodeficiency virus infection predicted by elevated cerebrospinal fluid levels of human immunodeficiency virus RNA. Arch Neurol. 2002 Jun;59(6):923-8. PubMed.

    . HIV-1-induced neuronal injury in the developing brain. J Leukoc Biol. 1999 Apr;65(4):453-7. PubMed.

    . Clinical-neuropathologic correlation in HIV-associated dementia. Neurology. 1993 Nov;43(11):2230-7. PubMed.

    . The neuropathogenesis of AIDS. Nat Rev Immunol. 2005 Jan;5(1):69-81. PubMed.

    . Human immunodeficiency virus-associated dementia. Semin Neurol. 1999;19(2):129-50. PubMed.

    . HIV-1-infected and/or immune activated macrophages regulate astrocyte SDF-1 production through IL-1beta. Glia. 2006 Nov 1;54(6):619-29. PubMed.

    . HIV-associated neurologic disease incidence changes:: Multicenter AIDS Cohort Study, 1990-1998. Neurology. 2001 Jan 23;56(2):257-60. PubMed.

    . Proteolytic processing of SDF-1alpha reveals a change in receptor specificity mediating HIV-associated neurodegeneration. Proc Natl Acad Sci U S A. 2006 Dec 12;103(50):19182-7. PubMed.

    . HIV-1 infected mononuclear phagocyte secretory products affect neuronal physiology leading to cellular demise: relevance for HIV-1-associated dementia. J Neurovirol. 2000 May;6 Suppl 1:S14-23. PubMed.

    . HIV-induced metalloproteinase processing of the chemokine stromal cell derived factor-1 causes neurodegeneration. Nat Neurosci. 2003 Oct;6(10):1064-71. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Loss of response to levodopa in Parkinson's disease and co-occurrence with dementia: role of D3 and not D2 receptors. Brain Res. 2002 Nov 15;955(1-2):138-52. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Arrestins and two receptor kinases are upregulated in Parkinson's disease with dementia. Neurobiol Aging. 2008 Mar;29(3):379-96. PubMed.
  2. . Proteolytic processing of SDF-1alpha reveals a change in receptor specificity mediating HIV-associated neurodegeneration. Proc Natl Acad Sci U S A. 2006 Dec 12;103(50):19182-7. PubMed.