It’s getting hot around calcium. A trend is afoot about something going awry with calcium regulation in Alzheimer neurons, and it makes this ion an increasingly attractive suspect in AD pathogenesis, even if important details remain murky. The trend just got another lift with a PNAS paper from the laboratory of Lennart Mucke at the Gladstone Institute of Neurologic Diseases at the University of California, San Francisco. The scientists report that the behavioral deficits of APP-transgenic mice—held up as evidence that these models of β-amyloidosis indeed have “AD-like” symptoms, but also sometimes criticized for their fickleness and variability—correlate surprisingly tightly with calbindin, a calcium-binding protein that buffers intracellular calcium levels. Mucke’s team found a similar correlation with c-fos, a calcium-dependent immediate-early gene thought to play a role in synaptic function.

If replicated, this work could be relevant in several respects. First, it establishes a molecular link between Aβ generation and behavioral deficits, although the mechanism of calbindin and c-fos’ action in these mice (and in AD) remains unclear. Second, it extends other recent work distinguishing between small, nonfibrillar Aβ species and deposited plaques by implicating only the former. Third, it suggests that calbindin could serve as a new functional marker because its levels are tightly linked to cognitive decline.

This is not calbindin’s first appearance in AD research. Just recently, Changiz Geula reported a series of papers about calbindin, including one describing a selective and age-related loss of calbindin from cholinergic neurons in non-human primates (Wu et al., 2003">). Among prior studies on calbindin is one from Mark Mattson’s lab reporting that calbindin protects cultured cells against Aβ-induced toxicity (Guo et al., 1998). A more distantly related paper in last week’s Neuron connects the protein huntingtin to calcium signaling: see ARF related news story), again linking intracellular calcium and neurodegeneration. The present work demonstrates a cause-effect relationship between Aβ42 levels and age-dependent reduction of calbindin (and c-fos) in neurons key to learning and memory, and at the same time, it correlates these levels with the learning and memory deficits seen in APP-transgenic mice.

First author Jorge Palop and colleagues analyzed the expression of calbindin and c-fos in the hippocampus of PDAPP mice and in postmortem dentate gyrus tissue of humans with sporadic AD. They found that calbindin immunocytochemistry was markedly down in granule cells of the dentate gyrus. This reduction was region-specific as pyramidal cells in the nearby CA1 region were unaffected, and it did not reflect death of neurons but reduced levels in existing neurons. Calbindin expression in these neurons declined with advancing age and increasing Aβ levels. Calbindin levels appeared to be in lockstep with c-fos levels. Reductions in both these calcium-related proteins correlated with the relative abundance of Aβ42, but not with APP expression or plaque deposition.

Does this drop in calbindin and c-fos matter functionally? Water-maze tests indicated as much. Mice with the lowest calbindin/c-fos levels also had the greatest trouble with tasks requiring spatial learning and memory retention. General motor deficits did not account for the performance differences. This is the paper’s key finding. “Behavioral deficits in hAPP FAD mice showed a striking relationship to neuronal reductions of calbindin and c-fos in the dentate gyrus,” the authors write.

Though the prime focus in AD research has long been the loss of neurons in particular brain areas, more recent attention has grown for the question of what causes neuronal and/or synaptic dysfunction in still-living neurons, either before massive neuron loss occurs or in those neurons that somehow survive an ongoing disease process nearby. Perhaps, these authors suggest, calbindin and c-fos reductions are part of larger calcium-related changes that impair the function of these hardy neurons. This notion implies that neurons molecularly altered in this way could be ultimately be saved or treated.

The present study also carries further a line of investigation that tries to show that not just Aβ fibrils and mature plaques, but particularly small nonfibrillar assemblies of Aβ are damaging and, indeed, may be the ones that cause cognitive dysfunction. Specific mechanisms linking these assemblies to calbindin and other calcium-related changes are unknown; speculation includes changes in calcium-channel function, the formation of Aβ pores in membranes, and influences on inflammation, the authors write. Co-authors included others at the Gladstone Institute and Eliezer Masliah at the University of California, San Diego.

The scientists also suggest that the field pursue a potentially important practical outcome of this work. Since behavioral testing is laborious and can produce maddeningly variable results, other, simpler and more expedient outcome measures are needed to test experimental drugs in mouse models. The standard pathological marker in current use-plaque load-correlates poorly with behavioral changes, Palop et al. write. Calbindin levels correlate tightly and could become the basis for a better surrogate marker of neuronal dysfunction in the assessment of novel treatments, perhaps with biochemical or radiological methods.

In the present study, calbindin assessment in human tissue included 15 AD cases and two nondemented controls. Many more cases need to be analyzed before one can establish how well calbindin reductions correlate with cognitive deficits in AD, the authors note.—Gabrielle Strobel


  1. Amyloid peptide-induced behavioral deficits in transgenic models of Alzheimer's disease are tightly linked to depletion of calcium-dependent proteins in granule cells of the dentate gyrus.

    This work builds on earlier observations that described the reduction in levels of calbindin D28K, a high-affinity calcium-binding protein, in the brains of patients with various neurodegenerative diseases.
    Transgenic mice carrying mutant forms of human APP accumulate amyloid plaques and exhibit learning deficits. Careful immunochemical studies show that there are also significant reductions in the levels of calbindin D28K (CB) and c-Fos in the brains of these mice, and that these seem to correlate more with the levels of soluble Ab peptide levels in brain extracts than with the amount of accumulated amyloid plaque. Reduced levels of CB were also found in brain samples from patients with AD.

    The strength of this study is the careful correlation of learning defects in genetically manipulated mice that show age-dependent
    changes in the amounts of a well-defined calcium-binding protein. Those changes correlate, albeit roughly, with Ab peptide levels rather than with recognizable plaque material. The results are consistent with the possibility that reduced levels of calcium "buffering" capacity may cause affected neurons to be more sensitive to calcium-induced cell death. The basis for the reduced levels of CB have not been determined, but the authors note that reduced CB levels could prove to be a useful surrogate marker for the detection of incipient forms of neurodegeneration.

  2. I would like to stress that already in 1993, Sutherland MK, Wong L, Somerville MJ, Yoong LK, Bergeron C, Parmentier M, McLachlan DR published in Brain Res Mol Brain Res. 1993 Apr;18(1-2):32-42 Reduction of calbindin-28k mRNA levels in Alzheimer as compared to Huntington hippocampus. Indeed the correlation between learning deficit and calbinbin is indeed appealing and deserve further experiments. What about cross breeding knock out mice for calbindin and APP transgenic mice?

Make a Comment

To make a comment you must login or register.


News Citations

  1. Expanded Huntingtin Heightens Calcium Signaling

Paper Citations

  1. . Selective age-related loss of calbindin-D28k from basal forebrain cholinergic neurons in the common marmoset (Callithrix jacchus). Neuroscience. 2003;120(1):249-59. PubMed.
  2. . Calbindin D28k blocks the proapoptotic actions of mutant presenilin 1: reduced oxidative stress and preserved mitochondrial function. Proc Natl Acad Sci U S A. 1998 Mar 17;95(6):3227-32. PubMed.

Further Reading

Primary Papers

  1. . Neuronal depletion of calcium-dependent proteins in the dentate gyrus is tightly linked to Alzheimer's disease-related cognitive deficits. Proc Natl Acad Sci U S A. 2003 Aug 5;100(16):9572-7. Epub 2003 Jul 24 PubMed.