A paper out today in Cell proposes a novel role for presenilins in calcium signaling—that of an ion channel. The work, from Ilya Bezprozvanny and colleagues at the University of Texas Medical Center at Dallas, shows that presenilins appear to function as calcium channels in the endoplasmic reticulum (ER) membrane, responsible for the passive leak of calcium from that organelle. Familial Alzheimer disease (FAD) mutations (PS1-M146V and PS2-N141I) or deletion of the presenilin genes rids cells of the calcium leak channel and leads to calcium overload in the ER. The observations could explain the abnormal calcium signaling seen in human FAD fibroblasts (reviewed in Smith et al., 2005).

The location of presenilins in the ER membrane, their nine transmembrane spanning domains, and the fact that mutations in presenilins result in deranged calcium movement in cells led lead author Huiping Tu and coworkers to ask if the protein directly mediates calcium transport. The researchers expressed both wild-type and mutant human presenilins in Sf9 insect cells, and tested the proteins for the ability to reconstitute divalent ion channels in lipid bilayers in vitro.

First, the researchers tested ER microsomes from the insect cells containing predominantly the uncleaved, holo form of presenilins. They fused purified microsomes with planar lipid bilayers and measured divalent ion current flow across the membrane. Microsomes from PS1 or PS2 expressing Sf9 cells, but not untransfected cells, supported current flow, suggesting that the proteins formed functional channels. Channel formation did not require γ-secretase activity, as the catalytically dead PS1-D257A still reconstituted a current. On the other hand, the FAD mutants PS1-M146V and PS2-N141I did not. Microsomes containing both wild-type PS1 and the M146V mutant also failed to conduct current in the bilayer experiments, suggesting that the mutant acted as a dominant negative to shut down the wild-type activity. The results were confirmed with purified PS1 and the M146V mutant, which formed channels of very low conductance when reconstituted into lipid bilayers. Not all FAD mutants were channel-negative, however—the δE9 FAD mutant showed enhanced current flow.

To ask if the presenilins functioned as physiological calcium channels in cells, the researchers looked at calcium signaling in mouse embryonic fibroblasts derived from coauthor Bart De Strooper’s presenilins 1 and 2 double knockout (DKO) mice. Using Fura2 calcium imaging, they found that resting cytosolic calcium levels in DKO cells were lower than wild-type cells. In addition, calcium mobilization from ER stores in response to IP3 signaling was much higher compared to wild-type cells. The DKO cells had larger ER calcium stores, as measured by higher and longer cytosolic calcium increases after treatment with the calcium ionophore ionomycin. Finally, blocking calcium uptake into the ER with the SERCA pump inhibitor thapsigargin resulted in a large increase in cytosolic calcium in wild-type cells, but not in DKO cells, presumably because they lacked a leak channel.

Further evidence that presenilins did indeed function as the leak channel came from rescue experiments, which showed that adding back PS1 or PS2 in DKO cells normalized calcium levels and mobilization. The ability of PS1 or PS2 wild-type and mutant proteins to restore normal calcium behavior in response to IP3, ionomycin or thapsigargin mirrored precisely their ability to function as calcium channels in the in vitro lipid bilayer experiments: Wild-type, PS1 δE9, and D257A mutants reconstituted normal calcium signaling, while PS1-M146V and PS2-N141I mutants did not. In addition, the M146V mutant acted as a dominant negative when coexpressed with wild-type protein in cells.

Together, the results suggest that presenilins form a passive ER calcium leak channel. For a final test of this idea, the researchers isolated ER microsomes and filled them with calcium. The calcium leaked out when the inflow was blocked with thapsigargin, and they found the leak was faster in microsomes from PS1-expressing Sf9 cells, but not from M146V-expressing cells. The leak was slower in microsomes from DKO cells compared to wild-type, but this was restored by expression of PS1. They also directly measured ER calcium in cells with the ER dye Mag-Fura2 and found that in agreement with their Fura2 studies, ER calcium levels were doubled in DKO MEFs. Transfection with PS1, PS1-δE9 or PS1-D257A (but not M146V) reduced calcium to wild-type levels.

The authors propose, based on all these results, that the uncleaved forms of PS1 and PS2 function as ER calcium leak channels. In DKO cells or in cells with FAD mutant presenilins, the lack of this leak leads to high calcium in the ER and exaggerated calcium release upon stimulation. The results are consistent, but the story is far from clear, as another group obtained exactly the opposite results from a recent study in DKO fibroblasts that used different methods (Kasri et al., 2006).

The current data provide support for the “Ca2+ hypothesis of AD,” which attributes AD pathophysiology to deranged calcium signaling in neurons. The failure of FAD mutants, at least the few tested so far, to form channels would be consistent with a loss-of-function model for the mutants. According to Malcolm Leissring of the Scripps Institute in Jupiter, Florida, “it is unclear if these or any other effects of presenilins will come anywhere near to dethroning the Aβ hypothesis, though they might go some way in explaining the different phenotypes associated with specific PS mutations.”—Pat McCaffrey

Tu H, Nelson O, Bezprozvanny A, Wang Z, Lee S, Hao Y, Serneels L, De Strooper B, Yu G, Bezprozvanny I. Presenilins form ER Ca2+ leak channels, a function disrupted by familial Alzheimer’s disease-linked mutations. 2006 September 8; Cell 126:981-993. Abstract


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. Comment from H. De Smedt and the IP3-team in Leuven

    Discrepancies in Two Recent Papers on ER Ca2+-leak Channels in Presenilin1, -2 Double Knockout Cells
    This paper describes presenilin (PS)-related mechanisms that affect Ca2+ leak from the endoplasmic reticulum (ER). However, it points to a very different mechanism—Ca2+-channel leak properties of presenilin—to that which we have recently published: upregulation of type 1 inositol 1,4,5-trisphosphate receptor (IP3R1) (Kasri et al., 2006). Although these two conclusions are not mutually exclusive, the niggling point is that both papers report very different and even sometimes opposing experimental findings. There is no obvious explanation for these discrepancies, but it is clear that all methodologies currently applied to evaluate ER Ca2+ concentrations and ER Ca2+ leak are imperfect and often lead to contradictory results. This was extensively discussed by Clark Distelhorst and Gordon Shore in their recent review of the conflicting findings regarding the effects of Bcl-2 proteins on ER Ca2+ (Distelhorst and Shore 2004).

    Before analyzing potential reasons why different experimental findings have been obtained in these two presenilin papers, I will first summarize the findings in each that are not disproved by findings in the other.

    The most important observation in our paper is that there is an isoform-specific fourfold upregulation of IP3R1 in murine embryonic fibroblast (MEF) PS double knockout (dko) cells. This observation was very solid, as it was done not only using isoform-specific antibodies against both isoforms of IP3R, but also using a common antibody that allows a simultaneous detection of both receptor isoforms. The latter allows a determination of the relative expression of IP3R-isoforms and is therefore independent of load controls that are problematic when comparing different cell types. One such control, actin, can be especially problematic because the dko cells are characterized by quite a different cellular morphology. Moreover, the fact that the enhanced Ca2+ leak could be reversed using siRNA-mediated downregulation of specifically IP3R1 demonstrates that the increased level in IP3R1 is the primary cause of this leak. The role of IP3R1 as an ER Ca2+-leak channel is in agreement with findings from other groups (Oakes et al., 2005).

    The major observation in the paper by Tu et al. is that wild-type presenilins, but not PS1-M146V and PS2-N141I FAD mutants, can form low-conductance divalent-cation-permeable ion channels in planar lipid bilayers. These channel properties of presenilins were confirmed using lipid bilayer reconstitution of the purified proteins, and it suggests a Ca2+ signaling function for presenilins which would provide further support for the “Ca2+ hypothesis of AD.”

    While the basic observations of both papers may point to two of the potential mechanisms for ER Ca2+ leak, it should be clear that many other leak pathways may coexist, and, as was adequately discussed by Tu et al., the exact identity of ER Ca2+ leak channels still largely remains an “enigma of Ca2+ signaling” (Camello et al., 2002).

    This is where the deviating observations and conflicting results come in. Even worse, the basic observation about the ER Ca2+ level is exactly the opposite in both papers. Although essentially the same immortalized mouse embryonic cell lines (MEF and MEF dko fibroblasts) were used, we found that MEF PS dko cells had a lower ER Ca2+ level, explained by increased IP3R1-mediated leak, whereas in the Tu et al. paper the opposite was found—as may be expected if the leak occurs via presenilins, which have been knocked out. There is no explanation for this discrepancy except that different techniques were used to measure ER (Ca2+) and the Ca2+ leak. In our hands, targeted aequorins were used to estimate ER (Ca2+) and saponin-permeabilized monolayers were used for estimating the leak rate. In the Tu et al. paper, Mag-Fura was used for ER Ca2+ measurement, and Fura-2 fluorescence was used for evaluating Ca2+ fluxes in microsomes.

    It is very clear that all these methods have their own drawbacks, and the main problem would appear to be what fraction of the ER is actually measured. In each of the methods used there are uncertainties as to whether only the ER is targeted and whether it may be disturbed by the preparation procedures. As a result it may very well be that different subfractions of the ER have been evaluated. The depletion of the ER in the aequorin method or the use of saponin may have affected the structure of the ER. The preparation of microsomes, on the other hand, certainly results in a mixture of membrane fractions, the distribution and purity of which may also be different for different cell types. A second drawback in these studies is the means used to deplete the ER: ionomycin is not specific and will deplete all Ca2+-containing compartments, whereas thapsigargin will target only those compartments filled up by the SERCA-type Ca2+ pump. In our work, the saponin-permeabilized monolayers largely reflect the thapsigargin-dependent Ca2+ stores, and the small thapsigargin-independent part was subtracted in the calculations. For the preparations in the Tu et al. paper, both for intact cells and for microsomes, large differences were observed between ionomycin-releasable and thapsigargin-releasable Ca2+. These data are interpreted as a measure of the total Ca2+ content and the rate of Ca2+ release, respectively. It is, however, not established that the ionomycin-derived Ca2+ content only reflects the ER. Furthermore, the thapsigargin-induced Ca2+ release rate in intact cells may not only reflect ER Ca2+ release but also the rate of Ca2+ efflux driven by PMCA or Na+/Ca2+ exchanger and Ca2+ uptake by the thapsigargin-independent stores (Golgi, mitochondria). Moreover, in microsomal preparations, the Ca2+-release rate will not only depend on the distribution of presenilin in the different microsomal fractions but also on their diameter, composition, and aggregation, and these parameters may be variable if preparations have to be made from different cell types. One should keep in mind that the MEF dko cells are defective cells that grow more slowly and have different morphology as compared to the wild-type cells. This may result in microsomal fractions with different biochemical and physical properties.

    In conclusion, both papers have provided evidence for new mechanisms of ER Ca2+ control, and these mechanisms are clearly related to presenilin expression and may therefore play a role in the pathology of AD. However, the quantitative significance of these leak pathways in intracellular compartments, and particularly in different ER fractions, is very difficult to evaluate. This is largely because there are no fool-proof methods for obtaining preparations that truly reflect and measure the properties of the ER in a real cellular context. Moreover, the cellular heterogeneity of the ER and the existence of other membrane compartments, where IP3Rs or presenilins may operate, remain difficult to fully appreciate. Finally, the molecular tools to evoke ER-related Ca2+ fluxes are imperfect and not equally reliable in all conditions. Appreciation of the significance of the above-mentioned mechanisms for neuronal function and dysfunction will have to wait until more adequate techniques for measuring cellular Ca2+ signals are available.


    . Up-regulation of inositol 1,4,5-trisphosphate receptor type 1 is responsible for a decreased endoplasmic-reticulum Ca2+ content in presenilin double knock-out cells. Cell Calcium. 2006 Jul;40(1):41-51. PubMed.

    . Bcl-2 and calcium: controversy beneath the surface. Oncogene. 2004 Apr 12;23(16):2875-80. PubMed.

    . Proapoptotic BAX and BAK regulate the type 1 inositol trisphosphate receptor and calcium leak from the endoplasmic reticulum. Proc Natl Acad Sci U S A. 2005 Jan 4;102(1):105-10. PubMed.

    . Calcium leak from intracellular stores--the enigma of calcium signalling. Cell Calcium. 2002 Nov-Dec;32(5-6):355-61. PubMed.

  2. This recent study by Tu et al. (2006) provides a much-needed advance toward understanding how presenilin (PS) mutations can alter ER Ca2+ signaling patterns. Cumulative data over the past several years have clearly shown that cells (both neurons and non-neuronal model systems) display marked increases in evoked Ca2+ release from the ER. However, the mechanism by which presenilin can influence Ca2+ stores has remained utterly elusive. An inherent hurdle has been the level at which the previous studies have been conducted: examining individual ER channel activity in biological preparations such as cell cultures and brain slices is rather intractable (with the exception of work from Kevin Foskett’s lab), while the biochemical and molecular biological approaches are too minimalist.

    The planar lipid bilayer approach was, therefore, an ideal preparation to start addressing presenilin function in membranes and its relation to the Ca2+ signaling dysregulation seen with certain AD-linked presenilin mutations. This technique allows one to insert specific channels of interest into a modified “model membrane” in order to observe and manipulate their function. Given that wild-type presenilin can form cation-permeable channels in these lipid bilayer models—and that the PS1-M146V and PS2-N141I mutants are impaired in this function—the extension to biological models using murine embryonic fibroblasts (MEFs) and rescue experiments in PS double knockouts (PS-DKOs) becomes easier to interpret and certainly more powerful. Hypothesizing that ER stores overfill due to impaired Ca2+ leak current through presenilin channels is a novel proposition, and it is backed up by clear mechanistic evidence in both model membranes and biological systems.

    This study is particularly elegant in that it contributes to our understanding of presenilin function at several levels. At the basic science level, we have new insight into the role of presenilin in the ER—why it is even located there (addressing the spatial paradox)—and a novel candidate for the leak channel—which has been inferred but never really seen. And, since the leak function is separate from its role in the γ-secretase complex, these results also imply an additional, separate, and parallel role of the presenilins in maintaining Ca2+ homeostasis. At the neuropathology level, this is the first real mechanistic study that can point to how AD-linked presenilin mutations can result in increased ER Ca2+ stores through a loss of function.

    At a more global level, there is still much to be explored regarding how mutant PS and ER Ca2+ signaling dysregulations are linked to the pathophysiology of AD. Primarily, can impaired Ca2+ leak channels be linked to Aβ plaque formation and neurofibrillary tangles that are diagnostic of AD, or are they a separate and independent phenomenon in the disease process? Interestingly, in Tu’s study, not all PS mutations generated the same channel phenotype, and this will ultimately need reconciling. The PS1-δE9 mutation resulted in an apparent gain of function with increased cation conductance—yet in humans, the M146V and δE9 mutations ultimately result in the same disease state. In the basic research realm, an additional point that needs reconciling is the conflicting data regarding SERCA pump blockers (e.g., thapsigargin). In several studies examining effects of mutant PS1, application of SERCA blockers results in enhanced Ca2+ release into the cytosol (Guo et al., 1997; Leissring et al., 2000; Herms et al., 2003; Stutzmann, personal observation in brain slice preparations), which is at odds with the proposed reduction in the PS-leak channel conductance and the raw data presented in the Tu et al., study.

    Determining if/how presenilin interacts with other ER Ca2+ channels such as the IP3 and ryanodine receptors is an important next step, particularly in light of several recent studies demonstrating an increase in ryanodine receptor number and function in PS1-M146V expressing neurons (Chan et al., 2000; Smith et al., 2005; Stutzmann et al., 2006). So, there is likely still more to the PS story that has yet to be uncovered, but this study provides vital information to both the basic science and AD fields, and infuses new life into the Ca2+ hypothesis of AD. And, perhaps most importantly, it provides a clear new direction with which to focus future PS-Ca2+ signaling studies.


    . Alzheimer's presenilin mutation sensitizes neural cells to apoptosis induced by trophic factor withdrawal and amyloid beta-peptide: involvement of calcium and oxyradicals. J Neurosci. 1997 Jun 1;17(11):4212-22. PubMed.

    . Capacitative calcium entry deficits and elevated luminal calcium content in mutant presenilin-1 knockin mice. J Cell Biol. 2000 May 15;149(4):793-8. PubMed.

    . Capacitive calcium entry is directly attenuated by mutant presenilin-1, independent of the expression of the amyloid precursor protein. J Biol Chem. 2003 Jan 24;278(4):2484-9. PubMed.

    . Presenilin-1 mutations increase levels of ryanodine receptors and calcium release in PC12 cells and cortical neurons. J Biol Chem. 2000 Jun 16;275(24):18195-200. PubMed.

    . Calcium dysregulation in Alzheimer's disease: recent advances gained from genetically modified animals. Cell Calcium. 2005 Sep-Oct;38(3-4):427-37. PubMed.

    . Enhanced ryanodine receptor recruitment contributes to Ca2+ disruptions in young, adult, and aged Alzheimer's disease mice. J Neurosci. 2006 May 10;26(19):5180-9. PubMed.

  3. The work by Bezprozvanny and colleagues is unquestionably a breath of fresh air in the field of AD, especially for those interested in the “Ca2+ overload” hypothesis for the pathogenesis of this devastating disease. It is particularly interesting given that an increasing number of groups are beginning to address this issue from the point of view of the internal stores. In fact, up until now only two papers focused the reader’s attention on Ca2+ levels inside the stores using direct approaches: one mentioned by Bezprozvanny and colleagues (Kasri et al., 2006), and one coming from our group (Zatti et al., 2006), which was not mentioned. These two papers, however, show results which need to be considered in a open discussion on the Cell’s paper.

    The first finding obtained by Bezprozvanny and colleagues, showing that PSs are leak channels, does not contradict our published data: we have repeatedly demonstrated that overexpression of wt-PS2 and, to a lesser extent, also of wt-PS1, reduces the ER Ca2+ level in different cell models (Zatti et al., 2004; Giacomello et al., 2005; Zatti et al., 2006). Surprisingly, what is not consistent with our findings is the fact that, in our models, the expression of various FAD-linked PS mutants often results in a “gain of function” if considering the effect of PSs on the ER leakage. In fact, the ability of wt PSs to reduce ER Ca2+ release is also shared by different PS mutants: PS2-M239I (Zatti et al., 2004); PS2-T122R (Giacomello et al., 2005); PS2-N141I, PS2-D366A, PS1-A246E, PS1-M146L, PS1-P117L (Zatti et al., 2006). Notably, these mutations include two mentioned by Bezprozvanny and colleagues (PS2-N141I and PS1-M146L/V), as well as one devoid of γ-secretase activity (PS2-D366A).

    We have published data (Zatti et al., 2006) showing that the store Ca2+ content is unchanged or even reduced when PSs are expressed in different cell models either stably (such as in human FAD fibroblasts, HEK293, and SH-SY5Y clones) or transiently (such as in HeLa and SH-SY5Y cells, MEFs, and primary cultures of rat neurons). These results were obtained by using two different methodological approaches, that is, by cytosolic Ca2+ imaging with fura-2 (as described by Bezprozvanny and colleagues) and by recombinant ER-targeted aequorin (as described by Kasri et al., 2006). No evidence of an exaggerated Ca2+ release was found in cells expressing any of the investigated PS2 (M239I, -T122R, -N141I, -D366A) or PS1 (-A246E, -L286V, -M146L, -P117L) mutations. Similarly, no Ca2+ overload was found when directly measuring ER and Golgi apparatus Ca2+ levels (using appropriately targeted aequorins) in HeLa and SH-SY5Y cells overexpressing the above-mentioned PS mutants (Zatti et al., 2006), as well as in the stable clones HEK293/PS1-M146L and SH-SY5Y/PS2-T122R and in DKO MEF cells (our unpublished data, and see also Kasri et al., 2006). Consistently, DKO MEFs did not show an increased Ca2+ store content if compared to MEFs expressing only the wt-PS1 when measuring the cytosolic Ca2+ changes induced by store depletion with cyclopiazonic acid (Zatti et al., 2006).

    Thus, the reasons for these discrepancies cannot merely be due to differences in the methodology employed for Ca2+ measurements. The true reasons for such discrepancies should indeed be sought if one wishes to shed light on this complex phenomenon. Conversely, ignoring them does not help the AD community and, more importantly, hinders scientific progress.

    We believe that, among the different models employed in this type of investigation, human fibroblasts from FAD patients should be given at least the same weight as MEFs, not least because we are interested in the human pathology. A reduced and not an exaggerated Ca2+ release was detected by cytosolic fura-2 measurements in human FAD-fibroblasts carrying the PS1-M146L (two patients) or the PS1-P117L (one patient), whose donors were presenting a devastating early-age-of-onset AD (30 years for the PS1-P117L-carrying subject; Zatti et al., 2006). A stronger reduction in ER Ca2+ content was inferred with the same technique in human FAD-fibroblasts carrying the PS2-M239I (two patients) or the PS2-T122R (two patients) when compared to healthy age-matched control subjects (Zatti et al., 2004; Giacomello et al., 2005).

    It is also worth noting that the “abnormal Ca2+ signaling” usually reported for human FAD fibroblasts not always means an increased Ca2+ load since a reduced Ca2+ release was also observed (Peterson et al., 1988; McCoy et al., 1993). The discussion on this issue is further complicated by the fact that the large majority of the studies with AD fibroblasts were carried out in the 1980s-1990s when Alzheimer donors were not genetically characterized. Interestingly, by using aequorin, McCoy et al. (1993) showed a reduced Ca2+release in human early-onset FAD fibroblasts from a Canadian family which was recently shown to carry the PS1-A246E mutation (Huang et al., 2005).

    Furthermore, we have to consider that, in FAD fibroblasts, at variance with the rescued DKO MEFs, the PS mutant exerts its effect in the presence of the endogenous wild-type proteins, as occurring also in the majority of the cell models tested. This fact makes the comparison even more problematic.

    Given the suggested protective role exerted by a low ER Ca2+ level (Scorrano et al., 2003), we proposed that PS mutations which strongly reduce the ER Ca2+ content (such as those in PS2) should attenuate the pathology, whereas other mutations that leave the ER Ca2+ content unchanged or mildly reduced (such as those in PS1) could be unable to compensate for other defects due to the mutations themselves. Indeed, oxidative stress induces a pronounced Ca2+ overload in PS1-A246E-FAD fibroblasts compared to aged controls (Huang et al., 2005). Our hypothesis is thus consistent with the later ages of onset and milder AD phenotypes observed in patients carrying PS2 mutations with respect to those carrying PS1 ones.

    In summary, as far as the physiological role of PSs is concerned, our results are in agreement with those reached by Bezprozvanny and colleagues. However, the presence of contrasting findings with the pathological mutations shows the limitations of the simple definition “gain or loss of function” for multifaceted proteins such as PSs, especially when considering how different are the backgrounds in which their effects are evaluated.


    . Reduction of Ca2+ stores and capacitative Ca2+ entry is associated with the familial Alzheimer's disease presenilin-2 T122R mutation and anticipates the onset of dementia. Neurobiol Dis. 2005 Apr;18(3):638-48. PubMed.

    . Up-regulation of inositol 1,4,5-trisphosphate receptor type 1 is responsible for a decreased endoplasmic-reticulum Ca2+ content in presenilin double knock-out cells. Cell Calcium. 2006 Jul;40(1):41-51. PubMed.

    . Modification of endoplasmic reticulum Ca2+ stores by select oxidants produces changes reminiscent of those in cells from patients with Alzheimer disease. Free Radic Biol Med. 2005 Oct 15;39(8):979-89. PubMed.

    . Serum- and bradykinin-induced calcium transients in familial Alzheimer's fibroblasts. Neurobiol Aging. 1993 Sep-Oct;14(5):447-55. PubMed.

    . Altered response of fibroblasts from aged and Alzheimer donors to drugs that elevate cytosolic free calcium. Neurobiol Aging. 1988 May-Jun;9(3):261-6. PubMed.

    . BAX and BAK regulation of endoplasmic reticulum Ca2+: a control point for apoptosis. Science. 2003 Apr 4;300(5616):135-9. PubMed.

    . Calcium dysregulation in Alzheimer's disease: recent advances gained from genetically modified animals. Cell Calcium. 2005 Sep-Oct;38(3-4):427-37. PubMed.

    . The presenilin 2 M239I mutation associated with familial Alzheimer's disease reduces Ca2+ release from intracellular stores. Neurobiol Dis. 2004 Mar;15(2):269-78. PubMed.

    . Presenilin mutations linked to familial Alzheimer's disease reduce endoplasmic reticulum and Golgi apparatus calcium levels. Cell Calcium. 2006 Jun;39(6):539-50. PubMed.

  4. I was quite interested in the regulation of calcium within the endoplasmic reticulum, and subsequent cell death apparently related to calcium toxicity. It appears the presenilin1 and 2 permit calcium regulation, and familial Alzheimer presenilin1 and 2 are not able to perform this function, probably leading to cell dysfunction and development of familial Alzheimer disease. This certainly is a lead to follow in determining the pathophysiology of sporadic Alzheimer disease. There may be multiple causes of endoplasmic reticulum dysfunction and calcium accumulation.

    I performed aluminum neurotoxicity experiments on hippocampal rat neurons several years ago and found dantrolene and dimethylsulfoxide reduced cell death from aluminum toxicity, indicating aluminum toxicity may be mediated through release of calcium from intracellular stores and oxidative stress (1).

    There may be multiple mechanisms disrupting calcium metabolism in the endoplasmic reticulum, including metals such as aluminum and other metals potentially capable of oxidation such as copper and iron. Oxidative stress might also be implicated as well.

    I am not sure how β amyloid could effect calcium metabolism within the endoplasmic reticulum and other intracellular stores, but amyloid precursor protein could be implicated as well, since it may be capable of forming ion channels.

    Disruption of calcium metabolism and β amyloid toxicity may act synergistically in causing cellular dysfunction and Alzheimer disease.

    If intracellular structures such as endoplasmic reticulum and sarcoplasmic reticulum are effected by disturbed calcium metabolism, protein assembly in intracellular structures may result in dysfunctional proteins unable to perform intracellular processes normally with subsequent cellular death and resulting Alzheimer disease.


    . Aluminum neurotoxicity is reduced by dantrolene and dimethyl sulfoxide in cultured rat hippocampal neurons. Biol Trace Elem Res. 2002 Apr;86(1):85-9. PubMed.

  5. Presenilin Is a New Endoplasmic Reticulum Membrane Protein Essential for Calcium Leak
    A long-standing mystery in the cell biology of calcium homeostasis is the molecular nature and the physiological role of “leak-channels” in the endoplasmic reticulum (ER) membrane. Indeed, the ER is the major calcium store, and the Ca2+ filling status of the ER controls many physiological processes ranging from gene expression to apoptosis and proliferation. Furthermore, more and more papers suggest that the abnormal luminal ER calcium concentration ([Ca2+]L) and deranged calcium signaling are associated with severe human pathologies such as cancer and neurodegenerative diseases.

    Under resting conditions, steady-state [Ca2+]L is determined by the dynamic equilibrium of two components: an active Ca2+ uptake mediated by ATP-dependent Ca2+ pumps of the SERCA family and passive Ca2+ efflux via leak channels. Even though this pump-leak cycle appears to be a common property of Ca2+-storing organelles, little is known about the proteins controlling the Ca2+ leak pathway. Several mechanisms involving quite different proteins have been previously suggested to explain the basal Ca2+ leak from ER, namely: 1) reverse Ca2+ flux through the pumps (Toyoshima et al., 2002); 2) Ca2+ leak in neutral complexes with small molecules by translocon channels (Lomax et al., 2002; Van Coppenolle et al., 2004); 3) the fluxes of Ca2+ through “natural” ionophores, such as bile acids (Zimniak et al., 1991); 4) an anti-apoptotic protein Bcl-2–mediated Ca2+ leak (Bassik et al., 2004); 5) IP3R- or RYR-mediated Ca2+ leak (Oakes et al., 2005) and, more recently, 6) pannexin 1-mediated calcium leak (Vanden Abeelle et al., 2006). However, “the drawing of these mechanisms is only a fantasy map of the leak terra incognita, and discovery of the exact mechanisms of calcium leak remains a challenge to scientists working in the calcium signaling field.” (Camello et al., 2002).

    The team of Ylya Bezprozvanny, using a multidisciplinary approach, clearly demonstrates that the nine transmembrane domain ER proteins, presenilins, account for almost 80 percent of passive calcium leak from the ER. The results of their study strongly suggest that presenilins can form calcium-permeable ion channels and, moreover, that the genetic deletion of presenilins (in double knockout, DKO, mice) resulted in a sixfold reduction in the rate of calcium leak across ER membrane. Heterologous expression of presenilins in DKO mouse embryonic fibroblasts was able to rescue calcium leakage defects observed in DKO cells, which is consistent with an ion “leak-channel” function of presenilins in the ER membrane.

    Even more intriguing is the finding that presenilin mutants associated with familial Alzheimer disease (FAD) were not able to form calcium leak channels. Thus, the results of this study strongly support the hypothesis of a crucial role of calcium homeostasis in Alzheimer disease, pointing out the specific function of presenilins.


    . Phosphorylation of BCL-2 regulates ER Ca2+ homeostasis and apoptosis. EMBO J. 2004 Mar 10;23(5):1207-16. PubMed.

    . Calcium leak from intracellular stores--the enigma of calcium signalling. Cell Calcium. 2002 Nov-Dec;32(5-6):355-61. PubMed.

    . Basal and physiological Ca(2+) leak from the endoplasmic reticulum of pancreatic acinar cells. Second messenger-activated channels and translocons. J Biol Chem. 2002 Jul 19;277(29):26479-85. PubMed.

    . Structural changes in the calcium pump accompanying the dissociation of calcium. Nature. 2002 Aug 8;418(6898):605-11. PubMed.

    . Bcl-2-dependent modulation of Ca(2+) homeostasis and store-operated channels in prostate cancer cells. Cancer Cell. 2002 Mar;1(2):169-79. PubMed.

    . Functional implications of calcium permeability of the channel formed by pannexin 1. J Cell Biol. 2006 Aug 14;174(4):535-46. PubMed.

    . Ribosome-translocon complex mediates calcium leakage from endoplasmic reticulum stores. J Cell Sci. 2004 Aug 15;117(Pt 18):4135-42. PubMed.

    . Taurine-conjugated bile acids act as Ca2+ ionophores. Biochemistry. 1991 Sep 3;30(35):8598-604. PubMed.

  6. Reply to Giuliano Binetti, Cristina Fasolato, Roberta Ghidoni, Paola Pizzo, and Sandro Sorbi
    We are thankful to Giuliano Binetti and his colleagues for the high praise given our paper and for their insightful comments. We apologize for not discussing their highly relevant paper, Zatti et al., 2006 [1], which appeared while our manuscript was in the final stages of review and we did not see it prior to publication of our paper.

    Binetti and colleagues raise interesting questions about the effects of presenilin FAD mutations on ER Ca2+ content and on inositol trisphosphate receptor (InsP3R)-mediated Ca2+ release. We attempted to reconcile our results with that of Zatti at al.; however, we ran into significant difficulties in interpreting their data.

    Let us consider an example of two PS1 FAD mutants for which extensive datasets are available from several laboratories. Zatti et al. reported that expression of PS1-M146L resulted in reduced Ca2+ response to cyclopiazonic acid (CPA) + histamine (Fig. 1C), no change in response to CPA + bradykinin (BK) (Fig. 1B), no change for response to CPA + carbamylcholine (CaCh) in HEK293 stable lines (Fig. 3B), no change in ER Ca2+ levels (Fig. 4C) and reduced Golgi Ca2+ levels (Fig. 5C). They also report that CPA response was reduced in human PS1-M146L fibroblasts (Fig. 1D). The response in neurons transfected with PS1-M146L was not tested in Zatti at al. paper.

    It is hard to reach a conclusion from these results about the effect of PS1-M146L on ER Ca2+ signaling. Most data in Zatti at al. suggest that PS1-M146L has either no effect or reduces the ER Ca2+ content and the InsP3R-mediated Ca2+ release. This conclusion directly contradicts our results with transfected DKO MEF cells [2] and the extensive characterization of the effects of PS1-M146V mutant on the InsP3R-mediated Ca2+ signals in Xenopus oocytes and in hippocampal neurons by Parker and La Ferla’s laboratories [3,4].

    A similar situation exists for PS1-A246E FAD mutant. Zatti et al. reported that expression of PS1-A246E resulted in reduced Ca2+ response to CPA + histamine (Fig. 1C), reduced Ca2+ response to CPA + BK (Fig. 2B), reduced ER Ca2+ levels in HeLa cells, but unchanged ER Ca2+ levels in SH-SY5Y cells (Fig. 4C), reduced Golgi Ca2+ levels in both HeLa and SH-SY5Y (Fig. 5C), and no change in CaCh-induced response in transfected rat cortical neurons (Fig. 6D).

    Once again, most of these results seem to indicate that PS1-A246E has either no effect or reduces the ER Ca2+ content and the InsP3R-mediated Ca2+ release. This conclusion directly contradicts our results with transfected DKO MEF cells and human A246E fibroblasts [5], and results from studies of the InsP3R-mediated Ca2+ signals in hippocampal neurons from A246E transgenic mouse performed by Jochen Herms’s laboratory [6].

    In summary, we agree with Giuliano Binetti and colleagues that effects of FAD mutations in presenilins on ER Ca2+ homeostasis is a very exciting and important area of AD research. However, as it is clear from the above discussion, much additional work by many laboratories will be required to clarify the exact nature of this interesting phenomenon.


    . Presenilin mutations linked to familial Alzheimer's disease reduce endoplasmic reticulum and Golgi apparatus calcium levels. Cell Calcium. 2006 Jun;39(6):539-50. PubMed.

    . Presenilins form ER Ca2+ leak channels, a function disrupted by familial Alzheimer's disease-linked mutations. Cell. 2006 Sep 8;126(5):981-93. PubMed.

    . Alzheimer's presenilin-1 mutation potentiates inositol 1,4,5-trisphosphate-mediated calcium signaling in Xenopus oocytes. J Neurochem. 1999 Mar;72(3):1061-8. PubMed.

    . Dysregulated IP3 signaling in cortical neurons of knock-in mice expressing an Alzheimer's-linked mutation in presenilin1 results in exaggerated Ca2+ signals and altered membrane excitability. J Neurosci. 2004 Jan 14;24(2):508-13. PubMed.

    . Familial Alzheimer disease-linked mutations specifically disrupt Ca2+ leak function of presenilin 1. J Clin Invest. 2007 May;117(5):1230-9. Epub 2007 Apr 12 PubMed.

    . Mutant presenilins disturb neuronal calcium homeostasis in the brain of transgenic mice, decreasing the threshold for excitotoxicity and facilitating long-term potentiation. J Biol Chem. 2001 Apr 13;276(15):11539-44. PubMed.

  7. The work by Bezprozvanny and colleagues undoubtedly adds considerably new information about the physiological function of presenilins as well as on their possible roles in AD pathogenesis at the molecular level. These data also add knowledge on the relationship among ER stress, presenilins, Aβ peptides, and derangement of calcium homeostasis in AD.

    In my opinion, the research by Bezprozvanny and colleagues emphasizes the importance of the fundamental role of free calcium modifications in cells undergoing biochemical changes underlying AD. While not questioning the key role of Aβ peptides in this disease, the data add another possible dimension to the key role performed by calcium in cellular stress and death following the biochemical modifications characterizing AD. Hence, some presenilin mutations affecting γ-secretase activity can impair cell viability by increasing Aβ peptide production or by shifting the latter towards the more amyloidogenic Aβ42, resulting in Aβ oligomerization and cell membrane(s) permeabilization. Other mutations that do not affect γ-secretase activity can disrupt calcium leakage from the ER, resulting in increased calcium stores in the ER and subsequent ER stress. It cannot be excluded that, for some specific PS mutations, the two effects may act synergistically with more severe cellular stress.

Comments on Primary Papers for this Article

  1. The puzzle of how presenilin (PS) mutations cause familial AD (FAD) is of great importance because, as many believe, it may relate to the origin of sporadic AD.

    Based on my critical analysis of the research data in comparison with AD features, I deduced a hypothesis for the mechanism of sporadic AD in 1998 [1]. This hypothesis, together with the report of the PS molecular structure [2], allowed me to predict that “presenilins most likely act as calcium channels in vivo and that their gene mutations may cause the disease by diminishing the Ca2+ channeling function” [3]. I also predicted that “Functional reconstitution and electrophysiological studies should directly reveal whether or not presenilins in artificial membranes could act as Ca2+ channels, and if so, whether the mutations would diminish the channeling function.” [2] It is thus encouraging to see, 8 years later, the elegant work by Tu et al. [3] showing that PS function as “calcium leak channels.” It has also shown that some FAD-linked mutations reduce the channel’s function [3].

    The comments made by various investigators in the Alzforum discussions (see above) make me hopeful that the important contributions of these findings to the understanding of FAD will be appreciated by the AD research field. However, given the field’s current enthusiasm for the “PS as γ-secretase” model, together with the extreme complexity of the Ca2+ signaling systems, this may take some time.

    I only have one minor question to the authors. Their report has shown that the FAD-linked PS mutants give rise to “lower cytosolic Ca2+ levels” at the steady-state compared to the wild-type PS [3, Fig. 7], a finding opposite to the premise of “calcium hypothesis” [4]. Why do they conclude that the findings “provide support for the ‘Ca2+ hypothesis of AD’”?

    View all comments by Ming Chen


Paper Citations

  1. . Calcium dysregulation in Alzheimer's disease: recent advances gained from genetically modified animals. Cell Calcium. 2005 Sep-Oct;38(3-4):427-37. PubMed.
  2. . Up-regulation of inositol 1,4,5-trisphosphate receptor type 1 is responsible for a decreased endoplasmic-reticulum Ca2+ content in presenilin double knock-out cells. Cell Calcium. 2006 Jul;40(1):41-51. PubMed.
  3. . Presenilins form ER Ca2+ leak channels, a function disrupted by familial Alzheimer's disease-linked mutations. Cell. 2006 Sep 8;126(5):981-93. PubMed.

Further Reading


  1. . Presenilins form ER Ca2+ leak channels, a function disrupted by familial Alzheimer's disease-linked mutations. Cell. 2006 Sep 8;126(5):981-93. PubMed.


  1. Aβ Assault on Neurons Targets ER, Calcium

Primary Papers

  1. . Presenilins form ER Ca2+ leak channels, a function disrupted by familial Alzheimer's disease-linked mutations. Cell. 2006 Sep 8;126(5):981-93. PubMed.