10 July 2012. Presenilins and calcium seem to share a close connection. Best known as constituents of the γ-secretase complex, the transmembrane proteins also function independently to regulate intracellular calcium stores. Now, in the July 9 Journal of Cell Biology, researchers led by Wim Annaert at KU Leuven, Belgium, report that presenilin influences calcium levels in lysosomes, the garbage disposers of the cell. Defects there may explain the poor functioning of the waste disposal system in presenilin knockouts, the authors propose. These new findings contradict a previous report in the field, from researchers led by Ralph Nixon at the Nathan Kline Institute in Orangeburg, New York, that autophagic problems in presenilin knockouts stemmed instead from a lack of acidity inside lysosomes. Annaert and colleagues challenge this view, failing to replicate key findings from Nixon’s paper. In the June 20 Journal of Neuroscience, researchers led by Sangram Sisodia at the University of Chicago, Illinois, also report they cannot reproduce Nixon’s results. Confusing the issue, however, is the fact that all three groups use somewhat different reagents, methodologies, and cell lines.
In contrast to their novel role in lysosomal calcium, presenilin mutants have long been known to heighten calcium release from the endoplasmic reticulum (ER). Most recently, researchers led by Grace Stutzmann at Rosalind Franklin University in North Chicago, Illinois, examined the consequences of altered calcium signaling in young, presymptomatic AD mice. In the June 13 Journal of Neuroscience, the authors report that, although synaptic responses in these animals appear still normal, the system is actually prone to a dampened excitability that can be uncovered by manipulating calcium stores. “There are profound signaling abnormalities going on under the radar in Alzheimer’s disease that far precede any overt cognitive deficits,” Stutzmann told Alzforum. In addition, she finds it exciting how the new lysosomal calcium findings dovetail with the ER work. “It is internally consistent that presenilins regulate calcium across a broad array of platforms,” she said. This, she added, may help explain what presenilin is doing in organelle membranes, when most of γ-secretase’s substrates are found in the plasma membrane. “Perhaps presenilin’s primary role as a holoprotein is to regulate calcium within organelles,” she speculated.
Cells normally dispose of unneeded components by sequestering them in autophagosomes, which then fuse with lysosomes for disposal. This system malfunctions in AD, with lysosomes accumulating in neurons (see, e.g., ARF related news story). In particular, cells lacking presenilin show this autophagic block. Using blastocyst-derived cells from presenilin 1 knockout mice, Nixon had found that presenilin was needed in the ER to help glycosylate a subunit of the lysosomal proton pump. Without glycosylation, the v0a1 subunit remains stuck in the ER and the proton pump malfunctions, so lysosomes lose their acidity, he reported (see ARF related news story on Lee et al., 2010).
These data took Annaert by surprise, because in his own studies he saw no changes in lysosomal pH in presenilin 1 knockout cells, he told Alzforum. First author Katrijn Coen attempted to replicate Nixon’s findings, using mostly mouse embryonic fibroblasts lacking both presenilin 1 and 2 or primary hippocampal neurons from presenilin 1 knockouts. Coen and colleagues measured lysosomal pH by three different methods, but found no difference between knockouts and control cells. Knocking down v0a1 levels did not change lysosomal acidity, they report. They also found that v0a1 is normally glycosylated and trafficked in the double knockout cells. Finally, the authors mutated the glycosylation site of the v0a1 ortholog in flies, and showed that the unglycosylated protein fully rescued lysosomal defects in photoreceptor cells lacking endogenous v0a1. This shows that the proton pump does not need glycosylation to function, the authors conclude.
What, then, might explain the lysosomal defects in these cells? Since lysosomes need both acidity and calcium, the authors measured calcium, finding lower levels in the lysosomes of the double knockouts compared to controls. Lysosomes in the knockouts also released less calcium when stimulated. Notably, lysosomal fusion requires calcium release (see Saftig and Klumperman, 2009). Annaert’s group previously reported that lysosomes struggled to fuse properly with autophagosomes in presenilin mutant cell lines (see Esselens et al., 2004). In ongoing work, Annaert is investigating the molecular basis of the lysosomal calcium deficit.
In their new paper, Sisodia and colleagues also describe unsuccessful attempts to replicate Nixon’s study, using mostly mouse embryonic stem cells lacking either both presenilins or presenilin 1 alone. First author Xulun Zhang found protein turnover to be normal in these cells, suggesting no problems with autophagy. Using a dye that enters all intracellular organelles, the authors saw no significant difference in average vesicular pH between wild-type and presenilin knockout cells. A labeled, overexpressed form of v0a1 was normally glycosylated in all cells, and endogenous v0a1 was processed normally in neurons from double knockout mice, the authors report. Looking for an alternative explanation for lysosomal abnormalities, the authors examined gene expression. They found that a suite of genes involved in lysosome production was more highly expressed in the knockouts, suggesting that more lysosomes are made in these cells. Sisodia told Alzforum he has no plans to pursue this research further.
Speaking with Alzforum, Nixon questioned the conclusions, noting that the other groups used different methods to measure pH than his. In particular, Nixon used the dye LysoSensor yellow/blue conjugated to dextran, which targets lysosomes exclusively and gives a more sensitive and specific measurement of pH, he said. The three groups also used different cell lines, although Nixon said he has repeated his work in the lines used by the other groups and still sees acidity differences, and Annaert reports using Nixon’s lines and finding no pH changes. Regarding v0a1, Nixon pointed out that the other groups primarily looked at its trafficking in overexpression systems, which could mask the behavior of the endogenous protein. In other words, with high enough levels of a protein, enough of it may reach the target to do the job. However, Nixon agreed that lysosomes have a calcium defect. He believes this is secondary to the pH change, as his unpublished data show that correcting lysosomal pH normalizes calcium, but not vice versa. Changes in pH are known to dysregulate calcium, he added.
In a commentary accompanying the JCB paper, Ilya Bezprozvanny at the University of Texas Southwestern Medical Center, Dallas, highlighted the varying methodologies as an explanation for the different outcomes. “One can expect that most of these technical issues will be sorted out and the sources of discrepancy identified,” he wrote. Nonetheless, the new data constitute a serious challenge to the lysosomal acidification hypothesis, he concluded.
Kim Green at the University of California, Irvine, noted that data from his own lab largely agree with Annaert’s. He previously reported no defect in lysosomal acidity (see ARF related news story), and has unpublished data showing the same calcium defect that Annaert shows, he told Alzforum. Green traced this effect to a role for presenilin in regulating calcium channels in the lysosome. Despite the conflicting data from different groups, “The big picture is that we are all showing that presenilin is important for aspects of autophagy,” Green said. Importantly, this activity is independent of presenilin’s role in γ-secretase. What is missing are more in-vivo studies, Green suggested, asking, “What role does presenilin actually play in the adult brain?”
One widely known consequence of mutant presenilin 1 is to flood neurons and synaptic compartments with calcium released from ER stores through ryanodine receptors (see, e.g., ARF related news story; ARF related news story; Bezprozvanny, 2009). In her paper, Stutzmann focused on the early synaptic consequences of this disrupted calcium signaling in 3xTg-AD mice, which carry mutant presenilin. At six to eight weeks of age, these mice have no measurable cognitive deficits or brain pathology, and synaptic signaling in hippocampal slices appears normal. However, by blocking ryanodine receptors, first author Shreaya Chakroborty uncovered markedly altered pre- and postsynaptic mechanisms that together push the system into a depressed state where neurons are less likely to fire.
On the presynaptic side, high levels of calcium increased spontaneous release of vesicles, which depleted stores and led to a weaker response to action potentials. On the postsynaptic side, the authors saw heightened activation of a calcium-activated potassium channel. This hyperpolarized the membrane and made the neuron less likely to fire in response to stimulation. Both these changes favor long-term depression and decrease the capacity for long-term potentiation, which is essential for storing new memories. Why, then, does the synaptic network behave normally until challenged by blocking ryanodine receptors? There must be a compensatory mechanism, Stutzmann hypothesizes. Since the ryanodine receptor block lowers calcium levels, Stutzmann speculated that the compensatory mechanism is calcium-regulated, and she is currently investigating what it might be. In ongoing work, she is also examining how altered calcium signaling affects the structure of synapses.
As AD advances, the compensatory mechanism may become overwhelmed by increasing metabolic stress or the growing load of β amyloid and tau pathology, and cognitive deficits ensue, Stutzmann speculated. Her data indicate that early synaptic changes are reversible, and she believes these synaptic alterations will be a key target for therapeutic intervention. Stutzmann also points out that although only a few people with AD have presenilin mutations, sporadic AD patients have been reported to lose presenilin expression in critical brain regions (see Davidsson et al., 2001). In addition, calcium can become dysregulated in sporadic AD in response to other insults, suggesting that this mechanism might be widely applicable, she noted.—Madolyn Bowman Rogers.
Coen K, Flannagan RS, Baron S, Carraro-Lacroix LR, Wang D, Vermeire W, Michiels C, Munck S, Baert V, Sugita S, Wuytack F, Hiesinger PR, Grinstein S, Annaert W. Lysosomal calcium homeostasis defects, not proton pumps defects, cause endo-lysosomal dysfunction in PSEN-deficient cells. J Cell Biol. 2012 Jul 9;198(1):23-35. Abstract
Zhang X, Garbett K, Veeraraghavalu K, Wilburn B, Gilmore R, Mirnics K, Sisodia SS. A role for presenilins in autophagy revisited: normal acidification of lysosomes in cells lacking PSEN1 and PSEN2. J Neurosci. 2012 Jun 20;32(25):8633-48. Abstract
Bezprozvanny I. Presenilins: a novel link between intracellular calcium signaling and lysosomal function? J Cell Biol. 2012 Jul 9;198(1):7-10. Abstract
Chakroborty S, Kim J, Schneider C, Jacobson C, Molgó J, Stutzmann GE. Early presynaptic and postsynaptic calcium signaling abnormalities mask underlying synaptic depression in presymptomatic Alzheimer’s disease mice. J Neurosci. 2012 Jun 13;32(24):8341-53. Abstract