Ca2+-dependent Activator Proteins of Secretion Promote Vesicular Monoamine Uptake*

Ca2+-dependent activator proteins of secretion (CAPS) 1 and 2 are essential regulators of synaptic vesicle and large dense core vesicle priming in mammalian neurons and neuroendocrine cells. CAPS1 appears to have an additional and as yet unexplained function in vesicular catecholamine uptake or storage as CAPS1-deficient chromaffin cells exhibit strongly reduced vesicular catecholamine levels. Here we describe a role of CAPS proteins in vesicular monoamine uptake. Both CAPS1 and CAPS2 promote monoamine uptake and storage mediated by the vesicular monoamine transporters VMAT1 and VMAT2. Monoamine uptake of vesicular preparations from embryonic brains of CAPS1 deletion mutants is decreased as compared with corresponding preparations from wild type littermates, and anti-CAPS1 or anti-CAPS2 antibodies inhibit monoamine sequestration by synaptic vesicles from adult mouse brain. In addition, overexpression of CAPS1 or CAPS2 enhances vesicular monoamine uptake in Chinese hamster ovary cells that stably express VMAT1 or VMAT2. CAPS function has been linked to the heterotrimeric GTPase Go, which modulates vesicular monoamine uptake. We found that the expression of CAPS1 is decreased in brain membrane preparations from mice lacking Go2α, which may explain the reduced monoamine uptake by Go2α-deficient synaptic vesicles. Accordingly, anti-CAPS1 antibodies do not further reduce monoamine uptake by Go2α-deficient synaptic vesicles, whereas antibodies directed against CAPS2, whose expression is not altered in Go2α-deficient brain, still reduce monoamine uptake into Go2α-deficient vesicles. We conclude that CAPS proteins are involved in optimizing vesicular monoamine uptake and storage mediated by VMAT1 and VMAT2.

Secretion of neurotransmitters is mediated by Ca 2ϩ -dependent fusion of secretory organelles with the plasma membrane. CAPS 3 proteins were initially identified as cytosolic regulators of Ca 2ϩ -dependent secretion in PC12 cells (1). Mammals express two isoforms of CAPS, CAPS1 and CAPS2, which have similar functions but differ in their spatiotemporal expression pattern (1)(2)(3)(4)(5)(6)(7). Subsequent studies on mammalian CAPS isoforms showed that CAPS proteins play an essential role in the priming step of synaptic vesicles and large dense core vesicles (1,2,(7)(8)(9)(10). In addition, CAPS1 appears to have a second and as yet unexplained function in vesicular catecholamine uptake or storage (10).
Transport of monoamines into secretory vesicles is mediated by VMATs. In mammals, two isoforms, termed VMAT1 and VMAT2, have been identified (11,12). VMAT1 expression is restricted to neuroendocrine cells, and VMAT2 is the only isoform expressed in neurons. Monoamine uptake via VMAT1 and VMAT2 is regulated by vesicle-associated heterotrimeric G-proteins, which is illustrated by the fact that in neurons and neuroendocrine cells, the ␣-subunit of G o2 inhibits monoamine uptake (13)(14)(15). The role of CAPS proteins in VMAT-mediated vesicular catecholamine uptake and storage is currently discussed controversially because corresponding studies undertaken so far have yielded contradictory results. Experiments on chromaffin cells of CAPS1-deficient mice indicated a role of CAPS1 in catecholamine loading as CAPS1-deficient chromaffin cells exhibit strongly reduced catecholamine levels in their secretory granules. Indeed, many fusing chromaffin granules of CAPS1-deficient chromaffin cells contain no catecholamines at all (10). Chromaffin cells of mice lacking both CAPS1 and CAPS2 exhibit an additional reduction in the refilling and maintenance of a pool of rapidly releasable chromaffin granules but no further reduction in vesicular catecholamine sequestration as compared with chromaffin cells of mice lacking only CAPS1 (16). In contrast to these findings, knock-down of CAPS1 expression in PC12 cells was found to cause increased endogenous catecholamine levels and enhanced accumulation of norepinephrine (17). Based on these indirect readouts, a role of CAPS proteins in vesicular monoamine uptake was categorically excluded (17).
In the study reported here, we directly examined the role of CAPS proteins in vesicular monoamine uptake and storage employing a set of complementary experimental approaches. Our results support the notion that CAPS proteins promote vesicular monoamine uptake and storage in brain cells and engineered cell lines.

EXPERIMENTAL PROCEDURES
Antibodies-Rabbit polyclonal antibodies directed against CAPS1 and CAPS2 were described previously (7). Rabbit polyclonal antibodies directed against actin were purchased from Sigma. Horseradish peroxidase-labeled goat anti-rabbit antibodies were obtained from Vector Laboratories (Burlingame, CA).
Mouse Lines-CAPS1-deficient mice were generated, bred, and genotyped as described previously (8,10). G o2 ␣ splice variant-specific deletion mutants were kindly provided by L. Birnbaumer (Research Triangle Park, NC) and bred and genotyped as described previously (19,20). For all experiments, mutant and wild type littermates obtained by interbreeding of heterozygous parents were used.
Serotonin Uptake by Vesicle Preparations from Embryonic Mouse Brain-Embryos of heterozygous CAPS1 deletion mutants were removed at E18. Brains were processed separately. After homogenization in 1 ml of water containing 10 mM HEPES and protease inhibitors, homogenates were centrifuged for 10 min at 1,000 ϫ g, and the postnuclear supernatant was spun down for 30 min at 360,000 ϫ g. The resulting pellets were resuspended in potassium-glutamate-ATP buffer (150 mM potassium glutamate; 20 mM 1,4-piperazinediethanesulfonic acid; 4 mM EGTA; 2.9 mM MgCl 2 (is equivalent to 1 mM free Mg 2ϩ ); and 2 mM ATP, adjusted to pH 7.0 with KOH). After genotyping, two samples of identical genotype were pooled for further analysis. Uptake assays were performed as described before (21)  Preincubation of Synaptic Vesicles with Anti-CAPS Antibodies and Subsequent Serotonin Uptake Assays-Anti-CAPS1 and anti-CAPS2 antibodies and the corresponding preimmune sera were dialyzed against potassium-glutamate-ATP buffer (150 mM potassium glutamate; 20 mM 1,4-piperazinediethane-sulfonic acid; 4 mM EGTA; 2.9 mM MgCl 2 (is equivalent to 1 mM free Mg 2ϩ ); and 2 mM ATP, adjusted to pH 7.0 with KOH) overnight at 4°C. Synaptic vesicles were prepared from mouse brains as described (22,23) and incubated with a 1:10 dilution of the indicated antibodies in potassium-glutamate-ATP buffer for 30 min on ice. After that, uptake assays were performed as described before (21). Final dilution of antibodies during uptake assays was 1:20.
Immunoblot Analysis-Subcellular fractions were prepared from mouse brains as described (22,23). After determination of protein content, samples were analyzed by SDS-PAGE and Western blotting using the ECL detection system (GE Healthcare). ECL-processed films were scanned, and protein bands were densitometrically quantified using the Labimage 1D program (KAPELAN, Halle, Germany). It was ensured that signals were in the linear range of the ECL detection system. Quantification of proteins comparing wild type and knock-out samples were performed from the same gel using actin as internal standard.
CHOVMAT1 and CHOVMAT2 Cell Lines-VMAT1-expressing CHO cells were cultured at 37°C, 5% CO 2 in Dulbecco's modified Eagle's medium/Ham's F-12 medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. CHOVMAT2 cells (21) were cultured in the same medium supplemented with 400 g/ml Geneticin. Cells were transfected with CAPS-DNA using the Lipofectamine TM transfection reagent (Invitrogen). 2 days after transfection, cells were analyzed by fluorescence microscopy and serotonin uptake assays. After selective permeabilization of plasma membranes by SLO, serotonin uptake into intracellular compartments by VMAT1 or VMAT2 was assayed as described (21).
Fluorescence Microscopy-Cells were grown on glass coverslips, washed twice with phosphate-buffered saline, and fixed in 4% formaline in 0.1 M phosphate buffer, pH 7.4, for 45 min at room temperature. After three rinses with phosphate-buffered saline and one rinse with water, coverslips were mounted on glass slides for fluorescence microscopic analysis.
Experimental Design-All experiments presented were repeated at least two times. Individual uptake experiments were performed in triplicate. Values are presented as mean Ϯ S.D.

Absence of CAPS Proteins Decreases Vesicular Serotonin
Uptake-Electrophysiological evidence indicates that the deletion of CAPS1 reduces vesicular monoamine uptake or storage in chromaffin granules (10). Efforts to obtain direct biochemical evidence for an effect of the CAPS1 deletion on vesicular monoamine uptake has been confounded by the fact that CAPS-deficient mice die at birth and thus yield only very small amounts of chromaffin granules that are not sufficient for biochemical uptake assays. To circumvent this problem, we tested first whether monoamine uptake assays into synaptic vesicles prepared from E18 embryonic mouse brains are feasible.
Using E18 embryonic synaptic vesicles, we found that serotonin uptake was 3-fold higher without additives as compared with uptake measured in the presence of 6 M of the VMAT blocker reserpine, indicating a specific uptake via VMAT2 (Fig.  1A, left panel). This allowed us to analyze monoamine uptake by synaptic vesicles of CAPS1 deletion mutants prepared at E18. Due to the low uptake seen in samples from embryonic brains, serotonin uptake assays using vesicular preparations from embryonic brains of wild type, CAPS1 ϩ/Ϫ , and CAPS1 Ϫ/Ϫ mice were performed with 400 nM serotonin instead of 40 nM, resulting in an increased uptake in general. Serotonin uptake was significantly decreased in vesicles from CAPS1 Ϫ/Ϫ mice as compared with wild type littermates (Fig. 1A, right panel). Unspecific accumulation of serotonin in the presence of reserpine was comparable for all genotypes and subtracted (see the legend for Fig. 1A). Vesicles from heterozygous mice did not exhibit a significant reduction of serotonin uptake. The reduced serotonin uptake of CAPS1-deficient vesicles is not due to reduced VMAT2 expression, which was not altered in CAPS1 knock-out vesicles, indicating an impact of the deletion mutation on VMAT function rather than expression (Fig. 1B). Likewise, expression of the vesicular proton pump is normal in CAPS1 knock-out vesicles (Fig. 1B), which indicates that vesicles from CAPS1 knock-out retain the protein machinery to generate a normal electrochemical ⌬H ϩ gradient. The above data indicate that genetic deletion of CAPS1 specifically perturbs vesicular monoamine uptake activity of embryonic synaptic vesicles.
Encouraged by these data, we tested next whether a reduction of uptake can also be seen when CAPS protein function in synaptic vesicles from adult brain is perturbed by specific antibodies. Synaptic vesicles were incubated with anti-CAPS1 or anti-CAPS2 antibodies and the respective preimmune sera prior to the uptake procedure. Although preincubation with preimmune sera showed no effect, incubation with both anti-CAPS antibodies reduced serotonin uptake by 40% (Fig. 1C). To demonstrate the specificity of the antibodies used in the perturbation assay, vesicular preparations from CAPS1 Ϫ/Ϫ and the corresponding wild type embryonic brains were incubated with the anti-CAPS1 antibody. As expected, a decrease in the uptake by the anti-CAPS1 antibody was only observed in the wild type but not in the CAPS1 Ϫ/Ϫ preparation (Fig. 1D). Uptake into CAPS1 Ϫ/Ϫ vesicles was reduced as compared with wild type vesicles, confirming the effect of the CAPS1 mutation seen before. These results indicate that both CAPS isoforms influence monoamine uptake and/or storage in a similar manner and that the reduced uptake activity seen in vesicles from CAPS1-deficient brain is most likely a direct effect of the CAPS1 loss and not due to a secondary developmental or homeostatic effect.
Expression of CAPS1 Is Decreased in Brains of G o2 ␣ Ϫ/Ϫ Mice-Both CAPS1 and the ␣-subunit of heterotrimeric G-protein G o2 are involved in monoamine uptake and storage (10,24), and an indirect functional link between G o proteins and the CAPS ortholog Unc-31 in Caenorhabditis elegans was reported previously (25). Thus, we analyzed the expression of CAPS1 and CAPS2 in brain fractions of wild type and G o2 ␣ Ϫ/Ϫ mice. CAPS1 levels were significantly reduced in homogenates and synaptosomal fractions of G o2 ␣ Ϫ/Ϫ deletion mutants as compared with wild type littermates ( Fig. 2A), whereas CAPS2 levels were similar in brains from both genotypes (Fig. 2B).
Subsequently, the effects of anti-CAPS1 and anti-CAPS2 antibodies on serotonin uptake of synaptic vesicles from G o2 ␣ Ϫ/Ϫ mice were determined and compared with the effect induced in vesicles of wild type mice. Specificity of the antibodies was demonstrated previously (7, 10) (Fig. 1D). In accordance with the lower expression levels of CAPS1 in G o2 ␣ Ϫ/Ϫ mutants, there was no effect of anti-CAPS1 antibodies on vesicular serotonin uptake into G o2 ␣ Ϫ/Ϫ vesicles (Fig. 2C, left panel). In contrast, the effect of anti-CAPS2 antibodies on serotonin uptake FIGURE 1. Influence of CAPS1 and CAPS2 on serotonin uptake by synaptic vesicles. A, left panel, serotonin uptake by a vesicular preparation from embryonic mouse brains (E18). Uptake in the absence of reserpine is 3-fold higher than in the presence of reserpine, indicating a specific accumulation of serotonin via VMAT2. Uptake assays were performed using 40 nM [ 3 H]serotonin. Right panel, serotonin uptake was assayed using vesicular preparations from embryonic brains of wild type (wt), CAPS1 ϩ/Ϫ , and CAPS1 Ϫ/Ϫ mice in the presence of 400 nM serotonin (40 nM [ 3 H]serotonin plus 360 nM serotonin), accounting for higher accumulation of serotonin. Serotonin uptake was significantly decreased in CAPS1 Ϫ/Ϫ samples. The reduction in uptake by preparations from CAPS1 ϩ/Ϫ mice was not statistically significant. Unspecific uptake determined in the presence of 6 M reserpine was similar in wild type (2.80 pmol/mg of protein), CAPS1 ϩ/Ϫ (2.63 pmol/mg of protein), and CAPS1 Ϫ/Ϫ samples (2.73 pmol/mg of protein) and was subtracted from total uptake values. Each experiment was repeated at least twice. The values given represent the means of three samples Ϯ S.D. B, vesicular preparations from brains of embryonic wild type and CAPS1 Ϫ/Ϫ littermates were analyzed by Western blotting using anti-VMAT2 and anti-actin antibodies. 5, 10, and 15 g of protein of the respective sample were loaded. Quantification is given as relative optical density (OD) and was assessed using actin as an internal control. Values represent the mean of four animals per genotype Ϯ S.D. C, mouse synaptic vesicles were incubated with anti-CAPS1 antibodies (left panel) or anti-CAPS2 antibodies (right panel) and the respective preimmune sera (PIS) for 30 min on ice. The serotonin uptake assay was then performed as described above. Incubation with both anti-CAPS1 and anti-CAPS2 antibodies decreased vesicular serotonin uptake significantly. D, serotonin uptake was assayed using vesicular preparations from embryonic brains of wild type and CAPS1 Ϫ/Ϫ mice using 400 nM serotonin as given in B. Serotonin uptake was significantly decreased in CAPS1 Ϫ/Ϫ samples but was not further decreased by application of the anti-CAPS1 antibody, which, however, decreased uptake into vesicular preparations from wild type mice. The values given are corrected for unspecific accumulation in the presence of reserpine and represent the means of three samples Ϯ S.D.
by synaptic vesicles was even more pronounced in G o2 ␣ Ϫ/Ϫ vesicles than in wild type samples (Fig. 2C, right panel). The differential effects of anti-CAPS1 and anti-CAPS2 antibodies on synaptic vesicles from G o2 ␣ Ϫ/Ϫ mice further confirms the specificity of the antibodies used. Generally, serotonin uptake was lower in synaptic vesicles from G o2 ␣ Ϫ/Ϫ animals as compared with vesicles of wild type animals, as observed previously, although VMAT expression is increased in G o2 ␣ Ϫ/Ϫ mice (15). Probably, the reduced expression of CAPS1 accounts for the reduced monoamine uptake in vesicles of G o2 ␣ Ϫ/Ϫ mice.
Transfection of VMAT-expressing CHO Cells with CAPS1 and CAPS2 Increases Vesicular Serotonin Uptake-The data reported above show that both genetic deletion of CAPS and perturbation of CAPS function by specific antibodies reduce monoamine uptake into synaptic vesicles. To see whether increased CAPS function leads to the opposite effect, i.e. increased vesicular monoamine uptake, CHO cell lines that permanently express either VMAT1 or VMAT2 were transiently transfected with CAPS-encoding cDNAs and assayed for VMAT activity. These VMAT-expressing cell lines are devoid of plasma membrane monoamine transporters and monoamine-synthesizing or -degrading enzymes, and VMATs were shown to reside on endosomal compartments when expressed in CHO cells (24). After permeabilization of the plasma membrane, this model system is well suited for studies on vesicular monoamine uptake and its regulation (21).
VMAT1-and VMAT2-expressing CHO cells were transiently transfected with CAPS1 or CAPS2 cDNA. The respective CAPS sequences were placed in front of an IRES-eGFP sequence for fluorescence analysis of transfection efficiency. An IRES-eGFP-vector cDNA was used for control transfections. Kinetics of serotonin transport by VMAT1 and VMAT2 were analyzed following SLO permeabilization in CAPS1-(  Table 1). These results confirm the notion that CAPS1 and CAPS2 promote vesicular monoamine uptake and/or storage.

DISCUSSION
Our results show that both CAPS isoforms promote vesicular monoamine uptake, irrespective of the VMAT isoform expressed. This function is fulfilled in both neuronal and nonneuronal cells and detectable in vesicle preparations from embryonic and adult mouse brain. The results are consistent with prior observations indicating a requirement of CAPS for filling of chromaffin granules (10,16). They also reveal a functional link between G o2 ␣, which was shown previously to regulate monoamine storage, and CAPS1 as CAPS1 expression is decreased in brains of G o2 ␣ Ϫ/Ϫ mice.
Role of CAPS1 and CAPS2 in Vesicular Monoamine Loading-Both CAPS isoforms were shown to regulate large dense core vesicle and synaptic vesicle secretion in mammalian cells (8,9). Previous investigations of an additional role of CAPS proteins in vesicular monoamine uptake and storage yielded contradictory results (10,17). Our present results are in line  1:20) and the respective preimmune sera (PIS). Afterward, serotonin uptake was assayed as described above. Left panel, in contrast to the results obtained with wild type samples, incubation with anti-CAPS1 antibodies did not change vesicular serotonin accumulation in G o2 ␣ Ϫ/Ϫ samples, probably due to diminished expression of CAPS1 in G o2 ␣ Ϫ/Ϫ mutants. Right panel, like in wild type animals, anti-CAPS2 antibodies decreased vesicular serotonin uptake in G o2 ␣ Ϫ/Ϫ mice. Although CAPS2-expression is unchanged in G o2 ␣ Ϫ/Ϫ as compared with wild type mice, the effect of anti-CAPS2 antibodies on vesicular serotonin uptake is even more pronounced in G o2 ␣ Ϫ/Ϫ samples than in wild type samples. with one of these earlier studies, which demonstrated a facilitating role of CAPS in large dense core vesicle loading in mouse chromaffin cells (10,16). In adrenal chromaffin cells from embryonic CAPS1 Ϫ/Ϫ mutants, up to 70% empty (i.e. catecholamine-free) chromaffin granules were shown to undergo exocytosis. Additionally, cytosolic levels of catecholamine metabolites were found to be strongly increased in adrenal glands of CAPS1 Ϫ/Ϫ mice, indicating a shift from vesicular to cytosolic catecholamine pools. These observa-tions were interpreted in the context of a dysfunction of vesicular catecholamine loading or storage in the absence of CAPS1 (10). A recent study on chromaffin cells of deletion mutant mice lacking both CAPS isoforms also suggested, besides a strong deficit in exocytosis, fusion of empty vesicles (16). However, knock-down of CAPS1 in PC12 cells led to a higher accumulation of exogenous norepinephrine, and the endogenous level of dopamine was increased (17). Although catecholamine filling of secretory vesicles was not tested in this study, e.g. by directly correlating high resolution membrane fusion measurements with amperometric recordings of released catecholamines as was done in the conflicting study (10,16), the phenomenon of increased norepinephrine accumulation and endogenous dopamine levels after CAPS1 knock-down was attributed solely to a reduction of (constitutive) secretion in CAPS1 knock-down cells. An additional direct role of CAPS proteins in vesicular monoamine uptake was excluded categorically (17). However, the endogenous amounts of dopamine and norepinephrine do not correlate in the various knock-down clones (i.e. high levels of dopamine were found associated with lower levels of norepinephrine and vice versa), whereas comparable amounts were seen in the control cells (17). Thus, it is possible that knock-down of CAPS1 also affects metabolism of dopamine or its turnover to norepinephrine. A change in the metabolic balance may also explain the selectively reduced amounts of dopamine as compared with the almost unchanged amounts of norepinephrine in the CAPS1 Ϫ/Ϫ chromaffin cells (10). Essentially, without direct information on vesicular uptake characteristics, changes in the metabolic balance of dopamine and its metabolites can only be used as a very indirect readout for vesicular uptake, especially in a cell culture system such as PC12 cells.
Our present study demonstrates that both CAPS isoforms promote vesicular monoamine uptake and storage in both VMAT1-expressing and VMAT2-expressing cells (Figs.  1-3), thus supporting the data obtained in chromaffin cells of CAPS deletion mutants (10,16). Overexpression of CAPS isoforms in CHOVMAT cells leads to effects, i.e. increased monoamine uptake (Fig. 3), that are opposite to those seen upon CAPS deficiency in deletion mutants (Fig. 1A) or after  (Table 1). loss of function due to antibody-mediated perturbation (Fig.  1C). In view of these findings and the fact that neither VMAT2 nor vesicular proton pump levels are changed in CAPS1-deficient vesicles (Fig. 1B), we conclude that CAPS proteins positively regulate monoamine uptake and that the reduced vesicular monoamine uptake in CAPS-deficient cells is a direct consequence of CAPS loss rather than a secondary, indirect effect of the mutation. CAPS proteins may stabilize synaptic vesicles, allowing them to store higher amounts of monoamines by shifting the electrical proton gradient (⌬H ϩ ) across the vesicle membrane to higher values. An increase of the ⌬pH component of ⌬H ϩ would allow more monoamines to be stored inside vesicles and would counteract nonspecific leakage (24,26). However, the protein levels of the proton pump is not affected in CAPS1 Ϫ/Ϫ vesicle preparations, and the loading of glutamatergic synaptic vesicles is only weakly affected by deletion of CAPS1 and CAPS2 (8), which indicates that the electrochemical gradient is functional in CAPS-deficient vesicles and that CAPS proteins preferentially regulate storage of monoamines in VMAT2-containing vesicles. Functional Relationship between CAPS1 and G o2 ␣-The ␣-subunit of G o2 was shown to regulate vesicular monoamine storage. Vesicular monoamine uptake is decreased in G o2 ␣ Ϫ/Ϫ mice (15) (Fig. 2C). Thus, both CAPS proteins and G o2 are involved in vesicular monoamine loading. In addition, loss-offunction mutation of the heterotrimeric G o protein of C. elegans partly suppresses phenotypic deficits of unc-31 (CAPS) mutants, which further indicates a functional interplay between Unc-31/CAPS and G o2 ␣ (25).
In view of these facts, a second objective of the present study was to investigate the functional relationship between CAPS and G o2 ␣. To this end, the expression levels of both CAPS isoforms were determined in brains of wild type and G o2 ␣ Ϫ/Ϫ mice (Fig. 2, A and B). Only the expression of CAPS1 was decreased in G o2 ␣ Ϫ/Ϫ brains as compared with wild type samples, and application of anti-CAPS1 antibodies did not further interfere with vesicular monoamine uptake into synaptic vesicles of G o2 ␣ Ϫ/Ϫ mice (Fig. 2C). On the other hand, CAPS2 levels were not altered in G o2 ␣ Ϫ/Ϫ brains, and the inhibitory effect of anti-CAPS2 antibodies on vesicular monoamine uptake was still pronounced in G o2 ␣ Ϫ/Ϫ samples (Fig. 2C).
In adult brains, CAPS1 and CAPS2 are expressed differentially (5,7). In the midbrain, CAPS2 is concentrated in dopaminergic neurons of the ventral tegmental area and substantia nigra (5). The nigro-striatal dopamine system was previously shown to be disturbed in G o2 ␣ Ϫ/Ϫ mice (15). Thus, the different effects of anti-CAPS1 and anti-CAPS2 antibodies on vesicular monoamine uptake of G o2 ␣ Ϫ/Ϫ vesicles seem to be related to alterations of dopaminergic signaling in G o2 ␣ Ϫ/Ϫ mice and to the fact that CAPS1 and CAPS2 are expressed in different subpopulations of monoaminergic neurons. Alternatively, the fact that anti-CAPS2 antibodies still inhibit vesicular monoamine uptake in G o2 ␣ Ϫ/Ϫ brain samples, whereas anti-CAPS1 antibodies do not, may be due to the fact that CAPS1 levels are already significantly reduced in G o2 ␣ Ϫ/Ϫ brain, whereas CAPS2 lev-els are not. Together, our findings are compatible with the notion that the reduced expression of CAPS1 accounts for the reduced monoamine uptake in vesicles of G o2 ␣ Ϫ/Ϫ mice.
So far, the molecular basis of the role of CAPS proteins in vesicle loading remains unclear. Using co-immunoprecipitation assays, chemical cross-linking experiments, or yeast two-hybrid assays, we were unable to detect a direct interaction between CAPS proteins and VMATs or G o2 ␣ (data not shown). Thus, either a direct interaction between CAPS and VMATs or G o2 ␣ does not occur in vivo or the corresponding interactions are weak and/or transient. Nevertheless, CAPS1 might be involved in a signaling cascade that is activated by the vesicular monoamine content, and that depends on the first intravesicular loop of VMAT and vesicle-associated G o2 ␣ (21).
Conclusions-In summary, the present study shows that CAPS proteins facilitate vesicular monoamine uptake and/or storage, most likely by exerting a direct modulatory effect on the uptake machinery. Although this finding is in striking contrast to a previous study in which the effects of CAPS knockdown was studied in cell lines (17), it provides strong biochemical support of previous studies on CAPS-deficient chromaffin cells, which detected a large population of chromaffin granules that are devoid of catecholamines (10,16) and led to the initial suggestion that CAPS proteins do not only play a key role in vesicle priming (1, 2, 7-10) but also regulate vesicular monoamine uptake and/or storage. In addition, our data indicate that reduced expression levels of CAPS1 may contribute to the decreased vesicular monoamine uptake seen in G o2 ␣ Ϫ/Ϫ mice (15).