Streptolysin O (SLO) Permeabilization, β-Hexosaminidase Secretion, and Lgp120 Surface Translocation Assays

Confluent monolayers of NRK cells were washed twice with 1 ml ice-cold buffer A (20 mM Hepes, 110 mM NaCl, 5.4 mM KCl, 0.9 mM Na₂HPO₄, 10 mM MgCl₂, 2 mM CaCl₂, and 11 mM glucose, pH 7.4) and incubated for 5 min at 4°C with SLO (Murex) at 0.5 U/ml in buffer A. After one wash with 1 ml ice-cold buffer B (20 mM Hepes, 100 mM K-glutamate, 40 mM KCl, and 5 mM EGTA, pH 7.2), and 0.5 ml of buffer B containing 2 mM ATP and 5 mM free Mg²⁺ (added as MgCl₂) with or without Ca²⁺ (final [Ca²⁺] 1 μM), and containing antibodies or recombinant Syt C₂ fragments, was added to the cells at 37°C for the indicated time points. The desired concentrations of free Mg²⁺ and Ca²⁺ were obtained with a Ca²⁺ or Mg²⁺/EGTA buffering system calculated using the software by Foehr and Warchol. Cell extracts were obtained by solubilization in buffer B 1% NP-40, followed by centrifugation at 11,000 g. For detection of β-hexosaminidase activity, 350 μl was incubated for 15 min at 37°C with 50 μl of 6 mM 4-methyl-umbellyferyl-N-acetyl-β-d-glucosaminide in sodium citrate-phosphate buffer, pH 4.5. The reaction was stopped by addition of 100 μl of 2 M Na₂CO₃, 1.1 M glycine, and the fluorescence was measured in a F-2000 spectrofluorimeter (Hitachi Instruments, Inc.) at excitation 365 nm, emmision 450 nm.

For detection of surface-translocated Lgp120, NRK cells plated on glass coverslips were permeabilized with SLO as described above and exposed to 100 μl of 1 μM Ca²⁺ buffer containing antibodies or recombinant Syt C₂A fragments for 10 min at 37°C. Cells were then incubated for 30 min at 4°C with culture supernatant from a mouse hybridoma line (Ly1C6) producing antibodies to rat Lgp120, washed with PBS at 4°C, and fixed with 2% PFA for 15 min. After fluorescent staining with DAPI (for detection of cell nuclei) and a goat anti–mouse IgG conjugate (for detection of the anti-Lgp120 mAb), coverslips were examined under a Zeiss axiovert 135 fluorescence microscope.

The C₂A Domain of Syt VII Regulates Ca²⁺-triggered Exocytosis of Lysosomes

Extensive evidence implicating Syt C₂A and C₂B domains in Ca²⁺-triggered exocytosis was obtained through inhibition of exocytosis after introduction of specific peptides, soluble recombinant fragments, or antibodies into neurons (Bommert et al. 1993; Elferink et al. 1993; Mikoshiba et al. 1995; Mochida et al. 1997) or certain endocrine cells (Lang et al. 1997; Mizuta et al. 1997; Ohara-Imaizumi et al. 1997). We took a similar approach to investigate the role of the C₂ domains of Syt VII on Ca²⁺-regulated exocytosis of lysosomes in NRK fibroblasts. Initially, a polyclonal rabbit antiserum was produced by imunization with the purified recombinant C₂A domain of Syt VII. The antibodies were affinity-purified, and shown by Western blot to recognize an ~66-kD band in rat NRK and mouse L fibroblast cell extracts, which comigrates with the band recognized in NRK fibroblasts and L6E9 myoblasts by the antibodies against the Syt VII NH₂ terminus peptide (Fig. 4 a).

On immunofluorescence assays with NRK cells, the anti-C₂A antibodies labeled a vesicular population that also contained the lysosomal glycoprotein, Lgp120 (Fig. 4 b). To confirm that these anti-Syt VII C₂A antibodies reacted with the same molecule recognized by the anti-Syt VII NH₂ terminus antibodies, immunoprecipitation was performed, followed by Western blot. NRK cell (Fig. 4 c, NRK) or rat brain extracts (Fig. 4 c, Brain) were immunoprecipitated with purified preimmune (Fig. 4 c, PI) or anti-C₂A (Fig. 4 c, α-C₂A) IgG. The immunoprecipitates were then run on SDS-PAGE, together with samples of NRK or brain extracts, and processed for Western blot detection with anti-Syt VII or anti-Syt I NH₂-terminal antibodies. Since all antibodies used in this experiment were generated in rabbits, the Western blot also revealed the heavy chain of rabbit IgG, migrating at ~55 kD. It is evident from Fig. 4 c that the preimmune antibodies, although loaded in higher amounts than the immune IgG, did not immunoprecipitate Syt VII or Syt I. On the other hand, Syt VII was clearly detected migrating above the IgG heavy-chain band, after immunoprecipitation of NRK extracts with anti-C₂A antibodies. Smaller amounts of Syt VII were also detected after immunoprecipitation of brain extracts with anti-C₂A antibodies (Fig. 4 c, top). Although Syt I was clearly detected on whole brain extracts, it was not immunoprecipitated by the anti-C₂A antibodies (Fig. 4 c, bottom). These results demonstrate that Syt VII is effectively recognized in NRK cells by the antibodies generated against the Syt VII C₂A domain, and suggest that these antibodies do not cross-react with the C₂A domain of Syt I. Again, similarly to what is shown in Fig. 2 a, Syt VII was not detected in 5 μg of rat brain extract, whereas it was clearly detected on the same amount of NRK cell extract (Fig. 4 c, top).

Previous studies established the predominantly neuronal expression of Syt I, and the ubiquitous distribution of Syt VII (Li et al. 1995; Sudhof and Rizo 1996). Therefore, it was initially puzzling to us the lack of detection of Syt VII in rat brain, particularly in view of reverse transcriptase (RT)-PCR and Western blot data that suggested a higher expression of Syt VII in brain than in other tissues (Li et al. 1995). To clarify this issue, we performed Western blots on increasing amounts of NRK or rat brain extracts, with the antibodies specific for the NH₂ terminus of Syt VII or Syt I. As shown in Fig. 4 d, top, Syt VII was only detected in brain when about eight times more material was loaded in the gels. Syt I, as expected, was detected in all concentrations of brain extract, but not in NRK cells (Fig. 4 d, bottom). These results reinforce our interpretation that this effect is due to the lower abundance of lysosomes in relation to synaptic vesicles in brain extracts. It is important to note that our experiments involved highly specific antibodies against the unique ectodomain of Syt VII, which exclusively detect full-length, transmembrane Syt VII. Li et al. 1995, on the other hand, generated antibodies against the cytosolic region of Syt VII containing the conserved C₂ domains, which, in addition to a possible cross-reactivity with other isoforms (Butz et al. 1999), do not distinguish between full-length and truncated forms generated by alternative splicing. A recent study reported that the vast majority of RT-PCR products of Syt VII obtained from rat brain lacked the transmembrane region, apparently due to exon skipping. Importantly, that study also showed that all the exon-skipped forms of Syt obtained by RT-PCR resulted in frame-shifted messages (Craxton and Goedert 1999). This implies that the original NH₂-terminal sequence of Syt VII would not be present in the truncated forms, explaining the lack of reactivity with our antibodies generated against the NH₂-terminal sequence of full-length Syt VII. It was suggested that unknown upstream exons may import novel NH₂-terminal sequences to these exon-skipped forms, or, alternatively, that frame-shifted forms may be initiated downstream, to produce double C₂ domain-containing proteins (Craxton and Goeder, 1999). Although the physiological function of these truncated forms of Syt VII remains to be determined, it is noteworthy that alternative splicing appears to be a mechanism commonly used in the brain of higher organisms for increasing functional diversity (Black 1998).

Having established that the anti-Syt VII C₂A antibodies recognize lysosome-associated Syt VII in NRK cells, β-hexosaminidase secretion assays (Rodríguez et al. 1997) were performed to verify if the antibodies interfered with Ca²⁺-triggered lysosome exocytosis. When SLO-permeabilized NRK cells were exposed to 1 μM [Ca²⁺] in the presence of 5 or 12 μg/ml affinity-purified anti-Syt VII C₂A antibodies, inhibition levels of 46 and 87%, respectively, were observed in relation to the level of β-hexosaminidase released in the presence of preimmune rabbit antibodies (Fig. 5 a). Next, we followed the translocation of lysosomal membrane proteins to the cell surface, as an additional assay to examine the inhibitory effect of the anti-Syt VII C₂A antibodies on Ca²⁺-triggered lysosome exocytosis. The antibodies generated against the Syt VII NH₂ terminus ectodomain peptide were not used for these surface-labeling assays because they only recognize Syt VII in cells fixed and permeabilized with methanol (possibly reflecting a need to extract other lysosomal proteins to expose the epitopes). Thus, we used the previously characterized assay of Ca²⁺-dependent surface translocation of the lumenal domain of Lgp120 (Rodríguez et al. 1997), which colocalizes with Syt VII in NRK cells (Fig. 3, a–c). The cells were permeabilized with SLO, exposed to buffer containing or not 1 μM Ca²⁺ and 12 μg/ml preimmune IgG or anti-Syt VII C₂A affinity-purified antibodies, and surface-labeled at 4°C with an mAb against a lumenal domain epitope of Lgp120. As shown in Fig. 5b–e, Fig. 1 μM Ca²⁺ triggers the translocation of Lgp120 to the plasma membrane of SLO-permeabilized NRK cells, resulting in a punctate surface fluorescence pattern identical to that previously reported for ionomycin-treated intact NRK cells (Rodríguez et al. 1997). When the cells were exposed to Ca²⁺ in the presence of anti-Syt VII C₂A antibodies, a significant reduction in surface-exposed Lgp120 was observed (Fig. 5, f–g).

These findings strongly suggested a functional role for the C₂A domain of Syt VII in the regulation of lysosome exocytosis. To confirm these results, we proceeded to verify, using the two assays described above, if introduction of soluble Syt C₂ domains had a specific and direct effect on lysosome exocytosis. Histidine-tagged C₂A, C₂B, or C₂A–C₂B domains of Syt VII and of the exclusively neuronal isoform Syt I were expressed in E. coli and purified (Fig. 6 a). NRK cells were permeabilized and exposed to Ca²⁺ for 10 min in the presence of increasing concentrations of the Syt VII C₂A domain, and the supernatant was assayed for released β-hexosaminidase. A clear dose-dependent inhibition in exocytosis was observed, reaching 90% at 200 μg/ml. The same concentrations of the Syt I C₂A domain had no effect (Fig. 6 b).

Open in a separate window

Figure 6

Syt VII C₂A domain inhibits lysosome exocytosis in NRK cells. a, SDS-PAGE (10 μg/lane) of recombinant Syt VII C₂A (1); Syt VII C₂B (2); Syt VII C₂A-B (3); Syt I C₂A (4); Syt I C₂B (5); and Syt I C₂A-B (6). b, Permeabilized cells were stimulated with 1 μM Ca²⁺ in the absence or presence of the indicated concentrations of Syt VII or Syt I C₂A domains, and the supernatant was assayed after 10 min for released β-hexosaminidase. c, Permeabilized cells were stimulated with Ca²⁺ in the absence or presence of 200 μg/ml Syt I fragments, and the supernatant was assayed for released β-hexosaminidase at the indicated time points. d, Same as c, except that 200 μg/ml Syt VII fragments were used. Op-Tc (d) corresponds to an unrelated protein expressed in E. coli and purified under the same conditions. The data is expressed as percentage of the total cellular content of β-hexosaminidase (mean of triplicates ± SD). e–h, Permeabilized cells were stained for surface Lgp120 after incubation in buffer containing: e, no Ca²⁺; f, 1 μM Ca²⁺; g, 1 μM Ca²⁺ plus 200 μg/ml recombinant Syt I C₂A fragments; and h, 1 μM Ca²⁺ plus 200 μg/ml recombinant Syt VII C₂A fragments.

Kinetic β-hexosaminidase secretion assays were then performed in the presence of 200 μg/ml recombinant Syt I and Syt VII C₂A, C₂B, and C₂A–C₂B fragments. No inhibition was observed with the C₂A or C₂B domains of Syt I, and only a partial reduction (of ~30%) in β-hexosaminidase release was detected in the presence of the larger Syt I fragment containing both the C₂A and C₂B domains (Fig. 5 c). In contrast, a strong inhibition, detectable after the initial five minutes, was observed in the presence of both constructs containing the Syt VII C₂A domain (Fig. 6 d). Only a small reduction in exocytosis (of <20%) was detected at the later time points in the presence of the Syt VII C₂B domain alone. An unrelated protein (OP-Tc), expressed and purified under the same conditions, had no effect (Fig. 6 d). Lgp120 surface-translocation assays confirmed the specific inhibitory effect of the Syt VII C₂A domain on lysosome exocytosis. Whereas the typical pattern of Ca²⁺-triggered Lgp120 surface immunofluorescence was observed in cells exposed or not to Syt I C₂A (Fig. 6, f–g), the presence of Syt VII C₂A fragments was strongly inhibitory (Fig. 6 h).

The results of these functional exocytosis assays suggest a specific role for Syt VII in the regulation of Ca²⁺-induced lysosome exocytosis. In particular, our data points to the involvement of C₂A domain-specific functions of Syt VII, which include the binding of syntaxin at low micromolar [Ca²⁺] (Li et al. 1995). Although our findings are consistent with a function for Syt VII as a high affinity Ca²⁺ sensor, additional studies are needed to clarify the basis for this activity, and to further characterize the structural and functional differences between the C₂ domains of different Syt isoforms.

The presence of Syt VII on the membrane of dense lysosomes may explain earlier observations of assembly of AP-2–containing clathrin coats on mature lysosomes of permeabilized NRK cells (Traub et al. 1996). AP-2 has been shown to bind to the C₂B domain of Syt I (Zhang et al. 1994) and several other Syt isoforms, including Syt VII (Li et al. 1995; Haucke and De Camilli 1999). It is thus conceivable that Syt VII is also involved in the retrieval of lysosomal components delivered to the plasma membrane by exocytosis, in an analogous manner to the proposed role of Syt I in synaptic vesicle recycling (Fukuda et al. 1995; Jorgensen et al. 1995).