ATP-containing vesicles in stria vascular marginal cell cytoplasms in neonatal rat cochlea are lysosomes

We confirmed that ATP is released from cochlear marginal cells in the stria vascular but the cell organelle in which ATP stores was not identified until now. Thus, we studied the ATP-containing cell organelles and suggest that these are lysosomes. Primary cultures of marginal cells of Sprague-Dawley rats aged 1–3 days was established. Vesicles within marginal cells stained with markers were identified under confocal laser scanning microscope and transmission electron microscope (TEM). Then ATP release from marginal cells was measured after glycyl-L-phenylalanine-ß- naphthylamide (GPN) treatment using a bioluminescent assay. Quinacrine-stained granules within marginal cells were labeled with LysoTracker, a lysosome tracer, and lysosomal-associated membrane protein 1(LAMP1), but not labeled with the mitochondrial tracer MitoTracker. Furthermore, LysoTracker-labelled puncta showed accumulation of Mant-ATP, an ATP analog. Treatment with 200 μM GPN quenched fluorescently labeled puncta after incubation with LysoTracker or quinacrine, but not MitoTracker. Quinacrine-labeled organelles observed by TEM were lysosomes, and an average 27.7 percent increase in ATP luminescence was observed in marginal cells extracellular fluid after GPN treatment. ATP-containing vesicles in cochlear marginal cells of the stria vascular from neonatal rats are likely lysosomes. ATP release from marginal cells may be via Ca2+-dependent lysosomal exocytosis.

However, the nature of ATP vesicles in the marginal cells was still unclear. Zhang and his colleagues 13 reported that lysosomes in the astrocyte contain abundant ATP that can be released in a stimulus-dependent manner. Selective lysis of lysosomes abolished both ATP release and Ca 2+ wave propagation among astrocytes, implicating physiological and pathological functions of regulated lysosome exocytosis in these cells. In addition, Wang 14 recently reported that autophagy permits immunogenic cell death (ICD)-associated secretion of ATP, which contributes to the maintenance of lysosomal ATP stores. Furthermore, ATP release in this setting is mediated by lysosomal-associated membrane protein 1 (LAMP1) and pannexin 1 (Panx1) -dependent lysosomal exocytosis. Given that the lysosome is ubiquitous across cell types, we suppose that lysosomal vesicles and ATP vesicles depicted by White and co-workers 11 in marginal cells of the stria vascular are the same, and that ATP release from the marginal cells is via Ca 2+ -dependent lysosomal exocytosis.
Next, we report that quinacrine selectively labeled lysosomes in marginal cells and confocal imaging of quinacrine-or Mant-ATP[2′ -/3′ -O-(N′ -Methylanthraniloyl) adenosine-5′ -O -triphosphate] -labeled vesicles indicated that these were lysosomes. Moreover, quinacrine-labeled electron dense precipitates within the cytoplasm in the marginal cells according to transmission electron microscopy (TEM) were identified as lysosomes. And ATP release was measured in the extracellular fluid of marginal cells after glycyl-L-phenylalanineß-naphthylamide (GPN) treatment. These data offered solid evidence for lysosomal ATP storage in cochlear marginal cells of neonatal rats. Our results may provide new insight into mechanisms underlying intracellular ATP storage and release in marginal cells as well.

Results
Primary culture of marginal cells and verification by flow cytometry. We first established a primary culture of marginal cells from cochlear explants of the stria vascular of neonatal rats (Fig. 1). Proliferated marginal cells grew outside the stria vascular explant and were arranged like polygonal paving stones, with individual large nuclei. The epithelial origin of cultured marginal cells in the stria vascular was previously confirmed by expression of cytokeratin 18 15 . Therefore, cytokeratin 18 antibody was used to verify the purity of the cultured marginal cells in the present study. Flow cytometry revealed that 85.3% of the cells were cytokeratin18-positive cells (Fig. 2).

Specific staining of cytoplasmic vesicles of marginal cells under confocal laser scanning microscope.
Several specific markers were used to verify vesicles within marginal cells. Incubation with quinacrine for 30 min at room temperature in the dark resulted in numerous granule-like fluorescent puncta in the cytoplasm in cultured marginal cells under confocal laser scanning microscope (Fig. 3a). Fluorescent puncta in the cytoplasm in 3T3 cells (negative control) was not observed at the same background fluorescence (Fig. 3b). Then, marginal cells loaded with quinacrine (green) were immunostained with LAMP1 (red) (Fig. 4a), a specific marker for lysosomes 16 . Average 89.8% of quinacrine-stained granules were confirmed to be immunopositive for LAMP1, indicating the co-localization of quinacrine-and LAMP1-positive puncta (n = 5, Fig. 4e).
When marginal cells were incubated with quinacrine (green) and labeled with LysoTracker ® Deep Red (red), the lysosome tracer, the co-localization of green and red was observed (Fig. 4b,e). LysoTracker-labelled puncta also showed accumulation of fluorescent ATP analog, Mant-ATP (green) (Fig. 4c,e). While marginal cells labeled with MitoTracker ® Red CMXRos (red), the mitochondria tracer, had not shown the co-localization of green and red after incubation with quinacrine (green) (Fig. 4d,e).

Selectively disrupting lysosomes or mitochondria in cultured marginal cells. After incubation
with quinacrine or LysoTracker ® Deep Red, treatment with 200 μ M GPN, a substrate of the lysosomal exopeptidase cathepsin C that selectively induces lysosome osmodialysis [17][18][19] , largely attenuated the appearance of labeled puncta within the marginal cells, however, this did not occur when staining with MitoTracker ® Red CMXRos ( Fig. 5a-c).

TEM.
The ultra-structure of the cultured marginal cells was observed by TEM which revealed many microvilli-like extensions, numerous coated and uncoated vesicles, coated omega-shaped invaginations, quinacrine labeled lysosomes, and unlabeled mitochondria (Fig. 7). Neighboring cells were connected with tight junctions. Several coated omega-shaped invaginations that may represent regions of active transport through exo-or endocytosis were found in the apical plasma membrane.  (Fig. 8b). Together with given that GPN can disrupt lysosomes in marginal cells, suggesting that these lysosomes most likely contained ATP. ATP release in marginal cells was greater than 3T3 controls after 200 μ M GPN treatment for 5, 10, and 15 min respectively (n = 12, *P < 0.01, independent samples t-test). Interestingly, ATP release in marginal cells was not different at any time point after GPN treatment (n = 12, *P > 0.05, independent samples t-test) (Fig. 8c). In addition, the high concentration of ATP was detectable after 5 min of treatment with 1%Triton X-100. Application of 10 μ M thapsigargin (TG), the endoplasmic reticulum (ER) calcium store inhibitor, also caused ATP release. However, GPN did not increase ATP release after 5 μ M TG treatment for 5 min (n = 12, *P < 0.01, independent samples t-test (Fig. 8d).

Discussion
First, we established a primary culture of marginal cells from cochlear explants of the stria vascular area in neonatal rats. The stria vascular was isolated by microdissection, and dissociated into single cells after tissue incubation for three days. Then marginal cells were purified by trypsin digestion combined with differential adhesion methods. As early as 1989, Rarey and Patterson 23 established primary cell culture of bovine stria vascular using dissociated cell techniques. In this study, ten days were required to purify marginal cells in our group compared to 14 days required by Kim's group 15 . Separation of the stria vascular from the spiral ligament is difficult and marginal cells proliferated toward the apical surface of the explant quickly in the first 3 days of tissue culture. Therefore, we developed a tissue culture technique that can simulate an in vivo environment and allows marginal cell growth. Specifically, because fibroblasts grew rapidly on the 4 th day of culture, 0.25% trypsin was used to digest fibroblasts and then a sterile 0.1% collagenase (type I) solution was used to dissociate the culture into single cells. The differential adhesion method widely used in primary cell cultures [24][25][26] was modified at the purification step which was repeated several times until purified marginal cells were obtained.
Vesicular storage of ATP in cochlear stria vascular marginal cells has been reported and ATP release through exocytosis in human monocytes 27 , astrocytes 13 and microglial cells 20 has been documented. Moreover, ATP release from marginal cells was detected in the previous report 9 . Therefore, in the present study, we explored lysosomal ATP stores in marginal cells of the stria vascular in the cochlea of neonatal rats. Staining of isolated marginal cells with quinacrine, a putative marker of ATP and an acridine derivative which reversibly binds to adenine nucleotide and DNA at a ratio of 1 quinacrine molecule to 4 nucleotides 28-30 , revealed numerous green  granule-like fluorescent puncta in the cytoplasm under confocal laser scanning microscopy. Furthermore, much of the co-localization was observed with double staining of quinacrine/LAMP1, quinacrine/LysoTracker or Mant-ATP/LysoTracker in which LAMP1 and LysoTracker are specific markers for lysosomes, while Mant-ATP is a fluorescent nucleotide analogue used for studying ATP stores and nucleotide-binding proteins 13,20 . Quinacrine-stained vesicles did not pick up the tracer dye of the mitochondria which is one of the potential sources of ATP in the cells. Moreover, TEM revealed that only lysosomes were labeled by quinacrine. A high concentration of ATP was detectable after treatment for 5 min with 1% Triton X-100, indicating that ATP was contained within membrane-enclosed compartments. Furthermore, treatment of marginal cells for 5 min with 200 μ M GPN, a reagent that selectively induces lysosome osmodialysis, increased ATP in the extracellular fluid in a Ca + -dependent manner. Collectively, these results indicate that the vesicles containing ATP are likely lysosomes.
Ultra-structural characteristics of marginal cells under TEM have been reported 15,[31][32][33] . Numerous cytoplasmic coated (CV) and uncoated vesicles (UV), lamellar bodies (LB) of unknown origin have been observed in our work. We suggested these objects may be different stages of lysosomes during cell metabolism. ATP has been reported to be secreted by exocytosis in nerves, human monocytes, astrocytes and microglial cells 13,20,27,34 . Coated omega-shaped invaginations representing regions of active transport were found in the apical plasma membrane, indicating exo-or endocytosis. The presence of lysosomal stores and the apparent abundant vesicular structures observed at the level of the luminal surface of the marginal cells of the stria vascular might imply that ATP was actively transported across the membrane of these cells. Therefore, it is likely that ATP release from marginal cells is through exocytosis.
To further characterize the lysosomal exocytosis in marginal cells, fluorescent FM1-43 dye was incubated with isolated cells. The FM1-43 dyes contain a cationic head, making them impermeable to membranes. When a secretory cell is stimulated to evoke exocytosis, FM1-43 molecules that were inserted in the membrane are internalized during compensatory endocytosis and newly formed secretory granules or vesicles become stained with dye (staining/endocytosis). If stained secretory granules or vesicles undergo exocytosis in dye-free medium, due to concentration gradient, FM1-43 molecules dissociate from the membrane and loose fluorescence (destaining/ excocytosis) 21,22,35 . In our work, increased fluorescence was observed after 2 minutes of GPN stimulation, following with attenuated staining. This recycling of secretory membrane could be monitored as an indication of exocytosis after treatment of GPN 21,22 . Furthermore, we found that AM1-43, a fixable FM1-43 analogue that is largely preserved after immunochemical procedures, labeled about 87.5% population of vesicles that were immunopositive for the lysosomal membrane protein LAMP1, but not for the endosomal marker EEA1. Thus, we suggest that an increased extracellular ATP resulting from treatment of cells with GPN is caused by lysosomal exocytosis. Although the mechanism of GPN-induced increase in extracellular ATP remains unknown, it should not be ruled out that ATP leak from cytoplasm is due to cell damage in this case. More experiments to confirm the ATP release pathways are needed in the future.
Reportedly, Ca 2+ signaling gives rise to disruption of lysosomes by exposure to GPN in human monocytes 27 . Zhang's group demonstrated that calcium wave propagation is the most regulator of lysosomal exocytosis in astrocytes that contain abundant ATP 13 . In our previous work 9 , one of the important characteristics of ATP release from marginal cells was found to be Ca 2+ dependent. In the present study, application of the ER calcium store inhibitor, TG, increased extracellular ATP. However, GPN was unable to increase extracellular ATP following TG treatment. This result suggests an interaction exists between ER and lysosomal compartments in marginal cells and ATP release from these cells is Ca 2+ dependent as well. TG is an inhibitor of the Ca 2+ -ATP enzyme, responsible for replenishing Ca 2+ into the ER. Sivaramakrishnan et al. 27 demonstrated that bidirectional communication exists between lysosomal and ER Ca 2+ stores via three pathways: (i) release of ER Ca 2+ triggers lysosome exocytosis (ER to lysosome); (ii) activation of ryanodine receptors (lysosome to ER) and (iii) release of ATP and mobilisation of ER calcium through activation of PLC-coupled P2Y receptors (lysosome to ER). Ying Dou et al. 20 reported that microglia cells release ATP through Ca 2+ dependent lysosomal exocytosis. Our work may suggest that GPN induces Ca 2+ leak from lysosomes and an increased intracellular Ca 2+ subsequently causes lysosomal exocytosis, thus intracellular ATP was released to the outside of the cell. The extracellular ATP activates cell surface P2Y receptors to generate Ins(1,4,5)P3(IP3) and bind to IP3 receptors, causing ER Ca 2+ release. Since present results are consistent with those previously reported, we speculated that ATP release from marginal cells may be via Ca 2+ -dependent lysosomal exocytosis.
The mechanism underlying ATP accumulation in the lysosomes is unknown. An acidic luminal pH (pHL) in lysosomes must be maintained to activate hydrolytic enzymes and to degrade internalized macromolecules 36,37 . Protons are pumped into the lumen of the lysosomes, which uses energy from ATP hydrolysis through a vacuolar-type H + -ATPase. ATP is negatively charged and ATP-binding-cassette (ABC) proteins are expressed on the lysosome membrane 38 . High-level activity of the proton pump in the lysosome membrane may maintain an even higher positive membrane potential in the lysosome, allowing lysosomes to accumulate more ATP through ABC transporters.
In conclusion, ATP-containing vesicles in cochlear marginal cells of the stria vascular from neonatal rats are likely lysosomes. ATP release from marginal cells may be via Ca 2+ -dependent lysosomal exocytosis.
The limitation of this study is that we could not identify ATP-containing vesicles from secretory vesicles, one kind of membranous vesicular organelles that perform a variety of biological functions ranging from secretion to cellular communication in eukaryotic cells 39 though all the evidence here points to ATP-containing vesicles as lysosomes. More experiments to address this issue are needed in the future.

Marginal cells culture.
Male and female neonatal Sprague-Dawley rats (1-3 days-of-age) were provided by the Shanghai Laboratory Animal Center of the Chinese Academy of Sciences. All parents of neonatal rats were tested for positive auricle reflexes. Rats were euthanized using a rapid guillotine method, approved by the Animal Care Committee. Cochlea was removed from the surrounding bone, and the outer coil containing the stria vascular and spiral ligament was isolated by microdissection in Dulbecco's phosphate-buffered saline (DPBS, pH 7.35, GIBCO). Isolated tissues were then plated on poly-L-lysine coated dishes with the marginal cell surface facing up. Cultures were maintained with growth medium: RPMI-1640 culture medium (Hyclone, NYL1024) was supplemented with 10% fetal bovine serum (FBS, GIBCO, 10099-141, Australia), and 100 U/ml penicillin (GIBCO) in an incubator (Thermo Scientific HERA CELL 150i CO 2 incubator) with 5% CO 2 at 37 o C.
After three days of culture, 0.25% trypsin (1× ) (GIBCO, 15050-065) was used to digest fibroblast cells, and then a sterile 0.1% collagenase (type I) (Sigma, C-0130) solution (w/vol; in serum-free 1640 culture medium) was used to dissociate the culture into single cells for 15 min in an incubator with 5% CO 2 at 37 °C. Cells were harvested by centrifugation at 1,000 × g for 3 min, then re-suspended in 10 ml growth medium and cultured in a 100 mm 3 culture plate in an incubator. Dead and non-adherent cells were removed by refreshing the culture medium after 12 h of culture. Culture medium was refreshed every 2 days and purified marginal cells were obtained after seven days of cell culture by 0.25% trypsin (1× ) (GIBCO, 15050-65) and 0.25% trypsin-EDTA (1× ) (GIBCO, 25200-056) digestion combined with differential adhesion methods. Adherent cells were tentatively identified as marginal cells of the cochlear stria vascular.
All animal experiments were approved by Shanghai Jiaotong University School of Medicine, and the experimental methods were carried out in accordance with the approved guidelines by    (red), Mito-tracker (red) and LAMP1 (red) were obtained with λ excitation = 594 nm. Fluorescence of FM1-43(red) and AM1-43 (red) were detected at 560-625nm with excitation at 488 nm. Fluorescent images of LAMP1 (green) and EEA1 (green) were obtained with λ excitation = 488 nm. A low laser power (less than 0.5% power) was used to avoid possible fluorescent bleaching. A decrease in fluorescent intensities of quinacrine due to photobleaching was less than 5% over 10 min. A total of 5 images obtained from 5 independent experiments were calculated for co-localization analysis in each group. Data analysis was performed with Image Pro Plus 6.0 (Media Cybernetics), SPSS 20.0 software (IBM SPSS Inc., NewYork, State of NewYork) and GraphPad Prism v5.0 (GraphPad Software, Inc., Santiago, California); error bars indicate SD.

TEM.
Marginal cell suspension aliquots were stained with quinacrine dihydrochloride (5 × 10 −6 mol/L; 1 × PBS) for 30 min at room temperature in the dark. Then, fractions were fixed by the slow addition of 4% glutaraldehyde in 0.1 M PBS for more than 2 h at room temperature in the dark. Marginal cells were post-fixed in 1% O s O 4 and dehydrated in ascending concentrations of ethanol and embedded in epoxy resin within the microcentrifuge tube. Sections were then cut with a diamond knife, stained with a saturated solution of uranyl acetate in 50% ethanol and lead citrate. Sections were examined and photographed with a Philips CM120 electron microscope at 80 KV. ATP measurement. To measure ATP release from marginal cells, an ATP bioluminescence assay kit (Promega, G7570) was used along with an opaque 96-well plate (Coring, 3912) to avoid optical cross-talk. Control wells containing culture medium without cells were used for background luminescence and 100 μ l samples per well were withdrawn at room temperature. Clarified samples were mixed 1:1 with luciferase reagent before measurements were made using a Modulus Luminometer (Centro XS 3 LB 960, BERTHOLD) with 1 s integration time. The contents were mixed for 2 min on an orbital shaker to induce cell lysis. Luminescence of 3 parallel replicates was recorded of each cell group and every experiment was repeated four times. Luminescent values represent Error bars indicate SD. means ± SD of 12 replicates for each cell group. After 200 μ M GPN was added to the 5 different dilutions of marginal cells, luminescence was measured before or 5, 10, or 15 min after stimuli, respectively. Treatment with 200 μ M GPN, 1% Triton X-100 (Sigma, T9284), 10 μ M TG (Sigma, T9033), 200 μ M GPN + 5 μ M TG and 5 μ M TG + 200 μ M GPN was performed in another experiment with 10,000 cells per well and 3T3 cells were controls. Data represent averaged results from at least four experiments for each group. Data analysis was performed with SPSS 20.0 software (IBM SPSS Inc., NewYork, State of NewYork) and GraphPad Prism v5.0 (GraphPad Software, Inc., Santiago, California); error bars indicate SD.

Conclusions
Vesicles containing ATP are likely lysosomes in marginal cells of the stria vascular in neonatal rat cochlea. ATP within these vesicles may be transported across membranes through Ca 2+ -dependent lysosomal exocytosis.