MCM triggers calcium transients essentially in resting microglia
In the present study, we stimulated cultured rat microglial cells with a pressure pulse (5 s) of MCM (Fig.1A). This cocktail of MCs mediators obtained by heat (53ºC, 1 h), showed a rapid and transient elevation of intracellular calcium ([Ca2+]i) in one out of two cells (167 out of 353 cells) measured by Fura–2 fluorescence (Fig. 1B, C). As MCs are temperature sensitive, the noxious physical stimuli (53 ºC) produced a clear degranulation verified by visual inspection under microscope (data not shown). This strategy elicits MC degranulation and avoids the use of stimuli such as compound 48/80, antigens, complement proteins or neuropeptides which can also activate microglia themselves. Interestingly, MCM obtained at 37 ºC, was also able to trigger intracellular Ca2+ elevation in 27% (6 out of 22) of microglial cells. At this temperature, MCs showed an intact appearance without clear degranulation. The increase in intracellular Ca2+ evoked in microglial cells by MCM at 37ºC was also transient and reached a peak of 0.23 ± 0.06 and area under curve (AUC) of 4.34 ± 1.12 that were similar to that induced by the MCM obtained by heat (53ºC) (F/F: 0.33 ± 0.01; AUC: 4.38 ± 0.24) (Fig. 1E, F).
Microglial cells with a ramified morphology with long, thin processes and small cell bodies are the major responders to the MCM (80.2 % at 53ºC and 100% at 37ºC) (Fig. 1G, H). These cells exhibit shapes typical of resting microglia and a weaker iba1 signal, a microglia-specific calcium-binding protein, in comparison with expression of iba1 in spindle, rod or amoeboid-shaped cells, corresponding to activated microglia (Fig. 1G). These data suggest that low secretion of mediators, such as the escape of biogenic amines from granules during basal activity of MCs (37ºC) is sufficient to evoke Ca2+ elevations mainly in those microglial cells with a resting phenotype.
Histamine is basically the mediator involved in microglia evoked Ca2+ transient
Next, we investigated which mediators were involved in this Ca2+ signaling. MCs contain a wide array of chemical mediators that can be released to the extracellular medium. We focused on the preformed products contained within the granules. To elucidate which mediators were involved in the promotion of microglia activation, we first directly stimulated microglial cells with bioactive monoamines (histamine, serotonin) and ATP. The effects of histamine (100 M) on Ca2+ signal in microglial cells were similar to those elicited by MCM 53ºC (Fig. 2A, B), with a 43% of response (83 out of 192 cells) (Fig. 2D) and, a [Ca2+]i peak and area under curve of 99.1 ± 0.6% and 121.2 ± 10.5 %, respectively (Fig. 2E, F). Histamine mostly affected intracellular Ca2+ of resting cells (81.9% of cells) in consonance with the effects observed with MCM (80.2% of cells). However, serotonin did not evoke any response. On the contrary, external application of ATP (100 M) caused a Ca2+ transient (Fig. 2C) in most of the stimulated cells (94.8 %; 128 out of 135 cells) basically in activated microglia (70.9%) (Fig. 2D). In addition, while [Ca2+]i peak was not different (88.6 ± 6 %) (Fig. 2E), the area under the curve was significantly lower (62.8 ± 3.1 %) (Fig. 2F) in ATP than in MCM-stimulated cells. When the selective antagonist of the histamine H1 receptor, cetirizine, was incubated within the cell chamber, 10 min before the application of histamine, Ca2+ transient response was completely abolished (Fig. 2B, D). Cetirizine also suppressed microglial activation induced by MCM, although a small pool of cells was resistant to the drug (10.3%; 21 out of 203 cells) (Fig. 2A, D). On the other hand, apyrase, an enzyme that catalyzes ATP hydrolysis, dramatically reduced ATP-response to 15.7 % (8 out of 51 cells); however, it did not affect MCM-microglia response (62.3%; 38 out of 61 cells) (Fig. 2A, C, D). Moreover, while [Ca2+]i peak and the area under the curve were abruptly reduced in cells under ATP + Apyrase (F/F: 12.5 ± 4.1%: AUC: 4.3 ± 1.7%), the effect of apyrase on microglia response mediated by MCM was smaller. [Ca2+]i peak was similar (107 ± 2.4 %), however, the area under the curve was lower (60.6 ± 10.7 %) than that obtained by application of MCM alone (Fig. 2E, F). Next, we selectively blocked PAR2 receptors with the antagonist FSLLRY-NH2 to determine the involvement of mast cell-specific proteases. We incubated cells for at least 10 min with FSLLRY-NH2 (400 M) before proceeding to apply MCM. The drug barely modified [Ca2+]i signal evoked by MCM (F: 107.5 ± 10.2%; AUC: 105.7 ± 8.6%; 38 out of 70 cells)(Fig. 3B, E, F). Lastly, to resolve the implication of highly anionic serglycin proteoglycans (PGs) containing glycosaminoglycan side chains of either heparin or chondroitin sulfate (CS) type in microglia activation, we incubated the MCM with heparinase or chondrotinase ABC for 1h at 37ºC to break down the PGs of heparin and CS, respectively. Chondroitinase ABC did not change the peak amplitude and the area under the curve of [Ca2+]i transients in respect to untreated MCM (F: 95.6 ± 9.25%; area under curve: 99.9 ± 9.1%; 16 out of 30 cells) (Fig. 3C,E,F). Similarly, heparinase neither reduced the amount of [Ca2+]i entry into the cell (F: 78.5 ± 7.7%; area under curve: 106 ± 9.1%; 32 out of 55 cells) (Fig. 3D, E, F). All these data suggest histamine H1 receptor mostly contributes to intracellular Ca2+ elevation in cultured microglia mediated by MCs. To a lesser extent, ATP also contributes in shaping the Ca2+ transient. Because histamine is the key mediator to initiate Ca2+ signaling that triggers microglia activation, we measured histamine concentration of used cocktails. On average, histamine concentration from MCM obtained at 53ºC (1 h) was 275 ± 90 M and at 37ºC (1h) of 17.5 ± 0.6 M.
Ca2+ dependent exocytosis in microglia stimulated by MCM and histamine
ATP is considered the major chemokine attracting microglia towards the injured brain regions [26] and microglia release ATP in response to stimuli which trigger intracellular calcium elevation [27]. Here, we incubated cells with quinacrine (10 μM for 10 min at room temperature), a fluorescent marker for intracellular ATP-enriched vesicles used to examine Ca2+ dependent regulated exocytosis of ATP [6][7]. We quantified exocytosis measuring fluorescence changes of cells after stimulation with MCM for 30s through a micro-perfusion pipette. As a control, application of an external solution produced a variable but gradual loss of fluorescence signal in 65% of cells which can be attributed to spontaneous vesicle fusion with the plasma membrane and dispersal of its fluorescent cargo (Fig. 4A, B, E). However, application of MCM induced a more abrupt loss of fluorescence (Fig. 4C, D, F) in 90,3% of cells to a significantly larger extent (F: 0.43 ± 0.02; n = 113) than that observed in control cells (F: 0.26 ± 0.04; n = 40). Furthermore, stimulation with histamine 100 M and LPS 1 g/ml also induced an important loss of fluorescence of 0.5 ± 0.04 and 0.49 ± 0.02, respectively, in more than 92% of cells (Fig. 4H). ATP did not increase significantly the exocytosis in 89.5% of cells affected by the nucleotide. Cells with an activated morphology prevail over those of resting phenotype when measuring exocytosis in all used treatments (Fig. 4G). These results suggest that despite few activated microglia were able to rise [Ca2+]I in response to MCM, the vast majority of them undergo exocytosis of quinacrine loaded-organelles, comparable to that observed by histamine or LPS.
Divergent microglia metamorphosis induced by the MCM
As microglial activation is characterized by morphological changes and upregulation of the specific calcium-binding protein iba1, we analyzed the area and iba1-fluorescence in cells iba1-positive from control cells and cells treated during 6 h and 48 h with MCM 10%, histamine 10 and 100 M, ATP 100 M and LPS 1 g/ml. MCM was diluted (1:10) because the pure extract (106 MCs/mL) would have required a wasteful volume of cells. Intriguingly, data analysis showed different results when using the two different cocktails of mediators assayed (named MCM1 and MCM2) so the results have been shown separately. Cells treated with histamine 100 M (HIS100) were not viable after 48 h of treatment and only 12 cells from 4 different cultures could be included in the analysis. MCM1 showed a reduction in area at 6 h (154 ± 10 m2; n = 143) but an increase in area at 48 h (348 ± 20 m2; n = 112) in respect to control cells (6h: 185 ± 9.5 m2, n = 218; 48h: 194 ± 10 m2, n = 238) (Fig. 5A, B, C). Nevertheless, MCM2 did not show differences in area at 6 h (201 ± 14 m2; n = 125) or 48 h (200 ± 15 m2; n = 133). Histamine 10 M (HIS10) showed an increase in area at 48 h (226 m ± 16 m2; n = 251) while HIS100 reduced the area at 6 h (171 ± 8 m2; n = 219) but it did not modify area at 48h (210 ± 41 m2; n = 12). LPS showed an important increment of area at 6 h (293 ± 35 m2; n = 166) and a reduction at 48 h (155 ± 9 m2; n = 215) in respect to control cells. Finally, ATP only showed an increase of area in respect to control cells at 6 h (210 ± 9 m2; n = 180) (Fig. 5A, B). On the other hand, the effects of both cocktails of mediators and both histamine concentrations did not change iba1 intensity at 6 h (Fig. 5D) but modified it significantly at 48h. MCM1 and HIS100 showed a similar tendency of reduction of iba1 fluorescence (MCM1: 26.1 ± 1.2; HIS100: 34.2 ± 2.3) in regard to control cells (49.6 ± 2.4) while MCM2 and HIS10, on the contrary, significantly raised iba1 intensity (MCM2: 56.6 ± 12.6; HIS10: 58.4 ± 2.3). The higher increment was obtained with LPS at 48 h (82.6 ± 3.5). ATP reduced iba1 intensity both at 6 h (48.8 ± 2.1) and 48 h (42.1 ± 1.7). Overall, MCM1 showed a similar pattern of area modification and Iba1 expression to histamine 100 M while MCM2 showed a similar response to histamine 10 M. The concentration of histamine was measured and resulted in 21 µM for the MCM1 and 3.7 µM for the MCM2. These results indicate that the histamine concentration regulated the level of iba1 expression, hence it could derive in two opposite microglia phenotypes, which was confirmed when phagocytosis was studied.
The phagocytotic phenotype of microglia is determined by the histamine concentration in the MCM
Phagocytosis was assessed by using fluorescent latex beads of 1m under the same conditions and incubation intervals as above. Both MCM1 and HIS100 showed a reduction in the engulfment of microspheres per cell at 6h (MCM1: 0.71 ± 0.1; HIS100: 1.1 ± 0.1) and 48h (MCM1: 0.66 ± 0.17; HIS100: 0.33 ± 0.22) regarding control cells (6h:1.5 ± 0.2; 48h:1.92 ± 0.2) (Fig. 6A, B). MCM2 and HIS10 also showed a similar value of uptake of fluorescent beads. Neither of them affected phagocytosis at 6 h yet increased the internalization of microspheres at 48h (MCM2: 3.07 ± 0.35; HIS10: 3.08 ± 0,2) to the same extent as LPS (2.91 ± 0.3). ATP did not modify phagocytosis at any time (1.8 ± 0.2, 6h; 1.94 ± 0.2, 48h). A clear concordance was again observed in the responses between MCM1 and HIS100 and between MCM2 and HIS10. In addition, once more, inverse effects on phagocytosis were observed between the two assayed cocktails with low and high concentrations of histamine. The percentage of total cells with phagocytosed beads was 49.8 % (6h) and 51.3% (48h) in control cells. Among these, a small number of cells were able to uptake more than 6 beads (7.3%, 6h; 10.1%, 48h). This number was almost non-existent in cells treated with MCM1 (0 %, 6h; 2.7%, 48h) and HIS100 (5%, 6h; 0%, 48h), but it increased in cells treated with MCM2 (6.4%, 6h; 18.1%, 48h) and HIS10 (7.7%, 6h; 20.3%, 48h) for 48 h (Fig. 6D, E). LPS was a positive activator of phagocytosis (12.1%, 6h; 20%, 48h) while ATP had a behavior similar to control cells (11.7%, 6h; 11%, 48h).
Finally, in order to understand how morphological changes and iba1-intensity can influence functional responses of microglia, we have related these two parameters with phagocytosed beads per cell. Interestingly, we observed an exponential relationship between the area and engulfed beads at 6 h. In short periods, larger cells are more active in phagocytosis (Fig. 6F). This relation was not observed at 48 h (Fig. 6G, inset). In contrast, iba1 expression was a better value to predict phagocytosis at longer times. There is a sigmoidal relation between iba1 intensity and the number of beads per cell in each condition at 48 h (Fig. 6G). Higher fluorescence intensity indicates higher intakes of beads. This relationship could not be observed at 6 h (Fig. 6F, inset).