Masitinib mediates TGF-Beta1 and Nitric Oxide Secretion and Ameliorates MPTP/Microglia-Induced Degeneration of Differentiated SH-SY5Y Cells

Introduction Microglia secretome includes not only growth factors and cytokines which support neuronal survival, it includes neurotoxic cytokines/enzymes, as well. MPTP is a neurotoxin which has degenerative effects on SH-SY5Y neuroblastoma cells. Masitinib mesylate is a tyrosine kinase inhibitor which has been shown to have beneficial effects in neurodegenerative diseases. Aim We first aimed to determine the most efficient microglial cell conditioned medium in terms of neurodegenerative effect. Next, we investigated the possible protective/therapeutic effects of masitinib against MPTP/microglia-induced degeneration of differentiated (d)-SH-SY5Y cells, and the role of transforming growth factor (TGF)-β1 and nitric oxide (NO) in these events. Material-Methods Non-stimulated/LPS-stimulated microglia cells were treated with masitinib or its solvent, DMSO. With or without MPTP-d-SH-SY5Y cell cultures were exposed to the conditioned media (CM) from microglia cell cultures, followed by cell survival analysis. Immunofluorescence staining of microglia and d-SH-SY5Y cells were performed with anti-CD-11b and anti-PGP9.5 antibody, respectively. TGF-β1/NO concentrations in CM of microglia/d-SH-SY5Y cell culture were measured. Results The initial 24 hrs CM of non-stimulated microglia cell culture was found to be the most detrimental microglial medium with lowest survival rates of treated d-SH-SY5Y cells. The toxicity of 48 and 72 hrs’ CM on d-SH-SY5Y cells were both lower than that of 24 hrs’ CM. Masitinib (0.5 µM), significantly prevented MPTP-related cell degeneration of d-SH-SY5Y cells. It also decreased the degenerative effects of both non-induced/LPS-induced microglia CM on with or without MPTP-d-SH-SY5Y cells. Although NO levels in microglia CM showed a negative correlation with survival rates of treated d-SH-SY5Y cells, a positive correlation was seen between TGF-β1 concentrations in microglial CM and rates of treated d-SH-SY5Y cell survival. Conclusion Masitinib ameliorates viability of with/without MPTP-d-SH-SY5Y cells. It does not only reverse the degenerative effects of its solvent, DMSO, but also prevents the degenerative effects of microglial secretions and MPTP. We suggest that masitinib begins to act as a neuroprotective agent via mediating TGF-β1 and NO secretion, as neurons are exposed to over-activated microglia or neurotoxins.


60
Neurodegeneration, which can be defined as the progressive loss of functional neurons, is a common characteristic of  Table I]. 129 Part 1 cells were stimulated with increasing doses of LPS (29). 135 Part 1-4 supernatants were later applied on Part 1'-4' d-SH-SY5Y cell groups [Table II]. 136

Treatment of different doses of LPS (Part 1) and LPS-/+ and Masitinib/DMSO (Part 2-4) on microglia cells 137
Considering the importance of timing of the onset of the experiments after cell plating and the effects of duration of 138 culture supernatant on its cytokine content, we followed 2 different protocols;  Microglia cells were seeded in 6-well plates as 10 5 cell/well. Groups were formed and the treatments described in 145  Table I were performed. 147 morphological analysis and to measure the intensity of microglial activity, cells were stained with anti-CD11b 148 antibody. Immunofluorescence (IF) analysis ( Figure 6), morphological assessment and cell counting [ Part 1 Part 2 Part 3 Part 4 24 hrs after plating microglia, cell media were refreshed, and groups were formed: 3 hours after plating microglia (once cells attached), groups were formed. No media change. I-No treatment II-1 µg/ml LPS III-5 µg/ml LPS IV-10 µg/ml LPS V-20 µg/ml LPS** I-No treatment II-M-HD followed by 10 µg/ml LPS III-10 µg/ml LPS followed by M-LD IV-10 µg/ml LPS followed by M-HD V-DMSO followed by 10 µg/ml LPS VI-10 µg/ml LPS followed by DMSO*** I-No treatment II-M-LD III-M-HD IV-1 µg/ml LPS V-M-LD followed by 1 µg/ml LPS VI-1 µg/ml LPS followed by M-LD VII-M-HD followed by 1 µg/ml LPS VIII-1 µg/ml LPS followed by M-HD IX-DMSO followed by 1 µg/ml LPS X-1 µg/ml LPS followed by DMSO**

174
Groups as Part 1'-6' were formed and treatments were performed as described in Table II: 175

210
The cells were fixed with paraformaldehyde for 20 min. and blocked with phosphate buffered saline (PBS) containing 211 0.1% Tween-20 (v/v) and 5% bovine serum albumine for 1 hr at room temperature.

224
Microglia cells were counted by staining these cells with antibody against CD11b, which is a marker commonly used 225 to evaluate microglial activation. Cytoplasmic staining intensity was scored 0 to 5. Microglia cell counts (MCCs) and 226 amoeboid cell counts (ACCs) were determined as follows.

227
First, the area with highest microglial cell number was identified at low power (40× and 100×). MCCs were then

251
In without MPTP-d-SH-SY5Y cells, microglia CM which were not activated with LPS (Group P1'-I) caused a higher 252 rate of cell degeneration when compared to the ones activated by increasing doses of LPS (Groups P1-II-V). When 253 the latter supernatants were applied, the dose of LPS that microglia were exposed showed a negative correlation with 254 the viability of SH-SY5Y cells (Groups P1'-II-V).

257
However 20 µg/ml LPS-treated microglia supernatant (Group P1'-V) caused a similiar degeneration with that of non- 287 The initial 24 hrs' CM of microglia which were activated with 1 µg/ml LPS caused also significant degeneration, with 288 around 26% cell viability of both with and without MPTP d-SH-SY5Y cells exposed to this CM (Group P3'-IV). This 289 initial 1 µg/ml-LPS-stimulated microglia CM was the second most detrimental medium -following initial medium of   Table III.
351 Figure 6. IF microscopy images of microglia cells treated with differant concentration of LPS and/or Masitinib.

353
DAPI is blue, CD-11b is GFP. Scale bar is 10 uM.

356
In MPTP-treated d-SH-SY5Y cells (II), PGP9.5 staining intensity was found to be lower than that of control (I). In In microglia cell cultures that were exposed to 1, 5 and 10 µg/ml LPS (II-IV), microglial cell numbers were found to 371 be increased in parallel to LPS dose. However, 20 µg/ml LPS (V) did not cause such an increase in microglial number.

372
In the latter group, amoeboid cell number was found to be significantly increased compared to all other groups. Thus, 373 these results may explain the possible cause of higher rates of cell viability that was seen in d-SH-SY5Y cells exposed 374 to CM of 20 µg/ml LPS induced microglia (V).

375
In 'LPS + M-LD treated microglia group (VI), microglial cell number and CD-11b intensity were found to be lower 376 than that of M-HD treated one (VIII). This may explain the higher rates of cell viability seen in d-SH-SY5Y cells that 377 were exposed to M-LD treated microglia CM compared to M-HD treated one.   Table I).

386 3.3.2. d-SH-SY5Y Cells 387
Morphological assessments showed prominent findings d-SH-SY5Y cell degeneration especially in the center of large 388 cell aggregates due to MPTP treatment, although cell proliferation and aggregate formation were not affected (

415
The highest mean level of TGF-β1 was reached in CM of microglia which were activated by 20 µg/ml LPS (Figure 9). finding was compatible with our morphological assessments, in which a significant degree of cell 504 degeneration was noted especially in the center of large cell aggregates (Figures 7, 8).

516
We also found that non-MPTP treated cells were more susceptible to non-induced microglia CM which were activated with 20 µg/ml LPS (Group P1-V), than that of aforementioned groups (Table   531 III).

532
In this context, in vitro exposure to LPS and/or interferon gamma (IFNγ) has been associated with 533 morphological alterations of microglia from ramified to amoeboid, an activated or "M1" phenotype 534 which has long been associated with neuroinflammation. LPS mainly initiates the production of pro-535 inflammatory cytokines such as, TNFα, IL-1β, IL-6, NO, and ROS in both glial cells and neurons 536 (15,38,39). Microglia can switch phenotype when exposed to specific growth factors or cytokines, as 537 well.

538
Conveniently, in our study, we also found that amoeboid cell count and intensity of CD-11b, a 539 microglial activation marker, showed positive correlations with increasing doses of LPS that microglia 540 were exposed to. Our in vitro neurodegenerative model may reflect and mimic the findings of

557
The treatment of d-SH-SY5Y cells with microglial supernatants in which masitinib-solvent, DMSO 558 was administered on activated/non-activated microglia (Groups P3'-IX and X, respectively) resulted 559 in a lesser percentage of cell viability when compared to masitinib-treated ones. We suggest that 560 masitinib ameliorates cell viability by not only reversing the degenerative effects of its solvent,

561
DMSO, but also preventing the degenerative effects of microglial secretions, as well ( Figure 3). treated ones in terms of TGF-β1 levels, it was found that DMSO significantly decreased the levels of 564 this cytokine (325 pg/ml). Despite that, masitinib which applied to non-induced microglia provided 565 significantly higher levels of TGF-β1, suggesting that masitinib reverses adverse effects of DMSO, 566 besides its own enhancer effect on TGF-β1.

567
Moreover, pro-inflammatory NO levels decrease, whereas TGF-β1 levels increase in microglia CM 568 with increasing doses of LPS. However, 20 µg/ml-LPS treated microglia CM showed high levels of 569 both mediators. Conveniently, this LPS dose was found to be less effective in terms of activating 570 microglial proliferation compared to the effects of 5 or 10 µg/ml LPS treatment, suggesting that LPS 571 may not be effective when applied at very high doses.

572
Furthermore, we found that the protective effects of masitinib in both with or without MPTP treated

587
When almost all the other groups of microglia supernatants were applied to d-SH-SY5Y cells, their 588 survival rates also showed a positive correlation with TGF-β1 levels of the related supernatants (Table   589 III). It suggests that decreased level of anti-inflammatory TGF-β1 may be one of the factors accounting for the decreased viability rates, besides several other cytokines. Conveniently, the 591 expression of TGF-β1 and its receptors were up-regulated in amoeboid microglial cells following 592 hypoxic exposure indicating an autoregulation of microglia in neuropathologies, as reported by Li et al 593 (Li et al., 2008). Additionally, binding of TGF-β1 to its receptors was shown to inhibit free radical 594 production and proliferation of microglia (44-46).

595
Furthermore, TGF-β superfamily, comprising of TGF-β1-3, has anti-inflammatory action and is 596 generally present in low levels in the brain until there is an inflammation (47). In inflammed CNS,

629
Therefore, when considered that masitinib tends to show its neuroprotective effect as the degenerative 630 effects of over-activated microglia or neurodegenerative stimuli are intervened, a biomarker is a 631 prerequisite for detecting the appropriate application time of treatment. Thus, a biomarker, which can 632 indicate that microglial activation or neurodegeneration has initiated, would be of great interest for 633 researchers to arrange a treatment protocol -with right timing-for masitinib.

634
In conclusion, here we highlight the neurotoxic potential of resting and activated microglia, and reveal 635 that these cells remarkably contribute to the increased degeneration of neurons. In our study, we also