ENT-A010, a Novel Steroid Derivative, Displays Neuroprotective Functions and Modulates Microglial Responses

Tackling neurodegeneration and neuroinflammation is particularly challenging due to the complexity of central nervous system (CNS) disorders, as well as the limited drug accessibility to the brain. The activation of tropomyosin-related kinase A (TRKA) receptor signaling by the nerve growth factor (NGF) or the neurosteroid dehydroepiandrosterone (DHEA) may combat neurodegeneration and regulate microglial function. In the present study, we synthesized a C-17-spiro-cyclopropyl DHEA derivative (ENT-A010), which was capable of activating TRKA. ENT-A010 protected PC12 cells against serum starvation-induced cell death, dorsal root ganglia (DRG) neurons against NGF deprivation-induced apoptosis and hippocampal neurons against Aβ-induced apoptosis. In addition, ENT-A010 pretreatment partially restored homeostatic features of microglia in the hippocampus of lipopolysaccharide (LPS)-treated mice, enhanced Aβ phagocytosis, and increased Ngf expression in microglia in vitro. In conclusion, the small molecule ENT-A010 elicited neuroprotective effects and modulated microglial function, thereby emerging as an interesting compound, which merits further study in the treatment of CNS disorders.


Introduction
Neurodegeneration and neuroinflammation are fundamental hallmarks of central nervous system (CNS) disorders [1]. Neurodegeneration refers to the impaired function and loss of neurons, which is often permanent due to the limited regenerative capacity of the adult neural system [2]. Neuroinflammation involves the inflammatory activation of glial cells, that is, microglia and astrocytes, and can have protective or destructive consequences for the neural system [1,3]. Microglia play a key role in the maintenance of homeostasis due to synaptic pruning and their clearing and neurotrophic function [3][4][5], while their inflammatory activation is a prerequisite for elimination of insults (infections 2.1.2. Synthesis of (E)-[3β-(t-butyldimethylsilyloxy)-5-androsten-17-ylidene]ethyl ester (2) To a solution of compound 1 (2.25 g, 6.28 mmol) in anhydrous THF (20 mL), imidazole (1.33 g, 19.5 mmol) and iodine (4.76 g, 37.5 mmol) were added at 0 • C and the mixture was stirred at 0 • C for 30 min. Subsequently, tert-butyldimethylsilyl chloride (1.05 g, 6.67 mmol) was added and the resulting mixture was stirred at 25 • C overnight. After completion of the reaction, the solvent was evaporated in vacuo and the residue was diluted and extracted with ethyl acetate (EtOAc). The organic layer was washed with sat. aq. Na 2 S 2 O 4 , brine and dried over anhydrous Na 2 SO 4 14.5, 18.4, 19.6, 21.1, 24.6, 26.1, 30.6, 31.7, 31.8, 32.2, 35.4, 36.8, 37.5, 42.9, 46.2, 50.5, 54.
Mp  2.1.5. Synthesis of (17S,20S)-3β-(t-butyldimethylsilyloxy)-17α,20-methan-5-pregnane-21-al (5) To a solution of compound 4 (0.85 g, 1.91 mmol) in dry CH 2 Cl 2 (100 mL) was added at 0 • C Dess-Martin periodinane (1.62 g, 3.82 mmol) and the reaction mixture was stirred at 25 • C for 1.5 h. After completion of the reaction (checked by TLC), a mixture of saturated aqueous NaHCO 3 and 10% aq. Na 2 S 2 O 4 (1:1) were added and the resulting mixture was stirred for 30 min. The reaction mixture was extracted with diethyl ether (Et 2 O) (30 mL × 3 times) and the combined organic layers washed with saturated aqueous NaHCO 3 and brine, dried over Na 2 SO 4 and the solvent was removed in vacuo. The residue was purified by FCC (petroleum ether 40-60 • C/EtOAc 98/2→96/4) to afford compound 5 as a white crystalline solid (0.71 g, yield = 84%  (6) To a suspension of NaH 60% in mineral oil (0.3 g, 6.5 mmol) in dry THF (4 mL) was added at 0 • C triethylphosphonoacetate (1.3 mL, 6.5 mmol) and the reaction mixture was stirred at RT for 30 min. Subsequently, a solution of compound 5 (0.72 g, 1.62 mmol) in dry THF (16.4 mL) was added at 0 • C and the reaction mixture was stirred at the same temperature for 30 min. The reaction was quenched with saturated aqueous NH 4 Cl and extracted with EtOAc (20 mL × 3 times). The combined organic layers were washed with brine, dried over anhydrous Na 2 SO 4 and the solvent was removed in vacuo. The residue was purified by FCC (petroleum ether 40-60 • C/ethyl acetate 99/1→98/2) to afford compound 6 as a white crystalline solid (0. 8  To a solution of compound 6 (0.8 g, 1.56 mmol) in anhydrous THF (47 mL) TBAF (1.0 M in THF) (5.4 mL, 5.4 mmol) was added dropwise at 0 • C and the reaction mixture was stirred at RT overnight. The reaction was quenched with water at 0 • C, and the resulting mixture was extracted with EtOAc (30 mL × 3 times). The organic layer was washed with brine, dried over anhydrous Na 2 SO 4 and the solvent was removed in vacuo.

Mice
Eight-to twelve-week-old C57BL/6J male mice were used (Charles River Laboratories, Sulzfeld, Germany). Mice had free access to food and water and were housed under a 12-h light-dark cycle. They were intraperitoneally (i.p.) injected at two consecutive days with freshly prepared ENT-A010 (70 mg/kg) in phosphate-buffered saline (PBS) with 4.5% ethanol and 1% bovine serum albumin (BSA) (Sigma-Aldrich, St. Louis, MO, USA), or vehicle control solution. One hour after the second injection, the mice were injected i.p. with lipopolysaccharide (3 mg/kg) (Ultrapure LPS, Escherichia coli 0111:B4, Invivogen, San Diego, CA, USA). After 16 h, they were deeply anesthetized with ketamine/xylazine and intracardially (i.c.) perfused with PBS. Afterward, brain regions were isolated and snap-frozen for further analyses, or brains were collected for histological analyses. For the study of ENT-A010 brain uptake, mice were administered i.p. 70 mg/kg ENT-A010 (in PBS containing 4.5% ethanol and 1% BSA) or vehicle solution and they were euthanized 1 and 2 h after injection with ENT-A010 and 1.5 h after injection with control solution. Different brain regions (hippocampus, hypothalamus, cortex, brainstem, cerebellum), livers and spleens were collected and snap-frozen. Animal experiments were approved by the Landesdirektion, Dresden, Germany.

TUNEL Assay
DRG neurons were maintained for 48 h without NGF and in the presence of a NGFneutralizing antibody (N8773, 1:500, Sigma-Aldrich, St. Louis, MO, USA) with ENT-A010 (500 nM) or the same amount of dimethyl sulfoxide (DMSO). Primary hippocampal neurons were treated for 48 h with 5 µM oligomeric Aβ (AnaSpec, Fremont, CA, USA) in the presence of ENT-A010 (500 nM, newly supplemented every 24 h) or vehicle control (DMSO). Cells were then fixed with 4% paraformaldehyde (PFA) and labelled with terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL, Roche, Hertfordshire, UK) following the manufacturer's protocol. Subsequently, cells were immunostained against TUJ1 (1:2000, 801201, Biolegend, San Diego, CA, USA) and anti-mouse Cy3 (1:1000, Invitrogen, A10521, Waltham, MA, USA) and imaged with a Leica SP8 confocal microscope. The percentage of apoptotic neurons was determined by normalizing the number of TUNEL + cells to the total number of Hoechst + TUJ1 + neuronal cells. The FIJI software was used to determine the numbers of TUNEL + and Hoechst + cells [50].
Aβ 1-42 peptide was prepared according to the manufacturer's instructions. Oligomeric Aβ, which is considered to be the toxic form of amyloid, was prepared as previously described with slight modifications [51]. Aβ peptide was diluted in DMEM at 5 µM and incubated for 24 h at 37 • C. It was then centrifuged for 5 min at 14,000× g and the supernatant containing oligomeric Aβ was collected.

CellTox Assay
PC12 cells were plated in 96-well plates, serum-deprived for 4 h and subsequently treated with or without NGF (100 ng/mL) or ENT-A010 (500 nM) in the presence or absence of GW441756 TRKA inhibitor (20 µM, G-190, Alomone Labs, Jerusalem, Israel) for another 24 h. The CellTox assay (Promega, Leiden, Belgium) was performed according to the manufacturer's instructions. Hoechst (1:10,000, Invitrogen, MA, USA) was added for 30 min along with CellTox reagent and cells were imaged with a Zeiss AXIO Vert A1 fluorescent microscope. The number of CellTox + (dead) cells was normalized to the number of Hoechst + cells, the latter depicting the total number of cells, per image. The numbers of CellTox + and Hoechst + cells were determined using the FIJI software.

Phagocytosis Assay
Primary microglial cells were treated on two consecutive days with 1 µM ENT-A010 or an equal volume of DMSO. One hour after the second dose, cells were stimulated or not for 24 h with 100 ng/mL LPS (Ultrapure from E. coli K12, Invivogen, San Diego, CA, USA). In some experiments, cells were treated with the AKT inhibitor MK2206 (2.5 µM, Cayman Chemical, MI, USA), 30 min prior to the first treatment with ENT-A010. Cells were then given for 2 h 750 nM HiLyte™ Fluor 555-labeled Aβ 1-40 (Anaspec, Fremont, CA, USA). Afterward, they were harvested by gentle scraping, washed twice with FACS buffer (3% FBS and 2 mM EDTA in PBS), re-suspended in FACS buffer and analyzed with a BD FACSCanto II flow cytometer using the FACSDiva 6.1.3 software. The mean fluorescent intensity was determined after debris and doublet exclusion in a total number of 10,000 events per sample.

Western Blotting
Proteins extracts were prepared in ice-cold RIPA buffer supplemented with protease and phosphatase inhibitors (Roche, Hertfordshire, UK). Protein concentration was determined with the BCA assay (Thermo Scientific, Rockford, IL, USA). Protein lysates were mixed with reducing Laemmli SDS sample buffer (Alfa Aesar, Haverhill, MA, USA), denatured at 95 • C for 5 min and 100 µg of protein were loaded on a polyacrylamide gel and separated with SDS-PAGE. Afterward, proteins were transferred onto nitrocellulose membranes and blocking was performed with 5% BSA in TBS-T buffer overnight at 4 • C. The next day, primary antibodies were added to the membranes and incubated overnight at 4 • C. Primary antibodies used were following: anti-phosphoTRKA (phospho Y490) (Abcam, ab1445), anti-TRKA (Abcam, ab76291), anti-phosphoAKT (Ser473) (Cell Signaling, #4060), anti-AKT (Cell Signaling, #4691) and anti-Vinculin (Cell Signaling, #4650), all diluted at 1:1000 in 5% BSA in TBS-T. On the third day, goat anti-rabbit IgG horseradish peroxidaseconjugated antibody (1:3000, R&D Systems, Minneapolis, MN, USA) was added to the membranes and incubated for 2 h at RT. Finally, membranes were washed with TBS-T and developed using SuperSignal West Pico Chemiluminescent Substrate (Life Technologies, Carlsbad, CA, USA) or SuperSignal West Fempto Chemiluminescent Substrate (Life Technologies) and the luminescent image analyzer LAS-3000 (Fujifilm, Dusseldorf, Germany). The intensity of the bands was quantified using the FIJI software [38,43].

Immunofluorescent Staining and Confocal Microscopy
Brains were post-fixed with 4% PFA in PBS for 4 h at 4 • C, immersed in a 30% sucrose solution in PBS and incubated overnight at 4 • C. Then, the tissues were embedded in OCT compound (Tissue-Tek), and frozen at −80 • C. Fourteen-µm-thick coronal sections were cut by cryosectioning and transferred onto slides. Antigen retrieval was performed in 0.1 M citrate buffer (pH 6.0) for 10 min in a pressure cooker. Tissue sections were blocked with a serum-free Protein Block (Dako, Denmark) for 2 h at RT, followed by overnight incubation with anti-IBA1 primary antibody (019-19741, Wako, Osaka, Japan) diluted 1:750 in antibody diluent (Dako, Denmark). The next day, sections were washed and incubated with donkey anti-rabbit IgG-Alexa Fluor 555 (Life Technologies, Carlsbad, CA, USA) diluted 1:350 in antibody diluent for 2 h. Then, the nuclei were counterstained for 5 min with DAPI (1:10,000 in PBS) and the slides were mounted using Fluoromount (Life technologies, Carlsbad, CA, USA).
Confocal imaging was performed using a Zeiss LSM880 system (Zeiss, Jena, Germany). Images were acquired with the ZEN 3.2 blue edition software. Sections were excited with two laser lines at 405 nm, and 561 nm and emissions were detected with photomultiplier tube detectors within a window of 415-478 nm (DAPI) and 571 nm-625 nm (Alexa-Fluor 555), respectively. An air immersion 20×/0.8NA Plan-Apochromat was used at zoom factor 1. Pixel size corresponded to 210 nm × 210 nm. Laser power, pinhole size and photomultiplier gain were determined for each fluorophore at the beginning and were kept constant throughout the imaging procedure. Six Z-stacks were taken with a step size of 1.57 µm, through a total thickness of 7.868 µm per section. Processing was performed with the FIJI software.

RNA Extraction and qRT-PCR
Total RNA was extracted from cells or tissues using the Nucleospin RNA isolation kit (Macherey-Nagel, Dueren, Germany), according to the manufacturer's instructions. cDNA was synthesized using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). qPCR was performed using the SsoFast Eva Green Supermix (Bio-Rad Hercules, CA, USA), a CFX384 real-time System C1000 Thermal Cycler (Bio-Rad), and the Bio-Rad CFX Manager 3.1 software, as previously described [36,38,43]. The relative amount of mRNA was calculated with the ∆∆Ct method, using 18s as a housekeeping gene. The primer sequences are listed in Table S1.

Statistical Analysis
All values are expressed as the mean ± SEM. Mann-Whitney U or Student's t-test was used for the comparison of two groups. One-way ANOVA followed by Tukey's multiplecomparison test was used for multiple group comparisons. A p < 0.05 was considered to mark statistical significance. Statistical analysis was performed using GraphPad Prism 7 (GraphPad Software Inc., San Diego, CA, USA).

Synthesis of ENT-A010
The synthesis of ENT-A010 starting from DHEA involves seven high-yielding steps as shown in Figure 1. The Horner-Emmons reaction of DHEA with triethyl phosphonoacetate in the presence of EtONa as a base gave the (E)-a,b-unsaturated ester 1 in 96% yield, which was, in turn, reacted with tert-butyldimethylsilyl chloride (TBSCl) to afford the tertbutyldimethylsilyl-protected alcohol 2 in 93% yield. Reduction of the ester group in 2 using DIBAL-H gave the allylic alcohol 3 in quantitative yield, which was subjected to a Simmons-Smith cyclopropanation reaction in the presence of CH 2 I 2 and Et 2 Zn to yield the (17S,20S)cyclopropyl derivative 4 in 60% yield [52]. Compound 4 was consequently oxidised with Dess-Martin periodinane in dichloromethane to afford the corresponding aldehyde 5 in 84% yield. Horner-Emmons reaction of aldehyde 5 with triethylphosphonoacetate in the presence of NaH afforded only the E isomer of the a,b-unsaturated ester 6 in 96% yield. Deprotection of the C3 alcohol using tetrabutylammonium fluoride (TBAF) 1.0 M in THF yielded quantitatively ENT-A010.

ENT-A010 Promotes Neuronal Survival in a TRKA-Dependent Manner
ENT-A010 was selected for further investigation from a library of novel synthetic C17spiro-DHEA derivatives (a manuscript describing their synthesis is in preparation) based on its capacity to induce TRKA phosphorylation. To test TRKA phosphorylation, PC12 cells were stimulated for 30 min with ENT-A010 or NGF, and proteins were immunoprecipitated for TRKA and immunoblotted against phosphorylated tyrosine. ENT-A010 induced TRKA phosphorylation, at levels similar to those of NGF ( Figure 2).

ENT-A010 Promotes Neuronal Survival in a TRKA-Dependent Manner
ENT-A010 was selected for further investigation from a library of novel synthetic C17-spiro-DHEA derivatives (a manuscript describing their synthesis is in preparation) based on its capacity to induce TRKA phosphorylation. To test TRKA phosphorylation, PC12 cells were stimulated for 30 min with ENT-A010 or NGF, and proteins were immunoprecipitated for TRKA and immunoblotted against phosphorylated tyrosine. ENT-A010 induced TRKA phosphorylation, at levels similar to those of NGF ( Figure 2).

ENT-A010 Promotes Neuronal Survival in a TRKA-Dependent Manner
ENT-A010 was selected for further investigation from a library of novel synthetic C17-spiro-DHEA derivatives (a manuscript describing their synthesis is in preparation) based on its capacity to induce TRKA phosphorylation. To test TRKA phosphorylation, PC12 cells were stimulated for 30 min with ENT-A010 or NGF, and proteins were immunoprecipitated for TRKA and immunoblotted against phosphorylated tyrosine. ENT-A010 induced TRKA phosphorylation, at levels similar to those of NGF ( Figure 2).  Then, we asked if ENT-A010, similarly to DHEA and NGF, may exert neuroprotective function [25,26,28,40,53]. To examine this, we used three different in vitro experimental models to induce cell apoptosis: PC12 cells cultured in serum-deprived conditions, NGFdeprived dorsal root ganglia (DRG) neurons and Aβ-induced cell death of hippocampal neurons. PC12 cells were cultured for 24 h in serum-free medium to undergo apoptosis and ultimately cell death [25,26,54], and treated or not with NGF or ENT-A010 for 24 h. ENT-A010, similarly to NGF, protected PC12 cells against serum deprivation-induced cell death, while TRKA inhibition with a selective TRKA inhibitor, GW174456, partially reversed the pro-survival effect of ENT-A010 (Figure 3). tive function [25,26,28,40,53]. To examine this, we used three different in vitro experimental models to induce cell apoptosis: PC12 cells cultured in serum-deprived conditions, NGF-deprived dorsal root ganglia (DRG) neurons and Aβ-induced cell death of hippocampal neurons. PC12 cells were cultured for 24 h in serum-free medium to undergo apoptosis and ultimately cell death [25,26,54], and treated or not with NGF or ENT-A010 for 24 h. ENT-A010, similarly to NGF, protected PC12 cells against serum deprivationinduced cell death, while TRKA inhibition with a selective TRKA inhibitor, GW174456, partially reversed the pro-survival effect of ENT-A010 (Figure 3). Next, we set out to examine whether ENT-A010 affects the survival of DRG neurons, a neuronal population the survival of which is known to be highly dependent on NGF [53]. Primary DRG neurons were isolated from P0-P1 mouse pups and cultured for 14 days in the presence of NGF. Then, an NGF-free and anti-NGF supplemented medium was added to the cells for 2 days, thereby promoting DRG neuron cell death ( Figure 4). ENT-A010, similarly to NGF treatment, prevented cell death of DRG neurons (Figure 4). Next, we set out to examine whether ENT-A010 affects the survival of DRG neurons, a neuronal population the survival of which is known to be highly dependent on NGF [53]. Primary DRG neurons were isolated from P0-P1 mouse pups and cultured for 14 days in the presence of NGF. Then, an NGF-free and anti-NGF supplemented medium was added to the cells for 2 days, thereby promoting DRG neuron cell death ( Figure 4). ENT-A010, similarly to NGF treatment, prevented cell death of DRG neurons (Figure 4).  Given that hippocampal neurons are prone to Aβ-induced cell death [55], we asked whether ENT-A010 can also preserve survival in toxic Aβ-challenged hippocampal neurons. We treated the latter with Aβ1-42 oligomers and ENT-A010 or control vehicles and found that ENT-A010 exerted strong protection against Aβ-induced cell death ( Figure  5A). Moreover, we examined whether ENT-A010 can reverse Aβ-induced synapse degen- Given that hippocampal neurons are prone to Aβ-induced cell death [55], we asked whether ENT-A010 can also preserve survival in toxic Aβ-challenged hippocampal neurons. We treated the latter with Aβ 1-42 oligomers and ENT-A010 or control vehicles and found that ENT-A010 exerted strong protection against Aβ-induced cell death ( Figure 5A). Moreover, we examined whether ENT-A010 can reverse Aβ-induced synapse degeneration [56]. To this end, hippocampal neurons were kept in culture for 16 days after isolation and treated for 4 h with Aβ 1-42 oligomers and ENT-A010, followed by immunostaining for Synaptophysin, a pre-synaptic marker. ENT-A010 restrained Aβ-induced loss of Synaptophysin ( Figure 5B). Synaptophysin-and TUJ1-positive areas were quantified and the ratio (Synaptophysin + area /TUJ1 + area) was calculated and normalized in each experiment to the control. Ten to twelve images were acquired per sample. Scale bar: 100 μm (applies to all non-zoomed photos), and 10 μm (applies to all zoomed-in photos). Data are shown as mean ± SEM, n = 3-4, *: p < 0.05, **: p < 0.01.

ENT-A010 Promotes Phagocytosis in Microglia
In vivo, the effects of drugs on neurons may be altered by their effects on microglia. Hence, we set out to examine whether ENT-A010, in addition to its neuroprotective effects, may also affect microglial functions, given that steroid hormones derive from and target microglia [11]. We previously showed that functional TRKA is expressed in microglia [38,43]. Here, we show that ENT-A010 induced TRKA and AKT phosphorylation in microglial cells ( Figure 6A,B, respectively). NGF was previously shown to promote macropinocytosis and Aβ clearance in microglia [42]. Therefore, we assessed the effects of ENT-A010 on Aβ engulfment and found that it increased Aβ uptake in LPS-treated microglia ( Figure 6C), while its effect was abolished by the AKT inhibitor MK2206 ( Figure 6D). Synaptophysin-and TUJ1-positive areas were quantified and the ratio (Synaptophysin + area /TUJ1 + area) was calculated and normalized in each experiment to the control. Ten to twelve images were acquired per sample. Scale bar: 100 µm (applies to all non-zoomed photos), and 10 µm (applies to all zoomed-in photos). Data are shown as mean ± SEM, n = 3-4, *: p < 0.05, **: p < 0.01.

ENT-A010 Promotes Phagocytosis in Microglia
In vivo, the effects of drugs on neurons may be altered by their effects on microglia. Hence, we set out to examine whether ENT-A010, in addition to its neuroprotective effects, may also affect microglial functions, given that steroid hormones derive from and target microglia [11]. We previously showed that functional TRKA is expressed in microglia [38,43]. Here, we show that ENT-A010 induced TRKA and AKT phosphorylation in microglial cells ( Figure 6A,B, respectively). NGF was previously shown to promote macropinocytosis and Aβ clearance in microglia [42]. Therefore, we assessed the effects of ENT-A010 on Aβ engulfment and found that it increased Aβ uptake in LPS-treated microglia ( Figure 6C), while its effect was abolished by the AKT inhibitor MK2206 ( Figure 6D).

ENT-A010 Promotes a Protective Microglial Phenotype
Next, we examined whether ENT-A010 modulates the expression of Triggering Receptor Expressed On Myeloid Cells 2 (Trem2) and MER Proto-Oncogene, Tyrosine Kinase (Mertk), which play a key role in phagocytosis [57][58][59][60]. ENT-A010 increased Trem2 and Mertk expression in LPS-treated primary microglia ( Figure 7A,B), standing in accordance with its effect on phagocytosis ( Figure 6C,D). Moreover, it increased the expression of Ngf ( Figure 7C), suggesting that ENT-A010-treated microglia may display enhanced neuroprotective function. In addition, ENT-A010 treatment enhanced the expression of Arginase 1 (Arg1) in LPS-treated microglia, indicating that it may modulate arginine metabolism and promote an M2-like microglial phenotype ( Figure 7D) [38]. Finally, ENT-A010-treated microglia showed a tendency for increased Cx3cr1 expression, suggesting that ENT-A010 might restore the expression of homeostatic genes in inflammatory microglia ( Figure 7E). We previously showed that TRKA activation by NGF or DHEA restrains inflammatory activation of LPS-stimulated microglia, as manifested by reduced secretion of TNF, IL-6 and IL-1β and decreased inducible nitric oxide synthase (iNOS) expression [38,43]. In contrast, ENT-A010 did not alter the LPS-induced expression of pro-inflammatory genes, such as Il-1β, Tnf, Il-6, or iNos (data not shown).  A,B). Signal intensities of pTRKA, total TRKA, pAKT and total AKT were measured and in each experiment, the ratio of pTRKA/TRKA and pAKT/AKT was set as 1 for Ctrl samples. (B) shows the quantification of pAKT/AKT at 60 min of treatment. (C,D) Primary microglia were treated with 1 µM ENT-A010 on two consecutive days, and 1 h after the second treatment, they were stimulated with 100 ng/mL LPS. In (D), cells were pre-treated with the AKT inhibitor MK2206 (2.5 µM) 30 min prior to the first ENT-A010 treatment. Twenty-four h after the LPS treatment, fluorescently labeled Aβ was applied for 2 h to the cells, followed by flow cytometry. Data are shown as mean ± SEM, n = 4 for (A), n = 4 for (B), n = 6 (C), n = 4-10 (D), *: p < 0.05, **: p < 0.01, ns: non-significant.

ENT-A010 Promotes a Protective Microglial Phenotype
Next, we examined whether ENT-A010 modulates the expression of Triggering Receptor Expressed On Myeloid Cells 2 (Trem2) and MER Proto-Oncogene, Tyrosine Kinase (Mertk), which play a key role in phagocytosis [57][58][59][60]. ENT-A010 increased Trem2 and Mertk expression in LPS-treated primary microglia ( Figure 7A,B), standing in accordance with its effect on phagocytosis ( Figure 6C,D). Moreover, it increased the expression of Ngf ( Figure 7C), suggesting that ENT-A010-treated microglia may display enhanced neuroprotective function. In addition, ENT-A010 treatment enhanced the expression of Arginase 1 (Arg1) in LPS-treated microglia, indicating that it may modulate arginine metabolism and promote an M2-like microglial phenotype ( Figure 7D) [38]. Finally, ENT-A010-treated microglia showed a tendency for increased Cx3cr1 expression, suggesting that ENT-A010 might restore the expression of homeostatic genes in inflammatory microglia ( Figure 7E). We previously showed that TRKA activation by NGF or DHEA restrains inflammatory activation of LPS-stimulated microglia, as manifested by reduced secretion of TNF, IL-6 and IL-1β and decreased inducible nitric oxide synthase (iNOS) expression [38,43]. In contrast, ENT-A010 did not alter the LPS-induced expression of pro-inflammatory genes, such as Il-1β, Tnf, Il-6, or iNos (data not shown).

Peripherally Administered ENT-A010 Is Detected in the Brain
Next, we set out to investigate the in vivo effects of ENT-A010. To assess whether peripherally administered ENT-A010 can reach the brain, mice were injected i.p. with 70 mg/kg ENT-A010 or control solution, and 1 or 2 h later, different brain regions (brainstem, frontal cortex, hypothalamus, hippocampus, cerebellum), livers and spleens were isolated and analyzed with UHPLC-MS. ENT-A010 was detected at both time points in all examined brain regions, as well as the liver and the spleen (Figure 8).

Peripherally Administered ENT-A010 Is Detected in the Brain
Next, we set out to investigate the in vivo effects of ENT-A010. To assess whether peripherally administered ENT-A010 can reach the brain, mice were injected i.p. with 70 mg/kg ENT-A010 or control solution, and 1 or 2 h later, different brain regions (brainstem, frontal cortex, hypothalamus, hippocampus, cerebellum), livers and spleens were isolated and analyzed with UHPLC-MS. ENT-A010 was detected at both time points in all examined brain regions, as well as the liver and the spleen (Figure 8). olecules 2022, 12, x FOR PEER REVIEW Figure 8. Peripherally administered ENT-A010 is detected in the brain. Mice were 70 mg/kg ENT-A010 and 1 and 2 h later indicated brain regions, the liver and the lected, snap-frozen and analyzed by UHPLC-MS. Control mice were injected with of carrier solution (4.5% ethanol, 1% BSA, PBS). Data are shown as mean ± SEM condition, *: adj p < 0.05 **: adj p < 0.01 ***: adj p < 0.005 ****: adj p < 0.0001.

ENT-A010 Preserves the Homeostatic Phenotype of Microglia in the Hippoc
Next, we asked whether peripherally administered ENT-A010 may the brain. To this end, mice received ENT-A010 i.p. on two consecutive da the second dose, they were treated i.p. with LPS for 16 h. We focused o ENT-A010 in the hippocampus, since the hippocampus is strongly affecte inflammation and prone to neurodegeneration [61][62][63], and asked whethe fected the phenotype of hippocampal microglia. Staining of brain sections hippocampal formation) from mice treated with ENT-A010, LPS or ENTthe microglial marker ionized calcium-binding adaptor molecule 1 (IBA-1 that LPS treatment induced thickening of the branches and enlargement of microglia, standing in accordance with previous studies [64,65]. These m were blunted in the mice, which were pretreated with ENT-A010 ( Figure  LPS   . Peripherally administered ENT-A010 is detected in the brain. Mice were i.p. injected with 70 mg/kg ENT-A010 and 1 and 2 h later indicated brain regions, the liver and the spleen were collected, snap-frozen and analyzed by UHPLC-MS. Control mice were injected with the same amount of carrier solution (4.5% ethanol, 1% BSA, PBS). Data are shown as mean ± SEM, n = 3 mice per condition, *: adj p < 0.05 **: adj p < 0.01 ***: adj p < 0.005 ****: adj p < 0.0001.

ENT-A010 Preserves the Homeostatic Phenotype of Microglia in the Hippocampus
Next, we asked whether peripherally administered ENT-A010 may exert effects on the brain. To this end, mice received ENT-A010 i.p. on two consecutive days and 1 h after the second dose, they were treated i.p. with LPS for 16 h. We focused on the effects of ENT-A010 in the hippocampus, since the hippocampus is strongly affected by peripheral inflammation and prone to neurodegeneration [61][62][63], and asked whether ENT-A010 affected the phenotype of hippocampal microglia. Staining of brain sections (including the hippocampal formation) from mice treated with ENT-A010, LPS or ENT-A010 + LPS for the microglial marker ionized calcium-binding adaptor molecule 1 (IBA-1) demonstrated that LPS treatment induced thickening of the branches and enlargement of the cell body of microglia, standing in accordance with previous studies [64,65]. These microglial traits were blunted in the mice, which were pretreated with ENT-A010 ( Figure 9A). Moreover, LPS decreased homeostatic gene expression signature, exemplified by reduced expression of Trem2, Transforming growth factor beta receptor 1 (Tgfbr1), G Protein-Coupled Receptor 34 (Gpr34) and Transmembrane Protein 119 (Tmem119) (Figure 7B-E) [3,66]. Essentially, ENT-A010 partially restored the expression of these genes, suggesting that it may preserve the homeostatic signature of microglia in vivo ( Figure 9B-E). In addition, it increased the expression of genes associated with the M2-like microglial phenotype, such as Chitinaselike protein 3 (Chil3), Resistin-like alpha (Retnla) and Arginase 1 (Arg1) (Figure 9F-H). In contrast, ENT-A010 treatment did not alter the expression of pro-inflammatory genes, such as Il-1β, iNos, Il-6, Hexokinase 2 (Hk2) or Hypoxia-inducible factor 1α (Hif1α) (Figure 9I-M). Similarly, ENT-A010 did not affect Il-1β, Il-6, Tnf and iNos expression in livers and spleens in LPS-treated mice (Supplementary Figure S2).  Hippocampi were isolated from mice treated as described in (A) and the whole RNA was analyzed by real-time PCR for indicated genes, using 18S as a housekeeping gene. Fold change of relative gene expression was calculated based on the gene expression in the 'LPS' samples. Data are shown as mean ± SEM, n = 3-10 mice, *: p < 0.05, **: p < 0.01, ns: non-significant.

Discussion
Despite the significant advancement in deciphering the pathophysiology of neurodegenerative diseases, their treatment remains extremely challenging [67][68][69]. Neurotrophins, such as NGF, can prevent or reverse neurodegeneration, increase neurite outgrowth and promote synaptic plasticity [70,71]. Moreover, we and others have shown that NGF can exert anti-inflammatory effects and promote Aβ clearance in microglia [42,43]. Therefore, activation of NGF signaling could have the potential for the treatment of neurodegenerative diseases. However, being a large-sized protein, NGF has low stability in the circulation, negligible blood-brain barrier penetration and little diffusion within the CNS parenchyma [70][71][72][73][74], while it may cause hyperalgesia [75,76], thus limiting its pharmacological use.
Along the same line, we now synthesized ENT-A010 by replacing the metabolically labile oxirane ring of BNN-27 with the more stable cyclopropane moiety. ENT-A010 was selected from a panel of newly synthesized DHEA analogs (manuscript in preparation), based on its potential to induce TRKA phosphorylation. We show that ENT-A010 increased cell survival at comparable levels to NGF in serum-starved PC12 cells and NGF-starved DRG neurons. It also reduced apoptosis and restored Synaptophysin expression in Aβ-treated hippocampal neurons. Moreover, it increased Aβ uptake, favored microglial homeostatic signature, and increased Ngf expression in primary microglia. Both its neuroprotective effect and its effect on Aβ engulfment in microglia were blunted by TRKA inhibition.
Next, we examined the potential of peripherally given ENT-A010 to target brain tissues. ENT-A010 administered intraperitoneally was detected one and two hours postinjection in all tested brain regions (brainstem, frontal cortex, hypothalamus, hippocampus, cerebellum). Its brain levels were similar to those in the liver, signifying its efficient uptake in the brain compared to peripheral organs. Moreover, systemically administered ENT-A010 affected the morphology and transcriptional profile of hippocampal microglia in LPS-treated mice. In the experiments described here, we employed a previously used experimental setting, in which DHEA given prior to LPS restrained microglia-mediated neuroinflammation [38].
In this LPS model, we did not observe any neuronal apoptosis in the hippocampus evidenced by the absence of cleaved caspase 3 staining and no changes in the expression of apoptosis-regulating genes, such as Bcl2 family genes, Casp3 and Apaf1 (data not shown); hence we did not study the neuroprotective effects of ENT-A010 in this model. AD, PD, or other neurodegenerative disease models should be used to delineate the neuroprotective role of ENT-A010 in vivo. Moreover, while pre-treatment relates to prevention, studies involving ENT-A010 post-treatment following an insult or disease onset will be important to examine the clinical relevance of these findings.
Single-cell RNA analyses recently revealed a subset of microglia called diseaseassociated microglia (DAM) in AD, amyotrophic lateral sclerosis, multiple sclerosis and aging or models thereof [84][85][86]. DAM is featured by downregulated expression of homeo-static genes, such as P2ry12, Tmem119 and Cx3cr1, and enhanced expression of phagocytic genes, including Trem2 [3,84,85,87]. The current notion is that the clearing function of microglia plays a protective role in AD [86]. TREM2 mediates Aβ engulfment by microglia [57,58], is required for DAM activation [84] and limits AD progression [87], while a TREM2 mutation is associated with increased AD risk [88].
The present study demonstrates that LPS-induced acute inflammation decreased the expression of homeostatic genes and Trem2 in the hippocampus, standing in agreement with previous reports [89], while these effects were partially reversed by pretreatment with systemically applied ENT-A010. In accordance, ENT-A010 increased phagocytosis and Aβ clearance and enhanced Trem2 and Mertk expression in microglia in vitro. These findings suggest that ENT-A010 may promote protective features in microglia. In contrast to NGF or DHEA, which also induce TRKA phosphorylation and exert anti-inflammatory effects in microglia [37,38,42,43], ENT-A010 did not affect Il-1β, Il-6, Tnf, iNos, or Hk2 expression in LPS-treated microglia, which could be attributed to potential activation of different TRKA-dependent or independent signaling pathways by ENT-A010.
In conclusion, we developed a C17-spiro-cyclopropyl-DHEA synthetic analogue capable of inducing TRKA phosphorylation and AKT signaling, reaching all tested brain regions when applied peripherally, promoting neuronal survival, enhancing microglial phagocytic capacity and shifting hippocampal microglia towards a homeostatic state under inflammatory conditions. These traits could render ENT-A010 an interesting candidate for the treatment of neurodegenerative conditions, in which neuronal survival and microglial clearance capacity are compromised and microglial homeostasis is disrupted. To this end, a detailed exploration of the effects of ENT-A010 in animal models mimicking neurodegenerative diseases is essential. These research outcomes will shed light on the role of neuroactive steroid and neurotrophin mimetics on neurodegenerative and inflammatory aspects of CNS diseases.

Institutional Review Board Statement:
The study was approved by the Landesdirektion, Dresden, Germany (protocol number TVV57/2018, approval 13 December 2018).

Informed Consent Statement: Not applicable.
Data Availability Statement: The data supporting the findings of this study are available upon request.