Repurposing HAMI3379 to Block GPR17 and Promote Rodent and Human Oligodendrocyte Differentiation

SUMMARY

Identification of additional uses for existing drugs is a hot topic in drug discovery and a viable alternative to de novo drug development. HAMI3379 is known as an antagonist of the cysteinyl-leukotriene CysLT₂ receptor, and was initially developed to treat cardiovascular and inflammatory disorders. In our study we identified HAMI3379 as an antagonist of the orphan G protein-coupled receptor GPR17. HAMI3379 inhibits signaling of recombinant human, rat, and mouse GPR17 across various cellular backgrounds, and of endogenous GPR17 in primary rodent oligodendrocytes. GPR17 blockade by HAMI3379 enhanced maturation of primary rat and mouse oligodendrocytes, but was without effect in oligodendrocytes from GPR17 knockout mice. In human oligodendrocytes prepared from inducible pluripotent stem cells, GPR17 is expressed and its activation impaired oligodendrocyte differentiation. HAMI3379, conversely, efficiently favored human oligodendrocyte differentiation. We propose that HAMI3379 holds promise for pharmacological exploitation of orphan GPR17 to enhance regenerative strategies for the promotion of remyelination in patients.

Graphical Abstract

In Brief

Identification of alternative uses for existing drugs is a hot topic in drug discovery. Merten et al. repurposed the experimental drug HAMI3379, originally developed to treat cardiovascular and inflammatory disorders, for pharmacological exploitation of orphan GPR17, and thereby enhance regenerative strategies for promotion of remyelination in patients.

INTRODUCTION

Myelin is a lipid-rich substance that surrounds the axon of nerve cells. In the CNS, myelin is made by oligodendrocytes, specialized cells that extend and modify their plasma membrane to wrap nerve axons in a spiral fashion. Myelin sheaths are essential for rapid propagation of action potentials but also provide metabolic and trophic factor support to stabilize and protect neuronal axons, thereby preventing axon degeneration (Funfschilling et al., 2012; Lee et al., 2012). Importance of myelin-forming oligodendrocytes is evident from diseases such as multiple sclerosis (MS) and traumatic CNS injury, in which demyelinated lesions cause impaired neural transmission and ultimately axonal loss. In MS, demyelination triggered by episodic autoimmune destruction of oligodendrocytes is followed by remyelination, an inherent attempt of the brain to recreate functional myelin sheaths (Duncan et al., 2017; Patrikios et al., 2006). Essential to this process are oligodendrocyte precursor cells (OPCs) distributed throughout the adult CNS, which maintain their potential to proliferate, migrate, and differentiate into mature oligodendrocytes. Notably, remyelination in MS is often incomplete, despite the presence of substantial numbers of OPCs in demyelinated regions (Boyd et al., 2013; Kuhlmann et al., 2008), indicating that failure of OPC differentiation and myelination, rather than recruitment and migration, is the principal barrier impeding remyelination. Consequently, strategies to overcome OPC differentiation failure to foster development of myelinating oligodendrocytes are of enormous clinical importance (Plemel et al., 2017).

Recent high-throughput phenotypic screenings conducted on rodent oligodendrocytes have identified a number of (experimental) drugs that enhance remyelination (Deshmukh et al., 2013; Mei et al., 2014; Najm et al., 2015). However, potential clinical use of these compounds is limited by serious side and/or off-target effects, explaining the need to search for both new cellular targets and alternative bioactive molecules. Extensive research has focused on identification and characterization of signal transduction cascades that orchestrate the complex process of (re)myelination. Among these are growth factor receptors and G protein-coupled receptors (GPCRs) (Nadeem et al., 2015), the latter only recently assigned key roles in regulating glial cell biology (Mogha et al., 2016). GPR17, an orphan GPCR, has attracted particular attention as an oligodendroglial maturation inhibitor, because knockout and transgenic mouse models have revealed intriguing phenotypes: overexpression of GPR17 in the oligodendrocyte lineage causes severe defects in myelinogenesis whereas GPR17 knockout is associated with precocious myelination at the neonatal stage (Chen et al., 2009).

Despite the absence of primary endogenous messengers for GPR17, synthetic ligands have become available and have aided in defining the biological function of this receptor. MDL29,951 (MDL), a small-molecule surrogate agonist, and pranlukast, an anti-asthma medicine and cysteinyl-leukotriene 1 receptor blocker, activate and inhibit GPR17, respectively, both in heterologous cell expression systems and in primary rodent oligodendrocyte cultures (Hennen et al., 2013; Simon et al., 2016, 2017). Stimulation with MDL arrests primary wild-type, but not GPR17-deficient, mouse oligodendrocytes at a less differentiated stage, resulting in a pronounced loss of myelin basic protein (MBP)-positive cells (Hennen et al., 2013). Vice versa, pharmacological inhibition of GPR17 with pranlukast promotes differentiation of primary mouse (Hennen et al., 2013) and rat (Ou et al., 2016) oligodendrocytes. Pranlukast even phenocopies the effect of GPR17 deletion in a lysolecithin model of focal demyelination because both GPR17 knockout and pranlukast-treated wild-type mice exhibit an earlier onset of remyelination (Ou et al., 2016). Hence, rodent preclinical data imply GPR17 inhibition as promising strategy for the treatment of human demyelinating diseases.

Drugs emerging from preclinical studies in rodent MS models are known for their poor record of success in human clinical trials (Baker and Amor, 2015) revealing a limitation of rodent models to accurately reflect human pathologies. Therefore, to bridge the translational gap it is essential to interrogate the biological role of GPR17 and its ligands in both rodent and human oligodendrocytes. In the present study, by utilizing specific GPR17 ligands, we address the role of this receptor for survival and differentiation of human oligodendrocytes that were derived from human induced pluripotent stem cells (hiPSCs). In addition, we take advantage of the GPR17 surrogate agonist MDL to identify HAMI3379 (HAMI), a cysteinyl-leukotriene 2 receptor antagonist, as a novel scaffold that attenuates the function of human, rat, and mouse GPR17. HAMI is an effective blocker of GPR17 in recombinant mammalian expression systems and primary cells, favoring differentiation of rodent and human oligodendrocytes. GPR17 activity of HAMI opens up an opportunity to repurpose this experimental drug for an indication that could not have been predicted, thereby providing an additional scaffold to be evaluated further as a remyelinating agent for potential use in humans.

RESULTS

hiPSC-Derived Oligodendrocytes Express GPR17 and Treatment with MDL29,951 Impairs Their Maturation and Survival

To explore the translational value and predictive power of rodent GPR17 findings, it is essential to understand whether human oligodendrocytes also express this receptor and respond to GPR17 ligands in a manner akin to that of rodent oligodendrocytes. Two transcripts of GPR17 have been identified in human brain: a short and a long isoform, generated by alternative splicing and differing in length by 28 amino acids at their NH₂ termini (Blasius et al., 1998; Pugliese et al., 2009). To characterize expression of both GPR17 isoforms in human oligodendrocytes, we derived these from hiPSCs using a previously established protocol (Gorris et al., 2015). We initially converted hiPSCs into radial glia-like neural precursor cells (RGL-NPCs), which, under defined differentiation conditions, retain the ability for tripotential differentiation into neurons, astrocytes, and oligodendrocytes (Gorris et al., 2015). RT-PCR analysis of RGL-NPC-derived neurons did not detect expression of both isoforms of GPR17 (Figure 1A). Aligned with this finding, no GPR17-positive immunostaining was identified (Figure 1B). Similarly, we did not detect GPR17 transcripts and protein in RGL-NPC-derived astrocytes (Figures 1C and 1D), thus confirming the absence of GPR17 in these two cell types. Conversely, we found a stepwise induction of both GPR17 transcripts during differentiation of RGL-NPCs into oligodendrocytes (Figure 1E). Highest GPR17 abundance was detected in cultures during the initial stage of terminal differentiation (Gorris et al., 2015) when cells exhibited a branched phenotype and co-expressed the immature oligodendrocyte marker O4 (Figure 1F). After 6 weeks of differentiation, terminally differentiated oligodendrocytes, which displayed a more complex morphology and expressed the mature myelin marker MBP, revealed diminished GPR17 expression (Figure 1E). To investigate whether activation of GPR17 prevents maturation of human oligodendrocytes, we differentiated the oligodendrocyte cultures in the presence or absence of the small-molecule GPR17 agonist MDL (Hennen et al., 2013; Simon et al., 2017). Immunocytochemical analyses showed that cells expressing the oligodendrocyte marker O4 were also immunopositive for MBP under control conditions, thus corroborating the capacity for myelin expression in our hiPSC-derived oligodendrocyte culture (Figure 1G). As anticipated, treatment with MDL approximately halved the amount of O4-positive cells co-expressing MBP (Figures 1H, 1I, and S1). Moreover, increasing concentrations of MDL further diminished the number of O4-positive oligodendrocytes (Figures 1J and S1), in agreement with recent findings demonstrating that GPR17 activation impairs rodent oligodendrocyte survival (Ou et al., 2016). Taken together, our data provide evidence that human oligodendroglial cells express GPR17 and that its activation leads to arrest of the oligodendrocyte maturation program, consistent with the findings previously observed in rodents (Hennen et al., 2013; Ou et al., 2016; Simon et al., 2016).

Figure 1. GPR17 Is Expressed in hiPSC-Derived Oligodendrocytes, and Treatment with MDL29,951 Impairs Their Maturation and Survival.

(A and B) Neurons differentiated from hiPSC-derived RGL-NPCs exhibited expression of GAPDH and the pan-neuronal marker βIII-tubulin (TUBB3) but complete absence of GPR17 in RT-PCR analysis (A) and lack of GPR17 protein expression in cells immunopositive for βIII-tubulin in immunostaining analysis (B).

(C and D) Examination of RGL-NPC-derived astrocytes showed presence of transcripts for GAPDH and the astroglial marker GFAP but lack of GPR17 in RT-PCR analysis (C), confirmed by immunocytochemical staining displaying absence of GPR17 in GFAP-positive astrocytes (D).

(E) RT-PCR analysis of GPR17 transcript expression in hiPSC-derived RGL-NPCs at different stages of oligodendroglial differentiation characterized by GAPDH and stepwise induction of the oligodendrocyte transcription factor Olig2 and the myelination stage-specific marker MBP.

(F) Immunostaining analysis of human-derived oligodendrocytes displayed colocalization of GPR17 with the oligodendrocyte-specific marker O4.

(G and H) Detection of O4 and MBP in human oligodendrocytes after 6 weeks of terminal differentiation in control conditions (G) and after treatment with 3 μM MDL29,951 (H).

(I) Quantification of normalized MBP⁺/O4⁺ ratios detected in human oligodendrocytes differentiated in the absence and presence of 3 μM MDL29,951 from three independent experiments.

(J) Quantification of O4-positive oligodendrocytes counted after treatment with 3 μM MDL29,951 and normalized for each experiment to the numbers counted in control conditions, from three independent experiments.

(K) Illustrative immunocytochemical staining of O4 in human oligodendrocytes after 5 weeks of terminal differentiation in the absence and presence of 10 μM pranlukast.

(L) Concentration-effect profile of the inhibition of 5 μM forskolin-stimulated cAMP production triggered by MDL29,951 in hGPR17-CHO-FITR cells in the absence and presence of increasing concentrations of cangrelor and ticagrelor from three independent experiments performed in duplicates. The 100% level represents the activation of the receptor by 0.1 nM MDL29,951 and 0% represents the level of activation of the receptor by 0.1 μM MDL29,951.

Data are presented as mean ± SEM. Statistical significance was analyzed by two-tailed Student’s t test: *p < 0.05, ***p < 0.001. Scale bars, 20 μm (B, D, and F) and 50 μm (G, H, and K). See also Figure S1.

Because GPR17 agonism prevents differentiation of hiPSC-derived oligodendrocytes, GPR17 antagonism should promote their maturation and, consequently, qualify as an effective strategy for remyelination therapy in humans. So far, pharmacological GPR17 inhibition has only been achieved with the cysteinyl-leukotriene 1 (CysLT₁) receptor antagonist pranlukast both in heterologous expression systems and primary rodent oligodendrocytes (Hennen et al., 2013; Ou et al., 2016). Surprisingly, pranlukast was incompetent to facilitate human oligodendrocyte differentiation. Rather, akin to the results observed with MDL, it reduced the viable oligodendrocyte population and diminished the number of O4-positive cells (Figure 1K), thereby uncovering an unexpected detrimental effect on cell survival. Note that pranlukast was applied at a concentration sufficient to achieve full GPR17 inhibition, in agreement with previous reports by us and others (Hennen et al., 2013; Ou et al., 2016). P2Y12 receptor inhibitors ticagrelor and cangrelor, previously purported to block GPR17 (Ciana et al., 2006; Martini et al., 2010), did not suppress receptor function in stable hGPR17-Chinese hamster ovary (CHO) cells (Figure 1L). As a consequence, none of the above compounds qualified as a pharmacological probe to abrogate GPR17 signaling in human oligodendrocytes and to be evaluated further as a remyelinating agent for potential use in humans.

The CysLT₂ Receptor Antagonists HAMI3379 and BayCysLT₂ Block Human GPR17 Signaling

Because GPR17 is phylogenetically intermediate between nucleotide P2Y and cysteinyl-leukotriene receptors, we expanded our antagonist search to the cysteinyl-leukotriene 2 (CysLT₂) receptor inhibitors HAMI (Wunder et al., 2010) and BayCysLT₂ (Bay) (Ni et al., 2011), and also included leukotriene B4 receptor blocker LY255283 (Lee and Kim, 2013) (Figure 2A). We assessed potential antagonism on human (h) GPR17 by pretreating hGPR17-HEK293 cells with increasing concentrations of inhibitors and recorded integrated cell responses after challenge with MDL at its EC₈₀ using label-free dynamic mass redistribution (DMR) biosensor technology. HAMI (Figure 2B) and Bay (Figure 2C), but not LY255283 (Figure 2D), efficiently blunted hGPR17 signaling in a concentration-dependent manner (Figure 2E). HAMI and Bay displayed similar antagonistic properties in assays measuring Gα_q-mediated inositol phosphate accumulation (Figure 2F) and intracellular Ca²⁺ mobilization (Figure 2G). Comparable results were obtained when we measured the ability of both compounds–each ineffective when applied alone–to inhibit MDL-induced interaction of hGPR17 with β-arrestin2 in HEK293 cells stably expressing both proteins (Figures 2H and 2I). HAMI and Bay also antagonized MDL-stimulated label-free DMR responses and Ca²⁺ mobilization via GPR17 but not via muscarinic M1 receptors in two additional cellular backgrounds engineered to stably express the human ortholog: hGPR17-CHO and hGPR17-human astrocytoma 1321N1 cells (Figures 2J–2M and S2). Thus, both CysLT₂ antagonists efficiently and specifically dampened GPR17-promoted signaling events across cell lines.

Figure 2. Two CysLT₂ Receptor Antagonists, HAMI3379 and BayCysLT₂, Block Human GPR17 Signaling.

(A) Chemical structure of HAMI3379, BayCysLT₂ and LY255283.

(B–D) Representative traces of 1 μM MDL29,951 determined by label-free DMR assays in HEK293 cells overexpressing hGPR17 in the presence of increasing concentrations of HAMI3379 (B), BayCysLT₂ (C), and LY255283 (D). Label-free signatures are shown as representative traces (mean + SEM), measured in triplicates.

(E) Inhibition curves of tested compounds calculated from maximum DMR response depicted in (B) to (D) as mean values ± SEM from three to six independent experiments.

(F and G) Effect of increasing concentrations of HAMI3379 and BayCysLT₂ on MDL29,951-mediated intracellular inositol phosphate (IP1) accumulation (F) and intracellular calcium mobilization (G) in HEK293 cells overexpressing hGPR17 (n = 3).

(H) Influence of HAMI3379 and BayCysLT₂ on MDL29,951 concentration-effect curve in β-arrestin2 recruitment assays performed in HEK293 cells stably expressing hGPR17-Rluc and GFP2-β-arrestin2 (n = 5).

(I) Both inhibitors did not reveal agonism in the β-arrestin2 readout (n = 4).

(J–M) HAMI3379 and BayCysLT₂ inhibition of hGPR17 activation assessed by label-free technology (J and L) (n = 4–5) or intracellular calcium mobilization (K and M) (n = 3) in CHO (J and K) or 1321N1 (L and M) cells. Quantified data are shown as mean values ± SEM of n independent experiments. See also Figure S2.

HAMI3379 and BayCysLT₂ Inhibit Function of Rodent GPR17 Orthologs

Changes of ligand potency are known to occur among species orthologs of GPCRs, precluding direct conversion of ligand pharmacology from human to rodent receptors and animal models (Hudson et al., 2013). Important for subsequent studies, we therefore examined whether HAMI and Bay also block the function of rodent GPR17 orthologs. Both compounds suppressed MDL-triggered intracellular Ca²⁺ mobilization in HEK293 cells stably transfected with rat (r) and mouse (m) GPR17 in a concentration-dependent manner (Figures 3A–3D). We corroborated suppression of rGPR17 and mGPR17 function using inositol phosphate accumulation (Figures 3E and 3F) and label-free DMR assays (Figures 3G–3J) with MDL applied at its EC₈₀ across all assays. While EC₈₀ inhibition experiments show effective blockade of GPR17 function by HAMI and Bay, they do not provide insight into the compound’s mode of action. We therefore probed the molecular mechanism underlying HAMI- and Bay-mediated blockade of all three GPR17 orthologs using cyclic AMP (cAMP) accumulation assays, which capture activation of Gα_i family proteins and thus provide functional estimates of antagonist action on the preferred signaling pathway of GPR17. Defined concentrations of preadded HAMI (Figures 4A–4C) and Bay (Figures 4D–4F) rightward shifted the MDL concentration-effect curves without depressing the maximal response at all species orthologs. Schild regressions yielded data points that were falling on straight lines, the slopes of which were not significantly different from one (Figures 4G–4I). This mode of surmountable antagonism is compatible with competitive interaction between HAMI or Bay and the MDL binding site within the GPR17 binding pocket. Note that we do not refer to an orthosteric binding mode because GPR17 presently lacks confirmed endogenous ligands (Simon et al., 2017). In cells expressing hGPR17, increasing concentrations of HAMI but not Bay consistently elevated basal cAMP levels (Figures 4A and 4D, insets). This suggests that hGPR17 is active in the absence of stimulating MDL, and that HAMI acts as inverse agonist to reduce this MDL-independent intrinsic receptor signaling. No such effect was apparent for HAMI in m/rGPR17-expressing cells (Figures 4B and 4C, insets). Altogether, our data reveal that HAMI and Bay efficiently block human and rodent GPR17 signaling in recombinant mammalian expression systems, and thus qualify as pharmacological tools to interrogate GPR17 function in rodent and human oligodendrocyte physiology.

Figure 3. HAMI3379 and BayCysLT₂ Inhibit the Function of Rodent GPR17 Orthologs.

(A and B) Impact of increasing concentrations of HAMI3379 and BayCysLT₂ on the MDL29,951-mediated release of intracellular calcium in HEK293 cells stably expressing rat (A) or mouse (B) GPR17. Calcium flux signatures are shown as representative traces (mean + SEM) measured in triplicate.

(C–F) Antagonist inhibition curves derived from calcium flux (C and D) (n = 4) and IP1 accumulation assays (E and F) (n = 3–4) on rodent GPR17.

(G and H) Representative label-free traces of rat (G) and mouse (H) GPR17 function, shown as means + SEM of triplicate determination, in the absence or presence of HAMI3379 and BayCysLT₂.

(I and J) Concentration-effect relationships calculated from DMR recordings showed inhibition of rat (I) and mouse (J) receptor by high antagonist concentrations in whole-cell assay (n = 3–4).

Concentration-effect curves represent mean values ± SEM of n independent experiments performed in triplicate.

Figure 4. HAMI3379 and BayCysLT₂ Inhibit Human, Mouse, and Rat GPR17-Mediated Gα_i Signaling.

Influence of increasing concentrations of HAMI3379 (A–C) and BayCysLT₂ (D–F) on human (A and D), mouse (B and E), and rat (C and F) GPR17-induced reduction of 5 μM forskolin-stimulated cAMP production triggered by MDL29,951, with corresponding Schild regression plots (G–I). The 100% level represents the activation of the receptor by 0.1 nM MDL29,951 and 0% represents the level of activation of the receptor by 0.1 μM MDL29,951. Insets in (A) to (F) show the enlarged beginning of the cAMP curves for improved visualization. Concentration-effect curves are depicted as mean values ± SEM from three independent experiments performed in duplicates.

HAMI3379 and BayCysLT₂ Dampen GPR17 Activity in Primary Rat Oligodendrocytes

In agreement with previous observations, MDL triggered GPR17-mediated release of Ca²⁺ from intracellular stores in primary rat differentiating oligodendrocytes (Hennen et al., 2013). This effect was diminished when HAMI or Bay were pre-added to cells (Figures 5A and 5B). No inhibition was observed when ATP or carbachol were applied as stimuli for endogenous P2Y and muscarinic receptors, respectively (Figures 5C and 5D), confirming specificity of antagonist action in the Ca²⁺ mobilization assay. Furthermore, incubation with MDL triggered robust and concentration-dependent whole-cell responses in DMR assays (Figure 5E). Pretreatment with either HAMI (Figures 5F and 5G) or Bay (Figures 5H and 5I) reduced GPR17 signaling but was without effect when oligodendrocytes were activated with ATP (Figures 5J–5N). These data are in keeping with our previous findings demonstrating specific activation of endogenous rGPR17 by MDL in primary rat oligodendrocyte cultures (Hennen et al., 2013; Simon et al., 2016), and corroborate HAMI and Bay as inhibitors of rGPR17 activity in these cells.

Figure 5. HAMI3379 and BayCysLT₂ Dampen GPR17 Activity in Primary Rat Oligodendrocytes.

(A and B) Measure of MDL29,951-induced release of intracellular calcium in oligodendrocytes in the absence and presence of 10 μM antagonist over time (A) and concentration-effect relations derived from the traces (B) (n = 5).

(C and D) Both antagonists did not abolish calcium mobilization by purinergic and muscarinic receptors activated with ATP and carbachol, respectively. Data are depicted as mean + SEM of five independent experiments performed in triplicate.

(E–I) Representative recording of label-free responses in primary rat oligodendrocytes treated with the indicated concentrations of MDL29,951 in the absence of antagonist (E) and after preincubation with 30 μM HAMI3379 (F) or BayCysLT₂ (H). Concentration-response curves derived from DMR signals indicate GPR17 inhibition in whole-cell assay by HAMI3379 (G) and BayCysLT₂ (I) (n = 4–6).

(J–N) Label-free responses induced by activation of purinergic receptors with ATP are not influenced by both CysLT₂ antagonists as depicted in (J), (K), and (M) by representative traces and in the resulting concentration-effect curves (L and N) (n = 4).

Representative calcium and DMR traces are shown as means + SEM of triplicates (A, E, F, H, J, K, and M). Concentration-effect curves (B, G, I, L, and N) are presented as means ± SEM of n independent experiments performed in triplicate.

HAMI3379 Promotes Maturation of Rat and Human Oligodendrocytes

To test whether HAMI and Bay affect terminal differentiation of oligodendrocytes, we incubated rat oligodendrocytes with the thyroid hormone triiodothyronine (T₃), a known promoter of oligodendrocyte maturation, in the presence or absence of either compound, and quantified the expression of the myelination marker MBP. Western blot and densitometric analysis showed that treatment with increasing concentrations of Bay did not change the expression of MBP, whereas incubation with HAMI augmented MBP levels (Figure 6A). This finding seems counterintuitive given that both ligands block GPR17 signaling in primary rodent oligodendrocytes (Figure 5). We therefore examined whether variations in MBP abundance are related to oligodendroglial cell viability. However, no change in cell viability was found after treatment with either inhibitor (Figure 6B). Abundance of intracellular cAMP is a critical determinant of oligodendrocyte maturation (Simon et al., 2016; Syed et al., 2013). Therefore, we analyzed in more detail whether both compounds alter cAMP concentrations when GPR17 is expressed. HAMI was without effect on intracellular cAMP levels in stable rGPR17 transfectants (Figure 6C). Conversely, Bay decreased the abundance of cAMP in a concentration-dependent manner, indicative of weak partial agonism on rat GPR17 in addition to its inhibitory properties (compare also Figure 4F, inset). These data likely explain the capacity of HAMI but not Bay to enhance MBP expression in rat oligodendrocytes. HAMI enhancement of MBP expression was blunted by MDL (Figure 6D), indicating that HAMI promotes differentiation via GPR17. We also examined the effect of HAMI on oligodendroglial morphology in cultured primary oligodendrocytes from GPR17 heterozygous (GPR17^+/−) and null (GPR17^−/−) mice (Figure 6E) by evaluating cell process branching using Sholl analyses. As previously reported, GPR17^+/− mice have similar abundance of GPR17 compared with wild-type littermates but contain nuclear localized GFP in the GPR17 locus to facilitate cell tracking in differentiation studies (Chen et al., 2009). Quantification of process branching in GPR17^+/− oligodendrocytes revealed augmentation by HAMI of intersections per given radius (Figure 6F, upper panel), and increased complexity of the oligodendrocyte process meshwork (Figure 6G, upper panel). In agreement with a role for GPR17 as inhibitor of oligodendrocyte differentiation, overall elaboration and branching was more efficient in cultures from GPR17^−/− compared with those from GPR17^+/− mice (compare Figures 6F and 6G). HAMI did not affect process complexity in GPR17^−/− oligodendrocytes (Figures 6F and 6G, lower panels), corroborating its specific action via GPR17.

Figure 6. HAMI3379 Promotes the Maturation of Rat and Human Oligodendrocytes.

(A) Illustrative western blot of MBP abundance (upper panel) showing that primary rat differentiating oligodendrocytes treated for 48–72 hr with HAMI3379, but not with BayCysLT₂, in the presence of 0.20 nM triiodothyronine (T₃) expressed higher MBP levels compared with untreated cells. Quantitative analysis of MBP-immunoreactive band corrected by β-actin from seven independent experiments (lower panel) showed that treatment with HAMI3379 significantly increased MBP expression in primary rat oligodendrocytes. Data are presented as mean + SEM. Statistical significance was analyzed by two-tailed Student’s t test, *p < 0.05.

(B) In contrast to the topoisomerase II inhibitor etoposide, both antagonists did not exhibit any cytotoxic effect on rat oligodendrocytes after 48 hr of treatment (n = 4–5). Error bars denote +SEM.

(C) Effect of increasing concentrations of HAMI3379 and BayCysLT₂ on intracellular cAMP levels upon stimulation with forskolin in CHO cells overexpressing rat GPR17. The 100% level represents the activation of the receptor at 0.1 μM HAMI3379 and 0% the level of activation of the receptor at 0.1 μM MDL29,951, in the presence of 0.3 μM forskolin. Data are depicted as mean ± SEM of three independent experiments.

(D) Representative western blot analyzing MBP expression in primary rat differentiating oligodendrocytes cultured for 48–72 hr with T₃ and HAMI3379 in the absence or presence of MDL29,951 (upper panel). Evaluation of MBP levels in rat oligodendrocytes show that MDL29,951 counteracted the HAMI3379-mediated enhancement of oligodendrocyte maturation (lower panel). Data are presented as mean + SEM from four independent experiments.

(E) OPCs isolated from GPR17^+/− and GPR17^−/− mice were cultured in the absence or presence of HAMI3379 for 3 days and stained for MBP (red) and GFP to identify nuclei (green). Scale bars, 60 μm.

(F and G) Sholl analyses of oligodendrocyte morphology. Treatment with HAMI3379 increased branching in oligodendrocytes from GPR17^+/−, but not from GPR17^−/−, mice (F). Analysis of process complexity (G) revealed that GPR17^+/− mouse oligodendrocytes treated with HAMI3379 displayed increased elaboration compared with oligodendrocytes cultured alone. No effect of HAMI3379 was observed in oligodendrocytes from GPR17^−/− mice (n = 22 cells/condition). Data are presented as mean + SD of three independent experiments. Statistical significance was analyzed by unpaired two-tailed Student’s t test, *p < 0.05.

(H) Representative micrographs of oligodendrocytes differentiated from hiPSC-derived RGL-NPCs after 5 weeks of terminal differentiation cultured in the absence or presence of HAMI3379 and subsequently stained for O4 and MBP. Scale bars, 50 μm.

(I) Quantification of normalized MBP⁺/O4⁺ ratios showed that HAMI3379 treatment efficiently increased the amount of MBP-positive human oligodendrocytes (mean + SEM; n = 3 independent experiments; two-tailed t test, *p < 0.05, **p < 0.01).

(J) Quantification of O4-positive human oligodendrocytes counted after treatment with increasing concentrations of HAMI3399 and normalized for each experiment to the numbers calculated in control conditions. Data are presented as mean + SEM of three independent experiments. See also Figure S3.

These features qualify HAMI to probe the biological role of GPR17 in hiPSC-derived oligodendrocytes. We treated RGL-NPC-derived OPCs with HAMI upon growth factor withdrawal and subsequently immunostained for O4 and MBP (Figures 6H and S3). Vehicle-treated control oligodendrocytes exhibited a ramified morphology, expressed O4, and initiated the expression of MBP. Exposure to HAMI increased the ratio of MBP- to O4-positive cells (Figure 6I) without affecting total O4-positive cell number (Figure 6J), indicating that antagonism of hGPR17 also accelerates differentiation of human oligodendroglial cells in vitro. Altogether, these data identify HAMI as alternative antagonist of GPR17 and likely explain how this molecule exerts control over rodent and human oligodendrocyte differentiation.

DISCUSSION

Myelination during adulthood is a process that is dynamically regulated. In adult rodents, new myelin sheaths are formed by oligodendrocytes newly differentiated from proliferating OPCs, which in turn were previously generated both by differentiation of neural progenitor cells in the subventricular zone and by existing OPCs (Gonzalez-Perez and Alvarez-Buylla, 2011). Therefore, a reservoir of OPCs is retained in the brain, which in rats is estimated to represent 5% of the cells of adult brain (Dawson et al., 2003). In humans, the notion that recently generated oligodendrocytes contribute to myelination as in rodents is favored by remyelination studies in which (1) newly generated oligodendrocytes engage in myelination in nonhuman primates (Yang et al., 2006) and (2) human OPCs generate oligodendrocytes that contribute to remyelination when transplanted into rodents (Dietz et al., 2016). Conversely, a recent report has shown that the oligodendrocyte population in humans is much more static than in rodents, with an oligodendrocyte turnover 100-fold lower than in mice, thus suggesting that remyelination in humans would be achieved with previously existing oligodendrocytes rather than oligodendrocytes originating from proliferating OPCs (Yeung et al., 2014). This possible discrepancy between rodents and humans might puzzle attempts to discover effective remyelinating drugs for clinical use in humans, especially those that try to promote OPC proliferation. Indeed, potential drugs emerging from studies in rodent models of MS have a poor record of success in clinical trials to date (Baker and Amor, 2015).

As a consequence, focus has recently shifted to discovering drugs that foster OPC differentiation and maturation. In this context, antagonism of the orphan GPCR GPR17 has been proposed as novel route for therapeutic remyelination. In rodents, GPR17 is expressed in immature non-myelinating oligodendrocytes (Fumagalli et al., 2011), where it acts as a negative regulator of oligodendrocyte maturation during postnatal development (Chen et al., 2009; Hennen et al., 2013). Because GPR17 expression is highly abundant in active white matter plaques of MS patients (Chen et al., 2009), it has been suggested that upregulation of GPR17 in demyelinating lesions is correlated with remyelination arrest. Increased GPR17 levels are also observed in adult mouse models of MS, such as experimental autoimmune encephalomyelitis (Chen et al., 2009) and lysophosphatidylcholine (LPC)-induced demyelination (Ou et al., 2016). Notably, pharmacological inhibition of GPR17 with pranlukast (Hennen et al., 2013) promoted generation of new myelin upon LPC-mediated demyelination injury in mice (Ou et al., 2016), thereby highlighting the potential for GPR17-targeted therapeutic approaches in human demyelinating diseases.

Because the likelihood of success for a drug candidate in clinical trials depends on whether drug behavior in rodent model systems is predictive for modulation of human remyelination, a first step to address this issue is to investigate in vitro whether human OPC differentiation is regulated similarly to that in rodents. Herein, we took advantage of the human precursor cells RGL-NPCs as a platform for generation of human oligodendrocytes. Under defined differentiation conditions, RGL-NPCs convert into OPCs expressing the major oligodendrocyte fate determinants, such as Olig1/2, Nkk6.2, Nkk2.2, Sox10, and NG2, which are subsequently capable of differentiating into myelinating oligodendrocytes (Gorris et al., 2015). We observed that GPR17 was exclusively expressed in human oligodendrocytes in a developmentally regulated manner, since its expression was increased during the initial steps of differentiation to then decline when myelin gene MBP expression was initiated. Moreover, activation of GPR17 with MDL inhibited terminal differentiation of human oligodendrocytes and compromised their survival. Therefore, our data demonstrate that also in humans GPR17 is restricted to oligodendrocyte-lineage cells and functions to negatively regulate oligodendrocyte differentiation and myelination. Hence, no disparity between rodent and human physiology is apparent that could deter us from pursuing GPR17 as a druggable target for remyelination in humans.

Agonists are employed to identify antagonist ligands that provide starting points for studying receptor pharmacology and physiology. By utilizing MDL, we have previously shown that pranlukast, an anti-asthma agent originally described as a CysLT₁ receptor antagonist, fully reversed the agonist action of MDL in cells heterologously expressing GPR17 as well as in primary rat oligodendrocytes (Hennen et al., 2013). Conversely, the leukotriene receptor antagonist montelukast, previously proposed to inhibit GPR17 by others (Ciana et al., 2006), does not antagonize MDL-induced GPR17 function (Hennen et al., 2013). The same applies to cangrelor and ticagrelor, two antiplatelet medicines also described to abrogate GPR17 signaling (Ciana et al., 2006; Martini et al., 2010). Therefore, pharmacological inhibition by pranlukast was the only means to investigate the therapeutic potential of GPR17 inhibition in human oligodendrocytes. However, the deleterious effect of pranlukast on RGL-NPC-derived oligodendrocyte survival compelled us to search for new GPR17 antagonists with consistent in vitro profiles across species. Here, we repurposed HAMI as alternative inhibitor for GPR17 and propose that GPR17 blockage by HAMI provides a molecular explanation as to how this molecule might exert control over rodent and human oligodendrocyte differentiation. Clearly, pharmacological inhibition of GPR17 by HAMI for therapeutic remyelination in humans merits further investigation and provides an alternative application for this experimental drug.

STAR★METHODS

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Evi Kostenis (kostenis@uni-bonn.de).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Cell Lines

Cell lines were kept in a humidified atmosphere with 5% CO₂ at 37°C. All media contained 10% (v/v) fetal bovine serum (FBS, PAN biotech), penicillin (100 U/ml) and streptomycin (100 μg/ml). For culture of stable mGPR17-HEK293, rGPR17-HEK293 and hGPR17–1321N1 Dulbecco’s modified Eagle’s medium (DMEM) was supplemented with G418 (500 μg/ml or 800 μg/ml) (InvivoGen), for 3HA-hGPR17-HEK293 (later named hGPR17-HEK293) with zeocin (56 μg/ml) (InvivoGen), for 3HA-hGPR17-Rluc / GFP²-β-arrestin2-HEK293 with zeocin (56 μg/ml) and G418 (500 μg/ml).

Flp-In™ T-REx CHO cells stably expressing hGPR17 (hGPR17-CHO-FITR) were maintained in DMEM with nutrient mixture F-12 (DMEM/F12) supplemented with hygromycin B (500 μg/ml) and blasticidin (30 μg/ml). Expression from the Flp-In locus was induced by treatment with doxycycline (1 μg/ml) for 16–20 h. mGPR17-CHO-Flp-In and rGPR17-CHO-Flp-In were cultivated in Ham’s F-12 nutrient mix with GlutaMAX and 600 mg/ml hygromycin B. For hM1-CHO cells, G418 (200 μg/ml) was added to Ham’s F-12 nutrient mix with GlutaMAX.

METHOD DETAILS

Cell Culture of Primary Rat Oligodendrocytes

Primary rat OPCs were isolated by a differential detachment method from mixed glial cultures of postnatal Wistar rat cerebra, as previously described (Hennen et al., 2013; Simon et al., 2016). OPCs were seeded in proliferation medium (Neurobasal medium supplemented with 2% (v/v) B27, 2 mM GlutaMAX, 100 units/ml penicillin, 100 mg/ml streptomycin, 10 μg/ml platelet-derived growth factor-AA (PDGF-AA, PeproTech) and 10 ng/ml basic fibroblast growth factor (bFGF, PeproTech). For induction of spontaneous in vitro differentiation and GPR17 protein expression, medium was switched to growth factor-free Neurobasal medium. For terminal differentiation and quantification of myelin basic protein (MBP) expression, after 24 h the growth factor-free medium was supplemented with 0.20 ng/ml triiodothyronine (T₃) and 10 ng/ml ciliary neurotrophic factor (CNTF, PeproTech) together with the analyzed compounds for additional 2–3 days.

Cell Culture of Primary Mouse Oligodendrocytes

Primary mouse preoligodendrocytes (O4+) were purified by immunopanning from mouse cerebral cortices. Briefly, dissociated cortex cells of P7 GPR17^+/− or GPR17^−/− mice were sequentially panned on Ran2 and GalC panning plates to deplete astrocytes, microglia, and mature oligodendrocytes. Finally, preoligodendrocytes were enriched on the O4 panning plate. The purified preoligodendrocytes were seeded onto poly-L-ornithine–coated glass coverslips and maintained in proliferation medium [Sato medium with PDGF-AA (10 ng/ml)]. When OPCs reached 70 – 80% confluence, PDGF-AA was removed from the medium and cultures were treated with drugs for 3 days.

Differentiation of Oligodendrocytes, Neurons and Astrocytes from Human Radial Glia-like Neural Stem Cells

Cultures enriched for oligodendrocytes from hiPSC-derived radial glia-like neural precursor cells (RGL-NPCs) were performed in a three step culture paradigm according to a previously published protocol (Gorris et al., 2015). Briefly, 800,000 RGL-NPCs were seeded onto matrigel-coated 3.5 cm tissue culture dishes in N2-medium (DMEM/F12, 1% N2, 0.5 mg/ml apotransferrin, 0.06 mg/ml insulin and 1.6% D-glucose) containing 20 ng/ml FGF2 and 20 ng/ml EGF. After cells had reached about 80% confluence, differentiation along the oligodendrocyte lineage was started. First, OPC generation (stage I) was carried out for 2 weeks in N2-medium containing 10 ng/ml EGF, 10 ng/ml PDGF-AA and 4 mg/ml forskolin. After that OPCs were allowed to proliferate (stage II) for another week in the presence of 10 ng/ml PDGF-AA, 30 ng/ml T₃, 20 ng/ml noggin and 200 mM ascorbic acid. The first two steps were performed to specifically support the development and proliferation of oligodendroglial progenitors. To reduce the differentiation of astroglial cells, noggin was added to the culture medium during the second step. Finally, terminal differentiation (stage III) for up to 5 to 6 weeks was carried out in the presence of 30 ng/ml T₃, 200 mM ascorbic acid and 1 mg/ml laminin. At each differentiation stage, half of the medium was changed every 2 to 3 days. Compounds were added after three weeks of terminal differentiation (stage III) with a medium change every 3 days, followed by analysis of O4 and MBP expression by immunocytochemistry. For neuronal differentiation hiPSC-RGL-NPCs were replated on matrigel-coated tissue culture dishes in N2 medium without growth factors supplemented with 20 ng/ml brain-derived neurotrophic factor (BDNF; R&D Systems) and 200 mM ascorbic acid (Sigma-Aldrich) for 3 weeks. Astrocytic differentiation was carried out in N2 medium containing 5% fetal calf serum (Thermo Fisher Scientific) for at least 14 days.

Reverse Transcriptase-Polymerase Chain Reaction

Total RNA samples were isolated using the Quick-RNA™ MiniPrep Kit (Zymo Research) according to the manufacturer’s instructions. The RNA samples were treated with DNase I (Thermo Fisher Scientific), and cDNA was synthesized with the iScript™ cDNA synthesis Kit (BioRad). RT-PCR reactions were done on a T3 Thermocycler (BioRad) using the GoTaq® Flexi DNA Polymerase (Promega) according to the supplier’s instructions. The forward primers for the two GPR17 amplicons were isoform-specific (human GPR17 long: 5′GCTGAAACTCTCAGGCTCTGAC-3′, human GPR17 short: 5′CCAGCAGCTAGAGGCTCTGA-3′) while the reverse primer was shared (5′GCCAGGGTATTGCCAACTAA-3′) giving transcript lengths of 189 bp for long and 187 bp for short version of GPR17 (Benned-Jensen and Rosenkilde, 2010). Further primers for defining the identity of oligodendrocytes were Olig2 (5′-GCTGCGTCTCAAGATCAACAG-3′, 5′-CACCAGTCGCTTCATCTCCTC-3′; product 196 bp) and MBP (5′-CTATAAATCGGCTCACAAGG-3′, 5′-AGGCGGTTATATTAAGAAGC-3′; product 166 bp), for astrocytes GFAP (5′-TCATCGCTCAGGAGGTCCTT-3′, 5′-CTGTTGCCAGAGATGGAGGTT-3′; product 383 bp) and for neurons TUBB3 (5′-GCAACTACGTGGGCGACT-3′, 5′-GGCCTGAAGAGATGTCCAAA-3′; product 160 bp). GAPDH expression was detected with 5′-ATGACCCCTTCATTGACCTCAACT-3′ and 5′-ATACTTCTCATGGTTCACACCCAT-3′, product 320 bp. Amplification conditions were as follows: initial denaturation at 95°C for 2 min followed by 30 to 35 cycles of denaturation at 95°C for 30 s, annealing at 50 to 55°C for 30 s, extension for 30 s at 72°C and a final polymerization at 72°C for 10 min.

Immunocytochemical Analysis

Cultured cells were fixed with 4% paraformaldehyde solution (Sigma-Aldrich) in PBS for 10 min at room temperature. For the detection of intracellular antigens, cell membranes were permeabilized with 0.1% triton X-100 (Sigma-Aldrich) in PBS for 20 min at room temperature. Cells were subsequently blocked with blocking solution (5% NGS in PBS) for 20 min at room temperature. The primary antibodies (rabbit anti-GFAP, 1:1000, DAKO; rabbit anti-GPR17, 1:1000, Cayman; rat anti-MBP, 1:25, Abcam; mouse anti-O4, 1:100, R&D Systems; rabbit anti-βIII-Tubulin, 1:1000, Covance) were diluted in blocking solution and incubated for at least 2 h at room temperature or overnight at 4°C. The secondary antibodies (appropriate Alexa-488, Alexa-555-conjugated antibodies, Thermo Fisher Scientific) were diluted in blocking solution and incubated for 1 h at room temperature. DAPI (4′,6-diamidino-2-phenylindole, 1:10,000; Sigma-Aldrich) was used for nuclear counterstaining. Labeled cells were preserved in Vectashield (Vector Laboratories) and analyzed using an Apotome Observer Z.1 inverted fluorescence microscope (Zeiss) with ZEN software. To quantify the number of MBP- and O4-positive cells, 100 fields of vision of a 3.5 cm dish were recorded at 20× magnification, number of positive cells in this area was counted using ImageJ software and the MBP⁺/O4⁺ ratio or the counted O4⁺ cells of dimethyl sulfoxide (DMSO) control were set to 100%.

Sholl Analyses of Primary Mouse Oligodendrocytes

Analyses of oligodendrocyte morphology was performed using the Sholl plugin for ImageJ on 2D segmented grayscale images, drawing concentric circles 6 mm apart around the soma center of mature MBP-positive oligodendrocytes. Total number of intersections per oligodendrocyte was counted as well as the number of intersections made by processes with each successive circle.

Calcium Mobilization Assays

Intracellular calcium mobilization was quantified with the Calcium 5 Assay kit in conjunction with the FlexStation 3 Multimode Microplate Reader (Molecular Devices). Cells were seeded (HEK293 60,000 cells/well, CHO and 1321N1 50,000 cells/well, oligodendrocytes 7,000 cells/well with 4 days proliferation and 1–2 days of differentiation) into black 96-well tissue culture plates with clear bottoms (poly-D-lysine-coated for HEK293 cells, poly-L-ornithine-coated for oligodendrocytes). On the assay day, cells were loaded with the Calcium 5 indicator dye for 45 min (for oligodendrocytes 25 min already in the presence of antagonist), then incubated with antagonist for 30 min prior agonist application and measurement of intracellular calcium flux.

HTRF-Based cAMP Accumulation Assays

Changes of the intracellular second messenger cAMP in hGPR17-CHO-FITR, mGPR17-CHO-Flp-In and rGPR17-CHO-Flp-In were assessed with the HTRF-cAMP dynamic 2 kit (Cisbio International) using a Envision multilabel plate reader (Perkin Elmer). The assays were performed in 384-well plates with 5,000 cells per well in a final volume of 40 μl. To estimate potency of inhibitors, cells were preincubated with antagonists in assay buffer (Hanks’ buffered salt solution (HBSS) supplemented with 20 mM HEPES) for 30 min at ambient temperature. Then cells were stimulated with varying concentrations of MDL29,951 in the presence of 0.1 mM isobutylmethylxanthine (IBMX) and 5 μM forskolin for 60 min. The reactions were stopped by addition of cAMP-d2 reagent and anti-cAMP antibody, and plates were incubated for 60 min. The level of cAMP was then determined by measuring the fluorescence ratio (665 nm/620 nm). All incubations were performed in duplicates.

Label-Free Dynamic Mass Redistribution (DMR) Assays

DMR was recorded as described previously in detail (Schröder et al., 2010, 2011). Briefly, cells were seeded (HEK293 18,000 cells/well, CHO and 1321N1 15,000 cells/well, oligodendrocytes 3,000 cells/well with 3 days proliferation and 1 day of differentiation) in 384-well fibronectin-coated EPIC biosensor plates (Corning). On the assay day, cells were washed twice with HBSS containing 20 mM HEPES adjusted to final DMSO content and incubated for 1 hour at 28°C in the EPIC benchtop reader (Corning). The sensor plate was scanned for a baseline read for about 5 min, and then buffer or antagonists were added with a semi-automated liquid handling system (Selma, CyBio®). After a 1 h antagonist read, MDL29,951 was applied and DMR changes were recorded for 1–2 h.

β-Arrestin2 Recruitment Assays

For β-arrestin2 recruitment assays on HEK293 cells stably expressing 3HA-hGPR17-Rluc as energy donor and GFP²-β-arrestin2 as energy acceptor, cells were detached and resuspended in HBSS with 20 mM HEPES at a density of 1.09 × 10⁶ cells per ml. Cell suspension (165 μl) was distributed into white 96-well microplates and incubated in the presence of 5 μl buffer or antagonist for 30 min before incubation with varying concentrations of MDL29,951 for 5 min. Then DeepBlueC (coelenterazine 400a; Gold Biotechnology; 20 μl per well) was injected by injector 3 to yield a final concentration of 5 μM. To detect BRET, light emissions at 400 and 515 nm were measured sequentially using a Mithras LB 940 instrument. The BRET signal (BRET ratio) was determined by calculating the ratio of the light emitted by the fluorescence acceptor (515 nm) and the light emitted by Rluc (400 nm).

Western Blotting

OPCs were seeded in proliferation medium into poly-L-ornithine-coated 12-well tissue culture plates (80,000–100,000 cells per well). After induction of GPR17 expression and terminal differentiation of oligodendrocytes in the presence of analyzed compounds for 2–3 days, cells were washed twice with ice-cold PBS and lysed in ice-cold lysis buffer (25 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% IGEPAL) supplemented with protease inhibitor mixture (Sigma-Aldrich). Lysates were rotated for 20 min at 4°C and centrifuged at 13,200 rpm at 4°C for 10 min. Protein concentration was determined using the Pierce BCA Protein Assay (Thermo Fisher Scientific) according to manufacturer’s instructions. 7.5–15 μg of protein were separated by 10% NuPAGE bis-tris gel electrophoresis (Thermo Fisher Scientific) and transferred to nitrocellulose membrane (HybondTM-C Extra, GE Healthcare) by wet blotting (XCell II™ Blot Module, Thermo Fisher Scientific). After washing, membranes were blocked with Roti-Block (1×; Carl Roth) for 1 h at room temperature and incubated overnight at 4°C in Roti-Block with mouse anti-MBP antibody (1:5000, LifeSpan BioSciences) and rabbit anti-β-actin (1:2500, BioLegend) to normalize for equal loading and protein transfer. Membranes were washed 3 times with PBS containing 0.1% Tween and then incubated for 1 h at room temperature with corresponding horseradish peroxidase-conjugated secondary antibody (goat anti-rabbit IgG antibody HRP (1:10000, antikoerper-online), goat anti-mouse IgG antibody HRP (1:10000, Sigma-Aldrich)) in Roti-Block. The immunoreactive proteins were visualized by chemiluminescence using Amersham Biosciences ECL Prime Western blotting detection reagent (GE Healthcare) and quantified by densitometry using Gelscan software (Bioscitec).

Cell Viability Assays

Cell viability was assessed using a fluorimetric detection of resorufin formation (CellTiter-Blue Cell Viability Assay, Promega). OPCs were seeded in proliferation medium at a density of 7,000 cells per well into black 96-well poly-L-ornithine-coated plates with clear bottom. After induction of GPR17 expression by growth factor removal, cells were treated with 0.1% DMSO or indicated compounds diluted in medium containing 0.20 ng/ml T₃ and 10 ng/ml CNTF for additional 2 days. To detect cell viability, CellTiter-Blue reagent (20 μl) was added, and cells were incubated for 3 h at 37°C as per manufacturer’s instructions. Fluorescence (excitation 560 nm, emission 590 nm) was measured using a FlexStation 3 Multimode Microplate Reader (Molecular Devices), and data were expressed as the percentage of cell viability relative to DMSO control.

QUANTIFICATION AND STATISTICAL ANALYSIS

Pharmacological Validation

Data points were fitted to both three-parameter (fixed Hill slope) and four-parameter nonlinear regression isotherms using Prism 6.05 (GraphPad Software). Quantification of buffer corrected DMR signals was performed by calculation of the maximum response or the area under the curve (AUC) between at least 0 and 1200 s. Calcium responses were analyzed using the maximal peak fluorescence within 30 s after agonist addition. All data presented are mean +/± SEM. Statistical analyses were performed using two-tailed Student’s t test. P value significance thresholds were *P<0.05, **P<0.01 and ***P<0.001.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
rabbit anti-GFAP	DAKO	Cat# Z0334; RRID: AB_10013382
rabbit anti-GPR17	Cayman	Cat# 10136; RRID: AB_10613826
rat anti-MBP	Abcam	Cat# ab7349; RRID: AB_305869
mouse anti-O4	R&D Systems	Cat# MAB1326; RRID AB_357617
rabbit anti-βIII-Tubulin	Covance	Cat# MMS-435P; RRID: AB_2313773
mouse anti-MBP	LifeSpan BioSciences	Cat# LS-B8056; RRID: AB_10946075
rabbit anti-β-actin	BioLegend	Cat# 622102; RRID: AB_315946
goat anti-rabbit IgG antibody HRP	antikoerper-online	Cat# ABIN102010; RRID: AB_10762386
goat anti-mouse IgG antibody HRP	Sigma-Aldrich	Cat# A4416; RRID: AB_258167
Bacterial and Virus Strains
DH5α Competent Cells	Thermo Fisher Scientific	Cat# 18265-017
XL1-Blue Competent Cells	Stratagene	Cat# 200130
Chemicals, Peptides, and Recombinant Proteins
MDL29,951	Maybridge	Cat# SEW06645
Ticagrelor	A gift from Sven Nylander, AstraZeneca	N/A
Cangrelor	Cayman Chemical	Cat# 22086
Pranlukast	Cayman Chemical	Cat# 10008319
BayCysLT2	Cayman Chemical	Cat# 10532
HAMI3379	Bertin Pharma	Cat# 10580
LY255283	Cayman Chemical	Cat# 70715
DNase I	Thermo Fisher Scientific	Cat# EN0521
GoTaq® Flexi DNA Polymerase	Promega	Cat# M8291
DeepBlueC	Gold Biotechnology	Cat# C-320-10
DMEM	Thermo Fisher Scientific	Cat# 41965062
DMEM/F12	Thermo Fisher Scientific	Cat# 31330095
Ham’s F-12 nutrient mix with GlutaMAX	Thermo Fisher Scientific	Cat# 31765068
FBS	PAN biotech	Cat# P30-3702
G418	InvivoGen	Cat# ant-gn-5
zeocin	InvivoGen	Cat# ant-zn-1
hygromycin B	InvivoGen	Cat# ant-hg-5
blasticidin	InvivoGen	Cat# ant-bl-1
doxycycline	Sigma-Aldrich	Cat# D9891
Neurobasal medium	Thermo Fisher Scientific	Cat# 21103049
B27	Thermo Fisher Scientific	Cat# 17504044
GlutaMAX	Thermo Fisher Scientific	Cat# 35050038
PDGF-AA	PeproTech	Cat# 100–13A-50UG
bFGF	PeproTech	Cat# 100–18B-50UG
CNTF	PeproTech	Cat# 450-50-25UG
Triiodothyronine (T3)	Sigma-Aldrich	Cat# T2752
poly-L-ornithine	Sigma-Aldrich	Cat# P3655
poly-D-lysine	Sigma-Aldrich	Cat# P2636
Hanks’ buffered salt solution (HBSS)	Thermo Fisher Scientific	Cat# 14175129
isobutylmethylxanthine (IBMX)	Sigma-Aldrich	Cat# I5879
forskolin	Bachem	Cat# TRC-F701800
Critical Commercial Assays
Quick-RNA™ MiniPrep Kit	Zymo Research	Cat# R1050
iScript™ cDNA synthesis Kit	BioRad	Cat# 1708890
FLIPR® Calcium 5 Assay kit	Molecular Devices	Cat# R8186
HTRF-cAMP dynamic 2 kit	Cisbio International	Cat# 62AM4PEC
Pierce BCA Protein Assay	Thermo Fisher Scientific	Cat# 23225
ECL Prime Western blotting detection reagent	GE Healthcare	Cat# RPN2236
CellTiter-Blue Cell Viability Assay	Promega	Cat# G8080
Experimental Models: Cell Lines
HEK293	ATCC	Cat# CRL-1573; RRID: CVCL_0045
hGPR17–1321N1	Prof. Christa E. Müller lab	N/A
Flp-In™ T-REx CHO	Thermo Fisher Scientific	Cat# R71807
CHO-Flp-In	Thermo Fisher Scientific	Cat# R75807; RRID: CVCL_U424
CHO	ATCC	Cat# CCL-61; RRID: CVCL_0214
hiPSC-derived radial glia-like neural precursor cells (RGL-NPCs)	Gorris et al., 2015	N/A
Experimental Models: Organisms/Strains
Wistar rat	Charles river	Strain code: 003; RRID: SCR_003792
GPR17+/− and GPR17−/− C57B1/6 mice	Chen et al., 2009	N/A
Oligonucleotides
RT-PCR primers see Method Details
Recombinant DNA
mGPR17-pcDNA3.1 (+)	This paper	N/A
rGPR17-pcDNA3.1 (+)	This paper	N/A
hGPR17-pLXSN	This paper	N/A
3HA-hGPR17-pcDNA3.1/Zeo (+)	This paper	N/A
mGPR17-pcDNA5/FRT	This paper	N/A
rGPR17-pcDNA5/FRT	This paper	N/A
3HA-hGPR17-Rluc-pcDNA3.1/Zeo (+)	This paper	N/A
GFP2-β-arrestin2-pcDNA3.1 (+)	This paper	N/A
Software and Algorithms
Prism 6.05	GraphPad	https://www.graphpad.com/scientific-software/prism/
ZEN software	Zeiss	https://www.zeiss.de/mikroskopie/downloads/zen.html
ImageJ	National Institutes of Health	http://imagej.nih.gov/ij/
Other
black 96-well tissue culture plates with clear bottoms	Corning	Cat# 3603
white 384-well plates	Greiner bio-one	Cat# 784080
384-well fibronectin-coated EPIC biosensor plates	Corning	Cat# 5042
white 96-well microplates	Corning	Cat# 3917

Highlights.

• Experimental drug HAMI3379 repurposed as inhibitor of orphan GPR17

• HAMI3379 favors differentiation of rodent and human oligodendrocytes in culture

• Pro-differentiation effects of HAMI3379 are mediated by GPR17

• Alternative-application discovery for HAMI3379 as strategy to treat demyelinating diseases

SIGNIFICANCE.

There is a great need for new drug therapies that promote myelin repair and regeneration to treat demyelinating diseases, such as multiple sclerosis (MS). However, potential drugs emerging from studies in rodent models of MS have a poor record of success in clinical trials, indicating limitations of rodent models to accurately simulate human pathologies. Our study attempts to bridge this translational gap by adding human iPSC-derived oligodendrocytes to the battery of test systems for assessment of pharmacology and function of the oligodendrocyte-specific GPR17. GPR17 is an orphan G protein-coupled receptor (GPCR) that has attracted particular attention as key player in and inhibitor of oligodendrocyte differentiation in rodents. Consequently, experimental drugs that block GPR17 function across species might be expected to also enhance human oligodendrocyte differentiation and, eventually, to promote remyelination in MS. Our results highlight species-conserved expression and function of GPR17 in oligodendrocyte precursor cell development, thereby minimizing potential mismatches between rodent and human oligodendroglial physiology. Furthermore, we identify HAMI3379, previously developed to treat cardiovascular and inflammatory conditions, as an experimental drug that favors differentiation of rodent and human oligodendrocytes in culture. Because HAMI3379 exerts its effect via inhibition of GPR17, and because GPCRs belong to the most successful protein families in the history of drug development, alternative-application discovery for HAMI3379 may be an important step toward more rapidly bringing benefits to patients in the field of demyelinating diseases, an area of high unmet medical need.