Evaluation of Urea-Based Inhibitors of the Dopamine Transporter Using the Experimental Autoimmune Encephalomyelitis Model of Multiple Sclerosis

Abstract

The dopaminergic system is involved in the regulation of immune responses in various homeostatic and disease conditions. For conditions such as Parkinson’s disease and multiple sclerosis (MS), pharmacological modulation of dopamine (DA) system activity is thought to have therapeutic relevance, providing the basis for using dopaminergic agents as a treatment of relevant states. In particular, it was proposed that restoration of DA levels may inhibit neuroinflammation. We have recently reported a new class of dopamine transporter (DAT) inhibitors with high selectivity to the DAT over other G-protein coupled receptors tested. Here, we continue their evaluation as monoamine transporter inhibitors. Furthermore, we show that the urea-like DAT inhibitor (compound 5) has statistically significant anti-inflammatory effects and attenuates motor deficits and pain behaviors in the experimental autoimmune encephalomyelitis model mimicking clinical signs of MS. To the best of our knowledge, this is the first study reporting the beneficial effects of DAT inhibitor-based treatment in animals with induced autoimmune encephalomyelitis, and the observed results provide additional support to the model of DA-related neuroinflammation.

Graphical Abstract

INTRODUCTION

Dopaminergic signaling is involved in regulating central and peripheral immunity in various homeostatic and pathological conditions including Parkinson’s disease and multiple sclerosis (MS).^1-3 Dopamine (DA) is one of the neurotransmitters that acts as a mediator of bidirectional communication between immune cells (central, including microglia and macrophages, and peripheral, like T-cells^4,5 and B-cells,^6,7 natural killer cells,⁷ and neurons).^8-10 This communication is thought to involve DA receptors (as the majority of their subtypes are expressed in both systems) and selected monoamine transporters, including the dopamine transporter (DAT), norepinephrine transporter (NET), and vesicular monoamine transporter 2.³ Immune responses regulated by the presence of DA include activation of immune cells, increased production of cytokines, chemotaxis, and others. As reviewed recently by Matt and Gaskill, levels of DA and produced dopaminergic tone are region-specific, defining the number of immune cells exposed to this mediator and the complexity of the response.³ Once triggered, dopaminergic degeneration and concurrent inflammation often contribute to a disease’s progress. Thus, it was suggested that dopaminergic agents could have therapeutic potential by restoring homeostasis.^3,10 In addition, dopaminergic signaling has been shown to regulate pain responses in animal models associated with chronic pain and inflammation, suggesting additional benefits in using DA signaling modulators.¹¹

The role of astrocytes in neurodegenerative diseases such as MS has gained more and more attention. In the healthy central nervous system (CNS), they contribute to an anti-inflammatory environment. During inflammation, triggered by oxidative stress, astrocytes become reactive and promote tissue damage as well as upregulation and secretion of cytokines, such as TNF-α, IL-1β, and IL-6; neurotrophic factors include nerve growth factor, brain-derived neurotrophic factor, vascular endothelial growth factor, and leukemia inhibitory factor.¹² Chronic inflammation and neurodegeneration have been intermittently correlated with clinical disability and MS.¹³

Among dopaminergic agents, DAT ligands have been extensively explored for clinical use as treatment for stimulant-use disorder,^14,15 depression,¹⁶ appetite dysfunction,¹⁷ fatigue,¹⁸ and sleep disorders.¹⁸ The heterogenicity of pharmacological responses induced in the presence of DAT ligands is attributed to the difference in the interaction of selected compounds with the transporter.¹⁹ Some compounds, atypical inhibitors, bind to the inward-facing isomer or induce the conformational transition of the metastable state to the transporter’s inward-occluded state.²⁰ In contrast, cocaine-like compounds prefer the outward-facing DAT conformation. The conformation-specific activity of atypical DAT ligands results in pharmacological responses and addiction liabilities distinctive from the ones produced by classical psychostimulants, warranting potential therapeutic use (Figure 1).²⁰ In addition, allosteric modulators were identified with the ability to selectively affect DA uptake or release or to inhibit both of these processes (Figure 1).¹⁹

Figure 1.

Representative selected structures of known DAT inhibitors: atypical (1,2), classical (3) ligands, and allosteric modulators (4).

We recently reported²¹ a new type of DAT inhibitor with a K_i value of 15 nM, discovered by serendipity (Table 1, compound 5). The achiral molecule 5 has some structural similarities to modafinil^14,18,22-26 and bears the bis-4-fluorophenyl moiety, an alkyl linker, and a 4-chlorophenylurea motif in the head region. Over the last decade, modafinil has been evaluated in various disease conditions and showed robust results as a controller of fatigue and daytime sleepiness and as an enhancer of cognitive function.¹⁸ Structure–activity relationship studies performed on modafinil analogues have focused on increasing affinity and selectivity to DAT over other monoamine transporters, as well as optimizing metabolic stability and the toxicity profile of the original molecule.^15,22,23

Table 1.

Monoamine Transporters Binding Data for the Prepared Urea Analogues^a

#	Structure	DAT (95% CI)	NET (95% CI)	SERT (95% CI)
5		15.0 nM (5.8 to 23.0)	NA	(1655.8 to 8810.5)
6		78.5 nM (46.6 to 146.7)	NA	NT
7		188.8 nM (> 15.5 nM)	NA	NT
8		29.5 nM (17.9 to 48.7)	185.9 nM (125.8 to 298.6)	NT
9		131.6 nM (72.1 to 239.4)	272.9 nM (175.3 to 476.7)	NT
10		62.3 nM (39.8 to 97.7)	NA	NA
11		111.4 nM (34.0 to 350.1)	NA	552.5 nM (327.6 to 1538.2)
12		>10,000 nM	NA	NT
13		NA	NA	NA
14		502.3 nM (285.5 to 1360.5)	1534.6 nM (1064.1 to 2710.2)	NA
15		NA	NA	NA
16		117.7 nM (56.9 to 305.6)	601.2 nM (368.8 to 1235.9)	NT
17		NA	1485.9 nM (659.5 to 22,689.5)	NA
18		188.8 nM (>15.5 nM)	2108.6 nM (>285.8 nM)	NA
19		1517.1 nM (970.1 to 5001.2)	NA	NA
20		80.0 nM (63.8 to 349.5)	NA	NA
21		NA	NA	NA
22		118.1 nM (69.9 to 257.1)	NA	NA
23		74.5 nM (48.4 to 124.9)	NA	NA
24		357.2 (195.6 to 1614.4)	NA	NA
25		535.0 nM	NA	NA
26		70.5 nM (53.8 to 93.3)	NA	NA
27		141.6 nM (92.7 to 404.0)	NA	NA
28		731.3 nM (272.5 to 24740)	NA	NA
29		1589.3 nM (>390.1 nM)	NA	NA
30		1836.5 nM (> 381.1 nM)	NA	NA

^aIn primary screening assays, compounds were tested in triplicate or quadruplicate at a final concentration of 10 μM. Compounds with a minimum of 50% antagonist activity were subjected to secondary screening (dose–response) assays. NA—not active (antagonist activity of compound was less than 50%). NT—not tested.

^*Receptor binding profiles were generously provided by the National Institute of Mental Health’s Psychoactive Drug Screening Program, Contract # HHSN-271-2013-00017-C (NIMH PDSP).

In this paper, we continue to explore the effect of performed structural changes on the affinity of compound 5 to the DAT. In addition, we extended testing of selected analogues for their affinity to norepinephrine transporter and serotonin transporter (SERT). To explore the role of dopaminergic signaling in neuroinflammation, we have performed in vivo evaluation of compound 5 (K_{i DAT} = 15.0 nM)²¹ for its ability to modulate chronic neuroinflammation and related pain state using the experimental autoimmune encephalomyelitis (EAE) animal model. During the course of our study, Zvejniece et al. have published a paper reporting neuroprotective and anti-inflammatory activity of a nootropic agent, R-phenylpiracetam (4R)-2-(4-phenyl-2-oxopyrrolidin-1-yl-acetamide) using lipopolysaccharide (LPS), carrageenan, and formalin models.²⁷ The observed effect was explained by the DAT inhibition properties of this compound, further supporting our original hypothesis.

CHEMISTRY

All the reported compounds were prepared and analyzed using chemical and analytical procedures recently published by our group.²¹

RESULTS AND DISCUSSION

In Vitro Activity of Selected Analogues against DAT, NET, and SERT.

Our previous results have shown that modifications in all three regions of the hit compound 5 can affect affinity to the DAT.

We have reported that different substitution patterns on the phenylurea moiety are tolerated, including mono and di-substitutions, electron-donating groups (EDGs), and electron-withdrawing groups (EWGs), as well as 2,4- and 3,4-di-substitutions.²¹ In general, the best activity was observed for compounds with the EWGs at the para- (5, 8) or meta- and para- positions (10, 23) of tested analogues. In a subset of monosubstituted compounds 5–9, we saw that introduction of para-EDGs (6, K_i 78.5 nM; 7, K_i 188.8 nM) or shift to the meta-position (9, K_i 131.6 nM) resulted in loss of affinity to DAT. Disubstituted phenyl urea compounds 10–12 continued the pattern we observed previously, where 2,4-analogues had lesser activity than 3,4-derivatives.

A correlation between the electronic character of para- substituents at the bis-phenyl moiety (compounds 13–16 and 23–25) and the observed activity patterns at DAT is not straightforward. Although EDGs are clearly more favorable (14 vs 13; 24 and 25 vs 15), the most active compounds of these urea-based analogues bear 4-fluoro substitutions. In addition, an abrogated activity can be caused by hydrogen-bond acceptor properties of nitro- (13) and trifluoromethoxy (15) functionalities, whereas hydrogen-bond acceptor/hydrogen-bond donor characteristics within 14 regained some of the affinity to DAT, affording compound 14 with Ki of 502.3 nM. Unsubstituted bis-phenyl analogue 16 has an affinity (Ki 117.7 nM) slightly less than the one seen for the original compound 5. Finally, analogue 17 confirmed the importance of the diphenyl-moiety for the observed inhibition of DAT.

Table 1 presents new data on the affinity of the urea-based compounds to SERT and NET transporters. Among tested analogues, only compound 5 (K_i of 2466.0 nM) and its thiourea analogue 11 (K_i of 552.5 nM) have shown affinity to SERT. A larger subset of derivatives (8, 9, 14, and 16–18) showed affinity to NET, while preserving selectivity to the DAT. In this group, monosubstituted phenylureas were more active, although not all monosubstituted analogues have shown affinity to NET. Interestingly, compound 17 acted as a selective NET ligand, with a Ki of 1485.89 nM, signifying that the presence/loss of the phenyl moiety can modulate DAT/NET selectivity.

In the prepared series, none of the alterations in the alkyl linker part, including elongation and shortening of the alkyl chain, and locking of this flexible moiety into a cycle correlated with the ability to inhibit NET and SERT transporters.

The inhibitory potency of compound 5 was compared to that of known inhibitors of monoamine transporters (Figure 2), showing potency of 5 at the DAT to be comparable with that of GBR12909.

Figure 2.

Inhibition of [3H]-DA uptake by the DAT in the presence of tested compounds. IC₅₀ values were calculated based on at least three independent experiments (n = 4). Tested compounds showed the following inhibitory potency: compound 5 IC₅₀ 17.8 nM (95% CI < 24.8 nM; r² 0.9060); cocaine IC₅₀ 408.9 nM (95% CI: 28.3–598.6 nM; r² 0.9309); nisoxetine IC₅₀ 1.43 μM (95% CI: 0.99–2.1 μM; r² 0.9178); fluoxetine IC₅₀ 16.1 μM (95% CI: 9.8–37.5 μM; r² 0.765); GBR12909 IC₅₀ 22.5 nM (95% CI: 16.1–31.3 nM; r² 0.9284).

Evaluation of DAT Inhibitors in the EAE Animal Model of MS.

Our next set of experiments evaluated compound 5 for its ability to inhibit CNS inflammation pathways in animals with EAE. EAE is a widely used experimental model that mimics complex interactions between immunopathological and neuropathological mechanisms and creates a versatile and robust platform for developing clinical agents for the treatment of MS.²⁸ In our experiments, EAE was induced by immunization of mice (CB57BL/6) with myelin oligodendrocyte protein (MOG35-55) in the presence of pertussis toxin to promote neuroinflammation. Because the EAE model has more consistent results in female mice, this study used only female mice (n = 46) for the initial assessment. Treatment of animals (days 1–10) included daily oral doses of compound 5, or reference compound methylprednisolone (Solu-Medrol), a potent anti-inflammatory corticosteroid that is used for the treatment of acute exacerbations in patients with MS.²⁹ To determine if the observed in vivo activity of compound 5 in this animal model is characteristic of other DAT inhibitors, we have included another group of animals treated with 10 mg/kg³⁰ of (±)-modafinil in our study. Our selection of modafinil as a control was based on its mechanism of action (DAT inhibitor) and clinical use in treating MS patients.^18,31 The clinical signs of MS in animals were characterized using a standard scale (see General Procedures for in vivo experiments), and monitoring/recording continued until day 23. Compound 5, methylprednisolone (Solu-Medrol), and (±)-modafinil were all dissolved in a vehicle solution (2.5% ethanol, 2.5% dimethyl sulfoxide (DMSO), 5% Tween 80, 25% PEG 400, and 65% PBS).

The control treatment EAE group of animals (EAE vehicle) showed characteristic clinical deficits that began as paralysis in the tail (clinical grade 1) from day 13 post-EAE induction (Figure 3). From day 13 to day 16 post-EAE induction, these symptoms progressed to more severe clinical deficits such as hindlimb weakness (clinical grades 2–3) and hindlimb paralysis (clinical grades 3–4) (Figure 3, ***P < 0.001, F_11,176 = 5.985, one-way analysis of variance (ANOVA)). Pretreatment with DAT inhibitors (compound 5 or (±)-modafinil) or Solu-Medrol significantly attenuated clinical deficits observed in the EAE-vehicle group (Figure 3, ++, +++P < 0.01–0.001, F_3,578 = 104.1, two-way ANOVA, Bonferroni post hoc test). There is no significant difference between compound 5 with (±)-modafinil group, and no significant difference between compound 5 with the Solu-Medrol group at most time points except at days 17–20 (Figure 3, #, ## P < 0.05–0.01, F_3,578 = 104.1, two-way ANOVA, Bonferroni post hoc test).

Figure 3.

Clinical scores of EAE mice. EAE mice treated with vehicle showed clinical deficits that began with clinical grade 1 from day 13 post-EAE induction and progressed to more severe clinical deficits clinical grades 2–3 from day 13 to day 16 post-EAE induction and clinical grades 3–4 from day 16 to day 23 post-EAE induction (***P < 0.001, F_11,176 = 5.985, one-way ANOVA, vehicle control, n = 12). A 10-day treatment with compound 5 (10 mg/kg, p.o., n = 11), (±)-modafinil (10 mg/kg, p.o., n = 6), or Solu-Medrol (50 mg/kg, p.o., n = 9) significantly reduced clinical deficits in EAE animals (++P < 0.01, +++P < 0.001, two-way ANOVA, Bonferroni post hoc test compared to vehicle control). #, ## P < 0.05–0.01, two-way ANOVA, Bonferroni post hoc test compared Solu-Medrol group to vehicle control.

On day 24 post-EAE induction, we performed pain behavioral testing in normal control and EAE groups treated with vehicle, compound 5, Solu-Medrol, and modafinil. The front left paw withdrawal thresholds were measured using an electronic von Frey plantar anesthesiometer to test for mechanical hypersensitivity (see General Procedures for in vivo experiments). Animals in the EAE-vehicle group, compound 5 group, and modafinil group showed robust tactile allodynia reflected in the decreased reflex thresholds (Figure 4, *, ** P < 0.05–0.01, one-way ANOVA). Only Solu-Medrol-treated animals showed an increased mechanical threshold (+P < 0.05, one-way ANOVA compared to the EAE-vehicle group), suggesting antinociceptive effects of Solu-Medrol, whereas the compound 5-treated group shows the tendency of antinociceptive effects but had no significant effect (P = 0.0621, one-way ANOVA compared to the EAE-vehicle group).

Figure 4.

Effect of the treatment with DAT inhibitors (compound 5 and (±)-modafinil) or Solu-Medrol on mechanical withdrawal thresholds in EAE mice. The front left paw withdrawal thresholds were measured using an electronic von Frey plantar anesthesiometer in normal control mice (n = 8) and EAE mice treated with vehicle (n = 10), compound 5 (n = 10), Solu-Medrol (n = 9), and modafinil (n = 6), Animals in the EAE-vehicle group, compound 5 group, and modafinil group showed decreased withdraw thresholds (*, ** P < 0.05–0.01, one-way ANOVA). Solu-Medrol-treated animals showed an increased mechanical threshold (+P < 0.05, one-way ANOVA compared to the EAE-vehicle group).

To measure supraspinally organized nocifensive and affective pain-like behaviors, audible and ultrasonic vocalizations (Figure 5) were evoked by brief (15 s) noxious mechanical stimulation of the left front paw (see General Procedures for in vivo experiments). Overall, EAE-vehicle mice had increased audible and ultrasonic vocalizations (duration) than normal control mice (P < 0.001), indicating increased nocifensive and affective pain responses in this group of animals. Treatment with DAT inhibitors 5 and (±)-modafinil significantly inhibited both types of vocalizations (P < 0.001 ANOVA with Bonferroni correction). However, treatment with Solu-Medrol affected audible vocalization levels in EAE mice, while changes in ultrasonic vocalizations were not statistically significant (audible P < 0.05 ANOVA with Bonferroni correction). The data can be interpreted to suggest that Solu-Medrol primarily affects nocifensive aspects of pain (von Frey thresholds and audible vocalization).

Figure 5.

Effect of the treatment with DAT inhibitors (compound 5 or modafinil) or corticosteroid Solu-Medrol on audible and ultrasonic vocalizations in EAE mice. Audible and ultrasonic vocalizations were recorded in normal control mice (n = 8) and EAE mice treated with vehicle (n = 10), compound 5 (n = 10), Solu-Medrol (n = 9), and modafinil (n = 6) groups on day 24 after EAE induction. Animals in the EAE-vehicle group showed increased audible and ultrasonic vocalizations (*** P < 0.001, one-way ANOVA compared to normal control mice). Solu-Medrol-treated animals showed an increased mechanical threshold (+P < 0.05, one-way ANOVA compared to the EAE-vehicle group). The compound 5 group and modafinil group showed decreased audible and ultrasonic vocalizations (+, ++, +++ P < 0.05–0.001, one-way ANOVA compared to the vehicle-treated group). The Solu-Medrol treated group showed decreased audible vocalizations (+P < 0.05, one-way ANOVA compared to the vehicle-treated group).

Evidence suggests that the level of neuroinflammation and glial cell reactivity in the CNS strongly correlate with the progress of MS as well as other disorders such as chronic pain, which is associated with MS. The amygdala has emerged as an important node of the emotional-affective aspects of pain and pain modulation. The central nucleus (CeA) in particular receives nociceptive information and serves major output functions.³² Therefore, we examined the astrocyte and macrophage reactivity in the amygdala in EAE mice and assessed the effect of treatments on these inflammation markers. In the EAE-vehicle group, significantly enhanced reactivity of astrocytes (Figure 6) was found in the CeA as evidenced by the higher levels of GFAP expression in the CeA of EAE-vehicle mice compared to normal controls (P < 0.001, ANOVA) (Figure 6B,C). DAT inhibitor 5 or Solu-Medrol reduced GFAP expression (P < 0.001, ANOVA) (Figure 6B,C), suggesting that these compounds prevent astrocyte reactivity and attenuate neuroinflammation.

Figure 6.

Effect of DAT inhibitor 5 and Solu-Medrol on the GFAP expression in amygdala of EAE mice. (A) Diagram of a coronal brain slice (2.30 mm caudal to bregma) indicates the amygdala (CeA) area shown in the confocal images (B) of GFAP expression in the CeA. C. The number of GFAP cells was significantly increased in EAE mice treated with vehicle (*** P < 0.001 compared to normal, n = 20–22 slices). A 10-day treatment with compound 5 (10 mg/kg, oral dosing) or Solu-Medrol (50 mg/kg, oral dosing) significantly decreased the number of GFAP cells (+++ P < 0.001 compared to EAE-vehicle, n = 15–20 slices). Scale bars, 100 μm.

Elevated levels of tumor necrosis factor (TNF), a potent inflammation mediator, in MS lesions, serum, and cerebrospinal fluid (CSF), were linked to the progression of MS and EAE.³³ Furthermore, Valentin-Torres et al. showed that blockage of TFN during the acute stage of disease prevents MS lesion progression and promotes remyelination.³³ Transforming growth factor-beta (TGFβ) is known to play a complex role in MS and EAE progression.³⁴ It can regulate the production of the inflammatory Th17 cells and, at the same time, enhances levels of T_reg population in MS patients, providing a beneficial effect.^34,35 Previously, higher DA levels were shown to inhibit TGFβ production³⁶ and promote Th17 phenotype expansion in patients.³⁷ Higher interferon-gamma (INFγ) levels in MS and EAE conditions correlate with the activation of the inflammation-induced T_H1 pathway.³⁸ Administration of INFγ was shown to worsen MS symptoms in patients and increase the relapse rate.³⁹ At the same time, lack of INFγ or INFγ receptors increased severity of symptoms in the EAE model,⁴⁰ suggesting a complex modulatory role of this cytokine in the inflammation response and MS progression.⁴¹ Recent data suggest that INFγ plays a stage-dependent role in the progression of the EAE and that therapies based on INFγ-signaling need to be patient-specific.⁴² As a therapy for MS, glucocorticoids, including Solu-Medrol, act through a variety of immunosuppressive effects, including preferential inhibition of T_H1 proinflammatory cytokine levels, such as TNFα, IL-1β, and IL-12.^43,44 At the same time, this class of agents³⁵ are known to inhibit the production of proinflammatory markers, including IL-6, IL-12, and IL-1β cytokines.

Our study evaluated if modulation of DA signaling with DAT inhibitor 5 and glucocorticoids has similar effects on immune responses in the amygdala and other tissues (lymph nodes and spinal cord) in the EAE model. Lymph nodes, as well as other lymphoid organs, are characterized by close interaction between DA (derived from the sympathetic terminals) and immune cells, suggesting that balanced DA levels are required to maintain homeostatic balance in these tissues. Similar interactions between immune cells and DA in CSF led us to test for a treatment effect in spinal cord-derived samples as well.

Our data showed that Solu-Medrol and DAT inhibitor 5 significantly attenuated IL-6 protein and proinflammatory cytokine mRNA expression in the spinal cord and amygdala (Figures 7 and 8A,B). The most significant inhibition was induced by 5 in the spinal cord of EAE animals, where levels of IL-1β declined by 90% (Figure 8B). However, no statistically significant changes in cytokine expression levels were detected in the lymph nodes. (Figure 8C). We also found that treatment with Solu-Medrol caused inhibition of TNF levels in the amygdala cells and to a lesser extent in the spinal cord (Figure 8A,B). No significant changes in TNF expression were noted in the lymph nodes (Figure 8C). DAT inhibitor 5 had similar effects, inducing a 75% reduction of TNF levels in the amygdala (Figure 8A). In our study, DAT inhibitor 5 and glucocorticoid Solu-Medrol decreased the level of TGFβ mRNA expression in amygdala cells and spinal cord cells. The effect was similar for both of these agents, with the highest reduction of 70% observed in the spinal cord. No significant changes in the expression of TGFβ levels in lymph nodes were observed. IL-12 marker levels were slightly inhibited by 5 only (amygdala and spinal cord), whereas compound 5 produced a 25–50% reduction in INFγ levels in amygdala and lymph nodes but no statistically significant effects in spinal cord samples (Figure 8). Furthermore, we evaluated the effect of both Solu-Medrol and DAT inhibitor 5 on the expression levels of anti-inflammatory cytokine IL-10. Both compounds increased IL-10 expression in the amygdala, but only DAT inhibitor 5 increased IL-10 gene expression levels in the spinal cord tissue (by ~4-fold), whereas Solu-Medrol had no notable effect in these cells. Except for a decrease of INFγ levels by compound 5, Solu-Medrol or compound 5 had no effect on any analyzed markers in the lymph node samples in the EAE model.

Figure 7.

Effect of DAT inhibitor 5 and Solu-Medrol on the IL-6 expression in the amygdala in EAE mice. (A) Diagram of a coronal brain slice (2.30 mm caudal to bregma) indicates the amygdala area shown in the confocal image (B) of IL-6-IR expression in the CeA. (C) The number of IL-6-IR cells was significantly increased in EAE mice treated with vehicle (*** P < 0.001 compared to normal, n = 10–12 slices). A 10-day treatment with compound 5 (10 mg/kg, oral dosing) or Solu-Medrol (50 mg/kg, oral dosing) significantly decreased the number of IL-6-IR cells (+++ P < 0.05 compared to the EAE vehicle, n = 9–13). Scale bars, 100 μm.

Figure 8.

Effect of tested compounds on mRNA expression of pro and anti-inflammatory cytokines in EAE mice. Quantitative polymerase chain reaction (qPCR) analyses of inflammatory cytokine expression in CeA (A), spinal cord (B), and lymph node (C) cells isolated from EAE mice. Animals were treated with vehicle, DAT inhibitor 5 (10 mg/kg, p.o.), or Solu-Medrol (50 mg/kg, p.o.). All data represent mean ± SE, and P-values were calculated by ANOVA with Bonferroni post hoc tests, *,**,***P < 0.05, 0.01, and 0.001.

These findings suggest that neuroinflammation in the brain (amygdala) and spinal cord and reactive microglia in the amygdala are key mechanisms affected by the administration of DAT inhibitor 5 to EAE animals. Regulation of dopaminergic signaling in the presence of this compound results in subsequent modulation of immune markers, providing further evidence for a potential role of dopaminergic drugs in the therapy of chronic inflammation of the CNS. A more robust decline in clinical deficits of Solu-Medrol-treated animals can be explained in part by a dose regimen (50 mg/kg of Solu-Medrol vs 10 mg/kg of 5) that provides low micromolar levels of the drug in the brain. This dose regimen was selected based on the previous publications evaluating the effect of Solu-Medrol on the EAE model.

In the next set of experiments, we have evaluated an anti-inflammatory effect of DAT inhibitors (compound 5 and (±)-modafinil) and Solu-Medrol on RAW 264.7 mouse macrophages⁴⁵ induced with LPS (Supporting Information, Figure S1). Our preliminary data showed that both DAT inhibitors showed minimal ability to modulate mRNA expression levels of the proinflammatory factors (TNF-α and IL-1β) and the anti-inflammatory marker interleukin-4 receptor α (IL-4Rα). Under the same experimental conditions, Solu-Medrol has shown strong inhibition of IL-1β mRNA levels.

CONCLUSIONS

This study extends the evaluation of a recently reported class of urea analogues as inhibitors of monoamine transporters. We show that tested analogues have high selectivity for the DAT, with only a few compounds having dual activity at SERT or NET. Our data highlighted structural features that promote high affinity to the DAT, with Ki values as low as 15.0 nM. Furthermore, this series of DAT inhibitors is achiral, providing the advantage of synthetic simplicity. Structural similarity with modafinil offers indirect evidence that inhibitor 5 and its analogue are also atypical DAT inhibitors characterized by low addictive potential. Indirectly, this statement is supported by the lack of behavioral changes observed in the group of animals treated with compound 5 (recorded video of in vivo experiment, data not shown). The corresponding studies targeting the classification of urea analogues as classical or atypical DAT inhibitors are underway.

Based on the multiple reports highlighting the DA-mediated immune cell function in many disease states, we hypothesized that DAT inhibitors could modulate the inflammation process in the brain.^{3,27,35,46,47} The favorable pharmacokinetic profile of selected compound 5, including its ability to cross the BBB in vivo and metabolic stability, allowed us to investigate its ability to modulate clinical symptoms in EAE animal models that closely mimics MS-relevant inflammation symptoms in mice. Both DAT inhibitors, urea-based molecule 5 and modafinil, showed statistically significant anti-inflammatory effects and attenuated motor deficits and pain behaviors in the EAE model, confirming the potential connection between dopaminergic and neuroimmune systems under experimental settings. Although the polypharmacological effect for the action of compound 5 is possible, we have confirmed that our compound has none or very limited affinity to all subtypes of 5-HT, DA, opioid, alpha, beta, sigma, muscarinic acetylcholine, histamine receptors, and GABAA. We also confirmed that this molecule has no affinity to FGFR1, a possible target for some of the urea analogues produced in this series. We have to specify that this study’s goal was not to show the therapeutic advantage of using the DAT inhibitor 5 over Solu-Medrol in the treatment of EAE-induced symptoms. Instead, we were interested in evaluating the tested compound’s anti-inflammatory effect and comparing its activity to that of the potent anti-inflammatory corticosteroid. Our data showed that another DAT inhibitor, (±)-modafinil, has shown potency in the same EAE model. In addition to the DA circuits, this compound has some effect on serotonin pathways. Because serotonin plays a role in the progression of MS, we have hypothesized that modafinil can have a more pronounced impact on the EAE animals.^31,48 Interestingly, the anti-inflammatory activity of both DAT inhibitors in stimulated murine monocytic cells RAW 264.7 was modest compared to in vivo results, indirectly showing the importance of DA-mediated pathways in the observed activity of modafinil and compound 5. We believe that our data and the previously published results by other groups provide additional support to the existing concept of DA-induced inflammation in the CNS tissues. To the best of our knowledge, this is the first study reporting beneficial effects of DAT inhibitor-based treatment in animals with induced autoimmune encephalomyelitis. Mechanisms of the observed effect of DAT inhibitors in CNS inflammation models and the role of monoaminergic systems in MS-related inflammation remain to be determined.

EXPERIMENTAL SECTION

Preparation and characterization of all included compounds were reported recently.²¹ However, for interested readers we have included characterization of these analogues in the Supporting Information.

General Chemistry Procedures.

All reactions were carried out in oven- or flame-dried glassware under positive nitrogen/argon pressure unless otherwise noted. All solvents and chemicals were of reagent grade. Unless otherwise noted, all reagents and solvents were purchased from commercial vendors and used as received. The purity and characterization of compounds were established by a combination of methods, including thin-layer chromatography (TLC), high-performance liquid chromatography, mass spectrometry, and nuclear magnetic resonance (NMR) analysis. ¹H and ¹³C NMR spectra were recorded on a Bruker 400 MHz Advance III HD spectrometer using chloroform-d, methanol-d, or DMSO-d₆ with tetramethyl (0.00 ppm) or solvent peaks as the internal standard. Chemical shifts (δ) are recorded in ppm relative to the reference signal, and coupling constant (J) values are recorded in hertz (Hz). Multiplicates are indicated by s (singlet), d (doublet), dd (doublet of doublets), t (triplet), q (quartet), m (multiplet), and br (broad). TLC was performed on EMD precoated silica gel 6-F254 plates, and spots were visualized with UV light or iodine staining. Flash column chromatography was performed with silica gel (40–63 μm, 60 Å) using the mobile phase indicated or on a Teledyne Isco (CombiFlash R_f UV/vis). High-resolution mass spectra were obtained using a TripleTOF 5600 mass spectrometer. The purity of all final compounds was greater than 95%. The purity was determined with a Waters Acquity UPLC using a C18 column (Cortecs, 1.6 μm, 2.1 × 50 mm): eluent A, 0.1% aqueous CF₃COOH and eluent B. CH₃CN containing 0.1% CF₃COOH, gradient elution (0 min: 95% A, 5% B; 2 min: 50% A, 50% B; 4 min: 5. 50% A, 50% B; 6 min: 10% A, 90% B; 9 min: 10% A, 90% B; 10 min: 95% A, 5%), with a flow rate of 0.2 mL/min.

General Procedures for In Vivo Experiments.

Female C57BL/6 mice (9–12 weeks old) were purchased from Charles River and housed in a temperature-controlled room under a 12 h light/dark cycle. Water and food were available without restriction. Animals were acclimatized to the laboratory for 7 days prior to immunization. On the day of the experiment, mice were transferred from the animal facility and allowed to acclimate for at least 1 h. All experimental procedures were approved by the Institutional Animal Care and Use Committee at Texas Tech University Health Sciences Center and conform to the International Association for the Study of Pain and National Institutes of Health guidelines.

EAE Model Induction.

The EAE model was induced using Hooke KitTM MOG35–55/CFA Emulsion PTX (EK-2110, Hooke Laboratories, Lawrence, MA, USA).⁴⁹ Antigen of MOG35–55/CFA Emulsion was administered subcutaneously at the upper back and the lower back of each mouse (0.1 mL/site, 0.2 mL/mouse in total). Intraperitoneal injection of 80 ng pertussis toxin (EK-2110, Hooke Laboratories, Lawrence, MA, USA) was performed 2 h after antigen injection and again 24 h later. Control mice were treated with PBS.

Seven days after antigen injection, mice were monitored daily, and the clinical signs of EAE were scored on a scale of 0 to 5⁴⁹ (see “Behavioral testing”).

Experimental Protocol.

A total of 46 female C57BL/6 mice (9–12 weeks old) were randomly divided into four groups: nonimmunized mice (non-EAE control) and three groups of EAE mice which received oral injection vehicle (EAE-vehicle), compound 5 (EAE-5), and Solu-Medrol (EAE-solumedrol), respectively. Oral injections were given daily from day 1 to day 10 after antigen injection. Mice were monitored daily, and the clinical signs of EAE were scored on days 7 to 23 after antigen injection. Pain behaviors, including nocifensive reflex thresholds and vocalizations, were measured 24 days after antigen injection. All behavioral testing was carried out by an experimenter blinded to the specific treatment groups. After the behavioral testing, some mice of each group were perfused with 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB), and brain tissues were collected for immunohistochemistry, while the remaining mice were sacrificed by decapitation to obtain brain tissues for PCR experiments.

Behavioral Testing.

EAE Scoring.

Seven days after antigen injection, mice were placed in an arena (50 cm × 50 cm) for 5 min. Clinical signs of EAE were scored and recorded using a computerized video tracking system (EthoVisionXT 11 software, Noldus Information Technology). The grading system for clinical assessment of EAE was as follows:⁴⁹ Score 0, no apparent changes in motor function compared to nonimmunized mice; Score 1, limp tail or hind limb weakness but not both; Score 2, limp tail and hind limb weakness; Score 3, limp tail and complete hind limb paralysis; Score 4, limp tail, complete hind limb paralysis and front limb weakness; Score 5, moribund state with severe paralysis and spontaneously rolling in the cage.⁴⁹

Mechanosensitivity.

Mechanical withdrawal thresholds were measured using a plantar electronic von Frey anesthesiometer (IITC Life Science, Woodland Hills, CA). The tip was applied perpendicularly to the front left paw base with increasing force until a flexion reflex was provoked, which was automatically recorded as the paw withdrawal threshold (in grams). The average of triplicate measurements at least 30 s apart was used.

Emotional Responses.

Vocalizations in the audible (20 Hz–16 kHz) and ultrasonic (25 ± 4 kHz) ranges were measured using a condenser microphone and a bat detector, respectively, connected to a filter and amplifier (UltraVox four-channel system; Noldus Information Technology, Leesburg, VA) as in our previous studies.^50-54 Animals were anesthetized briefly with isoflurane (2%, precision vaporizer) and placed in a custom-designed cloth hammock that permitted access to the front and hind limbs for mechanical test stimuli. After habituation, brief (15 s) innocuous (100 g/6 mm²) and noxious (300 g/6 mm²) mechanical stimuli were applied to the front paw using a calibrated forceps with a force transducer to monitor the applied force (in grams). Durations of audible and ultrasonic vocalizations were analyzed for 1 min following the onset of the mechanical stimulus using Ultravox 2.0 software (Noldus Information Technology).

Immunocytochemistry.

Mouse brain tissues for immunocytochemistry were taken on days 25–26 after antigen injection. Mice were deeply anesthetized with isoflurane (3–4%) and sacrificed by transcardiac perfusion with 4% PFA in 0.1 M PB. The brain was removed, postfixed for 3–4 h, and then transferred to a 30% sucrose solution in 0.1 M PB. Brains were embedded in optimal cutting temperature compound (Fisher Scientific, Edmonton, AB), frozen, and processed for cryostat sectioning (40 mm). Coronal brain slices (40 μm) containing the amygdala were obtained with a cryostat at −20 °C (Leica CM 1850, Leica Microsystems Nussloch GmbH, Nussloch, Germany). The sections were free-floating for blocking, antibody exposure, and washes. PBS with 0.1% Triton X-100 plus 5% nonfat dehydrated milk was used to block nonspecific binding for 1 h at 22 °C. The following antibodies were used: (i) mouse anti-GFAP monoclonal antibody (GA5), Alexa Fluor 488 (1:200, Invitrogen, Catalog # 53–9892-80); (ii) rabbit anti-IL-1beta polyclonal antibody (1:100, Novus Biological, Catalog # NB600–633); and (iii) rabbit anti-IL-6 polyclonal antibody (1:400, Novus Biological, Catalog # NB600–1131). Primary antibodies were visualized using goat antirabbit Alexa Fluor 488 secondary antibodies (1:500, Molecular Probes, Eugene, OR). Quantification of immunocytochemistry images was performed with an Olympus FV3000 confocal microscope (Olympus, Waltham, MA). Labeled cells were counted using ImageJ software (NIH). Primary antibodies were omitted for controls to verify signals and determine noise.

PCR Experiments.

Mouse lymph nodes, spinal cord, and amygdala RNA were extracted with RNAzolRT (Molecular Research Center, Inc. Cincinnati, OH); cDNA was synthesized from RNA using SuperScrip II reverse transcriptase according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA); the cDNA was used for performing for real-time PCR (RT-PCR) to confirm the fold changes. 7900HT Fast Real-Time PCR system, TaqManGene Expression Master Mix, and TaqManGene Expression Assays were purchased from ThermoFisher Scientifi c. Assays ID Mm00607939_s1, Mm00446190_m1, Mm00434228_m1, Mm00443258_m1, Mm01178820_m1, Mm01168134_m1, Mm00434169_m1, and Mm001288386_m1 were used for mouse β-actin, IL-6, IL-1β, TNFα, TGF-β1, IFN-γ, IL-12a, and IL-10. Fold change was calculated using the 2(-Delta Delta C(T)) method; β-actin was used as an internal marker.

RT-PCR Experiments.

To determine the mechanisms by which our experimental compound was working, we stimulated murine monocytic cells, RAW 264.7 with LPS (Sigma-Aldrich, USA) and treated them with our compound independently. In brief, 100,000 RAW 264.7 cells were plated in a well of a six-well plate in Dulbecco’s modified Eagle medium (Hyclone, USA) containing 10% fetal bovine serum (Hycone, USA) and 1% antibiotic/antimycotic (Gibco, USA) solution. The next morning, 1 μL of 1 mg/mL LPS was added to the experimental groups. Treatment groups received a final concentration of 10 and 1 μM of our experimental compound that was dissolved in DMSO (Thermo Fisher, USA). We also treated cells using the same concentrations of modafinil (Supelco, USA), which was dissolved in acetonitrile, and 6α-methylprednisolone (Sigma-Aldrich, USA), which was dissolved in DMSO. We also used two vehicle controls, adding only DMSO to one group and only acetonitrile to another, both in quantities that matched the volume of the drug solutions added in the drug-treatment groups. After 48 h, the media in each well were removed, and cells were lysed using 1 mL of TriZol reagent (Ambion, USA). We then isolated mRNA from each sample, synthesized cDNA from the mRNA, and used the cDNA to run RT-PCR. We then analyzed the mRNA expression of the proinflammatory factors such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), as well as the anti-inflammatory marker interleukin-4 receptor α (IL-4Rα).

Statistics.

All averaged values are given as the mean ± SE. Statistical significance was accepted at the level P < 0.05. GraphPad Prism 7.0 software was used for all statistical analyses. Statistical analysis was performed on the raw data. For multiple comparisons, ANOVA was used with Bonferroni post hoc tests.

ABREVIATIONS USED

BDNF

brain-derived neurotrophic factor

CeA

central nucleus

CSF

cerebrospinal fluid

CNS

central nervous system

DAT

dopamine transporter

EAE

experimental autoimmune encephalomyelitis

EDG

electron-donating group

EWG

electron-withdrawing group

GFAP

glial fibrillary acidic protein

GPCR

G-protein coupled receptor

IC₅₀

the half maximal inhibitory concentration

INFγ

interferon-gamma

K_i

inhibition constant

LIF

leukemia inhibitory factor

LPS

lipopolysaccharide

MOG35-55

myelin oligodendrocyte protein

MS

multiple sclerosis

NA

not active

NET

norepinephrine transporter

NGF

neurotrophic factors including nerve growth factor

NT

not tested

SERT

serotonin transporter

TGFβ

transforming growth factor-beta

TNF

tumor necrosis factor

VEGF

vascular endothelial growth factor

VMAT2

vesicular monoamine transporter 2