Regulation of Melanocortin-3 and -4 Receptors by Isoforms of Melanocortin-2 Receptor Accessory Protein 1 and 2

The neural melanocortin receptors (MCRs), melanocortin-3 and -4 receptors (MC3R and MC4R), play essential non-redundant roles in the regulation of energy homeostasis. Interaction of neural MCRs and melanocortin-2 receptor accessory proteins (MRAPs, MRAP1 and MRAP2) is suggested to play pivotal roles in MC3R and MC4R signaling. In the present study, we identified two new human (h) MRAP2 splice variants, MRAP2b (465 bp open reading frame) and MRAP2c (381 bp open reading frame). Human MRAP2s are different in C-termini. We investigated the effects of five isoforms of MRAPs, hMRAP1a, hMRAP1b, hMRAP2a, hMRAP2b, and hMRAP2c, on MC3R and MC4R pharmacology. At the hMC3R, hMRAP1a and hMRAP2c increased and hMRAP1b decreased the cell surface expression. hMRAP1a increased affinity to ACTH. Four MRAPs (hMRAP1a, hMRAP1b, hMRAP2a, and hMRAP2c) decreased the maximal responses in response to α-MSH and ACTH. For hMC4R, hMRAP1a, hMRAP2a, and hMRAP2c increased the cell surface expression of hMC4R. Human MRAP1b significantly increased affinity to ACTH while MRAP2a decreased affinity to ACTH. Human MRAP1a increased ACTH potency. MRAPs also affected hMC4R basal activities, with hMRAP1s increasing and hMRAP2s decreasing the basal activities. In summary, the newly identified splicing variants, hMRAP2b and hMRAP2c, could regulate MC3R and MC4R pharmacology. The two MRAP1s and three MRAP2s had differential effects on MC3R and MC4R trafficking, binding, and signaling. These findings led to a better understanding of the regulation of neural MCRs by MRAP1s and MRAP2s.

Melanocortin-2 receptor accessory protein 1 (MRAP1), first identified as low molecular weight protein from fat tissue [28], was the first MC2R accessory protein identified, as the specific molecular chaperone for MC2R in regulating receptor expression, ligand binding, and signaling [29][30][31][32]. MRAP1 mutations account for~20% of familial glucocorticoid deficiency cases [29,33]. There are two alternatively spliced isoforms of human (h) MRAP1, hMRAP1a and hMRAP1b, with similar effects on MC2R trafficking and signaling [29,30]. MRAP1a and MRAP1b are widely expressed, but their distribution patterns are distinct [29,34]. These results suggest that hMRAP1a and hMRAP1b might possess multiple functions beyond regulating MC2R (primarily expressed in the adrenal gland) [29]. Indeed, hMRAP1a has been shown to regulate all five hMCRs in distinct ways [35][36][37]. However, almost all of the investigations focus on MRAP1a and its regulation on MC3R/MC4R, and there are few studies on MRAP1b.
Recently, we identified two new human MRAP2 splice variants, MRAP2b and MRAP2c. Human MRAP2b and MRAP2c share the same amino acid sequences in N-termini and transmembrane domains (TMD) with hMRAP2a. However, whether MRAP2b and MRAP2c are involved in MC3R/MC4R regulation was unknown. Additionally, the regulation of MC3R/MC4R by MRAP1b is not clear. Hence, the potential effects of all five isoforms of hMRAPs, hMRAP1a, hMRAP1b, hMRAP2a, hMRAP2b, and hMRAP2c, on hMC3R and hMC4R pharmacology were systematically investigated in this study.

Flow Cytometry Assay
The influence of hMRAP1s or hMRAP2s on the total and cell surface expression of hMC3R and hMC4R was performed using flow cytometry (Accuri Cytometers, Ann Arbor, MI, USA) as described previously [56,57]. Cells (6-well plates) were transfected with hMC3R or hMC4R (N-terminal c-myc tag) and hMRAP1a, hMRAP1b, hMRAP2a, hMRAP2b or hMRAP2c plasmids at a ratio of 1:5. Fluorescence of cells transfected with empty vector (pcDNA3.1) was used for background staining. The expression of the hMC3R or hMC4R was calculated as the percentage of the cell transfected with hMC3R or hMC4R but without MRAPs set as 100% [56].

Statistical Analysis
All data were represented as mean ± S.E.M. The parameters and significance of differences were calculated by GraphPad Prism 8.3 software (GraphPad, San Diego, CA, USA). The significance of differences in ligand binding, cAMP signaling, and flow cytometry parameters were all determined by one-way ANOVA, with p < 0.05 set as significant.

Regulation of hMC3R Pharmacology by hMRAP1s and hMRAP2s
Flow cytometry was used to determine MRAP regulation of hMC3R expression ( Figure 2). The results showed that hMRAP1a and hMRAP2c significantly increased the cell surface expression, and hMRAP1b decreased the cell surface expression of hMC3R ( Figure 2A). Human MRAP2a and hMRAP2b had no effect on the cell surface expression of hMC3R ( Figure 2A). Only hMRAP1b decreased the total expression of hMC3R, and the other four MRAPs did not affect the total expression of hMC3R ( Figure 2B). Competitive ligand binding assays were performed to explore MRAP regulation of hMC3R binding properties. Different concentrations of unlabeled α-MSH or ACTH  were used to compete with a fixed amount of 125 I-NDP-MSH. Results showed that only hM-RAP1b significantly decreased the maximal binding value (B max ), and hMRAP1a, hMRAP2a, hMRAP2b, and hMRAP2c had no significant effect on B max s of hMC3R ( Figure 3 and Table 1). All MRAPs did not affect α-MSH affinities at hMC3R ( Figure 3A and Table 1). Only hM-RAP1a increased ACTH affinity of hMC3R, and the other MRAPs had no effect on affinities of hMC3R to ACTH ( Figure 3B and Table 1).  Values are expressed as the mean ± SEM of at least three independent experiments. a Significant difference from the parameter of hMC3R, p < 0.05.
The signaling properties of hMC3R modulated by MRAPs were determined using cAMP RIA. Results showed that all hMRAPs had no significant effect on potencies of hMC3R to α-MSH and ACTH ( Figure 4A,B and Table 2). Four hMRAPs (hMRAP1a, hMRAP1b, hMRAP2a, and hMRAP2c) markedly decreased maximal responses (R max ) in response to α-MSH and ACTH, and hMRAP2b deceased R max to ACTH but not α-MSH ( Figure 4A,B and Table 2). In addition, all MRAPs significantly decreased the basal activities of hMC3R (Table 2).

Regulation of hMC4R Pharmacology by hMRAP1s and hMRAP2s
Results of flow cytometry showed that hMRAP1a, hMRAP2a, and hMRAP2c significantly increased, while hMRAP1b and hMRAP2b had no significant effect on the cell surface and total expression of hMC4R ( Figure 5). Ligand binding assays indicated that at hMC4R, hMRAP1a and hMRAP1b significantly decreased B max s, while hMRAP2a increased B max s ( Figure 6 and Table 3). No significant effect was observed for hMRAP1a, hMRAP1b, hMRAP2a, hMRAP2b, and hMRAP2c on α-MSH affinities at the hMC4R ( Figure 6A and Table 3). Additionally, hMRAP1b increased affinity, whereas hMRAP2a decreased affinity of hMC4R to ACTH ( Figure 6B and Table 4). MRAP1a, MRAP2b, and MRAP2c had no effects on affinities of hMC4R to ACTH ( Figure 6B and Table 3).  . Data are expressed as % of hMC4R binding ± range from duplicate measurements within one experiment. All experiments were performed at least three times independently.  Modulation of hMC4R signaling by MRAP1s and MRAP2s was also studied. Data showed all MRAPs had no effects on α-MSH potencies of hMC4R ( Figure 7A and Table 4). Only hMRAP1a significantly increased ACTH potency, and the other MRAPs did not affect ACTH potency at hMC4R ( Figure 7B and Table 4). Both hMRAP1a and hMRAP1b significantly increased the basal cAMP levels, whereas all three MRAP2s decreased the basal activities of hMC4R (Table 4). Additionally, all MRAPs decreased R max s of hMC4R when α-MSH was used ( Figure 7A and Table 4). Only hMRAP1b decreased ACTH-stimulated cAMP generation, and the other MRAPs had no effect on R max s of hMC4R in response to ACTH ( Figure 7B and Table 4). Data are mean ± SEM from triplicate measurements within one experiment. All experiments were performed at least three times independently.

Discussion
Alternative splicing is prevalent in eukaryotes, resulting in a greatly increased diversity of proteins encoded by the genome [59]. Tissue-specific and developmentally regulated alternative splicing is also modulated by divergent stimulation. Approximately 95% of multi-exon genes are alternatively spliced in humans [60,61]. Isoforms produced by alternative splicing might have different functions. For example, two splice variants of receptor expression-enhancing protein 6 gene have distinct functions in the retina [62]. However, in the majority of cases, isoforms from alternative splicing have not been well investigated. In this study, we identified two human MRAP2 splice variants, MRAP2b and MRAP2c. Additional studies are needed to confirm which tissues express these alternative splicing variants. Human MRAP1 also has two isoforms: MRAP1a and MRAP1b. The potential effects of the two MRAP1 and three MRAP2 isoforms on hMC3R and hMC4R pharmacology were investigated herein.
Human MRAP1s and hMRAP2s have several similar structural features as MRAP1 and MRAP2 of other species. The conserved motif, LKAHKHS in hMRAP1 or LKAHKYS in hMRAP2, is required for reverse topology ( Figure 1B) [31,63,64], and the corresponding reverse topology motif is also observed in MRAP1 and MRAP2 orthologs of other species [65]. In addition, YEYY motif is apparent in both MRAP1 and MRAP2 of nearly every vertebrate examined [66] and plays an important role for MC2R activation [67]. However, the activation motif (LDYL) was only found in the hMRAP1 paralogs but not in hMRAP2s ( Figure 1B), which is a critical difference between MRAP1 and MRAP2 [65]. MRAP1 paralogs facilitate the activation of hMC2R, but MRAP2 paralogs (without this activation motif) cannot promote MC2R activation in teleosts and tetrapods [64,65,68,69].
Detailed pharmacological studies were performed on potential MRAP regulation of hMC3R. There was no report on the regulation of hMC3R by hMRAP1b, hMRAP2b, and hMRAP2c. Both hMRAP1a and hMRAP2a were reported to decrease the cell surface expression of hMC3R [35,50]. Our data showed that hMRAP1a and hMRAP2c increased, hMRAP1b decreased, and hMRAP2a and hMRAP2b had no effect on the cell surface expression of hMC3R (Figure 2A). Previously, it has been reported that both hMRAP1a and hMRAP2a decrease NDP-MSH-stimulated [35] or increase α-MSH-induced [37,50] cAMP production of hMC3R. The current study is the first to explore potential MRAP modulation of MC3R using ACTH. In this study, four MRAPs (hMRAP1a, hMRAP1b, hMRAP2a, and hMRAP2c) showed similar effects on MC3R signaling, resulting in decreased α-MSHand ACTH-stimulated cAMP levels of hMC3R (MRAP2b only decreased ACTH-stimulated signaling of hMC3R) ( Figure 4 and Table 2). Our findings indicated that hMRAP1b, hMRAP2b, and hMRAP2c might also be involved in regulating hMC3R in distinct ways compared with hMRAP1a and hMRAP2a. In addition, MRAP1 or MRAP2 has been reported to increase ACTH potency at chicken and frog MC3R [44,70]. However, MRAP2s have no effect on ACTH potency of fish (topmouth culter) MC3R [48]. Our current results showed that all MRAPs had no effect on ACTH potency at hMC3R (Table 2). Further studies in MC3Rs from other species are needed to address whether MRAPs change MC3R to an ACTH-preferring receptor.
The regulation of MRAP1s and MRAP2s on hMC4R was also studied. It was reported that hMRAP1a and hMRAP2a decrease the cell surface expression of hMC4R [35,50]. Our current results showed that hMRAP1a, hMRAP2a, and hMRAP2c increased the cell surface expression of hMC4R whereas hMRAP1b and hMRAP2b had no effect ( Figure 5A). For signaling, conflicting results were reported previously: hMRAP1a was reported to either decrease NDP-MSH-stimulated [35] or increases [50,71] or does not affect α-MSHstimulated [37] signaling of hMC4R. Our data showed that MRAP1b decreased α-MSHand ACTH-induced cAMP generation, while MRAP1a only decreased α-MSH-stimulated cAMP signaling of hMC4R (Figure 7 and Table 4). Inconsistent results were also reported on hMRAP2-regulated hMC4R signaling: MRAP2a has no effect [51] or increases [50,71] α-MSH-stimulated signaling of hMC4R. MRAP2a does not affect ACTH-induced [51] or decreases NDP-MSH-stimulated [35] cAMP levels of hMC4R. Our study demonstrated that all MRAP2s decreased α-MSH-stimulated cAMP signaling but had no effect on ACTHinduced signaling of hMC4R (Figure 7 and Table 4). Similar results were also observed in chicken MC4R, in which MRAP1 and MRAP2 do not affect ACTH-stimulated signaling but inhibit α-MSH-induced signaling [44]. Our findings suggested that the new isoforms studied herein, hMRAP1b, hMRAP2b, and hMRAP2c, could modulate MC4R signaling.
An interesting observation reported previously is that MRAPs might change MC4R preference to different endogenous ligands. Previous results showed that hMRAP1a or hMRAP2a increase [51,71] or do not affect α-MSH potency at hMC4R [37]. Two endogenous hormones, α-MSH and ACTH, are used in this study to investigate whether MRAPs change ligand potencies of hMC4R. Our results showed that MRAP1s and MRAP2s could not change α-MSH potencies of hMC4R (Table 4). For ACTH, there is no report on whether MRAP1 affects ACTH potency at hMC4R, and hMRAP2a was reported to increase ACTH potency at hMC4R [51]. MRAP2 increase of ACTH potency of MC4R has also been observed in several other species, including pig, chicken, frog, and zebrafish [44,46,51,70,72]. However, this phenomenon was not observed in several other species, such as orangespotted grouper [73], Nile tilapia [74], topmouth culter [47], and snakehead [49]. Our results showed that only hMRAP1a significantly increased ACTH potency, and the other MRAPs had no effect on ACTH potency at hMC4R (Table 4). We conclude that the MRAP effect on ACTH potency at MC4R might be species-dependent.
Human MC4R has modest basal cAMP signaling [58]. The loss of constitutive activity in MC4R mutations is considered as one cause of obesity [75,76]. The higher constitutive activity of hMC4R is pivotal in regulating energy homeostasis [77] and increased basal activity of MC4R might protect against obesity. Human MRAP1a was shown to increase [37,71] or have no significant effects on the constitutive activity of hMC4R [35,36]. The ratios between hMC4R and hMRAP1a have an important effect on the basal activity of hMC4R [37,71], which might result in the inconsistent results. Our finding showed that both hMRAP1a and hMRAP1b significantly increased hMC4R basal activity (Table 4).
Splicing variants with different specific domains provide a nature-made opportunity to study the functions of a specific domain. Lab-generated truncated MRAPs indicate that N-terminus, but not C-terminus, of hMRAP1 has crucial roles in regulating hMC2R trafficking and signaling [81], and similar phenomena have been observed in the cells heterologously expressing truncated mouse MRAP1 and hMC2R [31,63]. The present study found that hMRAP1a and hMRAP1b with different C-termini played different roles in regulating hMC3R or hMC4R pharmacology (Tables 2 and 4), indicating that C-termini of MRAP1 is important for modulation of MC3R and MC4R signaling. Results of MRAP2 deletion mutants and chimeras indicate that the C-terminus of MRAP2a is important for trafficking and signaling of GPCRs, such as ghrelin receptor, orexin receptor, and prokineticin receptor [79,82]. Similar to MRAP1s, hMRAP2a, hMRAP2b, and hMRAP2c are also different in C-termini. However, our study suggested that C-termini of hMRAP2s played distinct roles in regulating MC3R/MC4R trafficking with similar effects on signaling (Figures 2 and 5, Tables 2 and 4). Collectively, these results suggest that distinct regions of MRAP1s or MRAP2s might have different roles in regulating diverse GPCRs, resulting in increased complexity of MRAPs in modulating GPCRs.

Conclusions
In summary, MRAP1b and two newly identified MRAP2 splicing variants, hMRAP2b and hMRAP2c, had potential roles in regulating MC3R and MC4R pharmacology. All MRAPs except MRAP2b decreased α-MSHand ACTH-stimulated cAMP generation of hMC3R. MRAP1s and MRAP2s showed opposite effects on the basal activity of hMC4R, with MRAP1s increasing and MRAP2s decreasing the basal activities of hMC4R. MRAP1a conferred increased potency for ACTH at the hMC4R whereas the other MRAPs had no effect on ACTH potency. These findings suggest the complexity of MRAPs in modulating MC3R/MC4R and provide a new opportunity for regulating MC3R and MC4R signaling.

Data Availability Statement:
The raw data supporting the conclusions of this article will be made available by the authors upon request, without undue reservation.

Conflicts of Interest:
The authors declare that there is no conflict of interest that would prejudice the impartiality of this study. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.