MC₁ receptors are constitutively expressed on leucocyte subpopulations with antigen presenting and cytotoxic functions

Abstract

The expression of melanocortin MC₁ receptors on human peripheral lymphocyte subsets was analysed by flow cytometry using rabbit antibodies selective for the human MC₁ receptor and a panel of monoclonal antibodies against lymphocyte differentiation markers. The MC₁ receptor was found to be constitutively expressed on monocytes/macrophages, B-lymphocytes, natural killer (NK) cells and a subset of cytotoxic T-cells. Interestingly T-helper cells appeared to be essentially devoid of MC₁ receptors. The results were confirmed by RT-PCR which indicated strong expression of MC₁ receptor mRNA in CD14⁺, CD19⁺ and CD56⁺ cells. However, only a faint RT-PCR signal was seen in CD3⁺ cells, in line with the immuno-staining results that indicated that only part of the CD3⁺ cells (i.e. some of the CD8⁺ cells) expressed the MC₁ receptor. The MC₁ receptors' constitutive expression on immune cells with antigen-presenting and cytotoxic functions implies important roles for the melanocortic system in the modulation of immune responses.

Introduction

The neuroendocrine transmitter α-MSH exerts a large array of central and peripheral effects by virtue of its binding to five subtypes of melanocortin receptors, MC_1-5. The MC receptors are G-protein coupled and all five of them couple in stimulatory fashion to adenylate cyclase. The MC receptors are monomeric proteins thought to be built of seven transmembrane α-helices. The human MC receptors have a length of between 297 and 360 amino acids, with the proteins' extracellular N-terminus ranging 23–74 amino acids. The MC receptors show about 40–60% over all sequence homology, but share no obvious sequence similarity in their N-terminal domains [1]. The MC₁ receptor was originally demonstrated on melanocyte-derived melanoma cells [1,2] and on melanocytes [3], but there is now also evidence that it is expressed on monocytes/macrophages [4,5], neutrophils [6] and dermal microvascular endothelial cells [7]. The reported effects of α-MSH on the immune system are primarily anti-inflammatory and include the reduction of pro-inflammatory cytokines IL-1β and TNF-α [8], inhibition of NF-κB activation [9], induction of IL-10 synthesis [4], down-regulation of adhesion molecules on endothelial cells [10] and inhibition of neutrophil migration [11]. α-MSH is also shown to inhibit hapten-induced delayed-type hypersensitivity [12] and to reduce adjuvant arthritis [13]. A role for endogenous α-MSH in the control of the immune system appears probable. Many cells of the body, including the cells of the immune system express the α-MSH precursor pro-opiomelanocortin (POMC) and are shown to release MSH peptides [1]. Moreover, during various conditions of inflammation such release of MSH peptides is seen to be enhanced [1]. Still, the expression of MC₁ on the main subsets of human lymphocytes has not been examined.

In this report we demonstrate for the first time the presence of the MC₁ receptor on lymphocyte subsets by use of immunofluorescence labelling with polyclonal antibodies selective for the MC₁ receptor and flow cytometry. Moreover we confirm the immuno-labelling data by mRNA analysis using RT-PCR with specific primers for the MC₁ receptor. Our results show that, besides being expressed on the monocytes/macrophages, the MC₁ receptor is constitutively expressed on B-cells, NK-cells and a subset of cytotoxic T-cells.

Materials and methods

Antibodies

The isotypes, specificities and sources of the monoclonal antibodies (mAbs) against different phenotypic lymphocyte differentiation markers used in this study are listed in Table 1.

Table 1.

Specificities of mAbs used herein

Surface marker	mAb/clone	Isotype	Main reactivity	Source
CD45	2B11 and	IgG1	Leucocytes	Dako
	PD7/26
CD3	T3–4B5	IgG1	T cells and subsets of thymocytes	Dako
CD4	MT310	IgG1	MHC class II restricted T cells	Dako
CD8	DK25	IgG1	MHC class I restricted T cells	Dako
CD14	TUK4	IgG2a	Monocytes/macrophages	Dako
CD19	HD37	IgG1	B cells	Dako
CD56	MY 31	IgG1	Isoform of NCAM, NK-cells	Becton-Dickinson,
Negative control	DAK-G01	IgG1	A. niger glucose oxidase	Dako
Negative control	DAK-G05	IgG2a	A.niger glucose oxidase	Dako

The polyclonal M1-Y and M2-Y antibodies against the human MC₁ receptor were previously described [14]. They had been raised by immunization of rabbits with, respectively, two unique 15 and 11 amino acid long synthetic peptides corresponding to the N-terminal of the MC₁ receptor. Their specificity had been verified using various tests, which included their ability to stain cells expressing the recombinant MC₁ receptor, while not staining cells lacking the MC₁ receptor [14–16].

Isolation of peripheral blood mononuclear cells

Peripheral blood samples were donated by 5 healthy adult individuals, diluted 1: 2 with Tris-buffered Hank's salt solution, pH 7·2 (TH) and subjected to Ficoll-isopaque (Lymphoprep, Nycomed, Oslo, Norway) gradient centrifugation. The interface containing peripheral blood mononuclear cells (PBMC) was collected, washed in TH, supplemented with 0·2% human serum albumin (HSA) and antibiotics, and used in the immunofluorescence labelling experiments. Subpopulations of PBMC, were separated by Dynabeads (see below) and used for total RNA extraction.

Two-colour immunofluorescence staining and flow cytometric analysis of isolated PBMC

One hundred thousand living cells per well were plated into U-shaped microtitre plates in TH containing 0·2% HSA and 0·02% NaN₃ and incubated with an appropriate concentration of rabbit-anti human MC₁ antibodies for 30 min on ice. After the incubation the cells were washed three times in the same medium, containing 0·2% HSA and 0·02% NaN₃. During the first wash the cells were centrifuged through a layer of neat FCS. After washing, FITC-conjugated swine antirabbit antibodies (DAKO, Glostrup, Denmark) were applied for 30 min in the dark on ice followed by a further wash, as described above. A final 30 min incubation with a phycoerythrin-conjugated mAb for a lymphocyte differentiation antigen was then applied, followed by another washing step. Appropriately labelled isotype-matched irrelevant mAbs (Dako or Becton-Dickinson, Mountain View, CA; Table 1) and normal rabbit serum were used as controls for the determination of unspecific fluorescence. Flowcytometric analyses were performed using a FACScan flow-cytometer (Beckton Dickinson). The fluorescence data were displayed as dot plots and quantified using Cell Quest software. Anti CD45 mAbs were used to verify the staining of the whole leucocyte population.

Separation of subpopulations of PBMC

Subpopulations of PBMC from individual donors were obtained by positive selection using magnetic separation with goat antimouse immunomagnetic beads (Dynal) labelled with mAbs specific for T-cells (anti-CD3), B cells (anti-CD19), monocytes/macrophages (anti-CD14) and NK cells (anti-CD56), as previously described [17] and used for preparation of total RNA. The cells bound to the immunomagnetic beads were subjected to five washes with 10–20 times the initial cell/bead volume of ice-cold PBS, allowing nonbound cells to be completely removed. A minimum of 1000–1500 cells/sample were counted using a light microscope in order to ascertain that only cells bound to beads were present in the selected population [17]. The cells attached to the beads were then lysed, and the lysate was used for extraction of total RNA.

Total RNA extraction and RT-PCR analysis

Total RNA was extracted from cell lysates of positively selected subpopulations of PBMC (see above) by the acid guanidinum thiocyanate-phenol-chloroform method [18]. Single strand cDNA copies were made from 1 μg of total RNA using random hexamers and murine leukaemia virus reverse transcriptase (PE Biosystems Nordic, Stockholm, Sweden). The RT step was performed at 42°C for 15 min followed by denaturation at 99°C for 5 min. The cDNA was used as template for PCR with primer pairs specific for the human MC₁ receptor (MC₁ forward 5′-TGGTGAGCTTGGTGGAGAACGC-3′ and MC₁ reverse 5′-TCTTGAAGATGCAGCCGCACG-3′). The PCR reaction was performed in a total volume of 20 µl containing 20 pmol of each primer, 400 µm dNTPs and 3 mm MgCl₂, 100 mm KCl, 20 mm Tris pH 8·8, and 1 U Taq polymerase (Fermentas AB, Vilnius, Lithuania) and 2 µl cDNA. The PCR program was as follows: initial denaturation at 94°C for 5 min, followed by 35 amplification cycles of 45 s at 94°C, 45 s at 63°C, and 1 min at 72°C, followed by a final extension at 72°C for 10 min The primers used would amplify a 686-bp DNA fragment corresponding to the MC₁ receptor sequence. Samples containing the pE-11D plasmid with the cloned human MC₁ receptor DNA served as a positive control [2]. For β-actin primers and protocols were as described [17]. PCR products were analysed by 2% agarose gel electrophoresis, stained by ethidium bromide and visualized by UV illumination. Amplified products were sequenced using an ABI Prism 377 sequencer and ABI sequencing kit (Perkin Elmer).

Results

Expression of the MC₁ receptor on peripheral blood mononuclear cells

Peripheral blood lymphocytes were isolated from five donors and double-stained using monoclonal antibodies against lymphocyte differentiation markers and the polyclonal antibody M1-Y against the MC₁ receptor. The results from 5 individual donors are summarized in Table 2 and illustrated in Fig. 1. As can be seen practically all CD19 positive cells (B-lymphocytes) were stained by the MC₁ receptor antibody. Also the majority of CD14 positive cells (monocytes) and CD56 positive NK cells were stained by the MC₁ receptor antibody. Moreover, a subset of CD8⁺ cells expressed MC₁ receptors. However, very few (i.e. virtually none) of the CD4⁺ cells stained for the MC₁ receptor (Table 2 and Fig. 1b). Additional experiments, where the MC₁ receptor M2-Y antibody was used, showed essentially the same pattern for the frequency of MC₁R positive cells of the CD19, CD14, CD56, CD4 and CD8 positive leucocyte subsets (data not shown). Unspecific staining was excluded by using normal rabbit serum and isotype matched monoclonal antibodies (Table 1 and Fig. 1f). Thus, as can be seen from Fig. 1f an irrelevant IgG1 isotype-matched mAb did not bind to PBMC incubated with normal rabbit serum. Similar results as in Fig. 1f were obtained using an irrelevant IgG2a isotype-matched mAb and normal rabbit serum (data not shown).

Table 2.

Expression of the MC₁ receptor (MC₁R) on subpopulations of freshly isolated human PBMC. The table represents data from 5 different healthy adult donors using mAbs according to Table 1 for surface marker detection and the M1-Y antibody for MC₁ receptor detection in conjunction with FACS analysis. Values are expressed as the means ± SEM

Surface marker	Marker expression (% of total cell number)	Proportion of MC1R positive cells (% of marker positive cells)
CD14	10 ± 2	100
CD19	12 ± 4	95 ± 5
CD4	38 ± 6	4 ± 2
CD8	30 ± 7	39 ± 15
CD56	14 ± 2	86 ± 8

Fig. 1.

Dual colour flow cytometric analysis of MC₁ receptor (MC₁R) expression on subsets of isolated PBMC from a healthy adult donor. Anti-CD14 mAb was used to stain monocytes/macrophages (a), anti-CD4 mAb to stain T_H cells (b), anti-CD8 mAb to stain CTLs (c), anti-CD19 mAb to stain B-cells (d) and anti-CD56 mAb to stain NK cells (e). Staining of MC₁Rs was done using the M1-Y antiserum. An irrelevant IgG1 isotype mAb and normal rabbit serum served as one of the negative controls (f). (Similar results as in F were obtained using an irrelevant IgG2a isotype mAb and normal rabbit serum; data not shown).

Message for MC₁ receptors is expressed in isolated subpopulations of PBMC

In order to investigate the expression of MC₁ receptors in lymphocyte subsets RT-PCR was applied using specific primers for the human MC₁ receptor on cDNA obtained from lymphocytes positively selected for CD3, C14, CD19 and CD56 by mAb-coupled immunomagnetic beads. As shown in Fig. 2 CD14, CD19 and CD56 selected cells invariably gave PCR products of the size corresponding to that of the MC₁ receptor (686 bp). By contrast CD3⁺ cells gave only weak or faint signals (Fig. 2a). The lack of the MC₁ mRNA signal in the latter cells were not due to differences in cDNA levels. This can be seen from Fig. 2a where the bands are compared with the band for β-actin PCR product (540 bp) of the identical samples. Figure 2b compares the MC₁ receptor band of CD56 cells with the band obtained using a plasmid containing the human MC₁ receptor. As seen both bands show similar mobility on the agarose gel. The identity of the presumed MC₁ receptor DNA was verified by sequencing of the PCR product isolated from the CD56 cells. The sequence corresponding to nucleotides 138–523 of the PCR product could be read, and matched that of nucleotides 603–989 of the reported [2] MC₁ receptor DNA sequence.

Fig. 2.

Detection of MC₁ receptor transcripts in human peripheral blood mononuclear cells by RT-PCR. Subpopulations of PBMCs were obtained by positive selection and used for preparation of cDNA followed by RT-PCR amplification and detection on agarose gels using ethidium bromide staining. (A) From left to right lanes represent RT-PCR bands using MC₁R and β-actin selective primers for cells positively selected for CD56, CD19, CD14 and CD3. (B) RT-PCR bands obtained using an MC₁R plasmid or cDNA from cells positively selected for CD56. Indicated by arrows to the right are the mobility of standard DNA, nt 300–700. Indicated by the arrow to the left is the 686 nt size of the band amplified from the MC₁R plasmid.

Discussion

The present study gives unambiguous evidence for the expression of the MC₁ receptor on subsets of peripheral blood mononuclear cells of healthy adult humans. The identity of the MC₁ receptor is verified on several levels. Firstly, the same pattern for the distribution of the MC₁ receptor on the lymphocyte subsets was seen using two different antibodies that had been raised by immunizing rabbit with two different peptides corresponding to 15 and 11 amino acid long epitopes in the N-terminus of the human MC₁ receptor [14]. Simultaneous 2-colour immunofluorescense flow cytometry showed that all or almost all CD14⁺, CD19⁺ and CD56⁺ cells express MC₁ receptors on their surface. By contrast no or only very few CD4⁺ cells were MC₁ receptor positive, while approximately 1/3 of CD8⁺ cells stained for the MC₁ receptor. Secondly, the expression of the MC₁ receptor message was demonstrated in lymphocyte subsets with a distribution entirely compatible with the immunostaining. Thus, by use of RT-PCR a band with the correct size for an MC₁ receptor message was demonstrated in CD14, CD19 and CD56 positive cells. By contrast CD3 positive cells (i.e. a T-cell population consisting of the CD4⁺ and CD8⁺ cells) showed only a faint signal corresponding to the MC₁ receptor product, reflecting that only a part of the CD3⁺ population (CD8⁺ cells) expressed the message for the MC₁ receptor. Thirdly, unambiguous verification for the MC₁ receptor identity of the PCR amplified band was reached by the DNA sequencing.

According to the present results four important immune cell populations: monocytes/macrophages, B cells, CTLs and NK cells express MC₁ receptors constitutively. Thus, MC₁ receptors seem to be expressed in immune cells with antigen presenting and cytotoxic functions. Monocytes/macrophages participate both in the processing and presentation of the antigen to T-helper (T_H) cells. After the activation with the aid of T_H cells, macrophages mediate the delayed type hypersensitivity. B-cells are also important antigen presenting cells, besides their effector functions. CTLs and NK-cells are cytotoxic effector cells involved in the defence against viral infections and tumour cells.

The melanocortic peptides are known to affect immune and inflammatory responses, but their exact site(s) and mode of actions are still not very well defined. The melanocortic peptides act on the five subtypes of melanocortin receptors MC_1-5. Melanocortic peptides are derived from a larger POMC precursor that may generate the shorter ACTH, α-, β- and γ-MSH peptides. All of these peptides are capable of activating the MC₁ receptor, with the largest activity being shown by α-MSH. The MC₂ receptor on the other hand is responsive only to ACTH (and fragments of ACTH down to ACTH(1–24)). The MC_3-5 receptors are also responsive to the various MSH peptides with some variation in relative sensitivity compared to the MC₁ receptor.

Earlier studies indicate that the melanocortins have a complex and multifaceted role in regulating the immune system. It is well established that POMC is expressed in cells of the immune system and is processed to ACTH in cells such as lymphocytes and macrophages [19]. Processing into α-MSH may also occur and has been demonstrated in a monocytic cell line [20]. Interestingly the α-MSH production in these cells was shown to be strongly up-regulated by stimulation with IL-6, TNF-α or concanavalin A [20]. Moreover, γ-MSH immunoreactivity was demonstrated in neutrophils of human eosinophilic patients [21].

The overall effect of melanocortic peptides appears to be anti-inflammatory, something that has been demonstrated in a number of animal models of inflammation where α-MSH was administered exogenously [1]. Exogenously applied α-MSH also inhibits the release of immuno-stimulatory and pro-inflammatory cytokines IL-1β and TNF-α in whole blood, as well as in isolated peripheral blood mononuclear cells, after stimulation with bacterial lipopolysacharide [7,22]. These effects have earlier been attributed to the presence of MC₁ receptors in monocytes/macrophages [1], a location for the MC₁ receptor, which is also confirmed by the results of the present study. The fact that MC₁ receptors are expressed not only on monocytes/macrophages, but also on the antigen presenting B-cells and on cells with cytotoxic functions (CTLs and NK-cells) strongly suggest that anti-inflammatory and immuno-modulatory effects of the melanocortic peptides are also mediated by actions on these cells. E.g. a previous study report that, depending on the concentrations used, ACTH and α-MSH could either augment or inhibit the IgE synthesis in an in vitro mononuclear cell system [23]. Our finding that only about one third of the CD8⁺ population express the MC₁ receptors is also quite interesting. It thus appears that the cytotoxic T-lymphocyte subclass is heterogeneous with respect to MC₁ receptor expression, something that might be taken as an indication for quite specific roles of the melanocortic system in propagating cytotoxic T-cell responses. The presence of the MC₁ receptor on NK cells may also indicate a role of the melanocortic system for the regulation of MHC nonrestricted cell cytoxocicity.

It has been shown that the in vivo administration of α-MSH can induce hapten specific tolerance [12], thus further indicating very specific mode of actions of the POMC peptides in immune regulation. The essential complete absence of MC₁ receptors on T_H cells (results of present study) thus seems to disfavour a role of these cells in mediating the α-MSH effect in the delayed-hypersensitivity response and for the initiation of hapten specific tolerance, at least if these effects were to be propagated by the activation of MC₁ receptors. A role for monocytes/macrophages is supported by the observation that hapten-specific α-MSH induced tolerance is dependent on IL-10 [12], a cytokine produced by the monocytes/macrophages. It is thus tempting to speculate that it is indeed the MC₁ receptors on the monocytes/macrophages that are responsible for the development of the tolerance, rather than MC₁ receptor lacking CD4⁺T cells. Studies directed towards how the melanocortins affect the interplay of T_H cells and macrophages/monocytes might give insight into these questions.

There are some indications that other MC receptors besides the MC₁ receptor might also have roles in the regulation of the immune system. Thus, expression of MC₃ receptors has been demonstrated in murine macrophages [24]. There is also evidence for the expression of MC₅ receptors on mouse B-lymphocytes [25]. Moreover, the expression of both MC₁ and MC₅ receptors were reported on a human mast cell line [26]. Finally, it should be mentioned that part of the α-MSH mediated effects may even be conveyed via a nonmelancortin receptor mediated pathway that mediate anti-inflammatory effects via the inhibition of NF-κB activation [1].

In summary our study provides evidence that MC₁ receptors are expressed on subsets of lymphocytes, namely B-cells, NK cells and a subpopulation of cytotoxic T cells. By contrast the MC₁ receptor was absent on T-helper cells. Moreover we confirm that MC₁ receptors are present on monocytes/macrophages. The expression of MC₁ receptors on antigen presenting cells, and cells with cytotoxic functions, lends support to specific roles for the melanocortic system in the regulation of the immune system.