Alpha-melanocyte stimulating hormone (α-MSH) is a neuropeptide exhibiting anti-inflammatory activity in experimental models of autoimmune diseases. However, no studies thus far have examined the effects of α-MSH on systemic lupus erythematosus (SLE). This study aimed to determine the effects of an α-MSH agonist in induced murine lupus. Here we employed female Balb/cAn mice in which lupus was induced by pristane. Groups of lupus animals were treated daily with the α-MSH analogue [Nle4, DPhe7]-α-MSH (NDP–MSH) (1·25 mg/kg) injected intraperitoneally or saline for 180 days. Normal animals comprised the control group. Arthritis incidence, plasma immunoglobulin (Ig)G isotypes, anti-nuclear antibodies (ANA) and plasma cytokines were evaluated. Renal function was assessed by proteinuria and histopathological lesion. Glomerular levels of IgG, α-smooth muscle actin (α-SMA), inducible nitric oxide synthase (iNOS), C3, CD3, melanocortin receptors (MCR)1, corticotrophin-releasing factor (CRF) and α-MSH was estimated by immunohistochemistry. When compared with normal controls, lupus animals exhibited increased arthritis, IgG levels, ANA, interleukin (IL)-6, IL-10, proteinuria and mesangial cell proliferation together with glomerular expression of α-SMA and iNOS. Glomerular expression of MCR1 was reduced in lupus animals. NDP-MSH treatment reduced arthritis scores by 70% and also diminished IgG1 and IgG2a levels and ANA incidence. In the glomerulus, NDP–MSH treatment reduced cellularity by 50% together with reducing IgG deposits, and expression levels of α-SMA, iNOS and CRF were also all decreased. Taken together, our results suggest for the first time that α-MSH treatment improves several parameters of SLE disease activity in mice, and indicate that this hormone is an interesting potential future treatment option.
Systemic lupus erythematosus (SLE) is an autoimmune syndrome exhibiting heterogeneous outcomes that can affect multiple organs. The disease is characterized by tissue damage induced by chronic inflammation, the presence of autoantibodies and autoreactive lymphocytes and cytokine imbalances . The kidneys are particularly susceptible to injury, as almost 60% of SLE patients will present renal involvement at some point. Immune complex deposition and inflammatory milieu are the key factors that amplify kidney lesions [2, 3].
The neuro-immuno-endocrine system is known to participate in autoimmune diseases regulation. A disturbed hypothalamic–pituitary–adrenal (HPA) axis response has been associated with SLE in humans and in animal models [4, 5]. Accordingly, pituitary polypeptide adrenocorticotrophic hormone (ACTH), a melanocortin family component, was widely used during the 1950s to treat both lupus and nephrotic syndrome. However, ACTH was quickly replaced by oral synthetic corticoid treatment due to administration advantages .
Today, interest in melanocortin neuropeptides has re-emerged through α-melanocyte-stimulating hormone (α-MSH). This peptide, derived from ACTH, has arisen as a new tool to control the inflammatory process . Multiple experimental studies have described the anti-inflammatory properties of α-MSH during ischaemia–reperfusion injury  and in autoimmune diseases including uveitis , encephalomyelitis , contact hypersensitivity  and arthritis . Although anti-inflammatory effects have also been reported in models of kidney disease, improving haemodynamic failure and reducing the severity of renal lesions [13-15], limited effects in restoring impaired renal function and proteinuria have been described [14, 16, 17].
Melanocortin receptors (MCR) are present in multiple tissues and cell types, including the glomerulus, renal tubules, neutrophils, monocytes, dendritic cells and lymphocytes . MCR1 activation prevents nuclear factor kappa B (NF-κB) translocation and reduces the production of several inflammatory mediators . In experimental models of nephritis, α-MSH increases animal survival and improves glomerular filtration. Simultaneously, it decreases apoptosis, fibrosis, neutrophil infiltration, cytokine expression and reactive oxygen species production [13-15].
Induced lupus models that resemble the SLE human disease are of pivotal relevance because they allow the study of mechanisms involved in the disease onset . Among these models, intraperitoneal injection of mice with the hydrocarbon oil pristane (2,6,10,14-tetramethylpentadecane) is remarkable in that it faithfully exhibits the clinical manifestation and laboratorial abnormalities that are characteristic of SLE . In addition, the pristane lupus-like model can be induced in the Balb/c strain, which is a more reliable colony of animals compared with spontaneous lupus mice .
Considering that SLE patients present antigen presentation defects, cytokine release, arthritis and nephritis, and suffer alterations in the HPA axis, a plausible connection exists between melanocortins and lupus pathophysiology. However, no study discusses the effect of α-MSH on lupus. Therefore, the present study evaluated the effects of an α-MSH analogue treatment on pristane-induced murine lupus.
Materials and methods
This protocol was approved by the local Research Ethic Committee of Hospital Clinics/School of Medicine, University of São Paulo, under process number 0612/09·09. All animal care and experimental procedures were developed in strict conformity with Universities Federation for Animals Welfare and the Committee of Brazilian College of Experimental Animals (COBEA).
Because the pristane-induced model presents female preponderance as human disease , only adult female mice were employed. Twenty adult BALB/cAnUnib mice, 8·5 weeks old and weighing 20–25 g, were employed. The animals were kept in conventional colony rooms at the Rheumatology Division facility with controlled temperature (23 ± 1°C) and a 12-h light/dark cycle with food in pellets and water ad libitum. To avoid stress interference, no experiments were initiated until 15 days after animal arrival.
Experimental design and α-MSH treatment
Lupus was induced in 15 mice with a single intraperitoneal (i.p.) injection of 0·5 ml pristane (2,6,10,14 tetramethylpentadecane; Sigma Chemical Co., St Louis, MO, USA) that had been previously filtered through a 0·22 μm membrane (Millipore, Billerica, MA, USA) . Five control mice received the same volume of 0·9% saline. Peripheral blood (250 μl) was collected from the submandibular vein at baseline and 180 days after treatment.
Alpha-melanocyte-stimulating hormone analogue [Nle4, DPhe7]-α-MSH (NDP-MSH; Melanotan-I) was purchased from Peptides International (Louisville, KY, USA). The α-MSH analogue [Nle4, DPhe7]-α-MSH was chosen due to its higher stability, biodisponibility and resistance to proteolytic degradation compared with the native peptide. In addition, it does not interfere with endogenous glucocorticoid production [24, 25].
The NDP-MSH treatment was prepared according to the manufacturer's instructions. NDP-MSH was first dissolved in acetic acid solution and then added to 0·9% saline (final acid concentration 0·06%). Daily treatment with the hormone analogue started just after pristane injection and was always performed between 8:00 and 10:00 h [26, 27]. Mice with pristane-induced lupus (LM) were allocated randomly in two groups of five or 10 animals: LM treated with saline (0·9%, pH = 5, LM) or LM treated with the NDP-MSH 1·25 mg/kg day i.p. (LM-MSH). The control group (control) consisted of normal animals treated with saline i.p. Before killing, the animals were anaesthetized with xylazine (5 mg/kg) and ketamine (50 mg/kg) administered i.p.
Total leucocyte counts
Total peripheral blood leucocytes were counted in a Neubauer chamber diluted 1:20 with Turk.
Immunoglobulin (Ig)G1, IgG2a, IgG2b and IgG3 levels in diluted plasma (1:15 000) were measured by enzyme-linked immunosorbent assay (ELISA). IgG isotype standards from Southern Biotechnology (Birmingham, AL, USA) were used for standard curve fitting at concentrations of 15·62–1000 ng/ml.
Anti-nuclear and cytoplasmic autoantibodies (ANA) were detected by indirect immunofluorescence using a home-made HEp-2 cells assay. Slides were incubated for 40 min with 1:200 murine plasma in phosphate-buffered saline (PBS), washed and then incubated for 30 min with 1:50 fluorescein isothiocyanate-conjugated goat anti-mouse IgG H + L (Southern Biotechnology). After another wash, slides containing cells were analysed with a fluorescence microscope.
Quantification of plasma cytokines
Cytokine levels [interleukin (IL)-6, IL-10, tumour necrosis factor (TNF)-α and interferon (IFN)-γ] were analysed using a fluorokine mouse multi-analyte profile kit (R&D Systems, Minneapolis, MN, USA). According to the manufacture's protocol, plasma samples were diluted 1:2 and incubated with anti-cytokine beads overnight at 4°C. Events were counted on a luminex 200 system (Luminex, Austin, TX, USA). Data analyses were performed using the Milliplex Analyst version 3·5.5·0 (Vigenetech, Carlisle, MA, USA). A five-parameter regression formula was used to calculate the sample concentration. Data were expressed in pg/ml.
The arthritis severity was assessed 180 days after pristane inoculation, using a scoring system that considers oedema, erythema and number of affected joints as: 1 = small, 2 = moderate, 3 = marked and 4 = severe. Thus, the maximum total score per animal was 16 .
Renal function evaluation
Before killing, the animals were individually contained to collect spot urine. All samples were analysed using proteinuria measurements kits (Labtest Diagnosis, Minas Gerais, Brazil). To estimate urinary albumin concentration, 10 μl of urine was run on 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and was Coomassie-stained. A pattern of bovine serum albumin (BSA) (Sigma Chemical Co.) was used to adjust the standard curve with concentrations from 0·0625 to 0·5 mg/ml. The bands were measured by densitometric analyses with ImageJ software.
After killing, the kidneys were perfused in situ with saline, the right kidney was removed and fixed in 10% buffered formalin, embedded in paraffin and sectioned at 3 μm thickness in the transversal plane containing the renal long axis. Slides were stained using haematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) stains to highlight the glomerulus and pricrosirius red to stain collagen fibres. Cellularity was quantified on H&E-stained slides by counting the total glomerular cell nuclei. At least 30 glomeruli/slides were assessed, and the results were expressed as number of nuclei per glomerulus. Sections stained with PAS were graded as: 1+ = mild focal mesangial hypercellularity; 2+ = moderate mesangial hypercellularity; 3+ = complex endocapillary hypercellularity sometimes with mild sclerosis or necrosis; 4+ = severe endocapillary proliferative glomerulonephritis with necrosis or crescent formation. Scores ≥ 2+ were considered to be positive . Pricrosirius red-stained slides were analysed under polarized light using an Olympus camera attached to an Olympus BX-51 microscope (Center Valley, PA, USA), and the collagen area was determined based on positive staining in the image analyses system.
Glomerular expression of α-smooth muscle actin (α-SMA), T cell marker CD3, complement component 3 (C3), corticotrophin-releasing factor (CRF), inducible nitric oxide synthase (iNOS), α-MSH, MCR1 and IgG were determined by immunohistochemical analyses in deparaffinized kidney sections. After rehydration, the endogenous peroxidase activity was ablated by incubation in 3% hydrogen peroxide for 10 min. Next, the sections were incubated with Tris/ethylenediamine tetraacetic acid (EDTA) (10 mM/1 mM buffer, pH 9·0) for 25 min and incubated with a biotin/avidin blocking solution (Dako, Glostrup, Denmark). Primary antibodies anti-α-SMA (ab5694; 1:100; Abcam, Cambridge, UK), anti-CD3 (ab5690; 1:100; Abcam), anti-C3 (sc-28294; 1:50; Santa Cruz Biotech, Santa Cruz, CA, USA), anti-CRF (h-019-06; 1:100; Phoenix Pharmaceuticals, Burlingame, CA, USA), anti-α-MSH (h-043-01; 1:100; Phoenix Pharmaceuticals), anti-MCR1 (ABIN686287, 1:100; Antibodies-online, Aachen, Germany), anti-iNOS (PA5-16855; 1:200; Thermo Scientific, Rockford, IL, USA) and anti-IgG (LS-C59195; 1:400; LifeSpan, Bellevue, WA, USA) were added to each section and incubated overnight at 4°C. After PBS washing, the slides were incubated for 30 min with EnVision-horseradish peroxidase (HRP), labelled streptavidin–biotin (LSAB)-HRP or Advance-HRP (Dako). Enzyme sandwich reactions were developed using 3,3′-diaminobenzidine (Sigma Chemical Co.), and then the slides were washed, counter-coloured with haematoxylin and mounted with Permount.
The slides were digitally archived using a Pannoramic Scan instrument (software version 1·11·25·0; 3DHistech, Budapest, Hungary) with a ×20 objective and expanded focus. Total levels of collagen fibres, α-SMA, CD3, C3, CRF, α-MSH, MCR1, iNOS and IgG in the glomerulus were quantified using Image-ProPlus version 4·1 software for Windows (Media Cybernetics, Silver Spring, MD, USA).
The positively stained areas were determined by colour threshold. These procedures generated a file containing a colour selection data, which was applied afterwards to the kidney sections. The results of each marker were expressed as the ratio of positively stained area per total glomerular area (μm2).
Significance tests were performed with spss version 17·0.1 software (spss, Inc., Chicago, IL, USA) or GraphPad prism version 6·0 Software (GraphPad software, San Diego, CA, USA). Data were initially compared with the Gaussian curve through the K–S test. Parametric data were analysed using one-way analysis of variance (anova) with pairwise post-test comparisons by the Neuman–Keuls method. Non-parametric data were compared using anova followed by modified Tukey's test and results are expressed as median and interquartile ranges. The incidence of ANA, arthritis and renal lesion were compared using Pearson's χ2 test. P < 0·05 was considered statistically significant.
Inhibitory effect of NDP–MSH on arthritis development
Arthritis incidence and severity is shown in Fig. 1. LM mice exhibited a high incidence of arthritis (80%) and a mean score of 5·2. Treatment with NDP-MSH reduced both incidence (40%) and arthritis score (1·6, P < 0·05 versus LM).
When compared with control animals, both the LM and LM-MSH groups exhibited greater weight gain by day 180 of the experiment (P < 0·01). LM animals presented with leucocytosis 180 days after pristane administration. NDP-MSH treatment had no impact on the white blood cell count. The relationships between right kidney, spleen and liver weights and body weights were not different between the groups.
NDP–MSH treatment reduces IgG1, IgG2a levels and ANA frequency
As shown in Fig. 2, plasma levels of IgG1 and IgG2a increased 180 days after pristane inoculation compared with control animals. NDP-MSH treatment reduced IgG1 and IgG2a levels (P < 0·05 and P < 0·001 versus LM, respectively). IgG2b and IgG3 levels did not differ between the groups.
The plasma ANA assessment is presented in Fig. 3a. One hundred per cent (five of five) of LM exhibited ANA positivity at titres ≥1:200. None of the control animals (none of five) showed ANA positivity (P < 0·01) compared to 30% of LM-MSH animals (P < 0·01 versus LM). Figure 3b illustrates the absence of anti-nuclear antibodies registered in control animals and Fig. 3c shows a positive pattern verified in HEp-2 cells incubated with LM plasma.
NDP–MSH effect on cytokine production
As shown in Table 1, both plasma IL-6 and IL-10 levels increased 180 days after pristane inoculation (P < 0·05 versus control). When compared with the LM group, treatment with NDP-MSH slightly reduced IL-6 and IL-10 levels. The LM-MSH group showed increased TNF-α levels when compared with the LM group (P < 0·01). IFN-γ levels were below the detection range in all groups.
Table 1. Effect of [Nle4, DPhe7]-α-MSH (NDP-MSH treatment on cytokine production of lupus mice (LM) and control mice
(n = 5)
(n = 5)
(n = 10)
Results are expressed as mean ± standard error. *P < 0·05 versus control and #P < 0·05 versus LM. IL = interleukin; IFN = interferon; TNF = tumour necrosis factor.
Figure 4 shows the results of the renal function evaluation in spot urine samples. At the end of the experimental period LM had significant proteinuria (P < 0·001), which was not affected by the NDP-MSH treatment (Fig. 4a). The same pattern was observed in albuminuria (Fig. 4b).
Histopathological analysis data are presented in Fig. 5. LM showed increased glomerular cellularity (Fig. 5a) compared with the control group (P < 0·01). The treatment of LM with NDP-MSH reduced this cellularity (P < 0·05 versus LM). Figure 5b shows the renal damage assessed 180 days after lupus induction. Lupus animals exhibited a profile characterized by moderate histopathological mesangial hypercellularity (score = 2) that was higher than the degree of injury observed in the control animals (P < 0·05). NDP-MSH treatment reduced this score (P < 0·05 versus LM). Notably, 80% of the animals treated with NDP-MSH showed a degree of damage less than 2. The fibrosis evaluation in pricrosirius red-stained slides revealed no difference between the control and LM groups (data not shown).
Effect of NDP–MSH treatment on glomerular IgG deposits and the expression of α-SMA, iNOS, CRF, α-MSH and MCR1
Glomerular deposition of IgG and the expression of α-SMA and iNOS are shown in Fig. 6. When compared with the control group, LM showed increased expression of all markers, which were decreased by NDP-MSH treatment. The figures depict the glomerular appearance of control animals (Fig. 6b), LM (Fig. 6c) and LM-MSH animals (Fig. 6d). The expression of CD3 and C3 did not differ between the groups (data not shown).
Glomerular expression of MCR1 was reduced in LM (P < 0·01 versus control) and in LM-MSH animals (P < 0·001 versus control). The expression of neuropeptides, CRF and α-MSH were similar in control and LM. Treatment with NDP-MSH reduced the expression of glomerular CRF (P < 0·01) but did not affect α-MSH expression (P = 0·12) (Fig. 7).
This study shows, for the first time to our knowledge, that the α-MSH super-analogue NDP-MSH improves several parameters of lupus disease progression in mice.
The chosen experimental model was an important component of our protocol, as pristane presence in human blood suggests a positive correlation with autoimmune disease incidence . In this context, after pristane injection, mice developed a lupus-like syndrome characterized by the presence of hypergammaglobulinaemia, ANA, specific antibodies, arthritis, glomerulonephritis and abnormal cytokine production [21, 22, 29]. The incidence and severity of joint inflammation, one of the most common SLE manifestations, were suppressed by the treatment employed in this study. In accordance with our findings, neuropeptide therapy in other arthritis models was effective in controlling joint inflammation [12, 31]. In human disease, α-MSH has been related to rheumatoid arthritis pathogenesis as a possible compensatory mechanism . In the present study, the ability of NDP-MSH treatment reducing lupus activity was confirmed by the observed reductions in hypergammaglobulinaemia and ANA incidence. Accordingly, α-MSH treatment also reduced IgG1 and IgG2a levels in an airway inflammation model . The importance of this impairment is corroborated by findings that treating pristane-induced lupus with other immunomodulatory compounds also changes these parameters [34, 35].
Lupus immune dysfunction is characterized by uncontrolled cytokine production. Levels of IL-6 and IL-10 are tightly correlated with the activity and severity of the disease in humans  and in experimental models [37-39]. The importance of IL-6 in lupus is evidenced by the reductions in autoantibody production and renal injury observed in IL-6 knock-out mice injected with pristane . Furthermore, clinical studies show that patients exhibit reduced SLE activity after treatment with anti-IL-6R . Interestingly, patients with chronic renal disease present an inverse correlation between IL-6 and α-MSH levels . Accordingly, our lupus animals exhibited increased IL-6 production, and NDP-MSH treatment reduced this production by 37·9%, although not statistically significantly. IL-10 levels are also elevated in mice with pristane-induced lupus, and attenuation of the disease is accompanied by IL-10 reduction . Lupus patients treated with anti-IL10 therapy show impaired disease activity . In our study, we did not observe a significant effect of α-MSH on IL-10 production. In this context, the effect of this neuropeptide on IL-10 remains controversial, as previous studies have reported both reduced IL-10 levels  or no effect [45, 46]. As part of the mechanism of action of α-MSH is on NF-B preventing nucleus translocation, the anti-inflammatory activity certainly involves the reduction of proinflammatory markers [47, 48]. Surprisingly, our treatment with NDP-MSH increases TNF-α levels. Although TNF-α is a proinflammatory cytokine, the role of TNF-α in lupus has not been definitively established. In this context, clinical studies have shown that anti-TNF therapy worsens disease activity and triggers a lupus-like disease . In experimental models, the administration of high doses of TNF-α does not alter and sometimes delays nephritis development [50, 51].
The most severe complication of SLE is renal involvement. In the present study, as well as in humans, not all mice developed nephritis. Previous studies of pristane-induced lupus showed similar incidence rates and the same patterns of glomerular lesion [29, 38]. Regarding markers of renal injury, our schedule of treatment with NDP-MSH did not alter proteinuria/albuminuria, but it did improve other parameters of glomerular injury. Accordantly, similar findings were observed in other nephritis studies, where α-MSH did not affect plasma urea or creatinine but did improve histological markers [14, 15]. The inefficacy of α-MSH on proteinuria was also reported on New Zealand black/white F1 mice . Our data show that lupus is characterized by high glomerular α-SMA expression. In accordance with these results, mesangial cells of patients with LN class II also present increased α-SMA expression [52, 53]. The administration of α-MSH also reduced glomerular expression α-SMA glomerular in models of liver damage  and pulmonary fibrosis . In our view, the glomerular injury induced by pristane is not particularly severe. In our study, we detected no changes in CD3 and C3 expression or in glomerular fibrosis. In this context, only patients with proliferative NL (classes III and IV) present high infiltration of T lymphocytes in the kidney and necrosis .
One interesting point regarding the relationships between melanocortins is the negative feedback effect of α-MSH on CRF . Our study corroborates this finding, as we observed minor expression of CRF in the NDP-MSH-treated group.
The increased expression of iNOS and increases in nitric oxide are associated with renal damage in SLE patients  and mice . Pristane injection also increases iNOS expression in kidney . To our knowledge, this is the first study showing the ability of α-MSH to reduce glomerular iNOS expression in a lupus-like disease model. Studies in vivo and in vitro supplied experimental data to support our results [60, 61]. Indeed, iNOS expression is associated with inflammation and iNOS blockade ameliorates arthritis in animals .
Although the melanocortin receptor MCR1 is widely expressed in the kidney , no study has addressed its expression in human or experimental lupus. Our data show that NDP-MSH treatment reduces glomerular MCR1. In accordance with these results, down-regulation of the receptor has been documented previously following NDP-MSH treatment . Nevertheless, animals with lupus but not receiving treatment also presented minor MCR1 expression. Attenuated MCR1 expression has been described previously during ischaemic kidney disease in rats and in human liver [64, 65]. The reason for this effect remains unknown; however, the explanation could be receptor desensitization by high α-MSH levels during the development of the disease. The current study was unable to test this hypothesis, as we did not measure native α-MSH levels in our animals. In this regard, the effects of native α-MSH were completely overwhelmed by the high doses of its hormone analogue .
Altogether, our results show that α-MSH opposes several factors involved in lupus pathogenesis. The current treatment of lupus patients involves a large therapeutic arsenal that is often incapable of controlling multiple disease symptoms. We believe that the neuropeptide NDP-MSH shows potential to serve as an important new tool in treating this inflammatory condition.
The authors are grateful to Maria Aurora Gomes da Silva, Maria de Fátima de Almeida, Margarete Borges Galhardo Vendramini, Cleonice Bueno and Antonio dos Santos Filho for their skilful technical assistance and to Ana Maria de Lauro Castrucci, Anna Catania, Mauro Perretti, Minoru Satoh and Monique Kowalski Schimitid for helpful discussions. Isac de Castro performed the statistic analyses. This study was supported by grant 2009/54549-8 from the State of São Paulo Foundation for Research Support (FAPESP) and from the Brazilian Council of Scientific and Technologic Development (CNPq). The funders had no role in study design, data collection and analyses, decision to publish or preparation of the manuscript.
S. B. V. M., D. A. C. B. and I. L. N. conceived and designed the experiments. D. A. C. B., T. V. B. and D. M. A. C. M. performed the experiments. D. A. C. B. and S. B. V. M. analysed the data. S. B. V. M., I. L. N. and D. M. A. C. M. contributed reagents, materials and analysis tools. S. B. V. M. and D. A. C. B. wrote the paper.
The authors have declared that no conflicts of interest exist.