Address correspondence and reprint requests to Dr V. Hook, Buck Institute for Age Research, 8001 Redwood Blvd., Novato, CA 94945, USA. E-mail: email@example.com
Alpha-melanocyte-stimulating hormone (α-MSH) is a neuropeptide expressed in pituitary and brain that is known to regulate energy balance, appetite control, and neuroimmune functions. The biosynthesis of α-MSH requires proteolytic processing of the proopiomelanocortin (POMC) precursor. Therefore, this study investigated the in vivo role of the prohormone convertase 2 (PC2) processing enzyme for production of α-MSH in PC2-deficient mice. Specific detection of α-MSH utilized radioimmunoassay (RIA) that does not crossreact with the POMC precursor, and which does not crossreact with other adrenocorticotropin hormone (ACTH) and β-endorphin peptide products derived from POMC. α-MSH in PC2-deficient mice was essentially obliterated in pituitary, hypothalamus, cortex, and other brain regions (collectively), compared to wild-type controls. These results demonstrate the critical requirement of PC2 for the production of α-MSH. The absence of α-MSH was accompanied by accumulation of ACTH, ACTH-containing imtermediates, and POMC precursor. ACTH was increased in pituitary and hypothalamus of PC2-deficient mice, evaluated by RIA and reversed-phase high pressure liquid chromatography (RP-HPLC). Accumulation of ACTH demonstrates its role as a PC2 substrate that can be converted for α-MSH production. Further analyses of POMC-derived intermediates in pituitary, conducted by denaturing western blot conditions, showed accumulation of ACTH-containing intermediates in pituitaries of PC2-deficient mice, which implicate participation of such intermediates as PC2 substrates. Moreover, accumulation of POMC was observed in PC2-deficient mice by western blots with anti-ACTH and anti-β-endorphin. In addition, increased β-endorphin1−31 was observed in pituitary and hypothalamus of PC2-deficient mice, suggesting β-endorphin1−31 as a substrate for PC2 in these tissues. Overall, these studies demonstrated that the PC2 processing enzyme is critical for the in vivo production of α-MSH in pituitary and brain.
Therefore, this study analyzed α-MSH in pituitary and brains of mice that lack active PC2 (Furuta et al. 1997). Results showed that α-MSH was absent in pituitaries and brains of PC2-deficient mice, compared to wild-type controls. The absence of α-MSH was accompanied by increased ACTH in pituitary and brains of PC2-deficient mice, consistent with ACTH as a substrate for PC2. In addition, accumulation of ACTH-containing intermediates and POMC precursor in pituitary were demonstrated by western blots with anti-ACTH and anti-β-endorphin1−31 serum. Changes in levels of β-endorphin1−31 were also observed in pituitary and the hypothalamus of brains from PC2-deficient mice. These results provide in vivo evidence for the requirement of PC2 in the production of α-MSH in pituitary and brain.
Preparation of tissues from control and PC2-deficient mice for α-MSH determinations
PC2-deficient, null mice (Furuta et al. 1997), were obtained from the Jackson Laboratories (Bar Harbor, ME, USA). This strain of mice lacks active PC2, resulting from deletion of exon 3, which represents a ‘functional knockout’ of PC2 activity. PC2-deficient and control adult mice (littermates) were genotyped by Jackson Laboratories and in our laboratory to confirm mutant –/– and wild-type +/+ mice, as described previously (Furuta et al. 1997). Fresh pituitary and brain tissues (hypothalamus, cortex, and the remaining brain regions) were dissected from PC2-deficient and wild-type control mice, frozen on dry ice, and tissue homogenates were prepared by sonication in 0.1 n HCl on ice. The homogenate was centrifuged (15 000 × g for 10 min at 4°C) and the resultant supernatant was collected as the acid extract for assay of α-MSH and peptide hormone (ACTH and β-endorphin1−31) tissue content by radioimmunoassays (RIA). Protein content was determined utilizing bovine serum albumin as standard (according to protocol of Biorad DC protein assay kit). Neuropeptide tissue levels were expressed as pg peptide per µg protein.
α-MSH and neuropeptide radioimmunoassays
Quantitative measurements of α-MSH, as well as ACTH and β-endorphin1−31 neuropeptides, utilized RIAs conducted under neutral pH (pH 7.5) conditions. A specific RIA was utilized to measure levels of α-MSH in tissue extracts, with RIA kits and protocols from Phoenix Pharmaceuticals, Inc. (Mountain View, CA, USA). The α-MSH RIA, conducted under neutral and native buffer conditions, does not crossreact with full-length POMC that was prepared by expression and purification of recombinant bovine POMC as we have described previously (Hook et al. 1997). In addition, the RIA for α-MSH does not crossreact with other POMC-derived peptides that consist of ACTH, β-endorphin1−31, β-MSH, γ-MSH, or (Met)enkephalin. It is known that the primary sequences of α-MSH from mouse, human, rat, porcine, and bovine are identical (Nakanishi et al. 1979; Roberts et al. 1979; Chang et al. 1980; Oates and Herbert 1984).
Evaluation of α-MSH in PC2-deficient mice was accompanied by assessment of two other POMC-derived neuropeptides, ACTH and β-endorphin1−31. The ACTH RIA utilized anti-ACTH serum from NIDDK/NIH, and 125I-ACTH from Peninsula Laboratories (San Carlos, CA, USA), performed as we have described previously (Hook et al. 1982). The RIA for human ACTH does not crossreact with full-length POMC (bovine), obtained by expression and purification of POMC as we have described (Hook et al. 1997). Mouse and human ACTH possess 95% homology (Roberts et al. 1979; Chang et al. 1980). The ACTH RIA does not recognize α-MSH, β-MSH, γ-MSH, or β-endorphin1−31 that are derived from POMC.
The POMC product β-endorphin1−31 was also measured in tissue extracts by RIA, obtained as a kit (from Phoenix Pharmaceuticals), which does not recognize POMC. The RIA for β-endorphin1−31 also does not crossreact with α-MSH, ACTH, β-MSH, γ-MSH, β-LPH (Met)enkephalin (Leu)enkephalin, β-endorphin1−27, and β-endorphin1−26.
Reverse-phase high pressure liquid chromatography (RP-HPLC) of ACTH and β-endorphin1−31
The relative changes in tissue contents of ACTH and β-endorphin1−31, measured by RIA, were further evaluated by assessing coelution of ACTH and β-endorphin1−31 immunoreactivity with standard ACTH and β-endorphin1−31 peptides. Samples were subjected to RP-HPLC on a C-18 RP-HPLC column (3.2 × 150 mm, 5 µ, 300 Å, Vydac, Hesperia, CA, USA) with a C-18 guard column, equilibrated in 0.1% (v/v) trifluoroacetic acid. Separation was achieved with an acetonitrile gradient in 0.1% trifluoroacetic acid, from 25 to 60% over 8 min, isocratic at 60% for 3 min, and 60–99% over 2 min at a flow rate of 0.5 mL/min. Fractions of 1 min were collected, and concentrated under vacuum in a Speed-vac prior to RIA assays.
POMC products evaluated by reducing and denaturing conditions with western blots for ACTH and β-endorphin1−31 immunoreactivities
POMC-derived products in pituitary extracts were analyzed by western blots with preparation of tissue samples for sodium dodecyl sulfate–polyacrylamide gel electrophoresis under reducing and denaturing conditions with β-mercaptoethanol and sodium dodecyl sulfate, respectively, as we have previously described (Hook et al. 1997). Samples were subjected to electrophoresis on 16% Tricine gels (Invitrogen, Carlsbad, CA, USA), and proteins were transferred electrophoretically to nitrocellulose membranes (Amersham, Piscataway, NJ, USA). Membranes were blocked in 5% non-fat dried milk in TTBS buffer (20 mm Tris-HCl, pH 7.4, 0.5 m NaCl, 0.05% Tween 20), and were incubated with anti-ACTH or anti-β-endorphin1−31 (final dilution of 1 : 500) in TTBS for 2 h. After washing in TTBS and incubation with sheep anti-rabbit-horseradish peroxidase (Amersham) for 1 h followed by washing, the membrane was developed by enhanced chemiluminescence (according to the manufacturer's protocol, Amersham).
The sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels of western blots utilized reducing conditions by β-mercaptoethanol, and denaturation by sodium dodecyl sulfate, thereby providing denaturated POMC and intermediate products that were detected by anti-ACTH and anti-β-endorphin1−31 sera. In contrast, when these same antisera were utilized in RIAs conducted at neutral pH (without reduction or denaturation), these antisera did not recognize the POMC precursor. It is known that disulfide bonds and reduction of POMC alters its conformation (Cool et al. 1997; Loh et al. 2002) which can influence recognition of different conformational states of POMC by antibodies. These different conformational states of POMC suggest that the neutral pH of the RIAs provide conditions for the specificity of anti-ACTH and anti-β-endorphin sera to detect only processed ACTH and β-endorphin1−31 (not POMC precursor). However, the reducing and denaturing conditions of the western blots can modify the conformational features of POMC and derivatives (Cool et al. 1997; Loh et al. 2002) which allow these antisera to ACTH and β-endorphin to detect POMC.
α-MSH is absent in pituitary and brains of PC2-deficient mice
α-MSH is generated by proteolytic processing of the POMC precursor at paired basic and multibasic residues (Fig. 1). In this study, the in vivo role of the PC2 processing enzyme for production of α-MSH was evaluated in pituitary and brains of PC2-deficient mice and wild-type control mice (normal littermates) (Furuta et al. 1997).
In PC2-deficient mice, α-MSH levels were obliterated in pituitary, hypothalamus, cortex, and brain (remainder of the brain, with hypothalamus and cortex removed), compared to wild-type controls (Fig. 2). α-MSH in pituitaries from PC2-deficient mice was reduced to approximately 2% of controls (100%). In hypothalamus, cortex, and brains (without hypothalamus and cortex) of PC2-deficient mice, α-MSH was not detectable (0%) compared to normal controls (100%). In control wild-type mice, the high content of α-MSH in pituitary of approximately 2000 pg α-MSH/µg protein contrasts with lower α-MSH levels in brain regions containing less than 1 pg α-MSH/µg protein (Table 1). It is noted that quantitation of α-MSH utilized a specific RIA that detects processed α-MSH, and which does not detect the POMC precursor. These results demonstrate the requirement for PC2 in the production of α-MSH in pituitary and brain.
Table 1. Differential changes in POMC-derived neuropeptides in tissues from PC2-deficient mice
α-MSH (pg/µg protein)
ACTH (pg/µg protein)
β-Endorphin (pg/µg protein)
After dissection of hypothalamus and cortex, the resultant brain tissue was utilized as the remaining ‘brain’ tissue. Peptide tissue content is expressed as the mean ± SEM, with n = 6–9 for wild-type controls (WT), and n = 3–4 for PC2-deficient mice (Null).
a–d Indicate significant reduction of α-MSH in tissues from Null compared to WT with p-values of < 0.001, < 0.001, < 0.0011, and < 0.0103, respectively (by two-tailed t-tests). e, f Indicate significant increases in ACTH in tissues from Null compared to WT with p-values of < 0.0001 and < 0.0002. g,hIndicate significant increases in β-endorphin from Null compared to WT in pituitary and hypothalamus with p < 0.0001 and p < 0.0016, respectively.
Elevated ACTH in pituitary and hypothalamus of PC2-deficient mice
The POMC-derived product ACTH has been hypothesized to represent the immediate substrate of PC2 for proteolytic cleavage to generate α-MSH (Benjannet et al. 1991; Thomas et al. 1991). Thus, the absence of PC2 activity would be predicted to result in accumulation of ACTH. Indeed, ACTH levels (detected by specific RIA) in pituitary and hypothalamus regions of PC2-deficient mice were increased several-fold above controls (Fig. 3a). Increases in ACTH content occurred selectively in pituitary and hypothalamus, compared to cortex and other regions of brain (brain without hypothalamus or cortex); these results suggest tissue-selective roles for PC2-mediated metabolism of ACTH. It was noted that the pituitary contains high levels of ACTH of approximately 5400 pg ACTH/µg protein, compared to the lower levels of ACTH in brain regions of 1–5 pg ACTH/µg protein (Table 1).
The increased levels of ACTH in pituitary and hypothalamus were confirmed by RP-HPLC characterization of ACTH immunoreactivity from PC2-deficient mice (Figs 3b and c). In hypothalamus, the peak of ACTH immunoreactivity coeluted with standard ACTH from PC2-deficient mice, and was approximately twofold greater than wild-type controls (Fig. 3b). In pituitary, RP-HPLC demonstrated that the peak of ACTH immunoreactivity coeluted with standard ACTH (Fig. 3c). The peak of ACTH isolated by RP-HPLC was approximately 10-fold greater in pituitary extracts from PC2-deficient mice compared to wild-type controls. These RP-HPLC results provided evidence for elevated ACTH in pituitary and hypothalamus of PC2-deficient mice.
Accumulation of POMC-derived intermediates in pituitaries of PC2-deficient mice
The relatively high levels of ACTH in pituitary allowed analyses of ACTH-containing intermediates by anti-ACTH western blots. Anti-ACTH western blots of pituitary from PC2-deficient mice detected increased levels of ACTH immunoreactive bands of 33.5, 32, 22, 10, and 6 kDa, compared to wild-type controls (Fig. 4a). These results suggest that ACTH-positive bands represent POMC-derived intermediates that contain the ACTH domain (Fig. 1). The accumulation of these ACTH-containing intermediates in PC2-deficient mouse pituitary suggests that these putative POMC-derived intermediates may serve as substrates for PC2.
The identities of the pituitary ACTH-containing intermediates were predicted from anti-ACTH and anti-β-endorphin1−31 western blots (Fig. 4b). The 33.5, 32, 22, and 6 kDa ACTH-positive bands (Fig. 4a) were not recognized by anti-β-endorphin1−31. Thus, the 33.5 kDa ACTH-positive band may represent COOH-terminal truncated POMC that lacks β-endorphin (Fig. 1). The 32 kDa band may represent POMC without the β-MSH and β-endorphin domains, based on its apparent size and lack of detection by β-endorphin antisera. The 22 kDa band may represent 21–23 kDa ACTH that includes N-POMC, γ-MSH, and ACTH domains. The 10 kDa ACTH-positive band (Fig. 4a, lane 2) contains the ACTH domain. The 8 kDa β-endorphin-positive band contains the β-endorphin domain.
Evidence for accumulation of POMC in PC2-deficient mice
Western blots with the antisera to β-endorphin1−31 detected increased levels of POMC of 36–40 kDa apparent molecular weight in PC2-deficient mice compared to wild-type controls (Fig. 4b). Some accumulation of POMC was detected with anti-ACTH (Fig. 4a), but was more readily detected by anti-β-endorphin. It is possible that the anti-β-endorphin recognized denatured POMC more readily than anti-ACTH. Accumulation of POMC in PC2-deficient mice, compared to wild-type controls, is consistent with POMC as substrate for the PC2 enzyme.
In addition, western blots with both ACTH and β-endorphin antiserum also detected high molecular weight bands (approximately 70 kDa) of POMC that are consistent with dimer forms of POMC (Cawley et al. 2000) (Fig. 4a and b). These oligomeric forms of POMC were increased in PC2-deficient mice compared to wild-type controls. These results suggest that accumulation of POMC promoted the formation of dimer forms of POMC.
It is noted that the ACTH antisera utilized for western blots was identical to that used for RIA of ACTH. Whereas the ACTH antiserum recognized denatured POMC that resulted from the reducing and sodium dodecyl sulfate denaturing conditions of western blots, the RIA performed under native, neutral pH conditions showed no crossreactivity with POMC. It was apparent that the ACTH antiserum recognized denatured POMC that resulted from the reducing and sodium dodecyl sulfate denaturing conditions of western blots, but the antisera did not detect POMC under neutral conditions of the RIA for ACTH. It is known that reducing and denaturation conditions can modify the conformation of POMC (Cawley et al. 2000; Loh et al. 2002). In addition, these reducing and denaturing conditions also provide explanation for the detection of POMC by anti-β-endorphin1−31, whereas the RIA for β-endorphin1−31 does not detect POMC under native neutral conditions.
Beta-endorphin in pituitary and brains of PC2-deficient mice
Further analyses of POMC-derived products in PC2-deficient mice was achieved by measuring tissue levels of β-endorphin1−31 by RIA. Beta-endorphin1−31 levels in pituitary and hypothalamus were increased by approximately five-fold and three-fold, respectively, in PC2-deficient mice compared to controls (Fig. 5a). In control mice, β-endorphin1−31 levels were high in pituitary (approximately 400 pg/µg protein) compared to brain that contained less than 0.5 pg β-endorphin1−31/µg protein (Table 1).
Analyses of β-endorphin1−31 by RP-HPLC were conducted for pituitary and hypothalamus, to confirm that the increased levels of β-endorphin1−31 immunoreactivity corresponded in elution position to authentic β-endorphin1−31 standard. RP-HPLC showed that the peak of β-endorphin immunoreactivity in pituitary was fourfold greater in PC2-deficient mice than in wild-type controls (Fig. 5b). In hypothalamus, β-endorphin was approximately fourfold greater in PC2-deficient mice compared to controls (Fig. 5c). The accumulation of β-endorphin1−31 in PC2-deficient mice suggested β-endorphin1−31 as a PC2 substrate. Indeed, PC2 processing of β-endorphin1−31 has been shown to generate β-endorphin1−27 (Allen et al. 2001). These RP-HPLC and RIA analyses demonstrated increases in β-endorphin1−31 in pituitary and hypothalamus of PC2-deficient mice, which occurs concomitantly with obliteration of α-MSH.
This study investigated the in vivo role of the subtilisin-like PC2 prohormone convertase in the production α-MSH in PC2-deficient mice. α-MSH was absent in pituitary, hypothalamus, cortex, and other brain regions (collectively) of PC2-deficient mice compared to wild-type controls. These striking results demonstrate the critical requirement for PC2 in the production of α-MSH from its POMC precursor.
α-MSH is known to be an important regulator of appetite control, energy balance, and neuroimmune functions (Ichiyama et al. 2000; Williams et al. 2001; Pritchard et al. 2002; Zimanyi and Pelleymounter 2003). The PC2-deficient mice, however, have not yet been extensively tested for alterations in these physiological systems that are influenced by α-MSH. Nonetheless, it is known that PC2 mice show a moderate decrease in rate of growth (Furuta et al. 1997). They also show a reduced rise in blood glucose levels in a glucose tolerance test, which is predicted to be consistent with a deficiency in glucagon. With the finding of nearly complete obliteration of α-MSH in PC2-deficient mice, further examination of physiological changes in future studies may be of interest.
The absence of α-MSH was accompanied by accumulation of ACTH, ACTH-containing intermediates, and POMC precursor. PC2-deficient mice showed increased levels of ACTH in pituitary and hypothalamus, detected by specific RIA for ACTH and RP-HPLC. These findings suggest that ACTH serves as a substrate for its conversion by PC2 to α-MSH. Further analyses of POMC-derived intermediates in pituitary, conducted by reducing and denaturing western blot conditions, showed accumulation of ACTH-containing intermediates in pituitaries of PC2-deficient mice; these accumulated intermediates may serve as PC2 substrates. Western blots with anti-β-endorphin and anti-ACTH provided evidence for accumulation of the POMC precursor. In addition, increased levels of β-endorphin1−31 was observed in pituitary and brains of PC2-deficient mice; accumulation of β-endorphin1−31 is consistent with its role as a substrate for PC2, which is involved in generating β-endorphin1−27 from β-endorphin1−31 (Allen et al. 2001). Importantly, these in vivo studies demonstrate that PC2 is essential for the production of α-MSH.
Although pituitary α-MSH was reduced by 98% in PC2-deficient mice compared to wild-type controls, the pituitary content of α-MSH in the PC2-deficient mice of approximately 40 pg/µg protein is still nearly 40-fold greater than that in brain (Table 1). The significant amount of α-MSH remaining in pituitaries of PC2-deficient mice indicate that other proteases may provide production of a low amount of α-MSH. Thus, in addition, to PC2, POMC processing also involves PC1 (PC1 is also known as PC1/3) (1–3), a neuroendocrine member of the subtilisin-like family of prohormone convertases (Steiner et al. 1992; Hook et al. 1994; Cawley et al. 1998; Seidah et al. 1999). The preferential localization of PC1 to anterior pituitary and the expression of both PC1 and PC2 in intermediate pituitary (Day et al. 1992; Schafer et al. 1993) are consistent with a role for PC1 in the production of ACTH from POMC in anterior pituitary, and roles for the combined activities of PC1 and PC2 in intermediate pituitary to generate α-MSH and β-endorphin. Recently, a role for PC1 in the production of ACTH was demonstrated in PC1-deficient mice (Zhu et al. 2002).
POMC undergoes tissue-specific processing for selective production of α-MSH compared to ACTH and β-endorphin peptide products. Therefore, control mechanisms for PC2 or PC1 may be important in regulating the production of α-MSH and other neuropeptides derived from POMC. PC2 undergoes activation by 7B2, and inhibition by the C-terminal peptide derived from 7B2 (Braks and Martens 1994; Benjannet et al. 1995; Fortenberry et al. 1999; Westphal et al. 1999). In addition, PC1 can be inhibited by endogenous pro Ser-Ala-Ala-Ala-Ser (proSAAS) (Qian et al. 2000; Fortenberry et al. 2002). Thus, it may be of interest in future studies to examine the coordinate control of PC2 and PC1 by enzyme activators or inhibitors for regulating the production of α-MSH that is derived from POMC.
Numerous reports demonstrate that POMC processing in pituitary secretory vesicles is also achieved by an aspartyl protease known as ‘POMC converting enzyme’ (PCE) (Parish et al. 1986; Estivariz et al. 1992; Cawley et al. 1998). PCE cleaves POMC at the paired basic and multibasic residues to generate ACTH and β-endorphin products. However, the aspartyl protease gene for PCE has not yet been identified. It will be of interest in future studes to examine the effects of genetic ablation of the PCE gene for testing its in vivo role in POMC processing.
In summary, results from these studies demonstrate the critical requirement for PC2 in the production of α-MSH. Other studies of PC2-deficient mice have demonstrated that PC2 participates in the processing of several prohormones, including proinsulin, proglucagon, prodynorphin, proenkephalin, procholecystokinin (proCCK), and proneurotensin (Furuta et al. 1998; Johanning et al. 1998; Berman et al. 2000; Vishnuvardhan et al. 2000; Furuta et al. 2001; Villeneuve et al. 2002). PC2 represents a key processing enzyme for the conversion of propeptide precursors into α-MSH, as well as multiple peptide hormones and peptide neurotransmitters.
This study was supported by grants from NIDA, NINDS, and NHLBI of the National Institutes of Health to VYHH and MCB.