Molecular cloning and characterization of preproopiomelanocortin (prePOMC) cDNA from the ostrich (Struthio camelus)

General and Comparative Endocrinology 146 (2006) 310–317 www.elsevier.com/locate/ygcen

Communication in Genomics and Proteomics

Molecular cloning and characterization of preproopiomelanocortin (prePOMC) cDNA from the ostrich (Struthio camelus) Ryno Naudé a,¤, Willem Oelofsen a, Akiyoshi Takahashi b, Masafumi Amano b, Hiroshi Kawauchi b a

Department of Biochemistry and Microbiology, P.O. Box 77000, Nelson Mandela Metropolitan University, Port Elizabeth 6031, South Africa b School of Fisheries Sciences, Kitasato University, Sanrika, Iwate 022-0101, Japan Received 5 August 2005; revised 11 November 2005; accepted 21 November 2005 Available online 2 February 2006

Abstract To date proopiomelanocortin (POMC), the precursor protein for melanotropin (MSH), adrenocorticotropin (ACTH), lipotropins (LPH), and -endorphin (-END) in the pituitary gland, has been studied extensively over a wide spectrum of vertebrate classes. A paucity of information exists, however, with regard to POMC in the avian class, where to date POMC from only one species, the domestic chicken, appears to have been fully characterized. In the present study, we report the use of three clones of cDNA to provide the complete nucleotide sequence of ostrich prePOMC cDNA, consisting of 1072 bp (excluding the poly(A) tail). The deduced amino acid sequence of 253 amino acid residues includes the N-terminal signal peptide of 17 amino acid residues. The predicted amino acid sequence in the overall arrangement of its domains, conforms to that found in other tetrapods. Sequence domains for -MSH, ACTH, -MSH, -LPH, -MSH, and -END are located at positions 74–85, 134–172, 134–146, 175–220, 203–220, and 223–253, respectively, in ostrich prePOMC, but some of them may not be released in the ostrich pituitary gland, despite the presence of nine potential processing sites consisting of 2–4 dibasic amino acids each. Substitution of glutamic acid for a dibasic amino acid at position 202 in ostrich prePOMC could prevent release of -MSH. To date the release of pro--MSH, -LPH, ACTH, -LPH, and -END have been conWrmed by direct isolation and characterization from ostrich pituitary extracts. In the present study, we have also identiWed ACTH, -LPH and -END in a single frozen ostrich pituitary slice by means of MALDI-TOF mass spectrometry. When compared to a wide range of vertebrate prePOMC molecules, ostrich prePOMC revealed a high level of amino acid sequence identity (77%) with chicken prePOMC, which is the only other avian sequence available. As with other vertebrate classes, considerable intraclass diVerences were also evident between chicken and ostrich prePOMCs, which belong to diVerent avian orders. Identity of ostrich prePOMC with non-avian tetrapod counterparts is only moderate (53–56%), whereas lower identities (20–49%) are evident over a range of Wsh prePOMCs. © 2005 Elsevier Inc. All rights reserved. Keywords: Proopiomelanocortin; prePOMC; Ostrich; Pituitary; cDNA cloning; Avian; Mass spectrometry

1. Introduction Proopiomelanocortin (POMC) has become Wrmly established as the common precursor protein for the family of pituitary hormonal polypeptides which includes: adrenocorticotropin (ACTH), melanocyte-stimulating hormone

*

Corresponding author. Fax: +27 41 5042814. E-mail address: [email protected] (R. Naudé).

0016-6480/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygcen.2005.11.018

(MSH), the lipotropins (LPH), and -endorphin (-END) (Smith, 1980). The complete structure of a preproopiomelanocortin (prePOMC), which included the N-terminal signal sequence, was Wrst accomplished by molecular cloning of the bovine gene (Nakanishi et al., 1979). Since then POMC cDNAs have been cloned, and the deduced amino acid sequences have been reported for a wide range of vertebrates, including several tetrapod species (Takahashi et al., 2001). POMC has also been identiWed in invertebrates (Salzet et al., 1997; Stefano et al., 1999). Despite these impressive

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317

advances, a relative paucity of information appears to exist on POMC in avian systems. To our knowledge the Wrst and only complete study to date of an avian POMC, was the cloning and characterization of the prePOMC gene of the chicken (Takeuchi et al., 1999). The ostrich (Struthio camelus), a non-Xying bird of ancient lineage, appeared to be of considerable interest from the point of view of the molecular phylogenetics of POMC in avian tetrapods. Over a number of years several members of the POMC-related family of polypeptide hormones derived from the ostrich pituitary gland, have been investigated by the authors at the level of protein isolation and structural characterization, including the following: ACTH (Li et al., 1978; Naudé and Oelofsen, 1977), -lipotropin (Naudé and Oelofsen, 1981; Naudé et al., 1981a), -endorphin (Naudé et al., 1980; Naudé et al., 1981b), -lipotropins (Litthauer et al., 1984) and the NH2-terminal POMC fragment (pro--MSH) (Naudé et al., 1993). Together these isolated polypeptides constituted a total of 181 amino acid residues that represent a major portion of the total ostrich prePOMC. The remainder of the sequence is occupied mainly by the N-terminal signal peptide and by the region separating the -MSH domain from the ACTH domain (“Joining peptide” region). In the present study, we report the entire sequence of the cloned ostrich prePOMC cDNA and the deduced amino acid sequence. Furthermore, ACTH, -LPH and -END were identiWed by means of direct application of mass spectrometry to a frozen ostrich pituitary slice. This study conWrmed the amino acid sequences that we reported for the previously isolated hormonal polypeptides, and also allowed a comparison of ostrich prePOMC with its counterpart from the chicken, the only other fully characterized avian POMC available to date. 2. Materials and methods 2.1. Nucleic acid preparation The synthesis of oligonucleotides was performed by Nihon Gene Research Laboratories (Sendai, Japan). Total RNA was prepared from ostrich pituitaries using Isogen (Nippon Gene, Tokyo, Japan). The template for the polymerase chain reaction (PCR) was prepared from the total RNA at 37 °C for 60 min using a First-Strand cDNA Synthesis Kit (Amersham–Pharmacia Biotech, Buckinghamshire, UK). The Wrst-strand cDNA from the 5⬘-region was prepared using the 5⬘-RACE System for Rapid

AmpliWcation of cDNA Ends, Version 2 (Life Technologies, Tokyo, Japan). Primer 4 (5⬘-CGCCTGAACTCCAGCGGGAA-3⬘) was used for reverse transcription in the 5⬘-RACE.

2.2. Polymerase chain reaction for POMC cDNA cloning PCR was conducted using a thermal cycler (PC-808, Astec, Fukuoka, Japan). The reaction mixture (50 l) for reverse transcription (RT)-PCR was composed of 0.5 l AmpliTaq Gold (1 U, Applied Biosystems, Foster City, CA), 5 l buVer solution accompanied by the enzyme, 1 l of the Wrst strand cDNA, 1 l of forward primer (10 M), 1 l of reverse primer (10 M), and 5 l dNTP solution (2.5 mM each). PCR conditions were activation of the enzyme at 95 °C for 15 min, 35 cycles of denaturation (60 s at 94 °C)-annealing (60 s at 55 °C)-extension (90 s at 72 °C), followed by a Wnal extension at 72 °C for 7 min. AmpliWcation of the middle portion of the ostrich POMC cDNA was performed with primers 1 [5⬘-ATGGA (AG)CA(TC)TT(TC)(CA)GITGG-3⬘] and 2 [5⬘-CAT(GA)AAICCICC (GA)TA-3⬘]. SpeciWc primers used for 3⬘-RACE and 5⬘-RACE were designed based on partial nucleotide sequences of ostrich POMC cDNA [for clone 2: primer 3 (5⬘-TCAAGGTCTACCCCAAC-3⬘) and clone 3: primer 5 (5⬘-GTCTCCTCCTGCACGCCGTT-3⬘)].

2.3. MALDI-TOF-MS MALDI-TOF-MS was performed for direct proWling of a pituitary slice of the ostrich according to the method of Yasuda-Kamatani and Yasuda (2000). In brief, a slice (40 m) prepared from a frozen ostrich pituitary was placed on a MALDI sample plate, followed by rinsing twice using the matrix solution of -cyano-4-hydroxycinnamic acid (-CHCA) saturated in 50% acetonitrile in 0.1% triXuoroacetic acid to remove excess salts present in the sample. The fresh -CHCA matrix solution was added to the sample and dried under vacuum. The MALDI-TOF-MS spectra were obtained using an AXIMA-CFRplus mass spectrometer (Shimadzu, Kyoto, Japan). External calibration was performed using lamprey ACTH(1–31) and bovine insulin, of which the monoisotopic molecular weights were 3580.91 and 5730.61, respectively.

3. Results 3.1. IdentiWcation of ostrich POMC cDNAs Clone 1 was obtained by RT-PCR using primers 1 and 2 (Fig. 1). Based on the nucleotide sequence of this cDNA, primer 3 was synthesized for 3⬘-RACE and primers 4 and 5 for 5⬘-RACE. Using combinations of primers as shown in Fig. 1, clones 2 and 3 were obtained by PCR. The overlapping of clones 1, 2, and 3 provided the entire sequence of ostrich POMC cDNA consisting of 1072 bp excluding the poly(A) tail (Fig. 1). Ostrich prePOMC is composed of 253 amino acid residues (Fig. 2). Amino acid sequence compari-

802 (stop) 1072

1 43 94 POMC 469

SP Primer 1

1

AAAAAAA

708 primer 2

clone 1 517

Primer 3 kit primers

311

1072 clone 2

AAAAAAA

Not I adapter primer

513 543 clone 3

primer 4 primer 5

Fig. 1. Schematic diagram depicting the relative positions of cDNA fragments of the ostrich POMC cDNAs. Numbers show positions on POMC cDNAs. SP, signal peptide. Horizontal arrows show relative positions and direction of primers.

312

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317

Fig. 2. Structure of ostrich prePOMC cDNA. The nucleotide sequences of ostrich POMC cDNAs (not including the poly(A) tail) and deduced amino acid sequences are shown. Positions of nucleotide and amino acid sequences are indicated on both sides. The N-terminus of prePOMCs is designated as position 1. Signal peptide is indicated by a thin line. Fragments of POMC, namely ACTH, -LPH and -END, detected by MALDI-TOF-MS are indicated by bold lines. -MSH is shown by dotted line. Locations of primers are shown by bold letters. Position of primer 5 is shown by lower cases. Horizontal arrows show direction of primers. Asterisk shows stop codon. The accession number in the DDBJ/EMBL GenBank nucleotide sequence database is DQ100375.

son with ostrich pro--MSH (Naudé et al., 1993) reveals that the signal peptide is composed of 17 amino acid residues and, in turn, ostrich POMC of 236 amino acid residues. 3.2. MALDI-TOF-MS of frozen slices Direct application of MALDI-TOF-MS to a frozen slice from an ostrich pituitary using -CHCA as a matrix provided several peaks as shown in Fig. 3. Among these peaks, mass/ charge (m/z) values of 3404.09, 4624.66, and 5156.52 corresponded to -END, ACTH, and -LPH, respectively (Table 1). 4. Discussion The complete nucleotide sequence of ostrich prePOMC cDNA consists of 1072 bp (excluding the poly(A) tail), and the deduced amino acid sequence is comprised of 253

amino acid residues, including the 17 amino acids in the proposed N-terminal signal peptide (Fig. 2). In agreement with the chicken (Takeuchi et al., 1999), the ostrich POMC gene appears to be a single gene, showing the same general structural organization as that of other vertebrate classes. The cDNA predicted amino acid sequence conWrms the sequences previously reported for the various isolated hormonal polypeptides including ACTH (Li et al., 1978), -LPH (Naudé et al., 1981a), the two -LPH variants (Litthauer et al., 1984) and -END (Naudé et al., 1981b). Furthermore, through direct application of mass spectrometry to a frozen slice from an ostrich pituitary several peaks were revealed (Fig. 3). Amongst them three peaks of m/z values of 3404.09, 4624.66, and 5156.52, corresponding to -END, ACTH and -LPH, respectively (Table 1), independently conWrmed the presence of these previously isolated hormonal polypeptides in the ostrich pituitary gland.

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317

ACTH (4624.66)

313

γ-LPH (5156.52)

β-END (3404.09)

Fig. 3. Direct MALDI-TOF-MS spectrum of a slice from ostrich pituitary. Number in parenthesis shows observed monoisotopic mass.

Table 1 IdentiWcation of ostrich POMC-related peptides Peptides

Calculated monoisotopic mass

Observed monoisotopic mass (M+H)

ACTH -LPH -END

4623.32 5155.36 3402.82

4624.66 5156.52 3404.09

In the case of the NH2-terminal POMC fragment (pro--MSH) comprising 63 amino acid residues, the cDNA predicted sequence diVers at two positions from that previously proposed (Naudé et al., 1993). Serine is replaced by alanine at position 50 and alanine is replaced by serine at position 69. Ser69 could represent a potential site for glycosylation in ostrich prePOMC (Naudé et al., 1993). However, no evidence was found in favour of O-glycosylation at the equivalent Ser47 in chicken POMC (Berghman et al., 1998). Asn89 in ostrich prePOMC is considered to be a more likely site for glycosylation, which was conWrmed in the case of the equivalent Asn90 in chicken prePOMC (Takeuchi et al., 1999). However, the aspect of glycosylation sites in ostrich prePOMC still needs conWrmation. In accordance with all vertebrate POMCs (except the lamprey POM), four conserved cysteine residues are present in the equivalent positions 26, 32, 44, and 48 in ostrich prePOMC (Fig. 4), which is the N-terminal region of the pro--MSH domain. These cysteine residues may be involved in disulWde bridges stabilizing the tertiary structure of prePOMC and POMC. By comparison with ostrich pro--MSH (Naudé et al., 1993) the authors propose a signal peptide comprising the N-terminal 17 amino acid residues in ostrich prePOMC. This would leave a total of 236 amino acid residues for the POMC sequence. A comparison of the deduced ostrich prePOMC sequence with its vertebrate counterparts over a wide range (Fig. 4), reveals the diVerent functional domains to occur in the expected order. In accordance with all tetrapod (and

also lobe-Wnned Wsh) prePOMCs, the highly conserved HisPhe-Arg-Trp core sequence required for melanotropic activity appears in three regions of ostrich prePOMC: the -MSH domain (Lys74–Phe85), the -MSH domain (Ser134– Val146), which also represents the N-terminal tridecapeptide in the ACTH domain (Ser134–Phe172), and the -MSH domain (Asp203–Asp220) which also constitutes the C-terminal octadecapeptide sequence of the -LPH domain (Ala175–Asp220). Typical of all vertebrate classes, the -END domain (Tyr223–Gln253), with the Met-enkephalin pentapeptide core sequence for opiate activity (Tyr223– Gly224–Gly225–Phe226–Met227), also comprises the C-terminal sequence in ostrich prePOMC. However, the -MSH domain, which is typical of the chondrichtyes (cartilaginous Wsh) as in dogWsh (Amemiya et al., 1999a), stingray (Amemiya et al., 2000) and ratWsh (Takahashi et al., 2004), is absent from ostrich prePOMC (as in all vertebrates other than the chondrichtyes; Fig. 4). It was proposed that -MSH in the chondrichtyes was derived from -MSH and -END by a process of internal gene duplication, followed by mutations in the duplicated region, to give rise to MSH and the C-terminal extension of -MSH (CTED) (Takahashi et al., 2004). In sarcopterigians (lungWsh) and tetrapods, including the ostrich, the three classical MSH domains (, , and ) appear to have been well conserved during the evolutionary development of POMC from their early invertebrate via their early vertebrate ancestors (Takahashi et al., 2001). Nine potential processing sequences consisting of 2–4 dibasic amino acid residues each are present in ostrich prePOMC (Figs. 2 and 4). Four of these are conWrmed endoprotease cleavage sites, based on the corresponding polypeptide cleavage products that were previously isolated and characterized and which included pro--MSH (Naudé et al., 1993), ACTH (Li et al., 1978), -LPH (Naudé et al., 1981a), two -LPH variants (Litthauer et al., 1984), and -END (Naudé et al., 1981b).

314

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317

Fig. 4. Amino acid sequence comparison of ostrich prePOMC with prePOMCs from bovine (Nakanishi et al., 1979), chicken (Takeuchi et al., 1999), mud turtle (Shen et al., 2003), Xenopus A (Martens, 1986), African lungWsh (Amemiya et al., 1999b), white sturgeon A (Amemiya et al., 1997), barWn Xounder A (Takahashi et al., 2005), dogWsh (Amemiya et al., 1999a), ratWsh (Takahashi et al., 2004), and lamprey (Takahashi et al., 2001). Numbers of amino acid sequences are indicated on the right side. Component polypeptide domains of ostrich prePOMC are indicated by horizontal overlines. Shaded letters (blue colour) indicate amino acid residues identical to ostrich prePOMC. A heavily shaded area (green colour) indicates the minimal core sequence for melanotropic activity or remnants of it. Small letters indicate signal sequences. The presence of an asterisk indicates a conWrmed cleavage site in ostrich prePOMC. Hyphens represent missing residues. For interpretation of the references to color in this Wgure legend, the reader is referred to the web version of this paper.

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317

These polypeptides are Xanked in ostrich prePOMC by: Arg87–Arg88(at pro--MSH C-terminal end), Lys132– Arg133 and Arg173–Arg174 (for ACTH), Arg173–Arg174 and Lys221–Arg222 (for -LPH) and Lys221–Arg222 (at -END Nterminal end). The -MSH sequence, Xanked by the paired dibasic residues Arg73–Lys74 and Arg87–Arg88, could theoretically be released by endoprotease cleavage. However, to date release of -MSH has not been conWrmed in the pituitary glands of either the ostrich, the chicken (Takeuchi et al., 1999) or the reptilian mud turtle (Shen et al., 2003), despite the presence of identical or similar paired dibasic Xanking sequences in all of them (Fig. 4). As far as -MSH release from ostrich POMC is concerned, the substitution of a basic residue by the acidic glutamic acid residue in position 202, could prevent the release of -MSH, which to date has not been detected in ostrich pituitary extracts. This would limit the number of fragments obtainable from the C-terminal region of ostrich POMC to - and -LPH and -END, as was evident in the isolation studies cited above. However, in the case of lamprey (in the agnatha), an identical substitution is present at position 187 (Fig. 4) preceding the MSH-A sequence in proopiomelanotropin (POM) (Takahashi et al., 2001). Yet MSH-A could be isolated together with MSH-B from lamprey pituitary extracts. This could imply possible species diVerences in the endoprotease processing of POMC. The presence of a potential monobasic cleavage site in ostrich -END has been shown in a few other species, amongst others in the toad (Dores et al., 1994). Unlike the lamprey, the pars intermedia in all probability is absent in the ostrich pituitary as in other avian species. Whereas the pars intermedia is considered to be involved in further cleavage of pro--MSH, ACTH, and -LPH, leading to the release of smaller MSH molecules (-, -, and -MSH, respectively) (Takahashi et al., 2001), its absence could prevent such further cleavage in the avian pituitary. In the light of these considerations it would seem fair to conclude that the physiological signiWcance of

315

-, -, and -MSH of pituitary origin, remains uncertain in birds (Takeuchi et al., 1999). In addition, there is growing evidence for a possible autocrine/paracrine role of the avian melanocortin system. Widespread expression of the POMC gene in various non-pituitary tissues in the avian body is suggested (Takeuchi et al., 2003), with resultant local production of -MSH, which may allow it to act as an autocrine/paracrine hormone in birds, and thus compensate for the absence of the pars intermedia. In comparison to ostrich prePOMC the overall amino acid sequence identity across the vertebrate spectrum of prePOMCs ranges from the highest value for chicken prePOMC (76.9%) to the lowest for the primitive Wsh, the lamprey of the agnathans (19.8% for POC and 19.6% for POM; Fig. 4 and Table 2). When compared to its reptilian ancestry, ostrich prePOMC revealed only moderate identity of 56.3% in the mud turtle and a similar value of 55.9% in the leopard gecko (Endo and Park, 2004). Similar moderate levels of amino acid sequence identity is evident for the amphibian and mammalian prePOMCs (53.3% for Xenopus A prePOMC and 54.3% for the bovine molecule). Further comparison of the ostrich and chicken prePOMC sequences with those of the two reptilian species available, revealed that 57.7% of the ostrich sequence was identical to those of both the mud turtle and the leopard gecko. The chicken prePOMC sequence revealed a similar identity (58.2%) with the leopard gecko sequence and a slightly higher identity of 60.6% with the mud turtle sequence. However, when compared to each other, the two reptilian sequences revealed higher identities (66–69%). It is apparent, as was noted previously for chicken prePOMC (Endo and Park, 2004), that the more ancient ostrich prePOMC also does not seem to cluster any closer with its generally accepted reptilian ancestry, but rather is clustered with its known avian counterpart. Amongst the examples of Wsh prePOMCs the highest identities are seen in the most primitive ray-Wnned member, the sturgeon, a chondrostean (48.7%) and a similar value

Table 2 Amino acid sequence identity (%) of ostrich prePOMC and its various functional domains in comparison with a range of vertebrate prePOMCs (for references see legend to Fig. 4) Domain b

# of residues Bovine Chicken Mud Turtle Xenopus A Afr. LungWsh Sturgeon A bf Flounder A DogWsh RatWsh Lamprey POM Lamprey POC a b c d

Total prePOMC

Signal peptide

pro--MSHa

-MSH

-MSH

ACTH

-LPH

-MSH

-END

253 54.3 76.9 56.3 53.3 48.2 48.7 43.7 35.6 32.0 19.6 19.8

17 26.9 60.9 34.6 24.0 28.0 16.0 16.7 15.4 28.0 12.5 13.0

63 65.6 88.9 68.9 73.8 65.6 60.7 — 50.8 41.9 — —

12 75.0 100 100 91.7 75.0 75.0 — 83.3 83.3 — —

13 100 100 100 92.3 100 100 100 92.3 92.3 47.6c 40.1

39 79.5 82.1 76.9 76.9 66.7 82.1 75.6 66.7 69.2 — 31.7

46 35.0 67.4 40.8 32.6 32.6 36.2 34.9 14.9 12.3 19.3 —

18 61.1 94.4 72.2 64.7 64.7 70.6 47.1 61.1 60.0 36.8d —

31 74.2 80.6 74.2 67.7 64.7 58.8 39.5 58.3 39.4 28.6 31.3

NH2-terminal POMC (Naudé et al., 1993). Numbers refer to ostrich prePOMC. MSH-B (Takahashi et al., 2001). MSH-A (Takahashi et al., 2001).

316

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317

(48.2%) for the lobe-Wnned African lungWsh, which is considered to be amongst the closest living relatives of the tetrapods. Lower identities are revealed in the more advanced teleost, the bar-Wnned Xounder (43.7%) and in the chondrichtyes (35.6% for the dogWsh and 32.0% for the ratWsh). It should be noted that the -MSH domain is absent from prePOMC of the primitive agnathans (the lamprey) and of the advanced ray-Wnned teleosts of which the bar-Wnned Xounder is an example. This in part contributes to their lowered amino acid sequence identities with ostrich and other tetrapod prePOMCs where the -MSH domain is present. Likewise the presence of the -MSH domain in the chondrichtyes, the dogWsh and ratWsh, contributes to their lowered sequence identities with ostrich prePOMC where the latter domain and the accompanying CTED segment are absent. As can be expected, functionally signiWcant sequences in ostrich prePOMC tend to be more highly conserved (Fig. 4 and Table 2). More divergent regions, which reXect increased rates of mutation during the evolution of POMC, separate the highly conserved regions, for example the region between the - and -MSH domains that is the joining peptide segment between Asn89 and Ser131 in the ostrich POMC sequence. Likewise the N-terminal part of the ostrich -LPH sequence (Ala175–Glu200) separating the corticotropin-like intermediate lobe peptide (CLIP) region in ACTH from the -MSH domain, reXects relatively low levels of sequence conservation. Such highly divergent regions can be expected to imply low or no physiological signiWcance. The relative closeness in the evolutionary development of mammals, birds, reptiles and amphibians, is also reXected in the detailed primary structures of their respective prePOMC molecules that are compared in Fig. 4 and Table 2 for representatives of each of these four classes. Amino acid sequence identities in comparison to the ostrich prePOMC fall within a narrow range of approximately 53–56% for the mammalian (bovine), reptilian (mud turtle), and amphibian (Xenopus) examples, but steeply increase to 77% for the avian example, the chicken, as might have been expected. For the reptiles the complete primary sequence of prePOMC is available for the mud turtle (Fig. 4) and the leopard gecko (Endo and Park, 2004), and revealed sequence identities of 56.3 and 55.9%, respectively, as compared to ostrich prePOMC. The present study would seem to indicate that the sequence identities of the two reptilian prePOMCs reach only moderate levels (56%) as compared to ostrich prePOMC, whereas chicken prePOMC revealed a much elevated level of identity (77%) with the ostrich counterpart. Notwithstanding this apparent closeness in molecular architecture of the two avian molecules, it is nevertheless also true that, as noted within other vertebrate classes, considerable diVerences in molecular structure may arise during the evolutionary development of diVerent orders within the same class. This is evident also for chicken prePOMC (order: Galliformes) when compared to its ancient avian relative, ostrich prePOMC (order: Struthioniformes) from which it diVers in 57 of its total of 251 amino acid residues (a 22.7% diVerence).

In conclusion, by having cloned and characterized the ostrich prePOMC cDNA sequence, the authors have conWrmed the presence in its deduced amino acid sequence, of the various functional peptide domains that were previously characterized in ostrich pituitary isolates by means of direct polypeptide sequencing. The occurrence of the N-terminal POMC (including -MSH), ACTH (including -MSH), -LPH (including -MSH) and the C-terminal -END domains, conWrms that the overall molecular architecture of ostrich prePOMC Wts into the general scheme of tetrapod prePOMCs. Furthermore, the amino acid sequence of the ancient ostrich prePOMC, appears to be most closely related to that of its fellow avian counterpart, the domestic chicken. Acknowledgments The authors wish to acknowledge the technical assistance of Satoshi Noro and Takamasa Kumagai, Kitasato University. This study has been supported by grants from the South African National Research Foundation (NRF) and the Nelson Mandela Metropolitan University (formerly the University of Port Elizabeth). The secretarial assistance of Heather Hundleby and Hettie Oelofsen is appreciated. References Amemiya, Y., Takahashi, A., Dores, R.M., Kawauchi, H., 1997. Sturgeon proopiomelanocortin has a remnant of -melanotropin. Biochem. Biophys. Res. Commun. 230, 452–456. Amemiya, Y., Takahashi, A., Suzuki, N., Sasayama, Y., Kawauchi, H., 1999a. A newly characterized melanotropin in proopiomelanocortin in pituitaries of an elasmobranch, Squalus acanthias. Gen. Comp. Endocrinol. 114, 387–395. Amemiya, Y., Takahashi, A., Meguro, H., Kawauchi, H., 1999b. Molecular cloning of lungWsh proopiomelanocortin cDNA. Gen. Comp. Endocrinol. 115, 415–421. Amemiya, Y., Takahashi, A., Suzuki, N., Sasayama, Y., Kawauchi, H., 2000. Molecular cloning of proopiomelanocortin cDNA from an elasmobranch, the stingray, Dasyatis akajei. Gen. Comp. Endocrinol. 118, 105–112. Berghman, L.R., Devreese, B., Verhaert, P., Gerets, H., Arckens, L., VandenBroeck, J., Van Beeumen, J., Vaudry, H., Vandesande, F., 1998. The molecular characterization of chicken pituitary N-terminal proopiomelanocortin (POMC). Mol. Cell. Endocrinol. 142, 119–130. Dores, R.M., Gieseker, K., Steveson, T.C., 1994. The posttranslational modiWcation of -endorphin in the intermediate pituitary of the toad, Bufo marinus, includes processing at a monobasic cleavage site. Peptides 15, 1497–1504. Endo, D., Park, M.K., 2004. Molecular characterization of the leopard gecko POMC gene and expressional change in the testis by acclimation to low temperature and with a short photoperiod. Gen. Comp. Endocrinol. 138, 70–77. Li, C.H., Chung, D., Oelofsen, W., Naudé, R.J., 1978. Adrenocorticotropin 53. The amino acid sequence of the hormone from the ostrich pituitary gland. Biochem. Biophys. Res. Commun. 81, 900–906. Litthauer, D., Naudé, R.J., Oelofsen, W., 1984. Isolation, characterization and primary structure of two -LPH variants from ostrich pituitary glands. Int. J. Pept. Protein Res. 24, 309–315. Martens, G.J.M., 1986. Expression of two proopiomelanocortin genes in the pituitary gland of Xenopus laevis: complete structures of the two preprohormones. Nucleic acids Res. 14, 3791–3798.

R. Naudé et al. / General and Comparative Endocrinology 146 (2006) 310–317 Nakanishi, S., Inoue, A., Kita, T., Nakamura, M., Chang, A.C.Y., Cohen, S.N., Numa, S., 1979. Nucleotide sequence of cloned cDNA for bovine corticotropin--lipotropin precursor. Nature 278, 423–427. Naudé, R.J., Oelofsen, W., 1977. The isolation and characterization of corticotropin from the pituitary gland of the ostrich, Struthio camelus. Biochem. J. 165, 519–523. Naudé, R.J., Oelofsen, W., 1981. Isolation and characterization of -lipotropin from the pituitary gland of the ostrich, Struthio camelus. Int. J. Pept. Protein Res. 18, 135–137. Naudé, R.J., Oelofsen, W., Maske, R., 1980. Isolation, characterization and opiate activity of -endorphin from the pituitary gland of the ostrich, Struthio camelus. Biochem. J. 187, 245–248. Naudé, R.J., Chung, D., Li, C.H., Oelofsen, W., 1981a. -Lipotropin: primary structure of the hormone from the ostrich pituitary gland. Int. J. Pept. Protein Res. 18, 138–147. Naudé, R.J., Chung, D., Li, C.H., Oelofsen, W., 1981b. -Endorphin: primary structure of the hormone from the ostrich pituitary gland. Biochem. Biophys. Res. Commun. 98, 108–114. Naudé, R.J., Litthauer, D., Oelofsen, W., Chrétien, M., Lazure, C., 1993. The production of the ostrich NH2-terminal POMC fragment requires cleavage at a unique signal peptidase site. Peptides 14, 519–529. Salzet, M., Salzet-Raveillon, B., Cocquerelle, C., Verger-Bocquet, M., Pryor, S.C., Rialas, C.M., Laurent, V., Stefano, G.B., 1997. Leech immunocytes contain proopiomelanocortin: nitric oxide mediates hemolymph proopiomelanocortin processing. J. Immunol. 159, 5400–5411. Shen, S.-T., Lu, L.-M., Chen, J.-R., Chien, J.-T., Yu, J.Y.-L., 2003. Molecular cloning of proopiomelanocortin (POMC) cDNA from mud turtle, Pelodiscus sinensis. Gen. Comp. Endocrinol. 131, 192–201.

317

Smith, I., 1980. The ACTH-LPH family tree—a continuing story. Biochem. Educ. 8, 1–4. Stefano, G.B., Salzet-Raveillon, B., Salzet, M., 1999. Mytilus edulis hemolymph contains pro-opiomelanocortin: LPS and morphine stimulate diVerential processing. Mol. Brain Res. 63, 340–350. Takahashi, A., Amemiya, Y., Nozaki, M., Sower, S.A., Kawauchi, H., 2001. Evolutionary signiWcance of proopiomelanocortin in agnatha and chondrichtyes. Comp. Biochem. Physiol. B 129, 283–289. Takahashi, A., Itoh, T., Nakanishi, A., Amemiya, Y., Ida, H., Meguro, H., Kawauchi, H., 2004. Molecular cloning of proopiomelanocortin cDNA in the ratWsh, a holocephalan. Gen. Comp. Endocrinol. 135, 159–165. Takahashi, A., Amano, M., Itoh, T., Yasuda, A., Yamanome, T., Amemiya, Y., Sasaki, K., Sakai, M., Yamamori, K., Kawauchi, H., 2005. Nucleotide sequence and expression of three subtypes of proopiomelanocortin mRNA in barWn Xounder. Gen. Comp. Endocrinol. 141, 291–303. Takeuchi, S., Teshigawara, K., Takahashi, S., 1999. Molecular cloning and characterization of the chicken proopiomelanocortin (POMC) gene. Biochim. Biophys. Acta 1450, 452–459. Takeuchi, S., Takahashi, S., Okimoto, R., Schiöth, H.B., Boswell, T., 2003. Avian melanocortin system: -MSH may act as an autocrine/paracrine hormone. Ann. N.Y. Acad. Sci. 994, 366–372. Yasuda-Kamatani, Y., Yasuda, A., 2000. IdentiWcation of orcokinin generelated peptides in the brain of the crayWsh Procambarus clarkii by the combination of MALDI-TOF and on-line capillary HPLC/Q-Tof mass spectrometries and molecular cloning. Gen. Comp. Endocrinol. 118, 161–172.