THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 256, No. I, h u e of January 10, pp. 257-261. 1983 Prrnted in US.A.

Complete Amino Acid Sequence of Mouse Pro-opiomelanocortin Derived from the Nucleotide Sequence of Pro-opiomelanocortin cDNA*

(Received for publication, July 20, 1982)

Michael Uhler$ and Edward Herbert From the Department of Chemistry, University of Oregon, Eugene, Oregon 97403

Polyadenylated RNA was isolated from a mouse pi- tuitary tumor cell line (AtT-20/DL6v) which synthe- sizes and secretes adrenocorticotropic hormone and 8- endorphin. The RNA was used to construct a cDNA library by a double linker technique and the library was screened for pro-opiomelanocortin (POMC) se- quences. One recombinant plasmid, pMKSU16, con- tained a 923-base pair insert comprising the entire POMC-coding sequence as well as 98 bases of 5’ non- coding and 5 bases of 3’ noncoding sequence. The pro- tein sequence predicted by the cDNA shows mouse POMC to consist of 235 amino acids with seven poten- tial tryptic cleavage sites consisting of pairs of basic amino acid residues. Comparison with the previously published bovine POMC sequence suggests certain re- gions of POMC are highly conserved between the two species, particularly the regions corresponding to the a-, 8-, and y-melanocyte-stimulating hormones.

POMC’ is a 30,000-dalton polyprotein which contains within it the amino acid sequence ofthe polypeptide hormone ACTH, as well as the endogenous opiate peptide, /3-endorphin (1-4). Although POMC contains the sequences of several biologi- cally active peptides, none of these peptides are active in the form of the precursor. Biological activity is only achieved after the peptides are cleaved from the precursor and properly modified by glycosylation, amidation, or acetylation. Cleavage of the biologically active peptides from the precursor is usually a t sites in the protein sequence where pairs of basic amino acids occur, presumably by the action of a trypsin-like enzyme. Several other polyproteins which share structural features in common with POMC have been identified to date including the proenkephalin precursor (5-77, the vasopressin-neurophy- sin precursor (8), and the a-factor precursor (9). However, POMC is unique among these polyproteins in that its com- ponent peptides possess very diverse biological activities.

In the anterior and intermediate lobes of mouse and rat pituitary, POMC is proteolytically processed to different end products and its synthesis is under the control of different regulators in these two tissues (10-12). In the anterior lobe, POMC is processed to an NH,-tenninal fragment, ACTH, and PLPH, while in the intermediate lobe, these peptides are processed further (13). The NH2-terminal fragment is proc-

* This work was supported by National Institutes of Health Grant AM16879 and by National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Grant ROI AM30155. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “adcrertjsement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

+ Present address, Department of Genetics and Medicine, Stanford University Medical School, Stanford, CA 94305.

ACTH, adrenocorticotropic hormone; P-LE’H, P-lipotropin. ’ The abbreviations used are: POMC, pro-opiomelanocortin;

. ” ~ _ _ _ ___

essed to pro y-MSH; ACTH gives rise to a-MSH and corti- cotropin-like intermediate lobe peptide, and /3-LPH is cleaved to y-LPH and P-endorphin. For this reason the study of POMC gene structure and expression is of great interest.

Isolation and sequencing of the full length bovine POMC cDNA (4), the bovine (14) and human genes (15), as well as partial sequencing of the rat POMC (16) gene has been reported. We have chosen mouse as a species to study POMC gene expression at the nucleic acid level for three reasons. First, characterization of the pathways and regulation of pro- teolytic processing has been best studied in a mouse tumor cell line and in mouse anterior pituitary primary cultures. Second, a number of experimental systems are available for study of POMC expression in the mouse including the tumor cell line, primary cultures, and whole animals. Finally, the well developed genetic studies of the mouse make available such tools as mouse-Chinese hamster cell hybrids (17) with isolated mouse chromosomes and inbred mouse strains which may be altered in the expression of POMC (18).

In the work presented here, we report as an initial step in the study of mouse POMC gene expression the characteriza- tion of the mouse POMC mRNA and the use of the double linker technique of cDNA cloning for the isolation and se- quencing of a full length cDNA clone for mouse POMC.

MATERIALS AND METHODS

AtT-20/D16v cells were cultured in Dulbecco’s modified Eagle’s medium (Grand Island Biological) (11). Polysomal RNA was isolated by the magnesium precipitation procedure of Palmiter (19) and chro- matographed on oligo-dT cellulose (20). Solution hybridization using MI3 strand separation was performed as described (21). Northern blot analysis followed the procedure described by Thomas (22 ) . pMKSU16 and ME150 plasmid DNAs were prepared and, after digestion with restriction enzymes, the inserts were purified away from vector DNA by chromatography on Sepharose ZB-CL in 100 m~ NaCl, 10 mM Tris, pH 7.4, and I mM EDTA. Commercially available nick translation kits were used (New England Nuclear or Bethesda Research Laboratories).

The strategy and protocol of the double linker technique employed is described elsewhere (23). Briefly, double-stranded cDNA was syn- thesized and a mixture of EcoRI and Hind11 linkers were ligated to the cDNA. Excess linkers ligated to the cDNA were removed by digestion with EcoRl and HzndIII. Vector DNA was prepared by digestion with EcoRI and Hind l I , removal of the small 29-base pair fragment on a 10-40% sucrose gradient, and phosphatase treatment of the EcoRI/HindIII pBR322 vector. A 5-fold molar excess of vector over cDNA was used during the ligation with a vector concentration of 1.5 pg/ml. The ligated DNA was then used to transform the Escherichia coli strain HRI.

Ampicillin-resistant transformants were selected and the colonies were screened by hybridization (24) with nick-translated ME150 (25). The cloned inserts were characterized by preparing small amounts of the plasmid DNAs using the method of Birnboim and Doly (26) and digesting with EcoRI and HindIII. The pMKSU16 insert was se- quenced by digesting total plasmid DNA with either AZuI, FnuDII, HueIII, or RsuI and cloning these fragments into the SmuI site of M13mp8 (27). The resulting plaques were transferred to nitrocellulose and the fiters hybridized with nick-translated pMKSU16 insert.

257

258 Sequence of Mouse Pro-opiomelanocortin and Its mRNA

Hybridizing plaques were picked and grown for sequencing by the dideoxy termination technique.

RESULTS

Total and polyadenylated polysomal RNA was isolated from the mouse pituitary tumor cell line, AtT-20/D16v, as described by Palmiter (19). As shown in the Northern blot of Fig. 1, hybridization with nick-translated ME150 demon- strates that P-endorphin-related mRNA migrates with an apparent size of 1070-1 150 nucleotides. Furthermore, solution hybridization results with ME150 indicated this mRNA spe- cies comprised 0.11% of total polysomal RNA and 3.1% of polyadenylated polysomal RNA. Employing previously re- ported methods, double-stranded cDNA was synthesized us- ing the AtT-20/D16v polyadenylated polysomal RNA as tem- plate.

In order to optimize cloning efficiency of the double- stranded cDNA, the effect of various pretreatments of the cloning vector, pBR322, on background transformation was

-4512

POMC mRNA -

-2822

-1 775

-1153

c“ 831

-527

-316

FIG. 1. Northern blot analysis of AtT-20 polyadenylated RNA. Twenty micrograms of polyadenylated AtT-20 RNA (left) and size marker restriction fragments of pBR322 containing the ME150 insert (30) were size fractionated on a 1.7% agarose gel, transferred to nitrocellulose, and hybridized with nick-translated Me150 insert. The various size markers indicated were obtained by the following diges- tions of pBR322 containing the ME150 insert at the Hind11 site: 4512, EcoRI; 2822, BamHI and PuaII; 1775, HinfI; 1153, BamHI and PuuI; 831, RsaI; 527, EcoRI and BamHI; 316, EcoRI and RsaI.

investigated. The motivation behind these experiments was that by lowering the background of transformation, a large molar excess of vector could be used during the ligation without introduction of a substantial number of parental vector transformants. The results of these pretreatments are shown in Table I. By comparing the transformants/ng for the ligated EcoRI/HindII-digested pBR322 with EcoRI-digested pBR322, it can be seen that a double restriction digestion can reduce the transformation background approximately 100-fold compared to a single digestion. The higher background ob- tained with the EcoRI/PvuI digestion is presumably due to a greater contamination of the cloning vector by the 626-base pair fragment during sedimentation through the sucrose gra- dient than that which occurred with the 26-base pair EcoRI/ Hind11 fragment. In addition, phosphatase treatment of the EcoRI/HindII vector reduced the transformation back- ground another 100-fold. In light of these results, the double- stranded cDNA was ligated with mixed EcoRI and Hind11 linkers and the ligation products digested with both EcoRI and HindII. The cDNA was then inserted into the phospha- tase-treated EcoRI/HindII/pBR322 vector and the recom- binant plasmids used for transformation of E. coli RRI. Ap- proximately 60,000 independent clones were obtained and these were colony hybridized with nick-translated ME150. Initially, screening of several clones which hybridized to ME150 by size analysis of their inserts indicated that one plasmid, pMKSU16, contained a 923-base pair insert including linkers. Sequencing of this inserted DNA fragment showed that it contains a near full length cDNA for mouse POMC. The sequence of the pMKSU16 insert is shown in Fig. 2 compared to the published bovine POMC cDNA sequence (4), the only other POMC cDNA that has been sequenced to date.

The mouse POMC cDNA codes for a protein of 235 amino acids and contains 98 bases of 5’ noncoding and 5 bases of 3’ noncoding regions. The f i t 26 amino acids comprise the signal sequence of mouse POMC, as evidenced by the high percentage of hydrophobic amino acids in this region. The sequence of this region reported here is consistent with the partial protein sequence determined previously, with the ex- ception of a methionine reported at position number 3 of the

TABLE I Transformation backgrounds of various cloning vectors

Ten and 100 ng of the designated vectors, prepared as described under “Materials and Methods,” were used to transform E. coli RRI and transformants were transferred to plates containing 25 pg/ml of tetracycline or 50 p g / d of ampicillin. After incubation at 37 “C for 18 h, colonies were counted and transformants/ng of vector calculated.

Modification

Phos- Vector used pha- Ligase Ampicillin Tetracycline

treat- ment tase treat-

ment Number of resistant trans-

Supercoiled pBR322 formants/ng of vector

Blank control - - <0.01 t O . O 1 EcoRI-treated - - 2.7 3.8

- - 1030 1160

pBR322 - + 167 150 + + 0.40 0.43

EcoRI/HindII large - - 0.02 0.02

+ 1.2 0.23 + + t O . O 1 t O . O 1

EcoRI/PuuI large - - 0.43 0.60

fragment -

fragment - + 67 50 + + 0.1 0.23

POMC cDNA and protein sequence FIG. 2. Comparison of mouse

(upper) with bovine POMC cDNA (4) and protein sequence (lower). The amino acid numbers for mouse POMC are shown above the protein sequence, nucleotide differences between the mouse and bovine cDNAs are denoted with an asterisk and corresponding bo- vine nucleotide and amino acid sequence shown below. The dashed line in the mouse sequence indicates a gap intro- duced to show homology. When no base appears in the bovine sequence, it is be- cause the base is identical with the one in the mouse sequence.

precursor (28). The cleavage of the signal sequence from the precursor is predicted to take place between the Ser26 and Trp" residues by comparison with sequences reported for mature mouse POMC (28). Residues 27 through 121 corre- spond to the NHz-terminal region of mouse POMC. Compar- ison with two independent partial protein sequences shows the sequence predicted here is in total agreement with the exception of conflicting reports for the amino acid assignment at position 33 (28, 29). The sequence of the mouse POMC cDNA is consistent with a glutamine at this position rather than a valine. The dibasic amino acid sequence Ly~"~-Arg"~ separates the NHs-terminal region from the 39 amino acid mouse ACTH sequence, while a similar dibasic amino acid pair, Lys'"-Arg'"', separates ACTH from P-LPH. The se-

quences of the ACTH and P-LPH regions agree with the previously published protein sequences.

DISCUSSION

Three regions of homology exist in the nucleotide sequence of the pMKSU16 insert corresponding to the sequences of a-, p-, and y-MSH. These regions are amino acids 77-87 (y - MSH), 124-137 (a-MSH), and 188-199 (P-MSH). The se- quence CACTTCCGCTGG, corresponding to the protein se- quence His-Phe-Arg-Trp, is strictly conserved in all three regions. Although the function of these peptides is not known in mammals, the conservation of these sequences and the high degree of homology in surrounding sequences suggests these peptides interact with a common recognition protein, although

260 Sequence of Mouse Pro-opiomelanocortin and Its mRNA 27 121 124 162 165 235

N- termfinal pgtlda (95)

ACTH (39 )

p -LPH (71 T ANTERIOR LOBE PROCESSING

27

INTERMEDIATE LOBE PROCESSING

74 77 100 103 121 0 4 I37 142 162 ms 202 205 235

FIG. 3. Proteolytic processing sites of mouse POMC in the anterior (upper) and intermediate (lower) lobes of the pituitary.

in slightly different ways. In addition, these regions of homol- ogy as well as others suggest that the POMC gene has evolved by a series of duplications of a common ancestral DNA sequence followed by subsequent substitution, addition, or deletion of other sequences.

A comparison of the 5' end of the mouse cDNA with the same region of the bovine cDNA (4) suggests that the mouse cDNA terminates 19 bases before the 5' end of the mRNA. By comparing the 3' end of both sequences and assuming the poly(A) tail is added at the same position relative to the sequence, AAAATAAAA, thought to be involved in poly(A) addition, it can be deduced that approximately 8 bases are missing from the 3' end of the mouse cDNA. The length of the mRNA would be 935 bases not including the poly(A) tail. Since Northern blot analysis has shown the mouse POMC mRNA to range in size from 1070 to 1150 bases, this would imply an average length of 135-215 bases for the poly(A) tail, which is slightly larger than the poly(A) tails of other mRNA molecules. Alternatively, it is possible that the mouse POMC mRNA has a larger 5' noncoding sequence than the bovine mRNA.

Fig. 2 also shows that the bovine cDNA contains insertion sequences in both the 5' noncoding region (15 bases inserted between nucleotides 54 and 55 of the mouse sequence) and 3' noncoding region (60 bases inserted between nucleotides 835 and 836 in the mouse sequence) as well as two in the NH2- terminal coding sequence (9 bases inserted between nucleo- tides 365 and 366 and 15 bases inserted between nucleotides 440 and 441 in the mouse sequence) and one in the y-LPH region (66 bases inserted between nucleotides 611 and 612 in the mouse sequence). The y-MSH, ACTH, and /3-endorphin sequences all appear to be highly conserved between the two species, suggesting these may be the most functionally impor- tant regions of the precursor. The 5' noncoding region seems to be more highly conserved between the two species than the 3' noncoding region; however, a marked homology between the two sequences is noted near the poly(A) addition site.

Seven dibasic amino acid cleavage sites are predicted from the mouse POMC cDNA shown, although only six of these are known to be cleaved in uiuo (Fig. 3). The two sites marked on either side of the ACTH sequence are those that are cleaved during processing in the anterior lobe and separate the NHz-terminal fragment (71 amino acids) from ACTH. The other cleavage sites are sites of proteolysis recognized only in the intermediate lobe of the pituitary and would generate 3 peptides of undetermined function from the NH2- terminal peptide of 48, 24, and 19 amino acids, o-MSH (14 amino acids), and corticotropin-like intermediate lobe peptide (21 amino acids) from ACTH.

In conclusion, sequencing of the POMC-coding cDNA insert of pMKSU16 predicts an amino acid sequence entirely con- sistent with both the mouse POMC protein sequencing and characterization performed to date as well as the bovine POMC cDNA sequence previously published. Also, compari- son of the bovine and mouse cDNA sequences suggests that a-MSH, ACTH, and P-endorphin play an important enough role in hormonal and neuromodulator functions of the POMC precursor to be highly conserved between the two species. Finally, the availability of a full length mouse cDNA will allow the detailed molecular analysis of POMC gene expres- sion in an experimentally manipulable system. Most impor- tant among these questions will be that of whether the appar- ent tissue specific processing of POMC in the anterior and intermediate lobe of the pituitary is the result of expression of multiple POMC genes.

REFERENCES

1. Mains, R. E., Eipper, B. A,, and Ling, N. (1977) Proc. Natl. Acad.

2. Roberts, J . L., and Herbert, E. (1977) Proc. Natl. Acad. Sci. U. S.

3. Roberts, J . L., and Herbert, E. (1977) Proc. Natl. Acad. Sci. U. S.

4. Nakanishi, S., Inoue, A,, Kita, T., Nakamura, M., Chang, A. C. Y., Cohen, S. N., and Numa, S. (1979) Nature (Lond.) 278, 423- 427

5. Comb, M., Seeburg, P. H., Adelman, J., Eiden, L., and Herbert, E. (1982) Nature (Lond.) 295,663-666

6. Noda, M., Furutani, Y., Takahashi, H., Toyosata, M., Hirose, T., Inayama, S., Nakanishi, S., and Numa, S. (1982) Nature (Lond.) 295,202-206

7. Gubler, V., Seeburg, P., Hoffman, B. J., Gage, L. P., and Uden- friend, S. (1982) Nature (Lond.) 295, 206-209

8. Land, H., Schutz, G., Schmale, H., and Richter, D. (1982) Nature (Lond.) 295,299-303

9. Kurgan, J., and Herskowitz, I. (1982) Cell, in press

Sci. U. S. A. 74,3014-3018

A. 74,4826-4830

A . 74, 5300-5304

10. Roberts, J . L., Phillips, M., Rosa, P. A., and Herbert, E. (1978) Biochemistry 17,3609-3618

11. Herbert, E., m e n , R. G., and Paquette, T. L. (1978) Endocrinol- ogy 102,218-226

12. Rosa, P., Policastro, P., Budarf, M., and Herbert, E. (1980) in Biosynthesis, Modification, and Processing of Cellular and Viral Polyproteins (Koch, G., and Richter, D.. eds) pp. 127- 138, Academic Press, New York

13. Rosa, P., Policastro, P., and Herbert, E. (1980) J. Exp. Bioi. 89, 215-237

14. Nakanishi, S., Teranishi, Y., Watanabe, Y., Notake, M., Noda, M.. Kakidani, H., Jingami, H., and Numa, S. (1981) Eur. J.

15. Cochet, M., Chang, A. C. Y., and Cohen, S. N. (1982) Nature

16. Drouin, J., and Goodman, H. M. (1980) Nature (Land.) 288,610-

Biochem. 1,429-438 -

(Land.) 297, 335-339

613

Sequence of Mouse Pro-opiomelanocortin and Its mRNA 261

17. D’Eustachio, P., Bothwell, A. L. M., Takaro, T. K., Baltimore, D., and Ruddle. F. J. (1982) J. Exp. Med. 153, 793-800

18. Rosier, J., Rogers, J., Shibasaki, T., Guillemin, R., and Bloom, F. E. (1979) Proc. Natl. Adad. Sci. U. S. A. 76, 2077-2080

19. Palmiter, R. D. (1974) Biochemistry 13, 3606-3615 20. Aviv, H., and Leder, P. (1972) Proc. Natl. Acad. Sci. U. S. A. 69,

21. Beach, L. R., and Palmiter, P. D. (1981) Proc. Natl. Acad. Sci. U.

22. Thomas, P. S. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 5201-

23. Kurtz, D. T., and Nicodemus, C . F. (1981) Gene 13, 145-152 24. Granstein, M., and Hogness, D. (1975) Proc. Natl. Acad. Sci. U.

1408-1412

S. A. 78, 2110-2114

5205

S. A. 72,3961-3965

25. Civelli, O., Birnberg, N., and Herbert, E. (1982) J. Biol. Chem.

26. Birnboim, H. C., and Doly, J. (1979) Nucleic Acids Res. 7, 1513- 1523

27. Messing, J., Crea, R., and Seeburg, P. H. (1981) Nucleic Acids Res. 9,309-321

28. Policastro, P., Phillips, M., Oates, E., Herbert, E., Roberts, J. L.,

259 Seidah, N., and Chretien, M. (1981) Eur. J . Biochem. 116,255-

29. Keutmann, H. T., Lampman, G. W., Mains, R. E., and Eipper, B. A. (1981) Biochemistry 20, 4148-4155

30. Roberts, J. L., Seeburg, P. H., Shine, J., Herbert, E., Baster, J. D. and Goodman, H. M. (1979) Proc. Natl. Acad. Sci. U. S. A. 76,

257,6783-6787

2153-2157