456
`
`Nature Vol. 285 12 June 1980
`
`23. Bleux. C., Ventura, M. & Liacopoulos, P. Nature 267, 709-711 (1977).
`24. Abney, E. R., Cooper, M. D., Kearney, J. F., Lawton, A. R. & Parkhouse, R. M. E. J.
`/mmun. 120, 2041-2049 (1978).
`25. Kincade, P. W., Lawton, A. R., Bockman, D. E. & Cooper, M. D. Proc. natri. Acad. Sci.
`U.S.A. 67, 1918-1925 (1970).
`26. Pernis, B., Chiappino, G., Keus, A. S. & Gell, P. G. H.J. exp. Med. 122, 853--875 (1965).
`27. Cebra, J. J., Colberg, J.E. & Dray, S. J. exp. Med. 123, 547-558 (1966).
`28. Herzenberg, L. A. et al. Cold Spring Harb. Symp. quant. Biol. 16, 33-45 (1976).
`29. Hood, L., Kronenberg, M., Early, P. & Johnson, N. ICN-UCU Symp. molec. cell. Biol. 6,
`1-27 (1977).
`30. Williamson, A. R. & Fitzmaurice, L. C. in The Generation of Antibody Diversity. A New
`Look (ed Cunningham, A. J.) 183-211 (Academic, London, 1976).
`31. Rabbitts, T. H. Natu'C 275, 291 -296 (1978).
`32. Tonegawa, S., Maxam, A. M., Tizard, R., Bernard, 0. & Gilbert, W. Proc. natn. Acad. Sci.
`U.S.A. 74, 4406-4410 (1978).
`33. Marcu, K. B .. Schibler, U. & Perry, R. P. Science 204, 1087-1088 (1979).
`34. Dreyer, W. J., Gray, W. R. & Hood, L. Cold Spring Harb. Symp. quant. Biol. 32, 353-368
`(1967).
`35. Galley, J. A. & Edelman, G. M. Nature 227, 341-348 (1970).
`36. Kabat, D. Science 175, 134-140 (1972).
`37. Honjo, T. & Kataoka, T. Proc. natn. Acad. Sci. U.S.A. 75, 2140-2144 (1978).
`38. Adams, J. M., Gough, N. M., Webb, E. & Cory, S. Biochemistry (in the press).
`39. Dayhoff. M. 0. (ed.) Atlas of Protein Sequence and Structure 5 Suppl. 3, 227 (1978).
`
`40. Birshtein, B. K., Campbell, R. & Greenberg, M. L. Biochemistry (in the press).
`41. Yamawaki-Kataoka, Y. et al. Biochemistry 18, 490-497 (1979).
`42. Southern, E. M. J. molec. Biol. 98, 503 517 (1975).
`43. Sakano, H. et al. Nature 277, 627-632 (1979).
`44. Honjo, T. et al. Ce/118, 559-568 (1979).
`45. Kataoka, T., Yamawaki.Kataoka, Y., Yamagishi, H. & Honjo, T. Proc. natn. Acad. Sci.
`U.S.A. 76, 4240-4244 (1979).
`46. Tucker, P. W., Marcu, K. B., Newell, M., Richards, J. & Blattner, F. R. Science 206,
`1303-1306.
`47. Lis, J .. Prestidge, L. & Hogness, D.S. Ce/114, 901 919 (1978).
`48. Kemp, D. J., Cory, S. & Adams, J. M. Proc. nutn. Acad. Sci. U.S.A. 76, 4627 4631 ( 1979).
`49. Yoshida, T. H., Imai, H. T. & Potter, M. J. natn. Cancer Inst. 41, 1083-1098 (1968).
`50. Ferrarini, M., Corte, G., Viale, G., Durante, M. L. & Bargellesi, A. Fur. J. Jmmun. 6, 372-
`380 (1976).
`51. Sprent, J. & Miller, J. F. A. P. Eur. J. lmmun. 2, 384--387 (1972).
`52. Lenhard·Schuller, R., Hohn, B., Brack, C., Hirama, M. & Tonegawa, S. Proc natn. Acad.
`Sci. U.S.A. 75, 4709-4713 (1978).
`53. Steinmetz, M., Zachau, H. G. & Mach, B. Nucleic Acids Res. 6, 3213- 3229 (1979).
`54. Rabbitts, T. H., Forster, A., Dunnick, W. & Bentley, D. L. Nature 283, 351--356 (1980).
`55. Coleclough, C., Cooper, D. & Perry, R. P. Proc. natn. Acad. Sci. l.I.S.A. (in the press).
`56. Horibata, K. & Harris, A. Exp/ Cell Res. 60, 61-77 (1970).
`57. Rigby, P. J., Dieckmann, M., Rhodes, C. & Berg, P. J. molec. Biol. 113, 237 251 (1977).
`58. Laskey, R. A. & Mills, A. D. FEBS Lett. 82, 314-316 (1977).
`
`Expression of cloned P-endorphin gene
`sequences by Escherichia coli
`James L. Robertst
`John Shine*, Ivy Fettest,
`Nancy C. Y. Lant,
`John D. Baxtert§
`&
`*Molecular Biology .Unit, Research School of Biological Sciences, Australian National University, Canberra, Australia
`tHoward Hughes Medical Institute Laboratories, Departments of Medicine and Biochemistry and Biophysics, and the Metabolic Research Unit,
`University of California, San Francisco, California 94143
`tDepartment of Biochemistry, Columbia University, College of Physicians and Surgeons, New York, New York 10032
`
`DNA coding for the opiate peptide {3-endorphin has been cloned into bacterial plasmids in such a way as to direct the
`synthesis of a hybrid {3-galactosidase/{3-endorphin protein. This hybrid protein can readily be cleaved in vitro to release
`biologically active {3-endorphin.
`
`·----~---------
`
`(ACTH//J-LPH precursor) which also contains the sequences
`of corticotropin (ACTH) and P-lipotropin (/J-LPH)9 -P We
`report here the expression, in Escherichia coli, of cloned P(cid:173)
`endorphin sequences as a hybrid /J-galactosidase//j-endorphin
`protein and describe the release, from the hybrid protein, of
`mature, biologically active P-endorphin.
`
`the programmed
`to
`two basic approaches
`THERE are
`expression of mammalian gene sequences in bacteria. In one,
`the cloned mammalian coding sequences are spliced into the
`bacterial gene such that the bacterial initiation codon is
`followed directly by the codon for the first amino acid of the
`mammalian protein 1
`. In the other, the coding sequence is
`inserted into an internal position in a bacterial gene so that
`the sequences are in phase with the bacterial coding sequences,
`which results in the synthesis of a hybrid bacterial-mammalian
`protein, from which
`the mammalian protein has
`to be
`released 2 - 8 . Although release can be a problem, there may be
`several advantages to the hybrid protein approach. First, it
`the chances of a disturbance of the mRNA
`minimizes
`secondary structure in the region of the normal ribosome
`binding site and initiation codon, causing a reduced rate of
`initiation of protein synthesis. Second, degradation of the
`mammalian protein in the bacterium may be reduced when it
`1s
`fused. These
`factors may
`increase
`the yield of
`the
`mammalian protein. Third, construction of plasmids for
`synthesis of a hybrid protein is ordinarily simpler than for
`exact expression; for instance there may be less need for
`3 with
`chemically synthesized DNA 1
`the hybrid protein
`•
`approach. Finally, for direct expression of an exact protein, the
`first amino acid must be methionine which may not be
`cleaved.
`We have, therefore, chosen the hybrid protein approach in
`an attempt to persuade bacteria to synthesize P-endorphin, a
`31 amino acid endogenous opioid. This protein is generated by
`post-translational processing from a larger precursor protein
`
`Construction of a plasmid for expression
`of Jl-endorphin gene sequences
`To programme bacteria
`to
`express P-endorphin gene
`sequences, we used a previously cloned DNA fragment derived
`by reverse transcription of mRNA coding for the mouse
`ACTH//J-LPH precursor protein 18
`. This fragment contains the
`coding information for amino acids 44-90 of P-LPH (Fig. 1).
`This portion of P-LPH is composed of the carboxy-terminal
`15 amino acids of P-melanocyte stimulating hormone (MSH)
`and the complete P-endorphin sequence with the exception of
`the C-terminal glutamine. It was thus necessary to modify this
`cloned fragment in order to recreate the codon for the C(cid:173)
`terminal amino acid, insert a stop codon, link the fragment in
`phase to a bacterial gene, and devise a method for the release
`of mature P-endorphin from the hybrid protein.
`Our approach to these problems is outlined in Fig. 2. The
`the plasmid by H indlll
`cloned fragment,
`released from
`digestion, was first cleaved in the P-MSH coding region with
`the bacterial P(cid:173)
`to facilitate
`later
`'phasing' with
`Hpall
`galactosidase gene. The single stranded H pall and Hind III
`termm1 were
`then
`partially
`'filled-in'
`using
`reverse
`transcriptase, dATP and dCTP; this recreated the 3'-terminal
`glutamine codon. The remaining unpaired terminal nucleotides
`§To whom correspondence should be addressed.
`- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
`© 1980 Macmillan Journals Ltd
`0028--0836/80/240456-06$01.00
`
`Merck Ex. 1069, pg 1535
`
`

`
`Nature Vol. 285 12 June 1980
`
`457
`
`Protein
`
`N
`
`NH2-Termlnol Frooment
`
`ACTH
`
`a-MSH CLIP
`
`LPH
`
`c
`,B-MSH p-Endorphin
`
`mRNA 5'
`
`cONA
`
`Hoe Ill
`
`Hat1lll
`
`Hhol
`
`'
`
`'
`
`AAAAAAA 3'
`
`TTTTTTT
`AAAAAAA
`
`' '
`
`' '
`connectinQ''
`peptide
`-
`10
`15
`18
`5~n
`{ XGG CCC TAC CGG GTG GAG CAC TTC CGC TGG AGC AAC CCG CCC AAG GAC AAG CGT
`,B-MSH
`(Gly) Pro Tyr ArQ Vol Glu His Phe ArQ Trp Ser Asn Pro Pro Lys Asp Lys Arg
`
`5
`10
`15
`TAC GGU GGC TTC ATG ACC TCC GAG AAG AGC CAG ACG CCC CTG GTG ACG CTC TTC
`Tyr Gly Gly Phe Met Thr Ser Glu Lys Ser Gin Thr Pro Leu Vol Thr Leu Phe
`
`,B-Endorphin
`
`20
`25
`31
`30
`AAG AAC GCC ATC ATC AAG AAC GCG CAC AAG AAG GGC c
`Lys Asn Alo
`lie
`lie Lys Asn Alo His Lys Lys Gly
`
`----------------~
`
`Fig. 1 Nucleotide sequence of the previously cloned fragment of cDNA to ACTH ,8-endorphin mRNA used to construct a plasmid for
`expression of ,8-endorphin sequences (from ref. 18). The plasmid containing this fragment (pMAE-1) had been constructed by ligating chemically
`synthesized decamers containing the restriction sites for endonuclease Hindlll to an Haelll digest of the cDNA. This fragment was then cloned in
`the HindIII site of plasmid pBR322. The Haelll sites on either end of the cDNA are shown, but the Hindlll 'linker' sequences are not indicated.
`The protein structure of the ACTH-,8-endorphin precursor and its constituent hormones are shown and the amino acids of ,8-MSH and ,B(cid:173)
`endorphin are numbered. The Hpall endonuclease site at which this DNA was cleaved in the construction of the expression plasmid is also
`indicated (see Fig. 2).
`
`were removed by S1 nuclease digestion and the blunt-ended
`fragment ligated to a synthetic octanucleotide containing the
`recognition site for EcoRI. Apart from providing the EcoRI
`termini necessary for ligation to the /J-galactosidase gene, this
`octanucleotide provided a stop codon directly following the 3' -
`terminal glutamine codon. The modified fragment was then
`ligated into the EcoRI site of the plasmid p/J-gal (pBR322
`carrying the lac control region and the coding sequence for P(cid:173)
`galactosidase2 ). The unique Eco RI site in this plasmid occurs
`at the codon for amino acid 1004 of /J-galactosidase. This
`resulted in the formation of a recombinant plasmid (p/J-gal(cid:173)
`end) carrying a hybrid
`/J-galactosidase-{J-endorphin gene
`sequence.
`
`Expression of the hybrid P-galactosidase/
`P-endorphin gene
`E. coli RRl cells carrying the recombinant plasmids p{J-gal or
`p/J-gal-end were examined for expression of polypeptides under
`control of the lac promoter. Polyacrylamide gel electrophoresis
`of extracts from stationary- or log-phase cells containing p/J(cid:173)
`gal demonstrated the presence of large amounts of the P(cid:173)
`galactosidase peptide, which was absent from uninduced
`control cells (Fig. 3). This high copy number of this plasmid
`(30--50 copies per cell) renders host cells mostly constitutive
`for {J-galactosidase synthesis. Similar analysis of cells carrying
`p/J-gal-end demonstrated the absence of the /J-galactosidase
`peptide and the appearance of a new band reflecting a protein
`approximately 30 amino acids larger than /J-galactosidase-the
`size expected for
`the
`/J-galactosidase-{J-MSH-{J-endorphin
`hybrid protein (Fig. 3e ). This protein is also synthesized under
`the control of the lac promoter, as was demonstrated by its
`induction with isopropyl-{J-thiogalactoside (IPTG; Fig. 3f).
`Although it accounts for several per cent of the total cellular
`protein, the expression of the presumptive hybrid protein
`appears to be reduced in cells carrying the parental p{J-gal
`plasmid as compared to the synthesis of /J-galactosidase.
`However, this level seems to be increased by amplification of
`
`the plasmid DNA sequences (Fig. 3i). As was previously found
`with both the A and B chains of human insulin linked to the
`same /J-galactosidase fragment3, the hybrid protein is insoluble
`and can be recovered from a high speed pellet of cell extracts
`where it represents a substantial proportion of the total
`proteins (Fig. 3h ).
`
`Release and partial purification of biologically
`active P-endorphin
`the
`/J-endorphin from
`The steps used
`to release mature
`presumptive hybrid protein are shown in Fig. 4 and detailed in
`the legend. To do this, we made use of: (1) the arginine residue
`that precedes the /J-endorphin sequence (Fig. 1 ), which can be
`a site for proteolytic cleavage by trypsin; (2) the absence of
`any internal arginines in this hormone (Fig. 1); and (3) the
`observation that the lysine residues in /J-endorphin (Fig. I)
`could be protected from attack by trypsin by reaction with
`citraconic anhydride in vitro 25 • Thus, after dissolving the
`precipitated hybrid protein and treatment with citraconic
`anhydride at pH 9, the P-endorphin (containing the modified
`lysine groups) was released from
`the hybrid protein by
`cleavage with trypsin. Native P-endorphin was subsequently
`produced by removal of the citraconic groups at pH 3. Cell
`extracts prepared in this manner were assayed for /J-endorphin
`activity. The {J-endorphin was further purified from
`this
`preparation with the use of glass extraction as described in
`Fig. 4 legend.
`
`Immunological activity of partially purified
`bacterially synthesized P-endorphin
`To test the immunological activity of the released {J-endorphin,
`a heterologous radioimmunoassay was designed (Fig. 5). The
`antiserum had been raised against mouse pro-ACTH//j-LPH;
`radiolabelled synthetic human f:l-endorphin was used as the
`radioligand. The human hormone differs from that of mouse
`endorphin only at position 27 where a tyrosine is present
`
`
`
`© Nature Publishing Group1980
`
`Merck Ex. 1069, pg 1536
`
`

`
`458
`
`Nature Vol. 285 12 June 1980
`
`Cloned DNA fraoment
`p-endorphln
`~!::!P. ___ a_n~~~~~~~~~~~~~~~~~~~--Hindm
`t
`
`Hindm
`
`Val
`(5')CGG GTG
`C CAC ___ -
`
`- -
`
`- - - - - - - - - -
`
`-
`
`- - - - - - - -
`
`Gly
`GGC
`--CCG
`
`CA
`GTTCGA (5')
`
`dATP, dCTP
`Val
`Gly
`CGG GTG ___________ - - - - ______ - - - _GGC
`
`CC CAC
`
`CCG
`
`Gly
`Val
`GGC
`GG GTG
`CC CAC------- - - - - - - - - - - - - - - - - - --CCG
`
`SI nuclease
`
`Gly
`(TGAATTCA)
`Val
`TGAATTCAGG GTG - - - - - - - - - - - - - - - ___________ GGC
`ACTTAAGTCC CAC
`CCG
`
`Gly
`Aro Val
`AATTC AGG GTG ___________________________ GGC
`
`Gin
`CAA
`GTT CGA
`
`Gin
`CAA
`GTT
`
`Gin
`CAA TGAATTCA
`GTT ACTTAAGT
`
`Gin
`CAA TG
`GTT ACTTAA
`
`l H1>aII endonuclease
`l
`1 Reverse transcrlptase
`l T4 DNA lioaset'RI octomer'
`l Eco RI endonuclease
`l Clone Into Eco RI site
`
`Fig. 2 Construction of a plasmid for the
`expression of P-endorphin sequences by E.
`coli. All
`recombinant DNA experiments
`were
`carried out
`in
`accordance with
`ASCORD (Australia) or NIH guidelines.
`Recombinant plasmids were isolated from
`chloramphenicol-amplified
`cultures
`as
`previously described 19
`• The sources
`for
`restriction endonucleases, RNA-dependent
`DNA-nucleotidyl-transferase
`(reverse
`transcriptase), other enzymes and reagents,
`unless noted otherwise, are provided
`in
`18
`previous publications5
`2 0
`21
`. The cloned
`·
`•
`•
`150-base pair fragment from pMAE-1 (Fig.
`1, ref. 18) was isolated by HindIII digestion
`and electrophoresis in a 7 % polyacrylamide
`gel. Digestion of 500 ng DNA with 5 U of
`HpaIJ was carried out at 37 °C for 1 h in
`20µ1 6mM Tris (pH 7.5), 6mM MgCI and
`6 mM P-mercaptoethanol. Following phenol
`extraction and ethanol precipitation,
`the
`DNA was dissolved in 50µ1 of a reaction
`mixture containing 500 µM each of dATP
`and dCTP. Incubation was carried out at
`37 °C with 10 U of reverse transcriptase for
`15 min. Unincorporated triphosphates were
`then
`removed by
`chromatography on
`Sephadex G-50
`and
`after
`ethanol
`precipitation, the DNA was dissolved in
`50µ1 of 0.3 M NaCl, 30mM Na acetate pH
`4.5, 3 mM ZnC1 2 and incubated with 5 U of
`S1 nuclease at 20 °C for 5 min followed by
`phenol extraction and Sephadex G-50
`chromatography. The duplex DNA was then
`ligated to a 20-fold molar excess of octamers
`('linkers') containing the restriction site for
`endonuclease Eco RI in 40 µl of 60 mM Tris (pH 7.5), 8 mM MgC1 2 , lOmM P-mercaptoethanol, 1 mM ATP, using 2 µl of bacteriophage T 4
`DNA ligase (New England Biolabs) at 4 °C for 24 h. Following digestion with 50 units of Eco RI at 37 °C for 5 h, the linker fragments were
`removed by Sephadex G-50 chromatography and the DNA ligated to 1 µg of EcoRI-cleaved, and alkaline phosphatase treated 2 1 p/)-gal
`DNA2, and transformed in E. coli RR1 22 . Plasmid DNA was isolated from 3-ml cultures of several independent recombinant clones and
`cleaved with HaelII to determine the orientation of the cloned fragments . One of these clones (pfi-gal-end), with the insert in the correct
`orientation for readthrough of p-endorphin sequences from the /J-galactosidase gene, was selected for further analysis. The terminal nucleotide
`sequences of the inserted fragment were determined by digesting the plasmid pp-gal-end with EcoRI endonuclease, isolating the inserted P(cid:173)
`endorphin gene fragment by gel electrophoresis and electroelution, 3'- terminal labelling of the EcoRI termini with [a3 2P]dATP (Amersham,
`2,000Cimmol- 1 ) and reverse transcriptase,cleaving the fragment with Hhal , isolation of the two digestion products by electrophoresis in a
`10 % polyacrylamide gel, and subjecting these to DNA sequence analysis by the method of Maxam and Gilbert 2 3
`. The sequences were found
`to be identical to those shown.
`
`G TCC CAC
`
`CCC
`
`of ,e-oalactosldase gene
`carried on pBR322
`
`ro-creoted
`C·termlnol Gin
`~RI
`and stop codon
`le< Jj'-ende<phln
`1004 100!1
`Gly Gin Stop
`Glu Piie Aro Val
`- - - GAA TTC AGG GTG ____________________ GGC CAA TGA ATT C ..•
`CTT AAG TCC CAC
`CCG GTT ACT TAA C •. .
`
`lys-orQ
`,B-oalactosldase .... 1-,e-MsH-1-1--,e-endorphin-------•
`
`a
`
`b
`
`c
`
`d
`
`I
`
`g
`
`Ir
`
`Hybrid -(cid:173)
`p.gal --
`
`Fig. 3 Proteins synthesized from plasmids pfi-gal or pjl-gal-end.
`Bacteria harbouring p/)-gal or p/J-gal-end were grown at 37 "C to
`log phase
`in L-broth and collected by
`(A 650 of 0.25)
`centrifugation. Cells were dissolved in SOS sample buffer and the
`equivalent of 50 µl of original culture subjected to electrophoresis
`in a 7 % polyacrylamide gel and visualized by staining with
`Coomassie blue 24
`. The positions of the hybrid protein
`jl(cid:173)
`galactosidase- fi-MSH-jl-endorphin
`and
`/J-galactosidase
`are
`indicated by arrows. a, E. coli RRl; b, E.coli RR! induced with
`2 mM isopropyl-/J-thiogalactoside (IPTG) ; c, E. coli RRl carrying
`p/J-gal; d, E. coli carrying p/J-gal induced with IPTG ; e, E. coli
`RR! carrying pp-gal-end; f, E. coli RR! carrying pjl-gal-end
`induced with IPTG. Cells obtained in e were collected by
`centrifugation and the cell pellet resuspended in phosphate(cid:173)
`buffered saline. After disruption by sonication, the cell debris and
`insoluble protein were pelleted by centrifugation at 15,000r.p.m.
`for 30 min. g, Supernatant of p{J-gal-end cells after sonication ; h,
`pellet from pfi-gal-end cells after sonication; i, proteins from p/J(cid:173)
`gal-end cells after amplilicat1on of plasmid DNA in 50µg per ml
`chloramphenicol, followed by 3 h growth in fresh L-broth.
`
`
`
`© Nature Publishing Group1980
`
`Merck Ex. 1069, pg 1537
`
`

`
`Nature Vol. 285 12 June 1980
`
`instead of a histidine and at position 31 where a glutamic acid
`residue replaces a glutamine 28
`• These changes do not inhibit
`the ability of the human hormone to react to antiserum raised
`against
`the mouse hormone. Immunological activity was
`examined in partially purified extracts prepared as described in
`the legends to Figs 3 and 4 from fully induced E. coli RRl
`cells containing the recombinant plasmid pp-gal-end.
`The endorphin purified from E. coli RRl cells carrying the
`recombinant plasmid p{J-gal-end competitively inhibited the
`binding of human 1251 P-endorphin to the mouse endorphin
`antibody (Fig. 5). As based on the immunological activity in
`the glass-purified material and the recovery from the trypsin(cid:173)
`treated preparation (see the legends to Figs 4 and 5), it is
`estimated that 0.5 µg of released endorphin of molecular
`weight 3,900 can be obtained from 109 cells (8 x 104 molecules
`of endorphin per cell). By contrast, extracts of cells harbouring
`the plasmid p,8-gal which had been purified in a similar
`manner showed no competition for endorphin antibody
`binding sites. The activity demonstrated by extracts of these
`cells (over a series of 11 dilutions to 1: 104
`) was the same as
`that of the assay buffer.
`To determine if the material reactive to antiserum to P(cid:173)
`endorphin was of the anticipated size, 60 µl of the glass bead(cid:173)
`purified preparation was filtered on a Sephadex G-50 column
`(Isolab 3 ml propylene column) and the column fractions were
`assayed for immunological activity. The column had been pre(cid:173)
`calibrated with human P-endorphin28
`. The profile of
`immunological activity was
`included by the column and
`exhibited an elution pattern identical to the authentic P(cid:173)
`endorphin. No
`immunological activity was found
`in
`the
`excluded fractions . Thus, the released peptide rather than the
`fusion protein appears to account for the activity.
`
`Biological activity of bacterially
`synthesized /J- endorphin
`The opiate activity of endorphin prepared from extracts of
`fully induced E . coli RRl cells containing the recombinant
`plasmid p,8-gal-end was examined by testing its ability to bind
`to opiate receptors of rat brain membrane preparations31 and
`to elicit an opiate-like effect of inhibiting the stimulation by
`in
`the
`prostaglandin E 1 of cyclic AMP accumulation
`neuroblastoma-glioma hybrid cell line NG108-15 32
`.
`The receptor-binding activity of the material synthesized in
`the bacteria was examined (Table 1) by testing its ability to
`inhibit the binding of a radiolabelled agonist (3H-o-Ala2 -Met 5
`-
`enkephalinamide) or antagonist (3H-naloxone). Endorphin
`prepared according to Figs 3 and 4 legends (without glass
`extraction) inhibited both the agonist and antagonist binding.
`As might be expected with this crude preparation (purified
`material was not available for
`these experiments), some
`nonspecific inhibition of binding was observed in the control
`reaction, where an extract from pp-gal was used ; however, this
`was substantially less than in the reaction with an extract from
`pp-gal-end.
`The neuroblastoma-glioma hybrid cell line NG 108-15 is
`richly endowed with opiate receptors32 and possesses an
`adenylate cyclase which can be stimulated by prostaglandin
`E 1 (ref. 34). This stimulation is inhibited by morphine and the
`inhibitory effect of morphine can be prevented by naloxone 35
`•
`Endorphin prepared as described in Figs 3 and 4 legends
`(without glass extraction) showed a significant inhibition of the
`stimulation by prostaglandin E 1 of adenylate cyclase activity
`(Table 1). This inhibition was completely reversed by addition
`of the opiate antagonist naloxone, and it was much greater
`than the activity observed in the control reaction (Table 1).
`
`Discussion
`This report demonstrates the expression of cloned P-endorphin
`gene sequences by bacteria in a form from which substantial
`amounts of biologically active P-endorphin can be obtained.
`
`459
`
`Table I Opiate receptor binding and opiate-like actions of extracts
`of p/3-gal-end
`
`Competition for receptor binding
`3H-Ala2-Met5-enkephalinamide
`(agonist)
`3 H-naloxone (antagonist)
`Cyclic AMP production
`No antagonist present
`+Naloxone
`
`% Inhibition
`pp-gal-end
`p/J-gal
`
`84 (83- 85)
`68 (67--69)
`
`65 (63-71)
`25 (17-32)
`
`23.3 (21.9- 24.6) 5 1 (5.0-5.2)
`0
`0
`
`Rat brain membrane preparations were isolated according to the
`method of Snyder31 with slight modification. Male Sprague-Dawley
`rats (200-220 g) were killed by decapitation and the brains minus
`cerebellum were homogenized in 40 volumes of 0.05 M Tris-HCl buffer
`pH 7.7 at 4 °C. The homogenates were first centrifuged at 600g for 5
`min to remove the nuclei. The supernatant was then centrifuged at
`4 °C for 15 min at 49,000 g. The pellets were suspended in 30 ml (per
`rat brain) of the same Tris buffer. The resuspended pellets were then
`incubated at 37 °C for 30min and centrifuged for 15min at 49,000g.
`The final pellet from each rat brain was then resuspended in 40 ml of
`0.05 M Tris-HCI buffer containing 100 µg ml - 1 of bacitracin and used
`for the binding assays. The binding experiments were performed, as
`previously described at 25 °C for 20 min, or at 0 ''C for 3 h. Reaction
`mixtures contained 1 ml of tissue suspension and 1.5 nM of 3H-o-Ala 2
`-
`Met5-enkephalinamide (New England Nuclear, 38Cimmol- 1) or 3H(cid:173)
`naloxone (New England Nuclear, 25Cimmol - 1
`); 4.5µg of bacterial
`3 H-o-Ala 2-Met 5
`extract was
`incubated
`in
`the
`reaction with
`-
`enkephalinamide and 0.65 µg was used in the reaction with 3 H(cid:173)
`naloxone. In each case the extracts were from a separate preparation.
`Non-radioactive levorphanol or naloxone 31 was used as standard
`competitor for each assay. The reaction was terminated by vacuum
`filtration over Whatman glas fibre filters (GH-B). The filters were
`washed with a large volume of ice-cold Tris buffer and placed in 12 ml
`of Hydromix
`scintillation
`fluid
`(Yorktown Res.).
`In parallel
`experiments 10- 6 M of naloxone were incubated to estimate the
`background binding which was subtracted from the total binding to
`yield the specific binding. Results are reported as per cent inhibition of
`specific binding that was 5,613 c.p.m. per assay with 3H-o-Ala 2 -Met 5
`-
`enkephalinamide and 2,869c.p.m. per assay with 3 H-naloxone. Mean
`values and ranges (in parentheses) of triplicate incubations are shown.
`To test for biological activity, the neuroblastoma-glioma hybrid cells
`(line NG108-15) were used 32
`. Cells were grown in T-flasks (one T-75
`in Dulbecco's modified Eagle's medium
`flask per 60 assays)
`(DMEM)33 containing 10 % fetal calf serum, 0.1 mM hypoxanthine,
`10 µM aminopterin and 16 µM thymidine. At confluency, the cells were
`collected and homogenized in 0.32 M sucrose, 40 mM HEP ES and
`2mM EDTA, pH 7.6. Incubations 34
`3 5 were carried out at 30 °C for
`·
`15 min in a total volume of 100 µI containing 3 x 106 c.p.m. of 32 P(cid:173)
`ATP, 10 units creatine phosphokinase, 50µM prostaglandin E 1
`(PGE 1 ; Upjohn), 5 µI of opiate agonist or extract (3.5 µg) of pellet
`material (prepared from fully induced E. coli RRl as indicated in
`legends to Figs 3 and 4), 20mM HEPES, 5mM MgCl 2 , 1 mM cyclic
`AMP, 20 mM
`creatine
`phosphate, 0.1 mM ATP, 0.125 mM
`phosphodiesterase inhibitor (ZK 62711, Schilling) and 1 mM protease
`inhibitor (Sigma). The reaction was stopped by the addition of 150 µI
`of lM HCl0 4 and 0.3ml of H 20 containing ~30,000c.p.m. of
`tritiated cyclic nucleotide per tube. After mixing and centrifuging, the
`supernatant solutions were poured onto Dowex 50 columns. [a- 32P](cid:173)
`cyclic AMP was then separated from other nucleotides by the two(cid:173)
`column (Dowex 50 and alumina) method described by White and
`Karr 36 . The adenylate cyclase activity, which was stimulated by PGE,
`(level of cyclic AMP released increased from 30 to 180 pmol per mg
`protein per min) in the absence of opiate agonist or test material was
`34
`used as control 33
`. The activity of the bacterially synthesized
`•
`endorphin preparation is expressed as the ability to inhibit this
`stimulation. The agonist activity of the test material was further
`examined by the reversibility of the effect by the opiate antagonist
`naloxone 34. Mean values and ranges (in parentheses) of triplicate
`incubations are shown.
`
`These data, along with those recently demonstrating the
`6
`, insulin 3 and growth hormone 1
`synthesis of somatostatin2
`•
`show the potential usefulness of bacteria to synthesize large
`
`
`
`© Nature Publishing Group1980
`
`Merck Ex. 1069, pg 1538
`
`

`
`12 June 1980
`Nature Vol. 285
`·-- ··-- · - ·
`·~---------------- - -··- --
`-
`
`460
`
`economic
`and
`quantities of polypeptides of medical
`importance. The methodology used in this report differs from
`and therefore complements that previously used for the other
`proteins. In all of these studies in which a n exact hormone
`was produced, fairly large fragments of synthetic DNA were
`used: In the current study, small fragments of DNA containing
`restr~ct1on endonuclease sites were used to create the proper
`reading frame, to add the codon for o ne amino acid and a
`
`,B-11oloctosidose
`
`,B-MSH ,B-endorphln
`
`anhydride, pH 9
`
`Hybrid
`protein
`
`Modified protein
`(cltroconlc 11roups
`on lysine residues)
`
`Modified
`.J\J.r\.Jv- ,B-•ndorphin
`
`! Cltroconic
`t::: ~ ? 0 ? ?
`~ Jvt
`,8-11a1actosidte ! Acid
`
`hydrolysis, pH 3
`
`+ _,B· MSH peptides
`)
`\JV
`AF\.f
`
`~-_,B-endorphln
`
`I Glass
`
`~ extraction
`~-Partially-purified
`,8- endorphln
`
`the fl-galactosidase··/:i(cid:173)
`Fig. 4 Release of fl-endorphin from
`MSH- fl-endorphin hybrid protein. The pellet obtained, as in Fig.
`3 was dissolved in 10 ml of 6M guanidinium chloride, 1 ~~ fl(cid:173)
`mercaptoethanol and centrifuged at 20,000 r.p.m.
`for
`l h.
`Citraconic an hydride (Fisher) was added to the supernatant in
`three lots of 10µ1 over a 15-min period, during which time the
`pH was maintained between 9 and 11 by the addition of 2M
`Na0H 25 . The solution was
`then dialysed overnight against
`50mM ammonium bicarbonate followed by the addition of
`trypsin (Worthington) to 0.5 mg ml · 1 and incubation at 37 ''C for
`12 h.
`Trypsin was
`inactivated
`by
`the
`addition
`of
`phenylmethylsulphonyl fluoride (PMSF) to 1 mM for l h and the
`solution made 1 '.%', in formic acid before lyophilization. The dried
`protein was then dissolved in Tris (base) buffer pH 7.6 to yield a
`protein concentration 26 of 0.5- LO µg µI - 1. This preparation was
`tested for
`immunological and biological activity. To further
`purify /3-endorphin, a modification of the technique devised for
`the extraction and purification of ACTH by Rees et al. 27 was
`used. To 0.5 ml aliquots of the crude endorphin preparation were
`added 0.5 ml of horse serum (Grand Island Biological) and 50 mg
`glass powder (140 mesh, Corning Glass Works, washed once with
`water, heated for 24 h at 120 "C and then stored in a desiccator
`until use) in a 15-ml plastic centrifuge tube. The samples were
`vortexed for 30 s and then centrifuged for 5 min at 3,000 r.p.m.
`The supernatant was discarded and the glass adsorbant washed
`with 3 ml of water. The endorphin was removed from the glass
`by addition of 1 ml of 50 % acetone
`in 0.25-5.00 M HCI,
`vortexing for 30 s and then centrifuging as above. The endorphin(cid:173)
`rich supernatant was then transferred to J 0 x 75 mm polystyrene
`tubes (Falcon) and evaporated to dryness with a fine stream of
`in a water bath at 45 °C. Dried samples were
`nitrogen
`reconstituted with Tris buffer (pH 7.6) and serially diluted for
`radioimmunoassay. To monitor the recovery of /3-endorphin in
`the extraction procedure, a parallel experiment was performed in
`which the recovery of 1251-/Jh-endorphin was measured. The
`recovery was 60 ~; •. The human P-endorphin had been synthesized
`by the solid phase technique by Li et al. 28 and labelled with 125 1
`by a chloramine-T procedure29 . The iodinated peptide was then
`purified by adsorption to 35 mg of glass powder (Corning) in
`lOml of buffer (pH 7.4) containing 0.05 M sodium phosphate,
`0.25 ':%: human albumin and 0.5 /~ /3-mercaptoethanoL After
`mixing for 5 min, the suspension was spun at 3,500 r.p.m. for
`5 min and the pellet was washed once with water and then eluted
`with 2 ml of 40 ·-·~; acetone in 0.25 M HCL The 1251-/lh-endorphin
`was then diluted in the phosphate-containing buffer described
`above before use. Antiserum directed against mouse /J-endorphin
`had previously been generated in rabbits by Allen et a/. 3 0
`.
`
`10
`
`10 2
`Reciprocal of sample dilution
`
`10.1
`
`Immunological activity
`Fig. 5
`inhibition with
`(competitive
`human 1251-/Jh-endorphin for antibody sites) in partially purified
`extracts from E. coli RR 1 cells carrying the recombinant plasmid
`pfl-gal-end (eJ and p/J-gal (.&). Extracts from fully induced
`bacteria containing these plasmids were prepared as described in
`Figs 3 and 4 legends. Incubations were carried out at 4 "C for
`24 h in a total volume of 150 µl containing 75 µl of i 251-/J(cid:173)
`endorphin ( 10,000 c.p.m.) (prepared as descri bed for Fig. 4), 50 ~Li
`of endorphin antibody (final titre I: 1500 yielded 50 °;, binding;
`I: 150 yielded 83 % binding), 25 µl of standards or extracts (from
`lyophilizates at I µg µ1- 1 protein) at various concentrations in
`0.05 M sodium phosphate (pH 7.4), 0.25 ':%, human albumin and
`0.5 /~ /3-mercaptoethanoL Free radioligand was then separated
`from bound peptides by adding 100 µl of a dextran-charcoal
`suspension containing 3 % charcoal (Norit A, Pfanstiehl Chemical
`Co), 0.75 /·;, dextran (Schwartz-Mann) and 60'/;, horse serum
`(Grand Island