Recombinant

prolactin: expression in Escherichia coli, purification and biological activity mouse

M. Yamamoto, T. K. Nakashima

Harigaya,

T.

Ichikawa, K. Hoshino and

Central Research Laboratories, Shikibo Ltd, 103 Fushio-cho, Ikeda-shi, Osaka 563, Japan *Department of Anatomy, Faculty of Medicine, Kyoto University, Sakyo-ku, Kyoto 606, Japan fDepartment of Biochemistry, Mie University School of Medicine, Mie 514, Japan (T. Harigaya is now at School of Agriculture, Meiji University, 1-1-1 Higashimita, Tama-ku, Kawasaki 214, Japan) received 23 May 1991 ABSTRACT

Transformation of Escherichia coli cells with a recombinant plasmid containing modified mouse prolactin (mPRL) cDNA and a pKK223-3 vector resulted in efficient expression of mPRL protein. Cloned mPRL cDNA was modified by removing the 5' non-translating sequence as well as the sequence which encoded the signal peptide of preprolactin for recombination. In addition, approximately 100 nucleotides of the 5\m='\ x=req-\ terminal region of the cDNA, which include the ATG initiation codon and the following 31 codons of mature

chosen to be rich in AT without changing the amino acid sequence of the protein. The modified cDNA was finally inserted into the multicopy plasmid, pUC19, before high-level expression of mPRL in E. coli cells was obtained. Western blotting analysis of total protein from transformed E. coli cells showed that both 23 and 16kDa peptides were recognized by specific mPRL antisera. The purified and refolded 23 kDa protein exhibited a growth-stimulating effect on rat Nb 2 Node lymphoma cells, and was very similar to that of natural pituitary PRL. was

mPRL, were replaced by a chemically synthesized oligonucleotide duplex. The sequence of this duplex

Journal of Molecular Endocrinology (1992) 8, 165-172

INTRODUCTION

rats

Prolactin (PRL) is an hormone which exhibits

adenohypophysial peptide wide variety of activity in vertebrates, including growth and differentiation of mammary epithelium, lactation, osmoregulation and parental behaviour in teleosts, amphibian develop¬ ment, broodiness in hens and production of crop sac 'milk' in pigeons (for reviews see Banerjee & Menon, 1987; Horseman, 1987; Nicoll, Anderson, Hebert & Russell, 1987; Oka & Taga, 1987). Since this variety a

of actions of PRL is sometimes controversial when heterologous PRL is used in physiological exper¬ iments, it may be necessary to use homologous PRL in examining its physiological properties in each species. Prolactins are usually purified from the pituitary gland or its organ culture media (Kohmoto, 1975). It has therefore been difficult to prepare natural PRL from the pituitary gland of small animals such as mice and rats. Fortunately, however, cDNA for PRL was recently cloned and sequenced for several species, including mice and

(Cooke, Coit, Weiner et al. 1980; Harigaya, Nakayama, Ohkubo et al. 1986). Furthermore, the recent development of recombinant DNA tech¬ niques has enabled us to obtain mammalian proteins in Escherichia coli cells transformed with their

expression vectors. In this study, we have attempted to express a high level of mouse PRL (mPRL) in E. coli cells using a recombinant plasmid cDNA. Recombinant mPRL was efficiently expressed in E. coli and was purified and refolded to cross-react with specific mPRL antisera in Western blotting analysis. Preparations of this recombinant mPRL

were

also shown

to

exhibit

growth-promoting activity on Nb 2 Node lymphoma cells

(Tanaka, Shiu,

Gout

et

al.

1980).

MATERIALS AND METHODS

Plasmids and bacterial strains

The cDNA clone for mPRL, pmPRL6, has been cloned and sequenced by Harigaya et al. (1986).

Net Leu Pro Ile

Cys Ser Ala Gly Asp Cys Gin Thr Ser Leu (MPI)

Arg Glu Leu Phe Asp Arg Val Val He Leu Ser His Tyr Ile His Thr Leu Tyr

(ÜP3)

--

c c ccg cccg CG CT CCC CCTGC CCG) atg tta cca att tgt tct gct ggt gat tgt caa act tct aga gaa TTA TTT GAT AGA GTT GTT ATA TTA TCT CAT TAT ATT CAT ACT TTA T 3' AAT AAA CTA TCT CAA CAA TAT AAT AGA GTA ATA TAA GTA TGA AAT 3' tac aat ggt taa aca aga cga cca cta aca gtt tga aga a at

(cg

t[ta

5'|aatt

[EcoRI]

tct|ctt

(HP2)

W4'

ATA~|5'

[AccI]

figure 1. Synthesized DNA sequence encoding the N-terminus of mouse prolactin (mPRL). Four single strand oligonucleotides, MPI, MP2, MP3 and MP4, were synthesized independently, and then annealed and ligated. The lines denote the ends of the synthesized oligonucleotides and the respective overlapping regions between fragments. The constructed oligonucleotide duplex corresponded to approximately 100 nucleotides of the 5'-terminal region of the mPRL cDNA, which included the ATG initiation codon and the first 31 codons for mature mPRL protein. The sequence of this duplex was chosen to be rich in AT without changing the amino acid sequence of the protein. Nucleotides in parentheses indicate those which appear in native mPRL cDNA. The 5' and 3' termini of the constructed fragment are ligated to the EcoRI site in vector plasmid and the AccI site in mPRL cDNA respectively.

pKK223-3 plasmid containing tac promoter (De Boer, Comstock & Vasser, 1983) and rrnB ribosomal transcription terminators (Brosius, Dull, Sleeter & Noller, 1981) was purchased from Pharmacia LKB Biotechnology (Tokyo, Japan) and used as an ex¬ pression vector. This vector allows the synthesis of foreign proteins without generating any fusion with other peptide sequences. Since the vector contains the strong tac promoter in conjunction with the lac operator, the synthesis of the foreign protein is repressed in strains producing the lac repressor but can be induced in the presence of isopropylthio-ßgalactoside (IPTG). The high copy number plasmid pUC19 (Vieira & Messing, 1982) was obtained from Nippon Gene Co. (Toyama, Japan). The host strain used for the transformation of the recombinant plasmids was E. coli K-12 JM109

(recAl, endAl, GyrA96, thi, hsdRll, supE44, relAl, (lac-pro), F'(traD36, probAB, lad, lacZ MIS)) (Yanisch-Perron, Vieira & Messing, 1985). E. coli

cells were grown at 37°C in LB medium or in M9 medium containing 5.0 mg casamino acids/ml (M9CA) in the presence of 100 pg ampicillin/ml (Sambrook, Fritsch & Maniatis, 1989).

a signal sequence for pNMPl, or with a DNA sequence modified to AT-rich codons for pKKMPl (Fig. 1). The strategy of construction of expression vector pKKMPl is depicted in Fig. 2. After primary construction, pKKMPl was recon¬ structed with a high copy number vector, pUC19, to express a larger amount of prolactin. The recon¬ structed plasmid, whose cDNA insert possessed normal direction or reverse direction, was then

lacking

designated as pKCMPl or pKCMP-R respectively (Fig. 2). These constructed plasmids were used to transform E. coli JM109 cells (Hanahan, 1983), and the colonies resistant to ampicillin were screened by the colony immunoassay method of Helfman, Feramisco, Fiddes et al. (1983) with a rabbit antimPRL antibody. Plasmids were isolated from sev¬ eral positive clones and subjected to restriction enzyme mapping analysis to confirm the correct construction.

Expression of mPRL cDNA in E. coli For expression of mPRL cDNA, the transformed E. 2. Construction of mouse prolactin (mPRL) expression vectors. The plasmid containing mPRL cDNA, pmPRL6, was first digested with SphI and the resulting end was converted to a blunt end with T4 DNA polymerase. This plasmid was further digested with AccI. The pKK223-3 vector was digested with Smal and treated with bacterial alkaline phosphatase (BAP), and further digested with EcoRI. DNA frag¬ ments corresponding to the region upstream of the AccI site in mPRL cDNA were synthesized and prepared as shown in Fig. 1. The plasmid, pKKMPl, was obtained by ligation of the vector and these fragments. This plasmid was then digested with Sspl and treated with BAP. The Sspl fragment containing promoter and cDNA for mPRL was ligated with pUC19 that had been digested with PvuII and Sspl. The obtained plasmid whose cDNA insert possessed normal direction or re¬ verse direction was designated as pKCMPl or pKCMPR respectively. figure

Oligonucleotide synthesis Oligonucleotides were synthesized by the phosphite triester method (Beaucage & Caruthers, 1981) using phosphoramidite chemistry (Sinha, Biernat & Köster, 1983) and an Applied Biosystems 381A DNA synthesizer (Tokyo, Japan). The oligonucleotides were purified using Applied Biosystems OPC car¬ tridges. Construction of expression vectors The primary plasmids, pNMPl and pKKMPl, were constructed using the pKK223-3 vector and cloned mPRL cDNA fragment as follows. The 5' end of mPRL cDNA, from ATG to the AccI site, was replaced with synthetic DNA identical to the orig¬ inal nucleotide sequence for mature mPRL protein

Ace I

EcoRI

Smal

(MPI). (MP2)

[EcoRI] (MP3) (MP4)

SphI

SphI

Smal cut BAP treatment EcoRI cut

cut

Conversion Ace I cut

to

blunt end

P-

+ [AccI] Ligation

(excess)

Ligation

V

Sspl

Sspl

Sspl

BAP

Sspl

cut treatment

coli JM109 cells were grown at 37°C in LB medium containing 100 pg ampicillin/ml for approximately 3 h (optical density at 660nm 0.5) followed by a further 5 h of growth in the presence of 2 mivi IPTG (Wako Pure Chemical Industries Ltd, Osaka, Japan). Then 0.3 ml of each culture was centrifuged, and precipitates were analysed by sodium dodecyl =

sulphate-polyacrylamide gel electrophoresis (SDSPAGE) followed by Coomassie brilliant blue stain¬ ing or Western blotting analysis with the rabbit mPRL antiserum.

SDS-PAGE analysis E. coli proteins were dissolved in 40 pi sample buffer (62.5 mM Tris-HCl (pH 6.8)/5mM EDTA/10% (v/v) glycerol/2% (w/v) SDS/10% (v/v) 2-mer-

captoethanol (2-ME)/0.01% (w/v) bromophenol blue) and heated at 95°C for 5 min. The samples were then analysed by 15% SDS-PAGE (Laemmli, 1970). The gels were stained with Coomassie bril¬

liant blue. To determine the relative amount of synthetic mPRL in the total E. coli proteins, the

protein bands were scanned by a CS-9000 dualwavelength flying-spot scanner (Shimadzu Co., Kyoto, Japan).

Immunological analysis Immunoblot (Western blotting) analysis was per¬ formed according to the method of Towbin, Staehlin & Gordon (1979). Protein samples were electrophor-

mentioned above and then transferred electrophoretically to a Clearblot P-membrane (Atto Co., Tokyo, Japan). The membrane was treated with the rabbit mPRL antiserum ( 1:1000 dilution in 50mM Tris-HCl (pH 7.5)/0.9% (w/v) NaCl/ 0.05% (v/v) Tween 20) and then with protein Aconjugated horseradish peroxidase (HRP-protein A; 2pg/ml in the same buffer as described above) (E-Y Labs Inc., San Mateo, CA, U.S.A.). The band specific for mPRL was detected by staining with 0.05% (w/v) 4-chloro-l-naphthol and 0.05% (v/v) hydrogen peroxide in 50 mM Tris-HCl (pH 7.5)/ 0.9% (w/v) NaCl. esed

solved in a minimum amount of 0.1m Tris-HCl buffer (pH 8.0) containing 6m guanidine hydrochloride and 0.1m 2-mercaptoethanol (2-ME). After the solution had been incubated for 1 h at room temperature, the 2-ME was removed by passing through a Sephadex G-25 column equilibrated with 0.1m Tris-HCl buffer (pH 8.0) containing 6m

guanidine hydrochloride. Refolding of the denatured mPRL protein was performed according to the method of Tsuji, Nakagawa, Sugimoto & Fukuhara (1987). The solution containing the refolded mPRL was centrifuged, and the supernatant was desalted using a Sephadex G-25 column equilibrated with 50 mM ammonium bicar¬ bonate buffer (pH 8.0) containing 20 mM NaCl. The

recombinant mPRL in the desalted solution was further purified to homogeneity on a COSMOGEL DEAE column (Nacalai Tesque, Kyoto, Japan; 2x10cm) with a linear gradient from 0.05 to 0.6m ammonium bicarbonate buffer (pH 8.0).

Bioassay of mPRL activity Bioassay of mPRL was performed using rat Nb 2 Node lymphoma cells as described by Tanaka et al. (1980). Natural rat PRL (NIADDK B-4, 20IU/ mg) and mPRL purified from pituitary organ cul¬ ture (kindly provided by K. Kohmoto, University of Tokyo, Japan) were used as standards. RESULTS

as

Purification of mPRL from E. coli cells E. coli JM109 cells transformed with pKCMPl were induced by IPTG in a 2 litre culture of M9CA, and were centrifuged at 15 000 g1 at 4°C for 15 min and resuspended in 200 ml of a buffer consisting of 50mM Tris-HCl (pH 7.5), 5 mM EDTA and 1 mg Cell suspensions were sonicated in an lysozyme/ml. RUS-600T sonicator (Nihonseiki Kaisha Ltd, Tokyo, Japan) and expressed mPRL protein was harvested as inclusion bodies by centrifugation at 15 000 g and 4°C for 20 min. The pellet was dis-

The initial plasmid constructed in our study to obtain high-level expression of mPRL in E. coli was pNMPl. This plasmid contains the 197 codons that specify the entire amino acid sequence of mature PRL (lacking the signal sequence), and the initiation codon ATG. Nine nucleotides separate the ATG from the consensus Shine-Dalgarno (SD) sequence AGGA (Shine & Dalgarno, 1974) as follows: 5'AGGAAACAGAATTATG-3'. The E. coli cells transformed with pNMPl produced no detectable Met-mPRL (Fig. 3, lane 3). We assumed that the results with the expression vector pNMPl were probably due to the prevention of effective translation by the formation of secondary structures of mRNA around the ATG initiation codon and neighbouring nucleotides. We therefore tried to replace the sequence downstream of the initiation codon in pNMPl. The design of the replacement sequence was chosen to give the ATrich nucleotide sequence shown in Fig. 1 without changing the amino acid sequence of mPRL. Thus pKKMPl, in which the first 31 codons downstream of the ATG codon were modified, was constructed. When E. coli cells were transformed with pKKMPl,

very small amounts of MetmPRL. No Coomassie blue-stained band of protein with the electrophoretic mobility of standard mPRL was detected in polyacrylamide gels of total protein extracts of transformed E. coli cells (Fig. 3). This was true regardless of the time of harvest of the bacteria from the early log phase through to the stationary phase, the culture medium, or the pres¬ ence or absence of IPTG (data not shown). How¬ ever, very small amounts of recombinant mPRL were identified in the E. coli extracts in Western blotting analysis (Fig. 3). Calculations based on the intensity of the reacting bands, and the cell density of the cultures from which the proteins were extracted, showed that the number of PRL mol¬ ecules synthesized per E. coli cell was about 6000 to 7000. The protein of smaller size (approximately 16 kDa) than PRL, which also reacted with antibody (Fig. 3), appeared to be the product of a specific proteolytic cleavage. Since this protein was similar in size to that of a cleaved PRL reported by Mittra (1984), it would seem that proteases cleaved the large disulphide loop of PRL in E. coli. To clarify whether the low copy number of expression plasmid pKKMPl per E. coli cell was responsible for the low level of expression, the multi-copy expression plasmids pKCMPl and pKCMP-R were constructed (Fig. 2). E. coli cells harbouring these plasmids eventually synthesized large amounts of Met-mPRL. Figure 3 shows the results of SDS-PAGE analysis of proteins extracted from E. coli harbouring plasmids pKKMPl, pKCMPl or pKCMP-R. When the cells were grown in LB medium, those containing pKCMPl (Fig. 3, lanes 6 and 7) or pKCMP-R (Fig. 3, lanes 8 and 9) produced significant amounts of a protein with the same electrophoretic mobility as natural mPRL (Fig. 3, lane 2), corresponding to a molecular weight of 23 kDa. Moreover, the accumulation of this protein increased proportionally in response to increasing levels of IPTG in M9CA medium (data not shown), indicating that its synthesis was under the control of the lac operator. To confirm that mPRL was synthesized in the cells transformed by pKKMPl, pKCMPl and pKCMP-R, Western blotting analysis was per¬ formed with use of a specific mPRL antiserum. Figure 3 shows that the intensity of the Coomassie blue-stained protein bands of 23 kDa corresponded to that of the reaction products with mPRL antiserum. Calculations based on the intensity of the bands in SDS-PAGE and the density of E. coli cells containing pKCMPl in the cultures suggested that nearly 600 000 PRL molecules were present per cell. This indicated that an 80- to 100-fold increase in expression of mPRL was obtained in cells harbour-

they produced only

figure 3. Expression of mouse prolactin (mPRL) in Escherichia coli. (a) SDS-PAGE analysis. Samples of lysates containing protein from 0.3 ml media from cells grown to stationary phase in LB medium plus 2 mM isopropylthio-ßgalactoside were electrophoresed in 15% polyacrylamide gel. Proteins were stained with Coomassie brilliant blue, (b) Western blotting analysis of mPRL synthesis. Samples separ¬ ated by SDS-PAGE as described above were subjected to Western blotting analysis. The arrowheads show the position of recombinant mPRL. Lane 1, molecular markers: from top to bottom, phosphorylase b (97.4kDa), bovine serum albu¬ min (66.2kDa), ovalbumin (42.7kDa), carbonic anhydrase (31.0kDa), trypsin inhibitor (21.5 kDa) and lvsozvme (14.4kDa); lane 2, natural mPRL (6pg); lane 3, 4-5, 6-7 and 8-9, lysates of E. coli harbouring pNMPl, pKKMPl, pKCMPl and pKCMP-R expression vectors respectively.

ing pKCMPl compared with those with pKKMPl.

The amount of PRL that accumulated in E. coli cells with pKCMPl or pKCMP-R grown to stationary phase in the presence of 2mM IPTG accounted for approximately 15% of the total E. coli proteins. In addition, mPRL produced by E. coli containing pKCMPl formed inclusion bodies, and its amount was about 400mg/l. The protein was easily re¬ covered from sonicated cells by centrifugation and dissolved in 6 m guanidine hydrochloride solution. The mPRL protein obtained was then refolded and further purified to homogeneity by anion exchange chromatography. The recombinant mPRL was ana¬ lysed by SDS-PAGE and Western blotting, which revealed a single band at about 23 kDa, which showed a mobility identical to that of the band of natural mPRL, both in reducing and non-reducing conditions (Fig. 4). About 40mg of this purified preparation was obtained from 2 litres of cultured cells. The highly purified mPRL sample was used in the Nb 2 cell bioassay. The results clearly demonstrated

that the recombinant mPRL possesses growthpromoting activity for lymphoma cells, the activity being similar to that of natural PRL purified from rat pituitary glands (Fig. 5) and mouse pituitary organ culture (data not shown). There was essen¬ tially no growth in the control cultures over the 3day incubation period, however. DISCUSSION

The results of the present study show that E. coli cells harbouring the expression plasmid pKCMPl or pKCMP-R synthesized significant amounts of recombinant mPRL. The synthesized protein ap¬ peared to be stable in the bacterial cells. In several efficient expression plasmids for growth hormone that have been constructed, frequently used codons in E. coli (Gouy & Gautier, 1982; Grosjean & Fiers, 1982) have been substituted in part or in whole for the natural codons of the mammalian hormone gene (Goeddel, Heyneker,

figure 4. SDS-PAGE analysis of refolded recombinant mouse prolactin (mPRL). (a) Coomas¬ sie brilliant blue staining of recombinant and natural mPRLs. The samples were loaded onto a 15% polyacrylamide gel and electrophoresed. Lane 1, molecular markers: from top to bottom, phosphorylase b (97.4kDa), bovine serum albumin (66.2kDa), ovalbumin (42.7 kDa), carbonic anhydrase (31.0kDa), trypsin inhibitor (21.5 kDa) and lysozyme (14.4kDa); lane 2, natural mPRL (15 pg) heated in sample buffer containing 10% 2-mercaptoethanol (reduced); lane 3, recombinant mPRL (15 pg) heated in sample buffer containing 10% 2-mercaptoethanol (reduced); lanes 4 and 5, natural and recombinant mPRLs (25 pg) respectively in sample buffer without 2-mercaptoethanol (non-reduced), (b) Western blotting analysis of recombinant and natural mPRLs. The same samples as described above were subjected to SDS-PAGE and Western blotting, followed by reaction with rabbit mPRL antiserum, and protein A-conjugated horseradish peroxidase. Lanes 2-5 are identical to those shown in (a) except that 3 and 5 pg mPRLs were run in lanes 2-3 and 4-5 respectively.

o

ío6

p

--^8

significantly. Similar conclusions have been also reported by others (Gren, 1984; Munson, Stormo, Niece &

Reznikoff, 1984; Wood, Boss, Patel & et al. 1985). The consensus

Emtage, 1984; Buell

—i-ii-1-1-1_i_i_ " 0.005 0.05 0.5 O 50 5

(ng/ml) figure 5. Activation of growth of stationary Nb 2 node lymphoma cells by recombinant mouse prolactin (mPRL). PRL concentration

incubated for 72 h with recombinant mPRL PRL (NIADDK B-4; 20IU/mg; •) and without PRL (A). Each value represents the mean of triplicate Cells

were

(O),

rat

experiments. Hozumi

et al. 1979; Seeburg, Sias, Adelman et al. 1983; Ikehara, Ohtsuka, Tokunaga et al. 1984). More recently, however, it has been reported that

sequences constructed using codons which fre¬ quently appear in E. coli have been poorly expressed in E. coli. For example, Buell, Schulz, Selzer et al. (1985) found a very low level of synthesis of human somatomedin-C encoded by a synthetic gene consist¬ ing of E. coli codons inserted into a plasmid bearing the strong leftward promoter of bacteriophage X. In our expression plasmid pKKMPl, we changed 20 of the first 30 codons of the mPRL cDNA. In all instances, however, the criterion we used was the nucleotide-by-nucleotide consensus sequence and not codons appearing in E. coli. For example, the most frequent codon for proline in E. coli is CCG (Gouy & Gautier, 1982; Grosjean & Fiers, 1982), whereas we used CCA for proline after the ATG initiation codon. Interestingly, Schoner, Hsiung, Belagaje et al. (1984) observed that CCA is one of the codons which enhances the expression of bovine growth hormone. It was incorporated into the first cistron of their two-cistron expression system. Although it is not always clear why minor changes in the 5'-terminal regions of foreign or natural genes in E. coli exhibit profound effects on translation rates, secondary structures of mRNAs in these regions may possibly determine the efficiency of translation. In a series of experiments on plasmid construction, in which the E. coli chloramphenicol

acetyltransferase (cat) gene was transcribed from the lac promoter, Schottel, Sninsky & Cohen (1984) found that if the potential for base-pairing between the SD sequence and neighbouring nucleotides was increased, the initiation of translation was reduced

sequence consists mainly of As in the region of the SD sequence and the initiation codon ATG, and the energy of hydrogen bonds between A and T is less than that between G and C. Therefore, this consen¬ sus sequence may affect the destabilization of poss¬ ible secondary structures. Thus, to obtain high-level expression of Met-mPRL without the additional amino acids, we used an AT-rich sequence to construct the expression plasmid. No obvious local secondary structures could be predicted in the 5' region of the mRNA of mPRL encoded by pKKMPl in the present study. Mouse prolactin produced by E. coli containing pKCMPl was estimated to be about 15% of the total cellular protein, and the protein formed inclusion bodies within the cells. There are many cases of soluble proteins expressed in E. coli in an insoluble form (Harris, 1983). It is a great advantage to obtain proteins produced in E. coli as insoluble granules, because it is possible to sediment them with a single centrifugation step after disruption of the cells, eliminating impurities such as cell membrane pro¬ teins. Soluble mPRL was obtained after solubilization and renaturation of the granule proteins. It showed biological activity on the growth of rat Nb 2 Node lymphoma cells equivalent to that of natural PRL. The present expression vectors carrying mPRL cDNA were stably maintained by the host E. coli cells, and a constant production of recombinant mPRL was obtained. This system should prove to be useful in further work examining the physiological properties of mPRL and could be of general use in the expression of a wide spectrum of recombinant eukaryotic genes in E. coli. ACKNOWLEDGEMENTS

We thank Dr K. Kohmoto for providing the natural mPRL and the rabbit mPRL antiserum, and Dr N. Sensui for providing the Nb 2 Node lymphoma cell line. REFERENCES

Banerjee, M. R. & Menon, R. S. (1987). Synergistic actions of glucocorticoid and prolactin on milk protein gene expression. In Actions of Prolactin on Molecular Processes, pp. 121-136. Ed.

J.

A. Rillema. Boca Raton: CRC Press Inc.

Beaucage, S. L. & Caruthers, M. H. (1981). Deoxynucleoside phosphoramidites A new class of key intermediates for deoxypolynucleotide synthesis. Tetrahedron Letters 22, -

1859-1862.

Brosius, J., Dull, T. J., Sleeter, D. D. & Noller,

H. F.

(1981).

Gene organization and primary structure of a ribosomal RNA operon from Escherichia coli. Journal of Molecular Biology 148, 107-127. Buell, G., Schulz, M.-F., Selzer, G., Chollet, A., Movva, N. R., Semon, D., Escanez, S. & Kawashima, E. (1985). Optimizing the expression in E. coli of a synthetic gene encoding somatomedin-C (IGF-I). Nucleic Acids Research 13, 1923-1938. Cooke, N. E., Coit, D., Weiner, R. I., Baxter, J. D. & Martial, J. A. (1980). Structure of cloned DNA complementary to rat prolactin messenger RNA. Journal of Biological Chemistry 255, 6502-6510. De Boer, H. A., Comstock, L. J. & Vasser, M. (1983). The tac promoter: A functional hybrid derived from the trp and lac promoters. Proceedings of the National Academy of Sciences of the U.S.A. 80, 21-25. Goeddel, D. V., Heyneker, H. L., Hozumi, T., Arentzen, R., Itakura, K., Yansura, D. G., Ross, M. J., Miozzari, G., Crea, R. & Seeburg, P. H. (1979). Direct expression in Escherichia coli of a DNA sequence coding for human growth hormone. Nature 281, 544-548. Gouy, M. & Gautier, C. (1982). Codon usage in bacteria: correlation with gene expressivity. Nucleic Acids Research 10, 7055-7074. Gren, E. J. (1984). Recognition of messenger RNA during translational initiation in Escherichia coli. Biochimie 66, 1-29. Grosjean, H. & Fiers, W. (1982). Preferential codon usage in prokaryotic genes: the optimal codon-anticodon interaction energy and the selective codon usage in efficiently expressed genes. Gene 18, 199-209. Hanahan, D. (1983). Studies on transformation oí Escherichia coltwith plasmids. Journal of Molecular Biology 166, 557-580.

Harigaya, T., Nakayama, K., Ohkubo, H., Nakanishi, S., Seo, H. & Hoshino, K. (1986). Cloning and sequence analysis of cDNA for mouse prolactin. Biochimica et Biophvsica Ada

868, 30-38. Harris, T. J. R. (1983). Expression of eukaryotic genes in E. coli. In Genetic Engineering, vol. 4, pp.127-183. Ed. R.

Williamson. London: Academic Press Inc. D. M., Feramisco, J. R., Fiddes, J. C, Thomas, G. P. & Hughes, S. H. (1983). Identification of clones that encode chicken tropomyosin by direct immunological screening of a cDNA expression library. Proceedings of the National Academy of Sciences of the U.S.A. 80, 31-35. Horseman, N. D. (1987). Models of prolactin action in non mammalian vertebrates. In Actions of Prolactin on Molecular Processes, pp.41-68. Ed. J. A. Rillema. Boca Raton: CRC Press Inc.

Helfman,

Ikehara, M., Ohtsuka, T., Tokunaga, T., Taniyama, Y., Iwai, S., Kitano, K., Miyamoto, S., Ohgi, T., Sakuragawa, Y., Fujiyama, K., Ikari, T., Kobayashi, M., Miyaké, T., Shibahara, S., Ono, A., Ueda, T., Tanaka, T., Baba, H., Miki, T., Sakurai, A., Oishi, T., Chisaka, O. & Matsubara,

K. (1984). Synthesis of a gene for human growth hormone and its expression in Escherichia coli. Proceedings of the National Academy of Sciences of the U.S.A. 81, 5956-5960. Kohmoto, K. (1975). Mouse prolactin obtained by pituitary organ culture in a serum-free medium. Endocrinología Japónica 22, 465-469. Laemmli, U. K. (1970). Cleavage of structural proteins during

the assembly of the head of bacteriophage T4. Nature 227, 680-685. Mittra, I. (1984). Somatomedins and proteolytic bioactivation of prolactin and growth hormone. Cell 38, 347-348. Munson, L. M., Stormo, G. D., Niece, R. L. & Reznikoff, W. S. (1984). lac Z translation initiation mutations. Journal of Molecular Biology IT7', 663-683. Nicoll, C. S., Anderson, T. R., Hebert, N. J. & Russell, S. M. (1987). Prolactin, growth factors and cell growth. In Actions of Prolactin on Molecular Processes, pp. 199-212. Ed. J. A. Rillema. Boca Raton: CRC Press Inc. Oka, T. & Taga, M. (1987). Role of polyamines in prolactin actions. In Actions of Prolactin on Molecular Processes, pp.137-152. Ed. J. A. Rillema. Boca Raton: CRC Press Inc. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, edn 2. Cold Spring Harbor: Cold Spring Harbor Laboratory Press. Schoner, B. E., Hsiung, H. M., Belagaje, R. M., Mayne, N. G. & Schoner, R. G. (1984). Role of mRNA translational efficiency in bovine growth hormone expression in Escherichia coli. Proceedings of the National Academy of Sciences of the U.S.A. 81, 5403-5407. Schottel, J. L., Sninsky, J. J. & Cohen, S. N. (1984). Effects of alterations in the translation control region on bacterial gene expression: use of cat gene constructs transcribed from the lac promoter as a model system. Gene 28, 177-193. Seeburg, P. H., Sias, S., Adelman, J., De Boer, H. A., Hayflick, J., Jhurani, P., Goeddel, D. V. & Heyneker, H. L. (1983). Efficient bacterial expression of bovine and porcine growth hormones. DNA 2, 37-45. Shine, J. & Dalgarno, L. (1974). The 3'-terminal sequence of E. coli 16s ribosomal RNA complementary to nonsense triplets and ribosome binding sites. Proceedings of the National Academy of Sciences of the U.S.A. 71, 1342-1346. Sinha, N. D., Biernat, J. & Köster, H. (1983). ß-Cyanoethyl

N,N dialkylamino/ N-morpholinomonochloro phosphoamidites, new phosphitylating agents facilitating ease of deprotection and work-up of synthesized oligonucleotides.

Tetrahedron Letters 24, 5843-5846. Tanaka, T., Shiu, R. P. C, Gout, P. W., Beer, C. T., Noble, R. L. & Friesen, H. G. (1980). A new sensitive and specific bioassay for lactogenic hormones: Measurement of prolactin and growth hormone in human serum. Journal of Clinical Endocrinology and Metabolism 51, 1058-1063. Towbin, H., Staehlin, T. & Gordon, J. (1979). Electrophoretic

transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proceedings of the National Academy of Sciences of the U.S.A. 76, 4350-4354. Tsuji, T., Nakagawa, R., Sugimoto, N. & Fukuhara, K. (1987). Characterization of disulfide bonds in recombinant proteins: reduced human interleukin 2 in inclusion bodies and its oxidative refolding. Biochemistry 26, 3129-3134. Vieira, J. & Messing, J. (1982). The pUC plasmids, an M13 mp7 derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19, 259-268. Wood, C. R., Boss, M. A., Patel, T. P. & Emtage, J. S. (1984). The influence of messenger RNA secondary structure on expression of an immunoglobulin heavy chain in Escherichia coli. Nucleic Acids Research 12, 3937-3950. Yanisch-Perron, C, Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 33, 103-119.