`Vol. 78, No. 12, pp. 7398-7402, December 1981
`Biochemistry
`
`Expression of two human growth hormone genes in monkey cells
`infected by simian virus 40 recombinants
`(recombinant DNA/gene transfer/growth factors/protein transport)
`GEORGE N. PAVLAKIS*, NAOMI HIZUKAt, PHILLIP GORDENt, PETER SEEBURGt, AND DEAN H. HAMER*
`*Laboratory of Biochemistry, National Cancer Institute, and tDiabetes Branch, National Institute of Arthritis, Metabolism and Digestive Diseases, National Institutes
`of Health, Bethesda, Maryland 20205; and *Division of Molecular Biology, Genentech, Inc., South San Francisco, California 94080
`Communicated by Maxine F. Singer, August 26, 1981
`
`ABSTRACT
`We have constructed simian virus 40 recombi-
`nants carrying two different human growth hormone (hGH) genes.
`Monkey kidney cells infected with these recombinants synthesize,
`process, and secrete hGH. The product of gene 1, which has cod-
`ing sequences identical to those of a cloned hGH complementary
`DNA, is indistinguishable from pituitary hGH by several criteria.
`The product of gene 2, which is predicted to encode a variant pro-
`tein, is less immunoreactive than pituitary hGH but binds effi-
`ciently to hGH cell surface receptors. These results show that gene
`2 has the potential to be expressed into a previously unidentified
`form of hGH. They also demonstrate that it is possible to produce
`a mature hormone by gene transfer in eukaryotic cells and indicate
`the utility ofthe simian virus 40-monkey cell system for producing
`and characterizing secreted animal cell proteins.
`
`Human growth hormone (hGH) is -synthesized by the acidophil
`cells of the anterior pituitary as a prehormone containing a hy-
`drophobic amino-terminal signal sequence that is removed dur-
`ing secretion (1, 2). The mature hormone exhibits multiple bi-
`ological effects in vivo, including diabetogenic, insulin-like,
`lactogenic, and growth-promoting activities (reviewed in ref.
`3). Most hGH preparations can be separated into multiple bands
`by high-resolution isoelectric focusing or gel electrophoresis
`(refs. 4-6 and references therein). Some of these appear to re-
`sult from posttranslational modification (7, 8) and others rep-
`resent primary sequence variants. The best studied of these is
`the Mr 20,000 ("20K") variant, which is a single polypeptide
`chain identical to the major form of hGH except that it lacks
`amino acid residues 32-46 (9, 10). It is also known that hGH
`exists in heterogeneous forms in human plasma with respect to
`molecular size and biological properties (11-13). Whether these
`variants are important in clinical conditions in which hGH re-
`sponsiveness is inappropriate for its plasma concentration is
`unknown, and there is no information as to whether the differ-
`ent forms could represent different gene products.
`Recent DNA cloning experiments show that the human ge-
`nome contains at least seven different hGH-related genes (14,
`15). One ofthese, designated here as the hGH1 gene, has been
`shown by sequence analysis to be capable of encoding the pre-
`dominant form of pituitary hGH (F. DeNoto, D. Moore, and
`H. Goodman, personal communication). A second gene, re-
`ferred to as the hGH2 gene, is highly homologous to the hGH1
`gene but contains 14 point mutations that are expected to lead
`to amino acid substitutions in the mature hormone (unpublished
`data). This suggests that some of the heterogeneity of hGH
`might be due to the expression of multiple closely related but
`nonidentical genes.
`As a first step in testing this hypothesis we have inserted the
`
`The publication costs ofthis article were defrayed in part by page charge
`payment. This article must therefore be hereby marked "advertise-
`ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
`
`hGH1 and hGH2 genes into a simian virus 40 (SV40) cloning
`vector and have used the resulting recombinants to infect pri-
`mary cultures of monkey kidney cells. Previous experiments
`have demonstrated the utility of the SV40-monkey cell system
`for gene transfer and for the analysis of various steps in mRNA
`formation (reviewed in ref. 16). Here we have used this system
`to compare the synthesis, processing, and secretion of hGH1
`and hGH2 and to produce sufficient amounts of these proteins
`for initial characterization.
`MATERIALS AND METHODS
`Procedures for the construction, propagation, and character-
`ization of recombinant plasmids and viruses have been de-
`scribed (16, 17). The SV40 vector used to clone the hGH genes
`was derived from the viable deletion mutant d12005 (18), which
`has undergone a 250-base pair deletion in the viral early gene
`region. This allows us to insert approximately 2.7 kilobase pairs
`(kb) of foreign DNA into the late gene region as compared to
`2.45 kb for the equivalent wild-type vector (unpublished re-
`sults). Experiments with recombinant plasmids and viruses
`were carried out according to the National Institutes of Health
`guidelines.
`To label proteins, a culture of 2 X 107 infected or uninfected
`cells was incubated either for 3 hr in 5 ml ofmedium containing
`[3H]leucine [197 Ci/mmol, Amersham (1 Ci = 3.7 X 1010
`becquerels)] at 200 ttCi/ml as the only leucine or for 24 hr in
`20 ml ofmedium containing 10% the normal level ofleucine and
`[3H]leucine at 20 ,Ci/ml. After the medium had been removed
`the cells were lysed in 1 ml of 0.5 M Tris-HCl (pH 7.5)/0.15
`M NaCV1% Triton X-100/0.5% aprotinin (Sigma). The media
`and cell extracts were treated with excess rabbit anti-hGH an-
`tibody (National Pituitary Agency) followed by protein-A-bear-
`ing Staphylococcus aureus (19). Immunoprecipitated proteins
`were collected by centrifugation, eluted by boiling 5 min in 60
`mM Tris-HCl (pH 6.8)/1% NaDodSO1% 2-mercaptoethanol,
`then analyzed by electrophoresis through a 20% acrylamide/
`NaDodSO4 gel and fluorography (20, 21). 'Partial chymotrypsin
`digestions (22) of the immunoprecipitated proteins were per-
`formed in 50-,ul reaction mixtures containing 20 ugofunlabeled
`pituitary hGH and 2.5 ug of chymotrypsin. Aliquots were re-
`moved and boiled after 5, 10, and 60 min ofincubation at 370C,
`pooled, and then analyzed by 20% acrylamide/NaDodSO4 gel
`electrophoresis.
`Double antibody radioimmunoassays with rabbit anti-hGH
`antibody were carried out as previously described (23), using
`hGH (preparation HS 2243E) and antibody from the National
`Pituitary Agency. Radioimmunoassays with guinea pig anti-
`hGH antibody were performed by Hazleton Laboratories (Vi-
`
`Abbreviations: hGH, human growth hormone; SV40, simian virus 40;
`kb, kilobase pair(s).
`
`7398
`
`Merck Ex. 1061, pg 1461
`
`
`
`Biochemistry: Pavlakis et aL
`
`A
`
`hGH gene
`
`Eco RI
`t_
`
`Eco RI
`
`hGH mRNA
`
`5
`
`5'~~~~~_
`
`3~~'
`
`pre hGH
`
`hGH
`
`N
`
`JH2
`
`_
`
`NH2
`
`C02
`
`C02
`
`B
`
`SVhGH(L)
`
`SVhGH(E)
`
`FIG. 1. SV40-hGH recombinants. (A) Both hGH genes were cloned
`on 2.7-kb EcoPJ fragments containing five structural sequences (solid
`bars) and four intervening sequences (hollow bars) together with 500
`base pairs of 5' flanking sequences and 550 base pairs of 3' flanking
`sequences (thick lines) (14). Translation of hGH mRNA yields pre-hGH
`containing an amino-terminal signal sequence of 23 amino acids. This
`is processed to generate mature hGH. (B) The two hGH genes were
`inserted, in both possible orientations, into an SV40d12005 vector (thin
`lines) extending clockwise from theBamHI site at 0.14 map unit to the
`Hpa II site at 0.72 map unit. Both vector sites were converted to EcoRI
`sites by using oligonucleotide linkers (unpublished results). ori, Origin
`of replication.
`
`Proc. Natl. Acad. Sci. USA 78 (1981)
`
`7399
`
`enna, VA). Radioreceptor assays using either IM-9 cultured
`human lymphocytes (24) or pregnant rabbit liver membranes
`were performed by a modification of the method of Tsushima
`and Friesen (25). Our modification involved use of 50 mM
`Hepes buffer and an incubation at 40C for 16 hr.
`
`RESULTS
`SV404GH Recombinants. The hGHl and hGH2 genes
`were originally cloned in phage A as 2.7-kb EcoRI fragments of
`human placental DNA (14). The two genes are approximately
`95% homologous but can be readily distinguished from one an-
`other by the fact that gene 1 contains one BamHI site whereas
`gene 2 contains two sites. DNA sequence analysis shows that
`both the gene 1 and gene 2 EcoRI fragments contain approxi-
`mately 500 base pairs of 5' flanking sequences and 550 base
`pairs of 3' flanking sequences as well as five hGH structural
`sequences (exons) separated by four intervening sequences (in-
`trons) (Fig. 1A). The coding sequences of hGH1 are identical
`to those in cloned hGH complementary DNA, suggesting that
`this gene is expressed into the major form of pituitary hGH (F.
`DeNoto, D. Moore, and H. Goodman, personal communica-
`tion). In contrast, the hGH2 gene differs from the cDNA by
`several base changes, 14 ofwhich are expected to lead to amino
`acid substitutions in the mature hormone (unpublished data).
`The two different 2.7-kb hGH gene fragments were inserted,
`in both possible orientations, into a SV40 vector that retains the
`origin of viral DNA replication, a functional early gene region,
`and the extreme 5' and 3' termini of the late gene region (Fig.
`1B). In the SVhGH(L) recombinants the hGHl and hGH2
`genes are in the same orientation as SV40 late gene transcrip-
`tion, whereas in the SVhGH(E) recombinants they are in the
`
`A
`1 23 4 5 6
`
`B
`34 5 6
`
`1 2
`
`C
`1 2 3 4 56
`=.M
`ilg
`
`D
`1 2 3 4 5 6
`
`.4'Ws.
`
`_
`
`whGH
`
`Synthesis of hGH in infected monkey cells. Monolayers of 2 x 107 monkey kidney cells were labeled with [3H]leucine (200 ,uCi/ml) at
`FIG. 2.
`40-43 hr after infection. Cellular and media proteins were analyzed by immunoprecipitation with excess anti-hGH antibody and 20% acrylamide/
`NaDodSO4 electrophoresis. (A) Immunoprecipitates of cell extracts from 5 x 106 cells. (B) Immunoprecipitates of media from 106 cells. (C) Total
`cellular extracts from 105 cells. (D) Total media from 105 cells. In each panel the lanes represent cells infected with: 1, no virus; 2, wild-type SV40;
`3, SVhGHl(E) plus helper; 4, SVhGHl(L) plus helper; 5, SVhGH2(E) plus helper; 6, SVhGH2(L) plus helper. The symbols (E) and (L) refer to the
`orientation of the genes relative to the late promoter of the SV40 vector.
`
`Merck Ex. 1061, pg 1462
`
`
`
`uAn
`
`* 100
`0!t 80
`
`v
`.3
`.,
`
`60
`
`40
`
`20
`
`FIG. 4. Radioimmunoassay of hGH preparations. In A a rabbit anti-hGH serum was used and in B a guinea pig anti-hGH serum was used. For
`the assay shown in B the standard and 1"5I-labeled hGH ('25I-hGH, specific activity -30 ,uCi/gg) were the same as used in the radioreceptor assays
`shown in Fig. 5. The media used in this experiment were collected from approximately 2 x 107 cells that had been infected with SVhGH1(L) or
`SVhGH2(L) and labeled in 20 ml of medium containing [3H]leucine at 20 ,Ci/ml from 24 to 48 hr after infection. Analysis of these samples by
`acrylamide/NaDodSO4 gel electrophoresis and fluorography showed that they contained approximately equal amounts of [ Hileucine-labeled hGH.
`The lower scale refers to the concentration of pituitary hGH standard added to the assay and the upper scale refers to the amount of infected cell
`medium added. The total sample volume was 100 ,l in all assays. Controls of medium from uninfected cells and cells infected with wild-type SV40
`are shown. B, 125I-hGH bound; Bo '25I-hGH bound in the absence of competitor.
`
`Proc. Nad Acad. Sci. USA 78 (1981)
`
`synthesis of hGH was tested by labeling infected cells with
`[H]leucine and analyzing the cellular proteins by immunopre-
`cipitation and acrylamide/NaDodSO4 gel electrophoresis. As
`shown in Fig. 2A, cells infected with each of the four recom-
`binants synthesized a protein that comigrated with authentic
`pituitary hGH and was absent from uninfected and wild type
`SV40-infected controls. The amount ofthis protein synthesized
`was similar for SVhGHL(L) compared to SVhGH2(L) and for
`SVhGH1(E) compared to SVhGH2(E); in both cases about 3-
`fold more polypeptide was made in the late (L) than the early
`(E) orientation. Thus the two genes function equally well, but
`the level of expression depends upon their orientation relative
`to the vector.
`To determine whether the hGH was secreted from monkey
`cells, we repeated the immunoprecipitation and gel analysis on
`the media from the control and recombinant-infected cells. Fig.
`2B shows that both the hGH1 and hGH2 proteins are present
`in the media. Quantitation of the gel lanes by microdensito-
`metry showed that the secreted material represents approxi-
`mately 80% of the total hGH synthesized in a 3-hr pulse with
`[3H]leucine. To exclude the possibility ofcell leakiness or lysis,
`as compared to active transport, we also analyzed the total in-
`tracellular and media proteins without immunoprecipitation.
`The cell extract (Fig. 2C) showed a complex array of bands, as
`expected from the fact that SV40 does not shut down host cell
`synthesis, and it was not possible to resolve hGH from the back-
`ground of cellular proteins. In contrast, the media (Fig. 2D)
`contained a much more discrete set ofproteins, and hGH could
`readily be visualized as a predominant band in the recombinant-
`infected samples but not in the controls. No SV40 capsid pro-
`teins were detected in the media, indicating that cell lysis was
`not occurring at the time oflabeling (40 hr after infection). These
`results show that both hGHl and hGH2 are specifically and
`efficiently secreted from infected monkey cells.
`The structure ofthe secreted hGH1 and hGH2 was analyzed
`by partial chymotrypsin digestion of the NaDodSO4-denatured
`proteins followed by acrylamide/NaDodSO4 gel electropho-
`resis. Fig. 3 shows that both proteins gave rise to identical
`[3H]leucine-containing chymotryptic peptides and that these
`comigrated with the peptides obtained from unlabeled pituitary
`hGH. These data, in conjunction with the fact that the intact
`proteins comigrate with pituitary hGH on NaDodSO4 gels, sug-
`gest that the amino-terminal signal sequences have been ap-
`propriately removed. The hGH1 protein also comigrated with
`
`0.1
`
`Medium, ;LI
`1
`
`10
`
`100
`Uninfeted
`Wild type
`
`-
`
``13,...
`
`hGH2
`
`Ptity hGH
`
`hGH1
`
`B
`
`......
`0.1
`
`1.
`
`1
`hGH, ng/ml
`
`10
`
`100
`
`7400
`
`Biochemistry: Pavlakis et aL
`
`hGH _-
`_G'
`
`1.
`
`Partial chymotrypsin digests. [3H]Leucine-labeled hGH1
`FIG. 3.
`and hGH2 were purified by immunoprecipitation of the media from
`cells infected with SVhGH1(L) or SVhGH2(L). The precipitated pro-
`teins were eluted by boiling in NaDodSO4, mixed with 20 jtg of un-
`labeled pituitary hGH, partially digested with chymotrypsin, and ana-
`lyzedby 20% acrylamide/NaDodSO4 gel electrophoresis. (A) Fluorogram
`of undigested samples. (B) Fluorogram of chymotrypsin-digested sam-
`ples. (C) Photograph of panel B stained with Coomassie brilliant blue.
`In each panel lane 1 represents hGH1 and lane 2 represents hGH2.
`
`opposite or early orientation. These recombinant molecules
`were constructed by cloning in Escherichia coli, then propa-
`gated in monkey kidney cells as virions by mixed transfection
`with a temperature-sensitive early gene mutant of SV40 (SV40
`tsA239) as helper (26). The resulting stocks of virus contained
`approximately 10% SV40-hGH recombinant genomes and 90%
`helper genomes (results not shown). These viral stocks were
`used to infect fresh monolayers ofprimary monkey kidney cells
`for all subsequent studies.
`Synthesis, Processing, and Secretion of hGH in Infected
`Monkey Cells. The ability of these recombinants to direct the
`
`Medium, Al
`1
`
`10
`
`hGH, ng/ml
`
`Merck Ex. 1061, pg 1463
`
`
`
`Biochemistry: Pavlalds et al
`
`Proc. Natd Acad. Sci. USA 78 (1981)
`
`7401
`
`Medium, ,A
`
`Medium, Al
`
`tR
`1-41
`
`la
`
`9:
`
`0QQ1--,s
`12
`
`w0.
`
`4
`
`0410.-IC4
`
`0
`0
`tR
`
`,0
`_s
`0
`
`10
`hGH, ng/ml
`
`hGH, ng/ml
`
`Radioreceptor assay of hGH preparations. In A data from the IM-9 cultured human lymphocyte assay are shown and in B data from
`FIG. 5.
`the pregnant rabbit liver membrane assay are shown. In each instance the assay contained the receptor preparation, 1"I-hGH, and various amounts
`of the unknown hGH preparation. These ingredients were diluted in buffer to final volume of 500 gI and the incubation and separation of the bound
`and free components were carried out as described (24, 25). The media were the same samples used in Fig. 4. Note that the relationship of the lower
`and upper scales is different from that in Fig. 4.
`
`pituitary hGH in isoelectric focusing (27) and nonequilibrium
`pH gradient electrophoresis gels (28), but hGH2 could not be
`resolved in either system (results not shown).
`Binding of hGH1 and hGH2 to Antibodies and Receptors.
`Fig. 4 shows double antibody radioimmunoassays of the media
`from SVhGH1- and SVhGH2-infected cells using either rabbit
`(Fig. 4A) or guinea pig (Fig. 4B) anti-hGH antibody. In order
`to ensure that we compared equal quantities of hGH1 and
`hGH2 polypeptides, the media used in this experiment were
`collected from cells uniformly labeled with [3H]leucine. Scans
`ofthe NaDodSO4 gel fluorograms showed that the SVhGH1 and
`SVhGH2 samples contained equal amounts of hGH polypep-
`tide. The dose-response curves for hGH1 were parallel to the
`pituitary hGH standard in both assays. In contrast, hGH2 gave
`nonparallel curves with both antisera and it reacted less with
`the guinea pig antibody than with the rabbit antibody. Because
`of the nonidentical crossreactivity the most dilute samples of
`hGH2 give the highest values in radioimmunoassay. Under
`these conditions hGH2 had approximately 10% the immunoac-
`tivity of hGHl with the rabbit antibody and less than 5% with
`the guinea pig antibody.
`The ability of hGH1 and hGH2 to bind to hGH cell surface
`receptors was tested by radioreceptor assays using either the
`human lymphocyte line IM-9 (24) or pregnant rabbit liver mem-
`branes (25) as the receptor sources. In both systems hGH1 was
`indistinguishable from pituitary hGH (Fig. 5). Surprisingly,
`hGH2 was 50% as active as hGH1 in the lymphocyte assay and
`100% as active in the liver membrane assay and gave
`dose-response curves parallel to the standard in both systems.
`Thus the ratio of receptor to immunoassay activity is approxi-
`mately 1 for hGH1, whereas it is 10 or greater for hGH2 (Table
`
`Relationship of receptor binding activity (radioreceptor
`Table 1.
`assay; RRA) to immunoreactivity (radioimmunoassay; RIA) of the
`two different gene products
`
`RRA/RIA
`Assay system
`hGH1
`hGH2
`.10
`IM-9 cultured human lymphocytes
`0.87
`Pregnant rabbit liver membrane
`.20
`0.86
`The RIA potency of hGH2 was calculated from Fig. 4B at the greatest
`dilution that exhibited competition. Because of the different slope of
`the curve, this represents a maximum estimate. Accordingly, the RRA/
`RIA ratios calculated for hGH2 are minimum values.
`
`1). The identity ofthe hGHl and hGH2 preparations in the liver
`assay is in contrast to the behavior ofthe Mr 20,000 hGH variant,
`which is only 3-20% as potent as pituitary hGH in this system
`(29).
`
`DISCUSSION
`We have inserted two different hGH genes into SV40 and have
`shown that both genes can be efficiently expressed in infected
`monkey kidney cells. The hGHl protein, as predicted from the
`DNA sequence, appears identical in all respects to the major
`form of pituitary hGH. In contrast, the hGH2 protein differs
`from authentic hGH both in its behavior on isoelectric focusing
`gels and in its low immunoreactivity, yet it binds to hGH re-
`ceptors quite efficiently. These results show that, at least in the
`heterologous SV40-monkey cell system, gene 2 can be ex-
`pressed into a novel variant form of hGH. Whether or not the
`gene 2 product has biological activity in vivo is unknown; we
`refer to it as a "growth hormone" simply because ofits homology
`to pituitary hGH and its ability to bind to hGH receptors.
`Both hGH1 and hGH2 are processed and secreted from mon-
`key cells. Appropriate removal of the hydrophobic leader se-
`quence was not unexpected, because the necessary enzymes are
`present in a variety of cell types and species (30). Also, Gruss
`and Khoury (31) have presented evidence that preproinsulin is
`processed to proinsulin in monkey cells infected with an
`SV40-insulin gene recombinant; however, the mature hor-
`mone was not produced due to failure to remove the internal
`C peptide. In pituitary cells hGH is sequestered into secretory
`granules prior to release into the bloodstream. The specific and
`efficient secretion ofhGH from monkey kidney cells, which lack
`these granules, suggests that they are not essential for transport
`across the cell membrane.
`We do not know to what extent transcription of the hGH
`genes depends upon their own promoters as compared to ad-
`jacent viral promoters, but it is interesting to note that the
`amount ofprotein produced in the late orientation is only about
`3-fold greater than that in the early orientation. If transcription
`were completely dependent upon the SV40 promoters we
`would expect a ratio closer to 25:1 (unpublished results) because
`the later promoter is more active than the early promoter late
`in infection and because the early transcription unit is separated
`from the hGH genes by a polyadenylylation site.
`Because both hGH genes are interrupted by four intervening
`sequences it is clear that their signals for RNA splicing must be
`
`Merck Ex. 1061, pg 1464
`
`
`
`7402
`
`Proc. Natl. Acad. Sci. USA 78 (1981)
`Biochemistry: Pavlakis etalP
`
`functional in our system. Recently, Lewis and coworkers (9, 10)
`have described a Mr 20,000 variant of hGH that lacks amino
`acids 32-46, and it has been suggested that this protein might
`result from alternative splicing of hGH mRNA (32). We have
`not been able to detect this protein in SVhGH-infected monkey
`cells. Therefore, it appears either that these cells utilize dif-
`ferent RNA processing pathways than pituitary cells do or that
`the Mr 20,000 variant is the product of yet another hGH gene.
`Both pituitary and plasma hGH consist of heterogeneous
`forms. For the most part this heterogeneity relates to higher
`molecular weight forms as exhibited by gel filtration ofpituitary
`or plasma preparations (11, 12). The present genetic data sup-
`port the idea that these higher molecular weight components
`represent posttranslational modifications of the molecule rather
`than different gene products. On the other hand, there are some
`clinical situations that could, in principle, be explained by the
`existence of primary sequence variants of hGH such as hGH2.
`In acromegalic patients it has been found that the major hGH
`component of plasma ("little hGH") has a higher receptor to
`immunoassay ratio than does the material purified from normal
`serum (13). It is apparent from Table 1 that this could be ac-
`counted for by the presence ofhGH2 in the acromegalic serum.
`Thus, an increase in an hGH form similar to hGH2 would pro-
`duce a greater bioactive effect (as predicted from receptor ac-
`tivity) than would be predicted from the immunoreactive hGH
`concentration.
`Although the potency of an hGH preparation in the IM-9
`cultured lymphocyte radioreceptor assay correlates well with
`the bioactivity of the preparation as measured in the classic rat
`bioassay, other circumstances could exist in vivo (33). For in-
`stance, a genetic variant of insulin has been described in a pa-
`tient exhibiting insulin resistance (34). In this case the abnormal
`insulin molecule has decreased bioactivity but, more important,
`it acts as a partial agonist at the level of the insulin receptor and
`results in a shift in the insulin dose-response to the right.
`Whether a similar situation exists for variant short stature, a
`condition in which appropriate plasma concentrations of im-
`munoreactive hGH fail to produce an appropriate tissue re-
`sponse (35-37), is presently unknown.
`4) show that SVhGH-infected
`Radioimmunoassays (Fig.
`monkey kidney cells produce as much as 50 Ag per 2 X 107 cells
`per day (8 x 107 molecules per cell per day) ofhGH, a level that
`compares favorably with cultured pituitary cells. Furthermore,
`because the monkey cells export only a small fraction of their
`own proteins, the hGH can be collected from the media in
`highly enriched form. The ability to produce milligram quan-
`tities of hGH2 should facilitate the search for this protein in
`human tissues and sera and will also allow us to test the bio-
`logical activity of the variant protein by appropriate bioassays
`and animal tests. The high level of hGH production in our ex-
`periments suggests that the SV40-monkey system will be of
`general utility for characterizing secreted animal proteins, in-
`cluding those for which the gene is available but the function
`is unknown.
`We are grateful to Carla M. Hendricks for help with the radioim-
`munoassay, F. DeNoto, D. Moore, and H. Goodman for communicat-
`ing their unpublished results, and Phyllis Donoghue for preparation of
`the manuscript.
`
`1.
`
`2.
`
`3.
`
`Sussman, P. M., Tushinski, R. J. & Bancroft, F. C. (1976) Proc.
`Nati Acad. Sci. USA 73, 29-33.
`Martial, J. A., Hallewell, R. A., Baxter, J. D. & Goodman, H. M.
`(1979) Science 205, 602-607.
`Raiti, S., ed. (1973) Advances in Human Growth Hormone Re-
`search (Dept. Health, Ed., & Welf., Washington, DC), Publ.
`No. 74-612.
`
`4.
`
`5.
`
`6.
`
`7.
`
`8.
`
`9.
`
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`18.
`
`19.
`20.
`
`21.
`
`22.
`
`23.
`
`24.
`
`25.
`
`26.
`27.
`28.
`
`29.
`
`30.
`
`31.
`
`32.
`33.
`
`34.
`
`35.
`
`36.
`
`37.
`
`Bala, R. M., Ferguson, K. A. & Beck, J. C. (1973) in Advances
`in Human Growth Hormone Research ed. Raiti,
`S. (Dept.
`Health, Ed., & Welf., Washington, DC), Publ. No. 74-612, pp.
`494-516.
`Cheever, E. V. & Lewis, U. J. (1969) Endocrinology 85,
`465-470.
`Chrambach, A., Yadley, R. A., Ben-David, M. & Rodbard, D.
`(1973) Endocrinology 93, 848-857.
`Lewis, U. J. & Cheever, E. V. (1965) J. Biol Chem. 240,
`247-250.
`Lewis, U. J., Cheever, E. V. & Hopkins, W. C. (1970) Biochim.
`Biophys. Acta 214, 498-508.
`Lewis, U. J., Dunn, J. T., Bonewald, L. F., Seavey, B. K. &
`VanderLaan, W. P. (1978) J. Biol. Chem. 253, 2679-2687.
`Lewis, U. J., Bonewald, L. F. & Lewis, L. J. (1980) Biochem.
`Biophys. Res. Commun. 92, 511-516.
`Gorden, P., Hendricks, C. M. & Roth, J. (1973) J. Clin. Endo-
`crinol Metab. 36, 178-184.
`Goodman, A. D., Tanenbaum, R. & Rabinowitz, D. (1972) J.
`Clin. Endocrinol. Metab. 35, 868-878.
`Gorden, P., Lesniak, M. A., Eastman, R., Hendricks, C. M. &
`Roth, J. (1976) J. Clin. Endocrinol Metab. 43, 364-373.
`Fiddes, J. C., Seeburg, P. H., DeNoto, F. M., Hallewell, R. A.,
`Baxter, J. D. & Goodman, H. M. (1979) Proc. Natl. Acad. Sci.
`USA 76, 4294-4298.
`Goodman, H. M., DeNoto, F., Fiddes, J. C., Hallewell, R. A.,
`Page, G. S., Smith, S. & Tischer, E. (1980) in Mobilization and
`Reassembly of Genetic Information, Miami Winter Symposium,
`eds. Scott, W. A., Werner, R., Joseph, D. R. & Schultz, J. (Ac-
`ademic, New York), Vol. 17, pp. 155-179.
`Hamer, D. H. (1980) in Genetic Engineering, eds. Setlow, J. K.
`& Hollaender, A. (Plenum, New York), Vol. 2, pp. 83-101.
`Hamer, D. H., Kaehler, M. & Leder, P. (1980) Cell 21, 697-708.
`Seigh, M. J., Topp, W. C., Hanich, R. & Sambrook, J. F. (1978)
`Cell 14, 79-88.
`Kessler, S. W. J. (1975)J. Immunol. 115, 1617-1624.
`Maizel, J. V., Jr. (1971) in Methods in Virology (Academic, New
`York), Vol. 5, pp. 179-246.
`Bonner, W. M. & Laskey, R. A. (1974) Eur. J. Biochem. 46,
`83-88.
`Cleveland, D. W., Fischer, S. G., Kirschner, M. & Laemmli, U.
`K. (1977)J. Biol. Chem. 252, 1102-1106.
`Gorden, P. & Hendricks, C. M. (1974) in New Techniques in Tu-
`mor Localization and Radioimmunoassay, eds. Croll, M. N.,
`Brady, L. W., Honda, T. & Wallner, R. J. (Wiley, New York),
`pp. 17-24.
`Eastman, R. C., Lesniak, M. A., Roth, J., DeMeyts, P. & Gor-
`den, P. J. (1979) J. Clin. Endocrinol. Metab. 49, 262-268.
`Tsushima, T. & Friesen, H. G. (1973)J. Clin. Endocrinol. Metab.
`37, 334-337.
`Chow, J. Y. & Martin, R. G. (1974)J. Virol 13, 1101-1109.
`O'Farrell, P. H. (1975) J. Biol. Chem. 250, 4007-4021.
`O'Farrell, P. Z., Goodman, H. M. & O'Farrell, P. H. (1977) Cell
`12, 1133-1142.
`Sigel, M. B., Thorpe, N. A., Kobrin, M. S., Lewis, U. J. & Van-
`derlaan, W. P. (1981) Endocrinology 108, 1600-1603.
`Blobel, G., Walter, P., Chang, C. N., Goldman, B., Erickson,
`A. H. & Lingappa, V. R. (1979) in Symposium of the Society of
`Experimental Biology, eds. Hopkins, C. R. & Duncan, C. J.
`(Cambridge, Cambridge, England), Vol. 33, pp. 9-36.
`Gruss, P. & Khoury, G. (1981) Proc. Natl. Acad. Sci. USA 78,
`133-137.
`Wallis, M. (1980) Nature (London) 284, 512.
`Lesniak, M. A., Gorden, P., Roth, J. & Gavin, J. R., III (1974)
`J. Biol. Chem. 249, 1661-1667.
`Given, B. D., Mako, M. E., Tager, H. S., Baldwin, D.,
`Markese, J., Rubenstein, A. H., Olefsky, J., Kobayashi, M., Kol-
`terman, 0. & Poucher, R. (1980) N. EngI J. Med. 302, 129-135.
`Rudman, D., Kutner, M. H., Goldsmith, M. S., Kenny, J., Jen-
`nings, H. & Bain, R. P. (1980) J. Clin. Endocrinol. Metab. 51,
`1378-1384.
`Kowarski, A. A., Schnieder, J., Ben-Galim, E., Weldon, V. V.
`& Daughaday, W. H. (1978) J. Clin. Endocrinol. Metab. 47,
`461-464.
`Frazor, T. E., Gavin, J. R., Daughaday, W. H., Hillman, R. E.
`& Weldon, V. V. (1980) Pediatr. Res. 14, 478 (abstr.).
`
`Merck Ex. 1061, pg 1465