`Vol. 80, pp. 825--829, February 1983
`Immunology
`
`Immunoglobulin gene expression in transformed lymphoid cells
`(gpt/transformation)
`
`VERNON T. Or*, SHERIE L. MORRISONt, LEONARD A. HERZENBERG*, AND PAUL BERG:t
`Departments of •Genetics and tBiochemistry, Stanford University School of Medicine, Stanford, Califurnia 94305; and *Department of Microbiology and the Cancer
`Center, Institute of Cancer Research, CoUege of Physicians and Surgeons of Columbia University, New York, New York 10032
`
`Contribt.ted by Leonard A. Herunberg, October 15, 1982
`
`ABSTRACT Myeloma, hybridoma, and thymoma cell lines
`have been successfully transfected for the E~eherichia coli xan(cid:173)
`thine-guanine phosphorihosyltransferase gene (gpt) by using the
`plasmid vector pSV2-gpt. The transformed cells synthesize the
`bacterial enzyme 5-phospho-a-D-ribose-1-dipbosphate:xanthine
`phosphorihosyltransferase (XGPRT; EC 2.4.2.22) and have been
`maintained in selective medium for over 4 months. Lymphoid cell
`lines expressing a " immunoglobulin light chain were obtained by
`transfecting cells with pSV2-gpt containing a rearranged IC light
`chain genomic segment from the S107 myeloma ceU line. The S107
`light chain is synthesized in gpt-transformed J558L myeloma ceUs
`and is identical to the light chain synthesized by the Sl07 myeloma
`cell line, as judged by immunoprecipitation and two-dimensional
`gel electrophoresis. Furthermore, this light chain is synthesized
`and secreted as part of an intact antibody molecule by transformed
`bybridoma ceUs that normally secrete an IgGl (y,K) antibody
`molecule. No light chain synthesis was detected in a similarly
`transformed rat myeloma or a mouse thymoma line.
`
`Techniques to introduce novel genes into eukaryotic cells pro(cid:173)
`vide a powerful tool to study mechanisms of gene regulation and
`expression. Most studies on eukaryotic gene expression have
`been conducted in heterologous host cells-i.e., genes have
`been transfected into cell types (particularly human HeLa and
`mouse L cells) that nonnally do not express the gene of interest
`(1-3). Though a great deal has been learned about eukaryotic
`regulator sequences with these gene transfer experiments, it
`would be preferable to transfer genes encoding proteins ex(cid:173)
`pressed during differentiation back into the cell type that nor(cid:173)
`mally expresses the genes of interest. The appropriate cell type
`provides protein modification systems, such as glycosyltrans(cid:173)
`lerases, necessary to make fully biological functional products.
`In addition, the appropriate cell type may be used to study tis(cid:173)
`sue-specific regulation of gene expression.
`To undertake studies of (i) the regulation and expression of
`immunoglobulin genes, (ii) the biosynthesis, chain-assembly,
`and secretion of immunoglobulin heavy and light chains, and
`(iii) structure-function correlates of antibody molecules, we
`have explored techniques for transfection of lymphoid cells us(cid:173)
`ing the pSV2-gpt vector (4, 5). This DNA can express the Eco
`gpt gene encoding xanthine-guanine phosphoribosyltransferase
`(XGPRT; 5-phospho-a-o-ribose-l-diphosphate:xanthine phos(cid:173)
`phoribosyltransferase, EC 2.4.2.22). Cells synthesizingXGPRT
`can be grown with xanthine as the sole precursor of guanine
`nucleotide fonnation (4, 5). Successfully transfonned cells can
`be isolated by their ability to grow in medium containing xan(cid:173)
`thine and mycophenolic acid, an inhibitor of guanine nucleotide
`synthesis; if the transformed cell line is hypoxanthine phos(cid:173)
`phoribosyltransferase-negative (HPRr; IMP pyrophosphate
`phosphoribosyltransferase, EC 2.4.2.8), transfunnants can be
`
`The publication costs of this article were defrayed in part by page cbarge
`payment. This article must therefore be hereby marked "advertise(cid:173)
`ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
`
`825
`
`selected in hypoxanthine/aminopterin/thymidine (HAT) me(cid:173)
`dium (6). In the present experiments both calcium phosphate
`precipitation (7, · 8) and protoplast fusion (9) techniques have
`been used to transfect cells.
`pSV2-gpt containing a rearranged K light chain gene (10) was
`used to transfonn several cultured lymphoid cell lines. Among
`the gpt transfonnants were clones that produce a new immu(cid:173)
`noglobulin light chain. The light chain produced by these trans(cid:173)
`funned cell lines appears to be identical to the light chain syn(cid:173)
`thesized by the myeloma cell from which the rearranged gene
`was isolated. Furthennore, in transfonned hybridoma cells,
`this light chain is assembled with an immunoglobulin heavy
`chain and secreted as a complete antibody molecule.
`
`MATERIALS AND METHODS
`Cell Lines. J558L is a spontaneous heavy chain-loss-variant
`myeloma cell line obtained from the ]558 cell line [ a,A; anti-a 1-
`3 dextran (ll)] that synthesizes and secretes a A light chain. Y3-
`Ag1.2.3 is a HPRrrat myeloma cell line originally described
`by Galfre et aL (12) that synthesizes and secretes a rat K light
`chain. 27-44 is a HPRT- mouse IgG 1 anti-dansyl hybridoma cell
`line (13); and BW5147 is a HPRr, ouabain' AKR thymoma
`originally described by Hyman and Stallings (14). Cell lines
`were maintained in either 10% newborn calf serum in Dul(cid:173)
`becco' s modified minimal essential medium (DME medium) or
`10% fetal calf serum in alpha modified minimal essential me(cid:173)
`dium.
`Recombinant DNA Vectors. The plasmid vector pSV2-gpt
`has been described (4, 5). Fig. 1 shows a partial restriction en(cid:173)
`zyme map of this vector. A second vector, which is derived from
`pSV2-gpt, but contains the herpes simplex thymidine kinase
`promoter inserted 5' of the gpt gene, was constructed by J.-F.
`Nicolas (unpublished data). pSV2-S107 was constructed by in(cid:173)
`serting a BamHI fragment containing the entire rearranged
`phosphocholine-specific K chain gene from the S107 myeloma
`cell line (10) into the unique BamHl site in pSV2-gpt. The light
`chain gene is oriented so that the direction of transcription is
`opposite to the gpt gene (Fig. 1). The genomic rearranged Sl07
`K light chain DNA was a gift from M. Scharff.
`Transfection by Protoplast Fusion. Protoplasts were pre(cid:173)
`pared essentially as described by Sandri-Goldin et aL (9). &ch(cid:173)
`erichia coli K-12 strain HB10l, containing the appropriate plas(cid:173)
`mid, was grown at 37"C in Luria broth containing 1% glucose
`to an absorbance at 600 nm of0.6-0.8. Chloramphenicol was
`added to 125 ,u.g/ml, the culture was incubated at 37"C for 12-
`16 hr to amplify the plasmid copy number, and the cells were
`harvested by centrifugation. For every 25 ml of culture, 1.25
`ml of chilled 20% sucrose/0.05 M Tris•HCI (pH 8) was added;
`
`Abbreviations: XGPRT, xanthine-guanine phosphoribosyltransferase;
`HPRT, hypoxanthine phosphoribosyltransferase; HAT, hypoxanthine/
`aminopterin/thymidine; DME medium, Dulbecco's modified minimal
`essential medium.
`
`Sanofi/Regeneron Ex. 1031, pg 915
`
`Merck Ex. 1031, pg 941
`
`
`
`-----------~~ --~ --~
`
`826
`
`Immunology: Oi et aL
`
`Proc. NatL Acad. Sci. USA 80 (1983)
`
`FIG. 1. Structure of the vectors
`used for lymphoid cell transforma(cid:173)
`tions. The diagram of the parental
`p8V2-gpt plasmid vector was taken
`from Mulligan and Berg (4, 5): pBR322
`DNA is represented by the solid black
`lines and the plasmid's DNA replica(cid:173)
`tion origin and 1!-lactamase gene are
`indicated; the gpt gene sequence is
`represented by the hatched segments;
`simian virus 40 (SV40) sequences are
`the stippled segments. The SV40 ori(cid:173)
`gin of DNA replication (ori) and early
`promoter are located 5' of the gpt se-
`quences. pSV2-gptTKpr has an inser(cid:173)
`tion of 250 base pairs, containing the
`herpes simplex thymidine kinase pro(cid:173)
`moter, between the gpt gene and the
`SV40 early promoter (unpublished
`data). pSV2-8107 has a 7-kilobaae
`BamlD fragment, containing the en(cid:173)
`tire genomic 8107 light chain gene,
`inserted into the unique BamlD site
`of pSV2-gpt. This rearranged light
`chain gene is oriented in the opposite
`cfuection to gpt and contains the leader,
`V, and "constant region exons as well
`as flanking 5' and 3' sequences.
`
`Ec:o11Pt
`
`pBR322 O<i
`
`Pvull
`
`SV40 O<i
`
`the bacteria were suspended and 0.25 ml oflysozyme [a freshly
`prepared solution of5 mg/ml in 0.25 M Tris·HCI (pH 8)] was
`added. After5 minofincubationonice, 0.5 mlof0.25 M EDTA
`(pH 8) was added and incubation on ice was continued for an
`additional 5 min. After addition of 0.5 ml of 0.05 M Tris·HCI
`(pH 8), the bacteria were transferred to a 37"C water bath and
`were incubated for 10 min. At this time examination of the bac(cid:173)
`teria with a phase-contrast microscope showed that the vast
`majority had been converted to protoplasts. The bacteria were
`diluted with 10 ml of DME medium containing 10% sucrose
`and 10 mM MgCI2 that was warmed to 37"C. After further in(cid:173)
`cubation for 10 min at room temperature the protoplasts were
`ready for fusion.
`Fusion of protoplasts with suspension cells was effected with
`a procedure normally used in the production of hybridomas
`(15). Cell lines were grown to a density of0.3-l X 106 cells per
`ml in DME medium supplemented with 10% newborn calf
`serum. Five milliliters of the protoplast suspension was added
`to 2 x 106 cells in growth medium. The mixture was centrifuged
`for 5 min at room temperature at approximately 500 X g. The
`supernate was aspirated and the pellet was resuspended gently
`in 2 ml of a polyethylene glycol solution [50 g of polyethylene
`glycoll,500 (BDH) in 50 ml of DME medium] adjusted to pH
`8 with C02. After 3 min of centrifugation at 500 X g the poly(cid:173)
`ethylene glycol was diluted with 7 ml of DME medium while
`resuspending the pellet. After 5 min of centrifugation at 500
`X g, the supernate was removed carefully and the cells were
`resuspended in DME medium containing 10% newborn calf
`serum and garamycin at 100 ~-tg/ml and were plated either in
`96-well or 24-well plates. After 48 hr, cells were diluted with
`an equal volume of DME medium containing xanthine at 250
`~-tg/ml, hypoxanthine at 15~-tg/ml, mycophenolic acid at 6~-ti/
`ml, and 10% newborn calf serum. Every several days, as re(cid:173)
`quired, spent medium was aspirated carefully and was replaced
`with fresh medium containing the same supplements. Colonies
`of transformants were visible by 10 days. Transformants were
`maintained in selective medium.
`Transfection by Calcium Phosphate Precipitation. Lym(cid:173)
`phoid cell lines grown in suspension were transfected by cal(cid:173)
`cium phosphate precipitation as described by Chu and Sharp
`(7). Ten times concentrated HeBS buffer was stored at -20°C
`
`until used, whereupon it was diluted to two times concentrated
`and adjusted to pH 7.05. Plasmid DNA (80 ~-tg/ml) was made
`up in 125 mM CaCl2 which was stored as a 2 M stock solution
`at -200C. DNA-calcium phosphate precipitates were formed
`by dropwise addition of the DNA into the HeBS solution. The
`precipitate formed in 30 min at room temperature. The final
`DNA concentration was 40 ~-tg/ml.
`Cells were washed once in serum-free medium and were
`suspended directly in the DNA-calcium phosphate precipitate
`(lOS cells per 20 ~-tg of DNA per 0.5 ml). This suspension was
`incubated at 37"C for 30 min and then was diluted 1:10 in
`serum-eontaining medium. The cells were plated either into 24-
`well plates (2 X l(f cells per well) or 96-well plates (2 X 104
`cells per well). Transfection of Y3 cells was done as described
`by Graham and Vander Eb (8) for adherent cell lines. The DNA(cid:173)
`calcium phosphate precipitate was put directly onto the cell
`monolayer. After 30 min at 37"C, serum-eontaining medium
`was added. After 24 hr, half of the medium volume from each
`culture was removed and was replaced with fresh medium. On
`days 3, 4, 5, 8, 11, and 14, half the medium volume was re(cid:173)
`moved and HAT medium was added. Transformed colonies
`were visible between 10 and 21 days.
`Immunoprecipitations and Gel Electrophoresis. Immuno(cid:173)
`precipitations were done with [35S]methionine-labeled cellly(cid:173)
`sates and supernates. Biosynthetic labeling procedures have
`been described (16). Rabbit anti-mouse light chains, rabbit anti(cid:173)
`mouse K light chains, rabbit anti-mouse immunoglobulin, and
`a hybridoma anti-mouse IgGl allotype antibody were used for
`immunoprecipitations. Staphylococcus aureus, Cowan strain 1
`(IgGsorb; Enzyme Center, Boston) was used to coprecipitate
`the antigen-antibody complexes (16).
`One-dimensional NaDodS04/polyacrylamide slab electro(cid:173)
`phoresis and two-dimensional nonequilibrium gradient gel
`electrophoresis were done as described (17). Autoradiography
`of polyacrylamide gels was with preflashed XAR-5 film and fluo(cid:173)
`rography by using sodium salicylate (18).
`
`RESULTS
`Transfection Frequencies. The frequency at which stable
`transformed lymphoid cell lines were generated was influenced
`
`Sanofi/Regeneron Ex. 1031, pg 916
`
`Merck Ex. 1031, pg 942
`
`
`
`Immunology: Oi et al
`
`Proc. NatL Acad. Sci. USA 80 (1983)
`
`827
`
`by every parameter tested. Different cell lines and different
`vectors produced different transfOrmation frequencies. More(cid:173)
`over, the two DNA delivery procedures, protoplast fusion and
`calcium phosphate precipitation, yielded different transfor(cid:173)
`mation frequencies. Tables 1 and 2 summarize the results by·
`using protoplast fusion and calcium phosphate precipitation,
`respectively.
`Under the present experimental conditions, BW5147 ap(cid:173)
`pears to.be the least competent recipient of the cell lines tested,
`having a transformation frequency of approximately 10-6• Y3
`and 27-44 yielded frequencies in the rangeof0.3 to >5 X 10-6.
`In the present experiments, J558L yielded the highest fre(cid:173)
`quency with the range of 3 X 10-6 to > 10-4
`. Protoplast. fusion
`appears on balance to be a more efficient delivery system than ·
`calcium phosphate precipitation.
`A striking feature of these results is the enhanced transfor(cid:173)
`mation frequency for gpt obtained with the light chain-contain(cid:173)
`ing vector, pSV2-S107. This dramatic increase is evident when
`the pSV2-S107 vector was used with the J558L and Y3 myeloma
`cell lines; transformation with this recombinant was 5- to at least
`10-fold greater than that obtained with the other. vectors. Trans(cid:173)
`formation of the hybridoma 27-44 cell line was increased only
`about 2-fold with pSV2-S107. The sequence(s) in the pSV2-S107
`insert that is responsible for the enhanced transformation fre(cid:173)
`quency must yet be mapped. Transformation of the Y3 cell line
`was occasionally greater with pSV2-gptTKpr than with pSV2-
`gpt (fable 2). Regardless of which vector was used, BW5147
`transformants were detected only at very low frequencie:;. The
`amount of XGPRT activity in cell lines stably transformed by
`the three recombinant plasmids was not significantly different
`(Fig. 2 and data not shown).
`XGPRT Activity. The transformed cell lines expressed ·the
`Eco gpt gene, as measured.by the presence ofXGPRT activity
`in the celllysates. E. coli XGPRT can be distinguished from
`mammalian HPRT activity by its different electrophoretic mo(cid:173)
`bility (4, 5). 1n cells selected for resistance to mycophenolic acid,
`both the cellular HPRT and bacterial XGPRT activities were
`detectable (Fig. 2). Cells lacking their own HPRT activity and
`selected for gpt in HAT medium had only the bacterial enzyme
`activity (Fig. 2).
`Immunoglobulin Light Chain Expression. The organization
`of exons in the S107 genomic light chain gene is shown in Fig.
`1. To produce the S1071ight chain protein from this gene, two
`introns must be processed from the primary mRNA transcripts
`and the leader polypeptide removed by post-translational cleav(cid:173)
`age. For secretion of the light chain as part of an intact antibody
`molecule, the newly synthesized light chain must fold and as(cid:173)
`semble with an immunoglobulin heavy chain to form an H2Lz
`tetramer. This also involve; the formation of lnterchain disul(cid:173)
`fide bonds.
`
`Table 1. Transformation of lymphoid cell lines with the
`pSV2-gpt vectors by using .protoplast fusion
`Cell line
`
`Vector
`pSV2-gpt
`pSV2-gptTKpr
`pSV2-S107
`
`J558L
`21/288* 27 j36t
`10/190° 24 /36t
`186/192° 36/36t
`149/192'
`
`BW5147
`0/76° 1/48t
`o;so• 1/48t
`4/96* 8j48t
`
`Table 2. Transformation of lymphoid cell lines with the pSV2-
`gpt.vectors by using calcium phosphat& precipitation
`Cell line
`
`Vector
`pSV2-gpt
`pSV2-gptTKpr
`pSV2-S107
`
`Y3
`3/48°
`19/48°
`47/48°
`
`27-44
`10/192t
`10/192t
`43/288t
`
`BW5147
`1f192t
`1/192t
`0/192t
`
`Results are from three experiments with the Y3·cell line and two
`experiments with 27-44-and BW5147 celf lines.
`• Cells were plated in 24-well culture dishes at 2 x 1W cells per well.
`t Cells were plated. in 96-well culture dishes at 4 x 10' cells per well.
`
`Of the four cell lines stably transformed with pSV2-Sl07, the
`J558L cell line synthesized, but did not secrete, the S 107 K light
`chain. However, this cell line was not expected to secrete the
`newly made light chains because heavy chain-loss-variants of
`the SlQ7 myeloma cell line also do not secrete endogenous light
`chains (M. Scharff, personal communication). Transformants of
`the 27-44 hybri.doma cell line synthesized and secreted the S107
`light chain. Moreover, the Sl071ight chain was assembled into
`tetrameric H2~ immunoglobulin molecules with the endoge(cid:173)
`nous yl heavy chain and was secreted. Twelve independently
`transformed Y3 and seven BW5147 cell lines did not produce
`detectable amounts of the Sl07 light chain, as judged by im(cid:173)
`munoprecipitation and gel analyses. XGPRT analyses verified
`that these cells were, indeed, transformants.
`Autoradiograms of two-dimensional polyacrylamide gels
`showing the apparent M~ and charge of the light chains pro(cid:173)
`duced by J558L and 27-44 transformants are shown in Figs. 3
`and 4. The two-dimensional gel pattern. of the S 107 light chain
`synthesized by Sl07 myeloma cells is included to show that the
`transformed cell lines produced a light chain that is identical
`in apparent Mr and charge. The two-dimensional gel patterns
`also show that the leader polypeptide was removed in trans(cid:173)
`formed cell lines that expreS5ed the light chain. This indicates
`that proper transcription, mRNA, and protein processing occur
`in the transforrnants. Transcription of the Sl071ight chain gene
`probably occurs from its own promoter, because the light chain
`gene is oriented opposite to the direction of the SV40 early pro(cid:173)
`moter (see Fig. 1).
`The antibodies secreted by 27-44 transformants were im(cid:173)
`munoprecipitated with both hybridoma anti-IgG1 allotypic an(cid:173)
`tibody and rabbit anti-mouse light chain antisera. Both reagents
`precipitated the Sl07 light chain (data not shown). Sequential
`precipitation, first with the hybridoma anti-IgG 1 antibody and
`
`2
`
`3
`
`4
`
`5
`
`7
`
`XG!'RT,., ..
`
`Results are from three experiments and are expressed as the number
`of culture wells having stable transformants.
`• After protoplast fusion cells were plated in 96-well culture dishes at
`10' cells per well.
`t Cells were plated at 1W cells per 2.0 ml of culture in 24-well dishes.
`; Cells were plated at 5 x 103 cells per well in 96-well culture dishes.
`
`FIG. 2. XGPRT and HPRT production in transfDrmed lymphoid
`cell lines. Enzyme analyses were done as described by Mulligan and
`Berg (4, 5). Lanes: 1 and 2, electrophoreti~ mobility Df mammalian
`HPRT; 3-5, J558L cell transformants; and 6 and 7, tranaformants of
`the 27-44 cell line. Because 27-44·is a HPRT~ cell line, only XGPRT
`is present.
`
`Sanofi/Regeneron Ex. 1031, pg 917
`
`Merck Ex. 1031, pg 943
`
`
`
`828
`
`Immunology: Oi et al.
`
`Proc. NatL Acad. Sci. USA 80 (1983)
`
`AC!OIC
`
`A
`
`M. , 1D ,.
`
`BASIC
`
`ACIDIC
`A
`
`M
`
`.•
`
`'O • BASIC
`
`67~
`
`-
`
`30
`
`20
`
`67
`
`20-·
`
`67
`
`30
`
`45-····
`
`30 .....
`
`70
`
`3Q.,
`
`20·-
`
`20-
`
`30-
`
`3()~
`
`-
`
`S107t:
`
`c
`
`D
`
`FIG. 4. 8107light chain produced by transformed J558L cells. Au·
`torad.iograms of two-dimensional gels of light chains immunoprecip(cid:173)
`itated from celllysates of J558L transformed with pSV2-8I07 DNA
`are shown. Because J'558L does not produce a heavy chain, only the
`light chain portions of the two-dimensional gels are shown. (A) A light
`chain produced by the parental J558L cell line. (B) 8107 K light chain.
`(C) A mixture of the two light chains. The K and A light chains are
`distinguished on the basis of both charge and apparent Mr (J) and E)
`Two independently derived J558L cell lines transformed with pSV2-
`8107 DNA. The tranaformant examined in D only appears to produce
`larger quantities of the 8107 K light chain than the endogenous J558
`A light chain, because the 8107 K chain is synthesized and remains in
`the cytoplasm, while the J558 A chain is synthesized and secreted.
`
`compared. Amounts varied from barely detectable to quantities
`equal to endogenous light chain. This variation may be due to
`the chromosomal region where the light chain has integrated.
`It also could result from different copy number of the light chain
`gene in different transformants. Quite possibly, mutations or
`deletions of sequences needed for the expression of this gene
`could have occurred during transformation or subsequent to
`integration of the light chain sequence. Further studies are
`needed to determine the cause of this variation and why light
`chain expression does not occur in Y3 or BW5147 cell lines
`transformed with the same light chain gene vector.
`
`DISCUSSION
`These experiments show that it is possible to use two methods,
`calcium phosphate precipitation and protoplast fusion, to intro(cid:173)
`duce genes into lymphoid cells. With pSV2-gpt containing the
`gene for an immunoglobulin light chain (pSV2-Sl07) both meth(cid:173)
`ods give rise to transformants that synthesize bacterial XGPRT
`and the murine light chain. Higher transformation frequencies
`are seen following protoplast fusion. Indeed, by using protoplast
`fusion and the pSV2-Sl07 plasmid, transformants can be ob(cid:173)
`tained at a frequency of greater than w-•. Transformation fre(cid:173)
`quencies are lower when using the other plasmids or calcium
`phosphate precipitation. Because mycophenolic acid resistance
`or reversion of the HPRT- phenotype do not occur sponta(cid:173)
`neously in the cell lines used, stable transformation, at even low
`frequencies, can be detected.
`A surprising result is the increased frequency of gpt trans(cid:173)
`formation when the Sl07 light chain is incorporated into the
`
`Sanofi/Regeneron Ex. 1031, pg 918
`
`[)
`
`FIG. 3. 8107light chain produced by transformed 2744 cells. Au·
`toradiograms of two-dimensional gels of the light and heavy chains
`produced by parental and transformed cell lines are shown. (A) Pa(cid:173)
`rental2744 IgG1 anti-dansyl antibody immunoprecipitated with an
`anti-IgGl-specific hybridoma antibody. Both the -y1 heavy and K light
`chains can be seen. (B) 8107 IgA antibody immunoprecipitated with
`a rabbit anti-IgA antiserum. The a heavy chain was distinguished
`clearly by charge and apparent M, from the -yl heavy chain in A. (C)
`A mixture of the immunoprecipitates of A and B. The two K light
`chains can be seen as distinct .spota (indicated by arrows) having
`nearly identical charge but different apparent M,. (J)) Immunoprecip(cid:173)
`itate of a transformed 27-44 cell line. Only the -yl heavy chain was
`present, but two light chains can be seen. In this case the amount of
`8107light chain was considerably lower than in the artificial mixture
`shown in C.
`
`then with the rabbit anti-K antisera, indicated that little, if any,
`freeS 107light chain was secreted by these cells. This shows that
`the 5107 light chain is assembled with the -yl heavy chain into
`an intact antibody molecule.
`Different amounts ofS1071ight chains were produced when
`a number of independent J558L and 27-44 transformants were
`
`Merck Ex. 1031, pg 944
`
`
`
`Immunology: Oi et aL
`
`Proc. Natl Acad. Sci. USA 80 (1983)
`
`829
`
`pSV2-gpt vector. Ibis enhancing effect occurs with both the rat
`and mouse myelomas. A similar increased transformation fre(cid:173)
`quency has been observed with a bovine papillomavirus vector
`containing the human ~globin region sequences (19). At pres(cid:173)
`ent, the mechanism for the increased transformation frequency
`in both cases is obscure. Possibly, the chromosomal DNA pro(cid:173)
`vides an origin of DNA replication, which permits the plasmid
`to replicate within the transformed cell and increases the trans(cid:173)
`formation frequency. Transcription from the immunoglobulin
`promoter cannot be essential for the increased transformation
`frequencies because deletion of the fragments that are pre(cid:173)
`sumed to contain the immunoglobulin promoter region does not
`abolish the enhancement of transformation. It also is possible
`that pSV2-Sl07 is more efficient for transformation because of
`increased XGPRT production; this seems unlikely because
`there are no consistent differences in enzyme levels in the stable
`transformants obtained with either vector.
`DNA-mediated gene transfer into lymphoid cells may permit
`a study of the regulation and expression of immunoglobulin
`genes in cells in which they normally are synthesized. It may
`be possible to examine the basis lOr differential immunoglobulin
`gene expression at different stages of lymphocyte differentia(cid:173)
`tion. Cell lines in which immunoglobulin synthesis C8 .l be in(cid:173)
`duced (20-22) are suitable hosts to determine if the transduced
`immunoglobulin genes also are re11ponsive to those signals.
`Studies with cells transformed with genetic elements that are
`inducible by steroid hormones demonstrate that transduced
`DNA can respond, ifthe·cellcontains the appropriate receptors
`(23).
`A question of central importance is what determines the uti(cid:173)
`lization of various promoters and thus the synthesis of defined
`proteins in certain cell lines. In our experiments light chains
`are efficiently produced in both transformed mouse myeloma
`and hybridoma cell lines. However, light chain production did
`not occur in either a rat myeloma or a mouse thymoma. The
`inability of the immunoglobulin promoter to function in a dif(cid:173)
`ferent species has been reported by Falkner and Zachau (24).
`The lack of production of mouse immunoglobulin in a rat my(cid:173)
`eloma is surprising because mouse myelomas have been used
`to fuse to rat myelomas. to produce hybrid cells that synthesize
`both rat and mouse immunoglobulin molecules (25). The pos(cid:173)
`sibility that the S 107 light chain is synthesized but rapidly de(cid:173)
`graded in the Y3 myeloma has not been excluded.
`There is evidence that differentiated cell types express im(cid:173)
`munoglobulin genes to varying levels. For example, somatic cell
`hybridization of myelomas yields hybridomas that produce an(cid:173)
`tibodies, whereas thymomas yield hybrid cells with T-cell phe(cid:173)
`notypes (26). Furthermore, hybridization of myelomas with
`non-B cells results in cessation of immunoglobulin production
`(26, 27) .. The lack of light chain expression in the transformed
`thymoma may reflect tissue-specific gene regulation. It is im(cid:173)
`portant to determine if immunoglobulin gene expression in the
`nonexpressing mouse thymoma and rat myeloma cell lines is
`regulated at the level of transcription, RNA processing, trans(cid:173)
`lation, or rapid protein turnover.
`The study of the structure and function of the immunoglob(cid:173)
`ulin molecule has been of great interest, both because of the
`ability of immunoglobulin to react with a diverse liunily of li(cid:173)
`gands and also because of the biologic importance of antibody
`molecules. Initially, the study of immunoglobulins was limited
`to the study of heterogeneous serum pools after immunization.
`The advent of myelomas, and more recently hybridomas, has
`permitted the study of homogeneous populations of antibodies.
`DNA-mediated transfection and immunoglobulin gene expres(cid:173)
`sion is an important tool to permit the study of immunoglobulin
`
`molecules. By using this technique, it should be possible to
`study the function of both novel chain combinations and novel
`chain structures. In vitro site-specific mutagenesis techniques
`can be used to construct specific mutations in immunoglobulin
`genes that can be expressed after transfection. Because signif(cid:173)
`icant quantities of immunoglobulin are produced in the trans(cid:173)
`fonnants, sufficient quantities of protein necessary for detailed
`analyses should be obtained.
`
`Note Added in Proof. After this paper was submitted fur. publication,
`we learned that Douglas Rice and David Baltimore have reported sim(cid:173)
`ilar results with a different K light chain gene and different lymphoid
`ceU recipients (28).
`
`We thank Dr. P. Jones fur assistance with tw<Hiimensional gel elec(cid:173)
`trophoresis. This work was supported in part by National Institutes of
`Health Grants GM-13235, CA-15513 (P.B.), Al-08917 (L.A. H.), CA-
`16858 (S.L.M.), CA-22736 (S.L.M.), and CA-13696 to The Cancer Cen(cid:173)
`ter of Columbia University and by a grant from the Becton Dickinson
`FACS Systems (L.A.H.). S.L.M. is a recipient of Research Career
`Development Award Al..oo408.
`
`1. Mantei, N., Boll, W. & Weissmann, C. (1979) Nature (London)
`281, 40--46.
`2. Mellon, P., Parker, V., Gluzrnan, Y. & Maniatis, T. (1981) CeU
`27, 279-288.
`3. Mantei, N. & Weissmann, C. (1982) Nature (London) 297, 128-
`132.
`4. Mulligan, R. C. & Berg, P. (1980) Science 209, 1422-1427.
`5. Mulligan, R. C. 8t Berg, P. (1981) Proc. Natl Acad. Sci. USA 18,
`2072-2076.
`6. Littlefield, J. W. (1964) Science 145, 709--710.
`7. Chu, G. 8t Sharp, P. A. (1981) Gene 13, 197-202;
`8. Graham, F. L. 8t Van der Eb, A. J. (1973) J. Virol 52, 455--456.
`9. Sandri-Goldin, R. M., Goldin,. A. L., Levine, M. & Glorioso, J.
`C. (1981) Mol CelL Blol1, 743-752.
`10. K.wan, S.-P., Rudikoff, S., Seidman, J. G., Leder, P. & Scharff,
`M. D. (1981)]. Exp. Med. 153, 1366-1370:
`11. Lundblad, A., Steller, R., Kabat, E. A., Hirst, J. W., Weigert,
`M.G. & Coho, M. (1972) Immunochemistry 9, 535-544.
`12. Galfre, G., Milstein, C. & Wright, 8". (1979) Nature (London)
`277, 131-133.
`13. Dang], J. L., Parks, D. R., Oi, V. T. 8t Herzenberg, L.A. (1982)
`Cytometry 2, 395--401.
`14. Hyman, R. & Stallings, V. (1974) J. Natl Cancer Inst. 52, 429-
`436.
`15. Sharon, J., Morrison, S. L. & Kabat, E. A. (1979) Proc. Natl
`ACtJd. Sci. USA 76, 1420-1424.
`16. Oi, V. T., Bryan, V. M., Herzenberg, L. A. & Herzenberg, L.
`A. (1980)]. Exp. Med.151, 1260-1274.
`17. Oi, V. T., Jones, P. P., Goding, J. W., Herzenberg, L. A. &
`Herzenberg, L. A. (1978) Curr. Top. Microbiol Immunol 81,
`115-129.
`18. Moosic, J. P., Sung, E., Nilson, A., Jones, P. & McKean, D. J.
`(1982)]. Biol Chem. 257, 9684-9691.
`19. DiMaio, 0., Treisman, R. 8t Maniatis, T. (1982) Proc. Natl Acad.
`Sci. USA 19, 4030-4034.
`20. Paige, C. J., Kincade, P. W. & Ralph, P. (1978)]. Immurwl121,
`641-647.
`21. Knapp, M. R., Severinson-Gronowicz, E.,. Schroder, J. &
`Strober, S. (1979)]. lf11111Unol123, 1000-1006.
`22. Boyd, A. W., Goding, J. W. & Schrader, J. W. (1981)J.Immunol
`126, 1261-1265.
`23. Lee, F., Mulligan, R., Berg, P. & Ringold, G. (1981) Nature
`(London) 294, 228-232.
`24. Falkner, F. G. & Zachau, H. G. (1982) Nature·(I..ondon) 298,
`286-288.
`25. Cotton, R. G. H. & Milstein, C. (1973) Nature (London) 244, 42-
`43.
`26. Goldsby, R. A., Osborne, B. A., Simpson, E. & Herzenberg, L.
`A. (1977) Nature (London) 267, 707'-708.
`27. Coffino, P., Knowles,. B., Nathenson, S. G. & Scharff, M. D.
`(1971) Nature (London) 231, 87-90.
`28. Rice, D. 8t Baltimore, D. (1982) Proc. Natl Acad. Sci. USA 79,
`7862-7865.
`
`Sanofi/Regeneron Ex. 1031, pg 919
`
`Merck Ex. 1031, pg 945