Cell-type preference of immunoglobulin kappa and lambda gene promoters.

Immunoglobulin gene constant regions are known to be associated with strictly tissue-specific enhancer elements. Until recently the promoter of the variable region, which becomes linked to the constant region by somatic rearrangement, could have been viewed as a passive recipient of the enhancer stimulus. Here we show that the promoters of the immunoglobulin kappa and lambda light chain genes are approximately 20-30 times more active in lymphoid cells than in non-lymphoid cells. To avoid the problem of differential mRNA stability upon transfection of immunoglobulin genes into non-lymphoid cells we have constructed chimeric genes. All kappa mRNA sequences were progressively deleted to fuse the kappa gene promoter to a globin gene coding body. A similar chimeric gene was constructed with the promoter of the lambda gene. The cell-type preference of the promoter may be exploited during B-lymphocyte differentiation to regulate the immunoglobulin gene promoter independently from the enhancer.     

Differences in human alpha-, beta- and delta-globin gene expression in monkey kidney cells

We have compared the function of the human alpha-, beta- and delta-globin genes using various plasmid expression vectors derived from pBR322. Amplification of recombinants occurred after their introduction, by calcium-phosphate-mediated DNA transfer, into monkey kidney cells that constitutively produce T antigen (COS cells). The human alpha-globin gene promoter functioned independently, but the beta-globin gene promoter was nearly totally dependent on the enhancing activity of the 72 bp direct repeats from the SV40 genome. Furthermore, when the human alpha- and beta-globin genes were linked in the same vector, the alpha promoter was active but the beta promoter was not. Function of the delta-globin gene promoter also depended on the enhancer element. In vectors containing the 72 bp repeats and the beta- or delta-globin gene, the activity of the beta-globin gene was approximately 50 times greater than that of the delta-globin gene, approximating the ratio of beta and delta mRNA observed in normal human bone marrow cells.     

Multiple instability-regulating sites in the 3'' untranslated region of the urokinase-type plasminogen activator mRNA.

In LLC-PK1 cells urokinase-type plasminogen activator (uPA) mRNA has a short half-life. It is stabilized by inhibition of protein synthesis and by downregulation of protein kinase C (PKC). In the present study on uPA mRNA metabolism, we focused our attention on the 3' untranslated region (3'UTR) of the uPA mRNA, as this region is long and highly conserved among several mammalian species, including mice and humans. To investigate the possible role of the 3'UTR of uPA mRNA in mRNA metabolism, we inserted this region into the 3'UTR of the rabbit beta-globin gene that is linked to the cytomegalovirus promoter and stably transfected it into LLC-PK1 cells. While the parental globin mRNA was stable, the chimeric mRNA was degraded as rapidly as endogenous uPA mRNA, suggesting that the 3'UTR of uPA mRNA contains most of the information required for its rapid turnover. Further analysis showed that there are at least three independent determinants of instability in the 3'UTR; one is an AU-rich sequence located immediately 3' of the poly(A) addition signal, and one is a sequence containing a stem structure. One determinant seems to require ongoing RNA synthesis for its activity. All chimeric unstable globin mRNAs became stable in the presence of cycloheximide, a protein synthesis inhibitor, suggesting that the stabilization of mRNA by protein synthesis inhibition is not through a specific sequence in the mRNA. In PKC-downregulated cells, globin mRNAs with the complete 3'UTR or the AU-rich sequence were stabilized, suggesting that PKC downregulation stabilizes uPA mRNA through the AU-rich sequence. Here we discuss the significance of multiple, independently acting instability determinants in the regulation of uPA mRNA metabolism.     

Expression of adenovirus type 12 E1A gene in monkey cells, using a simian virus 40 vector.

Simian virus 40 (SV40) recombinants carrying the adenovirus type 12 E1A gene were constructed. The SV40 expression vector was constructed by removing most of the VP1 gene and an internal part of the intervening sequence for late 16S RNA and by joining the 5' and 3' splice sites into a small segment. The adenovirus type 12 E1A gene with or without its own promoter was inserted downstream from the SV40 late promoter and the splicing junctions. The recombinant DNA was propagated and packaged in monkey cells by cotransfection with an early temperature-sensitive mutant (tsA58) DNA as helper. Immunofluorescent staining of the monkey cells infected with the resulting virus stocks showed that up to 20% of the cells overproduced the E1A gene products in the nuclei. Two-dimensional gel electrophoresis of the products indicated that the products were very similar or identical to the authentic polypeptides synthesized in adenovirus type 12-infected human embryo kidney cells. The E1A mRNA was initiated at the SV40 late promoter irrespective of the presence of the E1A promoter and terminated at either the E1A or the SV40 polyadenylation signal. These hybrid mRNAs were correctly spliced in the E1A coding region.     

SV40 recombinants carrying a functional RNA splice junction and polyadenylation site from the chromosomal mouse beta maj globin gene.

We have introduced a fragment of the chromosomal mouse beta major globin gene into SV40 and used the resultant hybrid virus to infect cultured monkey kidney cells. The mouse DNA fragment, which contains an intervening sequence and a poly(A) addition site, has been inserted in both possible orientations relative to the SV40 late region promoter. While the fragment is transcribed regardless of orientation, the RNA splice signal and poly(A) addition site are utilized only when the fragment is inserted in the "sense" orientation. Thus genomic mouse signals for both splicing and polyadenylation are recognized across species boundaries. Furthermore, since only an 18 bp segment was included on the 5' border of the intervening sequence, we can define a maximum length for this splice signal.     

High-level expression of the bovine growth hormone gene in heterologous mammalian cells

The gene coding for bovine growth hormone (bGH) was isolated from a lambda-phage library constructed using bovine pituitary DNA partially digested with MboI. Expression of this gene transfected into mouse and monkey cells was studied. CV-1 monkey cells transfected with simian virus 40 (SV40) vectors containing the intact bGH gene, including the putative promoter region, did not express bGH. However, replacement of the bGH promoter with the mouse metallothionein-I (MT) promoter resulted in high-level synthesis and secretion of bGH. These results show that the bGH promoter functions poorly in CV-1 cells but CV-1 cells process and translate the bGH mRNA accurately. The MT-bGH chimeric gene was used to establish permanent bGH-secreting mouse C127 cell lines using the 69% transforming fragment of bovine papilloma virus (BPV) as the vector. One such cell line produced high levels of bGH and secreted it into the medium efficiently. Secreted bGH is processed accurately and is bioactive as judged by its ability to bind to rabbit liver membrane preparations.