Daniel L. Simmons and Dr. Orlan Kenneth Macdonald, Chemistry and Biochemistry
In order to create a unique clone of chicken COX-2 it was necessary to insert a synthetic DNA tag into the full-length version of the COX-2 cDNA. First, a 21-nucleotide oligo and its complement were produced using a DNA synthesizer. The oligos were purified, phosphorylated and then annealed.
The annealed oligo tag was then inserted into a full length, unspliced chicken COX-2 cDNA already in a pBluescript® KS- plasmid. The full length COX-2 in KS- was prepared by restriction digest. The prepared plasmid and oligos were then ligated together.
The successful ligation of the insert into the COX-2/plasmid combination was identified through restriction digest. Dideoxy sequencing of the clone positively identified the ligation of the annealed oligos into the PmaC I site (bp 782) of COX-2.
For expression in RSV-transformed mammalian cells, it was necessary to transfer the newly synthesized, tagged COX-2 gene from the KS- plasmid to a plasmid with a mammalian transcription promoter, the pRc/RSV plasmid. The RSV plasmid was prepared by restriction digest. The tagged COX-2 clone was removed from the KS- plasmid by restriction digest and then ligated into the prepared RSV plasmid.
The presence and orientation of the tagged COX-2 insert in the plasmid was determined initially by restriction digest with the restriction enzyme Hind III, followed by several, multiple restriction digests. Seven constructs with the insert in the forward orientation with respect to the RSV plasmid were positively identified. Additionally, five constructs in the reverse orientation were found. The new construct was called pRCP+21, for plasmid RSV with the full length COX-2 clone, restricted at the PmaC I site, adding (+) a 21-nucleotide oligo.
Rous sarcoma infected-chicken embryo fibroblasts (RSV-CEF) were transfected with the pRCP+21 construct. After stimulation of the cells for nine hours, the RNA was isolated from the cells. The isolated RNA was reverse transcribed using AMV reverse transcriptase. Then, the RT product was used as a template in polymerase chain-reaction in order to determine the presence of spliced, plasmid-derived mRNA. The PCR process was designed to specifically amplify the RNA derived only from the plasmid that was transiently introduced into the cell.
The PCR product corresponding to the spliced COX-2 plasmid RNA was isolated, prepared and inserted into a cloning plasmid, KS-. The ligation product was restriction digested to ascertain the presence of the tagged COX-2 insert and then sequenced to determine the splicing specificity. Unfortunately, the sequence of the spliced clone has been difficult to determine due to problems in sequencing and remains, still, to be positively identified. Northern blot analysis of the spliced product, probing for the inserted tag, reveals that the inserted, tagged COX-2 plasmid is transcribed in the cell at less than one tenth the expression of the endogenous COX-2 gene.
The second phase of my project involved determining whether stable insertion of three different COX-2/luciferase chimera into the genome of non-chicken cells would confer splicing of chicken COX-2 intron-1. The three different chimera were 1) pLuc I1, the full length luciferase gene in which its first intron was replaced with intron-1 from chicken COX-2; 2) PCLH-4, a fusion gene in which the first four exons, including intron-1 of chicken COX-2 is fused to the full length luciferase gene. Removal of COX-2 intron-1 from the transcribed gene would result in quantifiable luciferase activity; 3) PCLH-5, a spliced form of the PCLH-4 gene, in which intron-1 is no longer present, therefore transcription of this gene automatically produces luciferase activity. This construct was used as the positive control. Additionally, lysates from stimulated RSV-CEF cells transfected with PCLH-4 and stimulated were used as a control.
The stable insertion of these distinct constructs was done using Lipofectin©. LA-24, or NIH 3T3 cells transformed by RSV, were the cells used because they can be regulated by the same stimulation methods as the RSV-CEF, and they are immortal. After transfection, the cells were selected for stable insertion of the plasmid into the host genome using Geneticin© (G-418). After selection with the drug, the surviving colonies were individually transferred to separate plates for culture. The many colonies were cultured and prepared for assay.
The assay for luciferase activity showed that the cells transfected with pLuc I1, upon stimulation, showed no increase of splicing from the non-stimulated cells. The overall activity was determined statistically to be negligible. Assay of the PCLH-4 cells, after stimulation, showed an increase in luciferase activity over the non-stimulated and non-transfected cells. However, the activity measured was not as marked as what we had expected. Lastly, the assay of the positive control, PCLH-5 cells, measured an increase in luciferase activity from the PCLH-4 cells by more than seventy fold.
The assay results showed that stable insertion of these different genes into the genome of nonchicken cells, resulting in chromatin association, does not confer splicing of intron-1. We arrived at this conclusion because, first, the measured activity of the control from the constructs in the native chicken cells was more than several thousand fold over the pLuc I1 and PCLH-4 cells. Second, the measured activity of the spliced, positive control construct in the LA-24 cells was over seventy fold greater than the activity of the unspliced constructs, suggesting that intron-1 was not removed accurately.
My work has yielded several results. First, I have constructed a COX-2 clone that yields plasmid derived mRNA that is easily identifiable and differentiated from endogenous COX-2 mRNA. This construct will be of further use as studies continue in determining which portion(s) of the COX-2 mRNA is(are) involved in the regulation of the expression of COX-2 in chicken. Second, I have determined, through stable insertion of several constructs into the genome of transformed nonchicken cells, that chromatin association does not confer splicing of chicken COX-2 intron-1. Additionally, this verifies that all of the elements within the COX-2 gene necessary for accurate removal of intron-1 are not completely contained within the intron itself nor within the first four exons of the COX-2 gene, and that any trans elements necessary for accurate splicing are not functional in the LA-24 cells.