1"" 411-244. F: 2 '1! .f. ‘ =~ Lt", .. a . 3:. x3....a..“..fr 5. é... a! . lawn.) ., ., ,3: {.2 ,4 mr ivy? r grain, ‘ 3&3; J. 2...? .. 3 i ... .7 s, Tr. K: likawnr. Y... ..,.,...n.. . £35.58!" axnnnxm. .6 V 1. brflh.i..§.9 finflkvhkfi . . I... ... )r. .31.: . .. r... .332}; i’\ - mafinufift. .3 .. PH Liar... fin}. .1» .,3 . :3 ..! flaykvtawbm. ,1 v a 9\t . . 5.. 4 I. at: i .l - Jim‘s _. ... . um I. i....:.p.,.m.‘.,: .2 E. . , s . fimfingwge .3m...=.w f . ‘ , . . . v 3;. , , 1‘7”", w lg. i ,fimfiwfifi ‘ , . _ fixwfikfii . £%%.$rw.§ MICHIGAN ST T UNWER l I l’llllll’ll/ll x " ” “‘ ‘3 ll ll ill Ill ll ll THESIS 3 1293 01046 0529 This is to certify that the thesis entitled Inhibition of Fatty Acid Synthase by an Antisense Message in TAl Cells presented by Michelle Kay Mater has been accepted towards fulfillment of the requirements for Masters Animal Science degree in 7 ‘2’ /" 2) ,.. Major professor U DaugNovember 15, 1994 0-7639 MSU is an Affirmative Action/Equal Opportunity Institution LIBRARY l Mlchlgan State Unlverslty w PLACE It RETURN BOXtonmmfiIbclI-okoufmnmm TO AVOID FINES rotunonorbdmddodm. DATE DUE DATE DUE DATE DUE T—l—T MSU loMAfimnflvkom/EM Opporbnltylmwon INHIBITION OF FATTY ACID SYNTHASE BY AN ANTISENSE MESSAGE IN TAl CELLS BY Michelle Kay Mater A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Animal Science 1994 ABSTRACT INHIBITION OF FATTY ACID SYNTHASE BY AN ANTISENSE MESSAGE IN TAl CELLS BY Michelle Kay Mater Potential attenuation of lipogenesis by fatty acid synthase (FAS) antisense messenger RNA in adipocytes was examined. A portion of rat FAS message in the sense and antisense orientation was ligated into the selective expression vector, pcDNA3. These antisense and sense plasmids, as well as two controls (pcDNA3 and pcDNAB/CAT), were transfected into TAl cells using lipofectin, and stable polyclonal and monoclonal cell lines were established. Fatty acid synthase activity was measured in each stable cell line to determine whether FAS activity had occurred. Results indicated that the antisense polyclonal cells had lower PAS activity (P<.05) while the monoclonal antisense, sense and CAT cells all had lowered EAS activity (P<.05). Site of integration of the sense, antisense and control plasmids into the TAl genome was not determined. Integration of sense and control plasmids into genomic sites involved in lipogenesis may explain attenuation of EAS in monoclonal cell lines- Dedication my inspiration, Cindy iii Acknowledgments I would like to thank my major professor, Dr. Bergen, for all his help and advise throughout this project. His support is much appreciated. I would also like to thank Dr. Emery for his support and thoughtfulness. Thank-you to Dr. Helferich, who provided me with the opportunity to learn new techniques in his laboratory needed in this project. Thanks also to Dr. Ritchie, who served on my committee. I value all the time and guidance given to me by my committee. Thank—you. I would also like to sincerely thank Dr. Patty Weber. Patty has helped me with an untold number enzyme assays and cell culture techniques. Her patience, suggestions and support in all my lab work and writing have meant a great deal to me. Thanks for everything, Patty. Sharon DeBar also been a tremendous source of assistance throughout this project-thanks Sharon! Thank-you to Scott Kramer for his friendship and help. Thanks also to Daniel, Josep, Kim, Rita, Ron, and Shan. Most importantly, thank you to my closest friend and biggest cheerleader, Cindy. Thanks for all that listening, proof-reading and support for the last two years. Thanks to my Mom and Dad for all their love and support throughout the years as well as to all my brothers and sisters. These people have truly made this all possible. iv TABLE OF CONTENTS List of Tables List of Figures Introduction Antisense Literature Review Mechanisms of Antisense Actions Endogenous antisense in procaryotes Endogenous antisense in eucaryotes Early use of exogenous antisense Successful use of endogenous antisense Using exogenous antisense messages Problems with exogenous antisense messages Variations of results when using antisense messages Conclusions Fatty Acid Synthase Literature Review Overall reaction Origin and structure of FAS Gene structure of PAS Regulation of FAS PAS in adipocyte cell culture systems Conclusions Hypothesis and Objective Aim #1: Synthesis of Plasmids 1. Methods 2. Results and Discussion Aim #2: Establishment of Cell Lines 1. Methods 2. Results and Discussion Aim #3: Determination of FAS Inhibition 1. Methods 2. Results and Discussion Overall Conclusions vii viii 31 31 32 34 35 36 37 38 39 39 42 47 47 50 55 55 56 68 Appendix A 70 Appendix B 83 Appendix C 98 Appendix D 104 Bibliography 107 vi LIST OF TABLES Table Title Page Table 1. Predicted Lengths of Restriction Fragments 44 vii LIST OF FIGURES FIGURE TITLE PAGE Figure l .Antisense Action 4 Figure 2 .Action of micF RNA 7 Figure 3 Diagram of pcDNA3 40 Figure 4 Diagram of pCAT 40 Figure 5 Diagram of pASFASl 41 Figure 6 Diagram of pSFASl 41 Figure 7 Orientation of plasmids 43 Figure 8 Sequence Chromatogram of pSFASl 45 Figure 9 FASl Sequence 46 Figure 10 Secondary Structure 48 Figure 11 CAT Protein in Experiment #1 54 Figure 12 CAT Protein in Experiment #2 54 Figure 13 AP2 Cell Line 57 Figure 14 CP2 Cell Line 57 Figure 15 SP2 Cell Line 58 Figure 16 3P2 Cell Line 58 Figure 17 AMl Cell Line 59 Figure 18 CMl Cell Line 59 Figure 19 8M2 Cell Line 60 Figure 20 3M1 Cell Line 60 Figure 21 Nontransfected Cell Line (TA1) 61 Figure 22 FAS Activity in Polyclonal Cells 62 Figure 23 FAS Activity in Monoclonal Cells 63 Figure D1 FAS Activity - April 21 Assay 104 Figure D2 FAS Activity - May 9 Assay 104 Figure D3 FAS Activity - May 16 Assay 105 Figure D4 FAS Activity - June 3 Assay 105 Figure D5 FAS Activity - June 10 Assay 106 Figure D6 FAS Activity — June 17 Assay 106 viii INTRODUCTION The meat animal industry continues to strive toward producing a leaner product. Although nutrition, health and breeding programs have increased lean gains, fat gains have not decreased (Bergen and Merkel, 1991). Some attempts to decrease fat accumulation in swine have been successful, including feeding beta-agonists and injecting growth hormones (Mersmann, 1991). However, these methods are not yet approved for public use and involve exogenous agents. A more specific approach in lowering fat deposition needs to be created to meet consumer demands for a leaner, low fat product. One method of specifically inhibiting fat synthesis is to block a key enzyme in the fatty acid synthesis pathway. Another technique to block a specific gene product involves the use of an antisense message (antisense mRNA). Antisense mRNA is a specific complementary message against its normal cellular mRNA (sense mRNA). Due to this complementarity, it is thought that the mRNAs hybridize and prevent translation of the protein. The hybrid is then degraded to nucleic acids leaving no unnatural by-products in the cell (Izant and Weintraub, 1985). This thesis project develops a system to determine if antisense fatty acid synthase (FAS) mRNA can inhibit fat synthesis in the adipogenic cell line, TAl. Three specific 2 aims were undertaken:create an antisense plasmid, establish antisense stable cell lines and determine if the FAS enzyme was inhibited. To this end, a vector containing antisense FASl (a 3' portion of the rat FAS mRNA from Witowski et al, 1987) as well as control vectors were stably transfected into TAl cells. Following transfection, monoclonal and polyclonal cell lines were selected with the antibiotic geneticin. inhibition of fat synthesis was measured by FAS enzyme assays in each of the cell lines. The system developed will enable further investigation of antisense mechanisms and viability of its uses to inhibit lipogenic enzymes at the cellular level. Antisense Literature Review Understanding gene expression and its regulation is the goal of many scientists today. The cell has numerous mechanisms to regulate these processes, many of which are only starting to be comprehended. The fact that DNA is transcribed to RNA, which is translated to protein is easily accepted but when, where and how much transcription and translation occurs is a complex process. Antisense messages, whether RNA or DNA, are examples of how cells can control these processes. Transcription of DNA occurs in the 5' to 3' direction. Polymerases read the 3' to 5' antisense strand of DNA and 5' to 3' sense RNA is synthesized. Ribosomes initiate translation by binding to an AUG site present in mRNA. This AUG site is generally near the 5' end of the sense RNA. It is understood that most cellular mRNA is sense mRNA. The term "antisense" therefore refers to the directionality of a nucleic acid sequence. The sequence of the sense strand of DNA is the 5' to 3' strand and is usually the only sequence reported in the literature. It is also the sequence of the RNA transcribed from that DNA sequence, i.e. the sense RNA. The antisense strand of DNA is the template for sense RNA transcription. If RNA were synthesized from the sense strand, the RNA produced would then be antisense RNA. The two possible strands of RNA synthesized from a DNA template are exactly complementary to each other. This complementarity between sense and antisense nucleic acids theoretically could be used to block sense RNA translation or DNA transcription, thereby controlling gene expression (Figure 1). DNA 5' — 3' Plasmid or other source 3' ' J 5' \l’ J, Transcription mRNA Antisense mRNA 5' _3' 3' '— J5. i Translation \ J, 3 RNA hybrid . ' r . 5' Protein CQ 5' _ 3' Translation blocked Figure 1. Antisense Action. Double stranded DNA contains both a sense and antisense strand but transcription results in only sense mRNA. The presence of antisense RNA could disrupt the translation process by preventing ribosomal binding, translocation of sense RNA into cytoplasm or inhibition splicing. Antisense technology encompasses a wide scope of techniques. Procaryotes and eucaryotes alike employ the use of antisense messages to block either transcription or translation by the hybridization of sense and antisense messages. Researchers have used antisense methods to create mutants and cure diseases like the flu and AIDS, as well as 5 develop new plants such as the Flavr Savr tomato and treat cancer patients. Mechanisms of Antisense Actions In many systems, it is unknown how the antisense can inhibit a gene product. In theory, the action of antisense is based on the complementarity of sense and antisense nucleic acids. Eguchi et al (1991) defines antisense RNA as "an RNA that interferes with the activity of another RNA by binding to a complementary region of a target RNA and affecting its function". Antisense messages are specifically complementary to its sense target. This complementarity would allow the strands to hybridize in vivo. The hybrid would be unable to be transcribed (in rare cases of DNA:RNA hybridization) or be transported out of the nucleus and/or be translated into protein. This simple but elegant idea would not only be highly specific to one gene, but could also be transcriptionally regulated itself (Izant and Weintraub, 1985). Endogenous Antisense in Procaryotes Natural occurrences of antisense messages have been detected in procaryotics and to a lesser extent in eucaryotics. After discussing examples of each of these, 6 the remainder of this section will review the variety of uses and methods of applying antisense technology. A well studied example of a naturally occurring antisense mRNA used to inhibit a procaryotic gene product is in E. coli. ColEl, a common plasmid present in E. coli, has its number of copies per bacterium regulated to twenty to thirty by antisense RNA. To begin replication of ColEl, an RNA transcript is synthesized from a point 555 base pairs upstream of the origin and which has been designated RNA II. At the origin, RNase H cleaves this RNA II to give a 3'—OH end to be used as a primer for DNA replication. An antisense transcript, RNA I, is synthesized from the complementary strand of the DNA used to make RNA II. RNA I base pairs to the 5' end of RNA II, inactivating it. Since it is known that the primer can still be initiated and elongated at the 3' end, the antisense RNA I is thought to somehow prevent the RNase H cleavage and creation of the 3'- OH end. RNA I and RNA II have similar secondary structures predicting that they base pair in 3 loops, creating a conformational change at the cleavage site. This binding of RNA I to RNA II is sometimes referred to as a "kissing structure". RNA I therefore acts as a transcriptional regulator of ColEl copy number (Lewin, 1990; Tomizawa 1981 and 1982). Another example of a procaryotic antisense gene regulator is the micF gene (Figure 2). When osmolarity is increased in a bacterium's environment, the env2 gene is .. '. .i .A. rw'rWWM-‘WWW 529$”)? { J"\-"" "" renew-“2;; i Em- “wk—'3’“ m ' Figure 2. Action of micF RNA. Binding of micF RNA to OmpF inhibits the translation of OmpF protein. The lower portion of the diagram depicts the structure of the sense-antisense hybrid. activated. This in turn activates ompR which increases levels of on micF and ompC. OmpF, a constitutive gene, is inhibited by micF. The ompF mRNA has a complementary region to the 174 base micF mRNA. This region in the ompF contains a ribosomal binding site. This suggests that the micF mRNA inhibits translation of ompF, or that the duplex formed is unstable and degraded by ribonucleases. OmpF and ompC proteins levels are known to be inversely related in the bacteria. This relationship of protein levels implies that regulation of one is tied to the other (Coleman, 1984: Mizuno, 1984). In addition to the micF and ColEl antisense systems, other antisense examples exist in procaryotics. For instance, the protein transposase is regulated by an antisense mRNA in the transposon TnlO inhibiting translation of transposase and therefore inhibiting transposition (Simons, 1983; Eguchi et al, 1991). The crp gene in E. coli has a complementary region with an antisense transcript synthesized from the opposite strand. This antisense transcript inhibits transcription of crp by an action thought to be similar to rho-independent termination. Lambda and p22 phage as well as replication of ColE2, IncF, IncI, R6K and pT181 plasmids are also thought to be regulated by antisense transcripts (Green, 1986; Eguchi et al, 1991). Procaryotic organisms use antisense messages as transcriptional and translational regulators. These 9 examples have been used as models for using exogenous antisense in eucaryotic cells. Endogenous antisense messages in eucaryotics would be most desirable as references for using exogenous antisense in eucaryotics, but procaryotic antisense examples are both more numerous and better understood. Endogenous Antisense in Eucaryotes The existence of eucaryotic antisense was predicted by Green et al (1986) based on the action of spliceosomes. Spliceosomes are comprised of snRNPs and protein. Splicing of mRNA requires the U1 RNA of U1 snRNP (small nuclear ribonucleoprotein) to bind to a complementary region at the intron-exon junction. If this complementary region is removed splicing can not occur (Green et a1, 1986). Other occurrences of complementary genes appeared with the discovery of Alu sequences, studies of the c-myc gene, human e-globin and myosin heavy chain genes and sequencing of the erbA homologs. An endogenous antisense messages to c-myc exists in eucaryotes. Early researchers had already inhibited the c- myc gene with exogenous antisense oligonucleotides before the endogenous message was detected. Cooney et a1 (1988) synthesized a twenty-seven base antisense oligodeoxyribonucleotide that was complementary to the -115 region of the c-myc gene. The oligonucleotide bound to the 10 DNA duplex to form a triple helix and repressed c-myc transcription in vitro. Postel et al (1991) continued this work by repeating the experiment in vivo with HeLa cells. Celano et al (1992) determined that an endogenous antisense transcript was produced when the human colon cancer cell (COLO 320) line was deprived of polyamines. The antisense transcript was shown to be highly homologous to the second exon of the c-myc gene and is thought to originate from the opposite strand in an intron of the c-myc gene. Interestingly, it was also shown that this antisense mRNA is homologous to regions in other genes such as N-myc, thymidine kinase and p53 suggesting that this antisense message may regulate other gene products as well (Celano et al, 1992). McCarthy et al (1983) found that an endogenous antisense transcript existed in chick embryonic muscle. Translation of the chicken myosin heavy chain (MHC) mRNA is controlled by a 102 antisense nucleotide transcript. It binds specifically in vitro to MHC mRNA and blocks translation. McCarthy et a1 (1983) proposed that the area of binding of the two transcripts occurred in the poly(A) region although a sequence at the 5' end of the mRNA was also possible. The origin of the antisense transcript has not been identified. Although a few other examples of natural antisense messages in eucaryotics exist, the mechanism of these endogenous antisenses are not yet known. One example is ll Rev-ErbAa protein. Although this protein is a member of the thyroid/steroid hormone receptor superfamily, it does not form a dimer with retinoic acid or thyroid hormone. When the DNA sequence of Rev-ErbAa was characterized, it was found to contain a part of the C-erbAa gene. C—erbAa produces both r-erbAa-l and r-erbAa-2 depending on how the mRNA is spliced. R-erbAa—Z has an antisense region complementary to Rev-ErbAa. Some tissues contain both the antisense and sense mRNAs suggesting that perhaps translation of one or both of the transcripts could be inhibited (Lazar et al, 1989). Miyajima et a1 (1989) determined that two erbA homologs ear-1 and ear-7 were synthesized from overlapping exons. Ear-7 was further processed to ear-71 and ear-72. Ear-71 is complementary to c—erbA (the T3 receptor). Both the c-erbA and ear-71 mRNAs are made into similar functioning proteins. It is unknown if their expression is somehow regulated using this complementarity (Miyajima et al, 1989). Wu et al (1990) found that human e-globin gene contained alternative transcripts originating from within an Alu repetitive sequence. These transcripts were transcribed in opposite directions, located only in the nuclei, not polyadenylated and found in mature K562 cells or embryonic red blood cells. Wu et a1 (1990) used a transient expression experiment to determine whether e-globin could be down regulated using this antisense transcript and went on to suggest that the antisense transcript bound to the sense 12 mRNA as it was transcribing which somehow blocked the polymerase II from continuing its transcription. Other antisense transcriptions originating within Alu sequences were not unknown at the time. Adeniyi-Jones et al (1985) detected an antisense message to the alpha-fetoprotein, 210 nucleotides in length, perhaps being used to inhibit this protein either transcriptionally or translationally. While all of these antisense transcripts have been detected, the mechanism of action of each has yet to be determined. Further antisense transcripts are likely to be detected as more sequencing is completed. Even as more antisense messages are found, the major focus of this type of research is using exogenous antisense rather than determining the purpose of the endogenous messages. Early Use of Exogenous Antisense Paterson et al (1977) first inhibited translation of rabbit B-globin by hybridization of antisense mRNA with its complementary cDNA. Though it was referred to as "hybrid arrested cell free translation", it is similar to the antisense method. Paterson et al (1977) used linearized adenovirus containing the rabbit B-globin gene in a cell free system with isolated rabbit globin mRNA. The nucleic acids were denatured, combined with formamide (to prevent DNAzDNA rehybridization) and incubated to allow for hybridization. Upon completion of the translation assay, it 13 was determined that B-globin mRNA was not translated in samples where hybridization could have occurred. It was also determined that heat denaturing the hybrid reinstated the ability to translate B-globin. Though Paterson et al (1977) used this method to locate the translatable sections of the viral genome, they went on to suggest that this work could be used to study organization and expression of RNA viruses. Although Paterson et a1 (1977) may have initiated the use of exogenous antisense, the bridge between endogenous antisense to using exogenous antisense messages to manipulate gene function was built by Izant and Weintraub (1984). Their work resulted in several important discoveries about antisense specificity, target areas and methods of incorporation. These first experiments also fostered further research using antisense to attempt everything from inhibiting genes to treating cancer and AIDS . The first experiments by Izant and Weintraub (1984 and 1985) determined the specificity of the antisense message. Using two plasmids (one containing an antisense fragment of the herpes simplex virus thymidine kinase gene (TK) and the other with the sense TK gene), they cotransfected the plasmids into LTK- cells (these cells produce no thymidine kinase). Not only did the antisense transcript block TK, it did so specifically. Further experimental results showed that antisense chicken TK inhibited only chicken TK and not 14 herpes TK. When the antisense fragment was put in front of the inducible promoter murine mammary tumor virus (MMTV) and the cells activated by dexamethasone, the antisense inhibited the TK but uninduced cells did not. Using a cell line that produced endogenous TK, the antisense plasmid lowered endogenous TK activity six-fold. Izant and Weintraub (1985) also measured how much of the TK gene was needed to inhibit its target. Experiments with the TK gene showed that only a 52-base fragment from the untranslated 5' end of the gene was needed for the antisense transcript to inhibit TK. It did not contain the AUG start site. This work was repeated with antisense chloramphenicol acetyl transferase (CAT) expression vectors. CAT activity was lower in cells containing the antisense plasmids as compared to cells with other control plasmids. B-actin was also inhibited with an antisense transcript. Using pBR322 with antisense B-actin, transfected cells had reduced growth as a result of less actin production. Upon staining for actin filaments, only half of the antisense treated cells showed microfilaments. While an antisense fragment from the 3' or the 5' end of the CAT mRNA could block CAT, only the 5' ntisense fragment from B-actin would inhibit it (Izant and Weintraub, 1985). In conjunction with the above experiments, Izant and Weintraub (1985) used a variety of techniques to incorporate the antisense messages into the cells. They used microinjection, the DEAE transfection method for transient 15 expression and calcium phosphate transfection for selection of stable polyclonal cell lines. Each of these methods were successful. They determined that a ratio of 5:1 antisense to sense DNA was needed to block TK when transfection was used but 50:1 was needed when the antisense transcript was microinjected. Izant and Weintraub (1985) summarized this early antisense work with what has become general guidelines for using antisense messages to inhibit a gene product. They observed that a 5' piece may be more efficient in inhibiting genes (which is supported by the finding that procaryotes use 5' antisense fragments). It was noted that the antisense fragment did not necessarily have to include the translation start site, AUG. Although Izant and Weintraub (1985) did not determine the stability of antisense transcripts in vivo, the intracellular site where the hybrids were located, or if different domains of RNA were more susceptible to antisense hybridization, they nevertheless proposed this is a very powerful technique for gene analysis. Successful Use of Endogenous Antisense Many genes have been successfully blocked with an antisense message. One common use is the creation of mutants in order to analyze cell function of a particular gene product. Florini and Ewton (1990) did this with l6 myogenin to determine if differentiation still occurred in the presence of IGF-I. Using an oligodeoxynucleotide against the first twenty-five nucleotides of the myogenin gene, Florini and Ewton (1990) blocked myogenin from being translated. Stimulation of antisense treated cells with IGF-l resulted in no differentiation while mismatched oligomers (same bases as antisense oligomer but scrambled order) and control treated cells had normal differentiation. This effect was only seen when the antisense message was added at the same time as the IGF-l. Other cellular processes including proliferation were not changed in the treated cells. Using radioactive end-labelled oligomers, Florini and Ewton (1990) determined that incorporation of the antisense oligomers into the cells in culture continued slowly for seventy-two hours after addition. Thus, this antisense oligomer was highly specific and did not cause other cellular actions to be inhibited. Using antisense to create false "mutant" cells or animals has become an excellent way to study gene function. Antisense can also be used as a therapeutic tool against many diseases. Simons et a1 (1992) used antisense oligomers against the c-myb gene to inhibit smooth muscle cell (SMC) growth. Proliferation of smooth muscle cells are implicated in the cause of atherogenesis, failure of bypass grafts and stenosis establishment after artery angioplasty. Using rats with injury to the carotid artery, Simons et a1 (1992) injected antisense c-myb oligomers into the artery 17 after a balloon angioplasty. The 1 mg/ml antisense solution (200 pl total) gelled to the artery on contact. The antisense treatment resulted in reduced SMC growth as compared to sense, mismatched or gel treated. The reduction in growth occurred only in the area where the gel was placed; surrounding areas accumulated SMC. This technique may be a treatment for heart surgery patients in the future. Antisense messages have also been used to treat cancer. Han et al (1991) used transgenic mice containing the antisense gene to the retroviral packaging sequence of the Moloney murine leukemia virus. Upon birth, mice were infected with the virus. All control mice got leukemia symptoms but antisense transgenic mice did not. While Han et al (1991) did inhibit the leukemia virus, no mechanism for inhibiting the virus was established. Virus could not be detected in supernatants of stable cell lines containing the antisense gene, implying the packaging of the virus must have been blocked as expected. Plant and animal agriculture have benefitted from antisense as well. The Flavr Savr tomato just marketed contains an antisense gene to polygalacturonase. This gene is responsible for the softening of the tomato as it ripens. Smith et al (1988) originally reported the inhibition of softening by transforming tomatoes with a 730 bp fragment (containing the start site for translation). An plasmid containing an antisense ethylene gene was also transformed into tomatoes resulting in lower ethylene production 18 (Hamilton et al, 1990). Ethylene, thought to be involved in ripening, was reduced in the transgenic tomatoes. These tomatoes were less likely to become overripe or shrivel than nontransgenic tomatoes. Though an effect was seen, antisense RNA could not be detected. Hamilton et al (1990) explained this as a common occurrence in antisense systems and perhaps due to immediate degradation of the RNA hybrid or the antisense RNA inhibiting transcription. Hoffman (1993) reported on work done by Wong and Halawani using antisense prolactin. Prolactin is known to cause broodiness in turkeys which reduces further egg laying. .After antisense prolactin was injected into pituitary cells from turkeys, prolactin levels are lowered. Further work is currently being conducted to create transgenic turkeys carrying this antisense gene. It is hoped that these turkeys will produce more eggs for the producer. Using Exogenous Antisense Messages Procaryotes and eucaryotes have provided a variety of examples of antisense messages regulating a gene product. Though the regulation of antisense is not well understood, many researchers have used these few examples to create exogenous antisense messages to interfere with a variety of genes. Exogenous antisense messages have become a powerful tool to create mutants and interfere with cellular processes 19 as well as potential therapeutic agents against a variety of diseases. Three important points must be considered when using exogenous antisense messages to block a gene product. First, the target gene and its sequence for the antisense message must be known to create the antisense message. A variety of antisense messages can be used:DNA oligonucleotides, RNA oligonucleotides or plasmids containing a portion or the complete gene in the antisense orientation under the control of a promoter. This may depend on the cell, tissue or organism the antisense is used in. Finally, a method of incorporating the antisense message into the desired location must be ascertained. The sense targets for hybridization include every part of a gene or mRNA. Antisense messages are targeted against the initiation site for translation, 5' end, 3' end, exons, introns, splice sites or the entire gene as well as combinations of the above. Moroni et al (1991) used plasmids containing either the 5' end, 3' end or the entire EGF receptor and found that each was effective at inhibiting translation of EGF receptor mRNA. Kim and Wold (1985) determined that both the 3' untranslated region as well as a 5' antisense fragment of thymidine kinase blocked TK production. DNA is also a target for antisense, creating a triple helix formation which prevents transcription (Cooney et al, 1988). Different areas in different genes are more susceptible to antisense messages but vary within each 20 system it is used in. Izant and Weintraub (1985) found that only 52 bases of the 5' end of thymidine kinase (TK) was needed to block the TK levels while a portion of the 5' or 3' end of the chloramphenicol acetyl transferase (CAT) gene inhibits this product. It is generally considered advantageous to use the 5' portion of the gene, one which contains the AUG translational start site, as the target for antisense (Izant and Weintraub, 1985; Kim and Wold, 1985). However, Denhardt (1992) suggests targeting several different areas of a gene and comparing the results of each. The types of antisense created by researchers include a variety of structures and modifications of those structures. These include oligodeoxynucleotides, oligonucleotides and transfected plasmids or viruses containing antisense genes. Oligodeoxynucleotides are short strands of DNA while oligonucleotides are RNA strands (both are often called oligomers). Both of these short fragments have been synthesized with modified bases in attempts to increase stability and efficacy in the cell. For example, replacing an oxygen with a sulfur on the phosphate in the backbone of the nucleic acid (phosphorothioates) has been shown to increase the stability of the oligonucleotide in the cell (Akhtar et al, 1991). The length of the oligomer must be long enough to be specific while oligomers that are too lengthy are difficult to make and have reduced efficiency. An oligomer fifteen nucleotides long would occur naturally only once in a sequence 500 million base pairs in length 21 (Marcus-Sekura, 1988). Oligonucleotides have the advantage of being commercially available but the quantities needed for an experiment can be quite costly. Vectors containing an antisense gene are another type of antisense commonly used. Plasmids with an inserted gene or partial gene in the antisense orientation have been transfected transiently or stably into cells where they produce antisense mRNA. The length of the antisense transcript produced from the plasmid or virus ranges from as few as fifty bases to the entire length of a gene. Using a vector containing the antisense gene necessitates the use of a promoter. Strong promoters such as the cytomegalovirus (Han et al, 1991) or inducible promoters like the mouse mammary tumor virus (Izant and Weintraub, 1985) are some examples. Stable transfection of a plasmid with an antisense gene or gene fragment has an advantage over oligomers in vitro. Whether the vector uses an inducible or constitutive promoter, the antisense message is always present and more can be produced, unlike oligomers which have only a limited lifetime and must be continuously added to the system. The delivery method of antisense transcripts depends on the target and the researchers as well as the type of antisense used. Oligonucleotides have been placed in media above cells to be incorporated naturally as well as microinjected into tumor cells (Moroni et a1, 1992) or pumped subcutaneously into mice (Ratajczak et al 1992). Transgenic mice (Han et al, 1991; Pepin et a1, 1992; Richard 22 et al, 1993) as well as transgenic plants like tomatoes (Smith et al 1988) and flowers (Moffat, 1991) have been created with the incorporation of antisense genes. Stable or transient cell lines are made by transfecting a plasmid containing the antisense gene. Though calcium phosphate precipitation and protoplast fusion are commonly used for transfection in animals and plants respectively, many different methods are used including lipofection (Yeoman et al, 1992), DEAE-dextran (Han et al, 1991) and electroporation (Kaiser et al, 1992). After transfection with a plasmid containing a resistance gene such as neomycin, selective medium is utilized to obtain the transgenic cells line(s). Lin and Lane (1992) used electroporation to stably transfect 3T3-Ll cells with an plasmid containing antisense CCAAT/enhancer-binding protein. This plasmid also had the neomycin resistance gene. After transfection monoclonal cell lines were selected with geneticin (cells transfected with plasmids containing the neomycin gene are resistant to the antibiotic geneticin). Although many researchers isolate monoclonal cell lines, Kim and Wold (1985), Kaiser et al (1992), Wu et al (1992) and Izant and Weintraub (1985) used polyclonal cell lines. Direct microinjection of in vitro transcribed RNA has also been used in oocytes (Rosenburg et al, 1985; Melton et al, 1985, Fire et al, 1991) and embryos (Bevilacqua et al, 1988). I 23 Problems with Exogenous Antisense Messages Izant and Weintraub (1985) seemed to suggest that any gene product could be target for antisense messages and therefore be inhibited. Unfortunately, antisense inhibition is not as easy as Izant and Weintraub (1985) proposed it could be. Stability of the antisense transcripts, proper levels to use and detecting the messages have been causes of concern. While antisense works in some systems very well, in others it does not or requires high amounts of the antisense message. One of the difficulties of using antisense messages is the stability of transcripts. Oligonucleotides modified with a sulfur (phosphorthioates) or methyl groups (methyl phosphonates) or alternating methyl/phosphodiester groups have all been determined to be more stable than unmodified oligonucleotides (Akhtar et al, 1991). Krieg et al (1993) added a cholesteryl moiety to the oligonucleotide and found that these associated to the lipid bilayer via low density lipoproteins and increased efficiency of the antisense message. The presence of serum also destabilizes oligonucleotides. Calf serum in particular was the most destabilizing serum type when compared to human serum, nuclear extract and cytoplasmic extract (Akhtar et al, 1991). Antisense RNA, transcribed in vitro then injected, also has problems with stability. Bevilacqua et a1 (1988) 24 determined that when twenty picograms of antisense B- glucuronidase RNA (more than 20 picograms killed the embryos) was injected into embryos, only 20% of it remained after 36 hours. The sense-antisense hybrid was also made in vitro then injected into cells to measure the hybrid stability. It was determined that the hybrid was stable for only five hours. Even with such a short half life, they were able to block B-glucuronidase in the embryos. How much antisense message to use is different in every system. Izant and Weintraub (1985) reported that up to a fifty fold excess of antisense was required to yield a twenty fold inhibition of genes in xenopus oocytes but only a one to one ratio was needed to block B-galactosidase transcription. The required amount also varied depending on the method of introduction. Only a five fold excess of antisense thymidine kinase was required to block the gene in transfected cells but a fifty fold was needed in microinjected cells (Izant and Weintraub, 1985). Huge amounts of antisense oligonucleotides are often needed to get the effect desired. For example, Lemaitre et a1 (1987) used 100 nM of antisense oligomers to vesicular stomatitis virus N-protein to inhibit the virus while Offensperger et al (1993) used 1.5 uM antisense oligomers to duck hepatitis B virus to inhibit viral replication. Wang et al (1992) used 30 uM antisense alpha subunit to a G-protein to reduce its expression. 25 Secondary structure of the antisense messages may play a role in their stability. The secondary structure of the RNA 1:RNA II hybrid has been determined in the ColEI system (Lewin, 1990). The binding structure is comprised of three stem loops and is thought to aid in the hybridization of the remainder of the molecules. Eguchi et al (1991) discusses the presence of a protein, Rom, that stabilizes the hybrid by lowering the dissociation rate by a factor of one hundred. It is thought that Rom recognizes the structure of the hybrid, not the sequences involved. Other proteins in the cellular environment may affect stability in other systems as well. The sequences within the RNA hybrids may also affect stability of each RNA as well as the hybrid. While computer programs can predict secondary structure and the energy of the molecule, it is difficult to use these methods on large antisense molecules. Though mutating one base pair in RNA II of ColEI results in the same predicted secondary structure as the nonmutant, the mutant has a dramatic loss in function and a different susceptibility to RNases. To complicate things even further, tertiary and other structures cannot be accounted for in computer programs (Eguchi et al, 1991). Locating antisense transcripts in target cells has been a further challenge. Northern blots do not always reveal the antisense message which has resulted in the use of RNase protection assays. Celano et a1 (1992) used a protection assay to detect antisense c-myc synthesized in COLO 320 26 cells. Fernandez et al (1993) and Sklar et a1 (1991) also used nuclease protection assays to detect antisense messages. Fire at al (1991) used both a nuclease protection assay and PCR to amplify messages in order to detect antisense RNA. Pepin et a1 (1992) used PCR to amplify mRNA messages from transgenic mice as they were unable to find the antisense message in a Northern blot. Khokha et al (1989) were unable to detect any antisense mRNA in three out of four of their monoclonal cells lines by Northern blot analyses. Kim and Wold (1985) originally were unable to detect any appreciable level of antisense thymidine kinase message. They suggested that perhaps the antisense mRNA was being processed or transported incorrectly. After changing the promoter in their vector and the selection method, antisense RNA was detected with a protection assay. Although this implies antisense messages are hard to detect, Han et al (1991), Hamilton et a1 (1990) and Kaiser et a1 (1992) were able to detect antisense mRNA using either total RNA or mRNA in Northern blots with riboprobes. Finding antisense oligonucleotides in vivo is rarely attempted since these molecules are so small. Grigoriev et a1 (1993) UV crosslinked oligodexoynucleotides in vitro to the interleukin 2 receptor DNA to show a triple helix formed and prevented transcription. Holt et a1 (1988) end labelled oligonucleotides and using a 81 nuclease protection assay, showed that a hybridization did form between the antisense oligonucleotide and c-myc mRNA in vivo. 27 Once again, these differences in antisense systems have made it difficult to provide a single model for using antisense. Methods of detecting and measuring stability of antisense messages are sometimes not specific enough to determine how the antisense works. Even so, these problems have not halted the popularity of this relatively new technique to control gene expression. Variations of results when using Antisense Messages Antisense messages have not always worked as expected or not worked at all. The controls can also cause some inhibition as shown in some of the following examples. Leiter et al (1990) is a good example of confusing antisense results. Leiter et al (1990) tried to inhibit the flu virus (strain A.and C) with an antisense phosphorothioate oligomer. As controls, mismatched oligonucleotides were used as well as unmodified (phosphodiester) antisense and mismatched oligonucleotides. The phosphodiester oligomers did not inhibit either the C or A flu virus even at concentrations of 80 uM. Both C and A viruses were inhibited with phosphorothioate oligomers but at 20 uM and 1.25 uM concentrations respectively. Virus A was even inhibited with the mismatched sequences. These results show that antisense is not the same in each case, even when similar genes are targeted. 28 Ratajczak et al (1992) used an antisense phos- phorothioate oligomer to inhibit Cemyb. C-myb proto— oncogene synthesizes proteins needed in leukemia cell growth. Inhibiting this gene would theoretically lower leukemia cell proliferation. Ratajczak placed tiny pumps into the paraspinal space of leukemic mice. Antisense oligomers were pumped into the mice at 1 ul/hour at 4.2 ug/ul (total dose 100 ug/day) for 3 days. Although death was delayed, by day eleven all of the mice were dead. The experiment was repeated for seven days then again for fourteen days. In both cases, mice receiving antisense oligomers all lived longer than control, sense or mismatched oligomer treated mice. Eventually (by day 42) all mice had died. The authors suggested that antisense should perhaps be used in combination with other conventional treatments when treating cancer. Antisense therapy has been suggested as a method to treat AIDS. Several researchers have tried to use antisense to inhibit the various viral genes involved [Matsukura at al (1987), Agrawal et al (1988), Gilboa et al (1994)]. Lisziewicz et al (1993) tried antisense against the tat, gag and rev genes in cell culture. Although the tat antisense did not inhibit the virus, both the gag and rev antisense inhibited virus replication for more than eighty days when cells were treated twice a week with the antisense. If the oligomers were removed, virus was detected after only four 29 passages. The antisense therefore inhibited replication of virus, but did not get rid of it. Antisense oligonucleotides have been used in humans to treat cancer. Recently, an acute myeloblastic leukemia patient received antisense oligomers to p53, a tumor suppressor gene but involved in cancerous cell proliferation. This antisense message was successful in inhibiting p53 in cell culture even after treatment with the antisense was discontinued. The patient received 0.05 mg/kg/hour of phosphorothioate oligonucleotide for ten days (total dose 700 mg). The only side effect seen was higher levels of gamma glutamyl transpeptidase which returned to normal after treatment was complete. The patient did not respond to treatment though cells taken at day six and eleven had reduced growth in vitro. This is one of the few examples in which toxicity of the phosphorothioate modified antisense oligonucleotide was mentioned. Phosphorothioate modified nucleotides are released into the cell when the oligomer is broken down. Theoretically, these altered bases could be used in the synthesis of DNA. As there was no major toxicity detected in this case, this provides a definite advantage for using antisense oligonucleotides as a therapeutic (Bayever et al, 1992). Another unexpected finding is the fact that sense messages, usually used as controls in antisense experiments, sometimes also inhibit the target gene (Jorgensen, 1990 and Nellen et al, 1993). Jorgensen (1990) was working on 3O genetically engineering flowers with different colors. After adding an extra copy of the gene of the anthocyanin (purple pigment), his flowers turned white. Thus an extra sense gene knocked out the endogenous gene. These new colored flowers may be making money for the plant industry, but it is adding to the confusion about antisense. Conclusion Antisense RNA or DNA has been shown to affect many processes in gene expression including transcription, splicing, transport or translation. The function of endogenous antisense messages remains questionable in eucaryotes, but antisense undeniably has become very useful as an exogenous gene regulator. Further understanding of how these messages work may enable even more use of antisense as a treatment for diseases or as means to disrupt expression of other undesirable gene products. As of now, the action of antisense remains as Denhardt reports the "sometime inhibition of transcription, processing, transport or translation". Fatty Acid Synthase Literature Review Fatty acid synthase (FAS) is a key enzyme in the synthesis of lipids in an organism. In this project, FAS was targeted in order to inhibit fattening in TA1 cells in hopes that this method could be used to reduce fattening in a production animal. Due to the availability of the FASl fragment (Witowski et al, 1987) and the understanding of TA1 cell culture (Dickerson et al, 1992), the enzyme activity was measured in order to determine if the method was successful. Outlined below is brief summary of this multidimensional enzyme. FAS catalyzes the reactions needed to synthesize palmitate, the required fatty acid for further fatty acid biosynthesis. It is a very large protein, complex in its function, origin, sequence, gene structure and regulation. Each of these areas has been extensively studied. Overall Reaction FAS synthesizes palmitate by the addition of two carbon units (from acetyl CoA) to malonyl CoA. As a new 31 32 fatty acid is elongated, carbon dioxide and water are released. FAS utilizes NADPH as a reductant throughout the reactions. The overall reaction is: Acetyl CoA + 7 malonyl CoA + 14NADPH + 7 H+ => palmitate + 14mm+ + enzo + 7 co2 + 8 CoA Acetyl CoA is converted to Malonyl CoA.by acetyl CoA carboxylase and is the first committed step in de novo fatty acid synthesis. Malonyl CoA and acetyl CoA are then linked to the acyl carrier protein (ACP) subunit of FAS and fatty acid synthesis follows (Stryer, 1988). Fatty acid synthesis occurs in the cytoplasm of eucaryotic cells. Each intermediate in the pathway is linked to ACP, a coenzyme, by a sulfur bond (Stryer, 1988). ACP has prosthetic group of phosphopantetheine which enables all the acyl groups of each intermediate to remain attached as the reaction progresses. The end product, palmitoyl CoA inhibits the enzyme while citrate, the source of acetyl CoA, activates it in the cytoplasm (Stryer, 1988). Origin and Structure of FAS One monomer of FAS is comprised of seven separate enzymes in three domains. The active protein functions as a homodimer in eucaryotes. The individual enzymes include 6- 33 ketoacyl synthase (condensing enzyme), acetyl transacylase, malonyl transacylase, B-ketoacyl reductase, enoyl reductase, dehydrase, thioesterase and the acyl carrier protein (Stryer, 1988). In bacteria, the enzymes function as individual entities and as two separate units in fungi. It is thought that these individual enzymes became linked as animals evolved (Amy et al, 1992). The monomer has a molecular weight of 260,000, with the three domains of each protein arranged in an antiparallel fashion. The first domain contains the B-ketoacyl synthase and both transacylases. The substrates enter in domain one and are elongated at the interface of domain I and domains II and III of the opposite monomer. Domain II contains the reductases, dehydrase and the ACP. The ACP prosthetic group is positioned into the center of the three domains. Domain III (the thioesterase) releases the palmitate. Overall Domain I from one monomer acts in conjunction with Domains II and III from the second monomer forming the homodimer (Wakil, 1989). In chicken FAS, domain I, II and III have a molecular weight of 127,000, 107,000 and 33,000 respectively (Mattick et al, 1983) and the rat domains are 125,000, 95,000 and 12,000 daltons respectively (Rangan et al, 1991). The FAS amino acid sequence is also homologous among different species. Examination of homology between chicken, rat and yeast FAS was determined by Chang and Hammes (1989). Chicken and rat FAS amino acids were 67% similar. The highest degree of similarity was between active sites in 34 chicken and rat FAS. Yeast and chicken FAS was only 18.8% identical (Chang and Hammes 1989). Mildner and Clarke (1991) reported that rat and pig FAS was 68% homologous but pig and chicken had only 14% homologous amino acids. The thioesterase domains of FAS are similar in rats, mallards and rabbits. These similarities in the FAS protein also indicate a common point of origination (Witkowski et al, 1987). Gene Structure of FAS The rat FAS gene is a well studied example of eucaryotic FAS. This gene is a single gene over 18,000 base pairs in length and includes forty-two introns. Introns average 191 bases in FAS, unlike the 1127 base average in most vertebrate genes. Intron-exon connections match the established GT/AT standard. Two separate polyadenylation sites are present which give rise to two mRNAs of lengths 8.3 and 9.1 kilobases (Amy et al, 1992). The gene present only once in the eucaryotic genome. The rat FAS gene has been completely cloned including the transcriptional start site and regulatory region. Eighty-seven nucleotides present at the 5' end of the cloned gene are similar in lung, liver and mammary gland FAS mRNA. Sequence analysis revealed that the 5' end of the gene contains similar transcriptional regulatory sequences to the estrogen response element, glucocorticoid response element, thyroid 35 response element and the progesterone response element. Eight sequences of the cAMP transcriptional regulatory element are also present (Amy et al, 1990). Joshi and Smith (1993) used a baculoviral vector to show that FAS could be synthesized, folded and dimerized in an insect host cell. The prosthetic group was also successfully attached to the ACP. The FAS cDNA was spliced together then cloned into the baculovirus. Forty-eight hours after transfection, FAS accounted for twenty percent of the cytoplasm. Further studies on FAS will benefit from this system (Joshi and Smith, 1993). Regulation of FAS FAS is an enzyme with an abundance of regulators. Dietary components and changes as well as hormones all affect the transcription of FAS. Dietary changes can influence FAS transcription. Fasting an animal decreases transcription while feeding increases FAS levels. FAS mRNA.levels were positively related to rate of FAS synthesis, i.e. transcriptional regulation is involved (Morris et al, 1984). Fatty acids can inhibit lipogenic enzyme synthesis at the transcriptional level. Although polyunsaturated fatty acids in the diet can regulate transcription of FAS, monounsaturated or saturated fats seem to have no influence over FAS transcription (Armstrong et al, 1991). 36 Polyunsaturated fats have been shown to depress FAS mRNA levels in rat liver (Clarke et al, 1990a, Blake and Clarke, 1990). FAS mRNA levels also increased when the rats ate an elevated carbohydrate diet or glucose. High fat diets, however, dramatically lowered FAS mRNA levels. From these studies, it was estimated that FAS mRNA had a half life of eight hours (Clarke et al, 1990b). Hormones can also modify FAS mRNA levels. FAS mRNA was decreased in both adipose and liver tissue in pigs injected daily with porcine somatotropin. Pigs fed higher levels of protein also had depressed FAS mRNA levels in the adipose but not in the liver (Mildner and Clarke, 1991). In obese rats, FAS is overly abundant. Guichard et al (1992) determined that the rate of FAS transcription was much higher in obese rats as compared to lean rats. Amplification of the gene or mutations in the protein were ruled out. The difference in regulation of FAS between the lean and obese rats was not elucidated. The experiment did show that FAS was transcriptionally regulated in rat adipose tissue. FAS in adipocyte cell culture systems Several studies of FAS in cell culture have led to an increased understanding of FAS and its regulation. FAS mRNA levels dramatically increases in 3T3-L1 cells when they differentiate from preadipocytes to adipocytes. It has also 37 been shown that cAMP lowers and insulin raises FAS mRNA levels in these cells. Sequences present in the promoter region have been shown to be responsive to insulin (Moustaid et a1, 1993; Paulauskis and Sul, 1988). Triiodothyronine has proven to increase transcription rates of FAS mRNA as well as increase its stability in 3T3-L1 as well (Moustaid and Sul, 1991). In TA1 cells, another adipogenic cell line, FAS activity increased upon differentiation with dexamethasone and indomethacin. The B-adrenergic agonists ractopamine and isoproterenol also affect FAS by depressing its activity in TA1 cells (Dickerson et al, 1992). Conclusions These different examples demonstrate that FAS regulation remains as complex as the protein itself. Antisense may be an alternative method to study the regulation and function of FAS. Objective The meat animal industry continues to strive toward devoloping methods to lower fat accumulation in food animals that can be regulated, produce no by-products and could be implemented in a production animal in the future. This experiment was designed to research a very specific method of inhibiting a lipogenic enzyme. The objective was to determine if a plasmid containing a portion of the FAS gene in antisense orientation can inhibit fatty acid synthase activity in stably transfected TA1 cells. Three separate aims were required to accomplish this goal:create an antisense plasmid, transfect the plasmids into TA1 cells and create stable cells and subsequently determine if FAS activity was inhibited. Hypothesis Stably transfecting TA1 cells with a plasmid containing antisense fatty acid synthase can lower fatty acid synthase (FAS) activity. Aim #1 Synthesize a plasmid containing the antisense and sense piece of FAS. Aim #2 Establish stable TA1 cell lines with the antisense, sense and control plasmids. Aim #3 Determine if inhibition of endogenous FAS occurred in the stable antisense TA1 cell lines. 38 39 Aim #1: Plasmids Methods for Plasmid Synthesis The host vector required several characteristics:a strong constitutive promoter, a polyadenylation site, resistance genes for selection in bacteria as well in eucaryotes and the ability to integrate into a eucaryotic genome. For these reasons, pcDNA3 was purchased from Invitrogen, San Diego, California (Figure 3). This vector uses the cytomegalovirus as a strong constitutive promoter. It contains a polyadenylation sequence from the bovine growth hormone which immediately follows the multiple cloning site. Both the ampicillin and neomycin resistance genes are present for procaryotic and eucaryotic selection, respectively. The SV4O early promoter allows for integration into the host genome. Both T7 and SP6 promoters enable the plasmid to be used for in vitro transcription or for sequencing. The pcDNA3 plasmid, with the chloramphenicol acetyl transferase (CAT) gene ligated into the HindIII site, was also purchased from Invitrogen (Figure 4). Both of these plasmids were used as controls throughout the experiments. FASl (Witowski et al, 1987), was ligated into the vector to create the antisense and sense plasmids (Figures 5 and 6). This fragment is 1200 base pairs in length and codes for the acyl carrier protein domain and parts of the 4 .meHm HHHUCH$ .m2 emnmonecfi HHHUCHI orb ODCA omumofia ma muflm ocflcoHo maafluase ma e5 232.21 2.0 . :3... 52:29.5. 41 macaw ka Dacom .coflumucwfluo oncom :fi m<200d mo obflm Hmoom on» obcfl Umumofla ms Hm15 ug in each blot). This probe should detect FASl as well as the plasmids used in each cell line. In this case, there was some detection in each cell line, 52 including the nontransfected control. No distinct bands were detected. It is unknown why the plasmid could not be detected reproducibly in the transfected cells lines since the geneticin selection insures that the plasmid is present in the cells. If the plasmid was present in only one or two copies per genome, the system used to detect the plasmids may not have been sensitive enough. The riboprobe was used for this reason. Since the synthesis of riboprobes allows for more incorporation of radioactive nucleotides, this probe should have caused more exposure on the film when bound to a complementary fragment of DNA as compared to the random primed probe. However, the riboprobe did not detect anything except positive controls. The digoxigenin plasmid probe had a much wider range of DNA it could detect. By digesting the DNA with EcoRI, the probe should detect the 1200 base pair FASl as well as the remainder of the plasmid, 5440 base pairs in length. However only a similar smear of DNA was detected in each cell line. Antisense FAS mRNA was also unable to be detected by Northern blotting. Northern blots of total RNA (10—20 ug per lane), when probed with either the sense or antisense riboprobe, showed no differences between cell lines. Northern blots of mRNA (0.3 ug per lane) also failed to detect antisense mRNA. Only endogenous FAS was detected in each cell line. This result is not uncommon in antisense work; others have been unable to detect antisense mRNA 53 (Pepin et al, 1992; Fire et al, 1991;, Celano et a1, 1992; Kim and Wold, 1985). Poor or no detection of antisense mRNA may be due to instability of antisense mRNA or rapid degradation of the sense:antisense hybrid (Kim and Wold, 1985). Since antisense mRNA really has no purpose in the cell, it is likely that the cell would degrade it quite rapidly making it difficult to detect. The ELISA assay, a colormetric assay detecting CAT protein, showed a significantly greater amount of CAT protein in CP2 cells as compared to nontransfected cells (Figures 11 and 12). This experiment was repeated and verified this result, however numerically both the CP2 and nontransfected cells had four fold lower activity in the second experiment. This difference between experiments is due to differences between the ELISA kits used. The second kit showed much lower results in other experiments within the laboratory. The ELISA assay offers further evidence that this plasmid can transcribe RNA usable within the cell. The CP2 cell line produced the active CAT protein in significant amounts as compared to nontransfected cells. Though this can not be used to measure how much antisense or sense mRNA is produced in the other cell lines, it does show that this plasmid can be used to incorporate exogenous genes into the TA1 cell. 54 CAT Protein CAT Cells vs Control Cells 0.4 P<.Ol p I 9 m3 / 2 0 0.2‘ O u 0.1 l 0.0 [3 CAT Cells II Control Cells Figure 11. CAT Protein in Experiment #1. Amount of CAT protein in polyclonal cells transfected with pCAT compared to cells not transfected. Amount of protein is determined per 200 ul of cell extract. Significance is indicated on the chart itself. CAT PROTEIN CAT Cells vs Control Cells P<.Ol c>oro\\o'o o c: o c: o o CO C: O D O N 1: CAT Cells - Control Cells Figure 12. CAT Protein in Experiment #2. Amount of CAT protein in polyclonal cells transfected with pCAT compared to cells not transfected. Amount of protein is determined per 200 ul of cell extract. Significance is indicated on the chart itself ' 55 The ELISA assay offers further evidence that this plasmid can transcribe RNA usable within the cell. The CP2 cell line produced the active CAT protein in significant amounts as compared to nontransfected cells. Though this can not be used to measure how much antisense or sense mRNA is produced in the other cell lines, it does show that this plasmid can be used to incorporate exogenous genes into the TA1 cell. Aim #3 Methods of Measuring FAS Activity Cells were observed visually after staining with Oil Red 0 stain (Pollard et al, 1989). Oil Red O specifically stains lipids red and leaves the remainder of the cell colorless. FAS enzyme activity was measured on all cell lines. Cells were rinsed in PBS, incubated in the presence of digitonin to release the protein, then placed in potassium phosphate buffer. After centrifugation, the extract was assayed using the Cary 2200 recording spectrophotometer with the DS-15 enzyme program (Dickerson et a1, 1992). Protein was assayed using the Bradford procedure (Bradford et al, 1976). These procedures are written in detail in Appendix C. 56 Statistical analysis on enzyme assays were performed using Bonferoni t-tests for individual blocks and combined using the chi square test (Gill, 1978). Results and Discussion of Staining and FAS Activity Visually, the cell lines look very similar. Photographs were taken nine days after the cells were induced to differentiate (Figures 13-21). The Oil Red 0 stain is specific for lipid which appears as dark circular droplets in the photographs. The result of the staining with Oil Red 0 Stain is somewhat ambiguous. As photographs can be very subjective, any differences seen between the cell lines may not represent the true differences. Looking at the plates themselves, one may say_that the lipid droplets in the monoclonal antisense Cell lines look smaller but there is very little differences in amount of lipid droplets. The cells do differ somewhat among cell lines in size. The TA1 cells transfected with pCAT appear more cuboidal and do not generally grow as quickly the first day after plating. As seen in the photographs, the CAT cells have as much lipid accumulation as the others. .A more objective method for determining differences among the cells lines is the FAS enzyme activity assay. One monoclonal and each polyclonal cell line was assayed. Each treatment was assayed at three different times (blocks) and \ . e. .i 5. .. ... flaw“ .n .23.... .... mm A. .2 u. m. _ _ A .1 ~ — a .. M“ .i stained ) (AP2 Antisense polyclonal cell line Figure 13 with Oil Red 0 stain nine days after induction of differentiation. "#3: w l h l w m i. ._, M i J.. J 2”,... Wfldmfi a 3.0.1...” u .a n .. ‘89.” N r . Ott. :' v3.\ 13‘... 1 a. 3:. pCAT polyclonal cell line (CP2) stained with Oil Figure 14. Red 0 stain nine days after induction of differentiation. mm. .. .....z a . bemmxe . .. I. . ‘ . l 0.. ...\ ..t. . .O l\ . , .... ......w..3v. P.0...... . a... ”$0.5 m...» .0 4.%w&.;% . a..\ . . . . ....vi .... .. .v . ». .wfiéb .....s... in»... . a.» {irru .. fir ...»? .....-me. . . . .w .. . ....” 6 .wwww, .H£.y ., .. rmhgaovkaz ... . i3. .RwL . . _, ...,» ......wu. 31mg? I....I.o... a Z . .4 . . . ... . . é... ...... $1... 13... . . .. E£n.t.. 4%.”). .....‘fi (SP2) stained with Sense polyclonal cell line Oil Red 0 stain nine days after induction of differentiation. Figure 15 pcDNA3 polyclonal cell line (3P2) stained with Oil Red 0 stain nine days after induction of differentiation. Figure 16. ahfltlflm.&-tmfi . . tar ....ufiu ... f c... .. urn-K... spawwn.in.x qsnx stained l .9. . ......fi .. ..—v- w rvwv‘, L I .. . x. ......u. .35.... ...... , . 0.. .8 9”. . (AMI) days after induction of _.-.~ .... M ”M M . 1‘ ”mos-....“ n Au r ...p . .1... fl...» ; ..r.\ ... . ...-xiii. \&. - . 33...»... final... . fluctkfufiwwfiihfidfiwmmm s A.“ . . .. _. _._. st... Antisense monoclonal cell line .ann .uhw . ._ ., . WW: ..n., ,m-\ . . ... . . .:.a.r . tors ...4‘0‘... th Oil Red 0 stain nine differentiation. W1 Figure 17 ine (CMl) stained with Oil pCAT monoclonal cell 1 Red 0 stain nine days after induction of differentiation. Figure 18 60 t ‘- 63...: _-?av‘ ‘6. 531.3?“ .' .1! 1’, . ' 9‘ ’ 1.4 U x 35:. In [TV a ..IW 'gmgm[;{ Figure 19. Sense monoclonal cell line (SM2) stained with Oil Red 0 stain nine days after induction of differentiation. II 0 .‘ A . 6', ‘|( “.Lg‘bl’ '. .w‘...‘ ‘ ' 0‘, :' \9 . Ii); I 3-‘9 S -R .3.) dz‘ “)1 .ifl) fi.‘ 1‘ \‘t. ‘0 fi .. yr ' 5 a P u -l --"'-.-'.g:-;‘»-'1