Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
I FRONTIERS IN THE NUTRITION SCIENCES
MOLECULAR BIOLOGY AND NUTRITION RESEARCH Richard W. Hanson, Maria Hatzoglou, Mary M. McGrane, Fritz M. Rottman, and Thomas Wagner Metabolic research is the foundation of the science of nutrition, and progress in one area is linked to the vitality of the other. Metabolic research has not prospered over the past decade, when compared with research in fields such as genetics, due in part to a limitation in the techniques applied to metabolic problems. For too long metabolic research has focused on redefining problems of regulation without developing new methods powerful enough to provide definitive answers to these problems. The result has been a general demise of metabolism as an area of modern biology. With its research base contracting, nutritional science has become a more descriptive and less vital scientific field. The revolution in molecular biology has brought new opportunities and a fresh challenge for metabolic research. To date, research in molecular biology has focused on the characterization of specific genes and on the processes that alter their expression. Many genes have been sequenced and their promoter-regulatory regions have been delineated, and various mutations involved in disease processes have been described. However, the ability to isolate and characterize a gene of interest is the first step in a process that can involve the modification and expression of that gene in cells and animals. It is here that the opportunities for the direct application of molecular biology to the field of metabolic research has the potential for its greatest impact. This brief review outlines some of the techniques in this rapidly evolving field and speculates on their application to metabolic research. 3
ISOLATION AND CHARACTERIZATION OF GENES OF METABOLIC INTEREST The first area in which molecular biology will have an impact on metabolic research is the isolation and characterization of genes that code for proteins of metabolic interest. To date, understanding of the regulation of a key metabolic enzyme, such as the pyruvate dehydrogenase complex, phosphofructokinase, or P-enolpyruvate carboxykinase (PEPCK), ha's been limited by a total lack of knowledge of the primary o'r three-dimensional structures of these proteins or the configuration of the ligand-binding domains, which are critical to their functioning. As the list of genes of this type that have been'sequenced' "rows, there will soon be a complete series of structures on which to base experimental' approaches to understanding the regulation of key enzymes. This initial phase of characterization of metabolically important genes is well under way, and it is safe to predict that within the next 5 years, most of the major 'regulatory enzymes in important metabolic pathways such as glycolysis and gluconeogenesis will have been cloned, sequenced, and studied further. The availability of complementary DNA (cDNA) clones for specific regulatory enzymes will permit their overexpression in bacteria and purification of the protein. Coupling of these techniques with site-directed mutagenesis of the DNA will permit a detailed analysis of structure-activity relationships for individual enzymes. This level of analysis requires knowledge of the complete crystal structure of the protein to accurately assign specific amino acids within the protein. 'At this point, the determination of the crystal structure of a protein is the rate-limiting step in studying'its function by site-directed mutagenesis. This fact is reflected in the current demand for (end ' shortage of)''X-ray ' crystallographers. It is ironic that it is now easier to isolate, clone, and sequence a gene than it is to characterize it by established physical methods. ' An excellent example of the great power of the combined techniques discussed above is the work of Robert Fletterick and colleagues on the characterization of glycogen phosphorylase (Sprang et al., 1987; S. Sprang, E. Goldsmith, and R. Fletterick, unpublished data). This dimeric enzyme, which is present in muscle and liver 4
tissues (Krebs and Fisher, 1956), has been purified and its crystal structure determined (Sprang et al., 1987, ,~nn7,hl i chap ~! The It's is regulated by two parallel mechanisms. The first of these involves its phosphorylation by a glycogen, phosphorylase kinase, which converts the enzyme from the inactive lo-form to the active a-form (Krebs and Fisher, 1956~. Glycogen phosphorylase kinase is, in turn, phosphorylated (and activated) by a cyclic AMP (cAMP)-dependent protein kinase. This mechanism of liver glycogen phosphorylase regulation is directly linked to the level of blood glucose via alterations in the concentration of glucagon. The second type of regulation involves control of the activity of the enzyme by a series of intracellular intermediates, including AMP, ATE, and glucose-6-phosphate. These two mechanisms of regulation working together ensure a coordinated response to changes in energy metabolism and to the carbohydrate status of the organism. Such control is central to metabolic regulation and forms the basis of the nutritional response to carbohydrate intake. _ --~ ~ ~ J · ~ 4~_ AWLS ~ ~ ~ Glycogen phosphorylase b2 the unphosphorylated form of glycogen phosphorylase' binds AMP at a site on one of the subunits, which in turn promotes further binding of this ligand to a site on the second subunit. Glycogen phosphorylase b is inactive in the absence of AMP and is 80% as active as Glycogen phosphorylase a when AMP is bound to the enzyme (Green and Cori, 1943; Morgan and Parmeggiani, 1964~. Both ATP and glucose-6-phosphate inhibit glycogen phosphorylase b by competing with AMP and stabilizing the catalytically inactive conformation of the enzyme. An understanding of the nature of the complex interaction between these regulatory molecules has been limited by the absence of a detailed structure-function analysis of glycogen phosphorylase. In a recent publication, Sprang et al. (1987) presented a detailed analysis of the structure of the glycogen phosphorylase nucleotide activation switch. That study elegantly demonstrated the power of the combined techniques of molecular biology and protein chemistry when they are applied ~ ' ~ ~ interest. Since the c DNA for the enzyme was cloned and the primary sequence of the phosphorylase was deduced from the nucleotide sequence, a detailed map of the amino acids at the enzyme's active site was determined. From ~ co a regulatory enzyme or metabolic 5
the crystal structure of the protein, Sprang and coworkers showed that the subunit interface of glycogen phosphorylase is composed of a catalytic and a regulatory domain (Figure 1~. The catalytic site of the enzyme was B i\ / Dome, n (' ~ .: - \ r At . . . . . . .... . .. ... ..... ... ...... , .. .... / _~/Regulatory ~ W ~moi>> . ~ FIGURE 1 Schematic representation of changes in the dimeric glycogen phosphorylase a molecule. Abbreviations: S. L~gand-binding sites for glucose and glucose-l-Pi G. glycogen and oligosaccharides; and N. AMP and ATP. SOURCE: From Sprang et al. (1987), with permission of the authors. mapped to a crevice in the protein between the two domains, while the regulatory domain was found at the interface of the two subunits. The amino acid residues involved in nucleotide binding were identified, and their relative affinities for AMP and ATE were determined. A reaction mechanism based on the structure of both the 6
regulatory and catalytic sites on glycogen phosphorylase can thus be derived, and the confirmational changes that occur in the enzyme after ligand binding can be described. A detailed presentation of these studies is beyond the scope of this review; however, the work of Sprang et al. (1987) is an example of the refined analysis of enzyme regulation that will provide a molecular basis for the next generation of metabolic studies. Because of this work, it is now possible to discuss the regulation of glycogen phosphorylase in terms of its molecular interactions with regulatory molecules and to alter the metabolic function by site-directed mutagenesis based on an understanding of the reaction mechanism. Information of this type for other enzymes can be expected to increase rapidly as the sequencing and crystal structure analyses become available. ~ knowledge of the enzyme structure at this level also permits the rational design of modified proteins for metabolic studies involving gene transfer to cells and animals. This is discussed below in more detail. THE INTRODUCTION OF GENES INTO CELLS AND ANIMALS Construction of Chimeric Genes Many structural genes that code for proteins of metabolic interest have been isolated and characterized (Goodridge and Hanson, 1986), including genes that code for key regulatory enzymes in metabolic pathways, hormones such as growth hormone, and insulin and a variety of receptors, to name only a few genes of interest. The number of proteins that have been isolated and characterized is growing rapidly and is likely, ultimately, to include virtually all of the important proteins involved in metabolic processes. At the same time, the control regions within the 5-flanking gene sequences are being isolated and studied. It is this promoter-regulatory region of a gene that controls its expression. Specific sequences contained in this complex region of a gene interact with transcriptional regulatory factors' which are often intermediates in hormone action, to- control the tissue-specific expression of the gene in animals. It is
now possible to construct genes that contain a chimeric promoter-regulatory region that contains, for example, selected ho`mone-responsive elements, a promoter element with appropriate strength, and a tissue-specific element that directs expression of the gene to a tissue of interest. - This chimeric promoter-regulatory region can these ligated to a segment of DNA coding for a structural gene of interest and can be introduced into cells or animals. As the techniques~for stably introducing genes into cells and animals improve, it should be possible to reproducibly target genes for selected tissues in animals by using these chimeric genes. This is a new tool with great potential for metabolic studies, since it will permit, for the first time, a modification of an enzymatic step in a complex metabolic pathway without the use of inhibitors or other compounds that have a broad It also provides the potential to correct metabolic defects in humans, if the technology can be perfected to ensure a predicted site of integration into the human genome, as well as a normal level of expression of the newly introduced gene. We review here some aspects of this field and use as an example our own studies with the PEPCK gene, since its promoter-regulatory region has proven useful in driving the expression of a variety of structural genes in a regulated and tissue-specific manner. We also review the techniques currently being used to introduce genes into cells and animals in order to demonstrate both the. limitations of and potential for these techniques. spectrum of action in cells or animals. Properties of a Regulated Promoter The selection of an appropriate promoter-regulatory region for use with a linked structural gene in metabolic studies depends on the tissue in which the gene is to be expressed and the type and level of regulation of gene expression that is required. The promoter-regulatory region of the gene coding for the cytosolic form of PEPCK (GTP) (EC 4.1.1.32) contains a highly regulated promoter sequence, with regulatory elements for cAMP (Short et al., 1986; Wynshaw-Boris et al., 1984, 1986), glucocorticoids (Wynshaw-Boris et al., 1984, 1986), insulin (Magnuson et al., 1987; McGrane et al., 1988) and thyroid hormone (M. Hatzoglou, W. Lamers, A. 8
Wyushaw-Boris, and R.W. Hanson, unpublished observations), all within 550 base pairs (bp) of flanking DNA at the 5' end of the gene. This gene also has a very high level of transcription (3,500 parts per million after administration of cAMP) (tamers et al., 1982; Meisner et al., 1983) in transgenic mice and tissue-specific sequences are present in the promoter that directs gene expression, to the liver and kidney (McGrane et al., 1988~. The messenger RNA (mRNA) for PEPCK has a half-life of 30 minutes (Granner et al.. 1983 Til~hman et al ~ ~ , _ __, ~__= ~.^ ~ ., 1974), so that the level of mRNA for the enzyme is largely dependent on alterations in the transcription rate of the gene. Hormones such as glucagon (acting via cAMP), glucocorticoids, or insulin have a very marked and rapid effect on PEPCK gene transcription. a- _ _ The administration of Bt2cAMP to an animal causes an 8- to 10-fold induction in the rate of transcription of the Gene within 20 minutes (tamers et al., 19821. This effect can be blocked in cells if insulin is administered together with the cAMP (Granner et al., 1983; Wynshaw-Boris et al., 1986b). The PEPCK promoter thus has many advantages for use as a vehicle- to drive the expression of linked genes in animal tissues; it is expressed in a tissue-specific manner and is regulated acutely by hormones, and its promoter strength is great enough to ensure sufficiently high levels of gene product. ~, ~. ~. . _,_ ~ The various hormone response elements in the promoter have been identified by gene transfection experiments in which chimeric genes containing segments of the promoter-regulatory region of the PEPCK gene were ligated to the structural gene for a selectable marker, such as the amino-3'-glycosyl phosphotransferase gene, which makes cells resistant to the cytotoxic compound, G418, or the Herpes virus thymidine kinase (TK) gene (Short et al., 1986; Wynshaw-Boris et al., 1984,-1986b). Figure 2 shows the locations of sequences in the PEPCK promoter that confer sensitivity to cAMP and glucocorticoids in the chimeric PEPCK-TK gene. Two cAMP response elements have been mapped by DNA-protein footprinting to the region between positions -82 to -90 and -135 to -142 (Roesler et al., 1989~. The entire region of DNA between positions -61 and -416 can act as a hormonally sensitive enhancer and can regulate expression when it is 9
a, O a' ~ ~ o so a, - ~ to ~ o - ~Jo it ~o ~ ~ ~ / / ~v o ~ ~ no / ~no =^ ~ ~ <. ~ / ~ ~ ~ ~ ~ ~ ~ / ~ ~ ~ ~ o ~ ~ ~ u o ° L,~ R ~ -a 9 ~ _ ~ . ~ oJo~o~ ~ >~ i ; \ ~o ~ on ~ o - of ~ \ ~=m == ~ · \ mO o Cot ~ o \ ~Ed e 3 ~ m" :~ ~\ } ~ ~ ~ "m \ .~, ~ ~ al to Cal \ :~' Ha o c ~ ~ ~ .- \ ~ ~ ~ o ~.. ~04 8 8 ~ e e a) al a, a, u' ~ ~ a, ~ 10
introduced into the chimeric gene in either transcriptional orientation or at a distance from the start site of transcription of the gene (Figure 3; Wyushaw-Boris et al., 1986b). REGULAT10N cAMP DEX -6s0 ~1 -548 +1 +73 ,-. - -355 +1 +73 -~74 +1 +73 -109 +1~73 + + _ __ ~ -68 +1 +73 -109 +1 -109 -62/-l09+l + + ~_ _ _ _ _ _ ~+ + -62 -10~-l09+l -548 -10~-K9+' -416 -61/ - 109 +' ,. -61 -416/-109 +1 + ___c , . /, + + -109 +1 +1800/-416 -61 ___~, ,* + + -109 +1 +1800~-61 -416 -S48 +73i-650 +1 -S48/-650 ~1 FIGURE 3 Functional analysis of the PEPCK promoter-regulatory region for cAMP and glucocorticoid regulatory elements. Rat hepatoma cells deficient in TK (cont. next page) 11 +
(Fig. 3 caption cont.) were transfected with the genes shown on the left. These chimeric genes contain segments of the PEPCK promoter-regulatory region ligated to the structural gene for Herpes virus TK from the region noted in the figure. RNA was isolated from hypoxanthine-aminopterin-thymidine (HAT)-resistant cells that were either treated with B-t2AMP (cAMP) or dexamethasone (DEX) or not treated, and the level of TK mRNA was determined by quantitative S1 nuclease mapping. An induction (+) or lack of response (-) of TK mRNA within 3 hours after the addition of cAMP or DEX is indicated on the~right. The PEPCK promoter-regulatory region is represented by filled boxes, with arrows indicating the direction of ligation relative to the direction of PEPCK gene transcription. Open boxes are the TK promoter sequences retained in the chimeric gene in the indicated constructions. The direction of chimeric gene transcription is noted by the arrow at position +1. The numbers below either the PEPCK or the TK promoter-regulatory regions indicate the position of the fragment relative to the start site of the gene of origin. The thin line represents TK structural gene sequences. SOURCE: Wynshaw-Boris et al. (1986a); the figure is based on a variety of published studies (Short et al., 1986; Wynshaw-Boris et al., 1984, 1986b), in which the methods used are described in detail. ~, The number of tissue-specific promoters that have been isolated and characterized in both cells and transgenic animals is growing rapidly. One of the most intensively studied elements is the promoter for the rat elastase I gene, which is expressed in the acinar cells of the exocrine pancreas. Hammer et al. (1987) have reported that 134 bp of 5'-flanking sequence in the elastase I gene is sufficient to direct gene expression in the pancreas of transgenic mice. This element, which extends from positions -72 to -205, is highly conserved in all species studied and confers the correct developmental and regulatory properties on a linked structural gene. It has even been possible to induce neoplasia of the exocrine pancreas in fetal transgenic mice that contain a chimeric gene composed of 4.5 kilobases (kb) of the rat elastase I promoter, ligated to the structural gene for the onco gene c-H-ras (Quaife et al., 1987~. The 12
- tissue-specific expression of the immunoglobin, globin, actin, and a-fetoprotein genes has also been studied by using tranagenic animals; and sequences that are responsible for their tissue-specific expression have been identified in the 5'-flanking regions of these genes. It is thus possible to select promoters containing sequences that direct the expression of a linked structural gene to a tissue of choice in an animal. '~ Is Anus possible to select Promoters Methods for Introducing Genes into Cells and Animals Genes can be introduced into cells by transfection, particle-mediated gene transfer, microinjection, electroporation, or viral infection and can be introduced into animals by microinjection of DNA into the male pronucleus of fertilized eggs or retroviral infection of embryonic tissue prior to development. A detailed discussion of each of these techniques is presented in detail by Wynshaw-Boris et al. (1986a). We concentrate on three of the most widely used techniques relevant to metabolic studies: DNA transfection, retroviral infection of cells and animals, and nuclear DNA . . . . m~cro~nJect~on. DNA Transfection. DNA transfection is the most commonly used technique for introducing genes into cells. DNA is layered onto cells as a calcium phosphate precipitate; it is then taken up by endocytosis and is transported to the nucleus. The DNA is retained in the nucleus for a relatively short time (24 to 48 hours) before it is degraded. A small fraction of the gene introduced by this procedure is transiently expressed, and about 0.01% is integrated into the genome of the host cell. It is possible to use selection techniques (see Wynshaw-Boris et al. [1986a] for a detailed review) to isolate and subclone cells that have stably incorporated functional copies of the transfected gene into their genome and to establish clonally isolated lines of these cells. Alternatively, transient expression of the gene can be measured and used to monitor gene function. Despite its limitations, this procedure is widely used to measure the function and regulation of specific promoter elements in any cell line that can be maintained in culture. The usefulness of DNA transfection for metabolic studies is limited by its low efficiency of gene transfer and by the need to isolate and establish 13
cell lines in order to carry out the more long-term experiments required for the replication of metabolic measurements. Retroviral Vectors. Retroviruses have several features that make them ideal vectors for the introduction of genes into cells and animals. Retroviral infection is highly efficient and produces cells that have integrated into their genomes a single copy of the provirus containing the gene of interest. In certain cases, this obviates the need to subsequently select from a large population of cells individual cells that functionally express the transfected gene.- A wide variety of species-specific retroviral vectors are available that can accommodate DNA segments ranging from 1 to 15 kb. Finally, retroviral vectors can be used to infect both cells and animals, which makes retroviruses useful vectors for the introduction of genes into differentiating tissues in animals during development (Johnson et al., 1985; Soriano and Jaenisch, 1986; Van der Putten et al., 1985~. There have been several excellent recent reviews that document the usefulness of retroviral vectors for gene transfer (Bernstein et al., 1985; Gelboa et al., 1986~. The retroviral life cycle involves four steps. First, the virus binds to and infects the host cell. This process is species specific in that the virus recognizes determinants on the surface of the cell to be infected. Second, the RNA of the virus is converted to DNA by reverse transcriptase. This DNA enters the nucleus and integrates into the host cell genome as a provirus. Third, the provirus acts as a template for the synthesis of RNA, which yields viral proteins and additional copies of the retroviral genome. Finally, the genomic RNA and the-retroviral proteins recombine into a mature virus that is released from the cell to initiate a new round of infection. These steps are illustrated in Figure 4 and have been described by Bernstein et al. (1985) and Gelboa et al. (1986~. The retroviral genome contains the information needed for transcription of the proviral DNA (cis functions), as well as the genes coding for proteins that are necessary for the normal life cycle of the virus (bans functions), such as the envelope protein (env gene) or reverse transcriptase (~1 gene). The cis functions in the retrovirus include the promoter sequences required for 14
B.ndlaq ond m~ect~f~ ~-~ 44) - ~ - ~ his calm B`.dd`rg v.rus Q \ ~"A_Intect ous \~UCkuS [\ :t ~ ~ J \ c~omoso~na' ~ ~ ~ \ oN~ ~ \P,o~e'~s FIGURE 4 Retroviral life cycle. The process of retroviral infection is outlined. The figure traces the integration of proviral DNA into the genome of the host cell. Details are given in the text. (Reprinted with permission from Bernstein et al. [1985, p. 235]). the initiation of gene transcription, various enhancers, and the sequences needed to guide the integration of the provirus into the host cell genome. These are contained within regions of the provirus termed long terminal repeats (LTRs). There is also an important segment of DNA that encodes the information needed for encapsidation, or viral packaging ~ Sequences). Many retroviral vectors that retain these cis functions within the vector have been constructed. An example is the retroviral vector ply, shown in Figure SA, which is derived from the Maloney murine leukemia retrovirus (Korman et al., 1987~. To be functional, retroviral vectors of this type rely on the availability of bans functions from helper cells that have been infected with retroviruses or from a specially constructed cell that contains these functions stably integrated into its genome. One such cell line, t2, was constructed by Mann et al. (1983) and is widely used to produce a high titer of retrovirus that is capable of only a single round of infection. 15 -
- neo ;~;nd me ~ EcoRI pBR ori Sam HI SmaI Saline ~ _~ r. ._ I D ~ 1_ Pvu I~ y~-~-Xho I _~/25 MTV LTR Sma I j7Z' 3; Nhe I j -, lo 66 Kpn I ~ \~\ pLJ 4~\_ClaI A Tic \__NtleI \ ~_XbaI MLV l TR 2"S \ \~_SmaI ~ 48~ 5~] H~Kpni ~ JISO ~ \ ~1 Eco RI Hinds Pol~mo EGFIY Recion bGH '' ////////// \ 71 SAC I B SAC I 891 1 ~ no ~ ~ \/ I (13 IR PJ~ pLJ- PCKneo - PCKbGH ~5.6Kb FIGURE 5 The murine retroviral vectors pLJ and pLJPCKneo-PCKbGH used to introduce genes into cells and animals. (A) pLJ was a generous gift of Dr. Richard Mulligan (Massachusetts Institute of Technology, Boston, Mass.~. The vector is composed of the Moloney murine leukemia virus LTRs and includes the ~ sequences and the simian virus 40 (SV40) promoter linked to the structural gene for amino-3'-glycosyl phosphotransferase (neo). (B) pLJPCKneo-PCKbGH. The SV40 promoter from pLJ was removed and replaced by a 550-bp segment of the PEPCK promoter-regulatory region. A new chimeric gene composed of the some region of the PEPCK gene ligated to the bGH structural gene (without its own promoter) was inserted at the unique ClaI site in an orientation opposite the direction of transcription from that of the 5' LTR. 16
Alternatively, a series of avian retroviral vectors that are replication competent and that infect virtually all of the tissues of the bird are available (Hughes and Kosik, 1984~. Use of these retroviral vectors results in viremic birds. This type of virus is a powerful tool for the introduction of a foreign gene of interest into avian cells. DNA Microinjection. The methods for the development of transgenic mice are theoretically straightforward, but ~ ~ require a large financial investment in expensive, specialized equipment. A recent detailed review by Camper (1987) provides an excellent summary of both the technique used for producing transgenic animals and the equipment needed to they are technically demanding and initiate such an undertaking. . . . In practice, female animals are induced to superovulate by injecting follicle-stimulating hormone; they are mated with fertile males. The fertilized eggs are collected from the oviducts, and approximately 300 copies of the gene of interest, in a volume of 1 pi, is microinjected into the male pronucleus. The microinjected eggs are implanted into the oviducts of pseudopregnant females and allowed to develop to term (Wagner et al., 1981~. It is customary to wait until the animal is born and nutritionally independent from the mother before initiating the analysis of integration of the microinjected DNA in the genome of the animal. However, analysis of the host animal's genome can be carried out during fetal development or immediately after birth if specific developmental or metabolic questions are involved. Normally, a segment of the animal's tail is removed, and the DNA is extracted and analyzed by dot-blotting with a specific DNA probe, usually a fragment of the transgene that was microinjected. Camper (1987) estimated that of 60 eggs that survive microinjection, 6 of the implanted zygotes develop to term, and of these, 1 to 2 are tranagenic. METABOLIC STUDIES USING GENETICALLY MODIFIED CELLS AND ANIMALS The most common use of genetically modified cells and animals is for the analysis of gene expression control. This is relevant to metabolic studies, since many of the 17
adaptations caused by diet are mediated by hormonally induced alterations in the expression of genes of metabolic importance. Also, understanding of the mechanism of action of hormones such as insulin, glucagon, and glucocorticoids has advanced significantly since the advent of these new techniques for genetically modifying cells and animals. Elements in the promoter-regulatory regions of genes that are required for their response to these hormones have been identified, and specific binding proteins that regulate gene transcription are being isolated and characterized. The major impact of this new technology on nutrition, however, is likely to come in the next phase of research, when genes of metabolic interest are modified, reinserted, and studied in animals. This type of experimental model will be invaluable for research in a broad variety of areas of nutrition and metabolism. Introduction and Regulation of Genes Introduced into Hepatoma Cells via Retroviral Vectors A major point that must be established in proving the usefulness of the techniques available for programming metabolic processes in cells is the level of expression and regulation of genes introduced into cells and animals. The usefulness of retroviral vectors for this type of study is illustrated by a series of experiments in which two chimeric genes, PEPCK-neo and PEPCK-bovine , . _ growth hormone (bGH) contained in a single retroviral vector (Figure 5B), were infected into hepatoma cells in culture. The 550-bp segment of the promoter-regulatory region of the PEPCK gene used to construct these chimeric genes contains both the cAMP and glucocorticoid regulatory elements (Short et al., 1986; Wynshaw-Boris et al., 1984, 1986~. When Bt2cAMP or glucocorticoids were added to hepatoma cells (rat FTO-2B cells and mouse Hepa 1-6C cells) infected with this virus, there was a two- to threefold increase in the output of bGH in the culture medium within 2 hours (Figure 6~. Furthermore, the PEPCK-neo gene was regulated in a parallel fashion in the cells. It is worth noting that the concentration of bGH released by mouse hepatoma cells after Bt2cAMP stimulation is 0.5 ~g/ml of culture 18
70 60 50 FTO-2B 40 30 20 10 _ _ Hope 1- 6C j 700 ~e: CONTROL cAMP Dex cAMP CONTROL cAMP Dox cAMP Insulin 400 ~ c - 300 100 Insulin FIGURE 6 Effect of hormones on the production of bGH by FT0-2B and Hepa 1-6C hepatoma cells infected with the retroviral vector pLJPCKneo-PCKbGH. Cells were incubated with Bt2cAMP (cAMP), dexamethasone (DEX), and insulin (Ins) for 24 hours; the bGH in the culture medium was determined. medium. This is a high level of expression of an infected gene and indicates that the 550 bp segment of the PEPCK promoter-regulatory region used in this chimeric gene may be useful in future studies in which a strong, regulatable promoter is required for producing high levels of a specific gene product for metabolic studies. Use of Retroviral Vectors to Introduce Genes into Animals Retroviral vectors have been used to introduce genes into mammalian tissues, both during development and after 19
differentiation. Embryonic stem cells, such as the XY stem cell line EK.CCE (Robertson et al., 1986) or preimplantation mouse embryos (Soriano and Jaenisch, 1986), have been infected with various murine retroviruses, often containing a marker gene of interest. This has resulted in the detection of viral sequences in specific tissues of adult animals. This technique allows the construction of tranagenic animals containing a specific gene that was introduced into the germ line by retroviral infection (Van der Putten et al., 1985~. Retroviral vectors have also been used to introduce genes into hematopoietic stem cells, which have been reimplanted into the bone marrow of recipient mice (Magli et al., 1987; McSvor et al., 1987~. This allows investigators to track the fate of these cells during stem cell maturation (Magli et al., 1987~. These techniques have a clear application to Gene therapy, and ~ - - ~ ~ -- - - - - - -- - - C7 - - -- - - ~ of, · . ~ . . . _ . . . tnelr usefulness In treating a variety or Human diseases and is being studied intensively. Metabolic and nutritional research would benefit greatly from the development and general application of these techniques since they afford a potential method for introducing genes of metabolic interest into tissues of intact animals. It would be possible, for example, to design nutritional studies by using a modified gene that is responsive to dietary and hormonal stimuli in an animal species of interest. The potential for these techniques for metabolic studies is illustrated by a series of experiments in which the murine retroviral vector, pLJPCKneo (see Figure 5B), was injected as an infectious retrovirus into the peritoneal cavity of fetal rats between 16 and 20 days of gestation while the animals were still in utero (Hatzoglou et al., unpublished observations). During this period the liver is undergoing differentiation from a largely hematopoietic tissue into a hepatic organ (Jones, 1982~. This is accompanied bY a high level of cell division and hepatic growth. the liver of these animals, J _' The retroviral vector integrates into introducing the regulatable rerun promoter linked co a structural gene for a marker protein such as neo or bGH, the latter of which can be assayed in the blood of the animal. The chimeric PEPCK-neo and PEPCK-bGH genes are expressed in the livers of host animals; the level of expression of these genes appears to be regulated by 20
cAMP, glucocorticoids, and thyroid hormone (Hatzoglou et al., unpublished observations). The relative level of KNA synthesized from these newly introduced genes was an order~of magnitude lower than the level of PEPCK mRNA synthesized by the endogenous gene (Figure 7~. The ability to direct the expression of a chimeric gene into the tissues of an animal relatively late in development, however, holds promise as a technique of potential usefulness for broader metabolic studies aimed at introducing modified genes of interest into animals to alter various metabolic and nutritional parameters. INTO -28 ,.lj~ :~e ·~o 710 bp ~ ~ - 37S bp - ` fETAL INJECTION , . . . t 1 In ~ z z ~ _ ~ _ ~ 11 1 : ~ 1 . ., t . - -: . _L_ _ _ . .; ~ ; --- t.;c i .'." i,'. ~-_ 8C ~ provirus pLJ-PCKneo - I I I I I Bgl Ir ~* 32 p - 5'- end lobe led DNA probe 111111 viral LTR RNa ITIL ~ 2 3 4 S ~ 2 3 Fl - 3 ~1 -. 0710bp ~ 375 bp PCK- neo RNA FIGURE 7 Expression of the neo gene introduced into FTO-2B hepatoma cells and in rat liver by retroviral (cont. next page) 21
(fig. 7 caption cont.) infection and its induction by Bt2cAMP. The infectious retrovirus pLJPCKneo was introduced into FTO-2B hepatoma cells and into the peritoneal cavities of 19 fetal rats in utero. The virus had a titer of 8 x 107 G418-resistant colonies of NIH 3T3 cells per ml of virus-containing medium. Fetal rats were injected with 100 pi of virus of the same titer at 19 days of fetal life and were then killed at 3 months after birth. A quantitative S1 nuclease analysis of neo mRNA in FTO-2B cells after treatment with Bt2cAMP is presented on the left. The band present at 375 bp is a protected fragment corresponding to the predicted size of the neo gene probe used in these studies and indicates the mRNA transcript from the PEPCK promoter of the infected retrovirus. The diagram under the panels shows the orientation and size of the hybridization probe used, as well as the size of the predicted fragments protected from S1 nuclease. The 710-bp band represents mRNA transcribed from the 5' LTR of the vector. Bt2cAMP (0.1 ~M) was added to the medium, and RNA was extracted from the cells 3 hours later. Insulin (10-8 unitsj was Added at the same time as the Bt2cAMP or 2 hours later (next to the last lane in the left panel). Rats were injected later with Bt2cAMP 3 hours prior to death. Their livers were removed' and the RNA was\extracted and analyzed for neo sequences. The two panels on the right are the result of two separate exposures of the bands protected during S1 nucleate mapping of RNA from the liver. The three lanes on the far right were exposed to X-ray film for 48 hours i and the five bands on the left were exposed for 24 hours. The relative intensities of the bands are proportional to the concentration of neo mRNA present in the cells. SOURCE: Data from Hatzoglou et al. (unpublished observations). Tissue-Specific Expression and Dietary Regulation of a Chimeric PEPCK-bGH Gene Introduced into the Germ Line of Mice A series of transgenic mice were produced by the microinJection of a chimeric PEPCK-bGH gene into the male pronucleus of fertilized mouse eggs by the technique outlined above and as described by Wagner et al. (1981~. 22
The microinjection experiments were performed by Drs. June Yun and Thomas Wagner (Edison Animal Biotechnology Center, Ohio University, Athens, Ohio). The promoter-regulatory region of the PEPCK gene contained 450 bp of 5'-flanking sequence, which included the cAMP and glucocartic regulatory elements. The concentration of bGH in the serum of animals that contained the gene stably integrated into their genome and ranged from a low of 5 to more than 2,300 ng/ml of serum (McGrane et al., 1988~. Mice with high levels of bGH grew at twice the rate of their littermates but were normal in all other respects. A detailed Northern blot analysis of animal tissues indicated that the chimeric PEPCK-bGH gene was expressed in only the liver and the kidney. This mirrors the expression of the endogenous PEPCK gene, which is also highly expressed in liver and kidney and only marginally in other animal tissues (Hanson and Garbers, 1972~. The level of bGH mRNA in the livers of these transgenic animals was regulated by hormones and diet in a manner similar to that of the endogenous gene. When starved animals were fed a diet high in carbohydrate for 1 week, there was a 95% decrease in the concentration of bGH in the serum of the mice (Figure 8~. This suggests that the expression of the PEPCK-bGH chimeric gene in these animals is sensitive to the insulin released after glucose ingestion. When the mice were refed a diet high in protein but devoid of glucose, the levels of bGH in serum were increased 30-fold after 1 week on the diet. This inductive effect of a high protein and carbohydrate-free diet on the levels of hepatic PEPCK is part of the response of the animal to the need for enhanced gluconeogenesis. The transgenic mice also responded to the administration of Bt2cAMP by increasing the level of bGH in their serum two- to threefold in 90 minutes (see Figure 8~. 23
0.5 0.4 3 0.3 0.2 0.1 A B (24 hr) _ (1 week) (1 week) . _ 1 1 Bt2AMP (90 min) starved High High CHO Protein No CHO 3 2 1 FIGURE 8 Regulation of the bGH concentration in the blood of transgenic mice by diet and Bt2cAMP administration. (A) A transgenic mouse expressing 300 ng of bGH per ml was starved for 24 hours and then fed a diet composed of 82% sucrose, 12.2% casein, 0.3% dl-methionine, 4% cottonseed oil, 2% brewer's yeast, and a 1% mineral mix plus vitamins for one week. Blood was drawn from the tail vein; the animal was then fed a diet containing 64% casein, ~ 22% a-cell nutritive fiber, 11% And 1 & mineral mix with vegetable oil, 2% brewer's yeast vitamins (synthetic diets were from Nutritional Biochemical Corporation) for 1 week. The concentration of bGH in the blood was determined. (B) A mouse expressing 1.4 fig of bGH was injected with Bt2cAMP and theophylline (both 30 mg/kg of body weight) at three consecutive intervals. At 90 minutes the mouse was bled from the tail vein and the concentration of bGH was measured (McGrane et al., 1988~. 24 - 3: c,
The acute responsiveness of the PEPCK promoter- regulatory region to induction by diet and hormones and its tissue-specific expression in the liver and kidney makes it an ideal tool for targeting the expression of various structural genes of interest to these tissues. It is also possible to modulate the level of expression of the structural gene over a broad range by altering the carbohydrate content of the diet fed to transgenic animals. Since the gene for the cytosolic form of PEPCK is normally not expressed until birth (Ballard and Hanson, 1967), the developing fetus is not exposed to a high level of protein that would result from expression of the structural gene. This has clear advantages with hormones such as bGH, which have the potential of interfering with the normal development of the fetus. have noted a normal developmental pattern for the transgenic animals that we have studied to date. It should also be possible to use a segment of the PEPCK promoter, when it is linked to a structural gene of interest that contains its own core promoter but that lacks homologous tissue-specific elements, to direct the expression of this chimeric gene to the liver and kidney. POSSIBLE APPROACHES TO GENETICALLY BASED ALTERATIONS IN METABOLIC PROCESSES There are many metabolic models that can be used to test the potential of the techniques described above. An example is the role of the mitochondrial isozyme of PEPCK in the regulation of hepatic gluconeogenesis. Since the discovery of PEPCK in chicken liver mitochondria by (Utter and Kurahaski, 1954), as well as its cytosolic isozyme in the livers of a variety of species (Nordic and Lardy, 1963), the metabolic roles of the two forms of PEPCK have remained unresolved (Hanson and Mehlman, 1976~. This has been compounded by the fact that species vary in the relative amounts of the two forms of the enzyme that are present in their livers. In birds, for example, 100% of the hepatic PEPCK is the mitochondrial isozyme, while in rodent species such as rats and mice, 90% of the enzyme is the cytosolic enzyme. The majority of species studied to date, including humans, have equal amounts of the two forms of PEPCK (Hanson and Garbers, 1972~. These enzymes are distinct, but related proteins that have approximately 60% sequence identity and are coded for by different nuclear genes (S.M. Weldon, S. 25
Savon, W. C. Merrick, and R.W. Hanson, unpublished observations). The only gluconeogenic tissue known to have a single form of PEPCK is the liver of the chicken, which contains only the mitochondrial isozyme (Watford et al., 1981~. This distribution of the enzyme creates metabolic restrictions on the ability of the chicken to synthesize hepatic glucose. The lack of cytosolic PEPCK in the avian liver has been cited as a reason for the low rates of gluconeogenesis noted with oxidized substrates such as pyruvate or alanine in this tissue (Williamson, 1976~. Lactate conversion to pyruvate in the cytosol provides the NADH required by the glyceraldehyde-3-phosphate dehydrogenase reaction during gluconeogenesis. In the liver of the chicken, the restricted synthesis of P-enolpyruvate in the mitochondria necessitates a continuous supply of NADH from oxidation-reduction reactions in the cytosol. However, for gluconeogenesis from pyruvate or alanine, the transfer of reducing equivalents, as well as carbon, from the mitochondria is required. In tissues that possess a cytosolic form of PEPCK, this is accomplished by the transport of malate into the cytosol, which is then converted to oxalacetate, generating NADH (Williamson, 1976~. In tissues that lack cytosolic PEPCK, the carbon for gluconeogenes~s muse leave the mitochondria as P-enolpyruvate without the transfer of reducinz equivalents. . In chickens glucose _ , synthes~s ~n the liver is restricted to the Cori cycle, whereas net gluconeogenesis from amino acids occurs in the kidney, a tissue that contains both isozymic forms of PEPCK (Watford et al., 1981~. Thus, birds recycle lactate from the muscle and red blood cells to the liver for gluconeogenesis It should be possible to test the metabolic role of the two forms of PEPCK directly in the hepatic cells isolated from chickens, in which the gene for the cytosolic form of the enzyme has been introduced via an infectious retrovirus by the techniques described above. The gene for the cystosolic form of PEPCK from chickens has been isolated, and a full-length cDNA is available (Hod et al., 1984 a,b). Since the gene can be inserted into the liver via infection of fertilized eggs with the replication-competent retrovirus described above, the levels of gene expression can be measured both by determining the concentration of PEPCK mRNA present in 26
the liver and by assaying for enzyme activity in the cytosol present in the liver. The effects of this genetic modification of the cellular localization of the rate-limiting enzyme in hepatic gluconeogenesis should have dramatic effects on glucose homeostasis in chickens. Therefore, a method is available to test the predictions of the metabolic model discussed above. Alternatively, it should be possible to target the mitochondrial isozyme of PEPCK to the hepatic mitochondria in mice. This would involve the construction of a chimeric gene consisting of the promoter-regulatory region of the gene for the cytosolic form of PEPCK, which contains both hormone regulatory elements as well as sequences directing the tissue-specific expression of the gene to the liver and kidney, linked to the structural gene for the mitochondrial form of PEPCK from chickens, which also contains the signal sequences necessary to direct the protein to the mitochondria of murine liver. A cDNA coding for this isozyme has recently been isolated (Weldon et al., unpublished observations). Alternatively, the availability of a hepatoma cell line with minimum deviation that can grow in a glucose-free medium, such as the Fao cell line isolated and characterized by Weiss and colleagues (Deschatrette and Weiss, 1974), will be an invaluable tool in these types of studies, since it permits infection with the murine retroviral vectors already developed and has been shown to effectively introduce functional and regulatable genes into other hepatoma cells. Since the cells are able to synthesize glucose from the gluconeogenic precursors contained in the culture medium, their growth in the absence of added glucose should be directly dependent on the efficiency of gluconeogenesis. Finally, once these genetically modified cells are constructed and characterized, they are a permanent resource for other investigators who are interested in related metabolic problems. The scenario presented above represents only one of many metabolic questions that might be approached by using the techniques outlined here. We anticipate rapid growth in this field as the number of genes that are isolated and characterized increases. It should be possible, for example, to introduce receptors for hormones such as insulin into the membranes of cells in 27
which these receptors are either present in low number or missing. Such an approach has already been accomplished with the human placental insulin receptor, which has been introduced into rodent CHO cells (which have a low number of insulin receptors) and has been shown to respond to added insulin by greatly increasing the transport of glucose into the cells (Ellis et al., 1986~. However the metabolic consequences of altering the uptake of glucose in these cells was not considered. There currently is a unique opportunity to extend the usefulness of studies of this type into the area of metabolism and nutrition. THE NEED FOR NEW EDUCATIONAL PROGRAMS COMBINING METABOLISM AND MOLECULAR BIOLOGY The important advances that have occurred in molecular and cellular biology over the past 15 years have great potential for metabolic and nutritional research. A core technology centered around the ability to introduce specific genes into cells and animals in a directed and regulatable fashion is now developing. This technology will be generic in its usefulness in-biology and medicine and will extend from the possible correction of genetic defects to the genetic patterning of cells and animals for metabolic studies. The field of nutrition has a vast potential as a research area in which this technology will find application. As discussed above, it will soon be possible to alter specific steps in a complex metabolic pathway in intact animals to determine the animals' responses to individual dietary components or to target the expression of hormone receptors to tissues that are normally unresponsive to a specific hormone. For the potential of these new approaches to be fully realized in the nutritional sciences, a new generation of investigators who are familiar with molecular biology, as well as with metabolism and nutrition, must be trained. Without these individuals, the potential application of many of these techniques will be needlessly delayed. Unfortunately, among graduate students who are attracted to the fast pace and high visibility of molecular genetics, there is a diminished interest in metabolism and nutrition as a research area. There is also a gulf between faculty in the nutritional sciences and those in molecular biology in most universities. Two cultures 28
have arisen that have different languages, different traditions, and different approaches to science. Unfortunately, these differences continue to grow. The information contained in this volume based on the symposium "Frontiers in the Nutrition Sciences" could make an important and timely contribution to the development of a program for education in nutrition. We hope that the nutritional sciences will share in the excitement of the new advances in molecular and cellular biology by participating in a revitalization of metabolic research, the basis for all nutritional science. REFERENCES Ballard, F.J., and R.W. Hanson. 104:866-871. Bernstein, A., S. Berger, D. Huszar, and J. Dick. 1985. P. 235 in J. Stelow and A. Hollaendis, eds. Genetic Engineering: Principles and Methods, Vol. 7. Plenum, New York. 1967. Biochem. J. Camper, S.A. 1987. Biotechnology 5:638-650. Deschatrette, J., and M.C. Weiss. 1974. Biochemic 56:1603-1611. Ellis, L., De O. Morgan, D.E. Koshland, E. Clauser, G.R. Moe, G. Bollay, R.A. Roth, and W.J. Rutter. 1986. Proc. Natl. Acad. Sci. USA 83:8137-8141. Gelboa, E., M.A. Eglitis, P.W. Kantoff, and U.F. Anderson. 1986. Biotechnology 4:504-512. Goodridge, A.G., and R.W. Hanson, eds. 1986. Metabolic Regulation: Applications of Recombinant DNA Techniques. Annals of the New York Academy of Sciences, Vol. 478. N.Y. Academy of Sciences, New York. Granner, D., E. Andreone, K. Sasaki, 1983. Nature 305:549-551. Green, A.A., and C. Cori. 1943. J. Biol. Chem. 151:21-28. Hammer, R.E., G.H. Swift, D.M. Ornitz, C.J. Quaife, R.D. Palmiter, R.L. Brinster, and R.J. MacDonald. 1987. Mol. Cell. Biol. 7:2956-2967. Hanson, R.W., and A.J. Garbers. 1972. Am. J. Clin. Nutr. 25:1010-1021. Hanson, R.W., and M.H. Mehlman. 1976. Gluconeogenesis: Its Regulation in Mammalian Species. John Wiley & Sons, New York. and E. Beale. 29 Am. J. Clin
Hod, J., S. Morris, and R.W. Hanson. 1984a. J. Biol. Chem. 259:15603-15608. Hod, Y., H. Yoo-Warren, and R.W. Hanson. 1984b. J. Biol. 259:15609-15614. Hughes, S., and E. Kosik. 1984. Urology 136:89-99. Johnson, D., K. Hasse, R. Mulligan, and R. Jaenisch. 1985. Proc. Natl. Acad. Sci. USA 82:6927-6931. Jones, C.T. 1982. The development of the metabolism in the fetal liver. In C.T. Jones, ed. The Biochemical Development of the Fetus and the Neonate. Elsevier Biomedical Press, Amsterdam. Korman, A.J., J.D. Frantz, J.L. Strominger, and R.C. Mulligan. 1987. Proc. Natl. Acad. Sci. USA 87:2150-2154. Krebs, E.G., and E.H. Fisher. 1956. J. Biol. Chem. 216:121-128. l~merS, W.H., R.W. Hanson, and H.M. Meisner. 1982. Proc. Natl. Acad. Sci. USA 79:5137-5141. Magli, M.C., J.E. Dick, D. Huszar, and A. Bernstein. 1987. Proc. Natl. Acad. Sci. USA 84:789-793. Magnuson, M.A., P.G. Guinn, and D.K. Granner. 1987. J. Biol. Chem. 262:14917-14920. Mann, R., R.C., Mulligan, and D. Baltimore. 1983. Cell 38:153-159. McGrane, M.M., J. DeVente, J. Yun, J. Bloom, E. Park, A. Wynshaw-Boris, T. Wagner, F.M. Rottman, and R.W. Hanson. 1988. J. Biol. Chem. 263:11443-11451. McSvor, R.S., M.J. Johnson, A.D. Miller, S. Pitts, S.R. Williams, D. Valerio, D.W. Martin, and I.M. Verma. 1987. Cell. Mol. Biol. 7:838-846. Meisner, H.M., W.H. Lamers, and R.W. Hanson. 1983. Trends Biochem. Sci. 8:165-167. Morgan, H.E., and A. Parmeggiani. 1964. J. Biol. Chem. 239:2440-2448. Nordlie, R.C., and H.A. Lardy. 1963. J. Biol. Chem. 238:2259-2263. Quaife, C.J., C.A. Pinkert, D.M. Orintz, R.D. Palmiter, and R.L. Brinster. 1987. Cell 48:1023-1034. Robertson, E., A. Bradley, M. Kuehn, and M. Evans. 1986. Nature 323:445-448. Roesler, W., G.R. Vandenbark, and R.W. Hansen. 1989. J. Biol. Chem. 264:9657-9664. Short, J.M., S. Wynshaw-Boris, H.P. Short, and R.W. Hanson. 1986. J. Biol. Chem. 261:9721-9726. Soriano, P., and R. Jaenisch. 1986. Cell 46:19-29. 30
Sprang, S., E. Goldsmith, and R. Fletterick. 1987. Science 237:1012-1019. Tilghman, S.M., R.W. Hanson, L. Reshef, M.F. Hopgood, and I.J. Ballard. 1974. Proc. Natl. Acad. Sci. USA 71:1304-1308. Utter, M.F., and K. Kurahaski. 1954. J. Biol. Chem. 207:787-802. Van der Putten, H., F.M. Botteri, D.A. Miller, M.G. Rosenfeld, H. Fan, R.M. Evans, and I.M. Verma. Proc. Natl. Acad. Sci. USA 82:6148-6152. 1985. Wagner, T.E., P.C. Hoppe, J.D. Jollick, D.R. Scholl, R. Hodinka, and J.G. Gault. 1981. Proc. Natl. Acad. Sci. USA 78:6376-6380. Uatford, M., Y. Hod, M.F. Utter, and R.W. Hanson. 1981. J. Biol. Chem. 256:10023-10027. Williamson, J.R. 1976. R.W. Hanson and M.A. Mehlman, eds. Gluconeogenesis: Its Regulation in Mammalian Species. Wiley Interscience, New York. Wynshaw-Boris, A., T.G. Lugo, J.M. Short, R.E.K. Fournier, and R.W. Hanson. 1984.: J. Biol. Chem. 2S9: 12161-12169. Wynshaw-Boris, A., J.M. Short, and R.W. Hanson. 1986a. Biotechnology 4:104-119. Wyushaw-Boris, A., J.M. Short, D.S. Loose, and R.W. Hanson. 1986b. J. Biol. Chem. 261:9714-9720. 31
