![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
The Journal of Nutrition Vol. 128 No. 11 November 1998,
pp. 2052-2057
Department of Pathology and Laboratory Medicine, University of Cincinnati College of Medicine, Cincinnati, OH 45267-0529
The importance of many dietary constituents in maintenance of health is obvious. While deficiencies in dietary intake of specific nutrients may be detrimental to growth, reproduction and immunity, excessive amounts of specific nutrients in the diet can also lead to disease states. For example, increased intake of dietary fat and cholesterol is associated with hyperlipidemia and an increased risk of coronary heart disease (Keys 1970 Advances in molecular biology techniques during the past decade have led to an explosion of research aimed at understanding diet and gene interactions in health and diseases. While most of the earlier work focused on nutrient regulation of gene expression, transgenic technology has also been used to study the effect of overexpression of specific genes in modulation of dietary nutrient effects and on the metabolism of dietary components as they relate to normal health and pathogenesis of diseases. More recently, targeted gene inactivation, commonly known as gene knockout, has been employed to study the functional importance of specific genes and the impact of specific genetic mutations and deletions on complex metabolic processes which ultimately lead to various diseases (Melton 1990 Experimental approaches to modulate gene expression in vivo
INTRODUCTION
Introduction
References
). Excessive intake of vitamin D has also been shown to result in soft-tissue calcification and renal calculi. However, it must be noted that while the correlation exists between increase nutrient uptake and specific diseases, the response of a given individual is quite variable. These individual variations in dietary responsiveness are likely due to the different genetic composition of each individual. Numerous genetic factors are involved in determining responsiveness to specific dietary nutrients. These include genes important for nutrient absorption as well as those important for the metabolism and processing of the nutrient in the diet. Additionally, the amount of each nutrient in the diet also has an impact on the level of specific gene expression. Such regulatory mechanisms may also account for individual differences in susceptibility to diet-induced diseases.
). This review provides an overview of the utility and importance of the gene ablation technology to nutritional and metabolic research.
, Thomas and Capecchi 1987). Alternatively, antisense DNA or RNA, which inhibit gene expression by complementation to single-stranded mRNA (Izant and Weintraub 1985) and ribozymes (Haseloff and Gerlach 1988), which catalyze RNA hydrolysis in a sequence-specific manner, have also been used successfully to abolish gene expression in mammalian cells. However, technical difficulties in obtaining high level stable expression of antisense nucleotides and ribozymes have limited the usefulness of these approaches. Most of the research with antisense nucleotides and ribozymes were restricted to in vitro cell culture systems and, thus, were of limited value to nutritional and metabolic studies. In contrast, targeted gene disruption by homologous recombination in embryonic stem cells have been employed widely to produce animal models with specific gene deletions. Many of these models are quite useful for nutritional and metabolic research. Therefore, this review will focus on the use of this technique.
). In a typical experiment, the ES cells and the recipient embryo are obtained from animals carrying genes of different coat colors, such that the initial selection of chimeric mice can be based on the coat color of the offspring. The most commonly used ES cells to date are those derived from the mouse strain 129, which has an agouti coat color. These ES cells can then be microinjected into embryos obtained from C57BL/6J mice, which have a black coat color. Offspring with a high degree of agouti coat color, indicating the transmission of ES cell-derived genes, can then be crossbred with each other to obtain animals with a genetic background identical to that of the ES cells. Using this approach, mice with specific gene modifications can be obtained by manipulation of the ES cell genome.
). Although such targeted recombination is a rare event in comparison with nonhomologous integration of the transfected DNA, methodology has been developed to optimize the chance of homologous recombination and to rapidly screen and select ES cells in which such event has taken place. Most of the experiments to date utilized isogenic DNA for the targeting construct to maximize hybridization of the targeting DNA to the proper gene locus in the chromosome (Deng and Capecchi 1992). Furthermore, a minimum of 4 kB DNA was found to be required for effective homologous recombination to take place. A selectable gene marker, such as the neomycin-resistant gene, would then be inserted into an exon to disrupt the coding sequence of the gene of interest. The chimeric targeting gene construct can be used to transfect ES cells. Homologous recombination of the transfected DNA with chromosomal DNA at the target locus will result in the replacement of a portion of the endogenous gene with the targeting construct, thus disrupting the coding sequence and inactivation of the endogenous gene. The use of a selectable gene marker allows the selection for cells that have taken up and expressed the transfected DNA. Growth of the ES cells in the presence of antibiotic selection indicates the integration of the transfected DNA into the ES cell genome. In a successful experiment, approximately 0.01-0.001 of the antibiotic-resistant cells would have the transfected DNA targeted to the proper gene locus, while the remaining cells would have incorporated the DNA in a nonhomologous site. In some cases, investigators have included a negative selectable gene marker at the 5' or 3' end of the targeted construct to allow for selection against random insertion events. Homologous recombination at the targeted gene locus would result in deletion of the negative selectable gene marker, while integration at nonhomologous sites would have included this marker in the genome. The inclusion of a negative selection marker usually results in an additional 10-fold enrichment of homologous recombination clones.
Utility of gene knockout mice to explore physiologic functions and produce animal models for human disease
The utilities for production of gene knockout mice are many-fold. First, gene knockout mice can be produced as an animal model for human diseases. These gene knockout mice can then be used to explore potential intervention strategy, including pharmacological and genetic therapy approaches, for treatment of specific metabolic diseases. A classical example is the production of LDL receptor gene knockout mice as an animal model for Type II Familial Hypercholesterolemia (FH) (Ishibashi et al. 1994). Patients with this genetic disorder suffer coronary heart disease at an early age. The disease is characterized by severely elevated plasma LDL and cholesterol levels due to defective LDL receptor gene expression. The production of the LDL receptor gene knockout mice resulted in a rodent model that also displayed severe hypercholesterolemia and atherosclerosis when fed a Western type diet (Ishibashi et al. 1994). The availability of the LDL receptor knockout mice also provided investigators with the opportunity to assess the role of the LDL receptor in chylomicron remnant metabolism. Results of these studies revealed that while the initial hepatic removal of chylomicron remnants is independent of the LDL receptor, endocytosis of the majority of these lipoproteins requires the presence of functional LDL receptors on the hepatic cell membrane (Herz et al. 1995). The characteristics of the LDL receptor knockout mice are dramatically different from those observed in genetically unaltered mice, but are clinically and symptomatically identical to those observed in Type II FH human subjects (Ishibashi et al. 1994). Thus, the LDL receptor knockout mice can also be used as a model for the human disease. In fact, recent studies have begun to use this animal model to test for the efficacy of several genetic therapy approaches as potential treatment for Type II Familial Hypercholesterolemia. Early results from these studies are encouraging as hypercholesterolemia in the LDL receptor knockout mice could be overcome by intravenous injection of recombinant adenovirus encoding the human LDL receptor gene (Ishibashi et al. 1993).
, Zhang et al. 1992). Previous studies showed an elevation of plasma apoE level in animals when fed an atherogenic high fat/high cholesterol diet. The increase in apoE level appears to correlate with development of atherosclerotic lesions in several cholesterol-fed animal models. While these studies suggest a potential contributory role of apoE in atherogenesis, in vitro cell culture experiments showed that this apolipoprotein facilitates cholesterol efflux from lipid-laden macrophages/foam cells and increases catabolism of chylomicron remnants and cholesterol-rich VLDL (Mahley 1988). Thus, the in vitro studies suggest that apoE may serve a protective role against atherosclerosis, and its increased level after cholesterol feeding may represent a systematic response to the diet. The exact role of apoE in atherosclerosis was clarified by production and characterization of apoE knockout mice in two separate laboratories. Results of both studies indicated that apoE gene deletion resulted in an animal with severe atherosclerosis, thus confirming the protective role of this apolipoprotein against atherosclerosis (Plump et al. 1992, Zhang et al. 1992). Significantly, and in contrast to genetically unaltered mice in which even the "susceptible" strains only develop fatty streak lesions in response to an atherogenic diet, the apoE knockout mice developed atherosclerosis with severity similar to that observed in human patients (Plump et al. 1992, Zhang et al. 1992).
4 integrin and ICAM-1, but not the E-selectin, in macrophage recruitment to the atherosclerotic plaque (Patel et al. 1998). Other studies have also used the apoE knockout mice to show the role of inflammatory and immune response in the atherosclerotic process (Zhou et al. 1998). Important to investigators in nutritional research is the demonstration of the presence of oxidation specific epitopes in atherosclerotic lesion areas of apoE knockout mice (Palinski et al. 1994). This study not only confirmed the role of oxidation in atherogenesis, but the results also indicate that this animal model can be used to evaluate the potential value of various antioxidants, including vitamin E, as an intervention strategy against atherosclerosis.
Potential species differences in extrapolating data from knockout mouse studies to human health issues
Although the generation of gene knockout mice is clearly a valuable tool for producing a rodent model of human disease and to evaluate physiologic functions of specific genes, it must be noted that not all the knockout mice will necessarily produce a predictable phenotype similar to those observed in human subjects. For example, HDL and apoAI levels are negatively correlated to atherosclerosis in human subjects (Miller 1984). While high fat, high cholesterol feeding in selected mouse strains resulted in atherosclerosis and a concomitant decrease in apoAI (Ishida and Paigen 1989), suggesting that reduced apoAI level is a major risk factor for coronary heart disease, apoAI gene ablation in the mouse model resulted in plasma HDL deficiency without a parallel increase in atherosclerosis severity (Li et al. 1993, Williamson et al. 1992). These results suggest that a compensatory gene that serves similar functions as apoAI gene may exist in the mouse. Alternatively, in this animal model, apoAI may interact with other genetic factors in conferring susceptibility to atherosclerosis. Whether these possibilities also exist in human subjects remain unknown. Finally, species differences in metabolism and physiology may also explain some of the discrepancies in these results.
), mice with the apoB-null genotype died in utero (Farese et al. 1995, Huang et al. 1995). In contrast to humans, some apoB(+/
) heterozygotes developed severe neural tube defects shortly after gestation and the male apoB(+/) mice were infertile (Huang et al. 1995). Significantly, the apoB(+/) mice produced in one laboratory were resistant to diet-induced hypercholesterolemia (Farese et al. 1995), suggesting the involvement of the apoB protein in fat and cholesterol absorption through the intestine. However, apoB(+/) mice generated in a different laboratory did not show any abnormalities in lipid absorption, and their plasma lipoproteins responded normally to a high fat, high cholesterol challenge (Huang et al. 1995). Although several factors may account for the different results observed in the two laboratories, it is noteworthy that the genetic background of the mice reported in the two studies are different which may account for the different results.
Importance of genetic background in evaluating gene defects in physiology and pathophysiology
The importance of the genetic background in evaluating the effect of gene ablation in physiologic functions is best illustrated by comparing the angiotensinogen (AGT) knockout mice generated in two laboratories (Kim et al. 1995; Tanimoto et al. 1994). Tanimoto et al. (1994) used ES cells derived from a [C57BL/6 × CBA] F1 embryo for their genetic manipulation experiments and then bred their chimeric mice with outbred ICR mice to obtain heterozygous and homozygous AGT knockout mice. Analysis of their animals revealed that AGT gene disruption had no influence on the blood pressure of these animals. Unfortunately, ability to detect heterozygous effects are limited due to the mixed genetic background of the offsprings. In contrast, Smithies laboratory generated AGT gene knockout mice and mated their chimeras to a defined genotype (Kim et al. 1995). Their results showed a direct correlation between AGT gene copy number and arterial blood pressure. The latter study revealed that the plasma AGT level in their agt(+/) animals was 35% of normal, significantly less than the 50% level expected from their genotype. Thus, their results clearly indicated the involvement of a second gene. An interpretation that is consistent with the results with agt(/) mice with mixed genetic background.
). Previously, a possible role of the cholesterol esterase in mediating dietary cholesterol absorption was proposed. However, this hypothesis has been controversial due to different results obtained from various laboratories. Early studies with isolated intestinal cells showed that exogenously added cholesterol esterase increased cholesterol uptake by enterocytes, thus suggesting its role in cholesterol absorption (Bhat and Brockman 1982). However, one study reported normal cholesterol absorption in pancreatic-diverted, and thus cholesterol esterase-depleted, rats (Watt and Simmonds 1981). Another study showed that cholesterol absorption efficiency was reduced in pancreatic-diverted rats and this function could be restored by reinfusion of pancreatic juice with cholesterol esterase, but not juice devoided of cholesterol esterase (Gallo et al. 1977 and 1984). The discrepancies between these results could not be resolved by over 15 years of intense in vivo or in vitro studies. The use of specific inhibitors raised questions regarding the specificity of the inhibitor for this protein. The in vitro studies incubating an enterocyte-like tissue culture cell line with cholesterol in the presence or absence of cholesterol esterase also led to contradictory results. Although differences in experimental conditions may partially explain some of the discrepancies, no cell culture system is ideal for testing complex physiological issues such as the mechanism of nutrient transport. Thus, we have chosen to produce an animal model with defective expression of the cholesterol esterase gene and then study the physiologic function of this protein in nutrient uptake.
Candidate gene approach to identify proteins involved with a complex physiological trait
View this table:
Table 1.
Efficiency of Cholesterol Esterase Gene Targeting
in Embryonic Stem Cells
View this table:
Table 2.
Chimeric Animals from Cholesterol Esterase
Targeted ES Cells
-[14C]sitosterol by gavage. The ratio of the two radiolabels excreted in the feces over a 24-h period was similar in the control and cholesterol esterase-null mice. In contrast to these results, when experiments were performed with [3H]cholesteryl oleate instead of [3H]cholesterol, a higher amount of the 3H radiolabel was found excreted in feces and dramatically less of the radiolabel was detected in the serum of the cholesterol esterase-null mice in comparison with that detected in control animals. Thus, cholesterol esterase gene deletion resulted in decrease absorption of esterified cholesterol and has no effect on free cholesterol absorption through the gastrointestinal tract. Thus, our results indicate that cholesterol esterase is responsible for mediating intestinal absorption of cholesteryl esters but does not play a primary role in determining the total amount of unesterified cholesterol absorbed from a single bolus meal (Howles et al. 1996).
). The genes and gene products involved with the cholesterol absorption process can only be inferred from cell culture studies as well as studies with metabolic inhibitors. However, results of studies using the latter approaches may not always reflect physiological situations and candidate genes identified from these studies need to be verified in vivo. The cholesterol esterase studies described previously clearly illustrated the utility of gene knockout mice to test the physiological importance of candidate genes.
, Largis et al. 1989). The cloning of the first ACAT gene by Chang and his colleagues (Chang et al. 1993) led to the production of a gene knockout mouse with defective expression of this cholesterol esterification enzyme (Meiner et al. 1996). Cholesterol absorption efficiency was shown to be similar between wild-type and ACAT-null mice (Meiner et al. 1996). Interestingly, although cholesterol esterification activity in the adrenals and macrophages is reduced in the ACAT knockout mice, cholesterol esterification activity in the liver of these animals is within the normal range (Meiner et al. 1996). These results suggested that another form of ACAT may be active in the liver and intestine. Indeed, more than one ACAT enzyme has now been identified in yeast (Yang et al. 1996, Yu et al. 1996) and in mammals (Meiner et al. 1997). Whether these other ACAT enzymes play a role in mediating cholesterol absorption through the gastrointestinal tract remains to be determined. Regardless of the results of the cholesterol absorption studies, the production of knockout mice with defective expression of the various ACAT genes will provide valuable tools to identify the precise role of each of these ACAT enzymes in cholesterol metabolism. Readers with an interest in the ACAT gene family are referred to an excellent review article on this topic for addition information (Farese 1998).
, Schwarz et al. 1996). The ablation of the multidrug resistance gene 2, which encodes a P-glycoprotein, also yielded animals with reduced cholesterol absorption efficiency. The decrease in cholesterol absorption in the mdr2-null mice was shown to be related to the abnormal hepatic secretion of phospholipids to bile (Oude Elferink et al. 1996, Smit et al. 1993, Voshol et al. 1998). Thus, both of these animal models documented the importance of lipid composition in the bile in determining the efficiency of cholesterol transport through the gastrointestinal tract. Other enzymes that are involved with modifying lipid structure and composition in the intestinal lumen have also been suggested to play a significant role in mediating lipid and cholesterol absorption efficiency. These enzymes include the pancreatic lipase (Jeppesen 1997, Wickham et al. 1998) and phospholipase A2 (Homan and Hamelehle 1998, Mackay et al. 1997). The physiologic importance of these enzymes in dietary lipid absorption can be tested using the gene knockout approach.
Manuscript received 1 June 1998. Initial reviews completed . Revision accepted 3 August 1998.
, Hotamisligil et al. 1996, Peters et al. 1997) and diabetes (Katz et al. 1995, Terauchi et al. 1995, Tsao et al. 1997) that are useful for studying diet-gene interactions in these metabolic disorders. Selected examples of the more recently produced knockout mouse models that are relevant to nutritional research are listed as references on Table 3. A more complete listing of currently available knockout mice can be found at several Web Sites with the following addresses: (1) http://www.bioscience.org/knockout/knochome.htm; (2) http://www.ornl.gov/TechResources/Trans/hmepg.html; and (3) http://www.bis.med.jhmi.edu/Dan/tbase/tbase.html.
View this table:
Table 3.
Selected Examples of Recent Knockout Mouse Models for Nutritional Research
1
Supported by National Institutes of Health Grants DK-40917 and DK-46405.
FOOTNOTES
2
The costs of publication of this article were defrayed in part
by the payment of page charges. This article must therefore be hereby marked "advertisement"
in accordance with 18 USC section 1734 solely to indicate this fact.
3
Abbreviations used: ACAT, acyl-CoA cholesterol acyltransferase; AGT, angiotensinogen; apo, apolipoprotein; ES cells, embryonic stem cells; FH, Familial Hypercholesterolemia; ICAM, intracellular cell adhesion molecule; mdr, multidrug resistance.
| |
LITERATURE CITED |
|---|
Introduction
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  K. L. Donnelly, M. R. Margosian, S. S. Sheth, A. J. Lusis, and E. J. Parks Increased Lipogenesis and Fatty Acid Reesterification Contribute to Hepatic Triacylglycerol Stores in Hyperlipidemic Txnip-/- Mice J. Nutr., June 1, 2004; 134(6): 1475 - 1480. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
