Which of the following is the basis for the intestine specific expression of apoprotein b-48?

Murielle M. Véniant, Edward Kim, Sally McCormick, Jan Borén, Lars Bo Nielsen, Martin Raabe, Stephen G. Young, Insights into Apolipoprotein B Biology from Transgenic and Gene-Targeted Mice, The Journal of Nutrition, Volume 129, Issue 2, February 1999, Pages 451S–455S, https://doi.org/10.1093/jn/129.2.451S

Abstract

The B apolipoproteins (apo),4apoB48 and apoB100, are key structural proteins required for the assembly of triacylglycerol-rich lipoproteins; they are constituents of all classes of lipoproteins that are considered atherogenic (Young 1990). Apolipoprotein B100 is synthesized by the liver, where it is necessary for the formation of VLDL, and by the yolk sac during development, where it appears to play a role in the nutrition of the embryo (Farese et al. 1995 and 1996a,). As discussed in this short review, apoB100 is also made by cardiac myocytes. Apolipoprotein B48 is synthesized in the intestine as a result of an apoB mRNA-editing process that converts codon 2153 to a translational stop codon (Davidson 1994). Apolipoprotein B mRNA is extensively edited in the intestines of all mammals (Greeve et al. 1993). Thus, apoB48 is the principal structural protein of intestinally derived lipoproteins. Apolipoprotein B mRNA editing activity occurs in the livers of some mammalian species, including mice and rats (Greeve et al. 1993). Thus, the liver of these species secretes both apoB48 and apoB100.

Over the past five years, several laboratories have used gene-targeted mice as well as conventional transgenic mice to investigate various aspects of apoB biology. For example, human apoB transgenic mice have been used to create animal models to study atherogenesis (Callow et al. 1994, Linton et al. 1993a, Purcell-Huynh et al. 1995) and to study the DNA sequences that control the expression of the apoB gene (Nielsen et al. 1998a and 1998b,). Transgenic mice expressing mutant forms of human apoB have also been used to study the structure/function of the apoB molecule (Callow and Rubin 1995, McCormick et al. 1995, 1996 and 1997,). Gene knockout mice have been used to create animal models for the human apoB deficiency syndromes (Farese et al. 1995, Homanics et al. 1993, Huang et al. 1995) and to study the “physiologic rationale” for the existence of apoB48 and apoB100 in mammalian lipoprotein metabolism (Farese et al. 1996b). This review will not attempt to describe all of these studies but rather will focus on three recent examples of how the use of genetically modified mice has contributed to our understanding of apoB.

GENERATING AUTHENTIC MOUSE MODELS FOR THE HUMAN APO B DEFICIENCY SYNDROME, FAMILIAL HYPOBETALIPOPROTEINEMIA

Familial hypobetalipoproteinemia (FHb) is a human apoB deficiency syndrome caused by mutations in the apoB gene that interfere with the synthesis of the apoB molecule (Linton et al. 1993b). Most of the mutations causing hypobetalipoproteinemia have been point mutations, either deletions of one or two nucleotides or single nucleotide substitutions causing nonsense mutations (Linton et al. 1993b). These mutations often lead to the synthesis of a truncated form of apoB, which can be detected in the plasma lipoproteins. In these cases, the concentration of the truncated apoB in the plasma is invariably quite low. For example, in heterozygotes, the concentration of the truncated apoB is typically <5% of the concentration of apoB100 that is produced by the wild-type apoB allele. One of the goals of our laboratory has been to understand the basic mechanisms for the low plasma levels of the truncated apoB proteins in these syndromes.

To create a mouse model of FHb, we used gene targeting to insert an apoB83 mutation (Leu3798Stop, a mutation typical of those causing human FHb) into the mouse apoB gene (Kim et al. 1998). In addition to the Leu3798Stop mutation, the apoB83-only (Apob83) allele contained a mutation at codon 2153 that abrogated apoB48 formation. The latter mutation had no affect on apoB mRNA levels (Farese et al. 1996b). Unlike the apoB70 mutation described earlier by the laboratory of Dr. Nobuyo Maeda (Homanics et al. 1993), the apoB83-only allele did not alter the structure of the apoB transcript. Unlike the “apoB48-only” mutation generated earlier by our laboratory (Farese et al. 1996b), the apoB83 nonsense mutation was inserted into an “unnatural” site within the apoB gene (i.e., one that did not lead to the production of a physiologically normal apoB protein).

In mice heterozygous for the Apob83 allele, apoB83 was present only in trace levels in the plasma (∼2% of the level of apoB100 that was produced by the other allele), a phenotype that is strikingly similar to that observed in human FHb heterozygotes with an apoB83 mutation (Farese et al. 1992). An analysis of the plasma lipoproteins in the heterozygous apoB83-only mice revealed that apoB83 was present in low concentrations in the VLDL but was absent in the LDL—a pattern identical to the distribution of apoB83 in human FHb heterozygotes (Farese et al. 1992).

An analysis of heterozygous apoB83-only mice uncovered dual mechanisms for the low plasma levels of apoB83 (Kim et al. 1998). First, the Apob83 mRNA levels and apoB83 secretion by primary hepatocytes were reduced by ∼75%. Thus, unlike the nonsense mutation in the apoB48-only mice (Farese et al. 1996b), a nonsense mutation at this “unnatural” site within the apoB gene caused low apoB mRNA levels and correspondingly low apoB secretion rates. Of note, the Apob83 pre-mRNA levels were not reduced (Kim et al. 1998). Thus, we suspect that the Apob83 nonsense mutation affects either the stability or the processing of the Apob transcript. There are a number of precedents for low mRNA levels associated with nonsense mutations in other genetic diseases (Baserga and Benz 1988 and 1992, Kessler and Chasin 1996, Maquat 1995).

In addition to low synthesis rates, apoB83 was removed very rapidly from the plasma, much more rapidly than apoB100 (Kim et al. 1998). We estimate that the absolute concentration of apoB83 in the plasma of heterozygous apoB83-only mice is only ∼2% of the concentration of apoB100, despite the fact that apoB83 synthesis rates are ∼25% as high as the apoB100 synthetic rates. The rapid clearance of apoB83 was documented by genetic experiments in which the apoB83 mutation was placed on a background of apoE deficiency (Piedrahita et al. 1992) or LDL receptor deficiency (Ishibashi et al. 1993) and by metabolic experiments in which we blocked lipoprotein degradation with Triton WR-1339 (Li et al. 1996).

There are several explanations for the rapid clearance of apoB83. First, apoB83-containing lipoproteins, like apoB48-containing lipoproteins, might accommodate more apoE, which could be partially responsible for the rapid clearance of the apoB83-containing lipoproteins. However, this probably does not represent the entire story. Apolipoprotein B83 contains the portion of the apoB molecule that interacts with the LDL receptor, and we believe that apoB83 is likely to be more effective than apoB100 in mediating the uptake of LDL and VLDL by the LDL receptor. Enhanced uptake of apoB83-containing VLDL would also explain the extremely low levels of apoB83 in the plasma and why apoB83 is virtually absent from the LDL fraction of Apob83/+mice (and human apoB83 heterozygotes). Extremely rapid apoB83 clearance by the LDL receptor pathway would be consistent with data from Dr. G. Schonfeld′s laboratory (Parhofer et al. 1990); they found that apoB89 (a truncated apoB molecule in the plasma of a kindred with FHb) was cleared much more rapidly than apoB100.

Homozygous Apob83/83 embryos manifested multiple developmental abnormalities, including exencephalus and herniation of the liver into the umbilical cord; none survived more than a few days (Kim et al. 1998). Drs. R. Farese, R. Hamilton and co-workers proposed that apoB synthesis and secretion by the yolk sac plays a key role in mouse development (Farese et al. 1996a). The fact that Apob83/83 mice display severe developmental abnormalities indicates that the quarter-normal apoB synthesis rates fall below the threshold required for normal mouse development. Alternatively, the peculiar intrinsic metabolic properties of apoB83 (particularly the extremely rapid uptake of apoB83 by the LDL receptor pathway) could “short-circuit” normal embryonic lipoprotein delivery (e.g., prevent apoB83-containing lipoproteins from reaching critical sites within the developing mouse embryo).

In humans, the FHb phenotype is somewhat variable, depending on the length of the truncated apoB. The plasma levels of apoB37 in human apoB37 heterozygotes are probably about 10-fold greater than the plasma levels of apoB83 in human apoB83 heterozygotes. To obtain a broader view of metabolic mechanisms in FHb, we have used gene-targeting techniques to introduce another apoB gene mutation (Asn1785Stop) into the mouse genome, thereby generating apoB39-only mice. Interestingly, the plasma levels of apoB39 in heterozygous apoB39-only mice (Apob39/+ mice) were 10- to 15-fold higher than the plasma levels of apoB83 in Apob83/+ mice (E. Kim and S. Young, unpublished observations). In addition, many of the homozygous apoB39-only mice were born and appeared to be completely healthy, in contrast to the homozygous apoB83-only mice. During the next year, we will compare the phenotypes of our apoB39-only and apoB83-only mice and attempt to understand why the plasma levels of the truncated apoB are higher in the apoB39-only mice and why they do not exhibit the severe neurodevelopmental abnormalities.

USING LARGE GENOMIC CLONES TO GENERATE MUTANT FORMS OF HUMAN APO B

About 5 years ago, our laboratory and that of Dr. E. Rubin identified and mapped a P1 bacteriophage clone (p158) that spanned the entire human apoB gene and contained ∼19 kb of 5′ flanking sequences and 17.5 kb of 3′ flanking sequences. We used that clone to generate human apoB transgenic mice (Callow et al. 1994, Linton et al. 1993a). In transgenic lines with >10 copies of the transgene, the plasma levels of human apoB in hemizygous mice fed a standard laboratory diet were 60–80 mg/100 mL, which are similar to those in normolipidemic humans (Linton et al. 1993a). The development of the p158-human apoB transgenic mice represented a breakthrough as a “human apoB expression system.”

One of our primary goals was to transform this “human apoB expression system” into a system that would be useful for studies of apoB structure and function. In particular, we were interested in understanding the LDL receptor–binding domain and in defining the features of the apoB molecule that are important for its interaction with apo(a) in the assembly of lipoprotein(a) [Lp(a)]. To analyze the structure of human apoB, we wanted to be able to insert mutations into clone p158, express the mutant p158 construct in transgenic mice and then analyze the properties of the mutant human apoB protein.

We began our studies of apoB structure with a discrete issue, i.e., to define the structural features of the apoB molecule that are important for Lp(a) assembly. To address this issue, we initially interrupted the coding sequence of p158 with a transposon and then used the mutant p158 to create transgenic mice expressing a truncated form of apoB, apoB90. An analysis of the human apoB90 revealed that it was completely incapable of forming Lp(a), once again suggesting that the carboxyl terminal 10% of the apoB molecule was essential for Lp(a) formation (McCormick et al. 1994).

The transposon mutagenesis approach allowed us to make many constructs coding for truncated apoBs, but did not allow us to introduce point mutations into p158. To overcome this obstacle, we sought to introduce subtle mutations into the human apoB gene by using homologous recombination in yeast artificial chromosome (YAC) (McCormick et al. 1995 and 1996,). For these studies, the human apoB gene (the insert from p158) was cloned into a YAC, and “pop-in, pop-out” gene targeting was used to introduce subtle mutations into the human apoB gene. The YAC DNA was then used to generate transgenic mice expressing high levels of the mutant human apoB (McCormick et al. 1995).

To determine the utility of the YAC gene-targeting approach, we began by testing the hypothesis that the last cysteine residue of apoB100, cysteine-4326, was involved in the disulfide bond with apo(a). We used the YAC gene-targeting system to change cysteine-4326 to a glycine and then generated transgenic mice expressing the mutant human apoB. Our studies with the mutant apoB protein demonstrated that it completely lacked the capacity to bind to apo(a) and form Lp(a) (McCormick et al. 1995 and 1996,). These studies indicated that apoB cysteine-4326 is the site of attachment for apo(a). Callow and Rubin (1995) obtained the same findings.

Recently, we performed additional experiments to define the apoB sequences, aside from cysteine-4326, that are important for Lp(a) assembly. To address this issue, we used the YAC gene-targeting approach to introduce nonsense mutations into exon 29 of the apoB gene, generating two truncated forms of human apoB, apoB95 (4330 amino acids) and apoB97 (4397 amino acids). We then tested the ability of those truncated apoB's to form Lp(a) (McCormick et al. 1997). Our studies revealed that Lp(a) was formed extremely slowly and inefficiently with apoB95, even though that molecule contained the critical cysteine residue (cysteine-4326). In contrast, Lp(a) was formed rapidly and efficiently with apoB97. From these studies, we concluded that the sequences carboxyl-terminal to cysteine-4326, particularly residues 4331–4397, play an important role in the initial interaction of apoB with apo(a).

SECRETION OF APO B–CONTAINING LIPOPROTEINS BY THE HEART

The critical role of lipoprotein assembly and secretion in the liver and intestine has been well documented (Havel and Kane 1995, Kane and Havel 1989, Young 1990). During our initial characterization of the p158-human apoB transgenic mice, we made a surprising and provocative observation, i.e., the human apoB transgene was expressed in the heart. This finding was confirmed by both RNA slot-blot studies and RNase protection assays in many independent lines of the p158 transgenic mice. The amount of apoB mRNA in the heart was ∼4% of that in the liver (Linton et al. 1993a). Heart expression of the human apoB transgene was confirmed in independent experiments by Callow et al. (1994).

Initially, we were extremely skeptical of the finding that the human apoB gene was expressed in the heart. First, clone p158 clearly did not contain all of the regulatory sequences required for the intestinal expression of the human apoB gene (Nielsen et al. 1997). Thus, p158 might have represented an “incomplete” genomic clone from the perspective of the regulatory sequences controlling proper tissue-specific gene expression (i.e., we hypothesized that p158 lacked the regulatory sequences that would normally serve to silence the apoB gene in the heart, in addition to lacking the regulatory sequences required for expression of the apoB gene in the intestine). Second, the heart seemed a very unlikely site for apoB gene expression. From the perspective of lipoprotein metabolism, the heart has always been viewed as a “consumer” of lipoprotein-derived lipids and plasma fatty acids. Along with skeletal muscle and adipose tissue, the heart is one of the principal sites for uptake of plasma free fatty acids (Goldberg 1996). At first glance, it would seem extremely counterproductive for the heart to express the apoB gene, inasmuch as apoB secretion would serve to transport lipids away from the heart.

Subsequent studies have shown that apoB gene expression in the heart is not simply a peculiarity of the transgenic mice generated with clone p158. We have recently analyzed apoB gene expression in more than three dozen transgenic lines developed with a large array of bacterial artificial chromosome (BAC) clones (Nielsen et al. 1997) modified by RecA-assisted restriction endonuclease-cleavage (L. Nielsen and S. Young, unpublished observations). In each of the BAC transgenic lines, we documented human apoB gene expression in the heart. Like the p158 human apoB transgenic mice, the human apoB mRNA levels in the BAC transgenic mice averaged 3–4% of those observed in the liver and ∼10% of those in the intestine (L. Nielsen and S. Young, unpublished observations). Interestingly, apoB gene expression in the hearts of transgenic mice does not require distant gene-regulatory sequences. Bacterial artificial chromosome constructs containing as little as 5 kb of 5′ flanking sequences and 1.5 kb of 3′ flanking sequences were expressed in the heart (L. Nielsen and S. Young, unpublished observations).

We considered the possibility that the expression of human apoB in the hearts of the p158- and BAC-transgenic mice might represent an artifact resulting from the expression of a fragment of human genomic DNA in the mouse. We therefore used RNase protection assays to assess the expression of the endogenous apoB genes in human and mouse hearts. To test this issue, we prepared RNA from nontransgenic mouse hearts and from human hearts obtained from cardiac transplantation operations. Of note, the apoB mRNA was easily detectable in RNA prepared from five different human hearts and was present at levels similar to those observed in the human apoB transgenic mice (Nielsen et al.1998). In addition, the mouse apoB gene was expressed in the hearts of nontransgenic mice (Nielsen et al.1998). Finding apoB gene expression in the human heart led us to suspect that the human apoB cDNA would be encountered in Expressed Sequence Tags (EST) databases. This suspicion was borne out; the human apoB cDNA has been identified in two different heart cDNA libraries (http://www.ncbi.nlm.nih.gov/dbEST). The only other tissues in which the apoB cDNA was identified were liver and intestine.

To localize the cell type responsible for cardiac expression of the apoB gene, we performed initial immunohistochemical studies with both human and mouse apoB-specific antibodies. These studies revealed very intense staining of capillary endothelium and much less intense staining of the cardiac myocytes (L. Nielsen and S. Young, unpublished observations). These preliminary results led us to hypothesize that apoB was synthesized and secreted by the capillary endothelium. However, two lines of evidence suggested that this interpretation was not correct. First, the laboratory of Dr. R. Hamilton performed ultrastructural studies on the capillary endothelium of both human and mouse heart tissue. They found that the caveolae of the endothelial cells (the putative transcytotic vesicles) contained numerous lipoproteins, whereas the Golgi apparatus from these cells contained none (L. Nielsen, R. Hamilton and S. Young, unpublished observations). Thus, the ultrastructural studies suggested that the staining of the endothelial cells in the immunohistochemical studies was due to the presence of plasma lipoproteins that were trapped within the endothelial cells. Second, and probably more importantly, in situ hybridization studies with a human apoB-specific riboprobe demonstrated apoB gene expression in the cardiac myocytes (Nielsen et al. 1998). In those studies, surrounding “non-cardiac” tissues (including the aorta, pulmonary artery and lungs) manifested no apoB gene expression, whereas tissues known to secrete apoB (the liver and the absorptive enterocytes of intestine) were intensely positive. Interestingly, the signal was more intense in atrial myocytes than in ventricular myocytes.

In the liver and intestine, the microsomal triglyceride transfer protein (MTP) gene is expressed in association with the apoB gene and is essential for lipoprotein assembly (Gordon et al. 1994). We suspected that the MTP gene might also be expressed in the heart. This suspicion has been confirmed. A Western blot of human heart microsomes revealed the presence of the 97-kDa fragment of MTP (Borén et al. 1998), and Northern blots revealed the presence of the MTP mRNA in mouse hearts as well as human hearts (Nielsen et al. 1998). The finding that the MTP gene is expressed in the heart is actually consistent with a much earlier observation by Wetterau and Zilversmit (1986). Even before the MTP cDNA was cloned, they had documented MTP activity in microsomes prepared from rat heart (at levels that were ∼5% as high as those observed in liver microsomes).

The expression of both the apoB and MTP genes in the heart suggested that the heart might actually be a lipoprotein-secreting organ. To test this hypothesis, heart tissue from several different lines of human apoB transgenic mice was metabolically labeled with [35S]methionine/cysteine, and the media were fractionated by sucrose density gradient ultracentrifugation. Apolipoprotein B was immunoprecipitated from each fraction and examined by SDS-PAGE and autoradiography. The human apoB transgenic mouse hearts secreted apoB-containing lipoproteins, predominantly in the LDL fraction (Borén et al. 1998). Metabolic labeling experiments were also performed on fresh human heart tissue (obtained from the explanted diseased hearts at the time of cardiac transplantation) and on the hearts of nontransgenic mice (Borén et al. 1998). In both cases, we documented the synthesis and secretion of apoB100-containing lipoproteins. Further, metabolic labeling experiments performed in the presence of an MTP inhibitor drug (Gordon et al. 1996) revealed that the secretion of apoB100-containing lipoproteins by the heart is completely dependent on MTP (Borén et al. 1998).

Recently, Dr. R. Hamilton and Ms. J. Wong performed ultrastructural studies on human heart myocytes using special staining techniques that permit the visualization of lipoproteins (Nielsen et al. 1998). They found examples of VLDL- and intermediate density lipoprotein (IDL)-sized lipoproteins within the Golgi apparatus of heart myocytes from two different human hearts. The lipoproteins were typically observed singly or in very small groups in the Golgi cisternae or vesicles. Lipoproteins were also identified within the Golgi vesicles of ventricular myocytes from human apoB transgenic mice.

The conservation of apoB and MTP expression in the hearts of humans and mice, two species separated by 80 million years of evolution, suggests that heart lipoprotein secretion is important. But what is the precise purpose of heart lipoprotein secretion? Given the well-established role of apoB in assembling triglyceride-rich lipoproteins, we are attracted to a simple hypothesis, i.e., that lipoprotein secretion unloads surplus myocyte lipids, particularly triglycerides, into the extracellular space (and eventually into the bloodstream). Although the delivery of lipids to the heart is subject to metabolic regulation, it is easy to imagine that changing metabolic conditions (such as ischemia or a sudden change in physical activity) might occasionally lead to an accumulation of surplus fatty acids or triglycerides. Thus, the ability to secrete lipoproteins into the bloodstream might serve to limit both the accumulation of potentially toxic fatty acid intermediates and local storage of triglycerides. This hypothesis, which might reasonably be designated a “reverse triglyceride transport hypothesis,” awaits testing.

We thank S. Ordway and G. Howard for reviewing the manuscript.

LITERATURE CITED

, & , , Jr (

Nonsense mutations in the human β-globin gene affect mRNA metabolism

Proc. Natl. Acad. Sci. U.S.A.

, & , , Jr (

β-Globin nonsense mutation: deficient accumulation of mRNA occurs despite normal cytoplasmic stability

Proc. Natl. Acad. Sci. U.S.A.

, , , & , (

Apo B100–containing lipoproteins are secreted by the heart

, & , (

Site-specific mutagenesis demonstrates that cysteine 4326 of apolipoprotein B is required for covalent linkage with apolipoprotein(a) in vivo

, , , , , & , (

Expression of human apolipoprotein B and assembly of lipoprotein(a) in transgenic mice

Proc. Natl. Acad. Sci. U.S.A.

, (

RNA editing of the apolipoprotein B gene: A mechanism to regulate the atherogenic potential of intestinal lipoproteins?

, , Jr, , , , , , , , , , & , (

A novel function for apolipoprotein B: lipoprotein synthesis in the yolk sac is critical for maternal-fetal lipid transport in mice

, , Jr, , , , , , & , (

A truncated species of apolipoprotein B, B-83, associated with hypobetalipoproteinemia

, , Jr, , , , , , & , (

Knockout of the mouse apolipoprotein B gene results in embryonic lethality in homozygotes and protection against diet-induced hypercholesterolemia in heterozygotes

Proc. Natl. Acad. Sci. U.S.A.

, , Jr, , , , , , , , , , , , , , , , , , & , (

Phenotypic analysis of mice expressing exclusively apolipoprotein B48 or apolipoprotein B100

Proc. Natl. Acad. Sci. U.S.A.

, (

Lipoprotein lipase and lipolysis: central roles in lipoprotein metabolism and atherogenesis

, , , , , , , & , (

Inhibition of the microsomal triglyceride transfer protein blocks the first step of apolipoprotein B lipoprotein assembly but not the addition of bulk core lipids in the second step

, , , , , , , , , , , & , (

Secretion of apolipoprotein B-containing lipoproteins from HeLa cells is dependent on expression of the microsomal triglyceride transfer protein and is regulated by lipid availability

Proc. Natl. Acad. Sci. U.S.A.

, , , , , , , & , (

Apolipoprotein B mRNA editing in 12 different mammalian species: hepatic expression is reflected in low concentrations of apoB-containing plasma lipoproteins

, & , (

Introduction: structure and metabolism of plasma lipoproteins

The Metabolic and Molecular Bases of Inherited Disease

, , , , , , , , , & , (

Targeted modification of the apolipoprotein B gene results in hypobetalipoproteinemia and developmental abnormalities in mice

Proc. Natl. Acad. Sci. U.S.A.

, , , , , , , , , & , (

Apo B gene knockout in mice results in embryonic lethality in homozygotes and neural tube defects, male infertility, and reduced HDL cholesterol ester and apo A-I transport rates in heterozygotes

, , , , , , , , , & , (

Hypercholesterolemia in low density lipoprotein receptor knockout mice and its reversal by adenovirus-mediated gene delivery

, & , (

Disorders of the biogenesis and secretion of lipoproteins containing the B apolipoproteins

The Metabolic Basis of Inherited Disease

, & , (

Effects of nonsense mutations on nuclear and cytoplasmic adenine phosphoribosyltransferase RNA

, , , , , , , & , (

Dual mechanisms for the low plasma levels of truncated apolipoprotein B proteins in familial hypobetalipoproteinemia: Analysis of a new mouse model with a nonsense mutation in the Apob gene

, , , , , & , (

Method to measure apolipoprotein B-48 and B-100 secretion rates in an individual mouse: evidence for a very rapid turnover of VLDL and preferential removal of B-48- relative to B-100-containing lipoproteins

, , , , Jr, , , , , , , , , , & , (

Transgenic mice expressing high levels of human apolipoprotein B develop severe atherosclerotic lesions in response to a high-fat diet

Which of the following is the basis for the intestine specific expression of apoprotein B

What is the major function of apoB 48?

How is ApoB48 formed?

In which of the following is apo b48 found?