G. Liu, Institute of Cardiovascular Science, Peking University Health Science Center, Xueyuan Road 38, Haidian District, Beijing 100191, China Fax: +86 10 82802769 Tel: +86 10 82802769 E-mail: firstname.lastname@example.org; email@example.com X. Deng, College of Animal Science and Veterinary Medicine, Jilin University, Xi’an Road 5333, Changchun 130062, China Fax: +86 431 87836160 Tel: +86 431 87836161 E-mail: firstname.lastname@example.org; email@example.com
Hypertriglyceridemia has recently been considered to be an independent risk factor for coronary heart disease, in which apolipoprotein (Apo)CIII is one of the major contributory factors, as it is strongly correlated with plasma triglyceride levels. Although ApoCIII transgenic mice have been generated as an animal model for the study of hypertriglyceridemia, the features of lipoprotein metabolism in mice differ greatly from those in humans. Because of the great similarity between pigs and humans with respect to lipid metabolism and cardiovascular physiology, we generated transgenic miniature pigs expressing human ApoCIII by the transfection of somatic cells combined with nuclear transfer. The expression of human ApoCIII was detected in the liver and intestine of the transgenic pigs. As compared with nontransgenic controls, transgenic pigs showed significantly increased plasma triglyceride levels (83 ± 36 versus 38 ± 4 mg·dL−1, P < 0.01) when fed a chow diet. Plasma lipoprotein profiling by FPLC in transgenic animals showed a higher peak in large-particle fractions corresponding to very low-density lipoprotein/chylomicrons when triglyceride content in the fractions was assayed. There was not much difference in cholesterol content in FPLC fractions, although a large low-density lipoprotein peak was identified in both nontransgenic and transgenic animals, resembling that found in humans. Further analysis revealed markedly delayed clearance of plasma triglyceride, accompanied by significantly reduced lipoprotein lipase activity in post-heparin plasma, in transgenic pigs as compared with nontransgenic controls. In summary, we have successfully generated a novel hypertriglyceridemic ApoCIII transgenic miniature pig model that could be of great value for studies on hyperlipidemia in relation to atherosclerotic disorders.
Apolipoprotein (Apo)CIII plays an important role in the metabolism of plasma triglyceride-rich lipoproteins and their remnants, acting as an inhibitor of lipoprotein lipase (LPL) activity  and interfering with the binding of endothelial proteoglycans and specific lipoprotein receptors to lipoprotein particles for receptor-mediated endocytosis [2,3].
Mature ApoCIII is a 79-residue glycoprotein that exists in chylomicrons (CMs), very low-density lipoproteins (VLDLs), and high-density lipoproteins (HDLs). The ApoCIII enhancer is involved in maintaining an active chromatin subdomain in the ApoAI/CIII/AIV region and regulating liver-specific and intestine-specific expression of the three genes of the cluster, but not of ApoAV, at the chromatin level [4–8].
Large-scale clinical trials have indicated that hypertriglyceridemia is an independent risk factor for coronary artery disease (CAD) [9–12], and high levels of ApoCIII are often correlated with hypertriglyceridemia. Furthermore, mutations in the ApoCIII gene directly affect plasma lipid profiles, whereby people with hereditary deficiencies in ApoCIII generally have lower plasma triglyceride levels , and carriers of a null mutation in ApoCIII have cardioprotective plasma lipid profiles with lower low-density lipoprotein (LDL) cholesterol (LDL-C) and higher HDL chlolesterol (HDL-C) levels . Overexpression of human ApoCIII in transgenic mice leads to hypertriglyceridemia , and ApoCIII knockout mice have reduced levels of plasma triglyceride-containing lipoproteins . However, as the lipoprotein profile of mice is quite different from that of humans, the mouse model alone is not sufficient for evaluating the role of ApoCIII in atherosclerosis or other cardiovascular diseases.
To develop a suitable animal model overexpressing human ApoCIII for investigation of the role of ApoCIII in lipid metabolism, we have generated a human–ApoCIII transgenic miniature pig model through the transfection of somatic cells combined with nuclear transfer. The porcine model is appropriate for studying lipoprotein metabolism associated with hyperlipidemia, because pigs have both LDL and HDL particles present in plasma [16,17] and a human-like cardiovascular system [18,19]. Furthermore, experimental CAD induced in the porcine model has a similar natural progression to that in human patients [20,21].
A 10-kb human genomic DNA fragment spanning the ApoCIII gene was successfully used to generate human ApoCIII transgenic miniature pigs. Ten transgenic pigs were born, and five founders survived. In this study, the phenotypes of the five-first-generation transgenic miniature pigs were characterized for hypertriglyceridemia pertaining to human ApoCIII expression. These pigs will be bred to generate a line of transgenic miniature pigs for further study.
Generation of human ApoCIII transgenic miniature pigs
In this study, a 10-kb human genomic DNA fragment spanning the ApoCIII gene was constructed to generate transgenic miniature pigs. A total of 160 reconstructed embryos were transferred into eight recipients following natural estrus. Six of them showed early pregnancies, of which three gave normal birth. Fourteen male piglets were born through natural delivery on days 113–115 (Fig. 1), including two prenatal deaths. Postnatal death occurred in five piglets 1–10 days after birth, for undetermined reasons; the remaining pigs survived, and had grown healthily for up to 4 months at the completion of the relevant analysis.
Among the 14 natural-born piglets, 10 were confirmed as human ApoCIII transgenic pigs by PCR genotyping (Fig. 2C), and four were nontransgenic littermates. Southern blot analysis also confirmed the integration of human ApoCIII fragments into the transgenic pigs (Fig. 2D). Thus, transgenic miniature pigs expressing human ApoCIII were successfully generated. These first-generation transgenic miniature pigs were used in the present study.
During the preparation of the manuscript, several piglets died of various causes. At present, only one founder transgenic pig has survived, and it still expresses human ApoCIII. Because of the inability to breed, we have cloned four piglets from the fibroblasts of this founder, using similar cloning technique to those described in Experimental procedures. Human ApoCIII can be detected in all of these four cloned piglets.
Transgenic miniature pigs express human ApoCIII
Mature ApoCIII is a 79-residue glycoprotein that exists in CMs, VLDLs, and HDLs. The ApoCIII enhancer is involved in maintaining an active chromatin subdomain in the ApoAI/CIII/AIV region and in regulating liver-specific and intestine-specific expression of the three genes of the cluster, but not of ApoAV, at the chromatin level [4–8].
By the use of RT-PCR, we found that transgenic human ApoCIII was exclusively expressed in the liver and intestine in the transgenic miniature pigs (Fig. 3A). Human ApoCIII protein was also demonstrated by western blotting in the delipidated VLDLs isolated from human and transgenic pig plasma at the same levels of triglyceride (Fig. 3B). In addition, immunofluorescence staining (Fig. 3C) of the liver and intestine sections gave the same result as RT-PCR. Thus, we confirmed that human ApoCIII is indeed expressed in the transgenic miniature pigs.
Phenotype analysis of human ApoCIII transgenic miniature pigs
Transgenic pigs showed significant increases in plasma triglyceride levels (Fig. 4A), with 2.5-fold (94.9 ± 10.4 versus 38.4 ± 3.5 mg·dL−1) and 2.3-fold (40.6 ± 14.3 versus 17.5 ± 1.59 mg·dL−1) increases in the postprandial and fasting states, respectively, as compared with control pigs. However, total cholesterol (TC) and HDL levels were similar in wild-type and transgenic pigs (Figs 4B and 5A).
Subsequently, FPLC was used to assess the distribution of plasma lipoprotein. Transgenic pigs showed a striking increase in triglyceride content distributed in the triglyceride-rich lipoprotein fraction and a significant VLDL/CM peak (Fig. 5B).
The triglyceride levels in transgenic pigs took longer to reach peak levels during an oral fat-load test (2.5 versus 2 h), with the peak level being 2.3-fold greater than that in the control pigs (88 versus 39 mg·L−1) (Fig. 4C). This result suggests that transgenic pigs have delayed triglyceride absorbance and clearance.
Moreover, overexpression of human ApoCIII in pigs affected post-heparin plasma LPL activity: when this was initially assayed with a 10-μL sample as enzyme source incubated with 100 μL of radiolabeled substrate emulsion, a reduction of almost 60% was found in the post-heparin plasma from transgenic pigs as compared with wild-type controls (103.7 ± 31.0 versus 249.7 ± 69.2 mU·mL−1, n = 4). However, when the plasma was serially diluted, a dramatic increase in LPL activity in the transgenic sample occurred. The activity was increased from 104.6 mU·mL−1 at × 1 dilution to 494.6 mU·mL−1 at × 8 dilution (Fig. 6). This phenomenon suggests that the actual amount of LPL in the body is increased but that the catalytic activity is reversibly inhibited in vivo.
In epidemiological studies, human plasma ApoCIII levels were found to be associated with increased plasma triglyceride levels . Correspondingly, the human ApoCIII null mutation led to low plasma triglyceride levels . Therefore, human ApoCIII transgenic and knockout animal models are the preferred tools for studying the mechanisms of hypertriglyceridemia-associated diseases and for potential drug development. In the present study, we generated human ApoCIII transgenic pigs and analyzed the phenotype of this model with respect to lipid metabolism. Although the genetic modification of somatic cells followed by nuclear transfer is relatively inefficient, no other techniques can yet be utilized in pigs or other livestock for the generation of knockout and transgenic animals . This strategy was applied for the generation of human ApoCIII miniature pigs, and it was demonstrated that the human ApoCIII gene had been integrated into the pig genome, expressing human ApoCIII in the liver and intestine. We have also generated the same gene transgenic model through zygote microinjection in rabbit germlines . The overexpression of human ApoCIII in the two transgenic animals resulted in hypertriglyceridemia of varying degrees, but the rabbits expressed the human ApoCIII transgene only in the liver.
Pigs are preferred over rabbits and rodents as an animal model for cardiovascular disease, mainly because they have a human-like cardiovascular system [25,26]. In this study, the plasma triglyceride levels of human ApoCIII transgenic miniature pigs reached 110–130 mg·dL−1 on a chow diet; these levels are considered to represent mild or moderate hypertriglyceridemia, which is much more common clinically than severe hypertriglyceridemia. This feature is the same as in the transgenic rabbit model (∼ 190 mg·dL−1), but differs markedly from the situation in mice, which would have plasma triglyceride levels in the range of 300–1000 mg·dL−1 . Hence, pig and rabbit models better mimic their human counterparts in hypertriglyceridemia, because human patients commonly display mild or moderate increases in triglyceride levels. It should be noted that the plasma lipoprotein profile of the pig is closer to that of humans than that of rabbits. As shown in Fig. 5A, the LDL-C content is more than twice that of HDL-C, whereas in rabbits the HDL-C content is still higher than that of LDL-C . It is well known that more than two-thirds of plasma cholesterol is carried in LDLs in humans. It is possible that differences between species and integration sites are responsible for some of the discrepancies in plasma triglyceride levels between transgenic miniature pigs and transgenic mice.
The markedly delayed clearance of oral fat load in transgenic miniature pigs suggests that the elevation in triglyceride level may be attributable to the inhibition of triglyceride hydrolysis by LPL in vivo, as the LPL activity of post-heparin plasma from human ApoCIII transgenic pigs assayed in an in vitro system actually increased upon serial dilution of the sample from the initial low level, whereas no such dilution effect was observed in wild-type post-heparin plasma. This suggests that, although the actual amount of LPL is increased in vivo, the catalytic activity of LPL could be inhibited at the relatively high level of ApoCIII. The other possibility, of impaired uptake of VLDLs by hepatocytes as demonstrated in ApoCIII transgenic mice by Aalto-Setälä , could also contribute to reduced clearance of oral fat load in the transgenic pigs. This finding is consistent with the traditional view of ApoCIII as a reversible inhibitor of LPL.
The pig model shows a similar lipoprotein profile to that of humans, in that the plasma cholesterol is mainly carried in the LDL and HDL fractions. Combined with the similar susceptibilities of humans and pigs to atherosclerosis, the application of an ApoCIII transgenic pig model with moderate hypertriglyceridemia caused by increased VLDL and CM levels would certainly facilitate investigation of the role of ApoCIII in triglyceride metabolism and atherosclerosis.
In conclusion, we have successfully generated transgenic miniature pigs expressing human ApoCIII. Our model should provide a powerful tool with which to gain further understanding and insights into the in vivo functional roles of human ApoCIII in lipid metabolism, and the role of triglycerides in atherosclerosis or CAD. Human ApoCIII transgenic miniature pigs clearly represent a superior model than other germlines for evaluating the efficacy and pharmacology of new drugs for hypertriglyceridemia. These first-generation transgenic miniature pigs will be bred to yield a line of transgenic miniature pigs for further study.
The oocytes were collected from an abattoir. The sows were bred and maintained at the pig farm of Jilin University, and were provided with a chow diet and water ad libitum. All animal experiments were carried out in accordance with the European Communities Council Directive 86/609/EEC for animal welfare, and were approved by the Institutional Animal Care and Use Committee of Jilin University (protocol No. 2008-11). Age-matched and same-strain nontransgenic miniature pigs were used as controls.
Generation of transgenic miniature pigs
Transgenic pigs expressing human ApoCIII were generated according to the method described by Lai et al. [28,29]. For somatic cell transfection, we constructed a vector named pcDNA-CIII (Fig. 1) containing a human ApoCIII genomic fragment with 5′-flanking and 3′-flanking sequences and a neomycin resistance gene. Fibroblasts isolated from a miniature pig fetus at 35 days of gestation were transfected with pcDNA-CIII, using FUGene HD transfection reagent (Roche Applied Science, Mannheim, Germany) at a 3 : 1 ratio of FuGENE HD (μL) to DNA (μg), according to the manufacturer’s protocol. After 72 h, the cells were cultured in selection medium containing 350 μg·mL−1 G418 antibiotic (E859; AMRESCO) for an additional 7 days. The cell colonies were picked, propagated, and then confirmed by PCR and sequencing.
One positive cell clone was used in nuclear transfer. Cumulus–oocyte complexes aspirated from ovaries were washed three times in maturation medium, the oocytes were matured for 40 h at 39 °C, and cumulus cells were removed as the oocytes matured. Metaphase II oocytes were enucleated by removal of the polar body and the associated metaphase plate. A 45-μs electrical pulse of 95 V from an ElectroCell Manipulator 200 (Genetronics, San Diego, CA, USA) was used to fuse the membranes of the donor cell and oocyte to form a cytoplasmic hybrid, and the reconstructed oocytes were activated. Then they were cultured for 18–22 h, and the ones in a good growth state were surgically transferred into an oviduct of the surrogate.
The surrogates were kept in a conventional environment for housing pigs, and observed daily for confirmation of pregnancy by checking estrus. All of the transgenic miniature pigs were delivered by natural birth without any chemical induction.
For identification of positive transgenic miniature pigs, genomic DNA was extracted from the umbilical cord with a commercial kit (DNeasy Tissue; Qiagen, Hilden, Germany) for PCR amplification of an 888-bp fragment with the primers 5′-CCA G AA ATC ACC CAA AGA CA-3′ and 5′-GGA AAT GA G GGA TAA AAC GG-3′. We also performed Southern blotting with the 888-bp probe (Fig. 2B). Hybridization of EcoRI-digested genomic DNA with the 888-bp probe yielded a band of 3.1 kb.
To examine the expression of human ApoCIII in transgenic pig tissue, total RNA was isolated from different tissues with TRIzol reagent (Gibco BRL, Gaithersburg, MD, USA), and cDNA was transcripted with a reverse transcription kit from TaKaRa (DRR012A, RNA LA PCR Kit) with random primers. Subsequently, PCR was performed with primers 5′-TAG GAA TTC GAA CAG AGG TGC CAT GC-3′ and 5′-TAG CTC GAG TAT TGA GGT CTC AGG CA-3′.
Western blot analysis
To additionally confirm the expression of human ApoCIII, the VLDL fraction was separated from 4 mL of plasma by ultracentrifugation (at 284 000 g for 18 h) for western blot analysis. Delipidated VLDL samples were separated by 15% SDS/PAGE and transferred semidry onto a poly(vinylidene difluoride) membrane (Invitrogen, Carlsbad, CA). Goat anti-(human ApoCIII IgG) (1 : 1000; Merck-Frosst, Darmstadt, Germany) and anti-goat IgG–horseradish peroxidase (1 : 5000; Amersham Biosciences, Piscataway, NJ, USA) were used as the first and second antibody, respectively.
Immunofluorescence detection of tissue sections
Sections from the spleen, kidney, liver, heart and intestine were collected and fixed in 4% formaldehyde for 18 h, and then embedded in paraffin blocks by the use of routine protocols. Sections 7 μm in thickness were fixed in ice-cold acetone for 10 min, and washed in NaCl/Pi (pH 7.4). The sections were incubated with primary antibodies [goat anti-(human ApoCIII IgG), 1 : 300] at 4 °C overnight, and then incubated with Cy3-conjugated secondary antibody [donkey anti-(goat IgG), 1 : 500] at room temperature for 2 h. The sections were mounted in Gelvatol and analyzed with an Olympus FV-1000 confocal laser scanning microscope.
Plasma lipid and lipoprotein analysis
The plasma lipid and lipoprotein profiles of ApoCIII transgenic pigs were compared with those of age-matched controls at the age of 50 days. One milliliter of blood was collected after 16 h of food deprivation and after 6 h of refeeding on a chow diet. The triglyceride, TC and HDL-C levels in both fasting and postprandial states were determined with commercial kits (Sigma, St Louis, MO, USA). For determination of the lipids distributed in plasma lipoprotein, FPLC was performed with 200 μL of pooled plasma from four pigs per group, using a Superose 6 column (Amersham Bioscience) as described previously. Thirty-five fractions of 0.5 mL each were collected and enzymatically assayed for TC and triglyceride content.
LPL activity assay
Post-heparin plasma was prepared from a blood sample collected 10 min after intravenous injection of heparin (50 units·kg−1 of body weight). The enzymatic activity of LPL was determined with radioactive trioleoylglycerol emulsion substrate, according to an established method . LPL activity was calculated as the fraction of total lipolytic activity inhibited by 1 m NaCl, and expressed as mU·mL−1 (1 mU corresponds to 1 nmol of free fatty acid generated per minute).
Oral fat-load test
At 40 days of age, two groups of miniature pigs were orally given olive oil at 10 mL·kg−1 body weight after a 12-h fast. Approximately 1 mL of blood was collected into heparinized Eppendorf tubes via the precaval vein at baseline and at 30, 60, 90, 120, 150, 180 and 210 min. Plasma triglyceride was measured enzymatically as previously described.
Quantitative data were expressed as the mean ± standard error of the mean. Statistical analysis was performed with two-tailed Student’s t-tests or nonparametric Mann–Whitney U-tests. In all cases, differences were considered significant at P < 0.05.
We thank G. Tan for nuclear transfer, C. Guo, D. Li and Y, Zhou for surgery, and B. Sun for supporting transgenic pigs bred at Boar Farm of Jilin University. This research was funded by the National Natural Science Foundation of China (No. 31172172 and No. 30821001), the Major National Basic Research Program of the People’s Republic of China (No. 2011CB503900), and the Beijing Natural Science Foundation (No. 5101004).