COMPLETE PHENOTYPIC CHARACTERIZATION OF APOBEC-1 KNOCKOUT MICE WITH AWILD-TYPE GENETIC BACKGROUND AND A HUMAN APOLIPOPROTEIN-B TRANSGENIC - BACKGROUND, AND RESTORATION OF APOLIPOPROTEIN-B MESSENGER-RNA EDITING BY SOMATIC GENE-TRANSFER OF APOBEC-1

We have produced gene knockout mice by targeted disruption of the apob ec-1 gene. As recently reported by Hirano et al. (Hirano, K.-I., Young , S. G., Farese, R. V., Jr., Ng, J., Sande, E., Warburton, C., Powell- Braxton, L. M., and Davidson, N. O. (1996) J. Biol. Chem. 271, 9887-98 90), these animals do not edit apolipoprotein (ape) B mRNA or produce apoB-48. In this study we have performed a detailed analysis of the li poprotein phenotypic effects of apobec-1 gene disruption that were not examined in the previous study. We first analyzed the plasma lipoprot eins in knockout animals with a wild-type genetic background. Although there was no difference in plasma cholesterol between apobec-1(-/-), (+/-) (+/+) mice, there Or was a marked (176%) increase in plasma apoB -100, from 1.8 +/- 1.2 mg/dl in apobec-1(+/+) mice to 2.7 +/- 0.6 mg/d l in apobec-1(+/-) and 5.0 +/- 1.4 mg/dl in apobec-1(-/-) mice. Plasma apoE was similar in these animals. By fast protein liquid chromatogra phy (FPLC) analysis, there was a significant decrease in plasma high d ensity lipoprotein (HDL) cholesterol in apobec-1(-/-) mice. We further fractionated the plasma lipoproteins into d < 1.006, 1.006-1.02, 1.02 -1.05, 1.05-1.08, 1.08-1.10, and 1.10-1.21 g/ml classes, and found a m arked (30-40%) reduction in the cholesterol and protein content in the (d 1.08-1.10 and 1.10-1.21) HDL fractions, corroborating the FPLC dat a. SDS-gel analysis revealed an absence of apoB-48, an increase in apo B-100 in the very low density Lipoprotein (VLDL) and low density lipop rotein (LDL) fractions, and a small decrease in apoA-I in the HDL frac tions in the apobec-1(-/-) samples. We next raised the basal plasma ap oB levels in the apobec-1(-/-) animals by crossbreeding them with huma n apoB transgenic (TgB) mice. The plasma apoB-100 was 3-fold higher in apobec-1(-/-)/TgB(+/-) mice (26.6 +/- 18.3 mg/dl) than in apobec-1(+/ +)/TgB(+/-) mice (9.8 +/- 3.9 mg/dl, p < 0.05). The apobec-1(-/-)/TgB( +/-) mice had a plasma cholesterol levels of 170 +/- 28 mg/dl and trig lyceride levels of 106 +/- 31 mg/dl, which are 80% and 58% higher, res pectively, than the corresponding values of 94 +/- 21 mg/dl and 67 +/- 11 mg/dl in apobec(+/+)/TgB(+/-) mice. By FPLC, the apobec-1(-/-)/TgB (+/-) animals developed markedly elevated plasma LDL cholesterol (518. 5 +/- 329.5 mu g/ml) that is 373% that of apobec-1(+/+)/TgB(+/-) mice (139.0 +/- 87.0 mu g/ml) (p < 0.05). The elevated plasma triglyceride was accounted for mainly by a 97% increase in VLDL triglyceride in the apobec-1(-/-)/TgB(+/-) mice. We conclude that apobec-1(-/-) animals h ave a distinctive lipoprotein phenotype characterized by significant h yperapoB-100 and HDL deficiency in mice with a wild-type genetic backg round. Furthermore, the abolition of apoB mRNA editing elevates plasma total cholesterol and LDL cholesterol in apobec-1(-/-) animals with a TgB background. Finally, to exclude the possibility that absence of a poB mRNA editing was a secondary effect of chronic Apobec-1 deficiency , we treated apobec-1(-/-) mice with a replication-defective mouse Apo bec-1 adenoviral vector and found that we could acutely restore apoB m RNA editing in the Liver. These experiments indicate that Apobec-1 is an essential component of the apoB mRNA editing machinery and absence of editing in the knockout animals is a direct consequence of the abse nce of functional Apobec-1.