a -Tocopherol Transfer Protein Is Important for the Normal Development of Placental Labyrinthine Trophoblasts in Mice*

a -Tocopherol transfer protein ( a -TTP), a cytosolic protein that specifically binds a -tocopherol, is known as a product of the causative gene in patients with ataxia that is associated with vitamin E deficiency. Targeted disruption of the a - TTP gene revealed that a -tocopherol concentration in the circulation was regulated by a -TTP expression levels. Male a -TTP 2 / 2 mice were fertile; how-ever, placentas of pregnant a -TTP 2 / 2 females were se-verely impaired with marked reduction of labyrinthine trophoblasts, and the embryos died at mid-gestation even when fertilized eggs of a -TTP 1 / 1 mice were transferred into a -TTP 2 / 2 recipients. The use of excess a -to-copherol or a synthetic antioxidant (BO-653) dietary supplement by a -TTP 2 / 2 females prevented placental failure and allowed full-term pregnancies. In a -TTP 1 / 1 animals, a - TTP gene expression was observed in the uterus, and its level transiently increased after implantation

From the ‡Pharmaceutical Technology Laboratory, Chugai Pharmaceutical Co., Ltd., Gotemba,Shizuoka, Japan and the §Department of Health Chemistry, Graduate School of Pharmaceutical Sciences, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033 Japan ␣-Tocopherol transfer protein (␣-TTP), a cytosolic protein that specifically binds ␣-tocopherol, is known as a product of the causative gene in patients with ataxia that is associated with vitamin E deficiency. Targeted disruption of the ␣-TTP gene revealed that ␣-tocopherol concentration in the circulation was regulated by ␣-TTP expression levels. Male ␣-TTP ؊/؊ mice were fertile; however, placentas of pregnant ␣-TTP ؊/؊ females were severely impaired with marked reduction of labyrinthine trophoblasts, and the embryos died at mid-gestation even when fertilized eggs of ␣-TTP ؉/؉ mice were transferred into ␣-TTP ؊/؊ recipients. The use of excess ␣-tocopherol or a synthetic antioxidant (BO-653) dietary supplement by ␣-TTP ؊/؊ females prevented placental failure and allowed full-term pregnancies. In ␣-TTP ؉/؉ animals, ␣-TTP gene expression was observed in the uterus, and its level transiently increased after implantation (4.5 days postcoitum). Our results suggest that oxidative stress in the labyrinth region of the placenta is protected by vitamin E during development and that in addition to the hepatic ␣-TTP, which governs plasma ␣-tocopherol level, the uterine ␣-TTP may also play an important role in supplying this vitamin.
Vitamin E (␣-tocopherol) is the most potent lipid-soluble antioxidant in biological membranes, where it contributes to membrane stability. Patients with ataxia and isolated vitamin E deficiency (AVED) 1 have low or undetectable serum vitamin E concentrations and exhibit neurological dysfunction and muscular weakness. It is now established that ␣-tocopherol transfer protein (␣-TTP), a cytosolic liver protein known to specifically bind to ␣-tocopherol (1), is defective in AVED patients (2), indicating that ␣-TTP is a major determinant of plasma ␣-tocopherol level. Although ␣-tocopherol was initially identified as an anti-sterility factor to prevent abortion (3), the mechanism of action and the molecules responsible for its antisterility effect remain unknown. One of the reasons for this is that vitamin E is difficult to deplete from tissues and requires elaborate manipulations to cause deficiency symptoms to occur in experimental animals. In this study, we established a mouse model lacking ␣-TTP by targeted mutagenesis. This animal model for human AVED patients is suitable for examination of the complex pathophysiology of diseases associated with vitamin E deficiency and/or caused by oxidative stress. Here we examined the role of ␣-TTP in pregnancy and embryogenesis using our new animal model.

MATERIALS AND METHODS
Generation of ␣-TTP Knockout Mice-An ␣-TTP targeting vector was constructed from an 8.8-kb ␣-TTP genome fragment encompassing exon 1. We inserted a fragment of PGK-neo cassette into the SmaI-SmaI site positioned 5Ј and 3Ј to exon 1 and flanked a 1.8-kb fragment of HSV-tk gene downstream of exon 2. AB2.2-Prime ES cells (Lexicon Genetics) or A3-1 ES (4) cells were transfected by electroporation with a linearized targeting vector. G418/gancyclovir-resistant clones were screened by PCR, and then ES cells containing the disrupted allele were injected into C57BL/6J (CLEA, Japan) blastocysts as described previously (5). To obtain ␣-TTP ϩ/Ϫ mutants, chimeras were mated with C57BL/6J females. ␣-TTP Ϫ/Ϫ mutant mice were produced from ␣-TTP ϩ/Ϫ crosses. Genotypes were determined by PCR and confirmed by Southern blot analysis of DNA from tail tissue. The PCR primer pairs (ot198, 5Ј-AGCCCACACAAAAATGAAAAACGTCTCCAAG-3Ј and PGK-1, 5Ј-GCTAAAGCGCATGCTCCAGACTGCCTTG-3Ј) were used to detect the ␣-TTP mutant allele. PCR primer pairs (ot198 and TTPN17, 5Ј-TCTCT-GCAATGCCCGCCGTGCTGTCCCG-3Ј) were used to detect the ␣-TTP wild-type allele. After an initial hot start at 94°C for 1 min, 35 cycles (94°C for 30 s, 62°C for 1 min, and 72°C for 1 min and 20 s) were run using Takara EX Taq (TaKaRa, Japan). The expected PCR products of wild-type and mutant alleles were 990 and 950 bp, respectively. Genomic DNA from mutant mice were analyzed by Southern blotting using probe A including exon 1 and mouse ␣-TTP cDNA (open-reading frame) probe, after digestion with EcoRI. The resultant two fragments, which had approximately the same number of nucleotides, were mixed and used for probe A. Mouse cDNA probe for ␣-TTP was prepared by RT-PCR with mouse liver total RNA. In the next step, 15 g of genomic DNA was electrophoresed on a 0.7% agarose gel and transferred onto a Hybond Nϩ membrane (Amersham Pharmacia Biotech). The membranes were hybridized overnight at 42°C in a buffer containing 50% formamide, 5ϫ SSPE, 0.5% SDS, 5ϫ Denhardt's solution, and 250 g/ml denatured salmon sperm DNA with 32 P-labeled probe. The membranes were washed for 30 min in 2ϫ SSC, 0.2% SDS, and then in 0.5ϫ SSC, 0.2% SDS at 65°C for 30 min. The 3.75-kb EcoRI fragment represents the wild-type allele.
Analysis of ␣-TTP Expression by Northern Blotting-Total RNA was extracted from the liver of each adult mouse genotype and from uterus, placentas, and embryos of ␣-TTP ϩ/ϩ mice using ISOGEN (Nippon Gene, Japan). 10 g of total RNA from liver and 20 g of total RNA from uterus, placenta, and embryo were electrophoresed on a 1% agarose gel and transferred onto a Hybond Nϩ membrane. The membranes were hybridized and washed using the method described above for Southern blotting.
Determination of Plasma ␣-Tocopherol Concentrations-Mice were * 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 U.S.C. Section 1734 solely to indicate this fact.
fed a normal (36 mg of ␣-tocopherol/kg diet) or ␣-tocopherol-supplemented diet (600 mg of ␣-tocopherol/kg diet) after weaning. These diets were prepared from a vitamin E-deficient diet (Funabashi Farm, Chiba, Japan) supplemented with 5.0% (w/w) stripped corn oil (Tama Biochemical, Tokyo, Japan) and D-␣ tocopherol. D-␣-Tocopherol was kindly provided by Eisai Co. Ltd. (Tokyo, Japan). Blood samples were collected from overnight fasted animals, and plasma was separated from whole blood by centrifugation. Plasma (50 l) was diluted with 950 l of phosphate buffered saline and was used for the following procedure. Diluted plasma was mixed with 1 ml of 6% pyrogallol in ethanol, and 2.0 g of tocol was subsequently added as an internal standard and mixed vigorously. After incubation at 70°C for 2 min, 0.2 ml of 60% KOH was added, and the mixture was incubated at 70°C for 30 min. In the next step, 5 ml of n-hexane and 2.5 ml of water were added, and the mixture was mixed vigorously and then centrifuged at room temperature. The hexane layer was saved and the hexane extracts were evaporated under nitrogen. The residue was redissolved in 100 ml of ethanol and subjected to HPLC analysis and electrochemical detection. The HPLC system was an IRIKA P-530 (IRIKA, Kyoto) with an IRIKA RP-18 column (4 ϫ 250 mm). The eluent was methanol/water/NaClO 4 at a ratio of 1000:2:7 (v/v/w) and a flow rate of 10 ml/min. Detection was performed with an IRIKA Amperometric E-520 detector. The retention time was 6.88 min for tocol, which was used as an internal standard, and 10.88 min for ␣-tocopherol as described previously (6).
Viability of Embryos in the Uterus of ␣-TTP Mutant Mice-To determine the time of death, ␣-TTP ϩ/ϩ and ␣-TTP Ϫ/Ϫ mutant females were mated with C57BL/6J males, and then the pregnant females were sacrificed between 9.5 and 14.5 dpc. The death of embryos was confirmed by the absence of a heartbeat.
Morphological Appearance and Histology-Embryos and placentas with and/or without the uterine horns were fixed with 10% neutralbuffered formalin for up to 24 h. Embryos and uterine horn segments were subsequently processed into paraffin sections and deparaffinized for staining with hematoxylin/eosin before microscopic analysis.
All experiments described in the present study were conducted in accordance with the Guiding Principles for the Care and Use of Research Animals promulgated by Chugai Pharmaceutical, Shizuoka, Japan.

␣-TTP Mutant
Mice-To delete the initiation codon for ␣-TTP, a targeting vector was designed in which the entire exon 1 was replaced by a neomycin-resistance cassette (Fig.  1A). This targeting vector was introduced into ES cells by electroporation, and then the ES cells were used to introduce vector into the mouse germline. We obtained three independent mutant mouse lines. Two lines (clone nos. L236 and L254) were derived from AB2.2-Prime ES cells and one line (clone no. C229) was derived from A3-1 ES cells. The chimeras of these lines were bred with C57BL/6J to produce heterozygous mice for ␣-TTP. Mice from the L236 and C229 lines were bred and used for further analysis. When heterozygous mice were interbred, approximately one-fourth of the offspring were ␣-TTP Ϫ/Ϫ mutants as expected for a recessive mutation (␣-TTP ϩ/ϩ :␣-TTP ϩ/Ϫ :␣-TTP Ϫ/Ϫ ϭ 63:105:74; Fig. 1C). Both ␣-TTP ϩ/Ϫ and ␣-TTP Ϫ/Ϫ mice were normal in appearance and growth for at least 6 months. There were no significant differences among the genotypes in the plasma levels of VLDL, LDL, and HDL cholesterol as measured by HPLC (data not shown).
Infertility of Female ␣-TTP Ϫ/Ϫ Mice-As shown in Table I, ␣-TTP Ϫ/Ϫ males were fertile. The ␣-TTP Ϫ/Ϫ females became pregnant after mating, but none of the four or five tested delivered offspring (Table I). Because ␣-TTP Ϫ/Ϫ mutants were obtained from mating ␣-TTP ϩ/Ϫ males and females in a Mendelian fashion, the ␣-TTP Ϫ/Ϫ zygotes could develop to full-term. On the other hand, although fertilized eggs from ␣-TTP ϩ/ϩ mice could be successfully implanted into ␣-TTP Ϫ/Ϫ recipients, they failed to develop to full-term (Table II). The number of live embryos (as determined by the presence of a heartbeat) of ␣-TTP Ϫ/Ϫ mice markedly decreased between 11.5 and 14.5 dpc (Fig. 3).
The placentas and embryos of various maternal genotypes were not morphologically different at 9.5 dpc (data not shown). However, the embryos in the uteri of ␣-TTP Ϫ/Ϫ mutants FIG. 1. Generation of ␣-TTP null mice. A, mouse ␣-TTP locus, the targeting vector, and the predicted structure of the ␣-TTP locus after homologous recombination. The neomycin cassette was inserted into the SmaI-SmaI restriction site positioned 5Ј and 3Ј to exon 1. The PCR primer pairs ot198 and PGK-1 and ot198 and TTPN17 were used to detect ␣-TTP mutant and wild-type alleles, respectively. B, Southern blot analysis of EcoRI-digested genomic DNA. Probe A including exon 1 or a mouse ␣-TTP cDNA probe did not hybridize to a 3.75-kb fragment in homozygous mutant mouse genomes. C, genotyping of offspring from heterozygous F1 intercrosses were analyzed by PCR. D, expression of ␣-TTP was analyzed by Northern blot using total RNA from the liver. ␣-TTP Ϫ/Ϫ and ␣-TTP ϩ/Ϫ mice had undetectable or half-levels of ␣-TTP mRNA in the liver compared with ␣-TTP ϩ/ϩ mice, respectively. The blot was reprobed for cholesterol 7␣-hydroxylase, which was used as loading control.

␣-Tocopherol Transfer Protein During Pregnancy 1670
showed developmental failure from 10.5 dpc, and the majority of these embryos showed neural tube malformations (Fig. 4E). In normal pregnancy, the labyrinth region of the placenta starts development from around 9 -9.5 dpc and then functions as a nutrient transport unit (8). Under normal embryogenesis, the allantoic vessels are seen by about 10 dpc where they penetrate the chorionic plate, and the ectoplacental plate is transformed into the labyrinthine part of the placenta (8). At this stage, the placenta could be divided into several well defined layers such as the spongiotrophoblast layer and the labyrinth region. Histological examination showed a specific abnormality limited to the labyrinth region of the ␣-TTP Ϫ/Ϫ mutant at 10.5 dpc (Fig. 4, C and F). In these mice, there was a marked reduction in the number of trophoblast cells, resulting in an abnormally small labyrinth. Furthermore, embryonic blood vessels were virtually absent in the trophoblast.
Expression of the ␣-TTP Gene in the Uterus-In ␣-TTP ϩ/ϩ mice, expression of the ␣-TTP gene was observed in the uterus throughout pregnancy (Fig. 5), and the expression level of the ␣-TTP gene increased transiently after implantation on 4.5 dpc and gradually decreased by parturition. Because ␣-TTP expression did not increase in pseudopregnant mice at 4.5 dpc (data not shown), implantation of embryos or the development of embryos seems to have stimulated ␣-TTP gene expression. After about 4.5 dpc, the polar trophectoderm gives rise to extraembryonic ectoderm of the chorion, which later contributes to the trophoblast component of the labyrinth region, and the ectoplacental cone, which later produces the spongiotrophoblast layer (9). ␣-TTP was not expressed in the placenta at any time during development (Fig. 5). Although expression of the ␣-TTP gene in embryos was moderate, expression of this gene does not seem to be essential for embryonic development because ␣-TTP Ϫ/Ϫ eggs developed to full-term. These results suggest that ␣-TTP acts as a uterine factor and plays an important role in placental development.

␣-Tocopherol Transfer Protein During Pregnancy 1671
Rescue of Embryos in Uteri of ␣ -TTP Ϫ/Ϫ Mutants by Diet Containing Excess Amounts of ␣-Tocopherol or Synthetic Antioxidant-To examine the effect of ␣-tocopherol dietary supplementation on the development of the placenta and embryos in uteri of ␣-TTP Ϫ/Ϫ mutants, the diet was supplemented with ␣-tocopherol (567 mg/kg diet) either starting at 0.5 dpc after mating or throughout the experiment. With this diet, plasma ␣-tocopherol levels in ␣-TTP Ϫ/Ϫ mice were maintained within the normal range, which were close to the levels in ␣-TTP ϩ/Ϫ mice fed a normal diet (Fig. 2). This therapy, as well as supplementation of a synthetic antioxidant, BO-653 (10), had a pronounced effect on full-term development of embryos in the uteri of ␣-TTP Ϫ/Ϫ mutants (Table III). The delivered pups showed normal growth and behavior and were fertile at adulthood.

DISCUSSION
Vitamin E was identified in the 1920s as a substance required for animals to have offspring (3). In this study, we generated ␣-TTP Ϫ/Ϫ mice with undetectable levels of plasma vitamin E even upon feeding with normal diet. Using these mice, we analyzed the infertility caused by vitamin E deficiency and found that the ␣-TTP Ϫ/Ϫ female mice have defective labyrinthine trophoblast formation during embryogenesis. The placental failure was effectively abrogated by ␣-tocopherol or synthetic antioxidant dietary supplement, indicating that vitamin E or other antioxidants are essential for the formation of labyrinthine trophoblasts. It is well known that the feto-placental system is prone to the attack of oxidants and that placental brush border membrane is most susceptible to peroxidation (11,12). Oxygen-free radicals are also involved in the induction of fetal anomalies. For example, excess oxygen radical activity has been reported to be associated with disturbed embryogenesis in diabetic pregnancy (13). Other studies have also shown a reduction in the severity of these diseases with administration of vitamin E during early pregnancy (14,15). These findings, together with the present results, suggest that embryogenesis, especially the formation of the placental labyrinthine trophoblasts, is more susceptible to oxidative stress. Efficient functioning of the enzymic and nonenzymic reactive oxygen species scavengers ensures a normal intrauterine fetal growth and development (12,16). Mukherjea and co-workers (17,18) demonstrated that ␣-tocopherol content in the placental membrane increased as gestation progressed.
In this context, it is interesting to note that the expression of the ␣-TTP gene in the uterus of normal mice transiently increased around 4.5 dpc, possibly leading to an increase in ␣-tocopherol levels supplied to the embryo. On the other hand, it was also demonstrated that vitamin E crosses the placenta from the mother to the embryo, and interestingly, of the various forms of vitamin E transferred, the RRR-␣-tocopherol (best ligand for ␣-TTP), crossed most efficiently (19). ␣-TTP expressed in the uterus may explain stereospecific transport of tocopherols to the placenta, and up-regulation of ␣-TTP expression may result in the increase in the transport of ␣-tocopherol to the placenta during embryogenesis. In addition to the hepatic ␣-TTP, which governs plasma ␣-tocopherol levels, the uterine ␣-TTP may also be the important factor for feto-placental development. We have established ␣-TTP-disrupted mice as a model for vitamin E deficiency. This model should be a useful tool for the study of diseases caused by oxidation stress.