Targeted Expression of Cre Recombinase Provokes Placental-Specific DNA Recombination in Transgenic Mice

Background Inadequate placental development is associated with a high incidence of early embryonic lethality and serious pregnancy disorders in both humans and mice. However, the lack of well-defined trophoblast-specific gene regulatory elements has hampered investigations regarding the role of specific genes in placental development and fetal growth. Principal Findings By random assembly of placental enhancers from two previously characterized genes, trophoblast specific protein α (Tpbpa) and adenosine deaminase (Ada), we identified a chimeric Tpbpa/Ada enhancer that when combined with the basal Ada promoter provided the highest luciferase activity in cultured human trophoblast cells, in comparison with non-trophoblast cell lines. We used this chimeric enhancer arrangement to drive the expression of a Cre recombinase transgene in the placentas of transgenic mice. Cre transgene expression occurred throughout the placenta but not in maternal organs examined or in the fetus. Significance In conclusion, we have provided both in vitro and in vivo evidence for a novel genetic system to achieve placental transgene expression by the use of a chimeric Tpbpa/Ada enhancer driven transgene. The availability of this expression vector provides transgenic opportunities to direct the production of desired proteins to the placenta.


Introduction
The mammalian placenta is the first organ to be developed during gestation and carries out multiple functions required for normal embryonic development in the uterine environment [1]. Impaired placental development is associated with many complications to both moms and babies during pregnancy, including preeclampsia, intrauterine growth retardation (IUGR) and fetal loss [2,3]. Thus, a better understanding of gene function during placentation could provide new insights regarding normal placental development and fetal growth, which will in turn help guide the development of prevention strategies and new therapies for the treatment of diseases associated with pregnancy, including fetal abnormalities.
Genetic manipulation of the mouse is a powerful experimental approach to study the functional role of specific genes by gain of function or loss of function strategies [4]. However, the disruption of many genes results in embryonic lethality because of placental defects, making it difficult to evaluate the potential role the gene may play in extraplacental tissues [1]. In addition, the embryonic lethality of these mutant mice prevents their use to assess the role of specific genes in postnatal physiology and development. Over the past decade, the Cre/loxP system, utilizing Cre recombinase to catalyze a deletion event between two DNA fragments containing the 34 bp loxP recognition site, has been commonly used for conditional gene deletion strategies to assess biological function of genes in certain types of cells or tissues in vivo [5]. However, the lack of a robust placental specific transgene to efficiently and specifically express desired genes (including Cre recombinase) in placenta has hampered progress in a number of areas of placental developmental biology. For example, the inability to disrupt the expression of specific genes in the placenta has prevented the analysis of the exact role of these genes in placental development. In addition, the inability to reliably direct expression of transgenes to the placenta has prevented the use of transgenic strategies to correct placental defects, thereby making it possible to study the role of a specific gene in other (i.e., non-placental) aspects of embryonic development and/or postnatal development. To overcome this limitation, we set out to develop a mammalian expression vector containing strong placental enhancers to drive robust placental specific expression of desired genes.
To reach this goal, we focused on the placental enhancers from two genes known to drive the expression of reporter genes in placenta of transgenic mice. The first enhancer is a 340 bp regulatory fragment from the 59-flanking region of trophoblast specific protein a gene, Tpbpa, a gene encoding a putative inhibitor of placental cathepsins that is specifically expressed in the trophoblast lineage in mice [6,7,8]. The other is a 1.8 kb fragment from the 59 flanking region of the gene encoding adenosine deaminase (ADA), an enzyme present at high levels in trophoblast cell lineages throughout placental development [9]. Early studies identified a 1.8 kb fragment in the 59-flanking region of the Ada gene that is responsible for placenta specific expression [10,11]. However, the characterized placental regulatory elements for Ada generally provided relatively low placental expression. Therefore, we attempted to generate a robust placental specific enhancer by testing combinations of both Tpbpa and Ada enhancers.

Reagents
One-step RT-PCR kit, cell culture medium, antibiotics and fetal bovine serum (FBS) were purchased from Invitrogen (Carlsbad, CA). pCAG-CATZ and AdMA19 plasmids were a generous gift from Dr. M.D. Schneider (British Heart Foundation Centre of Research Excellence, National Heart and Lung Institute, Imperial College London, London SW7 2AZ, UK)). PKS-Tpbpa plasmid was a generous gift from Dr. Janet Rossant (University of Toronto, Toronto, Ontario, M5S 1A1 Canada). Cre recombinationreporter mice were purchased from Jackson laboratory. HEK-293, M1, A549, ML12, K562 and HeLa cells were purchased from the American Type Culture Collection (Manassas, VA).

Plasmid construction
The Tpbpa enhancer fragment was generated by Pfu DNA polymerase-based PCR using PKS-Tpbpa plasmid [6] and paired primers flanked with KpnI sequence, 59 primer, TAGGTACCG-TAGACTGTTCCTCAGTAGA; 39 primer, TAGGTACCCTC-GAGAGAGAA AGACACTT. PCRs were performed with an initial denaturation at 95uC, 2 min. PCR cycling conditions were 30 cycles of denaturation at 95uC for 30 sec, annealing at 60uC for 30 sec, extension at 72uC, 60 sec, and final extension at 72uC for 10 min. The PCR product of the Tpbpa enhancer was digested with KpnI and subcloned into KpnI restriction site of pGL3-basic luciferase vector to generate pTpbpa-Luc. In addition, the Tpbpa PCR fragment was blunted and subcloned into pAda-Luc vector containing the 0.8 kb basal Ada promoter (AdaP) or Ada enhancer/AdaP containing the 1.8 kb Ada enhancer and 0.8 kb basal Ada promoter to generate pTpbpa-AdaP-Luc and pTpbpa/Ada-AdaP-Luc vector in reverse or forward direction (pTpbpar/Adaf-AdaP-Luc and pTpbpaf/Adaf-AdaP-Luc). Finally, a second copy of Tpbpa was further subcloned into pTpbpaf-AdaP-Luc or pTpbpar/Adaf-AdaP-Luc constructs to generate pTpbpaf-Tpbpaf/AdaP-Luc and pTpbpaf-Tpbpar/Adaf-AdaP-Luc.
To generate Tpbpar/Adaf-AdaP-Cre vector, a Cre DNA recombinase fragment (1.9 kb) containing a nuclear localization sequence (NLS) was released from paMHC-Cre plasmid by digestion with KpnI, subsequently blunted with T7 DNA polymerase and then cut with Hind III. This fragment was subcloned into pTpbpar/Adaf-AdaP-Luc construct with replacement of luciferase fragment by digestion with XbaI, blunted with T7 DNA polymerase and then cut with Hind III. The accuracy of all construct sequences was confirmed by DNA sequencing.

PCR analysis of Cre recombinase activity in vitro
To determine Cre-mediated DNA recombination after transient transfection, PCR analysis was conducted as described [14]. Briefly, HTR cells were cotranfected with the Cre-dependent reporter gene, CAG-CATZ, with or without CMV-Cre or different amounts of Tpbpar/Adaf-AdaP-Cre construct, respectively. CAG-CATZ harbors a CAT gene flanked by loxP sites and driven by the chicken b-actin promoter. Downstream of CAT is E coli bgalactosidase (lacZ). DNA was isolated 48 h after transfection and PCR analysis was performed using oligonucleotides synthesized corresponding to the 39end of the chicken b-actin promoter (59-CTGCTAACCATGTTCATGCC-39; AG2) and 59 end of the LacZ gene (59-GGCCTCTTCGCTATTACG-39; Z3). In addition, additional oligonucleotides synthesized corresponding to the 39end of the CAT gene (59-CAGTCAGTTGCTCAATGTACC-39; CAT2) and 59 end of the CAT gene (59-ACTGGTGAAACT-CACCCA-39; CAT3) were used as an internal control for transfection efficiency and resulted in a 320 bp PCR fragment. Using AG2 and Z3 primers, in the absence of Cre, only the 2100 bp precursor PCR fragment was detected. However, in the presence of Cre, both 2100 bp (precursor) and 690 bp (product) PCR fragments were detected. Any cells transfected with the Cre reporter were screened by PCR using primer pair CAT2 and CAT3 which give a 320 bp PCR product.

Luciferase analysis of Cre recombinase activity in vitro
Similar to PCR analysis of Cre recombinase activity, HTR cells were transfected with AdMA19 luciferase construct containing a spacer interposed between two loxP sites precluding efficient lucifearse expression in the absence of Cre recombinase. Fortyeight hr after transfection, cellular extracts were collected and luciferase activity was measured as described [3,13,15].

Generation of transgenic mice
All animal manipulations in this study were reviewed and approved by the Animal Welfare Committee, University of Texas Houston Health Science Center (Protocol# HSC-AWC-09-159). The 5 kb Tpbpar/Adaf-AdaP-Cre gene containing the Tpbpar/Adaf-AdaP chimeric enhancer Ada promoter and Cre recombinase transgene was excised from Tpbpar/Adaf-AdaP-Cre plasmid vector backbone using NotI and HpaI. Linear Tpbpar/Adaf-AdaP-Cre DNA fragment was separated by electrophoresis through 1% agrose gel and purified using Qiaex II reagents (QIAGEN Inc., Chatsworth, CA). The linear Tpbpar/Adaf-AdaP-Cre gene was microinjected into the pronuclei of FVB zygotes [6,16] at a concentration of 2 ng/ml in 10 mM Tris-HCl (pH 7.4), 0.1 M EDTA and injected embryos were transferred to pseudopregnant FVB females.

RNA isolation, quantification of real time PCR (RT-qPCR) and semi-quantitative RT-PCR
Trizol reagent was used for the isolation of total RNA. The reverse transcriptase-polymerase chain reaction (RT-PCR) was performed according to the manufacturer's recommended protocol (Invitrogen, Carlsbad, CA). 1 mg of RNA was used per reaction and single strand cDNA was synthesized at 55uC for 30 minutes. The annealing temperature of PCR for Cre and GFP is 50uC. PCR conditions were as described [15]. Cre primer sequences were: sense primer, 59-CCCTGTTTCACTATCCAGGT, and antisense primer, 59-GGGTAACTAAACTGGTCGAG. GFP primer sequences were: reverse, 59-GGCCATGATATAGACGTT; forward, 39-AA-GTTCATCTGCACCACCG. b-actin was used as an internal control and primer sequences were: reverse, 59-CCACCGATC-CACACAGAGTAC and forward, 59-GCTCTGGCTCCTAG-CACCAT. RT-PCR products were revealed on 2% agarose gels. RT-qPCR was performed using SYBR green JumpStart Taq ReadyMix (Sigma) on an Applied Biosystem 7000 under the following conditions: 95uC, 2 min; 95uC, 15 s; 50uC, 15 s; 72uC, 30 s; 40 cycles. Each cDNA sample was run in triplicate. Relative Cre expression was calculated following normalization to b-actin.
Genomic DNA isolation, genotyping by PCR and gene copy number analysis Genomic DNA was isolated using a DNeasy tissue kit (Qiagen, Valencia, CA). Presence of Tpbpar/Adaf-AdaP-Cre transgenes in founders was assessed by amplification of genomic DNA from tail samples [6,16], using a sense primer at the 59 end and anti-sense primer for the Cre cDNA, respectively (59-CGGTCTCTGA-GAGCCATC-39 and 59-CCCTGAACATGTCCATCA-39) and resulting in a 340 bp band. To determine the genotype of placentas from pregnant F1 offspring, genomic DNA was isolated from embryos. Cre recombinase transgene gene copy number was determined by qPCR as described [17]. qPCR was performed with 200 ng of DNA in duplicate using green JumpStart Taq ReadyMix (Sigma) on an Applied Biosystem 7000. Endogenous mouse b-actin was used as an internal control for DNA input. The quantitative standard curve was generated using a series of standard samples containing 1, 2, 4, 8, 16 copies of the Cre gene respectively, which were prepared by mixing wild type DNA from FVB mice with the Tpbpar/Adaf-AdaP-Cre vector. Standard curve was drawn by plotting C tCre against the log of Cre gene copies of corresponding standard samples. Cre was amplified from genomic DNA extract from transgenic mouse tissues and the gene copy number was obtained by using the standard curve with the given sample C tCre .

LacZ staining and quantificaiton
Mouse placentas were bisected. Half of each placenta was frozen with liquid nitrogen and half was mounted in freezing medium and then frozen with liquid nitrogen. Sets of 4-6 mm cryostat sections were obtained and fixed in 4% formaldehyde for 10 min. 59-bromo-4-chloro-3-indolyl-b-D galactopyranoside (Xgal) staining was performed as described [6]. The counterstain was nuclear fast red. Quantification of the LacZ staining was performed using the Image-Pro Plus software. The density of each zone of blue staining (positive for LacZ) was measured. The average densities of each zone of placentas were averaged and the SEM is indicated.

Statistical analysis
All data are expressed as the mean 6 SEM. Statistical significance of the differences between the mean values of multiple groups was tested by one-way ANOVA, followed by Tukey-Kramer post-tests. Data were analyzed for statistical significance using GraphPad Prism 4 software (GraphPad Software, San Diego, CA). A value of P,0.05 was considered significant.

Results
Random assembly of Tpbpa and Ada placental enhancers upstream of the Ada basal promoter to drive expression of a luciferase reporter gene Tpbpa and Ada placental enhancers are among the few enhancers known to confer placenta-specific gene expression in vivo. However, the utility of these enhancers has been limited by low expression levels. Thus, to generate a potentially stronger placental expression vector we chose to randomly assemble placental enhancers from the Tpbpa and Ada genes in front of the Ada basal promoter (for details see Methods and Materials section). To quantify the expression of these chimeric enhancer promoter combinations, a luciferease reporter vector was used as shown in Figure 1A.

Screening for expression constructs with trophoblast specific transcriptional activity
The various constructs were analyzed by transfection into the human trophoblast cell line (HTR) followed by the measurement of luciferase activity in cell extracts. All of the constructs exhibited enhanced luciferase expression in the trophoblast cell line except the enhancerless basal luciferase vector and a construct lacking the Ada basal promoter ( Figure 1B). Among all of the constructs tested, a particular construct in which the Ada enhancer was in the forward orientation and the Tpbpa enhancer was in the reverse orientation (Tpbpar/Adaf-AdaP) showed the highest luciferase activity in the trophoblast cell line ( Figure 1B). We also tested this construct for luciferase expression in a variety of non-trophoblast cells, including HEK293 (renal cell), Hela (cervical carcinoma), M1 (renal tubular epithelial cell), A529 (human lung epithelial cell), ML12 (mouse lung epithelial cell) and K562 (human leukemia cell). Notably, the luciferase activity driven by the double enhancer construct, Tpbpar/Adaf-AdaP, was significantly higher in trophoblast cells (HTR) than in non-trophoblast cells ( Figure 1C), suggesting that this construct is capable of enhanced expression in trophoblast cells. Thus, the Tpbpar/Adaf enhancer combination provoked the highest transcriptional activity among the constructs tested and provided gene expression that was restricted to trophoblast cells.
Tpbpa/Ada enhancer construct drives Cre-mediated recombination in cultured trophoblast cells The Cre/loxP system provides a powerful investigative strategy to study gene function in specific types of cells or tissues. To achieve trophoblast-restricted expression of Cre recombinase, we modified the Tpbpar/Adaf-AdaP expression construct by replacement of the luciferase reporter gene with cDNA encoding Cre recombinase equipped with a nuclear localization signal (NLS) (Figure 2A). Next, the ability of the Tpbpar/Adaf-AdaP-Cre construct to direct the synthesis of Cre recombinase in trophoblast cells was determined by a loxP site specific recombination assay as described previously [14]. HTR cells were co-tranfected with Credependent reporter genes, CAG-CATZ for PCR analysis or AdMA19 for luciferase bioassay, together with CMV-Cre or different amounts of Tpbpar/Adaf-AdaP-Cre transgene (Figure 2 B  & D). The CAG-CATZ plasmid harbors a CAT gene flanked by loxP sites and driven by the chicken b-actin promoter. Down-stream of CAT is the E coli b-galactosidase gene (lacZ) ( Figure 2B). The AdMA19 reporter gene contains a CMV promoter driving a luciferase transgene. However, efficient expression of the luciferase reporter depends on a Cre-mediated loxP recombinational event that removes of a spacer region separating the CMV promoter from the luciferase coding region ( Figure 2D). A 320-bp band was detected by PCR (using internal control primers, CAT2 and CAT3) from all cells transfected with CAG-CATZ. A 690 bp PCR product (representing recombination between the two loxP sites) was detected by primers AG and Z3 when using DNA isolated from cells cotransfected with either CMV-Cre or Tpbpar/Adaf-AdaP-Cre construct and loxP construct (CAG-CATZ). Only the 2100 bp PCR product, representing the original unrecombined DNA sequence, was detected when cells were transfected with CAG-CATZ alone ( Figure 2C).
Similarly, luciferase activity was only observed in the cells transfected with either CMV-Cre or Tpbpar/Adaf-AdaP-Cre and AdMA19 but not AdMA19 alone ( Figure 2D). Thus, both assays indicated that Cre recombinase functions in a dosage-dependent manner to promote loxP-dependent DNA recombination. Notably, the lucifearse activity mediated by the Cre recombinase in human trophoblast cells transfected with the Tpbpar/Adaf-AdaP double enhancer construct was even higher than that achieved by transfection with the CMV promoter driven Cre. Thus, these findings demonstrate the Tpbpar/Adaf-AdaP expression vector drives enhanced expression of Cre recombinase in human trophoblast cells.

Tpbpar/Adaf-AdaP-Cre transgene directs placentalrestricted expression in mice
To determine whether the chimeric Tpbpar/Adaf-AdaP construct drives placenta-specific Cre recombinase expression, the construct was used to generate transgenic mice. Live births were genotyped for the Cre transgene. Seventeen Tpbpar/Adaf-AdaP-Cre transgenic mice were identified from 40 pups by PCR using primers specific for the chimeric Tpbpar/Adaf-AdaP construct. Eight transgenic lines were further characterized. Six out of eight transgenic lines (75%) showed the placental-restricted expression. As shown in Figure 3A, Tg 1 and Tg6 showed relatively lower copies of Cre transgenes than those of Tg 5 as judged by analysis of genomic DNA from tails using real time PCR. Real time PCR (qPCR) data indicated that Tg5 contains 8 copies of Tpbpar/Adaf-AdaP transgene.
Next, to determine whether Tpbpar/Adaf-AdaP transgenes were only expressed in the placenta we mated females from all six Tpbpar/Adaf-AdaP-Cre transgenic lines with wild type FVB males. On gestation day 16.5 the Tpbpar/Adaf-AdaP-Cre transgenic females were sacrificed and placentas, fetuses and multiple maternal organs were collected. We found that Cre mRNA was expressed in approximately half of the placentas and was not detected in any of the maternal organs tested or in fetuses of Tpbpar/Adaf-AdaP-Cre transgenic pregnant mice. All six transgenic lines showed very similar transgene expression patterns. The expression pattern observed for one transgenic female (line 5, with the highest copy-number of the transgenes) is presented in Figure 3C and 3D. Notably, we confirmed that all placentas positive for Cre mRNA also contained the Tpbpar/Adaf-AdaP-Cre transgenes, while the placentas negative for Cre mRNA were also negative for the transgene (Figure 3B &C). Although the genotyping analysis by PCR showed fetuses contained the Tpbpar/Adaf-AdaP-Cre transgene, no Cre mRNA was detected in those fetuses ( Figure 3D & E). Thus, Tpbpar/Adaf-AdaP-Cre constructs appear to confer placental specific Cre mRNA expression.
To validate our RT-qPCR results, we used immunochemical analysis to determine the expression profile of Cre recombinase in the placenta and multiple other organs obtained from pregnant Tpbpar/Adaf-AdaP-Cre transgenic mice. Immunochemistry staining by anti-Cre antibody confirmed that Cre recombinase was not observed in the liver, kidneys, heart or spleens of pregnant female transgenic mice ( Figure 3G). However, Cre recombinase was observed in placentas containing Tpbpar/Adaf-AdaP-Cre transgenes (Cr + (Tg)) but not in those lacking this trangene (Cre 2 ) ( Figure 3F). ADA (adenosine deaminase), a well-known trophoblast marker, is used to identify trophoblast cells in placenta. Consistent with previous studies, ADA was expressed throughout the placenta but with significantly higher expression in spongiotrophoblast region ( Figure 3F). In placentas showing Cre mRNA expression, we found that Cre protein was present in giant cells (indicated by long arrow), spongiotrophoblast cells (indicated by short arrow) and cells in labyrinthine zone (indicated by arrow head), with highest expression in the spongiotrophoblast zone ( Figure 3F). Of particular note, Cre recombinase was observed inside of the nuclei of trophoblast cells ( Figure 3F, inset). Overall, our studies provide in vivo evidence that Tpbpar/Adaf-AdaP chimeric constructs drive the expression of Cre recombinase in placentas of transgenic mice.
Characterization of Tpbpar/Adaf-AdaP-Cre transgene through pregnancy Next, we determined Tpbpar/Adaf-AdaP-Cre transgene expression at multiple time points of pregnancy., We observed Cre transgene expression in placentas as early as E9.5, with expression increasing through E14.5 and E16.5 (Figure 4 A-B). Thus, transgene expression appears to display a pattern of increasing expression from E9.5 through E16.5 and presumably reflects an increase in trophoblast cell number.
Assessment of Cre recombinase activity in Tpbpar/Adaf-AdaP-Cre transgenic mice To evaluate the Cre-dependent DNA recombination in Tpbpar/ Adaf-AdaP-Cre transgenic mice, female transgenic mice were mated with Z/EG male mice, a double reporter mouse line that expresses enhanced GFP upon Cre-mediated excision of lacZ ( Figure 5A). Pregnant mice were sacrificed at E16.5, placentas and multiple maternal organs were collected for gene expression analysis and histological studies. The results showed ( Figure 5C) Cre mRNA was detected in approximately half of the placentas where it was correlated with the presence of the Tpbpar/Adaf-AdaP-Cre genotype ( Figure 5B). However, Cre mRNA in other maternal organs was not detected (data not shown). These results provide additional evidence that Tpbpar/Adaf-Ada transgenes induce placental specific Cre expression in transgenic mice. Because Cre-mediated excision of the lacZ gene allows expression of the GFP reporter, GFP mRNA was only detected in the placentas that carried both Tpbpar/Adaf-AdaP-Cre and lacZ-GFP transgenes ( Figure 5B, D lanes 6 and 9), while the placentas carrying only the lacZ-GFP transgenes or the Tpbpar/Adaf-AdaP-Cre trangene (LacZ 2 /Cre + ) were negative for GFP mRNA (Figure 5B, D). This finding demonstrates that the in vivo placenta-specific expression of Cre recombinase results in loxPdependent DNA recombination.
Next, histochemical staining for lacZ was used to determine Cre-mediated loxP-dependent DNA recombination at the cellular level. Consistent with our RT-PCR analysis, lacZ staining showed that constitutive expression of the lacZ reporter (blue staining) was significantly decreased in the spongiotroblast zone of placentas with lacZ + /Cre + genotype ( Figure 5E) compared with placentas having only the lacZ + transgene alone. Because of high expression of Cre in spongiotrophoblast cells layer seen in Tpbpar/Adaf-AdaP-Cre transgenic mice ( Figure 3F), lacZ quantification showed the greatest decrease in these cells ( Figure 5E). Quantification of lacZ staining in spongiotrophoblast cell layer was showed as Figure 5F. Taken together, these data provide in vivo evidence that Tpbpar/ Adaf-AdaP-Cre transgenic mice with placenta specific expression of Cre recombinase are capable of conducting loxP mediated DNA recombination in vivo.

Discussion
Here we report the development and characterization of a Tpbpa/Ada-AdaP chimeric enhancer transgene that confers a high level of trophoblast specific expression in cultured cells and in transgenic mice. Using this double enhancer construct to drive the expression of Cre recombinase, we demonstrated Cre-mediated loxP-dependent DNA recombination in the placenta but not in the maternal organs tested or in fetuses of transgenic mice. This double enhancer construct should be a useful genetic tool to manipulate placental gene expression in mice. This placenta expression vector should provide investigative opportunities to understand the functional role of specific genes in placental development and placenta-related pregnancy disorders.   Numerous earlier reports have attempted to identify and characterize placental gene regulatory elements. Several groups [19,20,21,22] have used transgenic mouse approaches to show that a 5.4,6.0-kb promoter and 59-flanking sequence of HLA-G contains trophoblast-restricted regulatory elements. However, the level of reporter gene expression in transgenic placentas at day 12.5 was relatively low and 450 times less than endogenous bactin. Analysis of the murine Ada gene by Shi et al [10] showed that a placenta regulatory element resided within a 1.8 kb segment of DNA in the 59 flanking region and that this element provided consistent but variable placenta-specific expression. A trophoblast specific enhancer derived from the 59 flanking region of the Tpbpa gene was also inconsistent in driving placenta specific expression in transgenic mice. Calzonetti et al [6] showed that a 340 bp fragment provided placenta specific expression of a lacZ reporter gene in only 5 of 16 transgenic lines (31.2%) examined. In recent years, gene transfer strategies, aimed at targeting genes to the trophoblast lineages, have been used in efforts to overcome this limitation. For example, direct injection of gene-therapy vectors into placentas results in limited levels of gene expression in the placenta, but also results in serious injury and patchy expression [23,24]. Trophoblast-specific gene manipulation using lentivirusbased vectors has recently been developed and used in mice and rats [25,26]. However, the lentivirus-based vector mediated trophoblast-specific gene expression requires blastocyst isolation, incubation with lentivirus vectors and the microinjection of transduced blastocysts into pseudopregnant mice [25,26]. This approach is expensive, time consuming and inconvenient for general laboratory use. Thus, development of efficient, noninvasive and convenient genetic tools to specifically manipulate gene expression in placenta is desperately needed and would greatly facilitate efforts to understand placental formation and fetal development.
In order to construct a more robust and reliable placenta specific expression construct, we assembled a chimeric placental expression vector using placental enhancer elements from two genes, Tpbpa and Ada. Each of these genes has been previously characterized by prenatal expression in the placenta, with highest expression occurring in the spongiotrophoblast layer. Using this combinatorial strategy, we identified a novel enhancer combination containing Tpbpa and Ada regulatory elements driving transcription from the Ada-basal promoter. We prepared transgenes using a combination of Tpbpa and Ada enhancer elements. From seventeen transgenic mice were identified eight transgenic lines were developed by mating with nontransgenic FVB mice. Six of eight transgenic lines (75%) revealed Cre expression in placenta specifically. Thus this chimeric Tpbpar/Adaf-AdaP construct showed more robust and reliable placenta specific transcription activity in transgenic mice.
We assembled a chimeric placental expression vector using placental enhancer elements from the Tpbpa and Ada genes. We have used the newly characterized expression construct to achieve placenta specific loxP-dependent DNA recombination mediated by a Tpbpa/Ada chimeric enhancer transgene encoding Cre recombinase. Notably, this chimeric enhancer construct is capable of driving cre gene expression in the placenta as early as E9.5, and continued expression through 16.5. Cre recombinase was observed in giant cells, spongiotrophoblasts and labyrinthine region in Tpbpar/Adaf-AdaP-Cre transgenic mice. However, highest expression was observed in the spongiotrophoblast layer, in agreement with earlier studies of the Tpbpa and Ada placental regulatory elements.
Inadequate placenta development is associated with a high incidence of early embryonic lethality [1] and serious pregnancy disorders, such as preeclampsia [3,27], fetal defects, fetal loss and IUGR [2,13,28]. Thus, for some mutant mice it is difficult to determine whether a prenatal lethal phenotype is caused by placental defects, fetal defects, or both. In addition, many gene disruptions result in embryonic lethality because of abnormal placental development and thereby prevent research opportunities to study the role of that gene in other organs prenatally and postnatally. Thus, the use of placenta specific expression constructs to drive the expression of a gene of interest in the placenta allows us to differentiate the causative factors of knockout phenotypes. For example, genetically restoring ADA enzymatic activity to placentas of Ada-deficient fetuses corrected most of the prenatal purine metabolic disturbances, prevented serious fetal liver damage, and rescued ADA-deficient fetuses from perinatal lethality [16]. Therefore, the placental-specific expression vector reported here is likely to provide novel possibilities to genetically restore placental expression of a gene of interest and thereby rescue embryonic lethality of mutant mice caused by placental defects.