MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tanaka, H.
Right arrow Articles by Kataoka, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tanaka, H.
Right arrow Articles by Kataoka, H.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, July 2005, p. 5687-5698, Vol. 25, No. 13
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.13.5687-5698.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Hepatocyte Growth Factor Activator Inhibitor Type 1 (HAI-1) Is Required for Branching Morphogenesis in the Chorioallantoic Placenta

Hiroyuki Tanaka,1 Koki Nagaike,1 Naoki Takeda,2 Hiroshi Itoh,1 Kazuyo Kohama,1 Tsuyoshi Fukushima,1 Shiro Miyata,1 Shuichiro Uchiyama,1 Shunro Uchinokura,1 Takeshi Shimomura,3 Keiji Miyazawa,4 Naomi Kitamura,5 Gen Yamada,2 and Hiroaki Kataoka1*

Second Department of Pathology, Faculty of Medicine, University of Miyazaki, Miyazaki, Japan,1 Center for Animal Resources and Development, Graduate School of Molecular and Genomic Pharmacy, Kumamoto University, Kumamoto, Japan,2 Yokohama Research Center, Mitsubishi Pharma Corporation, Yokohama, Japan,3 Department of Molecular Pathology, Graduate School of Medicine, University of Tokyo, Tokyo, Japan,4 Department of Biological Sciences, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, Yokohama, Japan5

Received 16 December 2004/ Returned for modification 3 February 2005/ Accepted 31 March 2005


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Hepatocyte growth factor activator inhibitor type 1 (HAI-1) is a membrane-associated Kunitz-type serine proteinase inhibitor that was initially identified as a potent inhibitor of hepatocyte growth factor activator. HAI-1 is also a cognate inhibitor of matriptase, a membrane-associated serine proteinase. HAI-1 is expressed predominantly in epithelial cells in the human body. Its mRNA is also abundant in human placenta, with HAI-1 specifically expressed by villous cytotrophoblasts. In order to address the precise roles of HAI-1 in vivo, we generated HAI-1 mutant mice by homozygous recombination. Heterozygous HAI-1+/– mice underwent normal organ development. However, homozygous HAI-1–/– mice experienced embryonic lethality which became evident at embryonic day 10.5 postcoitum (E10.5). As early as E9.5, HAI-1–/– embryos showed growth retardation that did not reflect impaired cell proliferation but resulted instead from failed placental development and function. Histological analysis revealed severely impaired formation of the labyrinth layer, in contrast all other placental layers, such as the spongiotrophoblast layer and giant cell layer, which were formed. Our results indicate that mouse HAI-1 is essential for branching morphogenesis in the chorioallantoic placenta and lack of HAI-1 function may result in placental failure.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Hepatocyte growth factor activator inhibitor type 1 (HAI-1) is a membrane-associated Kunitz-type serine proteinase inhibitor that has five domains: two extracellular Kunitz-type serine proteinase inhibitor domains, an extracellular low-density lipoprotein (LDL) receptor class A domain, a transmembrane domain, and a short intracytoplasmic domain, including the carboxyl-terminal end (13). Thus, this inhibitor is a type 1 transmembrane protein that functions on the cellular surface (13). HAI-1 was originally identified as a potent inhibitor of hepatocyte growth factor activator (HGFA) (37). At present, two important serine proteinases have been identified as target enzymes of HAI-1, HGFA and matriptase (13).

HGFA is a serum serine proteinase that specifically converts the inactive proform of hepatocyte growth factor/scatter factor (pro-HGF/SF) to an active form, HGF/SF, in response to tissue injury (26, 27). Because HGF/SF is a pleiotropic factor with multifunctional roles in a variety of cells via its high-affinity receptor, c-Met receptor tyrosine kinase (42), HGFA might be involved in a number of pathophysiological phenomena in vivo (10, 12, 13). Matriptase is also a cognate proteinase for HAI-1 (22). This proteinase is a type 2 transmembrane protein with an extracellular catalytic domain (20, 38). Matriptase also activates pro-HGF/SF (18, 19), and thus, HGFA and matriptase might have compensatory roles in the activation of pro-HGF/SF in vivo. Consequently, HAI-1 would be a critical regulatory molecule in pericellular activation of pro-HGF/SF (13).

In addition to pro-HGF/SF, matriptase also demonstrates processing activities of pro-urokinase-type plasminogen activator and protease-activated receptor 2 (19, 38). Several extracellular matrices are also sensitive to matriptase (22), and processing of filaggrin by matriptase is essential for maturation of epidermal keratinocytes (24). Therefore, cellular matriptase has important roles in a number of cellular events, such as migration, remodeling of extracellular matrix, and cellular maturation, and it is possible that HAI-1 is involved in regulation of these events.

In addition to proteinase inhibitor domains, HAI-1 possesses an LDL receptor class A domain, a transmembrane domain, and a short intracytoplasmic domain (13). The LDL receptor class A domain appears to have a regulatory function in the activation of promatriptase on the cellular surface (30). On the other hand, little is known regarding the function of the intracytoplasmic domain, and it remains to be determined whether HAI-1 mediates signaling from outside to intracellular machineries.

Although several biochemical activities have been demonstrated for HAI-1 in vitro, its role in vivo has yet to be established. An RNA blot analysis indicated that human HAI-1 is expressed across multiple tissues, with the highest mRNA level in the placenta (37). Subsequent immunohistochemical analyses revealed localization of HAI-1 protein on the basolateral surface of epithelial cells (15). In the placenta, HAI-1 was specifically expressed in villous cytotrophoblasts (11). Indeed, specific antibody for HAI-1 was useful to isolate cytotrophoblasts in high purity from human placental tissue in vitro (32). Taken together, these findings suggest that HAI-1 may be important in epithelial function as well as in placental development and function.

In order to address the roles of HAI-1 in vivo, we have generated HAI-1-deficient mice and studied the effects of deficiency on placental and embryonic development.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Construction of HAI-1-targeting vector. The mouse HAI-1 gene was cloned from a mouse (129/SvJ) genomic bacterial artificial chromosome (BAC) library (Genome Systems), and two DNA fragments digested with BamHI were found to cover the entire mouse HAI-1 gene, as described previously (8). These clones were subcloned into a pBluescript II SK+ phagemid vector (Stratagene) and used to construct a targeting vector by replacing the coding regions (exons 5 to 11) of the mouse HAI-1 gene, which correspond to the first Kunitz domain, the LDL receptor class A domain, the second Kunitz domain, the transmembrane domain, and the intracytoplasmic domain of HAI-1, with the neomycin resistance gene cassette. This was accomplished by inserting 1.2 kbp of short-arm DNA fragment and 6 kbp of long-arm DNA into a modified pKO vector (Lexicon Genetics Inc.), which contains the neomycin resistance gene cassette as well as the herpes simplex virus thymidine kinase gene (Fig. 1A).



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 1. (A) Schematic representation of the targeting strategy for the mouse HAI-1 gene. Exons 5 to 11, corresponding to Kunitz domain 1 (KD1), the LDL receptor class A domain (LDLR), Kunitz domain 2 (KD2), the transmembrane domain (TM), and a part of the intracytoplasmic domain of mouse HAI-1 (mHAI-1) protein were replaced by the neomycin resistance gene (neo) cassette. The position of the epitope of the antibody used in this study is indicated. Restriction sites are indicated as follows: B, BamHI; C, ClaI; E, EcoRI; H, HindIII; K, KpnI; N, NotI; S, ScaI; X, XhoI. The positions of probes 1 and 2 for Southern blot analyses and the expected fragments detected by each probe were also indicated. HSV-TK, herpes simplex virus thymidine kinase gene. (B) PCR-based genotyping of wild-type and mutant alleles. The wild-type fragment was confirmed by using a PCR primer pair, P1 and P2 or P4 and P5, to amplify a 1,249-bp or 588-bp product, respectively. Mutant recombination was confirmed by using a PCR primer pair, P1 and P3, to amplify a 1,309-bp product. The position of each primer is shown in panel A. Representative results of wild-type (+/+), heterozygous mutant (+/–), and homozygous mutant (–/–) mice are shown. (C) Southern blot analysis of embryos. Ten µg of genomic DNA was digested with HindIII and probed by a 2.7-kbp HindIII/KpnI fragment (probe 1) to distinguish HAI-1 wild-type and mutant alleles, yielding 10.3- and 7.7-kbp bands for the wild-type and mutant alleles, respectively (upper panel). Ten µg of genomic DNA was digested with ScaI/ClaI and probed by probe 2 to distinguish HAI-1 wild-type and mutant alleles, yielding 3.4- and 4.7-kbp bands for the wild-type and mutant alleles, respectively (lower panel). (D) RT-PCR for the expression of mouse HAI-1 mRNA. Analyses were performed using intron-spanning primers and total cellular RNA extracted from HAI-1+/+, HAI-1+/–, and HAI-1–/– embryos at E9.5 and E10.5. Two independent experiments using different primer sets are shown. Expression of HAI-1 mRNA was absent in HAI-1–/– embryos. For an internal control, GAPDH mRNA was amplified.

 
Gene targeting in ES cells and generation of gene-disrupted mice. The targeting vector was linearized with NotI and electroporated into TT2 embryonic stem (ES) cells (41). A positive-negative selection strategy was then used to determine which ES cell clones underwent homologous recombination. ES cells were exposed to medium containing 200 µg/ml G418 (Sigma) and 1 µmol/liter ganciclovir (Nacalaitesque). DNAs isolated from surviving clones were then screened by PCR as described below, and positive clones were reconfirmed by Southern blot analysis with two external probes (probes 1 and 2; Fig. 1A) confirmed the 5' and 3' sites of the integration of the HAI-1 gene.

Positive clones from ICR donors were identified, expanded, and microinjected into the eight-cell stage, and high-percentage chimeric mice were obtained. The chimeras were bred to C57BL/6 mice, and the agouti-colored offspring (F0) with germ line transmission of the targeted mutation were crossed again to C57BL/6 wild-type mice, and heterozygous offspring were crossed to produce homozygous mutant offspring. The genotypes of these mice were assessed by PCR analysis for 45 cycles using primers 5'-GGGTTCTAGAAGTTCTGTGGGTGTAAGGAT-3'(forward primer for wild-type and mutant alleles: P1), 5'-GAGGTTGCCTGGGCAACAAGAACAA-3' (forward primer for wild-type allele: P4), 5'-GGCCCACCTTGTAGGATGCGAGGCAATA-3' (reverse primer for wild-type allele: P2), 5'-AGGGACCTAATAACTTCGTATAGGATACTT-3 (reverse primer for mutant allele: P3), and 5'-GAAGCCATCGATACAGCAGCCATTG-3' (reverse primer for wild-type allele: P5) in order to detect 1,249-bp (P1 versus P2) and 588-bp (P4 versus P5) products for the wild-type allele and a 1,309-bp (P1 versus P3) product for the mutant allele. The thermal cycle profile was 30 s at 95°C, 1 min at 60°C, and 2 min at 72°C.

The genotypes of embryos were determined by performing PCR of genomic DNA obtained from the tail or forelimb of each embryo. Manipulation of ES cells and generation of chimeric mice were achieved at the Center for Animal Resources and Development of Kumamoto University, while mice breeding was done at the Animal Center of the University of Miyazaki under guidelines for animal and recombinant DNA experiments.

Southern and Northern blotting analyses and reverse transcription-PCR. For Southern blot analysis, 10 µg of genomic DNA was digested with HindIII or ScaI/ClaI (New England Biolabs Inc.), electrophoresed on a 1% agarose gel, and transferred onto Hybond-N+ nylon membranes (Amersham). The membrane was probed with probe 1 (a 2.7-kbp HindIII/KpnI fragment) or probe 2 (Fig. 1A). Probe 2 was prepared by PCR using primers 5'-TTTGAGACAGGATCTCACGATATAGCTCAA-3' (forward primer) and 5'-ATAAAGGACTTCAGGACATTACCTGGGAT-3' (reverse primer), resulting in a 1,377-bp product (Fig. 1A).

Northern blot analysis for HAI-1 mRNA was performed by using 20 µg of total cellular RNA prepared from mouse embryo and placental tissues at each gestation period (E7.5 to E18.5) (Seegene, Inc.). For a probe, an 823-bp fragment, corresponding to bases 9 to 831 of mouse HAI-1 cDNA (9), was prepared by reverse transcription (RT)-PCR. Probes were radiolabeled by random priming with [{alpha}-32P]CTP. Membranes were autoradiographed with Kodak XAR-5 film (Eastman Kodak) at –80°C for 18 or 72 h. For internal control of loading, the Northern blot membranes were subsequently hybridized to a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe (Clontech).

For RT-PCR, 1 µg of total cellular RNA derived from mouse embryo or placenta was reverse-transcribed by random hexamer and oligo(dT)12 mixed primers and SuperScript II reverse transcriptase (Invitrogen). The resultant cDNAs derived from embryos at E9.5 and E10.5 were subjected to PCR for mouse HAI-1. Also, the resultant cDNAs derived from placentas at E9.5 and E10.5 were subjected to PCR for mouse GCM1 (glial cell missing 1), which is expressed in the labyrinth layer of mouse placenta (2, 33, 36), mouse Hand1, which is expressed in trophoblast giant cells (33), mouse Mash2, which is expressed in trophoblast stem cells, ectoplacental cone, and spongiotrophoblasts (33), mouse Tpbpa (trophoblast specific protein {alpha}; 4311), which is expressed in spongiotrophoblasts (20), or mouse placental lactogen-1 (Pl-1), which is expressed by trophoblast giant cells (6). For an internal control, RT-PCR for GAPDH mRNA was also performed. The thermal cycle profile was 30 s at 95°C, 1 min at 60°C, and 1 min at 72°C.

Primer sequences were as follows: HAI-1 forward#1, 5'-CTCGCATCCTACAAGGTGGGCCGCT-3', and reverse#1, 5'-TCCAGAACTGGCCTGAGTGCAGCTG-3' with an expected product size of 968 bp; HAI-1 forward#2, 5'-CTCAGACCAACCAGAGGAAA-3', and reverse#2, 5'-GAGATTCCTTGCACATCCTT-3, with a product size of 243 bp; Mash2 forward, 5'-GCCCGGAGCATGGAAGCACACCTT-3', and reverse, 5'-TCAGTAGCCCCCTAACCAACTGGA-3', with a product size of 798 bp; Tpbpa forward, 5'-GAGACATGACTCCTACAATCTTCC-3', and reverse, 5'-TTGCCTAACTTCATACTGCTGTCC-3', with a product size of 407 bp; Pl-1 forward, 5'-TCCGCAGGAATGCAATTGTTGCTG-3', and reverse, 5'-GGGAAAGCATTACAAGTCTGGTTC-3', with a product size of 711 bp; GCM1 forward, 5'-AAACACATCTACAGCTCGGACGACA-3', and reverse, 5'-CTCAGTGCTTCCCCCCAAATCATAA-3', with a product size of 624 bp; Hand1 forward, 5'-CTCTCCAACATGAACCTCGTGGGC-3', and reverse, 5'-GGTCTCACTGGTTTAGCTCCAGCG-3', with a product size of 664 bp; GAPDH forward, 5'-AAAATGGTGAAGGTCGGTGT-3', and reverse, 5'-TTTGATGTTAGTGGGGTCTC-3', with a product size of 255 bp.

For the quantification of GCM1 mRNA levels, real-time RT-PCR analysis was performed using total cellular RNA extracted from E10.5 placentas. Real-time PCR on the LightCycler (Roche Diagnostics) was performed with a master mix of LightCycler DNA Master SyberGreen I (Roche Diagnostics) according to the manufacturer's instructions. For an internal control, real-time RT-PCR for ß-actin mRNA was also performed. The thermal cycle profile for GCM1 was 10 s at 95°C, 10 s at 61°C, and 5 s at 72°C. The thermal cycle profile for ß-actin was 10 s at 95°C, 10 s at 54°C, and 5 s at 72°C.

Product amplification specificity was determined by melting curve analysis. Primer sequences were as follows: GCM1 forward, 5'-GGTCATTCCAGGAAGGCGTCCAACT-3', and reverse, 5'-CTCAGTGCTTCCCCCCAAATCATAA-3', with a product size of 131 bp; and ß-actin forward, 5'-AGAGGGAAATCGTGCGTGAC-3', and reverse, 5'-CAATAGTGATGACCTGGCCGT-3', with a product size of 138 bp.

In situ hybridization. In situ hybridization of digoxigenin-labeled probes to frozen sections was described previously (29). Briefly, 4-µm frozen sections were prepared and fixed in freshly prepared 4% paraformaldehyde in phosphate-buffered saline (PBS) for 16 h. Then the sections were air-dried and rinsed in nuclease-free water and used for the subsequent in situ hybridization study. Probes detecting trophoblast giant cells (expressing Pl-1) and spongiotrophoblast cells (expressing Tpbpa) were prepared to discriminate the different placental layers. The cDNAs of each Pl-1 and Tpbpa were generated by RT-PCR as described above, and subcloned into the TOPO TA cloning vector (Invitrogen). In vitro transcription to generate digoxigenin-labeled RNA probes was carried out according to the manufacturer's instructions (Roche Diagnostics). The same amount of each antisense or sense probe was used for hybridization, in which the sense probe was used as a negative control. The in situ hybridization reaction was performed using a fully automated in situ hybridization apparatus (Ventana HX System Discovery and RiboMap system; Ventana), according to the manufacturer's instruction. Hybridization was performed by using 1 ng per slide of digoxigenin-labeled RNA probe at 65°C for 6 h. After the hybridization, signals were detected by biotin-labeled antidigoxigenin antibody. The reaction was detected with a BlueMap kit (Ventana) and counterstained with nuclear fast red.

Histology and immunohistochemistry. After cervical dislocation of pregnant mice, the embryo and placenta tissue were fixed in 4% paraformaldehyde-PBS overnight and then dehydrated and embedded in paraffin. Four-µm-thick sections were prepared and stained with hematoxylin and eosin. For immunohistochemistry, the sections were processed for antigen retrieval (autoclaving in 10 mM citrate buffer, pH 6.0, or 1 mM EDTA, pH 8.0, for 5 min), followed by treatment with 3% H2O2 in PBS for 10 min and washed in PBS twice. After blocking in 3% bovine serum albumin and 10% normal goat serum in PBS for 1 h at room temperature, the sections were incubated with primary antibodies for 16 h at 4°C.

The anti-mouse HAI-1 was a rabbit polyclonal antibody prepared by immunizing a synthesized peptide which corresponds to the carboxyl terminus of intracytoplasmic domain of mouse HAI-1 (Pro484 to Leu507) (29). This epitope corresponded to the outside of the deleted portion by insertion of the neomycin resistance gene cassette (Fig. 1A). After immunization, immunoglobulin G (IgG) was prepared from the rabbit serum, followed by immunopurification using the antigen-peptide affinity column. The specificity of the antibody was verified by immunoblot analysis using cellular extracts of Chinese hamster ovary (CHO) cells transfected with the pCIneo expression vector (Promega) containing full-length mouse HAI-1 cDNA. The procedures for protein extraction and immunoblot were described previously (29). Ki-67 monoclonal rat anti-mouse antibody (clone TEC-3, DAKO) (1:50 dilution) and anti-cleaved caspase-3 (Asp175) rabbit polyclonal antibody (Cell Signaling Technology, Inc.) were also used. Negative controls consisted of omission of the primary antibody. The sections were then washed in PBS and incubated with Envision-labeled polymer reagent (DAKO) or with Histofine simple stain mouse MAX-PO (rat) reagent (NICHIREI) for 45 min at 37°C. The reaction was revealed with nickel, cobalt-3,3'-diaminobenzidine. (Immunopure metal enhanced DAB substrate kit; Pierce), and the sections were counterstained with hematoxylin.

BrdU labeling. Timed pregnant mice from heterozygous matings were injected intraperitoneally at E9.5 with 5-bromo-2-deoxyuridine (BrdU) (50 µg/g of body weight; Sigma). The mice were sacrificed 1.5 h after injection, and the uteri were removed and used for the immunohistochemical analysis with mouse monoclonal antibody against BrdU (clone BU33; Sigma). Formalin-fixed, paraffin-embedded sections were pretreated with 2 N HCl for 30 min at 37°C followed by treatment with 0.02% trypsin in CaCl2 for 20 min. Then the sections were immunostained with anti-BrdU antibody at a dilution of 1:200 for 16 h at 4°C. The sections were then washed in PBS and incubated with Envision-labeled polymer reagent (DAKO) for 45 min at 37°C. The reaction was revealed as described above, and the sections were counterstained with hematoxylin.

Transplacental passage of rhodamine 123. Timed pregnant mice from heterozygous matings were injected intraperitoneally at E9.5 with rhodamine 123 (1 µg/g of body weight; DOJINDO Laboratories). The mice were sacrificed 2 h or 6 h after the injection, and the embryos were removed from the uteri and analyzed with a fluorescence stereomicroscope (MZ16F; Leica Microsystems) with yellow fluorescent protein filter (excitation at 510 and 20 nm, emission at 560 and 40 nm). The fluorescence intensity in the entire embryo was measured with Leica FW4000TZ software (Leica Microsystems).


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Targeted disruption of the HAI-1 gene in mice. To disrupt the HAI-1 gene, a neo gene cassette was inserted that replaced exons 5 to 11, removing all functional domains (both Kunitz-type proteinase inhibitor domains, LDL receptor class A domain, transmembrane domain, and a part of the intracytoplasmic domain) of HAI-1 (Fig. 1A). We expected this insertion to result in a loss-of-function mutation in the HAI-1 gene. The targeting vector contained 1.2 kbp of genomic sequence at the 5' end and 6 kbp at the 3' end (Fig. 1A). The vector was electroporated into ES cells, and five homologous recombinants out of 244 G418-resistant clones were identified. PCR analyses of the genomic DNA from these clones revealed the presence of both the mutant allele and wild-type allele (HAI-1+). Southern blot analyses of these clones also confirmed the homologous recombination in the target locus (7.7-kbp fragment with probe 1 and 4.7-kbp fragment with probe 2; Fig. 1A and C) in addition to wild-type fragments (10.3 kbp with probe 1 and 3.4 kbp with probe 2; Fig. 1A and C). To generate mice carrying the targeted HAI-1 allele, two independent ES clones (clones A and B) were used successfully to establish germ line chimeras, and the resulting heterozygous mutant offspring were HAI-1+/–. No phenotypic differences were observed between heterozygous adult mice derived from the two independent clones.

Lack of functional HAI-1 resulted in retarded intrauterine growth and development and subsequent lethality around E10.5. Heterozygous (HAI-1+/–) mice were intercrossed to produce HAI-1 homozygous mutant (HAI-1–/–) offspring. Among 21 live-born progeny derived from this mating of two independent ES clones, no viable homozygous mutant offspring (HAI-1–/–) were identified, whereas 35 (32.7%) wild-type (HAI-1+/+) and 72 (68.3%) heterozygous (HAI-1+/–) offspring were obtained, indicating that HAI-1–/– mice died as embryos (data not shown). In order to determine the time point of lethality and to analyze the morphological phenotype of mutant embryos, embryos at different stages of gestation were analyzed for genotype by PCR (Fig. 1B) and Southern blot analyses (Fig. 1C). In parallel, we performed RT-PCR analysis with use of the total RNA of E9.5 and E10.5 embryos. The results indicated that normal HAI-1 mRNA was absent in homozygous HAI-1–/– embryos and appeared to be decreased relative to wild-type levels in HAI-1+/– embryos (Fig. 1D). Normal HAI-1 protein was also not detectable in HAI-1–/– embryos (see below at Fig. 4 and 6).



View larger version (129K):
[in this window]
[in a new window]
  		FIG. 4. Immunohistochemical analyses for the expression and localization of HAI-1 protein in mouse embryos. At E10.5 (A to D), immunoreactivity for HAI-1 was observed in the midgut of wild-type (+/+) embryos (A) but not of homozygous mutant (–/–) embryos (B). Higher magnifications of the indicated areas in panels A and B are shown in panels C and D, respectively. At E18.5 (E to H), mouse HAI-1 protein was expressed in the epithelial cells of the esophagus (arrow) (E), stomach (arrow), and pancreas (arrowhead) (F) and also of the intestine (G) of wild-type HAI-1+/+ embryos, showing its localization on the basolateral surface of epithelial cells (H). Weak reactivity was observed in the liver tissue (arrowhead) present in panel E. In the placenta tissue of the HAI-1+/+ embryo at E18.5 (I and J), HAI-1 protein was specifically expressed in labyrinthine trophoblasts and chorionic trophoblasts of the wild-type normal placenta. (K) Immunoblot analysis of mouse HAI-1 with the antibody used in the immunohistochemical study. Cellular extracts of CHO cells transfected with the mouse HAI-1 expression vector (lane 1) and with empty vector (lane 2) were used. Note that the 66-kDa mature HAI-1 and {approx}10-kDa carboxyl-terminal fragment of HAI-1 after ectodomain shedding were specifically labeled with the antibody.

 


View larger version (100K):
[in this window]
[in a new window]
  		FIG. 6. Expression of HAI-1 protein in placental tissue at E9.5. HAI-1+/+, HAI-1+/–, and HAI-1–/– placentas were simultaneously immunostained on the same slide. Distinct immunoreactivity was observed in the labyrinth layer of the HAI-1+/+ placenta (A). The HAI-1+/– placenta showed mildly decreased immunoreactivity (B). The immunoreactivity was absent in the HAI-1–/– placenta (C). Arrowheads indicate positively stained maternal uterine epithelial cells that served as an internal positive control for the immunostaining. Higher magnification of the slide revealed that HAI-1 is expressed by clusters of cuboidal trophoblasts at the interface between the chorionic plate and labyrinth layer as well as in differentiated labyrinthine syncytiotrophoblasts (D and E). The surface localization of HAI-1 is noted (E). Immunoreactivity was completely absent in the HAI-1–/– placenta (F).

 
Viable HAI-1–/– embryos were found from E6.5 to E9.5, and similar results were obtained from both independent ES clones (Tables 1 and 2). At E10.5, although viable homozygous embryos were still detectable, dead and degenerating embryos were intermingled, suggesting that the lethal point was around E10.5. Growth retardation was occasionally suggested as early as at E9.5 (Fig. 2A). Surviving HAI-1–/– embryos at E10.0 and E10.5 could be easily distinguished from their littermates by their smaller size, pale color, and enlarged pericardium (Fig. 2B). HAI-1–/– embryos could occasionally be isolated at E11.5 but were not alive; they showed strong growth retardation and were often already in the process of resorption (Fig. 2C). Histologically, retardation of development was prominent in HAI-1–/– mice at E10.0, producing the histological findings that are compatible with E9.5 (Fig. 2D). The same morphological findings were obtained in mice derived from both independent ES clones.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Summary of the genotyping of embryos derived from ES clone Aa

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Summary of the genotyping of embryos derived from ES clone Ba

 


View larger version (107K):
[in this window]
[in a new window]
  		FIG. 2. Morphology of the embryos. (A) Embryos of HAI-1+/+ (+/+) and HAI-1–/– (–/–) mice at E9.5. Note that the HAI-1–/– embryo is slightly smaller than that of the HAI-1+/+ embryo. Bar, 2 mm. (B) Embryos of HAI-1+/+ and HAI-1–/– mice at E10.5. The HAI-1–/– embryo is smaller than the HAI-1+/+ embryo. The arrow indicates the enlarged pericardium due to pericardial effusion in the HAI-1–/– embryos. Bar, 2 mm. (C) Embryos of HAI-1+/+ and HAI-1–/– mice at E11.5. The HAI-1–/– embryos were not viable at E11.5 and were apparently smaller than the HAI-1+/+ embryos. Bar, 2 mm. (D) Histology of HAI-1+/+ and HAI-1–/– embryos at E10.0. The HAI-1–/– embryo showed retardation of development compared with the HAI-1+/+ embryo. Bar, 0.1 mm.

 
Expression and localization of HAI-1 in embryo and placental tissue. The lethal phenotype of HAI-1–/– embryos indicated that HAI-1 has an indispensable function in fetal development, and we analyzed expression and localization of HAI-1 during the course of gestation. As early as E7.5, HAI-1 mRNA could be detected by Northern blot analysis by using total RNA obtained from combined embryonic and placental tissues (Fig. 3A). Expression levels in the embryos and also in placental tissue increased with the course of gestation (Fig. 3B and C). Immunohistochemically, HAI-1 protein was detectable in the midgut of wild-type embryos at E10.5 (Fig. 4A and C), but this signal was completely absent in HAI-1–/– embryos (Fig. 4B and D).



View larger version (47K):
[in this window]
[in a new window]
  		FIG. 3. Expression of HAI-1 mRNA in normal and wild-type (HAI-1+/+) embryos and placentas. (A) Northern blot analysis for HAI-1 mRNA in combined embryo and placental tissues at E7.5 to 9.5. (B) Northern blot analysis for HAI-1 mRNA in embryo at E10.5 to 18.5. (C) Northern blot analysis for HAI-1 mRNA in placentas at E10.5 to 18.5. The relative abundance of HAI-1 mRNA is expressed as a ratio of the integrated absorbance of the test band to that of the corresponding GAPDH band and shown in panels B and C as bar graphs.

 
Because the epitope of the antibody used was present in the carboxyl-terminal end of mouse HAI-1 outside the portion deleted by gene targeting (Fig. 1A), the total absence of immunoreactivity would indicate the absence of truncated HAI-1 derived from cryptic mRNA in HAI-1–/– embryos. At E18.5, the immunoreactivity of HAI-1 of wild-type embryos was significantly enhanced. Immunoreactivity was most abundant in the epithelial cells lining the digestive tract (Fig. 4E, F, and G) and skin (not shown). At higher magnification, basolateral cell surface localization of HAI-1 was confirmed (Fig. 4H). The distribution and subcellular localization of HAI-1 in mouse were similar to those in human tissues (15). In the term murine placenta (E18.5), HAI-1 was specifically expressed in labyrinthine syncytiotrophoblasts and chorionic trophoblasts (Fig. 4I and J), and this finding was also compatible with the previous observation that human HAI-1 was specifically expressed in villous cytotrophoblasts (11, 32). The specificity of the antibody used in this study was further confirmed by using CHO cells transfected with the mouse HAI-1 expression vector (Fig. 4K). A 66-kDa membrane-form mature HAI-1 band and a {approx}10-kDa carboxyl-terminal fragment produced by ectodomain shedding of HAI-1 (29) were specifically detected with this antibody by an immunoblot analysis.

Lack of functional HAI-1 resulted in impaired formation of labyrinth layer of placenta. In humans, HAI-1 mRNA is most abundant in placental tissue (11, 37). Because of similar placental expression of HAI-1 in mice, we assumed that the growth retardation and embryonic lethality of HAI-1 mutant embryos may be caused by impaired placental function due to HAI-1 deficiency. Therefore, we inspected the placental tissues of HAI-1–/– embryos histologically. At E8.5, fusion of the chorion and allantois had occurred in HAI-1–/– gestations, however, the chorionic plate was thin and branching of fetal vessels was insufficient compared with HAI-1+/– and HAI-1+/+ placentas (Fig. 5A, B, and C). Notable differences were not observed between HAI-1+/– and HAI-1+/+ placentas.



View larger version (145K):
[in this window]
[in a new window]
  		FIG. 5. Histological analyses of the placenta in wild-type (+/+), heterozygous (+/–), and homozygous (–/–) HAI-1 mutant placentas at E8.5 to E9.5. At E8.5 (A to C), chorioallantoic attachment in HAI-1–/– placenta (C, upper panel) had already occurred as well as in wild-type (A, upper panel) and heterozygous (B, upper panel) placentas. Nucleated red blood cells in fetal vessels of the allantois, indicating a functional connection with the chorion, were noted in all embryos. In addition, these vessels showed branching morphogenesis in HAI-1+/+ (A, lower panel) and HAI-1+/– embryos (B, lower panel). However, these vessels did not penetrate the labyrinth layer of the HAI-1–/– placenta (C, lower panel). At E9.5 (D to G), the HAI-1+/+ placenta had started to form labyrinth layer (D and E), whereas the HAI-1–/– placenta showed impaired formation of the labyrinth layer (F and G).

 
At E9.5, chorionic trophoblasts in HAI-1+/+ placentas had started to spread inwards, forming the labyrinth layer together with fetal vessels (Fig. 5D and E). In contrast, in HAI-1–/– placentas, chorionic trophoblasts were packed and did not show migration and differentiation to form the labyrinth layer (Fig. 5F and G). Immunohistochemically, HAI-1 protein was expressed in the labyrinth layer of the wild-type, normal placenta at E9.5 (Fig. 6A), whereas the immunoreactivity was modestly decreased in the placenta of heterozygous embryos (Fig. 6B). There was no immunoreactivity in homozygous mutant (HAI-1–/–) embryos (Fig. 6C). At higher magnification, HAI-1 was expressed on the surface of chorionic trophoblasts and also in the differentiated labyrinthine trophoblasts of HAI-1+/+ placentas (Fig. 6D and E). Impaired formation of the labyrinth layer and the absence of differentiated labyrinthine trophoblasts became evident in HAI-1–/– placentas at E10.5 (Fig. 7A, B, and C). In normal placenta at E10.5, HAI-1 protein was localized in chorionic trophoblasts present at the interface between the chorionic plate and labyrinth layer and also in labyrinthine trophoblasts, but not in spongiotrophoblasts (Fig. 7A) and trophoblast giant cells (data not shown).



View larger version (164K):
[in this window]
[in a new window]
  		FIG. 7. Formation of the labyrinth layer in the placentas at E10.5 (A to C) and E11.5 (D to J). Immunohistochemically, HAI-1 was expressed in chorionic trophoblasts and differentiated labyrinthine trophoblasts (A) but not spongiotrophoblasts (A) and trophoblast giant cells (not shown) of normal HAI-1+/+ (+/+) placenta. Ch, chorionic plate; La, labyrinth layer; Sp, spongiotrophoblast layer. Impaired formation of branching, organized labyrinthine structure became obvious in HAI-1–/– (–/–) placenta (C) compared with HAI-1+/+ placenta (B) at E10.5. The labyrinth layer of the HAI-1–/– placenta was abnormally thin, largely avascular, and composed predominantly of clusters of cuboidal trophoblasts lacking the formation of differentiated labyrinthine structure at E10.5 (C). At E11.5, defects in placental development become more evident in the HAI-1–/– placenta. Distinct layered structures consisting of the chorionic plate (Ch), labyrinth layer (La), spongiotrophoblast layer (Sp), trophoblast giant cell layer (Gi), and maternal decidual tissue (De) were observed in the HAI-1+/+ placenta (D), whereas development of the labyrinth layer was selectively and severely impaired in the HAI-1–/– placenta (E). In a higher magnification, fetal blood vessels were present in the labyrinth of the HAI-1+/+ placenta (F). In contrast, fetal blood vessels were rare and an organized labyrinth layer was not formed in the HAI-1–/– placenta (G). Some fetal vessels showed degenerative change (H and I) and thrombus formation (J), reflecting the death of the embryo.

 
At E11.5, the normal, wild-type placenta showed the typical three-layer appearance consisting of the innermost labyrinth layer, middle layer of spongiotrophoblasts, and outermost giant cell layer (Fig. 7D). In contrast, in HAI-1–/– gestations, the labyrinth layer failed to form (Fig. 7E). The fetal vessels (identified by their containing nucleated embryonic blood cells) had failed to elongate and bifurcate into the labyrinth layer and instead had collapsed (Fig. 7F and G). Degenerated fetal vessels and thrombus formation were observed at E11.5 in HAI-1–/– placentas, indicating death of the embryo (Fig. 7H, I, and J). Despite the obvious failure in labyrinth layer formation, the spongiotrophoblast and giant cell layers were formed. The yolk sac appeared normal at E9.5 (not shown).

Analysis of trophoblast marker gene expression in HAI-1–/– placentas. To further analyze the placental defects in HAI-1–/– gestations, we performed in situ hybridization and RT-PCR analysis for expression of several trophoblast markers (Fig. 8). At E9.5, trophoblast giant cells expressing Pl-1 (6) were present in a similar pattern in wild-type (HAI-1+/+) and mutant (HAI-1–/–) placentas (Fig. 8A). Spongiotrophoblasts expressing Tpbpa (20) were also detected in HAI-1–/– placenta (Fig. 8B). With RT-PCR analyses, preservation of expression of Pl-1 and Tpbpa mRNAs was confirmed in HAI-1–/– placenta; in addition, the levels of Mash2 and Hand1 mRNAs were not altered significantly (Fig. 8C). In contrast, mRNA for GCM1, expressed specifically by labyrinthine trophoblasts (2, 33, 36), was significantly decreased in HAI-1–/– placenta at E9.5 and E10.5 (Fig. 8D and E). These results confirmed that the differentiated labyrinthine trophoblasts were specifically and significantly affected by the absence of functional HAI-1.



View larger version (66K):
[in this window]
[in a new window]
  		FIG. 8. Analyses of the expression of trophoblast marker genes in placentas at E9.5 and E10.5. Serial sections of HAI-1+/+ (+/+) and HAI-1–/– (–/–) placentas were probed with the antisense RNA probe for Pl-1 (A) and Tpbpa (B). The expression of the trophoblast giant cell marker (Pl-1) and the spongiotrophoblast marker (Tpbpa) was comparable between HAI-1+/+ and HAI-1–/– placentas at E9.5, indicating the giant cell layer and the spongiotrophoblast layer were formed normally in HAI-1–/– placenta. The trophoblast marker genes were also analyzed by RT-PCR (C and D). The expression levels of Pl-1, Hand1, Tpbpa, and Mash2 were preserved in the HAI-1–/– placenta at E9.5 (C). On the other hand, the level of GCM1 mRNA (a marker of labyrinthine trophoblast) was significantly decreased in the HAI-1–/– placenta at E9.5 and E10.5. The results of two independent experiments are shown for E10.5. The decreased GCM1 mRNA level was further confirmed by quantitative real-time RT-PCR (E), showing significantly decreased GCM1 mRNA in the HAI-1–/– placenta at E10.5. Values are means ± standard deviation of three experiments.

 
Analysis of placental transport function in HAI-1 mutant mice. In order to assess placental function, we analyzed the efficiency of BrdU transport (28). At E9.5, pregnant female mice were injected with BrdU, and the litters were dissected 1.5 h later. Serial sections of each embryo were then immunostained with anti-BrdU antibody as well as with an antibody against nuclear antigen Ki-67, which is present in the nuclei of proliferating cells. As shown in Fig. 9A, HAI-1–/– embryos showed cellular proliferation comparable to that of wild-type embryos, as judged by Ki-67 immunoreactivity. However, BrdU incorporation was significantly decreased in the HAI-1–/– embryos, suggesting defects in placental function that resulted in greatly reduced availability of BrdU to the embryo (Fig. 9A).



View larger version (84K):
[in this window]
[in a new window]
  		FIG. 9. Impaired placental transport function in HAI-1–/– placentas. (A) Immunohistochemical study of Ki-67 labeling in the neuroepithelial and mesenchymal cells at E9.5 indicated that cell proliferation in the HAI-1–/– embryo is indistinguishable from that in HAI-1+/+ (+/+) and HAI-1+/– (+/–) embryos. However, BrdU labeling experiments revealed no labeled cells in the HAI-1–/– embryo, whereas HAI-1+/+ and HAI-1+/– embryos showed strong BrdU labeling. (B) Transplacental passage of rhodamine 123 dye. Two hours after intraperitoneal injection of rhodamine 123, passive passage of the dye from the mother to the embryo was observed in HAI-1+/+ gestations. This passage was inefficient in HAI-1–/– mice. Six hours after the injection, partial elimination of rhodamine 123 by embryo-to-maternal passage, probably mediated by P-glycoprotein of syncytiotrophoblasts (31), was suggested in HAI-1+/+ gestation but not in the HAI-1–/– embryo. (C) Immunohistochemical study for activated, cleaved caspase-3 revealed scattered positive cells (apoptotic cells) in HAI-1–/– embryos but not in wild-type embryos at E9.5 (inset). Cells around the second branchial arch are shown. At E10.5, HAI-1–/– embryos showed significantly enhanced labeling of active caspase-3 in many different tissues, indicating widespread apoptosis. Tissues around the second branchial arch and optic vesicles are shown.

 
We then checked the transplacental passage of the fluorescent dye rhodamine 123. At high concentrations, this dye shows passive mother-to-fetus passage (31). This dye is also a model substrate for P-glycoprotein present in syncytiotrophoblasts of the labyrinth layer, reflecting feto-maternal transplacental clearances (31). As shown in Fig. 9B, 2 h after injection of the dye into the mother, passive passage of maternal rhodamine 123 to the embryo was more evident in HAI-1+/+ placentas than HAI-1–/– placentas. However, 6 h after the injection, partial elimination of rhodamine from embryo circulation was suggested in HAI-1+/+ gestations but not in HAI-1–/– gestations (Fig. 9B).

Finally, we performed immunohistochemistry with an antibody that specifically recognizes the activated cleaved form of caspase-3. At E9.5, scattered cells that were positive for activated caspase-3 were observed in HAI-1–/– embryos but not in normal wild-type embryos (Fig. 9C, inset). At E10.5, HAI-1–/– embryos showed significantly enhanced labeling of active caspase-3 in many different tissues, indicating widespread apoptosis (Fig. 9C). The distribution of apoptotic cells was not related to the distribution of HAI-1-expressing cells observed in Fig. 4A. Taken together, these data suggest that death of HAI-1–/– embryos is caused by severely impaired placental function resulting in massive apoptosis of the embryonic cells.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study, we generated a mutation in the mouse HAI-1 gene by replacing most of its functional domains with the neo gene and then examined the phenotype of the HAI-1–/– genotype in the progeny of heterozygous intercrosses. The results indicate that HAI-1–/– embryos invariably die by midgestation (around E10.5). Histological examination revealed a specific abnormality in the labyrinth layer of the HAI-1–/– placenta. In normal pregnancy, the labyrinth layer starts to develop around E9.0 to 9.5 and subsequently functions as a nutrient transport unit (16). Under normal embryogenesis, allantoic vessels are seen by E10, when they penetrate the chorionic plate and the ectoplacental plate is transformed into the labyrinthine placenta (16). In HAI-1–/– gestations, the labyrinth layer is abnormally thin, embryonic blood vessels were scant in the trophoblast layer, and organized branching morphology of the labyrinth layer was never established. Therefore, the function of HAI-1 is essential during formation of the labyrinth layer, and lack of functional HAI-1 results in impaired placental development and function. Although histological analysis at E10.0 revealed growth and developmental retardation such as a thin-walled heart (Fig. 2D) in HAI-1–/– embryo, these findings appear to be secondary to insufficient placental function, as reported in other knockout mice (1, 4).

Death of embryos around E10.5 is consistent with primary placental failure, which is a common cause of midgestation embryonic lethality in mice (34). One common cause of placental failure is failure of chorioallantoic fusion (34). However, chorioallantoic attachment and fusion occurred in HAI-1–/– mice. Abnormal signaling in early trophoblast development is also a known cause of impaired placental development (34). This is also unlikely for HAI-1–/– mice, because both the spongiotrophoblast and giant cell layers were formed. Another common cause of placental failure is failure of chorioallantoic branching and labyrinthine development. To date, studies using mouse mutants have revealed that several molecules are critically involved in these steps (34). Among them, mice deficient for GCM1 appear to have the phenotype most similar to that of our HAI-1 knockout mice (2, 33, 36).

In GCM1–/– mice, the chorioallantoic interface remained flat up to 3 days after allantoic attachment and branching morphogenesis in the chorioallantoic placenta was severely impaired (2, 36). Moreover, similar to the findings seen with HAI-1–/– mice, the spongiotrophoblast layer and giant cell layer were normal in GCM1–/– mice (2, 36). Indeed, when a series of trophoblast markers were analyzed in HAI-1–/– placenta, the GCM1 mRNA level, a marker for labyrinthine trophoblasts, was significantly decreased, whereas the mRNA levels of other trophoblast markers such as Pl-1 (for trophoblast giant cell), Hand1 (trophoblast giant cell), Tpbpa (spongiotrophoblast), and Mash2 (trophoblast stem cell, ectoplacental cone and spongiotrophoblast) showed no apparent alteration. However, unlike GCM1–/– placentas, in which syncytiotrophoblasts are absent (2, 36), some syncytiotrophoblast-like cells were observed in HAI-1–/– placenta.

In human placenta, GCM1 is expressed by cytotrophoblasts (3), and this localization pattern also resembles that of HAI-1 (11). However, there are some differences between the localization of GCM1 and HAI-1 in human placenta in detail. Human HAI-1 is expressed by villous cytotrophoblasts at the base of villi immediately associated with the basement membrane but not by the extravillous cytotrophoblasts (11). On the other hand, human GCM1 is expressed by the extravillous cytotrophoblasts, and its mRNA is detectable only in a subset of villous cytotrophoblasts immediately at the base of villi (3). Mice lacking the 90-kDa heat shock protein ß (HSP90ß) may also show a phenotype similar to that of HAI-1–/– mice (40). In contrast to HAI-1 and GCM1, HSP90ß is expressed ubiquitously (40).

The question then is how the loss of functional HAI-1 induced this observed pathology of the labyrinth layer. HAI-1 belongs to a unique class of serine proteinase inhibitors that are synthesized as a type 1 transmembrane protein (13). HAI-1 was originally identified as a potent inhibitor of HGFA (37), a serine proteinase that specifically activates pro-HGF/SF (13, 26, 27). HAI-1 is also a cognate inhibitor of matriptase, another activator of pro-HGF/SF (18, 19). Importantly, evidence indicates that HAI-1 is not simply an inhibitor of HGFA and matriptase but also a critical regulator of the optimal activities of these enzymes in the pericellular microenvironment (14, 30). Therefore, the phenotype observed in HAI-1–/– mice might be a result of dysregulated activation of pro-HGF/SF.

The observations that both HGF/SF knockout and c-Met (HGF/SF receptor) knockout mice show abnormalities of the placental labyrinth layer (5, 35, 39) may be compatible with this hypothesis. However, the phenotypic manifestations of HGF/SF and c-Met knockouts are different from the effects of HAI-1 ablation in detail. Compared with the effects of the HAI-1–/– genotype, the effects of HGF/SF or c-Met deficiency become apparent at later periods of embryonic development, and the resulting defects in the labyrinth layer are less severe (5, 35, 39). The fact that the lethal point of HGF/SF–/– embryos (E12.5 to E14.5) is later than that of HAI-1–/– embryos (around E10.5) indicates that the placental abnormalities in HAI-1–/– are not simply caused by impaired regulation of HGF/SF activity. Moreover, the phenotypes of HGFA knockout (10) and matriptase knockout (23) mice are quite different from the phenotype seen with HAI-1 knockout mice, as neither HGFA nor matriptase knockout mice show apparent placental abnormalities. Taken altogether, it is reasonable to speculate that, in addition to the regulatory roles of HGFA and matriptase activities that are important in pro-HGF/SF activation, HAI-1 might have another unknown but important role in vivo.

To date, many genes have been reported to be critically involved in the early stage of placental development. Most are transcription factors, growth factors/growth factor receptors, and molecules involved in signal transduction (7, 33, 34). HAI-1 is the first example of a proteinase inhibitor that is critically involved in this stage. Because HAI-1 is a membrane-associated Kunitz-type proteinase inhibitor, the membrane form of HAI-1 may mediate important outside-in or inside-out signaling in labyrinthine trophoblasts acting on the cell surface. This issue will be the focus of future experiments and will help us to understand the role of this unique class of serine proteinase inhibitors.

Alternatively, HAI-1 may have an unknown target proteinase that is critically involved in the regulation of labyrinthine differentiation. As a protein homologous to HAI-1, HAI-2 has been reported as another inhibitor of HGFA (17). HAI-2 is also a membrane-associated Kunitz-type inhibitor. However, evidence suggests that the physiological role of HAI-2 is different from that of HAI-1 (13, 37). Importantly, mice lacking HAI-2 function show embryonic lethality around E8.5 accompanying a primary gastrulation defect (25). Clearly, HAI-1 and HAI-2 have separate, nonredundant functions in vivo, and these membrane-associated inhibitors might have yet undefined though very important biological functions in vivo. Although the placental labyrinth layer was the only tissue that showed obvious developmental defects in these HAI-1–/– gestations, a later role for HAI-1 in development of the embryo proper cannot be excluded; expression of HAI-1 increased in the normal embryo over the course of the gestation period, and expression was abundantly observed in the epithelial cells of the mature embryo. In this regard, further experiments using tetraploid rescue techniques or a conditional knockout strategy would be required.

In conclusion, HAI-1 is required at postimplantation stages for generation of the organized branching structure with differentiated trophoblasts that makes up the labyrinth layer of the placenta. This result is compatible with our previous observation that HAI-1 is specifically expressed in the villous cytotrophoblasts of human placenta (11, 32). Therefore, it will be very interesting to study whether this expression is affected in pathological conditions. Furthermore, understanding the molecular mechanism underlying the placental failure in HAI-1–/– mice in more detail may provide insight into the pathophysiology of intrauterine growth retardation and its treatment. Finally, because the effects of HAI-1 knockout on the development of the placenta are more severe than those observed in HGF/SF and c-Met knockouts, HAI-1 may have another important role in vivo that is independent of its regulatory function in the activation of pro-HGF/SF. Whether HAI-1 has a role at later times in the development of the embryo proper or in the adult will be clarified only when tetraploid rescue experiments or conditional knockout experiments have been performed.


   ACKNOWLEDGMENTS
 
We thank K. Araki and K. Yamamura, Department of Developmental Genetics, Institute of Molecular Embryology and Genetics, Kumamoto University, and K. Yoshinaga and T. Koshimoto, Faculty of Medicine, University of Miyazaki, for helpful suggestions and T. Miyamoto, Y. Nomura, and Y. Shiratani for skillful technical assistance.

This work was supported by Grants-in-Aid for Scientific Research (B) 14370079 and (C) 15590351 and the 21st Century COE program (Life Science) from the Ministry of Education, Science, Sports and Culture, Japan, and by a Grant-in-Aid for Cancer Research from the Ministry of Health, Labor and Welfare (15-13).


   FOOTNOTES
 
* Corresponding author. Mailing address: Second Department of Pathology, Faculty of Medicine, University of Miyazaki, 5200 Kihara, Kiyotake, Miyazaki 889-1692, Japan. Phone: 81 985 85 2809. Fax: 81 985 85 6003. E-mail: mejina{at}med.miyazaki-u.ac.jp. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Adams, R. H., A. Porras, G. Alonso, M. Jones, K. Vintersten, S. Panelli, A. Valladares, L. Perez, R. Klein, and A. R. Nebreda. 2000. Essential Role of p38{alpha} MAP kinase in placental but not embryonic cardiovascular development. Mol. Cell 6:109-116.[CrossRef][Medline]

2. Anson-Cartwright., L., K. Dawson, D. Holmyard, S. J. Fisher, R. A. Lazzarini, and J. C. Cross. 2000. The glial cells missing-1 protein is essential for branching morphogenesis in the chorioallantoic placenta. Nat. Genet. 25:311-314.[CrossRef][Medline]

3. Baczyk, D., A. Satkunaratnam, B. Nait-Oumensmar, B. Huppertz, J. C. Cross, and J. C. P. Kingdom. 2004. Complex patterns of GCM1 mRNA and protein in villous and extravillous trophoblast cells of the human placenta. Placenta 25:553-559.[CrossRef][Medline]

4. Barak, Y., M. C. Nelson, E. S. Ong, Y. Z. Jones, P. Ruiz-Lozano, K. R. Chien, A. Koder, and R. M. Evans. 1999. PPAR{gamma} is required for placental, cardiac, and adipose tissue development. Mol. Cell 4:585-595.[CrossRef][Medline]

5. Bladt, F., D. Reithmacher, S. Isenmann, A. Aguzzi, and C. Birchmeier. 1995. Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 376:768-771.[CrossRef][Medline]

6. Colosi, P., F. Talamantes, and D. I. Linzer. 1987. Molecular cloning and expression of mouse placental lactogen I complementary deoxyribonucleic acid. Mol. Endocrinol. 1:767-776.[Abstract]

7. Ihle, J. N. 2000. The challenges of translating knockout phenotype into gene function. Cell 102:131-134.[CrossRef][Medline]

8. Itoh, H., H. Kataoka, J. Y. Meng, R. Hamasuna, N. Kitamura, and M. Koono. 2001. Mouse hepatocyte growth factor activator inhibitor type 1 (HAI-1) and type 2 (HAI-2)/placental bikunin genes and their promoters. Biochem. Biophys. Acta 1519:92-95.[Medline]

9. Itoh, H., H. Kataoka, M. Tomita, R. Hamasuna, Y. Nawa, N. Kitamura, and M. Koono. 2000. Upregulation of HGF activator inhibitor type 1 but not type 2 along with regeneration of intestinal mucosa. Am. J. Physiol. Gastrointest. Liver Physiol. 278:G635-G643.[Abstract/Free Full Text]

10. Itoh, H., S. Naganuma, N. Takeda, S. Miyata, S. Uchinokura, T. Fukushima, S. Uchiyama, H. Tanaka, K. Nagaike, T. Shimomura, K. Miyazawa, G. Yamada, N. Kitamura, M. Koono, and H. Kataoka. 2004. Regeneration of injured intestinal mucosa is impaired in hepatocyte growth factor activator-deficient mice. Gastroenterology 127:1423-1435.[CrossRef][Medline]

11. Kataoka, H., J. Y. Meng, H. Itoh, R. Hamasuna, T. Shimomura, T. Suganuma, and M. Koono. 2000. Localization of hepatocyte growth factor activator inhibitor type 1 in Langhans' cells of human placenta. Histochem. Cell Biol. 114:469-475.[Medline]

12. Kataoka, H., R. Hamasuna, H. Itoh, N. Kitamura, and M. Koono. 2000. Activation of hepatocyte growth factor/scatter factor in colorectal carcinoma. Cancer Res. 60:6148-6159.[Abstract/Free Full Text]

13. Kataoka, H., S. Miyata, S. Uchinokura, and H. Itoh. 2003. Roles of hepatocyte growth factor (HGF) activator and HGF activator inhibitor in the pericellular activation of HGF/scatter factor. Cancer Metastasis Rev. 22:223-236.[CrossRef][Medline]

14. Kataoka, H., T. Shimomura, T. Kawaguchi, R. Hamasuna, H. Itoh, N. Kitamura, K. Miyazawa, and M. Koono. 2000. Hepatocyte growth factor activator inhibitor type 1 is a specific cell surface binding protein of hepatocyte growth factor activator(HGFA) and regulates HGFA activity in the pericellular microenvironment. J. Biol. Chem. 275:40453-40462.[Abstract/Free Full Text]

15. Kataoka, H., T. Suganuma, T. Shimomura, H. Itoh, N. Kitamura, K. Nabeshima, and M. Koono. 1999. Distribution of hepatocyte growth factor activator inhibitor type 1(HAI-1) in human tissues: cellular surface localization of HAI-1 in simple columnar epithelium and its modulated expression in injured and regenerative tissues. J. Histochem. Cytochem. 47:673-682.[Abstract/Free Full Text]

16. Kaufmann, M. H. 1992. Differentiation of the placenta, p. 472-477. In M. H. Kaufmann (ed.), The atlas of mouse development, 6th ed. Academic Press, Inc., New York, N.Y.

17. Kawaguchi, T., L. Qin, T. Shimomura, J. Kondo, K. Matsumoto, K. Denda, and N. Kitamura. 1997. Purification and cloning of hepatocyte growth factor activator inhibitor type 2, a Kunitz-type serine protease inhibitor. J. Biol. Chem. 272:27558-27564.[Abstract/Free Full Text]

18. Kirchhofer, D., M. Peek, W. Li, J. Stamos, C. Eigenbrot, S. Kadkhodayan, J. M. Elliot, R. T. Corpuz, R. A. Lazarus, and P. Moran. 2003. Tissue expression, protease specificity, and Kunitz domain functions of hepatocyte growth factor activator inhibitor-1B (HAI-1B), a new splice variant of HAI-1. J. Biol. Chem. 278:36341-36349.[Abstract/Free Full Text]

19. Lee, S. L., R. B. Dickson, and C. Y. Lin. 2000. Activation of hepatocyte growth factor and urokinase/plasminogen activator by matriptase, an epithelial membrane serine protease. J. Biol. Chem. 275:36720-36725.[Abstract/Free Full Text]

20. Lescin, K. R., S. Varmuza, and J. Rossant. 1988. Isolation and characterization of a novel trophoblast-specific cDNA in the mouse. Genes Dev. 2:1639-1646.[Abstract/Free Full Text]

21. Lin, C. Y., J. Anders, M. Johnson, and R. B. Dickson. 1999. Purification and characterization of a complex containing matriptase and a Kunitz-type serine protease inhibitor from human milk. J. Biol. Chem. 274:18237-18242.[Abstract/Free Full Text]

22. Lin, C. Y., J. Anders, M. Johnson, Q. A. Sang, and R. B. Dickson. 1999. Molecular cloning of cDNA for matriptase, a matrix-degrading serine protease with trypsin-like activity. J. Biol. Chem. 274:18231-18236.[Abstract/Free Full Text]

23. List, K., C. C. Haudenschild, R. Szabo, W. J. Chen, S. M. Wahl, W. Swaim, L. H. Engelholm, N. Behrend, and T. H. Bugge. 2002. Matriptase/MT-SP1 is required for postnatal survival, epidermal barrier function, hair follicle development, and thymic homeostasis. Oncogene 21:3765-3779.[CrossRef][Medline]

24. List, K., R. Szabo, P. W. Wertz, J. Segre, C. C. Haudenschild, S. Y. Kim, and T. H. Bugge. 2003. Loss of proteolytically processed filaggrin caused by epidermal deletion of Matriptase/MT-SP1. J. Cell Biol. 163:901-910.[Abstract/Free Full Text]

25. Mitchell, K. J., K. I. Pinson, O. G. Kelly, J. Brennan, J. Zupicich, P. Scherz, P. A. Leighton, L. V. Goodrich, X. Lu, B. J. Avery, P. Tate, K. Dill, E. Pangilian, P. Wakenight, M. Tessier-Lavigne, and W. C. Skarnes. 2001. Functional analysis of secreted and transmembrane proteins critical to mouse development. Nat. Genet. 28:241-249.[CrossRef][Medline]

26. Miyazawa, K., T. Shimomura, A. Kitamura, J. Kondo, Y. Morimoto, and N. Kitamura. 1993. Molecular cloning and sequence analysis of the cDNA for a human serine protease responsible for activation of hepatocyte growth factor. J. Biol. Chem. 268:10024-10028.[Abstract/Free Full Text]

27. Miyazawa, K., T. Shimomura, and N. Kitamura. 1996. Activation of hepatocyte growth factor in the injured tissues is mediated by hepatocyte growth factor activator. J. Biol. Chem. 271:3615-3618.[Abstract/Free Full Text]

28. Moser, M., Y. Li, K. Vaupel, D. Kretzschmar, R. Kluge, P. Glynn, and R. Buettner. 2004. Placental failure and impaired vasculogenesis result in embryonic lethality for neuropathy target esterase-deficient mice. Mol. Cell. Biol. 24:1667-1679.[Abstract/Free Full Text]

29. Nagaike, K., K. Kohama, S. Uchiyama, H. Tanaka, K. Chijiiwa, H. Itoh, and H. Kataoka. 2004. Paradoxically enhanced immunoreactivity of hepatocyte growth factor activator inhibitor type 1(HAI-1) in cancer cells at the invasion front. Cancer Sci. 95:728-735.[Medline]

30. Oberst, M. D., C. A. Williams, R. B. Dickson, M. D. Johnson, and C. Y. Lin. 2003. The activation of matriptase requires its noncatalytic domains, serine protease domain, and its cognate inhibitor. J. Biol. Chem. 278:26773-26779.[Abstract/Free Full Text]

31. Pavek, P., F. Staud, Z. Fendrich, H. Sklenarova, A. Libra, M. Novotna, M. Kopecky, M. Nobilis, and V. Semecky. 2003. Examination of the functional activity of P-glycoprotein in the rat placental barrier using rhodamine 123. J. Pharmacol. Exp. Ther. 305:1239-1250.[Abstract/Free Full Text]

32. Pötgens, A. J. G., H. Kataoka, S. Ferstl, H. G. Frank, and P. Kaufmann. 2003. A positive immunoselection method to isolate villous cytotrophoblast cells from first trimester and term placenta to high purity. Placenta 24:412-423.[CrossRef][Medline]

33. Roberts, R. M., T. Ezashi, and P. Das. 2004. Trophoblast gene expression: Transcription factors in the specification of early trophoblast. Reprod. Biol. Endocrinol. 2:1-9.[CrossRef][Medline]

34. Rossant, J., and J. C. Cross. 2001. Placental development: lessons from mouse mutants. Nat. Rev. Genet. 2:538-548.[Medline]

35. Schmidt, C., F. Bladt, S. Goedecke, V. Brinkmann, W. Zschiesche, M. Sharpe, E. Gherardi, and C. Birchmeler. 1995. Scatter factor/hepatocyte growth factor is essential for liver development. Nature 373:699-702.[CrossRef][Medline]

36. Schreiber, J., E. Riethmacher-Sonnenberg, D. Riethmacher, E. E. Tuerk, J. Enderich, M. R. Bösl, and M. Wegner. 2000. Placental failure in mice lacking the mammalian homolog of glial cell missing, GCMa. Mol. Cell. Biol. 20:2466-2474.

37. Shimomura, T., K. Denda, A. Kitamura, T. Kawaguchi, M. Kito, J. Kondo, S. Kagaya, L. Qin, H. Takata, K. Miyazawa, and N. Kitamura. 1997. Hepatocyte growth factor activator inhibitor, a novel Kunitz-type serine protease inhibitor. J. Biol. Chem. 272:6370-6376.[Abstract/Free Full Text]

38. Takeuchi, T., J. L. Harris, W. Huang, K. W. Yan, S. R. Coughlin, and C. S. Craik. 2000. Cellular localization of membrane-type serine protease 1 and identification of protease-activated receptor-2 and single-chain urokinase-type plasminogen activator as substrates. J. Biol. Chem. 275:26333-26342.[Abstract/Free Full Text]

39. Uehara, Y., O. Minowa, C. Mori, K. Shiota, J. Kuno, T. Noda, and N. Kitamura. 1995. Placental defect and embryonic lethality in mice lacking hepatocyte growth factor/scatter factor. Nature 373:702-705.[CrossRef][Medline]

40. Voss, A. K., T. Thomas, and P. Gruss. 2000. Mice lacking HSP90ßfail to develop a placental labyrinth. Development 127:1-11.[Abstract]

41. Yagi, T., T. Tokunaga, Y. Furuta, S. Nada, M. Yoshida, T. Tsukuda, Y. Saga, N. Takeda, Y. Ikawa, and S. Aizawa. 1993. A novel ES cell line, TT2, with high germline-differentiating potency. Anal. Biochem. 214:70-76.[CrossRef][Medline]

42. Zhang, Y. W., and G. F. Vande Wounde. 2003. HGF/SF-Met signaling in the control of branching morphogenesis and invasion. J. Cell Biochem. 88:408-417.[CrossRef][Medline]

Molecular and Cellular Biology, July 2005, p. 5687-5698, Vol. 25, No. 13
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.13.5687-5698.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted