Rescue of the embryonic lethal hematopoietic defect reveals a critical role for GATA-2 in urogenital development
1998; Springer Nature; Volume: 17; Issue: 22 Linguagem: Inglês
10.1093/emboj/17.22.6689
ISSN1460-2075
Autores Tópico(s)Prenatal Screening and Diagnostics
ResumoArticle16 November 1998free access Rescue of the embryonic lethal hematopoietic defect reveals a critical role for GATA-2 in urogenital development Yinghui Zhou Yinghui Zhou Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Kim-Chew Lim Kim-Chew Lim Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Ko Onodera Ko Onodera Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Satoru Takahashi Satoru Takahashi Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author Jun Ohta Jun Ohta Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author Naoko Minegishi Naoko Minegishi Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author Fong-Ying Tsai Fong-Ying Tsai Millennium Inc., Cambridge, MA, USA HHMI and Department of Pediatrics, Harvard Medical School, Boston, MA, USA Search for more papers by this author Stuart H. Orkin Stuart H. Orkin HHMI and Department of Pediatrics, Harvard Medical School, Boston, MA, USA Search for more papers by this author Masayuki Yamamoto Masayuki Yamamoto Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author James Douglas Engel Corresponding Author James Douglas Engel Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Yinghui Zhou Yinghui Zhou Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Kim-Chew Lim Kim-Chew Lim Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Ko Onodera Ko Onodera Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Satoru Takahashi Satoru Takahashi Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author Jun Ohta Jun Ohta Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author Naoko Minegishi Naoko Minegishi Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author Fong-Ying Tsai Fong-Ying Tsai Millennium Inc., Cambridge, MA, USA HHMI and Department of Pediatrics, Harvard Medical School, Boston, MA, USA Search for more papers by this author Stuart H. Orkin Stuart H. Orkin HHMI and Department of Pediatrics, Harvard Medical School, Boston, MA, USA Search for more papers by this author Masayuki Yamamoto Masayuki Yamamoto Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan Search for more papers by this author James Douglas Engel Corresponding Author James Douglas Engel Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA Search for more papers by this author Author Information Yinghui Zhou1, Kim-Chew Lim1, Ko Onodera1, Satoru Takahashi2, Jun Ohta2, Naoko Minegishi2, Fong-Ying Tsai3,4, Stuart H. Orkin4, Masayuki Yamamoto2 and James Douglas Engel 1 1Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, 60208-3500 USA 2Department of Molecular and Developmental Biology, Tsukuba University School of Medicine and Center for TARA, Tsukuba, Japan 3Millennium Inc., Cambridge, MA, USA 4HHMI and Department of Pediatrics, Harvard Medical School, Boston, MA, USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:6689-6700https://doi.org/10.1093/emboj/17.22.6689 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Mutations resulting in embryonic or early postnatal lethality could mask the activities of any gene in unrelated and temporally distinct developmental pathways. Targeted inactivation of the transcription factor GATA-2 gene leads to mid-gestational death as a consequence of hematopoietic failure. We show here that a 250 kbp GATA-2 yeast artificial chromosome (YAC) is expressed strongly in both the primitive and definitive hematopoietic compartments, while two smaller YACs are not. This largest YAC also rescues hematopoiesis in vitro and in vivo, thereby localizing the hematopoietic regulatory cis element(s) to between 100 and 150 kbp 5′ to the GATA-2 structural gene. Introducing the YAC transgene into the GATA-2−/− genetic background allows the embryos to complete gestation; however, newborn rescued pups quickly succumb to lethal hydroureternephrosis, and display a complex array of genitourinary abnormalities. These findings reveal that GATA-2 plays equally vital roles in urogenital and hematopoietic development. Introduction The GATA family of zinc finger proteins plays pivotal roles in cellular proliferation and differentiation in numerous lineages. The founding member of the family, GATA-1 (Evans and Felsenfeld, 1989; Tsai et al., 1989), is required in vivo for both erythropoiesis and megakaryopoiesis (Fujiwara et al., 1996; McDevitt et al., 1997; Takahashi et al., 1997). GATA-2 and GATA-3 (Yamamoto et al., 1990) have been shown to be crucial determinants of various developmental processes, including early hematopoiesis, neurogenesis and T lymphopoiesis (Kornhauser et al., 1994; Tsai et al., 1994; Pandolfi et al., 1995; Ting et al., 1996; Zheng and Flavell, 1997). GATA-4, -5 and -6, which comprise a structurally distinct subfamily, may play important roles in cardiac gene regulation (Kelley et al., 1993; Heikinheimo et al., 1994; Laverriere et al., 1994; Grepin et al., 1995; Gove et al., 1997; Molkentin et al., 1997), and GATA-4 additionally appears to be involved centrally in ventral specification (Molkentin et al., 1997). Among these factors, GATA-2 has been implicated in cellular proliferative responses in numerous, often unrelated, developmental pathways (Dorfman et al., 1992; Briegel et al., 1993; Nagai et al., 1994; Walmsley et al., 1994; Brewer et al., 1995; Ma et al., 1997). Verification of the paramount importance of GATA-2 to embryogenesis was demonstrated by targeted ablation of the gene. Mutant embryos lacking GATA-2 fail to generate a sufficient number of primitive erythrocytes and die at mid-gestation, showing that GATA-2 plays a crucial role in the earliest phases of blood formation (Tsai et al., 1994). Since its inception more than a decade ago, gene targeting in the mouse has provided biologists with an exceptionally versatile tool with which to manipulate the mammalian genome (Smithies et al., 1985; Thomas and Capecchi, 1987). Gene corrections, conversions, substitutions, temporal, spatial and conditional alterations and (most often) the generation of null mutations can now be accomplished in a rather straightforward manner, given the variety of specialized procedures that have been developed around the basic methodology. Despite its obvious success, the most frequently encountered secondary hurdle is that homozygous mutant embryos fail to survive gestation. While this establishes the functional consequences of gene loss as the first developmental block is encountered, it also conceals possible functions of the gene subsequent to this most proximal event. This complication, in turn, has hastened the development of practical surrogate tools to study gene function (e.g. by analysis of chimeras and hypomorphic alleles, and tissue-specific gene rescue or ablation). While powerful, these surrogate methods do not reveal mechanisms for transcriptional regulation within the locus, since they essentially focus only on the structural gene. To understand the details of the regulatory hierarchy that leads to tissue-specific gene expression, and to prove immediate epistatic relationships in genetic pathways, regulatory modules controlling the expression of a gene must be identified, dissected in molecular detail and tested for functionality in vivo. Here we show that after modification by insertion of a lacZ reporter gene at the translational initiation site, a 250 kbp GATA-2-containing yeast artificial chromosome (YAC) recapitulates GATA-2 expression during murine embryogenesis. Two smaller YACs, ∼200 or 120 kbp in size, respectively, fail to confer significant fetal liver (definitive hematopoietic) expression. We then show that the 250 kbp YAC can rescue the embryonic lethal defect in hematopoiesis that originally was attributed to targeted ablation of the GATA-2 gene. GATA-2 is expressed from the YAC at essentially wild-type levels, and normal erythroid and myeloid lineage cells are generated in vitro and in vivo. Thus the hematopoietic regulatory controls for GATA-2 lie within this YAC. Complementation of the hematopoietic embryonic lethal defect due to gene targeting revealed a new deficiency in a tissue that had not been implicated as one requiring GATA-2 function. The kidneys and ureters of GATA-2−/−::YAC compound mutant newborns are severely deformed, and neonates quickly succumb to fatal hydroureternephrosis. This abnormality was shown to result from a block in urine excretion due to an underlying mid-embryonic failure of the ureters to connect properly to the bladder. Other sex and excretory organs are also malformed in the YAC transgenic compound mutant mice, demonstrating that GATA-2 plays a broad role in urogenital development. Since these urogenital phenotypes were revealed only after hematopoietic rescue (thus allowing the animals to survive gestation), these experiments suggest that analogous strategies may be valuable in revealing vital activities for other genes whose later developmental functions presently are obscured because of embryonic lethality. Results Isolation and characterization of GATA-2 YACs Four YAC clones containing the mouse GATA-2 gene were identified originally from a YAC library (Research Genetics) by PCR. Extensive analysis of these YAC clones indicated that while all four contained the intact 14 kbp GATA-2 structural gene, only one, YAC 22-5E7, was intact (Figure 1A; data not shown). YAC 22-5E7 is 620 kbp in size, and contains 520 and 80 kbp of 5′ and 3′ sequence flanking the GATA-2 structural gene, respectively, and is oriented with the 5′ end of the gene closest to the YAC right (non-centromeric) vector arm (Figure 1A). To facilitate its manipulation and microinjection, a series of YAC deletion mutants, containing varying amounts of 5′ sequence, were generated using standard YAC deletion procedures (Heard et al., 1994; Emanuel et al., 1995; Figure 1B). YAC subclone d16, which is 250 kbp, contains large segments of 5′- and 3′-flanking sequence (150 and 80 kbp, respectively), while its overall size is still quite manageable for microinjection. Two smaller YACs, d27 and d18, which differ from d16 only in their 5′ ends, are ∼200 or 120 kbp in size, respectively (Figure 1B). Figure 1.Structures of the GATA-2 YACs and PFGE mapping of the d16Z YAC transgenic lines. (A) The orientation and composition of the four isolated murine GATA-2 YACs are depicted schematically. The top line represents the NotI restriction map of C57Bl/6 (B6) chromosome 6 in the vicinity of the GATA-2 locus (cross-hatched), while the next four lines depict the deduced structures of the four isolated YAC recombinants, with their sizes indicated in parentheses. All four YACs contain an intact mouse GATA-2 gene (black box). Clones 24-3F10 and 19-6B7 are chimeric, consisting of non-contiguous genomic DNA fragments (indicated by the open bar; data not shown). The YAC left and right vector arms are indicated by an open oval and an open rectangle, respectively. The size of each YAC recombinant is shown in parentheses. (B) B1 repeat recombinants of YAC 22-5E7 are illustrated diagramatically. Yeast cells bearing YAC 22-5E7 were electroporated with a linearized B1 repeat recombinant plasmid (Heard et al., 1994), and subclones bearing a new (LYS2) right vector arm (rectangle) contain different amounts of DNA deleted from the 5′ end of the YAC. (C) Integrity of d16Z transgenic lines: PFGE mapping of four GATA-2 YAC d16Z transgenic lines: d16Z5 (lanes 1), d16Z15 (lanes 2), d16Z31(lanes 3) and d16Z43 (lanes 4; see Materials and methods). Probes used for each panel are indicated at the bottom of each panel. Transgenic lines Z5, Z15 and Z31 all contain at least one intact YAC transgene copy, while line Z43 does not. Three additional d16Z transgenic lines examined (Z1, Z8 and Z28) were also found to be fragmented using the same assays (data not shown). Download figure Download PowerPoint GATA-2 transcription in the hematopoietic system is recapitulated by a GATA-2–lacZ YAC reporter transgene To define the transcriptional potential of these GATA-2 YAC subclones, each was first tagged by inserting the bacterial lacZ gene into the GATA-2 translation initiation site (Materials and methods; Lakshmanan et al., 1998; Minegishi et al., 1998). The lacZ-modified YACs (referred to below as d16Z, d27Z or d18Z) were isolated from pulsed-field gels, purified and then injected into fertilized CD1 ova to generate transgenic mice (Schedl et al., 1993; Bungert et al., 1995). Transgenic founders were identified initially by PCR using primer pairs specific for the YAC vector arms, as well as for the lacZ gene. Ten d16Z founders that were positive for all three PCR markers were identified, and seven of these transmitted the transgene through the germline. The integrity of the d16Z YACs in these seven lines was analyzed in F2 or subsequent generation animals using PCR, pulse-field gel electrophoresis (PFGE) and YAC telomere analysis by Southern blotting of thymus DNA (Liu et al., 1997). While several of the lines contained broken YACs, three contained at least one intact d16Z transgene copy (Figure 1C). In the hematopoietic system, we noted that staining in the fetal liver of d16Z embryos became apparent at 10.5 days post-coitus (d.p.c.) and remained intense until ∼13.5 d.p.c., but faded gradually thereafter (Figure 2A; data not shown). Detailed examination of thin sections from 12.5 d.p.c. embryos showed that a similar number and kind of fetal liver cells were stained either by β-galactosidase in the d16Z YAC transgenic animals (Figure 2D) or by a GATA-2-specific monoclonal antibody in wild-type embryos (Figure 2E). At earlier stages (9.0–9.5 d.p.c.), robust lacZ staining was detected in two parallel cables of mesenchymal cells flanking the dorsal aorta of d16Z embryos (Figure 2F). This staining was strikingly reminiscent of the aorta, gonads and mesonephros (AGM) region that defines the earliest definitive hematopoietic stem cell compartment in the embryo (Muller et al., 1994). Patchy staining was also detected in the visceral yolk sac at the same stage (Figure 2G), which upon detailed examination revealed exclusive labeling of primitive erythroid lineage cells (Figure 2H). The d16Z YAC is also expressed in the adult bone marrow as well as in a small fraction of hematopoietic cells in normal adult spleens (Figure 2I and J, respectively). Figure 2.YAC transgenes recapitulate the in vivo developmental expression of GATA-2. Embryos (11.5 d.p.c.) from transgenic lines d16Z5 (A), d27Z27 (B) and d18Z1 (C) were stained for β-galactosidase activity. All transgenic embryos displayed strong, identical staining patterns in the CNS (midbrain, hindbrain and spinal cord), the developing heart and the placenta (not shown). Only d16Z transgenic embryos displayed strong additional staining in the fetal liver (fl), which was conspicuously absent in the d27Z and d18Z transgenic lines. (D) Section of a 12.5 d.p.c. d16Z15 transgenic fetal liver stained for β-galactosidase activity. (E) Section of a wild-type 12.5 d.p.c. liver stained with a rat anti-GATA-2 monoclonal antibody RC1.1 (Materials and methods). (F) A 9.5 d.p.c. d16Z15 transgenic whole-mount embryo stained for lacZ activity. Note that the stained cells appear in a symmetrical pattern flanking both sides of the dorsal aorta. (G) Yolk sac of a 9.5 d.p.c. d16Z5 transgenic embryo. Distinct cells in the yolk sac were labeled, clearly identifiable as hematopoietic cells upon sectioning (H), which were not seen in d27Z or d18Z transgenic lines. (I) Bone marrow section from an adult d16Z15 mouse. (J) Spleen section of an adult d16Z15 transgenic mouse, displaying a strong punctate distribution of lacZ-positive cells in the red pulp. Download figure Download PowerPoint The two smaller GATA-2 YAC lacZ subclones (d27Z and d18Z; Figure 1B) were also examined in a similar fashion. Embryos from multiple transgenic lines bearing these two YACs exhibited staining patterns identical to the d16Z transgenic embryos at all developmental stages examined, except in one important respect. lacZ expression in hematopoietic tissues was virtually absent in the d27Z and d18Z transgenic embryos (Figure 2A–C; data not shown). In summary, a putative positive regulatory element(s) conferring high level expression to GATA-2 in the primitive and definitive hematopoietic compartments resides somewhere between the 5′ boundaries of the d27Z and d16Z YACs (which differ from one another by ∼50 kbp; Figure 1B). Thus the hematopoietic regulatory element(s) is located between 100 and 150 kbp 5′ to the GATA-2 gene (Minegishi et al., 1998). Since GATA-2−/− embryos die of hematopoietic failure, YAC d16 appeared to represent the minimal locus that might be able to rescue this GATA-2 deficiency. GATA-2 transcription outside the hematopoietic system We next determined whether the lacZ expression patterns in the intact transgenic lines coincided with normal GATA-2 expression in non-hematopoietic cells. Detailed histological examination of the intact d16Z, d27Z and d18Z transgenic lines confirmed the coincidence of β-galactosidase staining and multiple sites of previously established GATA-2 expression. Numerous tissues in the d16Z transgenic embryos stained quite prominently: first, strong expression in the ectoplacental cone and parietal yolk sac was found as early as 8.5 d.p.c., and persisted throughout gestation, coincident with normal GATA-2 expression in the placenta (Ng et al., 1994; Ma et al., 1997). A transient phase of expression is also detected in the endocardium of the developing heart between 9.5 and 12.5 d.p.c. (Figure 2A–C; data not shown), as well as in endothelial cells, another site of prominent GATA-2 expression (Dorfman et al., 1992; data not shown). Finally, β-galactosidase activity conferred by the d16Z GATA-2 YAC is quite robust in the embryonic central nervous system (CNS), including the midbrain, hindbrain and spinal cord, between 10.5 and 14.5 d.p.c. (Figure 2A–C). Further studies showed that the CNS lacZ expression pattern precisely overlaps endogenous GATA-2 expression there (Kornhauser et al., 1994; Y.Zhou, K.-C.Lim, K.Onodera and S.Takahashi, unpublished observations). We conclude that the elements conferring GATA-2 expression in the placenta, CNS and developing heart are all contained within the 250 kbp d16Z (Figure 2A), the 200 kbp d27Z (Figure 2B) and the 120 kbp d18Z transgenic YACs (Figure 2C). GATA-2 YAC transgenes rescue the hematopoietic defect in GATA-2−/− embryos GATA-2-deficient embryos die at ∼10.5 d.p.c. of embryogenesis, and current evidence suggests that this lethality is due to a failure in primitive erythroid proliferation (Tsai et al., 1994). Since the d16Z transgene conferred robust expression of the GATA-2-directed reporter in embryonic tissues representing the sites and times of most abundant fetal blood formation, while two smaller YACs did not, we next asked whether or not the 250 kbp d16 YAC could complement the hematopoietic defect encountered in GATA-2 homozygous mutant embryos. As the d16 YAC is structurally indistinguishable from the endogenous GATA-2 locus, except at the telomeric ends, we introduced a silent mutation into the YAC (by deleting a BamHI restriction site present within intron 3, thereby generating subclone d16B) so that all three alleles (wild-type, germline gene-targeted and YAC) of GATA-2 could be distinguished from one another. Purified YAC d16B DNA was injected into fertilized ova according to standard protocols (Materials and methods). Ten founders were recovered, of which only four transmitted the d16B transgene. The integrity of the integrated YACs was characterized by PFGE Southern blotting, which showed that the established lines d16B15, d16B63 and d16B89 all bear at least one intact YAC transgene copy (Figure 3A). Each of these animals was then intercrossed to GATA-2+/− mice to recover germline heterozygous mutant F2 that carried the YAC transgene. These GATA-2+/−::d16B YAC mice were intercrossed again with GATA-2+/− animals, and the resultant litters were recovered at different embryonic times and genotyped to determine whether or not GATA-2−/−::YAC d16B animals (abbreviated below as 'compound mutant' embryos or pups) were viable. Figure 3.YAC integrity and expression of GATA-2 in compound mutant transgenic mice. (A) Integrity of the GATA-2 d16B YAC in four transgenic lines. Lines d16B15 (lanes 1), d16B63 (lanes 2) and d16B89 (lanes 3) bear intact transgene copies (the band detected in line d16B15 by the GATA-2 5′ probe is >2 Mbp, not shown), while line d16B55 (lanes 4) is fragmented. (B) Southern blot of BamHI-digested tail DNA isolated from individual pups in a single litter from an F2 (GATA-2+/−::d16B15 YAC)×GATA-2+/− intercross. The expected BamHI fragment sizes of the YAC (YAC; 5.6 kbp), GATA-2 gene-targeted mutant (M; 4.6 kbp) and GATA-2 wild-type (WT; 3.8 kbp) alleles are indicated on the left. Pups 10.1 and 10.4 are both of genotype GATA-2−/−::d16B YAC. (C) RT–PCR analysis of GATA-2 expression in 10.5 d.p.c. d16B15 YAC compound mutant embryos, using murine S16 mRNA levels as internal control. The genotypes and expression level (expressed as a percentage of wild-type; lane 5) were: lane 1, (−/−) 0%; lane 2, (−/−::YAC) 58%; lane 3, (−/−::YAC) 71%; lane 4, (+/−) 60%; lane 5, (+/+) 100%; lane 6, (+/−::YAC) 130%; lane 7, (+/+::YAC) 140%. Download figure Download PowerPoint Since GATA-2−/− mutant animals die of hematopoietic failure at 10.5 d.p.c., we first asked if the hematopoietic defect was rescued by the transgene. As shown in Figure 4A and D, 10.5 d.p.c. GATA-2−/− embryos could be identified unequivocally by their pale yolk sac and vasculature containing very little blood. All of the compound mutant embryos, however, had a normal yolk sac which was well vascularized (Figure 4B and E) and indistinguishable from that of wild-type littermates (Figure 4C and F). Semi-quantitative RT–PCR analysis of total RNA recovered from compound mutant embryos at this stage showed that GATA-2 is expressed at essentially wild-type levels (Figure 3C). Hence, the 250 kbp YAC d16B, containing 150 and 80 kbp of sequence flanking the structural gene, is capable of restoring GATA-2 functions at this stage of embryonic development and reversing the effects of the null mutation. Figure 4.Hematopoiesis is normal in compound mutant pups. At 10.5 d.p.c., the yolk sac of GATA-2 null embryos (A) is paler than that of the compound mutant (B) or wild-type (C) embryos. The GATA-2 homozygous mutant embryos reflect the same deficiency (D), which is not found in either the compound mutant (E) or wild-type (F) embryos. (G and H) Micrographs of liver tissue sections of GATA-2+/−::YAC d16B or GATA-2−/−::YAC d16B P0 pups, respectively, taken at the same magnification. Arrows indicate typical megakaryocytes surrounded by other nucleated and enucleated (erythroid) hematopoietic cells. (I and J) Tissue sections of wild-type or GATA-2−/−::d16B compound mutant transgenic P0 pups, respectively, showing an approximately equal distribution of mast cells (arrows) near the surface of the skin. Download figure Download PowerPoint Next, we asked if the compound mutant mice survived embryogenesis. When examined at P0, compound mutant pups were found amongst litters generated using all three intact d16 YAC transgenic lines for rescue (e.g. Figure 3B), and at the expected Mendelian frequency. These data show (as expected from the restoration of embryonic hematopoiesis) that the embryonic lethality observed in GATA-2−/− embryos was indeed fully rescued by the d16B transgene integrated at any chromosomal position (Table I). Table 1. The 250 kbp d16B YAC rescues the GATA-2−/− embryonic lethal phenotype YAC transgenic line Without YAC With YAC +/+ +/− −/− +/+::YAC +/−::YAC −/−::YAC d16B15 (12 litters) 22 41 0 9 25 16 d16B63 (9 litters) 12 24 0 10 23 10 d16B89 (2 litters) 3 10 0 0 5 2 Total 37 (0.33) 75 (0.67) 0 19 (0.19) 53 (0.53) 28 (0.28) When the embryos and pups were examined in greater detail for hematopoietic phenotypes, we found that both the number and types of hematopoietic colonies produced in vitro from 13.5 d.p.c. fetal livers of compound mutant embryos were within the range of wild-type and heterozygous GATA-2 mutant controls (Figure 5). Peripheral blood smears taken from compound mutant pups were indistinguishable from those from wild-type littermates. Normal numbers of erythrocytes, megakaryocytes and leukocytes were detected in the bone marrow (data not shown) and liver (Figure 4H) of the newborn compound mutant pups. Finally, a normal number and distribution of mast cells was found in the skin of these animals (Figure 4J) in comparison with their wild-type littermate controls (Figure 4G and I, respectively). These data show that the d16B YAC transgene contains sequences sufficient for the rescue of myeloerythroid cell lineages in GATA-2−/− animals, and further that this rescue leads to the survival of the GATA-2-null mutant mice through full-term gestation. These data therefore support the contention that the original embryonic lethal mutant defect was indeed hematopoietic failure (Tsai et al., 1994). Figure 5.In vitro fetal liver hematopoietic progenitors in compound mutant mice. Two standard intercross matings were conducted between males (genotype GATA-2+/−::d16B YAC) bearing either the d16B15 or d16B63 YAC transgenes and GATA-2+/− females. Fetal livers were recovered from embryos at 13.5 d.p.c., dispersed and plated in methocel cultures containing IL-3, IL-6, SCF and Epo (see Materials and methods). Colonies were scored 7 and 10 days after seeding. The total number of fetal livers recovered, genotyped and assayed were: (+/+) = 4; (+/−) = 8; (−/−) = 0; (+/+)::YAC = 6; (+/−)::YAC = 3; (−/−)::YAC = 2. Download figure Download PowerPoint Rescued compound mutant pups develop a fatal urogenital disorder As described above, compound mutant pups were born in the expected Mendelian ratio (Table I) while no homozygous GATA-2 mutant animals survived gestation if they did not additionally carry the transgene. However, none of the compound mutant pups survived to weaning. Further examination revealed that compound mutant neonates rarely flourished beyond the first few days after birth, and they succumbed quickly to wasting. Necropsies of the compound mutant pups revealed a single consistent abnormality: GATA-2−/−::YAC newborns rescued with any of the three d16B transgenes developed hydroureternephrosis, a condition characterized by fluid-filled dilation of the kidneys and ureters. At mild to moderate manifestation, the kidneys were enlarged with a single, fluid-filled cavity, while at its most severe extreme, the kidneys were riddled with multiple large ancillary cysts causing the morphology to be completely disrupted (Figure 6A and B). Both ureters were extremely bloated, resembling a clinical condition in humans called megaureter (Figure 6A and H). Figure 6.GATA-2−/−::d16B YAC compound mutant mice develop hydroureternephrosis. (A) The kidneys and ureters from compound mutant (cm) P0 neonates (right) are enlarged dramatically in comparison with those of wild-type (wt) littermates (left). (B) An extremely dilated kidney from another compound mutant pup. Kidney sections of wild-type (C and D) or GATA-2−/−::YAC compound mutant (E and F) transgenic embryos at either 14.5 (C and E) or 17.5 d.p.c. (D and F). By 17.5 d.p.c., compound mutant kidneys have developed prominent central cysts (F). Injection of trypan blue into the renal pelvis of either wild-type (G) or compound mutant (H) neonates. Note that the colored dye does not enter the bladder (b) of the compound mutant animal, but rather accumulates in the kidneys (ki) and ureters (u). Download figure Download PowerPoint Multiple embryonic stages were examined in the hope of determining the etiology of this defect. At 14.5 d.p.c., the metanephros of the compound mutant embryos appeared normal, with regular numbers of S- and comma
Referência(s)