| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Graduate School of Comprehensive Human Sciences,
2 Center for Tsukuba Advanced Research Alliance, and 3 Environmental Response Project, Exploratory Research for Advanced Technology, Japan Science and Technology Corporation, University of Tsukuba, Tsukuba 305-8577, Japan
4 Department of Biochemistry, Tokyo Women's Medical University School of Medicine, Shinjuku-ku, Tokyo 162-8666, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionExperimental proceduresReferences |
|---|
NF mutant) was over-expressed in mice. These mice exhibited abnormal morphology in peripheral red blood cells (RBCs), reticulocytosis, splenomegaly, and erythroid hyperplasia, indicating compensated hemolysis. These mice were extremely sensitive to phenylhydrazine (PHZ), an agent that induces hemolysis, and their RBCs were osmotically fragile. Importantly, the hemolytic response to PHZ was partially restored by the simultaneous expression of wild-type GATA-1 with the NF mutant, supporting our contention that NF protein competitively inhibits the function of endogenous GATA-1. These data provide the first in vivo evidence that the NF domain contributes to the gene regulation that is critical for differentiation and survival of mature RBCs in postnatal erythropoiesis. |
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Circulatory red blood cells (RBCs) respond rapidly to changes in the environment, such as tissue hypoxia, oxidative stress, or dehydration. In order to withstand the shear stress of the circulation and to pass safely through capillary vessels, the unique structure of the erythroid membrane is composed of many erythroid-specific and ubiquitous proteins. Genetically mutated RBC membrane proteins reduce the resistance of RBC to share stress and shorten their life span, thereby leading to hereditary hemolytic diseases (reviewed in Delaunay 2002). Similarly, disrupted glucose and lipid metabolism and hemoglobin synthesis in RBCs causes human hemolytic disorders. The structural, biochemical, and physiological features of mature RBCs have been studied extensively. However, it is unknown how the expression of genes coding for proteins or enzymes required for the function of mature RBCs is regulated in response to environmental stimuli in vivo. It has been clearly shown that GATA-1 acts as a key regulator of erythroid-specific gene expression, but there is no direct evidence that GATA-1 contributes to the gene regulation required for the function of mature RBCs in vivo.
To elucidate the requirements for each functional domain of GATA-1 in embryonic hematopoiesis, we previously performed complementation rescue analyses of GATA-1 knockdown mice (Shimizu et al. 2001), exploiting transgenic mouse lines expressing various GATA-1 domain mutants under the regulation of the GATA-1 gene hematopoietic regulatory domain (GATA-1 HRD, Onodera et al. 1997; Nishimura et al. 2000; Ohneda et al. 2002). We refer to the GATA-1 knockdown allele as GATA-1.05, because the allele expresses GATA-1 mRNA at approximately 5% of that of the wild-type allele. The transgenic expression of wild-type GATA-1 under the control of GATA-1 HRD rescued the hemizygote knockdown (GATA-1.05/Y) mice from embryonic lethality (Takahashi et al. 2000; Shimizu et al. 2001), whereas the transgenic expression of the GATA-1 NF mutant did not. The GATA-1.05/Y::NF compound mutant embryos showed a maturation defect in definitive erythroid cells, revealing that the NF domain is essential for the function of GATA-1 in definitive erythropoiesis.
Here, we demonstrate that on certain occasions the genetically engineered modulation of GATA-1 leads to the fragility of RBCs and hemolytic syndrome and that the compensatory response accelerates erythropoiesis in adult hematopoietic tissues. This phenotype was found in the transgenic line of mice over-expressing GATA-1 NF mutant under the control of the GATA-1 HRD. Importantly, this phenotype was restored by the concomitant expression of wild-type GATA-1, demonstrating that transgene-derived GATA-1 NF protein competitively inhibits the function of endogenous GATA-1. These results suggest that, in addition to its critical contribution to definitive erythropoiesis in the embryonic liver, the NF domain plays important roles in vivo in the regulation of genes essential for the mature RBC functions. Our results thus suggest that during erythropoiesis, GATA-1 contributes to gene regulation in vivo in response to various environmental stresses.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
NF transgenic mice
We previously reported that transcriptional activation of the mouse 
-1globin promoter, a GATA-1 dependent reporter, was comparable in wild-type and NF GATA-1 transfected cells using non-hematopoietic QT6 cells (Shimizu et al. 2001). We performed similar transient transfection assay using various GATA-1-dependent reporter constructs and the mouse -1globin reporter with mutated GATA sites (Fig. 1A). Disruption of the GATA sites in the -1globin promoter completely abolished transcriptional activation by both wild-type and NF GATA-1, and no significant difference was observed in the reporter activity between wild-type and NF GATA-1 transfected cells in the other reporter constructs. These results suggest that function of NF domain should be assessed by alternative systems, such as transgenic complementation rescue study (Takahashi et al. 2000; Shimizu et al. 2001, 2004).
|
NT, NF, and CF, respectively) under the control of GATA-1 HRD (Shimizu et al. 2001). While the expression profiles of these transgenes almost fully and consistently recapitulated that of GATA-1, the expression levels of these transgenes varied from line to line due to integration position effect variegation. In particular, unlike the cases of the other GATA-1 mutants, we were unable to obtain lines of high NF producers. Therefore, to further examine the effect of NF mutant protein on the rescue of GATA-1.05/Y mutant mice, we attempted to generate additional lines of GATA-1 HRD-GATA-1 NF transgenic mice (NF TG; Fig. 1B). Of the 10 newly generated founder mice, we found two high NF expressors and established these mice as lines (Lines 1 and 2). The abundance of transgene-derived mRNA in the spleens of these mice was evaluated by RT-PCR (Fig. 1C). The primers were set at 5' and 3' to the NF region in order to discriminate between the endogenous (shown by triangle) and transgene (asterisk) derived GATA-1 mRNAs by size. The PCR product derived from the transgene was more abundant than that from the endogenous GATA-1 gene in transgenic mouse Lines 1 and 2, whereas the transgene expression level was comparable to the endogenous GATA-1 mRNA level in the transgenic line used in our previous study (Line M, Shimizu et al. 2001). Electrophoretic mobility shift assay (EMSA) using whole cell extracts prepared from the spleen of adult transgenic mice revealed that the transgene-derived NF protein (asterisk) binds to a GATA site in both transgenic lines (Fig. 1D). In both high producer lines, transgenic mice were born at the expected frequency and were fertile.
Next we tested the ability of NF TG Line 1 mice to rescue GATA-1.05/Y mutant mice from lethality, as the expression level of the transgene was most abundant in this line. Among 81 pups examined, no GATA-1.05/Y::NF compound mutants were identified. This indicates that, even at a high level of expression, NF GATA-1 cannot compensate for the lack of an NF functional domain in eliminating lethal anemia in GATA-1.05/Y embryos. This result supports our previous conclusion that the NF domain is vital for definitive erythropoiesis in the embryonic liver.
We noticed an intriguing feature during our analysis in that approximately 30% of the newborn pups with Line 1 NF TG was easily distinguishable from wild-type pups by their transient jaundice (Fig. 1E). These transgenic newborns recovered from jaundice within 3–4 days after birth, suggesting that their jaundice was caused by stress during the perinatal period. Even after the weaning, some NF TG mice were still distinguishable from their wild-type littermates by their smaller size (Fig. 1F). Indeed, the average body weight of this transgenic line was significantly lower than that of wild-type throughout the postnatal period (Fig. 1G). These observations led us to hypothesize that the high level of expression of NF protein affects the function of endogenous GATA-1 in adult erythropoiesis.
Irregular RBC morphology and accelerated erythropoiesis in NF TG mice
The hematopoietic indices of 9–13 week old NF TG mice showed no significant signs of anemia (Table 1) and their hematological parameters were almost similar to those observed in GATA-1HRD-GATA-1 high producer lines (GATA-1 TG, Takahashi et al. 2000) and wild-type mice. One exception was that there was a significant increase in the reticulocyte count in NF TG mice (Table 1). It is noteworthy that the increase in reticulocyte count was observed in both lines of transgenic mice (Line 1 and Line 2) and with a severity that correlated well with the NF transgene expression levels. Wright-Giemsa staining of peripheral blood smears from wild-type mice (Fig. 2A) and NF TG Line 1 mice (Fig. 2B) revealed the presence of anisocytosis and occasional spherocytes in NF TG blood smears. New methylene blue staining showed an increase in reticulocytes in the peripheral blood of NF TG mice (Fig. 2D) compared to wild-type peripheral blood (Fig. 2C). We also carried out scanning electron microscopy and found that RBCs of the NF transgenic mice are irregular in shape (Fig. 2F,G). We observed these irregularly shaped RBCs in the peripheral blood of Line 2 transgenic mice, albeit at a lower frequency (data not shown). Taken together, these results suggest that there is an increased fragility in the peripheral RBCs of NF TG mice. This fragility leads the NF TG mouse RBCs to hemolysis and is most likely the cause of the jaundice observed in the neonatal periods. The increase in reticulocytes may reflect compensatory erythropoiesis.
|
|
NF TG mice
We analysed hematopoietic cells in the bone marrow and spleen of NF TG mice by fluorescence-activated cell sorting (FACS). We stained cells obtained from the bone marrow (Fig. 3A) and spleen (Fig. 3B) with transferrin receptor (CD71) and Ter119 antibodies, since combined staining with CD71 and Ter119 can characterize the stages of erythroid lineage cells (Socolovsky et al. 2001). CD71 is expressed in proerythroblasts and early basophilic erythroblasts and its levels decrease with erythroid cell maturation, whereas Ter119 is highly expressed in terminally differentiated erythroblasts. In the bone marrow of NF TG mice, the population of CD71highTer119high cells (R2 region in Fig. 3), which correspond to early basophilic erythroblasts, was greater than in wild-type mice (wild-type, 27.1%; NF TG, 42.6%). A prominent increase in CD71highTer119high cells in the NF TG mice was also observed in the spleen (1.3% and 10.5%, respectively). In contrast, there was no significant difference in the CD71highTer119high cell population between GATA-1 TG and wild-type mice. These results indicate that the deleterious effect of NF TG on erythroid cell differentiation is specific for the transgene.
|
NF TG and wild-type mice (Table 2). Compared to wild-type mice, the erythroid colony formation in the bone marrow and spleen of NF TG mice was significantly increased. The augmented colony number was detected in the early erythroid colonies (BFU-E) more strikingly than in the late erythroid colonies (CFU-E), and this was clearly observed in the bone marrow. We surmise that this increase might be a physiological response to chronic hemolysis, although there remains a possibility that the high expression level of GATA-1 NF protein affects the differentiation process of erythroid progenitors. Thus, NF TG mice show an increase in erythroid progenitors and frequency of early stage erythroblasts in the bone marrow and spleen, suggesting a rapid turnover of erythroid cells and compensatory increase in erythropoiesis in the mutant mice.
|
NF TG mice are highly sensitive to oxidative stress-induced hemolysis
Hematological examinations suggest that the accelerated erythropoiesis in NF TG mice might overcome overt anemia under unstressed physiological conditions. To test whether these mice have the capacity to respond to further stimulation of erythropoiesis, phenylhydrazine (PHZ), that provokes acute hemolysis by inducing oxidative stress, was administered to NF TG, GATA-1 TG, and wild-type mice (Fig. 4). To this end, we initially adopted the standard experimental protocol for mice (Vannucchi et al. 2001), in which a standard dose of PHZ (60 mg/kg body weight) was administrated to mice for 2 consecutive days. To our surprise, two out of four NF TG mice died immediately after the second administration of PHZ, whereas none of the other genotypes died due to this PHZ treatment (Fig. 4A). The two surviving NF TG mice appeared inactive and pale, and the hematocrit of these mice dropped to as low as 15%. While the hematocrit values of the GATA-1 TG and wild-type mice also reached bottom by Day 2, they were still much higher than that of the NF TG mice (Fig. 4A). These mice recovered from anemia 8 days after the first PHZ treatment and these results were in very good agreement with our previous experiments utilizing standard PHZ administration. We repeated this experiment and examined the effects of acute hemolysis on kidney and liver. Neither hemorrhage nor necrotic change was observed in hematoxylin and eosin-stained sections of NF TG mice (data not shown). Furthermore, plasma indirect bilirubin concentration was significantly higher in the NF TG mice compared to wild-type mice treated with PHZ (0.93 ± 0.09 and 0.70 ± 0.10 mg/dL, respectively, P = 0.01, n = 3), while the values were similar in the NF TG mice and wild-type treated with saline (0.30 ± 0.06 and 0.27 ± 0.07 mg/dL, respectively, n = 3). These results indicate that some of the NF TG mice treated with PHZ died due to hemolytic anemia rather than functional failure of other organs.
|
NF TG mice survived beyond Day 3 (n = 4, data not shown). The NF TG mice could recover from hemolytic anemia when a quarter of the original dose of PHZ (15 mg/kg body weight) was administered (Fig. 4C,D). However, the hematocrit values of the NF TG mice on Day 2 and 4 were significantly lower than that of the wild-type mice (P < 0.05). Our initial reservation regarding this experiment was that the NF TG mice might not be able to respond to PHZ treatment fully, since erythropoiesis is accelerated even under unstressed conditions. Unexpectedly however, reticulocyte counts were sharply enhanced following PHZ administration, despite the initial reticulocyte count being significantly higher in NF TG mice than in wild-type mice (Fig. 4B,D). These results thus indicate that NF TG mice still retain the capacity to enhance erythropoiesis in response to hemolytic stress. However, in NF TG mice, hemolysis proceeded at an extraordinary speed owing to the high sensitivity of RBCs to PHZ, with the erythropoiesis activity failing to compensate for the hemolysis caused by PHZ.
The mean spleen weight of NF TG mice was 0.32 ± 0.04 g (n = 5) in the unstressed condition, which was significantly greater than the mean weight of wild-type mouse spleens (0.10 ± 0.01 g, n = 6; P = 0.01, Fig. 5A). The mean weight of wild-type mouse spleens (0.17 ± 0.01 g, n = 6) increased 2 days after the first PHZ administration. The mean weight of NF TG mouse spleen increased after PHZ treatment (0.46 ± 0.05 g on Day 2, n = 4), confirming the ability of these mice to respond to the stress placed on erythropoiesis.
|
NF TG and wild-type mice in the unstressed condition and on Day 2 of PHZ treatment (from the two surviving mice treated with 60 mg/kg body weight, see above) are shown in Fig. 5. The architecture of wild-type mouse spleen consists of white and red pulp areas (Fig. 5C) that are lost in the spleens of NF TG mice, even in the unstressed condition (Fig. 5D). While the architecture of wild-type spleen following PHZ treatment remained intact and the red pulp area was expanded (Fig. 5E), the spleen of NF TG mice was heavily affected and filled with erythroblasts (Fig. 5F). Similarly, the cellular density of the bone marrow from NF TG mice was increased (Fig. 5H) compared to that from wild-type mice (Fig. 5G).
Morphological examination of bone marrow cells further revealed that the ratio of myeloid to erythroid cells in NF TG mice was much smaller than in wild-type mice (0.93 and 1.52, respectively). Consistent with our FACS analysis (see Fig. 3), basophilic erythroblasts were significantly increased in the bone marrow of NF TG mice (Fig. 5J) compared to wild-type mice (Fig. 5I). The bone marrow of NF TG mice treated with PHZ showed further development of erythroid hyperplasia, with an increase in proerythroblasts and basophilic erythroblasts (Fig. 5L), while the bone marrow of wild-type mice treated with PHZ (Fig. 5K) showed erythroid hyperplasia similar to that in untreated NF TG mice (Fig. 5J). These results demonstrate that erythropoiesis is stimulated in the hematopoietic tissues of NF TG mice and that these mutant mice still have the capacity to increase erythropoiesis in response to stress signals of erythropoiesis.
Functional analysis of RBCs and gene expression in NF TG mice
To characterize functional properties of RBCs in NF TG mice in vitro, we worked on two experiments. First, we performed an osmotic fragility analysis and found that the osmotic fragility curve for NF TG mouse RBCs shifted to the right (Fig. 6A). In contrast, the RBCs of GATA-1 TG mice showed a similar curve to that of wild-type mice, indicating that NF TG mouse RBCs are sensitive to osmotic stresses. Secondly, we carried out a filtration test to examine whole cell deformability (Fig. 6B). Compared to that of GATA-1 TG mice (Fig. 6B, curve b), the RBCs of NF TG mice showed impairment in the velocity of flow through a 4-µm pore (Fig. 6B, curve c). The velocity of NF TG mouse RBCs at –35 mm H2O was less than half of that of GATA-1 TG mouse RBCs (34.4% and 76.0%, respectively, when the flow rate of buffer alone was set to 100%). In this analysis, a disturbance in flow rate generally indicates three pathological situations for erythrocytes: (1) a loss of surface membrane to volume, such as in spherocytes; (2) an enhanced viscosity, such as with dehydration; (3) impaired mechanical properties in erythroid cell membranes. In the case of NF TG mouse RBCs, the mean corpuscular volume and the mean corpuscular hemoglobin concentration was comparable to those of wild-type mice (Table 1). These observations rule out the first and the second possibility. Therefore, the impaired mechanical properties in cell membranes are the most plausible in NF TG mouse RBCs.
|
NF TG mice that causes hemolysis, we examined the protein expression of RBC ghosts by SDS-polyacrylamide gel electrophoresis (Fig. 6C). Against our expectations, however, we could not find any clear differences between NF TG and wild-type mice in the expressions of known membrane proteins, including spectrins, band 3, protein 4.2, protein 4.1, and actin. In addition, we found by immunoblot analysis that the expressions of ß-spectrin, ankyrin, and protein 4.1 in the RBC ghosts of NF TG mice were similar to those observed in the RBC ghosts of wild-type mice (Fig. 6D). The expression of adducin was detectable as a doublet band originating from two distinct isoforms, -adducin (MW 103 000) and ß-adducin (MW 97 000), the latter being the major isoform in RBC. We found that the expression of adducin was increased in the RBC ghosts of NF TG mice, but this seems unlikely to be the major cause of hemolysis (see Discussion).
We then examined the expression of GATA-2 and EKLF, two transcription factors regulated by GATA-1 (Grass et al. 2003; Xue et al. 2004), by semiquantitative RT-PCR analysis using spleen mRNAs prepared from NF TG and wild-type mice (Fig. 6E). The expression of GATA-2 is slightly higher in the NF TG mouse spleen, presumably due to the increased population of immature erythroid cells in which GATA-2 is highly expressed. The expression of EKLF is much higher in the NF TG mice, indicating that EKLF is regulated by GATA-1 activity that is independent of the NF domain. EKLF plays critical roles in the regulation of globin gene expression (Nuez et al. 1995; Perkins et al. 1995). Since defective hemoglobin production occasionally causes hemolytic anemia, we investigated the expression of ß-major globin in NF TG mice. However, the increase of EKLF did not affect the expression of ß-major globin. Moreover, immunohistochemical analysis revealed similar distribution of ß-major globin protein on the spleen sections prepared from NF TG and wild-type mice (Fig. 6F).
Hemolysis is partially restored by the expression of wild-type GATA-1 in NF TG mice
So far, our data suggest that the expression of GATA-1 NF protein affects the function of endogenous GATA-1 during erythroid cell differentiation. Of particular note, the GATA-1 TG mice did not show the phenotype observed with NF TG mice, but showed similar characteristics with wild-type mice in all experiments performed in this study (for instance, see Table 1 and Figs 3, 4 and 6A). This fact suggests that a defective NF domain may directly contribute to dysregulated gene expression. We therefore propose our competitive inhibition model (Fig. 8A) in which NF protein competitively inhibits DNA binding of endogenous GATA-1 protein, thereby modifying the gene expression profile. In this model, the expression of target genes requires certain cofactors that interact with GATA-1 through the NF domain, with NF protein affecting proper gene regulation. We also considered other possible models for explaining the molecular basis underlying the NF TG mouse phenotype (see Discussion).
|
NF TG mice with GATA-1 TG mice to generate compound mutant mice (NF::GATA-1). The scenario behind this experiment being that, if the transgenic expression of wild-type GATA-1 renders the phenotype of the NF TG mouse less severe, it may be due to the competition between the NF protein and wild-type GATA-1; NF protein binds to DNA less frequently in the presence of transgene-derived wild-type GATA-1 (Fig. 8B). To this end, we administered PHZ (15 mg/kg body weight) to the NF::GATA-1 compound, NF TG, and wild-type mice. We found that the hematocrit values of NF::GATA-1 mice were significantly higher than those of NF TG mice both 2 and 4 days after the first PHZ treatment, although they were still lower than those of wild-type mice (Fig. 7A). Similarly, the reticulocyte counts of the compound mutant mice were lower than those of the NF TG mice both before and after PHZ administration (Fig. 7B). Thus, the hemolytic response was partially restored in the compound mutant mice. These results thus support our contention that the NF TG mouse phenotype is caused by the lack of NF domain function.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
NF protein was expressed at a high level (Line 1), even in the presence of endogenous GATA-1. Line 2 with lower transgene expression than in Line 1 showed moderate reticulocytosis and milder hemolytic syndrome than in Line 1 (data not shown), indicating that the severity of the phenotype correlates well with the expression level of the transgenic NF protein. Importantly, hemolysis in Line 1 NF TG mice was restored by the transgenic expression of wild-type GATA-1. These results indicate that the NF domain is essential for GATA-1 to direct and maintain the proper functioning of erythroid cells in juvenile and adult hematopoietic tissues.
The NF domain of GATA-1 has been shown to stabilize the binding of CF to palindromic GATA sites, although NF itself is not capable of binding to the consensus GATA sequence (Trainor et al. 1996). The other important role of the NF domain is to recruit coregulatory proteins, such as FOG-1 (Tsang et al. 1997). We recently found that the expression of GATA-1V205G, a mutant protein in which a single amino acid substitution in the NF domain reduces the interaction of GATA-1 with FOG-1, still rescues the GATA-1 gene knockdown mice from embryonic lethality (V205GR mice; Shimizu et al. 2004). V205GR mice showed an apparent delay in recovery from PHZ-induced anemia (Shimizu et al. 2004). However, unlike the NF TG mice, the V205GR mice did not show such high sensitivity to or die from PHZ treatment. In addition, the osmotic fragility of RBCs from V205GR mice appeared to be within the normal range (data not shown). These results suggest that the hemolytic phenotype of the NF TG mice may be independent of FOG-1. Thus, an emerging model is that a certain NF-dependent cofactor, perhaps other than FOG-1, may play critical roles in the regulation of erythroid genes involved in the function of mature RBCs (Fig. 8A,B). A defective interaction of NF protein with such a cofactor may affect the activity of endogenous GATA-1. In this regard, it is worth noting that the NF domain is involved in the self-association of GATA-1 (Mackay et al. 1998). The self-association of GATA-1 has been shown to enhance transcriptional activity in zebra fish embryos (Nishikawa et al. 2003). Therefore, GATA-1 itself is a possible candidate for the NF-dependent cofactor required for the function of GATA-1 in mature RBCs.
To further clarify the molecular basis underlying the hemolytic phenotype of NF TG mice, we postulated three models in addition to the competitive inhibition model described above. The first model, based on the function of NF that modulates the specific DNA binding of CF, was considered in which the NF protein binds to a GATA site that is normally inactive and activates a silent gene (Fig. 8C). In this case, a double zinc finger structure or the NF domain itself is critical for determining the state of activity (active or inactive) of GATA sites. In the second model, NF protein acts as a ‘decoy’ to cofactors that can interact with both wild-type and NF GATA-1 (Fig. 8D). An abundance in NF protein leads to a deficiency in these cofactors. In the third model, NF protein activates gene expression to the same extent as endogenous GATA-1, leading to an extraordinary increase in target gene expression and severity of the hemolytic phenotype (Fig. 8E). In the latter two models, the total abundance of GATA-1 protein (i.e., the sum of endogenous and NF GATA-1 proteins), rather than the presence of NF protein, is a critical determinant of modified gene expression. There may be a variety of alternatives to the four models we proposed here. An important observation is that the hemolytic phenotype of the NF mice was enfeebled by the concomitant expression of wild-type GATA-1. The results strongly support the competitive inhibition model, because the phenotype ought not to be affected by the over-expression of wild-type GATA-1 in the first additional model, but ought to be more significant in the second and third additional models.
In our previous study on the transgenic complementation rescue of GATA-1 knockdown mice, we obtained high producer lines of CF, NT, and NTNF transgenes (Shimizu et al. 2001). Despite the fact that the NF domain was also deleted in NTNF TG mice, neither neonatal jaundice nor growth retardation was observed in this particular genotype. Moreover, the hematological indices of NTNF TG mice were not different to those of wild-type mice. The difference between the NF and the NTNF TG mice may indicate that the NT domain, the first assigned transactivation domain of GATA-1 based on reporter co-transfection assays (Martin & Orkin 1990; Yang & Evans 1992), contributes to the hemolytic phenotype of the NF TG mice. The results also suggest a different contribution of the NT and NF domains to postnatal erythropoiesis. Indeed, the rescue analysis showed that these two domains could not complement each other in the ability to support erythroid cell differentiation during embryonic definitive erythropoiesis (Shimizu et al. 2001).
Increasing lines of recent evidence unveiled an important role of the NT domain in vivo. Although the GATA-1 HRDNT transgene eliminated lethal anemia in GATA-1.05/Y embryos, the frequency of the pups rescued by this transgene was much less than rescued by the wild-type GATA-1 transgene (Shimizu et al. 2001). A very high incidence of mutations occurs in the second exon region of the GATA-1 gene in patients with Down's syndrome-associated acute megaloblastic leukaemia and transient myeloproliferative disorder (Wechsler et al. 2002; Xu et al. 2003; Gurbuxani et al. 2004). These mutations result in either a stop codon within the NT domain or disrupted splicing, producing NT mutant protein utilizing methionine at codon 84 as an alternative translation start site. The contribution of NT protein has been ascribed to the pathogenesis of acute megaloblastic leukaemia and transient myeloproliferative disorder in Down's syndrome patients.
We did not notice any significant difference between NF TG and control mice in RBC membrane protein expression, except for the increased -adducin expression in NF TG mice. We did not consider the increase in -adducin to be the major cause of hemolysis, because ß-adducin, an erythroid-specific isoform, has been reported to determine the accumulation of -adducin in RBC membranes through the formation of dimers or tetramers (Gilligan et al. 1999; Muro et al. 2000) and the expression of ß-adducin in RBC ghosts was similar in NF TG and wild-type mice. We also examined the expression of mRNAs encoding erythrocytic enzymes (glucose-6-phosphate dehydrogenase, glucose-6-phosphate isomerase and pyruvate kinase) by microarray analysis (n = 2, data not shown). In this analysis, mRNA from NF TG mouse spleen was compared to that from PHZ-treated wild-type mouse spleen, since histological analysis showed a similar degree of erythroid hyperplasia in these two samples. The expression of erythroid enzymes was comparable between NF TG and wild-type mice. Although abnormal hemoglobin production causes hemolytic anemia in human, our preliminary microarray analyses revealed no significant difference in -1globin expression level between NF TG and wild-type mice (data not shown). In addition, ß-major globin expression level in spleen was similar in NF TG and wild-type mice by semiquantitative RT-PCR.
Functional disorders and/or abnormal morphology in peripheral RBCs have been reported in mice with targeted disruption of the transcription factors p45 NF-E2 and small Maf proteins. RBCs from p45-deficient mice were sensitive to oxidative stress and had a reduced life span (Chan et al. 2001). RBCs from mafG::mafK compound null-mutant mice showed abnormal erythroid morphology and increased osmotic fragility (Onodera et al. 2000). Of note, GATA-1 has been shown to play an essential role in the gene regulation of NF-E2 p45 and MafK (Tsang et al. 1997; Katsuoka et al. 2000). Unexpectedly, however, neither microarray nor RT-PCR analyses using spleen and bone marrow mRNA revealed a reduced expression of these genes (data not shown). Therefore, despite our efforts, the identity of the genes responsible for the RBC fragility in NF TG mice is unclear at this point. Since hemolytic syndrome is caused by the impairment of a variety of genes, a comprehensive examination of gene expression profiles needs to be carried out.
In summary, the NF TG mice provided convincing evidence for the contribution of the NF domain of GATA-1 to the gene regulation required for resistance to stress of circulatory RBC in vivo. This mouse model may also be a potential tool for providing further insight into the functions of GATA-1 in homeostatic regulation during postnatal erythropoiesis.
|
Experimental procedures |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
The GATA-1NF (GATA-1197–232 amino acids) DNA fragment was generated by PCR as previously described (Shimizu et al. 2001). The GATA-1NF cDNA was then subcloned into a plasmid containing 8.5 kb of murine GATA-1 genomic locus spanning 3.9 kb 5' of the IE exon to the second exon (GATA-1 HRD). Transgenic mice were generated by microinjection of the construct into fertilized BDF1 eggs by standard procedures. Founders were screened by PCR and crossed with BDF1 mice. The body weights of the NF TG and wild-type mice were monitored at 1, 3, and 14 weeks of age.
Transfection analysis and gel mobility shift assay
The firefly luciferase reporter plasmids were constructed by inserting gene fragments containing GATA motifs originated from GATA-1 target genes into pRBGP3 plasmid (Igarashi et al. 1994). Transfection assay and electrophoretic gel mobility shift assay (EMSA) was performed as previously described (Shimizu et al. 2001).
Semi-quantitative RT-PCR
Total RNA was extracted from the spleens of 5–15 month old mice using the ISOGEN RNA extraction system (Nippon-Gene). The cDNAs were synthesized by reverse transcriptase (Clontech) and 24, 27, and 30 cycles of PCR amplifications were performed. The forward primer 5'-AAGTTTCATGGAAGCCTG-3' and the reverse primer 5'-CAGCTGGTCCTTCAGCCGCTCCCG-3' were used to amplify GATA-1. The endogenous and transgenic GATA-1 PCR amplicons were 430 and 320 base pairs, respectively. The GAPDH amplicon was obtained using the forward primer 5'-GAAGGTGAAGGTCGGAGTC-3' with the reverse primer 5'-GAAGATGGTGATGGGATTTC-3' and used to standardize the amount of cDNA loaded. Primer sequences used for GATA-2, EKLF and ß-major globin cDNA amplifications were as previously described (Suwabe et al. 1998).
Hematological and biochemical analysis
Mice were bled from the retro-orbital plexus into EDTA-containing tubes and the hematopoietic indices were determined using a hemocytometer (Nihon Kohden Co.). The data obtained from 14 to 29 adult mice were averaged. Peripheral blood and bone marrow smears were stained by Wright-Giemsa staining. Reticulocyte counts were examined manually using blood smears stained with new methylene blue. Total and direct bilirubin concentrations in plasma were determined by DRI-CHEM system (FUJIFILM) using TBIL-PIII and DBIL-PII slides, respectively, in accordance with the manufacturer's instructions. Difference between total and direct bilirubin concentration was regarded as indirect bilirubin concentration.
Scanning electron microscopy and immunohistochemistry
For scanning electron microscopy, RBC samples were fixed with 1.25% (v/v) glutaraldehyde and washed in 0.1 M phosphate buffer (pH 7.4). Samples were then fixed with 1% (v/v) osmium acid, followed by washing in 0.1 M phosphate buffer. Samples were dehydrated with graded ethanol solutions, freeze-dried, and processed for imaging. Spleens for immunohistochemistry were fixed with 4% (w/v) paraformaldehyde and were embedded in wax. Sections were stained using an antibody against mouse ß-major globin (Research Plus).
Flow cytometry analysis
Cells were prepared from bone marrow and spleen in phosphate-buffered saline with 2% bovine serum albumin (PBS/2% BSA) and the nonspecific binding of antibodies and the Fc
receptor was blocked with CD16/32 antibody (1 : 200 dilution) for 15 min on ice. Subsequently, cells were stained with phycoerythrin (PE)-conjugated anti-TER-119 (PharMingen) and fluorescein isothiocyanate (FITC)-conjugated anti-CD71 (PharMingen; 1 : 200 dilution) for 30 min on ice. PE- and FITC-conjugated rat IgG2b were used as isotype-matched controls. After the final wash with PBS/2% BSA, the samples were subjected to flow cytometric analysis using a FACS Caliber (Becton Dickinson). The data were analysed with the CellQuest program.
In vitro progenitor assay
Mononuclear cells from bone marrow and spleen were prepared using Lymphoprep (Nycomed) according to the manufacturer's instructions. Erythroid colony assays were performed by the standard procedure (Ohneda et al. 1990). For CFU-E assays, 2 x 104 cells were plated in methylcellulose medium (MethoCult M3231, StemCell Technologies) and cultured for 2 days in the presence of 1 U/mL erythropoietin. For BFU-E assays, 1 x 105 cells were plated and cultured for 7 days in the presence of 2 U/mL erythropoietin and 100 ng/mL stem cell factor (generous gifts from Chugai Pharmaceutical Co. and Kirin Brewery, respectively).
Stress induced erythropoiesis experiments
Adult male mice were administered PHZ (Sigma) by intraperitoneal injection for 2 consecutive days. Peripheral blood samples were taken from tail veins every 2 days to determine the hematocrit values and reticulocyte counts. In some experiments, mice were killed on day 2 for histological analysis. Spleen and bone marrow were fixed in 4% paraformaldehyde and embedded in polyester wax (BDH Laboratory). Sections were stained with hematoxylin and eosin.
Osmotic fragility test
Freshly prepared peripheral blood cells were resuspended in graded concentrations of NaCl at a final hematocrit of 1% for 30 min. The NaCl solutions used were at 0.1, 0.2, 0.3, 0.35, 0.4, 0.5, 0.55, 0.6, 0.65, 0.75, and 0.85% weight/volume, pH 7.4. Following low speed centrifugation, the optical density of the supernatant was measured at a wavelength of 540 nm. The optical densities of samples mixed with 0.1% and 0.85% NaCl were regarded as representing 100% and 0% lysis, respectively.
Filtration test
Blood samples were washed three times with PBS. RBCs were diluted with PBS to a hematocrit of 0.6% and filtered through the nickel mesh with micropores (4 µm in diameter) at 37 °C under various pressures (Koyama et al. 2003). The flow rates (ml/min) of RBCs were monitored. Experiments were performed on two separate occasions using five mice from each genotype and the results observed were reproducible.
Protein analysis of RBCs
An aliquot (40 µL) of washed RBCs was added to 1.2 mL of hypotonic buffer (5 mM sodium phosphate, pH 7.4), followed by centrifugation at 17 400 x g at 4 °C for 10 min. Pellets were resuspended in sample buffer, electrophoresed on a 9% polyacrylamide gel and visualized by standard Coomasie Blue staining. Immunoblot analysis was performed as previously described (Shimizu et al. 1995).
Statistical analysis
Statistical analysis was performed using the Student's t-test.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
* Correspondence: E-mail: masi{at}tara.tsukuba.ac.jp or kohneda{at}tara.tsukuba.ac.jp
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Delaunay, J. (2002) Molecular basis of red cell membrane disorders. Acta Haematol. 108, 210–218.[CrossRef][Medline]
Evans, T. & Felsenfeld, G. (1989) The erythroid-specific transcription factor Eryf1: a new finger protein. Cell 58, 877–885.[CrossRef][Medline]
Evans, T., Reitman, M. & Felsenfeld, G. (1988) An erythrocyte-specific DNA-binding factor recognizes a regulatory sequence common to all chicken globin genes. Proc. Natl. Acad. Sci. USA
85, 5976–5980.
Fujiwara, Y., Browne, C.P., Cunniff, K., Goff, S.C. & Orkin, S.H. (1996) Arrested development of embryonic red cell precursors in mouse embryos lacking transcription factor GATA-1. Proc. Natl. Acad. Sci. USA
93, 12355–12358.
Gilligan, D.M., Lozovatsky, L., Gwynn, B., Brugnara, C., Mohandas, N. & Peters, L.L. (1999) Targeted disruption of the beta adducin gene (Add2) causes red blood cell spherocytosis in mice. Proc. Natl. Acad. Sci. USA
96, 10717–10722.
Grass, A.G., Boyer, M.E., Pal, S., Wu, J., Weiss, J.W. & Bresnick, E.H. (2003) GATA-1-dependent transcriptional repression of GATA-2 via disruption of positive autoregulation and domain-wide chromatin remodeling. Proc. Natl. Acad. Sci. USA
100, 8811–8816.
Gurbuxani, S., Vyas, P. & Crispino, J.D. (2004) Recent insights into the mechanisms of myeloid leukemogenesis in Down syndrome. Blood
103, 399–406.
Igarashi, K., Kataoka, K., Itoh, K., Hayashi, N., Nishizawa, M. & Yamamoto, M. (1994) Regulation of transcription by dimerization of erythroid factor NF-E2 p45 with small Maf proteins. Nature 367, 568–572.[CrossRef][Medline]
Katsuoka, F., Motohashi, H., Onodera, K., Suwabe, N., Engel, J.D. & Yamamoto, M. (2000) One enhancer mediates mafK transcriptional activation in both hematopoietic and cardiac muscle cells. EMBO J. 19, 2980–2991.[CrossRef][Medline]
Koyama, T., Maruyama, T., Seki, R., O'Rear, E.A., Shio, H. & Uyesaka, N. (2003) A new quantitative filtration apparatus to evaluate erythrocyte deformability. Jpn. J. Physiol. 53, 481–486.[CrossRef][Medline]
Mackay, J.P., Kowalski, K., Fox, A.H., Czolij, R., King, G.F. & Crossley, M. (1998) Involvement of the N-finger in the self association of GATA-1. J. Biol. Chem.
273, 30560–30567.
Martin, D.I. & Orkin, S.H. (1990) Transcriptional activation and DNA binding by the erythroid factor GF-1/NF-E1/Eryf 1. Genes Dev.
4, 1886–1898.
Martin, D.I., Tsai, S.F. & Orkin, S.H. (1989) Increased gamma-globin expression in a nondeletion HPFH mediated by an erythroid-specific DNA-binding factor. Nature 338, 435–438.[CrossRef][Medline]
Muro, A.F., Marro, M.L., Gajovic, S., Porro, F., Luzzatto, L. & Baralle, F.E. (2000) Mild spherocytic hereditary elliptocytosis and altered levels of alpha- and gamma-adducins in beta-adducin-deficient mice. Blood
95, 3978–3985.
Nishikawa, K., Kobayashi, M., Masumi, A., et al. (2003) Self association of Gata1 enhances transcriptional activity in vivo in zebra fish embryos. Mol. Cell. Biol.
23, 8295–8305.
Nishimura, S., Takahashi, S., Kuroha, T., et al. (2000) A GATA box in the GATA-1 gene hematopoietic enhancer is a critical element in the network of GATA factors and sites that regulate this gene. Mol. Cell. Biol.
20, 713–723.
Nuez, B., Michalovich, D., Bygrave, A., Ploemacher, R. & Grosfeld, F. (1995) Defective haematopoiesis in fetal liver resulting from inactivation of the EKLF gene. Nature 375, 316–318.[CrossRef][Medline]
Ohneda, K. & Yamamoto, M. (2002) Roles of hematopoietic transcription factors GATA-1 and GATA-2 in the development of red blood cell lineage. Acta Haematol. 108, 237–245.[CrossRef][Medline]
Ohneda, O., Yanai, N. & Obinata, M. (1990) Microenvironment created by stromal cells is essential for a rapid expansion of erythroid cells in mouse fetal liver. Development
110, 379–384.
Ohneda, K., Shimizu, R., Nishimura, S., et al. (2002) A minigene containing four discrete cis elements recapitulates GATA-1 gene expression in vivo. Genes Cells 7, 1243–1254.[Abstract]
Onodera, K., Takahashi, S., Nishimura, S., et al. (1997) GATA-1 transcription is controlled by distinct regulatory mechanisms during primitive and definitive erythropoiesis. Proc. Natl. Acad. Sci. USA
94, 4487–4492.
Onodera, K., Shavit, J.A., Motohashi, H., Yamamoto, M. & Engel, J.D. (2000) Perinatal synthetic lethality and hematopoietic defects in compound mafG, mafK mutant mice. EMBO J. 19, 1335–1345.[CrossRef][Medline]
Perkins, A.C., Sharpe, A.H. & Orkin, S.H. (1995) Lethal beta-thalassaemia in mice lacking the erythroid CACCC-transcription factor EKLF. Nature 375, 318–322.[CrossRef][Medline]
Reitman, M. & Felsenfeld, G. (1988) Mutational analysis of the chicken beta-globin enhancer reveals two positive-acting domains. Proc. Natl. Acad. Sci. USA
85, 6267–6271.
Shimizu, T., Takakuwa, Y., Koizumi, H., Ishibashi, T. & Ohkawara, A. (1995) Localization of immuno-analogues of erythrocyte protein 4.1 and spectrin in epidermis of psoriasis vulgaris. Histochem. Cell Biol. 103, 363–368.[CrossRef][Medline]
Shimizu, R., Takahashi, S., Ohneda, K., Engel, J.D. & Yamamoto, M. (2001) In vivo requirements for GATA-1 functional domains during primitive and definitive erythropoiesis. EMBO J. 20, 5250–5260.[CrossRef][Medline]
Shimizu, R., Ohneda, K., Engel, J.D., Trainor, C.D. & Yamamoto, M. (2004) Transgenic rescue of GATA-1-deficient mice with GATA-1 lacking a FOG–1 association site phenocopies patients with X-linked thrombocytopenia. Blood
103, 2560–2567.
Socolovsky, M., Nam, H., Fleming, M.D., Haase, V.H., Brugnara, C. & Lodish, H.F. (2001) Ineffective erythropoiesis in Stat5a (-/-) 5b (-/-) mice due to decreased survival of early erythroblasts. Blood
98, 3261–3273.
Suwabe, N., Takahashi, S., Nakano, T. & Yamamoto, M. (1998) GATA-1 regulates growth and differentiation of definitive erythroid lineage cells during in vitro ES cell differentiation. Blood
92, 4108–4118.
Takahashi, S., Onodera, K., Motohashi, H., et al. (1997) Arrest in primitive erythroid cell development caused by promoter- specific disruption of the GATA-1 gene. J. Biol. Chem.
272, 12611–12615.
Takahashi, S., Shimizu, R., Suwabe, N., et al. (2000) GATA factor transgenes under GATA-1 locus control rescue germline GATA-1 mutant deficiencies. Blood
96, 910–916.
Trainor, C.D., Omichinski, J.G., Vandergon, T.L., Gronenborn, A.M., Clore, G.M. & Felsenfeld, G. (1996) A palindromic regulatory site within vertebrate GATA-1 promoters requires both zinc fingers of the GATA-1 DNA-binding domain for high–affinity interaction. Mol. Cell. Biol. 16, 2238–2247.[Abstract]
Tsai, S.F., Martin, D.I., Zon, L.I., D’Andrea, A.D., Wong, G.G. & Orkin, S.H. (1989) Cloning of cDNA for the major DNA-binding protein of the erythroid lineage through expression in mammalian cells. Nature 339, 446–451.[CrossRef][Medline]
Tsang, A.P., Visvader, J.E., Turner, C.A., et al. (1997) FOG, a multitype zinc finger protein, acts as a cofactor for transcription factor GATA-1 in erythroid and megakaryocytic differentiation. Cell 90, 109–119.[CrossRef][Medline]
Vannucchi, A.M., Bianchi, L., Cellai, C., et al. (2001) Accentuated response to phenylhydrazine and erythropoietin in mice genetically impaired for their GATA-1 expression (GATA-1low mice). Blood
97, 3040–3050.
Wall, L., deBoer, E. & Grosveld, F. (1988) The human beta-globin gene 3' enhancer contains multiple binding sites for an erythroid-specific protein. Genes Dev.
2, 1089–1100.
Wechsler, J., Greene, M., McDevitt, M.A., et al. (2002) Acquired mutations in GATA1 in the megakaryoblastic leukemia of Down syndrome. Nature Genet. 32, 148–152.[CrossRef][Medline]
Weiss, M.J., Keller, G. & Orkin, S.H. (1994) Novel insights into erythroid development revealed through in vitro differentiation of GATA-1 embryonic stem cells. Genes Dev.
8, 1184–1197.
Weiss, M.J. & Orkin, S.H. (1995) GATA transcription factors: key regulators of hematopoiesis. Exp. Hematol. 23, 99–107.
