Regulatory interferences at the iron transporter ferroportin 1 (Fpn1) cause transient defects in iron homeostasis and erythropoiesis in polycythaemia(Pcm) mutant mice. The present study identified decreased Fpn1 expression in placental syncytiotrophoblast cells at late gestation as the mechanism of neonatal iron deficiency in Pcm mutants. Tissue specificity of embryonic Fpn1 dysregulation was evident from concomitant decreases in Fpn1 mRNA and protein expression in placenta and liver, as opposed to upregulation of Fpn1 protein despite decreased transcript levels in spleen, implicating post-transcriptional regulation of Fpn1. Dysregulation of Fpn1 and decreased iron levels in Pcm mutant spleens correlated with apoptotic cell death in the stroma, resulting in a semidominant spleen regression. At 7 weeks of age, a transient increase in spleen size in Pcm heterozygotes reflected a transient erythropoietin-mediated polycythemia. Structurally, Pcm mutant spleens displayed a severe defect in red pulp formation, including disruption of the sinusoidal endothelium, as well as discrete defects in white pulp organization during postnatal development. Reduced functional competence of the Pcmmutant spleen was manifested by an impaired response to chemically induced hemolytic anemia. Thus, aberrant Fpn1 regulation and iron homeostasis interferes with development of the spleen stroma during embryogenesis,resulting in a novel defect in spleen architecture postnatally.

As a cofactor mediating oxidation-reduction chemistry (for a review, see Aisen et al., 2001), iron governs a plethora of pathways involved in cellular metabolism. For example,iron is essential for hemoglobin synthesis, which comprises the majority of daily iron utilization in humans. Organismal iron levels must be tightly controlled, because both iron deficiency and overload lead to severe consequences ranging from anemia to iron-induced tissue damage affecting multiple organs (for a review, see Andrews,2000). Mammalian iron homeostasis is achieved via the organismal interplay of multiple proteins that are regulated by both transcriptional and post-transcriptional mechanisms (for a review, see Hentze et al., 2004). In the duodenum, the pivotal site of organismal iron uptake, ferroportin 1 (Fpn1;also known as MTP1, IREG1, SLC11A3, SLC40A1) functions as the principal basolateral iron transporter in enterocytes(Abboud and Haile, 2000; Donovan et al., 2000; McKie et al., 2000). Furthermore, Fpn1 expression has been detected at other sites involved in iron homeostasis, including hepatocytes, reticuloendothelial macrophages of the liver and spleen, and placental syncytiotrophoblast cells (for a review, see McKie and Barlow, 2004).

We have recently identified a 58-base pair microdeletion in the promoter region of the Fpn1 locus in radiation-induced polycythaemia(Pcm) mice (Mok et al.,2004). This deletion, located four nucleotides upstream of the TATA box, conferred aberrant transcription initiation at the Fpn1locus, which resulted in the absence of the iron-responsive element (IRE) in the 5′ untranslated region (UTR) of the vast majority of hepatic Fpn1 transcripts in Pcm homozygotes. Consistent with a critical in vivo role for the IRE in translational regulation of the Fpn1 mRNA in response to cellular iron levels, Pcm mutant mice displayed significant elevations in Fpn1 protein levels in liver and duodenum, as well as organismal iron overload during early postnatal development. Strikingly, an erythropoietin (Epo)-dependent polycythemia was evident at 7 weeks of age in Pcm heterozygotes. Abrogation of the iron accumulation and polycythemia in adult Pcm mutant animals implicated hepcidin (Hamp), the hormonal regulator of iron homeostasis, in a superimposed regulatory mechanism on Fpn1 protein levels(Mok et al., 2004). Unexpectedly, Pcm mutant mice demonstrated a profound iron deficiency at birth, with a severe hypochromic, microcytic anemia in homozygotes. In the present study, we have identified dysregulated expression of placental Fpn1 during late gestation as the basis for neonatal iron deficiency in Pcm mutants. Furthermore, we have uncovered evidence for tissue specificity of developmental Fpn1 dysregulation, which associates with apoptotic cell death and semidominant defects in Pcm spleen organogenesis.

Surprisingly little is known about the development of the spleen, which is derived from the coelomic epithelium and mesenchyme of the dorsal mesogastrium(Green, 1967). The spleen subserves important organismal functions, which can be attributed to distinct anatomical regions within the organ. It represents the largest single lymphoid organ and performs a critical immunological role reflected in the highly organized arrangement of lymphoid and accessory cells within the white pulp(for a review, see Fu and Chaplin,1999). The lymphocytic composition of the white pulp is different from the red pulp, which also hosts a sizeable plasma cell population(Nolte et al., 2000; Garcia De Vinuesa et al.,1999). Whereas the white pulp is primarily engaged in mounting adaptive immune responses, the red pulp, composed of a stromal reticular network of endothelial sinusoids, macrophages, erythroid and accessory cells,performs a digestive function, clearing damaged and senescent red blood cells from the circulation. This composite function of the spleen is a direct reflection of embryonic spleen development, and results from colonization of a stromal infrastructure, consisting of intrinsic mesenchyme, by extrinsic hematopoietic cells at approximately embryonic (E) day 15 of murine embryogenesis (Sasaki and Matsumura,1988).

Aside from the relative dearth of well-characterized splenic markers,insight into the mechanisms regulating spleen development has been hampered by a limited number of mutant alleles. Over the past decades, several murine models of aberrant spleen development have been characterized, including dominant hemimelia (Searle,1959; Green,1967), and null mutants for transcription factors Hox11(Roberts et al., 1994; Dear et al., 1995; Koehler et al., 2000; Kanzler and Dear, 2001), Wt1 (Herzer et al.,1999), Bapx1 (Lettice et al., 1999; Tribioli and Lufkin, 1999; Akazawa et al.,2000), and capsulin (Lu et al., 2000). Following specification of the spleen primordium around E11.5 (Dear et al.,1995), these mutants display progressive spleen regression,resulting in its complete absence as early as E13.5, but no later than E15.5. Therefore, based on the early regression of the spleen, these models are ill-suited to address hematopoietic cell colonization and potential cellular interactions between intrinsic and extrinsic cell populations required for structural and functional competence of the spleen. Here, we report a novel defect in late embryonic spleen development in the context of aberrant fetal iron homeostasis in Pcm mutant mice, which manifests as a severe disruption of the red pulp sinusoidal endothelium in the postnatal spleen.

Mice and genotyping

Pcm mice originated from radiation mutagenesis of(101/HeH×C3H/HeH) F1 hybrids(Cattanach, 1995), and were crossed onto the A/J inbred background to generate partial congenics(Mok et al., 2004). In the present study, analyses were performed on animals from N5 and subsequent backcross generations. Age-matched embryos were generated by timed matings;noon of the day of vaginal plug detection was considered E0.5. PCR-based genotyping was performed using primers flanking the Pcm deletion, as described previously (Mok et al.,2004). All animal experiments in this study were approved by the Institutional Animal Care and Use Committee of Baylor College of Medicine.

Western blot analysis

Protein lysates from embryonic spleen, placenta and liver were prepared and quantified as described previously (Mok et al., 2004). Five μg of unboiled protein lysates from liver and placenta and 2.5 μg of unboiled spleen lysate were separated by electrophoresis on 8% SDS polyacrylamide gels and transferred to PVDF membrane(BioRad). Blocking was achieved by incubation in TRIS-buffered saline containing 5% BSA and 0.1% Tween 20. Membranes were incubated overnight at 4°C using primary antibodies against mouse Fpn1 (kindly provided by D. Haile) at 1:2000 and mouse actin (Santa Cruz Biotechnology) at 1:5000. Following washing, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000) against rabbit (for Fpn1)or goat (for actin), and signals were detected using luminol reagent (Santa Cruz Biotechnology).

mRNA analyses

Semi-quantitative RT-PCR for Hamp expression on E16.5 liver mRNA,and 5′ rapid amplification of cDNA ends (5′RACE) from E16.5 placenta and liver as well as E15.5 spleen were performed as described previously (Mok et al., 2004). Fpn1 mRNA expression was quantified via real-time RT-PCR on total RNA samples isolated from E16.5 placental and liver as well as E15.5 spleen. Reactions and signal detection were performed on an ABI Prism 7000 Sequence Detection System (Applied Biosystems). Commercially available Fpn1primers and probe were employed (Mm00489837_m1, Applied Biosystems). An 18S rRNA assay was performed using primers 5′-TCGAGGCCCTGTAATTGGAA-3′ (forward),5′-CCCTCCAATGGATCCTCGTT-3′ (reverse), and TaqMan MGB probe 5′-AGTCCACTTTAAATCCTT-3′ labeled with VIC (Applied Biosystems). Relative amounts of mRNA were expressed as a ratio to 18S rRNA levels, and normalized to an arbitrary wild-type (WT) ratio of 1.

Histology and immunohistochemistry

For histology and immunohistochemistry for Fpn1, F4/80 and Wt1, spleen,placenta and liver were fixed in Bouin's reagent (postnatal stages) or 4%PFA/PBS (embryonic stages), dehydrated in a graded series of ethanol, embedded in paraffin and sectioned at 5-10 μm. Prussian blue staining for iron was performed using the Accustain iron staining kit, according to the manufacturer(Sigma). For Fpn1 and F4/80 antigen retrieval, sections were treated with 3%H2O2 in methanol, whereas for Wt1, sections were boiled in 0.01 M citric acid, pH 6.0. Blocking was achieved with goat (Fpn1 and Wt1)or rabbit serum (F4/80). Primary antibody incubations were performed overnight at 4°C using antibodies against Fpn1 at 1:100, F4/80 (clone CI:A3-1,Serotec) at 1:100, or Wt1 (C-19, Santa Cruz Biotechnology) at 1:1000 dilution. Secondary antibodies were peroxidase-conjugated with the Vectastain Elite ABC Kit (Vector Laboratories), followed by signal detection with Vector NovaRED substrate (Vector Laboratories). Immunohistochemistry for B220, CD3, Ter119,MAdCAM-1, IBL-7/1, IBL-9/2 and IBL-7/22 was performed on frozen sections of 12-week-old spleens. Cryostat sections were stained with rat hybridoma supernatants containing IBL-7/1, IBL-9/2 and IBL-7/22 antibodies followed by biotin-amplified alkaline phosphatase detection as described(Balázs et al., 2001). Briefly, the sections were fixed in chilled acetone for 10 minutes, dried, and were encircled with water-repellent wax pen. Following rehydration with PBS containing 10% BSA for 20 minutes and removal of excess buffer, the sections were incubated with undiluted hybridoma supernatants against lymphocyte subsets and endothelial markers for 45 minutes. The Ly-76 erythroid antigen was detected by the monoclonal antibody Ter119 (BD Pharmingen) at 1 μg/ml concentration in PBS. The reaction was developed with biotinylated mouse anti-rat kappa chain reagent (clone MRK-1, BD Pharmingen) at 1 μg/ml, and ExtraAvidine-alkaline phosphatase conjugate (Sigma-Aldrich) using NBT/BCIP in the presence of 1 mg/ml levamisole. Sections were counterstained with methylene green. Rat monoclonal antibodies used: anti-B220 (clone RA3-6B2);anti-CD3 (clone KT-3); anti-MAdCAM-1 (clone MECA-367); IBL-7/1, IBL-7/22 and IBL-9/2 monoclonal antibodies were produced in the Department of Immunology and Biotechnology, University of Pécs, Pécs, Hungary(Balázs et al.,2001).

Determination of iron levels in embryonic tissues

The non-heme iron content of individual E15.5 liver samples was determined as described previously (Mok et al.,2004). Because of the small amount of tissue, E15.5 spleens were pooled by genotype, dried, and quantified similar to liver samples.

TUNEL analysis

Embryonic spleen and liver were dissected out from E15.5 and E16.5 embryos,fixed in 4% PFA/PBS for 2 hours at 4°C, dehydrated in a graded series of ethanol, embedded in paraffin and sectioned at 5 μm. Sections were processed for detection of TUNEL-positive cells using the In Situ Cell Death Detection Kit (Roche), and mounted with the Vectashield Hard Set mounting medium, containing DAPI (Vector Laboratories). Images were acquired via fluorescence microscopy for fluorescein (TUNEL) and DAPI signals using separate channels, and merged using Adobe Photoshop version 6.0.

Phenylhydrazine treatment of mice

To induce hemolytic anemia in 3- and 7-week-old wild-type and Pcmmutant mice, phenylhydrazine hydrochloride dissolved in sterile saline was administered intraperitoneally at a dose of 60 mg/kg body weight twice at a 24-hour interval on day 0 and day 1. Mice were euthanized on day 4, and spleens were dissected out and weighed. Baseline spleen values were obtained by comparison of spleen weights from untreated wild-type and heterozygous mutant mice at 3 and 7 weeks of age.

Statistical analyses

All data are reported as the mean±s.d. All comparisons were performed versus wild-type cohorts, and analyzed for statistically significant differences using the Student's unpaired t-test.

Decreased placental and liver Fpn1 protein expression in Pcmmutant embryos

We have previously demonstrated that Pcm mutant animals are iron deficient at birth (Mok et al.,2004), consistent with the putative role of placental Fpn1 in maternal-to-fetal iron transport (Donovan et al., 2000; McKie et al.,2000). In support of this hypothesis, we observed a marked decrease in placental Fpn1 expression in Pcm mutants at E16.5 by western blot analysis (Fig. 1A). Likewise, Pcm mutant liver showed a similar decrease in Fpn1 protein levels (Fig. 1B). Fpn1 expression has been demonstrated at the basolateral surface of syncytiotrophoblast cells in human placenta, where it is implicated in iron transport to the embryonic circulation(Donovan et al., 2000). Indeed, by immunohistochemistry, high levels of Fpn1 expression were detected in syncytiotrophoblast cells in the labyrinth cell layer in wild-type mouse placenta (Fig. 1D). In contrast, Fpn1 expression was decreased to near background levels in Pcm homozygous placenta at E16.5(Fig. 1D), consistent with western analysis (Fig. 1A). Furthermore, the architecture of the labyrinth cell layer appeared to be preserved in mutant placenta (data not shown).

Fig. 1.

Decreased Fpn1 protein expression in Pcm mutant placenta and embryonic liver is consistent with fetal iron deficiency, and contrasts with increased Fpn1 expression in embryonic spleen. At E16.5, placenta (A) and liver (B) show a graded decrease in Fpn1 protein expression on western blot. Approximate molecular mass: Fpn1, 68 kDa; actin, 41 kDa. (C) Western blot analysis reveals increased Fpn1 protein expression in E15.5 spleen. (D)Immunohistochemistry demonstrates a significant reduction of labyrinth cell expression of Fpn1 in E16.5 placenta from Pcm homozygotes. Original magnification 400×. (E) Immunohistochemistry reveals a widespread increase in Fpn1 expression in E15.5 Pcm mutant spleens. Original magnification 400×. (F) Analysis of non-heme iron demonstrates significant reduction of iron content in E15.5 Pcm livers, more severe in homozygotes. *, P<0.0001. (G)Semiquantitative RT-PCR of E16.5 liver indicates lower Hamp mRNA expression in Pcm mutants. Band sizes: Hamp, 171 bp;β-actin, 250 bp.

Fig. 1.

Decreased Fpn1 protein expression in Pcm mutant placenta and embryonic liver is consistent with fetal iron deficiency, and contrasts with increased Fpn1 expression in embryonic spleen. At E16.5, placenta (A) and liver (B) show a graded decrease in Fpn1 protein expression on western blot. Approximate molecular mass: Fpn1, 68 kDa; actin, 41 kDa. (C) Western blot analysis reveals increased Fpn1 protein expression in E15.5 spleen. (D)Immunohistochemistry demonstrates a significant reduction of labyrinth cell expression of Fpn1 in E16.5 placenta from Pcm homozygotes. Original magnification 400×. (E) Immunohistochemistry reveals a widespread increase in Fpn1 expression in E15.5 Pcm mutant spleens. Original magnification 400×. (F) Analysis of non-heme iron demonstrates significant reduction of iron content in E15.5 Pcm livers, more severe in homozygotes. *, P<0.0001. (G)Semiquantitative RT-PCR of E16.5 liver indicates lower Hamp mRNA expression in Pcm mutants. Band sizes: Hamp, 171 bp;β-actin, 250 bp.

Iron balance in Pcm mutant embryos

Given the maximal maternal iron transport to the fetus during late gestation (for a review, see Srai et al.,2002), decreased placental expression of Fpn1 should result in iron deficiency in Pcm mutant embryos. Indeed, a significant reduction of hepatic iron content in Pcm mutant embryos was observed,more severe in homozygotes (Fig. 1F). Corollary with the decreased hepatic iron content(Fig. 1F), Pcm mutant spleens exhibited a graded decrease in spleen iron levels at E15.5. Because of the small size of the spleens, only multiple samples pooled by genotype yielded readings above background level (n=4 +/+, 13.3 ng iron/spleen; n=5, Pcm/+, 9.9 ng iron/spleen; n=5, Pcm/Pcm, 4.9 ng iron/spleen). Nonetheless, it appears likely that,despite pooling of samples, the colorimetric assay may underestimate the degree of iron deficiency in Pcm mutant spleens. Furthermore, at all stages analyzed, i.e. E15.5 until birth, homozygous mutant embryos exhibited marked pallor (data not shown). Thus, decreased placental expression of Fpn1 probably represents the major contributing factor to neonatal iron deficiency and anemia in Pcm mutant mice(Mok et al., 2004).

Interestingly, Hamp mRNA levels were significantly reduced in E16.5 Pcm mutant liver by semi-quantitative RT-PCR(Fig. 1G), indicating that Hamp, the principal negative hormonal regulator of iron balance(Park et al., 2001; Pigeon et al., 2001) (for a review, see Ganz, 2003), is also responsive to iron deficiency during embryogenesis.

Increased Fpn1 protein expression in the spleen of Pcmmutant embryos

In striking contrast to placenta and liver, Fpn1 protein levels were higher in embryonic spleen from Pcm mutant embryos(Fig. 1C). Importantly,immunohistochemistry detected high levels of Fpn1 expression throughout the spleen of Pcm homozygous embryos, consistent with upregulation of Fpn1 expression in stromal cells (Fig. 1E). Thus, the present study uncovered evidence for tissue-specific regulation of embryonic Fpn1 expression by virtue of differential protein levels in Pcm mutant placenta and liver compared with spleen.

Decreased Fpn1 mRNA levels in placenta, liver and spleen

To determine whether transcript abundance caused the observed changes in embryonic Fpn1 protein levels, we performed real-time RT-PCR analysis for Fpn1 mRNA expression. Fpn1 mRNA levels were statistically significantly reduced in both placenta and liver from Pcm mutants(Fig. 2A,B), consistent with the decreased Fpn1 protein expression (Fig. 1A,B). Surprisingly, Fpn1 mRNA levels were also significantly reduced in Pcm homozygous mutant spleen(Fig. 2C), despite increased Fpn1 protein levels (Fig. 1C,E). This discordance between Fpn1 mRNA and protein expression in the spleen indicated that the upregulation of Fpn1 resulted from a post-transcriptional mechanism.

Fig. 2.

Embryonic Fpn1 mRNA expression is decreased in Pcm mutant mice. (A) Real-time RT-PCR analysis of E15.5 spleen shows reduction of Fpn1 mRNA in Pcm homozygous mutants. **, P<0.01. Decreased mRNA levels in placenta (B) and liver (C) in E16.5 Pcm mutants. *, P<0.05; **, P<0.01; ***, P<0.001. 5′RACE PCR of E15.5 spleen (D) and E16.5 placenta mRNA (E) demonstrates aberrant transcripts in Pcm homozygous mutants. Wild-type band, 630 bp.

Fig. 2.

Embryonic Fpn1 mRNA expression is decreased in Pcm mutant mice. (A) Real-time RT-PCR analysis of E15.5 spleen shows reduction of Fpn1 mRNA in Pcm homozygous mutants. **, P<0.01. Decreased mRNA levels in placenta (B) and liver (C) in E16.5 Pcm mutants. *, P<0.05; **, P<0.01; ***, P<0.001. 5′RACE PCR of E15.5 spleen (D) and E16.5 placenta mRNA (E) demonstrates aberrant transcripts in Pcm homozygous mutants. Wild-type band, 630 bp.

Aberrant transcription initiation at the Fpn1 locus leads to disruption of the IRE-mediated post-transcriptional regulation during postnatal development in Pcm mutant mice(Mok et al., 2004). Similar to postnatal liver, analysis of Fpn1 transcripts by 5′RACE from wild-type E15.5 spleen and E16.5 placenta showed a single, predominant band,reflecting homogeneous transcription start sites(Fig. 2D,E). However, analysis of Pcm/Pcm placenta and embryonic spleen revealed a significant reduction in the intensity of this band, as well as multiple, additional transcripts (Fig. 2D,E). To test whether transcript identity could explain the divergent Fpn1 protein expression levels in placenta and spleen, we performed DNA sequence analysis of the 5′RACE clones. Sequence data from wild-type transcripts from both the placenta and spleen reiterated published transcription start sites(Liu et al., 2002; Mok et al., 2004). In contrast, cDNAs recovered from Pcm mutant tissues demonstrated significant transcript heterogeneity and were aberrant with respect to transcription start site and mRNA splicing. Fpn1 transcripts devoid of the IRE, which were predominant in homozygous mutant tissue postnatally(Mok et al., 2004), comprised less than half of sequenced clones from mutant placenta and embryonic spleen(data not shown). Importantly, despite transcript heterogeneity, no evidence for aberrant Fpn1 protein isoforms was observed by western analysis (data not shown). Therefore, neither Fpn1 transcript abundance nor sequence appeared sufficient to explain the increased Fpn1 protein levels in Pcm mutant embryonic spleen, indicating an unknown,post-transcriptional regulatory mechanism.

Semidominant defect in Pcm mutant spleens during late embryogenesis

To ascertain whether aberrant Fpn1 protein expression in E15.5 spleen associated with an organ phenotype, we conducted a time-course analysis of spleen development. At E15.5, the spleen appeared to be consistent across all genotypes with regard to organ size and gross morphology(Fig. 3A). However, beginning at E16.5 (data not shown) and increasingly evident at E17.5, Pcmmutant spleens were reduced in size relative to wild type, more severe in homozygotes (Fig. 3B). It is noteworthy that the spleen size in homozygotes effectively regressed, rather than merely stalled, as evident from comparison of spleen size at E17.5 and E15.5 (Fig. 3A,B). At birth, a completely penetrant, semidominant defect in spleen size was observed in Pcm mutant pups (Fig. 3C). Thus, unlike previous models of defective splenogenesis(Searle, 1959; Roberts et al., 1994; Herzer et al., 1999; Lettice et al., 1999; Lu et al., 2000), we observed grossly intact spleens in Pcm mutants at E15.5(Fig. 3A), indicating that Pcm represents a novel murine model for disrupted splenogenesis.

Fig. 3.

Pcm mutant mice show a semidominant defect in spleen development during late embryogenesis. (A) Spleen development and size appear consistent across all genotypes at E15.5. Original magnification 40×. (B) By E17.5,significant reduction in spleen size is observed in Pcm mutants, more severe in homozygotes. Compare with A and note that spleen size in homozygotes regresses. Original magnification 40×. At P0 (C) and 7 weeks (D),spleens (arrowheads) of homozygous mutants maintain consistent decrease in size, whereas heterozygotes undergo a significant increase in size from P0 to 7 weeks. Original magnification (C) 16× and (D) 6×. H&E staining of spleen sections at P0 (E) and 7 weeks (F) reveals red and white pulp compartmentalization of wild-type and mutant spleens; arrowhead indicates white pulp follicle. Original magnification (E) 200× and (F)50×.

Fig. 3.

Pcm mutant mice show a semidominant defect in spleen development during late embryogenesis. (A) Spleen development and size appear consistent across all genotypes at E15.5. Original magnification 40×. (B) By E17.5,significant reduction in spleen size is observed in Pcm mutants, more severe in homozygotes. Compare with A and note that spleen size in homozygotes regresses. Original magnification 40×. At P0 (C) and 7 weeks (D),spleens (arrowheads) of homozygous mutants maintain consistent decrease in size, whereas heterozygotes undergo a significant increase in size from P0 to 7 weeks. Original magnification (C) 16× and (D) 6×. H&E staining of spleen sections at P0 (E) and 7 weeks (F) reveals red and white pulp compartmentalization of wild-type and mutant spleens; arrowhead indicates white pulp follicle. Original magnification (E) 200× and (F)50×.

Increased apoptosis in spleens from Pcm mutant embryos

The regression of spleen size in Pcm homozygous mutants during late gestation (Fig. 3A,B)implicated cell death by apoptosis as the likely mechanism. Indeed, a significant increase in TUNEL-positive foci was detected in E15.5 and E16.5 homozygous mutant spleens (Fig. 4A,B). In addition, heterozygous mutants exhibited moderately increased TUNEL foci relative to wild-type spleens, which demonstrated minimal baseline apoptosis. Importantly, significantly increased apoptotic cell death appeared to be restricted to Pcm mutant spleens, as the liver displayed similar levels of TUNEL-positive foci across all genotypes at these stages (data not shown).

Fig. 4.

Increased apoptosis detected in Pcm mutant spleens. TUNEL analysis at E15.5 (A) and E16.5 (B) indicates a significant increase in TUNEL-positive foci in Pcm mutant spleens, higher in homozygotes. Images represent merged composites of TUNEL (fluorescein) and DAPI-counterstained sections,original magnification 200×. (C) Immunohistochemistry for F4/80 at E15.5 reveals similar distribution of positive cells within spleens across all genotypes. Original magnification, 400×. (D) Immunohistochemistry for Wt1 at E15.5 demonstrates widespread stromal cell staining within the spleen. Original magnification, 400×.

Fig. 4.

Increased apoptosis detected in Pcm mutant spleens. TUNEL analysis at E15.5 (A) and E16.5 (B) indicates a significant increase in TUNEL-positive foci in Pcm mutant spleens, higher in homozygotes. Images represent merged composites of TUNEL (fluorescein) and DAPI-counterstained sections,original magnification 200×. (C) Immunohistochemistry for F4/80 at E15.5 reveals similar distribution of positive cells within spleens across all genotypes. Original magnification, 400×. (D) Immunohistochemistry for Wt1 at E15.5 demonstrates widespread stromal cell staining within the spleen. Original magnification, 400×.

At E15, the spleen constitutes a pre-hematopoietic organ, consisting of predominantly mesenchymal cells and few, scattered mononuclear cells(Sasaki and Matsumura, 1988). Using a monoclonal antibody against a macrophage/monocyte-associated surface antigen, F4/80 (Morris et al.,1991), we observed no difference in the number of macrophage/monocyte lineage cells among the genotypes in E15.5 spleens(Fig. 4C). Strikingly,F4/80-positive cells appeared to be less frequent than TUNEL-positive foci in homozygous mutant spleens (Fig. 4A,B). Furthermore, macrophage/monocyte lineage cells were observed throughout all later stages of embryonic spleen development at a similar frequency and distribution across all genotypes (data not shown).

The tumor suppressor Wt1 plays an essential role in organ development, including that of the kidney, gonad and mesothelial structures(Kreidberg et al., 1993). In addition, Wt1 regulates spleen development, as Wt1 null mice exhibit complete regression of the spleen by E15.5(Herzer et al., 1999). Using Wt1 as a marker for splenic stromal cells, analysis at E15.5 demonstrated a generalized pattern of Wt1 expression encompassing the majority of cells present at this stage in both wild-type and Pcm mutant spleens(Fig. 4D). At birth, similar to other putative stromal cell markers in the spleen, such as Hox11(Dear et al., 1995; Kanzler and Dear, 2001) and Bapx1 (Akazawa et al.,2000), Wt1 expression levels appeared significantly reduced and were primarily restricted to the capsule and subcapsular region in wild type(data not shown). Because of the low levels of Wt1 expression and abnormal spleen structure in postnatal day 0 (P0) Pcm homozygotes(Fig. 3C), Wt1 analysis at this stage was not informative (data not shown). Importantly, at E15.5, the pattern of Wt1 expression (Fig. 4D)correlated with the Fpn1 expression pattern in Pcm mutant spleens(Fig. 1E).

Taken together, the regression of Pcm spleens could not be reconciled with primarily macrophage cell death during spleen development. Rather, increased Fpn1 protein expression in stromal cells, in the context of organismal iron deficiency, associates with apoptotic death of stromal cells,resulting in the semidominant defect in spleen size at birth.

Postnatal changes in Pcm mutant spleens

Homozygous Pcm mutants maintained a decreased spleen size, as observed at 7 weeks of age (Fig. 3D). Conversely, heterozygous spleens gradually increased in size from birth, reaching wild-type size at 7 weeks of age(Fig. 3D). Histologically, at P0, Pcm heterozygous spleens appeared comparable to wild type,whereas homozygous spleens displayed discrete alterations in cellular composition, such as a decreased prevalence of basophilic cells(Fig. 3E). Postnatal organization and development of the white pulp appeared to occur to a significant extent in both heterozygous and homozygous mutant spleens, as distinct white pulp follicles were clearly identifiable at 7 weeks(Fig. 3F). In addition,heterozygous spleens were notable for increased eosinophilic cells within the red pulp as well as the marginal sinus surrounding the white pulp follicles. Seven-week-old homozygous spleens exhibited an increased white pulp to red pulp ratio relative to wild type (Fig. 3F). Given the increased rate of red blood cell production characteristic of Pcm heterozygotes at 7 weeks of age(Mok et al., 2004), the histological changes observed in the red pulp were consistent with increased erythropoiesis in Pcm heterozygous spleens.

Impaired red pulp function in Pcm heterozygous spleen

Given the regression in spleen size during late gestation(Fig. 3C,D), Pcmhomozygotes are likely to become functionally asplenic. In contrast, Pcm heterozygotes retained a significant amount of spleen stroma, and exhibited a significant increase in relative spleen size from during postnatal development (Fig. 3C,D). Consistent with its reduced size at P0(Fig. 3C), heterozygous mutant spleens demonstrated a significantly lower weight at 3 weeks of age as compared with wild type (Fig. 5A). However, by 7 weeks of age, and correlating with the gross appearance (Fig. 3D), similar spleen weights were measured in wild-type and heterozygous mutant mice(Fig. 5A). This increase in relative spleen size was transient in nature, as by 12 weeks of age,heterozygous spleen weights were again statistically significantly reduced as compared with wild type (Fig. 5A).

Fig. 5.

Pcm heterozygous mutant spleens exhibit functional competence for red pulp hyperplasia and red blood cell clearance. (A) Spleen weights demonstrate a transient increase in Pcm heterozygotes at 7 weeks of age. *, P<0.01. Phenylhydrazine treatment of 3-week (B)and 7-week-old mice (C) demonstrates reduced splenic hyperplasia in Pcm heterozygotes. **, P<0.0001. (D) Prussian blue staining of P0 spleen reveals no iron accumulation in wild-type or Pcm heterozygous mutants. Original magnification, 200×. (E) At 12 weeks of age, Pcm heterozygotes display significant red pulp iron accumulation. Original magnification, 100×.

Fig. 5.

Pcm heterozygous mutant spleens exhibit functional competence for red pulp hyperplasia and red blood cell clearance. (A) Spleen weights demonstrate a transient increase in Pcm heterozygotes at 7 weeks of age. *, P<0.01. Phenylhydrazine treatment of 3-week (B)and 7-week-old mice (C) demonstrates reduced splenic hyperplasia in Pcm heterozygotes. **, P<0.0001. (D) Prussian blue staining of P0 spleen reveals no iron accumulation in wild-type or Pcm heterozygous mutants. Original magnification, 200×. (E) At 12 weeks of age, Pcm heterozygotes display significant red pulp iron accumulation. Original magnification, 100×.

Recently, we showed that Pcm heterozygotes displayed a transient,Epo-stimulated polycythemia at 7 weeks of age(Mok et al., 2004). Concordant with the normalization of both Epo expression and hematocrit in 12-week old Pcm mutants (Mok et al.,2004), the splenic weight declined at this age(Fig. 5A). Furthermore, the prevalence of eosinophilic cells within the red pulp and marginal sinus, which was prominent at 7 weeks of age, restored to wild-type pattern (data not shown). Therefore, in agreement with the significant function of the postnatal spleen as an erythropoietic organ in the mouse(Brodsky et al., 1966; Bozzini et al., 1970), the transient nature of the increase in spleen size and histological changes in Pcm heterozygotes probably reflects augmented splenic erythropoiesis because of Epo stimulation.

To quantify the functional erythropoietic capacity of Pcmheterozygous spleens, we performed phenylhydrazine treatment of mice to induce hemolytic anemia (Itano et al.,1975). Strikingly, Pcm heterozygous spleens were notable for a significantly reduced hyperplastic response at both 3 and 7 weeks of age(Fig. 5B,C), indicating a decreased functional capacity to foster erythroid progenitor and precursor proliferation. No increase in spleen size was evident in phenylhydrazine-treated Pcm homozygotes at 3 weeks of age, consistent with functional asplenia (data not shown).

Additional evidence for the functional capacity of Pcmheterozygous spleens was gleaned from the analysis of iron distribution in the context of erythrophagocytosis. At P0, no iron staining was observed within the Pcm heterozygous spleen, reflecting the perinatal iron deficiency in Pcm mutants (Fig. 5D). However, by 12 weeks of age, Pcm heterozygotes displayed a significant, punctate pattern of iron accumulation confined to the red pulp region (Fig. 5E). Given the polycythemia at 7 weeks of age and hematocrit normalization at 12 weeks of age in Pcm heterozygotes, this staining pattern probably reflects erythrophagocytosis followed by iron scavenging within the reticuloendothelial macrophages of the red pulp.

Taken together, the Pcm heterozygous mutant spleens appear to retain red pulp function with regard to erythrophagocytosis, as well as extramedullary hematopoiesis in response to stimulation, albeit at a significantly reduced level relative to wild type.

Characterization of hematopoietic cell lineages in postnatal Pcm mutant spleens

Ter119 represents a marker for late erythroid lineage cells(Kina et al., 2000). Within the red pulp, Ter119-positive erythroid cells were observed at a similar density and distribution in wild-type and heterozygous spleens at 12 weeks of age (Fig. 6A). Homozygotes displayed a moderately reduced density of Ter119-positive cells in red pulp regions (Fig. 6A). The mature white pulp follicle consists of a T-cell rich, periarteriolar lymphoid sheath(PALS) region, which is surrounded by primary follicles comprised of B cells,the predominant lymphoid cell population of the spleen(Veerman and van Ewijk, 1975)(for a review, see Fu and Chaplin,1999). Pcm heterozygous spleens displayed a similar appearance of B- and T-cell regions, as demonstrated by B220 and CD3 expression, respectively (Fig. 6B,C). Interestingly, Pcm homozygotes exhibited an increased ratio of B to T cells. In addition, the central arteriole, normally positioned within a central location of the T-cell PALS region, was located more peripherally in heterozygotes, and appeared within the B-cell region in Pcm homozygotes (Fig. 6B,C). Thus, although the overall follicular structure of the white pulp appeared largely intact, lymphocyte composition and positioning of the central arteriole were abnormal in Pcm mutant spleens, more severe in homozygotes.

Fig. 6.

Intact white pulp follicles but aberrant red pulp sinusoidal architecture in Pcm mutant spleens. Immunohistochemistry of hematopoietic lineage markers Ter119 (A), B220 (B), and CD3 (C) within the spleen at 12 weeks of age. Arrowheads denote central arteriole. (D) MAdCAM-1 immunohistochemistry indicates the presence of marginal sinuses in Pcm mutant spleens at 12 weeks of age. Immunohistochemistry of IBL-7/1 (E), IBL-9/2 (F), and IBL-7/22 (G) demonstrates abnormal arrangement of the microvasculature endothelial components of the red pulp in Pcm mutant spleens at 12 weeks of age. Original magnification, 200×.

Fig. 6.

Intact white pulp follicles but aberrant red pulp sinusoidal architecture in Pcm mutant spleens. Immunohistochemistry of hematopoietic lineage markers Ter119 (A), B220 (B), and CD3 (C) within the spleen at 12 weeks of age. Arrowheads denote central arteriole. (D) MAdCAM-1 immunohistochemistry indicates the presence of marginal sinuses in Pcm mutant spleens at 12 weeks of age. Immunohistochemistry of IBL-7/1 (E), IBL-9/2 (F), and IBL-7/22 (G) demonstrates abnormal arrangement of the microvasculature endothelial components of the red pulp in Pcm mutant spleens at 12 weeks of age. Original magnification, 200×.

Red pulp sinusoid abnormalities in Pcm mutant spleens

The marginal sinus forms a discontinuous endothelial network surrounding the white pulp, which, in conjunction with the marginal zone, represents the site of immunological interactions and cellular extravasation from peripheral blood into the white pulp (Schmidt et al.,1993) (for a review, see Fu and Chaplin, 1999). The expression of cellular adhesion molecule MAdCAM-1 marks endothelial cells of the marginal sinus in the spleen(Kraal et al., 1995). We observed MAdCAM-1 expression surrounding white pulp follicles in Pcmmutant spleens at 12 weeks of age, delineating the presence of the marginal sinus (Fig. 6D). IBL-7/1, which also marks the marginal sinus (Balazs et al., 1999), similarly encompassed the white pulp follicles(Fig. 6E). Thus, the marginal sinus appears to be largely intact in Pcm mutant spleens. Immunoreactivity to IBL-7/1 has also been demonstrated in a heterogeneous pattern within the red pulp sinuses (Balazs et al., 2001), as observed in wild-type spleens(Fig. 6E). Strikingly, this red pulp staining pattern was markedly reduced in both heterozygous and homozygous mutant spleens, indicating an abnormality of the red pulp sinusoids. The IBL-9/2 antibody identifies a complementary subset of red pulp sinusoidal endothelium (Balazs et al.,2001). Likewise, the orderly distribution of IBL-9/2 staining endothelium was severely disrupted in Pcm mutant spleens(Fig. 6F). The IBL-7/22 antibody recognizes sinusoidal components in a pan-endothelial pattern, as well as reticular fibroblasts of the PALS and red pulp(Balazs et al., 1999). Heterozygous spleens displayed a decrease in IBL-7/22 reactivity(Fig. 6G), consistent with the reduced IBL-7/1- and IBL-9/2-positive red pulp sinusoid endothelial subpopulations (Fig. 6E,F). Interestingly, Pcm homozygotes displayed a higher level of IBL-7/22 reactivity than heterozygotes, but were disorganized relative to wild type(Fig. 6G). Given the reduction in IBL/7-1 and IBL-9/2 reactivity in Pcm homozygotes(Fig. 6E,F), this probably reflects an increased prevalence of reticular fibroblasts within the sinusoidal meshwork of the red pulp.

In summary, significant abnormalities in red pulp sinusoidal architecture were observed in Pcm mutant spleens, which are consistent with the reduced functional competence of the red pulp during Epo-stimulated erythropoiesis in Pcm heterozygous spleens.

The hypermorphic Pcm mutation causes upregulation of Fpn1 expression, resulting in increased organismal iron uptake during postnatal development (Mok et al.,2004). Because mutant Fpn1 transcripts lacked an IRE,which inhibits mRNA translation in response to cellular iron levels(Abboud and Haile, 2000; McKie et al., 2000; Liu et al., 2002), our data provided the first in vivo evidence for the importance of posttranscriptional regulation of Fpn1 expression. Surprisingly, Pcm mutant animals exhibited a severe iron deficiency at birth, more severe in homozygotes(Mok et al., 2004). Because fetal iron and other nutrient requirements are met by placental transport from the maternal to fetal blood circulation, maternal iron deficiency exerts profound effects on fetal as well as postnatal development(Crowe et al., 1995) (for a review, see McArdle et al.,2003). Interestingly, fetuses exhibit decreased severity of iron deficiency in comparison to the mother, indicating the presence of regulatory mechanisms to sustain an adequate supply of iron to the fetus(Gambling et al., 2001). Characterization of an Fpn1 mutation in zebrafish identified a role for this iron transporter in maternal-to-fetal efflux of iron via the placenta(Donovan et al., 2000). Our data confirm that decreased placental expression of Fpn1 mediates the fetal iron deficiency, as Pcm homozygous mutants displayed greatly diminished syncytiotrophoblast cell expression of Fpn1 during late gestation(for a model, see Fig. 7). Furthermore, transgenic overexpression studies implicated Hamp in the upregulation of placental iron transport(Nicolas et al., 2002). Conversely, the present results provide evidence that Hamp expression is responsive to the fetal iron status, as Pcm mutant liver exhibited decreased Hamp mRNA levels. Thus, the Hamp-Fpn1 regulatory axis appears to govern iron homeostasis during both embryonic (the present study)and postnatal development (Mok et al.,2004) via its function in placenta, liver and duodenum.

Fig. 7.

Model for Fpn1-mediated embryonic iron deficiency and spleen stromal defects in Pcm mice. Decreased Fpn1 mRNA and protein expression in placental syncytiotrophoblast cells leads to decreased maternal-to-fetal iron transport and embryonic iron deficiency. Although the spleen exhibits decreased Fpn1 mRNA, increased Fpn1 protein is observed, which should mediate cellular iron efflux from stromal cells. Under the influence of iron deficiency and/or hypoxia, spleen stromal cell death translates to decreased organ size and defects of the red pulp vascular endothelium. The broken line indicates the uncertain relationship between decreased Fpn1 expression in the fetal liver and the functional consequences, for instance with regard to fetal erythropoiesis.

Fig. 7.

Model for Fpn1-mediated embryonic iron deficiency and spleen stromal defects in Pcm mice. Decreased Fpn1 mRNA and protein expression in placental syncytiotrophoblast cells leads to decreased maternal-to-fetal iron transport and embryonic iron deficiency. Although the spleen exhibits decreased Fpn1 mRNA, increased Fpn1 protein is observed, which should mediate cellular iron efflux from stromal cells. Under the influence of iron deficiency and/or hypoxia, spleen stromal cell death translates to decreased organ size and defects of the red pulp vascular endothelium. The broken line indicates the uncertain relationship between decreased Fpn1 expression in the fetal liver and the functional consequences, for instance with regard to fetal erythropoiesis.

Placental Fpn1 expression has been hypothesized to be developmentally regulated in the mouse, with highest levels during late gestation (McKie et al.,2000). At this critical developmental phase, placental and liver Fpn1 transcript levels are significantly reduced in Pcmmutants, correlating with reduced Fpn1 protein expression. These data strongly suggest that a transcriptional mechanism governs developmental Fpn1 regulation in these tissues. Given the proximity of the Pcm microdeletion to a putative TATA box (Mok et al.,2004), it is conceivable that aberrant transcript levels result from a reduction of transcriptional efficiency and/or basal promoter activity. Alternatively, the microdeletion could disrupt putative transcription factor binding sites. Clearly, further investigation is warranted into the precise mechanisms by which the Pcm microdeletion dysregulates Fpn1transcription.

In striking contrast to placenta and liver, Fpn1 protein levels were increased in Pcm mutant spleen, despite significantly decreased Fpn1 mRNA levels. Based on Fpn1 transcript levels and sequence, it appears unlikely that the increased Fpn1 protein levels result from a transcriptional mechanism. Furthermore, although decreased Hamp expression is consistent with increased Fpn1 protein levels in the spleen, decreased Fpn1 protein expression in the placenta and liver renders a systemic regulatory effect unlikely. Thus, similar to postnatal development (Mok et al.,2004), an unknown post-transcriptional mechanism appears to govern Fpn1 protein levels during late gestation, which warrants additional studies. In this context, it would be of significant interest to characterize further the basis for decreased Fpn1 protein expression in embryonic liver, which,strikingly, transitions to increased hepatic Fpn1 protein levels at birth(Mok et al., 2004). Regardless of the mechanisms of tissue-specific Fpn1 regulation, increased Fpn1-mediated iron efflux in stromal cells should potentiate cellular iron deficiency in the spleen resulting from decreased embryonic iron levels. This pathophysiology correlates with dramatic consequences at the cellular and organ level, which become manifest as a semidominant defect in spleen development during late gestation (for a model, see Fig. 7).

Consistent with its pleiotropic role in cellular metabolism, iron has long been known to modulate pathways that regulate proliferation and cell death(for a review, see Le and Richardson,2002). For example, iron chelation has been demonstrated to induce apoptosis (Fukuchi et al.,1994; Haq et al.,1995) and promote cell-cycle arrest(Lederman et al., 1984). Similarly, cellular hypoxia represents a potent stimulus for the apoptotic cascade (for a review, see Brunelle and Chandel, 2002). The effects of iron chelation mimic cellular hypoxia (Wang and Semenza,1993), presumably via iron-dependent regulators of the hypoxia signaling cascade, such as a prolyl hydroxylase(Ivan et al., 2001; Jaakkola et al., 2001), which regulates hypoxia-inducible factor-1α (HIF-1α) (for a review, see Semenza, 2003). Therefore, it is conceivable that apoptotic cell death detected in Pcm mutant spleens results from a synergistic effect of iron deficiency and anemia during development. Recently, upregulation of the pleomorphic adenomas gene-like 2(PLAGL2) protein, a putative zinc-finger transcription factor, has been demonstrated in response to hypoxia and iron chelation(Furukawa et al., 2001). In turn, PLAGL2 induces apoptosis and upregulation of a proapoptotic factor BNip3(Mizutani et al., 2002). Interestingly, BNip3 is itself induced by hypoxia, a unique feature of regulation among the Bcl-2 family of apoptotic factors(Bruick, 2000). Thus, given the relative specificity of these factors in response to iron and hypoxia stimuli,we are currently assessing their potential involvement in the stromal cell apoptosis in Pcm mutant spleens. Furthermore, to the best of our knowledge, defects in spleen development have not been reported in extant mouse models of iron deficiency and anemia. A detailed analysis of these mutants would corroborate whether iron deficiency and/or Fpn1 protein upregulation induce apoptosis in spleen stromal cells.

The defects in spleen organogenesis in Pcm mutant mice are distinct from existing genetic mouse models of aberrant spleen development. Complete asplenia is observed in dominant hemimelia(Searle, 1959; Green, 1967), Hox11(Roberts et al., 1994; Dear et al., 1995; Koehler et al., 2000; Kanzler and Dear, 2001), Wt1 (Herzer et al.,1999), Bapx1 (Lettice et al., 1999; Tribioli and Lufkin, 1999; Akazawa et al.,2000) and capsulin mutant mice(Lu et al., 2000). In these mutants, primary induction of the splenic primordium is followed by complete involution of the organ before E15.5. Evidence for apoptotic cell death within the spleen has been observed in Hox11(Dear et al., 1995), Wt1 (Herzer et al.,1999) and capsulin mutant animals(Lu et al., 2000). Because Pcm mutant spleens appear intact at E15.5, this implicates Fpn1 and iron homeostasis in the disruption of a distinct, subsequent developmental phase in spleen organogenesis. Attempts to determine genetic interaction of Hox11 with Wt1 (Koehler et al., 2000), Bapx1(Akazawa et al., 2000) and capsulin (Lu et al., 2000)provided evidence that Wt1 functions downstream of Hox11(Koehler et al., 2000). Interestingly, Pcm mutant spleens at E15.5 exhibit similar Wt1 expression levels and patterns compared with wild type, suggesting that the defects in Pcm mice are independent of Hox11 and Wt1 function. As Pcm heterozygotes display a reduced severity of splenic disruption and homozygotes retain residual spleen tissue, Pcm mice represent a complementary resource to the existing mutants that will enable a more comprehensive understanding of the mechanisms and pathways of spleen organogenesis throughout development.

In the mouse, the red pulp constitutes a significant erythropoietic organ during normal development as well as in response to hypoxia, phlebotomy, or exogenous Epo administration (Brodsky et al., 1966; Bozzini et al.,1970). Pcm heterozygotes demonstrate an Epo-dependent polycythemia at 7 weeks of age (Mok et al., 2004). Here, we show that Pcm heterozygous spleens demonstrate changes consistent with elevated Epo levels, including a transient increase in spleen weight and red pulp hyperplasia, which temporally coincide with the transient polycythemia. In contrast, similar to genetically asplenic or splenectomized mice, the spleen rudiment in Pcm homozygotes should not respond to factors stimulating erythropoiesis(Bozzini et al., 1970; Lozzio, 1972). Therefore,despite elevated Epo levels, functional asplenia and severe perinatal iron deficiency probably represent the limiting factors toward the diminished rate of productive erythropoiesis during postnatal development in Pcmhomozygotes, resulting in the lower peak hematocrit as compared with heterozygous mutants.

The stromal defects during organogenesis of Pcm mutant spleens correlate well with the severe red pulp abnormalities postnatally, such as aberrant sinusoidal endothelial cell populations. In turn, these defects lead to decreased functional competence of Pcm mutant spleens, as reflected by impaired splenic hyperplasia in response to phenylhydrazine treatment. Furthermore, discrete abnormalities of the white pulp, as well as the marginal zone, have also been detected. Thus, it is likely that interactions between the intrinsic, stromal cell population, and extrinsic,hematopoietic cell lineages mediate proper structural and functional organization of the spleen during postnatal development. Therefore, further characterization of the patterning and organization of the white pulp in Pcm mutant spleens should yield additional insight into the mechanisms of spleen organogenesis in mammals.

We thank David Haile for generously providing the Fpn1 antibody reagent used in this study; we are grateful to Jaroslav Jelinek and William Craigen,as well as members of our laboratory, for helpful discussions and critical reading of the manuscript. We acknowledge Sonia Pai and Salomon Durrani for excellent technical assistance, and Gerard Karsenty for generous microscope access. H.M. is supported by an individual National Research Service Award from the National Institute of Environmental Health Sciences, NIH; has received NIH training grant support through the Department of Molecular and Human Genetics; and is a member of the Medical Scientist Training Program of Baylor College of Medicine funded by the NIH. P.B. was supported by the Széchenyi István Postdoctoral Fellowship of the Hungarian Academy of Sciences and ETT grant No. 592/2003. This work was supported by a research grant from the NIH to A.S.

Abboud, S. and Haile, D. J. (
2000
). A novel mammalian iron-regulated protein involved in intracellular iron metabolism.
J. Biol. Chem.
275
,
19906
-19912.
Aisen, P., Enns, C. and Wessling-Resnick, M.(
2001
). Chemistry and biology of eukaryotic iron metabolism.
Int. J. Biochem. Cell. Biol.
33
,
940
-959.
Akazawa, H., Komuro, I., Sugitani, Y., Yazaki, Y., Nagai, R. and Noda, T. (
2000
). Targeted disruption of the homeobox transcription factor Bapx1 results in lethal skeletal dysplasia with asplenia and gastroduodenal malformation.
Genes Cells
5
,
499
-513.
Andrews, N. C. (
2000
). Iron metabolism: iron deficiency and iron overload.
Annu. Rev. Genomics Hum. Genet.
1
,
75
-98.
Balazs, M., Grama, L. and Balogh, P. (
1999
). Detection of phenotypic heterogeneity within the murine splenic vasculature using rat monoclonal antibodies IBL-7/1 and IBL-7/22.
Hybridoma
18
,
177
-182.
Balazs, M., Horvath, G., Grama, L. and Balogh, P.(
2001
). Phenotypic identification and development of distinct microvascular compartments in the postnatal mouse spleen.
Cell. Immunol.
212
,
126
-137.
Bozzini, C. E., Barrio Rendo, M. E., Devoto, F. C. and Epper, C. E. (
1970
). Studies on medullary and extramedullary erythropoiesis in the adult mouse.
Am. J. Physiol.
219
,
724
-728.
Brodsky, I., Dennis, L. H., Kahn, S. B. and Brady, L. W.(
1966
). Normal mouse erythropoiesis. I. The role of the spleen in mouse erythropoiesis.
Cancer Res.
26
,
198
-201.
Bruick, R. K. (
2000
). Expression of the gene encoding the proapoptotic Nip3 protein is induced by hypoxia.
Proc. Natl. Acad. Sci. USA
97
,
9082
-9087.
Brunelle, J. K. and Chandel, N. S. (
2002
). Oxygen deprivation induced cell death: an update.
Apoptosis
7
,
475
-482.
Cattanach, B. M. (
1995
). A dominant polycythaemia.
Mouse Genome
93
,
1027
-1028.
Crowe, C., Dandekar, P., Fox, M., Dhingra, K., Bennet, L. and Hanson, M. A. (
1995
). The effects of anaemia on heart,placenta and body weight, and blood pressure in fetal and neonatal rats.
J. Physiol.
488
,
515
-519.
Dear, T. N., Colledge, W. H., Carlton, M. B., Lavenir, I.,Larson, T., Smith, A. J., Warren, A. J., Evans, M. J., Sofroniew, M. V. and Rabbitts, T. H. (
1995
). The Hox11 gene is essential for cell survival during spleen development.
Development
121
,
2909
-2915.
Donovan, A., Brownlie, A., Zhou, Y., Shepard, J., Pratt, S. J.,Moynihan, J., Paw, B. H., Drejer, A., Barut, B., Zapata, A. et al.(
2000
). Positional cloning of zebrafish ferroportin1 identifies a conserved vertebrate iron exporter.
Nature
403
,
776
-781.
Fu, Y. X. and Chaplin, D. D. (
1999
). Development and maturation of secondary lymphoid tissues.
Annu. Rev. Immunol.
17
,
399
-433.
Fukuchi, K., Tomoyasu, S., Tsuruoka, N. and Gomi, K.(
1994
). Iron deprivation-induced apoptosis in HL-60 cells.
FEBS Lett.
350
,
139
-142.
Furukawa, T., Adachi, Y., Fujisawa, J., Kambe, T.,Yamaguchi-Iwai, Y., Sasaki, R., Kuwahara, J., Ikehara, S., Tokunaga, R. and Taketani, S. (
2001
). Involvement of PLAGL2 in activation of iron deficient- and hypoxia-induced gene expression in mouse cell lines.
Oncogene
20
,
4718
-4727.
Gambling, L., Danzeisen, R., Gair, S., Lea, R. G., Charania, Z.,Solanky, N., Joory, K. D., Srai, S. K. and McArdle, H. J.(
2001
). Effect of iron deficiency on placental transfer of iron and expression of iron transport proteins in vivo and in vitro.
Biochem. J.
356
,
883
-889.
Ganz, T. (
2003
). Hepcidin, a key regulator of iron metabolism and mediator of anemia of inflammation.
Blood
102
,
783
-788.
Garcia De Vinuesa, C., Gulbranson-Judge, A., Khan, M., O'Leary,P., Cascalho, M., Wabl, M., Klaus, G. G., Owen, M. J. and MacLennan, I. C.(
1999
). Dendritic cells associated with plasmablast survival.
Eur. J. Immunol.
29
,
3712
-3721.
Green, M. C. (
1967
). A defect of the splanchnic mesoderm caused by the mutant gene dominant hemimelia in the mouse.
Dev. Biol.
15
,
62
-89.
Haq, R. U., Wereley, J. P. and Chitambar, C. R.(
1995
). Induction of apoptosis by iron deprivation in human leukemic CCRF-CEM cells.
Exp. Hematol.
23
,
428
-432.
Hentze, M. W., Muckenthaler, M. U. and Andrews, N. C.(
2004
). Balancing acts: molecular control of mammalian iron metabolism.
Cell
117
,
285
-297.
Herzer, U., Crocoll, A., Barton, D., Howells, N. and Englert,C. (
1999
). The Wilms tumor suppressor gene wt1 is required for development of the spleen.
Curr. Biol.
9
,
837
-840.
Itano, H. A., Hirota, K. and Hosokawa, K.(
1975
). Mechanism of induction of haemolytic anaemia by phenylhydrazine.
Nature
256
,
665
-667.
Ivan, M., Kondo, K., Yang, H., Kim, W., Valiando, J., Ohh, M.,Salic, A., Asara, J. M., Lane, W. S. and Kaelin, W. G., Jr(
2001
). HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing.
Science
292
,
464
-468.
Jaakkola, P., Mole, D. R., Tian, Y. M., Wilson, M. I., Gielbert,J., Gaskell, S. J., Kriegsheim, A. V., Hebestreit, H. F., Mukherji, M.,Schofield, C. J. et al. (
2001
). Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation.
Science
292
,
468
-472.
Kanzler, B. and Dear, T. N. (
2001
). Hox11 acts cell autonomously in spleen development and its absence results in altered cell fate of mesenchymal spleen precursors.
Dev. Biol.
234
,
231
-243.
Kina, T., Ikuta, K., Takayama, E., Wada, K., Majumdar, A. S.,Weissman, I. L. and Katsura, Y. (
2000
). The monoclonal antibody TER-119 recognizes a molecule associated with glycophorin A and specifically marks the late stages of murine erythroid lineage.
Br. J. Haematol.
109
,
280
-287.
Koehler, K., Franz, T. and Dear, T. N. (
2000
). Hox11 is required to maintain normal Wt1 mRNA levels in the developing spleen.
Dev. Dyn.
218
,
201
-206.
Kraal, G., Schornagel, K., Streeter, P. R., Holzmann, B. and Butcher, E. C. (
1995
). Expression of the mucosal vascular addressin, MAdCAM-1, on sinus-lining cells in the spleen.
Am. J. Pathol.
147
,
763
-771.
Kreidberg, J. A., Sariola, H., Loring, J. M., Maeda, M.,Pelletier, J., Housman, D. and Jaenisch, R. (
1993
). WT-1 is required for early kidney development.
Cell
74
,
679
-691.
Le, N. T. and Richardson, D. R. (
2002
). The role of iron in cell cycle progression and the proliferation of neoplastic cells.
Biochim. Biophys. Acta
.
1603
,
31
-46.
Lederman, H. M., Cohen, A., Lee, J. W., Freedman, M. H. and Gelfand, E. W. (
1984
). Deferoxamine: a reversible S-phase inhibitor of human lymphocyte proliferation.
Blood
64
,
748
-753.
Lettice, L. A., Purdie, L. A., Carlson, G. J., Kilanowski, F.,Dorin, J. and Hill, R. E. (
1999
). The mouse bagpipe gene controls development of axial skeleton, skull, and spleen.
Proc. Natl. Acad. Sci. USA
96
,
9695
-9700.
Liu, X., Hill, P. and Haile, D. J. (
2002
). Role of the ferroportin iron-responsive element in iron and nitric oxide dependent gene regulation.
Blood Cells Mol. Dis.
29
,
315
-326.
Lozzio, B. B. (
1972
). Hematopoiesis in congenitally asplenic mice.
Am. J. Physiol.
222
,
290
-295.
Lu, J., Chang, P., Richardson, J. A., Gan, L., Weiler, H. and Olson, E. N. (
2000
). The basic helix-loop-helix transcription factor capsulin controls spleen organogenesis.
Proc. Natl. Acad. Sci. USA
97
,
9525
-9530.
McArdle, H. J., Danzeisen, R., Fosset, C. and Gambling, L.(
2003
). The role of the placenta in iron transfer from mother to fetus and the relationship between iron status and fetal outcome.
Biometals
16
,
161
-167.
McKie, A. T. and Barlow, D. J. (
2004
). The SLC40 basolateral iron transporter family (IREG1/ferroportin/MTP1).
Pflügers Arch.
447
,
801
-806.
McKie, A. T., Marciani, P., Rolfs, A., Brennan, K., Wehr, K.,Barrow, D., Miret, S., Bomford, A., Peters, T. J., Farzaneh, F. et al.(
2000
). A novel duodenal iron-regulated transporter, IREG1,implicated in the basolateral transfer of iron to the circulation.
Mol. Cell
.
5
,
299
-309.
Mizutani, A., Furukawa, T., Adachi, Y., Ikehara, S. and Taketani, S. (
2002
). A zinc-finger protein, PLAGL2, induces the expression of a proapoptotic protein Nip3, leading to cellular apoptosis.
J. Biol. Chem.
277
,
15851
-15858.
Mok, H., Jelinek, J., Pai, S., Cattanach, B. M., Prchal, J. T.,Youssoufian, H. and Schumacher, A. (
2004
). Disruption of ferroportin 1 regulation causes dynamic alterations in iron homeostasis and erythropoiesis in polycythaemia mice.
Development
131
,
1859
-1868.
Morris, L., Graham, C. F. and Gordon, S.(
1991
). Macrophages in haemopoietic and other tissues of the developing mouse detected by the monoclonal antibody F4/80.
Development
112
,
517
-526.
Nicolas, G., Bennoun, M., Porteu, A., Mativet, S., Beaumont, C.,Grandchamp, B., Sirito, M., Sawadogo, M., Kahn, A. and Vaulont, S.(
2002
). Severe iron deficiency anemia in transgenic mice expressing liver hepcidin.
Proc. Natl. Acad. Sci. USA
99
,
4596
-4601.
Nolte, M. A., Hoen, E. N., van Stijn, A., Kraal, G. and Mebius,R. E. (
2000
). Isolation of the intact white pulp. Quantitative and qualitative analysis of the cellular composition of the splenic compartments.
Eur. J. Immunol.
30
,
626
-634.
Park, C. H., Valore, E. V., Waring, A. J. and Ganz, T.(
2001
). Hepcidin, a urinary antimicrobial peptide synthesized in the liver.
J. Biol. Chem.
276
,
7806
-7810.
Pigeon, C., Ilyin, G., Courselaud, B., Leroyer, P., Turlin, B.,Brissot, P. and Loreal, O. (
2001
). A new mouse liver-specific gene, encoding a protein homologous to human antimicrobial peptide hepcidin,is overexpressed during iron overload.
J. Biol. Chem.
276
,
7811
-7819.
Roberts, C. W., Shutter, J. R. and Korsmeyer, S. J.(
1994
). Hox11 controls the genesis of the spleen.
Nature
368
,
747
-749.
Sasaki, K. and Matsumura, G. (
1988
). Spleen lymphocytes and haemopoiesis in the mouse embryo.
J. Anat.
160
,
27
-37.
Schmidt, E. E., MacDonald, I. C. and Groom, A. C.(
1993
). Comparative aspects of splenic microcirculatory pathways in mammals: the region bordering the white pulp.
Scanning Microsc.
7
,
613
-628.
Searle, A. G. (
1959
). Hereditary absence of spleen in the mouse.
Nature
184
,
1419
-1420.
Semenza, G. L. (
2003
). Targeting HIF-1 for cancer therapy.
Nat. Rev. Cancer
3
,
721
-732.
Srai, S. K., Bomford, A. and McArdle, H. J.(
2002
). Iron transport across cell membranes: molecular understanding of duodenal and placental iron uptake.
Best Pract. Res. Clin. Haematol.
15
,
243
-259.
Tribioli, C. and Lufkin, T. (
1999
). The murine Bapx1 homeobox gene plays a critical role in embryonic development of the axial skeleton and spleen.
Development
126
,
5699
-5711.
Veerman, A. J. and van Ewijk, W. (
1975
). White pulp compartments in the spleen of rats and mice. A light and electron microscopic study of lymphoid and non-lymphoid celltypes in T- and B-areas.
Cell Tissue Res.
156
,
417
-441.
Wang, G. L. and Semenza, G. L. (
1993
). Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction.
Blood
82
,
3610
-3615.