SEARCH

SEARCH BY CITATION

Keywords:

  • thymus;
  • organogenesis;
  • development;
  • FGF10;
  • FGF7;
  • FGFR2IIIb;
  • thymic epithelium

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES

Heterogeneous epithelial populations comprising the thymic environment influence early and late stages of T-cell development. The processes that regulate the differentiation of thymic epithelium and that are responsible for this heterogeneity are not well understood, although mesenchymal/epithelial interactions are clearly involved. Here, we show that targeted expression by thymocytes of an fibroblast growth factor receptor-2IIIb (FGFR2IIIb) ligand, FGF10, profoundly alters the differentiation and function of thymic epithelium (TE). Reconstitution of irradiated lckFGF10 mice with normal bone marrow restores normal thymic organization and function, while wild-type mice reconstituted with lckFGF10 bone marrow recapitulates some of the thymic alterations seen in lckFGF10 mice. We also demonstrate that interference with FGFR2IIIb signaling in the thymus with a soluble FGFR2IIIb dominant-negative fusion protein leads to precocious reductions in thymic size and cellularity that resemble age-related thymic involution. These findings indicate that TE compartments are dynamically maintained and that FGF signals are involved in this process. Developmental Dynamics 236:3459–3471, 2007. © 2007 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES

During initial thymic organogenesis, the thymic rudiment emerges from the invaginations of third pharyngeal pouch endoderm. By approximately day 11 of embryonic development, the murine thymic anlagen consists of a collection of epithelium surrounded by neural crest-derived mesenchyme. After the colonization of this epithelial anlagen by lymphoid progenitor cells, the thymic epithelium (TE) expands dramatically and gives rise to distinct cortical and medullary TE types. Cortical TE supports the progressive differentiation of immature CD4CD8 thymocytes, the emergence of CD4+CD8+ thymocytes, and the subsequent positive selection of a subset of these cells. Medullary TE contributes to the establishment of self-tolerance by presenting a wide array of self-antigens, participating in the elimination of self-reactive mature thymocytes and perhaps supporting further postselection maturation of these cells. There is continued postnatal maturation of the medullary TE compartment, where multiple medullary foci (Rodewald et al., 2001) expand and coalesce to form the complex medullary structure of the adult thymus (Anderson et al., 2000).

Although the heterogeneity of TE is well documented, the program of differentiation that gives rise to this heterogeneity is poorly defined. Several genes that are involved in thymic organogenesis have been identified. Hoxa3, Six1, and Eya1 are required for proper patterning of the pharyngeal pouch endoderm (Manley and Capecchi, 1998; Zou et al., 2006) and the absence of Pax1 (Wallin et al., 1996) or Pax9 (Hetzer-Egger et al., 2002) leads to hypoplastic and/or aberrantly developed thymi. The Foxn1 transcription factor is also critically important for the differentiation or specification of thymic epithelium (Nehls et al., 1996; Dooley et al., 2005b). Recently, p63 activity has been shown to play an important role in maintaining TE progenitor cells (Senoo et al., 2007). It was reported that a fetal TE precursor population able to give rise to cortical and medullary TE could be identified on the basis of staining with the MTS24 antibody (Bennett et al., 2002; Gill et al., 2002), although the validity of the MTS24 epitope as a marker of TE precursor is questionable (Rossi et al., 2007a). Individual fetal thymic epithelial cells with potential to generate both cortical and medullary TE have been demonstrated (Rossi et al., 2006).

In the postnatal thymus, subsets of TE cells that simultaneously express both cortical and medullary markers have been proposed to be the immediate precursors to the cortical and medullary TE populations (Ropke et al., 1995; Klug et al., 1998) and progenitor epithelial cells within the postnatal thymus capable of giving rise to a complete thymic environment have been demonstrated functionally (Bleul et al., 2006).

The signals that promote TE differentiation have also not been clearly defined. Wnt signaling has been implicated in the regulation of Foxn1 expression (Balciunaite et al., 2002), while signaling pathways that regulate NFkB activity are known to play an important role in the development of a functional medullary TE compartment (Burkly et al., 1995; Boehm et al., 2003; Rossi et al., 2007c). Signaling mediated by bone morphogenetic proteins (BMPs) and fibroblast growth factors (FGFs) has also been implicated in proper TE differentiation (Erickson et al., 2002; Jenkinson et al., 2003; Bleul and Boehm, 2005; Rossi et al., 2007c).

FGFs represent a large family of secreted proteins that are mediators of mesenchymal/epithelial interactions through binding to high affinity transmembrane receptors (reviewed in Powers et al., 2000). A subset of the fibroblast growth family, FGF1, FGF3, FGF7 (keratinocyte growth factor; KGF), FGF10, and FGF22 (Umemori et al., 2004) serve as ligands for the IIIb isoform of FGFR2 that is primarily expressed by epithelial cells (reviewed in Powers et al., 2000). Targeted disruption of the FGFR2IIIb gene or expression of a soluble dominant-negative form of this receptor results in a spectrum of developmental defects affecting multiple tissues/organs, including markedly hypoplastic thymi (Celli et al., 1998; De Moerlooze et al., 2000; Revest et al., 2001). Many of the sequelae of FGFR2IIIb deficiency are recapitulated in FGF10-deficient mice (Revest et al., 2001), but not FGF7-deficient mice (Guo et al., 1996), indicating a nonredundant role for FGF10 in these developmental processes.

Persistence of the FGFR2IIIb, FGF7, and FGF10 in the postnatal thymus (Erickson et al., 2002; Gray et al., 2007; Rossi et al., 2007b) provides circumstantial evidence that this signaling pathway continues to play a role in the maintenance of the adult thymic environment. Exogenous FGF7 can ameliorate the deleterious effects of pretransplantation conditioning regimens on thymic function, increasing thymic cellularity and restoration of the peripheral T-cell pool (Panoskaltsis-Mortari et al., 1998; Min et al., 2002; Alpdogan et al., 2006) and can also partially reverse the age-related declines of thymic cellularity and function in aged mice (Alpdogan et al., 2006; Min et al., 2007). Administration of FGF7 to adult mice also transiently alters aspects of thymic epithelial cells (TECs) that have been implicated in TE differentiation (Rossi et al., 2007b).

To examine the impact of persistently elevated intrathymic levels of FGF10 on TE differentiation, we generated transgenic mice where FGF10 expression was targeted to thymocytes. This was accomplished by putting the a human FGF10 transgene under control of the proximal lck promoter, which is preferentially active in thymocytes (Allen et al., 1992). This promoter was also used to direct expression of a soluble dominant-negative FGFR2IIIb receptor (FGFR2DN) to evaluate the impact of impaired thymic FGFR2IIIb signaling. With this approach, we hoped to avoid the neonatal lethality associated with targeted deletion of FGFR2IIIb or systemic elevated expression of soluble FGFR2 receptors (Celli et al., 1998; De Moerlooze et al., 2000) and to avoid the pleiomorphic effects that resulted from FGF7 expression directed by keratin 14 promoter (Guo et al., 1993). Here, we describe the consequences resulting from these alterations of the FGFR2IIIb pathway in the postnatal thymus, consequences that point to an ongoing role for this signaling pathway in the behavior of progenitor thymic epithelial cells in the postnatal thymus.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES

Activity of the proximal lck promoter used to direct thymocyte expression of human FGF10 is first detected at the CD44+CD25 stage of thymocyte development and is expressed by the majority of thymocytes (Shimizu et al., 2001), indicating that transgenes regulated by this promoter will be expressed early during thymic ontogeny. Two independent lines of mice established for this transgene displayed indistinguishable phenotypes. Adult heterozygous TG mice were indistinguishable from normal littermates, displayed no discernable pathology and produced TG offspring in a Mendelian ratio.

Elevated FGFR2IIIb Signaling Disrupted the Normal Pattern of TE Development and Organization

An obvious impact of targeted overexpression of FGF10 on thymic organization was the appearance of large fluid-filled thymic epithelial cysts (Fig. 1a–c). These cysts were microscopically identifiable at birth and macroscopically evident by approximately 2–3 weeks of age, becoming progressively larger with advancing age. While there was some variability in size and frequency, the cysts usually first appeared in the medial aspect of the cephalic pole of the thymus. As shown in Figure 1d, there was morphological heterogeneity among the epithelial cells lining the cysts. Cuboidal epithelial cells lined some of the cysts; others contained ciliated columnar cells. Cells with ultrastructural features of mucous secretion were interspersed among the ciliated cells in the cyst lining, and collectively resembled respiratory epithelium (Fig. 1e). Consistent with a respiratory character, some epithelial cells lining these cysts were reactive with antibodies specific for CC10, a product of Clara cells normally found in conducting portions of the respiratory system (Cardoso et al., 1993; Fig. 1f). Macrophages and thymocytes were found within the lumen of some cysts, but were not a consistent feature.

thumbnail image

Figure 1. Thymic alterations in the lckFGF10 TG thymus. a: Individual thymic lobes from wild-type (WT) and TG mice. Some cysts in lckFGF10 TG thymic lobes are indicated by arrows. b: Organization of WT thymus. Smaller vessels distributed within the cortex (Ctx) and larger vessels restricted to the medulla (Med). c: Organization of lckFGF10 TG thymus. Cysts indicated by asterisks. d: Portions of two cysts with intervening parenchyma in a lckFGF10 TG thymus. Upper cyst (**) contains ciliated epithelial cells (arrows), while the lower cyst (*) does not. e: Some cyst epithelial cells are ciliated and others have cytoplasmic inclusions indicative of mucous secretion (*).f: Two cysts with intervening thymic parenchyma of a lckFGF10 TG thymus. Some of the epithelial cells of the upper cyst (**) are express CC10 (green fluorescence), while the epithelial cells lining the lower cyst (*) do not. Red fluorescence = K8. g: Adipose tissue associated with the capsule of the lckFGF10 TG thymus (white asterisks) and lipid accumulations in subcapsular stromal cells (white arrowheads). h,i: Lipid droplets within the cytoplasm of thymic stromal cells (arrows). j: Vascular cast of normal thymus. Note larger vessels in medulla and smaller vessels radiating into the cortex. k: Vascular cast of lckFGF10 TG thymus. Larger vessels of TG thymi are more prominent in the capsule/subcapsular regions of the thymus, with smaller vessels radiating into the parenchyma. A cyst devoid of blood vessels is indicated (*). l: Demonstration of human fibroblast growth factor-10 (FGF10) expression by thymocytes from lckFGF10 TG mice.

Download figure to PowerPoint

Expansion of perithymic adipose tissue in the TG mice was accompanied by the accumulation of lipid droplets in some stromal cells found in the subcapsular and outer cortical regions of the thymus (Fig. 1g–i), indicating that these cells were acquiring features of adipocytes. This finding may reflect the reported ability of FGF10 to act as a growth factor for preadipocytes (Sakaue et al., 2002) and is likely related to the expansion of the thymic mesenchymal compartment (described below). Concomitant with these stromal alterations, elevated FGFR2IIIb signaling also had a profound effect on the vascular organization of the thymus. In the normal thymus, larger vessels are found in the medulla, with postcapillary venules at the corticomedullary junction and capillary loops extending out into the cortex (Raviola and Karnovsky, 1972; Fig. 1j). This vascular pattern was reversed in the TG thymus, where larger vessels were usually running along the capsule, with capillary loops projecting into the parenchyma (Fig. 1k). The expression of human FGF10 by thymocytes from TG mice is shown in Figure 1l.

PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES

Patterns of keratin expression by TE have been described as indicators of TE phenotype and differentiation (Klug et al., 1998). We examined the keratin expression patterns in wild-type (WT) and TG thymi to determine whether or not chronic hyperstimulation of the FGFR2IIIb pathway would result in altered keratin expression patterns indicative of an altered program of TE differentiation. As shown in Figure 2a–c and in agreement with the literature (Klug et al., 1998), we observed prominent expression of K5 by medullary TE and a small subset of cortical TE in WT thymus, with the majority of the cortical TE expressing a K8+K5 phenotype. We found K8 expression to be more widespread among medullary TE than has been previously reported (Klug et al., 1998, 2002) and observed that most of the medullary TE was K5+K8+. As described previously (Klug et al., 1998), K14 expression was more restricted to medullary TE in the WT thymus when compared with K5; few cortical TE were K8+ K14+, and the corticomedullary boundary of K14 expression was more distinct than that of K5 (Fig. 2d–f). In both cases, the organization of TE can define cortical and medullary compartments, because medullary TE lacks the radial reticular organization displayed by cortical TE.

thumbnail image

Figure 2. Expression of Keratin 5 and Keratin 8. a–f: Wild-type (WT) thymus. a–c: K5 (red) and K8 (green) expression. a, merged; b, K5 only; c, K8 only. d–f: K14 (red) and K8 (green) expression. d, merged; e, K14 only; f, K8 only. g–l: lckFGF10 TG thymus; cysts indicated by white asterisks. g–i: K5 (red) and K8 (green) expression. g, merged; h, K5 only; i, K8 only. j–l: K14 (red) and K8 (green) expression. j, merged; k, K14 only; l, K8 only. m–p: Demonstration of secondary antibody specificity with WT thymus. m: Rabbit anti-K5 antibody detected with Alexa 555 donkey anti-rabbit IgG antibody conjugate. n: Rabbit anti-K5 detected with Alexa 488 donkey anti-rat IgG antibody conjugate. o: Rat anti-K8 detected with Alexa 488 donkey anti-rat IgG antibody conjugate. p: Rabbit anti-K5 detected with Alexa 488 donkey anti-rat IgG antibody conjugate. In m–p, all samples were viewed with appropriate filter combinations for the secondary antibodies used.

Download figure to PowerPoint

The profound alterations that we observed in the epithelial organization of the TG thymus are shown in Figure 2g–l. These samples and the samples shown in Figure 3 are from 8-week-old mice; similar results were obtained at all ages of mice examined (neonatal [RIGHTWARDS ARROW] 12 weeks), although they became more prominent with advancing age. A common feature was the effacement of the distinctive organization of the cortical and medullary stromal compartments, with densely packed TE surrounding cystic structures. Very few K8+K5 TE cells could be identified in FGF10 thymi; the majority of the TE cells were K8+K5+ (Fig. 2g–i). Comparison of single and merged channel images demonstrated that most regions of cells that appeared to be K8+K5 also expressed low levels of K5. However, the epithelial cells lining cysts in the TG thymi were predominantly K8+K5, with occasional K8+K5+ cells interspersed among them.

thumbnail image

Figure 3. Altered thymic environment in lckFGF10 TG mice. a,b: Ep-Cam (green) and K5 (red) in wild-type (WT; a) and TG (b) thymus. c,d: UEA1 (green) and K5 (red) in WT (c) and TG (d) thymus. e,f: DEC205 (green) and K5 (red) in WT (e) and TG (f) thymus. g,h: Ly51 (green) and K14 (red) in WT (g) and TG (h) thymus. i,j: TR7 (green) and K14 (red) in WT (i) and TG (j) thymus. k,l: CD3 (green) and K8 (red) in WT (k) and TG (l) thymus. m,n: N418 (green) and K8 (red) in WT (m) and TG (n) thymus. o–q: F4/80 (green) and K14 (red) in WT (o,p) and TG (q) thymus. C, cortex; M, medulla; asterisk, cyst; FGF, fibroblast growth factor.

Download figure to PowerPoint

Expression of K14 was also widespread in TG thymi, with higher levels in medullary areas and lower levels of expression in the cortex (Fig. 2j–l). Although more TE expressed low levels of K14 compared with K5, there were very few regions of cortex that were K8+K14. The cysts were an exception; most of the TE lining cysts was K8+K14. The coexpression of K8 with K5 or K14 was not due to cross-reactivity of the secondary reagents as evidenced by the single color control staining experiments shown in panels m–p of Figure 2. The Alexa 555-labeled goat anti-rabbit IgG antibodies and the Alexa 488-labeled goat anti-rat IgG antibodies bound their appropriate primary antibodies with no significant cross-reactivity.

The expression of Ep-CAM, which is preferentially associated with K5+ medullary TE in the WT thymus (Fig. 3a), was widespread throughout the lckFGF10 TG thymus, with the highest levels of expression associated with epithelium lining large cysts and smaller, duct-like epithelial structures that also expressed low levels of K5 (Fig. 3b). The widespread expression of Ep-Cam was reminiscent of the normal embryonic day 14 thymus, where most of the TE are Ep-Cam+ (Farr et al., 1991). The fucose-specific lectin, UEA1, which reacts with a subset of mature TE in the WT thymus (Farr and Anderson, 1985; Fig. 3c), also reacts with a small subset of cells in the lckFGF10 TG thymus (Fig. 3d), indicating that the K8+K5+ TE that predominate in the lckFGF10 TG thymus probably does not represent an expansion of terminally differentiated medullary TE.

The clear distinction between cortical and medullary compartments defined by the DEC205 and K14 expression, respectively, in the WT thymus (Kraal et al., 1986; Fig. 2e) was obscured in the TG thymus by a markedly expanded range of K14 expression and scattered DEC205+ cells throughout the thymus (Fig. 3f), although regions of epithelium that were DEC205+/K14low/negative were still evident in subcapsular regions that corresponded to the location of K8+K5/14low/negative cells. It is difficult to interpret the significance of the DEC205 expression within the areas of strong K5 expression, as dendritic cells also express DEC205 (Kraal et al., 1986; see below). The expression of Ly51, a widely used marker of cortical TE (Rouse et al., 1988; Fig. 3g), was dramatically reduced in the TG thymus and restricted to a few isolated cells in subcapsular areas and epithelial cells lining cysts (Fig. 3h). The mesenchymal thymic compartment (as defined by reactivity with ER-TR7 antibodies; Van Vliet et al., 1984), which is restricted to the thymic capsule, thymic vasculature, and scattered medullary cells in the WT thymus (Fig. 3i), was a conspicuous component throughout the TG thymus, with scattered “islands” of epithelium relatively devoid of mesenchymal stromal cells (Fig. 3j). This expansion of mesenchymal elements may be related to the alterations in the vascular organization and expansion of preadipocytes described above. Thus, chronically elevated FGFR2IIIb signaling resulted in global alterations of the thymic environment, perturbing epithelial, mesenchymal, and vascular compartments.

The pleiomorphic effects of elevated thymic levels of FGF10 extended to perturbed thymic distribution of hematogenously derived cells in the thymus. Mature thymocytes, defined by high levels of CD3 expression, were preferentially localized to the medulla of the WT thymus, with lower levels of CD3 expression in the cortex (Fig. 3k). In striking contrast, the TG thymi displayed scattered CD3+ cells throughout the parenchyma (Fig. 3l). Dendritic cells (defined by N418 staining; Metlay et al., 1990) displayed a pronounced medullary localization in the WT thymus (Fig. 3m), but were widely and uniformly distributed in the lckFGF10 TG thymus (Fig. 3n). Macrophages (defined by F4/80 staining; Austyn and Gordon, 1981) were slightly more pronounced in the cortex in the WT thymus (Fig. 3o,p) and were also widely distributed in the TG thymus (Fig. 3q).

Targeted Expression of FGF10 to the Thymus Perturbs Thymocyte Differentiation

The lckFGF10 TG thymi were hypoplastic and yielded significantly fewer thymocytes upon dissociation (Fig. 4a). Samples were from 8-week-old mice. Reductions in thymic cellularity were progressive during the postnatal phase of thymic growth; whereas splenic cellularity was equivalent between WT and TG mice (Fig. 4b), TG mice displayed a marked peripheral T-lymphopenia that was compensated by an increase in the splenic B-cell compartment (data not shown). The impact of altered FGFR2IIIb signaling in the peripheral lymphoid tissues will be presented elsewhere.

thumbnail image

Figure 4. Perturbed thymocyte development in the lckFGF10 TG thymus. a: Thymus and spleen cellularity (n = 6 wild-type [WT] & 6 TG). b: Flow cytometric analyses of thymocyte differentiation in WT mice. c: Flow cytometric analyses of thymocyte differentiation in TG mice. Numbers associated with flow cytometric profiles indicate percentages.

Download figure to PowerPoint

Thymocyte development—defined by CD4 and CD8 expression—was impaired, with a marked and preferential decrease in the representation of CD4+CD8+ thymocytes and reduced levels of CD3 expression (Fig. 4b,c). The representation of CD45R+, CD3 cells in TG thymi was increased; many of these were B cells, as defined by expression of IgM and IgD. Some of the increased representation of thymic B cells in the TG thymus was secondary to the dramatic reduction in thymocyte number, although the absolute numbers of thymic B cells did increase three- to fivefold in the TG thymus. The severity of these alterations correlated well with the progressive disorganization of the normal thymic organization.

The Thymic Phenotype Is Reversible and Dependent on the Source of Thymocytes

The alterations of the thymic environment caused by elevated FGFR2IIIb signaling could represent disruptions in thymic organogenesis and irreversible alterations of the thymic environment or may reflect alterations in the pattern of TEC turnover and ongoing epithelial differentiation that are dictated by persistently elevated thymic levels of FGFR2IIIb ligands. To distinguish between these two possibilities, we examined reciprocal radiation bone marrow chimeras made by reconstituting WT and TG mice with bone marrow from WT or TG mice. In this way, normal levels of FGFR2IIIb ligands could be restored in the postnatal TG thymus and elevated levels of FGFR2IIIb ligands could be introduced into a thymus that had undergone normal organogenesis. Four weeks after irradiation and bone marrow reconstitution, chimeric mice were analyzed (10–12 weeks of age at analysis).

Thymi of TG mice reconstituted with TG bone marrow retained the prominent cystic component and coexpression of K5 and K8 remained widespread, such that cortical and medullary compartments could not be easily identified with this reagent pair (Fig. 5a–c). Similar to the unmanipulated TG thymus, there was less coexpression of K14 with K8, scattered regions of K8+K14−/low TE could be identified (Fig. 5d,e) and the frequency of DEC205+ cortical cells was quite low (Fig. 5f). Reconstitution of irradiated WT mice with TG bone marrow resulted in the generation of prominent epithelial cysts similar to those seen in the unmanipulated TG thymus. However, in contrast to unmanipulated TG thymi or TG[RIGHTWARDS ARROW]TG chimeric thymi, the TG[RIGHTWARDS ARROW]WT chimeric thymi retained organized cortical and medullary epithelial compartments defined by patterns of K5/K8 (Fig. 5g–i) or K14/K8 (Fig. 5j,k) expression. Despite the persistence of these cortical and medullary TE compartments, the cortical marker DEC205 was variably expressed and DEC205+ cells did not display the typical organization of cortical TE (Fig. 5l).

thumbnail image

Figure 5. Alterations in thymic architecture of radiation bone marrow chimeras. Cysts indicated by asterisks. C, cortex; M, medulla. a–f: Thymus of TG[RIGHTWARDS ARROW]TG chimera. a–c: K5 (red) and k8 (green) expression: a, merged; b, K5 only; c, K8 only. d,e: K8 (green) & k14 (red) expression: d, merged; e,K14 only. f: DEC205 (green) and K14 (red) expression (merged). g–l: Thymus of TG[RIGHTWARDS ARROW]WT chimeras. g–i: K5 (red) and K8 (green) expression: g, merged; h, K5 only; i, K8 only. j,k: K8 (green) and K14 (red) expression. j, merged; k, k14 only. l: DEC205 (green) and K14 (red) expression (merged). m–r: Thymus of WT[RIGHTWARDS ARROW]WT chimeras. m–o: K8 (green) and K5 (red) expression: m, merged; n, K5 only; o, k8 only. p,q: K8 (green) and K14 (red) expression: p, merged; q, K14 only. r: DEC205 (green) and K14 (red) expression (merged). s–z: Thymus of WT[RIGHTWARDS ARROW]TG chimera. s–u: K8 (green) and K5 (red) expression: s, merged; t, K5 only; u, K8 only. v,w: K8 (green) and K14 (red) expression. v, merged; w, k14 only. x: DEC205 (green) and K14 (red) expression. y,z: Medulla of WT[RIGHTWARDS ARROW]TG chimeric thymus. Merged K8 (green) and K14 (red).

Download figure to PowerPoint

As expected, thymi of lethally irradiated WT mice reconstituted with WT bone marrow retained the normal compartmentalization of TE subsets. The medullary and cortical epithelial compartments were well defined by K5, K14, and NLDC expression (Fig. 5m–r). Scattered K14+K8+ and K5+K14+ cells were observed in the cortex. Remarkably, reconstitution of TG mice with WT bone marrow resulted in a dramatic normalization of the thymic architecture. A well-defined K8+ K14 NLDC145+ cortical epithelial compartment was evident and the K14+ medullary compartment was normal in size and well circumscribed. In contrast to the normalization of the K14 expression pattern, widespread K5 expression in the WT[RIGHTWARDS ARROW]TG chimeric thymi persisted throughout the cortical and medullary compartments, with many K5+K8+ cortical epithelial cells (Fig. 5s–x). While cystic epithelial structures in the WT[RIGHTWARDS ARROW]G thymi were more prevalent than in the unmanipulated normal thymus, they were much less prominent compared with TG thymi and were smaller, irregular in profile, lined by K8+K14 epithelial cells, and, similar to thymic cysts previously described in the WT thymus (Dooley et al., 2005a), were associated with the medullary TE compartment (Fig. 5y,z).

Ability of the Thymic Environment to Support T-Cell Development Is Also Reversible

The thymic cellularity of WT or TG mice receiving TG bone marrow was reduced comparably (Fig. 6a). In the TG[RIGHTWARDS ARROW]TG chimeras, thymocyte development was essentially abrogated and the majority of the cells recovered from these thymi were B cells (Fig. 6b–d). We attribute the lack of T lymphopoiesis in the TG[RIGHTWARDS ARROW]TG chimeric thymus to reflect the cumulative impact of irradiation and FGF10-dependent alterations of the thymic environment, as opposed to an intrinsic inability of lckFGF10 TG bone marrow cells to follow a normal pattern of T-lineage development. With mixed—95%WT (CD45.1) / 5%TG (CD45.2)—bone marrow chimeras, lymphocyte progenitor cells from lckFGF10 transgenic bone marrow could undergo a program of thymocyte development (Fig. 6e). Thymocyte development in TG[RIGHTWARDS ARROW]WT chimeric mice resembled that of unmanipulated TG mice described above, where the representation of CD4+CD8+ thymocytes was reduced and there was elevated representation of thymic B cells. Because these data represent a “snapshot” of the thymic environment 4 weeks after bone marrow transplantation, it remains to be determined whether or not the thymic alterations that we observed represent a new steady state situation. It will be important to define the kinetics of epithelial alterations after bone marrow transplantation and to assess thymic architecture and function in additional TG[RIGHTWARDS ARROW]WT chimeras at longer intervals after bone marrow transplantation.

thumbnail image

Figure 6. Thymic function in radiation bone marrow chimeras. a: Thymic and splenic cellularity of chimeric thymi. b: Expression of CD4 and CD8 by thymocytes from chimeric thymi. c: Expression of CD3 by thymocytes from chimeric thymi. d: Expression of IgD and IgM thymic B cells from chimeric thymi. Numbers associated with flow cytometric profiles indicate percentages.

Download figure to PowerPoint

Consistent with the restoration of normal thymic architecture in WT[RIGHTWARDS ARROW]TG chimeras, normal thymocyte development was re-established in TG mice after reconstitution with WT bone marrow, although thymic cellularity in WT[RIGHTWARDS ARROW]TG chimeras was reproducibly elevated (four- to eightfold) compared with WT[RIGHTWARDS ARROW]WT bone marrow chimeras (Fig. 6a). This rebound in thymic cellularity also occurred in TG thymic lobes implanted under the kidney capsule of WT recipients (data not shown).

Interference With FGFR2IIIb Signaling in the Postnatal Thymus Leads to Precocious Thymic Involution

While there is convincing evidence that FGFR2IIIb signaling is important for fetal thymic development, the neonatal lethality associated with genetic FGFR2IIIb or FGF10 deficiency precluded a straightforward evaluation of FGFR2IIIb signaling past birth. By targeting expression of a soluble dominant-negative form of FGFR2IIIb expression to the thymus, we were able to avoid deleterious systemic effects and could assess of the consequences of impaired FGFR2IIIb signaling in the context of the postnatal thymus. The ability of soluble FGFR2 to function as a dominant-negative signaling regulator has been previously established (Celli et al., 1998), where the soluble form of the receptor is thought to compete with the membrane forms for available ligand. Three independently generated lines of lckFGFR2IIIb dominant-negative (lckFGFR2DN) TG mice displayed identical phenotypes and one is described here. As expected from the restricted activity of the proximal lck promoter in thymocytes (Allen et al., 1992), expression of the transgene was readily detected in thymocytes from lckFGFR2DN mice, using polymerase chain reaction (PCR) primers that were specific for mouse FGFR2IIIb and mouse IgG1 heavy chain (Fig. 7a). The lckFGFR2DN mice displayed none of the developmental defects that were reported when soluble FGFR2IIIb decoy proteins were expressed systemically (Celli et al., 1998). However, they did display reduced thymic cellularity that was evident at birth and that became progressively more pronounced with advancing age (Fig. 7b). Thymocyte development in the hypoplastic FGFR2DN TG thymus, as defined by expression of CD3 or CD4 and CD8, was indistinguishable from WT littermates (Fig. 7c). Thus, it appears that impaired FGFR2IIIb signaling allowed for a qualitatively normal pattern of thymocyte differentiation that was quantitatively reduced.

thumbnail image

Figure 7. Alterations of thymic function and architecture in mice with thymocyte-targeted expression of a soluble dominant-negative fibroblast growth factor receptor-2IIIb receptor (FGFR2DN) receptor. a: Message for the chimeric FGFR2-immunoglobulin fusion protein is expression by TG but not wild-type (WT) thymocytes. b: Thymic cellularity. c: Expression of CD4 and CD8 by WT and TG thymocytes. Numbers associated with flow cytometric profiles indicate percentages. d–g: WT thymus. d,e: K8 (green) and K14 (red) expression: d, merged; e, K14 only. f,g: K8 (green) and K5 (red) expression: f, merged; g, K5 only. h–k: TG thymus. h,i: K8 (green) and K14 (red) expression: h, merged; i, K14 only. j,k: K8 (green) and K5 (red) expression: j, merged; k, K5 only. l,m: WT thymus. Merged UEA-1 (green) and K5 (red) expression. n,o: TG thymus. Merged UEA-1 (green) and K5 (red) expression.

Download figure to PowerPoint

Organization of the TE Compartment Is Relatively Unaffected by Antagonism of FGFR2IIIb Signaling

Immunohistochemistry was used to examine the organization of the thymic epithelial compartments. In WT littermate control thymi, patterns of K8, K5, and K14 expression were as described earlier (Fig. 7d–g). The thymi from lckFGFR2DN TG mice displayed a typical pattern of K14 expression that clearly delineated cortical and medullary TE compartments (Fig. 7 h,i). However, we noted widespread expression of K5 by cortical TE in the lckFGFR2DN thymus, compared with the much more limited coexpression of K5 and K8 in the WT thymus (Fig. 7f,g vs. 7j,k).

The composition of the medullary compartment was evaluated on the basis of K5 expression and reactivity with the fucose-specific lectin Ulex europeus (Farr and Anderson, 1985). Representation of K5+UEA+ medullary TE was less abundant in the lckFGFR2DN TG mice than in WT littermates (compare Figs. 7l,m and 7n,o), indicating that the expanded K5 expression did not reflect an expansion of differentiated medullary TE in the lckFGFR2DN TG thymus. The thymic phenotype of lckFGFR2DN mice hypocellularity, expansion of K5 expression, and normal thymocyte development, recapitulated the major alterations described in neonatal FGFR2IIIb-deficient mice (Revest et al., 2001).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES

In this study, we have examined the response of postnatal TE to altered levels of FGFR2IIIb signaling. Chronically elevated thymic expression of FGF10-disrupted normal thymic epithelial organization and resulted in abnormal thymocyte differentiation. The generation of reciprocal bone marrow chimeras, which allowed “resetting” of the intrathymic levels of FGF10 in the postnatal thymus, demonstrated the considerable plasticity of postnatal TE in response to altered FGFR2IIIb-mediated signals; thymic phenotype/function was perturbed in WT mice reconstituted with TG bone marrow and normal thymic phenotype/function was restored in TG mice reconstituted with WT bone marrow. While we have described the phenotype of lckFGF10 TG mice here, essentially identical results were obtained with lckFGF7 TG mice (Dooley and Farr, unpublished observations). Finally, expression of a FGFR2DN receptor regulated by the proximal lck promoter resulted in thymic hypoplasia and precocious thymic involution that were associated with altered TE differentiation but a normal pattern of thymocyte development. Although the lckFGFR2DN transgenic mice did allow an assessment of the thymic response of FGFR2DN postnatally and essentially recapitulated the thymic phenotype of FGFR2IIIb-deficient thymus (Revest et al., 2001), we cannot strictly attribute the thymic phenotype to a postnatal response, because proximal lck promoter is active in CD44+CD25 T-cell progenitors which are present in the fetal thymus by approximately 13 days of development. This leaves open the formal possibility that the postnatal thymic phenotype of these mice reflects an irreversible alteration of TE development during fetal development and the postnatal effects that we observed are secondary to a restricted fetal event. Unfortunately, because of the well-established deleterious impact of irradiation on thymic recovery after bone marrow reconstitution, reciprocal WT and lckFGFR2DN radiation bone marrow chimeras did not clarify the issue. Because the readout of FGFR2DN activity was reduced thymic cellularity, it was difficult to discriminate between radiation effects and effects due to transgene expression. Generation of transgenic mice with inducible thymic expression of FGFR2DN or induced thymic deletion of FGFR2IIIb, by avoiding problems associated with irradiation, will be a better approach to this question.

Elevated FGFR2IIIb signaling perturbed several aspects of the thymic environment. One was the emergence of prominent epithelial cysts, some of which displayed features of respiratory epithelium, another endodermal derivative. Thymic cysts also occur in homozygous CD3ϵ26 mice with an early block in thymocyte development, where cyst formation has been attributed to loss of thymocyte-derived signals that participate in normal TE differentiation (Hollander et al., 1995). In contrast to the plasticity of postnatal TE differentiation described here, restoration of normal thymic architecture in the CD3E26 mice by bone marrow transplantation could only be accomplished during the fetal/neonatal period. The requirement for homozygosity of the CD3ϵ26 transgene to see this effect suggests that the mechanism for the thymic phenomena in these mice may be more complex than originally proposed.

Thymic cysts are also a prominent feature in a model where the BMP4 signaling pathway is compromised by thymic expression of noggin, an inhibitor of BMP signaling (Bleul and Boehm, 2005). Because BMP and FGF signaling play opposing roles in some developmental settings (Weaver et al., 2000), the thymic cysts that arise with impaired BMP signaling or with elevated FGFR2IIIb signaling may reflect perturbations of a common pathway of TE differentiation. Although this may seem contradictory to the observation that short-term FGF7 administration up-regulates thymic BMP expression (Rossi et al., 2007b), cathepsin H, which degrades BMP4 and is thought to play a role in limiting BMP4 activity, is also elevated in response to FGFR2IIIb signaling (Lu et al., 2007).

Based on the respiratory character of the epithelium comprising the nude thymic rudiment (Dooley et al., 2005b) and the presence of small, but reproducible epithelial cysts in the normal thymus that also display features of respiratory epithelium (Dooley et al., 2005a), we have previously suggested that the emergence of respiratory type epithelium in the thymus may reflect an alternative developmental pathway available to progenitor epithelial cells resident within the thymus. The dramatic expansion of cyst epithelium and the respiratory character of some of the epithelial cells lining these cysts in response to elevated FGFR2IIIb signaling may reflect an impact of this signaling pathway on the behavior of this putative progenitor epithelial population.

The major alterations of TE composition evident in lckFGF10 TG mice are consistent with a role for this signaling pathway in TE differentiation. We have shown here that, in addition to expansion of epithelial cysts, elevated intrathymic levels of FGF10 also resulted in both the dramatic reduction of L51+ cortical TE and the emergence of glandular epithelial structures expressing high levels of Ep-Cam. The pattern of K5 and K8 expression by TE was also profoundly altered. The significance of this last finding is based on the proposition that the small subset of TE in the cortex and along the corticomedullary junction with a K8+K5+ phenotype in the normal thymus represents a progenitor population that gives rise to K8+K5 cortical or K8K5+ medullary TE (Klug et al., 1998). According to this model, elevated FGFR2IIIb signaling results in a large expansion of K8+K5+ progenitor TE and/or inhibition of TE differentiation. However, this interpretation is complicated by our finding of widespread and variable K8 expression by medullary TE, where few medullary epithelial cells displayed a K8K5+ or K8K14+ phenotype (this report and Gillard et al., 2007). Consequently, the TE alterations in the lckFGF10 thymus could also reflect an expansion of K8+K5+ medullary TE. However, the lack of an expansion in the TE population reactive with UEA, which has long been considered to be a marker of more mature medullary TE, suggests that the expanded K8+K5+ TE population represents an early or intermediate stage of TE differentiation. The presence of a sizable population of K8+K5/K14+ medullary TE in the normal thymus suggests that TE differentiation may be more complex than the binary (K8+K5+[RIGHTWARDS ARROW]K8+K5 and K8K5+) model of TE differentiation that has been put forth. For example, the small population of K8+K5 medullary TE could be precursors to the K8+K5+ TE population, which in turn may encompass the subset of TE cells that eventually express a K8+K5 phenotype in the cortex. These findings highlight the need for additional means and markers that can be used to characterize TE heterogeneity and stages of TE differentiation.

The basis for differences in K8 expression by medullary TE reported here and elsewhere (Klug et al., 1998, 2002) is not obvious and may reflect the use of different fluorochomes (FITC vs. Alexa 488) or differences in instrumentation. Furthermore, because shifts in fluorescence color (green + red [RIGHTWARDS ARROW] yellow) are very dependent on the relative intensity of the fluorescence signals involved, these spectral shifts may not accurately reflect the true extent of colocalization.

The responsiveness of the TE in the postnatal thymus to alterations in intrathymic levels of FGFR2IIIb ligands is significant as this finding clearly indicates that TE heterogeneity and organization are not static consequences established during the initial process of thymic organogenesis, but are rather dynamically maintained throughout life. Based on the widespread expression of FGFR2IIIb by TE (Rossi et al., 2007b), this process could reflect either an impact on the differentiation program of TE progenitor cells or a direct response of differentiated TE subsets to altered intrathymic ligand concentration. The inability of FGF7 or FGF10 to impact the differentiation of TE in the short-term context of fetal thymic organ culture (Jenkinson et al., 2003) seems incompatible with a direct effect on differentiated TE but would be consistent with a role for FGFR2IIIb signaling in regulating the behavior of TE progenitor cells that would become evident over longer time intervals.

Progenitor thymic epithelial cells physically demonstrated in the fetal thymus (Rossi et al., 2006), also persist in the postnatal thymus (Bleul et al., 2006); however, the phenotypic characteristics of these postnatal progenitor cells has yet been determined. The growing appreciation that postnatal thymic epithelium is mitotically active (Gillard and Farr, 2006; Gray et al., 2006; Yang et al., 2006; Rossi et al., 2007b) lends additional support to the proposition that the representation of different epithelial compartments in the postnatal thymus is the result of a dynamic equilibrium between a progenitor population and its differentiating progeny. This means that alterations in the behavior of the progenitor TE population would later become evident in more differentiated TE populations as a consequence of epithelial turnover.

The proposition that progenitor TE could be a target of FGFR2IIIb signaling has a precedent in the organogenesis of the pancreas, where the mesenchyme surrounding the FGFR2IIIb+ endodermal buds destined to become pancreas express FGF10, similar to the relationship between thymic epithelial progenitor endoderm and neural crest-derived mesenchyme (Revest et al., 2001; Jenkinson et al., 2003). In FGF10−/− mice, the pancreas is hypoplastic but contains all of the cellular constituents (Bhushan et al., 2001), analogous to the hypoplastic thymus of FGF10R2IIIb/FGF10-deficient mice (Revest et al., 2001) and the lckFGFR2 DN mice described here.

The hypoplastic nature of the FGF10−/− pancreas has been attributed to an inability of progenitor epithelium to expand, resulting in a smaller progenitor pool; this hypoplasia was corrected when explants of FGF10−/− pancreas primordia were cultured in the presence of exogenous FGF10 (Bhushan et al., 2001), similar to the ability of exogenous FGF7 or FGF10 to restore the expansion of TE progenitors in fetal thymic grafts stripped of mesenchyme (Jenkinson et al., 2003). It is significant for the discussion here that the pancreatic progenitor population that expands in response to FGFR2IIIb ligands in vitro also displays an arrest in further differentiation (Hart et al., 2003; Norgaard et al., 2003). This arrest is reversible; upon removal of FGF7, pancreatic progenitor cells expanded by this cytokine in vitro differentiate en mass to become endocrine cells (Elghazi et al., 2002).

Based on these data, we suggest that impaired FGFR2IIIb signaling compromises the normal maintenance of progenitor TE, which could account for the coordinate reduction of in the numbers of differentiated cortical and medullary TE and for the accumulation of K8+K5+ cortical TE that occurs in different models where this signaling pathway has been compromised (this report and Revest et al., 2001). Conversely, expansion of progenitor TE cells and antagonism of their subsequent differentiation by elevated levels of FGFR2IIIb signaling could account for the thymic phenotype of lckFGF10 mice where immature K8+K5+ TE predominate and epithelial cells resembling other endodermal derivatives emerge. In the context of steady-state TE turnover, the expansion of progenitor TE and the concomitant blockade of their subsequent differentiation in response to elevated FGFR2IIIb signaling provides a reasonable explanation for the plasticity exhibited by TE of the postnatal thymus, particularly the increased thymic size observed in the WT[RIGHTWARDS ARROW]TG chimeric thymi. In the WT[RIGHTWARDS ARROW]TG chimeric thymi, the elevated FGFR2IIIb signaling before irradiation and bone marrow reconstitution may have expanded the pool of TE progenitor cells and thus the restoration of physiological levels of thymic FGFR2IIIb ligands by transplantation with WT bone marrow may give rise to larger numbers of differentiated progeny and thereby provide a larger supportive niche for thymocyte development.

The profound alterations effected by persistent elevations of thymic levels of FGFFGF10 described are superficially contradictory to the well-documented therapeutic effects of limited FGF7 administration in the context of bone marrow transplantation. Some of this effect has been attributed to the ability of exogenous FGF7 to reduce the radiosensitivity of clonogenic stem cells in the small intestine and other proliferating epithelia (Khan et al., 1997; Farrell et al., 1998), rendering them more resistant to the effects of irradiation and/or chemotherapy treatments required to ablate the bone marrow compartment. Indeed, exogenous FGF7 increases the number of S-phase epithelial cells in the regions of intestinal crypts where the stem cell population is thought to reside (Potten et al., 2001). In addition to this protective effect, administration of FGF7 has been shown to directly improve thymic function after bone marrow transplantation (Panoskaltsis-Mortari et al., 1998; Min et al., 2002; Alpdogan et al., 2006) and can partially reverse the loss of thymic function associated with age-related thymic involution (Alpdogan et al., 2006; Min et al., 2007). A detailed analysis of the response of TE to a single course of thee injections of FGF7 to adult mice revealed transient alterations that included the elaboration of chemokines involved in recruitment of progenitor cells to the thymus, stimulated TE proliferation, and elevated expression of molecules that participate in the Wnt and BMP signaling pathways (Rossi et al., 2007b) previously implicated in the development of TE (Balciunaite et al., 2002; Bleul and Boehm, 2005). The increased thymic cellularity observed after transient administration of FGF7 and in lckFGF10 TG mice transplanted with WT bone marrow may actually be the result of similar responses that differ in the magnitude of the progenitor TE pool involved. This would predict an eventual normalization of thymic cellularity in the WT[RIGHTWARDS ARROW]TG chimeras as the differentiation program and representation of TE subsets equilibrate in response to a return to physiological intrathymic levels of FGFR2IIIb ligands.

The proposal here that levels of FGFR2IIIb signaling influence the behavior of TE progenitor cells is consistent with recent demonstrations that p63 plays a central role in maintaining TE progenitor cells (Candi et al., 2007; Senoo et al., 2007). Fetal and/or neonatal mice lacking p63 display a thymic phenotype very similar to that described for the thymus of mice lacking FGFR2IIIb or FGF10 (Revest et al., 2001); although dramatically reduced in size, the TE appears to undergo a normal pattern of differentiation and can support a normal pattern of thymocyte differentiation. The mechanistic basis for this defect appears to be an inability to maintain the progenitor epithelial population, as evidenced by increased apoptosis of TE (Candi et al., 2007; Senoo et al., 2007). The recent demonstration that FGFR2IIIb is a direct target of p63 suggests that the maintenance of the progenitor TE population by the action of p63 is mediated by the FGFR2IIIb pathway (Candi et al., 2007).

This model developed here, although speculative, points to several testable hypotheses that could facilitate a better understanding of TE differentiation. The proposal that FGFR2IIIb signaling can affect the size and activity of TE progenitor cell populations in the postnatal thymus may have important implications in the process of age-related thymic involution, where the thymic epithelial compartment is progressively lost and there is an accompanying reduction in the numbers of newly generated T cells entering the periphery. According to this perspective, age-related thymic involution could be a secondary result of progressive alterations in the mesenchymal–epithelial interactions that ultimately regulate the availability of FGFR2IIIIb ligands in the thymus or TE responses to them.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES

Mice

Coding sequences of human FGF7, FGF10, and a murineFGFR2IIIb-Ig fusion construct (Cheon et al., 1994) were inserted into the lck proximal promoter cassette that efficiently directs transgene expression to thymocytes (Abraham et al., 1991). Transgenic mice were generated by the Transgenic Resources Program operated by the Department of Comparative Medicine at the University of Washington and maintained on a C57Bl/6 background. Adult C57Bl/6 mice were obtained from Charles River (Wilmington, MA) or from our colony. Congenic C57Bl/6 CD45.1 mice were obtained from Dr. Alexander Rudensky. All mice were maintained in the University of Washington Specific Pathogen Free facility and used in accordance with protocols approved by the University of Washington Institutional Animal Care and Use Committee.

Bone Marrow Chimeras

Bone marrow cells were flushed from femurs and tibias of donor mice, washed by centrifugation, and then injected by means of the lateral tail veins into recipient 6- to 8-week-old mice that had previously received 900 rad of Cesium irradiation. Mice were maintained on water containing antibiotics and analyzed 4 weeks after bone marrow transplantation. Mixed bone marrow chimeras used a mixture of WT and TG bone marrow (95%: 5% ratio); using CD45.1 mice as the source of WT bone marrow and as transplant recipients.

Reagents

Primary antibodies for immunohistochemistry included rabbit anti–CC-10 (Cardoso et al., 1993) and the monoclonal anti–E-cadherin antibody (Ab) ECCD2 (Shirayoshi et al., 1986). Other anti-stromal cell Abs include ER-TR series developed by van Vliet et al. (1984), anti-EpCAM (G8.8; Farr et al., 1991; available from the Developmental Studies Hybridoma Bank; http://dshb.biology.uiowa.edu/), anti-DEC205 (NLDC145; Kraal et al., 1986), and CDR1 (Rouse et al., 1988), and F4/80 (Austyn and Gordon, 1981). Polyclonal rabbit anti-K5 and K14 antibodies were purchased from Covance (Princeton, NJ); monoclonal anti-K8 (Troma-1) was obtained from the Developmental Studies Hybridoma Bank (http://dshb.biology.uiowa.edu/). Biotinylated Ulex europaeus agglutinin (UEA), which binds a subset of medullary thymic epithelial cells (Farr and Anderson, 1985), was from Vector Laboratories (Burlingame, CA). Secondary reagents for immunofluorescence microscopy (goat anti-rabbit IgG, goat anti-rat IgG, chicken anti-rat IgG and streptavidin conjugated with Alexa 488, Alexa 546, or Alexa 647) were purchased from Molecular Probes (Carlsbad, CA).

Immunohistochemistry

Fluorescence-based immunohistochemistry was performed on thymi from multiple littermates representing all lines of TG mice and their non-TG littermates as previously described (Dooley et al., 2005a). Images were captured with a monochrome digital CCD camera (Orca-ER, Hamamatsu, Bridgewater, NJ) and assembled into RGB images with Photoshop (Adobe, San Jose, CA).

Flow Cytometry

Expression of CD3, CD4, B220, IgD, and IgM by cells isolated from TG and WT littermates was performed as previously described (Nelson et al., 1998). Antibodies to CD45.1 and CD45.2 were purchased from eBioscience (San Diego, CA).

Vascular Casting

Mice were killed by CO2 asphyxiation and then perfused by means of the left ventricle with Microfil V120 (Flow Tech, Carver, MA) and processed according to the manufacturer's directions. Thymi were cleared in a mixture of methyl and benzoyl benzoate and photographed.

Electron Microscopy

After CO2 euthanasia, mice were perfused by means of the left ventricle with a mixed aldehyde fixative and thymus tissue was processed for conventional electron microscopy as previously described (Farr and Anderson, 1985).

PCR Analyses

Thymocytes were harvested from lckFGF10, lckFGFR2DN, and WT mice. Isolated RNA was treated with DNAse (TurboDNAse, Ambion, Austin, TX) before generation of cDNA. Primer sequences were as follows: Human FGF10 fwd TCACCTTCAAGGAGATGTCCG; Human FGF10 rvs TTCCCCTTCTTGTTCATGGC; Mouse FGFR2IIIb fwd TGTTCAATGTGACGGAGATGG (exon 8); mouse IgG1 heavy chain rvs TGAGCTGTGTGCACCTCCAC (exon 3 of IgG1 heavy chain constant region); HPRT fwd GTTGGATACAGGCCAGACTTTGTTG; HPRT rvs GAGGGTAGGCTGGCCTATGGCT. PCR was carried out in 25- or 50-ml reactions with a final concentration of 10 mM Tris, 2.5 mM MgCl2, 0.2 mM dNTPs, 0.4 mM primers, and 0.625 U of Taq DNA polymerase in buffer B (Promega, Madison WI). Cycling conditions were 94°C for 5 min; cycles of 94°C for 30 sec, 60°C for 30 sec, 72°C for 45 sec, 72°C for 10 min; 4°C hold. No-reverse transcriptase controls were included to demonstrate that reaction products reflected cDNA and were not due to genomic DNA contamination.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. PATTERNS OF KERATIN EXPRESSION BY TE ARE ALTERED IN THE lckFGF10 TG THYMUS
  6. DISCUSSION
  7. EXPERIMENTAL PROCEDURES
  8. REFERENCES
  • Abraham KM, Levin SD, Marth JD, Forbush KA, Perlmutter RM. 1991. Delayed thymocyte development induced by augmented expression of p56lck. J Exp Med 173: 14211432.
  • Allen JM, Forbush KA, Perlmutter RM. 1992. Functional dissection of the lck proximal promoter. Mol Cell Biol 12: 27582768.
  • Alpdogan O, Hubbard VM, Smith OM, Patel N, Lu S, Goldberg GL, Gray DH, Feinman J, Kochman AA, Eng JM, Suh D, Muriglan SJ, Boyd RL, van den Brink MR. 2006. Keratinocyte growth factor (KGF) is required for postnatal thymic regeneration. Blood 107: 24532460.
  • Anderson M, Anderson SK, Farr AG. 2000. Thymic vasculature: organizer of the medullary epithelial compartment? Int Immunol 12: 11051110.
  • Austyn JM, Gordon S. 1981. F4/80, a monoclonal antibody directed specifically against the mouse macrophage. Eur J Immunol 11: 805815.
  • Balciunaite G, Keller MP, Balciunaite E, Piali L, Zuklys S, Mathieu YD, Gill J, Boyd R, Sussman DJ, Hollander GA. 2002. Wnt glycoproteins regulate the expression of FoxN1, the gene defective in nude mice. Nat Immunol 3: 11021108.
  • Bennett AR, Farley A, Blair NF, Gordon J, Sharp L, Blackburn CC. 2002. Identification and characterization of thymic epithelial progenitor cells. Immunity 16: 803814.
  • Bhushan A, Itoh N, Kato S, Thiery JP, Czernichow P, Bellusci S, Scharfmann R. 2001. Fgf10 is essential for maintaining the proliferative capacity of epithelial progenitor cells during early pancreatic organogenesis. Development 128: 51095117.
  • Bleul CC, Boehm T. 2005. BMP signaling is required for normal thymus development. J Immunol 175: 52135221.
  • Bleul CC, Corbeaux T, Reuter A, Fisch P, Monting JS, Boehm T. 2006. Formation of a functional thymus initiated by a postnatal epithelial progenitor cell. Nature 441: 992996.
  • Boehm T, Scheu S, Pfeffer K, Bleul CC. 2003. Thymic medullary epithelial cell differentiation, thymocyte emigration, and the control of autoimmunity require lympho-epithelial cross talk via LTbetaR. J Exp Med 198: 757769.
  • Burkly L, Hession C, Ogata L, Reilly C, Marconi LA, Olson D, Tizard R, Cate R, Lo D. 1995. Expression of relB is required for the development of thymic medulla and dendritic cells. Nature 373: 531536.
  • Candi E, Rufini A, Terrinoni A, Giamboi-Miraglia A, Lena AM, Mantovani R, Knight R, Melino G. 2007. DeltaNp63 regulates thymic development through enhanced expression of FgfR2 and Jag2. Proc Natl Acad Sci U S A 104: 1199912004.
  • Cardoso WV, Stewart LG, Pinkerton KE, Ji C, Hook GE, Singh G, Katyal SL, Thurlbeck WM, Plopper CG. 1993. Secretory product expression during Clara cell differentiation in the rabbit and rat. Am J Physiol 264: L543L552.
  • Celli G, LaRochelle WJ, Mackem S, Sharp R, Merlino G. 1998. Soluble dominant-negative receptor uncovers essential roles for fibroblast growth factors in multi-organ induction and patterning. EMBO J 17: 16421655.
  • Cheon HG, LaRochelle WJ, Bottaro DP, Burgess WH, Aaronson SA. 1994. High-affinity binding sites for related fibroblast growth factor ligands reside within different receptor immunoglobulin-like domains. Proc Natl Acad Sci U S A 91: 989993.
  • De Moerlooze L, Spencer-Dene B, Revest J, Hajihosseini M, Rosewell I, Dickson C. 2000. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 127: 483492.
  • Dooley J, Erickson M, Farr AG. 2005a. An organized medullary epithelial structure in the normal thymus expresses molecules of respiratory epithelium and resembles the epithelial thymic rudiment of nude mice. J Immunol 175: 43314337.
  • Dooley J, Erickson M, Roelink H, Farr AG. 2005b. Nude thymic rudiment lacking functional foxn1 resembles respiratory epithelium. Dev Dyn 233: 16051612.
  • Elghazi L, Cras-Meneur C, Czernichow P, Scharfmann R. 2002. Role for FGFR2IIIb-mediated signals in controlling pancreatic endocrine progenitor cell proliferation. Proc Natl Acad Sci U S A 99: 38843889.
  • Erickson M, Morkowski S, Lehar S, Gillard G, Beers C, Dooley J, Rubin JS, Rudensky A, Farr AG. 2002. Regulation of thymic epithelium by keratinocyte growth factor. Blood 100: 32693278.
  • Farr A, Nelson A, Truex J, Hosier S. 1991. Epithelial heterogeneity in the murine thymus: a cell surface glycoprotein expressed by subcapsular and medullary epithelium. J Histochem Cytochem 39: 645653.
  • Farr AG, Anderson SK. 1985. Epithelial heterogeneity in the murine thymus: fucose-specific lectins bind medullary epithelial cells. J Immunol 134: 29712977.
  • Farrell CL, Bready JV, Rex KL, Chen JN, DiPalma CR, Whitcomb KL, Yin S, Hill DC, Wiemann B, Starnes CO, Havill AM, Lu ZN, Aukerman SL, Pierce GF, Thomason A, Potten CS, Ulich TR, Lacey DL. 1998. Keratinocyte growth factor protects mice from chemotherapy and radiation-induced gastrointestinal injury and mortality. Cancer Res 58: 933939.
  • Gill J, Malin M, Hollander GA, Boyd R. 2002. Generation of a complete thymic microenvironment by MTS24(+) thymic epithelial cells. Nat Immunol 3: 635642.
  • Gillard GO, Farr AG. 2006. Features of medullary thymic epithelium implicate postnatal development in maintaining epithelial heterogeneity and tissue-restricted antigen expression. J Immunol 176: 58155824.
  • Gillard GO, Dooley J, Erickson M, Peltonen L, Farr AG. 2007. Aire-dependent alterations in medullary thymic epithelium indicate a role for Aire in thymic epithelial differentiation. J Immunol 178: 30073015.
  • Gray DH, Seach N, Ueno T, Milton MK, Liston A, Lew AM, Goodnow CC, Boyd RL. 2006. Developmental kinetics, turnover, and stimulatory capacity of thymic epithelial cells. Blood 108: 37773785.
  • Gray DH, Tull D, Ueno T, Seach N, Classon BJ, Chidgey A, McConville MJ, Boyd RL. 2007. A unique thymic fibroblast population revealed by the monoclonal antibody MTS-15. J Immunol 178: 49564965.
  • Guo L, Yu QC, Fuchs E. 1993. Targeting expression of keratinocyte growth factor to keratinocytes elicits striking changes in epithelial differentiation in transgenic mice. EMBO J 12: 973986.
  • Guo L, Degenstein L, Fuchs E. 1996. Keratinocyte growth factor is required for hair development but not for wound healing. Genes Dev 10: 165175.
  • Hart A, Papadopoulou S, Edlund H. 2003. Fgf10 maintains notch activation, stimulates proliferation, and blocks differentiation of pancreatic epithelial cells. Dev Dyn 228: 185193.
  • Hetzer-Egger C, Schorpp M, Haas-Assenbaum A, Balling R, Peters H, Boehm T. 2002. Thymopoiesis requires Pax9 function in thymic epithelial cells. Eur J Immunol 32: 11751181.
  • Hollander GA, Wang B, Nichogiannopoulou A, Platenburg PP, van Ewijk W, Burakoff SJ, Gutierrez-Ramos JC, Terhorst C. 1995. Developmental control point in induction of thymic cortex regulated by a subpopulation of prothymocytes. Nature 373: 350353.
  • Jenkinson WE, Jenkinson EJ, Anderson G. 2003. Differential requirement for mesenchyme in the proliferation and maturation of thymic epithelial progenitors. J Exp Med 198: 325332.
  • Khan WB, Shui C, Ning S, Knox SJ. 1997. Enhancement of murine intestinal stem cell survival after irradiation by keratinocyte growth factor. Radiat Res 148: 248253.
  • Klug DB, Carter C, Crouch E, Roop D, Conti CJ, Richie ER. 1998. Interdependence of cortical thymic epithelial cell differentiation and T-lineage commitment. Proc Natl Acad Sci U S A 95: 1182211827.
  • Klug DB, Carter C, Gimenez-Conti IB, Richie ER. 2002. Cutting edge: thymocyte-independent and thymocyte-dependent phases of epithelial patterning in the fetal thymus. J Immunol 169: 28422845.
  • Kraal G, Breel M, Janse M, Bruin G. 1986. Langerhans' cells, veiled cells, and interdigitating cells in the mouse recognized by a monoclonal antibody. J Exp Med 163: 981997.
  • Lu J, Qian J, Keppler D, Cardoso WV. 2007. Cathespin H is an FGF10 target involved in BMP4 degradation during lung branching morphogenesis. J Biol Chem 282: 2217622184.
  • Manley NR, Capecchi MR. 1998. Hox group 3 paralogs regulate the development and migration of the thymus, thyroid, and parathyroid glands. Dev Biol 195: 115.
  • Metlay JP, Witmer-Pack MD, Agger R, Crowley MT, Lawless D, Steinman RM. 1990. The distinct leukocyte integrins of mouse spleen dendritic cells as identified with new hamster monoclonal antibodies. J Exp Med 171: 17531771.
  • Min D, Taylor PA, Panoskaltsis-Mortari A, Chung B, Danilenko DM, Farrell C, Lacey DL, Blazar BR, Weinberg KI. 2002. Protection from thymic epithelial cell injury by keratinocyte growth factor: a new approach to improve thymic and peripheral T-cell reconstitution after bone marrow transplantation. Blood 99: 45924600.
  • Min D, Panoskaltsis-Mortari A, Kuro OM, Hollander GA, Blazar BR, Weinberg KI. 2007. Sustained thymopoiesis and improvement in functional immunity induced by exogenous KGF administration in murine models of aging. Blood 109: 25292537.
  • Nehls M, Kyewski B, Messerle M, Waldschutz R, Schuddekopf K, Smith AJ, Boehm T. 1996. Two genetically separable steps in the differentiation of thymic epithelium. Science 272: 886889.
  • Nelson AJ, Clegg CH, Farr AG. 1998. In vitro positive selection and anergy induction of class II-restricted TCR transgenic thymocytes by a cortical thymic epithelial cell line. Int Immunol 10: 13351346.
  • Norgaard GA, Jensen JN, Jensen J. 2003. FGF10 signaling maintains the pancreatic progenitor cell state revealing a novel role of Notch in organ development. Dev Biol 264: 323338.
  • Panoskaltsis-Mortari A, Lacey DL, Vallera DA, Blazar BR. 1998. Keratinocyte growth factor administered before conditioning ameliorates graft-versus-host disease after allogeneic bone marrow transplantation in mice. Blood 92: 39603967.
  • Potten CS, O'Shea JA, Farrell CL, Rex K, Booth C. 2001. The effects of repeated doses of keratinocyte growth factor on cell proliferation in the cellular hierarchy of the crypts of the murine small intestine. Cell Growth Differ 12: 265275.
  • Powers CJ, McLeskey SW, Wellstein A. 2000. Fibroblast growth factors, their receptors and signaling. Endocr Relat Cancer 7: 165197.
  • Raviola E, Karnovsky MJ. 1972. Evidence for a blood-thymus barrier using electron-opaque tracers. J Exp Med 136: 466498.
  • Revest JM, Suniara RK, Kerr K, Owen JJ, Dickson C. 2001. Development of the thymus requires signaling through the fibroblast growth factor receptor R2-IIIb. J Immunol 167: 19541961.
  • Rodewald HR, Paul S, Haller C, Bluethmann H, Blum C. 2001. Thymus medulla consisting of epithelial islets each derived from a single progenitor. Nature 414: 763768.
  • Röpke C, Van Soest P, Platenburg PP, Van Ewijk W. 1995. A common stem cell for murine cortical and medullary thymic epithelial cells? Dev Immunol 4: 149156.
  • Rossi SW, Jenkinson WE, Anderson G, Jenkinson EJ. 2006. Clonal analysis reveals a common progenitor for thymic cortical and medullary epithelium. Nature 441: 988991.
  • Rossi SW, Chidgey AP, Parnell SM, Jenkinson WE, Scott HS, Boyd RL, Jenkinson EJ, Anderson G. 2007a. Redefining epithelial progenitor potential in the developing thymus. Eur J Immunol 37: 24112418.
  • Rossi SW, Jeker LT, Ueno T, Kuse S, Keller MP, Zuklys S, Gudkov AV, Takahama Y, Krenger W, Blazar BR, Hollander GA. 2007b. Keratinocyte growth factor (KGF) enhances postnatal T-cell development via enhancements in proliferation and function of thymic epithelial cells. Blood 109: 38033811.
  • Rossi SW, Kim MY, Leibbrandt A, Parnell SM, Jenkinson WE, Glanville SH, McConnell FM, Scott HS, Penninger JM, Jenkinson EJ, Lane PJ, Anderson G. 2007c. RANK signals from CD4+3- inducer cells regulate development of Aire-expressing epithelial cells in the thymic medulla. J Exp Med 204: 12671272.
  • Rouse RV, Bolin LM, Bender JR, Kyewski BA. 1988. Monoclonal antibodies reactive with subsets of mouse and human thymic epithelial cells. J Histochem Cytochem 36: 15111517.
  • Sakaue H, Konishi M, Ogawa W, Asaki T, Mori T, Yamasaki M, Takata M, Ueno H, Kato S, Kasuga M, Itoh N. 2002. Requirement of fibroblast growth factor 10 in development of white adipose tissue. Genes Dev 16: 908912.
  • Senoo M, Pinto F, Crum CP, McKeon F. 2007. p63 Is essential for the proliferative potential of stem cells in stratified epithelia. Cell 129: 523536.
  • Shimizu C, Kawamoto H, Yamashita M, Kimura M, Kondou E, Kaneko Y, Okada S, Tokuhisa T, Yokoyama M, Taniguchi M, Katsura Y, Nakayama T. 2001. Progression of T cell lineage restriction in the earliest subpopulation of murine adult thymus visualized by the expression of lck proximal promoter activity. Int Immunol 13: 105117.
  • Shirayoshi Y, Nose A, Iwasaki K, Takeichi M. 1986. N-linked oligosaccharides are not involved in the function of a cell-cell binding glycoprotein E-cadherin. Cell Struct Funct 11: 245252.
  • Umemori H, Linhoff MW, Ornitz DM, Sanes JR. 2004. FGF22 and its close relatives are presynaptic organizing molecules in the mammalian brain. Cell 118: 257270.
  • Van Vliet E, Melis M, Van Ewijk W. 1984. Monoclonal antibodies to stromal cell types of the mouse thymus. Eur J Immunol 14: 524529.
  • Wallin J, Eibel H, Neubuser A, Wilting J, Koseki H, Balling R. 1996. Pax1 is expressed during development of the thymus epithelium and is required for normal T-cell maturation. Development 122: 2330.
  • Weaver M, Dunn NR, Hogan BL. 2000. Bmp4 and Fgf10 play opposing roles during lung bud morphogenesis. Development 127: 26952704.
  • Yang SJ, Ahn S, Park CS, Holmes KL, Westrup J, Chang CH, Kim MG. 2006. The quantitative assessment of MHC II on thymic epithelium: implications in cortical thymocyte development. Int Immunol 18: 729739.
  • Zou D, Silvius D, Davenport J, Grifone R, Maire P, Xu PX. 2006. Patterning of the third pharyngeal pouch into thymus/parathyroid by Six and Eya1. Dev Biol 293: 499512.