HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Su, D.-m. Articles by Manley, N. R. Search for Related Content PubMed PubMed Citation Articles by Su, D.-m. Articles by Manley, N. R.

Hoxa3 and Pax1 Transcription Factors Regulate the Ability of Fetal Thymic Epithelial Cells to Promote Thymocyte Development¹

Institute of Molecular Medicine and Genetics and Department of Pediatrics, Medical College of Georgia, Augusta, GA 30912

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Thymocyte maturation into T cells depends on interactions between thymocytes and thymic epithelial cells. In this study, we show that mutations in two transcription factors, Hoxa3 and Pax1, act synergistically to cause defective thymic epithelial cell development, resulting in thymic ectopia and hypoplasia. Hoxa3^+/-Pax1^-/- compound mutant mice exhibited more severe thymus defects than Pax1^-/- single mutants. Fetal liver adoptive transfer experiments revealed that the defect resided in radio-resistant stromal cells and not in hematopoietic cells. Compound mutants have fewer MHC class II⁺ epithelial cells, and the level of MHC expression detected was lower. Thymic epithelial cells in these mutants have reduced ability to promote thymocyte development, causing a specific block in thymocyte maturation at an early stage that resulted in a dramatic reduction in the number of CD4⁺8⁺ thymocytes. This phenotype was accompanied by increased apoptosis of CD4⁺8⁺ thymocytes and their immediate precursors, CD44^-25^-(CD3^-4^-8^-) cells. Our results identify a transcriptional regulatory pathway required for thymic epithelial cell development and define multiple roles for epithelial cell regulation of thymocyte maturation at the CD4^-8^- to CD4⁺8⁺ transition.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The vertebrate thymus is responsible for the production of self-restricted, self-tolerant T cells. Within the mature thymus, T cell precursors (thymocytes) undergo a series of developmental stages via interactions with thymic stromal cells, resulting in the generation of mature T lymphocytes that are then exported into the periphery. Epithelial cells constitute a major component of the thymic stroma and are required both for thymus organogenesis and for thymocyte maturation (1, 2, 3, 4, 5, 6, 7). Thymocyte development is characterized by the sequential expression of cell surface markers at particular stages of maturation. The main stages are CD4^-8^- (double negative (DN)³) to CD4⁺8⁺ (double positive (DP)) to mature CD4⁺ or CD8⁺ (single positive (SP)) cells. Interactions with thymic epithelial cells are required for both the DN to DP and DP to SP transitions (1, 2, 4, 8, 9). Thymic epithelial cells play a variety of roles in promoting thymocyte development, from providing maturation signals such as cytokines, chemokines, adhesion molecules, and matrix components to promoting MHC-mediated positive selection. However, the molecular mechanisms regulating thymic epithelial cell differentiation and function are poorly understood.

Although the role of epithelial cells in positive selection at the DP to SP transition has been studied extensively, their role in the DN to DP transition is less clear. Thymocyte development stages within the DN population are defined by their expression of the CD44 and CD25 cell surface markers, in stages progressing from CD44⁺25^- to CD44⁺25⁺, to CD44^-25⁺, and to CD44^-25^- (10). Many if not all of these stages require epithelial cell interactions to proliferate and/or differentiate to the next stage. MHC class II-positive epithelial cells are required for thymocytes to develop from DN to DP stages in reaggregate cultures (2) and for induction of ß and TCR rearrangements (11). Epithelial cells are known to provide a number of cytokines that are necessary both for early expansion of DN cells and at later thymocyte development stages (12, 13). However, for the most part the signals provided by epithelial cells to promote specific stages of early thymocyte differentiation and proliferation have not been defined.

Although the central role of epithelial cells in thymic development and function is widely acknowledged, little is known about the molecular pathways by which these cells become competent to perform these functions. Mutations in several transcription factors, including relB, whn, Pax9, and Hoxa3, have been shown to cause severe defects in thymus organogenesis and thymic epithelial cell development, leading to the absence of a functional thymus or deletion of a specific epithelial cell type (14, 15, 16, 17). Although all of these transcription factors have profound effects on thymic epithelial cell development, the severity of their phenotypes has made it difficult to use these mutants to investigate epithelial cell function. In contrast to these more severe phenotypes, mutation of the Pax1 gene has more subtle effects on thymus development and function. Pax1 mutants have a hypoplastic thymus that is deficient in thymocyte development (18, 19). Pax1 is expressed in the third pharyngeal pouch endoderm where thymus organogenesis originates, in a subset of thymic epithelial cells throughout fetal development, and in the adult thymus (19). Our previous studies have suggested that there is a functional link between Hoxa3 and Pax1 in thymus development (20). Hoxa3 is a member of the Hox family of transcription factors, which specify positional identity in the developing embryo (21). Mutation of the Hoxa3 gene in mice results in athymia, as part of a spectrum of pharyngeal region defects (17, 20, 22, 23, 24). Hoxa3 is expressed in both the third pharyngeal pouch endoderm and the neural crest mesenchyme, which contribute to early thymus organogenesis and may act in both of these cell types (6, 20, 24, 25, 26). Pax1 is specifically down-regulated in the third pharyngeal pouch endoderm in Hoxa3^-/- embryos at embryonic day 10.5 (E10.5) (20). This result indicates that Hoxa3 and Pax1 may be in a common pathway regulating thymus development.

To investigate this possible link between Hoxa3 and Pax1, we generated mice carrying mutations in both genes. To detect an interaction between these genes, we looked for the ability of the Hoxa3 mutation to enhance the Pax1^un-ex thymus phenotype in double heterozygotes (Hoxa3^+/-Pax1^+/un-ex) and in Hoxa3^+/-Pax1^un-ex/un-ex compound mutants. The thymus phenotype of Hoxa3^+/-Pax1^+/un-ex double heterozygotes was indistinguishable from that of Pax1^un-ex/un-ex single mutants, indicating that mutations in these genes show nonallelic noncomplementation. In addition, the Hoxa3^+/-Pax1^un-ex/un-ex compound mutant mice had a strikingly more severe thymus phenotype than Pax1^un-ex/un-ex single mutants did, displaying ectopic thymus lobes, increased thymic hypoplasia, and more severe effects on thymocyte maturation. Hoxa3^+/-Pax1^un-ex/un-ex compound mutants have fewer MHC class II⁺ thymic epithelial cells and reduced levels of epithelial I-A^b expression. Thymocyte maturation defects in these compound mutants included a 10-fold reduction in CD4⁺8⁺ thymocytes. Cell death analysis indicated that this decrease is associated with increased apoptosis in both CD44^-25^- and DP cells. This analysis also identified an apparent effect of the Pax1 mutation on specific subsets of triple negative (TN) cells. Our results may define multiple roles for epithelial cell regulation of thymocyte maturation at the DN to DP transition.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Strains and genotyping

The Hoxa3 mutant strain and genotyping by PCR have been described (17, 20). Pax1^un-ex mice and embryos were genotyped by Southern blot with the pSacII-3 probe, as described (18), or by PCR. PCR primers were designed by sequence analysis of the un-ex deletion breakpoint, which was cloned from a size-selected genomic library probed with the pSacII-3 probe (D.-m.S., unpublished data). The sequences of the oligos were as follows: common forward primer, 5'-GCCACCCAAAGGTCTCTGTGC-3'; wild-type reverse primer, 5'-CCACCAAAGGAGAGGGAATAGGAAGG-3'; and un-ex specific reverse primer, 5'-GGGTAGTTCATAGAGATCAAACCCC-3'. The wild-type allele yields a 300-bp product, and the un-ex allele gives a 430-bp product. Both the Hoxa3 and Pax1^un-ex strains are congenic on the C57BL/6J genetic background. Compound heterozygotes were viable and fertile. Noon of the day of vaginal plug was considered E0.5; embryos were also staged using somite number and morphological criteria. All animal use procedures were approved by the Medical College of Georgia Committee on Animal Use for Research and Education.

Histology

Embryos were collected at E17.5, fixed in 4% formaldehyde, and processed for paraffin embedding using standard techniques. Sections were cut at 6 µm and stained with hematoxylin and eosin.

Flow cytometry and Abs

Thymocytes were collected by filtering through a cell strainer. Three-color flow cytometry was performed using the following mAbs (PharMingen, San Diego, CA): biotin or PE anti-mouse CD3 (clone 145-2C11), R-PE anti-mouse CD4 (clone RM4-5), FITC or PE anti-mouse CD8a (clone 53-6.7), biotin anti-mouse CD69 (clone H1.2F3), biotin anti-mouse ßTCR (clone H57-597), biotin anti-mouse TCR (clone GL-3), biotin anti-mouse CD44 (clone Pgp-1), and FITC anti-mouse CD25 (clone 7D4). Biotinylated Abs were visualized using streptavidin-conjugated Cy-chrom (PharMingen). Data generation and analyses were performed using a FACStar^Plus flow cytometer. All data was analyzed using CellQuest software. Significant differences were determined by the unpaired Student’s t test or ANOVA analysis.

Adoptive transfer

Adoptive transfers of fetal hematopoietic stem cells were performed as previously described (27). Briefly, 5- to 8-wk-old Rag1^-/- mice were lethally irradiated and injected with 2.5 x 10⁷ E15.5 fetal liver cells from control, Pax1^un-ex/un-ex, and Hoxa3^+/-Pax1^un-ex/un-ex embryos. After 2 mo, the thymus, spleen, and peripheral lymph nodes were analyzed by flow cytometry for CD4, CD8, and CD3 expression as described above.

TUNEL analysis

TUNEL analysis was modified from the previously described method (28, 29). Briefly, for three-color analysis of CD4 and CD8 populations, E17.5 thymocytes were collected and stained with PE-CD4 and FITC-CD8. Cells were then fixed in 2% PFA for 1 h, permeabilized with 0.1% Triton X-100, and equilibrated in 1x TdT buffer. Terminal transferase reactions were conducted in 50-µl reactions containing 1x TdT buffer, 5 µM biotin-dUTP, and 100 U/ml TdT (Boehringer Mannheim, Indianapolis, IN). TUNEL⁺ cells were identified with streptavidin-peridinin chlorophyl protein (PharMingen). Four-color analysis of TUNEL⁺ cells in TN subsets was performed using cell surface staining with PE-CD4, PE-CD8, PE-CD3, APC-CD44, and FITC-CD25. Cells were then fixed, permeabilized, and the TdT reaction and streptavidin staining conducted as described above. Results were analyzed by gating on cells negative for staining with PE-CD4, PE-CD8, and PE-CD3 and measuring the percentage of TUNEL⁺ cells within each quadrant defined by labeling with APC-CD44 and FITC-CD25 as described above.

Fetal thymic organ culture and epithelial cell staining

Thymic lobes from E15.5 embryos were isolated and cultured at 37°C in high oxygen submersion culture as described previously (30, 31). Lobes were cultured in 1 ml RPMI 1640 medium with 10% FBS and 5 x 10^-5 M 2-ME containing 1.35 µM deoxyguanosine (dGuo) for 5 days to deplete thymocytes and dendritic cells (32, 33). After depletion, thymic lobes were washed twice with PBS and then were incubated with 0.25% Trypsin/0.02% EDTA for 30 min at 37°C. Cells were dissociated by pipetting, washed and counted, stained using the anti-I-A^b mAb M5/114 and a FITC-conjugated anti-rat IgG secondary Ab (PharMingen), and analyzed by flow cytometry. The M5/114 hybridoma cell line was obtained from the American Type Culture Collection (Manassas, VA); culture supernatant was purified using a protein G column. Ab staining was analyzed by flow cytometry as described above.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Hoxa3 enhances the Pax1 mutant phenotype

As a genetic test of whether Hoxa3 and Pax1 are in the same pathway, we investigated whether mutation of one copy of Hoxa3 could enhance the Pax1 mutant phenotype. At E17.5, control mice had two large thymic lobes located directly cranial to the heart (Fig. 1a). Pax1^un-ex/un-ex mice were as previously described (18, 19), with somewhat smaller thymic lobes that often contained cysts (Fig. 1b). Thymic lobes from Hoxa3^+/-Pax1^+/un-ex and Pax1^un-ex/un-ex mice were indistinguishable in size and morphology (Fig. 1, b and c), indicating that these two mutations do show a genetic interaction. Even more strikingly, Hoxa3^+/-Pax1^un-ex/un-ex mice at E17.5 had a dramatically more severe thymus phenotype than Pax1^un-ex/un-ex single mutants. Histological analysis showed that all five Hoxa3^+/-Pax1^un-ex/un-ex mice examined had very small thymic lobes that were often ectopically located anterior to their normal position (Fig. 1, d–f). Ectopic lobes were never seen in wild-type, single heterozygote Hoxa3^+/-Pax1^+/un-ex or Pax1^un-ex/un-ex animals. Both ectopic and normally placed Hoxa3^+/-Pax1^un-ex/un-ex thymic lobes at E17.5 had numerous cysts that were often very large relative to the size of the thymic lobe (Fig. 1, d and e). These cysts were morphologically similar to those seen in Pax1^un-ex/un-ex thymic lobes. All mutant and control thymuses had identifiable cortical and medullary development. However, in the Hoxa3^+/-Pax1^un-ex/un-ex mutants the density of cortical thymocytes was noticeably reduced (Fig. 1, g and h). These results indicated a surprisingly strong Hoxa3-Pax1 genetic interaction.

View larger version (144K): [in this window] [in a new window]   FIGURE 1. Histological analysis of E17.5 Hoxa3+/-Pax1un-ex/un-ex fetal thymus. Transverse sections of formaldehyde-fixed, paraffin-embedded embryos stained with hematoxylin and eosin. a, Control Hoxa3+/- thymus (th) has two large lobes located in the anterior thoracic cavity. b, Pax1un-ex/un-ex thymus is smaller and somewhat cystic. c, Hoxa3+/-Pax1+/un-ex thymus is similar to the Pax1un-ex/un-ex phenotype. d, Hoxa3+/-Pax1un-ex/un-ex embryo with bilateral ectopic thymic lobes. e, Hoxa3+/-Pax1un-ex/un-ex embryo with a single small thymic lobe (th) in the correct location, with two large cysts. f, Ectopic Hoxa3+/-Pax1un-ex/un-ex thymus lobe (th) with two large cysts, located lateral and dorsal to the hyoid bone. g, Higher magnification of control thymus showing light staining medullary region (me) and darker-staining cortical region (co) densely populated with lymphocytes. h, Higher magnification of Hoxa3+/-Pax1un-ex/un-ex mutant thymus also showing cortical (co) and medullary (me) differentiation but with lower density of cortical thymocytes. Scale bar: a–d, 750 µm; g and h, 50 µm.

The reduced thymocyte density and smaller thymus size suggested that there could be defects in thymocyte maturation in Hoxa3^+/-Pax1^un-ex/un-ex compound mutants. In addition, a previous study presented a preliminary analysis of thymocyte maturation in Pax1^un-ex mutants that suggested decreased total thymocyte cell number and DP cells at late fetal stages (19). We assayed defects in epithelial cell function by changes in thymocyte cell number and/or abnormal accumulations of cells in one of the more immature stages of thymocyte maturation using flow cytometric analysis. We compared the CD4 and CD8 expression of E17.5 fetal thymocytes from Pax1^un-ex/un-ex, Hoxa3^+/-Pax1^+/un-ex, and Hoxa3^+/-Pax1^un-ex/un-ex mice (Fig. 2). Hoxa3 and Pax1 single heterozygotes were used as controls. Hoxa3^+/-Pax1^+/un-ex double heterozygotes had the same CD4 and CD8 expression profile as the Pax1^un-ex/un-ex single mutants (Fig. 2, a and b). Both double heterozygotes and Pax1^un-ex/un-ex mutants had significantly fewer total thymocytes than did controls (p < 0.001). This reduction was due to a specific decrease in the number of CD4⁺CD8⁺ cells, as the number of more immature CD4^-CD8^- cells was the same as it was in controls. Hoxa3^+/-Pax1^un-ex/un-ex mice had even more severe reductions in total thymocyte cell number and the CD4⁺CD8⁺ compartment than did controls or Pax1 mutants. Cell number was reduced to 25–30% of controls, caused by a 10-fold reduction in CD4⁺CD8⁺ cell number (5- to 6-fold reduction compared with that of Pax1^un-ex/un-ex). Although the number of CD4^-CD8^- cells in Hoxa3^+/-Pax1^un-ex/un-ex mutants was consistently lower than in Pax1^un-ex mutants, this decrease was not significant (p > 0.05). The percentage of CD4⁺CD8⁺ cells was significantly decreased, with a compensatory increase in the percentage of CD4^-CD8^- cells (Fig. 2b).

View larger version (37K): [in this window] [in a new window]   FIGURE 2. Flow cytometry analysis of control and mutant fetal thymocytes at E17.5. Values are expressed as mean ± SEM. Legend for a and b is shown with a. a, Decreases in total cell number and DP cells relative to controls for single and double mutant combinations. b, DN and DP cell populations expressed as a percent of total thymocytes in E17.5 fetal thymus. c, Flow cytometry of fetal thymocytes from representative control, Pax1un-ex/un-ex, and Hoxa3+/-Pax1un-ex/un-ex at E17.5, stained for expression of CD4 and CD8. Values shown are percentage of total cell number; note the percent of DP cells in the upper right quadrants.

We investigated whether the expression of TCR had been affected in these mutants (Table I). Developing thymocytes rearrange and express either TCR or ßTCR, with the ßTCR⁺ lineage representing the majority of thymocytes during late fetal and adult stages. Expression of ßTCR is coincident with development from DN to DP cells. Although the total number of cells expressing ßTCR was reduced significantly in Hoxa3^+/-Pax1^un-ex/un-ex mutants, the percentage of ßTCR⁺ cells within the combined DP and SP cell populations was unchanged at E17.5 relative to controls (Table I; p > 0.05). The percentage of TCR⁺ cells among CD4^-8^- cells was also not significantly different (Table I; p > 0.05). To further investigate any effects on T cell selection, we used CD69 as a marker for CD4⁺8⁺ cells undergoing positive selection. A decrease in CD4⁺8⁺69⁺ cells was previously reported for the Pax1 undulated shorttail allele (19). Again, although the total number of CD4⁺8⁺69⁺ cells was significantly decreased in Hoxa3^+/-Pax1^un-ex/un-ex mice, the percentage of DP and SP cells expressing CD69 was the same in controls and mutants (Table I; p = 0.87). Taken together, these results suggest that there is a defect in the generation and/or proliferation of CD4⁺8⁺ cells; however, those CD4⁺8⁺ cells that are produced can undergo positive selection and generate mature SP cells. In support of this conclusion, although the numbers of mature CD4⁺ and CD8⁺ SP thymocytes were significantly reduced in these mutants, the percentages of these SP cells were not significantly decreased (data not shown).

It is well-documented that hematopoietic cells in the thymus play an important role in early thymus organogenesis (5). Although neither Hoxa3 nor Pax1 has been shown to be expressed in hematopoietic cells, other Hox genes, including the closely related gene Hoxb3, are expressed in hematopoietic stem cells and can affect hematopoiesis when overexpressed (34). To exclude any contribution to the observed phenotype by hematopoietic cell defects, we irradiated 5- to 8-wk-old Rag1^-/- mice and reconstituted their immune systems with fetal liver cells from E15.5 control and Hoxa3^+/-Pax1^un-ex/un-ex embryos. After 2 mo, the thymus, spleen, and peripheral lymph nodes were analyzed by flow cytometry for CD3, CD4, and CD8 expression. Our analysis showed no difference in the ability of control and Hoxa3^+/-Pax1^un-ex/un-ex fetal liver cells to reconstitute the thymus and peripheral lymphoid organs (Fig. 3). Total thymocyte cell numbers in the reconstituted thymic lobes were similar to normal adult cell numbers (data not shown). These results showed that the thymocyte maturation defects seen were not due to intrinsic defects in hematopoietic cells.

View larger version (55K): [in this window] [in a new window]   FIGURE 3. Flow cytometry analysis of thymocytes and splenocytes from irradiated Rag1-/- host mice reconstituted with E15.5 fetal liver cells. The genotypes of the donor embryos are listed to the left. No differences between control and mutant donor cells were seen. Values shown are percentage of total cell number in each quadrant.

To better define the thymocyte differentiation block at the DN to DP transition, we investigated DN cell development by assaying CD44 and CD25 expression in CD3^-4^-8^- TN cells (Refs. 10, 35 and Fig. 4). No dramatic changes in the percentage of cells in each subset were seen. However, small (<2x) but significant changes in the distribution of thymocytes were seen in both Pax1^un-ex/un-ex and Hoxa3^+/-Pax1^un-ex/un-ex mice. The percentage of CD44⁺25⁺ cells was reduced by 40% relative to single heterozygote controls (p < 0.01), whereas the percentage of CD44^-25⁺ cells was increased by 25% (p < 0.001). The percentage of CD44^-25^- cells was not significantly different. Because these differences were seen in both Pax1^un-ex/un-ex and Hoxa3^+/-Pax1^un-ex/un-ex mutants, these changes are likely due to the Pax1 mutation alone and cannot account for the exacerbated decrease in DP cell numbers seen in the compound mutants.

View larger version (31K): [in this window] [in a new window]   FIGURE 4. CD44 and CD25 expression in Pax1un-ex/un-ex and Hoxa3+/-Pax1un-ex/un-ex mutants at E17.5. Graph of compiled data from multiple litters; n values are shown on legend. Pax1-/- and Hoxa3+/-Pax1-/- average values in the CD44+25+ and CD44-25+ subpopulations were significantly different from other genotypes but were not different from each other (*).

View larger version (24K): [in this window] [in a new window]   FIGURE 5. Increased thymocyte apoptosis in Hoxa3+/-Pax1un-ex/un-ex compound mutants. Each data point represents data from an individual mouse; bars are average values. a, Percentage of TUNEL-positive thymocytes in CD4-8- and CD4+8+ populations. In both populations, Hoxa3+/-Pax1un-ex/un-ex average values were significantly higher than in control mice (p < 0.05). Values for Pax1un-ex/un-ex mutants were not significantly different from controls. b, Percentage of TUNEL-positive cells in CD3-4-8- subpopulations defined by expression of CD44 and CD25. Average values in CD44+25+ and CD44-25+ subpopulations were not significantly different between genotypes. Hoxa3+/-Pax1un-ex/un-ex compound mutants had significantly higher levels of TUNEL+ cells in the CD44-25- subpopulation.

As an initial assay for thymic epithelial cell differentiation, we cultured E15.5 thymic lobes in the presence of dGuo for 5 days to deplete thymocytes and dendritic cells (32, 33) and then measured expression of MHC class II (anti-I-A^b). Epithelial cells normally begin expression of MHC class II at E13–14; this expression is independent of interactions with thymocytes (36, 37). Interestingly, thymic lobes from double heterozygote and Pax1^un-ex/un-ex embryos were similar to controls in both epithelial cell number and MHC class II expression after the culture period (Fig. 6). In contrast, the total number of epithelial cells present in these depleted lobes was significantly decreased in the Hoxa3^+/-Pax1^un-ex/un-ex thymic lobes (Fig. 6). Hoxa3^+/-Pax1^un-ex/un-ex epithelial cells also had a significantly lower number and percentage of MHC class II⁺ cells than any other genotype class (Fig. 6, b and c). In addition, the level of MHC class II expression as measured by fluorescence intensity was reduced in the compound mutants (Fig. 6d). These results indicate a new defect in epithelial cell differentiation in the Hoxa3^+/-Pax1^un-ex/un-ex compound mutants compared with that of Pax1 single mutants.

View larger version (50K): [in this window] [in a new window]   FIGURE 6. Hoxa3+/-Pax1-/- epithelial cells at E15.5 have reduced MHC class II expression. E15.5 thymic lobes were cultured in the presence of dGuo for 5 days, dissociated, and assayed for the expression of MHC class II (I-Ab). a, The total number of epithelial cells was significantly decreased in the depleted lobes from Hoxa3+/-Pax1-/- mutants (*). Hoxa3+/-Pax1-/- epithelial cells also had a significantly lower number (b) and percentage (c) of MHC class II+ cells than any other genotype class (*). d, Histograms showing levels of MHC class II expression in representative cultures of selected genotypes. Dashed vertical line shows the peak of fluorescence intensity in control and Pax1un-ex/un-ex cells. Filled bar (*) shows shift in fluorescence intensity of MHC class II staining in Hoxa3+/-Pax1un-ex/un-ex cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our analysis also identified defects within the CD3^-4^-8^- TN cell compartment that appear to be specific to the Pax1 mutation alone. In both Pax1^un-ex/un-ex and Hoxa3^+/-Pax1^un-ex/un-ex mutants, the percentage of CD44⁺25⁺ cells was reduced and the percentage of CD44^-25⁺ cells was increased relative to controls. These changes in the TN compartment are identical in Pax1 single mutants and Hoxa3-Pax1 compound mutants, indicating that these defects are due only to the loss of Pax1. During the CD44^-25⁺ stage, ßTCR lineage thymocytes are selected for their ability to express a functionally rearranged TCRß molecule and to assemble with the pre-T molecule to form the pre-TCR complex (43, 44). The Pax1 phenotype is similar in some respects to previously described mutations in components of the pre-TCR complex, which cause partial or complete blocks in thymocyte maturation at the CD44^-25⁺ stage (45). Similarities include decreases in total thymocyte number and in the number and percentage of DP cells as well as increases in the percentages of DN cells and of CD44^-25⁺ cells in the TN compartment and specific defects in the ßTCR lineage but not in TCR⁺ cells. Despite these similarities, there are also differences indicating that the Pax1 mutant phenotype may not be explained entirely by a partial block at the CD44^-25⁺ stage. The magnitude of the differences seen is relatively small (<2x), and there is no change in the percentage of cells in the next stage, CD44^-25^- pre-DP cells. Pre-DP cells are strikingly reduced in mutants directly affecting pre-TCR expression (46), although it should be noted that not all mutants that affect pre-TCR signaling show a decrease in CD44^-25^- cells (47). The role of epithelial cells at this step has not been defined, and a specific ligand for the pre-TCR is unknown (48). However, because thymocytes cannot mature beyond this step in suspension culture, some as yet undefined interaction or signal is required to proceed beyond this stage (40), possibly involving cortical epithelial cells. Our results raise the intriguing possibility that there is an interaction between thymic epithelial cells and developing thymocytes at the CD44^-25⁺ stage that is disrupted in these mutants.

Some candidates for epithelial cell-produced factors that could mediate the observed phenotypes include cytokines and glucocorticoids. Thymic epithelial cells express a number of cytokines, beginning as early as E12.5 (49, 50). Thymocytes at different stages of development have the capacity to respond to different cytokines, and specific combinations of cytokines can have additional effects beyond those of individual factors (12, 13, 35, 50, 51). Although the null phenotypes for a number of cytokines have been described, none match the Hoxa3^+/-Pax1^un-ex/un-ex phenotype (12, 13). It should be emphasized that multiple cytokines could be reduced or absent in a specific subpopulation of epithelial cells, so the null phenotypes for the individual candidates could be different from the Hoxa3-Pax1 compound mutants. Another candidate is a decrease in glucocorticoid synthesis. Inhibition of corticosterone synthesis by thymic epithelial cells or of glucocorticoid receptor expression in thymocytes results in a phenotype that is strikingly similar to the Hoxa3^+/-Pax1^-/- phenotype, with a 95% decrease in DP cell number and no effect on DN cell number or CD44/CD25 subpopulations (52, 53). These effects were proposed to be caused both by a defect at the DN to DP transition and increased DP cell death due to increased negative selection (53). We are currently investigating possible changes in both cytokines and glucocorticoids in Hoxa3-Pax1 mutants.

Our results suggest that there are both fewer epithelial cells and a defect in epithelial cell differentiation in Hoxa3^+/-Pax1^un-ex/un-ex embryos. The fetal thymic organ culture experiments showed a significant reduction in epithelial cell number, specifically in the Hoxa3^+/-Pax1^un-ex/un-ex mutant thymic lobes. Hoxa3-Pax1 compound mutants also have a smaller thymic rudiment as early as E11.5–12.5 (data not shown), when thymocytes normally constitute a small percentage of the thymus. This result supports the conclusion that there are fewer epithelial cells from the earliest stages of organogenesis. A reduction in epithelial cell number alone could account at least in part for the observed effects on thymocyte differentiation by restricting the availability of epithelial cells capable of supporting thymocyte maturation.

In addition to reduced epithelial cell number, the reduction in MHC class II expression levels is an indication of defective epithelial cell differentiation. Although the role of MHC class II-expressing epithelial cells in promoting positive selection and the maturation of DP cells to single positive cells is well-documented (1, 2), the role of these cells in the development of DN cells into DP cells is less well-understood. Evidence from reaggregate thymic organ cultures indicates that interactions between thymocytes and MHC class II⁺ thymic epithelial cells are required for progression from late DN to DP stages (54). Induction of TCR rearrangements and commitment to the T cell lineage is also controlled by interactions between IL-7-producing MHC class II⁺ epithelial cells and immature DN thymocytes (11). The reduction in MHC class II expression itself is unlikely to be functionally related to the observed phenotype in that loss of this expression on thymic epithelial cells is associated with defects in positive selection of CD4⁺ cells but not in generation of DP cells (4, 8). We did not see a reduction in the percentage of CD4⁺ SP cells or in positive selection as measured by CD69 expression on DP cells. It should be noted that relatively little positive selection normally occurs at these fetal stages. Unfortunately, Hoxa3^+/-Pax1^-/- mice on the C57BL6 genetic background die at or soon after birth due to other effects of the mutations, and we have been unable to recover any survivors of this genotype for postnatal analysis. Therefore, we cannot determine as of yet whether the defects in thymocyte maturation we observed reflect a permanent deficit or if they are due to a developmental delay or whether defects in positive selection would be apparent postnatally.

Our results show that mutations in Hoxa3 and Pax1 act in a synergistic and dosage-dependent fashion to affect thymus development. Our results represent the first demonstration of genetic interactions between members of the Hox and Pax transcription factor families. This result is particularly striking because Hoxa3 and Pax1 show a stronger genetic interaction than Hoxa3 and its paralogs, Hoxb3 and Hoxd3 (22, 23, 24). The nature of this interaction is not yet determined. Although our previous studies suggested that Pax1 might be downstream of Hoxa3 (20), it is clear from the difference between the Hoxa3^+/-Pax1^un-ex/un-ex and Pax1^un-ex/un-ex phenotypes that there is an interaction beyond any transcriptional regulation of Pax1 by Hoxa3. Hoxa3 and Pax1 may act together to regulate common downstream target genes. Alternatively, other genes in the pathway could also be involved, including Pax9 (16, 55).

Another transcription factor important for early thymic epithelial cell development is whn (Hfh11), the gene mutated in the nude mouse (15). Loss of whn causes a thymic epithelial cell defect that results in arrest in epithelial cell differentiation and failure of lymphocytes to populate the thymus. Nehls et al. (56) proposed that whn activity functions as a "control point" dividing thymic epithelial cell development into initial formation of the alymphoid thymic rudiment and further differentiation in the presence of hematopoietic cells. The interaction between Hoxa3 and Pax1 may act downstream of this whn checkpoint or in a separate pathway from whn because the epithelial cell defect in Hoxa3-Pax1 mutants appears to occur before E12.5 (data not shown), but hematopoietic precursors do invade the thymic rudiment and undergo maturation. In support of a separate pathway, our preliminary data suggest that Hoxa3 and whn do not show a genetic interaction, in contrast to the results presented in this paper (C. Raines and Nancy R. Manley, unpublished data). Additional genetic studies and investigation of molecular interactions will be needed to determine the relationships between these transcription factors during early thymus development.

	Acknowledgments

 
We thank R. Balling and members of his laboratory for generously supplying the Pax1^un-ex mice and the probe for Pax1 Southern analysis and for sharing data before publication. We also thank A. Mellor, L. Ignatowicz, and B. Chmielowski for assistance with the chimera analysis and A. Mellor, M. Iwashima, and L. Ignatowicz for critical reading of the manuscript. B. Headrick in the Institute of Molecular Medicine and Genetics Electron Microscopy/Histology Core Facility supplied technical assistance with histology. Special thanks to B. Condie for advice and helpful discussion.

	Footnotes

 
¹ This work was supported in part by the Medical College of Georgia Research Institute Grants Program and by National Institutes of Health Grant #R01-HD35920.

² Address correspondence and reprint requests to Dr. Nancy R. Manley, Medical College of Georgia, Institute for Molecular Medicine and Genetics, 1120 15th Street, CA4010, Augusta, GA 30912.

³ Abbreviations used in this paper: DN, double negative; DP, double positive; SP, single positive; E, embryonic day; TN, triple negative; dGuo, deoxyguanosine.

Received for publication February 3, 2000. Accepted for publication March 20, 2000.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Anderson, G., J. J. Owen, N. C. Moore, E. J. Jenkinson. 1994. Thymic epithelial cells provide unique signals for positive selection of CD4⁺ CD8⁺ thymocytes in vitro. J. Exp. Med. 179:2027.[Abstract/Free Full Text]
2. Anderson, G., E. J. Jenkinson, N. C. Moore, J. J. Owen. 1993. MHC class II-positive epithelium and mesenchyme cells are both required for T-cell development in the thymus. Nature 362:70.[Medline]
3. Holländer, G. A., B. Wang, A. Nichogiannopoulou, P. P. Platenburg, W. van Ewijk, S. J. Burakoff, J. C. Gutierrez-Ramos, C. Terhorst. 1995. Developmental control point in induction of thymic cortex regulated by a subpopulation of prothymocytes. Nature 373:350.[Medline]
4. Markowitz, J. S., Jr H. Auchincloss, M. J. Grusby, L. H. Glimcher. 1993. Class II-positive hematopoietic cells cannot mediate positive selection of CD4⁺ T lymphocytes in class II-deficient mice. Proc. Natl. Acad. Sci. USA 90:2779.[Abstract/Free Full Text]
5. Ritter, M. A., R. L. Boyd. 1993. Development in the thymus: it takes two to tango. Immunol. Today 14:462.[Medline]
6. Le Douarin, N. M., F. Dieterlen-Lièvre, P. D. Oliver. 1984. Ontogeny of primary lymphoid organs and lymphoid stem cells. Am. J. Anat. 170:261.[Medline]
7. Boyd, R. L., C. L. Tucek, D. I. Godfrey, D. J. Izon, T. J. Wilson, N. J. Davidson, A. G. Bean, H. M. Ladyman, M. A. Ritter, P. Hugo. 1993. The thymic microenvironment. Immunol. Today 14:445.[Medline]
8. Grusby, M. J., R. S. Johnson, V. E. Papaioannou, L. H. Glimcher. 1991. Depletion of CD4⁺ T cells in major histocompatibility complex class II-deficient mice. Science 253:1417.[Abstract/Free Full Text]
9. Laufer, T. M., J. DeKoning, J. S. Markowitz, D. Lo, L. H. Glimcher. 1996. Unopposed positive selection and autoreactivity in mice expressing class II MHC only on thymic cortex. Nature 383:81.[Medline]
10. Godfrey, D. I., J. Kennedy, T. Suda, A. Zlotnik. 1993. A developmental pathway involving four phenotypically and functionally distinct subsets of CD3^-CD4^-CD8^- triple-negative adult mouse thymocytes defined by CD44 and CD25 expression. J. Immunol. 150:4244.[Abstract]
11. Oosterwegel, M. A., M. C. Haks, U. Jeffry, R. Murray, A. M. Kruisbeek. 1997. Induction of TCR gene rearrangements in uncommitted stem cells by a subset of IL-7 producing, MHC class II-expressing thymic stromal cells. Immunity 6:351.[Medline]
12. Di Santo, J. P., H.-R. Rodewald. 1998. In vivo roles of receptor tyrosine kinases and cytokine receptors in early thymocyte development. Curr. Opin. Immunol. 10:196.[Medline]
13. Malek, T. R., B. O. Porter, Y.-W. He. 1999. Multiple c-dependent cytokines regulate T-cell development. Immunol. Today 20:71.[Medline]
14. Burkly, L., C. Hession, L. Ogata, C. Reilly, L. A. Marconi, D. Olson, R. Tizard, R. Cate, D. Lo. 1995. Expression of relB is required for the development of thymic medulla and dendritic cells. Nature 373:531.[Medline]
15. Nehls, M., D. Pfeifer, M. Schorpp, H. Hedrich, T. Boehm. 1994. New member of the winged-helix protein family disrupted in mouse and rat nude mutations. Nature 372:103.[Medline]
16. Peters, H., A. Neubüser, K. Kratochwil, R. Balling. 1998. Pax9-deficient mice lack pharyngeal pouch derivatives and teeth and exhibit craniofacial and limb abnormalities. Genes Dev. 12:2735.[Abstract/Free Full Text]
17. Chisaka, O., M. R. Capecchi. 1991. Regionally restricted developmental defects resulting from targeted disruption of the mouse homeobox gene hox-1.5. Nature 350:473.[Medline]
18. Dietrich, S., P. Gruss. 1995. Undulated phenotypes suggest a role of Pax-1 for the development of vertebral and extravertebral structures. Dev. Biol. 167:529.[Medline]
19. Wallin, J., H. Eibel, A. Neubüser, J. Wilting, H. Koseki, R. Balling. 1996. Pax1 is expressed during development of the thymus epithelium and is required for normal T-cell maturation. Development 122:23.[Abstract]
20. Manley, N. R., M. R. Capecchi. 1995. The role of hoxa-3 in mouse thymus and thyroid development. Development 121:1989.[Abstract]
21. Krumlauf, R.. 1994. Hox genes in vertebrate development. Cell 78:191.[Medline]
22. Condie, B. G., M. R. Capecchi. 1994. Mice with targeted disruptions in the paralogous genes hoxa-3 and hoxd-3 reveal synergistic interactions. Nature 370:304.[Medline]
23. Manley, N. R., M. R. Capecchi. 1997. Hox group 3 paralogous genes act synergistically in the formation of somitic and neural crest-derived structures. Dev. Biol. 192:274.[Medline]
24. Manley, N. R., M. R. Capecchi. 1998. Hox group 3 paralogs regulate the development and migration of the thymus, thyroid and parathyroid glands. Dev. Biol. 195:1.[Medline]
25. Cordier, A. C., S. M. Haumont. 1980. Development of thymus, parathyroids, and ultimobranchial bodies in NMRI and nude mice. Am. J. Anat. 157:227.[Medline]
26. LeLievre, C. S., N. M. LeDouarin. 1975. Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos. J. Embryol. Exp. Morphol. 34:125.[Medline]
27. Washburn, T., E. Schweighoffer, T. Gridley, D. Chang, B. J. Fowlkes, D. Cado, E. Robey. 1997. Notch activity influences the ß versus T cell lineage decision. Cell 88:833.[Medline]
28. Gold, R., M. Schmied, G. Rothe, H. Zischler, H. Breitschopf, H. Wekerle, H. Lassmann. 1993. Detection of DNA fragmentation in apoptosis: application of in situ nick translation to cell culture systems and tissue sections. J. Histochem. Cytochem. 41:1023.[Abstract]
29. Listman, J. A., Y. Wang, J. E. Castro, I. J. Rimm, P. W. Finn, D. L. Perkins. 1998. Detection of rare apoptotic T cells in vivo. Cytometry 33:340.[Medline]
30. Su, D.-m., J. Wang, Q. Lin, M. D. Cooper, T. Watanabe. 1997. Interferons /ß inhibit IL-7-induced proliferation of CD4^-CD8^-CD3^-CD44⁺CD25⁺ thymocytes, but do not inhibit that of CD4^-CD8^-CD3^-CD44^-CD25^- thymocytes. Immunology 90:543.[Medline]
31. Kishi, H., D.-m. Su, A. Muraguchi, T. Watanabe. 1995. A novel cell surface antigen, immature thymocyte antigen-1, is involved in the differentiation of murine thymocytes. J. Immunol. 155:568.[Abstract]
32. Jenkinson, E. J., L. L. Franchi, R. Kingston, J. J. Owen. 1982. Effect of deoxyguanosine on lymphopoiesis in the developing thymus rudiment in vitro: application in the production of chimeric thymus rudiments. Eur. J. Immunol. 12:583.[Medline]
33. Ready, A. R., E. J. Jenkinson. 1987. Deoxyguanosine-induced survival of class II-expressing fetal mouse thymic allografts is associated with the depletion of donor lymphoid and dendritic cells. Transplant. Proc. 19:4225.[Medline]
34. Sauvageau, G., U. Thorsteinsdottir, M. R. Hough, P. Hugo, H. J. Lawrence, C. Largman, R. K. Humphries. 1997. Overexpression of HOXB3 in hematopoietic cells causes defective lymphoid development and progressive myeloproliferation. Immunity 6:13.[Medline]
35. Godfrey, D. I., A. Zlotnik. 1993. Control points in early T-cell development. Immunol. Today 14:547.[Medline]
36. Jenkinson, E. J., W. Van Ewijk, J. J. Owen. 1981. Major histocompatibility complex antigen expression on the epithelium of the developing thymus in normal and nude mice. J. Exp. Med. 153:280.[Abstract/Free Full Text]
37. Anderson, K. L., N. C. Moore, D. E. McLoughlin, E. J. Jenkinson, J. J. Owen. 1998. Studies on thymic epithelial cells in vitro. Dev. Comp. Immunol. 22:367.[Medline]
38. MacDonald, H., R. Budd, R. Howe. 1988. A CD3^- subset of CD4^-8⁺ thymocytes: a rapidly cycling intermediate in the generation of CD4⁺8⁺ cells. Eur. J. Immunol. 18:519.[Medline]
39. Nikolic-Zugic, J., M. W. Moore, M. J. Bevan. 1989. Characterization of the subset of immature thymocytes which can undergo rapid in vitro differentiation. Eur. J. Immunol. 19:649.[Medline]
40. Petrie, H. T., P. Hugo, R. Scollay, K. Shortman. 1990. Lineage relationships and developmental kinetics of immature thymocytes: CD3, CD4, and CD8 acquisition in vivo and in vitro. J. Exp. Med. 172:1583.[Abstract/Free Full Text]
41. Penit, C., B. Lucas, F. Vasseur. 1995. Cell expansion and growth arrest phases during the transition from precursor (CD4^-8^-) to immature (CD4⁺8⁺) thymoctytes in normal and genetically modified mice. J. Immunol. 154:5103.[Abstract]
42. Takahama, Y., J. J. Letterio, H. Suzuki, A. G. Farr, A. Singer. 1994. Early progression of thymocytes along the CD4/CD8 developmental pathway is regulated by a subset of thymic epithelial cells expressing transforming growth factor ß. J. Exp. Med. 179:1495.[Abstract/Free Full Text]
43. Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H. J. Fehling, H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor gene. Science 266:1208.[Abstract/Free Full Text]
44. Groettrup, M., H. von Boehmer. 1993. A role for a pre-T-cell receptor in T-cell development. Immunol. Today 14:610.[Medline]
45. Fehling, H. J., H. von Boehmer. 1997. Early ß T cell development in the thymus of normal and genetically altered mice. Curr. Opin. Immunol. 9:263.[Medline]
46. Molina, T. J., K. Kishihara, D. P. Siderovski, W. van Ewijk, A. Narendran, E. Timms, C. J. A. Wakeham, K. U. Paige, A. Hartmann, A. Veillette, et al 1992. Profound block in thymocyte development in mice lacking p56^lck. Nature 357:161.[Medline]
47. Turner, M., P. J. Mee, A. E. Walters, M. E. Quinn, A. L. Mellor, R. Zamoyska, V. L. Tybulewicz. 1997. A requirement for the Rho-family GTP exchange factor Vav in positive and negative selection of thymocytes. Immunity 7:451.[Medline]
48. von Boehmer, H., I. Aifantis, J. Feinberg, O. Lechner, C. Saint-Ruf, U. Walter, J. Buer, O. Azogui. 1999. Pleiotropic changes controlled by the pre-T-cell receptor. Curr. Opin. Immunol. 11:135.[Medline]
49. Wiles, M. V., P. Ruiz, B. A. Imhof. 1992. Interleukin-7 expression during mouse thymus development. Eur. J. Immunol. 22:1037.[Medline]
50. Palacios, R., S.-I. Nishikawa. 1992. Developmentally regulated cell surface expression and function of c-kit receptor during lymphocyte ontogeny in the embryo and adult mice. Development 115:1133.[Abstract]
51. Moore, T. A., A. Zlotnik. 1995. T-cell lineage commitment and cytokine responses of thymic progenitors. Blood 86:1850.[Abstract/Free Full Text]
52. Vacchio, M. S., V. Papadopulos, J. D. Ashwell. 1994. Steroid production in the thymus: implications for thymocyte selection. J. Exp. Med. 179:1835.[Abstract/Free Full Text]
53. King, L. B., M. S. Vacchio, K. Dixon, R. Hunziker, D. H. Margulies, J. D. Ashwell. 1995. A targeted glucocorticoid receptor antisense transgene increases thymocyte apoptosis and alters thymocyte development. Immunity 3:647.[Medline]
54. Anderson, G., K. L. Anderson, E. Z. Tchilian, J. J. Owen, E. J. Jenkinson. 1997. Fibroblast dependency during early thymocyte development maps to the CD25⁺ CD44⁺ stage and involves interactions with fibroblast matrix molecules. Eur. J. Immunol. 27:1200.[Medline]
55. Neubüser, A., H. Koseki, R. Balling. 1995. Characterization and developmental expression of Pax9, a paired-box-containing gene related to Pax1. Dev. Biol. 170:701.[Medline]
56. Nehls, M., B. Kyewski, M. Messerle, R. Waldschütz, K. Schüddekopf, A. J. Smith, T. Boehm. 1996. Two genetically separable steps in the differentiation of thymic epithelium. Science 272:886.[Abstract]

This article has been cited by other articles:

				S. Xiao, D.-m. Su, and N. R. Manley Atypical Memory Phenotype T Cells with Low Homeostatic Potential and Impaired TCR Signaling and Regulatory T Cell Function in Foxn1{Delta}/{Delta} Mutant Mice J. Immunol., December 15, 2007; 179(12): 8153 - 8163. [Abstract] [Full Text] [PDF]

				I. Popa, I. Zubkova, M. Medvedovic, T. Romantseva, H. Mostowski, R. Boyd, and M. Zaitseva Regeneration of the adult thymus is preceded by the expansion of K5+K8+ epithelial cell progenitors and by increased expression of Trp63, cMyc and Tcf3 transcription factors in the thymic stroma Int. Immunol., November 1, 2007; 19(11): 1249 - 1260. [Abstract] [Full Text] [PDF]

				J. Gui, X. Zhu, J. Dohkan, L. Cheng, P. F. Barnes, and D.-M. Su The aged thymus shows normal recruitment of lymphohematopoietic progenitors but has defects in thymic epithelial cells Int. Immunol., October 1, 2007; 19(10): 1201 - 1211. [Abstract] [Full Text] [PDF]

				R. W. Georgantas III, V. Tanadve, M. Malehorn, S. Heimfeld, C. Chen, L. Carr, F. Martinez-Murillo, G. Riggins, J. Kowalski, and C. I. Civin Microarray and Serial Analysis of Gene Expression Analyses Identify Known and Novel Transcripts Overexpressed in Hematopoietic Stem Cells Cancer Res., July 1, 2004; 64(13): 4434 - 4441. [Abstract] [Full Text] [PDF]

				R. Drut and C. Galliani Thymic Tissue in the Skin: A Clue to the Diagnosis of the Branchio-oculo-Facial Syndrome: Report of Two Cases International Journal of Surgical Pathology, January 1, 2003; 11(1): 25 - 28. [Abstract] [PDF]

				J. Kirchner, K. A. Forbush, and M. J. Bevan Identification and Characterization of Thymus LIM Protein: Targeted Disruption Reduces Thymus Cellularity Mol. Cell. Biol., December 15, 2001; 21(24): 8592 - 8604. [Abstract] [Full Text] [PDF]

				J.-M. Revest, R. K. Suniara, K. Kerr, J. J. T. Owen, and C. Dickson Development of the Thymus Requires Signaling Through the Fibroblast Growth Factor Receptor R2-IIIb J. Immunol., August 15, 2001; 167(4): 1954 - 1961. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Su, D.-m. Articles by Manley, N. R. Search for Related Content PubMed PubMed Citation Articles by Su, D.-m. Articles by Manley, N. R.