Pepro Tech
Originally published online as doi:10.1189/jlb.0507297 on October 26, 2007

Published online before print October 26, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
jlb.0507297v1
83/2/419    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ramsey, K.
Right arrow Articles by Clements, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ramsey, K.
Right arrow Articles by Clements, J. L.
(Journal of Leukocyte Biology. 2008;83:419-429.)
© 2008 by Society for Leukocyte Biology

Impaired thymic selection in mice expressing altered levels of the SLP-76 adaptor protein

Kimberley Ramsey*, Nancy Luckashenak{dagger},1, Gary A. Koretzky{ddagger} and James L. Clements{dagger},2

Departments of
* Cell Stress Biology and
{dagger} Immunology, Roswell Park Cancer Institute, Buffalo, New York, USA; and
{ddagger} Departments of Medicine and Pathology and Laboratory Medicine and Abramson Family Cancer Research Institute, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA

2 Correspondence: Department of Immunology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263, USA. E-mail: james.clements{at}roswellpark.org


arrow
ABSTRACT
 
Intracellular signaling initiated by ligation of the TCR influences cell fate at multiple points during the lifespan of a T cell. This is especially evident during thymic selection, where the nature of TCR-dependent signaling helps to establish a MHC-restricted, self-tolerant T cell repertoire. The Src homology 2 domain-containing leukocyte-specific phosphoprotein of 76 kDa (SLP-76) adaptor protein is a required intermediate in multiple signaling pathways triggered by TCR engagement, several of which have been implicated in dictating the outcome of thymic selection (e.g., intracellular calcium flux and activation of ERK family MAPKs). To determine if thymocyte maturation and selection at later stages of development are sensitive to perturbations in SLP-76 levels, we analyzed these crucial events using several transgenic (Tg) lines of mice expressing altered levels of SLP-76 in the thymus. In Tg mice expressing low levels of SLP-76 in preselection thymocytes, the CD4:CD8 ratio in the thymus and spleen was skewed in a manner consistent with impaired selection and/or maturation of CD4+ thymocytes. Low SLP-76 expression also correlated with reduced CD5 expression on immature thymocytes, consistent with reduced TCR signaling potential. In contrast, reconstitution of SLP-76 at higher levels resulted in normal thymic CD5 expression and CD4:CD8 ratios in the thymus and periphery. It is curious that thymic deletion of TCR-Tg (HY) thymocytes was markedly impaired in both lines of Tg-reconstituted SLP-76–/– mice. Studies using chimeric mice indicate that the defect in deletion of HY+ thymocytes is intrinsic to the developing thymocyte, suggesting that maintenance of sufficient SLP-76 expression from the endogenous locus is a key element in the selection process.

Key Words: thymocyte • tolerance


arrow
INTRODUCTION
 
Thymic selection makes a major contribution to the establishment of immunologic self-tolerance. Central to this process is the nature of the intracellular signals elicited by the TCR expressed on immature thymocytes following ligation with self-peptide:MHC (pMHC) complexes presented by cellular elements of the thymic stroma (e.g., cortical epithelial cells). Data obtained from multiple studies suggest that the establishment and maintenance of a defined intracellular signaling potential in preselection thymocytes are critical for preventing the escape of potentially autoreactive T cell clones. In support of this idea, perturbing the structure or expression of several key signaling intermediates functioning downstream of the TCR can have profound effects on the outcome of selection. For example, deletion of ITAMs in the cytosolic tail of the TCR{gamma} chain compromises signaling potential and impairs positive selection [1 ]. A spontaneous mutation in the Syk family protein tyrosine kinase ZAP-70 results in impaired thymic deletion and the onset of autoimmunity in mice [2 ]. Targeted deletion of the Tec family tyrosine kinase Itk alone or in combination with Rlk (an Itk homologue) has profound effects on positive and negative selection, respectively [3 4 5 ]. Perhaps related, biochemical signals, initiated as a consequence of increases in the second messenger calcium, have also been implicated in the selection process. Inhibiting the activity of the calcium-dependent serine-threonine kinase calcineurin with pharmacological agents impairs positive and negative selection in mice [6 ]. It is interesting that mice treated with low doses of cyclosporin A, a potent inhibitor of calcineurin, develop autoimmunity with age, most likely as a consequence of failed negative selection [7 ]. Collectively, these reports support the idea that the outcome of thymocyte selection and establishment of immunological tolerance are critically dependent on intracellular signaling potential.

One confounding issue that remains regarding thymic selection is why the same self-pMHC complexes that trigger TCR-dependent, positive selection fail to activate the selected repertoire in the periphery. One potential answer to this question lies in the observation that the signaling potential of preselection CD4+CD8+ double-positive (DP) thymocytes is different from that of mature, peripheral T cells. This is supported by experimental evidence indicating that preselection DP thymocytes are more readily activated than mature T cells in response to suboptimal stimuli, despite low levels of TCR/CD3 expression [8 9 ]. More recently, the expression of a specific microRNA (miR-181a) was shown to be elevated in DP thymocytes when compared directly with more mature, single-positive (SP) thymocytes or naïve peripheral T cells [10 ]. As miR-181a functions normally to augment TCR signaling, transient expression in DP thymocytes likely elevates TCR signaling potential and provides a relatively narrow developmental window where self-pMHC complexes trigger sufficient signaling to permit selection.

Although multiple distal molecular events (e.g., calcium release) that direct thymic selection have been identified, the more proximal signaling intermediates that regulate positive and negative selection have not been defined fully. In mature T cells, a biochemical signaling circuit comprised of the Src homology 2 domain-containing leukocyte-specific phosphoprotein of 76 kDa (SLP-76), Grb2-related adaptor downstream of Shc (Gads), and linker of activated T cell (LAT) adaptor proteins links TCR ligation (and subsequent activation of ZAP-70) directly with more distal events following TCR ligation [11 ]. By virtue of constitutive and inducible interactions with multiple signaling proteins, these adaptors cooperate to optimally activate phospholipase C (PLC){gamma}1, calcium flux, and ultimately, cytokine production following TCR engagement. It therefore seems likely that these same adaptors may also contribute to biochemical signaling pathways that dictate the outcome of selection in developing thymocytes. Indeed, Gads-deficient mice display severe defects in positive and negative selection when bred onto a TCR-transgenic (Tg) background [12 ]. More recently, it has been shown that mice expressing a mutant LAT protein incapable of binding PLC{gamma}1 also manifest defects in thymic-positive and -negative selection [13 ]. The selection defects observed in Gads-deficient and LAT mutant "knock-in" mice implicate SLP-76 as a potentially important regulator of the selection process. The fact that SLP-76 functions in signaling pathways regulating intracellular calcium flux and ERK1/2 activation suggests further that SLP-76 is biochemically positioned to relay signals dictating the outcome of selection. In support of this idea, inducible deletion of SLP-76, just prior to selection, markedly impairs positive selection and generation of CD4 and CD8 SP thymocytes [14 ].

In this report, we describe two effects of altering the normal pattern of SLP-76 expression on thymocyte development. First, reduced expression of SLP-76 appears to have a more profound effect on the generation of CD4 SP thymocytes when compared with CD8 SP thymocytes. Second, thymic deletion and selection of TCR-Tg thymocytes are less efficient in mice where the endogenous SLP-76 alleles are disrupted, and SLP-76 expression is Tg-dependent. A similar result was achieved in chimeric mice, suggesting that the temporal pattern of SLP-76 expression, intrinsic to the developing thymocyte, plays an important role in the outcome of thymic selection.


arrow
MATERIALS AND METHODS
 
Mice
Generation of SLP-76+/– and SLP-76–/– mice has been described previously [15 ]. To generate Tg mice expressing SLP-76 under transcriptional control of the Lck proximal promoter, a full-length cDNA encoding wild-type murine SLP-76 was generated by RT-PCR and cloned into the BamHI site present in the plasmid p1017 (kindly provided by Roger Perlmutter, Amgen, Inc.). The resulting construct was sequence-verified and used for subsequent microinjection into fertilized mouse embryos derived from an F2 cross of C57BL/6J x SJL/J (B6.SJL) mice (Transgenic Animal Facility, University of Iowa, Coralville, IA, USA). Five independent founder lines were established, and the relative copy numbers of each were established by Southern blotting. Three founder mice were chosen based on differing levels of Tg integration and backcrossed a minimum of four generations onto the SLP-76+/– background. SLP-76+/– Tg+ progeny were identified based on PCR analysis of genomic DNA and intercrossed to generate Tg+ and Tg– mice on SLP-76+/+, +/–, or –/– genetic backgrounds. HY TCR-Tg mice were purchased from Taconic (Germantown, NY, USA) and bred onto the appropriate backgrounds. All mice were maintained and cared for in accordance with the policies set forth by the Roswell Park Cancer Institute (RPCI) and Animal Care and Use Committee (Buffalo, NY, USA).

Chimeric mice were generated by first exposing wild-type, congenic B6.SJL mice (CD45.1+, Taconic) to a lethal dose of irradiation (1100 rad, delivered in two separate doses of 550 rad, 4 h apart) and subsequently, injecting the mice i.v. with bone marrow harvested from donor mice of the indicated genotypes. Chimeric mice were provided neomycin sulfate (2 mg/ml, Sigma Chemical Co., St. Louis, MO, USA) in drinking water for 2 weeks following irradiation and reconstitution.

Tissue isolation and cell preparation
Thymus or spleens were harvested from killed mice and single cell suspensions prepared by gently homogenizing each tissue in RPMI supplemented with 5% FCS using a plastic disposable pestle and a 1.5-ml microfuge tube. Tissue debris was removed by filtering the suspension through a nylon cell strainer (70 uM). RBCs were lysed using an ammonium chloride (0.15 M) lysis buffer. Viable cells (trypan blue-negative) were enumerated using a hemacytometer.

Flow cytometry and antibodies
Total thymocytes or splenocytes were isolated and washed once in media (RPMI/5% FCS) prior to staining. Cells (0.1–0.5x106 per sample) were incubated with Fc BlockTM (anti-CD16/32) for 10 min at 4°C prior to addition of one or more of the following fluorochrome-conjugated antibodies specific for: CD3{epsilon} (clone 145-2C11), CD4 (GK1.5), CD5 (53-7.3), CD8 (53-6.7), CD44 (IM7), CD62 ligand (CD62L; MEL-14), HY TCR (T3.70, eBiosciences, San Diego, CA, USA). All antibodies were purchased from BD Biosciences/PharMingen (San Diego, CA, USA) or Biolegend (San Diego, CA, USA), unless otherwise noted. For each experiment, background fluorescence or staining was determined using unstained cells or cells incubated with the appropriate fluorochrome-conjugated, isotype-matched control antibody. Cells were incubated with the antibody or antibodies for 30 min at 4°C, washed once, and resuspended in PBS/2% paraformaldehyde prior to analysis. For intracellular detection of SLP-76, cells were first surface-stained with the appropriate antibody and then fixed and permeabilized using 1x IC-PermTM buffer (Biosource, Worcester, MA, USA). FITC-conjugated antibody specific for murine SLP-76 (clone MS76, eBiosciences) was then added in IC-Perm buffer (final dilution, 1:200). Stained cells were then washed once and resuspended in PBS prior to analysis. All samples were analyzed at the RPCI Flow Cytometry Core Facility using a FACSCaliberTM or FACScanTM instrument (BD Biosciences/PharMingen), and data were acquired using CellQuestTM software. Collected events were analyzed further using WinMDI software (Version 2.8).

In vivo labeling with BrdU
Mice of the indicated genotypes (n=3/genotype) were injected i.p. with 0.8 mg BrdU (Sigma Chemical Co.), diluted in sterile PBS once daily for 5 consecutive days. Two days after the last injection, mice were killed, and single cell suspensions were prepared from harvested spleens. Cells were surface-stained for flow cytometry as described and then fixed and treated with DNase I prior to staining with a FITC-conjugated anti-BrdU antibody (Sigma Chemical Co.).


arrow
RESULTS
 
Thymocyte development and selection are impaired in Tg mice expressing lower than normal levels of SLP-76
Mice deficient for SLP-76 (–/–) are virtually devoid of peripheral {alpha}β or {gamma}{delta} TCR+ T cells as a result of a severe and early block in thymocyte development, precluding the analysis of SLP-76 function at later stages of thymocyte maturation and selection in this mutant strain [15 16 ]. As one means to reconstitute SLP-76 expression in SLP-76–/– mice, we engineered a Tg construct comprised of the proximal Lck promoter and a full-length, wild-type murine SLP-76 cDNA (pLckSLP). We reasoned that given the restricted activity of the proximal promoter to thymocytes [17 ], Tg-encoded SLP-76 would be expressed at early stages of thymocyte development and be maintained at least until the developing T cells acquired a mature status. It was also anticipated that Tg lines expressing a range of SLP-76 levels would be obtained as a result of differences in Tg copy number and/or integration site. To this end, three independent founder lines of pLckSLP Tg mice were produced and intercrossed with SLP-76+/– mice. Subsequent matings were established to generate Tg mice on SLP-76+/+, +/–, and –/– genetic backgrounds. In each case, the presence of the Tg+ was sufficient to rescue thymocyte development and promote maturation beyond the CD4–CD8– double-negative (DN) stage of development in the SLP-76–/– background, albeit with different efficiencies (Fig. 1 A , and data not shown). Tg-reconstituted SLP-76–/– mice derived from Line 1 Tg founders displayed a thymic CD4/CD8 profile that was indistinguishable from control mice (SLP-76+/+ or +/–) and elevated thymic cellularity when compared with wild-type mice (Fig. 1A and 1B) . In contrast, SLP-76–/– Tg+ mice derived from Line 2 founder mice manifested an increased frequency of DN thymocytes (Fig. 1A) and generally reduced thymic cellularity when compared with SLP-76+/+ mice (Fig. 1B) , suggesting that thymocyte development was suboptimal in this background. Similar results (i.e., increased frequency of DN thymocytes and reduced thymic yield) were obtained using SLP-76–/– Tg+ mice derived from a third independent line of Tg founders (data not shown).


Figure 1
View larger version (52K):
[in this window]
[in a new window]

  		Figure 1. The pLckSLP-76 Tg restores thymocyte development and SLP-76 expression in SLP-76–/– mice to varying degrees. (A) Two-color dot-plots showing the CD4 and CD8 expression on total thymocytes (gated first on size and scatter characteristics) isolated from mice of the indicated genotypes. NLC, Normal littermate control. The percentage of cells in each quadrant is indicated. (B) Thymocyte yield from mice (n≥5) of the indicated genotypes. P values obtained from comparing wild-type (+/+) samples with each of the other genotypes using the Student’s unpaired t-test are shown. (C) Intracellular SLP-76 expression (shaded histograms) in gated DN, DP, CD4 SP, and CD8 SP isolated from mice of the indicated genotypes. The mean fluorescence intensity (MFI) with background staining (isotype) values subtracted (dashed histograms) is provided for each histogram. Data are representative of three separate experiments.

Analysis of SLP-76 expression in specific thymocyte subsets (defined by CD4 and CD8 expression) provided one potential explanation for the reduced thymic cellularity and increased percentage of DN thymocytes observed in Line 2 SLP-76–/– Tg+ mice. When compared with the level of SLP-76 expressed in wild-type or SLP-76+/– thymocytes (Fig. 1C) , expression of Tg-encoded SLP-76 in Line 2 SLP-76–/– Tg+ mice was revealed to be quite low in the CD4–CD8– (DN) and CD4+CD8+ (DP) thymocyte subsets. Virtually identical data were obtained using SLP-76–/– Tg+ mice generated from Line 3 founders (data not shown). It is interesting that despite low expression in the DN and DP subsets, the level of SLP-76 expression in the CD4 SP or CD8 SP thymocyte subsets isolated from Line 2 SLP-76–/– Tg mice was more comparable with that observed in SLP-76+/– (non-Tg) mice (Fig. 1C) . SLP-76 expression in DN or DP thymocytes isolated from SLP-76–/– Tg+ mice derived from Line 1 was visibly higher than that observed in Line 2 SLP-76–/– Tg+ mice and was similar to or slightly elevated compared with SLP-76+/– non-Tg controls. The low level of SLP-76 expression observed in Line 2 (and Line 3) SLP-76–/– Tg+ mice likely accounts for the increased percentage of DN thymocytes and the reduced thymocyte yield obtained from these mice, as low SLP-76 expression is predicted to compromise the efficiency of pre-TCR signaling and thymocyte differentiation beyond the DN stage of development.

Impaired generation of CD4 SP thymocytes and decreased expression of CD5 in mice expressing low levels of SLP-76
To determine the impact of the reduced SLP-76 expression observed in Line 2 SLP-76–/– Tg+ DP thymocytes on subsequent development and selection, we examined the phenotype of thymocytes isolated from these mice in more detail. To this end, we analyzed the CD4/CD8 profiles of thymocytes after gating only on the more mature (CD3high) thymocyte subset (Fig. 2 A ). In parallel with a modest reduction in the percentage of CD3high thymocytes observed in Line 2 SLP-76–/– Tg+ mice, a skewing of the thymic CD4:CD8 ratio was reproducibly observed in these mice (Fig. 2A and 2B) . This ratio routinely fell in the range of 1.5–2.0 in Line 2 SLP-76–/– Tg+ mice, whereas the CD4:CD8 ratio within the CD3high subset fell between 3.0 and 4.0 in non-Tg SLP-76+/+ or +/– littermate controls (Fig. 2B) . A normal CD4:CD8 ratio within the CD3high subset was always observed in thymocytes isolated from Line 1 SLP-76–/– Tg+ mice, consistent with the level of thymic SLP-76 expression observed in this strain. The reduced CD4:CD8 ratio observed in Line 2 SLP-76–/– Tg+ mice was not simply a result of the random integration of the Tg, as SLP-76+/– Tg+ mice derived from Line 2 displayed a normal ratio of CD4:CD8 mature thymocytes within the CD3high population (~3:1; data not shown).


Figure 2
View larger version (33K):
[in this window]
[in a new window]

  		Figure 2. Skewing of the CD4:CD8 ratio and reduced surface expression of CD5 in thymocytes expressing low levels of SLP-76. (A) Thymocytes isolated from mice of the indicated genotypes were analyzed for surface TCR (CD3{epsilon}) expression. The percentage of CD3high cells in each single-color histogram is indicated. CD3high cells were then analyzed further for surface expression of CD4 and CD8. The percentage of cells in each quadrant is indicated for each dot-plot. (B) The ratio of CD4 to CD8 SP (CD3high) thymocytes was calculated for individual mice of the indicated genotype (n=3–5/group). *, The difference between the average value obtained for the Line 2 SLP-76–/– Tg+ samples and the average values calculated for each of the other samples was statistically significant (P<0.05, unpaired Student’s t-test). (C) Thymocytes were harvested from wild-type mice and stained for surface CD4, CD8, and CD5. The expression of CD5 (shaded histograms) on DN, DP, and CD4 and CD8 SP thymocytes is shown. The MFI of CD5 staining with background (control Ig, dashed histograms) subtracted is provided for each histogram. (D) Comparison of surface CD5 expression between mice of the indicated genotypes. Data for each thymocyte subset was acquired as in C and is normalized to the MFI obtained from the corresponding thymocyte subset isolated from a wild-type (+/+) control mouse. Data are representative of four separate experiments.

In wild-type mice, the level of CD5 expression has been reported to correlate in a positive manner with TCR signaling potential [18 ]. We therefore compared the surface expression of CD5 on thymocyte subsets isolated from Line 1 or Line 2 SLP-76–/– Tg+ mice with thymocytes obtained from wild-type control mice (Fig. 2C and 2D) . The DN and DP thymocyte subsets isolated from Line 2 SLP-76–/– Tg+ mice reproducibly displayed low surface levels of CD5 when compared with SLP-76+/+ or +/– or Line 1 SLP-76–/– Tg+ mice. The low level of surface CD5 expression is likely a consequence of less-efficient pre-TCR signaling, compromised TCR signaling at the DP stage, or a combination of both. It is worth noting that despite the low levels of CD5 expressed on the DN and DP thymocyte subsets isolated from Line 2 SLP-76–/– Tg+ mice, the level of CD5 expression on the CD4 SP and CD8 SP subsets was more comparable with that observed in control or Line 1 SLP-76–/– Tg+ mice (Fig. 2D) . These data suggest that SP thymocytes that have been selected successfully in Line 2 SLP-76–/– Tg+ mice have received sufficient TCR-dependent signals to express CD5 at normal levels, an idea consistent with our hypothesis that selected thymocytes represent those cells expressing sufficient, Tg-encoded SLP-76 to support positive selection and further maturation. The low levels of SLP-76 expressed in DP thymocytes in Line 2 SLP-76–/– Tg+ mice therefore appear to result in reduced signaling potential and visibly impaired selection of MHC class II-restricted thymocytes, consistent with a report published previously, describing severely compromised generation of CD4 and CD8 SP thymocytes when SLP-76 is deleted in an inducible manner, as thymocytes acquire a DP phenotype [14 ].

A high proportion of peripheral T cells arising in mice expressing low levels of Tg-reconstituted SLP-76 displays a memory phenotype
It was anticipated that the levels of Tg-encoded SLP-76 would fall in mature T cells arising in SLP-76–/– Tg+ mice (as a result of reduced promoter activity), facilitating the analysis of the SLP-76-dependent T cell function using primary murine T cells. Therefore, we examined the surface phenotype and levels of SLP-76 expression in peripheral T cells isolated from each of the Tg-reconstituted SLP-76–/– lines. Consistent with rescued thymocyte development, Tg expression proved sufficient to restore the peripheral T cell compartment as well (Fig. 3 A ). In contrast to SLP-76–/– mice, the presence of CD4+ and CD8+ T cells was readily detectable in spleens isolated from SLP-76–/– Tg+ mice derived from Line 1 or Line 2. However, we routinely found decreased percentages of CD4 and CD8 T cells and a skewing of the CD4:CD8 ratio in splenic suspensions prepared from Line 2 SLP-76–/– Tg+ mice (Fig. 3A) consistent with impaired selection and/or differentiation of CD4 SP thymocytes in the thymus. In addition, the CD4+ and CD8+ T cell population isolated from the spleens of Line 2 SLP-76–/– Tg+ contained an elevated frequency of cells with central memory (CD62L+CD44+) or effector memory (CD62L–CD44+) phenotypes (Fig. 3B) . These effects are not solely a result of Tg integration, as Line 2 SLP-76+/– Tg+ littermates display a normal distribution of these subsets (data not shown). Furthermore, the CD62L/CD44 phenotype of splenic T cells obtained from Line 1 SLP-76–/– Tg+ mice was consistently similar to that observed for control (SLP-76+/+ or +/–) mice (Fig. 3B) . Despite the differences in surface phenotype, the level of SLP-76 expressed in CD4+ and CD8+ T cells isolated from Line 1 or Line 2 SLP-76–/– Tg+ mice was generally comparable (Fig. 3C) . It is somewhat surprising that the levels of Tg-encoded SLP-76 were actually found to be elevated approximately two- to threefold in splenic T cells isolated from the SLP-76–/– Tg+ mice (Lines 1 and 2), compared with SLP-76+/+ non-Tg controls. These data indicate that the phenotypic differences observed between splenic T cells isolated from Line 1 or Line 2 SLP-76–/– Tg+ mice are not solely a result of differences in Tg-encoded SLP-76 expression, which was maintained at similar levels in peripheral T cells isolated from these mice.


Figure 3
View larger version (46K):
[in this window]
[in a new window]

  		Figure 3. Surface phenotype and SLP-76 expression in peripheral T cells isolated from Tg-reconstituted SLP-76–/– mice. (A) Splenocytes were isolated from mice of the indicated genotypes and analyzed for surface CD4 and CD8 expression. The percentage of CD4+ and CD8+ splenocytes in each dot-plot is indicated. (B) Same as in A, except the CD4+ and CD8+ cells were analyzed further for surface expression of CD44 and CD62L. A similar analysis was not done (ND) for the SLP-76–/– mouse, as there are not sufficient numbers of splenic CD4+ or CD8+ cells to analyze. (C) Intracellular SLP-76 expression was analyzed in the CD4 or CD8 splenocyte subsets isolated from mice of the indicated genotype. The MFI of SLP-76 staining (shaded histograms) is indicated with background (dashed histograms) staining subtracted. In each case, one representative experiment of at least three is shown.

We reasoned that the increased frequency of peripheral T cells with an effector or central memory phenotype in the Line 2 SLP-76–/– Tg+ line was a result of homeostatic expansion as a consequence of reduced thymic output. To test this directly, we measured incorporation of BrdU following five i.p. injections (once daily). As shown in Figure 4 A , an increased percentage of CD62L+CD44+ or CD62L–CD44+ cells isolated from Line 2 control (SLP-76+/– Tg+) or SLP-76–/– Tg+ mice labeled with BrdU when compared with naïve CD62L+CD44– T cells, consistent with homeostatic expansion in each case. To determine if the elevated frequency of CD44+ CD8+ peripheral T cells observed in the Line 2 SLP-76–/– Tg+ mice was TCR-dependent, we crossed this strain of mice onto an HY TCR Tg background. In female mice, the (H-2b-restricted) male antigen-reactive HY TCR is positively selected but never encounters cognate antigen in the periphery. As a result, the vast majority of HY TCR+ CD8+ T cells arising in wild-type female mice maintains a naïve phenotype (i.e., CD44low). This held true in the SLP-76+/– Tg+ and SLP-76–/– Tg+ backgrounds (Fig. 4B) . In both cases, the CD44+ population was found predominantly in the CD8+ HY TCRneg population when compared with the HY TCRhigh subset. This was evident especially in the SLP-76–/– Tg+ background. These data are consistent with TCR-dependent homeostatic expansion of peripheral CD8+ T cells in Line 2 SLP-76–/– Tg+ mice, most likely as a consequence of reduced thymic output.


Figure 4
View larger version (40K):
[in this window]
[in a new window]

  		Figure 4. The elevated frequency of peripheral CD8+ T cells arising in Line 2 SLP-76–/– Tg+ mice with a memory phenotype is a result of homeostatic expansion. (A) Splenocytes were harvested from mice of the indicated genotypes (Line 2) following five subsequent injections with BrdU (once daily) and analyzed for BrdU incorporation by flow cytometry. Average values of the relative percentage of BrdU-positive cells were calculated for the indicated T cell populations (n=3 mice). (B) Surface phenotype of peripheral T cells (spleen) isolated from HY+ Line 2 SLP-76+/– or –/– Tg+ mice. Splenocytes were harvested from mice of the indicated genotypes and analyzed for surface expression of CD8, HY TCR, and CD44 by flow cytometry. Histograms represent CD44 staining on the indicated populations. The percentage of cells in a given quadrant or gate is indicated. One representative experiment of two is shown.

Thymic selection is impaired in a SLP-76–/– background when restoration of SLP-76 expression is restricted to T cells
Breeding Line 1 and Line 2 SLP-76–/– Tg+ mice onto the HY TCR Tg background afforded us the opportunity to evaluate thymic positive and negative selection in the context of differing levels of SLP-76. We reasoned that the low levels of SLP-76 expressed in DP thymocytes in Line 2 SLP-76–/– Tg+ mice would reduce the efficiency of negative selection, and the level of SLP-76 expression achieved in Line 1 SLP-76–/– Tg+ mice should support efficient deletion. We were therefore surprised when we found that deletion of HY TCR+ thymocytes was visibly impaired in Line 1 and Line 2 SLP-76–/– Tg+ male mice (Fig. 5A and 5B ), as indicated by the presence of a substantial population of DP thymocytes maintaining low levels of HY TCR and in the case of Line 1 SLP-76–/– Tg+ mice, increased thymic cellularity. This phenotype is similar to that observed in HY TCR female mice, where HY TCR expression is reduced on DP thymocytes when compared with the DN thymocyte subset. The maintenance of HY TCRlow DP thymocytes in both lines of Tg-reconstituted SLP-76–/– mice may also have been attributed to altered HY TCR expression at earlier stages of development. However, HY TCR expression on the DN subset in Line 1 SLP-76–/– Tg+ HY+ mice was comparable with control mice (SLP-76+/– HY+), where deletion was intact (Fig. 5 B). A population of HY TCRhigh DN thymocytes was also readily visible in Line 2 SLP-76–/– Tg+ HY+ male mice, although the percentage was reduced compared with SLP-76+/– HY+ or Line 1 SLP-76–/– Tg+ HY+ mice. Despite maintenance of HY TCRlow DP thymocytes in the Tg-reconstituted SLP-76–/– background, it should be noted that no subsequent selection of CD8 SP HY TCR+ thymocytes was readily apparent (Fig. 5A) . Peripheral (splenic) T cells expressing low levels of CD8 and high levels of HY TCR were evident in Line 1 SLP-76–/– Tg+ HY+ male mice at a frequency comparable with that observed in control SLP-76+/– Tg+ HY+ mice (data not shown). The origin of such cells has not been defined fully, but the low levels of CD8 expressed on the peripheral cells may reduce signaling potential and allow for subsequent selection and retention.


Figure 5
View larger version (53K):
[in this window]
[in a new window]

  		Figure 5. Impaired thymic deletion and positive selection in Tg-reconstituted SLP-76–/– mice. (A) The extent of thymic deletion of HY TCR+ thymocytes in male mice was measured by analyzing surface expression of CD4 and CD8 on total thymocytes isolated from mice of the indicated genotypes. The numbers above each dot-plot represent total thymocyte yield. The percentage of cells in each quadrant is indicated. Data are representative of three separate experiments. (B) Thymocytes were prepared and stained as in A and then analyzed further for surface expression of the clonotypic HY TCR. Data are representative of three separate experiments. (C) Impaired positive selection of HY TCR+ thymocytes, which were isolated from female mice of the indicated genotypes and analyzed for surface CD4 and CD8 expression. Histograms represent HY TCR expression on DP and SP thymocytes harvested from mice of the indicated genotypes. DN thymocytes, which express high levels of the HY TCR, are excluded from the analysis. One representative experiment of three is shown.

When positive selection of HY TCR+ thymocytes was assessed in female mice, a subtle but reproducible impairment of positive selection was also noted in both lines of Tg-reconstituted mice (Fig. 5C) . In each case, a 30–40% reduction in the percentage of mature HY TCRhigh thymocytes was detected (after gating out the HY TCRhigh DN thymocyte compartment). Again, we had predicted a more profound effect on positive selection in the Line 2 SLP-76–/– Tg+ background and were therefore surprised to find a similar phenotype in both lines of Tg-reconstituted SLP-76–/– mice.

Impaired deletion observed in SLP-76–/– Tg+ HY+ mice is thymocyte-intrinsic
To determine if the impaired deletion of HY+ thymocytes in Tg-reconstituted SLP-76–/– mice was thymocyte-intrinsic or -extrinsic, we generated chimeric mice using irradiated wild-type hosts and bone marrow isolated from Line 1 Tg mice on a SLP-76+/– or –/– background. In addition, bone marrow harvested from Line 1 SLP-76–/– Tg+ mice was mixed 1:1 with RAG-2–/– bone marrow prior to injection, with the intent of reconstituting the hematopoietic contribution to the thymocyte stromal microenvironment and SLP-76+/+ cells were incapable of seeding the mature (DP and beyond) thymocyte compartment directly. In either case, impaired deletion of HY TCR+ SLP-76–/– Tg+ donor thymocytes derived from Line 1 mice was still observed (Fig. 6 A ), suggesting that the deletion defect was thymocyte-intrinsic and not solely a result of a SLP-76-deficient thymic microenvironment. As seen in nonchimeric mice (Fig. 5) , the DN thymocytes generated from the Line 1 SLP-76–/– Tg+ donor expressed high levels of HY TCR, and the DP thymocytes maintained low but detectable levels of HY TCR expression (Fig. 6B) . The population of HY TCRneg thymocytes within the DN subset isolated from chimeric mice reconstituted with a 1:1 mixture of bone marrow likely reflects the contribution from RAG-2-deficient bone marrow (Fig. 6B , upper-right histogram).


Figure 6
View larger version (48K):
[in this window]
[in a new window]

  		Figure 6. Impaired deletion in Tg-reconstituted SLP-76–/– HY+ male mice is thymocyte-intrinsic. (A) CD4 and CD8 profiles of thymocytes isolated from chimeric mice generated using wild-type hosts and bone marrow harvested from the indicated donors. Host- and donor-derived thymocytes were identified based on CD45.1 (host) expression. In each case, a small population of host-derived thymocytes (upper panels) was maintained in the chimeric background. (B) HY TCR expression on gated donor DN and DP thymocytes isolated from chimeric mice. The HY TCRpos DP thymocyte population visible in chimeric mice generated using SLP-76+/– Tg+ HY+ bone marrow is contained predominantly within the CD4lowCD8low subset and most likely represents a population of cells at a stage of development just prior to deletion. Similar data were obtained from two independent sets of chimeric mice.


arrow
DISCUSSION
 
The models and mechanisms proposed to account for thymic selection and lineage commitment have been modified continuously as new and innovative approaches are applied to this intriguing area. One common element to each of these models has been the central role played by intracellular signals elicited by TCR engagement. Given the requisite role ascribed to the SLP-76 adaptor in promoting TCR-dependent signaling, we hypothesized that SLP-76 is likely a key contributor to the process of TCR-dependent thymic selection. The studies presented here support this idea and reveal two key elements regarding the role SLP-76 plays during thymocyte selection. First, globally reduced SLP-76 expression in the T cell compartment has a more pronounced effect on generation of CD4 SP thymocytes, consistent with the notion that positive selection of MHC class II-restricted thymocytes is normally more efficient than MHC class I-restricted thymocytes as a result of the more robust nature of CD4 coreceptor signaling when compared with CD8 [19 ]. Second, a substantial portion of HY TCR+ thymocytes escapes the early deletion, normally observed in male mice, when SLP-76 expression is governed solely by Tg-mediated reconstitution, highlighting the importance of maintaining SLP-76 expression from the endogenous loci to the selection process.

Although we propose that the skewed CD4:CD8 ratio arising in Line 2 SLP-76–/– Tg+ mice is most likely a result of a reduced efficiency of selection of MHC class II-restricted DP thymocytes, we cannot rule out enhanced selection of thymocytes bearing MHC class I-restricted TCRs under these conditions. This is likely not the case for several reasons. First, it is generally accepted that thymocytes expressing a MHC class II-restricted TCR recruit more Lck into the signaling complex via CD4 coreceptor engagement, as a higher proportion of CD4 molecules associates with Lck when compared with CD8 [20 ]. This is thought to result in a stronger or more sustained signal, which supports efficient positive selection. The weaker signals associated with CD8 coreceptor engagement are also sufficient to support positive selection, albeit with less efficiency than CD4-associated signaling [19 ]. The low levels of SLP-76 expression achieved in Line 2 SLP-76–/– Tg+ mice are predicted to depress signaling potential in the DP thymocyte compartment, an idea supported by our observation that reduced surface expression of CD5 on DP thymocytes correlates with reduced SLP-76 expression. It is therefore difficult to envision how reduced signaling potential would enhance positive selection of MHC class I-restricted thymocytes and generation of CD8 SP thymocytes, resulting in the skewed CD4:CD8 ratio observed in Tg-reconstituted SLP-76–/– mice derived from Line 2 Tg founders. It remains possible that reduced SLP-76 expression could affect the outcome of lineage commitment, especially if sustained signaling via CD4 coreceptor engagement is required to fully commit selected thymocytes to the CD4 lineage [19 ]. Compromised signaling potential as a consequence of reduced SLP-76 expression may mimic the developmental cues that normally promote adoption of a CD8 SP fate, resulting in redirection of MHC class II-restricted thymocytes into the CD8 lineage.

Our conclusion that positive selection of MHC class II-restricted thymocytes is more sensitive to reduced SLP-76 expression than selection of MHC class I-restricted thymocytes is supported further by experiments performed in mice harboring a mutant allele of SLP-76 that can be deleted in an inducible manner. Under conditions where SLP-76 expression is extinguished as developing thymocytes acquire a DP phenotype, positive selection and generation of CD4 and CD8 SP thymocytes are markedly impaired [14 ]. Still, a small subset of mature (heat-stable antigenlow) CD8 SP thymocytes was generated in these mice, and the CD4 SP thymocyte subset was virtually absent, consistent with the notion that positive selection of MHC class I-restricted thymocytes can occur in the context of reduced or even absent SLP-76 expression. In addition, Tg-mediated reconstitution of SLP-76–/– mice with signaling-compromised SLP-76 mutants results in reduced thymic cellularity and a skewing of the thymic CD4:CD8 ratio in a manner similar to that observed in Line 2 SLP-76–/– Tg+ mice [21 22 ].

Similar to our observations, the peripheral CD4+ T cells arising in the mice where SLP-76 expression was abolished in an inducible manner contained an elevated frequency of CD44+ cells, consistent with homeostatic expansion. It is curious that these peripheral CD4+ T cells still lacked SLP-76, suggesting the possibility that TCR-dependent homeostatic expansion in response to low-affinity self-ligands may not be as dependent on SLP-76 expression. In our model system, we noticed that the SP thymocytes generated in Line 2 SLP-76–/– Tg+ mice expressed elevated levels of SLP-76 when compared with DP thymocytes, suggesting two possible scenarios. First, only those thymocytes expressing a certain "threshold" level of Tg-encoded SLP-76 may have been selected efficiently. Second, Tg-encoded SLP-76 expression may have increased following positive selection, thus reflecting the expression pattern of endogenous SLP-76 [23 ]. Given the variegated expression of most Tgs, we currently favor the first scenario. Still, it is possible that SLP-76 expression may be regulated in a post-transcriptional manner during the selection process. A more detailed, quantitative analysis of SLP-76-specific transcripts encoded by the endogenous SLP-76 gene or the pLckSLP-76 Tg at distinct stages of thymocyte development may provide additional clues regarding how SLP-76 expression is regulated normally during the selection process.

In contrast to the inducible strain, the majority of our studies was conducted in the context of a SLP-76-deficient microenvironment. Although this complicates interpretation of the observed results, the observation that defective deletion of HY+ thymocytes is maintained in chimeric mice indicates a thymocyte-intrinsic effect. Restricting SLP-76 expression potential to the Tg could have several possible effects on the developing thymocyte. First, maintenance of the temporal pattern of SLP-76 expression achieved from the endogenous gene may be an important aspect of the selection process, a pattern that would likely not be supported by the Tg. It is worth noting that the 3' UTR of the SLP-76 mRNA contains multiple adenosine uridine-rich elements, which are not preserved in the Tg, suggesting at least one level of potential post-transcriptional regulation of the endogenous transcript. Second, mosaic expression of the Tg likely results in a range of SLP-76 expression among individual thymocytes. A common element to both lines of Tg-reconstituted SLP-76–/– mice is likely to be the presence of immature thymocytes that express low levels of SLP-76, although presumably, the frequency of such cells would be less in Line 1 SLP-76–/– Tg+ mice. This notion is supported by data presented in Figure 1C , where DN and DP thymocytes expressing low to negligible levels of SLP-76 are visible in both lines of Tg-reconstituted SLP-76–/– mice. In preliminary experiments, we have found that the DP thymocytes maintained in SLP-76–/– Tg+ HY+ male mice express low levels of SLP-76 (data not shown). In the context of the HY TCR Tg background, thymocytes expressing the lowest levels of Tg-encoded SLP-76 may not support efficient, early deletion of HY TCRhigh thymocytes. However, the same SLP-76low cells that fail to undergo deletion would likely fail positive selection as well, a notion consistent with the lack of CD8 SP thymocytes in both lines of Tg-reconstituted SLP-76–/– Tg+ HY+ (male) mice (Figs. 5A and 6) . It is also possible that the population of DN thymocytes expressing low levels of SLP-76 may not support efficient allelic exclusion, giving rise to a higher frequency of thymocytes expressing endogenous TCR chains and reduced levels of the clonotypic HY TCR. Although DN thymocytes expressing reduced levels of HY TCR (compared with control mice) are difficult to detect in Line 1 SLP-76–/– Tg+ mice, such cells are evident in the Line 2 SLP-76–/– Tg+ background (Fig. 5B) . Reduced expression of the HY TCR may therefore be limited to only those cells expressing low levels of SLP-76 and would be predicted to shift deletion to a more physiological stage of maturation, as has been observed when the HY TCR is expressed later during thymocyte maturation [24 ]. Collectively, the data presented here add to the growing body of evidence supporting an important role for SLP-76 in maintaining the integrity of thymic selection and establishment of self-tolerance. It will now be of substantial interest to determine if the temporal pattern of SLP-76 expression observed during thymocyte development helps to establish signaling potential and maintain the integrity of the selection process.


arrow
ACKNOWLEDGEMENTS
 
This work was supported by an Arthritis Foundation Investigator Award (J. L. C.) and a National Cancer Institute Core grant (RPCI). The authors thank Drs. Sandra Gollnick and Martha Jordan for helpful suggestions and critical evaluation of the manuscript.


arrow
FOOTNOTES
 
1 Current address: Institute for Immunology, Ludwig-Maximilians University, Munich, Germany. Back

Received May 11, 2007; revised August 23, 2007; accepted September 26, 2007.


arrow
REFERENCES
 
    1
  1. Haks, M. C., Pepin, E., van den Brakel, J. H., Smeele, S. A., Belkowski, S. M., Kessels, H. W., Krimpenfort, P., Kruisbeek, A. M. (2002) Contributions of the T cell receptor-associated CD3{gamma}-ITAM to thymocyte selection J. Exp. Med. 196,1-13[Abstract/Free Full Text]
    2. 2
  2. Sakaguchi, N., Takahashi, T., Hata, H., Nomura, T., Tagami, T., Yamazaki, S., Sakihama, T., Matsutani, T., Negishi, I., Nakatsuru, S., Sakaguchi, S. (2003) Altered thymic T-cell selection due to a mutation of the ZAP-70 gene causes autoimmune arthritis in mice Nature 426,454-460[CrossRef][Medline]
    3. 3
  3. Berg, L. J., Finkelstein, L. D., Lucas, J. A., Schwartzberg, P. L. (2005) Tec family kinases in T lymphocyte development and function Annu. Rev. Immunol. 23,549-600[CrossRef][Medline]
    4. 4
  4. Lucas, J. A., Atherly, L. O., Berg, L. J. (2002) The absence of Itk inhibits positive selection without changing lineage commitment J. Immunol. 168,6142-6151[Abstract/Free Full Text]
    5. 5
  5. Schaeffer, E. M., Broussard, C., Debnath, J., Anderson, S., McVicar, D. W., Schwartzberg, P. L. (2000) Tec family kinases modulate thresholds for thymocyte development and selection J. Exp. Med. 192,987-1000[Abstract/Free Full Text]
    6. 6
  6. Urdahl, K. B., Pardoll, D. M., Jenkins, M. K. (1994) Cyclosporin A inhibits positive selection and delays negative selection in {alpha} β TCR transgenic mice J. Immunol. 152,2853-2859[Abstract]
    7. 7
  7. Urdahl, K. B., Pardoll, D. M., Jenkins, M. K. (1992) Self-reactive T cells are present in the peripheral lymphoid tissues of cyclosporin A-treated mice Int. Immunol. 4,1341-1349[Abstract/Free Full Text]
    8. 8
  8. Davey, G. M., Schober, S. L., Endrizzi, B. T., Dutcher, A. K., Jameson, S. C., Hogquist, K. A. (1998) Preselection thymocytes are more sensitive to T cell receptor stimulation than mature T cells J. Exp. Med. 188,1867-1874[Abstract/Free Full Text]
    9. 9
  9. Pircher, H., Rohrer, U. H., Moskophidis, D., Zinkernagel, R. M., Hengartner, H. (1991) Lower receptor avidity required for thymic clonal deletion than for effector T-cell function Nature 351,482-485[CrossRef][Medline]
    10. 10
  10. Li, Q. J., Chau, J., Ebert, P. J., Sylvester, G., Min, H., Liu, G., Braich, R., Manoharan, M., Soutschek, J., Skare, P., Klein, L. O., Davis, M. M., Chen, C. Z. (2007) miR-181a is an intrinsic modulator of T cell sensitivity and selection Cell 129,147-161[CrossRef][Medline]
    11. 11
  11. Burack, W. R., Cheng, A. M., Shaw, A. S. (2002) Scaffolds, adaptors and linkers of TCR signaling: theory and practice Curr. Opin. Immunol. 14,312-316[CrossRef][Medline]
    12. 12
  12. Yoder, J., Pham, C., Iizuka, Y. M., Kanagawa, O., Liu, S. K., McGlade, J., Cheng, A. M. (2001) Requirement for the SLP-76 adaptor GADS in T cell development Science 291,1987-1991[Abstract/Free Full Text]
    13. 13
  13. Sommers, C. L., Lee, J., Steiner, K. L., Gurson, J. M., Depersis, C. L., El Khoury, D., Fuller, C. L., Shores, E. W., Love, P. E., Samelson, L. E. (2005) Mutation of the phospholipase C-{gamma}1-binding site of LAT affects both positive and negative thymocyte selection J. Exp. Med. 201,1125-1134[Abstract/Free Full Text]
    14. 14
  14. Maltzman, J. S., Kovoor, L., Clements, J. L., Koretzky, G. A. (2005) Conditional deletion reveals a cell-autonomous requirement of SLP-76 for thymocyte selection J. Exp. Med. 202,893-900[Abstract/Free Full Text]
    15. 15
  15. Clements, J. L., Yang, B., Ross-Barta, S. E., Eliason, S. L., Hrstka, R. F., Williamson, R. A., Koretzky, G. A. (1998) Requirement for the leukocyte-specific adapter protein SLP-76 for normal T cell development Science 281,416-419[Abstract/Free Full Text]
    16. 16
  16. Pivniouk, V., Tsitsikov, E., Swinton, P., Rathbun, G., Alt, F. W., Geha, R. S. (1998) Impaired viability and profound block in thymocyte development in mice lacking the adaptor protein SLP-76 Cell 94,229-238[CrossRef][Medline]
    17. 17
  17. Wildin, R. S., Garvin, A. M., Pawar, S., Lewis, D. B., Abraham, K. M., Forbush, K. A., Ziegler, S. F., Allen, J. M., Perlmutter, R. M. (1991) Developmental regulation of lck gene expression in T lymphocytes J. Exp. Med. 173,383-393[Abstract/Free Full Text]
    18. 18
  18. Azzam, H. S., Grinberg, A., Lui, K., Shen, H., Shores, E. W., Love, P. E. (1998) CD5 expression is developmentally regulated by T cell receptor (TCR) signals and TCR avidity J. Exp. Med. 188,2301-2311[Abstract/Free Full Text]
    19. 19
  19. Erman, B., Alag, A. S., Dahle, O., van Laethem, F., Sarafova, S. D., Guinter, T. I., Sharrow, S. O., Grinberg, A., Love, P. E., Singer, A. (2006) Coreceptor signal strength regulates positive selection but does not determine CD4/CD8 lineage choice in a physiologic in vivo model J. Immunol. 177,6613-6625[Abstract/Free Full Text]
    20. 20
  20. Wiest, D. L., Yuan, L., Jefferson, J., Benveniste, P., Tsokos, M., Klausner, R. D., Glimcher, L. H., Samelson, L. E., Singer, A. (1993) Regulation of T cell receptor expression in immature CD4+CD8+ thymocytes by p56lck tyrosine kinase: basis for differential signaling by CD4 and CD8 in immature thymocytes expressing both coreceptor molecules J. Exp. Med. 178,1701-1712[Abstract/Free Full Text]
    21. 21
  21. Kumar, L., Pivniouk, V., de la Fuente, M. A., Laouini, D., Geha, R. S. (2002) Differential role of SLP-76 domains in T cell development and function Proc. Natl. Acad. Sci. USA 99,884-889[Abstract/Free Full Text]
    22. 22
  22. Myung, P. S., Derimanov, G. S., Jordan, M. S., Punt, J. A., Liu, Q. H., Judd, B. A., Meyers, E. E., Sigmund, C. D., Freedman, B. D., Koretzky, G. A. (2001) Differential requirement for SLP-76 domains in T cell development and function Immunity 15,1011-1026[CrossRef][Medline]
    23. 23
  23. Clements, J. L., Ross-Barta, S. E., Tygrett, L. T., Waldschmidt, T. J., Koretzky, G. A. (1998) SLP-76 expression is restricted to hemopoietic cells of monocyte, granulocyte, and T lymphocyte lineage and is regulated during T cell maturation and activation J. Immunol. 161,3880-3889[Abstract/Free Full Text]
    24. 24
  24. Baldwin, T. A., Sandau, M. M., Jameson, S. C., Hogquist, K. A. (2005) The timing of TCR {alpha} expression critically influences T cell development and selection J. Exp. Med. 202,111-121[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
jlb.0507297v1
83/2/419    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ramsey, K.
Right arrow Articles by Clements, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ramsey, K.
Right arrow Articles by Clements, J. L.