|
|
||||||||
CUTTING EDGE |





* Department of Immunobiology, New Guys House, Guys, Kings and St. Thomass School of Medicine, London, United Kingdom;
Department Anatomy, Medical Research Council Center for Immune Regulation University of Birmingham, Birmingham, United Kingdom; and
Department of Cellular and Molecular Pharmacology, University of California San Francisco, CA 94143
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
An improved understanding ideally requires a capacity to regulate specific kinase activities within complex mammalian systems. In this regard, a recent technological innovation described ATP analog-sensitive kinase alleles (ASKAs)4 that can accommodate ATP and function normally, but that additionally have high affinity for large ATP analogs, such as 4-amino-1-tert-butyl-3-(1'-naphthyl)pyrazolo[3,4-d]pyrimidine (NaPP1), that compete for entry into the ATP binding site (3, 4). By contrast, the active site of natural kinases is too small to accommodate such inhibitors (3) (Fig. 1a). Thus, ASKA technology theoretically permits one to regulate the biological effects of any chosen kinase.
This notwithstanding, the application of ASKA technology to complex mammalian biology faces potentially serious uncertainties. These include the expression levels of ASKAs achievable in primary cells and whether the doses of inhibitor needed to gain access to relevant cells within a multicellular organ cause nonspecific biological dysregulation: in sum, will the achievable expression levels of the ASKAs and the practical concentrations of inhibitor combine to regulate kinase activity across a biologically relevant range? To resolve these issues, this article examines the practicality of ASKA technology for studying T cell development in reaggregate fetal thymic organ culture (RTOC).
Stages of thymocyte development are readily marked by CD4 and CD8 expression (5). Early CD4-CD8- "double-negative" (DN) thymocytes mature into CD8low cells (immature single-positive cells (ISPs)) that become CD4CD8 double-positive (DP) (6), and finally CD4+ or CD8+ single-positive (SP) cells. The DN-ISP-DP transition depends on the TCR
-chain contributing to a pre-TCR (7, 8, 9), while the DP-SP transition depends on TCR
contributing to a mature 
TCR. The pre-TCR and TCR
are each associated with the src-related kinases, Lck and Fyn (10, 11, 12).
Several lines of transgenic and gene knockout mice have indicated that the pre-TCR-dependent DN-ISP-DP transition is inhibited by Lck overexpression as well as by Lck deficiency or expression of a dominant-negative Lck (13, 14, 15). Nonetheless, conclusions are inevitably complicated by variability in the onset and durability of Lck transgene expression in the different mouse strains and by the uncertain relationship between Lck expression and Lck activity. Instead, it would be useful to show that different biological effects occur within a single system as Lck activity is incrementally altered across a continuous range. This seemed an ideal context in which to test the application of ASKA technology to a complex immunological process.
| Materials and Methods |
|---|
|
|
|---|
Lck alleles generated using QuikChange Site-directed mutagenesis (Stratagene, La Jolla, CA) were subcloned into the pMX vector (16). Briefly, 2 x 106 phoenix-E packaging cells were transfected with 20 µg of vector DNA in CaCl2 and 2x HBS (pH 7.05). Medium was changed 8 h after infection, and supernatant viral particles were harvested at 48, 72, and 96 h, passed through 45-µm filters; aliquoted, and stored at -70°C (17).
RTOC, transduction, and formation
Embryonic thymocytes and stromal cells were isolated and prepared as described previously (18). E14 thymocytes were centrifuged at 2100 rpm for 1 h, incubated with virus for 3 h, and pelleted by centrifugation with thymic epithelial cells at a ratio of 1:3. The resultant slurry was transferred by a finely drawn glass pipette to the surface of a 0.8-µm nucleopore filter (19). Cultures were harvested at indicated times and thymocytes were released from reaggregates using fine cataract knives.
Transduction and Western blot of thymoma BW5147
Briefly, 1.2 x 106 cells were transduced with control vector or Lcka-as and, after 48 h, gfp+ cells were flow cytometry sorted and expanded in culture with 0, 0.1, 1, 5, or 10 µM NaPP1 (Cellular Genomics, Branford, CT) for another 48 h before lysis in Nonidet P-40 buffer. One hundred micrograms of protein lysate was separated on 10% SDS gels, blotted onto nylon membranes, and probed using Abs against total Lck (Santa Cruz Biotechnology, Santa Cruz, CA) and phosphorylated Ysrc416 (Cell Signaling, Beverly, MA), respectively.
| Results and Discussion |
|---|
|
|
|---|
Mutation to glycine of Thr316 in the ATP binding site of Lck creates Lck-as (Fig. 1a), an allele predicted to bind ATP and function as a normal tyrosine kinase (20, 21), but that should also accommodate NaPP1, an enlarged analog of the Src family kinase inhibitor PP1 that competes with ATP for entry into the ATP binding site (3). Other naturally occurring kinases contain a large amino acid corresponding to T316 that precludes NaPP1. Following construction of Lck-as, we combined the T316 G mutation in cis with a Y505 F mutation (Lcka) that renders Lck constitutively active (22), to create Lcka-as
|
Lcka-as is biologically active and specifically inhibitable in wild-type (wt) RTOC
DN thymocytes were transduced with a virus expressing gfp only, or coexpressing Lcka, and Lcka-as, respectively, and incubated in RTOC for
5 days, by which time >50% of gfp+ thymocytes transduced with a gfp-only virus developed to the DP stage (Fig. 2a; in all plots, the x-axis depicts CD8 expression, the y-axis depicts CD4). Although DP development varied from 50% to 82% in different experiments (e.g., because of age variation in the fetuses used as sources of thymocytes and stroma, or variable time of harvest), results within single experiments were more consistent. Moreover, development was largely comparable in nontransduced cultures, "gfp-only" transduced cultures and nontransduced (gfp-) cells recovered from RTOC supporting gfp+ transduced cells (Fig. 2). Thus, neither transduction nor gfp expression substantially affects thymocyte development in RTOC.
|
The inhibitory effect of activated Lck is poorly understood, but it is not readily attributed to apoptosis since transgenic mice expressing increased levels of wt or activated Lck showed no obvious decrease in thymocyte numbers (15). Similarly, comparable numbers of gfp+ thymocytes were recovered from the gfp-only and from the Lcka-as-transduced cultures. The recovery of similar numbers of cells from cultures containing 10 µM NaPP1 indicated that inhibitor per se also does not affect cell viability (see Fig. 2a).
To determine whether T cell development shows a graded or an all-or-nothing threshold response to variation in Lck activity, different doses of inhibitor were added to RTOCs. In this experiment, the representation of DPs in control vector-transduced cultures ranged from 53 to 71%, whereas for Lcka-as-transduced cells it was
9% (Fig. 2c). Increasing NaPP1 gradually reduced Lck overactivity, incrementally restoring DP cells to 11% (0.5 µM NaPP1), 20% (2 µM), and
45% (10 µM) (Fig. 2c). The development of gfp- cells from the same RTOC was within the range of the controls (Fig. 2c) and comparable cell yields again suggested that cell viability was not greatly affected by Lcka-as or by 10 µM inhibitor (see legends to Figs. 2 and 3). Thus, the dose-dependent regulation of a Lck ASKA in RTOC readily exposed a quantitative correlation of T cell development with Lck activity across a continuous range.
|
42%) is seen in gfplow cultures with no inhibitor and in gfphigh cultures with 5 µM inhibitor. However, the enhanced thymocyte maturation, even of gfplow cells, at high inhibitor concentrations (
70% DPs) demonstrates that ASKA methodology can reduce Lck biological activity to below that achieved by gating for gfplow expression. In sum, the application of ASKA technology permits Lck activity to be regulated over an unprecedentedly broad, biologically relevant range in a complex organ culture. Lcka-as is biologically active and specifically inhibitable in mutant RTOC
The biological function of Lcka-as was independently assayed in RTOC using thymocytes from RAG1-deficient mice that cannot rearrange TCR
genes and therefore cannot undergo DN-ISP-DP maturation (23). Thus, gfp-only vector-transduced thymocytes remain >90% DN (Fig. 3a), whereas enforced Lck expression mimics pre-TCR signaling, promoting the appearance of ISPs and increasing DP representation from 0.2 to 16% (Fig. 3a). Although transgenic Lck was reported to rescue DP development more fully than this (24), those studies used a debilitated Lck. Instead, Lck overexpression in transgenic mice led primarily to ISP accumulation (15). This distortion in thymocyte development was exaggerated in Lcka and Lcka-as-transduced RTOCs in which DP rescue was only 57% (Fig. 3a). However, increasing doses of inhibitor achieved an incremental rescue of Lcka-as-transduced RAG1-/- DPs (13, 31, 37, 43%; Fig. 3b). Thus, ASKA technology permitted the dose-dependent regulation of thymocyte development by Lck in wt and mutant forms of RTOC.
Chemical genetic complementation of Lck deficiency across a complete range
Finally, the Lck-ASKA was examined for its capacity to complement the defective development of Lck-/- thymocytes. Lck-/- mice show partial inhibition of the DN-ISP-DP transition (
35% DP cells) because the defect in pre-TCR signaling is partly compensated by fyn (Fig. 4a and Ref13). Rather than rescue this deficiency, strong Lck overexpression permitted only 1% DPs, consistent with its inhibition of thymocyte development (Fig. 4d). However, by selecting for lower gfp expression (lower levels of Lck), DP representation went from 7% (Fig. 4c) to 27% (Fig. 4b) and to 47% (Fig. 4a). However, substantial rescue of DPs (46%) was also achievable among gfphigh cells, simply by adding 10 µM NaPP1 (Fig. 4d). Indeed, within a single NaPP1-treated RTOC, the complementation of Lck deficiency was achieved across a full biologic range. Thus, cells gated for "medium" gfp in the presence of inhibitor showed essentially complete rescue (
60% DP; Fig. 4, b and c), whereas gfplow cells defined thymocytes in which Lck activity was now insufficient, and in which DP representation fell back to 38% (Fig. 4a).
|
Although overexpressed Lck inhibits TCR
chain gene rearrangement (25), the inhibition of T cell development in Lck-overexpressing mice does not phenocopy RAG deficiency and its underlying nature has not been fully clarified. There is little evidence for excess apoptosis (see above), and it has been noted in transgenic studies and here that there is an associated increase in ISPs (15). The higher the Lck expression (gfphigh and 0 inhibitor; Fig. 4d), the greater was ISP accumulation (60%), whereas in the presence of inhibitor, DP representation exceeded ISP representation in all but the gfplow cells (Fig. 4). Analysis of forward scatter (Fig. 4) and other properties showed that the ISPs are activated blast cells, whereas DPs are mostly smaller and resting. Of note, as Lck activity was gradually increased, the ISPs were more activated, as reflected by increasing forward scatter (600750; Fig. 4). Therefore, increasing Lck activity in early thymocytes appears, at least in part, to skew pre-TCR signaling toward cell activation at the expense of differentiation. This seems consistent with the role of activated Lck in malignancy (22). The capacity to sort and analyze cells expressing different levels of Lck from ASKA-transduced cells cultured in different inhibitor concentrations now offers the potential to identify protein phosphorylation and/or gene expression patterns that correlate with particular levels of kinase activity and their associated biological outcomes.
In sum, we have combined a novel chemical genetic approach with retroviral transduction of RTOC to achieve rapid and efficient analysis of a kinase playing a crucial role in mammalian cells. Previous study of the dose-dependent effects of Lck on T cell development required the labor-intensive generation of multiple lines of transgenic mice expressing different levels of an Lck transgene (15). Although those studies provided a firm foundation to validate the use of ASKA as a tool in developmental immunology, it is already clear from Fig. 2d that the range of biological effects of Lck achievable with ASKA technology exceeds that obtained by selecting cells stochastically expressing different levels of a kinase. Indeed, whereas identification of cell lines or transgenic mice expressing a biologically relevant level of a transgene can be problematic, the capacity to use ASKAs to regulate kinase activity may allow one to use mice or cell lines grossly overexpressing a particular kinase. The relative ease of ASKA construction (successfully applied to v-Src, CDK2, Fus3, CAMKII
(4)), the specificity of the inhibitors effects; the broad range of kinase activity achievable, and the closer correlation of biological outcome with kinase activity, as opposed to inducible levels of gene expression (26, 27) suggest that ASKA technology should be usefully applied to many kinases emerging from the expression profiling of lymphoid cells (28, 29) and from genome sequencing.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 G.A. and A.H. contributed equally to this study. ![]()
3 Address correspondence and reprint requests to Dr. Adrian Hayday, Department of Immunobiology, New Guys House, Guys Kings, and St. Thomass School of Medicine, London SE1 9RT, United Kingdom. E-mail address: adrian.hayday{at}kcl.ac.uk ![]()
4 Abbreviations used in this paper: ASKA, analog-sensitive kinase allele; NaPP, 4-amino-1-tert-butyl-3-(1'-naphthyl)pyrazolo[3,4-d]pyrimidine; RTOC, reaggregate fetal thymic organ culture; DN, double negative; ISP, immature single positive; DP, double positive; SP, single positive; wt, wild type. ![]()
Received for publication January 15, 2003. Accepted for publication May 13, 2003.
| References |
|---|
|
|
|---|
chain genes in thymocytes: a critical role for the
chain in development. Cell 73:513.[Medline]
gene in development of 
but not 
T cells. Nature 375:795.[Medline]
-chain gene rearrangement by overexpression of the non-receptor protein tyrosine kinase p56lck. EMBO J. 11:4877.[Medline]

+ and TCR
+ intraepithelial lymphocytes provided by serial analysis of gene expression (SAGE). Immunity 15:419.[Medline]

intraepithelial lymphocytes as suggested by their transcriptional profile. Proc. Natl. Acad. Sci. USA 98:10261.This article has been cited by other articles:
![]() |
S. F. Steinberg Structural Basis of Protein Kinase C Isoform Function Physiol Rev, October 1, 2008; 88(4): 1341 - 1378. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Wong, J. McLaughlin, D. Cheng, C. Zhang, K. M. Shokat, and O. N. Witte Sole BCR-ABL inhibition is insufficient to eliminate all myeloproliferative disorder cell populations PNAS, December 14, 2004; 101(50): 17456 - 17461. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Giannini and M.-J. Bijlmakers Regulation of the Src Family Kinase Lck by Hsp90 and Ubiquitination Mol. Cell. Biol., July 1, 2004; 24(13): 5667 - 5676. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |