of October 27, 2017.This information is current as

ProgressionIL-2-Dependent Regulation of Cell CycleExpansion Via IL-2-Independent and CD28 Costimulation Mediates T Cell

Vassiliki A. BoussiotisLeonard J. Appleman, Alla Berezovskaya, Isabelle Grass and

http://www.jimmunol.org/content/164/1/144doi: 10.4049/jimmunol.164.1.144

2000; 164:144-151; ;J Immunol 

        average*  

4 weeks from acceptance to publicationSpeedy Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

?The JIWhy

Referenceshttp://www.jimmunol.org/content/164/1/144.full#ref-list-1

, 44 of which you can access for free at: cites 77 articlesThis article

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2000 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

CD28 Costimulation Mediates T Cell Expansion ViaIL-2-Independent and IL-2-Dependent Regulation of Cell CycleProgression1

Leonard J. Appleman, Alla Berezovskaya, Isabelle Grass, and Vassiliki A. Boussiotis2

In the presence of TCR ligation by Ag, CD28 pathway mediates the most potent costimulatory signal for T cell activation, cytokinesecretion, and T cell expansion. Although CD28 costimulation promotes T cell expansion due to IL-2 secretion and subsequentsignaling via the IL-2 receptor, recent studies indicate that the dramatic T cell expansion mediated through the unopposed CD28stimulation in CTLA4-deficient mice is IL-2 independent. Therefore, we sought to dissect the effects of CD28 and IL-2 receptorpathways on cell cycle progression and determine the molecular mechanisms by which the CD28 pathway regulates T cellexpansion. Here we show that CD28 costimulation directly regulates T cell cycle entry and progression through the G1 phase inan IL-2-independent manner resulting in activation of cyclin D2-associated cdk4/cdk6 and cyclin E-associated cdk2. Subsequentprogression into the S phase is mediated via both IL-2-dependent and IL-2-independent mechanisms and, although in the absenceof IL-2 the majority of T cells are arrested at the G1/S transition, a significant fraction of them progresses into the S phase. Thekey regulatory mechanism for the activation of cyclin-cdk complexes and cell cycle progression is the down-regulation of p27kip1

cdk inhibitor, which is mediated at the posttranscriptional level by its ubiquitin-dependent degradation in the proteasome path-way. Therefore, CD28 costimulation mediates T cell expansion in an IL-2-independent and IL-2 dependent manner and regulatescell cycle progression at two distinct points: at the early G1 phase and at the G1/S transition. The Journal of Immunology,2000,164: 144–151.

T he most critical costimulatory signal for the productiveoutcome of the immune response is provided by the mem-bers of B7 family, B7-1 (CD80) and B7-2 (CD86) (1–4).

B7 costimulation in the presence of a suboptimal TCR signal re-sults in increased transcription and translation of multiple cyto-kines, up-regulation of IL-2 receptora-, b-, andg-chains (5–7), Tcell expansion, and effector function (8). The biologic significanceof this pathway has been well established in multiple in vivo mu-rine models, clearly demonstrating an active role of B7 in thegeneration of autoimmunity (9–11), tumor immunity (12–14), andallograft rejection (15). Moreover, blockade of the B7:CD28 co-stimulatory pathway has been shown to ameliorate autoimmunediseases (16), and inhibit humoral immunity (17) and alloreactivity(18, 19).

Although the functional role of B7:CD28 costimulation is wellestablished, the molecular mechanisms by which this pathway reg-ulates T cell expansion have not been determined. Because CD28costimulation induces increased IL-2 secretion, it has been hypoth-esized that CD28 mediates clonal expansion through accumulationof IL-2 and subsequent signaling via the IL-2 receptor pathway(20). However, several lines of evidence accumulated from studies

on CD28-, IL-2-, and CTLA4-deficient mice suggest that addi-tional IL-2-independent cell cycle regulatory mechanisms mayalso be mediated via CD28 costimulation. T cells from IL-2-defi-cient mice have reduced, but significant, proliferative T cell re-sponses to T cell lectin Con A, which can be fully restored byaddition of IL-2 (21). In contrast, T cells from CD28-deficientmice have dramatically impaired proliferative response and IL-2secretion in response to Con A, which is only partially restored bythe addition of exogenous IL-2 (22), suggesting that the profoundloss of the ability for T cell expansion in CD28-deficient mice isnot simply due to the lack of IL-2 production. More importantly,the dramatic proliferation and activation of T cells in the CTLA4-deficient mice, which illustrates the physiologic consequences ofunrelenting B7/CD28-mediated T cell activation, is not due to in-creased production of IL-2 protein or mRNA as compared withnormal control mice (23). These observations strongly suggest thatIL-2-independent mechanisms are involved in CD28-mediated Tcell expansion. In light of the above observations, we sought todissect the effects of CD28 and IL-2 receptor-mediated pathwayson the cell cycle and determine the mechanisms by which CD28costimulation regulates T cell expansion.

Cell cycle progression is a complex process that is activated bycyclins that associate with catalytically active cyclin-dependentkinases (cdks)3 to form active holoenzymes and is inhibited by cdkinhibitors (24). Induction of D-type cyclins occurs during G1

phase, induction of cyclin E at the late G1 restriction point, andinduction of cyclin A at the S phase entry (24, 25). The orderlyprogression of the cells through the cell cycle is controlled by thetimely expression of cyclins, the activation of cdk enzymaticactivity, and the subsequent phosphorylation of the relevant

Department of Adult Oncology, Dana-Farber Cancer Institute, Division of MedicalOncology, Brigham and Women’s Hospital, Department of Medicine, Harvard Med-ical School, Boston, MA 02115

Received for publication August 24, 1999. Accepted for publication October19, 1999.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Institutes of Health Grants AI 43552, HL54785, and AI 41584.2 Address correspondence and reprint requests to Dr. Vassiliki A. Boussiotis, Dana-Farber Cancer Institute, Smith 852, 44 Binney Street, Boston, MA 02115. E-mailaddress: vassiliki_boussiotis@dfci.harvard.edu

3 Abbreviations used in this paper: cdk, cyclin-dependent kinase; Rb, retinoblastoma;PVDF, polyvinylidene difluoride; Ubal, ubiquitin aldehyde; MAPK, mitogen-acti-vated protein kinase; PTK, protein tyrosine kinase; PI-1, proteinase inhibitor-1.

Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

substrates, one of which is the retinoblastoma (Rb) gene product(26). Hyperphosphorylation renders Rb protein incapable of bind-ing E2F-type transcription factors, resulting in activation of tran-scription of S phase genes (27–29).

In the results to be reported below, we demonstrate that CD28costimulation directly regulates T cell cycle entry and progressionthrough the G1 phase in an IL-2-independent manner by down-regulating p27kip1 cdk inhibitor, resulting in activation of cyclinD2-associated cdk4/cdk6 and cyclin E-associated cdk2. Subse-quent progression into the S phase is mediated via both IL-2-de-pendent and IL-2-independent mechanisms, indicated by the factthat, although in the absence of IL-2 the majority of T cells arearrested at the G1/S transition, a significant fraction of themprogresses into the S phase. These results show that CD28 co-stimulatory signals mediate entry of T cells into the cell cycle andrender them competent for progression into the S phase and clonalexpansion in an IL-2 dependent and IL-2-independent manner.

Materials and MethodsT cell preparation and culture

Leukocytes were obtained from normal healthy volunteers by leukaphere-sis. Mononuclear cells were isolated by Ficoll/Hypaque gradient centrifu-gation (Pharmacia LKB Biotechnology, Piscataway, NJ). Monocytes weredepleted by adherence on plastic. The CD281 T cell population was furtherenriched by separation from residual monocytes, B cells, and NK cells bymAb and magnetic bead depletion using mAbs anti-CD14 (Mo2), anti-CD11b (Mo1), anti-CD20 (B1), anti-CD16 (3G8), and anti-IL-2Ra(CD25), which have been previously described and were produced in ourlaboratory (30). The efficiency of purification was analyzed in each case byflow cytometry (Epics Elite; Coulter Electronics, Hialeah, FL), using anti-CD3, anti-CD14, and anti-CD28 mAbs, followed by FITC-conjugated goatanti-mouse Ig (Fisher Scientific, Pittsburgh, PA). After separation, T cellswere cultured in 24-well plates at 23 106 cells/ml in complete mediumconsisted of RPMI 1640 supplemented with 10% heat inactivated FCS, 2%glutamine, 1% penicillin/streptomycin at 37°C with 5% CO2. Anti-CD3(OKT3, IgG1; American Type Culture Collection, Manassas, VA) mAbwas precoated on plastic at a concentration of 0.5mg/ml. Anti-CD28 mAb(3D10, IgG1) was coated on affigel beads and was used at a final concen-tration of 1:10,000. IL-2 was used at 50 U/ml, and anti-IL-2-neutralizingmAb (R&D Systems, Minneapolis, MN) was used at 1mg/ml, which was10-fold the concentration required to inhibit the maximum IL-2-mediatedproliferation (plateau response).

Immunoblotting, immunoprecipitation, and in vitro kinasereactions

Following the indicated conditions and time intervals of culture, cell ly-sates were prepared, and equal amounts of protein (50mg/sample) wereanalyzed by 10% SDS-PAGE, transferred onto nitrocellulose membranes,and immunoblotted with the indicated mAbs or antiserum. Cyclin A, cyclinD2, cyclin E, cdk2, cdk4, cdk6, p15INK4b, and p16INK4a antiserum werepurchased from Santa Cruz Biotechnology (Santa Cruz, CA), p21cip1 mAbfrom Upstate Biotechnology (Lake placid, NY), and p27kip1 mAb fromTransduction Laboratories (Lexington, KY). To examine the phosphory-lation status of Rb, proteins were analyzed by 6% SDS-PAGE, transferredonto nitrocellulose membrane, and blotted with Rb-specific mAb (PharM-ingen, San Diego, CA). After immunoblotting with mAbs or antiserum,immunodetection was performed by incubation with HRP-conjugated anti-mouse IgG (1:5000) or anti-rabbit IgG (1:10,000) (Promega, Madison, WI)as indicated by the host origin of the primary Ab and developed by chemi-luminescence (NEN, Boston, MA). Stripping and reprobing of the immu-noblots were done as described (31). Where indicated, quantitation of theproteins was performed by computing densitometry using AlphaImagerSoftware (San Leandro, CA).

For in vitro kinase reactions, immunoprecipitations were done usingequal amounts of protein (500mg/sample) with anti-cdk2-specific anti-serum agarose conjugate (Santa Cruz Biotechnology), and in vitro kinasereactions were performed using histone H1 (Sigma, St. Louis, MO) asexogenous substrate, according to described protocol (32). After immuno-precipitation with anti-cdk4-specific antiserum, in vitro kinase reactionswere performed using Rb-GST (Santa Cruz Biotechnology) as exogenoussubstrate. Reactions were analyzed by 10% SDS-PAGE, transferred toPVDF membrane, and exposed to x-ray film.

Ubiquitin aldehyde (Ubal) and the proteasome inhibitors MG-132 andproteinase inhibitor-1 (PI-1) were reconstituted in DMSO according to themanufacturer’s instructions (Calbiochem, La Jolla, CA) and used at thefollowing concentrations: Ubal at 5mM, MG-132 at 10mM, and PI-1 at15 mM.

RT-PCR

T cells were cultured in 24-well plates with anti-CD3 and anti-CD28 mAbsas described above, and, after 12 and 24 h of culture, T cells were isolatedand used for RNA preparation. Two micrograms of RNA were used forreverse transcription, and PCR amplification of cDNA from 2mg of mRNAwas performed as previously described (7).Quantitative PCR (multiplex)was conducted as previously described (33, 34) with two sets of primers withsequences as follows: p27kip1, 59-CCATGTCAAACGTGCGAGTGT-39 and39-CGTTTGACGTCTTCTGAGG-59; G3PDH, 59-TGAAGGTCGGAGTCAACGGATTTGGT-39 and 59-CATGTGGGCCATGAGGTCCACCAC-39.PCR products were electrophoresed through 2% agarosegels containingethidium bromide and visualized under UV light. Images were digitalizedand quantitated with image analysis software. Intensity of the PCR productderived from p27kip1 was divided over G3PDH, which was the internalcontrol for the quantitative PCR.

Cell cycle analysis by flow cytometry

Cell cycle analysis was performed as described before (35). Briefly, afterculture under various conditions, cells were harvested at the indicated timepoints, and 53 105 cells per sample were resuspended in 1 ml PBS. Cellswere fixed with ethanol, propidium iodide was added at a final concentra-tion of 2.5mg/ml, ribonuclease was added at 50mg/ml, and samples wereincubated for 30 min at 37°C in the dark. Analysis was performed by FACS(Lysis II software; Becton Dickinson, Mountain View, CA).

ResultsCD28 costimulation mediates IL-2-independent entry to the G1

phase of the cell cycle

Early stages of G1 phase are controlled by D-type cyclins and theircdks, cdk4 and its homologue cdk6 (25, 36). Among theD-typecyclins, T cells constitutively express very low levels of cyclin D2,which is rapidly up-regulated after entry into the G1 phase of thecell cycle (25). To examine whether CD28 and IL-2 receptor path-ways could independently induce entry into the G1 phase, expres-sion of cyclin D2 was examined. T cells were stimulated withsubmitogenic concentrations of anti-CD3 mAb alone or with eitherexogenous IL-2 or anti-CD28 mAb. Expression of cyclin D2 wasinduced at low levels (2- to 3-fold) by submitogenic anti-CD3mAb (Fig. 1A, top panel, lanes 2-4) and was significantly aug-mented (8- to 12-fold) in the presence of IL-2 (Fig. 1A, top panel,lanes 5-7). Increase of cyclin D2 by IL-2 peaked at 72 h of culture,and the augmentation was reverted in the presence of anti-IL-2-neutralizing mAb (Fig. 1A, top panel, lanes 8-10). Cyclin D2 wasalso up-regulated (7- to 10-fold) by costimulation with anti-CD28mAb (Fig. 1A, top panel, lanes 11-13), which was only slightlyinhibited in the presence of anti-IL-2-neutralizing Ab (Fig. 1A, toppanel, lanes 14-16). Neither IL-2 nor anti-CD28 mAb alone in-duced cyclin D2 expression in the absence of anti-CD3 mAb (datanot shown). These results show that, in the presence of submito-genic stimulation via the TCR/CD3 complex, additional signalsmediated independently through the IL-2 and the CD28 pathwayactivate entry into the G1 phase of the cell cycle, as indicated bysynthesis of cyclin D2.

To drive cell cycle progression, cyclins have to associate withthe activated isoforms of specific cdks to form active holoenzymes(24). Cdks are constitutively expressed at low levels in unstimu-lated T cells and are up-regulated after activation (37).D-type cy-clins associate with cdk4 and its homologue cdk6. In light of theresults described above, we sought to determine whether CD28costimulation could mediate IL-2-independent signals to up-regu-late and activate cdk4 and cdk6, which could then facilitate pro-gression through the G1 phase of the cell cycle. Up-regulation of

145The Journal of Immunology

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

cdk4 (2- to 3-fold) and cdk6 (2- to 3-fold) protein expression wasinduced by submitogenic concentration of anti-CD3 mAb (Fig. 1A,middleandbottom panels,lanes 2-4). However, under these con-ditions, although cdk4 expression was up-regulated, its kinase ac-tivity was not activated (Fig. 1B, lanes 2-4). Similarly, cdk6 kinaseactivity was not activated (data not shown). Addition of IL-2slightly augmented the expression of cdk4 (2- to 3-fold) and cdk6(#2-fold), which peaked at 72 h of culture (Fig. 1A, middleandbottom panels, lane 7) but more importantly, induced cdk4 enzy-matic activity (Fig. 1B, lanes 6and7), which peaked at 72 h ofculture. Addition of anti-IL-2-neutralizing mAb markedly reducedIL-2-induced cdk4 and cdk6 up-regulation (Fig. 1A, middle andbottom panels,lanes 8-10) but more importantly abolished cdk4activation (Fig. 1B,lanes 8-10). Similar results were observedwhen cdk6 activation was examined (data not shown).

CD28 costimulation not only up-regulated (4- to 5-fold) cdk4expression (Fig. 1A, middle panel,lanes 11-13) but also inducedits activation (Fig. 1B,lanes 11-13), which peaked at 48 h of cul-ture. Anti-IL-2 mAb only slightly inhibited CD28-mediated up-regulation of cdk4 expression (Fig. 1A, middle panel,lanes 14-16)and activation (Fig. 1B, lanes 14-16). Similar results were ob-served when expression (Fig. 1A, bottom panel,lanes 11-16) andactivation (data not shown) of cdk6 were examined, suggestingthat IL-2-independent mechanisms that augment the expressionand regulate the enzymatic activation of these cdks are mediatedby CD28 costimulation. Taken together, these results indicate thatCD28-mediated signals can induce IL-2-independent entry to thecell cycle characterized by synthesis of cyclin D2. Moreover,CD28-mediated signals can induce IL-2-independent enzymaticactivation of cyclin D2-associated cdks, cdk4, and cdk6, therebyinducing progression of the T cells through the early stages of G1

phase of the cell cycle.

CD28 costimulation induces IL-2-independent passage of T cellsthrough the G1 restriction point into late G1

Although cyclin D2 and its associated cdks, cdk4 and cdk6, reg-ulate entry and progression through the early stages of the G1

phase of the cell cycle, passage through the G1 restriction point

into late G1 requires synthesis of cyclin E and activation of cdk2,resulting in the formation of enzymatically active cyclin E-cdk2complexes (38). Expression of cyclin E was induced by prolongedincubation (72 h) with submitogenic concentration of anti-CD3mAb (Fig. 2A, top panel, lane 4). Induction of cyclin E was sig-nificantly augmented in the presence of IL-2 and peaked at 72 h ofculture (Fig. 2A, top panel,lanes 5-7). Expression of cdk2 (Fig.2A, bottom panel,lanes 2-4) but not activation (Fig. 2B,lanes 2-4)was induced by 72 h of culture with submitogenic concentration ofanti-CD3 mAb. IL-2 augmented (2- to 4-fold) the expression (Fig.2A, bottom panel,lane 7) and induced the activation (Fig. 2B,lanes 5-7) of cdk2, which peaked at 72 h of culture. CD28 co-stimulation up-regulated expression of cyclin E (Fig. 2A, toppanel, lanes 11-13) as well as expression (3- to 5-fold) and acti-vation of cdk2 (Fig. 2A, bottom panel, lanes 11-13and Fig. 2B,lanes 11-13) within a shorter time interval of culture and resultedin a peak at 48 h.

Expression of cyclin E and cdk2 as well as activation of cdk2induced by IL-2 were inhibited in the presence of anti-IL-2 mAb(Fig. 2, A andB, lanes 8-10). In contrast, anti-IL-2 mAb had al-most no effect on CD28-mediated expression of cyclin E and cdk2(Fig. 2A, topandbottom panels, comparelanes 11-13with lanes14-16), but induced a partial but reproducible inhibition on CD28-mediated activation of cdk2 (Fig. 2B, comparelanes 11-13withlanes 14-16). The induction of cyclin E and enzymatically activecdk2 by CD28 costimulation, which is only partially inhibited byanti-IL-2 neutralizing mAb, is consistent with the ability of CD28pathway to provide an IL-2-independent signal, sufficient to induceT cells to transit through the G1 restriction point into the late G1phase.

CD28 costimulation is capable of inducinghyperphosphorylation of Rb in vivo

Activated cdk2 synergizes with cdk4 and cdk6 to hyperphospho-rylate and inactivate Rb, thereby releasing E2F-type transcriptionfactors and allowing synthesis of S phase genes (26–29). To de-termine whether our in vitro findings on cdk4, cdk6, and cdk2activation correspond to the in vivo events, the independent effectof CD28 and IL-2 receptor pathways on phosphorylation of en-dogenous Rb was examined. Submitogenic concentration of anti-CD3 up-regulated Rb protein expression but did not induce its

FIGURE 1. CD28 costimulation induces expression of cyclin D2 andactivation of cyclin D2-associated cdks.A, CD281 T cells were cultured asdescribed inMaterials and Methodsfor the indicated time intervals andculture conditions. Cell lysates were prepared, and equal amounts of pro-tein were analyzed by 10% SDS-PAGE and blotted with cyclin D2-specificAb. Blots were stripped and reblotted with antiserum specific for cdk4 andcdk6. Results are representative of four experiments.B, Equal amounts ofprotein (500mg/sample) were immunoprecipitated with antiserum specificfor cyclin cdk4, and in vitro kinase reactions were performed usingGST-Rb as exogenous substrate. Samples were analyzed by 10% SDS-PAGE, transferred to PVDF membrane, and exposed to x-ray film. Resultsare representative of three experiments.

FIGURE 2. CD28 costimulation induces expression cyclin E and acti-vation of cyclin E-associated cdk2.A, CD281 T cells were cultured asindicated, cell lysates were prepared, and equal amounts of protein wereanalyzed by 10% SDS-PAGE and blotted with antiserum specific for cyclinE. Blots were stripped and reblotted with antiserum specific for cdk2. Re-sults are representative of four experiments.B, Equal amounts of protein(500 mg/sample) were immunoprecipitated with antiserum specific forcdk2, and in vitro kinase reactions were performed using histone H1 asexogenous substrate. Samples were analyzed by 10% SDS-PAGE, trans-ferred to PVDF membrane, and exposed to x-ray film. Results are repre-sentative of three experiments.

146 CELL CYCLE PROGRESSION BY CD28 COSTIMULATION

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

phosphorylation (Fig. 3,lanes 2-4). In the presence of submito-genic concentration of anti-CD3, IL-2 induced a significant phos-phorylation of Rb, as determined by the shift of its electrophoreticmobility (Fig. 3, lanes 5-7), which was prevented by neutralizinganti-IL-2 mAb (Fig. 3,lanes 8-10). The effect of IL-2 on Rb hy-perphosphorylation peaked at 72 h of culture (Fig. 3,lane 7),which temporally coincided with the maximum simultaneous ex-pression and activation of cdk4, cdk2 (Fig. 1,A andB, and Fig. 2,A andB), and cdk6 (data not shown), induced by IL-2.

CD28 costimulation also induced hyperphosphorylation of Rb,which peaked at 48 h of culture (Fig. 3,lanes 11-13), which tem-porally coincided with the simultaneous expression and activationof all cdks induced by CD28 costimulation (Fig. 1,A andB, andFig. 2,A andB) and was significantly but not completely inhibitedby the presence of the anti-IL-2-neutralizing Ab (Fig. 3,lanes14-16). It is of note that the most potent inhibitory effect of anti-IL-2-neutralizing mAb on CD28-mediated phosphorylation of Rbwas detected at 72 h of culture (Fig. 3, comparelanes 13and16),which temporally coincided with the inhibitory effect of anti-IL-2-neutralizing mAb on CD28-mediated activation of cdk2 (Fig.2B, lane 16). Taken together, the above results indicate that CD28costimulation is capable of providing a signal sufficient to inducehyperphosphorylation of Rb and, therefore, primes T cells for en-try into the S phase of the cell cycle.

CD28 costimulation induces down-regulation of p27kip1 cdkinhibitor

Activation of cyclin/cdk holoenzyme is regulated by the presenceof specific cdk inhibitors that associate with the cyclin/cdk com-plex (39). The major inhibitors of cyclin/cdk complexes are themembers of the INK (p15INK4b, p16INK4a, p18INK4c, and p19INK4d)and the cip/kip (p21cip1, p27kip1 and p57kip2) families (40–46).INK family members specifically bind to and inhibit D-type cyc-lins complexed with cdk4 and cdk6, whereas p21cip1 and p27kip1

can inhibit many cyclin/cdk complexes. Because submitogenic anti-CD3 stimulation induced expression but not activation of cdks,whereas IL-2 receptor and CD28 pathway induced expression andalso activation of these cdks (Figs. 1 and 2), we sought to deter-mine whether the expression of cdk inhibitors under the variousconditions would account for this difference. The representativemembers of the INK family, p15INK4b and p16INK4a, were notdetected in any of the conditions tested (Fig. 4,first and secondpanels). Both IL-2 receptor and CD28-mediated signals in thepresence of submitogenic TCR stimulation resulted in increasedlevels of p21cip1 expression (Fig. 4,third panel, lanes 5-7 and11-13), which temporally coincided with the maximum activationof cdks and phosphorylation of Rb (Fig. 1,A andB, and Fig. 3).The paradoxical increase in the cell cycle inhibitor p21cip1 duringT cell expansion is consistent with results from other experimental

systems. Because p21cip1 also interacts with the proliferating cellnuclear Ag (PCNA), leading to inhibition of DNA elongation viapolymerased, the increase of p21cip1 in proliferating cells has beenhypothesized to represent a brief arrest of the cycling cells to allowDNA repair (47–50). Neutralizing anti-IL-2 mAb induced com-plete blockade of IL-2 (Fig. 4,third panel, comparelanes 5-7withlanes 8-10) but only partial blockade of CD28-mediated inductionof p21cip1 expression (Fig. 4,third panel, comparelanes 11-13with lanes 14-16).

Submitogenic stimulation by anti-CD3 did not affect the expres-sion of p27kip1 (Fig. 4, third panel, lanes 2-4). Addition of IL-2under these conditions resulted in down-regulation of p27kip1 pro-tein expression, which was most profound at 72 h of culture (Fig.4, forth panel, lanes 5-7), and this down-regulation was preventedby anti-IL-2-neutralizing mAb (Fig. 4,forth panel, comparelanes5-7with lanes 8-10). Costimulation through CD28 induced a morerapid reduction of p27kip1 protein expression (Fig. 4,fourth panel,lanes 11-13), which was detectable by 24 h of culture, and wasonly partially inhibited by anti-IL-2-neutralizing Ab (Fig. 4,fourthpanel, comparelanes 11-13to lanes 14-16).

These results indicate that, although submitogenic anti-CD3stimulation induces expression of cdk4, cdk6, and cdk2, it fails toinduce enzymatic activity of these cdks, because the expression ofp27kip1 remains high. IL-2- and CD28-mediated enzymatic acti-vation of cdks, which is required for the formation of active cyclin-cdk complexes and cell cycle progression, is not simply due to theup-regulation of the expression of cyclins and cdks but also due tothe down-regulation of p27kip1 cdk inhibitor, which releases cyclinD2-cdk4/6 and cyclin E/cdk2 complexes from this constraint, re-sulting in their enzymatic activation. Moreover, these results showthat, in addition to the well known IL-2-mediated down-regulationof p27kip1, CD28 pathway activates IL-2-independent mechanismsthat result in down-regulation of this cell cycle inhibitor.

Down-regulation of p27 by CD28 costimulation is controlled atposttranscriptional level and is mediated via its ubiquitinationand degradation in the proteasome pathway

There have been conflicting reports as to whether expression ofp27kip1 during the cell cycle is regulated at the level of protein ormRNA (51–55). It is possible that distinct regulatory mechanismsare operative in different cell types. Therefore, we sought to de-termine the mechanism by which CD28 costimulation regulates

FIGURE 3. CD28 costimulation induces hyperphosphorylation of Rb invivo. T cells were cultured for the indicated time intervals and cultureconditions, lysates were prepared, and equal amounts of proteins from eachsample were analyzed by 6% SDS-PAGE. Proteins were transferred ontonitrocellulose membrane, and blots were incubated with anti-Rb mAb, fol-lowed by incubation with HRP-conjugated anti-mouse IgG (1:5000) (Pro-mega) and detection by chemiluminescence (NEN). Results are represen-tative of three experiments.

FIGURE 4. CD28 costimulation induces down-regulation of p27kip1

cdk inhibitor. Immunoblots used in Fig. 2A were stripped and reblottedwith the indicated Abs or antiserum specific for p15INK4b, p16INK4a,p21cip1, and p27kip1 followed by incubation with HRP-conjugated anti-mouse IgG (1:5,000) or peroxidase-conjugated anti-rabbit IgG (1:10,000)(Promega) as indicated by the primary Ab used followed by detection withchemiluminescence (NEN).

147The Journal of Immunology

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

the expression of p27kip1 protein in T cells. Quantitative RT-PCRanalysis demonstrated that p27kip1 mRNA was not reduced (Fig.5A) by CD3- and CD28-mediated stimulation although this causedsignificant decrease in p27kip1 protein expression. These data sug-gest that reduction in p27kip1 protein by CD28 costimulation ismediated through posttranscriptional mechanisms.

In other experimental systems, posttranscriptional regulation ofp27kip1 has been shown to occur through enzymatic ubiquitinationfollowed by targeted proteolysis in the proteasome pathway (54,56). To examine whether CD28-mediated down-regulation ofp27kip1 involves ubiquitin-targeted proteolysis, cultures of T cellswith anti-CD3 and anti-CD28 were established in either mediaalone or in the presence of Ubal. Ubal is a potent and specificinhibitor of multiple ubiquitin hydrolases involved in pathways ofintracellular protein ubiquitin-dependent modification and turn-over and decreases the rate of ubiquitin-dependent degradation(57). The presence of Ubal prevented the down-regulation ofp27kip1 (Fig. 5B), suggesting that ubiquitination of p27kip1 is amandatory step for its down-regulation by TCR- and CD28-me-diated signals.

To examine whether CD28 costimulation mediated degradationof p27kip1 in the proteasome complex, two different proteasomeinhibitors (MG-132 and PI-1) were used. Stimulation of T cells byCD3 and CD28 in the presence of either inhibitor resulted in thegeneration of multiple bands detected by the p27kip1-specific Ab(Fig. 5C). The electrophoretic mobility of these proteins was con-sistent with previously reported ubiquitinated forms of p27kip1 (54,56). Importantly, the generation of these ubiquitinated forms wasobserved even in the presence of neutralizing anti-IL-2 Ab (Fig.5C, lanes 6-8). These results show that signals mediated directlythrough CD28 costimulation are capable of inducing degradationof p27kip1 in the ubiquitin-proteasome pathway and strongly sug-gest that CD28 costimulation regulates cell cycle progression of Tcells via a mechanism that is independent and upstream of IL-2receptor-mediated signaling.

CD28 costimulation mediates IL-2-independent progression tothe S phase of the cell cycle

p27kip1 has a unique role on the control of the cell cycle becauseit integrates extracellular and intracellular signals during the earlyG1 phase and regulates the activation of cdk-cyclin holoenzymecomplexes, which determine the ability of the cell to progressthrough the G1 phase, pass the G1 restriction point, and enter theS phase. In light of the above results, the question arises whetherCD28 costimulation can promote progression only into late G1

phase of the cell cycle characterized by synthesis of cyclin E andactivation of cdk2 or whether it is sufficient to induce entry to theS phase and clonal expansion in the absence of IL-2. Althoughsubmitogenic anti-CD3 activation did not induce entry to the cellcycle (data not shown), addition of IL-2 resulted in increase ofcells at the S phase of the cell cycle, which peaked at 72 h ofculture (Fig. 6A) and temporally coincided with expression of cy-clin A (Fig. 6B). CD28-mediated signals also increased the per-centage of cells at the S phase of the cell cycle (Fig. 6A) andinduced expression of cyclin A, which peaked at 48 h of culture(Fig. 6B). Anti-IL-2-neutralizing mAb abrogated IL-2-induced in-crease of cells in S phase of the cell cycle and expression of cyclinA (Fig. 6, A andB). Anti-IL-2-neutralizing mAb induced a signif-icant reduction in the percentage of the cells at the S phase andlevels of cyclin A expression induced by CD28 costimulation (Fig.6, A and B). Because these cells express cyclin E (Fig. 2A, toppanel, lanes 15and 16), they pass the G1 restriction point, and,therefore, CD28-costimulated cells that fail to enter S phase in theabsence of IL-2 are blocked at the G1/S transition. Importantly, inthe presence of anti-IL-2-eutralizing mAb, a significant fraction,equivalent to 37% of the CD28-costimulated cells entering thecycle in the absence of anti-IL-2 mAb, could progress into the Sphase (Fig. 6A) and synthesize cyclin A (Fig. 6B), indicating thatCD28 costimulation can mediate IL-2-independent T cellexpansion.

FIGURE 5. Down-regulation of p27 by CD28 costimulation is mediated via ubiquitin-dependent degradation in the proteasome pathway.A, T cells werecultured as indicated, and after 12 and 24 h of culture T cells were isolated and used for RNA preparation and reverse transcription. Quantitation of p27kip1

mRNA was done by quantitative multiplex PCR using two sets of oligonucleotides specific for p27kip1 and G3PDH.B, T cells were cultured for theindicated time intervals with anti-CD31 anti-CD28 in the presence or the absence of Ubal. Lysates were prepared, and equal amounts of protein wereanalyzed by 10% SDS-PAGE and immunoblotted with p27kip1-specific mAb.C, T cells were cultured for 36 h in either media, anti-CD31 anti-CD28,or anti-CD31 anti-CD281 anti-IL-2 neutralizing mAb, either alone or in the presence of two different proteasome inhibitors (MG-132, PI-1). DMSO wasused as a vehicle control for the reconstitution of the proteasome inhibitors. Viability of the cells at the end of the incubation period was determined bytrypan blue exclusion. Lysates were prepared, and equal amounts of protein were analyzed by 10% SDS-PAGE and immunoblotted with p27kip1-specificmAb. For the conditions of TCR1 CD28 culture, 50mg of protein per sample were used, and, for the TCR1 CD28 1 anti-IL-2 mAb culture, 100mgof protein per sample were used. Results are representative of three experiments.

148 CELL CYCLE PROGRESSION BY CD28 COSTIMULATION

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

DiscussionThe physiologic consequences of unrelenting B7/CD28-mediatedT cell activation are most clearly illustrated in the CTLA42/2 mice(23, 58). CTLA42/2 mice have a massive lymphoproliferation,and their T cells are activated in vivo in the absence of overtstimulation by Ag. These events are mediated exclusively via theCD28 pathway since they are reversed by CTLA4Ig (59), whicheliminates B7-1 and B7-2 signaling via CD28 as well as inCTLA4-deficient mice also deficient for B7-1 and B7-2 gene ex-pression (60). These abundantly proliferating T cells from CTLA4-deficient mice do not produce increased amounts of IL-2 protein ormRNA as compared with normal control mice (23), demonstratingthat the dramatic T cell expansion mediated through the unopposedCD28 stimulation in these mice is mediated via IL-2-independentmechanisms. Additional information suggesting the existence ofIL-2-independent CD28 mediated mechanism for T cell expansionis provided by experiments on IL-2-and CD28-deficient mice. Al-though addition of exogenous IL-2 fully restores the reduced T cellproliferative response to lectins in IL-2-deficient mice (21), it canonly partially restore the impaired T cell response in CD28-defi-cient mice (22). In light of these observations, we dissected theeffects of CD28 and IL-2 receptor pathways and attempted to de-termine the mechanisms by which CD28 costimulation regulates Tcell expansion.

Here we show that signals mediated via the CD28 pathway aresufficient to induce entry into the G1 phase, activation of cdks,passage through the G1 restriction point, and progression to the Sphase of the cell cycle in an IL-2-independent manner. The keyregulatory mechanism for the activation of cdks and cell cycleprogression is the down-regulation of p27kip1 cdk inhibitor. Ourresults provide a potential molecular mechanism explaining theIL-2-independent, CD28-mediated extensive lymphoproliferationin the CTLA4-deficient mice and suggest that the unopposedCD28-mediated signaling in these mice may lead to constitutivedown-regulation of p27kip1,resulting in massive T cell expansion.The striking similarity of the CTLA4-deficient and the p27kip1-deficient mice (61, 62), both of which are characterized by T cell

hyperplasia and increased percentage of cells in S phase in theabsence of obvious stimuli, suggest that mechanisms of the cellcycle regulatory machinery in the T lymphocytes may be similarlyaffected in these two types of deficient mice. Although cytokinesother than IL-2 may also contribute to the lymphoproliferationobserved in the CTLA4-deficient mice, this cannot be the case inthe results observed here using unprimed T cells from peripheralblood, since such cells require multiple rounds of restimulation andnumbers of cell divisions to differentiate into effectors producingother types of cytokines (7, 63).

Mechanistically, CD28-mediated down-regulation of p27kip1 iscontrolled at the posttranscriptional level and is mediated via itsubiquitination and degradation in the proteasome pathway. Al-though the signals that activate the proteasome-mediated degrada-tion of p27kip1 have not been fully determined, several observa-tions implicate Ras and the mitogen-activated protein kinase(MAPK) signaling in this process (64–67). Expression of a dom-inant negative Ras allele in NIH 3T3 cells abolished down-regu-lation of p27kip1 in response to epidermal growth factor (65). Con-versely, expression of an activated allele of Ras decreased p27kip1

levels in rat fibroblasts growth-arrested in G1 (64). In that system,a chemical inhibitor of MAPK blocked degradation of p27kip1,hyperphosphorylation of Rb, and activation of cdk2. Ubiquitin-dependent degradation of p27kip1 requires its prior phosphoryla-tion (56, 68). MAPK and cyclin E-cdk2 holoenzyme have beenshown to phosphorylate p27kip1 in vitro (64, 69). Moreover, pro-tein tyrosine kinase (PTK) activity is required for the induction ofprotein ubiquitination (70, 71). Importantly, although p27kip1 doesnot undergo tyrosine phosphorylation, PTK activity is required forsubsequent ubiquitination of p27kip1 (56). CD28-mediated co-stimulation has a direct effect on the activation of PTKs (72–75),Ras, and the MAPK pathway (76), thereby regulating the phos-phorylation of various intracellular substrates. Therefore, theseevents may directly link CD28-mediated signals to p27kip1 degra-dation. Protein ubiquitination is a dynamic process that is con-trolled by the coordinate action of multiple ubiquitin-conjugatingenzymes and deubiquitinating enzymes (77). Whether CD28 co-stimulation directly controls ubiquitination of p27kip1 by regulat-ing the activity of ubiquitinating or deubiquitinating enzymes re-mains to be determined.

Regardless of the level and the mechanism of regulation, ourresults show that CD28 is capable of mediating down-regulation ofp27kip1 in a direct, IL-2-independent, and a secondary IL-2-depen-dent manner. This is supported by the observation 1) that ubiq-uitination of p27kip1 is detected even in the presence of neutraliz-ing anti-IL-2 mAb and 2) that neutralizing anti-IL-2 mAb onlypartially inhibits CD28-mediated down-regulation of p27kip1 ex-pression whereas it efficiently prevents down-regulation of p27kip1

induced by IL-2 added in 100-fold higher concentration than thatproduced by CD28 costimulation. Therefore, CD28 costimulationregulates T cell cycle progression via two distinct mechanisms.First, it directly primes for entry to the cycle by down-regulatingp27kip1 before IL-2 accumulation. CD28 mediates the initial dropof p27kip1, which releases cyclin D2/cdk4-cdk6 and cyclin E/cdk2from the constraint of this inhibitor and, therefore, allows the ac-tivation of its enzymatic activity and progression to the late G1

phase. Second, since significant amounts of IL-2 accumulate as aconsequence of CD28-mediated increased IL-2 secretion and sinceT cells are ready to uptake them because expression of IL-2 re-ceptora-, b-, andg-chains has been induced, IL-2 receptor-me-diated signals result in enhanced and prolonged activation of cy-clin E/cdk2, which induces phosphorylation and furtherdegradation of p27kip1, amplification of the activation signal, andclonal expansion. This hypothesis is supported by the observation

FIGURE 6. CD28 costimulation mediates IL-2-independent progres-sion to the S phase of the cell cycle.A, After culture under various con-ditions, cells were harvested at the indicated time points, and cell cycleanalysis was performed as described inMaterials and Methods. B, Underthe same conditions, cell lysates were prepared, and equal amounts ofprotein (50mg/sample) were analyzed by 10% SDS-PAGE and immuno-blotted by cyclin A-specific antiserum. Results are representative of fourexperiments.

149The Journal of Immunology

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

that, although anti-IL-2 mAb does not affect CD28-mediated ac-tivation of cdk4 and cdk6, it partially inhibits CD28-mediated ac-tivation of cdk2. Most importantly, despite the partial inhibition ofCD28-mediated cdk2 activation by anti-IL-2 mAb, a significantfraction of cells can enter S phase, indicating that signals directlymediated through CD28 can be sufficient for T cell clonal expan-sion. Therefore, CD28 costimulatory signals mediate entry of Tcells into the cell cycle and render them competent for progressioninto the S phase and clonal expansion in an IL-2-dependent andIL-2-independent manner.

AcknowledgmentsWe thank Dr. Lee Nadler for helpful discussions and critical reading of themanuscript.

References1. Azuma, M., H. Yssel, J. H. Phillips, H. Spits, and L. L. Lanier. 1993. Functional

expression of B7/BB1 on activated T lymphocytes.J. Exp. Med. 177:845.2. Freeman, G. J., A. S. Freedman, J. M. Segil, G. Lee, J. F. Whitman, and

L. M. Nadler. 1989. B7, a new member of the Ig superfamily with unique ex-pression on activated and neoplastic B cells.J. Immunol. 143:2714.

3. Freeman, G. J., J. G. Gribben, V. A. Boussiotis, J. W. Ng, V. Restivo,L. Lombard, G. S. Gray, and L. M. Nadler. 1993. Cloning of B7-2: a CTLA4counter-receptor that costimulates human T cell proliferation.Science 262:909.

4. Linsley, P. S., E. A. Clark, and J. A. Ledbetter. 1990. T-cell antigen CD28mediates adhesion with B cells by interacting with activation antigen B7/BB-1.Proc. Natl. Acad. Sci. USA 87:5031.

5. Cerdan, C., Y. Martin, M. Courcoul, H. Brailly, C. Mawas, F. Birg, and D. Olive.1992. Prolonged IL-2 receptora/CD25 expression after T cell activation via theadhesion molecules CD2 and CD28: demonstration of combined transcriptionaland post-transcriptional regulation.J. Immunol. 149:2255.

6. Cerdan, C., Y. Martin, M. Courcoul, C. Mawas, F. Birg, and D. Olive. 1995.CD28 costimulation up-regulates long-term IL-2Rb expression in human T cellsthrough combined transcriptional and post-transcriptional regulation.J. Immunol154:1007.

7. Freeman, G. J., V. A. Boussiotis, A. Anumanthan, G. M. Bernstein, X.-Y. Ke,P. D. Rennert, G. S. Gray, J. G. Gribben, and L. M. Nadler. 1995. B7-1 and B7-2do not deliver identical costimulatory signals since B7-2 but not B7-1 preferen-tially costimulates the initial production of IL-4.Immunity 2:523.

8. June, C. H., J. A. Bluestone, L. M. Nadler, and C. B. Thompson. 1994. The B7and CD28 receptor families.Immunol. Today 15:321.

9. Guerder, S., J. Meyerhoff, and R. Flavell. 1994. The role of the T cell costimu-lator B7-1 in autoimmunity and the induction and maintenance of tolerance toperipheral antigen.Immunity 1:155.

10. Harlan, D. M., H. Hengartner, M. L. Huang, Y. H. Kang, R. Abe,R. W. Moreadith, H. Pircher, G. S. Gray, P. S. Ohashi, G. J. Freeman,L. M. Nadler, C. H. June, and P. Aichele. 1994. Transgenic mice expressing bothB7 and viral glycoprotein on pancreaticb cells along with glycoprotein-specifictransgenic T cells develop diabetes due to a breakdown of T lymphocyte unre-sponsiveness.Proc. Natl. Acad. Sci. USA 91:3137.

11. Verwilghen, J., R. Lovis, M. De Boer, P. Linsley, G. Haines, A. Koch, andR. Pope. 1994. Expression of functional B7 and CTLA4 on rheumatoid synovialT cells.J. Immunol. 153:1378.

12. Chen, L., A. Ashe, W. A. Brady, I. Hellstrom, K. E. Hellstrom, J. A. Ledbetter,P. McGowan, and P. Linsley. 1992. Costimulation of antitumor immunity by theB7 counter receptor for the T lymphocyte molecule CD28 and CTLA-4.Cell71:1093.

13. Townsend, S. E., and J. P. Allison. 1993. Tumor rejection after direct costimu-lation of CD81 T cells by B7-transfected melanoma cells.Science 259:368.

14. Matulonis, U., C. Dosiou, C. Lamont, G. J. Freeman, P. Mauch, L. M. Nadler,and J. D. Griffin. 1995. Role of B7-1 in mediating an immune response to myeloidleukemia cells.Blood 85:2507.

15. Turka, L. A., P. S. Linsley, H. Lin, W. Brady, J. M. Leiden, R. Q. Wei,M. L. Gibson, X. G. Zheng, S. Myrdal, D. Gordon, T. Bailey, S. F. Bolling, andC. B. Thompson. 1992. T-cell activation by the CD28 ligand B7 is required forcardiac allograft rejection in vivo.Proc. Natl. Acad. Sci. USA 89:11102.

16. Finck, B., P. Linsley, and D. Wofsy. 1994. Treatment of murine lupus withCTLA4Ig. Science 265:1225.

17. Linsley, P. S., P. M. Wallace, J. Johnson, M. G. Gibson, J. L. Greene,J. A. Ledbetter, C. Singh, and M. A. Tepper. 1992. Immunosuppression in vivoby a soluble form of the CTLA-4 T cell activation molecule.Science 257:792.

18. Blazar, B. R., P. A. Taylor, P. S. Linsley, and P. A. Vallera. 1994. In vivoblockade of CD28/CTLA4: B7/BB1 interaction with CTLA4-Ig reduces lethalmurine graft-versus-host disease across the major histocompatibility complexbarrier in mice.Blood 83:3815.

19. Guinan, E. C., V. A. Boussiotis, D. Neuberg, L. LaVita Brennan, N. Hirano,L. M. Nadler, and J. G. Gribben. 1999. Transplantation of anergic histoincom-patible bone marrow allografts.N. Engl. J. Med. 340:1704.

20. Thompson, C. B., T. Lindsten, J. A. Ledbetter, S. L. Kunkel, H. A. Young,S. G. Emerson, J. M. Leiden, and C. H. June. 1989. CD28 activation pathway

regulates the production of multiple T-cell-derived lymphokines/cytokines.Proc.Natl. Acad. Sci. USA 86:1333.

21. Schorle, H., T. Holtschke, T. Hunig, A. Schimpl, and I. Horak. 1991. Develop-ment and function of T cells in mice rendered interleukin-2 deficient by genetargeting.Nature 352:621.

22. Shahinian, A., K. Pfeffer, K. P. Lee, T. M. Kundig, K. Kishihara, A. Wakeham,K. Kawai, P. S. Ohashi, C. B. Thompson, and T. W. Mak. 1993. Differential Tcell costimulatory requirements in CD28-deficient mice.Science 261:609.

23. Tivol, E. A., F. Borriello, N. A. Schweitzer, W. P. Lynch, J. A. Bluestone, andA. H. Sharpe. 1995. Loss of CTLA4 leads to massive lymphoproliferation andfatal multiorgan tissue destruction, revealing a critical negative regulatory role ofCTLA4. Immunity 3:541.

24. Sherr, C. J. 1994. G1 phase progression: cyclin on cue.Cell 79:551.25. Ajchenbaum, F., K. Ando, J. A. DeCaprio, and J. D. Griffin. 1993. Independent

regulation of human D-type cyclin gene expression during G1 phase in primaryhuman T lymphocytes.J. Biol. Chem. 268:4113.

26. Hatakeyama, M., J. A. Brill, G. R. Fink, and R. A. Weinberg. 1994. Collaborationof G1 cyclins in the functional inactivation of the retinoblastoma protein.GenesDev. 1994:1759.

27. Dynlacht, B. D., O. Flores, J. A. Lees, and E. Harlow. 1994. Differential regu-lation of E2F transactivation by cyclin/cdk2 complexes.Genes Dev. 8:1772.

28. Chellappan, S. P., S. Hielbert, M. Mudryj, J. M. Horowitz, and J. R. Nevins.1991. The E2F transcription factor is a cellular target for the Rb protein.Cell65:1053.

29. Weintraub, S. J., C. A. Prater, and D. C. Dean. 1992. Retinoblastoma proteinswitches the E2F site from positive to negative element.Nature 358:259.

30. Boussiotis, V. A., G. J. Freeman, G. Gray, J. Gribben, and L. M. Nadler. 1993.B7 but not ICAM-1 costimulation prevents the induction of human alloantigenspecific tolerance.J. Exp. Med. 178:1753.

31. Boussiotis, V. A., D. L. Barber, T. Nakarai, G. J. Freeman, J. G. Gribben,G. M. Bernstein, A. L. D’Andrea, J. Ritz, and L. M. Nadler. 1994. Prevention ofT cell anergy by signaling through thegc chain of the IL-2 receptor.Science266:1039.

32. Matsushime, H., D. E. Quekke, S. A. Shurtleff, M. Shibuya, C. J. Sheer, andJ. Y. Kato. 1994. D-type cyclin-dependent kinase activity in mammalian cells.Mol. Cell. Biol. 14:2066.

33. Lee, C. R., S. Yamamoto, H. Wanibuchi, S. Wada, K. Sigamura, T. Kishimoto,and S. Fukushima. 1997. Cyclin D1 overexpression in rat two-stage bladder car-cinogenesis and its relationship with oncogenes, tumor suppressor genes and cellproliferation.Cancer Res. 57:4765.

34. Lee, S. R., T. Ichihara, S. Yamamoto, H. Wanibuchi, K. Sugimura, S. Wada,T. Kishimoto, and S. Fukushima. 1999. Reduced expression of the cdk inhibitorp27kip1 in rat two-stage bladder carcinogenesis and its association with expres-sion profiles of p21waf1/cip1 and p53.Carcinogenesis 20:1697.

35. Boussiotis, V. A., B. J. Lee, G. J. Freeman, J. G. Gribben, and L. M. Nadler.1997. Induction of T cell clonal anergy results in resistance, whereas CD28-mediated costimulation primes for susceptibility to Fas- and Bax-mediated pro-grammed cell death.J. Immunol. 159:3156.

36. Meyerson, M., and E. Harlow. 1994. Identification of G1 kinase activity for cdk6,a novel cyclin D partner.Mol. Cell. Biol. 14:2077.

37. Kwon, T. K., M. A. Buchholz, P. Ponsalle, F. J. Chrest, and A. A. Nordin. 1997.The regulation of p27Kip1 expression following the polyclonal activation of mu-rine G0 T cells.J. Immunol. 158:5642.

38. Sherr, C. J. 1996. Cancer cell cycles.Science 274:1672.39. Sherr, C. J., and J. M. Roberts. 1995. Inhibitors of mammalian G1 cyclin-depen-

dent kinases.Genes Dev. 9:1149.40. Serrano, M., G. J. Hannon, and D. Beach. 1993. A new regulatory motif in

cell-cycle control causing specific inhibition of cyclin D/cdk4.Nature 366:704.41. Chan, F. K., J. Zhang, L. Cheng, D. N. Shapiro, and A. Winoto. 1995. Identifi-

cation of human and mouse p19, a novel CDK4 and CDK6 inhibitor with ho-mology to p16ink4.Mol. Cell. Biol. 15:2682.

42. Hirai, H., M. F. Roussel, J. Y. Kato, R. A. Ashmun, and C. J. Sherr. 1995. NovelINK4 proteins, p19 and p18, are specific inhibitors of the cyclin D-dependentkinase cdk4 and cdk6.Mol. Cell. Biol. 15:2672.

43. Xiong, Y., G. J. Hannon, H. Zhang, D. Casso, R. Kobayashi, and D. Beach. 1993.p21 is a universal inhibitor of cyclin kinases.Nature 366:701.

44. Harper, J. W., G. R. Adami, N. Wei, K. Keyomarsi, and S. J. Ellegdge. 1993. Thep21 cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependentkinases.Cell 75:805.

45. Polyak, K., M. Lee, H. Erdjument-Bromage, A. Koff, J. M. Roberts, P. Tempst,and J. Massague. 1994. Cloning of p27kip1, a cyclin-dependent kinase inhibitorand a potential mediator of extracellular antimitogenic signals.Cell 78:59.

46. Toyoshima, H., and T. Hunter. 1994. p27, a novel inhibitor of G1 cyclin-cdkprotein kinase activity is related to p21.Cell 78:67.

47. Nourse, J., E. Firpo, M. W. Flanagan, S. Coats, K. Polyak, M. Lee, J. Massague,J. R. Crabtree, and J. M. Roberts. 1994. Interleukin-2-mediated elimination of thep27kip1 cyclin-dependent kinase inhibitor prevented by rapamycin.Nature 372:570.

48. Flores-Rozas, H., Z. Kelman, F. Dean, Z. Pan, J. W. Harper, S. Elledge, M.O’Donnell, and J. Hurwitz. 1994. cdk-interacting protein 1 directly binds withproliferating cell nuclear antigen and inhibits DNA replication catalyzed by theDNA polymerase and holoenzyme.Proc. Natl. Acad. Sci. USA 91:8655.

49. Waga, S., G. J. Hannon, D. Beach, and B. Stillman. 1994. The p21 inhibitor ofcyclin-dependent kinases controls DNA replication by interaction with PCNA.Nature 369:574.

50. Luo, Y., J. Hurwitz, and J. Massague. 1995. Cell-cycle inhibition by independentCDK and PCNA binding domains in p21 Cip1.Nature 375:159.

150 CELL CYCLE PROGRESSION BY CD28 COSTIMULATION

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

51. Hengst, L., and S. I. Reed. 1996. Translational control of p27kip1 accumulationduring the cell cycle.Science 271:1861.

52. Polyak, K., J. Kato, M. J. Solomon, C. J. Sherr, J. Massague, J. M. Roberts, andA. Koff. 1994. p27kip1 a cyclin-cdk inhibitor, links transforming factor-b andcontact inhibition to cell cycle arrest.Genes Dev. 8:9.

53. Kato, J. Y., M. Matsuoka, K. Polyak, J. Massague, and C. J. Sherr. 1994. CyclicAMP-induced G1 phase arrest mediated by an inhibitor (p27kip1) of cyclin-de-pendent kinase 4 activation.Cell 79:487.

54. Pagano, M., S. W. Tam, A. M. Theodoras, P. Beer-Romero, G. Del Sal, V. Chau,P. R. Yew, G. F. Draetta, and M. Rolfe. 1995. Role of ubiquitin-proteasomepathway in regulating abundance of the cyclin-dependent kinase inhibitor p27.Science 269:682.

55. Kamesaki, H., K. Nishizawa, G. Y. Michaud, J. Cossman, and T. Kiyono. 1998.TGF-b1 induces the cyclin-dependent kinase inhibitor p27Kip1 mRNA and pro-tein in murine B cells.J. Immunol. 160:770.

56. Alessandrini, A., D. S. Chiaur, and M. Pagano. 1997. Regulation of the cyclin-dependent kinase inhibitor p27 by degradation and phosphorylation.Leukemia11:342.

57. Jonston, N. L., and R. E. Cohen. 1991. Uncoupling ubiquitin-protein conjugationfrom ubiquitin-dependent proteolysis by use ofb,g-nonhydrolyzable ATP ana-logues.Biochemistry 30:7514.

58. Waterhouse, P., J. M. Penninger, E. Timms, A. Wakenam, A. Shahinian,K. P. Lee, C. B. Thompson, H. Griesser, and T. W. Mak. 1995. Lymphoprolif-erative disorders with early lethality in mice deficient in CTLA-4.Science 270:985.

59. Tivol, E. A., S. D. Boyd, S. McKeon, F. Borriello, P. Nickerson, T. B. Strom, andA. H. Sharpe. 1997. CTLA4Ig prevents lymphoproliferation and fatal multiorgantissue destruction in CTLA-4-deficient mice.J. Immunol. 158:5091.

60. Mandelbrot, D. A., A. J. McAdam, and A. H. Sharpe. 1999. B7-1 or B7-2 isrequired to produce the lymphoproliferative phenotype in mice lacking cytotoxicT lymphocyte-associated antigen 4 (CTLA-4).J. Exp. Med. 189:435.

61. Fero, M. L., M. Rivkin, M. Tasch, P. Porter, C. E. Carrow, E. Firpo, K. Polyak,L. H. Tsai, V. Broudy, R. M. Perlmutter, K. Kaushansky, and J. M. Roberts.1996. A syndrome with multiorgan hyperplasia with features of gigantism, tu-morigenesis and female sterility in p27Kip1-deficient mice.Cell 85:733.

62. Kiyokawa, H., R. D. Kineman, K. O. Monova-Todorova, V. C. Soares,E. S. Hoffman, M. Ono, D. Khanam, A. C. Hayday, L. A. Frohman, and A. Koff.1996. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitorfunction of p27kip1. Cell 85:721.

63. Bird, J. J., D. R. Brown, A. C. Mullen, N. H. Moskowitz, N. H. Mahowadl,M. A. Sider, T. F. Gajewski, C. R. Wang, and S. L. Reiner. 1998. Helper T celldifferentiation is controlled by the cell cycle.Immunity 9:225.

64. Kawada, M., S. Yamagoe, Y. Marakami, K. Suzuki, S. Mizumo, and Y. Uehara.1997. Induction of p27kip1 degradation and anchorage independence by Rasthrough the MAP kinase signaling pathway.Oncogene 15:629.

65. Takuwa, N., and Y. Takuwa. 1997. Ras activity in G1 phase required for p27kip1

down-regulation, passage through the restriction point and entry into S phase ingrowth factor-stimulated NIH 3T3 fibroblasts.Mol. Cell. Biol. 17:5348.

66. Feramisco, J. R., M. Gross, T. Kamata, M. Rosengberg, and R. W. Sweet. 1984.Microinjection of the oncogene form of the human H-ras (T-24) protein results inrapid proliferation of quiescent cells.Cell 38:1009.

67. Peeper, D. S., T. M. Upton, M. H. Ladha, E. Neuman, J. Zalvide, R. Bernards,J. A. DeCaprio, and M. E. Ewen. 1997. Ras signalling linked to the cell-cyclemachinery by the retinoblastoma protein.Nature 386:177.

68. Vlach, J., S. Hennecke, and B. Amati. 1997. Phosphorylation-dependent degra-dation of the cyclin-dependent kinase inhibitor p27.EMBO J. 16:5334.

69. Sheaff, R. J., M. Groudine, M. Gordon, J. M. Roberts, and B. E. Clurman. 1997.Cyclin E-cdk2 is a regulator of p27kip1. Genes Dev. 11:1468.

70. Cenciarelli, C., K. G. Wilhelm, A. Guo, and A. M. Weissman. 1996. T cellantigen receptor ubiquitination is a consequence of receptor-mediated tyrosinekinase activation.J. Biol. Chem. 15:8709.

71. Hou, D., C. Cenciarelli, J. P. Jensen, H. B. Nguyen, and A. M. Weissman. 1994.Activation-dependent ubiquitination of a T cell antigen receptor subunit on mul-tiple intracellular lysines.J. Biol. Chem. 269:14244.

72. August, A., S. Gibson, K. Kawakami, T. Kawakami, G. B. Mills, and B. Dupont.1994. CD28 is associated with and induces the immediate tyrosine phosphory-lation and activation of the Tec family kinase ITK/EMT in the human Jurkatleukemic T-cell line.Proc. Natl. Acad. Sci. USA 91:9347.

73. August, A., and B. Dupont. 1994. Activation of src family kinaselck followingCD28 crosslinking in the Jurkat leukemic cell line.Biochem. Biophys. Res. 199:1466.

74. Raab, M., Y. C. Cai, S. C. Bunnell, S. D. Heyeck, L. J. Berg, and C. E. Rudd.1995. p56Lck and p59Fyn regulate CD28 binding to phosphatidylinositol 3-kinase,growth factor receptor-bound protein GRB-2, and T cell-specific protein-tyrosinekinase ITK: implications for T-cell costimulation.Proc. Natl. Acad. Sci. USA92:8891.

75. Boussiotis, V. A., D. L. Barber, B. J. Lee, G. J. Freeman, J. G. Gribben, andL. M. Nadler. 1996. Differential association of protein tyrosine kinases with T cellreceptor is linked to the induction of anergy and its prevention by B7 family-mediated costimulation.J. Exp. Med. 184:365.

76. Nunes, J. A., Y. Collette, A. Truneh, D. Olive, and D. A. Cantrell. 1994. The roleof p21rasin CD28 signal transduction: triggering of CD28 with antibodies, but notthe ligand B7-1, activates p21ras. J. Exp. Med. 180:1067.

77. D’Andrea, A., and D. Pellman. 1998. Deubiquitinating enzymes: a new class ofbiological regulators.Crit. Rev. Biochem. Mol. Biol. 33:337.

151The Journal of Immunology

by guest on October 27, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from