Differentiation Regulated Differently During Human B Cell Cyclin

The Expression of p18INK4 and p27kip1 Cyclin-Dependent Kinase Inhibitors Is Regulated Differently During Human B Cell Differentiation1

Nicolas Schrantz,2 Georges Edouard Beney,2 Marie The´re`se Auffredou, Marie Franc¸oise Bourgeade, Ge´rald Leca, and Aime´Vazquez3

Cell cycle progression is under the control of cyclin-dependent kinases (cdks), the activity of which is dependent on the expression of specific cdk inhibitors. In this paper we report that the two cdk inhibitors, p27Kip1 and p18INK4c, are differently expressed and control different steps of human B lymphocyte activation. Resting B cells contain large amounts of p27Kip1 and no p18INK4c.In vitro stimulation by Staphylococcus aureus Cowan 1 strain or CD40 ligand associated with IL-10 and IL-2 induces a rapid decrease in p27Kip1 expression combined with cell cycle entry and progression. In contrast, in vitro Ig production correlates with specific INK4c expression of p18 and early G1 arrest. This G1 arrest is associated with inhibition of cyclin D3/cdk6-mediated retinoblastoma protein phosphorylation by p18INK4c. A similar contrasting pattern of p18INK4c and p27Kip1 expression is observed both in B cells activated in vivo and in various leukemic cells. Expression of p18INK4c was also detected in various Ig-secreting cell lines in which INK4c Kip1 both maximum Ig secretion and specific p18 expression were observed during the G1 phase. Our study shows that p27 and p18INK4c have different roles in B cell activation; p27Kip1 is involved in the control of cell cycle entry, and p18INK4c is involved

in the subsequent early G1 arrest necessary for terminal B lymphocyte differentiation. The Journal of Immunology, 2000, 165: 4346–4352.

n most cell lineages, ﬁnal differentiation is associated with clin D and cdk2/cyclin E complexes releases the transcription fac-

loss of proliferation and cell cycle arrest in G1, which may or tor E2F (8–14). E2F, in turn, activates genes required for the S I not be followed by a withdrawal from cell cycle. In response phase (8). Although the activity of each complex is governed by to Ag-specific and T-derived signals, human B lymphocytes pro- the periodic accumulation of cyclins, additional levels of control liferate before undergoing final differentiation into plasma secret- include the phosphorylation and dephosphorylation of critical ty- ing cells (1, 2). This differentiation takes place in nonproliferative rosine and serine/threonine residues and the presence of specific cells and is therefore dependent on cell cycle regulation. cdk inhibitors (CKI) (3, 15). These inhibitors belong to two dif-

by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. Cell cycle progression is under the control of a family of serine/ ferent families: one is the INK4 family, consisting of p16INK4a 4 threonine kinases named the cyclin-dependent kinases (cdk). Ac- (p16), p15INK4b (p15), p18INK4c (p18), and p19INK4d, and the other tivation of cdk is regulated by their association with regulatory is the Cip/Kip family, consisting of p21Cip1/Waf1/Sdi1 (p21), subunits (the cyclins), the expression of which is tightly regulated p27Kip1 (p27), and p57Kip2 (p57) (16–25). CKI belonging to the during the various steps of the cell cycle (3, 4). For instance, G1 INK4 family interact preferentially with cdk4 and cdk6, preventing progression is controlled by cdk4 and cdk6 associated with mem- their association with the cyclins D, and therefore are specific for bers of the cyclin D family. G1/S transition and S progression are G1. The Cip/Kip proteins regulate cdk activity by forming ternary controlled by cdk2 associated with cyclin E and cyclin A, respec- complexes with cdk and cyclins. They can regulate multiple cdk tively (5–7). One of their substrates is the retinoblastoma protein enzymes, including cdk4/6, throughout the cell cycle (15). There- https://www.jimmunol.org (pRb), which upon sequential phosphorylation by the cdk4/6/cy- fore, cdk4/6 are the only cdk to be regulated by both INK4 and Cip/Kip CKI, and this feature renders them pivotal in the control of both cell cycle entry and exit. Institut National de la Santeét de la Recherche Me´dicale, Unite´131, and Centre de Studies of CKI-deficient mice suggest that despite common cdk Recherche Claude Bernard, Clamart, France targets shared by both families of CKI, these different inhibitors Received for publication February 28, 2000. Accepted for publication July 26, 2000. regulate various cell functions. For instance, disruption of p16 is Downloaded from The costs of publication of this article were defrayed in part by the payment of page associated with the development of spontaneous tumors at an early charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. stage in various cell types, whereas p21-defective mice do not 1 This work was supported by Institut National de la Santeét de la Recherche Me´di- develop spontaneous tumors and are developmentally normal (20, cale and grants from the Association pour la Recherche sur le Cancer (Villejuif, 26–31). Both p18- and p27-deficient mice develop gigantism, dis- France). N.S. is a recipient of a fellowship from Fondation pour la Recherche Me´dicale. proportionally enlarged thymus and spleen, and pituitary tumori- 2 N.S. and G.E.B. contributed equally to this paper. genesis, which is regulated by p18 and p27 through two separate pathways controlling the function of pRb (31). In normal mice, 3 Address correspondence and reprint requests to Dr. Aime´Vazquez, Institut National de la Sante´ et de la Recherche Me´dicale, Unite´ 131, 32 rue des Carnets, 92140 p18 is widely expressed during embryogenesis and accumulates to Clamart, France. E-mail address: [email protected] high levels in a number of terminally differentiated tissues, 4 Abbreviations used in this paper: cdk, cyclin-dependent kinase; CD40L, CD40 li- whereas p27 is preferentially expressed in quiescent cells (12, 20, gand; CKI, cyclin-dependent kinase inhibitor; p18, p18INK4c; p27, p27Kip1; pRb, retinoblastoma protein; SAC, Staphylococcus aureus strain A; B-CLL, chronic lympho- 32–37), suggesting that p18 and p27 regulate different aspects of cytic leukemia B. cell activation.

We previously reported that cell cycle entry and G1 progression iodide fluorescence of individual nuclei using a FACScan flow cytometer of in vitro activated human B lymphocytes were dependent on both (Becton Dickinson, Mountain View, CA). Cell cycle compartments, including G /G , S, and G /M phases, and the percentage of the cells in the activation of the cdk6/cyclin D3 complex and down-regulation of 0 1 2 compartments were determined using the computer program Cell-Fit DNA p27 expression (12). Nevertheless, the mechanisms that control software with SFIT calculations (Becton Dickinson). differentiation-coupled cell cycle arrest in normal human B cells Western blotting are still poorly understood. In this paper we investigate the roles of different CKI during in vitro activation leading to final differenti- Cells were lysed by incubation in lysis buffer (50 mM Tris (pH 8), 150 mM ation of normal B cell into Ig-secreting cells. We observed oppo- NaCl, 0.1% Nonidet P-40, 5 mM EDTA, 50 mM NaF, 2 mM sodium ␮ ␮ site patterns of p27 and p18 expression in resting and in B lym- vanadate, 1 g/ml aprotinin, 2 mM PMSF, 1 mM DTT, 10 g/ml leupeptin, and 2 ␮g/ml pepsatin A; all from Sigma) on ice for 60 min. The lysate phocytes activated either in vitro or in vivo as well as in different was centrifuged for 30 min at 13,000 ϫ g at 4°C, and the supernatant was B cell lines. We show that human B differentiation is dependent on assayed for protein concentration (micro-bicinchoninic acid protein assay; G cell cycle arrest, which is associated with inhibition of cyclin Pierce, Rockford, IL). Cell lysate proteins (50 ␮g) were boiled for 5 min 1 ϫ D3/cdk6-mediated pRb phosphorylation by p18. Thus, these data in 2 sample buffer and resolved on 10 or 13% SDS-polyacrylamide gels by electrophoresis. Proteins were electroblotted onto 0.45-␮m pore size suggest that p27 and p18 are differentially involved in the control nitrocellulose filters, and the filters were blocked for1hatroom temper- of the sequential cell cycle entry and exit observed during human ature with 5% nonfat milk in PBS and 1% Tween-20. Filters were then B lymphocyte differentiation. incubated for 1 h with anti-cyclin D3 (2 ␮g/ml; Santa Cruz Biotechnology, Santa Cruz, CA), anti-cyclin E (2 ␮g/ml; C-198, Santa Cruz Biotechnol- ogy), anti-cdk2 (2 ␮g/ml; C-163, Santa Cruz Biotechnology), anti-cdk6 (2 Materials and Methods ␮g/ml; C-21, Santa Cruz Biotechnology), anti-pRb (2 ␮g/ml; clone G3- Reagents 245, PharMingen, San Diego, CA), anti-p18 (2 ␮g/ml; Santa Cruz Biotech- ␮ Recombinant IL-10 was purchased from R&D Systems (Minneapolis, nology), or anti-actin (1 g/ml). The blots were washed three times (10 min MN). Recombinant IL-2 was provided by Chiron Laboratory-France each time) with 0.1% Tween-20 in PBS and incubated (60 min) with perox- (Suresnes, France). Staphylococcus aureus Cowan 1 strain (SAC; pan- idase-labeled swine anti-rabbit or sheep anti-mouse Ig (1/5000; Amersham, Les Ulis, France). Blots were developed using an enhanced chemilumines- sorbin) was obtained from Calbiochem (La Jolla, CA). z-Val-Ala-D,L-Asp- fluoromethylketone was supplied by Bachem Biochimie SARL (Voisin le cence detection system (ECL, Amersham). Films were exposed for 1–30 min. Bretonneux, France). Murine fibroblastic cells expressing both human Immunoprecipitation and kinase assay CD40 ligand (CD40L) and CD32/Fc␥RII (CD40L/CD32L cells) were gifts from Dr. Saeland (Schering-Plough, Dardilly, France). Cells were lysed for 60 min on ice. After centrifugation for 30 min at 13,000 ϫ g, cell lysate was immunoprecipitated by incubation at 4°C for Cell preparations 3 h with 5 ␮g of Ab coated on 25 ␮l of protein G-Sepharose beads (Phar- macia). For retinoblastoma kinase assays, immune complexes were col- Normal B cells were isolated from human tonsils as previously described lected by centrifugation at 4°C and washed three times with buffer (50 mM (12). Briefly, single-cell suspensions were depleted of T cell by two cycles HEPES (pH 7.2), 10 mM MgCl2, 5 mM MnCl2, 1 mM DTT, 50 mM NaF, of rosetting with 2-aminoethyl-isothiouronium-treated sheep RBC and 2 mM sodium vanadate, 1 ␮g/ml aprotinin, 2 mM PMSF, 1 mM DTT, 10 were depleted of monocytes by adherence to plastic. B cells were then ␮g/ml leupeptin, and 2 ␮g/ml pepsatin) and twice with kinase buffer (150 separated into density fractions on discontinuous Percoll gradients (Phar- mM NaCl, 50 mM HEPES (pH 7.50), 1 mM EDTA, 1 mM EGTA, 50 mM macia, Uppsala, Sweden). Cells suspended below the 60% Percoll fraction NaF, 2 mM sodium vanadate, 1 ␮g/ml aprotinin, 2 mM PMSF, 1 mM DTT, Ͼ ϩ Ϫ (a homogeneous population of small cells, 97 IgD /CD38 cells) were 10 ␮g/ml leupeptin, and 2 ␮g/ml pepsatin A). They were then suspended considered to be resting cells, and cells collected at the 55% layer (large ␮ ␮ by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. in a total reaction volume of 40 l of kinase buffer containing 40 Ci of Ͼ Ϫ ϩ cells, 95% IgD /CD38 cells) were considered to be in vivo activated B [␥-32P]ATP (3000 Ci/nmol; Amersham) and 0.5 ␮g of truncated retinoblas- cells. Leukemic B cells were isolated by negative selection after two cycles toma p56 pRb protein (QED Bioscience, San Diego, CA). After1hat37°C, of T cell rosetting from blood obtained from patients with chronic lym- SDS-PAGE sample buffer was added, and the samples were heated to 100°C phocytic leukemia or plasma cell leukemia, provided by Dr. De Revel for 5 min, run on a 10% SDS polyacrylamide gel, electroblotted onto 0.45-␮m (Hoˆpital Percy, Clamart, France). Ramos, BL41, Daudi, Raji, and U266 pore size nitrocellulose filters, and autoradiographed for 16 h. cell lines were obtained from American Type Culture Collection (Manas- sas, VA), LP1 was obtained from the European Collection of Cell Cultures Results (Salisbury, U.K.), and the Burkitt cell line Capa-2 was obtained from Dr. Terminal B cell differentiation is associated with G arrest S. Sharma (Brown University, Providence, RI) (38). The human T cell 1 chronic lymphocytic leukemia-derived, IL-2-dependent Kitt 225 cell line To investigate the involvement of cell cycle regulatory molecules https://www.jimmunol.org was provided by Dr. Kori (Kyoto University, Kyoto, Japan) (39). during human B lymphocyte differentiation, we first studied cell B cell culture cycle progression during this process. For this, resting tonsillar B cells were activated in the presence of SAC (1/10,000), IL-2 (10 B cells (106/ml) were cultured in RPMI 1640 culture medium (Life Tech- nologies, Grand Island, NY) supplemented with 1% glutamine, 1% anti- ng/ml), and IL-10 (100 ng/ml). Every day, progression of the ac- biotics, 2␤-ME (0.5 ϫ 10Ϫ5 M), and 10% FBS (Life Technologies) in the tivated B cells through the cell cycle was determined by flow cy- presence of SAC (1/10,000), IL-10 (100 ng/ml), and IL-2 (10 ng/ml). For tometric analysis, and B cell differentiation was monitored as Ig 5 Downloaded from the experiment presented in Fig. 6, B cells (10 /ml) were cultured in the production by measuring the concentration in the culture superna- 4 same medium with irradiated CD40L/CD32L cells (10 /ml) in the presence tant. Following activation, we observed a decrease in the percent- of IL-10 (100 ng/ml). Ig concentrations in supernatants were determined by ELISA. Cell supernatants were incubated in microtiter plates coated with age of G0 /G1 cells to a minimum on day 3 (Fig. 1). A large fraction a mouse anti-human Ig Ab. Bound human Ig was revealed using alkaline of these cells (55%) was progressing through S phase, as evi- phosphatase-conjugated goat anti-human Ig and phosphatase substrate ( p- denced by flow cytometry and thymidine uptake (Fig. 1 and data nitrophenyl phosphate from Sigma, St. Louis, MO). The mouse anti-human not shown). By day 5 cells again accumulated in G , and this was Ig and the alkaline phosphatase-conjugated goat anti-human Ig were pur- 1 chased from Biosys (Compiegne, France). The proliferative response was completed by days 7–8. This progression through the cell cycle measured by addition of 0.5 ␮Ci of [3H]TdR (Centre Energie Atomique, can also be monitored by analysis of pRb phosphorylation. As we Saclay, France) to 105 cells for the last 16 h of culture. previously reported, resting B cells do not express pRb (12); in contrast, after 24 h of culture, activated B cells expressed the ret- Cell cycle analysis inoblastoma protein, which was phosphorylated (visualized by the appearance of an slower migrating band corresponding to phos- Approximately 1.5 ϫ 106 cells were washed in PBS and resuspended in 1.5 ml of hypotonic fluorochrome solution (50 ␮g/ml propidium iodide in phorylated pRb) from days 1–3 of stimulation. On day 4 the dis- 0.1% sodium citrate plus 0.1% Triton X-100; Sigma). Samples were stored appearance of the upper phosphorylated band of pRb was associ- for1hat4°Cinthedark before flow cytometric analysis of propidium ated with G1 arrest of activated B cells. By days 7 and 8, when all 4348 REGULATION OF p18INK4 EXPRESSION DURING HUMAN B CELL DIFFERENTIATION

shown). Because pRb phosphorylation during G1 was promoted by the cyclin D/cdk4 or cyclin D/cdk6 complexes, we next tested whether expression of cdk4 or cdk6 was down-regulated in acti-

vated cells re-entering G1. No signiﬁcant regulation of cdk6 (Fig. 2) or cdk4 (data not shown) expression was observed during the

time course of activation. Thus, G1 accumulation and the absence of pRb dephosphorylation observed in long term activated B lymphocytes did not appear to be due to the reduced expression of either cyclin D3 or cdk6. FIGURE 1. Ig production is associated with G cell cycle arrest. Human 1 Study of p27 and p18 expression during B cell differentiation tonsillar B lymphocytes (106/ml) were cultured for various times in the presence of SAC (1/10,000), IL-2 (10 ng/ml), and IL-10 (100 ng/ml). Ev- The activity of cdk is also regulated by cdk inhibitors belonging to ery day, culture supernatant was harvested and tested for Ig content, cell the INK4 and Cip/Kip families. We therefore investigated whether cycle analysis was performed on cells of each sample, and the percentage one or more of these inhibitors could be involved in the arrest of of cells in G1 phase was determined as described in Materials and Meth- cell cycle progression observed during in vitro B cell differentia- ods. In parallel, cell lysates (50 ␮g) were tested for pRb by immunoblotting tion. One member of the Cip/Kip family, p27, was strongly ex- with anti-pRb Ab. The same membrane was then stripped and probed with pressed in resting cells, and its expression decreased as long as anti-human actin Ab to assess the amount of protein loaded in each lane. these cells entered and progressed through the cell cycle. Its level Results are representative of at least three different experiments. of expression remained low even on days 6 and 7, when cells again

accumulated in G1 (Fig. 3). None of the INK4 family inhibitors tested (p15, p16, p18, and p19) was present at a signiﬁcant level in cells were arrested in G1 and terminal differentiation could be evidenced by the presence of Ig in the culture supernatants, the level resting B cells (Fig. 3 and data not shown). Interestingly, upon of total pRb protein expression was greatly diminished, returning activation with the combination of SAC/IL-2 and IL-10, an in- to a very low basal level close to that in resting cells. In addition, crease in p18 expression was observed. This up-regulation of p18 this decrease in pRb expression in our experimental conditions was expression was apparent on day 6 and was almost maximal after 7 not compensated by an increase in the amounts of other members days of activation (Fig. 3). In contrast, p15, p16, and p19 expres- of the pRb family, such as p107 or p130 (data not shown). sion levels were not modiﬁed (data not shown). Thus, these data show that p18 is highly expressed in activated cells that accumu-

G1 arrest is not associated with down-regulation of cyclin D3 or late in G1 and suggest that p18 may be involved in and responsible cdk6 expression for the absence of pRb phosphorylation in these stimulated cells. Cell cycle progression is tightly regulated by specific cyclin/cdk Inhibition of pRb phosphorylation is associated with formation complexes. We reported that the cyclin D3/cdk6 complex is the of cdk6/p18 complexes in vivo major regulator of the progression of human B cells to late G1. Indeed, this complex is directly responsible for pRb phosphoryla- Because the inhibitory activity of the INK4 inhibitors is specifi- by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. cally restricted to the cyclin D/cdk complexes controlling progres- tion, which is a key marker of late G and the G /S transition in B 1 1 sion in G , we investigated whether p18 is involved in the absence cells activated in vitro with mitogenic stimuli (12). We therefore 1 of pRb phosphorylation observed in G1-arrested cells. B cells were investigated whether the G1 accumulation and expression of non- phosphorylated pRb observed in long term activated B cells was cultured for 48 h or 7 days with SAC, IL-2, and IL-10, and cell due to down-regulation of either cyclin D3 or cdk6 expression. In lysates were immunoprecipitated with either control rabbit Ig or contrast to resting B cells, cyclin D3 was present in 24-h-activated anti-cdk6 Ab. The precipitates were analyzed for in vitro kinase B cells (Fig. 2). The level of cyclin D3 expression remained con- activity using recombinant pRb as the substrate (Fig. 4). The cdk6 stant until day 7. In parallel experiments we verified that among immune complex prepared on day 2 expressed a strong kinase activity compared with control Ig. In contrast, cdk6 immune com-

https://www.jimmunol.org the cyclin D family, only cyclin D3 expression was up-regulated in our experimental conditions, and that cyclin D1 or D2 was not plex isolated from cells activated for 7 days was not able to sig- detectable in SAC/IL-2/IL-10-activated B cells (12) (data not niﬁcantly phosphorylate pRb in vitro. These observations strongly Downloaded from

FIGURE 2. G1 arrest is not associated with down-regulation of cyclin FIGURE 3. Opposite expression of p27 and p18 during B cell differ- D3 or cdk6 expression. Human tonsillar B lymphocytes (106/ml) were entiation. Human tonsillar B lymphocytes (106/ml) were activated for var- activated for various times in the presence of SAC (1/10,000), IL-2 (10 ious times in the presence of SAC (1/10,000), IL-2 (10 ng/ml), and IL-10 ng/ml), and IL-10 (100 ng/ml). Cell lysates (50 ␮g) were tested for cyclin (100 ng/ml). Cell lysates (50 ␮g) were tested for p27 and p18 by immu- D3 and cdk6 by immunoblotting with specific anti-cyclin D3 or anti-cdk6 noblotting with specific anti-p27 or anti-p18 Ab (A and B, respectively). Ab (A and B, respectively). The amount of protein loaded in each lane was The amount of protein loaded in each lane was assessed by rehybridization assessed by reprobing the filter with an Ab specific for human actin. Re- of the filter with an Ab specific for human actin. Results are representative sults are representative of two different experiments. of four different experiments. The Journal of Immunology 4349

up-regulation of p18 expression, which binds to cdk6, dissociates the cyclin D3/cdk6 complex, and inhibits its activity. Such inhibition of cdk6 results in the absence of pRb phosphorylation and therefore prevents activated B lymphocytes from re-entering the cell cycle.

CD40-mediated B cell differentiation is also associated with up- regulation of p18 expression FIGURE 4. Inhibition of pRb phosphorylation is associated with formation of cdk6/p18 complexes in vivo. Human B cells (106/ml) were ac- To validate our previous observations, we next investigated tivated for 2 or 8 days in the presence of SAC (1/10,000), IL-2 (10 ng/ml), whether in vitro differentiation of human B lymphocytes following and IL-10 (100 ng/ml). Cell lysates were immunoprecipitated with either stimulation with CD40 was also associated with up-regulation of control rabbit serum (Ig) or anti-cdk6 Ab (cdk6), and kinase activity was p18 expression (Fig. 6). Tonsillar resting B cells (105 cells/ml) ␥ 32 assayed in vitro using recombinant pRb ( P ). After disappearance of were cultured for 7 days in the presence of irradiated murine radioactivity, the membrane was sequentially probed with anti-cdk6, anti- CD40L/CD32L fibroblastic cells (5 ϫ 103 cells/ml) and IL-10 p18, and anti-cyclin D3 Abs (Western blot). Results are representative of two different experiments. (100 ng/ml). Culture supernatants were then tested for human Ig, and cells were harvested. A fraction of the cells was analyzed by Western blotting for cdk inhibitor expression, and cell proliferation suggest that the in vivo pRb phosphorylation observed on day 2 was quantified by measuring radioactive thymidine uptake over 5 was dependent on the kinase activity of the cyclin D3/cdk6 com- 16hby10 cells (Fig. 6A)orbyG1 cell cycle analysis, quantified plex, whereas by day 7, cdk6 had lost its pRb phosphorylation by propidium iodide labeling and flow cytometric analysis of pro- activity. Because INK4 inhibitors abolish G cdk activity by dis- pidium iodide fluorescence of individual nuclei as well as cell 1 5 rupting the formation of cyclin D/cdk4/6 complexes, we investi- counting (Fig. 6B). The remaining cells (10 /ml) were cultured for gated whether p18, present in day 8 activated B cells, could asso- one (day 14; Fig. 6, A and B) or two (day 21; Fig. 6B) additional ciate with cdk6. Indeed, p18 was found in cdk6 immune periods of 7 days with CD40L/CD32L cells and IL-10 in fresh complexes prepared from 7-day-activated cells (Fig. 4). As ex- medium. In these conditions these cells undergo several prolifer- pected, the cyclin D3/cdk6 complex was dissociated when p18 was ative rounds, which can be detected by quantifying cells in G1. The associated with cdk6 on day 7 (Fig. 4), and the kinetics of the number of cells capable of re-entering the cycle (as measured by dissociation of the cyclin D3/cdk6 complex correlated with the kinetics of p18 expression (Figs. 3 and 5). In addition, we checked for the expression and activation of cdk2 which is involved in late

G1 progression and, when associated with cyclin E, additionally phosphorylates pRb. As previously reported, no cdk2 was ex-

pressed in G0 or early G1 B lymphocytes (12). Cdk2 was preferentially expressed by day 3 of activation when most of the cells are by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. cycling, and the level of cdk2 decreased to levels observed during

the ﬁrst 48 h of activation when cells are mostly in G0/G1 (Figs. 1 and 5). This expression of cdk2 was associated with activation, as indicated by the appearance of active cdk2 (visualized as a 33-kDa, faster migrating band corresponding to dephosphorylated cdk2) (19) and association with cyclin E (day 3 of Fig. 5). These data show that upon activation with the combination of SAC, IL-2, and IL-10, resting B cells enter the cell cycle, and their progression

through G1 is associated with pRb expression and its phosphory- https://www.jimmunol.org lation by the cyclin D3/cdk6 complex. Accumulation of these cells

in early G1, observed after 7 days of activation, is associated with Downloaded from

FIGURE 6. CD40-mediated B cell differentiation is also associated with up-regulation of p18 expression. Human tonsillar B cells (105/ml) were cultured for 7, 14, or 21 days in the presence of irradiated CDw32-CD40L cells and IL-10 (100 ng/ml). At the indicated time, supernatants were harvested and assayed for Ig content (A). Cells were counted, and the per-

FIGURE 5. Kinetics of cdk2/cyclin E and cdk6/cyclin D3 complexes centage of cells in G1 phase was determined as described in Materials and formation. Human tonsillar B lymphocytes (106/ml) were cultured for var- Methods and further cultured with freshly irradiated CD40L/CD32-L cells ious times in the presence of SAC (1/10,000), IL-2 (10 ng/ml), and IL-10 and IL-10 in fresh medium (B). In parallel, DNA synthesis was quantiﬁed (100 ng/ml). Every day, cell lysates were immunoprecipitated with either by measuring [3H]thymidine incorporation over 16 h for 105 cells from anti-cdk2 or anti-cdk6 Ab and subjected to Western analysis with anti-cdk2 each group. Cell lysates (50 ␮g) were tested for p18 by immunoblotting or cyclin E Ab for anti-cdk2 immunoprecipitations and with cyclin D3 for with anti-p18 Ab. The same membrane was sequentially stripped and anti-cdk6 immunoprecipitations, respectively. Results are representative of probed with anti-p27 and anti-human actin Ab (A). Results are represen- two different experiments. tative of at least three different experiments. 4350 REGULATION OF p18INK4 EXPRESSION DURING HUMAN B CELL DIFFERENTIATION

determining the percentage of G1 cells) decreased during the sec- etone to prevent apoptotic cell death). G1 arrest of Ramos cells ond period of culture (from days 7–14), and during the third period (assessed by both cell cycle analysis and pRb dephosphorylation)

(days 14–21) cells remained in G1, resulting in the absence of cell was associated with up-regulated p18 expression (Fig. 8, compare growth. In contrast, Ig secretion was detected on day 7, and Ig lane 2 to control nonsynchronized cells in lane 1). Concomitantly, concentrations were much higher in supernatants from 7 to 14 days the relative amounts of Ig in the supernatant also increased, sug-

of culture, when cell growth was slow (Fig. 6), and remained stable gesting that maximum Ig secretion was correlated to G1 cell cycle on day 21 (data not shown). Once again, terminal differentiation arrest and expression of p18. Furthermore, following induction of into Ig-secreting cells was correlated with a large regulation of p18 cell cycle progression by the addition of FCS, p18 expression and expression. In contrast, p27, which was highly expressed in resting Ig secretion were reduced in Ramos cells (Fig. 8, lanes 3 and 4). B cells, disappeared from activated cells. No major regulation of No signiﬁcant modulation of p27 was observed in cycling and

other cdk inhibitors was observed in these cells (data not shown). G1-arrested Ramos cells, whereas p27, but not p18, was highly up-regulated in the IL-2-dependent Kitt 225 T cell line arrested in p27 and p18 expression in normal B cells activated in vivo and G1 by IL-2 deprivation (Fig. 8, lanes 5 and 6). leukemic B cells To assess the particular expression of p18 and p27 during B cell Discussion activation and differentiation, we analyzed the expression levels of In this paper we investigate the interplay between the regulation of these two CKI in both resting and in vivo activated tonsillar B cells cell cycle progression and the ﬁnal differentiation process in hu- isolated through Percoll gradient. As shown in Fig. 7, activated B man B lymphocytes. Using different stimulatory combinations, we cells (CD38ϩ/IgDϪ; lane 2) expressed higher levels of p18 than observed that Ig secretion by normal tonsillar B cells occurs when Ϫ ϩ did resting B cells (CD38 /IgD ; lane 1). In contrast, p27 expres- cells stop progressing through the cell cycle and accumulate in G1. sion was restricted to resting B cells. We next studied the expres- The number of cell cycle divisions is variable according to the sion of p27 and p18 in two different non-Ig-secreting chronic lym- nature of the stimulus. For instance, after stimulation with the phocytic leukemia B cells (lanes 3 and 4) and in peripheral blood combination of the B cell mitogen SAC, IL-2, and IL-10, accu-

from a patient with Ig-secreting plasma cell leukemia (lane 5). We mulation of activated cells in G1 and subsequent Ig secretion take observed an opposite pattern of p18 and p27 expression in these place after a single round of proliferation, whereas G1 arrest and Ig different leukemia cells. p27 was highly expressed in chronic lym- secretion observed after activation with CD40 and IL-10 occur phocytic leukemia B (B-CLL) cells and to a lesser extent in leu- after several rounds of proliferation, leading to quantifiable cell kemic plasma cells, whereas p18 was very highly expressed in growth (Fig. 6B) (40, 41). Hasbold et al. reported that after CD40 Ig-secreting leukemic cells and was absent from the two B-CLL and IL-4 stimulation of murine B cells, IgG secretion was observed cells. after three rounds of proliferation, whereas IgE production was only evidenced after five proliferative rounds (42). This strongly p27 and p18 expression in different cell lines suggests that, at least in vitro, isotype specificity of secreted Ig is Various cell lines, including Burkitt (Ramos, BL41, Daudi, Raji, related to the number of proliferative rounds, consistent with the and CAPA-2) and myeloma (U266 and LP1) cells, were examined assumption that isotype switching preferentially takes place during

by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. for the expression of p18 and p27. All cells expressed signiﬁcant S progression (43, 44). Nevertheless, Ig secretion was always de- amounts of p18. In contrast, p27 expression was either low or too tected in our experimental conditions when cells were either ar-

low to be quantifiable in these cells (data not shown). All these rested in early G1 or leaving the cell cycle. The behavior of pRb cells, except for Daudi and Raji, which express, but do not secrete, may be a marker of this cell cycle exit. Thus, we observed that the Ig, are Ig-secreting cells. Thus, p18 expression correlated with the ability of B cell lines to produce Ig, although, as suggested by the expression of p18 in Daudi and Raji cells, its expression is not sufficient to ensure Ig secretion. In addition, p18 expression was observed in ␮ (BL41, Ramos, Daudi, Raji, and CAPA-2), ␥ (LP1), https://www.jimmunol.org and ⑀ (U266) Ig-expressing cells and thus was not correlated to a specific isotype. We next verified whether p18 expression was observed in these cells during all phases of the cell cycle or was

restricted to the G1 phase. For this, Ramos cells were arrested in G1 by a 3-day culture in the absence of FCS (but in the presence of the broad caspase inhibitor z-Val-Ala-D,L-Asp-ﬂuoromethylk- Downloaded from

FIGURE 8. G1 arrest of Ramos cells is associated with up-regulation of p18 expression. Ramos cells were cultured for 3 days in the presence (lane

1) or the absence (lane 2) of FCS. G1-arrested Ramos cells (lane 3) were cultured for an additional 3 days in the presence of FCS (lane 4). IL-2- dependent Kitt 225 T cells were cultured in the presence (lane 5)orthe FIGURE 7. High p18 expression in in vivo activated normal B cells and absence (lane 6) of IL-2 (10 ng/ml), such that they were synchronized or

plasma cell leukemia cells. Resting and in vivo activated B cells were not synchronized in G1, respectively. p18 was revealed by immunoblotting puriﬁed by Percoll fractionation as described in Materials and Methods. with anti-p18 Ab in cell lysates (50 ␮g). The same membrane was sequen- p18 and p27 expression was analyzed in resting (lane 1) or in vivo acti- tially stripped and probed with anti-p27, anti-pRb, and anti-human actin vated normal B cells (lane 2) and in chronic lymphocytic leukemia B cells Abs. Culture supernatants were harvested, and cells were enumerated. Ig (lanes 3 and 4) and leukemic plasma cells (lane 5). Cell lysates (50 ␮g) content is expressed as nanograms of total Ig per 106 cells. Cell cycle were tested for p27 by immunoblotting with anti-p27 Ab. The same mem- analysis was performed on cells of each sample, and the percentage of cells

brane was sequentially stripped and probed with anti-p18 and anti-human in the G1 phase was determined as described in Materials and Methods. actin Ab. Results are representative of two different experiments. Results are representative of at least two different experiments. The Journal of Immunology 4351

G1 accumulation observed from day 5 was associated with the arrested Ramos cells was associated with a speciﬁc increase in p18 dephosphorylation of pRb, which is the pRb pattern in early G1 with no apparent regulation of other CKI, underlining the partic- (Fig. 1). Moreover, terminal differentiation, evidenced by Ig se- ular pattern of p18 expression in differentiating cells already ob- cretion observed on days 7 and 8, was associated with a signiﬁcant served in normal B cells. decrease in the total pRb level, which was not compensated by an Interestingly, p18 expression and p27 expression appear to be increase in the amounts of other members of the pRb family such mutually exclusive in the cell types we studied. Thus, normal B as p130 or p107. This absence or lower pRb expression was also cells activated either in vitro or in vivo expressed large amounts of observed in resting B cells that had not yet entered the cell cycle p18 and almost no p27, whereas resting B cells expressed p27 but and is probably a property of B lymphocytes that are not cycling not p18. A similar dichotomy was observed between B-CLL cells

(45). Ig secretion also takes place during the G1 phase in cycling and leukemic plasma cells as well as in the different cell lines B cell lines, because Ig secretion by Ramos cells was signiﬁcantly analyzed. The signiﬁcance of this dissociation is not yet clear. A

increased when they were arrested by FCS deprivation in early G1. possible explanation for the differential pattern of p18 and p27 G1 progression is regulated by the kinases cdk4 and cdk6, which expression may be the speciﬁcities of their inhibitory activities are able, after association with cyclin D, to phosphorylate different during the cell cycle. Indeed, p18 is restricted to G1 cdk and as- substrates, including pRb. Such phosphorylation of pRb is corre- sociates preferentially with cdk6, which is the principal cdk con-

lated with the passage through the G1 restriction point (R1). Ad- trolling G1 progression in human B cells (12, 20). p27 can interact ditional phosphorylations of pRb by the cdk2/cyclin E complex are with various cdks controlling the different cell cycle steps. Thus,

necessary for the G1 to S transition (8–10, 14). In human B lym- p27 overexpression would be more potent in preventing resting B phocytes, cell cycle entry, triggered by mitogenic activation by cells from entering the cell cycle. In contrast, as B cell differen- ␮ IL-4 in combination with either anti- or anti-CD40 Ab, is depen- tiation requires a specific and mandatory stop in early G1, such dent on expression of cyclin D3 and, to a lesser degree, of cyclin arrest may be more specifically controlled by the specific inhibi- D2. Cyclin D3 associates preferentially with cdk6, and the cyclin tion of cyclin D3/cdk6-mediated-pRb phosphorylation by p18. The D3/cdk6 complex is responsible for the pRb phosphorylation ob- differential involvement of p18 and p27 during the control of B cell served both in vivo and in vitro (12). A similar pattern was ob- activation can also be observed in p18- and p27-deficient mice. served when B cells were stimulated with IL-10 in combination Indeed, although both deficient mice exhibited disproportionately with IL-2 and SAC. Cyclin D3, which is not expressed in resting enlarged and hyperplastic spleens, only B cells from p18-deficient B cells, was rapidly induced in 24-h-activated cells and was asso- mice showed a higher proliferative rate upon in vitro mitogenic ciated with cdk6 to form a cyclin D3/cdk6 complex responsible for activation with CD40L (31). Alternatively, as terminal differenti- pRb phosphorylation (Figs. 2 and 4). Nevertheless, although total ation is preferentially associated with apoptosis (see discussion amounts of cdk6 and cyclin D3 were not significantly modified, above), the different patterns of p18 and p27 expression observed cdk6 was not able to phosphorylate pRb on day 7 of culture when in B cells could be related to differential involvement of these two

these cells were arrested in early G1 (Figs. 1, 2, and 4). In contrast, CKI in the control of the apoptotic process. Indeed, high expres- expression levels of cdk2 and cyclin E are highly regulated during sion of p18 was very often observed in cells highly sensitive to cell cycle, and neither early (0–48 h) nor late (6–7 days) activated apoptosis, whereas expression of p27 was observed in cells, such

by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. cells express cyclin E and cdk2. This observation is in accordance as resting cells or the Kitt 225 T cells, more resistant to sponta- with the hypothesis that cell cycle arrest observed in our experi- neous in vitro apoptosis (data not shown). Additional experiments

mental conditions occurs in early G1. In addition to regulatory are necessary to elucidate the exact roles of p18 and p27 in the phosphorylations, the activity of cdks is closely dependent on the regulation of the sensitivity of normal B cells to apoptosis. CKI proteins (3, 15). This mechanism, involving CKI activity, is In conclusion, our data show that the two CKI, p18 and p27, are probably responsible for the cdk6 inhibition observed on day 7 of involved in different steps of B cell activation. Whereas p27 is culture, because 1) in contrast to p27, the expression of which associated with control of cell cycle entry of resting cells, p18 decreases upon activation, p18 expression was strongly up-regu- (through inhibition of cyclin D3/cdk6-mediated pRb phosphoryla-

lated in G1-arrested cells (day 7); and 2) the inactive cdk6 isolated tion) is more directly involved in induction of the G1 cell cycle https://www.jimmunol.org from these cells was associated with p18 (Figs. 3 and 4). A similar arrest necessary for terminal B lymphocyte differentiation. up-regulation of p18 expression was observed in both B cells activated in vitro with IL-10 and anti-CD40 Ab or B cells activated Acknowledgments in vivo (Figs. 6 and 7). In addition, high p18 expression was ob- We thank Dr. Sealand for the gift of CD40L/CD32L cells and Dr. De Revel served in cells from a patient with plasma cell leukemia, whereas for providing us with leukemic B cells. no p18 was present in two different nonsecreting B-CLL cell lines

Downloaded from (Fig. 7). These observations show that p18 is preferentially expressed in activated B cells engaged in ﬁnal differentiation. This References conclusion is in accordance with the previous observations that 1. Vazquez, A., J. P. Gerard, D. Olive, M. T. Auffredou, B. Dugas, S. Karray, J. F. Delfraissy, and P. Galanaud. 1986. Different human B cell subsets respond p18 is also up-regulated in EBV-positive CESS cells upon stimu- to interleukin 2 and to a high molecular weight B cell growth factor (BCGF). Eur. lation with IL-6 (which leads to cell cycle arrest and enhanced Ig J. Immunol. 16:1503. secretion) as well as during myogenesis (46, 47). 2. Paul, W. E., and R. A. Seder. 1994. Lymphocyte responses and cytokines. Cell 76:241. Beside their role in regulating cdk, CKI may also be involved in 3. Morgan, D. O. 1995. Principles of CDK regulation. Nature 374:131. other cellular functions. For instance, different groups reported that 4. Fisher, R. P. 1997. CDKs and cyclins in transition(s). Curr. Opin. Genet. Dev. p21 can exhibit anti-caspase properties (48). Therefore, it would be 7:32. useful to determine whether in addition to its involvement in cell 5. Rosenblatt, J., Y. Gu, and D. O. Morgan. 1992. Human cyclin-dependent kinase 2 is activated during the S and G2 phases of the cell cycle and associates with cycle regulation, p18 is directly involved in the differentiation pro- cyclin A. Proc. Natl. Acad. Sci. USA 89:2824. cess itself. Additional experiments, including transfection of dif- 6. Dulic, V., E. Lees, and S. I. Reed. 1992. Association of human cyclin E with a ferent B cell types with p18, are in progress to determine whether periodic G1-S phase protein kinase. Science 257:1958. 7. Sherr, C. J. 1993. Mammalian G1 cyclins. Cell 73:1059. p18 may directly regulate the process of Ig heavy chain produc- 8. Nevins, J. R. 1992. E2F: a link between the Rb tumor suppressor protein and viral tion. Nevertheless, the augmented Ig production observed in G1- oncoproteins. Science 258:424. 4352 REGULATION OF p18INK4 EXPRESSION DURING HUMAN B CELL DIFFERENTIATION

9. Dowdy, S. F., P. W. Hinds, K. Louie, S. I. Reed, A. Arnold, and R. A. Weinberg. 30. Nakayama, K., N. Ishida, M. Shirane, A. Inomata, T. Inoue, N. Shishido, I. Horii, 1993. Physical interaction of the retinoblastoma protein with human D cyclins. and D. Y. Loh. 1996. Mice lacking p27(Kip1) display increased body size, mul- Cell 73:499. tiple organ hyperplasia, retinal dysplasia, and pituitary tumors. Cell 85:707. 10. Ewen, M. E., H. K. Sluss, L. L. Whitehouse, and D. M. Livingston. 1993. TGF 31. Franklin, D. S., V. L. Godfrey, H. Lee, G. I. Kovalev, R. Schoonhoven, ␤ inhibition of Cdk4 synthesis is linked to cell cycle arrest. Cell 74:1009. S. Chen-Kiang, L. Su, and Y. Xiong. 1998. CDK inhibitors p18(INK4c) and 11. Reid, S., and E. C. Snow. 1996. The regulated expression of cell cycle-related p27(Kip1) mediate two separate pathways to collaboratively suppress pituitary proteins as B-lymphocytes enter and progress through the G1 cell cycle stage tumorigenesis. Genes Dev. 12:2899. following delivery of complete versus partial activation stimuli. Mol. Immunol. 32. Nourse, J., E. Firpo, W. M. Flanagan, S. Coats, K. Polyak, M. H. Lee, 33:1139. J. Massague, G. R. Crabtree, and J. M. Roberts. 1994. Interleukin-2-mediated 12. Blanchard, D. A., M. T. Affredou, and A. Vazquez. 1997. Modulation of the elimination of the p27Kip1 cyclin-dependent kinase inhibitor prevented by rapa- p27kip1 cyclin-dependent kinase inhibitor expression during IL-4-mediated hu- mycin. Nature 372:570. man B cell activation. J. Immunol. 158:3054. 33. Solvason, N., W. W. Wu, N. Kabra, X. Wu, E. Lees, and M. C. Howard. 1996. 13. Wagner, E. F., M. Hleb, N. Hanna, and S. Sharma. 1998. A pivotal role of cyclin Induction of cell cycle regulatory proteins in anti-immunoglobulin-stimulated D3 and cyclin-dependent kinase inhibitor p27 in the regulation of IL-2-, IL-4-, or mature B lymphocytes. J. Exp. Med. 184:407. IL-10-mediated human B cell proliferation. J. Immunol. 161:1123. 34. Franklin, D. S., and Y. Xiong. 1996. Induction of p18INK4c and its predominant 14. Brugarolas, J., K. Moberg, S. D. Boyd, Y. Taya, T. Jacks, and J. A. Lees. 1999. association with CDK4 and CDK6 during myogenic differentiation. Mol. Biol. Inhibition of cyclin-dependent kinase 2 by p21 is necessary for retinoblastoma Cell 7:1587. protein-mediated G1 arrest after gamma-irradiation. Proc. Natl. Acad. Sci. USA 35. Zindy, F., H. Soares, K. H. Herzog, J. Morgan, C. J. Sherr, and M. F. Roussel. 96:1002. 1997. Expression of INK4 inhibitors of cyclin D-dependent kinases during mouse 15. Sherr, C. J., and J. M. Roberts. 1995. Inhibitors of mammalian G1 cyclin-depen- brain development. Cell Growth Differ. 8:1139. dent kinases. Genes Dev. 9:1149. 36. Sanchez-Beato, M., A. I. Saez, J. C. Martinez-Montero, M. Sol Mateo, 16. Serrano, M., G. J. Hannon, and D. Beach. 1993. A new regulatory motif in L. Sanchez-Verde, R. Villuendas, G. Troncone, and M. A. Piris. 1997. Cyclin- cell-cycle control causing specific inhibition of cyclin D/CDK4. Nature 366:704. dependent kinase inhibitor p27KIP1 in lymphoid tissue: p27KIP1 expression is 17. Harper, J. W., G. R. Adami, N. Wei, K. Keyomarsi, and S. J. Elledge. 1993. The inversely proportional to the proliferative index. Am. J. Pathol. 151:151. p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent 37. Ciaparrone, M., H. Yamamoto, Y. Yao, A. Sgambato, G. Cattoretti, N. Tomita, kinases. Cell 75:805. T. Monden, H. Rotterdam, and I. B. Weinstein. 1998. Localization and expression 18. Xiong, Y., G. J. Hannon, H. Zhang, D. Casso, R. Kobayashi, and D. Beach. 1993. of p27KIP1 in multistage colorectal carcinogenesis. Cancer Res. 58:114. p21 is a universal inhibitor of cyclin kinases. Nature 366:701. 38. Chaouchi, N., L. Arvanitakis, M. T. Auffredou, D. A. Blanchard, A. Vazquez, and 19. Gu, Y., C. W. Turck, and D. O. Morgan. 1993. Inhibition of CDK2 activity in S. Sharma. 1995. Characterization of transforming growth factor-␤1 induced ap- vivo by an associated 20K regulatory subunit. Nature 366:707. optosis in normal human B cells and lymphoma B cell lines. Oncogene 11:1615. 20. Guan, K. L., C. W. Jenkins, Y. Li, M. A. Nichols, X. Wu, C. L. O’Keefe, 39. Hori, T., T. Uchiyama, M. Tsudo, H. Umadome, H. Ohno, S. Fukuhara, K. Kita, A. G. Matera, and Y. Xiong. 1994. Growth suppression by p18, a p16INK4/ and H. Uchino. 1987. Establishment of an interleukin 2-dependent human T cell MTS1- and p14INK4B/MTS2-related CDK6 inhibitor, correlates with wild-type line from a patient with T cell chronic lymphocytic leukemia who is not infected pRb function. Genes Dev. 8:2939. with human T cell leukemia/lymphoma virus. Blood 70:1069. 21. Polyak, K., M. H. Lee, H. Erdjument-Bromage, A. Koff, J. M. Roberts, 40. Rousset, F., E. Garcia, T. Defrance, C. Peronne, N. Vezzio, D. H. Hsu, P. Tempst, and J. Massague. 1994. Cloning of p27Kip1, a cyclin-dependent kinase R. Kastelein, K. W. Moore, and J. Banchereau. 1992. Interleukin 10 is a potent inhibitor and a potential mediator of extracellular antimitogenic signals. Cell growth and differentiation factor for activated human B lymphocytes. Proc. Natl. 78:59. Acad. Sci. USA 89:1890. 22. Toyoshima, H., and T. Hunter. 1994. p27, a novel inhibitor of G1 cyclin-Cdk 41. Rousset, F., S. Peyrol, E. Garcia, N. Vezzio, M. Andujar, J. A. Grimaud, and protein kinase activity, is related to p21. Cell 78:67. J. Banchereau. 1995. Long-term cultured CD40-activated B lymphocytes differ- 23. Hirai, H., M. F. Roussel, J. Y. Kato, R. A. Ashmun, and C. J. Sherr. 1995. Novel entiate into plasma cells in response to IL-10 but not IL-4. Int. Immunol. 7:1243. INK4 proteins, p19 and p18, are specific inhibitors of the cyclin D-dependent 42. Hasbold, J., A. B. Lyons, M. R. Kehry, and P. D. Hodgkin. 1998. Cell division kinases CDK4 and CDK6. Mol. Cell. Biol. 15:2672. number regulates IgG1 and IgE switching of B cells following stimulation by 24. Lee, M. H., I. Reynisdottir, and J. Massague. 1995. Cloning of p57KIP2, a cyclin- CD40 ligand and IL-4. Eur. J. Immunol. 28:1040. dependent kinase inhibitor with unique domain structure and tissue distribution. 43. Kenter, A. L., and J. V. Watson. 1987. Cell cycle kinetics model of LPS-stim- Genes Dev. 9:639. ulated spleen cells correlates switch region rearrangements with S phase. J. Im- 25. Matsuoka, S., M. C. Edwards, C. Bai, S. Parker, P. Zhang, A. Baldini, munol. Methods 97:111. KIP2 by guest on October 1, 2021. Copyright 2000 Pageant Media Ltd. J. W. Harper, and S. J. Elledge. 1995. p57 , a structurally distinct member of 44. Gronowicz, E. S., C. Doss, F. Assisi, E. S. Vitetta, R. L. Coffman, and S. Strober. the p21CIP1 Cdk inhibitor family, is a candidate tumor suppressor gene. Genes 1979. Surface Ig isotypes on cells responding to lipopolysaccharide by IgM and Dev. 9:650. IgG secretion. J. Immunol. 123:2049. 26. Deng, C., P. Zhang, J. W. Harper, S. J. Elledge, and P. Leder. 1995. Mice lacking 45. van der Sman, J., N. S. Thomas, and E. W. Lam. 1999. Modulation of E2F CIP1/WAF1 p21 undergo normal development, but are defective in G1 checkpoint complexes during G0 to S phase transition in human primary B-lymphocytes. control. Cell 82:675. J. Biol. Chem. 274:12009. 27. Serrano, M., H. Lee, L. Chin, C. Cordon-Cardo, D. Beach, and R. A. DePinho. 46. Morse, L., D. Chen, D. Franklin, Y. Xiong, and S. Chen-Kiang. 1997. Induction 1996. Role of the INK4a locus in tumor suppression and cell mortality. Cell of cell cycle arrest and B cell terminal differentiation by CDK inhibitor 85:27. p18(INK4c) and IL-6. Immunity 6:47. 28. Fero, M. L., M. Rivkin, M. Tasch, P. Porter, C. E. Carow, E. Firpo, K. Polyak, 47. Phelps, D. E., K. M. Hsiao, Y. Li, N. Hu, D. S. Franklin, E. Westphal, E. Y. Lee, L. H. Tsai, V. Broudy, R. M. Perlmutter, et al. 1996. A syndrome of multiorgan and Y. Xiong. 1998. Coupled transcriptional and translational control of cyclin- hyperplasia with features of gigantism, tumorigenesis, and female sterility in dependent kinase inhibitor p18INK4c expression during myogenesis. Mol. Cell. https://www.jimmunol.org p27(Kip1)-deficient mice. Cell 85:733. Biol. 18:2334. 29. Kiyokawa, H., R. D. Kineman, K. O. Manova-Todorova, V. C. Soares, 48. Suzuki, A., Y. Tsutomi, N. Yamamoto, T. Shibutani, and K. Akahane. 1999. E. S. Hoffman, M. Ono, D. Khanam, A. C. Hayday, L. A. Frohman, and A. Koff. Mitochondrial regulation of cell death: mitochondria are essential for procaspase 1996. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor 3-p21 complex formation to resist Fas-mediated cell death. Mol. Cell. Biol. function of p27(Kip1). Cell 85:721. 19:3842. Downloaded from