The INK4 family of proteins consists of four members which can block progression from the G1-to-S phase of the cell cycle by inhibiting the activity of cyclin dependent kinases (cdks) 4 and 6. Although the gene encoding p16INK4a is commonly inactivated in human tumors, p18INK4c is rarely altered. We show here that overexpression of p18INK4c does not block cell cycle progression in a T-cell acute lymphocytic leukemia cell line (CEM) sensitive to p16INK4a-mediated G1 arrest. A chimera consisting of the kinase-binding region of p16INK4a fused to the COOH terminus of p18INK4c is active in all known biochemical assays for INK4 function, but it does not arrest CEM cells. These data imply a novel level of p18INK4c regulation mediated through the COOH terminus and suggest that functional differences might underlie the distinct mutational profiles observed for p16INK4a and p18INK4c in tumors.

The inactivation of CKIs3 is an important event in the development of human cancer (1, 2). The INK4 family of proteins currently consists of four CKIs: p16, p15, p18, and p19. INK4 proteins share a common structural organization and all show in vitro binding specificity for only cdk4 and cdk6 among the known cdks. Although each member behaves identically in the established in vitro assays for INK4 function (3, 4, 5, 6), only the genes encoding p15 and p16 are inactivated in a significant proportion of human cancers (7, 8, 9). Inactivation of p16 appears to be particularly important in the development of melanoma and T-cell ALL (reviewed in Refs. 10 and 11). The p16 gene is mutated in a high proportion (>80%) of primary human T-cell ALLs and is, in fact, the most commonly mutated known gene in this tumor type (12, 13). The human p18 protein shares 38% amino acid identity with p16, binds the same target cdks as p16, and, like p16, can provoke cell cycle arrest when exogenously expressed in cell lines retaining wild-type pRb (3, 9). Because of these similarities, it has generally been assumed that the functions of p18 and p16 are essentially redundant (1). However, several leukemia cell lines express high levels of endogenous p18, a situation that should not be tolerated in these rapidly dividing cells if p18 and p16 indeed perform identical biological functions. CEM cells, which are derived from a T-cell ALL and are p16 null, express high levels of p18 (3, 14). Although p16 and p18 behave identically in in vitro assays, p18 has been reported to preferentially associate with cdk6 in vivo(3, 4). Others have found that p18 associates with cdk4 in vivo, but that this association is less stable than the association of p16 with cdk4 (14). The goal of this study was to determine whether p16 and p18 function differently in CEM cells, and, if so, whether this difference is attributable to differential association with cdk4. Functional differences between the two proteins could help explain the different mutational profiles observed for each CKI in human cancer.

Plasmid Construction.

The p16/18 chimera was constructed by PCR. The 5′ part of the cDNA was generated using the primers 5′mycp16 (5′-GGGAATTCAAATGGAGCCGGCGGCGGGG-3′) with 3′p16/18 (5′-GCACGGTAGCTGGTCAAGCACACGGCCAGC-3′) and the human p16 cDNA as a template. PCR conditions were 30 cycles of 94° for 1 min and 72° for 1 min. The 3′ piece of cDNA was amplified with the primers 5′p16/18 (5′-GCTGGCCGTGTGCTTGACCAGGTACCGTGC-3′) with 3′mycp18 (5′-TGCCTCGAGTTATTGAAGATTTGTGGC-3′) and the human p18 cDNA as a template. PCR conditions were 30 cycles of 92° for 1 min, 55° for 1 min, and 72° for 1 min. The resulting PCR products were gel-purified, combined, and used as a template to generate a full-length chimeric cDNA by PCR with the primers 5′mycp16 and 3′mycp18. PCR conditions were 94° for 1 min, 57° for 1 min, 72° for 1 min. p16 and p18 were amplified in a similar manner and cloned using a 5′ EcoRI site and a 3′ XhoI site into a version of pCDNA3 (Invitrogen, Carlsbad, CA) containing a Myc epitope tag. DNA for transfections was prepared using a cesium chloride gradient or with an EndoFree kit (Qiagen, Valencia, CA). The integrity of all constructs was verified by sequencing.

Cell Culture and Transfections.

U2OS and CEM cells were obtained from the American Type Culture Collection and maintained as suggested. U2OS cells were transfected using the Effectene transfection kit (Qiagen) according to manufacturer’s instructions. CEM cells were transfected by electroporation in an Electro Square Porator (BTX, San Diego, CA) on the low-voltage setting using a single pulse of 65 ms in duration and a field strength of 550 V/cm. Just before transfection, the CEM cells were washed twice in PBS and resuspended at a density of 30 × 106 cells/ml in RPMI + l-glutamine (2 mm). After electroporation, the cells were incubated 15 min at room temperature before plating in complete medium.

Flow Cytometry.

Cells were cotransfected with a plasmid encoding a membrane-targeted GFP-F and the construct of interest at a molar ratio of 1:5. Forty-eight h after transfection, cells were fixed in 80% ethanol for a minimum of 30 min at −20°C. Cells were washed once with PBS and once with PBS + 1% fetal bovine serum and then resuspended at ≤1 × 106cells/ml in a PBS + 1% fetal bovine serum solution containing RNase (250 μg/ml) and propidium iodide [diluted from a 50× stock of 0.5 mg/ml in 38 mm sodium citrate (pH 7.00)]. The cells were incubated at 37°C for at least 30 min, then stored at 4°C overnight before analysis. Data acquisition and analysis was performed on a Coulter EPICS XL-MCL flow cytometer. Gates were set to restrict analysis to GFP-positive cells, and peak area/peak height ratios were used to discriminate doublets. Cell cycle distribution was determined using the Modfit program (Verity Software, Topsham, ME).

In Vitro Kinase Assays.

Bacterially expressed INK4 proteins were prepared and added to an in vitro kinase assay as described (5). Recombinant, active cyclin D1/cdk4 was a generous gift from Dr. Robert Booher (Onyx Pharmaceuticals, Richmond, CA).

Immunoprecipitation and Immunoblotting.

Cell lysates were prepared as described (5). Briefly, cells were lysed in EIA lysis buffer + inhibitors [50 mm HEPES (pH 7.0), 250 mm NaCl, 0.1% NP 40, and 5 mm EDTA with 1 mm 4-(2-aminoethyl)benzenesulfonylfluoride, 1 mm DTT, 1 μg/ml aprotinin, 1 μg/ml pepstatin A, 1 μg/ml leupeptin, 0.1 mm sodium vanadate, and 1 mm sodium fluoride], incubated 10 min on ice, and spun 20,000 × g at 4°C for 10 min before the supernatant was transferred to a clean tube. Samples were mixed with an equal volume of 2× Laemmli sample buffer, boiled 5 min, and separated by SDS-PAGE before immunoblotting with antibodies versus the myc tag (9E10; purified from hybridoma supernatant), cdk4 (Santa Cruz Biotechnology H-22, Santa Cruz, CA), or pRB (PharMingen, San Diego, CA).

For immunoprecipitation, lysates were incubated 1 h with primary antibody on ice. Immune complexes were captured by the addition of protein G-Sepharose (Amersham-Pharmacia Biotech, Inc., Piscataway, NJ) and then 30 min of rocking at 4°C. The beads were washed four times with lysis buffer and then boiled in sample buffer before electrophoresis.

Pulse Chase Analysis.

Twenty-four h after transfection, cells were washed twice in PBS and incubated for 20 min in medium free of methionine and cysteine. 35S-methionine-labeling mix (New England Nuclear, Boston, MA) was added to the medium and the cells were incubated for 4 h at 37°C. The cells were then washed twice with PBS and fresh, unlabeled medium was added for the indicated times. The cells were lysed as above. The lysates were precleared by the addition of protein G-Sepharose 4 gel beads which were prebound to normal rabbit serum. Immunoprecipitation was carried out as above. Immunoprecipitates were separated by SDS-PAGE and visualized by autoradiography.

Isolation of Transfected Cells.

The pHOOK system (Invitrogen, Carlsbad, CA) was used to separate transfected cells from nontransfected cells. Briefly, pHOOK encodes a single chain antibody which recognizes the hapten antigen phOx. Cells expressing the pHOOK antibody are captured by binding to the phOX hapten conjugated to magnetic beads. Cells were cotransfected with the pHOOK plasmid and the DNA of interest at a ratio of 2:1. Twenty-four h after transfection, the cells were resuspended in 1 ml of medium, to which 15 μl of the bead solution were added. The cells were rocked gently for 30 min at 37°C. A magnet was used to collect cells bound to the magnetic beads. The bound cells were washed three times in whole media before replating. Sixteen h later, the cells were harvested and lysates were prepared as described above.

p16 and p18 Are Functionally Distinct.

The T-cell ALL cell line CEM has been reported to express high levels of the protein p18 (3, 14). Because the vast majority of T-cell ALLs are p16-null and sensitive to p16-mediated cell cycle arrest (11), we wanted to determine whether the CEM cell line responded differently to p16 and p18. CEM cells were transiently transfected with myc-tagged versions of either p16 or p18 under control of the powerful CMV promoter. The cell cycle distribution of the transfected cells was subsequently analyzed by flow cytometry. A plasmid encoding the GFP was cotransfected as a marker to allow for identification of successfully transfected cells. As shown in Fig. 1, wild-type p16 induced a reproducible accumulation of CEM cells in the G1 phase of the cell cycle under these conditions. In contrast, p18 failed to cause a G1 arrest and behaved similarly to the nonfunctional, tumor-derived p16 mutant P114L (5). Equivalent expression of the proteins in each cell line was verified by Western blotting (Fig. 2,C and 3,A and data not shown). The intrinsic activity of p18 is clearly not limiting, as it is capable of arresting cell cycle progression to the same extent as p16 in U2OS cells (Fig. 1). These data demonstrate that p16 and p18 are not functionally redundant and that their respective activities are cell type-dependent. This result is in agreement with reports that adenoviral expression of p16 and p18 produced differing results with respect to cell cycle arrest according to the cell type used (15).

A Functional Chimera with the p16 cdk-binding Domain Does Not Arrest CEM Cells.

The INK4 family members consist of a series of ankyrin repeats. p15 and p16 contain four ankyrin repeats, whereas p18 and p19 each consist of five ( Ref. 9 and Fig. 2,A). The crystal structures of p16 bound to cdk6 and of p19 bound to cdk6 have been solved, and they suggest that the residues necessary for kinase binding are contained in the first four ankyrin repeats of the INK4 proteins (16). Nuclear magnetic resonance data suggest that the fifth ankyrin repeat of p18 is not necessary for kinase binding, but that it might be important for stabilizing the protein (17). Moreover, the crystal structure of an inactive ternary structure containing cyclin K (a viral d-type cyclin), p18, and cdk6 confirms that the fourth and fifth ankyrin repeats of p18 do not participate in association with the kinase (18). Consistent with these observations, the domains of greatest divergence between INK4 family members lie in the COOH-terminal regions of the proteins beyond the fourth ankyrin repeat (Ref. 9 and Fig. 2 A).

To determine whether differential association with cdk4 mediates differences between p16 and p18 in CEM cells, we constructed chimeras in which the COOH-terminus of p18, including its fifth ankyrin repeat, was fused to the first four ankyrin repeats of p16 (p16/18; see Fig. 2 A for the site of exchange). Because this chimera contains the p16 residues important in kinase association, we predicted that the protein would mimic p16 in its kinase preference. The reverse chimera was also constructed in which the COOH terminus of p16 was fused to p18 after the fourth ankyrin repeat, as was a truncated version of p18 consisting of only its first four ankyrin repeats. p18 constructs lacking the fifth ankyrin repeat were not recovered from mammalian cell extracts in quantities large enough to be detected by Western blot, in agreement with the assertion that this part of the protein contributes to the stability of p18 (17). Additionally, the p18/16 chimera was recovered only in minute quantities from bacterial expression lysates, leading us to believe that the instability of this protein is not limited to mammalian cells (data not shown).

We verified the functional integrity of the p16/18 chimera using the three established assays for INK4 function: (a) binding to cdk4 and cdk6; (b) inhibition of cdk4/6 activity, and (c) the ability to arrest cell cycle progression in certain cell lines (19). To test the ability of each construct to inhibit cyclin D1/cdk4 complexes in vitro, His-tagged INK4 proteins were expressed in bacteria and purified to apparent homogeneity. These proteins were added to an in vitro kinase reaction containing active cyclin D1/cdk4 complexes and a COOH-terminal fragment of the pRb as a substrate. p16, p18, and p16/18 all inhibited 32P incorporation onto pRb in a dose-dependent manner, with each CKI requiring a similar molar excess relative to cdk4 for full inhibition (Fig. 2 B). This effect is specific, because the addition of equivalent amounts of p16 P114L did not inhibit cyclin D1/cdk4 activity.

To determine whether the p16/18 chimera was capable of binding cdk4 in mammalian cells, U2OS cells were transfected with myc-tagged versions of p16, p18, p16/18, and p16 P114L. As shown in Fig. 2,C, immunoprecipitation through the myc tag recovered roughly equivalent amounts of each of the transfected inhibitors. Western blotting the immunoprecipitated complexes for cdk4 confirmed that p16, p18, and p16/18 were each associated with cdk4 in U2OS cells (Fig. 2 C). Again, the specificity of this interaction was confirmed using the p16 P114L variant as a negative control.

A third assay for INK4 function is the ability to arrest certain pRb-positive cell lines. We chose to analyze the activity of the chimera in the osteosarcoma cell line U2OS because it is an environment in which p16 and p18 have been shown to behave similarly (3, 5). As expected, the expression of both p16 and p18 caused a pronounced accumulation of U20S cells in the G0/G1 phase of the cell cycle (Fig. 2,D). p16/18 is capable of arresting these cells to the same extent as either p16 or p18. Immunoprecipitation followed by Western blotting of lysates from this experiment verified equivalent expression of each construct (Fig. 2 C). In all of these assays, which collectively represent the currently accepted definitive proof of INK4 activity, both p18 and the p16/18 chimera are indistinguishable from p16.

When transfected into CEM cells, however, the p16/18 chimera did not arrest cell cycle progression. As shown in Fig. 3,A, expression of p16 resulted in a marked depletion of the proportion of cells in S-phase, whereas p18, the p16/18 chimera, and the p16 mutant P114L had little effect on cell cycle distribution. Equivalent expression of the constructs was verified by Western blotting lysates prepared from an aliquot of the same cells used for flow cytometry (Fig. 3,A, bottom right). The input amount of each plasmid construct required to yield approximately equivalent protein expression was determined empirically and was as follows: vector, 25 μg; p16, 25 μg; p18, 25 μg; p16/18, 75 μg; and p16 P114L, 50 μg. In all experiments where equal amounts of DNA were introduced into CEM cells, transfection with the p16 construct resulted in G1 arrest, whereas the expression of p18 and p16/18 did not. The experiment shown here was performed exactly as those depicted in Fig. 1, and the data are representative of at least four independent replicates.

To verify that the p16/18 chimera is incapable of causing a G1 arrest in CEM cells, we repeated this experiment in the presence of the mitotic spindle inhibitor nocodazole. Nocodazole was added to the culture media 16 h before harvest to impose a secondary block at the G2-M boundary. Cells that are not arrested at the G1-S transition will proceed through the cell cycle and accumulate at the nocodazole block point in G2. As shown in Fig. 3 B, only wild-type p16 was able to impose a G1 block in these cells, despite the equivalent protein expression levels of each test construct. These data clearly indicate that expression of the p16/18 chimera does not block cell cycle progression in CEM cells. The plasmid DNA amounts used in this experiment were adjusted as before to produce equivalent levels of protein expression, and were as follows: vector, 25 μg; p16, 30 μg; p18, 25 μg; p16/18, 75 μg; and p16 P114L, 75 μg.

It has been proposed that differences between p16 and p18 are attributable to either the relatively decreased in vivo affinity of p18 for cdk4 (3, 4) or to a decreased in vivo stability of p18/cdk4 interactions (14). Because our chimera contains all of the residues of p16 thought to participate in kinase binding (16, 20), we reasoned that the chimera should associate with cdks just as p16 does. Therefore, the functional differences we observed should not be attributable to an intrinsic preference of p16/18 for cdk6 over cdk4. However, because the loop connecting the fourth and fifth ankyrin repeats of p18 might play a role in p18 function (21), and because this loop is part of the chimera, we cannot definitively rule out a difference in kinase preference.

To determine whether the p16/18 chimera associates stably with cdk4 in CEM cells, we performed a pulse chase analysis (14). CEM cells were transfected with either an empty vector or with constructs encoding the myc-tagged inhibitor of interest. Twenty h after transfection, the cells were pulse-labeled with 35S-methionine for 4 h and chased for 0, 2, 4, or 19 h with cold medium. At each time point, lysates were prepared and immunoprecipitated with the 9E10 antibody against the myc-tag. Proteins recovered in the immunocomplexes were visualized by autoradiography. Two bands corresponding to cdk4 and cdk6 were immunoprecipitated with p16, p18, and p16/18, but were not present in the vector control (Fig. 4). In contrast to a previous report we found no gross difference between p16 and p18 with respect to the stability of the complexes formed with cdk4 (14). Subtle differences in the stability of kinase/inhibitor complexes cannot be ruled out, as evidenced by examination of time points later than those reported in the previous study (see the 19-h time points). A likely explanation for the difference in results lies in our use of an overexpression system versus the examination of endogenous proteins. An enforced excess of p18 protein could drive the equilibrium in favor of the formation of complexes with cdk4 regardless of the inherent stability of the association. Still, it is difficult to understand why CEM cells continue to divide in the presence of overexpressed p18 when the protein readily associates with cdk4. Similar binding data were obtained when this experiment was repeated in U2OS cells (data not shown), with the exception that these cells contain significantly less cdk6 (14).

Because it appears that both p18 and p16/18 associate readily with cdk4 in CEM cells, we assessed whether phosphorylation of the endogenous pRb was affected under these conditions. Hyperphosphorylated pRb can be distinguished from hypophosphorylated pRb on a Western blot by the slower relative mobility of the phosphorylated form (22). CEM cells were transfected with the construct of interest along with a plasmid encoding a single chain antibody against the hapten phOx. Transfected cells were separated from nontransfected cells using magnetic beads conjugated to the phOx antigen. Lysates prepared from the separated cells were probed with an antibody that recognizes pRb in all of its various phospho-forms. Expression of p16 caused an accumulation of pRb in its hypophosphorylated, active form (Fig. 4 B). The expression of p18 did not completely prevent the phosphorylation of pRb, although there seems to be an accumulation of the lower band relative to vector-transfected cells. The expression of the p16/18 chimera had no effect on the phosphorylation status of pRb. These results suggest that the failure of p18 and the p16/18 chimera to arrest CEM cells is attributable to an inability to inhibit phosphorylation of pRb.

Taken together, our results are consistent with three principal conclusions: (a) p16 and p18 are not functionally equivalent; (b) p18 is subject to a cell type-specific regulation that prevents this inhibitor from arresting CEM cells despite the intrinsic capability of p18 to inhibit the same target kinases as p16; and (c) this regulatory influence can be conferred onto p16 by transferral of the fifth ankyrin repeat of p18. Previous reports suggest that p16 and p18 might be distinguished by their relative affinities for cdk4 and cdk6 (3, 4). Our results argue against this explanation for functional differences between the inhibitors, because the p16/18 chimera should have the same affinity for cdk4 as does p16. If the proposed in vivo destabilization of p18/cdk4 complexes (14) is the cause of phenotypic differences between p16 and p18, our results would indicate that: (a) very transient dissociation of p18 and cdk4 is sufficient to allow kinase activity and to promote the inactivation of pRb, and (b) that destabilization of the p18/cdk complex is mediated by the COOH-terminal region of p18 (2). This seems the most likely explanation in light of the fact that p18 exists primarily as a monomer or in a heterodimeric complex with cdk4 or cdk6 in CEM cells (14), and these complexes are catalytically inactive in vitro.

One vexing issue regarding p18 has been how to resolve the apparent contradiction between the complexes this inhibitor forms in cells and the measured biochemical affinities observed in vitro. In vitro, p18 can form stable complexes with cdk4 with affinity equal to that of p16, but in vivo p18 is found to associate preferentially with cdk6 (3, 4, 23, 24). Moreover, endogenous p18/cdk4 complexes appear to be unstable, implying that currently undefined cellular components may influence the turnover rate of the complex (14). In the current study, we have provided additional evidence that p18 must be subject to in vivo posttranslational regulatory mechanisms that modulate its activity. Our data implicate a region of p18 that does not participate in kinase binding.

Our results and those of others cited above highlight a fundamental discrepancy between the behavior of p18 under in vitro and in vivo conditions. Because cellular INK4 complexes contain only an INK4 protein and a partner cdk (14), we conclude that some property resident in the fifth ankyrin repeat of p18 allows for modulation of the stability of the INK4/cdk complex through a transient interaction with additional cellular factors. We believe that the transferrable nature of the difference between p18 and p16 strongly suggests the existence of an active regulatory process mediated through the COOH terminus of p18. Such a level of regulation would be an additional function for the fifth ankyrin repeat of p18, as it has been suggested previously that this region may contribute to the stability of the protein (17, 21).

Our model draws a distinction between INK4 proteins which contain five ankyrin repeats, such as p18, and those which contain four repeats, such as p16. We hypothesize that the five-repeat INK4 proteins may act as reversible inhibitors of cell cycle progression, whereas four-repeat INK4 proteins impose an irreversible cell cycle block. In this model, p18 can act as a conditional regulator of cell cycle progression in settings where reentry into a proliferative state may be required despite a prior withdrawal from the cell cycle. For example, p18 is believed to participate in the cessation of cell division that accompanies differentiation in certain cell types (25, 26). Under some circumstances, these differentiated cells may be required to reenter the cell cycle, such as in the proliferation of specific hematopoietic cell lineages. We propose that the fifth ankyrin repeat of p18 provides a domain that participates in conditional regulatory interactions that can destabilize preexisting p18/cdk complexes to release the cdk for activation. Because overexpressed p18 associates with cdk4 in CEM cells (Fig. 4 A), we predict that this destabilization would be transient and occur only when a cell is stimulated to increase cdk4 activity. In contrast, we propose that four-repeat INK4 proteins such as p16 are involved primarily in imposing irreversible cell cycle arrest, such as occurs during cellular senescence (27). Because exit from the cycle under these conditions is believed to be essential for ensuring that cells undergo only a finite number of cell divisions, p16 activity would not be subject to cellular modifications that could alter the stability of p16/cdk complexes.

This model could explain why p16 is targeted for genetic alterations in cancer whereas p18 is not. In the progenitors for cancers such as T-cell ALL, cellular mechanisms for the posttranslational inactivation of p18 may already exist. The regulatory interactions that occur through the fifth ankyrin repeat may selectively down-regulate the activity of p18 in this setting. Phosphorylation of p18 may be a plausible candidate regulatory mechanism, inasmuch as p18 has been shown to be phosphorylated on serine residues in vivo(24). Because an incipient cancer cell may thus have alternative means for bypassing p18 function, there would be minimal selective pressure for direct mutation of the p18 gene. Conversely, there would be significant selective pressure for direct mutational inactivation of p16 (or components in the p16 pathway) during tumorigenesis, because p16 does not contain the regulatory fifth ankyrin repeat, and therefore the cell has no alternative means for bypassing the p16-imposed cell cycle arrest. Currently, studies are under way to test this model and to elucidate further the mechanisms by which p16 differs from p18 in vivo. Future experiments will be required to determine more precisely the role these inhibitors play in the transformed environment and to define the biological activities that are targeted for inactivation in human tumors.

Fig. 1.

p16 and p18 are not functionally redundant. The indicated inhibitors were transfected into either CEM cells (□) or U2OS cells (▪). The change in G1 population is expressed relative to vector-transfected controls. Bars, SD of at least three experiments for each condition. Approximately equivalent expression of each construct was verified by Western blotting for an N NH2-terminal myc-tag (Fig. 2,C, 3 A, and data not shown).

Fig. 1.

p16 and p18 are not functionally redundant. The indicated inhibitors were transfected into either CEM cells (□) or U2OS cells (▪). The change in G1 population is expressed relative to vector-transfected controls. Bars, SD of at least three experiments for each condition. Approximately equivalent expression of each construct was verified by Western blotting for an N NH2-terminal myc-tag (Fig. 2,C, 3 A, and data not shown).

Close modal
Fig. 2.

Characterization of the p16/18 chimera. A, alignment of p16 and p18. Regions of amino acid identity are shaded. Labeled boxes above the primary sequence, the first and second helices of each ankyrin repeat (e.g., the first helix of the third ankyrin repeat is labeled 3A). Arrow, the site of exchange for the p16/18 chimera. B, inhibition of recombinant cyclin D1/cdk4. Purified INK4 proteins were added at the indicated molar excess relative to cdk4 in in vitro kinase assays using a COOH-terminal fragment of pRb as a substrate. The kinase reactions were resolved on an 8% polyacrylamide gel, and the phosphorylated substrate was visualized by autoradiography. C, association with cdk4. Constructs encoding the indicated myc-tagged INK4 proteins were cotransfected with GFP into U2OS cells. Cell lysates prepared 48 h after transfection were subjected to immunoprecipitation using the anti-myc antibody 9E10. Immunoprecipitates were resolved on Western blots and probed with antibodies against either the myc-tag (9E10) or cdk4. D, cell cycle distribution of cells expressing INK4 proteins. U2OS cells harvested 48 h after transfection were fixed and stained with propidium iodide for flow cytometry. Gates were set to restrict analysis to cells expressing the cotransfected GFP marker. Each histogram profile depicts data from at least 5000 GFP-positive cells. The proportion of cells in the G1 phase of the cell cycle was calculated using the ModFit flow cytometry analysis software.

Fig. 2.

Characterization of the p16/18 chimera. A, alignment of p16 and p18. Regions of amino acid identity are shaded. Labeled boxes above the primary sequence, the first and second helices of each ankyrin repeat (e.g., the first helix of the third ankyrin repeat is labeled 3A). Arrow, the site of exchange for the p16/18 chimera. B, inhibition of recombinant cyclin D1/cdk4. Purified INK4 proteins were added at the indicated molar excess relative to cdk4 in in vitro kinase assays using a COOH-terminal fragment of pRb as a substrate. The kinase reactions were resolved on an 8% polyacrylamide gel, and the phosphorylated substrate was visualized by autoradiography. C, association with cdk4. Constructs encoding the indicated myc-tagged INK4 proteins were cotransfected with GFP into U2OS cells. Cell lysates prepared 48 h after transfection were subjected to immunoprecipitation using the anti-myc antibody 9E10. Immunoprecipitates were resolved on Western blots and probed with antibodies against either the myc-tag (9E10) or cdk4. D, cell cycle distribution of cells expressing INK4 proteins. U2OS cells harvested 48 h after transfection were fixed and stained with propidium iodide for flow cytometry. Gates were set to restrict analysis to cells expressing the cotransfected GFP marker. Each histogram profile depicts data from at least 5000 GFP-positive cells. The proportion of cells in the G1 phase of the cell cycle was calculated using the ModFit flow cytometry analysis software.

Close modal
Fig. 3.

The p16/18 chimera does not arrest CEM cells. A, CEM cells were cotransfected by electroporation with a GFP marker and the indicated constructs. Plasmid DNA quantities for the electroporation of the indicated constructs were determined empirically to give approximately equivalent expression by Western blotting (lower right). B, cells were transfected as in A, except that nocodazole was added to the cultures for 16 h before collecting the cells for flow cytometry analysis. The proportion of cells in G1 phase is as follows: vector, 2%; p16, 45%; p18, 18%; p16/18, 18%; and p16P114 L, 15%. Each histogram profile depicts data from at least 5000 GFP-positive cells.

Fig. 3.

The p16/18 chimera does not arrest CEM cells. A, CEM cells were cotransfected by electroporation with a GFP marker and the indicated constructs. Plasmid DNA quantities for the electroporation of the indicated constructs were determined empirically to give approximately equivalent expression by Western blotting (lower right). B, cells were transfected as in A, except that nocodazole was added to the cultures for 16 h before collecting the cells for flow cytometry analysis. The proportion of cells in G1 phase is as follows: vector, 2%; p16, 45%; p18, 18%; p16/18, 18%; and p16P114 L, 15%. Each histogram profile depicts data from at least 5000 GFP-positive cells.

Close modal
Fig. 4.

p18 and p16/18 associate with cdk4 but do not prevent pRb phosphorylation. A, CEM cells transfected with an empty vector control or with the indicated myc-tagged INK4 constructs were pulse-labeled with 35S-methionine and then switched to unlabeled media for 0, 2, 4, or 19 h before lysates were prepared. Lysates from each time point were incubated with the anti-myc antibody 9E10, and the immunoprecipitated material was resolved on a 12% SDS-PAGE gel and visualized by autoradiography. Upper panel, cdk4 and cdk6; lower panel, the various INK4 proteins. B, p18 and the p16/18 chimera do not block pRb phosphorylation in CEM cells. CEM cells transfected with the indicated INK4 constructs were separated from nontransfected cells using the pHOOK system. Cell lysates prepared from the purified transfected cells were probed by Western blotting with an antibody capable of detecting pRb in all phosphorylation states.

Fig. 4.

p18 and p16/18 associate with cdk4 but do not prevent pRb phosphorylation. A, CEM cells transfected with an empty vector control or with the indicated myc-tagged INK4 constructs were pulse-labeled with 35S-methionine and then switched to unlabeled media for 0, 2, 4, or 19 h before lysates were prepared. Lysates from each time point were incubated with the anti-myc antibody 9E10, and the immunoprecipitated material was resolved on a 12% SDS-PAGE gel and visualized by autoradiography. Upper panel, cdk4 and cdk6; lower panel, the various INK4 proteins. B, p18 and the p16/18 chimera do not block pRb phosphorylation in CEM cells. CEM cells transfected with the indicated INK4 constructs were separated from nontransfected cells using the pHOOK system. Cell lysates prepared from the purified transfected cells were probed by Western blotting with an antibody capable of detecting pRb in all phosphorylation states.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported in part by NIH Grant RO1 ES 09673. J. G. is supported by a teaching assistantship from the Department of Molecular Physiology and Biophysics at the University of Vermont. J. K. is a V Foundation Scholar and gratefully acknowledges the support of this organization.

3

The abbreviations used are: CKI, cyclin-dependent kinase inhibitor; INK4, inhibitors of cdk4; p16, p16INK4A; p15, p15INK4B; p18, p18INK4C; p19, p19INK4D; cdk, cyclin-dependent kinase; ALL, acute lymphocytic leukemia; pRb, retinoblastoma protein; GFP, green fluorescent protein; GFP-F, farnesylated GFP.

We thank Dr. Robert Booher of Onyx Pharmaceuticals (Richmond, CA) for graciously providing recombinant cyclin D1/cdk4 and Peter Burch (University of Vermont) for thoughtful comments and discussions. We also thank Dr. Jeff Bond (University of Vermont) for his assistance with molecular modeling.

1
Sherr C. J., Roberts J. M. CDK inhibitors: positive and negative regulators of G1-phase progression.
Genes Dev.
,
13
:
1501
-1512,  
1999
.
2
Hanahan D., Weinberg R. A. The hallmarks of cancer.
Cell
,
100
:
57
-70,  
2000
.
3
Guan K. L., Jenkins C. W., Li Y., Nichols M. A., Wu X., O’Keefe C. L., Matera A. G., Xiong Y. Growth suppression by p18, a p16INK4/MTS1- and p14INK4B/MTS2-related CDK6 inhibitor, correlates with wild-type pRb function.
Genes Dev.
,
8
:
2939
-2952,  
1994
.
4
Hirai H., Roussel M. F., Kato J. Y., Ashmun R. A., Sherr C. J. Novel INK4 proteins, p19 and p18, are specific inhibitors of the cyclin D-dependent kinases CDK4 and CDK6.
Mol. Cell. Biol.
,
15
:
2672
-2681,  
1995
.
5
Koh J., Enders G. H., Dynlacht B. D., Harlow E. Tumour-derived p16 alleles encoding proteins defective in cell-cycle inhibition.
Nature (Lond.)
,
375
:
506
-510,  
1995
.
6
Serrano M., Hannon G. J., Beach D. A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4.
Nature (Lond.)
,
366
:
704
-707,  
1993
.
7
Kamb A., Gruis N. A., Weaver-Feldhaus J., Liu Q., Harshman K., Tavtigian S. V., Stockert E., Day R. S., III, Johnson B. E., Skolnick M. H. A cell cycle regulator potentially involved in genesis of many tumor types.
Science
,
264
:
436
-440,  
1994
.
8
Otsuki T., Jaffe E. S., Wellmann A., Kumar S., Condron K. S., Raffeld M. Absence of p18 mutations or deletions in lymphoid malignancies.
Leukemia
,
10
:
356
-360,  
1996
.
9
Carnero A., Hannon G. J. The INK4 family of CDK inhibitors.
Curr. Top. Microbiol. Immunol.
,
227
:
43
-55,  
1998
.
10
Kamb A. Cyclin-dependent kinase inhibitors and human cancer.
Curr. Top. Microbiol. Immunol.
,
227
:
139
-148,  
1998
.
11
Tsihlias J., Kapusta L., Slingerland J. The prognostic significance of altered cyclin-dependent kinase inhibitors in human cancer.
Annu. Rev. Med.
,
50
:
401
-423,  
1999
.
12
Hirama T., Koeffler H. P. Role of the cyclin-dependent kinase inhibitors in the development of cancer.
Blood
,
86
:
841
-854,  
1995
.
13
Ragione F. D., Iolascon A. Inactivation of cyclin-dependent kinase inhibitor genes and development of human acute leukemias.
Leuk. Lymphoma
,
25
:
23
-35,  
1997
.
14
Parry D., Mahony D., Wills K., Lees E. Cyclin d-CDK subunit arrangement is dependent on the availability of competing INK4 and p21 class inhibitors.
Mol. Cell. Biol.
,
19
:
1775
-1783,  
1999
.
15
Schreiber M., Muller W. J., Singh G., Graham F. L. Comparison of the effectiveness of adenovirus vectors expressing cyclin kinase inhibitors p16INK4A, p18INK4C, p19INK4D, p21(WAF1/CIP1), and p27KIP1 in inducing cell cycle arrest, apoptosis, and inhibition of tumorigenicity.
Oncogene
,
18
:
1663
-1676,  
1999
.
16
Russo A. A., Tong L., Lee J. O., Jeffrey P. D., Pavletich N. P. Structural basis for inhibition of the cyclin-dependent kinase Cdk6 by the tumour suppressor p16INK4a.
Nature (Lond.)
,
395
:
237
-243,  
1998
.
17
Li J., Poi M. J., Qin D., Selby T. L., Byeon I. J., Tsai M. D. Tumor suppressor INK4: quantitative structure-function analyses of p18INK4C as an inhibitor of cyclin-dependent kinase 4.
Biochemistry
,
39
:
649
-657,  
2000
.
18
Jeffrey P. D., Tong L., Pavletich N. P. Structural basis of inhibition of CDK-cyclin complexes by INK4 inhibitors.
Genes Dev.
,
14
:
3115
-3125,  
2000
.
19
Yarbrough W. G., Buckmire R. A., Bessho M., Liu E. T. Biologic and biochemical analyses of p16(INK4a) mutations from primary tumors.
J. Natl. Cancer Inst.
,
91
:
1569
-1574,  
1999
.
20
Byeon I. J., Li J., Ericson K., Selby T. L., Tevelev A., Kim H. J., O’Maille P., Tsai M. D. Tumor suppressor p16INK4A: determination of solution structure and analyses of its interaction with cyclin-dependent kinase 4.
Mol. Cell
,
1
:
421
-431,  
1998
.
21
Li J., Byeon I. J., Ericson K., Poi M. J., O’Maille P., Selby T., sai M. D. Tumor suppressor INK4: determination of the solution structure of p18INK4C and demonstration of the functional significance of loops in p18INK4C and p16INK4A.
Biochemistry
,
38
:
2930
-2940,  
1999
.
22
Buchkovich K., Duffy L. A., Harlow E. The retinoblastoma protein is phosphorylated during specific phases of the cell cycle.
Cell
,
58
:
1097
-1105,  
1989
.
23
Parry D., Bates S., Mann D. J., Peters G. Lack of cyclin D-Cdk complexes in Rb-negative cells correlates with high levels of p16INK4/MTS1 tumour suppressor gene product.
EMBO J.
,
14
:
503
-511,  
1995
.
24
Thullberg M., Bartkova J., Khan S., Hansen K., Ronnstrand L., Lukas J., Strauss M., Bartek J. Distinct versus redundant properties among members of the INK4 family of cyclin-dependent kinase inhibitors.
FEBS Lett.
,
470
:
161
-166,  
2000
.
25
Phelps D. E., Xiong Y. Regulation of cyclin-dependent kinase 4 during adipogenesis involves switching of cyclin D subunits and concurrent binding of p18INK4c and p27Kip1.
Cell Growth Differ.
,
9
:
595
-610,  
1998
.
26
Franklin D. S., Xiong Y. Induction of p18INK4c and its predominant association with CDK4 and CDK6 during myogenic differentiation.
Mol. Biol. Cell
,
7
:
1587
-1599,  
1996
.
27
Alcorta D. A., Xiong Y., Phelps D., Hannon G., Beach D., Barrett J. C. Involvement of the cyclin-dependent kinase inhibitor p16 (INK4a) in replicative senescence of normal human fibroblasts.
Proc. Natl. Acad. Sci. USA
,
93
:
13742
-13747,  
1996
.