Proteolytic processing by cathepsin L generates p110 Cut homeobox 1 (CUX1) at the end of the G1 phase, whereas an alternative transcript encodes p75 CUX1. These short CUX1 isoforms were reported to be overexpressed in cancer cells, and transgenic mice overexpressing the p75 isoform were found to develop myeloproliferative disease–like myeloid leukemias. In the present study, we report that the neutrophil elastase can also generate a short CUX1 isoform in the MV4;11 acute myeloid leukemia cell line. Proteolytic processing was so efficient that the full-length CUX1 protein was detected only when cells were maintained in the presence of the specific elastase inhibitor III. In agreement with these findings, higher levels of the processed cyclin E isoforms were also detected in MV4;11 cells. Reappearance of full-length cyclin E and CUX1 could be induced upon the treatment of MV4;11 cells with the differentiation inducer phorbol 12-myristate 13-acetate or, unexpectedly, following overexpression of a short recombinant CUX1 protein. In both cases, the mechanism involved transcriptional repression of the neutrophil elastase gene. This result revealed a negative feedback loop whereby CUX1 shuts down the expression of the protease that cleaves it. Overall, the findings in MV4;11 and other cancer cells suggest that various mechanisms are used in cancer to favor the expression of short CUX1 isoforms. (Mol Cancer Res 2008;6(4):644–53)

Proteolytic processing has emerged as an important posttranslational mechanism that modulates the subcellular localization and/or biochemical activities of proteins with functions as diverse as receptor, ligand, signaling molecules, and transcription factors. In the case of transcription factors, the consequence of proteolytic processing is often to redirect the localization of proteins from a cytoplasmic site to the nucleus (reviewed in refs. 1, 2). However, in some instances, the peptides generated by processing exhibit a half-life, biochemical properties, and biological functions that are distinct from that of their precursors (reviewed in ref. 3). For example, the promyelocytic leukemia–retinoic acid receptor α (PML-RARα) fusion protein was shown to be processed by neutrophil elastase (4). Myeloid expansion and delayed myeloid maturation could not be induced by PML-RARα expression alone but required, in addition, the activity of neutrophil elastase (5). Moreover, in transgenic mice carrying the PML-RARα fusion gene, inactivation of the neutrophil elastase gene conferred some protection against leukemia: Penetrance was reduced and latency was increased (4). Altogether, these findings suggested that the presence of neutrophil elastase in promyelocytes could explain why the expression of the PML-RARα fusion protein is invariably associated with the promyelocytic leukemia phenotype (4).

Cut homeobox 1 (CUX1) belongs to a family of transcription factors present in all metazoans and involved in the control of proliferation and differentiation (reviewed in ref. 6). The literature includes a variety of terms: Cut in Drosophila melanogaster, CCAAT displacement protein, and Cut-like (CUTL) in human and cux in mouse (7-10). The Human Genome Organization recently proposed to change from the gene root of CUTL# (CUT-Like #) to CUX#. Thus, the term CUX1 will be used thereafter in the text to designate the human or mouse protein. Four CUX1 protein isoforms have been described thus far: p200, p150, p110, and p75. The full-length p200 protein carries the CCAAT displacement activity by binding rapidly but transiently to DNA, whereas p150 does not bind to DNA and may act as a dominant negative (11, 12). In contrast, the short p110 and p75 isoforms behave like classic transcription factors that engage in slow but stable interactions with DNA and can function as repressors or activators depending on promoter context (13-16). Although a tissue-specific transcript that is initiated within intron 20 encodes the p75 isoform, the p110 isoform is generated by a proteolytic processing event that is mediated by cathepsin L at the end of the G1 phase (13, 14, 17). At the cellular level, the p110 isoform was shown to stimulate cell proliferation and to accelerate entry into the S phase (18). These results extended early observations linking CUX1 to the control of gene expression in S phase (19, 20).

Microarray analysis revealed that CUX1 was one of the most up-regulated genes in malignant plasma cells from multiple myelomas compared with nonmalignant plasma cells (21). Moreover, in a survey of human breast cancers, CUX1 expression was found to be significantly increased in high-grade carcinomas and was inversely correlated with survival (22). In uterine leiomyomas, p110 CUX1 was frequently expressed at a higher level in the tumor than in the adjacent myometrium (23). In many cancer cell lines, increased expression and activity of nuclear cathepsin L was found to result in a higher ratio of processed over the full-length CUX1 protein (24). Interestingly, the intron 20–mRNA and p75 isoforms were found to be expressed only weakly or not at all in primary human mammary epithelial cells and normal breast tissues, but were detected in many breast tumor cell lines and human breast tumors (13). In invasive breast tumors, a significant association was established between higher intron 20–mRNA expression and a diffuse infiltrative growth pattern (13).

Collectively, expression studies in primary tumors and cancer cell lines suggested that increased expression of CUX1, notably its short isoforms, could contribute to tumor development and/or progression. Indeed, transgenic mice engineered to express p75 CUX1 displayed enhanced susceptibility to various cancers. In particular, the p75 CUX1 mice frequently succumbed to a disease defined as a myeloproliferative disease–like myeloid leukemia and characterized by splenomegaly, hepatomegaly, and frequent infiltration of leukocytes into nonhematopoietic organs like the kidneys and lungs (25). The cell type of origin of these tumors came as a surprise because the transgene was placed under the control of the mouse mammary tumor virus long terminal repeat and was inserted into the mouse hypoxanthine phosphoribosyltransferase (Hprt) locus using the method of targeted transgenesis (26). The unexpected phenotype of transgenic mice suggested a particular tropism of the p75 CUX1 isoform toward myeloid cells. This led us to investigate CUX1 expression and activity in a panel of acute myeloid leukemia cell lines. We identified a cell line in which the full-length CUX1 protein seemed not to be expressed. Our analysis of CUX1 expression in the MV4;11 cell line revealed a novel mechanism by which short CUX1 isoforms can be generated.

The Full-length CUX1 Protein Is Not Detected in the MV4;11 Acute Myeloid Leukemia Cell Line

We analyzed CUX1 expression in a panel of myeloid leukemia cell lines and found one cell line, MV4;11, which did not express the full-length p200 isoform but expressed only shorter isoforms. The full-length protein was easily detected in RS4;11 and U937 cells, but not in MV4;11 cells (Fig. 1A, lanes 1-3). Both the MV4;11 and RS4;11 cell lines derive from a myeloid leukemia that bears the 4;11 chromosomal translocation; we further compared CUX1 expression in these two cell lines by performing immunoprecipitation and immunoblotting with various CUX1 antibodies (27). Again, p200 was detected in RS4;11 but not in MV4;11 cells (Fig. 1B, lanes 1-3 and 8-10). Results from electrophoretic mobility shift assays were consistent with those from immunoblotting. In RS4;11 samples, we observed the slow migrating p200-complex that was supershifted with CUX1 antibodies (Fig. 1C, compare lanes 4, 8, and 12 with lanes 2, 6, 10, and 14). This complex was not detected in the MV4;11 samples; however, faster migrating complexes were observed (Fig. 1C, every odd number lane). These faster migrating complexes involved CUX1 protein species because incubation in the presence of various CUX1 antibodies generated distinct, albeit weak, supershifts (Fig. 1C, lanes 1, 5, 9, and 13). A pairwise comparison of the supershifts generated by each antibody revealed that some supershifts were unique to MV4;11, whereas others were unique to RS4;11 (Fig. 1C, compare lane 1 with 2, 5 with 6, 9 with 10, and 13 with 14). These findings suggest that MV4;11 and RS4;11 cells express distinct short CUX1 isoforms.

FIGURE 1.

The full-length CUX1 protein is not detected in the MV4;11 cell line. A. Nuclear extracts were prepared from RS4;11, U937, and MV4;11 cells and submitted to Western blot analysis using 861 CUX1 and actin antibodies. Arrows, CUX1 p200 and shorter isoforms. B. Whole-cell extracts (2 mg) from RS4;11 and MV4;11 cells were submitted to immunoprecipitation using the indicated antibodies and to immunoblotting using 861 and 1300 CUX1 and actin antibodies. In the “WCE” lanes, 30 μg of proteins were loaded directly onto the gel. C. Nuclear extracts from MV4;11 cells and from RS4;11 cells were analyzed by electrophoretic mobility shift assay using oligonucleotides containing a CUX1 consensus binding site. Where indicated, antibodies were added. Arrows, supershifted retarded complexes in MV4;11 cells. D. Schematic representation of the full-length CUX1 protein, p200, and the proteolytically processed isoform, p110. The regions recognized by CUX1 antibodies are indicated. Arrow, cleavage sites.

FIGURE 1.

The full-length CUX1 protein is not detected in the MV4;11 cell line. A. Nuclear extracts were prepared from RS4;11, U937, and MV4;11 cells and submitted to Western blot analysis using 861 CUX1 and actin antibodies. Arrows, CUX1 p200 and shorter isoforms. B. Whole-cell extracts (2 mg) from RS4;11 and MV4;11 cells were submitted to immunoprecipitation using the indicated antibodies and to immunoblotting using 861 and 1300 CUX1 and actin antibodies. In the “WCE” lanes, 30 μg of proteins were loaded directly onto the gel. C. Nuclear extracts from MV4;11 cells and from RS4;11 cells were analyzed by electrophoretic mobility shift assay using oligonucleotides containing a CUX1 consensus binding site. Where indicated, antibodies were added. Arrows, supershifted retarded complexes in MV4;11 cells. D. Schematic representation of the full-length CUX1 protein, p200, and the proteolytically processed isoform, p110. The regions recognized by CUX1 antibodies are indicated. Arrow, cleavage sites.

Close modal

MV4;11 Cells Express the Full-length p200 following Phorbol 12-Myristate 13-Acetate Treatment

Analysis of CUX1 mRNA expression did not reveal any difference between the MV4;11 and RS4;11 cells (data not shown). Moreover, a DNA rearrangement involving the CUTL1 locus at 7q22 was not detected in MV4;11 cells (ref. 27 and data not shown). Yet, we could not exclude that more subtle changes in DNA might have occurred in these cells. Convincing evidence that the gene was intact came from the observation that the full-length CUX1 protein was expressed in certain culture conditions. Upon treatment with phorbol 12-myristate 13-acetate (PMA), a commonly used differentiation inducer, MV4;11 cells stopped proliferating and started to adhere to the plate (Fig. 2A). Interestingly, with increasing amounts of PMA, we began to detect the full-length p200 CUX1 protein (Fig. 2B, compare lanes 1 to 4). This band was specific to CUX1 because it was competed out with the glutathione S-transferase/CUX1-861 immune peptide but not with the glutathione S-transferase peptide alone (Fig. 2B, compare lanes 1-4 with lanes 5-8 and 9-12). Of note, we did not detect changes in CUX1 mRNA expression upon treatment of MV4;11 cells with PMA (data not shown; see also Fig. 6A, lanes 2 and 3). These results suggest that DNA rearrangements or point mutations cannot account for the expression of short CUX1 isoforms in MV4;11 cells.

FIGURE 2.

MV4;11 cells express the full-length CUX1 protein following PMA treatment. A. MV4;11 cells (5 × 105/mL) were treated with ethanol (vehicle) or 1 μmol/L PMA for up to 3 d. A picture of the cell was taken after 48 h (left). Proliferation curves were obtained by counting cells in triplicate every day (right). B. MV4;11 cells were treated with 10 nmol/L, 100 nmol/L, or 1 μmol/L PMA for 48 h. Nuclear extracts were prepared and analyzed (40 μg) by immunoblotting with 861 CUX1 antibodies in the presence of no competitor or excess glutathione S-transferase protein or glutathione S-transferase/CUX1 protein containing amino acids 861 to 936.

FIGURE 2.

MV4;11 cells express the full-length CUX1 protein following PMA treatment. A. MV4;11 cells (5 × 105/mL) were treated with ethanol (vehicle) or 1 μmol/L PMA for up to 3 d. A picture of the cell was taken after 48 h (left). Proliferation curves were obtained by counting cells in triplicate every day (right). B. MV4;11 cells were treated with 10 nmol/L, 100 nmol/L, or 1 μmol/L PMA for 48 h. Nuclear extracts were prepared and analyzed (40 μg) by immunoblotting with 861 CUX1 antibodies in the presence of no competitor or excess glutathione S-transferase protein or glutathione S-transferase/CUX1 protein containing amino acids 861 to 936.

Close modal

A Recombinant Full-length CUX1 Protein Is Efficiently Processed in MV4;11 Cells

One approach to verify whether proteolytic processing generates a short isoform is to express a recombinant full-length protein using the entire coding sequence from the corresponding cDNA. We established stable populations of RS4;11 and MV4;11 cells carrying a retroviral vector that codes for a full-length CUX1 protein, herein called 1-1505. In MV4;11 cells, the full-length protein could now be detected, albeit at a much weaker level than in RS4;11 cells (Fig. 3A, compare lanes 1 and 3). Also, we observed an increase in a short CUX1 isoform that comigrated with an endogenous protein present in the parental MV4;11 cells (Fig. 3A, compare lanes 2 and 3; Fig. 3B, compare lanes 1 and 3). We interpret this band to represent a product of proteolytic processing. To confirm that MV4;11 cells cannot sustain elevated expression of the full-length CUX1 protein, MV4;11, 293 and HeLa cells were transiently infected with a retroviral vector expressing p200 CUX1 with a COOH-terminal hemagglutinin (HA) tag. In immunoblotting with an anti-HA antibody, the full-length CUX1 protein was easily detected in 293 and HeLa cells but not in MV4;11 cells (Fig. 3C, lanes 1-6). Yet, following treatment of MV4;11 cells with PMA, the protein was visible with the HA antibody (Fig. 3C, lane 8). Using the CUX1-861 antibody that recognizes an internal epitope revealed that addition of PMA to the stable population increased the steady-state level of the full-length CUX1 protein and reduced the amount of the processed isoform in both the parental and transduced MV4;11 cells (Fig. 3B, compare lanes 2 with 1 and 4 with 3). These findings confirmed that proteolytic processing generates a short CUX1 isoform in MV4;11 cells and that PMA treatment causes a decrease in processing.

FIGURE 3.

A recombinant full-length CUX1 protein is efficiently processed in MV4;11 cells. A. MV4;11 cells were stably infected with retroviral vectors expressing recombinant CUX1 proteins containing amino acids 1-1505 or 878-1505. Nuclear extracts were analyzed by immunoblotting using the 861 CUX1 and actin antibodies. Arrows, the recombinant proteins and the processed isoforms generated in MV4;11 cells. B. The MV4;11 cell lines stably expressing recombinant CUX1 proteins were treated or not with PMA for 48 h, and nuclear extracts were analyzed by immunoblotting as in A. C. MV4;11, 293 and HeLa cells were infected with a retroviral vector expressing the full-length CUX1 protein with a HA epitope tag at its COOH terminus. Lanes 6 and 7, MV4;11 cells were treated with 1 μmol/L PMA for 48 h. Nuclear extracts were analyzed by immunoblotting using antibodies against HA and actin.

FIGURE 3.

A recombinant full-length CUX1 protein is efficiently processed in MV4;11 cells. A. MV4;11 cells were stably infected with retroviral vectors expressing recombinant CUX1 proteins containing amino acids 1-1505 or 878-1505. Nuclear extracts were analyzed by immunoblotting using the 861 CUX1 and actin antibodies. Arrows, the recombinant proteins and the processed isoforms generated in MV4;11 cells. B. The MV4;11 cell lines stably expressing recombinant CUX1 proteins were treated or not with PMA for 48 h, and nuclear extracts were analyzed by immunoblotting as in A. C. MV4;11, 293 and HeLa cells were infected with a retroviral vector expressing the full-length CUX1 protein with a HA epitope tag at its COOH terminus. Lanes 6 and 7, MV4;11 cells were treated with 1 μmol/L PMA for 48 h. Nuclear extracts were analyzed by immunoblotting using antibodies against HA and actin.

Close modal

Proteolytic Processing of CUX1 Using Nuclear Extracts from MV4;11 Is Inhibited by Serine Protease Inhibitors

We did in vitro processing assays using a bacterially expressed CUX1 protein as a substrate and nuclear extracts from RS4;11 and MV4;11 cells as a source of proteases. This assay was previously shown to reproduce in vitro the proteolytic processing of CUX1 (17, 28). Two novel cleavage products were generated when the CUX1 protein was incubated with nuclear extracts from MV4;11 cells (Fig. 4A, lane 3). This findings suggested that CUX1 might be proteolytically processed by a novel protease in MV4;11 cells. The assay was therefore repeated in the presence of various protease inhibitors. Interestingly, processing was not inhibited in the presence of E64, a cysteine protease inhibitor previously shown to inhibit CUX1 processing by cathepsin L in other cell lines (Fig. 4B, compare lane 3 with 2; refs. 17, 28). However, two serine protease inhibitors, leupeptin and phenylmethylsulfonyl fluoride, were able to inhibit processing by MV4;11 nuclear extracts (Fig. 4B, compare lane 2 with 5 and lanes 8 and 9 with 7). In contrast, as expected, phenylmethylsulfonyl fluoride did not inhibit processing when added to a reaction containing nuclear extracts from the Hs578T breast tumor cell line (data not shown). These results suggest that nuclear extracts from MV4;11 cells contain a serine protease that can cleave CUX1.

FIGURE 4.

CUX1 is proteolytically processed by a serine protease in MV4;11 cells. A.In vitro processing assay. A recombinant CUX1 protein containing amino acids 612 to 1328 was expressed in bacteria, purified by affinity chromatography over nickel beads, and incubated in the presence of nuclear extracts (0.5 μg) isolated from RS4;11 or MV4;11 cells. The substrate and products of the reaction were analyzed by immunoblotting using anti-HA antibodies. *, contaminating protein present in the purified sample that is recognized by the HA antibody; note that this protein is not detected when the Myc antibody was used (see Fig. 3B). B. An in vitro processing assay was done as in A in the presence of the indicated protease inhibitors and analyzed by immunoblotting using anti-Myc antibodies. C. An in vitro processing assay was done as in A using nuclear extracts from MV4;11 cells that had been treated or not with 1 μmol/L PMA for 48 h. D. Total RNA was isolated from RS4;11, MV4;11 cells and PMA-treated MV4;11 cells. RT-PCR analysis was done using primers specific for various hematopoietic serine proteases, cathepsin L, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

FIGURE 4.

CUX1 is proteolytically processed by a serine protease in MV4;11 cells. A.In vitro processing assay. A recombinant CUX1 protein containing amino acids 612 to 1328 was expressed in bacteria, purified by affinity chromatography over nickel beads, and incubated in the presence of nuclear extracts (0.5 μg) isolated from RS4;11 or MV4;11 cells. The substrate and products of the reaction were analyzed by immunoblotting using anti-HA antibodies. *, contaminating protein present in the purified sample that is recognized by the HA antibody; note that this protein is not detected when the Myc antibody was used (see Fig. 3B). B. An in vitro processing assay was done as in A in the presence of the indicated protease inhibitors and analyzed by immunoblotting using anti-Myc antibodies. C. An in vitro processing assay was done as in A using nuclear extracts from MV4;11 cells that had been treated or not with 1 μmol/L PMA for 48 h. D. Total RNA was isolated from RS4;11, MV4;11 cells and PMA-treated MV4;11 cells. RT-PCR analysis was done using primers specific for various hematopoietic serine proteases, cathepsin L, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

Close modal

Expression of Neutrophil Elastase Is Elevated in MV4;11 Cells and Is Reduced upon PMA treatment

As p200 CUX1 was detected in PMA-treated MV4;11 cells, we reasoned that proteolytic processing of CUX1 might be reduced in these conditions (Fig. 2B). To confirm this hypothesis, the in vitro processing assay was repeated using as a source of proteases nuclear extracts from MV4;11 cells that were treated or not with PMA. Indeed, processing in vitro was reduced by the treatment of cells with PMA (Fig. 4C, lanes 2 and 3). As the substrate in this reaction was purified from bacteria, this result suggested that the effect of PMA was not mediated through modification of the substrate. Instead, PMA must affect the expression or activity of the protease in MV4;11 cells. We therefore investigated the effect of PMA on the expression of candidate proteases. RNA was isolated from RS4;11 and MV4;11 cells, treated or not with PMA, and protease mRNA expression was measured using a semiquantitative reverse transcription-PCR (RT-PCR) assay. In this assay, we were looking for a serine protease whose expression was higher in MV4;11 than in RS4;11 cells, and was reduced upon treatment with PMA. Expression of the neutrophil elastase mRNA fulfilled these criteria: The expression was much higher in MV4;11 than in RS4;11 cells; moreover, the expression was reduced in the presence of PMA (Fig. 4D, top). These results are in agreement with the previously reported transcriptional down-regulation of the neutrophil elastase gene upon treatment with PMA (29). Expression of cathepsin G mRNA, on the other hand, was not significantly different in the three samples (Fig. 4D, second panel). Interestingly, proteinase 3 mRNA expression was higher in MV4;11 than in RS4;11 cells, but its expression was not affected by PMA (Fig. 4D, third panel). Expression of cathepsin L mRNA was very weak in both cell lines but was induced in the presence of PMA (Fig. 4D, fourth panel). Indeed, PMA was previously shown to induce cathepsin L expression in various cell lines (30, 31). In summary, results from expression analysis revealed that neutrophil elastase was the only serine protease whose expression was both elevated in MV4;11 cells and reduced upon treatment of cells with PMA.

CUX1 Is Proteolytically Processed by the Human Neutrophil Elastase Serine Protease in MV4;11 Cells

The in vitro processing assay was repeated this time using a specific inhibitor for neutrophil elastase, elastase inhibitor III. The amount of the two cleavage products was reduced in the presence of elastase inhibitor III (Fig. 5A, compare lanes 3 and 4 with lane 2). This result suggests that neutrophil elastase is the active protease in nuclear extracts from MV4;11 cells. Indeed, in vitro processing using a purified preparation of elastase generated two cleavage products of identical mobility (Fig. 5A, compare lane 7 with lane 10). Moreover, the addition of elastase inhibitor III to the reaction inhibited the production of cleavage products whether the reaction was done with purified neutrophil elastase or nuclear extracts from MV4;11 (Fig. 5A, compare lane 7 with 8, and lane 11 with 10). As elastase inhibitor III is a cell-permeable compound, we tested its effect when applied directly onto cells in tissue culture. Interestingly, the effect of elastase inhibitor III mimicked that of PMA. After 48 hours, the full-length CUX1 protein was detected (Fig. 5B, compare lane 5 with 2). Altogether, these results show that neutrophil elastase is overexpressed and cleaves CUX1 in MV4;11 cells.

FIGURE 5.

Human neutrophil elastase is the serine protease responsible for the proteolytic processing of CUX1 in MV4;11 cells. A. An in vitro processing assay was done as in Fig. 3A using as a source of protease either nuclear extracts from MV4;11 or purified porcine pancreatic elastase. Where indicated, the assay was done in the presence of elastase inhibitor III (EI III) or DMSO as a carrier control. B. MV4;11 cells were cultured in the presence of either 1 μmol/L PMA, 100 μmol/L elastase inhibitor III, or DMSO for the indicated periods of time. Nuclear extracts were isolated and submitted to Western blot analysis using 861 CUX1 antibodies. C. MV4;11 cells were treated or not with PMA or elastase inhibitor III for 48 h. Nuclear extracts were isolated from these cells and Hs578T, RS4;11, and U937 cells, and were submitted to Western blot analysis using a cyclin E antibody. D. Total RNA was isolated from RS4;11, U937, and MV4;11 cells that were treated or not with PMA for 48 h. RT-PCR analysis was done using primers specific for neutrophil elastase, proteinase 3, and glyceraldehyde-3-phosphate dehydrogenase.

FIGURE 5.

Human neutrophil elastase is the serine protease responsible for the proteolytic processing of CUX1 in MV4;11 cells. A. An in vitro processing assay was done as in Fig. 3A using as a source of protease either nuclear extracts from MV4;11 or purified porcine pancreatic elastase. Where indicated, the assay was done in the presence of elastase inhibitor III (EI III) or DMSO as a carrier control. B. MV4;11 cells were cultured in the presence of either 1 μmol/L PMA, 100 μmol/L elastase inhibitor III, or DMSO for the indicated periods of time. Nuclear extracts were isolated and submitted to Western blot analysis using 861 CUX1 antibodies. C. MV4;11 cells were treated or not with PMA or elastase inhibitor III for 48 h. Nuclear extracts were isolated from these cells and Hs578T, RS4;11, and U937 cells, and were submitted to Western blot analysis using a cyclin E antibody. D. Total RNA was isolated from RS4;11, U937, and MV4;11 cells that were treated or not with PMA for 48 h. RT-PCR analysis was done using primers specific for neutrophil elastase, proteinase 3, and glyceraldehyde-3-phosphate dehydrogenase.

Close modal

Cyclin E Is Also Proteolytically Processed by the Neutrophil Elastase in MV4;11 Cells

Cyclin E has previously been identified as a target of elastase in breast cancer cells, and truncation of cyclin E was found to cause an increase in cyclin E/Cdk2 kinase activity (32, 33). We therefore predicted that cyclin E should be processed in MV4;11 cells. Indeed, the ratio of the full-length over the processed cyclin E isoforms was much lower in MV4;11 than in RS4;11, Hs578T, and U937 cells (Fig. 5C, compare lane 3 with lanes 1, 2, and 6). Moreover, treatment of MV4;11 cells with either elastase inhibitor III or PMA reduced the amount of processed cyclin E and increased the amount of full-length cyclin E (Fig. 5C, compare lanes 4 and 5 with lane 3). These results prompted us to compare the expression levels of neutrophil elastase in MV4;11 and U937 cells. Neutrophil elastase mRNA expression was much higher in MV4;11 than in U937 cells (Fig. 5D, lanes 1 and 3). This finding is consistent with the processing pattern of cyclin E and CUX1 in these two cell lines (cyclin E, Fig. 5C, lanes 3 and 6; CUX1, Fig. 1A, lanes 2 and 3). We noted again that PMA caused a reduction in the expression of the neutrophil elastase but not of proteinase 3 mRNA (Fig. 5D, lanes 3 and 4). These results suggested that overexpression of neutrophil elastase is responsible for the enhanced proteolytic processing of both CUX1 and cyclin E in MV4;11 cells.

The Recombinant CUX1 878-1505 Protein Stimulates Expression of the Endogenous Full-length CUX1 Protein by Repressing Expression of the Neutrophil Elastase Gene

Infection of MV4;11 cells with a vector expressing a short recombinant CUX1 protein, called 878-1505, revealed an interesting phenomenon: The full-length p200 protein could now be detected in these cells (Fig. 3A, compare lanes 2 and 4; Fig. 3B, compare lanes 1 and 5). Further tests were done to confirm that these cells only carried the CUX1 878-1505 vector and were not mixed with other cells (data not shown). We investigated a number of mechanisms that could account for the induction of full-length CUX1 by a short recombinant CUX1 protein. RT-PCR analysis indicated that expression of endogenous CUX1 mRNA was not significantly increased in the MV4;11 cells that either carried the 878-1505 vector or were treated with PMA (Fig. 6A, lanes 2-5). However, we found that expression of the neutrophil elastase mRNA was reduced in MV4;11 cells that expressed the recombinant 878-1505 protein (Fig. 6A, compare lane 9 with 11). In agreement with this finding, processing of cyclin E was also reduced in MV4;11 cells expressing recombinant CUX1 proteins (Fig. 6B, compare lane 2 with 3 and 4). Altogether, these results reveal a negative feedback loop whereby CUX1 represses transcription of the neutrophil elastase gene, whereas neutrophil elastase cleaves the CUX1 protein.

FIGURE 6.

CUX1 represses expression of the neutrophil elastase gene. A. Total RNA was isolated from RS4;11, MV4;11, and MV4;11/878-1505 cells, treated or not with PMA. RT-PCR analysis was done using primers specific for CUX1, neutrophil elastase, proteinase 3, cathepsin G, and glyceraldehyde-3-phosphate dehydrogenase. B. Nuclear extracts from RS4;11 cells and MV4;11 cells stably expressing CUX1 1-1505 or 878-1505 were analyzed by immunoblotting using anti–cyclin E antibodies.

FIGURE 6.

CUX1 represses expression of the neutrophil elastase gene. A. Total RNA was isolated from RS4;11, MV4;11, and MV4;11/878-1505 cells, treated or not with PMA. RT-PCR analysis was done using primers specific for CUX1, neutrophil elastase, proteinase 3, cathepsin G, and glyceraldehyde-3-phosphate dehydrogenase. B. Nuclear extracts from RS4;11 cells and MV4;11 cells stably expressing CUX1 1-1505 or 878-1505 were analyzed by immunoblotting using anti–cyclin E antibodies.

Close modal

Inhibition of Neutrophil Elastase in MV4;11 Cells Slows Down Cell Proliferation

To verify whether the activity of neutrophil elastase may confer a growth advantage to the MV4;11 cells, we measured cell proliferation in the presence of the elastase inhibitor III. Using two independent assays, cell counting and carboxyfluorescein diacetate succinimidyl ester staining, we observed a decrease in cell proliferation when MV4;11 cells were treated with the elastase inhibitor III (Fig. 7). These results are consistent with the notion that increased activity of neutrophil elastase in MV4;11 cells enhances their proliferation potential.

FIGURE 7.

Inhibition of neutrophil elastase in MV4;11 cells slows down cell proliferation. MV4;11 cells were treated with the elastase inhibitor III or the vehicle (DMSO) and their proliferation was measured with two assays. A. Proliferation curves. Cells were seeded in triplicate in complete medium containing elastase inhibitor III or the vehicle (DMSO) and counted every day for 5 d. Points, average; bars, SD. Growth medium with elastase inhibitor III or the vehicle was replaced every 48 h. B. Cells were stained with CellTrace carboxyfluorescein diacetate succinimidyl ester (CFSE) on day 0. A portion of the population was fixed immediately as the “0” generation (white peak). The rest of the cells was allowed to proliferate for 5 d in complete medium in the presence of elastase inhibitor III or the vehicle (DMSO). Cells were fixed and analyzed by flow cytometry. Solid peaks within the CFSE profiles represent successive generations, as indicated above the peaks.

FIGURE 7.

Inhibition of neutrophil elastase in MV4;11 cells slows down cell proliferation. MV4;11 cells were treated with the elastase inhibitor III or the vehicle (DMSO) and their proliferation was measured with two assays. A. Proliferation curves. Cells were seeded in triplicate in complete medium containing elastase inhibitor III or the vehicle (DMSO) and counted every day for 5 d. Points, average; bars, SD. Growth medium with elastase inhibitor III or the vehicle was replaced every 48 h. B. Cells were stained with CellTrace carboxyfluorescein diacetate succinimidyl ester (CFSE) on day 0. A portion of the population was fixed immediately as the “0” generation (white peak). The rest of the cells was allowed to proliferate for 5 d in complete medium in the presence of elastase inhibitor III or the vehicle (DMSO). Cells were fixed and analyzed by flow cytometry. Solid peaks within the CFSE profiles represent successive generations, as indicated above the peaks.

Close modal

The MV4;11 myeloid leukemia cell line represents the first case of cells in which the full-length CUX1 protein is not detected. Preliminary experiments as well as karyotyping analysis did not provide any evidence of genomic DNA rearrangement or alternative mRNA expression in MV4;11 cells (27). Ultimately, the best evidence against the notion of DNA rearrangement or point mutations was the finding that MV4;11 cells could be induced to reexpress the full-length CUX1 protein when cultured in the presence of either PMA (Figs. 2, 3B, and 5B) or the elastase inhibitor III (Fig. 5B), or when infected with a vector expressing the short CUX1 878-1505 protein (Fig. 3A and B). Of note, we did not detect any difference in the expression of CUX1 mRNA between the MV4;11 and RS4;11 cells nor after any of the manipulations cited above (Fig. 6A and data not shown). Evidence demonstrating that CUX1 is proteolytically processed by neutrophil elastase in MV4;11 cells was provided from various experimental approaches. In an in vitro processing assay, we showed that a recombinant CUX1 protein was processed differently when incubated in the presence of nuclear extracts from RS4;11 and MV4;11 cells (Fig. 4A). Processing with the MV4;11 extracts was inhibited in the presence of serine protease inhibitors like leupeptin and phenylmethylsulfonyl fluoride, but not with the E64 cysteine protease inhibitor (Fig. 4B). CUX1 peptides of similar electrophoretic mobility were generated by a purified preparation of neutrophil elastase and the MV4;11 nuclear extracts (Fig. 5A, compare lanes 6, 7, and 10). With both sources of protease, processing was inhibited in the presence of elastase inhibitor III (Fig. 5A, compare lane 7 with 8, and lane 10 with 11). In tissue culture, the endogenous full-length CUX1 protein was detected following the incubation of MV4;11 cells with either the elastase inhibitor III (Fig. 5B) or PMA (Figs. 2B, 3B, and 5B). Importantly, reexpression of full-length CUX1 protein correlated with a decrease in the expression of neutrophil elastase mRNA either when cells were treated with PMA (Fig. 5D) or when cells were infected with vectors expressing CUX1 proteins (Fig. 6A, lanes 9-11). In agreement with these findings, cyclin E was processed by neutrophil elastase in MV4;11 cells in a manner similar to what was previously reported in other cancer cells (Figs. 5C and 6B; ref. 33). Proteinase 3 was also expressed at a higher level in MV4;11 than in RS4;11 cells, but the fact that its expression was not affected by PMA or CUX1 878-1505 argues against this protease being the main serine protease that cleaves CUX1 (Figs. 5D and 6A). Collectively, our data indicate that the predominant activity responsible for the proteolytic processing of CUX1 and cyclin E in MV4;11 cells is that of neutrophil elastase.

The proliferation of MV4;11 cells was reduced in the presence of elastase inhibitor III (Fig. 7). These results indicate that increased neutrophil elastase activity indeed confers a growth advantage to cells. Although it is likely that other substrates of neutrophil elastase also contribute to increase the proliferation potential of MV4;11 cells, several lines of evidence suggest that the efficient cleavage of CUX1 may have some functional significance. Proteolytic processing was shown previously to relieve the protein from the inhibitory effect of an NH2-terminal autoinhibitory domain that can prevent DNA binding by all Cut repeats and the Cut homeodomain (16). Another consequence of proteolytic processing is to remove the Cut repeat 1, thereby changing the DNA binding and transcriptional properties of the protein. Although a protein that contains three Cut repeats and the Cut homeodomain can only engage in transient interactions with DNA and act as a repressor, CUX1 isoforms that do not include Cut repeat 1 can bind stably to DNA and function either as repressors or activators depending on promoter context (12-15, 17). Importantly, molecular changes imparted by proteolytic processing clearly have important consequences on the behavior of cells. Constitutive expression of a short recombinant CUX1 protein was shown to accelerate S-phase entry and stimulate cell proliferation (18). In contrast, the full-length protein did not have any effect in the same assays.5

5

L. Sansregret and B. Goulet, unpublished observations.

Whereas proteolytic processing of CUX1 was shown to be tightly regulated during cell cycle progression in normal cells, increased processing of CUX1 was reported both in primary human uterine leiomyomas and several cancer cell lines (23, 24). In transgenic mice, forced expression of short CUX1 proteins enhanced the susceptibility to various malignancies, in particular in the myeloid cell lineage (25). In contrast, a predisposition to cancer was not reported for the transgenic mice that express the full-length protein (34). Altogether, these results suggest that any event that leads to an increase in the level of short CUX1 isoforms is likely to provide cells with a proliferation advantage.

The findings reported in the present study add to the increasing evidence linking the aberrant activity of certain proteases with the altered localization or change in biochemical properties of proteins in cancer cells (33, 35-40). Cancer is often described as a genetic disease. However, nongenetic mechanisms also play a role in cancer. Like DNA mutations, proteolytic processing can lead to proto-oncogene activation and tumor suppressor inactivation depending on the domain that is cleaved off. Nongenetic changes in turn can lead to genetic changes that can be stably transmitted to daughter cells. Although aberrant proteolytic processing is much more difficult to assess than genetic changes because proteins are less stable molecules that DNA, it will be crucial in the future to establish a repertoire of cancers in which such mechanism plays a role. The importance of this is not only conceptual but also practical, because serine and cysteine proteases are a type of enzyme for which it is feasible in practice to develop efficient and specific inhibitors that can be used as therapeutic agents.

Cell Culture

MV4;11 cells were grown in Iscove's modified Dulbecco's medium supplemented with 10% fetal bovine serum. RS4;11 and U937 were grown in RPMI 1640 supplemented with 10% fetal bovine serum. Cells were infected by the addition of virus containing supernatant from 293VSV producer cells in the presence of 8 μg/mL polybrene (Sigma). For PMA (Sigma) and elastase inhibitor III (Calbiochem) treatments, MV4-11 cells were plated at 5 × 105/mL. Growth curves were done by plating in triplicate 2 × 105 cells/35 cm, and viable cells were counted every day. Live cell images were taken with a Retiga 1300 digital camera (QIMAGING) and a Zeiss AxioVert 135 microscope (Carl Zeiss Canada).

Carboxyfluorescein Diacetate Succinimidyl Ester Staining

Cells were stained using the CellTrace CFSE Cell Proliferation Kit and were analyzed by flow cytometry with 488-nm excitation and emission filters appropriate for fluorescein, according to the manufacturer's instruction (Molecular Probes). Carboxyfluorescein diacetate succinimidyl ester profiles were analyzed using the FlowJo software (Tree Star Software).

CUX1 Antibodies

To generate polyclonal antibodies α1032 and α1061, rabbits were injected with either 500 μg of purified bacterial fusion protein containing the regions of CUX1 (amino acids 1032-1106) or of the peptide MSSSESVKC-COOH coupled to keyhole limpet hemocyanin (amino acids 1061-1069), in Freund's complete adjuvant. Polyclonal antibodies were purified by affinity chromatography as previously described (14). The other antibodies used have been described (14).

Immunoprecipitations

Total protein extracts were prepared in modified radioimmunoprecipitation assay buffer [20 mmol/L Tris (pH 8.0), 150 mmol/L NaCl, 1% NP40, 0.5% deoxycholate, 2 mmol/L EDTA, 2 mmol/L EGTA]. Lysates were precleared for 1 h with 40 μL protein A–Sepharose. Immunoprecipitation was done with the indicated antibodies overnight at 4°C. Protein A–Sepharose was then added for 1 h at 4°C. Immunoprecipitates were washed twice with wash buffer 1 [20 mmol/L Tris (pH 8.0), 150 mmol/L NaCl, 1% NP40, 0.5% deoxycholate, 2 mmol/L EDTA, 2 mmol/L EGTA, 0.2% SDS] and twice with wash buffer 2 [20 mmol/L Tris (pH 8.0), 500 mmol/L NaCl, 1% NP40, 0.5% deoxycholate, 2 mmol/L EDTA, 2 mmol/L EGTA, 0.2% SDS] and resuspended in 2× SDS sample buffer.

Electrophoretic Mobility Shift Assay

Electrophoretic mobility shift assays were done as previously described with 6 or 3 μg of nuclear extracts (14).

Reverse Transcription-PCR

RT-PCR was done as previously described (13).

Purification of His-612-1328-HA-Myc

pTriex2 vector encoding His-612-1328-HA-Myc proteins were introduced into the BL21(DE3) strain of Escherichia coli and induced with 1 mmol/L isopropyl-l-thio-β-d-galactopyranoside for 3 h. Proteins were purified on Ni-NTA agarose (Qiagen) according to the manufacturer's instructions.

In vitro Processing Assay

Fifty nanograms of purified His-612-1328-HA-Myc were incubated with 0.5 μg of nuclear extracts from MV4;11 or RS4;11 in buffer C in the presence or absence of various protease inhibitors for 8 min at 37°C. SDS (3×) loading buffer was added to stop the reaction.

Grant support: A scholarship from the Fonds de la Recherche en Santé du Québec and grant 014288 from the National Cancer Institute of Canada (A. Nepveu).

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.

Note: B. Goulet and Y. Markovic contributed equally to this work.

1
Brown MS, Ye J, Rawson RB, Goldstein JL. Regulated intramembrane proteolysis: a control mechanism conserved from bacteria to humans.
Cell
2000
;
100
:
391
–8.
2
Hoppe T, Rape M, Jentsch S. Membrane-bound transcription factors: regulated release by RIP or RUP.
Curr Opin Cell Biol
2001
;
13
:
344
–8.
3
Goulet B, Nepveu A. Complete and limited proteolysis in cell cycle progression.
Cell Cycle
2004
;
3
:
986
–9.
4
Lane AA, Ley TJ. Neutrophil elastase cleaves PML-RARα and is important for the development of acute promyelocytic leukemia in mice.
Cell
2003
;
115
:
305
–18.
5
Lane AA, Ley TJ. Neutrophil elastase is important for PML-retinoic acid receptor α activities in early myeloid cells.
Mol Cell Biol
2005
;
25
:
23
–33.
6
Nepveu A. Role of the multifunctional CDP/Cut/Cux homeodomain transcription factor in regulating differentiation, cell growth and development.
Gene
2001
;
270
:
1
–15.
7
Scherer SW, Rommens JM, Soder S, et al. Refined localization and yeast artificial chromosome (YAC) contig-mapping of genes and DNA segments in the 7q21-32 region.
Hum Mol Genet
1993
;
2
:
751
–60.
8
Blochlinger K, Bodmer R, Jack J, Jan LY, Jan YN. Primary structure and expression of a product from cut, a locus involved in specifying sensory organ identity in Drosophila.
Nature
1988
;
333
:
629
–35.
9
Neufeld EJ, Skalnik DG, Lievens PM, Orkin SH. Human CCAAT displacement protein is homologous to the Drosophila homeoprotein, cut.
Nat Genet
1992
;
1
:
50
–5.
10
Valarche I, Tissier-Seta JP, Hirsch MR, Martinez S, Goridis C, Brunet JF. The mouse homeodomain protein Phox2 regulates Ncam promoter activity in concert with Cux/CDP and is a putative determinant of neurotransmitter phenotype.
Development
1993
;
119
:
881
–96.
11
Maitra U, Seo J, Lozano MM, Dudley JP. Differentiation-induced cleavage of Cutl1/CDP generates a novel dominant-negative isoform that regulates mammary gene expression.
Mol Cell Biol
2006
;
26
:
7466
–78.
12
Moon NS, Berube G, Nepveu A. CCAAT displacement activity involves Cut repeats 1 and 2, not the Cut homeodomain.
J Biol Chem
2000
;
275
:
31325
–34.
13
Goulet B, Watson P, Poirier M, et al. Characterization of a tissue-specific CDP/Cux isoform, p75, activated in breast tumor cells.
Cancer Res
2002
;
62
:
6625
–33.
14
Moon NS, Premdas P, Truscott M, Leduy L, Berube G, Nepveu A. S Phase-specific proteolytic cleavage is required to activate stable DNA binding by the CDP/Cut homeodomain protein.
Mol Cell Biol
2001
;
21
:
6332
–45.
15
Truscott M, Raynal L, Premdas P, et al. CDP/Cux stimulates transcription from the DNA polymerase α gene promoter.
Mol Cell Biol
2003
;
23
:
3013
–28.
16
Truscott M, Raynal L, Wang Y, Berube G, Leduy L, Nepveu A. The N-terminal region of the CCAAT displacement protein (CDP)/Cux transcription factor functions as an autoinhibitory domain that modulates DNA binding.
J Biol Chem
2004
;
279
:
49787
–94.
17
Goulet B, Baruch A, Moon NS, et al. A cathepsin L isoform that is devoid of a signal peptide localizes to the nucleus in S phase and processes the CDP/Cux transcription factor.
Mol Cell
2004
;
14
:
207
–19.
18
Sansregret L, Goulet B, Harada R, et al. The p110 isoform of the CDP/Cux transcription factor accelerates entry into S phase.
Mol Cell Biol
2006
;
26
:
2441
–55.
19
Holthuis J, Owen TA, van Wijnen AJ, et al. Tumor cells exhibit deregulation of the cell cycle histone gene promoter factor HiNF-D.
Science
1990
;
247
:
1454
–7.
20
van Wijnen AJ, van Gurp MF, de Ridder MC, et al. CDP/cut is the DNA-binding subunit of histone gene transcription factor HiNF-D: a mechanism for gene regulation at the G1/S phase cell cycle transition point independent of transcription factor E2F.
Proc Natl Acad Sci U S A
1996
;
93
:
11516
–21.
21
De Vos J, Thykjaer T, Tarte K, et al. Comparison of gene expression profiling between malignant and normal plasma cells with oligonucleotide arrays.
Oncogene
2002
;
21
:
6848
–57.
22
Michl P, Ramjaun AR, Pardo OE, et al. CUTL1 is a target of TGF(β) signaling that enhances cancer cell motility and invasiveness.
Cancer Cell
2005
;
7
:
521
–32.
23
Moon NS, Rong Zeng W, Premdas P, Santaguida M, Berube G, Nepveu A. Expression of N-terminally truncated isoforms of CDP/CUX is increased in human uterine leiomyomas.
Int J Cancer
2002
;
100
:
429
–32.
24
Goulet B, Sansregret L, Leduy L, et al. Increased expression and activity of nuclear cathepsin L in cancer cells suggests a novel mechanism of cell transformation.
Mol Cancer Res
2007
;
5
:
899
–907.
25
Cadieux C, Fournier S, Peterson AC, Bédard C, Bedell BJ, Nepveu A. Transgenic mice expressing the p75 CCAAT-displacement protein/Cut homeobox isoform develop a myeloproliferative disease-like myeloid leukemia.
Cancer Res
2006
;
66
:
9492
–501.
26
Bronson SK, Plaehn EG, Kluckman KD, Hagaman JR, Maeda N, Smithies O. Single-copy transgenic mice with chosen-site integration [see comments].
Proc Natl Acad Sci U S A
1996
;
93
:
9067
–72.
27
Lange B, Valtieri M, Santoli D, et al. Growth factor requirements of childhood acute leukemia: establishment of GM-CSF-dependent cell lines.
Blood
1987
;
70
:
192
–9.
28
Hebert S, Bérubé G, Nepveu A. Development of an in vitro assay for the proteolytic processing of the CDP/Cux transcription factor.
J Biochem Mol Biol
2003
;
36
:
390
–8.
29
Yoshimura K, Crystal RG. Transcriptional and posttranscriptional modulation of human neutrophil elastase gene expression.
Blood
1992
;
79
:
2733
–40.
30
Reilly JJ, Jr., Chen P, Sailor LZ, Mason RW, Chapman HA, Jr. Uptake of extracellular enzyme by a novel pathway is a major determinant of cathepsin L levels in human macrophages.
J Clin Invest
1990
;
86
:
176
–83.
31
Schuermann M, Jager R, Salge U, Risse-Hackl G, Havemann K, Heidtmann HH. Control of proteinase expression by phorbol-ester- and Fos-dependent pathways in human non-small-cell lung-cancer cells.
Int J Cancer
1997
;
71
:
275
–83.
32
Akli S, Keyomarsi K. Cyclin E and its low molecular weight forms in human cancer and as targets for cancer therapy.
Cancer Biol Ther
2003
;
2
:
S38
–47.
33
Porter DC, Zhang N, Danes C, et al. Tumor-specific proteolytic processing of cyclin E generates hyperactive lower-molecular-weight forms.
Mol Cell Biol
2001
;
21
:
6254
–69.
34
Ledford AW, Brantley JG, Kemeny G, et al. Deregulated expression of the homeobox gene Cux-1 in transgenic mice results in downregulation of p27(kip1) expression during nephrogenesis, glomerular abnormalities, and multiorgan hyperplasia.
Dev Biol
2002
;
245
:
157
–71.
35
Weijzen S, Rizzo P, Braid M, et al. Activation of Notch-1 signaling maintains the neoplastic phenotype in human Ras-transformed cells.
Nat Med
2002
;
8
:
979
–86.
36
Rios-Doria J, Day KC, Kuefer R, et al. The role of calpain in the proteolytic cleavage of E-cadherin in prostate and mammary epithelial cells.
J Biol Chem
2003
;
278
:
1372
–9.
37
Marambaud P, Shioi J, Serban G, et al. A presenilin-1/γ-secretase cleavage releases the E-cadherin intracellular domain and regulates disassembly of adherens junctions.
EMBO J
2002
;
21
:
1948
–56.
38
Wang XD, Rosales JL, Magliocco A, Gnanakumar R, Lee KY. Cyclin E in breast tumors is cleaved into its low molecular weight forms by calpain.
Oncogene
2003
;
22
:
769
–74.
39
Keyomarsi K, Tucker SL, Buchholz TA, et al. Cyclin E and survival in patients with breast cancer.
N Engl J Med
2002
;
347
:
1566
–75.
40
Carragher NO, Westhoff MA, Riley D, et al. v-Src-induced modulation of the calpain-calpastatin proteolytic system regulates transformation.
Mol Cell Biol
2002
;
22
:
257
–69.