SEARCH

SEARCH BY CITATION

Keywords:

  • cell cortex;
  • KIN1/PAR-1/MARK family;
  • mitosis;
  • maternal embryonic leucine-zipper kinase (Melk);
  • pEg3;
  • Xenopus XL2 cell

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References

Background information. Protein kinase pEg3 belongs to the evolutionarily conserved KIN1/PAR-1/MARK family, whose members are involved in a variety of functions, including cell polarity, microtubule stability, intracellular signalling and the cell cycle. Activity and phosphorylation of pEg3 are cell-cycle dependent and rise to maximum levels during mitosis. pEg3 was shown to interact with and phosphorylate phosphatase CDC25B, and to potentially control cell-cycle progression. Subcellular localization of pEg3 was investigated in Xenopus and human cultured cells.

Results. By expression of GFP (green fluorescent protein)-tagged pEg3 and indirect immunofluorescence with specific antibodies, pEg3 was found to be localized in the cytoplasm and the nucleus in interphase cells. During mitosis pEg3 was also found in the cytoplasm. From anaphase to telophase, a proportion of the protein was detected at the cell cortex. The cortical distribution in mitotic cells was dependent on F-actin, because the actin-depolymerization-inducing drugs cytochalasin D or latrunculin A prevented pEg3 cortical localization. The protein lacking the conserved C-terminal domain was not detected at the cell cortex, whereas the C-terminal domain was targeted to the cell periphery. In contrast with full-length pEg3, the cortical localization of the C-terminal domain and construct lacking the N-terminal domain was cell-cycle independent, and these constructs were found at the cell periphery in interphase cells.

Conclusions. pEg3 is localized at the cell periphery specifically during mitosis. The C-terminal domain is the only pEg3 domain found to be necessary and sufficient for cortical targeting. Cortical distribution of pEg3 also requires the F-actin cytoskeleton. The cell-cycle-independent cortical localization of the pEg3 C-terminal domain and a construct lacking the N-terminal domain indicates that a negative control mechanism involving the pEg3 catalytic N-terminal domain probably acts to prevent pEg3 cortical distribution during interphase. These results suggest that pEg3 might play a role at the cell cortex during mitosis.


Abbreviations used:
C-TAK1

Cdc25C-associated kinase 1

DAPI

4,6-diamidino-2-phenylindole

GFP

green fluorescent protein

HspEg3

human pEg3

KA1 domain

kinase-associated domain 1

KSR1

kinase suppressor of Ras 1

MAP

microtubule-associated protein

MBP

maltose-binding protein

M domain

medium domain

MARK

microtubule-associated protein/microtubule affinity-regulating kinase

PAR-1

partitioning-defective 1

XlpEg3

Xenopus pEg3

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References

We have previously shown that phosphorylation and activity of pEg3, a protein kinase of the KIN1/PAR-1/MARK family (where PAR-1 is partitioning-defective 1 and MARK is microtubule-associated protein/microtubule affinity-regulating kinase), is regulated during the cell cycle (Blot et al., 2002). In Xenopus, pEg3 accumulates and is phosphorylated during oocyte maturation. After fertilization, pEg3 is rapidly dephosphorylated, phosphorylated again at a maximum during mitosis and dephosphorylated upon exit from mitosis. In embryos, the kinase activity of pEg3 is correlated to its phosphorylation state and reaches a maximum during mitosis (Blot et al., 2002). In synchronized cultured Xenopus cells, phosphorylation of pEg3 is detected as early as G2 phase and increases until mitosis. Similarly, in cultured human cells pEg3 kinase activity is correlated to its phosphorylation and is maximal during mitosis (Davezac et al., 2002). In addition, pEg3 associates with and phosphorylates CDC25B protein phosphatase which activates Cdc2-cyclin B complex, a key regulator of entry into mitosis. Experiments involving ectopic expression of recombinant pEg3 in cultured cells suggested that pEg3 may have a role in cell-cycle progression (Davezac et al., 2002). Recently, a pEg3 phosphorylation site on CDC25B was identified. During mitosis CDC25B phosphorylated at this site was found to be localized at the centrosomes and the spindle poles (Mirey et al., 2005). From these results it appears that mitosis represents a critical step in the regulation of pEg3.

Members of the KIN1/PAR-1/MARK protein kinase family are conserved from yeast to humans. These kinases share a common primary structure in which the conserved N-terminal catalytic domain and C-terminal domain are separated by a poorly conserved medium domain (M domain). At the C-terminal end is the KA1 domain (kinase-associated domain 1; Pfam 02149), corresponding to the last 40 amino acids. The KA1 domain is conserved in all members of the KIN1/PAR-1/MARK family, and thus is a signature of these kinases (Barral et al., 1999). Kinases of the KIN1/PAR-1/MARK family are involved in a variety of different biological functions, including cell polarity, cell-cycle control, intracellular signalling, microtubule and protein stability (reviewed in Tassan and Le Goff, 2004). The cell-cycle-dependent regulation of pEg3 appears to be unique in the KIN1/PAR-1/MARK protein kinase family. Interestingly, pEg3 and pEg3 orthologues, named in mouse Melk (maternal embryonic leucine zipper kinase; Heyer et al. 1997) or MPK38 (murine protein kinase 38; Gil et al., 1997), in humans KIAA0175 (Nagase et al., 1996) and in Caenorhabditis elegans W03G1.6 (Wormpep, identified by the C. elegans genome project), seem to constitute a distinct group in the KIN1/PAR-1/MARK family (Blot et al., 2002).

The subcellular localization was determined for several kinases of the KIN1/PAR-1/MARK family. Mammalian and Drosophila homologues of PAR-1 are asymmetrically localized to the basolateral membrane in polarized epithelial cells (Böhm et al., 1997; Shulman et al., 2000; Cox et al., 2001). Interestingly, subcellular distribution was reported to be a dynamic process for several kinases. For example, the PAR-1 protein involved in the antero-posterior axis determination in C. elegans is localized to the cytoplasm of the oocyte and is partially redistributed to the posterior cell cortex after fertilization (Guo and Kemphues, 1995). In Schizosaccharomyces pombe, the Kin1p protein is found at the cell ends (Drewes and Nurse, 2003; La Carbona et al., 2004). Subcellular localization of Kin1p was also proposed to be a dynamic process, as Kin1p disappears during mitosis without being degraded and its recruitment at the cell ends is cell-cycle dependent (Drewes and Nurse, 2003). Redistribution is also observed for C-TAK1 (Cdc25C-associated kinase 1; also named MARK3 or Par-1a), which constitutively associates with mammalian KSR1 (kinase suppressor of Ras 1). In resting cells, C-TAK1 maintains the phosphorylation state of KSR1 on Ser392, generating a 14-3-3 binding site that results in cytoplasmic localization. In response to growth factor treatment, the phosphorylation of Ser392 decreases and the KSR1 complex, including C-TAK1, is translocated from the cytoplasm to the cell surface where it serves as a docking platform for components of the Ras/MAPK (mitogen-activated protein kinase) pathway (Müller et al., 2001).

In the present study, we report the cortical localization of GFP (green fluorescent protein)-tagged pEg3 in transfected Xenopus and human mitotic cells. Similarly, indirect immunofluorescence allowed the detection of endogenous pEg3 at the cortex during anaphase and telophase. This localization relies on the presence of intact actin filaments and the pEg3 C-terminal domain. Transfection experiments also allowed the identification of a temporal control mechanism that restricts pEg3 localization to the cell cortex during mitosis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References

Accumulation of GFP—pEg3 at the cell cortex in transfected mitotic Xenopus XL2 cells

To determine the subcellular localization of pEg3, Xenopus XL2 cells were transiently transfected with a plasmid expressing the Xenopus full-length pEg3 protein fused to the N-terminal end of the GFP (pEGFPN1-XlEg3 plasmid). In interphase cells, the GFP—pEg3 fusion protein was detected in the cytoplasm and associated with short filamentous structures composed of F-actin, as shown by staining with rhodamine-conjugated phalloidin (Figures 1a1c and Supplementary Figure S1, http:www.biolcell.orgboc098boc0980253add.htm). In 30% of transfected cells, GFP—pEg3 was also detected in the nucleus (data not shown). In cells transfected with the empty pEGFPN1 plasmid, GFP accumulated in the nucleus (Figure 1a′) and no staining of actin bundles was observed. Distribution of GFP—pEg3 was also examined in mitotic cells. In prophase and metaphase cells, GFP—pEg3 was distributed in the cytoplasm (data not shown and Figure 1d respectively). In anaphase and telophase, GFP—pEg3 was detected at the cell periphery (Figures 1e and 1f, and additional cells are shown in Supplementary Figure S2, http:www.biolcell.orgboc098boc0980253add.htm). In contrast, in mitotic cells expressing GFP alone, the protein was distributed uniformly in the cytoplasm and was not detected at the cell periphery (Figure 1e′), indicating that the distribution of GFP—pEg3 is dependent on pEg3. As shown in Figure 1(g), the distribution pattern of the GFP—pEg3 fusion protein was similar to that of pEg3 fused to the C-terminal end of GFP (pEGFPC1-XlEg3 plasmid). Labelling of F-actin with rhodamine-conjugated phalloidin showed that peripheral GFP—pEg3 localized with cortical F-actin in anaphase cells (Figures 1h and 1i). These results suggest that pEg3 is recruited to the cell cortex during mitosis.

image

Figure 1. Exogenous GFP—pEg3 is recruited to the cell cortex in mitotic XL2 cells

XL2 cells were transiently transfected with the pEGFPN1-XlEg3 (af), pEGFPC1-XlEg3 (gi) and pEGFP (a′ and e′) plasmids. Subcellular distribution of GFP—pEg3 was examined during interphase (ac) and mitosis (df, gi). Localization of GFP alone was observed during interphase (a′) and anaphase (e′). F-actin was detected with rhodamine-conjugated phalloidin (red, b and h; merge, c and i), and DNA was stained with DAPI (blue, insets). Scale bar, 10 μm.

Download figure to PowerPoint

Endogenous pEg3 is localized to the cortex of mitotic Xenopus XL2 and human HeLa cells

In order to examine the localization profile of endogenous pEg3, indirect immunofluorescence was performed on Xenopus XL2 cells. For this purpose immunoglobulins from a rabbit polyclonal serum (L2; Blot et al., 2002) raised against the last 305 amino acids of the non-catalytic domain of XlpEg3 (Xenopus pEg3) were affinity purified. In interphase XL2 cells, anti-pEg3 antibody stained the cytoplasm and the nucleus (Figure 2a). Notably, some cell nuclei were not stained, as reported for the GFP—pEg3-expressing cells, whereas other cells displayed a nuclear staining pattern. The significance of this variability requires more investigation to be understood. Contrary to the observation seen in GFP—pEg3-transfected cells, actin bundles were not detected, suggesting a mis-localization of GFP—pEg3 which may be the con-sequence of over-expression. In mitotic cells, the cyto-plasm was stained with the affinity-purified L2 immunoglobulins (Figures 2c2e). In addition, a thin fluorescent layer appeared at the periphery of cells from anaphase to telophase (Figures 2d and 2e, and Supplementary Figure S3, http:www.biolcell.orgboc098boc0980253add.htm), which can be super-imposed on to labelling of the actin filaments with rhodamine-conjugated phalloidin (Figures 2j2l). The same amount of immunoglobulins purified from the L2 preimmune serum gave a very weak general staining in interphase and mitotic cells (Figures 2b and 2f respectively). In order to ensure that this staining was specific to pEg3, the affinity-purified L2 antibody was preincubated with recombinant full-length XlpEg3 fused to MBP (maltose-binding protein) (MBP—XlpEg3). Since the L2 antibody was raised against a fusion protein of the last 305 amino acids of the non-catalytic domain of XlpEg3 with MBP, the purified L2 antibody was incubated with MBP alone. Detection of pEg3 in asynchronous and mitotic XL2 cells was blocked by the prior incubation of anti-pEg3 antibody with MBP—XlpEg3, but not with MBP alone (see Figure 2i for Western blot analysis, and Figures 2g and 2h for immunofluorescence results; only anaphase cells are shown). Taken together, these results show that, as observed with GFP—pEg3, a subset of endogenous XlpEg3 displayed a cortical localization from anaphase to telophase.

image

Figure 2. XlpEg3 is localized to the cell cortex during mitosis

Affinity purified anti-XlpEg3 L2 antibody (a, c, d and e) and Protein A-purified immunoglobulins of the corresponding preimmune serum (b and f) were used to visualize proteins with indirect immunofluorescence in XL2 cells during interphase (a and b) and mitosis (cf). The reactivity of the L2 antibody, after preincubation with PBS and MBP used as controls or MBP—XlpEg3 to block immunoglobulins, was tested by Western blotting (i) with asynchronously growing XL2 cells (As) or mitotic (M) XL2 cells. The cell periphery was stained by the L2 antibody preincubated with MBP (g), but not by the L2 antibody blocked by MBP—XlpEg3 (h). Cortical co-localization (orange) in mitotic cells of pEg3 revealed with the L2 antibody (j), and F-actin stained with rhodamine-conjugated phalloidin (k) is shown in (l). DNA was stained with DAPI (blue, insets). Scale bar, 10 μm.

Download figure to PowerPoint

Similar to XlpEg3, the kinase activity of HspEg3 (human pEg3) is cell-cycle dependent and reaches a maximum during mitosis (Davezac et al., 2002). This common regulatory mechanism suggests that HspEg3 could also have a cell-cycle-dependent subcellular localization. To explore the localization of pEg3 in cultured human cells, we used affinity-purified immunoglobulins of a rabbit polyclonal antibody (M1) raised against HspEg3 M domain (Davezac et al., 2002). In exponentially growing HeLa cells (‘As’), the M1 antibody revealed a protein of 75 kDa that corresponded to the predicted molecular mass of pEg3 calculated from the amino acid sequence. We have previously shown that the electrophoretic mobility of pEg3 is decreased during mitosis (Blot et al., 2002; Davezac et al., 2002). As expected the M1 antibody detected several proteins with reduced electrophoretic mobility in HeLa cells blocked in mitosis with nocodazole (Figure 3a). These bands were not detected with the corresponding preimmune serum. The M1 antibody stained the cortical layer of HeLa cells in anaphase and telophase, but not in metaphase (Figure 3c and Supplementary Figure S4, http:www.biolcell.orgboc098boc0980253add.htm). There was no immunostaining of the cell cortex with the preimmune serum (Figure 3b). The cortical staining disappeared when affinity purified M1 antibody was blocked by preincubation with HspEg3 (Figures 3a and 3e), whereas it was clearly visible when antibody was incubated with PBS (Figure 3d). The cellular cortex of HeLa cells was also stained with the anti-XlpEg3 L2 serum (Figure 3f), which cross-reacts with HspEg3 via the conserved C-terminal domain of the protein (Davezac et al., 2002). In addition, in HeLa cells transfected with pEGFP-HsEg3FL plasmid, a proportion of GFP—pEg3 was detected at the cell periphery and co-localized with F-actin (Figures 3g3i). These results show that the mitotic cortical localization of pEg3 is conserved in human cells.

image

Figure 3. Endogenous HspEg3 is localized to the cortex of mitotic HeLa cells

Proteins extracted from asynchronous (As) and mitotic (M) HeLa cells were probed (a) with affinity-purified M1 immunoglobulins against HspEg3 (M1) or the corresponding preimmune serum (PiM1). Arrows indicate the position of pEg3 in asynchronous and mitotic cells. Reactivity of M1 antibody preincubated with PBS or with HspEg3 was tested by Western blotting with the same protein extracts (As and M). HeLa cells in anaphase showed no cortical staining with the preimmune serum (b). Cortical localization of pEg3 was detected with affinity-purified M1 antibody (c), with the anti-XlpEg3 L2 antibody (f) or with M1 antibody preincubated in PBS (d), but not with M1 antibody blocked by preincubation with HspEg3 (e). HeLa cells were transfected with the plasmid pEGFP-HsEg3FL. GFP—pEg3 (green, g) and F-actin detected with rhodamine-conjugated phalloidin (red, h) were co-localized (orange, i). DNA was stained with DAPI (blue, insets). Scale bar, 10 μm.

Download figure to PowerPoint

pEg3 cortical localization is dependent on actin microfilaments

During mitosis, pEg3 was found to co-localize with cortical actin. To test the importance of the actin cytoskeleton for pEg3 cortical localization, cells were treated with cytochalasin D to disrupt actin filaments. After incubation, HeLa cells were fixed and immunostained with purified M1 antibody to examine pEg3 localization, and rhodamine-conjugated phalloidin was used to detect actin filaments. In treated cells, cortical microfilaments were greatly perturbed compared with untreated cells and the cortical localization of pEg3 disappeared (Figure 4 and Supplementary Figure S5, http:www.biolcell.orgboc098boc0980253add.htm). The same result was obtained when cells were incubated with latrunculin A, an inhibitor of F-actin assembly (Figure 4 and Supplementary Figure S5, http:www.biolcell.orgboc098boc0980253add.htm). The mitotic spindle and chromosomes did not show any apparent perturbation after cytochalasin D or latrunculin A treatment (Supplementary Figure S6, http:www.biolcell.orgboc098boc0980253add.htm). The cortical localization of pEg3 was not affected by microtubule depolymerization (Supplementary Figure S7, http:www.biolcell.orgboc098boc0980253add.htm). Thus the association of pEg3 with the cortex during mitosis depends on the integrity of the filamentous actin network.

image

Figure 4. Cortical localization of pEg3 depends on intact actin filaments

The actin cytoskeleton of HeLa cells was disrupted by treatment with cytochalasin D or latrunculin A for 30 min at 37°C. DMSO was used as a control. After fixation, cells were incubated with rhodamine-conjugated phalloidin to stain F-actin (red; a, c, e) or with purified M1 antibody (green; b, d, f). DNA was stained with DAPI (blue, insets). Scale bar, 10 μm.

Download figure to PowerPoint

pEg3 cortical localization depends on the presence of its C-terminal domain and is temporally restricted to mitosis

There is a correlation between the cell-cycle-dependent regulation of pEg3 and its cortical localization during mitosis. This led to the proposal that the cortical targeting of pEg3 may be dependent on its kinase activity. As for Xenopus GFP—pEg3 expressed in XL2 cells, transiently expressed GFP-tagged human full-length pEg3 (GFP—pEg3FL) localized at the cell cortex of HeLa cells during anaphase (Figures 3g and 5a) and telophase (data not shown). To determine if the kinase activity was necessary for the cortical localization of pEg3, HeLa cells were transfected with a plasmid expressing a catalytically inactive HspEg3 fused to GFP (pEGFPC1-HsEg3FLK/R). Similar to the active pEg3, the mutant protein accumulated at the cell cortex in anaphase (Figure 5b) and telophase (data not shown). This result indicates that the cortical localization of pEg3 during mitosis does not depend on its kinase activity.

image

Figure 5. C-terminal domain of pEg3 is necessary and sufficient for cortical localization

Primary structure of pEg3 is represented by a box showing the three pEg3 domains N (N-terminal domain; corresponds to the catalytic domain of the kinase), M and C (C-terminal domain). K/R indicates the position of amino acid substitution, Lys[RIGHTWARDS ARROW]Arg, in the catalytic domain. GFP-tagged full-length HspEg3 (FL), catalytically inactive form of the protein (FLK/R) and various truncated fragments of pEg3 were transiently transfected in HeLa cells. GFP alone was used as control. Subcellular localization was examined in mitotic cells (af, only anaphase is shown) and interphase cells (gl). DNA was stained with DAPI (blue, insets). Scale bar, 10 μm.

Download figure to PowerPoint

Therefore, involvement of the different domains of the protein in the cortical recruitment of pEg3 was analysed by the expression of various truncated GFP-tagged pEg3 constructs in human cells. The GFP—pEg3MC fusion protein that lacked the N-terminal catalytic domain maintained a mitotic cortical localization (Figure 5c and Supplementary Figure S8, http:www.biolcell.orgboc098boc0980253add.htm). In contrast, when the conserved C-terminal domain was deleted, the resulting fusion protein (GFP—pEg3NM) was distributed in the cytoplasm during mitosis, but did not localize to the cell cortex (Figure 5d). Moreover, expression of the C-terminal domain (GFP—pEg3C) resulted in the cortical staining of cells in anaphase (Figure 5e and Supplementary Figure S8, http:www.biolcell.orgboc098boc0980253add.htm) and telophase (data not shown). These data highlight the role of the pEg3 C-terminal domain in the cortical localization of the kinase.

The subcellular localization of GFP constructs described above was also examined in interphase HeLa cells. GFP—pEg3FL (full-length pEg3) and GFP—pEg3K/R (catalytically inactive pEg3) were detected in the cytoplasm (Figures 5g and 5h). As mentioned above for XlpEg3, we observed a cell-to-cell variability in the nuclear staining (some nuclei were stained and others were not). Interestingly, we observed that GFP—pEg3C was present at the periphery of interphase cells (Figure 5k and Supplementary Figure S8, http:www.biolcell.orgboc098boc0980253add.htm). This result is in agreement with the finding that the pEg3 C-terminal domain is necessary and sufficient to target pEg3 to the cell cortex. Accordingly, GFP—pEg3NM did not show cortical localization during interphase (Figure 5j). This result revealed that a temporal control exists which restricts the cortical localization of pEg3 to mitosis. This control was lost in the truncated construction GFP—pEg3C, since this construct was no longer restricted to mitotic cells. The same observation was made with GFP—pEg3MC (Figure 5i and Supplementary Figure S8, http:www.biolcell.orgboc098boc0980253add.htm), indicating that the M domain is not sufficient to restore the temporal regulation of pEg3 cortical localization and that the catalytic domain may be involved in this control mechanism.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References

We have previously shown that pEg3 regulation is cell-cycle dependent. pEg3 kinase activity correlates with its phosphorylation and reaches a maximum during mitosis (Blot et al., 2002; Davezac et al., 2002). In the present study we report the cortical localization of Xenopus and human pEg3 during mitosis. Evidence for the cortical distribution of pEg3 is supported by ectopic expression of GFP-tagged pEg3 and by immunostaining of the endogenous protein. Using both approaches, cortical distribution of pEg3 was observed during anaphase and telophase in Xenopus and human cells.

The catalytically inactive GFP-tagged pEg3 accumulates at the cell cortex, as does the active protein, demonstrating that pEg3 kinase activity is not required for cortical localization. In addition, in the absence of the catalytic domain, the non-catalytic MC construct still displays a cortical distribution during mitosis, showing that the catalytic domain is not involved in pEg3 cortical localization. In contrast, the C-terminal domain is necessary and sufficient for cortical targeting of the protein.

The cell cortex is a complex and undefined structure where actin filament networks are concentrated. The use of cytoskeletal inhibitors showed that the cortical localization of pEg3 during mitosis is dependent on the integrity of actin filaments. It seems unlikely that pEg3 binds directly to microfilaments, since no actin-binding motif was found in the pEg3 sequence. In C. elegans early embryos, a C-terminal fragment of PAR-1 interacts with the non-muscle myosin NMY2, and this interaction is necessary for the localization of PAR-1 to the cell cortex (Guo and Kemphues, 1996). Similarly, the pEg3 C-terminal domain might be important for interaction with putative partners localized at the cell cortex during mitosis.

Experiments with truncated pEg3 show that, in contrast to GFP—pEg3FL, the peripheral localization of GFP—pEg3MC and GFP—pEg3C is not restricted to mitotic cells. This result indicates that a negative control involving the N-terminal domain exists to restrict the cortical localization of pEg3 to mitosis. The mode of action of this control is unknown, but an exclusion mechanism acting on mammalian PAR-1 has been recently identified. This involves aPKC (atypical protein kinase C), which phosphorylates PAR-1 on the M domain, therefore excluding it from the cell membrane (Hurov et al., 2004). Releasing the negative control might be sufficient for pEg3 to localize at the cortex at anaphase. However, we cannot exclude the possibility that an additional positive signal activated during mitosis may be necessary to trigger pEg3 cortical localization. A correlation exists between the cortical localization of pEg3 and its phosphorylation state during mitosis. Indeed, pEg3 is highly phosphorylated during mitosis in XL2 cells (Blot et al., 2002, and Figure 2b), and the non-catalytic MC fragment is phosphorylated in Xenopus egg extracts (Blot et al., 2002). It is likely that phosphorylation has an important role in the regulation of pEg3 cortical localization during mitosis. Since pEg3 is first detected at the cortex at anaphase, it could be possible that phosphorylation of pEg3 before or after anaphase prevents or induces pEg3 cortical localization respectively.

By expression of GFP-tagged pEg3 and indirect immunofluorescence, pEg3 was also detected in the cytoplasm and in the nucleus of interphase cells, and during mitosis pEg3 showed a cytoplasmic distribution in addition to cortical localization. In this respect, our data are not in contradiction with the previous reports that pEg3 interacts with CDC25B (Davezac et al., 2002), the zinc-finger-like protein ZPR9 (Seong et al., 2002) and the splicing factor NIPP1 (Vulsteke et al., 2004), none of which have been described as cortically localized proteins.

Cortical localization has already been described for several members of the KIN1/PAR-1/MARK protein kinase family to which pEg3 belongs. Distribution of Kin1p to cell ends has already been reported to vary during the cell cycle in S. pombe (Drewes and Nurse, 2003). In C. elegans, the intensity of PAR-1 labelling varies during mitosis of the first embryonic cell cycle (Guo and Kemphues, 1995). However, for higher eukaryotic KIN1/PAR-1/MARK family members, pEg3 is the first kinase described to show a cell-cycle-dependent cortical localization in cultured somatic cells. The functional signification of this localization remains unknown. However, the finding that the cortical distribution of pEg3 during mitosis was conserved from Xenopus to humans indicates that this subcellular localization may be important for pEg3 function. The fact that pEg3 kinase activity is maximal in mitotic cells indicates that pEg3 might regulate cortical substrate(s) during mitosis. The uniform cortical pEg3 distribution suggests that pEg3 might exert its function all around the cell cortex. Substrate(s) might correspond to constitutive component(s) of the cell cortex specifically regulated during mitosis or to protein(s) localized at the cortex only during mitosis. One of the potential pEg3 substrates might be the mammalian protein LGN, an orthologue of the Drosophila Pins protein which shows mitosis-specific cortical localization (Kaushik et al., 2003). pEg3 may have a role in the maintenance of the spatial positioning of the mitotic spindle in cells during anaphase and telophase. It is known that the mammalian MARK proteins phosphorylate MAPs (microtubule-associated proteins) and destabilize the microtubule network (Drewes et al., 1997). By analogy, pEg3 might phosphorylate MAP(s) at the vicinity of the cell cortex, therefore modifying the stability of microtubules controlling the spindle position. Alternatively, in certain tissue types, pEg3 might be asymmetrically localized, and thus may have an additional role during ‘unequal’ mitosis, such as the control of asymmetric cortical localization of cell fate determinants. Undoubtedly, identification of the cortical substrates of pEg3 will help to explain its function at the cell cortex.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References

Cell culture, cell synchronization and drug treatment

Xenopus XL2 cells (Anizet et al., 1981) were cultured in L15 medium at 25°C and HeLa cells in DMEM (Dulbecco's modified Eagle's medium) at 37°C at 5% CO2. Media were supplemented with 10% foetal calf serum, 50 units/ml penicillin and 50 μg/ml streptomycin (all culture reagents were purchased from Gibco BRL). Synchronization of XL2 cells was performed as previously described (Blot et al., 2002). Mitotic HeLa cells were obtained after an overnight incubation with 0.5 μg/ml nocodazole (Sigma) at 37°C. Asynchronous HeLa cells were harvested by trypsinization. To disrupt the actin cytoskeleton, cells were incubated with 2.5 μg/ml cytochalasin D (Sigma) or 0.2 μg/ml latrunculin A (Molecular Probes) for 30 min at 37°C and protein localization was revealed by indirect immunofluorescence.

Plasmid construction

Plasmids pmal3c-XlEg3, pmal3c-XlEg3K/R, pET21a-XlEg3, pET21a-HsEg3 and pET21a-HsEg3K/R for production of MBP—XlpEg3, MBP—XlpEg3K/R, and His-tagged XlpEg3 (His6–XlpEg3), HspEg3 (His6–HspEg3) and HspEg3K/R (His6–HspEg3K/R) respectively were constructed as described previously (Blot et al., 2002; Davezac et al., 2002). To generate fusion protein of GFP with XlpEg3, the coding sequence of XlpEg3 cDNA was amplified by PCR with the primers 5′-CGCAGATCTGTCATGGCTGTGGATG-3′ (sense, BglII restriction site underlined) and 5′-CGGTCTAGATTACACCTTACAGCTGGA-3′ (antisense, XbaI restriction site underlined) using pBluescript-XlEg3 (Blot et al., 2002) as a template. The PCR product was digested with BglII and XbaI and ligated into pEGFP-C1 (Clontech Laboratories) at the BglII and XbaI sites to produce the plasmid pEGFPC1-XlEg3. The pEGFPN1-XlEg3 plasmid was produced by PCR amplification of the XlpEg3 gene from pT7T-XlEg3 (Blot et al., 2002) with the primers 5′-CGGGAGCTCGCTTGTTCTTTTTGCAGAAG-3′ (sense, SacI) and 5′-GCCCCGCGGCACCTTACAGCTGGAGAG-3′ (antisense, SacII). After digestion with SacI and SacII, the PCR product was introduced into pEGFP-N1 vector.

Preparation of the plasmids pEGFP-HsEg3FL and pEGFP-HsEg3FLK/R has been described previously (Davezac et al., 2002). Fragments of HspEg3 were generated by PCR of pBluescript-HsEg3 plasmid (Nagase et al., 1996). The MC noncatalytic domain was amplified with the primers 5′-GAAGAT-CTCACCTCGATGATGATTG-3′ (sense, BglII) and 5′-CGGAATTCACCTTGCAGCTAGATAGG-3′ (antisense, EcoRI). The PCR fragment was digested with BglII and EcoRI enzymes and ligated into pEGFP-C1 vector to generate pEGFP-HsEg3MC plasmid. Truncated HsEg3NM was amplified with primers 5′-CGCGGATCCAAAGATTATGATGAACTTC-3′ (sense, BamHI) and 5′-CCGAATTCTTATATTTTAATTGGAGTTT-3′ (antisense, EcoRI; and introduction of a UAA stop codon indicated in italic), digested with BamHI and EcoRI enzymes and ligated into BglII and EcoRI sites of pEGFP-C1, generating pEGFPC1-HsEg3NM plasmid. The C-terminal domain was amplified with primers 5′-CGGAATTCTGTCATGTGCCGCTCAGTGGA-3′ (sense, EcoRI; and introduction of an in-frame AUG initiator codon indicated in italic) and 5′-CGGGATCCCGTACCTTGCAGCTAGATAGGA-3′ (antisense, BamHI). The PCR fragment was digested with EcoRI and BamHI enzymes, and introduced into the pEGFP-N1 vector, which was disgested with the same enzymes to generate pEGFPN1-HsEg3C plasmid.

All constructs were subsequently verified by DNA sequencing. DNA was purifed using an Endofree Plasmid kit (Qiagen).

Transfection

Cells were grown from 50 to 70% confluency on glass coverslips. Plasmid DNA (0.5 μg) was transfected into XL2 cells using Transfast™ transfection reagent (Promega) or Fugene 6 (Roche) and into HeLa cells using Exgen 500 (Euromedex), according to the manufacturers' instructions.

Production, purification and blocking of anti-pEg3 antibody

Anti-XlpEg3 (L2) and anti-HspEg3 (M1) antibodies have been described previously (Blot et al., 2002; Davezac et al., 2002). The L2 antibody was affinity purified on a His-tagged XlpEg3 column. For this purpose His6–XlpEg3 was immobilized on CNBr-activated Sepharose 4B according to the manufacturer's instructions (Amersham Biosciences). M1 immunoglobulins were affinity purified on a His6–HspEg3K/R column using the same method. To block the L2 antibody, 0.5 μg of affinity-purified L2 immunoglobulin was incubated with 4 μg of MBP—XlpEg3K/R or MBP alone as a control in PBS at room temperature for 3 h. The M1 antibody was blocked with His6–HspEg3 using the same method. Antibodies were used for Western blotting or indirect immunofluorescence at the same concentration as non-blocked antibody. Immunoglobulins of the L2 pre-immune serum were purified with Protein A—Sepharose, as described by Harlow and Lane (1988), and used at the same concentration as affinity-purified L2 antibody.

Indirect immunofluorescence microscopy

Cells were grown on glass coverslips, fixed with 3.7% formaldehyde solution (Sigma) in PBS for 10 min at room temperature and permeabilized in PBS/0.1% Triton X-100 for 3 min at room temperature. Non-specific binding sites were blocked with PBS containing 1% BSA for 30 min. The cells were then incubated with 2 μg/ml affinity-purified L2 antibody or 2 μg/ml affinity-purified M1 antibody overnight at 10°C. Primary antibodies were revealed with FITC-conjugated goat anti-rabbit IgG (Jackson Immunoresearch) for 1 h at room temperature. All antibody reagents were diluted in PBS containing 1% BSA. F-actin was stained with rhodamine-conjugated phalloidin (1:100; Molecular Probes) for 30 min at room temperature. DNA was visualized with 0.5 μg/ml DAPI (4,6-diamidino-2-phenylindole; Sigma) added to the cells with the secondary antibody or with rhodamine-conjugated phalloidin. The coverslips were mounted on to glass slides in Vectashield mounting medium (Vector). Epifluorescence was monitored with a DM RXA Leica microscope coupled to a Q550 CW image analysis system (Leica) (microscopy platform, IFR140 GFAS ‘Génomique Fonctionnelle, Agronomie et Santé’).

Western blotting analysis

Proteins were separated by SDS/PAGE (Laemmli, 1970) and were electrotransferred on to a PVDF membrane (Immobilon-P, Millipore). Membrane blocking and antibody incubation were performed as described previously (Blot et al., 2002). Affinity-purified L2 and M1 antibodies were used at 0.1 μg/ml and 0.2 μg/ml respectively.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References

We thank Xavier Le Goff and Michel Philippe for critical reading of this manuscript and lab members for helpful discussions. We are grateful to Fabien Cubizolles for technical advice for the affinity purification of the antibodies. We thank Jean-Philippe Gagné for proofreading of the manuscript prior to submission. This work was supported by Ligue Nationale Contre le Cancer (équipe labellisée) and Cancéropôle Grand Ouest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. References
  • Anizet M.P. Huwe B. Pays A. Picard J.J. Characterization of a new cell line, XL2, obtained from Xenopus laevis and determination of optimal culture conditions In Vitro 1981 17 267274.
  • Barral Y. Parra M. Bidlingmaier S. Snyder M. Nim1-related kinases coordinate cell cycle progression with the organization of the peripheral cytoskeleton in yeast Genes Dev. 1999 13 176187.
  • Blot J. Chartrain I. Roghi C. Philippe M. Tassan J.-P. Cell cycle regulation of pEg3, a new Xenopus protein kinase of the KIN1/PAR-1/MARK family Dev. Biol. 2002 241 327338.
  • Böhm H. Brinkmann V. Drab M. Henske A. Kurzchalia T.V. Mammalian homologues of C. elegans PAR-1 are asymmetrically localized in epithelial cells and may influence their polarity Curr. Biol. 1997 7 603606.
  • Cox D.N. Lu B. Sun T.-Q. Williams L.T. Jan Y.N. Drosophila par-1 is required for oocyte differenciation and microtubule organization Curr. Biol. 2001 11 7587.
  • Davezac N. Baldin V. Blot J. Ducommun B. Tassan J.P. Human pEg3 kinase associates with and phosphorylates CDC25B phosphatase: a potential role for pEg3 in cell cycle regulation Oncogene 2002 21 76307641.
  • Drewes G. Nurse P. The protein kinase kin1, the fission yeast orthologue of mammalian MARK/PAR-1, localises to new cell ends after mitosis and is important for bipolar growth FEBS Lett. 2003 554 4549.
  • Drewes G. Ebneth A. Preuss U. Mandelkow E.M. Mandelkow E. MARK, a novel family of protein kinases that phosphorylate microtubule-associated proteins and trigger microtubule disruption Cell (Cambridge, Mass.) 1997 89 297308.
  • Gil M. Yang Y. Lee Y. Choi I. Ha H. Cloning and expression of a cDNA encoding a novel protein serine/threonine kinase predominantly expressed in hematopoietic cells Gene 1997 195 295301.
  • Guo S. Kemphues K.J. par-1, a gene required for establishing polarity in C. elegans embryos, encodes a putative Ser/Thr kinase that is asymmetrically distributed Cell (Cambridge, Mass.) 1995 81 611620.
  • Harlow E. Lane D. Antibodies: A Laboratory Manual 1988 Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press.
  • Heyer B.S. Warsowe J. Solter D. Knowles B.B. Ackerman S.L. New member of the Snf1/AMPK kinase family, Melk, is expressed in the mouse egg and preimplantation embryo Mol. Reprod. Dev. 1997 47 148156.
  • Hurov J.B. Watkins J.L. Piwnica-Worms H. Atypical PKC phosphorylates PAR-1 kinases to regulate localization and activity Curr. Biol. 2004 14 736741.
  • Kaushik R. Yu F. Chia W. Yang X. Bahri S. Subcellular localization of LGN during mitosis: evidence for its cortical localization in mitotic cell culture systems and its requirement for normal cell cycle progression Mol. Biol. Cell 2003 14 31443155.
  • La Carbona S. Allix C. Philippe M. Le Goff X. The protein kinase kin1 is required for cellular symmetry in fission yeast Biol. Cell 2004 96 169179.
  • Laemmli U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature (London) 1970 227 680685.
  • Mirey G. Chartrain I. Froment C. Quaranta M. Bouche J.P. Monsarrat B. Tassan J.P. Ducommun B. CDC25B phosphorylated by pEg3 localizes to the centrosome and the spindle poles at mitosis Cell Cycle 2005 4 806811.
  • Müller J. Ory S. Copeland T. Piwnica-Worms H. Morrison D.K. C-TAK1 regulates Ras signaling by phosphorylating the MAPK scaffold, KSR1 Mol. Cell 2001 8 983993.
  • Nagase T. Seki N. Ishikawa K. Tanaka A. Nomura N. Prediction of the coding sequences of unidentified human genes. V. The coding sequences of 40 new genes (KIAA0161–KIAA0200) deduced by analysis of cDNA clones from human cell line KG-1 DNA Res. 1996 3 1724.
  • Seong H.A. Gil M. Kim K.T. Kim S.J. Ha H. Phosphorylation of a novel zinc-finger-like protein, ZPR9, by murine protein serine/threonine kinase 38 (MPK38) Biochem. J. 2002 361 597604.
  • Shulman J.M. Benton R. St. Johnston D. The Drosophila homolog of C. elegans PAR-1 organizes the oocyte cytoskeleton and directs oskar mRNA localization to the posterior pole Cell (Cambridge, Mass.) 2000 101 377388.
  • Tassan J.P. Le Goff X. An overview of the KIN1/PAR-1/MARK kinase family Biol. Cell 2004 96 193199.
  • Vulsteke V. Beullens M. Boudrez A. Keppens S. Van Eynde A. Rider M.H. Stalmans W. Bollen M. Inhibition of spliceosome assembly by the cell cycle-regulated protein kinase MELK and involvement of splicing factor NIPP1 J. Biol. Chem. 2004 279 86428647.