Functional analysis of the LIM kinase 1 and its role in cell cycle progression

LIMK1 (LIM-kinase 1) is a member of the LIMKs family of serine/threonine kinases, comprised of LIMK1, LIMK2 and testicular specific kinases, TESK1 and TESK2. These enzymes catalyze phosphorylation of an actin depolymerizing factor cofilin, thereby inactivate its depolymerizing activity, that leads to actin stabilization as filamentous actin (F-actin). As a consequence, LIMK1 plays a key role in actin cytoskeleton remodeling, notably in response to many extracellular cues that trigger the activation of the small Rho GTPases Rho, Rac and Cdc42, these latter being able to activate LIMK1 through their downstream effectors ROCK, PAK1 and PAK4 respectively. Moreover, on a pathological point of view, dysregulation of LIMK1 has being associated to several diseases: indeed, partial loss of LIMK1 is associated to a neurological disorder called William’s syndrome, whereas LIMK1 upregulation and/or hyperactivation is linked to cancer metastasis. The introduction chapter of this thesis details the current knowledge about LIMK1 and its function in cell migration, cell cycle, neuronal differentiation, and phagocytosis. Additionally, the contribution of LIMK1 dysregulation in pathological circumstances, notably in WilliamÕs syndrome and in cancer metastasis is discussed. The results chapter summarizes the work undertaken as partial fulfilment of this doctoral study. In the first part of the results, the endogenous localization of LIMK1 is described in cell lines and tissues. We showed that, in different cell lines, LIMK1 and LIMK2 localize mainly to cytoplasm with enrichment in cell membrane in the direction of movement. Moreover, we observed that
active LIMK1 and LIMK2 localize to membrane ruffles. Interstingly, a more detailed analysis of
LIMK1 by immunostaining and fractionation shows that LIMK1 localizes to mitochondria, suggesting
a possible new function or regulation of LIMK1 in this compartment. In contrast to cell
lines, LIMK1 is shown to localize to both nucleus and cytoplasm in kidney and prostate tissues.
However, in mouse kidney and prostate cancer models, we observed that LIMK1 relocalized to
cytoplasm. Altogether, these suggest that, in exponentially growing cells (cultured cells), LIMK1
localizes to cytoplasm, whereas in differentiated cells (in tissues), its localization is both nuclear
and cytoplasmic. Moreover, during cancer development, LIMK1 is thought to relocalize mainly
to the cytoplasm, which might represent a new marker of cancer, proliferating cells.
In the second part of results, functional inactivation studies of LIMK1 by knockout mice and
RNA interference (siRNA and lentivector-mediated shRNA) are described. For knockout mice
part, the identification of embryonic stem (ES) cells with one allele of total deleted or floxed
LIMK1 gene has been completed. In RNA interference part, the characterization of a LIMK1
siRNA, which efficiently targets endogenous LIMK1, has been performed. This siRNA has been
later on used for downregulation experiments aiming at exploring the effect of LIMK1 in cell
cycle regulation. Additionally, the generation of a powerful drug- inducible lentivector-based
LIMK1 shRNA, is explained.
The third part of results outlines, the accumulation of LIMK1 protein upon loss of the tumor
suppressor Von Hippel- Lindau (VHL) protein. It is demonstrated at the molecular level that the
increase in LIMK1 expression is due to the transcriptional activation of its gene. Interestingly,
preliminary data suggest that, this might be a HIF- independent mechanism. Thus, these observations
suggest that LIMK1 might be a novel downstream effector of VHL, which might participate
in tumor progression or metastasis upon VHL loss.
Finally, the fourth part of results focused on LIMK1-dependent cell cycle regulation. Especially,
we demonstrated a premature exit from M phase upon LIMK1 depletion is described.
Eventhough no timing difference is detected in the first part of mitosis (nuclear envelope breakdown
to anaphase onset), a 20 % increase in spindle misorientation is observed in LIMK1 depleted
cells by using time-lapse microscopy. It is suggested that the premature M phase exit may
be due to alterations of later stages of mitosis.
Altogether, this doctoral study provides new insights regarding endogenous LIMK1 localization
in normal versus cancer cells and tissues, and highlights a novel role of LIMK1 in proper mitosis
progression, which requires additional work to understand the precise molecular mechanisms underlying
this phenotype. These data open new perspectives for a better understanding of LIMK1
contribution to cancer progression and metastasis, notably in terms of mislocalization of LIMK1
in cancer cells, but also, as LIMK1 has been identified as a novel effector of VHL, concerning
the contribution of LIMK-dependent processes upregulation to VHL-associated tumors growth
and metastasis.