Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jul 1.
Published in final edited form as: Cancer Lett. 2023 May 2;565:216207. doi: 10.1016/j.canlet.2023.216207

LIMK2: A Multifaceted Kinase with pleiotropic roles in human physiology and pathologies

Kavita Shah 1,*, Mason Cook 1
PMCID: PMC10316521  NIHMSID: NIHMS1898763  PMID: 37141984

Abstract

LIMK2, a serine-specific kinase, was discovered as an actin dynamics regulating kinase. Emerging studies have shown its pivotal role in numerous human malignancies and neurodevelopmental disorder. Inducible knockdown of LIMK2 fully reverses tumorigenesis, underscoring its potential as a clinical target. However, the molecular mechanisms leading to its upregulation and its deregulated activity in various diseases largely remain unknown. Similarly, LIMK2’s peptide substrate specificity has not been analyzed. This is particularly important for LIMK2, a kinase almost three decades old, as only a handful of its substrates are known to date. As a result, most of LIMK2’s physiological and pathological roles have been assigned to its regulation of actin dynamics via cofilin. This review focuses on LIMK2’s unique catalytic mechanism, substrate specificity and its upstream regulators at transcriptional, post-transcriptional and post-translational stages. Moreover, emerging studies have unveiled a few tumor suppressors and oncogenes as LIMK2’s direct substrates, which in turn have uncovered novel molecular mechanisms by which it plays pleiotropic roles in human physiology and pathologies independent of actin dynamics.

Keywords: LIMK2, Castration-resistant prostate cancer, Actin Dynamics, AURKA

1. Introduction

The human kinome consists of 538 protein kinases (1). Based on kinase domain sequence similarity, they are divided into eight major groups (Figure 1). Among these, Tyrosine Kinase-Like (TKL) Group is the most diverse and recently identified group. The TKL group is found in almost all eukaryotes, including plants, Dictyostelium, and vertebrates, but is absent in yeast kinome. TKL group has eight families including LISK family, which consists of two subfamilies-LIMK (LIM domain Kinase) and TESK (Testis Expressed Serine Kinase). LIM kinase 1 (LIMK1) and LIM kinase 2 (LIMK2) are the only two members of the LIMK family (Figure 1). Thus, in the human kinome phylogenetic tree, LIMK1 and LIMK2 belong to Tyrosine Kinase-Like (TKL) Group, LISK family and LIMK subfamily (TKL:LISK:LIMK) (Figure 1). Importantly, LIMKs are only present in vertebrates, Anopheles and Drosphila, but are absent in Caenorhabditis elegans, yeast, Dictyostelium and plants (2).

Figure 1.

Figure 1.

Classification of LIMK2 in kinome phylogenetic tree. LIMK2 belongs to Tyrosine Kinase-Like (TKL) Group, LISK family and LIMK subfamily (TKL:LISK:LIMK) in the kinome phylogenetic tree. CAMK: Calmodulin/Calcium regulated kinases; AGC: Protein Kinase A, G, and C families; CK1: Casein Kinase 1 aka Cell Kinase 1; CMGC: including cyclin-dependent kinases (CDKs), mitogen-activated protein kinases (MAP kinases), glycogen synthase kinases (GSK) and CDK-like kinases; RGC: Receptor Guanylate Cyclases; STE: Homologs of the yeast STE7, STE11 and STE20 genes; TK: Tyrosine Kinase; MLK: Mixed Lineage Kinases; MLKL: Mixed Lineage Kinase-Like; RAF: Rapidly Accelerated Fibrosarcoma; STKR: Serine Threonine Kinase Receptors; LRRK: Leucine Rich Repeat Kinases; LISK: consists of LIMK and TESK subfamilies; IRAK: IL1 Receptor Associated Kinase; RIPK: Receptor Interacting Protein Kinase; TESK: Testis Expressed Serine Kinase.

Despite the structural homology between LIMK1 and LIMK2, they show different expression and subcellular localization. LIMK1 is predominantly expressed in the skeletal muscles, heart, and brain, whereas LIMK2 shows more ubiquitous expression. While LIMK1 is detected at focal adhesions, LIMK2 is detected predominantly in the cytoplasm with some nuclear localization. Their distinct subcellular localization indicates that LIMK1 and LIMK2 might have both overlapping and unique cellular functions. This review focuses on LIMK2-its discovery, structure, upstream (transcriptional and post-translational) regulation and downstream roles under physiological and pathological conditions.

Discovery of LIMK2-A Unique Serine Kinase

In 1994, Nakamura et al. identified a 647-amino-acid polypeptide harboring a C-terminal kinase domain from human hepatoma HepG2 cDNA library, which they named LIM-kinase (LIMK) (3). A year later, they discovered another family member, which was named LIMK2, and the original LIMK was rechristened as LIMK1 (4). Human LIMK1 and LIMK2 genes were mapped to 7q11.23 and 22q12.2 chromosomes, respectively (3, 4). LIMK1 and LIMK2 have 50% overall identity and 70% identity in the kinase domains..

Although LIMKs belong to tyrosine kinase like (TKL) family, they predominantly phosphorylate serine residues. This is presumably due to their unique catalytic loop sequence. A typical kinase catalytic core is divided into 12 sub-domains (SD): SD I (aka P-loop or glycine rich loop), SD II, SD III (aka αC-helix), SD IV, SD V (hinge region), SD VIA, SD VIB (catalytic loop), SD VII (activation loop or segment), SD VIII (P+l loop), SD IX, SD X, SD XI, and SD XII (5) (Figure 2). The catalytic loop (SD VIB) of tyrosine kinases possesses arginine at third or fifth position of DLRAAN and DLAARN consensus sequences, whereas serine/threonine kinases use lysine or histidine at third or fifth positions of DLKXXN or DLXX(K/H)N consensus sequences (6). However, LIMK family has a DLNSHN consensus motif, which matches neither serine/threonine kinases nor tyrosine kinases (3, 4) (Figure 2, shown in yellow). Thus, LIM kinases were considered dual specificity kinases (2). Despite this classification, most LIMK2 substrates known to date are phosphorylated at serine residues. No LIMK2 substrate has been shown to be phosphorylated at a threonine residue as yet, indicating that this family of kinases could only be serine-specific.

Figure 2.

Figure 2.

Structural features of the kinase domain of LIMK2. Kinase domain of LIMK2 showing the residues which form catalytic spine (C-spine) and regulatory spine (R-spine). We also show various catalytic subdomains of LIMK2, including αC loop, αG loop, αF-αG loop, glycine rich loop, catalytic loop, and partial activation segment. The catalytic loop of LIMK2 contains DLNSHN sequence, which matches neither tyrosine kinases nor Ser/Thr kinases. The R-spine is a hallmark of every active kinase, and consists of four residues, two from the C-lobe and two from the N-lobe. H449 residue is from the HRD motif in the catalytic loop. F391 is from the DFG motif in the activation segment. M380 is the conserved aliphatic residue from the αC-helix, and F470 is an aliphatic aminoacid from the β4-strand. The C-spine consists of a series of hydrophobic residues and is completed after ATP binds. The structure was created using AlphaFold2.

LIMK2 Isoforms in humans and rodents

Currently, three splicing variants of human LIMK2 are known, namely LIMK2a, LIMK2b and LIMK2-1. All isoforms share N-terminal LIM domains, central PDZ and S/P (serine/proline rich) domains and a C-terminal kinase domain, but they differ at their boundaries. Thus, these isoforms have different lengths, expression, and subcellular localization pattern, and may have distinct functions.

LIMK2a encodes two tandem LIM domains, a central PDZ domain, a serine/proline rich (S/P) domain and a C-terminal kinase domain (Figure 2A). LIMK2b is identical to LIMK2a, except LIMK2b possesses a unique 16 amino acid sequence at its N-terminus, followed by a half LIM domain, and then by a full LIM domain. LIMK2-1 has identical N-terminus as LIMK2b as it also has one and half LIM domains and 16 amino acid unique sequence. However, LIMK2-1 lacks a few amino acids in the kinase domain, but contains a C-terminal protein phosphatase 1 inhibitory (PP1i) domain, which makes it the largest protein among LIMK2 isoforms (Figure 3A) (7). In silico analysis indicates that LIMK2-1 is only present in Hominidae primate (8). The authors further confirmed that LIMK2-1 is indeed expressed in human fetal brain and in specific areas of adult brain including adult cortex (8). However, its expression was much lower in adult cerebellum and absent in adult hippocampus.

Figure 3.

Figure 3.

Human and rodent LIMK2 Isoforms. (A) Domain structures of three isoforms of human LIMK2. (B) Domain structures of rodent-specific isoforms of LIMK2. In addition, rodents also express LIMK2a and LIMK2b isoforms.

In rodents, a few other LIMK2 isoforms have been identified in addition to LIMK2a and LIMK2b. In 1995, Nunoue et al. reported LIMK2d isoform in rat brains, which possesses one and half LIM domain and a truncated PDZ domain, but lacks S/P and kinase domains (Figure 3B) (9). Rat LIMK2a, LIMK2b and LIMK2d are expressed in adult cortex, hippocampus and cerebellum. Interestingly, LIMK2a and LIMK2d expression increases upon neuronal differentiation, but LIMK2b levels remain unchanged, indicating that LIMK2 may have an important role in the central nervous system. Another less studied mouse LIMK2 isoform is tLIMK2, a testis-specific isoform that lacks both LIM domains and part of the PDZ domain, but possesses S/P and kinase domains. tLIMK2 plays an important role in spermatogenesis in mice (Figure 3B). As this isoform has not been reported in humans, whether LIMK2 regulates spermatogenesis remains unknown (1012). Ikebe et al also identified a brain-specific Limk2c isoform in mouse brains, which encodes a 6-amino-acid insert within the protein kinase domain, although this isoform was not fully characterized (10). The human orthologs of these rodent isoforms are unknown to date.

Tissue-specific Expression of human LIMK2 isoforms

LIMK2a and LIMK2b are differentially expressed during development and display unique tissue expression. LIMK2a mRNA levels are more abundant in digestive organs (liver, pancreas, colon and stomach) than LIMK2b in both human fetal and adult tissues (13). LIMK2a transcripts were also abundant in fetal lung and brain, whereas in adult lung, both isoforms were equally expressed (14). In contrast, LIMK2b isoform was more prominent in the placenta, kidney and adrenal gland, as compared to LIMK2a transcripts. These results underscore tissue-specific regulation of LIMK2a and LIMK2b expression. As noted before, LIMK2-1 is expressed in human fetal brain and in adult cortex, indicating that LIMK2-1 may be involved in neurodevelopment and cognitive disorders (8).

Subcellular localization of LIMK2 isoforms

LIMK2a and LIMK2b display differential subcellular localization, presumably due to their distinct LIM domains (Figure 3A). While LIMK2a is found in the cytoplasm and the nucleus, LIMK2b is mainly localized in the cytoplasm. The subcellular localization of LIMK2-1 is not well studied. Most studies on LIMK2 did not specify which isoform was under investigation and functional differences between these isoforms remain a still mystery in the LIMK2 field.

Protein structure of LIMK2-Unique Substrate Recognition and Catalysis Mechanisms

LIM is an acronym derived from Lin-11, Isl-1, and Mec-3. LIM domains contain a double zinc-finger motif consisting of (C-X2-C-X16–23-H-X2-C)-X2-(C-X2-C-X16–21-C-X2-H/D/C) sequence. For LIMK2a, first LIM domain (12–65) consists of C-X2-C-X18-H-X2-C-X2-C-X2-C-X16-C-X2-D-X2 (amino acids shown in bold interact with zinc) (Figure 4A). Thus, cysteine, aspartic acid and histidine form two tetrahedral zinc-binding pockets that stabilize LIMK2 structure (15). The second LIM domain (67–125) is made of X5-C-X2-C-X15-H-X2-C-X2-C-X2-C-X19-C-X2-C-X (Figure 4A). Cysteine and histidine coordinate to bind zinc in the second LIM domain.

Figure 4.

Figure 4.

LIMK2 may exhibit a unique substrate recognition mechanism similar to LIMK1. (A) Zn finger binding residues in LIM1 and LIM2 domains of LIMK2a isoform. (B) A typical kinase possesses a relatively deep pocket that binds ATP and the phosphoacceptor loop (shown in red) of the substrate. The kinases typically recognize at least 1–3 residues flanking the site of phosphorylation in this loop, leading to site-specific selectivity. Red star (*) represents site of phosphorylation. (C) LIMK1 mode of substrate selection is based on LIMK1-cofilin co-crystal structure. Since cofilin phosphorylation site is the third residue from its N-terminus, this end pokes into the active site of LIMK1, which is relatively shallow and the residue is phosphorylated. The specificity of LIMK1 towards cofilin arises from a large binding interface, which includes parts of the activation loop and a large αF-αG loop insertion (shown as yellow star, also highlighted in Figure 2).

LIM domains are known to bind DNA, but LIM domains of LIMK have not been shown to bind DNA as yet. Instead, they play significant roles in protein-protein interactions. Interestingly, the LIM domains of LIMK1 bind to its kinase domain and sequesters the kinase activity (16). Such a mechanism has not been shown for LIMK2. Instead, the two LIM domains of LIMK2 are believed to inhibit its nuclear shuttling (17). We have shown that LIMK2 binds Aurora kinase A (AURKA) using its LIM domains (18). Given the differences between LIMK1 and LIMK2, the LIM domains of LIMK2 may have different functional consequences.

PDZ acronym was derived from PSD-95 (a 95 kDa post-synaptic density protein), DLG (the Drosophila melanogaster Discs Large protein), and ZO-1 (the zonula occludens 1 protein), all of which contain this domain. The PDZ domain has six ß-strands and two α-helices and interacts with binding proteins (19). LIMK2 associates with Neurofibromin (NF1) using its PDZ and S/P domains (20).

Although LIMK2 has a canonical protein kinase domain containing 12 conserved subdomains, it is unique in at least two ways: first, as noted before, the catalytic loop of LIMK1/2 has a distinctive consensus sequence (DLNSHN; residues 451–456 in LIMK2a), which matches neither serine/threonine kinases nor tyrosine kinases (Figure 2). Second, the kinase domain of LIMK is unusually shallow, indicating distinctive substrate recognition and catalysis mechanisms, which differ substantially from conventional kinases (21).

Typically, kinase-substrate specificity is governed by several mechanisms including co-localization, direct interactions, phosphoacceptor residue (S, T or Y) and the corresponding flanking residues in the phosphoacceptor loop. Direct or indirect kinase-substrate docking interactions often occur at sites distal to the active site, which may involve the kinase domain, flanking domains or associated scaffolding proteins. These interactions bring the phosphoacceptor loop of the substrate in the correct orientation to get phosphorylated in the active site of the kinase. The second level of specificity resides in the phosphoacceptor loop, which contains the phosphoacceptor site and the flanking residues (Figure 4B). The kinases typically recognize and bind at least 1–3 residues flanking the site of phosphorylation, leading to site-specific selectivity in the phosphoacceptor loop. The third tier of specificity arises from the depth of the catalytic site. Kinases with deep clefts phosphorylate tyrosine residues, and the ones with shallower clefts prefer serine/threonine residues. A few kinases possess another tier of regulation, which allows them to selectively phosphorylate serine, but not threonine. This discriminatory mechanism relies on DFG +1 residue in the activation loop, where a hydrophobic residue such as L, F or M offers steric hindrance to branched side chain of Thr residue, thereby conferring selectivity for serine residue (22).

The mechanism by which LIMK2 recognizes its substrate has not been reported as yet; however, two recent crystal structures of LIMK1 in complex with cofilin (a bonafide substrate of LIMK1 and LIMK2) revealed a few distinctive features (21, 23): First, LIMK1 and cofilin directly interact via a large binding interface, which includes parts of the activation loop and a large αF-αG loop insertion (Figures 2 and 4C) (21, 23). Second, the docking interaction positions the phosphoacceptor residue perpendicular to the catalytic site as compared to a typical canonical substrate binding mode (Figure 4C). The perpendicular position also appears to preclude interactions with the flanking residues in the phosphoacceptor loop. Third, the crystal structures demonstrated that cofilin could rock back and forth around the anchor helix resulting in the phosphoacceptor serine poking in and out towards the ATP γ phosphate to establish a productive orientation for phosphoryl transfer, which the authors referred to as a “rock and poke” mechanism. This mechanism is predicted to be unique to LIMK1, and perhaps for LIMK2. This rock-and-poke mechanism is further expected to allow variable phosphoacceptor positions conferring LIMKs with dual substrate specificity. Accordingly, LIMK1 was shown to phosphorylate both wild-type cofilin and mutant S3Y-cofilin at serine and tyrosine sites, respectively. Interestingly, LIMK1 did not phosphorylate the corresponding threonine residue on mutant cofilin (S3T), which is probably due to the presence of a leucine at activation loop DFG +1 position (Leu 472 in LIMK2a) (Figure 2).

Importantly, recent studies have uncovered many new substrates of LIMK2 namely TWIST1, SPOP, PTEN and NKX3.1 (2427). Unlike cofilin, which has a phosphoacceptor serine residue at third position from the N-terminus (MAS*A), none of the other substrates have phosphoacceptor sites close to the N-terminus, indicating the existence of additional mechanisms by which LIMK2 may selectively choose its substrates. Notably, peptide library screening revealed no substrate preference for LIMK1 (28). LIMK2 peptide specificity has not been analyzed. This can be rationalized based on LIMK1-cofilin structure, which shows that LIMK1 recognition of its substrates depends upon distal residues, which become proximal to the phosphoacceptor site in structured or folded form rather than their position within a linear sequence (Figure 4C) (21). Nevertheless, all the sites identified on TWIST1, SPOP, PTEN and NKX3.1 have serines followed by small residues (Gly/Ala) at +1 position. As LIMK2 displays only 70% sequence similarity with LIMK1 catalytic domain, LIMK2 may have discriminatory residues in the catalytic site allowing it to make non-covalent interactions with the phosphoacceptor loop, similar to canonical kinases. LIMK2 substrate specificity may also come from its other domains, particularly its LIM domains which may recruit substrates and position them for optimal phosphorylation. Future studies are required to uncover these mechanisms.

2. Regulation of LIMK2

Post-transcriptional Regulation of LIMK2 mRNA

Regulation by microRNAs (miR) and long ncRNAs (lncRNA):

MicroRNAs are short non-coding RNAs (ncRNAs) of 22–25 nucleotides length, which are critical regulators of gene expression. miR binds argonaute protein and commonly target sites in the 3′ untranslated region (UTR) of mRNA causing gene silencing through translation repression and/or mRNA decay. Deregulation of miRNA is associated with many diseases including cancer (29). Not surprisingly, many recent studies have identified LIMK2 dysregulation by miRNAs in different cancers.

Wang et al showed that miR-135a downregulates LIMK2 under physiological conditions by binding to its 3’-UTR. However, in bladder cancer patients, a single nucleotide polymorphism (G-to-A allele) in the 3’-UTR of LIMK2 disrupts this binding, causing LIMK2 overexpression and higher clinical grade (30) (Figure 5A).

Figure 5.

Figure 5.

Post-transcriptional Regulation of LIMK2 mRNA. (A) Regulation by miR135a. miR135a degrades LIMK2 mRNA under physiological conditions by binding to its 3’UTR (top panel). A single nucleotide polymorphism in the 3’-UTR of LIMK2 disrupts this binding, causing LIMK2 overexpression in bladder cancer. (B) Similarly, miR-939-5p binds LIMK2 and degrades it. LINC00460, a lncRNA, sequesters miR-939-5p in CRC causing LIMK2 overexpression. (C) NUDT21 increases the length of 3’UTR of LIMK2 by alternative polyadenylation, which results in its degradation. (D) TUG1 lncRNA promotes the binding of EZH2 across the LIMK2b promoter causing LIMK2b upregulation in small lung cancer. p53 activation upon DNA damage also promotes LIMK2b transcription. (E) MED12 binds LIMK2 promoter and inhibits its transcription.

A recent study uncovered the relationship between miR-939-5p and LIMK2 in colorectal cancer (CRC) (30). The authors initially focused on LINC00460, a lncRNA, which displayed positive correlation with both LIMK2 mRNA levels and poor prognosis in CRC patients (Figure 5B). Further unraveling revealed that LINC00460 and miR-939-5p negatively regulate each other, and miR-939-5p targets LIMK2 mRNA. As a result, LINC00460 upregulated LIMK2 by sponging miR-939-5p in CRC cells. The exact binding site of miR-939-5p on LIMK2 transcript was not revealed in the study (31).

Regulation of LIMK2 mRNA by alternative polyadenylation:

Nudix Hydrolase 21 (NUDT21) is a crucial protein involved in alternative polyadenylation (APA). APA plays a fundamental role in gene expression. About 70% mammalian protein-coding genes possess numerous polyA sites, which results in different mRNA isoforms with varying 3’-UTR lengths (32). Usually transcripts with proximal polyA sites and therefore shorter 3′-UTRs are believed to be more stable due to exclusion of regulatory sequences that mediate degradation. Thus, transcripts with shorter 3’-UTR are expected to produce more proteins and vice versa. Nevertheless, this process is highly context-dependent and requires the expression of regulatory microRNAs and RNA-binding proteins (33). A recent study showed that NUDT21 increases the length of 3’-UTR of LIMK2, which results in its degradation (Figure 5C). Thus, NUDT21 knockdown increased LIMK2 levels, which in turn activated Wnt/β-catenin pathway, leading to bladder cancer (34).

Transcriptional Regulation of LIMK2

LIMK2b is upregulated by TUG1:

Taurine-upregulated gene 1 (TUG1) is a 7100 nt long lncRNA. It is overexpressed in many cancers including small cell lung cancer (SCLC) and promotes aggressive oncogenic phenotypes. Notably, TUG1 overexpression positively correlated with higher LIMK2b expression in tissues from SCLC patients. Nevertheless, RNA immunoprecipitation studies did not show a direct interaction between LIMK2 and TUG1. Further analysis revealed that TUG1 promotes the binding of EZH2 across the LIMK2b promoter causing the upregulation of the latter (Figure 5D) (35).

LIMK2b is a direct transcriptional target of p53 upon DNA damage:

Two distinct promoters drive the expression of LIMK2 transcript variants 2a and 2b. Two independent studies showed that LIMK2b is a direct transcriptional target of p53 upon DNA damage (36, 37). P53 is a master orchestrator of DNA damage response, which regulates cell cycle arrest, survival and apoptosis upon genotoxic stress. P53 promotes cell survival, chemoresistance and actin cytosketal reorganization by directly upregulating LIMK2b, but not LIMK2a (Figure 5D) (36, 37).

MED12 inhibits LIMK2 transcription:

MED12 is a member of the multi-protein Mediator complex, which is essential for transcription by RNA polymerase II. The Mediator has four distinct modules namely the head, middle, tail and CDK8 kinase module. While head, middle and tail modules form the core mediator (total 26 subunits), CDK8 module associates reversibly with the core mediator subunit in a context-dependent manner. CDK8 consists of four components (MED12, MED13, CCNC, and CDK8). MED12 is a part of CDK8 kinase module, and is crucial for kinase assembly and function. Independently, MED12 also regulates multiple complex transcriptional programs, including WNT, SHH, Nanog, REST, AKT-mTOR, TGFb, RTK, SOX9 and estrogen pathways (38). Thus, MED12 is a transcriptional hub that controls numerous pathways. MED12 directly binds to the promotor of LIMK2 and inhibits its transcription. In NSCLC cells, the authors showed that LIMK2-mediated cofilin phosphorylation led to multinucleation of the cells and cytokinesis defects. Thus, by downregulating LIMK2, MED12 promoted cytokinesis and eventually tumorigenesis (Figure 5E) (39).

Hypoxia upregulates LIMK2 transcription:

Tumor-hypoxia is associated with increased EMT, cancer stem cell phenotype, metastasis, angiogenesis, metabolic reprogramming, immune evasion, chemoresistance and radioresistance (40). We previously reported that exposing castration-resistant prostate cancer (CRPC) cells to hypoxia increases LIMK2 transcription (24). Similarly, upregulation of HIF1α or HIF2α led to strong increase in LIMK2 transcription in these cells (24). We further observed that castration increased LIMK2 levels in normal mice. As both castration and androgen-deprivation therapy (ADT) increase hypoxia stress in prostate cancer causing CRPC (41), we exposed cells to ADT-mimicking conditions (using charcoal-stripped media), which did not increase LIMK2 levels. As tissue culture cells do not experience hypoxia under normal growth conditions despite androgen depletion, it shows that LIMK2 transcription is positively regulated by hypoxia in cells and in vivo (24).

Post-translational Regulation of LIMK2 protein

Protein kinases are regulated by a plethora of mechanisms including intra- and inter-molecular associations, oligomerization, post-translational modifications, subcellular localization and protein-protein interactions. In this respect, LIMK2 is a conventional kinase, regulated by protein-protein interactions and many upstream regulators and downstream effectors.

Regulation of LIMK2 kinase activity via phosphorylation at T505:

Most kinases autoactivate themselves by autophosphorylating their catalytic loop. However, LIMK2 (and LIMK1) are unable to autophosphorylate their activation loops, and are activated by other upstream kinases. LIMK2 is activated downstream of Rho and CDC42 GTPases (42) (Figure 6A). Rho activates ROCK1/2 kinases, which in turn activate LIMK2 by phosphorylating at T505 in the activation loop. CDC42 activates MRCKa, which too phosphorylates LIMK2 at T505 activating it. We showed that AURKA phosphorylates T505 (along with S283 and T494), which activates LIMK2 as well (18) (Figure 6B). Phospho-T505 stabilizes the N-terminal domain of LIMK2 in the active conformation by forming a salt-bridge with the arginine at DFG + 3 position.

Figure 6.

Figure 6.

Regulation of LIMK2 stability and kinase activity. (A) Rho and CDC42 GTPases activate ROCK and MRCKs respectively, which in turn phosphorylate LIMK2 at T505 causing its activation. LIMK2 phosphorylates Cofilin at Ser3, inactivating it causing actin polymerization, which has far reaching consequences in different tissues. (B) AURKA phosphorylates LIMK2 at three sites increasing its stability and kinase activity. (C) Par-3 directly binds the LIM domain of LIMK2 and inhibits its kinase activity promoting tight junction assembly. Neurofibromin (NF1) binds LIMK2 via its SecPH domain, which prevents LIMK2 phosphorylation by ROCK, thereby inhibiting its activation. LATS1/2 phosphorylates YAP at S127, which sequesters it in the cytoplasm. LATS1 binds LIMK2, which inhibits both their activity, resulting in YAP nuclear translocation.

Regulation of LIMK2 kinase activity via protein-protein interactions:

Par-3, a member of the polarity proteins, inhibits LIMK2 kinase activity by directly binding to its LIM domain, thereby decreasing the pool of phosphorylated cofilin, and promoting tight junction assembly (43) (Figure 6C). NF1 is a neuronal tumor-suppressor protein, that binds the PDZ and S/P domains of LIMK2 using its Sec14-Pleckstrin homology domain (SecPH). NF1 binding prevents LIMK2 phosphorylation by ROCK, thereby inhibiting its activation (Figure 6C) (20). Indeed, Nf1 depletion activates this pathway, regulating actin cytoskeleton reorganization and cell motility (20).

LATS1 (large tumor suppressor), a Ser/Thr kinase, is a central player of Hippo-LATS/Warts pathway that inhibits tumor growth by regulating cell proliferation, growth, and death. LATS1/2 inactivates YAP oncogenic functions by phosphorylating it at S127, which sequesters it in the cytoplasm, thereby suppressing its transcription regulation of cellular genes. A recent study showed that LATS1 binds LIMK2, which inhibits both their activity, resulting in YAP nuclear translocation (44). However, when LATS1 levels are low, LIMK2 is activated, and YAP is retained in the cytoplasm causing trophoblast stem cells (TSCs) differentiation (44) (Figure 6C).

Regulation of LIMK2 subcellular localization:

The LIM domains of LIMK2 are critical for its cytoplasmic localization. Accordingly, their deletion causes LIMK2 to be exclusively nuclear localized. In addition, LIMK2 has a basic amino acid-rich sequence from 491–503 in its kinase domain, which functions as both nuclear and nucleolar signal (17). Phosphorylation of LIMK2 at S283 and/or T494 by PKC inhibits its nuclear import. AURKA was also shown to phosphorylate LIMK2 at these sites, but its consequence on LIMK2’s subcellular localization was not analyzed (18).

Regulation of LIMK2 protein stability:

We uncovered a highly oncogenic loop between AURKA and LIMK2. AURKA phosphorylates LIMK2 at S283, T494 and T505, which increases its stability and its kinase activity (8). In turn, LIMK2 directly binds AURKA predominantly through its LIM domains and prevent its degradation. Thus, LIMK2 stabilizes AURKA and vice versa, triggering a critical reciprocal loop which promotes aggressive phenotypes (18).

Many kinases are known to be regulated by their substrates in a feedback loop. Recent studies have shown that LIMK2 directly phosphorylates TWIST1, SPOP, NKX3.1 and PTEN (2427). While TWIST1 is an oncogene, SPOP, NKX3.1 and PTEN are tumor-suppressors. While LIMK2 positively regulates TWIST1 protein (24), it promotes the degradation of SPOP, NKX3.1 and PTEN (2527). Notably, TWIST1 stabilizes LIMK2 in a feedback loop, whereas SPOP, NKX3.1 and PTEN promote LIMK2’s ubiquitylation. Overall, LIMK2 protein stability is regulated by multiple mechanisms in the cells (Figure 7 and Table 1).

Figure 7.

Figure 7.

Oncogenic Substrates of LIMK2 in CRPC. LIMK2 phosphorylates TWIST1 at four sites and stabilizes it. TWIST1 also stabilizes LIMK2 in turn. LIMK2-TWIST1 loop causes EMT and CSC in CRPC. LIMK2 also directly phosphorylates PTEN, NKX3.1 and SPOP, which degrades them. PTEN, NKX3.1 and SPOP in turn degrade LIMK2.

Table 1.

Oncogenic Substrates of LIMK2: LIMK2 directly phosphorylates TWIST1, SPOP, NKX3.1 and PTEN at multiple sites.

Substrate Role Cancer Phosphorylation Sites Consequence of phosphorylation Consequence on LIMK2
TWIST1 Oncogene CRPC S45, S78, S95, S199 Protein stabilized LIMK2 Stabi ized
SPOP Tumor Suppressor CRPC S59, S171, S226 Protein degraded LIMK2 Degraded
NKX3.1 Tumor Suppressor CRPC S185 Protein degraded LIMK2 Degraded
PTEN Tumor Suppressor CRPC S207, S226, S360, S361, S362 Protein degraded and activity inhibited LIMK2 Degraded

3. Physiological Roles of LIMK2

LIMK2 and Actin Dynamics

The most famous substrate of LIMK2 is cofilin. Cofilin regulates actin dynamics in a concentration-dependent manner (45). At a molar ratio to actin, cofilin binds to F-actin causing actin depolymerization. However, filament severing by cofilin generates free barbed ends, which in turn promotes actin polymerization. Thus, submicromolar concentrations of cofilin generate free barbed ends by severing existing filaments, which is used to nucleate actin polymerization. LIMK2 phosphorylates cofilin at Ser3, which inactivates it, promoting actin polymerization (Figure 6A). LIMK2-mediated actin cytoskeleton has wide-spread consequences in different tissues. Using a LIMK2a knock-out mouse model, Bernard et al showed that LIMK2a is required for platelet functions, including platelet adhesion, aggregation, spreading and thrombus formation (Figure 6A) (46). Interestingly, LIMK2-1 does not phosphorylate cofilin, but inhibits PP1 using its PP1i domain, thereby increasing phosphocofilin level and controlling actin dynamics (8, 9).

While a role of LIMK2 in actin dynamics is extensively studied, recent findings also suggest that it may regulate microtubule dynamics, although whether it is direct or indirect remains unclear (47).

LIMK2 in the brain

The actin cytoskeleton in the neurons play a critical role in their polarized morphology, axonal growth, motility, and in the formation, maturation and plasticity of dendritic spines, which in turn regulates synaptic plasticity (48). A recent study showed that embryos from LIMK2 knockout (KO) mice have reduced number of pyramidal neurons in upper cortical layers due to impaired neural progenitor cells proliferation and migration, indicating that LIMK2-mediated actin reorganization is essential for cortical development in mice (49) (Figure 6A). These phenotypes are similar to those found in, p21-activated kinase 1 (PAK1)-knockout (KO) mice suggesting that at the molecular level, PAK1-dependent activation of LIMK2 drives cortical development by promoting neuronal migration and proliferation of neural progenitor cells in mice (50). In addition, LIMK2 is also important for normal mitotic spindle formation, presumably by regulating AURKA (18). Thus, LIMK2-mediated correct spindle formation may play a significant role in neurogenesis. As the expression level of LIMK2 in subpallium is higher that in cortex, it may also aid in the migration of interneurons through the subpallium. In adult mice, LIMK1 expression is higher in the mice cortex than LIMK2 (14), suggesting that LIMK1 may play a more important role in regulating synaptic plasticity.

4. LIMK2 in Human Pathologies

LIMK2 deregulation occurs in many human diseases including CRPC (23), CRC (30), SCLC (35), bladder cancer (30), lung squamous cell carcinoma (LUSC) (51), triple negative breast cancer (52) and melanoma (53) (Table 2). In cancers, LIMK2 overexpression or hyperactivation promote highly oncogenic pathways including tumorigenesis, metastasis, EMT, drug resistance and cancer stem cell phenotypes (a detailed review is beyond the scope of this review). We recently identified LIMK2 as a potential clinical therapeutic target for CRPC in human patients and in vivo (24). LIMK2 is a disease-specific target that is highly abundant in CRPC patients, but is expressed at much lower levels in other stages of prostate cancer. Importantly, targeting LIMK2 completely reverses tumorigenesis in vivo, emphasizing its potential as a clinical target. Nevertheless, the molecular mechanisms by which LIMK2 promotes aggressive phenotypes remain unknown. Recent studies have identified several cancer-specific substrates of LIMK2, particularly in CRPC, which have uncovered its pleiotropic roles in activating oncogenic signaling cascades (Figure 7 and Table 2). LIMK2 also promotes necrotic neuronal death in CA1 neurons upon status epilepticus (SE). SE can be caused by a single seizure lasting >5 minutes or multiple successive seizures without recovery in between. LIMK2 is overexpressed upon SE, which impairs dynamic-related protein-1 (DRP1)-mediated mitochondrial fission causing mitochondrial elongation and neurotoxicity (54, Table 2). A rare missense variant in the PP1i domain of LIMK2-1 (S668P) has been shown to be associated with Intellectual Disability (ID) (9, Table 2). However, the molecular mechanisms by which it promotes ID remain incompletely understood. Future studies are needed to uncover the critical signaling nodes and pathways.

Table 2:

LIMK2 is known to be associated with many human diseases. G3BP1: G3BP Stress Granule Assembly Factor 1; ESM1: Endothelial Cell Specific Molecule 1.

LIMK2 Disease Consequences Downstream Molecular Mechanisms
Overexpression CRPC Tumorigenesis, Drug Resistance, Poor prognosis Phosphorylation of SPOP, PTEN, NKX3.1, TWIST1
Overexpression Breast cancer including TNBC Tumorigenesis and Metastasis Positively regulates AURKA and SRPK1
Overexpression LUSC Negatively correlates with Immune cell Infiltration Not known
Overexpression SE Necrotic Degeneration upon SE by impairing DRP1-mediated mitochondrial fission
Missense mutation in LIMK2-1 Intellectual Disability (ID) Rare LIMK2 mutation associated with ID Not known
Overexpression due to 3´-UTR SNP Bladder Cancer Correlates with risks of high-grade and highstage bladder cancer Not known
Overexpression due to LINC00460 Colorectal Cancer Correlates with poor prognosis Not known
LIMK2b overexpression due to TUG1 Small Cell Lung Cancer Cell growth and chemoresistance Not known
Overexpression Melanoma Tumorigenesis and Metastasis Via G3BP1-ESM1 pathway

5. Conclusion

LIMK2 is emerging as a key orchestrator in various human diseases including cancer and neurodevelopmental disorders. Although different avenues of LIMK2 regulation have been uncovered, a comprehensive understanding of LIMK2 signaling under physiological and pathological condition still remain largely unclear. Most importantly, a vast majority of studies target LIMK2 in general, and its isoform-specific roles are unknown particularly in humans. Importantly, rodents express specific LIMK2 isoforms including LIMK2d isoform in rat brains (11), brain-specific Limk2c isoform in mice, and tLIMK2 in testis. Thus, the results obtained from LIMK2 knockout mice should be carefully interpreted in the human context, as the latter lack these isoforms. Future studies are expected to uncover these isoform-specific mechanisms to fully exploit the potential of LIMK2 as a drug target.

Highlights.

  • LIMK2 is a predominantly a serine-specific kinase due to its unusual active site.

  • LIMK2 is believed to recognize its substrates by a unique rock and poke mechanism.

  • LIMK2 is deregulated at transcriptional, post-transcriptional and post-translational stages in cancer.

  • LIMK2 promotes EMT and cancer stem cell phenotypes by degrading tumor-suppressors and stabilizing oncogenes by direct phosphorylation.

Acknowledgements

This work was supported by NIH award 1R01-CA237660.

List of abbreviations:

LIMK2

LIMK Kinase-2

AURKA

Aurora Kinase A

CRPC

Castration-resistant prostate cancer

ADT

androgen-deprivation therapy

EMT

Epithelial to mesenchymal transition

Footnotes

Declaration of Interest Statement

The authors declare no conflict of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. (2002) The protein kinase complement of the human genome. Science. 298(5600), 1912–34. [DOI] [PubMed] [Google Scholar]
  • 2.Scott RW, Olson MF. (2007) LIM kinases: Function, regulation and association with human disease. J. Mol. Med. 85, 555–568. [DOI] [PubMed] [Google Scholar]
  • 3.Mizuno K, Okano I, Ohashi K, Nunoue K, Kuma K, Miyata T, Nakamura T. (1994) Identification of a Human Cdna-Encoding a Novel Protein-Kinase with 2 Repeats of the Lim Double Zinc-Finger Motif. Oncogene. 9 (6), 1605–1612. [PubMed] [Google Scholar]
  • 4.Okano I, Hiraoka J, Otera H, Nunoue K, Ohashi K, Iwashita S, Hirai M, Mizuno K. (1995) Identification and characterization of a novel family of serine threonine kinases containing two N-terminal LIM motifs. J Biol Chem. 270 (52), 31321–31330. [DOI] [PubMed] [Google Scholar]
  • 5.Dixit A, Yi L, Gowthaman R, Torkamani A, Schork NJ, Verkhivker GM. (2009) Sequence and structure signatures of cancer mutation hotspots in protein kinases. PLoS One. 4(10), e7485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Toshima J, Ohashi K, Okano I, Nunoue K, Kishioka M, Kuma K, Miyata T, Hirai M, Baba T, Mizuno K. (1995) Identification and characterization of a novel protein kinase, TESK1, specifically expressed in testicular germ cells. J Biol Chem. 270(52), 31331–7. [DOI] [PubMed] [Google Scholar]
  • 7.Vallee B, Cuberos H, Doudeau M, Godin F, Gosset D, Vourc’h P, Andres CR, Benedetti H. (2018) LIMK2-1, a new isoform of human LIMK2, regulates actin cytoskeleton remodeling via a different signaling pathway than that of its two homologs, LIMK2a and LIMK2b. Biochem J. 475 (23), 3745–3761. [DOI] [PubMed] [Google Scholar]
  • 8.Tastet J, Cuberos H, Vallee B, Toutain A, Raynaud M, Marouillat S, Thepault RA, Laumonnier F, Bonnet-Brilhault F, Vourc’h P, Andres CR, Benedetti H. (2019) LIMK2-1 is a Hominidae-Specific Isoform of LIMK2 Expressed in Central Nervous System and Associated with Intellectual Disability. Neuroscience. 399, 199–210. [DOI] [PubMed] [Google Scholar]
  • 9.Nunoue K, Ohashi K, Okano I, Mizuno K. (1995) LIMK-1 and LIMK-2, two members of a LIM motif-containing protein kinase family. Oncogene. 11 (4), 701–10. [PubMed] [Google Scholar]
  • 10.Ikebe C, Ohashi K, Mizuno K. (1998) Identification of testis-specific (Limk2t) and brain-specific (Limk2c) isoforms of mouse LIM-kinase 2 gene transcripts. Biochem Biophys Res Commun. 246 (2), 307–12. [DOI] [PubMed] [Google Scholar]
  • 11.Takahashi H, Koshimizu U, Nakamura T. (1998) A novel transcript encoding truncated LIM kinase 2 is specifically expressed in male germ cells undergoing meiosis. Biochem Biophys Res Commun. 249(1), 138–45. [DOI] [PubMed] [Google Scholar]
  • 12.Takahashi H, Koshimizu U, Miyazaki J, Nakamura T. (2002) Impaired spermatogenic ability of testicular germ cells in mice deficient in the LIM-kinase 2 gene. Dev Biol. 241(2), 259–72. [DOI] [PubMed] [Google Scholar]
  • 13.Mori T, Okano I, Mizuno K, Tohyama M, Wanaka A. (1997) Comparison of tissue distribution of two novel serine/threonine kinase genes containing the LIM motif (LIMK-1 and LIMK-2) in the developing rat. Mol. Brain Res. 45, 247–254. [DOI] [PubMed] [Google Scholar]
  • 14.Ikebe C, Ohashi K, Fujimori T, Bernard O, Noda T, Robertson EJ, Mizuno K. (1997) Mouse LIM-kinase 2 gene: cDNA cloning, genomic organization, and tissue-specific expression of two alternatively initiated transcripts. Genomics. 46, 504–508. [DOI] [PubMed] [Google Scholar]
  • 15.Brown S, Coghill ID, McGrath MJ, Robinson PA. (2001) Role of LIM domains in mediating signaling protein interactions. IUBMB Life. 51(6), 359–64. [DOI] [PubMed] [Google Scholar]
  • 16.Hiraoka J, Okano I, Higuchi O, Yang N, Mizuno K. (1996) Self-association of LIM-kinase 1 mediated by the interaction between an N-terminal LIM domain and a C-terminal kinase domain. Febs Lett. 399(1–2), 117–21. [DOI] [PubMed] [Google Scholar]
  • 17.Goyal P, Pandey D, Behring A, Siess W. (2005) Inhibition of nuclear import of LIMK2 in endothelial cells by protein kinase C-dependent phosphorylation at Ser-283. J Biol Chem. 280(30), 27569–77. [DOI] [PubMed] [Google Scholar]
  • 18.Johnson EO, Chang KH, Ghosh S, Venkatesh C, Giger K, Low PS, Shah K. (2012) LIMK2 is a crucial regulator and effector of Aurora-A-kinase-mediated malignancy. J Cell Sci. 125 (Pt 5), 1204–16. [DOI] [PubMed] [Google Scholar]
  • 19.Harris BZ, Lim WA. (2001) Mechanism and role of PDZ domains in signaling complex assembly. J Cell Sci. 114(18), 3219–31. [DOI] [PubMed] [Google Scholar]
  • 20.Vallée B, Doudeau M, Godin F, Gombault A, Tchalikian A, de Tauzia ML, Bénédetti H. (2012) Nf1 RasGAP inhibition of LIMK2 mediates a new cross-talk between Ras and Rho pathways. PLoS One. 7(10), e47283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Hamill S, Lou HJ, Turk BE, Boggon TJ. (2016) Structural Basis for Noncanonical Substrate Recognition of Cofilin/ADF Proteins by LIM Kinases. Mol Cell. 62(3), 397–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Chen C, Ha BH, Thévenin AF, Lou HJ, Zhang R, Yip KY, Peterson JR, Gerstein M, Kim PM, Filippakopoulos P, Knapp S, Boggon TJ, Turk BE. (2014) Identification of a major determinant for serine-threonine kinase phosphoacceptor specificity. Mol Cell. 53(1), 140–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Salah E, Chatterjee D, Beltrami A, Tumber A, Preuss F, Canning P, Chaikuad A, Knaus P, Knapp S, Bullock AN, Mathea S. (2019) Lessons from LIMK1 enzymology and their impact on inhibitor design. Biochem J. 476(21), 3197–3209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Nikhil K, Chang L, Viccaro K, Jacobsen M, McGuire C, Satapathy SR, Tandiary M, Broman MM, Cresswell G, He YJ, Sandusky GE, Ratliff TL, Chowdhury D, Shah K. (2019) Identification of LIMK2 as a therapeutic target in castration resistant prostate cancer. Cancer Lett. 448, 182–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Nikhil K, Haymour HS, Kamra M, Shah K. (2020) Phosphorylation-Dependent Regulation of SPOP by LIMK2 Promotes Castration Resistant Prostate Cancer. British J Cancer. 124(5), 995–1008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Nikhil K, Kamra M, Raza A, Shah K. (2021) Negative cross talk between LIMK2 and PTEN promotes castration resistant prostate cancer pathogenesis in cells and in vivo. Cancer Lett. 498, 1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Sooreshjani MA, Nikhil K, Kamra M, Nguyen D, Kumar D, Shah K. (2021) LIMK2-NKX3.1 Engagement Promotes Castration Resistant Prostate Cancer. Cancers (Basel). 13(10), 2324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mezna M, Wong AC, Ainger M, Scott RW, Hammonds T, Olson MF. (2012) Development of a high-throughput screening method for LIM kinase 1 using a luciferase-based assay of ATP consumption. J Biomol Screen. (4), 460–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Gebert LFR, MacRae IJ. (2019) Regulation of microRNA function in animals. Nat Rev Mol Cell Biol. 20 (1), 21–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Wang W, Yang C, Nie H, Qiu X, Zhang L, Xiao Y, Zhou W, Zeng Q, Zhang X, Wu Y, Liu J, Ying M. (2019) LIMK2 acts as an oncogene in bladder cancer and its functional SNP in the microRNA-135a binding site affects bladder cancer risk. Int J Cancer. 144(6), 1345–1355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Zhang Y, Liu X, Li Q, Zhang Y. (2019) lncRNA LINC00460 promoted colorectal cancer cells metastasis via miR-939-5p sponging. Cancer Manag Res. 11, 1779–1789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Tian B, Manley JL. (2017) Alternative polyadenylation of mRNA precursors. Nat Rev Mol Cell Biol. 18(1), 18–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Spies N, Burge CB, Bartel DP. (2013) 3′ UTR-isoform choice has limited influence on the stability and translational efficiency of most mRNAs in mouse fibroblasts. Genome Res. 23, 2078–2090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Xiong M, Chen L, Zhou L, Ding Y, Kazobinka G, Chen Z, Hou T. (2019) NUDT21 inhibits bladder cancer progression through ANXA2 and LIMK2 by alternative polyadenylation. Theranostics. 9(24), 7156–7167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Niu Y, Ma F, Huang W, Fang S, Li M, Wei T, Guo L. (2017) Long non-coding RNA TUG1 is involved in cell growth and chemoresistance of small cell lung cancer by regulating LIMK2b via EZH2. Mol Cancer. 16(1), 5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Croft DR, Crighton D, Samuel MS, Lourenco FC, Munro J, Wood J, Bensaad K, Vousden KH, Sansom OJ, Ryan KM, Olson MF. (2011) p53-mediated transcriptional regulation and activation of the actin cytoskeleton regulatory RhoC to LIMK2 signaling pathway promotes cell survival. Cell Res. 21(4), 666–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Hsu FF, Lin TY, Chen JY, Shieh SY. (2010) p53-Mediated transactivation of LIMK2b links actin dynamics to cell cycle checkpoint control. Oncogene. 29(19), 2864–76. [DOI] [PubMed] [Google Scholar]
  • 38.Srivastava S, Kulshreshtha R. (2021) Insights into the regulatory role and clinical relevance of mediator subunit, MED12, in human diseases. J Cell Physiol. 236(5), 3163–3177. [DOI] [PubMed] [Google Scholar]
  • 39.Xu M, Wang F, Li G, Wang X, Fang X, Jin H, Chen Z, Zhang J, Fu L. (2019) MED12 exerts an emerging role in actin-mediated cytokinesis via LIMK2/cofilin pathway in NSCLC. Mol Cancer. 18(1), 93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Shi R, Liao C, Zhang Q. (2021) Hypoxia-Driven Effects in Cancer: Characterization, Mechanisms, and Therapeutic Implications. Cells. 10(3), 678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Shabsigh A, Ghafar MA, de la Taille A, Burchardt M, Kaplan SA, Anastasiadis AG, Buttyan R. (2001) Biomarker analysis demonstrates a hypoxic environment in the castrated rat ventral prostate gland. J Cell Biochem. 81, 437–444. [PubMed] [Google Scholar]
  • 42.Sumi T, Matsumoto K, Takai Y, Nakamura T. (1999) Cofilin phosphorylation and actin cytoskeletal dynamics regulated by Rho- and Cdc42-activated LIM-kinase 2. J Cell Biol 147 (7), 1519–1532. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Chen X, Macara IG. (2006) Par-3 mediates the inhibition of LIM kinase 2 to regulate cofilin phosphorylation and tight junction assembly. J Cell Biol. 172(5), 671–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Basak T, Ain R. (2022) Molecular regulation of trophoblast stem cell self-renewal and giant cell differentiation by the Hippo components YAP and LATS1. Stem Cell Res Ther. 13(1):189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.DesMarais V, Ghosh M, Eddy R, Condeelis J. (2005) Cofilin takes the lead. J Cell Sci. 118 (1), 19. [DOI] [PubMed] [Google Scholar]
  • 46.Antonipillai J, Mittelstaedt K, Rigby S, Bassler N, Bernard O. (2020) LIM kinase 2 (LIMK2) may play an essential role in platelet function. Exp Cell Res. 388(1), 111822. [DOI] [PubMed] [Google Scholar]
  • 47.Po’uha ST, Shum MS, Goebel A, Bernard O, Kavallaris M. (2010) LIM-kinase 2, a regulator of actin dynamics, is involved in mitotic spindle integrity and sensitivity to microtubule-destabilizing drugs. Oncogene. 29, 597–607. [DOI] [PubMed] [Google Scholar]
  • 48.Hotulainen P, Hoogenraad CC. (2010) Actin in dendritic spines: connecting dynamics to function. J. Cell Biol. 189 (4), 619–629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Mao R, Deng R, Wei Y, Han L, Meng Y, Xie W, Jia Z. (2019) LIMK1 and LIMK2 regulate cortical development through affecting neural progenitor cell proliferation and migration. Mol Brain. 12(1), 67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Pan X, Chang X, Leung C, Zhou Z, Cao F, Xie W, Jia Z. (2015) PAK1 regulates cortical development via promoting neuronal migration and progenitor cell proliferation. Mol Brain. 8, 36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Su Y, Xu B, Shen Q, Lei Z, Zhang W, Hu T. (2022) LIMK2 Is a Novel Prognostic Biomarker and Correlates with tumor immune cell infiltration in Lung Squamous Cell Carcinoma. Front Immunol. 13, 788375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Malvi P, Janostiak R, Chava S, Manrai P, Yoon E, Singh K, Harigopal M, Gupta R, Wajapeyee N. (2020) LIMK2 promotes the metastatic progression of triple-negative breast cancer by activating SRPK1. Oncogenesis. 9(8), 77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Malvi P, Reddy DS, Kumar R, Chava S, Burela S, Parajuli K, Zhang X, Wajapeyee N. (2023) LIMK2 promotes melanoma tumor growth and metastasis through G3BP1-ESM1 pathway-mediated apoptosis inhibition. Oncogene. 2023 Mar 16. doi: 10.1038/s41388-023-02658-x. [DOI] [PubMed] [Google Scholar]
  • 54.Kim JE, Ryu HJ, Kim MJ, Kang TC. (2014) LIM kinase-2 induces programmed necrotic neuronal death via dysfunction of DRP1-mediated mitochondrial fission. Cell Death Differ. 21(7), 1036–49. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES