|Home | About | Journals | Submit | Contact Us | Français|
The Raf kinase inhibitory protein 1 (RKIP-1) and its orthologs are conserved throughout evolution and widely expressed in eukaryotic organisms. In its non-phosphorylated form RKIP-1 negatively regulates the Raf/MEK/ERK pathway by interfering with the activity of Raf-1. In its phosphorylated state, RKIP-1 dissociates from Raf-1 and inhibits GRK-2, a negative regulator of G-protein coupled receptors (GPCRs). Available data indicate that the phosphorylation of RKIP-1 by PKC can stimulate both the Raf/MEK/ERK and GPCR pathways. RKIP-1 has also been implicated as a negative regulator of the NF-κB pathway. Recent studies have shown that phosphorylated RKIP-1 binds to the centrosomal and kinetochore regions of metaphase chromosomes, where it may be involved in regulating the partitioning of chromosomes and the progression through mitosis. The collective evidence indicates that RKIP-1 regulates the activity and mediates the crosstalk between several important cellular signaling pathways. A variety of ablative interventions suggest that reduced RKIP-1 function may influence metastasis, angiogenesis, resistance to apoptosis, and genome integrity. Attenuation of RKIP-1 may also affect cardiac and neurological functions, spermatogenesis, sperm decapactiation, and reproductive behavior. In this review, the role of RKIP-1 in cellular signaling, and especially its functions revealed using a mouse knockout model, are discussed.
The history of Raf Kinase Inhibitory Protein (RKIP-1) is more than two decades long and goes back to the work of Bernier et al.  who, having isolated the protein based on its ability to bind phospholipids, coined it’s original name: phosphatidylethanolamine-binding protein (PEBP). Later, it was found that RKIP-1 is a widely expressed and highly conserved protein that does not share significant homology with any known protein family. In recent years, there has been an increased interest in RKIP-1 due to the discovery of its ability to influence intracellular signaling cascades, cell cycle regulation, the suppression of metastasis, neurodegenerative processes, the modulation of emotions, and reproduction. This minireview addresses selected aspects of RKIP-1 biology, and in particular its role in intracellular signaling, neurological function and reproduction. Excellent reviews have been published recently on other RKIP-1 functions [2–5].
The RKIP gene family is evolutionarily conserved, and its members can be found across distant species . In humans, in addition to RKIP-1, there is one other actively transcribed family member, hPEBP4 . In the mouse, there are three known active RKIP genes: RKIP-1 which is ubiquitously expressed, RKIP-2, whose expression is limited to testes [8, 9], and mRKIP-4, a homolog of hPEBP4 located on mouse chromosome 19, whose expression is limited to retinal ganglion cells of the eye . The fruit fly genome encodes six different paralogs of RKIP, all of which appear to be actively transcribed . The gene also has been found in nematodes [12, 13], parasites  and plants [15, 16].
Prior to the realization that the mammalian protein is involved in cellular signaling, a number of functions were proposed for RKIP-1. In yeast, the homolog of RKIP-1, Tfs1p, inhibits carboxypeptidase Y, which is a vacuolar enzyme and an inhibitor of yRas GAP. In the nematode, RKIP-1 is present in membranes shed from cell surfaces, suggesting that it has a role in protecting the worm from detection by the host immune system. Drosophila RKIP-1 isoforms  have been localized in the lumen of olfactory hairs and in antennae, where they appear to be involved in odorant binding . In plants, RKIP-1 has a role in shoot growth and flowering .
In higher animals, RKIP-1 has been reported to inhibit serine proteases such as thrombin, neuropsin, and chymotrypsin, despite the fact that it has no apparent homology to any known family of serine protease inhibitors [6, 18]. Ojika  has suggested that RKIP-1 may be the precursor of a hippocampal cholinergic neurostimulatory peptide (HCNP), which has been implicated in acetylcholine synthesis and secretion in the brain [19–21]. Ojika’s hypothesis is based on the observation that the first eleven amino acids of the RKIP-1 protein are identical to those of HCNP, suggesting that HCNP may be processed from RKIP-1. More recent work suggests HCNP may also have an endocrine function that affects the physiology of the heart [22, 23].
Structural and immunohistochemical studies in the late 1990’s enhanced our understanding of RKIP-1 and provided the first clues concerning its mechanism of action at the molecular level. These analyses revealed that RKIP-1 is sometimes localized at extracellular cell surfaces, even though it has no obvious secretion signal . Structural analyses indicated that RKIP-1 has a highly conserved phosphate-binding pocket that may allow it to bind to other phosphoproteins [24–26]. More recently, there has been an unprecedented interest in RKIP-1, largely due to the realization that in mammals its activity seems to be capable of suppressing the metastatic invasion of cancer cells.
It is now well established that RKIP-1 is a member of a large, evolutionarily conserved group of proteins involved in MAP kinase (MAPK) signal transduction. This signaling machinery evolved to rapidly activate nuclear transcription factors in response to extracellular stimuli [5, 27–29], and it can influence diverse cellular functions including cell proliferation, differentiation and apoptosis. MAP kinase pathways are a three component kinase module comprised of a MAP kinase kinase kinase (MKKK) that is able to phosphorylate and activate a MAP kinase kinase (MKK), which in turn phosphorylates and stimulates a MAP kinase (MAPK) . The terminal kinases in the pathway include the extracellular regulated kinases (ERKs) that can stimulate transcription factors such as AP1. MAP kinase signaling can be initiated by a number of stimuli including EGF, TPA and ultraviolet light.
The Raf kinases belong to the family of MAPKKKs . A-Raf, B-Raf, and Raf-1 (or c-Raf) are the three known isoforms of Raf in mammalian cells. While A and B-Raf display a tissue-specific pattern of expression, Raf-1 is widely expressed. The events that lead to Raf activation are only partially understood, but they almost certainly involve multiple phosphorylations and dephosphorylations [31–33] that trigger conformational changes and the exposure of the N-terminal regulatory domain of the protein. Raf has the ability to interact with a large number of proteins, including Hsp90, PP2A and Cdc25 phosphatases, Akt, PKCs, Jac kinases, MAPK ERK5, SUR-8, CNK, KSR and Grb10 adaptors, 14-3-3 adapter protein, and RKIP-1 . This elaborate network of Raf-interacting proteins constitutes a key regulatory element of the MAP kinase cascade .
Of all the Raf-binding proteins, RKIP-1 is the only known inhibitor of this MAP kinase pathway (Fig. 1). Using a two hybrid approach, Yeung et al. demonstrated that RKIP-1 could bind Raf-1 , and additional data from this group indicate that RKIP-1 also binds MEK and ERK. Interestingly, RKIP-1 does not interact with AP-1, and AP-1 activity induced by a constitutively active MEK is not affected by RKIP-1 . Furthermore, it appears that RKIP-1 cannot bind Raf-1 and MEK simultaneously since the binding sites for MEK and Raf-1 overlap on RKIP-1.
Signaling downstream of MEK is attenuated when RKIP-1 blocks the interaction between Raf-1 and MEK . Recently, Eves et al. demonstrated that RKIP-1 indirectly influences the Aurora B kinase and spindle checkpoints, and thus cell cycle, through its action on Raf-1 and the MAP kinase pathway .
The action of RKIP-1 is not limited to a single signaling cascade. Yeung et al. demonstrated that inhibition of RKIP-1 enhances NF-κB-mediated transcription, while over-expression of RKIP-1 reduces it . RKIP-1’s ability to control the transcriptional activity of NF-κB is due in part to its capacity to negatively regulate IKK, an activator of NF-κB transcription. RKIP-1 accomplishes this by controlling the upstream regulators of IKK, namely TAK-1 and NIK, and it has been suggested that TAK-1, NIK, IKKα, and IKKβ form part of the 700-kDa TNFα-induced IKK complex  (Fig. 1). Thus, RKIP-1 can influence NF-κB-regulated processes such as those that mediate the production of cytokines, cytokine receptors, cell adhesion molecules, and apoptotic effectors.
GPCR signaling is yet another intracellular control system that is influenced by this protein (Fig. 1). RKIP-1 controls GPCR activity by interacting with the amino-terminal part of the G-protein coupled receptor kinase-2 (GRK2) and inhibiting its ability to phosphorylate its target [39, 40]. In its active state, GRK2 phosphorylates GPCRs, uncoupling them from their associated G-proteins and marking them for degradation. By blocking the activity of GRK2, RKIP-1 stimulates signaling through the GPCRs and influences processes such as cardiac physiology.
RKIP-1is a phosphoprotein and its one known phosphorylation site (serine 153) has been shown to be a target of PKC . The phosphorylated form of RKIP-1 has been localized to the centrosomal and kinetochore regions of the prometaphase chromosome, where it may be involved in regulating spindle check point proteins and movement through the cell cycle . Protein Kinase C (PKC)-mediated phosphorylation of RKIP-1 decreases RKIP-1’s affinity for Raf-1 and increases its affinity for GRK2 . These data suggest that cell cycle progression and the activity of two biologically important cellular signaling systems are linked by the phosphorylation-dependent activity of the RKIP-1 protein.
Signaling proteins, including those from the MAP kinase superfamily, are often linked to disease states such as cancer . In recent years, RKIP-1 has been identified as a member of a novel class of molecules that suppress the metastatic spread of tumors.
Molecular pathways involved in the detachment, migration of malignant cells from the primary tumor site, and invasive colonization of distant organs are poorly understood. Factors contributing to this process can be classified as those that enhance positively, or regulate negatively the three basic phases of metastasis: metastatic initiation, progression, and virulence. The basic concepts of such metastatic changes have been recently reviewed [43–46]. Here, it is worth noting that the extensive search for genes that influence metastatic processes is a very important part of the overall effort to find cures for cancer. Understandably, genes that promote or suppress metastasis are of great interest for their clinical relevance and hope for pharmaceutical advances.
A growing body of evidence suggests that RKIP-1 is a novel candidate gene on the expanding list of metastasis suppressors (Table I).
The first evidence came from cell lines derived from metastatic prostate cancers, which display decreased levels of RKIP-1 mRNA and protein as compared with primary tumor cell lines . Furthermore, over-expression of RKIP-1 in metastatic cancer cells can decrease their invasiveness. Treatment of breast and prostate cancer cells with chemotherapeutic agents induces RKIP-1 expression and predisposes these cancer cells to apoptotic death . Recently, it has been proposed that the level of RKIP-1 in the blood can be used as a prognostic marker for prostate cancer patients . Consistent with the notion that RKIP-1 is a potent suppressor of metastases, experiments from several laboratories have demonstrated that malignant melanomas , breast cancer lymph node metastases , insulinomas , colorectal cancer  and hepatocarcinoma cells  frequently display a marked decrease in RKIP-1 expression.
It has been suggested that the absence of RKIP-1 may enhance the invasive characteristics and angiogenic properties of prostate cells . By influencing the Raf kinase and NF-κB pathways, RKIP-1 may make some cell types particularly sensitive to apoptotic signals , and it is possible that the absence of RKIP-1 promotes breast cell migration and metastasis by inducing the expression of chemokine receptors [37, 70]. Recent work by Eves et al. also indicates that the absence of RKIP-1 may increase the genetic instability of the cell  while work with hepatoma cells suggests that the absence of RKIP-1 may increase the rate of cell division . Taken together, these data suggest that attenuation of RKIP-1 expression may induce several of the characteristics generally associated with cancerous growth and spread.
Future studies may provide tools to harness RKIP-1’s potential anticancer properties. Drug-induced modulation of RKIP-1 expression may provide a potent means to control metastases. A small molecule, locostatin, has already been shown to abrogate RKIP-1’s ability to inhibit Raf-1 , and it may be possible to develop new drugs that enhance the activity of the RKIP-1 protein. Interventions capable of enhancing RKIP-1 activity would be particularly useful for the control of metastatic cells that display attenuated steady-state levels of RKIP-1.
RKIP-1 has been implicated as a factor in Alzheimer’s disease (AD), the most common form of dementia [72, 73]. AD manifests clinically through a progressive decline in multiple cognitive functions. These include memory impairment, aphasia, apraxia, agnosia and/or the loss of ability to plan and organize routine activities. There appear to be two basic forms of AD: early-onset, which affects individuals younger than 65 years of age, and late onset, which takes its toll on individuals older than 65 .
Early onset, or familial AD, has been linked to mutations in three genes: the amyloid precursor protein (APP) , presenilin 1 (PS1)  and presenilin 2 (PS2) . Autosomal dominant mutations in these genes occur infrequently, but they display extremely high penetrance. Early onset AD accounts for approximately 10% of all cases, and mutations in APP, PS1 and PS2 can be found in ~30% of these patients .
Genes casually involved in late-onset of the disease remain to be identified. A family history of dementia appears to have some prognostic value , while another recognized risk factor is apolipoprotein E (Apoe) [79–81]. However, it is important to note that AD is a multifactorial disease that likely results from the altered expression of many different genes.
A significant decrease in RKIP-1 mRNA levels was documented in brains of patients displaying late-onset AD . The hippocampal cholinergic neurostimulating peptide (HCNP) and nerve growth factor (NGF) are believed to play a role in maintaining elevated levels of choline acetyltransferase in the brain, and choline acetyltransferase is necessary for the development of cholinergic medial septal nuclei . As a precursor to HCNP, RKIP-1 thus emerges as an important component that affects the highly regulated action of neuronal choline acetyltransferase. Cholinergic dysfunction of the brain is known to be associated with the age-related memory loss, one of the symptoms that is particularly characteristic in AD .
Additional evidence from a mouse model supports the involvement of RKIP-1 in AD. Using the AD Tg2576 mouse, a significant correlation was demonstrated between Aβ plaque formation and decreased levels of RKIP-1 in hippocampal formations .
Using a RKIP-1 knockout mouse, we demonstrated that RKIP-1 expression is ubiquitous in the limbic formations including the olfactory neuclei, preoptic area, hypocampus, hypothalamic nuclei, amygdala and neocortices (Fig. 2). We also noted that RKIP-1−/− mice older than four months of age display learning and olfaction deficits . A decline in olfactory perception is a characteristic feature of AD patients , and tests for loss of olfactory discrimination have value for distinguishing between AD patients and depressed, non-AD individuals, whose sense of smell seems to be less affected.
The mammalian olfactory network is elaborate and molecular mechanisms by which RKIP-1 may interact with it remain unclear. Olfaction relies, in part, on seven-transmembrane members of the G-protein coupled receptor (GPCR) superfamily that includes vomeronasal type-1 receptor like (V1R-like) genes [87–89]. There are about 1500 olfactory V1Rs in the mouse and about 900 in humans . The smaller number of V1R genes in humans probably reflects the poorer sense of smell in humans. The other class of sensory GPCRs [88, 91], V2Rs, are more similar to metabotropic glutamate and calcium-sensing receptors, yet all GPCRs share a similar ligand-binding 7TM domain . RKIP-1 is a known molecular regulator of GPCR signaling [39, 40] and, in efforts to better understand mechanisms by which RKIP-1 affects the sense of smell, this pathway offers one attractive venue for future molecular analyses.
The role of RKIP-1 in reproduction is an intriguing and yet to be thoroughly addressed question. In rodents, RKIP-1 and its isoform RKIP-2, are expressed at high levels in the seminiferous tubules, and elongated spermatids [8, 93, 94]. Spermatozoa are highly differentiated cells that require complex and extensive cellular remodeling during spermatogenesis. The acquisition of well-defined protein surface domains begins early during spermiogenesis . Spermiogenetic restructuring takes place mainly in the epididymis of the male but also later, during the capacitation processes that culminate in the female reproductive tract. Lipids and phospholipids are highly organized on the membrane surfaces , and it has been suggested that formation of these domains arises in part from the affinity of the protein components to specific classes of phospholipids and fatty acids [97, 98]. RKIP-1 is a phosphadidylethanolamine-binding protein that is ubiquitous in mammalian sperm , and testicular and epididymal fluids [100, 101]. It has been suggested that RKIP-1 may play a role in membrane biogenesis and in the maintenance of antigen segregation in spermatozoa . RKIP-1 was found to be highly-expressed in seminiferous tubules, steroid-producing Leydig cells, and in elongating spermatid droplets of the rat . In more recent work it was found highly expressed in mouse seminiferous tubules .
However, the specific role of RKIP-1 in reproduction remains far from clear. The mouse model provides some important initial clues to RKIP-1’s function in these processes. We  and others [8, 99] found that the expression of the RKIP-1 in seminiferous tubules of testes is strong, dependent on the stage of spermatogenesis, and that the protein acumulates in the cytoplasmic droplets of sperm (Fig. 3A–E). In RKIP-1 knockout mice, the overall progression of spermatogenesis and fertility of adult RKIP-1−/− animals was found to be close to normal . Mature homozygous mutants are viable, appear normal, and the gene product thus seems to be largely dispensable during embryonic development (Fig. 3F–G). However, one prominent feature of RKIP-1−/− animals is that the reproduction rates in matings between mature RKIP-1−/− males and females are reduced by more than 90%. In addition, there is a decreased reproduction rate for matings between RKIP-1−/− males and RKIP-1+/− females, but normal rates for matings between RKIP-1−/− females and RKIP-1+/− males. This phenotype suggests the presence of fertility defects within the male .
An intriguing hypothesis put forth by Gibbons et al.  has implicated RKIP-1 in sperm capacitation. Mammalian sperm cells released from the male reproductive tract are non-fertilizing and must complete maturation before they gain the ability to fertilize an oocyte. This process is termed capacitation, and it involves a loss of proteins from the sperm surface. In vitro, capacitation can be accomplished simply by incubating sperm cells over a period of time. The process is reversible, and if decapacitating proteins are added to the capacitated sperm, spermatozoa revert to the non-fertilizing state. However, after exposure to external decapacitating proteins, the sperm cells are able to re-capacitate again. The surface proteins that reversibly prevent fertilization are called decapacitation factors (DFs). It is believed that DFs interact with the plasma membrane indirectly through specific receptors. Gibbons et al.  have isolated one such receptor, which, after purification and sequencing, they determined to be RKIP-1. This work provided the first strong evidence in support of a role for RKIP-1 in mammalian fertility. In another study, RKIP-1 has been identified as one of the key decapacitation factors .
Using RKIP-1 knockout mice, we tested the RKIP-1−/− and wild-type sperm cells for their capacitation status . In a chlortetracycline assay, RKIP-1−/− caudal epididymal spermatozoa were found to be over 90% capacitated compared to roughly 50% in the wild-type counterparts. The premature capacitation suggests that RKIP-1 is a modulator of the sperm surface properties important in the process of fertilization. However, RKIP-1−/− and wild-type mice did not differ in their response to X-ray-induced testicular injury and apoptosis. Additional studies are needed to determine whether RKIP-1 acts primarily as a decapacitation factor receptor or as a decapacitating factor. Nevertheless, RKIP-1 may have practical applications as a contraceptive drug if external sources of it can prevent egg fertilization over extended periods of time.
The Raf kinase inhibitory protein has been studied for more than two decades. Given its multifaceted roles within the cell (Fig. 4) and its ability to suppress metastasis, it is not surprising that RKIP-1 has captured the attention of many laboratories over the last several years. Future studies are under way and the genetic, cellular and molecular properties of RKIP-1 should lead to a more complete understanding of its function. The recently established mouse model should play an important role in these analyzes. New models will be needed as well to address the role of the highly similar RKIP-2, that operates exclusively in the mouse testes, or the function of N-terminal HCNP peptide, which is not ablated and may be normally processed in the available knockout mice.
This work was supported by grants 5P42ES013660-02 and 5P20RR015578-07 from the National Institute of Health.
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 citable 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.