TCL/RhoJ membrane localization and nucleotide exchange

TCL/RhoJ is a Cdc42-related Rho GTPase with reported activities in endothelial cell biology and angiogenesis, metastatic melanoma, and corneal epithelial cells; however, less is known about how it is inherently regulated in comparison to its closest homologues TC10 and Cdc42. TCL has an N-terminal extension of 18 amino acids in comparison to Cdc42, but the function of this amino acid sequence has not been elucidated. A truncation mutant lacking the N-terminus (ΔN) was found to alter TCL plasma membrane localization and nucleotide binding, and additional truncation and point mutants mapped the alterations of TCL biochemistry to amino acids 17-20. Interestingly, while the TCL ΔN mutant clearly influenced nucleotide exchange, deletion of the N-terminus from its closest homologue, TC10, did not have a similar effect. Chimeras of TCL and TC10 revealed amino acids 121-129 of TCL contributed to the differences in nucleotide loading. Together, these results identify amino acids within the N-terminus and a loop region distal to the nucleotide binding pocket of TCL capable of allosterically regulating nucleotide exchange and thus influence membrane association of the protein. INTRODUCTION TCL/RhoJ was first characterized about fifteen years ago as a Rho GTPase with closest sequence homology to TC10 and Cdc42 and so was termed TCL for being “TC10-like.” Its primary sequence shares the G-domains found within the Rho-family GTPases that facilitate binding of guanine nucleotides and hydrolysis of the gamma phosphate from guanosine triphosphate (1). Its similarity to Cdc42 indicated at the time of its initial characterization that it had some potential for redundant functionality, particularly since TCL interacts with some of the same downstream effector proteins such as PAK and WASP (1, 2). These initial observations pointed to TCL behaving as a typical Rho family GTPase, with the GDP-loaded form of TCL functioning as the inactive form of the protein, while the GTP-loaded form stimulates downstream signaling by interacting with appropriate actin remodeling effector proteins (3). Early attempts to frame the cellular context of TCL’s function implicated it in vesicle transport and recycling pathways associated with clathrin-mediated endocytosis, with TCL facilitating the return of receptors to the plasma membrane after internalization (4). More recently, TCL expression was shown to be amplified in endothelial cells, potentially through the of influence of the transcription http://www.jbc.org/cgi/doi/10.1074/jbc.M116.750026 The latest version is at JBC Papers in Press. Published on September 22, 2016 as Manuscript M116.750026 Copyright 2016 by The American Society for Biochemistry and Molecular Biology, Inc. by gest on N ovem er 6, 2017 hp://w w w .jb.org/ D ow nladed from TCL/RhoJ membrane localization and nucleotide exchange 2 factor ERG, where it participates in the projection and lumenization of new vessels (5–9). TCL also has a unique role in tumorassociated angiogenesis and in the maintenance of a stable tumor vasculature suggesting TCL may prove to be a reasonable chemotherapeutic target. This notion has been supported using directed delivery of siRNAs against TCL to endothelial cells of nude mice bearing human tumor explants. TCL suppression in these experiments inhibited pro-angiogenic vessel infiltration of tumors and promoted a loss of vessel integrity within the tumor mass (10). TCL expression and activity has also been implicated in metastatic melanomas, conferring both chemo-resistance and enhanced motility to the cancer cells (11, 12). These unique niches of TCL activity suggest it is a strong candidate for directed chemotherapy, with few anticipated off-target effects, particularly as TCL knock out mice show little phenotypic impairments except minor defects in corneal development (10, 13). Little is known about the molecular pathways that facilitate TCL’s function, and except for a few investigations, there is limited evidence regarding how nucleotide binding and hydrolysis may be influenced by GEFs and GAPs (8, 14, 15). In terms of relevant binding partners, TCL has been shown to directly interact with the GIT-PIX protein complex, leading to enhanced stabilization of focal adhesions. The interaction may be required to enhance endothelial tube formation during angiogenesis, as either GIT-PIX or TCL suppression resulted in reduced HUVEC tubule loop formations in vitro. However, it is unclear how the interaction is initiated and whether TCL is required for other proangiogenesis processes (16). To better understand the inherent biochemical properties of TCL that lead to its cellular activation, we sought to perform a more thorough structure/function analysis of the GTPase, partly in the hope of uncovering unique, targetable sites for anti-angiogenesis therapies (10). Our strategy was to pursue obvious primary sequence differences between TCL and its closest relatives within the Cdc42 family of GTPases. One obvious region of divergence is the C-terminal hypervariable domains which target Cdc42 family GTPases to different membrane environments; however, we were intrigued by the additional differences in signaling specificity provided by variations in their N-termini (17). For example, TC10 contains a 14 amino acid Nterminal extension in comparison to Cdc42, and two short GAG and GPG amino acid sequences within the N-terminus help localize TC10 to plasma membrane lipid rafts (18). Another example is provided by the Nterminal polyproline sequences found within the Cdc42 homologue Wrch1. The N-terminal region normally autoinhibits Wrch1, but binding of SH3 domain proteins to the polyproline sequence within the N-terminus permits spontaneous binding of GTP and activation of Wrch (19, 20). TCL differs from the other Cdc42-family GTPases in that it possesses a unique N-terminal extension of 18 amino acids in comparison to Cdc42; however, its function has not been investigated. The results in this paper will show the Nterminus plays an important role in TCL localization, and the deletion or mutation of amino acids 17-20 within the N-terminus causes the protein to redistribute from predominantly a plasma membrane association to intracellular vesicles. Interestingly, we found that effective GTP-loading of TCL also requires the presence of these amino acids, indicating this region allosterically regulates the nucleotide-binding pocket. This effect was not recapitulated with TCL’s closest homologue TC10. Using this difference, we were able to generate a series of TCL/TC10 chimeras to isolate the region of TCL that coordinates with its N-terminus to allosterically regulate nucleotide exchange. Taken together, these results highlight unique attributes of TCL’s primary structure that will aid in further elucidation of its function and provide an important framework to understanding TCL’s cellular biochemistry. by gest on N ovem er 6, 2017 hp://w w w .jb.org/ D ow nladed from TCL/RhoJ membrane localization and nucleotide exchange