| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 36, 34380-34386, September 5, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
**
From the
Departments of Pharmacology and Cell
Biology, Skirball Institute of Biomolecular Medicine, New York University
School of Medicine, New York, New York 10016, the
Department of Molecular and Cellular Biology,
University of Arizona, Tucson, Arizona 85721, the
¶Department of Molecular Biology and
Biochemistry, Rutgers University, Piscatatway, New Jersey 08854, and the
||Department of Medical Genetics, University of
British Columbia, Vancouver, British Columbia V6T 1Z3, Canada
Received for publication, March 28, 2003 , and in revised form, June 3, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The MTM family is conserved with at least 10 members identified in mammals (1, 2). Sequence analysis and phylogenetic comparisons have led to the division of MTMs into at least four subgroups, of which one lacks phosphatase activity due to an alteration in an essential catalytic residue. Although MTM proteins with catalytic activity were initially thought to function as dual-function protein phosphatases (3, 5, 6), recent evidence indicates that MTMs function primarily as lipid phosphatases to dephosphorylate phosphatidylinositol (PI) (3)P (7). In particular, recombinant MTMs are 20–500,000x more active using PI(3)P as a substrate compared with protein substrates, and overexpression of a substrate-trapping MTM mutant led to a 2-fold increase in PI(3)P levels (7). Similarly, a deletion of the MTM gene in Schizosaccharomyces pombe (Sp-MTM) led to an increase in PI(3)P levels, whereas the overexpression of hMTM1 led to a reduction of PI(3)P levels (8).
PI(3)P is one of four known lipid products of phosphatidylinositol 3-kinase (PI3K). PI(3)P is primarily regulated by class III PI3-kinases (9, 10). Class III PI3-kinases are evolutionarily conserved with the yeast Saccharomyces cerevisiae enzyme Vps34p, the prototypic member of this group (11). In contrast to other phospho-lipid products, PI(3)P is present constitutively in all mammalian cells. Genetic studies of vps34 in yeast and its mammalian homologue have demonstrated critical roles for VPS34 in protein trafficking, endocytosis, pinocytosis, and autophagy (12–14). Formation of PI(3)P on specific endosomal membranes acts to recruit proteins containing FYVE or Phox homology domains that can directly bind PI(3)P and can function as downstream effectors of class III PI3Ks (15–18). PI(3)P levels can also be directly generated by the class II PI3K as well as by the dephosphorylation of the class I PI3-kinase products PI3,4(P)2 and PI3,4,5(P)3 by 4 and 5 PI-phosphatase (9, 10).
MTMs have been proposed to function as negative regulators of PI(3)P levels in a manner analogous to the regulation of PI(3,4,5)P3 and PI(3,4,5)P3 by the PI3 phosphatase PTEN (1, 2). However, although only a single PTEN isoform has been identified, MTMs constitute a large family of PI(3)P phosphatase with at least some family members performing non-redundant functions. The mechanism(s) that accounts for the specific non-redundant functions of different MTMs is unknown. In addition, although PTEN is the only enzyme capable of dephosphorylating the 3 phosphate in PI(3,4,5)P2 and PI(3,4,5)P3, PI(3)P signaling can be terminated by several mechanisms that are independent of MTM function. For example, PI(3)P signaling in yeast can be inhibited by the phosphorylation of PI(3)P at the 5 position by Fab1p, a PI3,5-kinase, as well as by hydrolysis of PI(3)P in the yeast vacuole (19). The multitude of MTMs and alternative mechanisms for modulating PI(3)P levels suggest that MTMs might function in the regulation of only specific pools of PI(3)P in specific subcellular contexts. Presently, a direct link between an MTM and the regulation of specific subcellular pools of PI(3)P or a specific biologic function has not been made.
Several possibilities could explain the specific functions of MTMs and why MTM1 and MTMR2 are not functionally redundant to either each other or to other MTM family members. MTMs could function non-redundantly to regulate PI(3)P if they exhibited unique cell expression patterns or localize to unique subcellular compartments within a given cell. In fact, it has recently been shown that MTM1 and MTM2 exhibit over-lapping but distinct patterns of subcellular localization when overexpressed in a muscle cell line (20). Alternatively, MTMs could dephosphorylate targets other than PI(3)P (5, 8). In addition to dephosphorylating PI(3)P, hMTM1 has been shown to dephosphorylate Vps34 when expressed in S. pombe, and MTMR3 has been shown to dephosphorylate both PI(3,5)P2 and PI(3)P (8). To gain further insight into the function and regulation of MTMs, we explored the roles of MTMs in the nematode Caenorhabditis elegans.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Transgenic Worms—Green fluorescence protein (GFP) transcriptional reporter constructs were generated by first amplifying 2–4 kb of DNA 5' to the predicted ATG start sites using PCR and wild type C. elegans genomic DNA and then cloning this fragment into the vector pPD96.62 (gift of A. Fire) upstream and in-frame to the nuclear localization signal-GFP coding sequences The promoter region amplified for mtm-6 (F53A2.8) corresponded to nucleotides 25,981–29,300 of cosmid F53A2, mtm-3 (T24A11.1) corresponded to nucleotides 20,213–22,695 of T24A11, and mtm-1 (Y110A7A.5) corresponded to nucleotides 33304–34510 of YAC Y110A7A. Extrachromosomal arrays were generated by co-injecting GFP reporter constructs with the lin-15 rescuing plasmid, pSK1 (22), into lin-15(n765ts) animals. MTM-6::GFP full-length (amino acids 1–677) and MTM-1::GFP (amino acids 1–592) fusion proteins were generated by amplifying the complete coding region from the cDNA by PCR and cloning in-frame with the GFP coding sequences in the plasmid pPD95.75 (gift of A. Fire) containing the mtm-6 promoter sequences. The GFP-MTM-6 mutant lacking the GRAM (glucosyltransferase, Rab-like GTPase activators and myotubularins) domain was deleted in the first 38 amino acids (delta 3–38), whereas the GFP-MTM-6 mutant lacking the FYVE domain was truncated at amino acids 596 (amino acids 1–596) using PCR. cDNA clones, YK31h5 and YK114h02 for MTM-6 and MTM-1, respectively, were provided by Y. Kohara (National Institute of Genetics, Mishima, Japan). GFP expression was visualized using an epifluorescence microscopy using the Leica DMRE or by the Zeiss 510 laser scanning confocal microscope.
RNA Inhibition (RNAi)—RNAi was performed by injecting
double-stranded RNA or by the method of RNAi feeding
(23). Coding regions of MTMs
were amplified using PCR and genomic DNA and then were cloned into topo 1
(Invitrogen). For RNAi by injection, RNA was synthesized in vitro and
injected into L4 animals. The F1 progeny were screened. For RNAi by
feeding, the appropriate coding regions of each gene were cloned into the
vector L4440 (gift of A. Fire). HT115(DE3) E. coli containing the
RNAi plasmids were grown overnight, and one drop of bacteria was plated onto
nematode growth media plates containing 6 mM
isopropyl-1-thio-
-D-galactopyranoside and carbenicillin (25
µg/ml) per plate. After drying, L4 stage hermaphrodites were placed onto
each plate and F1 progeny were scored. The region amplified for
mtm-6 corresponded to nucleotides 25,981–29,300 of cosmid
F53A2, mtm-3 corresponded to nucleotides 13658–14397 of T24A11,
mtm-1 corresponded to nucleotides 35648–36121 of YAC Y110A7A,
mtm-5 (ceSBF1) corresponded to nucleotides 38975–39395 of
H28G03.6, cup-10 (mtmr-9) corresponded to nucleotides
38975–39395 of YAC Y39H10A
(1), and class II PI3-kinase
corresponded to nucleotides 11235–11764 of F39B1.
Localization of GFP::2XT10G3.5-FVYE—C. elegans expressing the tandem FYVE domains of T10G3.5 were a gift from F. Muller and have been previously described (24). Expression of GFP::2XT10G3.5-FVYE was induced in transgenic worms by heat shocking worms at 30 °C for 30 min.
Coelomocyte Uptake of GFP—arIs37(pmyo-3::ssGFP) animals express GFP with a signal sequence in the body wall muscles that is secreted into the pseudocoelomic cavity (25). The secreted GFP is endocytosed by coelomocytes and subsequently degraded. Worms with defective endocytotosis accumulate GFP in the pseudocoelomic cavity and exhibit diffuse green fluorescence with only low levels of fluorescence observed in the coelomocytes, whereas wild type worms exhibit bright green intracellular staining of coelomocytes with low levels of fluoresence seen in the pseudocoelomic cavity.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
|
To investigate the function of the MTMs in C elegans, MTM gene activity was reduced by RNAi. mtm-3 (RNAi) animals exhibited a mild egg laying defect, whereas animals injected with RNAi against the other myotubularins appeared unaffected. The egg laying defect observed in mtm-3 (RNAi) animals was not enhanced if one or more of the other MTMs were co-inactivated by RNAi. Simultaneous RNAi of two or more MTMs also failed to elicit obvious phenotypes, aside from that seen with mtm-3 alone. Together, these results suggest that other MTMs do not act redundantly with mtm-3 in normal egg laying. However, these results do not exclude the possibility that MTMs perform some functions that are not seen by RNAi injection due to the failure of RNAi to inactivate all MTM function.
Interestingly, two other MTMs, mtm-6 and mtmr-9, were found in a separate set of experiments to function in endocytosis in C. elegans. Mutations in both genes were identified in a screen to identify genes required for the endocytosis by coelomocytes of GFP secreted from body wall muscles into the body cavity and have been referred to previously as coelomocyte uptake-defective genes (25). Similarly, RNAi of vps34 as well as of mtm-3 have been shown to inhibit celomocyte uptake of GFP using this assay (Ref. 25 and data not shown). These findings indicate that both vps34 and MTM family members play essential roles in fluid phase endocytosis in coelomocytes.
RNAi of C. elegans mtm-6 Rescues Larval Lethality of let-512 Mutants—let-512 encodes the only C. elegans vps34 homologue. let-512 mutants typically die during the L3 to L4 larval stages (24). We tested whether inactivation of MTM function by RNAi could rescue the late larval lethality of let-512 mutants to investigate whether MTMs function as PI(3)P phosphatases. Only inhibition of mtm-6 (F53A2.8) function by RNAi rescued let-512 larval lethality (Fig. 2). We used the let-512(h797) strain KR1440, which is homozygous for dpy-5, let-512, and unc-13 on chromosome I and contains the free duplication sDP2 that covers dpy-5 and let-512. Animals that harbor sDP2 are rescued for the dpy-5 and let-512 phenotypes and are viable, whereas animals that lack sDP2 are Dpy, Let, and Unc die as larvae. Rescue of the lethal phenotype produce animals that are Dpy Unc adults. Therefore we assayed rescue of the let-512 phenotype by the presence of adult Dpy animals. We failed to identify any surviving adult Let-512 Dpy-5 (Unc) animals from either uninjected (0 Dpy/112 non-Dpy) or control injected (0 Dpy/86 non-Dpy) animals. By contrast, inactivation of mtm-6 by RNAi injection allowed let-512 mutants to survive to the adult stage, as evidenced by the presence of adult Dpy Unc animals. From several injected animals, 45 adult Dpy Unc animals and 91 non-Dpy Unc animals were recovered (45 Dpy/91 non-Dpy) (Fig. 2). The rescued let-512 mutants produced embyros that failed to hatch. We also tested four other let-512 alleles and observed similar results, indicating that rescue was specific for the let-512 mutation. Using the same RNAi protocol, the inactivation of mtm-1 or mtm-3, either alone or in combination, did not rescue the lethality of the let-512 mutant (Fig. 1 and data not shown). The inactivation of mtm-6 and a second MTM, either mtm-1 or mtm-3, produced a similar level of rescue to that observed by inactivating mtm-6 alone (Fig. 1). These findings suggest that only mtm-6 functions in the let-512-mediated pathway required for larval development, as only inactivation of mtm-6 rescued lethality.
|
Sequence analysis of the coding region of one of the let-512 mutants rescued by mtm-6 (RNAi), KR890: dpy-5(e61) let-512(h510) unc-13(e450)I; sDp2(I;f), revealed an A to G transition mutation that is predicted to generate a termination codon at residue 168 in the LET-512 protein. As this lesion would produce a truncated LET-512 protein that lacks the PI3-kinase catalytic domain, it likely eliminates all activity and would represent a null allele. This data indicates that LET-512 cannot be the source responsible for generating PI(3)P in the let512 mutants rescued by mtm-6 (RNAi) and would indicate that PI(3)P must be generated by other PI3-kinases in mtm-6(RNAi) let-512 rescued animals.
C. elegans has a class I PI3-kinase, encoded by age-1, and a class II PI3-kinase, encoded by F39B1.1. Class II PI3-kinases can generate PI(3)P both directly by acting on PI and indirectly by generating PI(3,4)P2, which can then be dephosphorylated at the 4 position to generate PI(3)P. To determine whether the F39B1.1 might provide PI3K activity in the let-512 mutants we determined whether rescue of let-512 mutants in worms injected with mtmt-6 (RNAi) is inhibited by co-injecting F39B1.1 (RNAi). Injection of F39B1.1 (RNAi) together with mtm-6 (RNAi) reduced the rescue of let-512 mutants by 80% (8 adult Dpy Let-512 Unc animals and 76 non-Dpy non-Let Unc Dpy) compared with worms injected with mtm-6 (RNAi) alone (45 Dpy Let-512 Unc animals and 91 non-Dpy non-Let Unc Dpy). The failure to rescue was not due to toxicity induced by F39B1.1 (RNAi) because injection of F39B1.1 (RNAi) by itself in control N2 worms did not result in toxicity. Thus, these findings suggest that in the setting of reduced MTM function, a class II PI3K can compensate for the loss of let-512 function.
Expression Pattern of the MTMs—As most mammalian MTMs are expressed ubiquitously (1, 2), the functional differences observed for each MTM is unlikely a consequence of expression pattern differences. We examined the pattern of MTM gene expression in C. elegans using transcriptional GFP reporters. DNA sequences upstream of the predicted ATGs of the three catalytically active MTMs were used to drive expression of a nuclear localization signal-GFP fusion protein, and the GFP expression was visualized by fluoresence microscopy (Fig. 3). We found that each MTM exhibited a relatively specific pattern of expression that overlapped minimally with other MTMs. mtm-6::gfp was expressed predominantly in intestinal cells, whereas mtm-3::gfp was expressed predominantly in neurons in the head with a low level of expression also detected in posterior intestinal cells which is a common artifact for GFP expression vectors. Reporter expression of mtm-1 was seen mostly in a few neurons in the head and in the tail. These observations reveal that each of the phosphatase active MTMs exhibit a unique expression pattern. Interestingly, we observed the highest levels of let-512::gfp expression in the intestine, which corresponds to the tissue with the high levels of mtm-6::gfp expression (Fig. 3). Thus, the rescue of let-512 mutants by mtm-6 (RNAi) likely reflects a rescue of intestinal defect. Although this analysis suggests that these MTMs might have non-overlapping expression patterns, the caveat is that these reporter constructs are not showing all the tissues in which these genes are expressed. This is exemplified by the fact that we know that both MTM-1 and MTM-6 function in coelomocytes, yet we do not see GFP in this tissue using these reporters.
|
Cellular Localization of MTMs—MTMs contain several domains that might function in subcellular targeting of MTMs via either protein-protein or protein-lipid interactions. As such, individual MTMs might be targeted to different subcellular domains and regulate discrete pools of PI(3)P. To determine the subcellular localization of mtm-6, a full-length MTM-6-GFP fusion protein was expressed under the mtm-6 promoter. We observed the MTM-6-GFP fusion protein at high levels in a uniform pattern associated with the apical membrane of the intestinal cell (Fig. 4). By contrast, expression of GFP alone, or a GFP fusion protein containing only the MTM-6 FYVE domain, exhibited a diffuse cytosolic expression in the intestine (Fig. 4). We also expressed a full-length MTM-1-GFP fusion protein in the intestine using the mtm-6 promoter and found that it exhibited a cytosolic localization, which was indistinguishable from GFP alone (Fig. 4). These results indicate that MTM-6 is localized to the apical membrane and that MTMs can have distinct subcellular distributions within the same cell.
|
To ascertain which domain of MTM-6 F53A2.8 is needed for apical localization, we examined the subcellular localization of MTM-6-GFP fusion proteins containing various regions of the MTM-6 protein. MTM-6 contains a putative amino-terminal GRAM domain, a PTP catalytic domain, a set interaction domain (SID), and a carboxyl-terminal FYVE domain. The GRAM domain is a novel 70 amino acid domain located at the amino terminus of several MTMs and has been proposed to act as an intracellular protein or lipid-binding signaling module (26). The SID was identified initially on the basis of its interaction with the SET domain of the Hrx protooncogene (5) and might also mediate protein-protein interactions. Although both MTM-6 and MTM-1 contain putative GRAM and SID domains, only MTM-6 contains a FYVE domain, suggesting that this domain might target MTM-6 to the plasma membrane. However, neither the GRAM nor the FYVE domains are required for apical targeting as MTM-6-GFP fusion proteins lacking either the FYVE or the GRAM domain target to apical membrane (Fig. 4). The role the PTP or SID domains could not be assessed because mutants that lack either the PTP or SID domains are not expressed, presumably because this results in the production of an unstable protein.
mtm-6 Inhibits PI(3)P Generation at the Apical Membrane—If the function of mtm-6 is to dephosphorylate PI(3)P, inhibition of mtm-6 by RNAi should lead to increased levels of PI(3)P at the apical membrane in intestinal cells. Previous work has shown that 2X-FYVE-GFP containing tandem FVYE domains specifically label PI(3)P in vivo (27), and previous experiments had demonstrated that tandem FYVE domains of T10G3.5 specifically label PI(3)P in C. elelgans (24) and that expression of a gfp::2XT10G3.5-FVYE fusion under the control of heat shock promoter specifically labels a number of intracellular vesicles with strongest expression seen in the intestine and hypodermal cells. To assess whether mtm-6 negatively regulates PI(3)P levels in C. elelgans, we determined the subcellular localization of GFP::2XT10G3.5-FVYE in worms treated with mtm-6 (RNAi). As previously reported, GFP::2XT10G3.5-FVYE localized to a number of subcellular vesicles at the periphery of intestinal cells in wild type worms (Fig. 5). Following treatment with mtm-6 (RNAi), a portion of GFP::2XT10G3.5-FVYE was found to re-localize to apical membranes (Fig. 5). The apical membrane staining was not seen in control worms or in worms treated with RNAi to mtm-1 or mtm-3 (Fig. 5 and data not shown). This finding provides direct evidence that endogenous mtm-6 negatively regulates PI(3)P levels and/or localization in vivo.
|
let-512 Coelomocyte Uptake Defect Is Rescued by Inhibiting mtm-1 (RNAi)—As stated above, the coelomocytes (six in hermaphrodites; five in males) are scavenger cells located in the pseudocoelomic cavity that continuously and nonspecifically endocytose fluid from the pseudocoelom (25). We generated let-512; myo-3::ssGFP animals and tested the role of MTMs in let-512-mediated endocytosis. In wild type myo-3::ssGFP animals, secreted GFP is efficiently endocytosed by coelomyctes, producing animals with bright green coelomocytes; whereas in let-512; myo-3::ssGFP animals, GFP accumulates in the pseudocoelom, yielding animals with a diffuse green pseudocoelom and faint green coelomocytes (Fig. 5 and data not shown). In contrast to the knockout results showing that worms mutant for mtm-6 have defective uptake of GFP into coelomocytes,2 mtm-6 (RNAi) did not lead to a similar defect in coelomcyte uptake of GFP (data not shown), presumably due to the failure of mtm-6 (RNAi) to completely inhibit mtm-6 in coelomocytes. In addition, although mtm-6 (RNAi) rescued the larval lethality of let-512 mutants, it failed to rescue the GFP uptake defect (Fig. 6). Because adult animals are easier to score for GFP uptake, to assess whether mtm-1 negatively regulates PI(3)P levels in coelomocytes, mtm-1 (RNAi) was co-injected with mtm-6 (RNAi) into let-512; myo-3::ssGFP animals. Inactivation of both mtm-6 and mtm-1 by RNAi, produced surviving animals that had a high level of green fluorescence in coelomoctyes and a low level in pseudocoelom typical of wild type animals. By contrast, inactivation of mtm-6 and mtm-3 yield surviving animals with high levels of fluorescence in the pseudocoelom, similar to mtm-6(RNAi) animals. These results indicate that like mtm-3, mtm-6, and mtmr-9, mtm-1 also functions in coelomocyte endocytosis, and that either MTM-1 alone, or MTM-6 with MTM-1, anatagonize LET-512 function in endocytosis in coelomocytes.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
We found that the expression pattern as well as the subcellular localization of MTMs are regulated differentially. The ability of MTM-6 to rescue the let-512 lethality correlated with the expression pattern of both let-512 and mtm-6. In contrast to the expression of the other MTMs, both let-512 and mtm-6 were expressed at highest levels in the intestine. Thus, inhibition of mtm-6, but not the other MTMs, would be expected to elevate PI(3)P levels in intestinal cells where let-512 likely plays an important role in intestinal function. At present, we do not know the exact function of PI(3)P in intestinal cells. The finding that large vacuoles accumulate in intestinal cells in let-512 mutants (data not shown) is consistent with the possibility that let-512 is important in feeding and in nutrient acquisition in C. elegans; let-512 could function by either stimulating the transport of acid hydrolyzes to the lysosome or by stimulating the transport and secretion of digestive enzymes into the gut lumen. However, a recent report (24) has shown that let-512 mutants fail to shed their old cuticles and accumulate nuclear envelope leading to an expansion of the perinuclear space. Based on these findings they proposed that the nuclear expansion in let-512 mutants may lead to breakdown of the nuclear pores leading to the loss of nuclear RNA export and protein synthesis in intestinal and hypodermal cells (24).
We also found that MTMs exhibit distinct subcellular targeting in cells; MTM-6 but not MTM-1 targets to the apical membrane in intestinal cells. Our initial expectation was that the FYVE domain, which is present in MTM-6 but not MTM-1, would mediate localization to the apical membrane. However, mutational analysis of MTM-6 indicated that the FYVE and a putative GRAM domain as well as VPS-34-kinase activity (data not shown) are not required for MTM-6 targeting. At present, the functions the FYVE and GRAM domains play in MTM-6 function are not known. One possibility is that these domains may regulate the localization of MTM-6 within apical membrane subdomains such as lipid rafts. In support of this idea, a recent report (28) examining the localization of the FYVE domain-containing proteins HRS1 and EEA1 demonstrated that whereas each protein localized to the same early endosome, they each localized to distinct locations within the same vacuole. We have not yet identified a critical region that mediates the localization of MTM-6 to apical membranes. A recent report (29) has indicated that a conserved region between the GRAM and PTP domain, which was named Rac-induced localization domain, is responsible for targeting MTM1 to Rac-induced ruffles. Although this region is also conserved in C. elegans MTM-6, this region is not responsible for targeting C. elegans MTM-6 to the plasma membrane because an MTM-6 mutant lacking this region still targets to apical intestinal membrane (data not shown). We have been unable to assess the role the PTP or SID domains because mutants that lack either the PTP or SID are not expressed presumably because this results in the production of an unstable protein.
Both FYVE and PX domains have been identified as PI(3)P binding domains, and a number of proteins containing these domains that bind PI(3)P have now been identified (16–18). Under most circumstances, the mechanism whereby VPS34 is activated as well as the specific intracellular pool of PI(3)P that mediates a specific biological function is poorly understood. Using a 2XFYVE-GFP construct to specifically identify PI(3)P pools in vivo, studies in mammalian cells and yeast as well as a recent study in C. elegans has demonstrated that most PI(3)P localizes to early endosomes (27). This has led to a model in which activated Rab5 generated on early endosomal membranes functions to recruit an active VPS34 leading to PI(3)P modification of the endosome (30). At least one function of PI(3)P and GTP-bound Rab5 is to recruit EEA1 to the limiting membrane of the endosome thereby enabling EEA1 to facilitate early endosomal fusion (12, 13). One of the interesting questions that arise from these studies is what function localizing MTM-6 to apical membranes in intestinal cells serves. One function MTM-6 may serve is to rapidly turnover PI(3)P levels at the PM leading to the turning off a PI(3)P signaling event. Alternatively, MTM-6 could function to prevent aberrant localization of PI(3)P at the apical membrane, which may arise from fusion of a PI(3)P-containing vacuole with the PM. The finding that inhibition of mtm-6 rescues the lethality in let-512 mutants argues in favor of mtm-6 turning off a positive signal. Consistent with the idea that mtm-6 regulates PI(3)P levels at the PM, we found that a 2XFYVE-GFP fusion localizes to the PM in intestinal cells when mtm-6 is inhibited by RNAi.
Further progress in understanding the specificity and function of MTMs will depend on identifying the specific pools of PI(3)P that are regulated by each MTM and the specific biological functions that specific PI(3)P pools mediate. Identifying the mechanism whereby MTMs localize to specific subcellular compartments as well as the mechanism whereby MTM phosphatase activity is regulated will likely be critical to this endeavor. However, our studies also indicate that the function regulated by MTMs is more complicated than initially appreciated and are not limited to simply antagonizing vps34 signaling by dephosphorylating PI(3)P. This is supported by the finding that three ceMTMs, including a presumed phosphatase-defective MTM, mtmr-9, play essential roles in coelomocyte endocytosis, a process that also requires vps34. Thus, MTMs resemble protein tyrosine phosphatases and function as both positive and negative regulators of biological processes. The identification of a genetic system in C. elegans to study MTMs should provide a valuable system to determine genetically the specific functions of MTMs as well as the molecules that regulate specific MTMs.
|
FOOTNOTES |
|---|
** To whom correspondence should be addressed. Tel.: 212-263-7458; Fax: 212-263-5711.
1 The abbreviations used are: MTM, myotubularins; PI, phosphatidylinositol;
RNAi, RNA inhibition; PI3K, phosphatidylinositol 3-kinase; SID, set
interaction domain; GRAM, glucosyltransferase, Rab-like
GTPase activators and myotubularins.
2 H. Fares, unpublished observations.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  F. L. Robinson, I. R. Niesman, K. K. Beiswenger, and J. E. Dixon Loss of the inactive myotubularin-related phosphatase Mtmr13 leads to a Charcot-Marie-Tooth 4B2-like peripheral neuropathy in mice PNAS, March 25, 2008; 105(12): 4916 - 4921. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Choudhury, S. Srivastava, Z. Li, K. Ko, M. Albaqumi, K. Narayan, W. A. Coetzee, M. A. Lemmon, and E. Y. Skolnik Specificity of the Myotubularin Family of Phosphatidylinositol-3-phosphatase Is Determined by the PH/GRAM Domain J. Biol. Chem., October 20, 2006; 281(42): 31762 - 31769. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Schaheen, G. Patton, and H. Fares Suppression of the cup-5 mucolipidosis type IV-related lysosomal dysfunction by the inactivation of an ABC transporter in C. elegans Development, October 1, 2006; 133(19): 3939 - 3948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Srivastava, K. Ko, P. Choudhury, Z. Li, A. K. Johnson, V. Nadkarni, D. Unutmaz, W. A. Coetzee, and E. Y. Skolnik Phosphatidylinositol-3 Phosphatase Myotubularin-Related Protein 6 Negatively Regulates CD4 T Cells Mol. Cell. Biol., August 1, 2006; 26(15): 5595 - 5602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bonneick, M. Boentert, P. Berger, S. Atanasoski, N. Mantei, C. Wessig, K. V. Toyka, P. Young, and U. Suter An animal model for Charcot-Marie-Tooth disease type 4B1 Hum. Mol. Genet., December 1, 2005; 14(23): 3685 - 3695. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. L. Robinson and J. E. Dixon The Phosphoinositide-3-phosphatase MTMR2 Associates with MTMR13, a Membrane-associated Pseudophosphatase Also Mutated in Type 4B Charcot-Marie-Tooth Disease J. Biol. Chem., September 9, 2005; 280(36): 31699 - 31707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Srivastava, Z. Li, L. Lin, G. Liu, K. Ko, W. A. Coetzee, and E. Y. Skolnik The Phosphatidylinositol 3-Phosphate Phosphatase Myotubularin- Related Protein 6 (MTMR6) Is a Negative Regulator of the Ca2+-Activated K+ Channel KCa3.1 Mol. Cell. Biol., May 1, 2005; 25(9): 3630 - 3638. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Dang, Z. Li, E. Y. Skolnik, and H. Fares Disease-related Myotubularins Function in Endocytic Traffic in Caenorhabditis elegans Mol. Biol. Cell, January 1, 2004; 15(1): 189 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Chaussade, L. Pirola, S. Bonnafous, F. Blondeau, S. Brenz-Verca, H. Tronchere, F. Portis, S. Rusconi, B. Payrastre, J. Laporte, et al. Expression of Myotubularin by an Adenoviral Vector Demonstrates Its Function as a Phosphatidylinositol 3-Phosphate [PtdIns(3)P] Phosphatase in Muscle Cell Lines: Involvement of PtdIns(3)P in Insulin-Stimulated Glucose Transport Mol. Endocrinol., December 1, 2003; 17(12): 2448 - 2460. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
