MCDB Home > Faculty > Dzwokai "Zach" Ma > Current Research

Since I arrived at UCSB in Febuary of 2004, my research focus has been gradually shifting from studying the mechanism of clathrin-independent internalization and recycling of receptors and channels toward two new projects. The current major project relates to the characterization of an unexpected novel link between membrane trafficking and histone methyltransferase. The other project regards the elucidation of the cellular function of a G-protein signaling regulatory protein, and the mechanism by which this protein controls drug addiction.

(I) Characterization of a novel link between endosomal trafficking and histone methyltransferases.

Introduction:

Covalent modifications of histones regulate the structure and function of chromatin. One such form of histone modification is the methylation of lysine residues within histones H3 and H4 by histone lysine methyltransferase complexes. Thus far, methyltransferase complexes have been found to target five lysines in H3 and one lysine in H4, each of which can potentially be mono-, di-, or tri-methylated. Depending on the site and the extent, these modifications can signal for either the activation or the repression of gene transcription. Emerging evidence suggests an intimate link between abnormal histone methylation and human disease including cancers, psychiatric disorders as well as autoimmune diseases. In mammals, there are at least 10 different histone H3 lysine 4 methyltransferase (H3K4MT) complexes, with six of them belonging to the MLL (mixed lineage leukemia) family. While the MLL complexes contain distinct catalytic subunits, they share some common components, including Ash2L, RbBP5, WDR5, and mDpy-30. Ash2L, RbBP5 and WDR5 form a stable core complex which confers substrate specificity and controls the enzymatic activity of the catalytic subunits. Dpy-30 was originally identified as an essential component of the dosage compensation machinery in C. elegans. However, Dpy-30 mutant males also exhibit growth and development defects indicating a general function for this protein. Subsequent studies have demonstrated that the yeast and mammalian orthologs of Dpy-30, Sdc1 and mDpy-30, respectively, are a common subunit of H3K4MT complexes, and that deletion of Sdc1 from yeast leads to a greatly reduced level of H3K4 trimethylation. Despite being a highly conserved subunit of H3K4MT complexes, the molecular function of mDpy-30 has remained unknown since it was first discovered from C. elegans in 1994. Moreover, whereas a plethora of information has been obtained regarding the machinery that writes and reads H3K4 methylation, little is known about the physiological roles of this modification.

Progress:

We have recently shown that mDpy-30 exhibits unexpected dual nuclear and TGN localization. Depletion of mDpy-30 by siRNA slows the endosome-to-TGN transport of internalized CIMPR and causes an accumulation of these receptors at the endosomes near cell protrusions. The same siRNA treatment did not affect the subcellular distribution of TGN46 or TfnR, two other cargo proteins which cycle between the plasma membrane and the endosomes or the TGN. Interestingly, suppression of either Ash2L or RbBP5, two other H3K4MT accessory subunits primarily found in the nucleus, causes a similar enrichment of CIMPR at cell protrusions, implying the involvement of H3K4MT in the endosomal trafficking of CIMPR. As a further support for a role of H3K4MT in endosomal trafficking, overexpression of mDpy-30, Ash2L or RbBP5 all result in an increased dispersal of CIMPR. In addition to CIMPR, Rab4 and Rab11, two small GTPases regulating the endosomal recycling, are also enriched at the protrusions of mDpy-30 knockdown cells. These observations suggest that mDpy-30 and other H3K4MT subunits regulate the endosomal trafficking, including the endosome-to-TGN transport and protrusion-targeting, of specific cargo proteins. Based on our knowledge, our study represents the first report to document the presence of a histone lysine methyltransferase subunit in the intracellular vesicular transport pathway as well as the function of H3K4MT components in membrane trafficking.

Although the above results indicate a role of H3K4MT in CIMPR trafficking, RbBP5 and Ash2L reside in the nucleus, and unlike mDpy-30 we have not been able to detect their TGN localization despite using multiple antibodies and GFP fusions. To better understand the function of cytoplasmic/TGN mDpy-30, we utilized a combination of immune-purification and mass spectroscopy to identify BIG1, a TGN-localized guanine nucleotide exchange factor (GEF) for the ARF small GTPases, as an mDpy-30 interacting proteins. Subsequent analyses demonstrate that BIG1 directly binds to mDpy-30 and this interaction is essential to recruit mDpy-30 to the TGN. Importantly, knockdown of BIG1 also leads to an enrichment of internalized CIMPR at cell protrusions. This finding is consistent with a recent report by Ishizaki et al. that BIG1 has a role in endosome-to-TGN transport. When combined with the above results, our study shows that mDpy-30 and its interacting proteins, including other H3K4MT components and TGN-localized BIG1, inhibit the targeting of internalized CIMPR to cell protrusions.

Given that the formation of the protrusion is the first step during cell movement and that emerging evidence implicates a function of CIMPR in cell migration, we wonder whether mDpy-30 also regulates the endosomal trafficking of integrins, a family of proteins which play a key role in cell adhesion/migration. For this purpose, we chose β1 integrin since it is widely expressed and has the capacity to control adhesion/migration on many extracellular matrix substrates by forming a functional heterodimer with multiple α integrins. Indeed, knockdown of mDpy-30, its TGN-interacting partner BIG1, or its nuclear interacting partner RbBP5, all lead to an accumulation of internalized β1 integrin at cell protrusions. Moreover, both CIMPR and β1 integrin were found to be enriched in the Rab4- and/or Rab11-positive recycling compartments near the protrusions. We next assessed the role of mDpy-30 and its interacting proteins in adhesion/migration and found that knockdown of mDpy-30 enhances cell adhesion/migration, whereas overexpression of mDpy-30 inhibits it. Furthermore, BIG1 or RbBP5 depletion also causes increased adhesion/migration. Taken together, these data show that mDpy-30 and its interacting proteins form a new class of negative modulators of cell adhesion/migration. We further suspect that these proteins exert their influence, at least partially, via preventing the endosomal recycling of specific cargo proteins to cell protrusions. Our findings thus demonstrate a physiological role of mDpy-30 and its associated proteins, H3K4MT and BIG1, in cell adhesion/migration and reveal a potential novel link among histone methyltransferase, endosomal transport, and cell adhesion/migration.

Future Directions:

The functional coordination among various subcellular compartments is of paramount importance for a eukaryotic cell. Whereas the "Unfolded Protein Response" is well documented as a cellular response communicating ER stress to gene regulation, it is not known whether a similar coordination/communication mechanism exists between other intracellular transport compartments and the nucleus. Our studies above have suggested a potential role of mDpy-30 and its interacting proteins in the modulation of cell adhesion/migration by coordinating endosomal/TGN recycling and gene regulation. Cell adhesion/migration is essential to many physiological and pathological events including The elucidation of novel regulatory mechanisms of cell migration will advance our knowledge regarding the control of cell adhesion/migration and may lead to alternative treatments for human diseases. There are three future aims for this project. Aim1 and Aim2 will focus on dissecting the functions of the cytoplasmic/TGN and the nuclear pools of mDpy-30 in endosomal trafficking and gene transcription, respectively. Aim3 is a long range goal in which we will explore whether and how a cell coordinates the function of two pools of mDpy-30 during cell adhesion/migration.

(II) Elucidation of the cellular function of AGS3, a G-protein regulator, and the mechanism by which AGS3 controls drug addiction (01/2006-present).

Introduction:

Drug abuse places an enormous burden on our society. A major difficulty in the treatment of drug addiction is the prevention of relapse. Many addictive drugs induce long-term changes in intracellular signaling through G protein-coupled receptors in the brain that contribute to the neural maladaptations underlying addiction. Recent studies further demonstrated the intimate link between G protein signaling and drug addiction by identifying a key role of one G protein regulator, Activator of G protein Signaling 3 (AGS3) in recurring cocaine- and alcohol-seeking behavior in animal models of addiction. Whereas biochemical studies have shown that AGS3 functions as a GDI (guanine dissociation inhibitor) of Gαi subunits, the cellular pathways modulated by AGS3 remain largely unknown.

Progress:

Earlier research from others has identified a G-protein signaling cascade as a major mechanism regulating TGN-to-plasma membrane transport at the Golgi. Moreover, a principal Golgi-associated G-protein subunit, Gαi3, is known to bind AGS3. In addition, several published reports suggest that a pool of AGS3 is localized to the Golgi apparatus. These observations prompted us to examine the role of AGS3 in the function of the Golgi apparatus.

We have found that the overexpression of AGS3 in cell cultures alters the surface-to-total protein ratios of a subset of heterologously-expressed plasma membrane receptors and channels in a way independent of their internalization and recycling. Moreover, overexpression and siRNA-mediated knockdown of AGS3 both result in the mislocalization/dispersal of TGN46 or CIMPR, two TGN-associated proteins which cycle between the TGN and PM via endosomes or between the TGN and endosomes. Similar treatments exert less impact on the distributions of the cis- or medial-Golgi marker proteins. Finally, adding a TGN-localization signal to a CD4-derived reporter renders the trafficking of fusion protein sensitive to AGS3. Taken together, these data support a model wherein AGS3 may modulate protein trafficking via the TGN.

Overexpression of Gαi3 on the Golgi membrane has been previously reported by others to retard the secretion of a heparin sulfate proteoglycan. Interestingly, Gαi3 in colon carcinoma cells has also been shown to control autophagic sequestration, a cellular process whereby the cell sequesters and recycles cytosolic constituents in a lysosomal-dependent manner. Given the biochemical interaction between Gαi3 and AGS3, Pattingre et al. have overexpressed AGS3 in human colon cancer HT-29 cells and found that overexpression of AGS3 stimulates autophagy. However, whether endogenously expressed AGS3 plays a role in autophagy remains to be elucidated. We have recently obtained evidence, via our collaboration with Dr. Hilde. Abrahamsen, that endogenous AGS3 indeed functions as a modulator of autophagy. Both overexpression and knockdown of AGS3 induce autophagy. Also, AGS3 phosphorylation is sensitive to the induction of autophagy mediated via the inhibition of the mammalian target of rapamycin (mTOR), suggesting that this modification of AGS3 is likely involved in the regulation of autophagy. Collectively, our study has indicated that autophagy is a cellular pathway modulated by AGS3 and we are currently collaborating with Drs. Stephen Lanier and Joel Blum (MUSC) to investigate the potential role of AGS3 in autophagy using AGS3 null mice. Moreover, a moderately increased AGS3 level observed during protracted cocaine withdrawal coincides with a decreased level of a marker associated with autophagic activity, implying a potential link between the AGS3-mediated autophagy and addiction processes.

To better understand the function of AGS3, we combined immuno-purification and mass spectroscopy to identify a major AGS3-intercating protein, USP9x. USP9x is a substrate-specific de-ubiquitinating enzyme and has been implicated in brain function and synaptic regulation from others' studies. Interestingly, as AGS3 does, a pool of USP9x resides at the Golgi apparatus. We have found that overexpression or knockdown of USP9x also leads to the mislocalization/dispersal of TGN46, similar to the effect seen when AGS3 levels are perturbed. Moreover, as has been demonstrated for AGS3, the level of USP9x is up-regulated within the brain following protracted (3 weeks) withdrawal from repeated cocaine treatment. These data indicate that USP9x and AGS3 may work in concert to regulate the cocaine-induced neuroplasticity.

Future Directions:

Recent studies have provided evidence that several proteins essential for autophagosome formation localize to the Golgi apparatus, implying a possible link between Golgi trafficking and autophagy. This concept would be consistent with our observation that AGS3 affects both autophagy and Golgi-associated membrane trafficking. Since our data so far have indicated a clear role of AGS3 in autophagy in both cell cultures and animals, our studies in the near future will be focused on the following two aspects. First, we will elucidate the function of the AGS3/USP9x interaction and determine whether and how USP9x regulates the autophagic activity. Second, we will collaborate with the research groups of Dr. Steve Lanier and Dr. Karen Szumlinski to investigate whether AGS3-mediated autophagy plays a role in the intensification of cocaine-seeking behavior during abstinence using an animal model of addiction. We will also explore whether USP9x is involved in the above process. The Lanier's group has generated AGS3 null mice and the Szumlinski's group has the expertise in animal behavior study.