Chapter Characterization of CD38 Expressed in Different Cellular Compartments CHAPTER CHARACTERIZATION OF CD38 EXPRESSED IN DIFFERENT CELLULAR COMPARTMENTS Synopsis CD38, a 42-45 kDa type II transmembrane glycoprotein, is a bifunctional ectoenzyme exhibiting both ADP-ribosyl cyclase and cyclic ADP-ribose hydrolase activities. Cellular localization of CD38 is normally found in plasma membrane, yet there are an increasing number of reports showing that localization of CD38 in various subcellular localizations like RER, nuclear envelope, small vesicles and mitochondria. To characterize and investigate the functional role of CD38 in different organelles, CD38 was transiently expressed in different cell compartment such as nucleus, endoplasmic reticulum (ER), mitochondria and plasma membrane. The subcellular localization of the recombinant CD38 with a Myc tag at its C-terminal (CD38-myc) was investigated by immunofluorescence studies. Immunostaining with anti-myc and anti-CD38 antibody separately have affirmed the CD38-myc expression in the respective cellular compartments. This was further supported by co-localization of CD38-myc with mitochondria tracker, MitoTracker Red and ER tracker, DioC6. ADP-ribosyl cyclase assay indicated a relatively high cyclase activity in the mitochondria, which is comparable to the ectocellular CD38, the predominant form of CD38 localized on plasma membrane. The topological study of mitochondria expressed CD38 was investigated using proteinase K treatment. Subcellular fractionated mitochondria from the transfected cells were subjected to protease protection assay and analyzed by subsequent ADP-ribosyl cyclase activity assay and Western blotting. The proteinase K treatment suggested a specific topology of the molecule with the carboxyl catalytic domain protruding into the cytosolic region. 91 Chapter Characterization of CD38 Expressed in Different Cellular Compartments cADPR produced by mitochondrial CD38 elicited a rapid calcium release from the Ca2+ loaded endoplasmic reticulum. This response is sensitive to treatment by 8Bromo-cADPR, antagonist of cADPR. Collectively, the present data has directly demonstrated the expression of functionally active CD38 in mitochondria. Based on the high cyclase activity and specific topology of CD38 and its role in Ca2+-release assay observed from the data, this suggest that mitochondrial CD38 plays a role in cADPR synthesis and may participate in a novel pathway of intracellular Ca2+ signaling. 3.1 Introduction 3.1.1 Topological Paradox of CD38/cADPR/Ca2+ Signaling System The ectocellular localization of CD38 raises two fundamental questions. The first question concerns if and how the two major functions of CD38, i.e., the receptorial properties and the enzymatic nature of CD38, are interrelated. A general conclusion was drafted from the 5th Torino CD38 meeting 2006, based on a large number of functions mediated by CD38 and its homologue, CD157, are found independent of their enyzmatic activities. It is reasonable to assume that the large extracellular domains of ectoenzymes and their association with other molecules can mediate response without the involvement of the catalytic activities (Malavasi et al., 2006). Moreever, it was reported that both molecules are frequently shed from the cell membrane through cleavage or other mechanisms producing soluble forms (Lee et al., 1996). As a result, a single model combining the characteristic of enzyme and receptor was not identified. One interpretation is that the two functions are independent with each other (Malavasi et al., 2006). 92 Chapter Characterization of CD38 Expressed in Different Cellular Compartments The second unresolved question concerns another apparent contradiction of CD38 functions, i.e., ectocellular generation of cADPR by CD38, for which only intracellular, Ca2+- related activities (Figure 3.1) have been identified in many cellular systems (De Flora et al., 1997, Malavasi et al., 2006, 2008; Davis et al., 2008). A related problem is the availability of extracellular NAD+ to the catalytic region localized in the extracellular domain of the molecule situated at the outer surface of CD38+ cells (extracellular region of the plasma membrane). So how can the cADPR produced by CD38, which is localized at the cell surface, exert its known intracellular functions? Several models have been proposed to explain this paradoxical topology. In order to put the model proposed in this study in perspective, a brief discussion on these alternate models is therefore essential. Figure 3.1 The topological paradox of CD38-catalyzed ectocellular formation of cADPR (cADPRE) and intracellular Ca2+-releasing activity of cADPR (cADPRI) on responsive stores (Adapted from Zocchi et al., 1993). cADPRE- extracellular cADPR cADPRI- intracellular cADPR 93 Chapter Characterization of CD38 Expressed in Different Cellular Compartments It was proposed by De Flora’s group that the surface CD38 may itself serve as a transporter to internalize the cADPR (Figure 3.2). This model involves transmembrane juxtaposition of two or four CD38 monomers to generate a catalytically active channel to bring about influx of cADPR to reach cADPRresponsive intracellular Ca2+ stores (Franco et al., 1998). However, the study by da Silva et al. (1998) has shown that there was no direct involvement of ectocellular synthesis of cADPR on the regulation of the cADPR-mediated intracellular Ca2+ signaling in T-lymphocytes and observed no increase of intracellular cADPR when the intact cells were incubated with NAD+. Therefore, the feasibility of this model is still debated and requires more investigation to clarify the paradoxical results. It was proposed that there is a NAD+-dependent two-step process which involved the oligomerization of cell surface CD38 followed by the internalization of CD38 oligomers. This process presents a means of shifting cADPR metabolism from the extracellular cell surface environment to an intracellular localization. It was shown that in the CD38 internalized cells, there was a corresponding increase in cADPR levels as well (Zocchi et al., 1996). Based on this observation, it was concluded that availability of NAD+ to the catalytically active site of the intravascular localized CD38 probably derived from the permeation of NAD+ across the endocytotic CD38-containing vesicles. 94 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Figure 3.2 Alternative mechanism of ectocellular cADPR (cADPRE) in releasing Ca2+ from responsive intracellular stores. Two mechanisms are depicted. (1) Influx of cADPRE to reach the Ca2+ stores on which it (cADPRI) can bind and release Ca2+ via active transportation across membrane by homodimeric CD38. (2) Binding of cADPRE to a cell surface receptor followed by still undefined signal transduction events ultimately resulting in the release of Ca2+ from target intracellular stores (Adapted from Zocchi et al., 1993). 95 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Zocchi et al., (1999) further showed that the effect of CD38-internalizing ligands on intracellular Ca2+ levels involved the following steps. Firstly, an influx of cytosolic NAD+ into the endocytotic vesicles takes place, which is mediated by a then recognized NAD+ transporter, connexin 43 (Cx43) hemichannel, (Bruzzone et al., 2000), that was showed to be able to mediate transmembrane fluxes of a nucleotide in whole cells. Secondly, an intravesicular CD38-catalyzed conversion of NAD+ to cADPR took place, which was finally followed by out pumping of the cyclic nucleotide via nucleoside transporter (Guida et al., 2002) into the cytosol and subsequent release of Ca2+ from thapsigargin-sensitive stores. Bruzzone et al. (2001) further demonstrated that the NAD+ transporter is sensitive to the [Ca2+]i level, showing a low transport of the nucleotide pyridine if [Ca2+]i level is high. Thus restriction of further mobilization of Ca2+ from intracellular stores by cADPR, formed by influx of NAD+, is achieved when cytosolic [Ca2+] levels are sufficiently high to reduce the activity of the NAD+ transporter (Figure 3.3). This system in principle represents a solution for the topological paradox and has been well demonstrated in specific cell types such as astrocytes. However, there are several issues here that require resolution: 1) Connexin 43 hemichannels appear to be open for NAD+ transport only at [Ca2+]i ≈ 100nM, indicating that this system may not operate when the [Ca2+]i is elevated above normal basal levels (Guse, 2005); 2) The NAD+ transporter and nucleoside transporter system seems to be restricted to several cell types; 3) The identity of the factors that initiate the efflux and influx of pyridine nucleotides in vivo remains unclear. Taken together, this mechanism would be a relatively slow and inefficient one for triggering Ca2+ release from intracellular cADPR-sensitive stores. 96 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Following investigation into endocytosis of human CD38 molecule in normal lymphocytes and a number of leukemia- and lymphoma-derived cell lines, Funaro et al. (1998) postulated that internalization might represent an alternative mechanism of intracellular signaling unrelated to its enzymatic properties and the Ca2+-releasing properties of cADPR. The data showed that the dynamic internalization is a much slower process in cellular signaling. The group proposed that instead of serving as the key step in triggering intracellular signaling, the internalization step may represent a negative feedback control mechanism which interrupts signal transduction processes or cell-cell cross-talk mediated by the surface membrane CD38. In agreement with this, Zocchi et al. (1995) reported that self-aggregation and internalization of CD38 in response to NAD+, β-mercaptoethanol and GSH (reduced glutathione), is accompanied by extensive inactivation of its ADP-ribosyl cyclase and NAD+ glycohydrolase activities. This may regulate the activity of protein but if internalized protein suffers from loss of enzymatic activity, then the question becomes: does the internalized CD38 possess sufficient activity to carry out intracellular signaling? 3.1.2 Ubiquitous Expression of CD38 in Different Cellular Compartments A key question is whether CD38 can catalyze cADPR formation at other more favourable locations, such as the endoplasmic reticulum, nucleus and mitochondria. The idea that CD38 can play a role in cADPR formation at any of these locations is indeed very tempting and reasonable one as all these subcellular compartments are in fact in close spatial proximity to target ryanodine receptor (RYR), which is widely recognised to be localized on endoplasmic reticulum. In addition, the NAD+ concentration is much higher intracellularly than extracellularly (Hasmann and Schemainda, 2003; Billington et al., 2008); therefore, even with a lower intracellular 97 Chapter Characterization of CD38 Expressed in Different Cellular Compartments expression of CD38, cADPR synthesis at these sites may contribute significantly to intracellular cADPR concentration. cADPR produced would then be conveniently transported to the RYR in close vicinity and thus trigger the downstream Ca2+ signaling. Moreover, intracellular /organelle-localized CD38 may gain access to the substrate within the organelles and produces metabolites that regulate calcium homeostasis directly within the organelles. Indeed a number of recent reports have shown exciting findings that in addition to being located on the plasma membrane, functional CD38 molecule is found to be associated with the cytosolic fraction, rough endoplasmic reticulum, nuclear membranes, and mitochondrial membrane (Figure 3.4, (Mizuguchi et al., 1995; Yamada et al., 1997; Meszaros et al., 1997; Matsumura et al., 1998; Liang et al., 1999; Adebanjo et al., 1999; Khoo et al., 2000; Brailoiu et al., 2000; Sun et al., 2002; Munshi et al., 2002;Khoo et al., 2002; Sternfeld et al., 2003; Yalcintepe et al., 2005; Sun et al., 2006). Interestingly, in recent findings, ryanodine receptors were found localized in those cellular compartments which coincide with CD38 distribution (Adebanjo et al., 1999; Beutner et al., 2003). It has been shown that CD38 activity increases in response to incubation with retinoic acid results in a manifold increase of intracellular cADPR content in cultured HL-60 cells (Takahashi et al., 1995). In sea urchin eggs the catalytic site of the cyclase faces the interior of the cell (Lee, 1997). The most recent finding reported by Davis and co-workers showed that enzymatic active intracellular ADP-ribosyl cyclase in sea urchin has a role in Ca2+ signaling via production of second messengers (Davis et al., 2008). In T-lymphocytes CD38 was detected ectocellularly and intracellular and both intracellular and extracellular synthesis of cADPR from NAD+ was confirmed (de Silva et al., 1998). The natural substrate for cADPR synthesis, β- 98 Chapter Characterization of CD38 Expressed in Different Cellular Compartments NAD+, is quint-essentially an intracellular nucleotide, and only minute concentrations of NAD+ were detected in extracellular space (De Flora et al., 1996); these NAD+ levels are far below KmNAD of ADPR-cyclase (Takahashi et al., 1995; Lee., 1997). It has demonstrated that NAD+-induced Ca2+ release requires CD38 and that it occurs through the activation of ryanodine-sensitive Ca2+ release channels. Also, evidence was provided through the expression of several mutated CD38 constructs that plasma membrane localization of the cyclase is not required for NAD+-induced Ca2+ release. As a result, Sun and co-workers has demonstrated that a full cytosolic Ca2+ response to NAD+ can be triggered by a solely intracellular CD38 expression (Sun et al., 2002). In view of all these interesting examples of intracellular CD38, the present study was carried to characterize the functional role of specific organelle targetedCD38 in an overexpression system. Mitochondria and ER targeted CD38 were successfully expressed in respective organelle of CD38- cells. Mitochondrialexpressed CD38 further showed significantly high ADP-ribosyl cyclase activity comparable to surface expressed CD38. The isolated mitochondria from the CD38+ cells showed enriched in CD38 amount as compare to whole cell lysate. The enzymatic active mitochondrial-expressed CD38 demonstrated a role in Ca2+ mobilization studies performed in an in vitro system. Upon addition of 8-BromocADPR, the Ca2+ mobilizing response of cADPR, generated from β-NAD+ catalyzed by mitochondrial CD38, was abolished. 99 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Figure 3.3 Topological and functional interactions between Cx43 hemichannels, CD38, and the cADPR transporter (Franco et al., 2001) regulate the intracellular NAD+ and cADPR metabolism at the level of vesicles/cytosol. Permeability of cytosolic NAD+ across nonphosphorylated Cx43 is followed by intravesicular generation of cADPR and by its efflux to the cytosol to reach the target calcium stores. The subsequent increase of [Ca2+]i triggers Ca2+-dependent processes including PKC-mediated phosphorylation of Cx43 hemichannels in the vesicle. This results in the impermeability of Cx43 to cytosolic NAD+ and accordingly in the blockade of further [Ca2+]i increases. This self regulatory loop providing a decreased NAD+ and cADPR metabolism sets the threshold of [Ca2+]i above which Ca2+dependent cytotoxic effects would be switched on (Adapted from Bruzzone et al., 2001). 100 Chapter Characterization of CD38 Expressed in Different Cellular Compartments B 4µM Ryanodine 5µM IP3 Fluorescence 490/535 20µM cADPR 5min Figure 3.21B cADPR-induced Ca2+-release from rat whole brain microsomes measured by Fluo-3 fluorescence. Ca2+-loaded microsomes were prepared as described in Materials & Methods and used at a final protein concentration of 0.5 mg/ml for flourimetry. Addition of 20µM cADPR caused a rapid release of Ca2+ from the loaded microsomes vesicles. Subsequent addition of 4µM ryanodine failed to elicit further Ca2+ flux across the membrane vesicles. Further addition of 5µM IP3 caused a rapid Ca2+ release. This is a representative data out of 2-3 experiments. 134 Chapter Characterization of CD38 Expressed in Different Cellular Compartments C µM IP3 µM Ryanodine 20 µM cADPR 5µM IP3 20µM cADPR Fluorescence 490/535 4µM Ryanodine 5min Figure 3.21C cADPR-induced Ca2+-release from rat whole brain microsomes measured by Fluo-3 fluorescence. Ca2+-loaded microsomes were prepared as described in Materials & Methods and used at a final protein concentration of 0.5 mg/ml for flourimetry. Addition of 4µM Ryanodine caused a rapid release of Ca2+ from the loaded microsomes vesicles. Subsequent addition of 20µM cADPR failed to elicit further Ca2+ flux across the membrane vesicles. Further addition of 5µM IP3 caused a rapid Ca2+ release. This is a representative data out of 2-3 experiments. 135 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Fluorescence 490/535 D 20µM 10µM Control cADPR 5min Figure 3.21D cADPR-induced Ca2+-release from rat whole brain microsomes measured by Fluo-3 fluorescence. Ca2+-loaded microsomes were prepared as described in Materials & Methods and used at a final protein concentration of 0.5 mg/ml for flourimetry. cADPR induced Ca2+ release from microsomal vesicles in a dose-dependent manner. In the control, 20µM heat-treated cADPR (95ºC, 50min) was added which did not cause Ca2+ release. This is a representative data out of 2-3 experiments. 136 Chapter E Characterization of CD38 Expressed in Different Cellular Compartments 100µM 8-Br-cADPR Fluorescence 490/535 CD38+mitochondria incubated with β-NAD+ 5min Figure 3.21E cADPR-induced Ca2+-release from rat whole brain microsomes measured by Fluo-3 fluorescence. No Ca2+ release observed upon the addition of incubated CD38+mitochondria with 2mM β-NAD+ to the system after 1hr incubation of the microsomal vesicles with 100µM 8-Bromo-cADPR (8-Br-cADPR). This is a representative data out of 2-3 experiments. 137 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Fluorescence 490/535 F a- CD38-mitochondria incubated with NAD+ b- CD38+mitochondria incubated with NAD+ 5min Figure 3.21F mitochondrial CD38 catalyzed the conversion of β-NAD+ to cADPR which induces Ca2+ release from the brain microsomes. Vesicles were loaded with Ca2+ as described in Materials & Methods. Prior incubation of isolated CD38+mitochondria with 2mM β-NAD+ (37ºC, 15min) was carried out before introduction into the system. Rapid released of Ca2+ from the loaded vesicles upon addition of the incubated samples (b). CD38-mitochondria isolated from vectortransfected COS-7 cells was subjected to the same treatment and failed to mobilize Ca2+ from the loaded vesicles (a). This is a representative data out of 2-3 experiments. 138 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Fluorescence 490/535 G a- in situ incubation of CD38-mitochondria with NAD+ a b- in situ incubation of CD38+mitochondria with NAD+ b 5min Figure 3.21G in situ mitochondrial CD38 catalyzed the conversion of β-NAD+ to cADPR which induces Ca2+ release from the brain microsomes. Vesicles were loaded with Ca2+ as described in Materials & Methods. Introduction of isolated CD38+mitochondria together with 2mM β-NAD+ into the system results in a slow released of Ca2+ from the loaded vesicles (b). CD38-mitochondria isolated from vector transfected COS-7 cells was subjected to the same treatment and failed to mobilize Ca2+ from the loaded vesicles (a). This is a representative data out of 2-3 experiments. 139 Chapter Characterization of CD38 Expressed in Different Cellular Compartments 3.3 Discussion While CD38 and its isoform, CD157, appear to be primarily ectocellular enzymes i.e., with the catalytic site facing the extracellular environment at the same time possessing the ability to participate in surface membrane signal transduction events, the mechanism by which its produced metabolite, cADPR, is transported into the intracellular medium to exert its calcium mobilizing properties has yet to be established conclusively. The experimental information and the published data gathered in all these years call on a revision of the mechanism of the enzymatic role of CD38. This includes the notion, which needs to be stressed, that CD38 should no longer or not only be considered as an ectoenzyme, but a full-fledged enzyme (Deaglio et al., 2001). This is indeed proven true by several independent research groups that identified the expression of functional intracellular CD38 which is either localized in the cytoplasm i.e, mitochondria (Mizuguchi et al., 1995; Yamada et al., 1997; Liang et al., 1999: Sternfeld et al., 2003), ER (Sun et al., 2002) and nucleus (Adebanjo et al., 1999; Khoo et al., 2000; 2002; Yalcintepe et al., 2005; Aksoy et al., 2006; Trubiani et al., 2007; Orciani et al., 2008). Deaglio et al. (2001) hypothesized that the ectoenzymatic functions of CD38 could be coincidental. It was proposed that while it has evolved into a receptor and an adhesion molecule, CD38 acquired its plasma membrane location. Even so it is known that CD38 has uncommon ability to recirculate from the plasma membrane to the cytoplasm i.e, nucleus. It was postulated that transfer of cytoplasmic pool of CD38 to the plasma membrane would switch off the enzymatic activity and switch on the receptorial function (Figure 3.22). This unusual ability could be a means of modulating the two activities following the principle of economy of a cell. 140 Chapter Characterization of CD38 Expressed in Different Cellular Compartments Figure 3.22 Multifunctionality of the CD38 molecule. CD38 may perform as an enzyme regulating the metabolism of nucleotides and as a receptor ruling a complex signal transduction pathway. There is no apparent relationship between the two activities (Adapted from Deaglio et al., 2001). In line with this view, Munoz et al. (2008) showed during immunologic synapse formation, human CD38 redistributes to the contact area of T cell-antigen-presenting cell (APC) conjugates in an antigen-dependent manner. APC conjugates with human T cells or B cells overexpressed CD38 revealed the presence of distinct pools of CD38; the cell surface CD38 and CD38 localized in recycling endosomes and recruited both pools of CD38 molecules to the T cell/APC contact sites (Munoz et al., 2008). This study also showed that cADPR produced outside the cells does not contribute to the increased [Ca2+]i induced by antigen binding. It was suggested that this may be due to increased surface CD38 by combining with intracellular pool of 141 Chapter Characterization of CD38 Expressed in Different Cellular Compartments CD38 which as a result causes cells to be more excitable and prone to respond to different stimuli by increasing [Ca2+]i . It is interesting to find that the cell surface CD38 and intracellular pool of CD38 work in tandem to bring about complex cellular signaling. Indeed, intracellular CD38 has been gaining much attention with recent publications involving CD38 research. Intracellular CD38 localized in specific organelle compartment may have different characteristics and functional roles in intracellular signaling due to the different level of control i.e, accessibility of substrates in different microenvironment. The present study investigates the intracellular expression of CD38 in a specific locale such as mitochondria, ER as well as nucleus in comparison to CD38 expressed in cell surface. Emerging data and reports have suggested possible mechanism(s) that would depend on the presence of functional intracellular CD38 (Adebanjo et al., 1999; Khoo et al., 2000; Deaglio et al., 2001; Sun et al., 2002; Aksoy et al., 2006; Partida Sanchez et al. 2007). Here in this study, a possible solution to this paradoxical situation was provided by investigating the intracellularly expressed CD38 in different subcellular locations. CD38 was targeted and expressed in the mitochondria and showed that immunoreactive CD38 could be localized in both plasma membrane and mitochondria compartments using cell lines. Functionality of the expressed mitochondria-targeted CD38 was further assessed in terms of enzymatic activity and shown indeed to possess full cyclase activity (Figure 3.8), with the ability to trigger Ca2+ release through cADPR generation (section 3.2.6). Having shown that mitochondria CD38 was successfully targeted and expressed in mitochondria by immunofluorescence study, the present study also confirmed mitochondria targeted CD38 resides in the outer mitochondrial membrane 142 Chapter Characterization of CD38 Expressed in Different Cellular Compartments as well as the specificity of this targeting system by employing both immunofluorescence and subcellular fractionation studies. Under the microscope, the typical long, thread-like pattern was not observed, instead mitochondria were seen as dicrete bodies in the peripheral cytoplasm in ovoid shape (Figure 3.9B, 3.11 and 3.12). This observation is not unprecedented. Posakony et al. (1975) showed by DAB staining the broad spectrum of mitochondrial shapes observed in HeLa cells. This observation underscores the complex and dynamic character of the structure of mitochondria. Their finding of dissimilar forms as ovoid and branched filaments support the concept that functional mitochondrial membrane may exist in a wide variety of configurations in cells (Posakony et al., 1975). Such structural diversity has also been observed in the electron-microscopic analyses of serial thin sections of rat liver cells (Brandt et al., 1974) and rat heart fibres (Berger et al., 1974), which have revealed rod-shaped, ovoid, V-shaped, diskoid, and filamentous organelles, as well as more complex branched forms. Similar patterns were observed in vector-transfected mitochondria (data not shown). This observation suggests that localization of CD38 on mitochondria does not affect the morphology of mitochondria. All targeted proteins were expressed and can be detected by CD38 antibodies under denaturing condition via Western blot. The molecular weight of the molecules were determined and verified by running alongside with a purified rat liver CD38 protein which would accurately show a 45kDa molecular weight (data not shown). However, the comparison of the immunofluorescence staining between differentially targeted CD38 was unable to reveal the expected staining pattern that symbolized nucleus localization of CD38 (data not shown). As for CD38 expressed in ER, it did not achieve a near 100% co-localization with specific ER marker as compared to close 100% merge between CD38 localized in mitochondria with MitoTracker Red. 143 Chapter Characterization of CD38 Expressed in Different Cellular Compartments The ADP-Ribosyl cyclase activities for each specific organelle targeted CD38 was further examined and low cyclase activities attained for targeted CD38 in both ER and nucleus was observed compared with targeted CD38 in mitochondria and plasma membrane. The ADP-Ribosyl cyclase activity of mitochondria targeted CD38 was comparable to its native form expressed in plasma membrane, indicating good insertion and folding of the protein to give a native, functional molecule. Lower cyclase activities observed in ER and nucleus targeted CD38 could be a reason of improper protein folding. It is also well understood that CD38 is a transmembrane glycoprotein with a type II orientation (i.e, it has a single transmembrane domain with its NH2 terminus in the cytoplasm and its COOH terminus protruding outside the cell). Based on up to date findings, most intracellular CD38 reported in specific cell compartments are membrane associated, it is therefore of interest to find out how all these intracellular membrane associated CD38 are related to the cell surface CD38. A full-fledged CD38 protein was then employed in the present study (retaining the N-terminal sequence and the transmembrane domain) for generating Mito-CD38, ER-CD38 and NucCD38 constructs. To ensure a proper protein expression and thus generate functional nucleus and mitochondrial CD38, removal of the N-terminal sequence and the transmembrane domain and factor in the specific organelle targeting sequence in constructing the plasmid was required. However, when full fledged CD38 is used, the retaining Nterminal sequence and the transmembrane domain could serve as the ER targeting signal, and thus predominate over the nuclear localization signal as well as the passive cytoplasmic localization. Hence proper localization of nucleus targeted CD38 could not be observed. 144 Chapter Characterization of CD38 Expressed in Different Cellular Compartments ER is in membranous continuity with the nuclear envelope, due to the high protein expression in this system, ER targeted CD38 staining in nuclear envelope was observed (Figures 3.9 C, 3.11). It has been shown that misfolded proteins might not enter the secretory pathway and are retained in the ER. The formation of perinuclear cluster (Figure 3.9C) is likely caused by high levels of protein overexpression. Eukaryotic cells use the cytoplasmic proteosome to degrade misfolded secretory glycoproteins, implying a retrograde translocation of glycoproteins from the lumen of ER to the cytosol (Sommer et al., 1993; Hiller et al., 1996; McCracken et al., 1996). It was also reported that the aberrant folded proteins resulting from the overexpression system must be retrogradely transported from the ER into the cytosol so as to become accessible to degrading proteosomes (Jarosch et al., 2002). The movement of the misfolded proteins into cytosol may explain the non-specific ER targeted CD38 staining observed in the co-localization studies compared with ER marker staining (Figure 3.11). This may explain the observation of extensive cytosolic staining of ER targeted CD38. In addition, Umar et al. (1996) demonstrated that all-trans-retinoic acid (RA) induced CD38 protein in HL-60 cells undergoes posttranslational modification into a high molecular weight form (crosslink form of CD38, p190). This post-translational modification of CD38 resulted in at least threefold increase in ADP-ribosyl cyclase activity as well as a 50% decrease in cADPR hydrolase activity as compared to ~45kDa CD38. The data obtained here did not indicate the presence of CD38 dimeric/multimeric complexes in either CD38+ COS-7 cells or mouse brain samples (Chapter 4) on Western blot. This may be due to more drastic SDS-PAGE conditions used here, which could potentially affect the disulphide linkages and thus dissociate the dimers (Bruzzone et al., 1998). On the other hand, it is known that extensive self- 145 Chapter Characterization of CD38 Expressed in Different Cellular Compartments aggregation and inactivation of CD38 can take place following exposure to βmercaptoethanol (Franco et al., 1994; Zocchi et al., 1995; Guida et al., 1995). This may not be applicable here as large molecular structure > 45kDa form was not observed. β-mercaptoethanol was not employed in both ADP-ribosyl cyclase activity assay and immunofluorescence staining (Methods and Materials). It is interesting to note that immunoreactive CD38 expressed on mitochondria showed comparatively high ADP-ribosyl cyclase activity compared to plasma membrane CD38, possibly implying functionality. The results showed that the enzymatic site is facing the cytosolic site which is unprecedented as this group has previously shown that localization of CD38 on the nucleus envelope with its catalytic site facing the nucleoplasm instead of nuclear lumen (Khoo et al., 2000). This served as the starting point for further investigation using mitochondria isolated from brain tissues, as detailed in the following chapter. The initial intention of this study was to target CD38 to specific locations to investigate the functionality of the protein in each specific locale. Following the expression profile showed in the data, expression of CD38 in location such as ER and nucleus may not be as straightforward. Indeed, immunoreactive CD38 was reportedly present in ER and nucleus; in each instance it has demonstrated a role in intracellular calcium signaling (Adebanjo et al., 1999; Khoo et al., 2000; Sun et al., 2002). It is interesting that this system showed the expected localization of CD38, which was targeted to mitochondria, while its topology with the C terminal on the cytosolic side of the outer mitochondrial membrane was unexpected. The localization process for CD38 and the means by which this protein is sorted and targeted to specific organelles remain poorly understood. 146 Chapter Characterization of CD38 Expressed in Different Cellular Compartments By examining the structure and sequence of CD38 protein, Sun et al. (2002) proposed that the NH2 termini of the CD38 proteins from rat, human and rabbit mimic the chimeric signals of cytochrome P450 which contain cryptic mitochondrial targeting signals. It was postulated that targeting of endogenous CD38 to mitochondria may follow the mitochondrial targeting pathway of cytochcrome P450 that needs its cryptic signal fully exposed by process like proteinase-mediated protein cleavage (Sun et al., 2002). However, the present study showed immunoblot results with single band at ~45kDa for CD38 expressed in mitochondria which is similar to the CD38 expressed in plasma membrane. There was no observation of CD38 protein band lower than 45kDa in the current study, given that to expose the cryptic sequence, protein has to go through enzymatic cleavage which will result in smaller molecular weight. A 45kDa fully glycosylated form of CD38, which is identical to that for glycosylated surface membrane CD38, was observed on isolated mitochondria from Mito-CD38 transfected cells as well as isolated mitochondria from brain tissues. Deglycosylation studies using EndoH (endoglycosidase H) showed similar glycosylated patterns in both mitochondrial CD38 and microsomal CD38 (data not shown). Given that CD38 carries glycosylation sites in the extracellular domain; it appears as fully N-glycosylated moiety as ~45kDa form. Removal of these glycosyl moieties completely from the molecule results in a ~33 kDa form (Koguma et al., 1994; Chidambaram et al., 1998). The presence of CD38 on mitochondria with similar molecular weight to plasma membrane CD38 is intriguing because it is well documented that most mitochondrial proteins, including soluble and membranebound proteins, not follow the ER-Golgi secretory pathway but are synthesized in the cytosol then transported to mitochondria (Neupert., 1997; Rapaport, 2003). Given 147 Chapter Characterization of CD38 Expressed in Different Cellular Compartments that mitochondrial proteins not follow the secretory pathway, it is expected to observe a 33kDa CD38 molecule rather than a 45kDa form. The puzzle of the mechanism by which a membrane glycoprotein such as CD38 can be translocated to an intracellular compartment such as mitochondria is an intriguing and important one. In fact, a number of reports have clearly shown that in contrast to previously held beliefs, glycosylated proteins are indeed present in the nucleus, mitochondria as well as the cytoplasm (Hart., 1997; Camici and Corazzi, 1997; Chandra et al., 1998; Schwer et al., 2004; Babakhanian et al., 2007). Most recent studies have provided evidence of trafficking of HCMV protein UL37 from ER, where it will be N-glycosylated in ER and subsequently imported into mitochondria (Mavinakere et al., 2006; Bozidis et al., 2008). This sequential trafficking from ER to mitochondria was shown to occur through a continuous lipid layer connecting the two organelles. High-resolution electron tomography has demonstrated that the ER and mitochondria are in close proximity and have numerous contacts (Marsh et al., 2001). It is then postulated that mitochondria-associated membrane (MAM), a subdomain of the ER, allows direct physical contact to mitochondria, as membrane bridges. Contacts between ER and mitochondria have shown to be exchanging sites for transfer of lipids (Stone et al., 2000) and potentially of Ca2+ (Pinton et al., 2001; Filippin et al., 2003; Szabadkai and Rizzuto, 2004; Yi and Hajnoczky, 2004) between organelles. The mitochondria-associated membrane (MAM) has recently been shown to function as a pre-Golgi compartment of the secretory route (Rusinol et al., 1994; Bozidis et al., 2008). Collectively, it is indeed tempting to postulate that ER enzymatic machinery appears to be responsible for the N-glycosylation of mitochondrial glycoproteins. It is therefore conceivable to suppose 148 Chapter Characterization of CD38 Expressed in Different Cellular Compartments that fully glycosylated CD38 (45kDa) can be translocated to mitochondria and expressed on the organelle. Ca2+ release experiments will be discussed in detail in the following chapter. About 30 years ago, Moser et al. (1983) showed the first definite report on the occurrence of an unidentified NAD+ glycohydrolase in rat liver mitochondria. Later publications further showed that this glycoprotein exhibit both cADPR and NAADP production activities, suggesting that these activities are a result of the presence of CD38 (Liang et al., 1999). The results obtained here in this present study supported these findings, and presented a novel localization of functional CD38 in outer mitochondria membrane. The present studies further showed that mitochondrial CD38 with its enzymatic site facing the cytosolic site as its novel topological appearance. The nature of this discovery suggests that CD38 might be involved in multiple signaling roles in the cell depending on its localization. However, the current studies have been performed under nonphysiological conditions in which CD38 were targeted with specific organelle targeting signal and overexpressed. It remains to be seen if endogenous CD38 display the same characterization and functional role as proposed in this study. The investigation was then continuing by examine the endogenously expressed CD38 on mitochondria isolated from mouse brain tissue (Chapter 4). 149 [...]... Mito -CD38 transfected COS-7 cells (See Material & Methods) It is known that NA+/K+ ATPase is an intrinsic enzyme on the plasma membrane of most cells and therefore commonly used as a plasma membrane marker In order to eliminate the possibility of plasma membrane contamination during mitochondria fractionation, the presence of NA+/K+ ATPase α form (α subunit of NA+/K+ ATPase) was investigated It can... to be extruding to the cytosolic region Four distinct compartments can be distinguished within mitochondria: the outer membrane, the inter membrane space, the inner membrane, and the matrix In order to investigate the subcellular compartmentalization of CD38 proteins, subcellular and partial submitochondrial fractionation experiments were carried out The precise topology of mitochondrial CD38 was investigated... that CD38 was successfully expressed and localized specifically to mitochondria, with the use of the pShooter vector system with a mitochondrial targeting signal Further investigation of subcellular localization of expressed CD38 in mitochondria in parallel with the ADPribosyl cyclase assay had demonstrated localization on the outer mitochondrial membrane with a specific topology in which the catalytic... Cruz), anti -CD38 peptide Ab, 39 (b, lab customized Ab), and CDA233 (c, Bio-product) using standard Western blot techniques as described under Materials and Methods A distinct ~45kDa band is detected (red arrow) for isolated CD38+ mitochondria Total protein loaded was 20 µg for each lane Mito -CD38 extracted mitochondria expressing recombinant CD38 Control Mito– extracted mitochondria from vector-transfected... cellular localization of CD38 expressed in mitochondria and to further characterize this protein expressed on the mitochondria, subcellular fractionation of the mitochondria from the Mito -CD38 transfected cells was carried out Simultaneously, to confirm the identity of CD38 localized on mitochondria, and to verify the specificity of the antibody used, the extracted mitochondria prepared from Mito -CD38. .. amount of CD38 in the proteinase K treated samples prior to hypotonic treatment (Figure 3.18 b) To further analyze the submitochondrial localization of CD38 in mitochondria, subfractions of mitochondria were identified by markers: Bcl-xL for outer membrane (Gonzalez-Garcia et al., 1994, Kaufmann et al., 20 03, Stewart et al., 20 05), prohibitin 122 Chapter 3 Characterization of CD38 Expressed in Different... the surface membrane of the CD38+ COS-7 cells with no cytoplasmic staining This is clearly identified as plasma membrane staining Figure 3.9 B demonstrated the localization of immunoreactive CD38 in the mitochondria, which shows diffuse cytoplasmic staining pattern throughout the cell body; and distinctly different from the staining pattern of plasma membrane The CD38 immunostaining pattern in ER in Figure... seen that 115 Chapter 3 Characterization of CD38 Expressed in Different Cellular Compartments there was an absence of NA+/K+ ATPase in the isolated mitochondrial fraction as compared to the whole cell lysate, thus attesting to the purity of the mitochondrial fraction (Figure 3.15 A & B) In comparison with the whole cell lysate, (by using an antibody against prohibitin, constitute mitochondria protein),... was susceptible to protease digestion Prohibitin, CoxIV and mtHSP70 reside in the inner mitochondrial membrane was protected from the proteinase K treatment Lane (a) showed intact CD38+ mitochondrial samples before proteinase K treatment Intact CD38+ mitochondria was subjected to proteinase K treatment prior to hypotonic treatment (b), hypotonic treated CD38+ mitochondria with no proteinase K (c) and... the Whole cell lysate (WCL) A) Immunoblot using NA+/K+ ATPase α subunit 1 (a) , nucleoporin (b), calretigulin (c) and prohibitin antibodies (d) Subfractionation of mitochondrial fraction and preparation of whole cell lysate were carried out as described in Materials & Methods Total protein loaded was 10µg for each lane B) Image J software was used to analyze the various bands to quantitatively identify . specific topology of CD38 and its role in Ca 2+ -release assay observed from the data, this suggest that mitochondrial CD38 plays a role in cADPR synthesis and may participate in a novel pathway of. In a separate experiment, the immunostaining pattern of CD38 targeted to ER and mitochondria using anti -CD38 antibody, C1586 and anti-myc antibody displayed a match staining pattern to organelle,. idea that CD38 can play a role in cADPR formation at any of these locations is indeed very tempting and reasonable one as all these subcellular compartments are in fact in close spatial proximity