Subcellular compartmentalization of CD38 in non hematopoietic cells a study to characterize its functional role in mitochondria 3

Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain CHAPTER CHARACTERIZATION OF CD38 EXPRESSED IN MITOCHONDRIA FROM MURINE BRAIN Synopsis CD38 is a type II transmembrane glycoprotein found on both hematopoietic and non-hematopoietic cells. It is well known for its involvement in the metabolism of cyclic ADP-ribose (cADPR), a cyclic nucleotide with calcium mobilizing activity independent of inositol trisphosphate. It is generally believed that CD38 is an integral protein with ectoenzymatic activities found mainly on the plasma membrane. Most of its known functions were derived from studies on the receptor and enzymatic properties of CD38 on the surface membrane. Thus far, no study has been performed to conclusively show the functional role of CD38 on mitochondria. Here, for the first time, enzymatically active CD38 was shown to be present intracellularly on the outer mitochondrial membrane (OMM) of CD38+mitochondria isolated from COS-7 cells transiently transfected with Mito-CD38 (CHAPTER 3) and supported by experimental results acquired using mouse brain tissue. Immunolabeling on mouse brain vibratome sections and isolated mitochondria using both TEM and SEM determined the localization of CD38 to the outer mitochondrial membrane with its Cterminal region (domain that contains the catalytic site) facing the cytosolic site. This specific topology of CD38 was further verified and supported by performing digitonin/proteinase K assay. Together with the evidence presented in this chapter, this study has directly demonstrated the expression of functionally active CD38 in mitochondria, which may therefore participate in a novel pathway of intracellular Ca2+ signalling in non-hematopoietic cells. 150 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain 4.1 Introduction 4.1.1 CD38 in Brain Mammalian brain tissues, the cerebellum in particular, seem to be suitable for studies on the CD38/cADPR/ADPR signaling system. CD38 is highly expressed in brain and accounts for the majority of the in vitro cyclase activities using brain tissue extracts (Lee et al., 1994). Furthermore, isolated rat brain (White et al., 1993) and cerebellum (Takasawa et al., 1993) microsomes have been proven to release Ca2+ when exposed to cADPR, indicating the presence of the complex receptor machinery that is responsible for the opening of the Ca2+ channels in these stores. In mouse brain, CD38 was found in both neurons and glial cells, showing a predominant intracellular location, and was enriched in neuronal perikarya (Ceni et al., 2003; Jin et al., 2007). In human brain, CD38 immunoreactivity was demonstrated in the perikarya and dendrites of many neurons (Mizuguchi et al., 1995). In rat astrocytes, ADP-ribosyl cyclase has been reported to have both intracellular and extracellular actions (Hotta et al., 2000). Moreover, co-culture of astrocytes with neurons resulted in significant overexpression of astrocyte CD38 both on the plasma membrane and intracellularly, and this effect was attributed to neuronreleased glutamate action on astrocytes (Bruzzone et al., 2004). Interestingly, Mizuguchi group’s immunohistochemical investigations using human brain tissues localized most of the brain CD38 immunoreactivity into the perikaryal and dendritic cytoplasm of neurons. The granular staining profiles suggest an association with intracellular organelles (Mizuguchi et al., 1995). In a separate work by Yamada group, CD38 was observed to be expressed in specific population of rat CNS (central nervous system) neurons, all of which showed labeling of the plasma 151 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain membrane as well as cell organelles such as rough ER, small vesicles, mitochondria and nuclear envelope (Yamada et al., 1997). Yamada group observed that in the rat cerebral cortex, CD38 immunoreactivity was demonstrated in a subset of pyramidal neurons, and was distributed preferentially in the perikarya and dendritic arbors. The subcellular labeling was in the region close to the plasmalemma, including the postsynaptic densities, implying that CD38 is involved in signal transduction via the plasma membrane of certain CNS neurons. In the cerebellar cortex, the immunoreactivity was recognized in a rather wide range of neuronal types such as granule, Golgi and basket cells. CD38 expression in rat Purkinje cell bodies and dendrites were quite weak compared to those in the human cerebellum, probably due to species differences. Except for the Purkinje cells, the subcellular localization of CD38 in cerebellar neurons was noted in the perikarya, axon terminals and dendrites, in order of decreasing intensity. Although the labeling of the plasmalemma and several organelles in these structures was similar to that in the cerebrum, the association of immunoreactivity with synaptic vesicles is rather unique to cerebellar neurons, suggesting an additional functional role of CD38 in certain types of presynaptic region. The present study also demonstrated the expression of CD38 in astrocytes. The immunolabeling was more intense compared to that in the neurons, and was distributed ubiquitously in the perikarya and processes, suggesting an essential role of this molecule in astrocytes. The neuronal distribution of CD38 revealed in this study showed a high level of conformity with that of ryanodine receptors in the mouse CNS (Nakanishi et al., 1992), implying the involvement of CD38 in the cADPR-mediated Ca2+-mobilizing system in neurons (Yamada et al., 1997). 152 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain Aksoy group presented the latest finding providing evidence that high cADPR levels were indicated in brain tissues as compared to other tissues such as heart, liver, spleen and kidney, conforming to the past findings (Aksoy et al., 2006). This group then proposed CD38 as the major NAD+ glycohydrolase/NADase in mouse brain. The term NAD+ glycohydrolase /NADase was used for the reason that the majority of CD38 catalytic activity is the degradation of NAD+ to the final product, ADPR (Introduction 1.3). Using separate techniques they have observed that the NAD+ glycohydrolase activity was present in WT brain extracts but was significantly reduced in CD38 KO mouse. The finding was further supported by the observation that a majority of the NAD+ glycohydrolase activity in the WT brain extracts can be immunoprecipitated by CD38 antibody. In addition, they also found that CD38 are not present solely in the plasma membrane but localized in intracellular membranes of organelles such as mitochondria and nuclei (Aksoy et al., 2006). 4.1.2 Brain Mitochondria and CD38 Mitochondria are unique organelles central for various cellular processes that include ATP production via oxidative phosphorylation, intracellular Ca2+ homeostasis, steroid synthesis, forms of apoptotic cell death and generation of reactive oxygen species. Neurons critically depend on mitochondrial functions to establish membrane excitability and to execute the complex processes of neurotransmission and plasticity (Wallace, 2005). The cell modulates increased cytosolic calcium either by extruding it across the plasma membrane or by compartmentalizing it to the endoplasmic reticulum and/or the mitochondria. Although the accumulation of calcium by mitochondria has been shown to be relatively slower, involving a Ca2+ uniporter of lower affinity when 153 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain compared with that of the endoplasmic reticulum (Carafoli, 1987), mitochondria possess a very high capacity for calcium and thus may be of significance when a sustained rise in intracellular calcium above 1µM occurs (Burgess et al., 1983). So, normal Ca2+ cycling [at low resting mitochondrial Ca2+ concentration ([Ca2+]m)] occurs by the movement of Ca2+ into mitochondria via the Ca2+ uniporter and slow extrusion via the Na+/Ca2+ exchanger or by Na+-independent mechanisms involved mitochondrial H+/Ca2+ exchanger (Nicholls et al., 2000; Bernardi, 1999). Both antiporters are of low capacity thus the transport rate can be easily surpassed by the Ca2+ uniporter, leading to net Ca2+ accumulation in mitochondria. Isolated mitochondria in the presence of phosphate take up Ca2+ to a fixed capacity, in a membrane potential (∆ψm) - dependent fashion (Nicholls and Akerman, 1982; Gunter et al., 1994; Chalmers et al., 2003). An alternative pathway for Ca2+ is via the transient low-conductance opening of the mitochondrial permeability transition pore (mPTP), which may release the toxic Ca2+ loads from mitochondria (Zoratti et al., 1995; Ichas et al., 1997). Mitochondrial Ca2+ uptake was observed in intact cells at cytosolic Ca2+ concentration ([Ca2+]c) as low as 150-300nM (Pitter et al., 2002). In the event of high local Ca2+ domains of tens of micromolar in the vicinity of voltage-operated Ca2+ channels or Ca2+ release sites of the endoplasmic reticulum (Rizzuto and Pozzan, 2006), rapid uptake Ca2+ into mitochondria might be necessary (Pacher et al., 2002; Gerencser et al., 2005). The ability of the mitochondrion to accumulate as well as to retain calcium could be of critical importance to the survival of the cell. An inability of the mitochondrion to retain sequestered calcium may lead to sustained elevated levels of cytosolic calcium, perhaps in concentrations high enough to activate calcium-dependent degradative processes (Nicotera et al., 1992). 154 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain The pathways involved in the movement of calcium from the matrix space are less well characterized. It is clear that efflux of accumulated calcium from respiring mitochondria can be induced by a variety of prooxidants (Lotscher et al., 1979; Bellomo et al., 1982; Boquist, 1984; Graf et al., 1985; Frei and Richter, 1986; Richter et al., 1987; Fagian et al., 1990). Several hypotheses have been presented to explain the efflux of calcium from respiring mitochondria, including changes in the redox status of the mitochondrial pyridine nucleotide pool (Lehninger et al., 1978), oxidantmediated formation of a nonspecific pore in the inner mitochondrial membrane (Crompton et al., 1988), or hydrolysis of intramitochondrial oxidized pyridine nucleotides by an NAD+ glycohydrolase /NADase (Lotscher et al., 1979; Lotscher et al., 1980; Richter et al., 1987). NAD+ has recently emerged as a crucial regulator of the signaling pathways implicated in multiple physiological conditions (refer to review by Chini, 2009). Important signaling roles of NAD+ such as its direct consumption for the synthesis of ADPR/ADPR polymers (ADP-ribosylation) (Jacobson et al., 1983; Shah et al., 1996; Vu et al., 1997), which appears to be involved with responses that can lead to normal cellular recovery, apoptosis or necrosis. cADPR along with ADPR is potent Ca2+ releasing agent involved in many signaling pathways leading to apoptosis or necrosis (refer to review by Chini, 2009) as well as a role as a substrate and regulator of the NAD+ dependent deacetylases sirtuins (Ziegler and Niere, 2004; Baur et al., 2006; Lagouge et al., 2006). Since NADP can be generated de novo from NAD+, conversion of NADP to NAADP by the NAD+ glycohydrolase serves as another Ca2+ mobilizing agents in cell (Lemer et al., 2001). All these signaling pathways have been shown to be very important in many physiological conditions ranging from egg 155 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain fertilization to the cellular mechanism of aging; longevity and death (refer to review by Chini, 2009). It is not unexpected that a significant proportion of the cellular NAD+ pool is reported to be compartmentalized within the mitochondria (Tischler et al., 1977; Lisa et al., 2001). It appears that NAD+ can be synthesized within this organelle and accumulated at much higher concentrations than seen in the cytosol (Lisa et al., 2001). The nuclear enzyme NMNAT-1 is the key enzyme in both de novo and salvage pathways of NAD+ synthesis (Kolb et al., 1999), until recently two newly reported isoforms, NMNAT-2 and NMNAT-3, which are located separately in the Golgi complex and mitochondria, are added to the family (Raffaelli et al., 2002; Magni et al., 2004; Berger et al., 2005). This suggests independent NAD+ metabolism in the nucleus, the Golgi complex and mitochondria (Figure 4.1). It is well recognized that NAD+ can mediate calcium homeostasis (Figure 4.2) through pathways such as the following: a) ADP-ribosyl cyclases/ NAD+ glycohydrolases produced cADPR from NAD+, which serves as a potent endogenous agonist of ryanodine receptor mediated calcium channels; b) ADPR, another Ca2+ mobilizing molecule generated from NAD+ by NAD+ glycohydrolases can activate TRPM2 receptors leading to Ca2+ influx (Guse, 2005) as well as substrate used for ADP-ribosylation, a posttranslational protein modification process as briefly introduced above. Several lines of evidence suggest that NAD+-dependent signaling events may occur in mitochondria. Dating back to the 1980s intensive studies were carried out on mitochondria NAD+ glycohydrolase /NADase in brain, liver and heart tissues. Since then this data has always been linked with ADP-ribosylation process on the inner membrane in mitochondrial calcium homeostasis due to the initial finding that NAD+ glycohydrolase is found on the inner mitochondrial membrane. Recent 156 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain findings found otherwise that this molecule is in fact localized on the outer mitochondrial membrane (Satrustegui et al., 1984; Boyers et al., 1993; Lisa et al., 2001). It was then postulated that instead of the previous belief of mitochondrial Ca2+ release upon the ADP-ribosylation of an intrinsic protein of the inner mitochondrial membrane following the pyridine nucleotide hydrolysis by a Ca2+-stimulated matrix NAD+ glycohydrolase (Boyer CS and Peterson DR, 1991; Masmoudi and Mandel, 1987), new evidence has shown that NAD+ hydrolysis is only possible after the PTP opening and Ca2+ release is the cause rather than consequence of NAD+ hydrolysis (Lisa et al., 2001). Recent report has further identified that bovine liver mitochondrial NAD+ glycohydrolase as the ADP-ribosyl cyclase (Ziegler et al., 1997). Following the finding by Liang et al. (1999) which showed a minimal amount of CD38 localized in rat liver mitochondria fraction, it is of interest to investigate the presence of CD38 in other tissues that have been shown to have robust ADP-ribosyl cyclase activities, i.e. brain (Aksoy et al., 2006). To have a closer investigation into the Ca2+ interplay by this molecule, the exact submitochondrial location of this activity has to be determined. 157 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain Figure 4.1 NAD+ metabolisms in cells. NAD+ metabolism occurs both intracellularly in various subcellular organelles and extracellularly. The key NAD+ synthesizing enzymes NMNAT-1, NMNAT-2, and NMNAT-3 are located at the nucleus, the Golgi complex, and mitochondria, respectively. There are NAD+-consuming enzymes in these organelles, including poly (ADP-ribose) polymerase-1 (PARP-1), PARP-2, and certain sirtuins in the nucleus, tankyrase in the Golgi complex, and NAD+ glycohydrolases in mitochondria. On plasma membranes, mono (ADP-ribosyl) transferases (ARTs) and ADP-ribosyl cyclases (ARCs) produce mono (ADPribosylation) on target proteins and generate cyclic ADP-ribose, respectively. Nicotinamide phosphoribosyltransferase (Nampt) may exist extracellularly and produce its biological effects by generating nicotinamide mononucleotide from nicotinamide. (Adapted from Ying, 2008) 158 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain Figure 4.2 Pathways by which NAD+ and NADP can affect calcium homeostasis. ADP-ribosyl cyclases (ARCs), poly(ADPribose) polymerases (PARPs)/poly(ADPribose) glycohydrolase (PARG), and sirtuins use NAD+ as a substrate to generate several Ca2+-mobilizing second messengers, including cADPR, ADP-ribose, and Oacetyl-ADP-ribose (O-acetyl-ADPR), which can activate TRPM2 receptors and ryanodine receptors (RyR). NAD+-dependent mono (ADP-ribosyl) transferases (ARTs) can also affect Ca2+ homeostasis by producing mono-ADP-ribosylation of P2X7 receptors (ADPR- P2X7R). NADH could modulate Ca2+ homeostasis by affecting IP3-gated Ca2+ channels, mitochondrial permeability transition (mPTP) and RyR. NAADP generated from NADP can also mobilize intracellular NAADPdependent Ca2+stores. NADPH may affect calcium homeostasis by its major effects on antioxidation and ROS generation, which can affect Ca2+pumps and Ca2+channels (Adapted from Ying, 2008). The encircled area is the major focus in this study. 159 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain The protease accessibility of other proteins of the intermembrane space and the inner mitochondrial membrane was analyzed to determine the integrity of the mitochondria. OPA1, a protein of the intermembrane space, was protease resistant in mitochondria but sensitive to proteinase K digestion when the outer membrane become solubilised (Figure 4.7). In contrast, the matrix protein mtHsp70 required relatively higher digitonin/protein ratio before it become accessible to protease digestion. Together, these results confirmed the mitochondria employed in these experiments remained intact prior to solubilisation by digitonin. While it was surprising to observe partial digestion of Tim23 by proteinase K in mitochondria sample even before application of digitonin, this data in fact is in agreement with those reported by Donzeau et al. (2000). They have proposed that with the unique localization of the N-terminal domain of Tim23 in outer membrane that Tim23 spans both mitochondria membranes, which explains the accessibility of the protein to unspecific protease such as proteinase K in intact mitochondria. At low digitonin/protein ratio, i.e, 0.1, the outer mitochondrial membrane started to become ‘leaky’, as shown by near 90% degradation of Tim23 and the onset of OPA1 digestion (Figure 4.7). Tim23 was completely removed at digitonin/protein ratio 0.2 onward as well as a nearly 60% protein removal of OPA1. Interestingly, prohibitin showed resistance to digitonin solubilisation as well as proteinase K digestion. Again, this could be explained by the deep association of the protein in the membrane, specific protein folding as well as the lack of specific enzyme digestion site. Cytochrome c again showed resistance to digitonin solubilisation as well as proteinase K digestion. This could be due to the reasons mentioned previously. The present data has shown that the levels of Cytochrome c, OPA1 and prohibitin were not affected by the treatment of intact mitochondria with proteinase K 169 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain (under non digitonin-solubilisation), despite the noticeable loss of CD38 (crossreference with outer mitochondrial membrane resident proteins, Tom20 and Bcl-xL), indicating that CD38 was not located in the matrix, inner mitochondrial membrane, or intermembrane space of mitochondria. It is noteworthy that anti-Tom20 antibody employed in this study targets to the protein’s C-terminal region which is known to extrude out in the cytosol (Iwahashi et al., 1997). As expected, the protein was susceptible to low digitonin solubilisation as well as protease digestion where intact mitochondria was employed in the study. Similarly, the anti-CD38 antibody sc-7049 (M19) that was used recognized the epitope that mapped to the C-terminal of CD38. Taken together, it is tempting to postulate that CD38 is expressed exclusively on the surface of the outer mitochondrial membrane with its bulky carboxyl catalytic domain facing the cytosol. 170 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain C 0.1 0.25 0.3 0.4 mg/ml dig/prot kDa 250 100 75 50 CD38 (A) 37 25 20 Tom20 (B) Bcl-xL (C) OPA1 (D) Tim23 (E) Prohibitin (F) mtHSP70 (G) Cytochrome c (H) 171 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain Figure 4.6 The topology of the CD38 molecule was investigated by digitonin titration of the Percoll purified mitochondrial fractions. Extraction of CD38 from Percoll purified mitochondrial fractions were carried out by a series of digitonin (dig)/protein (prot) ratio of 0.1, 0.25, 0.3 and 0.4. CD38, Tom20 and Bcl-xL were extracted from the mitochondrial fractions at 0.25 dig/prot ratio onward. Intermembrane space and matrix proteins were not affected by the digitonin treatment. Proteins that remained in the mitochondrial pellets were analyzed by SDS-PAGE and Western blotting which were probed separately with (A) anti-CD38 (B) anti-Tom20 (C) anti- Bcl-xL (D) antiOPA (E) anti-Tim23 (F) anti-prohibitin (G) anti-mtHSP70 (H) anti-cytochrome c antibody (Materials & Methods). Equal protein loading of the mitochondrial pellet lanes was monitored using anti-Cytochrome c antibodies. The results shown are representative of two independent experiments. Total proteins loaded for purified mitochondrial fraction were approximately 100µg. arrow-CD38 Figure 4.7 The topology of the CD38 molecule was investigated further by digitonin titration of the mitochondria and protease protection assay. Accessibility to protease digestion was determined by treatments of Percoll purified mitochondrial fractions with proteinase K and increasing concentrations of digitonin (dig). After 30 min, the reaction was stopped with PMSF, and the proteins (prot) were separated and analyzed by SDS-PAGE and Western blotting which were probed separately with (A) antiCD38 (B) anti-Tom20 (C) anti- Bcl-xL (D) anti-OPA (E) anti-Tim23 (F) antiprohibitin (G) anti-mtHSP70 (H) anti-cytochrome c antibody (Materials & Methods). CD38, Bcl-xL and Tom20 were sensitive to proteinase K digestion prior to digitonin treatment. Intermembrane space and matrix proteins were vulnerable to proteinase K digestion at the respective dig/prot ratio. Equal protein loading of the mitochondrial pellet lanes was monitored using anti-Cytochrome c antibodies. The results shown are representative of two independent experiments. Total proteins loaded for purified mitochondrial fraction were approximately 100µg. arrow-CD38 PK-Proteinase K 172 Chapter kDa Characterization of CD38 Expressed in Mitochondia from Murine Brain C - C + 0.1 + 0.2 + 0.25 + 0.3 + 0.35 + 0.4 + mg/ml dig/prot µg/ml PK 250 100 75 50 CD38 (A) 37 25 20 Tom20 (B) Bcl-xL (C) OPA1 (D) Tim23 (E) Prohibitin (F) mtHSP70 (G) Cytochrome c (H) 173 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain 4.2.4 Determination of the enzymatic activities of brain mitochondrial CD38 4.2.4.1 Determination of ADP-ribosyl cyclase activity of mitochondrial CD38 The purified mitochondrion was characterized further by examining the ADPribosyl cyclase activity of CD38. As expected, high ADP-ribosyl cyclase activity from both microsome and crude mitochondrial fraction was observed. Usually crude mitochondrial fraction is heavily contaminated by microsomal CD38 which accounts for high ADP-ribosyl cyclase activity. In contrast, a small but relatively distinct ADPribosyl cyclase activity from the purified mitochondrial fraction was observed. This result indicated for the first time the presence of a functionally active ADP-ribosyl cyclase in the brain mitochondrial fraction. The findings using purified mitochondria are therefore consistent with those reported previously using isolated mitochondria with targeted CD38 from Mito-CD38 transfected COS-7 cells, and thus suggest a role of mitochondrial CD38 in Ca2+ homeostasis (Chapter 3). Intact mitochondria were used in the protease protection assay previously to examine the protein topology on mitochondria. The proteinase K proteolysis of the intact mitochondria resulted in a loss in CD38 protein observed in Western blot analysis (Figures 4.7) suggested a corresponding possible loss in enzymatic activity, given that ADP-ribosyl catalytic site is located in the protein’s extracellular domain. ADP-ribosyl cyclase assay was therefore carried out to examine the enzymatic activity on the proteinase K treated mitochondrial samples. As expected, a corresponding decrease in cyclase activity was observed for proteinase K treated Percoll purified mitochondrial fraction as well as crude mitochondrial and microsomal fractions (Figure 4.8). Crude mitochondrial and microsomal fractions were used as positive control in this experiment. The relative susceptibility of 174 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain mitochondrial CD38 protein to protease digestion as well as its activity provided solid evidence that the catalytic domain of CD38 is in fact located on the outer mitochondrial membrane with a specific orientation that the catalytic damain facing cytosolic side. * * Cyclase activity (% )Control * PK - + Microsomes - + Crude Mito - + Purified Mito Figure 4.8 ADP-ribosyl cyclase activities of microsome, crude mitochondria and purified mitochondrial fractions. Microsomes, crude mitochondria (Crude Mito) and Percoll purified mitochondria (Purified Mito) fractions were prepared and subjected to proteinase K (PK) treatment. Fractions with/without PK treatment were then assayed for cyclase activity using NGD as the substrate under the conditions described in section Materials & Methods. Basal cyclase activity for Microsomes, Crude Mito and Purified Mito, respectively, was 284±6.39, 266±7.19, 51±1.92 nmoles min-1 mg-1. Values are mean ± SD of independent experiments performed (n=5). * represents significant differences in cyclase activities with respect to the untreated (-) group (P< 0.05). 175 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain 4.2.4.1 Determination of NAD+ glycohydrolase activity in mitochondria fraction. It was reported that the major enzymatic activity of CD38 is NAD+ hydrolysis and this enzyme appears to be very inefficient in catalyzing the conversion of NAD+ to cADPR (Chini, 2009). In fact, CD38 will generate one molecule of cADPR for every 100 molecules of NAD+ hydrolyzed, as reported by Aksoy (2006). It was then necessary to determine the NAD+ glycohydrolase activity of the mitochondrial CD38. A fluorimetric method was employed using 1, N6-etheno-NAD as the substrate. NAD+ glycohydrolase will act on the non fluorescent 1, N6-etheno-NAD (e-NAD+) and catalyze the substrate hydrolysis to a fluorescent compound, 1, N6-etheno-ADPR. Table 4.1 clearly demonstrated there were high levels of NAD+ glycohydrolase activity in the crude mitochondrial fractions and Percoll purified mitochondrial fractions. Significantly higher e-NAD+ hydrolysis activity observed in crude mitochondrial fraction as compared to the Percoll purified mitochondrial fraction (Table 4.1, 1157±24.1, 397±7.3, P < 0.05) is not surprising, given that crude mitochondrial fractions contain contaminants such as microsomal fractions (Figure 4.4). However, compared with the NAD+ glycohydrolase activity, significantly lower ADP-ribosyl cyclase activity was observed in all mitochondrial fractions (Table 4.1). The present data is consistent with the findings previously reported by Aksoy et al. (2006), which concluded that CD38 is the major NAD+ glycohydrolase present in brain tissues, along with the observation that NAD+ glycohydrolase activity was present in mitochondrial fraction of WT mice brain tissues, while essentially absent from CD38KO mice (also refer to Figure 4.9). 176 Chapter Characterization of CD38 Expressed in Mitochondia from Murine Brain Table 4.1 Determination of cyclase and NAD+ glycohydrolase activity in both crude mitochondrial fractions and Percoll purified mitochondrial fractions from mouse brain tissues. GDP-ribosyl Cyclasea, b (nmol min-1mg-1) NAD+ glycohydrolasea,b (nmol min-1mg-1) Crude mitochondria *270±2.8 1157±14.1 GDP-ribosyl cyclase /NAD+ glycohydrolase 0.233 Percoll purified mitochondria *40±1.30 187±7.3 0.214 a The enzyme activities are expressed as a net change in fluorescence (excitation 300nm, emission 410nm) as described in Materials and Methods. 50µg protein was used for each assay. b The data shown represent the average ± SD of five independent observations. *significantly lower in cyclase activities as compared to NAD+ glycohydrolase activities (P [...]... mitochondria with targeted CD38 from Mito -CD38 transfected COS-7 cells, and thus suggest a role of mitochondrial CD38 in Ca2+ homeostasis (Chapter 3) Intact mitochondria were used in the protease protection assay previously to examine the protein topology on mitochondria The proteinase K proteolysis of the intact mitochondria resulted in a loss in CD38 protein observed in Western blot analysis (Figures... cytosolic side as observed previously in experiments using cell lines (Chapter 3) 168 Chapter 4 Characterization of CD38 Expressed in Mitochondia from Murine Brain The protease accessibility of other proteins of the intermembrane space and the inner mitochondrial membrane was analyzed to determine the integrity of the mitochondria OPA1, a protein of the intermembrane space, was protease resistant in. .. Characterization of CD38 Expressed in Mitochondia from Murine Brain 4.2.4 Determination of the localization of mitochondrial CD38 using Transmission Electron Microscopy (TEM) 4.2.4.1 Localization of mitochondrial CD38 on mouse brain sections with DAB staining To determine the localization of the mitochondrial CD38, three CD38 antibodies were employed, namely, Santa Cruz antibody (sc-7049), customized... Murine Brain B A C C Figure 4.10 CD38 immunoreactivity in a Golgi cell soma, labeled with goat polyclonal CD38 antibody-sc-7049 Intracellular CD38 staining observed was most intense on mitochondria followed by ER (encircled areas A- C) The positive stained 182 Chapter 4 Characterization of CD38 Expressed in Mitochondia from Murine Brain mitochondria (arrow) was in close proximity with ER (inset) Scale... considering that cellular localization of CD38 is found predominantly in the plasma membrane (Jackson et al., 1990; Gelman et al., 19 93; Hara-Yokoyama et al., 1996; Deaglio et al., 1996; Konopleva et al., 1998; Franco et al., 1998) 1 63 Percollpurified mitochondria Crude mitochondria kDa 250 Microsomes Characterization of CD38 Expressed in Mitochondia from Murine Brain Homogenate Chapter 4 150 100 75 50 CD38. .. recovery of active and pure brain mitochondria (Nishadi et al., 2001) The isolation of mitochondria based on this method has been shown to result in an enriched mitochondrial preparation with negligible microsomal, nucleus, endoplasmic reticulum and plasma membrane contamination as well as microsomal-mitochondrial association Mitochondria were isolated from mouse brain tissues using the combination of gravity... protease such as proteinase K in intact mitochondria At low digitonin/protein ratio, i.e, 0.1, the outer mitochondrial membrane started to become ‘leaky’, as shown by near 90% degradation of Tim 23 and the onset of OPA1 digestion (Figure 4.7) Tim 23 was completely removed at digitonin/protein ratio 0.2 onward as well as a nearly 60% protein removal of OPA1 Interestingly, prohibitin showed resistance to. .. employed in the study Similarly, the anti -CD38 antibody sc-7049 (M19) that was used recognized the epitope that mapped to the C-terminal of CD38 Taken together, it is tempting to postulate that CD38 is expressed exclusively on the surface of the outer mitochondrial membrane with its bulky carboxyl catalytic domain facing the cytosol 170 Chapter 4 Characterization of CD38 Expressed in Mitochondia from Murine... As a result, protein digestion of CD38, Tom20 and Bcl-xL were observed prior to addition of digitonin Because the anti -CD38 antibody employed in this study was raised against the C-terminal half of this protein, it can be concluded that at least the C terminal of the protein is exposed to the cytosol region This data supported a specific topology of CD38 with its bulky C-terminal region extruding to. .. Chapter 4 Characterization of CD38 Expressed in Mitochondia from Murine Brain A B kDa 250 150 100 75 50 37 25 1 2 3 4 1a 2a 3a 4a Figure 4.5 Detection of CD38 from purified mitochondrial fraction prepared from mouse brain tissues Immunoblots comprised of the purified mitochondrial fractions at 100µg (lanes 3 and 3a) and 50µg (lanes 4 and 4a) were probed with the anti -CD38 antibody (sc-7049) and compared with . study has directly demonstrated the expression of functionally active CD38 in mitochondria, which may therefore participate in a novel pathway of intracellular Ca 2+ signalling in non -hematopoietic. Chapter 4 Characterization of CD38 Expressed in Mitochondia from Murine Brain 150 CHAPTER 4 CHARACTERIZATION OF CD38 EXPRESSED IN MITOCHONDRIA FROM MURINE BRAIN Synopsis CD38 is a type. releasing agent involved in many signaling pathways leading to apoptosis or necrosis (refer to review by Chini, 2009) as well as a role as a substrate and regulator of the NAD + dependent deacetylases