Properties of a highly purified mitochondrial deoxyguanosine kinase

Deoxyguanosine kinase, purified over 6000-fold from beef liver mitochondria by means of deoxyguanosine-3'-(4-aminophenyl phosphate)-Sepharose affinity chromatography, was nearly homogeneous. It phosphorylates only deoxyguanosine and deoxyinosine among the natural nucleosides, with apparent Km values of 4.7 and 21 microM, respectively. Among nucleoside analogs tested, only arabinosylguanine (Ki = 125 microM) and 8-aza-deoxyguanosine (Ki = 450 microM) competed with deoxyguanosine. The relative molecular mass of the enzyme is 56,000, as determined by equilibrium sedimentation, and sodium dodecyl sulfate-gel electrophoresis suggests two subunits of Mr 28,000. The pH optimum for enzyme activity is 5.5, but optimum enzyme stability is seen at pH 7.0. Triton X-100 increased the stability of the enzyme markedly. ATP is the best phosphate donor at pH 5.5, but pyrimidine triphosphates such as dTTP and UTP are more efficient donors at pH 7.4. The activation energy, at pH 5.5, was estimated to be 10.9 kcal/mol. Amino acid modification experiments suggest the involvement of arginine, cysteine, and probably histidine. The inactivation of the enzyme by modification of these amino acid residues was time and pH dependent. Both substrates protected the enzyme from inactivation in every case but that of photooxidation by Rose Bengal, where only deoxyguanosine prevented inactivation.     

Mitochondrial deoxyguanosine kinase mutations and mitochondrial DNA depletion syndrome

Mitochondrial deoxyguanosine kinase (dGK) catalyzes the initial phosphorylation of purine deoxynucleosides. Mutations in the dGK gene leading to deficiency in dGK activity is one of the causes of severe mitochondrial DNA depletion diseases. We used site-directed mutagenesis to introduce the clinically observed genetic alterations in the dGK gene and characterized the recombinant enzymes. The R142K enzyme had very low activity with deoxyguanosine and no activity with deoxyadenosine. The E227K mutant enzyme had unchanged K(m) values for all its substrates but very low V(max) values. C-terminal truncated dGK proteins were inactive. These results may help to define the role of dGK in mitochondrial DNA (mtDNA) precursor synthesis.     

Down-regulation of mitochondrial thymidine kinase 2 and deoxyguanosine kinase by didanosine: Implication for mitochondrial toxicities of anti-HIV nucleoside analogs

Mitochondrial thymidine kinase 2 (TK2) and deoxyguanosine kinase (dGK) catalyze the initial rate limiting phosphorylation of deoxynucleosides and are essential enzymes for mitochondrial function. Chemotherapy using nucleoside analogs is often associated with mitochondrial toxicities. Here we showed that incubation of U2OS cells with didanosine (ddI, 2′,3′-dideoxyinosine), a purine nucleoside analog used in the highly active antiretroviral therapy (HAART), led to selective degradation of both mitochondrial TK2 and dGK while the cytosolic deoxycytidine kinase (dCK) and thymidine kinase 1 (TK1) were not affected. Addition of guanosine to the ddI-treated cells prevented the degradation of mitochondrial TK2 and dGK. The levels of intracellular reactive oxygen species and protein oxidation in ddI-treated and control cells were also measured. The results suggest that down-regulation of mitochondrial TK2 and dGK may be a mechanism of mitochondrial toxicity caused by antiviral and anticancer nucleoside analogs.     

Human cytomegalovirus induces a cellular deoxyguanosine kinase, also interacting with Acyclovir

Abstract A cytosol deoxyguanosine kinase (dGK) is induced in either growing or human cytomegalovirus (HCMV, AD169)-infected human fibroblasts (HEF). Data obtained from polyacrylamide gel electrophoresis, heat inactivation and phosphorylation kinetic experiments proved that these dGKs are identical, but completely differ from HCMV-induced thymidine kinase (TK) or deoxycytidine kinase (dCK). In contrast to TK or dCK, only dGK interacts with Acyclovir ( K _i = 590 μ M). It is suggested that dGK is an important enzyme determining the antiviral activity of Acyclovir.     

Depletion of mitochondrial DNA by down-regulation of deoxyguanosine kinase expression in non-proliferating HeLa cells

Purine deoxyribonucleotides required for mitochondrial DNA replication are either imported from the cytosol or derived from phosphorylation of deoxyadenosine or deoxyguanosine catalyzed by mitochondrial deoxyguanosine kinase (DGUOK). DGUOK deficiency has been linked to mitochondrial DNA depletion syndromes suggesting an important role for this enzyme in dNTP supply. We have generated HeLa cell lines with 20-30% decreased levels of DGUOK mRNA by the expression of small interfering RNAs directed towards the DGUOK mRNA. The cells with decreased expression of the enzyme showed similar levels of mtDNA as control cells when grown exponentially in culture. However, mtDNA levels rapidly decreased in the cells when cell cycle arrest was induced by serum starvation. DNA incorporation of 9-beta-d-arabino-furanosylguanine (araG) was lower in the cells with decreased deoxyguanosine kinase expression, but the total rate of araG phosphorylation was increased in the cells. The increase in araG phosphorylation was shown to be due to increased expression of deoxycytidine kinase. In summary, our findings show that DGUOK is required for mitochondrial DNA replication in resting cells and that small changes in expression of this enzyme may cause mitochondrial DNA depletion. Our data also suggest that alterations in the expression level of DGUOK may induce compensatory changes in the expression of other nucleoside kinases.     

Adenylate kinase 2, a mitochondrial enzyme

The subcellular compartmentalization of the isozymes of ATP: AMP phosphotransferase (adenylate kinase) was analyzed in HeLa cells, RAG cells, and RAG-human hybrids that express human AK-2. In HeLa cells and in the hybrids, human AK-2 was present in a mitochondrial fraction prepared from cell extracts and in mitochondria purified by density gradient centrifugation. Human AK-1 was, as expected, distributed in the soluble cytoplasmic fraction of the cells. The rodent isozymes which are homologous to human AK-1 and AK-2 have been determined.This work was supported by NIH Grants HD 04807-07 and HD 06285-04.     

Low level of mitochondrial deoxyguanosine kinase is the dominant factor in acquired resistance to 9-beta-D-arabinofuranosylguanine cytotoxicity

9-beta-D-arabinofuranosylguanine (Ara-G) is an important and relatively new guanosiue analog with activity in patients with T-cell malignancies. The biochemical and molecular events leading to resistance to Ara-G are not fully understood. Therefore we generated two Ara-G-resistant human MOLT-4 leukemic cell lines with different levels of resistance. The mitochondrial enzyme deoxyguanosine kinase (dGK) and the nuclear/cytosol enzyme deoxycytidine kinase (dCK) are key enzymes in the activation of Ara-G. Decreased levels of dGK protein and mRNA were found in both resistant cell sublines. The activity of dCK was decreased in the subline with higher resistance to Ara-G and these cells were highly cross-resistant to other nucleosides activated by dCK. Increased activity of the mitochondrial enzyme thymidine kinase 2 was observed in both resistant sublines and this could be related to the dGK deficiency. In search for other resistance mechanisms it was found that the resistant cells overexpress the mdr1 gene, while no changes were detected in the levels of multidrug resistance-associated protein 1 through 6, lung resistance-associated protein or topoisomerase IIalpha or IIbeta. Taken together, our findings demonstrate that multiple mechanisms are involved in the acquired resistance to Ara-G. However, low expression of dGK is the most apparent alteration in both resistant cell lines. Partial deficiency of dCK was found in the subline cells with higher resistance to Ara-G. Furthermore, Ara-G may select for high expression of the multidrug resistance (mdr1) which could be a specific resistance mechanism but more likely part of an overall cellular stress response.     

Sphingosine kinase 1 overexpression induces MFN2 fragmentation and alters mitochondrial matrix Ca2+ handling in HeLa cells

Sphingosine kinase 1 (SK1) converts sphingosine to the bioactive lipid sphingosine 1-phosphate (S1P). S1P binds to G-protein-coupled receptors (S1PR_1–5) to regulate cellular events, including Ca²⁺ signaling. The SK1/S1P axis and Ca²⁺ signaling both play important roles in health and disease. In this respect, Ca²⁺ microdomains at the mitochondria-associated endoplasmic reticulum (ER) membranes (MAMs) are of importance in oncogenesis. Mitofusin 2 (MFN2) modulates ER-mitochondria contacts, and dysregulation of MFN2 is associated with malignancies. We show that overexpression of SK1 augments agonist-induced Ca²⁺ release from the ER resulting in increased mitochondrial matrix Ca²⁺. Also, overexpression of SK1 induces MFN2 fragmentation, likely through increased calpain activity. Further, expressing putative calpain-cleaved MFN2 N- and C-terminal fragments increases mitochondrial matrix Ca²⁺ during agonist stimulation, mimicking the SK1 overexpression in cells. Moreover, SK1 overexpression enhances cellular respiration and cell migration. Thus, SK1 regulates MFN2 fragmentation resulting in increased mitochondrial Ca²⁺ and downstream cellular effects.     

Regulation of human placental deoxyguanosine kinase by nucleotides

The activity of deoxyguanosine kinase purified from human placenta was regulated by various nucleotides. dTTP, an activator, only increased the Vmax value of the enzyme. The feedback inhibition by dGTP, dGDP and dGMP were competitive with respect to deoxyguanosine. Both the activation by dTTP and the inhibition by dGTP were reversible.     

Pre-ornithine transcarbamylase. Properties of the cytoplasmic precursor of a mitochondrial matrix enzyme

Biogenesis of the mitochondrial matrix enzyme, ornithine transcarbamylase, has been shown to begin with synthesis on cytoplasmic ribosomes of a precursor, designated pre-ornithine transcarbamylase, which is approximately 4000 daltons larger than its corresponding mitochondrial subunit, followed by post-translational uptake and proteolytic processing of the precursor to its mature counterpart by mitochondria. We now report initial studies on the structure and properties of preornithine transcarbamylase. When this precursor is labeled at the NH2 terminus with N-formyl[35S]methionine and processed by mitochondria, no label is recovered with the mature subunit. This demonstrates that the amino acid extension which is characteristic of the precursor and which is removed during mitochondrial processing is NH2-terminal. This NH2-terminal extension is found intact in two peptides produced by limited proteolysis of the labeled precursor. Moreover, this amino acid extension modifies the behavior of the precursor during immunoprecipitation in the presence of ionic detergents and plays a critical role in facilitating uptake of the precursor by mitochondria.     

Purification of mitochondrial creatine kinase: Two interconvertible forms of the active enzyme

The purification of creatine kinase from beef heart mitochondria is described. The purified enzyme appears as a single band after electrophoresis on SDS gels. Electrophoresis on cellulose acetate followed by staining for creatine kinase activity shows two forms of the enzyme. The slower migrating (m-1) form upon concentration is converted to the more rapidly migrating form (m-2). The reverse conversion occurs if the m-2 is incubated with β-mercaptoethanol. These results are consistent with a reversible oxidation of protein sulfhydryl group (s).     

Mitochondrial and submitochondrial localization of human deoxyguanosine kinase.

Deoxyguanosine kinase and thymidine kinase 2 are responsible for catalysing the first step in the salvage of deoxynucleosides in mitochondria. These enzymes also play an important role in activating several antiviral and anticancer nucleoside analogs, which may lead to unwanted side-effects when the resulting nucleotides are incorporated into the mitochondrial genome. We studied deoxyguanosine kinase in submitochondrial fractions from human placental mitochondria. It was localized in the mitochondrial matrix fraction by Western blotting using a purified polyclonal antibody. This antibody was also used in an immunohistochemical in situ experiment with human embryonic kidney 293 cells, in which the deoxyguanosine kinase antibody colocalized with a mitochondrion-specific fluorescent probe and there was no significant cytosolic staining.     

Divergent enzyme kinetics and structural properties of the two human mitochondrial creatine kinase isoenzymes

The mitochondrial isoenzymes of creatine kinase (MtCK), ubiquitous uMtCK and sarcomeric sMtCK, are key enzymes of oxidative cellular energy metabolism and play an important role in human health and disease. Very little is known about uMtCK in general, or about sMtCK of human origin. Here we have heterologously expressed and purified both human MtCK isoenzymes to perform a biochemical, kinetic and structural characterization. Both isoenzymes occurred as octamers, which can dissociate into dimers. Distinct Stokes' radii of uMtCK and sMtCK in solution were indicative for conformational differences between these equally sized proteins. Both human MtCKs formed 2D-crystals on cardiolipin layers, which revealed further subtle differences in octamer structure and stability. Octameric human sMtCK displayed p4 symmetry with lattice parameters of 145 A, indicating a 'flattening' of the octamer on the phospholipid layer. pH optima and enzyme kinetic constants of the two human isoenzymes were significantly different. A pronounced substrate binding synergism (Kd > Km) was observed for all substrates, but was most pronounced in the forward reaction (PCr production) of uMtCK and led to a significantly lower Km for creatine (1.01 mM) and ATP (0.11 mM) as compared to sMtCK (creatine, 7.31 mM; ATP, 0.68 mM).     

Three-dimensional structure of the complex between the mitochondrial matrix adenylate kinase and its substrate AMP

Crystals of adenylate kinase from beef heart mitochondrial matrix (EC 2.7.4.10) complexed with its substrate AMP were analyzed by X-ray diffraction. The crystal structure was solved by multiple isomorphous replacement and solvent flattening at a resolution of 3.0 A. There are two enzyme-substrate molecules in the asymmetric unit. The resolution was extended to 1.9 A by model building and refinement using simulated annealing. The current R-factor is 28.4%. The model is given as a backbone tracing for residues 5-218. The enzyme can be subdivided into three domains, the relative arrangements of which differ slightly but significantly between the two crystallographically independent molecules. When compared with other adenylate kinase structures, the chain fold is similar but the observed domain arrangement differs grossly, suggesting that large parts of the enzyme move during catalysis. The observed binding site of AMP is described. Its location in conjunction with data from homologous proteins clarifies the nucleotide-binding sites of the adenylate kinases. Previous assignments of these sites derived from X-ray crystallographic and nuclear magnetic resonance analyses are discussed.     

Hypoxia induces epidermal keratinocyte matrix metalloproteinase-9 secretion via the protein kinase C pathway

Hypoxia promotes keratinocyte migration on wound bed connective tissues and is a profound biological signal that transforms a basal keratinocyte, destined to differentiate, into a motile cell that is essential for re-epithelialization. In this study, we examined the effect of hypoxia on keratinocyte-derived collagenases associated with keratinocyte migration. Cells plated on various connective tissue matrices under normoxic and hypoxic conditions, demonstrated a two-fold increase in the 92 kDa, type IV collagenase (MMP-9) when examined by quantitative zymography and ELISA. Western blotting and ELISA demonstrated a two-fold increase in tissue inhibitor of metalloproteinase (TIMP-1), an enzyme that binds to MMP-9 and inhibits its activity. The hypoxia-induced increase in cell motility could be inhibited by a neutralizing antibody to MMP-9. Northern blotting demonstrated that MMP-9 and TIMP-1 mRNA increased 2.5- to 4-fold, 2-12 h after the cells were made hypoxic. The hypoxia-induced changes in MMP-9 and TIMP-1 were inhibited by staurosporine and bisindolylmaleimide, inhibitors of protein kinase C (PKC), but not by inhibitors of tyrosine phosphorylation and the mitogen-activated protein kinase pathway. Inhibition of PKC also inhibited hypoxia-induced keratinocyte migration on type I collagen. These data provide evidence that hypoxia-induced keratinocyte migration is mediated by increased cellular secretion of MMP-9 via the PKC pathway.     

A highly basic N-terminal extension of the mitochondrial matrix enzyme ornithine transcarbamylase from rat liver

We have deduced the amino acid sequence of the N-terminal leader peptide of the mitochondrial enzyme ornithine transcarbamylase from a cDNA clone obtained from a rat liver cDNA library. The sequence is remarkable in being highly basic, having 4 arginine, 3 lysine and 1 histidine with no acidic residues in a total of 32 residues. The leader sequence has no extensive hydrophobic stretches, has 72% homology with the leader peptide of human ornithine transcarbamylase [1], and in terms of its basic character resembles the N-terminal extensions on a number of fungal mitochondrial [2-5] and pea chloroplast [6] proteins. Thus the basic nature of these leader peptides may constitute the signal for mitochondrial import.     

Purification and characterization of human mitochondrial creatine kinase. A single enzyme form

Purification of human mitochondrial creatine kinase has been difficult and procedures that were highly successful in purifying canine enzyme failed for human mitochondrial creatine kinase. In the present study, we employed ultracentrifugation to remove the lipid, urea to prevent aggregation, followed by a final step of chromatofocusing which yielded a preparation of human mitochondrial creatine kinase with a specific enzyme activity of greater than 400 IU/mg. Biochemical and immunological characterization showed the preparation to be highly pure and free of even trace amounts of other creatine kinase isoenzymes. Antiserum specific for mitochondrial creatine kinase was developed which exhibited no cross-reactivity to cytosolic creatine kinase and mitochondrial creatine kinase did not cross-react with antiserum to the cytosolic forms. Marked differences were noted, both biochemically and immunologically, between mitochondrial creatine kinase and the cytosolic forms. Human mitochondrial creatine kinase was shown to have a molecular weight of around 82,000 and to be composed of two subunits of equal molecular weights around 41,000. Aggregates of mitochondrial creatine kinase were observed with molecular weights of around 200,000 in the absence of urea or if isolated from material after having undergone proteolysis. Isolation from fresh material or in the presence of urea inhibited aggregate formation for both canine and human mitochondrial creatine kinase. Despite claims of several investigators that mitochondrial creatine kinase exhibits two to three forms with varying molecular weights, our data indicate a single enzyme form made up of a subunit with a molecular weight of 41,000 and the high molecular weight aggregates appear to be induced artifacts. A radioimmunoassay was developed for human mitochondrial creatine kinase which, with appropriate modifications, should detect mitochondrial creatine kinase in human plasma.     

Mutations within the putative active site of heterodimeric deoxyguanosine kinase block the allosteric activation of the deoxyadenosine kinase subunit

Replacement of the Asp-84 residue of the deoxyguanosine kinase subunit of the tandem deoxyadenosine kinase/ deoxyguanosine kinase (dAK/dGK) from Lactobacillus acidophilus R-26 by Ala, Asn, or Glu produced increased Km values for deoxyguanosine on dGK. However, it did not seem to affect the binding of Mg-ATP. The Asp-84 dGK replacements had no apparent effect on the binding of deoxyadenosine by dAK. However, the mutant dGKs were no longer inhibited by dGTP, normally a potent distal endproduct inhibitor of dGK. Moreover, the allosteric activation of dAK activity by dGTP or dGuo was lost in the modified heterodimeric dAK/dGK enzyme. Therefore, it seems very likely that Asp-84 participates in dGuo binding at the active site of the dGK subunit of dAK/dGK from Lactobacillus acidophilus R-26.