Purification and characterization of wild-type and mutant TK1 type kinases from Caenorhabditis elegans.

Caenorhabditis elegans has a single deoxynucleoside kinase-like gene. The sequence is similar to that of human TK1, but besides accepting thymidine as a substrate, the C. elegans TK1 (CeTK1) also phosphorylates deoxyguanosine. In contrast to human TK1, the CeTK1 exclusively exists as a dimer with a molecular mass of approximately 60 kDa, even if incubated with ATP. Incubation with ATP induces a transition into a more active enzyme with a higher kcat but unchanged Km. This activation only occurs at an enzyme concentration in the incubation buffer of 0.5 micro g/ml (8.42 nM) or higher. C-terminal deletion of the enzyme results in lower catalytic efficiency and stability.

The role of mitochondrial dNTP levels in cells with reduced TK2 activity.

Both the nuclear and mitochondrial DNA (mtDNA) depend on separate balanced pools of dNTPs for correct function of DNA replication and repair of DNA damage. Import of dNTPs from the cytosolic compartment to the mitochondria has been suggested to have the potential of rectifying a mitochondrial dNTP imbalance. Reduced TK2 activity has been demonstrated to result in mitochondrial dNTP imbalance and consequently mutations of mtDNA in non-dividing cells. In this study, the consequences of a reduced thymidine kinase 2 (TK2) activity were measured in proliferating HeLa cells, on both whole-cell as well as mitochondrial dNTP levels. With the exception of increased mitochondrial dCTP level no significant difference was found in cells with reduced TK2 activity. Our results suggest that import of cytosolic dNTPs in mitochondria of proliferating cells can compensate a TK2 induced imbalance of the mitochondrial dNTP pool.

Thymidine kinase diversity in bacteria.

Thymidine kinases (TKs) appear to be almost ubiquitous and are found in nearly all prokaryotes, eukaryotes, and several viruses. They are the key enzymes in thymidine salvage and activation of several anti-cancer and antiviral drugs. We show that bacterial TKs can be subdivided into 2 groups. The TKs from Gram-positive bacteria are more closely related to the eukaryotic TK1 enzymes than are TKs from Gram-negative bacteria.

Effect of C-terminal of human cytosolic thymidine kinase (TK1) on in vitro stability and enzymatic properties.

Thymidine kinase (TK1) is a key enzyme in the salvage pathway of nucleotide metabolism and catalyzes the first rate-limiting step in the synthesis of dTTP, transfer of a gamma-phosphate group from a nucleoside triphosphate to the 5'-hydroxyl group of thymidine, thus forming dTMP. TK1 is cytosolic and its activity fluctuates during cell cycle coinciding with the DNA synthesis rate and disappears during mitosis. This fluctuation is important for providing a balanced supply of dTTP for DNA replication.The cell cycle specific activity of TK1 is regulated at the transcriptional level, but posttranslational mechanisms seem to play an important role for the level of functional TK1 protein as well. Thus, the C-terminal of TK1 is known to be essential for the specific degradation of the enzyme at the G2/M phase. In this work, we have studied the effect of deletion of the C-terminal 20, 40, and 44 amino acids of TK1 on in vitro stability, oligomerization, and enzyme kinetics. We found that deletion of the C-terminal fold markedly increased the stability as well as the catalytic activity.

Induction of thymidine kinases in phytohaemagglutinin-stimulated human lymphocytes.

Thymidine kinase (ATP:thymidine 5'-phosphotransferase, EC 2.7.1.21) activity increased approx. 200-fold during transformation of human lymphocytes with phytohaemagglutinin. Two peaks of thymidine kinase (peak IS and peak IIS) were separated from stimulated lymphocytes on DEAE-Sephadex. The activity in peak IS was 20-fold the activity in peak IIS. Only one peak was obtained from normal lymphocytes (peak IIN). The elution volume of this peak was identical to that of peak IIS. The Km values for ATP were 1 mM for peak IS, 0.2 mM for peak IIS and 0.3 mM for peak IIN. 90 micronM dTTP gave 50% inhibition of the activity in peak IS, while the same inhibition of the activities in peak IIS and peak IIN was obtained with only 15 micronM dTTP. Km for thymidine was about 6 micronM for peak IS. The kinetic relation between thymidine and the activity in peak IIS was complex but very similar to that of peak IIN. It is suggested, that a new form of thymidine kinase appears in the lymphocytes due to phytohaemagglutinin stimulation.

Different affinity of the two forms of human cytosolic thymidine kinase towards pyrimidine analogs.

Recent results showed that ATP enables a kinetically slow shift from a low affinity form to a high affinity form of human cytosolic thymidine kinase (TK1), as reflected by the respective apparent Km values for thymidine of 15 microM and 0.7 microM. The shift is dependent on the concentration of enzyme protein, and calculations indicate that the low affinity form is predominant in G1 cells, and the high affinity form is predominant in S-phase cells. Here, we report that the two forms of TK1 differ manyfold in affinity to the substrate ATP, to the inhibitor dTTP and to various analogs of thymidine substituted in the pyrimidine or sugar. Furthermore, the kinetic reaction mechanisms suggest that the nucleoside analog. 3'-azidothymidine, used for treatment of infections with human immune deficiency virus (HIV), is not a substrate for the low affinity form of TK1.

Identification of residues involved in the specificity and regulation of the highly efficient multisubstrate deoxyribonucleoside kinase from Drosophila melanogaster.

In contrast to all known deoxyribonucleoside kinases, a single highly efficient deoxyribonucleoside kinase from Drosophila melanogaster (Dm-dNK) is able to phosphorylate all precursor nucleosides for DNA synthesis. Dm-dNK was mutated in vitro by high-frequency random mutagenesis, expressed in the thymidine kinase-deficient Escherichia coli strain KY895 and clones were selected for sensitivity to the nucleoside analogs 1-beta-d-arabinofuranosylcytosine (AraC, Cytarabine), 3'-azido-2', 3'-dideoxythymidine (AZT, Zidovudine, Retrovir, 2', 3'-dideoxyadenosine (ddA) and 2',3'-dideoxycytidine (ddC, Zalcitabine, Hivid. Thirteen mutants with increased sensitivity compared to the wild-type Dm-dNK were isolated from a relatively small pool of less than 10,000 clones. Eight mutant Dm-dNKs increased the sensitivity of KY895 to more than one analog, and two of these mutants even to all four nucleoside analogs. Surprisingly, the mutations did not map to the five regions which are highly conserved among deoxyribonucleoside kinases. The molecular background of improved sensitivity was characterized for the double-mutant MuD (N45D, N64D), where the LD(100) value of transformed KY895 decreased 316-fold for AZT and more than 11-fold for ddC when compared to wild-type Dm-dNK. Purified recombinant MuD displayed higher K(m) values for the native substrates than wild-type Dm-dNK and the V(max) values were substantially lower. On the other hand, the K(m) and V(max) values for AZT and the K(m) value for ddC were nearly unchanged between MuD and wild-type Dm-dNK. Additionally, a decrease in feedback inhibition of MuD by thymidine triphosphate (TTP) was found. This study demonstrates how high-frequency mutagenesis combined with a parallel selection for desired properties provides an insight into the structure-function relationships of the multisubstrate kinase from D. melanogaster. At the same time these mutant enzymes exhibit properties useful in biotechnological and medical applications.

Human thymidine kinase 2: molecular cloning and characterisation of the enzyme activity with antiviral and cytostatic nucleoside substrates.

Based on amino acid sequence information from purified mitochondrial thymidine kinase (TK2), a cDNA of 1930 bp was cloned, containing an open reading frame encoding 232 amino acid residues starting with the N-terminal sequence determined from the native human protein preparation. Northern blot analysis with the cDNA coding region demonstrated several TK2 mRNAs, with 2 and 4 kb forms present in many tissues. We also characterised N-terminally truncated (starting at position 18) human TK2 with pharmacologically important antiviral and cytostatic nucleoside analogues. Results were highly similar to those with the native TK2 preparation. The anti-leukaemic drug arabinosyl cytosine is phosphorylated. The antitumour drug difluorodeoxycytidine and its metabolite difluorodeoxyuridine are good substrates, with K(m) values of 66 and 29 microM, respectively, and a relative Vmax of 0.6 compared to that of thymidine. Negative cooperativity was found with thymidine and the anti-HIV drug 3'-azidothymidine, but the reaction followed Michaelis-Menten kinetics with deoxycytidine, arabinosyl cytosine, and arabinosyl thymine. The results demonstrate a broad substrate specificity and complex kinetics, and suggest a role for TK2 in the activation of chemotherapeutic nucleoside analogues.

Gene technology-based antimetabolite design: the use of an in vitro protein expression system to facilitate antimetabolite design for virally-induced human diseases and malignant conditions.

A precondition for the chemotherapeutic treatment of a variety of virally-induced human diseases and malignant conditions is a highly selective interaction of the drug molecule to be used with it's biological target. To ensure the development of novel, effective drugs, it is essential that the biological target is well characterised with regard to it's structure and activity. Such characterisation relies upon adequate amounts of pure target being available. One of the most important enzymatic importers for antimetabolites is the enzyme thymidine kinase. In this article an in vitro protein expression system is described which facilitates the production of milligram amounts of pure and biologically active thymidine kinase, from a number of important biological sources. Results have shown that the in vitro produced enzyme has the exact biochemical propeties of the in vivo enzyme. Thus the in vitro protein expression system is an ideal vechicle to facilitate an in depth investigation of the enzyme's biological properties.

Collateral sensitivity to gemcitabine (2',2'-difluorodeoxycytidine) and cytosine arabinoside of daunorubicin- and VM-26-resistant variants of human small cell lung cancer cell lines.

Multidrug resistance (MDR), characterized by a cross-resistance to many natural toxin-related compounds, may be caused either by overexpression of a drug efflux pump such as P-glycoprotein, (P-gP), multidrug resistance proteins MRP1-3, or BCRP/MXR or, in the case of DNA topoisomerase II active drugs, by a decrease in the enzymatic activity of the target molecule termed altered topoisomerase MDR (at-MDR). However, human small cell lung carcinoma (SCLC) cell lines showed a collateral sensitivity to 2',2'-difluorodeoxycytidine (gemcitabine, dFdC) and 1-beta-D-arabinofuranosylcytosine (ara-C). H69/DAU, a daunorubicin (DAU)-resistant variant of H69 with a P-gP overexpression, and NYH/VM, a VM-26 (teniposide)-resistant variant of NYH with an at-MDR, were both 2-fold more sensitive to gemcitabine and 7- and 2-fold more sensitive to ara-C, respectively. MDR variants had a 4.3- and 2.0-fold increased activity of deoxycytidine kinase (dCK), respectively. dCK catalyzes the first rate-limiting activation step of both gemcitabine and ara-C. In addition, deoxycytidine deaminase, responsible for inactivation of dFdC and ara-C, was 9.0-fold lower in H69/DAU cells. The level of thymidine kinase 2, a mitochondrial enzyme that can also phosphorylate deoxycytidine and gemcitabine, was not significantly different between the variants. These differences most likely caused an increased accumulation of the active metabolites (dFdCTP, 2.1- and 1.6-fold in NYH/VM and H69/DAU cells, respectively) and of ara-CTP (1.3-fold in NYH/VM cells). Ara-CTP accumulation was not detectable in either H69 variant. The pools of all ribonucleoside and deoxyribonucleoside triphosphates were at least 3- to 4-fold higher in the NYH variants compared to the H69 variants; for dCTP and dGTP this difference was even larger. The higher ribonucleotide pools might explain the >10-fold higher accumulation of dFdCTP in NYH compared to H69 variants. Since dCTP is low, H69 cells might not need a high ara-CTP accumulation to inhibit DNA polymerase. This might be related to the lack of ara-CTP in H69 variants. In addition, the increased CTP, ATP, and UTP pools in the MDR variants might explain the increased ara-CTP and dFdCTP accumulation. In conclusion, the MDR variants of the human SCLC cell lines were collaterally sensitive due to an increased dCK activity, and consequently an increased ara-CTP and dFdCTP accumulation.

Thymidine kinase in human leukemia--expression of three isoenzyme variants in six patients with chronic myelocytic leukemia.

In six patients with untreated, chronic myelocytic leukemia (CML), the dominating thymidine kinase (TK) activity was compared with the fetal form, TK 1, from mitogen stimulated and the adult form TK 2 from unstimulated normal human lymphocytes, and with TK-1-onc, TK-3-onc and TK-4-onc. This was done in human acute, myelocytic and monocytic leukemias, using the combined thymidine/dTTP enzyme kinetics for isoenzyme characterization. TK-1-onc was found in one, TK-2-onc in two and TK-3-onc in three CML patients. The suffix -onc indicates the difference in ATP kinetics and molecular weights between the normal and the leukemic thymidine kinases. A possible relation between the isoenzyme forms and the types of leukemias is discussed.

Thymidine kinase in human leukemia. Expression of the lymphoblastic isoenzyme in three patients with acute myelocytic leukemia.

The dominating thymidine kinase activity in mononuclear white blood cells from three patients with untreated acute myelocytic leukemia (AML) was compared with TK 1 from phytohemagglutinin-stimulated and TK 2 from unstimulated, normal lymphocytes. The enzyme activity in the AML cells and the stimulated lymphocytes was found to be in the same range. Regarding the combined thymidine and dTTP kinetics, the enzymes from the three AML patients resembled TK 1, but the ATP kinetics were different and the molecular weights were lower, as previously found for thymidine kinases from other leukemic cells. Therefore, the designation TK-1-onc is suggested for the thymidine kinases from the AML cells.

Thymidine kinase isoenzymes in human acute and chronic lymphatic leukemia.

The dominating thymidine kinase isoenzyme was examined in mononuclear leucocytes from two patients, one with acute and one with chronic lymphatic leukemia. The two isoenzymes exhibited Michaelis-Menten substrate kinetics with ATP and cooperative inhibition kinetics with dTTP. The substrate kinetics with thymidine were different. According to the enzymatic properties the isoenzymes from the acute and chronic lymphatic cells were designated TK 3 and TK 4, respectively. Comparison with the isoenzymes in normal lymphocytes (TK 1, TK 2) and in acute monocytic leukemic cells (TK 3, TK 4) indicated the existence of three thymidine kinase isoenzymes in human leukemic cells which differed from the two isoenzymes in normal human lymphocytes.

Overexpression of human thymidine kinase mRNA without corresponding enzymatic activity in patients with chronic lymphatic leukemia.

The level of cytosolic thymidine kinase (TK1) mRNA in lymphocytes from six healthy people and in lymphocytes from five patients with untreated chronic lymphatic leukemia (CLL) was determined with competitive polymerase chain reaction (competitive PCR). Using this procedure we have shown that in patients with CLL, there is an overexpression of TK1 mRNA without corresponding enzymatic activity. The TK1 mRNA level is approximately 100-fold higher in lymphocytes from CLL patients than in lymphocytes from healthy persons. A high level of TK1 mRNA without corresponding enzyme activity may indicate a defect in the processing of the enzyme. This may disturb the cells' normal feedback system and thereby influence the development of malignant conditions.

Increased ratio between deoxycytidine kinase and thymidine kinase 2 in CLL lymphocytes compared to normal lymphocytes.

Deoxycytidine kinase (dCK) is important in the 5'-phosphorylation of deoxynucleoside analogs. Like dCK, thymidine kinase 2 (TK2) catalyzes the initial step of the phosphorylation of dcyd to dCTP. Thymidine is a strong inhibitor of the dCK activity of TK2. We examined the ratio of the dcyd phosphorylation carried out by dCK and by TK2 (dCK/TK2-dcyd) in lymphocytes from CLL patients and from donors. In the CLL lymphocytes we found a 3.5-fold average increase. Therefore, we conclude that addition of thymidine in the treatment of CLL with deoxynucleoside analogs will not be of any advantage. Furthermore, our results can explain earlier findings in CML and AML lymphocytes where the ara-C phosphorylation was twice the dcyd phosphorylation.

X-ray and UV-radiation sensitivity of circulating lymphocytes in multiple epidermal cancer in relation to previous radiation exposure.

The cellular sensitivity to X rays (200 kV, 16 mA) and UV radiation (254 nm) was examined in lymphocytes from three groups of patients with multiple epidermal malignant tumors, selected by their clinical history of carcinogenesis. Eight patients previously exposed to low energy ionizing radiation (less than or equal to 12 kV) had an increased cellular sensitivity to UV radiation as well as X rays compared with 24 age and sex matched controls. This indicates the existence of a cellular cross-sensitivity to UV radiation and ionizing radiation not previously established for human cells. In contrast six patients previously exposed to high energy ionizing radiation (between 25 and 170 kV) had normal cellular response to both UV radiation and X rays, indicating a different biologic effect of low and high energy ionizing radiation. In the third group of patients, previously exposed to therapeutic UV radiation/excess sunlight, the lymphocytes had a normal response to X rays, but an increased sensitivity to UV radiation. The possibility of evaluating the individual risk at radiation exposure is suggested.

Human thymidine kinase 1. Regulation in normal and malignant cells.

In mammalian cells, salvage pathway phosphorylation of thymidine is catalyzed by two thymidine kinases: the cell-cycle regulated cytoplasmic TK1 and the constitutively expressed mitochondrial TK2. Since TK1 is virtually absent in non-dividing cells, TK2 is probably the only thymidine kinase present in these cells. In cellular metabolism, TK1 and TK2 presumably serve to maintain sufficient dTTP for DNA replication and repair. TK1 purified from phytohemagglutinin-stimulated human lymphocytes is a dimer in the absence and a tetramer in the presence of ATP. In addition to the molecular weight transition, incubation with ATP at 4 degrees C or storage with ATP induces a reversible, enzyme concentration-dependent, kinetically slow transition from a low to a high affinity form of TK1, with Km values of 14 microM and 0.5 microM, respectively. This affinity difference implies that at cellular thymidine concentrations, the difference in catalytic activity between the two TK1 forms will be 3-5-fold. Calculations of cellular TK1 concentration suggested that the low affinity dimer form was dominant in G0/G1 cells and the high affinity tetramer form in S-phase cells. Hence, the transition may serve to fine-tune the cell-cycle regulation of thymidine kinase activity on the post-translational level. To study the ATP effect on the molecular level, an IPTG inducible T7 RNA polymerase-dependent expression system for the entire human TK1 polypeptide in E. coli was established. The recombinant TK1 has the same subunit mass and specific activity as the native enzyme. However, the recombinant TK1 solely displayed the kinetics of the high affinity form, with Km values of 0.3-0.4 microM regardless of pre-exposure to ATP, indicating that the ATP effect may be dependent on post-translational modifications absent in E. coli. Surprisingly, we did not observe any effect of ATP on TK1 purified from bone-marrow cells from a patient with acute monocytic leukemia (AMOL). Furthermore, the Km values of TK1 from these cells were 45 microM for the ATP-free enzyme and 65 microM for the ATP-incubated enzyme. With TK1 purified from HL-60 cells, we obtained the same pattern and kinetic values as for TK1 from lymphocytes. In the light of the results with the recombinant TK1, we presume that the lack of ATP effect and very high Km values observed for the AMOL TK1 may be due to changes in post-translational regulatory mechanisms in acute monocytic cells.

Normal sister chromatid exchange in lymphocytes from patients with multiple epidermal cancer?

Exposure to toxins in the environment and due to personal habits, e.g., tobacco smoking, may increase the rate of spontaneous sister chromatid exchange (SCE). The SCE in lymphocytes from a group of 31 patients with multiple epidermal cancer, who in the past had been exposed to various skin carcinogens, as a whole exceeded that of a control group--matched by sex, age, and smoking habits--but the difference was not statistically significant (p = 0.08). The individual SCE in these patients was also statistically independent of the nature of the carcinogenic exposure. We were unable to detect correlations between the SCE and UVC-radiation induced DNA synthesis, UVC-radiation tolerance, or rate of X-ray damage repair. This suggests that the molecular mechanisms involved in SCE induction and in repair of radiation damage are basically independent.

Thymidine in the micromolar range promotes rejoining of UVC-induced DNA strand breaks and prevents azidothymidine from inhibiting the rejoining in quiescent human lymphocytes.

The effect and inter-individual variation in the effect of exogenously added deoxynucleosides (2 x 10(-6) M) on rejoining of UVC-induced DNA strand breaks was examined in quiescent human lymphocytes from 25 healthy persons. Thymidine at concentrations below 2 x 10(-6) M, effectively and with statistically extreme significance, increased rejoining of UVC-induced DNA strand breaks in the lymphocytes of every one of the 25 persons tested (p < 0.0001, Wilcoxon's signed ranks test). The mean stimulation after 20 h of postirradiation repair was 48% (range 18-78%) with an inter-individual variation of 30% (coefficient of variation, CV). Deoxyguanosine stimulated rejoining in 16, but inhibited in three of 19 test persons (mean stimulation 28%, range -31 to 71%). The stimulating effect of deoxyguanosine was also extremely significant (p < 0.0004). Deoxycytidine and deoxyadenosine stimulated rejoining in some persons and inhibited it in others, and without statistical significance (p values above 0.5). The stimulating effect of thymidine was significantly inhibited by deoxycytidine (p < 0.05, n = 12) whereas deoxyguanosine neither promoted or inhibited the stimulation by thymidine (p = 1, n = 12). Rejoining of DNA strand breaks induced by methyl methanesulfonate did not appear significantly stimulated or inhibited by any of the four deoxynucleosides. Finally, the inhibiting effect of azidothymidine (AZT) on rejoining of UVC-induced DNA strand breaks was nullified by the addition of thymidine. In three donors examined, 10(-4) M AZT inhibited the rejoining by about 40-50%. The presence of less than 10(-5) M thymidine reduced the level of UVC-induced DNA strand breaks to below the level in control lymphocytes allowed to repair without AZT. These results indicate that among the four deoxynucleoside triphosphates, dTTP has a crucial role on the repair of UVC-induced DNA damage in quiescent lymphocytes. The results also indicate that an expansion of the dTTP pool may counteract the inhibiting effect of AZT on DNA repair in quiescent lymphocytes.

Substrate/inhibitor specificities of human deoxycytidine kinase (dCK) and thymidine kinases (TK1 and TK2).

Substrate/inhibitor specificities of nucleoside analogues with modified sugar moieties toward highly purified deoxycytidine kinase (dCK) and thymidine kinases (TK1 and TK2) from human leukemic spleen have been examined. Substrate activities of cytosine nucleosides vs dCK were as follows: 2'-fluoro-dC > 2'-O-methyl-C > araC > 2'-fluoro-2'-deoxy-araC > 3'-O-methyl-dC = 3'-fluoro-2',3'-ddC > cytosine beta-L-riboside > 2',3'-ddC > C = 1-(4-hydroxy-1,2,-butadienyl)-cytosine (cytalene) = 2'-azido-dC. Modified purine nucleosides were only feeble substrates: ara-A > 2'-fluoro-2',3'-dideoxy-araA = 2'-O-methyl-A. With TK1 and TK2, similar sugar-modified analogues of dU and dT were feeble substrates. Surprisingly alpha-dT was a relatively good substrate, as well some beta-L-ribonucleo-sides. Several 5'-substituted analogues of dC were good non-substrate inhibitors of dCK and, to a lesser extent, of TK2. The overall data are relevant to the role of these enzymes in "activation" (by phosporylation) of nucleoside analogues with antiviral and antitumor activities.