A Nuclear Spin Selective Control over the DNA Repair Key Enzyme Might Renovate the Cancer–Fight Paradigm. DNA Polymerase Beta to Engage with a Magnetic Isotope Effect

ions replaced with the stable 43 Ca 2+ isotopes inside the enzyme catalytic sites. The isotope mentioned is the only paramagnetic species of the Calcium isotopic set with a 0.135 natural abundance value and the negative 7/2 nuclear spin providing a nuclear magnetic moment equal to 1.317 Bohr magnetons. As compared to the Mg/ 40 Ca substitution, a 2.25-fold enzyme inhibition has been shown to provethe 43 Ca-MIE dependent mode of the catalysis turning down.An ion-radical mechanism based on the singlet – triplet conversion of the enzyme generated intermediates (ion-radical pairs) is found to be engaged once the paramagnetic metal isotope involved into the catalysis studied.The MIE promotes a primary reaction in DNA synthesis constituting in electron transfer between the ion – radical forming partners, [Ca(H2O)n 2+ ] and [Ca 2+ (dNTP)]. Once the metal isotope substitution takes place inside just one of two DNA Polymerase Beta catalytic sites, a consequent 43 Ca – promoted inhibition leads to a residual synthesis of shorted DNA fragments that counts 25 – 35 nucleotides in length contrasting with the 180n – 210n DNA produced by either intact or 40 Ca – loaded polymerase. Being occurred simultaneously with a marked MIE – promoted enzyme inhibition, this fact itself makes possible to consider these short (“size-invalid”) DNA segments hardly efficient in the DNA base – excision repair. The latter is a survival factor in leukemic cells where the DNApolβ was found overexpressed. That supports a standpoint considering theDNApolβ a legitimate target for antitumor agents since its inhibition deprives the malignant cell from a DNA base – excision repair in neoplasma. A possible trend making role of these data in the current developments on a novel concept establishing chemical background for cancer therapies is in a focus.

Taking into account a known fact of the cell/tumorspecific structural diversity of the DNApolB superfamily (Beard &Willson, 2006;Rechkunova&Lavrik, 2010), some efforts were made to employ DNApolB inhibitors as the anticancer agents (Matsubara et al, 20007;Ljungmann, 2009;Mizushina, 2009). In other words, enzymes of this group might be considered the legitimate targets for cytostatics as long as the latters are sufficient to meet such pharmacophore related requirements as the low toxicity, high selectivity and the efficiency of the enzyme inhibitory activity (Martin et al, 2010;Sanjeev et al, 2011).
One of the obstacles on a route of the DNApolBtargeting pharmacophores administration is their toxicity (Matsubara et al, 2007;Mizushina, 2009). Being Mg 2+ -dependent matalloenzymes (two Mg 2+ ions loosely coordinated inside two separate catalytic sites), DNApolB species were found to be controllable by a so called magnetic isotope effects (MIE) of the nontoxic bivalent metal ions, 25 Mg, and 67 Zn (Buchachenko et al, 2013). This nuclear spin selective path of the enzyme activity regulation has been revealed for a number of metalloenzymes (Buchachenko, 2009;Buchachenko et al, 2010Buchachenko et al, -2012. The mechanism beyond involves a singlettriplet conversion of the ionradical intermediates formed during the enzymatic intermolecular transfer of phosphate (Buchachenko, 2009;Buchachenko et al, 2012). As a result, these spin -possessing (magnetic) ions, 25 Mg 2+ and 67 Zn 2+ , work as the far better suppressors for DNApolB reaction compared to the spinless (nonmagnetic) ions of the very same metals (Buchachenko et al, 2013). Noteworthy, these inhibitorsare no doubt the nontoxic ones (Orlova et al, 2012).
Furthermore, the endoosmotic interenzyme substitution of endogenous magnesium with other magnetic bivalent metal ions like 43 Ca 2+ or 67 Zn 2+ leads to an essential MIE expressed in several kinases directed phosphate transfer enzymatic reactionsin both pure enzyme tests and leukemia cell cultures (Buchachenko et al, 2011;Orlova et al, 2012). In a later case, some porphyrinebased amphililicnanocationites were used to deliver the magnetic ions into the cell, regardless the ion concentration gradient vector (Amirshahi et al, 2008;Kuznetsov et al, 2010;Orlova et al, 2012).
It would be safe to say, therefore, that the MIEpromoting pharmacophores could engage the DNApolB molecular targets in anticancer therapies. This itself requires a firm biochemical, to be exact -chemico-enzymological, background.
Since the blasticleukemias, acute myeloblast ones (AML) including, are known for their DNApolB overexpression (Ljungmann, 2009(Ljungmann, , 2010Bukhvostov et al, 2013;Carceres -Corties, 2013), it makes sense to investigate the MIE pharmacological potential using enzyme isolated from these peculiar cancer cells. Earlier, we have purified and detailcharacterized the unique tumorspecific DNApolB overexpressed in AML/HL-60 cells . In a present study, we're evaluating the 43 Ca-MIE impact on this enzyme function with a special respect to further perspectives in antileukemia therapy preclinical developments.

Enzyme preincubation loading with exogenous Ca isotopes
To convert 40 CaO and 43 CaO into chloride salts, a conventional acidic treatment has been performed first (Buchachenko et al, 2011). Once the isotopic-pure salt sample solution adjusted to a stock 1.0 M concentration by the flame atomic absorption spectrophotometry (LQ600 AAA System, Shimadzu Corp, Japan), a routine incubation mixture has been composed as described by Buchachenko et al, 2013. This mixture was employed then to replace (endo -O c t o b e r 1 8 , 2 0 1 3 osmotic ion substitution, EOIS) the enzyme endogenous magnesium with an exogenous Ca 2+ under the following conditions: 15 mMTris -HCl (pH 8.0) / 20 mM CaCl2 / 1.5 mM EDTA / 60 -80 microgram pure enzyme per 1.0 mL / +37 0 C, 2 hrs (Buchachenko et al, 2013). To monitor the enzymebound Calcium level values estimated in pg Ca per 1.0 mg protein, the fiberglass filter retained acetoneprecipitated pellets (acetone washed, 100 mL/cm 3 ) were analyzed by X-ray fluorescent spectrometry in SL420 XF Analytical System, Bruker GmbH, Germany (Buchachenko et al, 2011;Svistunov et al, 2013). For protein measurements, a Bradfordlike routine colorimetric procedure has been applied (Katoch, 2011;Bukhvostov et al, 2013).

DNApolB catalytic activity measurements
To detect and quantify the enzymespecific catalytic activity, our original technique has been employed . The resulted activity values were expressed in amounts of the labeled DNA precursor incorporated into the enzymedirected nascent single strand DNA sequences corrected to 1.0 mg pure enzyme, i.e. [ 3 H]cpmDNA/mg protein. 220LX Liquid Scintillation Counter (Wallac OY, Finland) employed. The protein ultramicroamounts were estimated according to Itzhaki& Gill, 1964. The postincubation mixtures were subjected to a quantitative extraction of the DNA ultramicroamounts using an AccuPrep Genomic DNA Extraction Kit (Bioneer, Korea) as described by Mikami et al, 2004, for further DNApolB activity and DNA size detection electrophoresis tests. In diluted solutions, the DNA ultramicroamount measurements were performed according to Bukhvostov et al, 2013.

DNA electrophoresis
A standard 1.7% agarose gel electrophoretic technique has been employed (Reichman et al, 1957, modification: Bukhvostov et al, 2013. Prior to the samples supply, the routine colorimetric DNA measurements were performed (Burton, 1956). To observe an immediate result, the ethidium bromide treated unfixed gels were studied in the ImageQuant LAS4000 UV Scanner -Visualizer, GE HealthCare Life Sciences, Inc., USA . The gel [3H]radioautographs were registered as described in Katoch, 2011.

Statistics
The experiment data repetition preciseness as well as a significance of the experiment / control data differences were elucidated by a nonparametric standard technique for n lower than 6 (Brown & Hollander, 2007). The data were processed in HP700E analytical module (Hewlett Packard, Inc., USA) using the VaryLab-6 software package (Litekh Ltd, Russia).

RESULTS AND DISCUSSION
In a number of previously conducted studies, a universal mode of the nuclear spin dependent effects (MIE) expressing in several metalloenzyme directed reactions was clearly proven and specified (Amirshahi et al, 2008;Buchachenko et al, 2009Buchachenko et al, -2013Kuznetsov et al, 2010;Orlova et al, 2012). Thus, the DFT-method provided simulation models show why magnetic and nonmagnetic bivalent metal ions act differently once involved into the phosphate carrying enzymatic catalysis (Buchachenko et al, 2010(Buchachenko et al, , 2012(Buchachenko et al, , 2013. A wide diversity of the phosphate transferring metalloenzymes, including the HL-60 DNApolB, was found operable (controllable) by those above mentioned magnetic isotopes (Buchachenko et al, 2012(Buchachenko et al, , 2013 which makes a scheme presented in Figure 1 a quite legitimate way to interpret the 43 Ca-MIE/DNApolB interaction we're about to describe. Being a selfsufficient kind of illustration, this scheme ( Figure 1) deserves, nonetheless, a brief comment. The magnetic ( 43 Ca) nuclei induced singlettriplet conversion of the ionradical pair is a key element of this quantum mechanical mechanism (Buchachenko, 2009;Buchachenko et al, 2012). Consequently, this nonnucleophilic, very fast ionradical reaction is to desynchronize the orchestrated work of two separate metalpossessing DNApolB catalytic sites. This, in turn, might deprive one of them of an appropriate functional interaction with a second one (Buchachenko et al, 2013). This statement is in accordance with the data on unequal ion accessibility of different catalytic sites in enzyme studied (Cowan, 2002;Kornberg & Baker, 2005;Beard &Willson, 2006;Rechkunova&Lavrik, 2010).
However, there is one attention catching point in the 43 Ca-MIE we observed which is a simultaneous (a) enzyme inhibition and (b) an essential decrease of a maximum size of the DNA fragments produced, from nearly 200n to about 30n (Figures 2 and 3). A marked difference between the results obtained in 40 Ca (nonmagnetic) and 43 Ca (magnetic) isotope tests is enough to exclude any doubt regarding the MIE nature of the phenomenon we've seen here (Buchachenko, 2009;Buchachenko et al, 2012).
Noteworthy, this is the first report ever on the MIEpromoted changes in lengths of DNA blocks processed by DNApolB, a target for magnetic ions. Fast UV -EtBr visualization of key results (Figure 2) was then completed with the detailed radiometric tests conducted in both DNA electrophoresis gel autioradiography and the enzyme catalytic activity measurements (Figure 3).
Taking into account that the DNApolB synthesized DNA segment length was found to be one of the critical parameters for the DNApolBdirected DNA repair and, hence, to the cell survival (Sobol et al, 1996;Kornberg & Baker, 2005;Beard &Willson, 2006;Ljungmann, 2010;Rechkunova&Lavril, 2010), we might assume a certain pharmacological potential beyond the data presented in Figures 1 -3. To our opinion, this assumption makes sense due to a known fact of the DNApolB overexpression occurred in most leukemic cells (Shadan& Villarreal, 1996;Albertella et al, 2005;Matsubara et al, 2007;Martin et al, 2010) including the HL-60 ones . O c t o b e r 1 8 , 2 0 1 3 Looking at this biochemical data through the "pharmacology spectacles", we're facing two major problems, both related to the task of making the leukemic DNApolB a real molecular target for magnetic metal ions. First, the latters must be administered using a targeted delivery system of some sort. Second, the in situ intracellular ferromagnetic ions content (Fe 2+ , first of all) must be considered as long as a target compartment (organella, cell, tissue) is known.
Concerning the targeted delivery of bivalent metal ions, some low toxic nanocationites were qualified as the efficient Me 2+ -carriers in both in vitro and in vivo studies. Thus, 25 Mg 2+ ions were selectively delivered towards myocardiocytes and lymphocytes in rats owing to nanocationites based on watersoluble porphyrine adducts of fullerene-C60; the porphyrineaffine signaling mitochondria membrane proteins were playing a role of traps catching and then retaining these pharmacophores (Amirshahi et al, 2008;Kuznetsov et al, 2010;Orlova et al, 2013 a, b). Several oligoheterocyclicaminothiolateswere also found efficient to provide the cell plasmatic membranesl and nuclei permeability for Ca 2+ , Mn 2+ , Cu 2+ and Zn 2+ ions in leukemic cells in vitro (Shimanovsky et al, 2010). All these and related medicinal nano -particles are to release the cations they're loaded with in response to a slight metabolic acidosis, so common for the most malignant tumors known (Amirshahi et al, 2008;Kuznetsov et al, 2010;Shimanovsky et al, 2010;Orlova et al, 2013 b).
In our previous studies,we have revealed that MIE are hardly inducible in ironrich mammalian tissues (spleen, liver) as compared to the iron -poor ones (Shatalov et al, 2012;Svistunov et al, 2013). Obviously, these were the easily predictable results. There is no way for 43 Ca-MIE to get expressed in a "heavily iron -polluted" environment (Buchachenko et al, 2012). That's why it is important to find out a general regularity of such a dependence as this: . This is what our findings (Figure 4) are all about. The regularity shown here is worthy of being used to correct pharmacodynamics patterns in further 43 Ca-MIE studies.
A non -Markoff population dynamics model allows to predict a cytostatic efficiency for 43 Ca 2+ -loaded PMC16 nanoparticles (porphyrylcyclohexylfullerene), cation releasing pharmacophore . Such cytostatic effect was observed by Orlova et al (2012) in experiments with [ 67 Zn]4PMC16 carriers in normal lymphocytes and human leukemia cell lines.
To summarize the results presented in Figures 1 -4, we may propose a biochemical path for the DNApolB engaging MIE of 43 Ca. As a matter of fact, this scheme ( Figure 5) is based on the ionradical concept of the phosphatetransferring metalloenzyme function which involves a numerous DFTcalculations and experimental data (Buchachenko et al, 2010(Buchachenko et al, -2013Orlova et al, 2012Shatalov et al, 2012;Svistunov et al, 2013). As seen from this chart, the only path for the dNTP-oxyradical decay in a 43 Ca 2+ -possessing DNApolB is a fast and irreversible ionradical move leading to the nascent DNA chain growth (compare Figures 1 and 5).
Considering that one of the Mg 2+ ions belonging to bicationicDNApolB molecule is hardly accessible to the endoosmotic isotope substitution, i. e. to the enzyme intermolecular metal ion replacement procedure (Sobol et al, 1996;Cowan, 2002;Kornberg & Baker, 2005;Beard &Willson, 2006;Rechkunova&Lavrik, 2010), it would be reasonable to expect a maximum Ca 2+ -enzyme incorporation level to be close to 50%.
As we have confirmed (Figures 6 and 7), about a half of endogenous enzymebound magnesium was indeed replaced with calcium in our experiments. Keeping the intra-enzyme metal substitution conditions optimal (see Materials and Methods), we realized that not more than 54 -58% of the proteinbound magnesium was in fact replaced with 43 Ca 2+ ( Figure 6). Two hour long incubation time scale was found optimal to reach the ion exchange (replacement) equilibrium point in 20 mM CaCl2 media (Figure 7). Most likely, the 43 Ca-MIE we observed had indeed caused by dis-synchronization of functioning of two metalcontaining enzyme catalytic sites which is a consequence of the insertion of magnetic 43 Ca 2+ into just onethe only accessible oneof the above mentioned two metal possessing domains.
Last not least, DNApolB makes a contribution to the cell chromatin stability/flexibility transitions. Being located in both deep and superficial strata of chromatin (Bergoglio et al, 2002;Matsubara et al, 2007;Ljungmann, 2009Ljungmann, , 2010Martin et al, 2010;Shimanovsky et al, 2010), this enzyme fits the requirements we have for a molecular target accessibility for magnetic metal ions delivered by an appropriate ionreleasing pharmacophore.

CONCLUSION
The Ca -Mg substitution occurred inside the AML/HL-60 chromatin DNApolB catalytic sites leads to a sharp inhibition of this enzyme once the magnetic 43 Ca isotope got engaged. This phenomenon had caused by the 43 Ca-related magnetic isotope effect known for its ionradical mechanism specified earlier by Buchachenko et al (2009Buchachenko et al ( -2013. Moreover, the 43 Ca-MIE manifests itself not only in the DNApolB inhibition but in a residual synthesis of abnormally short, DNA-repair hardly sufficient, "invalid" DNA fragments. Since the isotope substitution performed was efficiently managed in the simple endoosmotic pressure tests, the data presented are worthy of further study on their pharmacological potential. This potential derives also from the DNApolB properties making it a promising candidate to a role of the molecular target for 43 Ca 2+ -inhibitor in pre-clinical safe cytostatics developments.