MUTANT HUMAN DEOXYCYTIDINE KINASE

Abstract

An isolated nucleic acid includes a sequence encoding a mutant human deoxycytidine kinase (hdCK) capable of converting prodrugs, such as a nucleoside analogue, into cytotoxic drugs. An isolated vector can include the nucleic acid and an isolated host cell can be genetically engineered with the isolated vector. The polypeptides can be obtained by a procedure using recombinant techniques. A pharmaceutical composition, which includes the isolated nucleic acid, the expression vector, the host cell, or an isolated mutant hdCK, can be used as a medicament, such as for the treatment of cancer or for the prevention of a viral infection. The polypeptides and nucleic acids can be used for the treatment of malignancies and viral infections, in methods of sensitizing cells to prodrugs, in methods of gene therapy, in methods of non-invasive nuclear imaging and in methods of inhibiting pathogenic agents in a subject.

Claims

1.-30. (canceled)

31. An isolated nucleic acid comprising a sequence encoding a mutant human deoxycytidine kinase (hdCK), wherein said mutant hdCK comprises a polypeptide sequence which differs from wild-type hdCK of sequence SEQ ID NO: 1 by at least two mutations, wherein the at least two mutations are a) at the amino acid positions S169 and E171, or b) at the amino acid positions S169 and E247, or c) at the amino acid positions S169 and L249, and wherein the encoded mutant hdCK is at least 80% identical to sequence SEQ ID NO: 1 of the wild-type hdCK.

32. The isolated nucleic acid according to claim 31, wherein said mutant hdCK comprises a polypeptide sequence which differs from wild-type hdCK of sequence SEQ ID NO: 1 by at least four mutations, wherein the at least four mutations are at the amino acid positions S169, E171, E247 and L249, and wherein the encoded mutant hdCK is at least 80% identical to sequence SEQ ID NO: 1 of the wild-type hdCK.

33. The isolated nucleic acid according to claim 31, wherein the encoded mutated hdCK increases the phosphorylation of a nucleoside analogue.

34. The isolated nucleic acid according to claim 31, which comprises a nucleotide sequence encoding a mutant hdCK of sequence SEQ ID NO: 5, or proteins having at least 80% amino acid sequence identity with SEQ ID NO: 5, provided that said sequence contains the at least four mutations at the amino acid positions S169, E171, E247 and L249.

35. An isolated expression vector comprising the nucleic acid according to claim 31.

36. An isolated host cell genetically engineered with the vector according to claim 35.

37. A process for producing a mutant human deoxycytidine kinase (hdCK) comprising in vitro culturing a host cell of claim 36 and recovering the expressed mutant hdCK from the cultured host cells and/or culture medium.

38. An isolated mutant human deoxycytidine kinase (hdCK) comprising a polypeptide sequence which differs from wild-type hdCK of sequence SEQ ID NO: 1 by at least two mutations, wherein the at least two mutations are: a) at the amino acid positions S169 and E171, or b) at the amino acid positions S169 and E247, or c) at the amino acid positions S169 and L249, and wherein the encoded mutant hdCK is at least 80% identical to sequence SEQ ID NO: 1 of the wild-type hdCK.

39. A pharmaceutical composition comprising: (a) an isolated nucleic acid comprising a sequence encoding a mutant human deoxycytidine kinase (hdCK), wherein said mutant hdCK comprises a polypeptide sequence which differs from wild-type hdCK of sequence SEQ ID NO: 1 by at least two mutations, wherein the at least two mutations are: a) at the amino acid positions S169 and E171, or b) at the amino acid positions S169 and E247, or c) at the amino acid positions S169 and L249, and wherein the encoded mutant hdCK is at least 80% identical to sequence SEQ ID NO: 1 of the wild-type hdCK, (b) an isolated expression vector comprising the isolated nucleic acid, (c) an isolated host cell genetically engineered with the vector, or (d) an isolated mutant human deoxycytidine kinase (hdCK) comprising a polypeptide sequence which differs from wild-type hdCK of sequence SEQ ID NO: 1 by at least two mutations, wherein the at least two mutations are: a) at the amino acid positions S169 and E171, or b) at the amino acid positions S169 and E247, or c) at the amino acid positions S169 and L249, and wherein the encoded mutant hdCK is at least 80% identical to sequence SEQ ID NO: 1 of the wild-type hdCK; and (e) a pharmaceutically acceptable carrier or diluent.

40. A method of treating a subject comprising administering to the subject the isolated nucleic acid, the expression vector, the host cell, the isolated mutant hdCK, or the pharmaceutical composition according to claim 39.

41. The method according to claim 40, wherein the subject is treated for cancer or for the prevention of a viral infection.

42. A kit comprising a pharmaceutical composition according to claim 39 and a nucleoside analogue.

43. An ex-vivo method of sensitizing a cell to a nucleoside analogue prodrug, which method comprises the steps of: (i) transfecting or transducing cells with: a nucleic acid of claim 1, or an expression vector comprising the nucleic acid according to claim 1.

44. The method according to claim 43, wherein the nucleic acid functions as a safety gene in gene therapy or cell therapy.

45. A method of non-invasive imaging of transfected or transduced cells in a subject, and which method comprises the steps of (i) contacting the cells in a subject with a detectable compound; and (ii) non-invasively monitoring the quantity of said detectable compound in said cell or subject, wherein the cells in the subject comprise a nucleic acid comprising a mutant hdCK and a reporter gene or a nucleic acid comprising mutant hdCK and one nucleic acid comprising a reporter gene, wherein said mutant hdCK comprises a polypeptide sequence which differs from wild-type hdCK of sequence SEQ ID NO: 1 by at least two mutations, wherein the at least two mutations are a) at the amino acid positions S169 and E171, or b) at the amino acid positions S169 and E247, or c) at the amino acid positions S169 and L249, and wherein the encoded mutant hdCK is at least 80% identical to sequence SEQ ID NO: 1 of the wild-type hdCK.

Description

FIGURES

[0329] FIG. 1. Graph representing the sensitivity to gemcitabine of Messa10K cells induced by M36. The percentage of alive cells over the total amount of cells is plotted as a function of the concentration of gemcitabine. Diamonds, Messa 10K cells; triangles wt dCK; squares G12; circles M36.

[0330] FIG. 2. Graph representing the sensitivity to gemcitabine of Messa10K cells induced by M36 to a wider range of gemcitabine than used in FIG. 1. Diamonds, Messa 10K cells; triangles wt dCK; squares G12; circles M36.

[0331] FIG. 3. Schematic representation of the localisation of the mutations characterising M36 in the human dCK. The position of the first and of the last amino acids are indicated. The location and numbering of the alpha helixes and of the beta sheets is given in green/light grey and blue/dark grey, respectively. The position and nature of the four mutations present in M36 is given below the drawing.

[0332] FIG. 4. Schematic representation of the experimental procedure used to test the level of resistance to infection by HIV-derived vectors of cells expressing constitutively M36 and treated in the presence of the anti-HIV compounds ddC and 3TC.

[0333] FIG. 5. Graph representing the sensitivity to infection of HEK 293T/CD4+ cells to HIV in the presence of varying concentrations of ddC of cells encoding M36 (black circles) or an extra copy of the wt dCK gene (white circles).

[0334] FIG. 6. Graph representing the sensitivity to infection of HEK 293T/CD4+ cells to HIV in the presence of varying concentrations of 3TC (Lamivudine) of cells encoding M36 (black circles) or an extra copy of the wt dCK gene (white circles).

[0335] FIG. 7. Graph representing the sensitisation to anti cancer compound other than gemcitabine obtained by M36. The percentage of alive cells over the total amount of cells is plotted as function of the concentration of the anticancer compound. White circles, Messa 10K cells; grey diamonds, wt dCK; black squares, M36. Panel A, sensitivity to Fludarabine. Panel B, sensitivity to AraC. The x axis is given in log scale.

[0336] FIG. 8. Graphs representing the biochemical comparative characterisation of phosphorylation of AraC by wt dCK and M36. Phsphorylation kinetics of wt dCK (grey circles) and M36 (white circles) overexpressed in E. coli. Steady state kinetic data are fitted according to the Michaelis-Menten equation. Panel A, phosphorylation of dC. Panel B, phosphorylation of gemcitabine. Panel C, phosphorylation of AraC.

[0337] FIG. 9. Tables representing the biochemical comparative characterisation of phosphorylation of AraC by wt dCK and M36. Panel A, ratio of 1/Km and of Kcat for M36 vs wt dCK, with respect to the natural substrate dC (black), to gemcitabine (grey), and to AraC (white). Panel B, ratio of Km (black) and of Kcat (grey) for M36 vs G12, with respect to the gemcitabine and dC. The dotted line gives the reference of a ratio of 1.

EXAMPLE

Materials and Methods

[0338] Tests of sensitivity to anticancer drugs. Cells were seeded at 5000 cells per well in a 96-well plate and grown overnight at 37 C. Two rows were used for each population. 12 h after seeding, increasing concentrations of Gemcitabine (0-400 nM or 0-100 M), or AraC (0-10 mM) were added to each well and left in culture for further 72 h. Cell viability was measured by MTT test (CellTiter 96 Non-Radioactive Cell Proliferation Assay, Promega) and the number of living cells in each well was evaluated by measuring the OD at 570 nm. For each population, the fraction of alive cells was calculated as OD570 concentration X Gem/OD570 concentration 0 Gem, to estimate the sensitivity of the population to the prodrug.
Tests of sensitivity to antiviral drugs. Viral vectors were generated by transfection of HEK-293T cells with 10 g of pCMVAR8.91 plasmid, 5 g of pHCMV-G plasmid, and 10 g of genomic plasmid SDY-dCK (described in Rossolillo et al. 2012) encoding either for the wild type sequence of the human dCK or the M36 variant. Transduction was performed on 110.sup.6 HEK-293T-CD4+ cells with viral vectors. Transduced cells were selected in the presence of 0.6 g/ml of puromycin, added 24 h after transduction. 50 to 200 individual puromycin-resistant clones were pooled and expanded obtaining the polyclonal wt-dCK cell population (wtdCK-pcP, for wt dCK polyclonal population) and the M36 wt-dCK cell population (M36-pcP). Vectors for challenging infection in the presence of antiviral drugs were generated by transfection with 10 g of pTopo plasmid encoding for the HIV envelope ADA (ref, plus Hamoudi 2013), 5 g of pHCMV-G plasmid, and 10 g of genomic plasmid SDY-Luc, which is a variant of the pSDY plasmid encoding for the firefly luciferase. These viruses were used to transduce 110.sup.6 HEK-293T-CD4+-wtdCK or HEK-293T-CD4+-M36 cells. Viral infection was then monitored 48 hours after transduction. For this the medium was removed, the cells were washed twice in PBS, lysed, centrifuged, and the supernatant was used to measure luciferase activity according to the manufacturer's protocol (Promega, Fitchburg, Wis., USA) with a Glomax luminometer (Promega, Fitchburg, Wis., USA).
Western blot analyses. Messa10K cell lines expressing either the wt dCK or M36, both in the context of a proviral DNA with functional LTR or with an inactivated LTR were lysed in 1RIPA buffer, and 30, 60, 120 g of total protein (evaluated by Bradford) for each cell type and loaded on a 12% bis-tricine gel (Invitrogen). After transfer on a PVDF membrane, the dCK proteins were analysed by western Blot with 1:4000 dilution of a polyclonal anti-dCK antibody (rabbit, Sigma-Aldrich) and 1:3000 anti-rabbit HRP (BioRad) conjugated secondary antibody and detected by autoradiography.
Production and purification of recombinant dCK proteins. The wt-dCK and M36 sequences were cloned in the pET14b plasmid and expressed in E. coli cells BL21 DE3 pLysE. Protein expression was induced by adding 0.1 mM IPTG, and cells were collected after 4 h of growth at 37 C. His-tagged proteins were eluted with 250 mM Imidazole from His-Trap TM FF Columns (GE HealthCare), the Histidine tag was removed using the S-Tag Thrombin Purification Kit (Novagen), and dCK and G12 were further purified by gel filtration on S-200 Sephacryl columns (GE Healthcare).
Phosphorylation tests in vitro. The efficiency of phosphorylation of the natural substrate deoxycytidine and of the prodrugs gemcitabine and AraC were measured for purified wt-dCK and M36 in a NADH-based assay as previously described. All reagents were purchased from Sigma (France) except Gemcitabine (Lilly France SAS). Enzymes were assayed at RT at a concentration of 0.9 M for dC, 0.3 M for Gemcitabine, and X for AraC. dC was used at concentrations between 5 and 50 M, Gemcitabine at concentrations between 10 M and 1 mM, and AraC. ATP was 4 mM. All experiments were performed in triplicate.

Results

[0339] Comparison of Wild-Type hdCK and G36 and G12 Mutant hdCK

[0340] Messa10K cells were induced with a wild-type hdCK viral vector, a G12 hdCK viral vector and G36 hdCK viral vector. Briefly, after transduction with the viral vectors the different Messa 10K populations were seeded into 96-well dishes, exposed to varying concentrations of Gemcitabine, and cell survival was measured by an MTT test as previously described (Rossolillo et al. 2012). Surprisingly, the M36 mutant hdCK demonstrated a further improved sensitisation phenotype compared to G12 (FIG. 1). After the first comparative experiment the phenotype conferred by M36 was confirmed and compared to the G12 one in experiments with a wider range of gemcitabine concentrations (FIG. 2). With respect to G12, M36 contains an additional mutation (S169N), located in the vicinity one of the mutations present in G12 (E171K). M36 therefore presents four mutations clustered in pairs in the 247-249 (E247K, L249M) and 169-171 region (S169N, E171K) (FIG. 3).

Characterisation of M36 in Cell Culture with a LTR Lentiviral Vector

[0341] The use of SIN vectors is required for biomedical applications. To this end M36 has been inserted in a SIN version of pSDY, and the comparison between efficiency of M36 and G12 has been done in polyclonal populations of Messa 10K cells generated by transduction with M36 or with G12 SIN vectors. The difference between the structure of the proviral DNA generated with these two types of vector is the presence of a functional or inactivated (for SIN vectors) promoter to drive the expression of the integrated genomic RNA. Also using SIN vectors M36 was more efficient than G12 with a reduction of the IC50 from 75 nM to 20 nM. In parallel, with respect to wt LTR, a higher proportion of cells was induced to death, since the plateau of maximum mortality observed was set at 80% of death instead of the 60%. Finally, for both variants, G12 and M36, the use of SIN vectors appeared more efficient than that of wt vectors (data not shown).

[0342] In order to understand whether the enhanced effect observed with the SIN vectors could be due to a higher production of M36 in the presence of a LTR variant, we investigated whether an increased production of M36 was observed in Messa10K cells encoding the M36 in the context of a proviral DNA carrying a functional LTR or a its partially deleted counterpart. An increased amount of M36 was detected by Western blot analyses from cells harbouring a provirus with an inactive LTR, possibly accounting for the increased sensitization observed with M36 in these cells (Data not shown).

M36 and Sensitisation to Antiviral and Other Anticancer Compounds

[0343] The biochemical characterisation of M36, the increased sensitivity of Messa 10K cells to gemcitabine treatment is mostly due to a decrease in the ability to phosphorylate the natural substrate (dC) while maintaining a good level of phosphorylation of the gemcitabine. This is expected to lead to a reduced competition between dC and the drug, inside the cell. We reasoned that, if the ability to phosphorylate other drugs were retained in the mutant, the decreased phosphorylation of dC should also enhance sensitisation to other drugs. Two antiviral and two anti-cancer drugs that are activated through phosphorylation by dCK were tested, 3TC and ddC, and Fludarabine and AraC, respectively.

[0344] For the antiviral test, the following assay was developed es demonstrated in FIG. 4. Firstly HEK 293T/CD4.sup.+ cells were transduced at a low MOI of by the vector encoding the M36 variant and selection of the successfully transduced cells was performed using puromycin. The selected cell line was then transduced with lentiviral vectors, as described in Materials and Methods, encoding the luciferase gene and luciferase levels ere measured 48 hours after infection. A comparison with the level of luciferase detected in the parallel infection of control HEK 293T/CD4.sup.+ cells expressing as a transgene the wt dCK gene was performed and the data plotted. As shown in the figure, no significant increase in resistance was observed with ddC (FIG. 5), while a lower infectivity (in the 2-fold range) was observed for 3TC (FIG. 6).

[0345] The test with the anticancer drugs was performed as previously described, for gemcitabine. The results were markedly different with the two compounds tested. If, in the case of fludarabine, no significant sensitisation by M36 of Messa 10K cells was observed (FIG. 7A), in sharp contrast, for AraC, a strong sensitisation is observed in M36-expressing cells with respect to control cells expressing an exogenous copy of the wt dCK gene (FIG. 7B). In this case, a decrease in the IC50 of approximately 10 000 times was observed for cells containing M36 with respect to cells containing the wt dCK as exogenous gene.

Biochemical Characterisation of Phosphorylation of Gemcitabine and AraC by M36

[0346] To characterise its biochemical properties M36 and, as a control, the wt dCK protein, were produced as recombinant, C-ter His tagged proteins in E. coli, purified and tested for their efficiency of phosphorylation in vitro. As it was the case for the G12 mutant (Rossolillo et al. 2012) also for M36 the substitutions present in the mutant led to a decrease in the V.sub.max of phosphorylation as function of the concentration of substrate, for both dC and for gemcitabine, but while the decrease is dramatic for the dC it is only modest for gemcitabine (FIGS. 8A-B). This observation is strictly similar to what seen with G12 (Rossolillo et al 2012) suggesting that, also for M36, the balance between phosphorylation of dC with respect to the gemcitabine is more in favour of the drug than for the wild type dCK. The results obtained with AraC follow the same trend, with a very modest reduction of the V.sub.max of phosphorylation as function of the concentration of this substrate for M36, compared to what observed for wt dCK (FIG. 8C).

[0347] A comparison of Km and Kcat values for dC, gemcitabine and AraC found with M36 and with wt dCK shows that with M36 the values of Kcat decrease to approximately 50% those of wt dCK, for the three substrates (FIG. 9A). Differences between the natural substrate and gemcitabine or AraC were instead observed when considering Km values. Indeed an increased Km is observed for gemcitabine and AraC (higher 1/Km values in FIG. 9A) while, for the natural substrate, the situation is reversed. This observation suggests that the sensitisation observed in cell culture is due to an increased relative affinity of the mutant for both anticancer compounds and a parallel decreased relative affinity for the natural substrate.

[0348] In order to address why, despite the overall similarity of the biochemical properties of M36 and G12, M36 induced a stronger sensitisation to Gemcitabine than G12, Km and Kcat values for dC and gemcitabine were also compared between M36 and G12, by dividing the values found with M36 by those reported for G12 by Rossolillo and colleagues (Rossolillo et al. 2012). For gemcitabine, both parameters were strictly comparable between the two mutants (FIG. 9B), with a ratio of 1 (dotted line). For the natural substrate, instead, an increase both in Km (by a 3.5 factor) and in Kcat (by a 9.5 factor) was observed for M36. This implies that this mutant has a lower affinity for the natural substrate but an improved catalytic activity. The more efficient sensitisation induced by M36 with respect to G12 in cell culture, thereby suggests that, within the cell, the affinity for the substrate is more important than the catalytic activity itself.

Discussion

[0349] Suicide gene therapy is an approach of potential interest in many fields as clinical treatment of cancers with poor prognosis and as security gene in cell therapy approaches. We previously described the identification of a triple mutant of the human dCK gene that sensitises a panel of cancer cells (HEK 293T, Messa 10K, and BxPC3) to treatment with low doses of gemcitabine. Here a new mutant was identified that leads to an improved sensitisation to gemcitabine. Furthermore, the M36 mutant not only increases sensitivity to gemcitabine but also to AraC, another anticancer compound currently used in clinics. The sensitisation to AraC was even much stronger than what observed for gemcitabine. In addition, we report that the use of SIN lentiviral vectors further magnifies the effect sensitisation observed.

[0350] The biochemical characterisation of M36 depicts a similar pattern of enzymatic properties compared to G12, with a more drastic decrease in the efficiency of phosphorylation of the natural substrate than for gemcitabine when comparing M36 to wt dCK. A comparison between the values found here for M36 and those reported for G12 suggests, interestingly, that the parameter that can account for the improved sensitisation to gemcitabine observed in cell culture with M36 with respect to G12 is an increased affinity for the drug. The lower efficiency of phosphorylation per se observed with M36 with respect to G12 does not seem to be critical in vivo, though.

[0351] Interestingly the use of the SIN vectors further sensitises cells. It is further shown here that this is probably due to relieved transcriptional interference occurring when the U3 promoter is functional. This translates in a higher amount of M36 synthesised that is expected to increase the intracellular concentration of activated drug. The loss of the potential enhancing effect exerted by the functional LTR on the expression of the neighbouring sequences, including the internal promoter EF1- that drives the expression of M36, does not seem to impact the amount of M36 protein present in the cell, instead.