The role of herpes simplex virus-1 thymidine kinase alanine 168 in substrate specificity

Abstract

Herpes simplex virus type 1 (HSV) thymidine kinase (TK) has been widely used in suicide gene therapy for the treatment of cancer due to its broad substrate specificity and the inability of the endogenous human TK to phosphorylate guanosine analogs such as ganciclovir (GCV). The basis of suicide gene therapy is the introduction of a gene that encodes a prodrug-activating enzyme into tumor cells. After administration, the prodrug is selectively converted to a toxic drug by the suicide gene product thereby bringing about the eradication of the cancer cells. A major drawback to this therapy is the low activity the enzyme displays towards the prodrugs, requiring high prodrug doses that result in adverse side effects. Earlier studies revealed two HSV TK variants (SR39 and mutant 30) derived by random mutagenesis with enhanced activities towards GCV in vitro and in vivo. While these mutants contain multiple amino acid substitutions, molecular modeling suggests that substitutions at alanine 168 (A168) may be responsible for the observed increase in prodrug sensitivity. To evaluate this, site-directed mutagenesis was used to individually substitute A168 with phenylalanine or tyrosine to reflect the mutations found in SR39 and mutant 30, respectively. Additionally, kinetic parameters and the ability of these mutants to sensitize tumor cells to GCV in comparison to wild-type thymidine kinase were determined.

Keywords: Herpes Simplex Virus 1 Thymidine Kinase; ganciclovir; gene therapy.; substrate specificity.

Fig. (1)

Deduced amino acid sequence of the two HSV TK mutants, mutant 30 and SR39. The top line shows the amino acid sequence of the residues 159-174 of wild-type HSV TK. Two highly conserved tripeptide motifs are boxed and denoted as Sites 3 and 4. The amino acids in bold in the top line denote codons that were targeted for the creation for the randomized libraries and subsequent mutations are shown in the following lines [15,16].

Fig. (2)

Mutant enzyme relative specificity differences compared to wild-type enzyme for ganciclovir. Using the equation, (k_cat/K_{m (GCV)})/((k_cat/K_{m (GCV)}) + (k_cat/K_{m (dT)})), the relative specificities were calculated. This equation takes into account the presence of endogenous thymidine that could compete with prodrug for the active site. The data are plotted as the fold increase from wild-type enzyme. Mutant 30 and SR39 values have been taken from Kokoris et al. [17] and [22], respectively.

Fig. (3)

Sensitivity of TK-expressing rat C6 constructs to GCV. Pools of stable transfectants containing vector only (pUB), wild-type TK, A168F, A168Y, Mutant 30 and SR39 were constructed and transfected in rat C6 glioma cells and evaluated for prodrug sensitivity as described in Materials and Methods. After six days of prodrug treatment, the growth inhibition was determined by staining with Alamar Blue and fluorescence recorded at 530/590 nm. Each data point (mean ± SEM, n=3 performed with at least fifteen replicates) is expressed as a percentage of the value for control wells with no prodrug treatment.

Fig. (4)

Superposition of HSV TK active site residues (residue numbers 58, 128-132,168-172, 222-225 (unlabeled in figure)). All three panels have consistent coloring: wild-type: yellow; A168F: blue; and A168Y: green. Models of wild-type and mutant proteins in complex with: A. Thymidine (thy); B. GCV.