Serum TK1 activity measurement has been used as tumor proliferation marker for the diagnosis and prognosis of various hematological malignancies [
11,
12] and in some solid-tumor diseases [
8,
15]. Development of TK1 antibody-based methods demonstrated that the concentration of STK1 protein may be used for the prognosis and monitoring of tumor therapy in several types of solid tumor diseases [
19,
21,
23,
36]. In this study, we attempted to clarify the reasons underlying the apparently different information in different forms of malignant diseases that measurement of STK1 activities and TK1 protein levels provide. We took advantage of the fact that we have sensitive assays for determining STK1 activities and STK1-25-kDa protein levels. Thus far, no studies have been conducted where the specific activities of STK1 or the molecular forms of TK1 in different types of tumor diseases have been analyzed in detail.
Patients with myelodysplastic syndrome (MDS) are classified into many subtypes but all have a dysfunctional blood-forming capacity, more specifically in regeneration of myeloid linages, with a large fraction of patients progressing to acute myeloid leukemia. In the patients tested here, increased STK1 activities and 25 kDa protein, above the cut-off (12.5 ng/mL) of the blood donor control group, was observed in 15 out of 22 cases. This number is surprisingly high, particularly because the control group was not age-matched and was of a mean age of 27 compared with the mean age of 76 y in the MDS group. Earlier studies have shown a decrease in mean STK1 activity and protein levels with increasing age [
19]. The specific activity of STK1 in the MDS patients varied from 30 to 858 nmoles per min per mg TK 25 kDa protein; however, the mean value was similar to that of the blood donor group. The mechanism for the variation and similarities between blood donors and MDS patients in specific activities and molecular form will be discussed below. Here, we have studied 111 clinical samples from patients who were diagnosed with MDS, breast or prostate cancer. Particularly in breast cancer patients, the availability of clinical data on staging and histological grad enabled comparisons between the two assays. Similarly, Gleason scores (GS) of the prostate cancer patients made a preliminary clinical assessment of the STK1 protein assay possible. Furthermore, these results indicate that the STK1 protein assay may distinguish early stages of cancer diseases more efficiently compared with the TK activity assay. The principles and basic performance of the two STK1 assays used here have been described previously [
9,
25].
Consistent with earlier studies using TK-Liaison and Divitum TK activity assays [
37], we observed that STK1 activity values in sera from breast and prostate cancer were higher than the healthy cut-off in approximately 25% of samples. This finding was also evident from our ROC curve analysis, which showed a sensitivity of 26% with 96% specificity in the case of breast cancer. However, with respect to STK1 protein levels, immunoaffinity/Western blotting assays distinguished breast cancer from blood donors with sensitivity close to 79%at 96% specificity. In the case of prostate cancer patients, the results were similar, with the STK1 protein assay improving sensitivity 3-fold compared with the activity assay. Previous studies demonstrated that in sera from patients with leukemia, gastric and breast cancer and healthy controls, there was a significant correlation between TK1 protein concentration and TK1 activity [
24]. Similar results were found in this study, and a possible explanation for these correlations may be that the protein assay measures both active and inactive serum TK1, whereas the activity assay measures only the active high-MW TK1 protein complexes. The results presented here are consistent with our recently published results on STK1 determination in dogs with malignant diseases. We observed no significant difference in STK1 activities in sera from dogs with solid tumors compared with healthy dogs. However, there was a significant difference in STK1 protein levels between these two groups [
25]. Although the clinical application of TK1 as a biomarker has increased in recent years, very few studies have been conducted to determine the molecular forms of TK1 in different forms of malignancies [
32,
33]. Here, we used size-exclusion chromatography to investigate this question. The most significant and unexpected finding was that serum TK1 activity and TK1 protein did not clearly co-elute in any of the sera analyzed from patients with malignancies. The situation in blood donors was not resolved due to the fact that TK1 protein was not detected in the fractions containing enzyme activity. Sera from MDS patients showed two peaks of TK1 activity and the TK1 protein; however, the patterns were not identical. The discrepancies between TK1 activity and protein-elution profiles were even more apparent in sera from breast and prostate cancer patients, where STK1 activity was primarily observed as an active high-MW complex, whereas the TK1 polypeptide was observed in almost all serum fractions in breast cancer. In sera from prostate cancer patients, the majority of TK1 protein eluted with a peak at approximately 100 kDa, and only a minor fraction eluted in the high-MW form. These results strongly suggest that serum TK1 is present in many oligomeric forms containing both active and inactive TK1 protein. The high-MW oligomers apparently have a higher proportion of active TK1 protein compared with those with lower molecular weights. The results clearly show that there is a larger fraction of inactive TK1 protein in sera from breast and prostate cancer patients; however, even in this case, there is apparently a mixture of both active and inactive TK1 in the fractions, thus explaining the lack of co-elution between TK1 activity and protein. The varying compositions of the TK1 oligomers in sera from different patients may also explain the differences in specific activities of TK1 observed in sera from blood donors and MDS patients compared with those in sera from breast or prostate cancer patients. Specific activity indicates the active enzyme in a particular concentration of protein. To determine the specific activity of TK1 in different malignancies, both STK1 activity and TK1 sub unit of 25 kDa protein levels were measured in clinical samples. In the MDS sera, STK1 activity follows the STK1 protein profile, but there was no complete correlation between the activity and 25 kDa protein in all fractions (Additional file
1: Figure S
5A). In case of healthy, there was only one peak in a high MW complex form and we could not observe any protein bands in the western blot analysis, indicating that the protein levels were very low in the fractions. These results indicate that TK1 elution profile based on TK1 activity or TK1 protein are similar and that most of the TK1 oligomers are active, resulting in high specific activity (active TK1/ mg of TK1 25 kDa protein). Whereas in breast and prostate cancer sera, the TK1 elution profile was different and the TK1 activity eluted as a major peak with high MW, while the TK1 protein eluted in multimeric forms with different MW, representing both active and inactive forms of TK1 (Additional file
1: Figure S
5B and S
5C). A recent study was done on comparison of TK1 molecular forms in sera from healthy dogs, dogs with leukemia and mammary tumors. The results showed that TK1 exits as high MW oligomers which are active in both healthy and leukemia sera. TK1 protein also follows the similar pattern in both cases. However, in mammary tumor sera, TK1 activity eluted as a major peak with high MW complex, while the TK1 protein eluted in multimeric forms with different MW, representing both active and inactive forms of TK1 [
38]. These results are similar to that what we observed in human TK1 forms in different sera. These differences in molecular forms lead to lower specific activity as shown in Table
2. These findings are most likely due to modifications of TK1 in sera from patients with different cancer diseases. Attempts to define these presumed structural differences are ongoing and may lead to the development of an in vitro diagnostic test specific for serum TK1 in patients with different tumor diseases.