Article Text
Abstract
Aim: Transketolase-like enzyme 1 (TKTL1) is a glycolytic enzyme that has been found to be upregulated in several tumours, and it is associated with tumour progression. Nephroblastoma is the commonest paediatric renal malignancy and has a good prognosis except for those with anaplasia. To the best of the authors’ knowledge, the expression of TKTL1 in nephroblastomas has not been studied before and the aim of this study was to compare the immunoexpression of TKTL1 in anaplastic and non-anaplastic nephroblastomas.
Methods: Twenty-eight patients who had nephrectomies for nephroblastomas were studied. Archival formalin-fixed paraffin-wax-embedded tissue sections were stained with monoclonal TKTL1 antibody.
Results: Six of the 15 anaplastic nephroblastomas showed staining in 80–100% of the tumour (p = 0.36). None of the 13 non-anaplastic nephroblastomas showed TKTL1 staining in >80% of the tumour.
Conclusion: TKTL1 expression is associated with the presence of anaplasia and may be a mechanism via which anaplastic tumour cells thrive under different conditions. Glycolytic inhibitors may play a role in anaplastic nephroblastomas.
Statistics from Altmetric.com
Take-home messages
Transketolase-like enzyme I (TKTL1) expression is associated with the presence of anaplasia.
TKTL1 expression may be a mechanism via which anaplastic tumour cells thrive.
Glycolytic inhibitors and ketogenic diet (which limits the intake of glucose) may play a role in treating anaplastic nephroblastomas, but this requires further study.
The pentose phosphate pathway, also known as the hexose monophosphate pathway, generates lactate and key substrates for DNA/RNA synthesis. In healthy epithelial cells around 30% of the glucose is metabolised via this pathway, while the other 70% is metabolised via the Emden–Meyerhof pathway for complete aerobic combustion. In contrast, approximately 90% of the glucose in invasive tumour cells is metabolised via the pentose phosphate pathway. These cancer cells metabolise up to 20 times more glucose and produce large amounts of lactate even in aerobic conditions (“aerobic glycolysis” or “Warburg effect”).1
Transketolase (TKT) controls the non-oxidative part of this pathway and in certain tumour types more than 85% of the ribose recovered from nucleic acids is derived directly or indirectly from the non-oxidative part of this pathway. Recently two genes (TKTL1 and TKTL2) with transketolase-like (TKTL) properties have been characterised. In particular TKTL1 has been implicated as the enzyme that drives the pentose phosphate pathway in cancer metabolism, with increased expression in breast,2 pulmonary non-small cell3 and nasopharyngeal carcinomas,4 as well as in ovarian granulosa cell tumours.5
In addition, TKTL1 expression correlates with tumour grade, and with higher expression in poorly differentiated (grade 3) gastric adenocarcinomas6 and glioblastomas (as compared with grade II and III astrocytomas),7 and in invasive carcinomas and high-grade squamous intraepithelial lesions (SILs) of the cervix (as compared with low-grade SILs).8 It is associated with increased Federation of Gynecology and Obstetrics (FIGO) stage in ovarian serous papillary carcinomas.9 Elevated TKTL1 expression is present in metastasising renal cell carcinomas10 and is an independent prognostic factor for reduced survival in laryngeal squamous cell carcinomas,11 urothelial carcinomas and colonic carcinomas.12 In thyroid papillary carcinomas, a significant association has been found between TKTL1 protein expression and the presence of lymph node metastases.13
Nephroblastoma (Wilms tumour) is the fourth commonest childhood malignancy after leukaemia, lymphomas and central nervous system tumours, and the commonest paediatric renal malignancy.14 It is treated by a combination of surgery, chemotherapy and radiotherapy, and has a relatively good prognosis. However, around 5% of nephroblastomas show “unfavourable” histology characterised by nuclear “anaplasia”. Tumours with diffuse anaplasia are resistant to chemotherapy and have a worse prognosis (48.5% 5-year survival versus 92.3% for non-anaplastic Wilms tumours).15
TKTL1 expression is associated with high tumour grade and poor prognosis in many tumours. Anaplastic nephroblastomas are associated with a poor prognosis. To the best of our knowledge, TKTL1 expression has not been investigated in nephroblastomas. With the availability of the novel antibody against TKTL1 we embarked on a study to see if TKTL1 immunoexpression correlates with the presence of anaplasia in nephroblastomas.
MATERIALS AND METHODS
This study was approved by the Research Ethics Committee of the Faculty of Health Sciences, University of Cape Town, South Africa. The slides and tissue blocks of 28 cases of nephroblastoma were retrieved from the archives of the Division of Anatomical Pathology, Red Cross War Memorial Children’s Hospital. The 28 cases included 15 anaplastic nephroblastomas and 13 non-anaplastic nephroblastomas. There were no cases of cystic partially differentiated nephroblastoma. All H&E-stained slides were reviewed to confirm the original diagnosis and to select an appropriate tumour block for immunohistochemical staining. Blocks with no or minimal tumour necrosis were preferred.
Immunohistochemical staining
Using routine techniques, sections (2–3 μm thick) were floated onto 3-aminopropyltriethoxysilane (APES)-coated slides and incubated overnight at 60°C. Sections were dewaxed in xylene, rehydrated in graded ethanols, and washed in water. Sections were then blocked for endogeneous peroxidase in 3% H2O2 water solution for 5 min and, after washing in water, heat-induced antigen retrieval was performed using a pressure cooker at full pressure for 2 min in citrate buffer at pH 6. Subsequently, sections were rinsed in phosphate-buffered saline (PBS) pH 7.6 and then blocked for non-specific binding using a 5% goat serum solution at room temperature for 10 min (Dako X0907; Dako, Glostrup, Denmark). The goat serum was drained off and the sections were incubated with the monoclonal mouse antihuman TKTL1 antibody (RIDA PentoCheck IHC; clone JF12T10; R-Biopharm AG, Darmstadt, Germany) overnight at room temperature. The sections were then washed in PBS and then treated with goat antimouse immunoglobulin labelled with horseradish peroxidase (Envision: Dako K4001; Dako) at room temperature for 30 min. Sections were washed in PBS. Positivity was developed by applying 3,3′-diaminobenzidene (Dako K3466) at room temperature for 5–10 min. Sections were washed in water, counterstained in an aqueous haematoxylin solution, followed by dehydration using graded ethanols and xylene before mounting with Entellan. A case of high-grade invasive transitional cell carcinoma of the bladder was used as the positive control. A negative control, in which the primary antibody was replaced by PBS, was run simultaneously.
The immunohistochemical scoring was done according to the manufacturer’s recommendations as follows: negative, 0–20% of tumour cells stained; 1+, 20–50% of tumour cells stained; 2+, 50–80% of tumour cells stained; and 3+, >80% of tumour cells stained. The numbers of tumour cells that were immunopositive was expressed as percentages of the total number of tumour cells per high-power field (HPF). Fifty HPFs were counted per slide and the final percentage was an average of the 50 HPFs. The scoring was done manually using an Olympus BX41 microscope. Immunohistochemical stains were analysed independently by two observers (H-TW and DG) and there was agreement on all immunohistochemical scores. Anaplasia was defined as the presence of nuclear enlargement that was more than three times larger than the adjacent non-anaplastic cells in two axes, nuclear hyperchromasia, and multipolar mitotic figures (fig 1A).16 Focal and diffuse anaplasia were defined using the topographic criteria published.17
Data were analysed using Stata Version 10 (Stata Corporation, College Station, Texas, USA). Fisher’s exact test was used to compare TKTL1 frequencies in specimens with and without anaplasia, with two-sided tests at alpha = 0.05.
RESULTS
Review of all the H&E-stained slides confirmed the original diagnosis in each case. The median age in our series was 43 months. A female preponderance (female:male 1.9:1) was present.
TKTL1 expression was less than 50% in 12 out of 13 non-anaplastic nephroblastomas; TKTL1 expression was more than 50% in eight out of 15 anaplastic nephroblastomas. None of the non-anaplastic nephroblastomas had more than 80% TKTL1 expression, whereas 40% of the anaplastic nephroblastomas expressed TKTL1 in more than 80% of the tumour cells.
Anaplastic nephroblastomas were found to express TKTL1 in a greater proportion of the cells (table 1). This difference was statistically significant (p = 0.036). Within the anaplastic group, there was no statistically significant difference in TKTL1 staining between those with focal anaplasia versus those with diffuse anaplasia (table 2). For the anaplastic tumours that showed >20% positivity (TKTL1 score ⩾1), the staining was granular cytoplasmic in seven cases (fig 1B), nuclear in five cases (fig 1C), and cytoplasmic and nuclear in three cases. In anaplastic tumours with a TKTL score ⩾1, the anaplastic and non-anaplastic cells were positive. TKTL1 expression in non-anaplastic tumours was patchy overall and less intense than anaplastic tumours (fig 1D). There was a significant decrease in TKTL1 expression in anaplastic tumours (p = 0.013) and non-anaplastic tumours (p = 0.046) treated with pre-operative chemotherapy (⇓tables 3 and 4).
There was no significant association between TKTL1 expression and stage in anaplastic (p = 0.078) and non-anaplastic nephroblastomas (p = 0.076).
DISCUSSION
To the best of our knowledge, this is the first study to investigate the expression of TKTL1 in nephroblastomas. In the present study we demonstrate a significantly increased expression of TKTL1 in anaplastic nephroblastomas compared with non-anaplastic nephroblastomas. However, this study is limited by the relatively small number of cases. As our aim was to compare anaplastic with non-anaplastic cases, we selected approximately equal numbers of each so that the results do not reflect the prevalence of TKTL1 expression of nephroblastomas overall (where anaplasia is only present in approximately 5% of nephroblastomas).
The presence of anaplasia is an indicator of chemoresistance, and hence the association with a poor prognosis. The mechanisms of chemoresistance in anaplastic nephroblastomas are currently under investigation. Expression of p53 is associated with anaplasia and a poor prognosis.18 P-glycoprotein is an ATP-dependent efflux pump with broad substrate specificity, including chemotherapeutic agents such as tacrolimus. Its expression in endothelial cells may act as blood–tumour barrier and its expression in tumour cells is induced by chemotherapy.19 Loss of heterozygosity at chromosomes 1p and 16q is associated with a poor prognosis. E-cadherin (16q22.1) has been proposed as the candidate gene involved in loss of heterozygosity of 16q.20 Many molecular markers (eg, alterations in carbonic anhydrase IX (CA9), Dickkopf 1 (DKK1), epidermal growth factor receptor 1 (EGFR1), HEY2, MYC, MYCN, telomerase reverse transcriptase (TERT), topoisomerase IIα (TOP2A), tripartite motif-containing 22 (TRIM22) and vascular endothelial growth factor (VEGF)) have been found by gene expression profiling to be upregulated in nephroblastomas, some of which are associated with a poor prognosis.21 22 TOP2A amplification has been detected only in anaplastic Wilms tumours and correlated strongly with proliferation, vascular invasion, metastases and adverse clinical outcomes, but TOP2A amplification may cause an increased sensitivity to topoisomerase inhibitors such as anthracyclines and actinomycin D.23
The increased expression of TKTL1 may indicate an increased capacity of the tumour cells for non-oxidative glycolysis, and this has several growth advantages: generation of ribose for nucleic acid synthesis, a fast and mitochondrial-independent production of ATP, protection against oxidative damage (via production of NADPH, an antioxidant), and production of large amounts of lactic acid leading to tissue acidosis with death of healthy cells via p53-mediated apoptosis but survival of p53-mutated tumour cells.12 TKTL1 in particular also generates acetyl-CoA that can be used in the synthesis of fatty acids and cholesterol.1
We postulate that the above advantages may complement several aspects of anaplastic nephroblastomas. The nuclear hyperchromasia and marked nuclear enlargement seen in anaplastic nephroblastomas indicate increased nuclear material, the synthesis of which may be facilitated by the substrates provided by an enhanced non-oxidative glycolysis. The ability to produce energy independent of mitochondria may allow the tumour cells to survive under hypoxic conditions. The production of NAPDH may allow the tumour cells to resist oxygen free radicals, such as that induced by radiotherapy. The lactic acidosis allows the survival of anaplastic tumour cells which are associated with p53 mutation. The capacity for lipid biosynthesis is of uncertain significance, but WT1, loss of heterozygosity of which is common in nephroblastomas, has been suggested to target the cholesterol/fatty acid synthesis pathway.24
The presence of TKTL1 immunoexpression in anaplastic cells and background non-anaplastic cells indicates that anaerobic glycolysis is not limited to the anaplastic cells in anaplastic nephroblastomas. This suggests that cells in the company of anaplastic cells also have subcellular metabolic alterations similar to anaplastic cells and may share similar genetic mutations.
It has been demonstrated that addition of thiamine, a cofactor in anaerobic glycolysis, activates transketolase and stimulates tumour growth,25 while application of transketolase inhibitors (such as oxythiamine) leads to a dramatic reduction in tumour proliferation in vitro (pancreatic carcinoma cell line)26 and in vivo (mice harbouring Ehrlich ascites tumour cells).27 However, overexpression of TKTL1 may not predict response to oxythiamine as demonstrated by one study where oxythiamine showed a weak effect in a TKTL1-expressing thyroid follicular carcinoma cell line.28 Gene silencing of TKTL1 by RNAi has also been shown to inhibit cell proliferation in human cancer cell lines including hepatoma,29 colon carcinoma,30 and nasopharyngeal carcinoma.4 Limitation of glucose substrate by using a ketogenic diet has also shown a significant reduction in growth of xenograft prostate cancer,31 gastric cancer32 and malignant brain tumours33 in mice.
We were unable to find any literature to link TKTL1 with chemotherapeutic agents used in the treatment of nephroblastomas. The main agents administered include vincristine, dactinomycin and doxorubicin. Their mechanisms of action include inhibition of microtubule assembly, elongation of RNA polymerase, and progression of TOP2A, respectively, and so they appear to target DNA synthesis rather than energy production.
In conclusion, we demonstrate in this study a significant correlation between TKTL1 immunoexpression and anaplasia in nephroblastomas. Therapeutic regimens that limit glycolysis may play a role in treating anaplastic nephroblastomas. However, further studies to confirm our findings are necessary.
REFERENCES
Footnotes
Funding: This study was funded by the National Health Laboratory Service of South Africa.
Competing interests: None.
Ethics approval: Obtained.