Downloaded from bjo.bmj.com on September 29, 2014 - Published by group.bmj.com
Laboratory science
Dysregulation of human apurinic/apyrimidinic endonuclease 1 (APE1) expression in advanced retinoblastoma Job Sudhakar,1,2 Vikas Khetan,3 Srinivasan Madhusudan,4 Subramanian Krishnakumar1 1
Larsen and Toubro Ocular Pathology Department, Vision Research Foundation, Sankara Nethralaya, Chennai, India 2 Birla Institute of technology and Science (BITS), Pilani, Rajasthan, India 3 Bhagwan Mahaveer Vitreoretinal Services, Medical Research Foundation, Sankara Nethralaya, Chennai, India 4 Laboratory of Molecular Oncology, Academic Unit of Oncology, School of Molecular Medical Sciences, University of Nottingham, Nottingham University Hospitals, Nottingham, UK Correspondence to Dr S Krishnakumar, Department of Ocular Pathology, Vision Research Foundation, Sankara Nethralaya, Chennai 600006, India; drkrishnakumar_2000@yahoo. com Received 14 August 2013 Revised 15 November 2013 Accepted 2 December 2013 Published Online First 2 January 2014
ABSTRACT Background Retinoblastoma (RB) is a childhood eye tumour. Dysregulation of DNA repair may not only influence pathogenesis but could also adversely impact on response to cytotoxic chemotherapy frequently used in RB therapy. We studied the expression of human apurinic/apyrimidinic endonuclease (APE1), a key multifunctional protein involved in DNA base excision repair in RB. Methods Expression of APE1 was evaluated by immunohistochemistry in a series of 55 RBs and in retina. In tumours, APE1 expression was analysed in cytoplasm and nucleus independently and correlated with histopathological features, including invasion, differentiation and International Intraocular Retinoblastoma Classification groups. Relative APE1 mRNA and protein expressions were evaluated by realtime PCR and western blot. The expression of APE1 in tumour groups was compared with retinal tissue. Results APE1 cytoplasmic expression was observed in 98% and nuclear positivity was observed in 83% of tumours analysed. Tumour cells invading the optic nerve showed predominant cytoplasmic immunoreactivity. An inverse correlation between cytoplasmic and nuclear positivity was observed. Real-time PCR revealed an increase in APE1 transcripts compared with retina. Western blot revealed a decreased protein concentration compared with retinal tissue. Conclusions This is the first study of APE1 expression in RB. Our observation suggests that subcellular localisation of APE1 is altered in RB. APE1 could be a potential drug target in RB.
INTRODUCTION
To cite: Sudhakar J, Khetan V, Madhusudan S, et al. Br J Ophthalmol 2014;98:402–407. 402
Retinoblastoma (RB) is the most common primary intraocular tumour usually seen in children below 5 years. Untreated RB invades sclera and optic nerve and eventually leads to death of the patients. The most common chemotherapeutic regimen constitutes a triple therapy, including carboplatin, vincristine and etoposide in multiple cycles;1 however, therapeutic failure is not uncommon in RB. The mechanisms of drug resistance in RB are complex2 and are yet to be fully understood. Recent evidence suggests that proficient DNA repair in cancer cells that allow repair of DNA damage induced by cytotoxic agents is an important factor determining therapeutic resistance.3 DNA base excision repair (BER) is involved in the repair of DNA bases that have been damaged by alkylation, oxidation or ring saturation and also in processing deaminated bases and DNA
single-strand breaks. Although there is more than one subpathway of BER, in most cases excision of a damaged base by a DNA glycosylase enzyme leads to the formation of a potentially cytotoxic apurinic/apyrimidinic (AP) site intermediate. This is a target for AP endonuclease, which cleaves the phosphodiester backbone on the 50 site of the AP site via a hydrolytic mechanism. The major AP endonuclease in human cells, apurinic/apyrimidinic endonuclease-1 (APE1), accounts for over 95% of the total AP endonuclease activity in most cultured human cell lines.4 APE1 is a ubiquitous multifunctional protein involved in DNA repair activity and also in the regulation of redox function and transcription.5 6 Depletion of intracellular APE1 by siRNAs or shRNAs sensitises mammalian cells to a variety of DNA-damaging agents.7–9 On the other hand, overexpression of APE1 conferred protective effect against chemotherapeutics and radiation.10 In human tumours, APE1 expression has prognostic/ predictive significance in several solid tumours,7 11 12 and its overexpression was associated with cisplatin resistance.13 As APE1 is a promising drug target, several groups have developed small molecule inhibitors for therapeutic application. Preclinical studies of inhibitors targeting either the DNA repair domain or the redox domain of APE1 potentiate the cytotoxicity of chemotherapeutics.13 14 In the current study, we investigated APE1 expression in RB. We provide the first evidence of APE1 overexpression in the RB cytoplasm compared with retinal tissue, implying that APE1 may be involved in the pathogenesis of RB. Given the recent interest in the development of small molecule inhibitors of APE1 as an anticancer strategy and based on the expression pattern in RB, our study provides preliminary evidence that APE1 is a potential target in RB.
MATERIALS AND METHODS This study was reviewed and approved by the local ethics committee of Vision Research Foundation, Sankara Nethralaya, India, and the committee deemed it in conformation with the generally accepted principles of research and in accordance with the Helsinki declaration (study code 301-2012P). RB tissues were collected from globe of patients who underwent enucleation as primary therapy. A written consent was obtained from patient’s parents for the use of tumour for research. Normal retina was collected from cadaveric eyes received at our institute’s eye bank. After removal of cornea for
Sudhakar J, et al. Br J Ophthalmol 2014;98:402–407. doi:10.1136/bjophthalmol-2013-304166
Downloaded from bjo.bmj.com on September 29, 2014 - Published by group.bmj.com
Laboratory science transplant, the retina was carefully detached and used for RNA and protein extraction.
Clinical information A total of 55 tumours were evaluated in this study. Among the tumours analysed, there were 17 invasive and 38 non-invasive tumours. Based on differentiation, there were 17 undifferentiated, 14 poorly differentiated, 7 moderately differentiated and 17 well-differentiated tumours.15 Based on the clinical presentation, the tumours were grouped as per International Intraocular Retinoblastoma Classification (IIRC).16 There were 36 group E and 19 group D tumours. Among group E tumours, there were 9 undifferentiated, 11 poorly differentiated, 4 moderately differentiated, 12 well differentiated, 14 high-risk invasive and 22 non-invasive tumours. Similarly, among group D tumours, there were 9 undifferentiated, 2 poorly differentiated, 3 moderately differentiated, 5 well differentiated, 5 high-risk invasive and 14 non-invasive tumours. Demography of clinical and histopathological characteristics is given in table 1, and the respective clinical details are summarised in table 2.
Immunohistochemistry RB paraffin sections (5 mm thick) were dewaxed and rehydrated. Antigen retrieval was performed by pressure cooker method in citrate buffer ( pH 6.0). Immunohistochemistry was performed using Novolink Mini polymer detection system kit (RE7290-K) supplied by Leica Microsystems, Newcastle, UK. In brief, the deparaffinised slides were flooded with deionised water for 5 min to neutralise proteins, washed in 50 mM Tris buffered saline (TBS) pH 7.4, incubated with protein block, washed in TBS and primary antibody (anti-APE1 rabbit polyclonal antibody, Novus Biologicals) diluted in TBS (1/400) was added and incubated for 2 hours at room temperature. The sections were washed in TBS and subjected to postprimary blocking and further incubated with polymer and washed in TBS. The reaction was revealed by 3,30 diaminobenzidine tetrahydrochloride and counterstained with haematoxylin. Negative control included the omission of the primary antibody and substitution with non-immune serum.17 18
Immunochemistry analysis Evaluations of immunostaining in RB tissues were objectively performed by two investigators independently and were blinded to the clinicopathological characteristics of patients. Immunoreactivity was recorded in percentage for both nuclear and cytoplasmic staining independently. In brief, randomly 10
Table 1 Clinical and histopathological demography of the cohort Tumour features
Invasive RB (n=17)
Non-invasive RB (n=38)
Group D (n=19)
Group E (n=36)
WD MD PD UD IIRC Grp D IIRC Grp E Inv NI
5 1 5 6 3 14 – –
12 6 8 12 16 22 – –
5 3 2 9 – – 3 16
12 4 11 9 – – 14 22
Grp, group; IIRC, International Intraocular Retinoblastoma Classification; Inv, invasion; MD, moderately differentiated; NI, no invasion; PD, poorly differentiated; RB, retinoblastoma; UD, undifferentiated; WD, well differentiated.
viable tumour fields were scanned for immunoreactivity under high-power objective (400×). Any appreciable brown colour was considered positive immunoreactivity. Difference in immunostaining evaluation between investigators was observed in four tumours, which were reviewed and settled over consensus. The non-parametric Mann–Whitney U test was used to determine the significance of immunoreactivity with tumour invasion and differentiation groups. Pearson’s correlation was used to determine the correlation. p Values
Laboratory science
Dysregulation of human apurinic/apyrimidinic endonuclease 1 (APE1) expression in advanced retinoblastoma Job Sudhakar,1,2 Vikas Khetan,3 Srinivasan Madhusudan,4 Subramanian Krishnakumar1 1
Larsen and Toubro Ocular Pathology Department, Vision Research Foundation, Sankara Nethralaya, Chennai, India 2 Birla Institute of technology and Science (BITS), Pilani, Rajasthan, India 3 Bhagwan Mahaveer Vitreoretinal Services, Medical Research Foundation, Sankara Nethralaya, Chennai, India 4 Laboratory of Molecular Oncology, Academic Unit of Oncology, School of Molecular Medical Sciences, University of Nottingham, Nottingham University Hospitals, Nottingham, UK Correspondence to Dr S Krishnakumar, Department of Ocular Pathology, Vision Research Foundation, Sankara Nethralaya, Chennai 600006, India; drkrishnakumar_2000@yahoo. com Received 14 August 2013 Revised 15 November 2013 Accepted 2 December 2013 Published Online First 2 January 2014
ABSTRACT Background Retinoblastoma (RB) is a childhood eye tumour. Dysregulation of DNA repair may not only influence pathogenesis but could also adversely impact on response to cytotoxic chemotherapy frequently used in RB therapy. We studied the expression of human apurinic/apyrimidinic endonuclease (APE1), a key multifunctional protein involved in DNA base excision repair in RB. Methods Expression of APE1 was evaluated by immunohistochemistry in a series of 55 RBs and in retina. In tumours, APE1 expression was analysed in cytoplasm and nucleus independently and correlated with histopathological features, including invasion, differentiation and International Intraocular Retinoblastoma Classification groups. Relative APE1 mRNA and protein expressions were evaluated by realtime PCR and western blot. The expression of APE1 in tumour groups was compared with retinal tissue. Results APE1 cytoplasmic expression was observed in 98% and nuclear positivity was observed in 83% of tumours analysed. Tumour cells invading the optic nerve showed predominant cytoplasmic immunoreactivity. An inverse correlation between cytoplasmic and nuclear positivity was observed. Real-time PCR revealed an increase in APE1 transcripts compared with retina. Western blot revealed a decreased protein concentration compared with retinal tissue. Conclusions This is the first study of APE1 expression in RB. Our observation suggests that subcellular localisation of APE1 is altered in RB. APE1 could be a potential drug target in RB.
INTRODUCTION
To cite: Sudhakar J, Khetan V, Madhusudan S, et al. Br J Ophthalmol 2014;98:402–407. 402
Retinoblastoma (RB) is the most common primary intraocular tumour usually seen in children below 5 years. Untreated RB invades sclera and optic nerve and eventually leads to death of the patients. The most common chemotherapeutic regimen constitutes a triple therapy, including carboplatin, vincristine and etoposide in multiple cycles;1 however, therapeutic failure is not uncommon in RB. The mechanisms of drug resistance in RB are complex2 and are yet to be fully understood. Recent evidence suggests that proficient DNA repair in cancer cells that allow repair of DNA damage induced by cytotoxic agents is an important factor determining therapeutic resistance.3 DNA base excision repair (BER) is involved in the repair of DNA bases that have been damaged by alkylation, oxidation or ring saturation and also in processing deaminated bases and DNA
single-strand breaks. Although there is more than one subpathway of BER, in most cases excision of a damaged base by a DNA glycosylase enzyme leads to the formation of a potentially cytotoxic apurinic/apyrimidinic (AP) site intermediate. This is a target for AP endonuclease, which cleaves the phosphodiester backbone on the 50 site of the AP site via a hydrolytic mechanism. The major AP endonuclease in human cells, apurinic/apyrimidinic endonuclease-1 (APE1), accounts for over 95% of the total AP endonuclease activity in most cultured human cell lines.4 APE1 is a ubiquitous multifunctional protein involved in DNA repair activity and also in the regulation of redox function and transcription.5 6 Depletion of intracellular APE1 by siRNAs or shRNAs sensitises mammalian cells to a variety of DNA-damaging agents.7–9 On the other hand, overexpression of APE1 conferred protective effect against chemotherapeutics and radiation.10 In human tumours, APE1 expression has prognostic/ predictive significance in several solid tumours,7 11 12 and its overexpression was associated with cisplatin resistance.13 As APE1 is a promising drug target, several groups have developed small molecule inhibitors for therapeutic application. Preclinical studies of inhibitors targeting either the DNA repair domain or the redox domain of APE1 potentiate the cytotoxicity of chemotherapeutics.13 14 In the current study, we investigated APE1 expression in RB. We provide the first evidence of APE1 overexpression in the RB cytoplasm compared with retinal tissue, implying that APE1 may be involved in the pathogenesis of RB. Given the recent interest in the development of small molecule inhibitors of APE1 as an anticancer strategy and based on the expression pattern in RB, our study provides preliminary evidence that APE1 is a potential target in RB.
MATERIALS AND METHODS This study was reviewed and approved by the local ethics committee of Vision Research Foundation, Sankara Nethralaya, India, and the committee deemed it in conformation with the generally accepted principles of research and in accordance with the Helsinki declaration (study code 301-2012P). RB tissues were collected from globe of patients who underwent enucleation as primary therapy. A written consent was obtained from patient’s parents for the use of tumour for research. Normal retina was collected from cadaveric eyes received at our institute’s eye bank. After removal of cornea for
Sudhakar J, et al. Br J Ophthalmol 2014;98:402–407. doi:10.1136/bjophthalmol-2013-304166
Downloaded from bjo.bmj.com on September 29, 2014 - Published by group.bmj.com
Laboratory science transplant, the retina was carefully detached and used for RNA and protein extraction.
Clinical information A total of 55 tumours were evaluated in this study. Among the tumours analysed, there were 17 invasive and 38 non-invasive tumours. Based on differentiation, there were 17 undifferentiated, 14 poorly differentiated, 7 moderately differentiated and 17 well-differentiated tumours.15 Based on the clinical presentation, the tumours were grouped as per International Intraocular Retinoblastoma Classification (IIRC).16 There were 36 group E and 19 group D tumours. Among group E tumours, there were 9 undifferentiated, 11 poorly differentiated, 4 moderately differentiated, 12 well differentiated, 14 high-risk invasive and 22 non-invasive tumours. Similarly, among group D tumours, there were 9 undifferentiated, 2 poorly differentiated, 3 moderately differentiated, 5 well differentiated, 5 high-risk invasive and 14 non-invasive tumours. Demography of clinical and histopathological characteristics is given in table 1, and the respective clinical details are summarised in table 2.
Immunohistochemistry RB paraffin sections (5 mm thick) were dewaxed and rehydrated. Antigen retrieval was performed by pressure cooker method in citrate buffer ( pH 6.0). Immunohistochemistry was performed using Novolink Mini polymer detection system kit (RE7290-K) supplied by Leica Microsystems, Newcastle, UK. In brief, the deparaffinised slides were flooded with deionised water for 5 min to neutralise proteins, washed in 50 mM Tris buffered saline (TBS) pH 7.4, incubated with protein block, washed in TBS and primary antibody (anti-APE1 rabbit polyclonal antibody, Novus Biologicals) diluted in TBS (1/400) was added and incubated for 2 hours at room temperature. The sections were washed in TBS and subjected to postprimary blocking and further incubated with polymer and washed in TBS. The reaction was revealed by 3,30 diaminobenzidine tetrahydrochloride and counterstained with haematoxylin. Negative control included the omission of the primary antibody and substitution with non-immune serum.17 18
Immunochemistry analysis Evaluations of immunostaining in RB tissues were objectively performed by two investigators independently and were blinded to the clinicopathological characteristics of patients. Immunoreactivity was recorded in percentage for both nuclear and cytoplasmic staining independently. In brief, randomly 10
Table 1 Clinical and histopathological demography of the cohort Tumour features
Invasive RB (n=17)
Non-invasive RB (n=38)
Group D (n=19)
Group E (n=36)
WD MD PD UD IIRC Grp D IIRC Grp E Inv NI
5 1 5 6 3 14 – –
12 6 8 12 16 22 – –
5 3 2 9 – – 3 16
12 4 11 9 – – 14 22
Grp, group; IIRC, International Intraocular Retinoblastoma Classification; Inv, invasion; MD, moderately differentiated; NI, no invasion; PD, poorly differentiated; RB, retinoblastoma; UD, undifferentiated; WD, well differentiated.
viable tumour fields were scanned for immunoreactivity under high-power objective (400×). Any appreciable brown colour was considered positive immunoreactivity. Difference in immunostaining evaluation between investigators was observed in four tumours, which were reviewed and settled over consensus. The non-parametric Mann–Whitney U test was used to determine the significance of immunoreactivity with tumour invasion and differentiation groups. Pearson’s correlation was used to determine the correlation. p Values
