Author's Accepted Manuscript
Dicentric chromosome aberration analysis using Giemsa and centromere specific fluorescence in-situ hybridization for biological dosimetry: An inter- and intra-laboratory comparison in Indian laboratories M. Bhavani, G. Tamizh Selvan, Harpreet Kaur, J. S. Adhikari, J. Vijayalakshmi, P. Venkatachalam, N.K. Chaudhury www.elsevier.com/locate/apradiso
PII: DOI: Reference:
S0969-8043(14)00232-2 http://dx.doi.org/10.1016/j.apradiso.2014.06.004 ARI6706
To appear in:
Applied Radiation and Isotopes
Received date: 19 February 2014 Revised date: 19 May 2014 Accepted date: 4 June 2014 Cite this article as: M. Bhavani, G. Tamizh Selvan, Harpreet Kaur, J.S. Adhikari, J. Vijayalakshmi, P. Venkatachalam, N.K. Chaudhury, Dicentric chromosome aberration analysis using Giemsa and centromere specific fluorescence in-situ hybridization for biological dosimetry: An inter- and intra-laboratory comparison in Indian laboratories, Applied Radiation and Isotopes, http://dx.doi. org/10.1016/j.apradiso.2014.06.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Dicentric chromosome aberration analysis using Giemsa and centromere specific fluorescence in-situ hybridization for biological dosimetry: An inter- and intra-laboratory comparison in Indian laboratories. M. Bhavani1, G. Tamizh Selvan1,2, Harpreet Kaur1, J.S. Adhikari2, J. Vijayalakshmi1, P. Venkatachalam1*, N.K. Chaudhury2 *. 1
Sri Ramachandra University, Porur, Chennai 600 116, Tamil Nadu, India.
2
Institute of Nuclear Medicine and Allied sciences, DRDO, Timarpur, New Delhi 110 054, India.
BHAVANI MANIVANNAN - Master of Science, Research Scholar, Department of Human Genetics, College of Biomedical Science Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] TAMIZH SELVAN GNANA SEKARAN- Master of Science, Research Scholar, Department of Human Genetics, College of Biomedical Science Technology and Research, Sri Ramachandra University, Porur, Chennai- 600 116 and Institute of Nuclear Medicine and Allied Sciences Brig Mazumdar Road, Timarpur, Delhi-110 054, India. [email protected] Dr. HARPREET KAUR - Ph.D. Assistant Professor, Department of Human Genetics, College of Biomedical Science, Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] Dr. JAWAHAR SINGH ADHIKARI - Ph.D. Scientist E, Chemical Radioprotector and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences, Brig Mazumdar Road, Timarpur, Delhi, India-110 054. [email protected]
1
Dr. VIJAYALAKSHMI JAGANATHAN - Ph.D. Assistant Professor, Department of Human Genetics, College of Biomedical Sciences, Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] Dr. Perumal Venkatachalam- Ph.D. Professor, Department of Human Genetics, College of Biomedical Science Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] Dr. NABO KUMAR CHAUDHURY- Ph.D. Scientist F, Group Head, Chemical Radioprotector and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences, Brig Mazumdar Road, Timarpur, Delhi, India-110 054. [email protected] *Correspondence (1): Dr. NABO KUMAR CHAUDHURY- Ph.D. Scientist F, Group Head, Chemical Radioprotector and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences, Brig Mazumdar Road, Timarpur, Delhi, India-110 054. [email protected] Tel No: 011-23905131/2397820. Fax: 011-23919509 *Correspondence (2): Dr. PERUMAL VENKATACHALAM- Ph.D. Professor, Department of Human Genetics, College of Biomedical Sciences, Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected]; Tel No: 044-24768027 Extn 237. Fax: 044-24767008
Running title: Networking of biodosimetry laboratories
2
Abstract: To facilitate efficient handling of large samples, an attempt towards networking of laboratories in India for biological dosimetry was carried out. Human peripheral blood samples were exposed to 60
Co -radiation for ten different doses (0-5Gy) at a dose rate of 0.7 and 2Gy/min. The
chromosomal aberrations (CA) were scored in Giemsa-stained and fluorescence in-situ hybridization with centromere-specific probes. No significant difference (p>0.05) was observed in the CA yield for given doses except 4 and 5Gy, between the laboratories, among the scorers and also staining methods adapted suggest the reliability and validates the inter-lab comparisons exercise for triage applications.
Key words: Biodosimetry; Dicentric chromosome; FISH and Dose-response curve.
1. Introduction The threat of radiological accidents or nuclear terrorism is of concern because many people can be exposed to radiation for a wide range of doses. The Fukushima nuclear power plant accident was the most recent (March 2011). In 2010, an orphaned 60Co source from a scrap yard in Mayapuri, New Delhi, India, led to radiation exposure related death of one individual and exposure of seven others. In such scenarios, the number of persons exposed to radiation could be large, and assessment of the exposure levels by a single biodosimetry lab will be difficult. Preparedness for the medical management of radiation events during a nuclear emergency will require immediate assessment of absorbed doses as per the requirement and guidelines of the International Atomic Energy Agency (IAEA) and the World Health Organization (WHO) (IAEA
3
2011; William et al. 2009; Timins et al. 2004). Quantification of the absorbed radiation dose by measuring dicentric chromosomes (DC) from peripheral blood lymphocytes of exposed individuals is the gold standard and is recommended for all biodosimetry laboratories (Bender et al. 1988). While biodosimetry based cytogenetic assays like the cytokinesis-blocked micronucleus (CBMN) technique (Prosser et al. 1988), fluorescence in-situ hybridization (FISH) (Pinkel et al. 1986), and premature chromosome condensation (PCC) (IAEA 1986) have advantages as well as limitations, measurement of DC is both specific and sensitive to the absorbed radiation dose. Further, this technique also allows differentiation between whole and partial body exposures (IAEA 1986). To quantify the absorbed radiation dose, DC frequency obtained from exposed individuals is then inferred from a standard in vitro reference doseresponse curve of any biodosimetry lab validated by the IAEA or by the authorized regulatory agency of the respective country. Many countries have developed capabilities in biological dosimetry that form an integral part of national emergency response systems in the event of individual or mass radiation casualties (Roy et al. 2004). Further, to meet the requirement for triage dosimetry, the inter-lab comparison exercise has been conducted in many parts of the world to investigate the sensitivity of the methodology employed for absorbed dose quantification and reduce the uncertainties related to dose estimation (Garcia et al. 1995). Therefore, the IAEA recommends that each service lab develop a quality programme to ensure the robustness, accuracy and reproducibility of its procedures in the form of inter- and intra-lab comparisons. International accreditation bodies like the International Organization for Standardization (ISO) have contributed immensely to this goal by providing standard performance criteria (pertaining to quality assurance and quality control) for cytogenetic service laboratories undertaking biodosimetry (Roy et al. 2004).
4
Lab inter-comparisons also help to harmonize protocols with regard to culture conditions, scoring criteria and statistical analysis. This harmonization is essential if laboratories are to set up networks to respond to a mass casualty event in which the number of potentially exposed individuals to be analysed exceeds the response capabilities of the local responders (Di Giorgio et al. 2011). The mutual assistance of several laboratories is required in such cases to increase the number of samples to be processed and achieve faster availability of results. To accomplish this, in 2007, the WHO initiated “BioDoseNet”, a network of more than 30 laboratories around the world and implemented revised regulations pertaining to human health, including the field of radio-nuclear incidents (Christie et al. 2010). The need for networking and quality assurance in biodosimetry is imperative and this is well established in the USA (Wilkins et al. 2011), Canada (Miller et al. 2007), Japan (Yoshida et al. 2007), Europe (Romm et al. 2008; Wojcik et al. 2010) and Portuguese (Martins et al. 2013). To our knowledge, no networking among biodosimetry laboratories exists in India. In this regard, the present study was attempted to compare doseresponse curves generated independently by two Indian laboratories so that a step towards networking be achieved. Further, the DC frequency obtained by two different techniques: Giemsa-stained metaphase preparations were confirmed with centromere-specific FISH for further reliability and validity. 2. Materials and methods 2.1. Participating Laboratories Lab-I: Department of Human Genetics, Sri Ramachandra University (SRU), Porur, Chennai, Tamilnadu, India. Lab-II: Chemical Radio Protection and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences (INMAS), Timarpur, New Delhi, India.
5
2.2. Blood sample collection All necessary human ethical clearance for this study was obtained prior to undertaking this study at the coordinating lab (IEC-NI/11/OCT/25/49). Informed consent of three healthy donors with no known history of smoking or genotoxic exposure was obtained. Blood samples from volunteers were collected by standard procedure in lithium-heparin containers. 2.3. In vitro irradiation of blood lymphocytes Blood samples from each volunteer were divided into aliquots and exposed to nine or ten different doses (0.05, 0.1, 0.25, 0.5, 0.75, 1.0, 2.0, 3.0, 4.0 and 5.0 Gy) of radiation at two different dose-rates.
Whole-blood irradiation was performed using a
60
Co teletherapy unit
(Theratron Phoenix Cobalt-60, 2.0 Gy/min) at the Bernard Institute of Radiology, Govt. Hospital, Chennai, India, and a teletherapy unit available at INMAS (Bhabhatron II, Panacea Med. Tech. Pvt. Ltd., Bangalore, India, 0.7Gy/min). A sham-irradiated aliquot of blood sample was used as control. After irradiation, blood samples were transported back to the lab, maintained at 37°C for an hour, after which the samples were cultured in duplicates for metaphase chromosome preparation. (The radiation sources of the teletherapy units are routinely calibrated by a team of radiation safety personnel using a 0.6cc ionization chamber (Scanditronix, Germany). 2.4. Cell culture and metaphase chromosome preparation Blood cultures were initiated using 1ml whole blood in 80% RPMI-1640 (Invitrogen, USA) medium, supplemented with 20% FBS (Invitrogen, USA), 40μg/ml phytohemagglutinin (Invitrogen, USA) and antibiotics (Penicillin and Streptomycin (Invitrogen, USA) at final concentrations of 100IU/mL and 100μg/mL respectively) at 37°C in a 5% CO2 incubator. At 24 hours post-incubation, colchicine (0.02μg/ml) was added aseptically and incubation was
6
continued for another 24 hours. The cells were harvested by treating with hypotonic solution (0.075M KCl) and fixing with Carnoy’s fixative (3:1 methanol and acetic acid). Finally, the cells were washed twice with fixative and dropped onto clean glass slides. The slides were air-dried and later coded used for analysis. 2.5. Giemsa staining and analysis of slides The slides were immersed in 4% Giemsa stain (pH 6.8 phosphate buffer for 5 min), rinsed in distilled water, dried and sealed under glass cover slips with DPX slide mountant. One set of coded slides was transported to the other lab participating in the inter-lab comparison exercise. Scoring was carried out at both the laboratories as per international recommended IAEA guidelines (IAEA 2011). CA was scored manually by two persons in each lab to construct dose-response curves. This data was also used for intra-lab comparison to determine scorer variation / efficiency. The frequency of aberrations and standard error were also calculated for the number of cells scored. 2.6. Fluorescence in-situ hybridization (FISH) and chromosome analysis Slides were prepared as detailed earlier. An area with good quality and quantity of metaphase chromosomes was marked on the slides prepared. In brief, prepared slides were treated in a series of ethanol concentrations (80%, 90% and 100%) and air dried. The pancentromeric probe (StarFISH, Cambio, Cambridge) employed for dicentric detection was diluted (1μl in 12.5 μl hybridization buffer) and loaded on the marked area which was then sealed with a glass cover slip. Each slide was placed in a moistened hybridization chamber (HyBriteTM, Vysis) which was programmed to perform co-denaturation at 73°C for 5 mins and hybridization at 37°C for 16-18 hrs. The slides were then washed in 0.4×SSC/0.3% IGEPAL solution at 68±1°C for 15-20 seconds and 2×SSC/0.1% IGEPAL solution at room temperature for 5-10 seconds. Cover
7
slips were then applied to the marked area with antifade-mountant containing DAPI (Vysis Inc, Downers Grove, USA) and the slides were stored at -20°C for an hour to enhance the probe signal. Chromosomal aberrations, in the form of dicentrics and rings, were scored manually to construct dose-response curves. For both the methods, only metaphase chromosomes with 46 centromeres were considered and each DC and ring chromosome (RC) scored was included only if accompanied by its respective acentric fragment. The slides were analyzed using a fluorescence microscope with appropriate filters. 2.7. Statistical analysis 2.7.1. Comparison of aberration yields The mean frequency of DC and RC scored at each dose by the two laboratories was compared to assess lab performance. The unpaired‘t’ test was used to compare mean aberration frequencies obtained at each dose by individual scorers for intra-lab comparison. The Kolmogorov-Smirnov and analysis of variance (ANOVA) tests were used to check scorer variation on the overall aberration frequencies obtained by each lab for inter-lab comparison. 2.7.2. Dose-response curve fitting From the aberration yields, the distribution pattern in cells was ascertained by the standard ‘u’ test as described in a recent IAEA publication (IAEA 2011). Curve fitting was done with the help of statistical software known as “Poly Fit,” developed by the National Radiation Protection Board (NRPB), UK (Edwards et al. 1973). The essence of this program is to fit the data points by a weighted least squares method, taking into account the scoring effort for each point. 2.7.3. Comparison of calibration curve co-efficient
8
Differences in calibration curves across the full dose range were analyzed using the t-test with Bonferroni correction to check whether , ß and c among the two laboratories fell within the respective 99.2% confidence interval (CI). 3 RESULTS 3.1. Intra- and inter-lab comparisons of DC frequency obtained from blood samples exposed to gamma radiation Scoring was carried out by two trained cytogeneticists, designated scorers A and B who scored 1 or more slides for each dose. In Lab-I, scorer A scored cells in excess of 500 for each dose, while scorer B scored 250 cells for low doses (upto 1Gy) and 100 cells for high doses (25Gy). In Lab-II, both scorers scored 1000 cells or more for each dose. The number of cells scored and the frequency of CA obtained by scorers in both labs are given in tables 1 and 2. Comparison of mean aberration frequencies obtained by scorers A and B at both labs using the ttest did not show any significant difference for most doses except at 1Gy and 2Gy in lab-I and 1Gy and 5Gy in lab-II (p< 0.05). However, when overall aberration frequencies were compared using the Kolmogorov-Smirnov test, the difference in the yield of aberrations was not statistically significant among scorers in both labs (p=0.96). As there was no significant difference in the aberration frequency obtained by scorers within the labs, the data on the aberration frequency obtained by the two scorers in each lab was pooled together and used for inter-lab comparison. Inter-lab comparison using the unpaired ‘t’ test revealed that aberration yields did not differ significantly between the two labs for doses up to 3Gy; however, at 4 and 5Gy doses, a significant difference was observed (p0.008). The coefficients from lab I and II are comparable with those obtained by other laboratories (Wilkins et al. 2011; Lloyd et al. 2006; Beinke et al. 2011), which shows the ability of the two participating laboratories in undertaking or sharing resources at
14
times of large-scale radionuclear emergencies. The overall results suggest that the obtained doseresponse curves can be readily used for dose estimation in triage scenarios to share the workload between laboratories. In a similar comparative exercise conducted by the Armed Forces Radiobiology Research Institute (AFRRI) where four other laboratories performed irradiation and analysis independently, dose-response coefficients of only two labs matched well with the AFRRI curves (Wilkins et al. 2011). The dose rate employed for irradiation was suggested as a contributing factor for the observed difference. It has been reported that dose-response curves for the same sample by different laboratories may vary considerably due to the contribution of external influencing factors such as culture conditions, slide preparation, metaphase cell selection and scoring (IAEA 2001). Therefore, we strongly feel that in scenarios where sample irradiation is performed by a single lab and scoring carried out by two laboratories, adherence to similar and consistent scoring criteria can reduce the observed variation to non-significant level for reliable dose estimation, which is critical in triage scenarios. Though many international networks exist for improvisation of dose estimation in Latin America (Garcia et al. 1995), Canada (Miller et al. 2007), Japan (Yoshida et al. 2007) and Europe (Romm et al. 2008; Wojcik et al. 2010) and inter-comparison exercises have been conducted in many countries, this is the first published report from India. It is hoped that the outcome from such studies and many others that foster collaborative research (Garcia et al. 1995, Miller et al. 2007; Yoshida et al. 2007) will ease cooperation among biological dosimetry laboratories, giving rise to more validating networks across the country and globally. 5 CONCLUSIONS The overall results suggest that despite minor variations in DC yield at higher doses (4 and 5 Gy), obtained from blood lymphocytes exposed to gamma radiation, the obtained co-
15
efficient, slope and intercept of dose-response curves do not show any significant difference between the labs. Further, as the frequency of DC did not show any significant difference when scored by conventional Giemsa-staining or centromere-specific FISH at any given dose, DC scoring can be readily used for dose estimation during triage scenarios to share workload among laboratories. FUNDING INFORMATION This work was supported by a grant from INMAS, DRDO to Sri Ramachandra University, Chennai (INM-311/1.7). ACKNOWLEDGEMENTS We wish to acknowledge Dr. K. Thayalan (Chief Medical Physicist, Dr. Kamakshi Memorial Hospital, Pallikaranai, Chennai) and radiotherapy unit staff (Bernard Institute of Radiology, Government General Hospital, Chennai) for permitting and helping us to use the teletherapy unit. We thank Dr. R.C. Wilkins and co-workers (AFRRI, USA) for helping us understand the Bonferroni correction. The authors also express their thanks to Dr. Solomon F.D. Paul, Professor and Head, Department of Human Genetics, College of Biomedical Sciences, Technology and Research, Sri Ramachandra University, Chennai, for his constructive inputs on the data analysis and interpretation. CONFLICT OF INTEREST STATEMENT The authors declare no conflict of interest.
Figure legends
16
Figure 1: Dose-response calibration curve obtained from two labs for
60
Co gamma rays (dose
rate: 2 Gy/min; Dose: 0-5Gy) using dicentric and centric ring yields as fitted by a linear quadratic model, Y = c + D + D2. (Lab-I is SRU and Lab-II is INMAS) Figure 2: Comparison of the dicentric and centric ring yields in peripheral blood lymphocytes exposed to
60
Co gamma radiation (Dose rate: 0.7 Gy/min; Dose: 0-5Gy) as obtained by two
methods (conventional and FISH). Each bar represents the mean ± SE of CA frequency obtained by two methods (Insert: Bar graph represents the mean ± SE of CA frequency obtained for 00.75 Gy by two methods). Significance levels of p
Dicentric chromosome aberration analysis using Giemsa and centromere specific fluorescence in-situ hybridization for biological dosimetry: An inter- and intra-laboratory comparison in Indian laboratories M. Bhavani, G. Tamizh Selvan, Harpreet Kaur, J. S. Adhikari, J. Vijayalakshmi, P. Venkatachalam, N.K. Chaudhury www.elsevier.com/locate/apradiso
PII: DOI: Reference:
S0969-8043(14)00232-2 http://dx.doi.org/10.1016/j.apradiso.2014.06.004 ARI6706
To appear in:
Applied Radiation and Isotopes
Received date: 19 February 2014 Revised date: 19 May 2014 Accepted date: 4 June 2014 Cite this article as: M. Bhavani, G. Tamizh Selvan, Harpreet Kaur, J.S. Adhikari, J. Vijayalakshmi, P. Venkatachalam, N.K. Chaudhury, Dicentric chromosome aberration analysis using Giemsa and centromere specific fluorescence in-situ hybridization for biological dosimetry: An inter- and intra-laboratory comparison in Indian laboratories, Applied Radiation and Isotopes, http://dx.doi. org/10.1016/j.apradiso.2014.06.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Dicentric chromosome aberration analysis using Giemsa and centromere specific fluorescence in-situ hybridization for biological dosimetry: An inter- and intra-laboratory comparison in Indian laboratories. M. Bhavani1, G. Tamizh Selvan1,2, Harpreet Kaur1, J.S. Adhikari2, J. Vijayalakshmi1, P. Venkatachalam1*, N.K. Chaudhury2 *. 1
Sri Ramachandra University, Porur, Chennai 600 116, Tamil Nadu, India.
2
Institute of Nuclear Medicine and Allied sciences, DRDO, Timarpur, New Delhi 110 054, India.
BHAVANI MANIVANNAN - Master of Science, Research Scholar, Department of Human Genetics, College of Biomedical Science Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] TAMIZH SELVAN GNANA SEKARAN- Master of Science, Research Scholar, Department of Human Genetics, College of Biomedical Science Technology and Research, Sri Ramachandra University, Porur, Chennai- 600 116 and Institute of Nuclear Medicine and Allied Sciences Brig Mazumdar Road, Timarpur, Delhi-110 054, India. [email protected] Dr. HARPREET KAUR - Ph.D. Assistant Professor, Department of Human Genetics, College of Biomedical Science, Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] Dr. JAWAHAR SINGH ADHIKARI - Ph.D. Scientist E, Chemical Radioprotector and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences, Brig Mazumdar Road, Timarpur, Delhi, India-110 054. [email protected]
1
Dr. VIJAYALAKSHMI JAGANATHAN - Ph.D. Assistant Professor, Department of Human Genetics, College of Biomedical Sciences, Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] Dr. Perumal Venkatachalam- Ph.D. Professor, Department of Human Genetics, College of Biomedical Science Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected] Dr. NABO KUMAR CHAUDHURY- Ph.D. Scientist F, Group Head, Chemical Radioprotector and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences, Brig Mazumdar Road, Timarpur, Delhi, India-110 054. [email protected] *Correspondence (1): Dr. NABO KUMAR CHAUDHURY- Ph.D. Scientist F, Group Head, Chemical Radioprotector and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences, Brig Mazumdar Road, Timarpur, Delhi, India-110 054. [email protected] Tel No: 011-23905131/2397820. Fax: 011-23919509 *Correspondence (2): Dr. PERUMAL VENKATACHALAM- Ph.D. Professor, Department of Human Genetics, College of Biomedical Sciences, Technology and Research, Sri Ramachandra University, Porur, Chennai, India - 600 116. [email protected]; Tel No: 044-24768027 Extn 237. Fax: 044-24767008
Running title: Networking of biodosimetry laboratories
2
Abstract: To facilitate efficient handling of large samples, an attempt towards networking of laboratories in India for biological dosimetry was carried out. Human peripheral blood samples were exposed to 60
Co -radiation for ten different doses (0-5Gy) at a dose rate of 0.7 and 2Gy/min. The
chromosomal aberrations (CA) were scored in Giemsa-stained and fluorescence in-situ hybridization with centromere-specific probes. No significant difference (p>0.05) was observed in the CA yield for given doses except 4 and 5Gy, between the laboratories, among the scorers and also staining methods adapted suggest the reliability and validates the inter-lab comparisons exercise for triage applications.
Key words: Biodosimetry; Dicentric chromosome; FISH and Dose-response curve.
1. Introduction The threat of radiological accidents or nuclear terrorism is of concern because many people can be exposed to radiation for a wide range of doses. The Fukushima nuclear power plant accident was the most recent (March 2011). In 2010, an orphaned 60Co source from a scrap yard in Mayapuri, New Delhi, India, led to radiation exposure related death of one individual and exposure of seven others. In such scenarios, the number of persons exposed to radiation could be large, and assessment of the exposure levels by a single biodosimetry lab will be difficult. Preparedness for the medical management of radiation events during a nuclear emergency will require immediate assessment of absorbed doses as per the requirement and guidelines of the International Atomic Energy Agency (IAEA) and the World Health Organization (WHO) (IAEA
3
2011; William et al. 2009; Timins et al. 2004). Quantification of the absorbed radiation dose by measuring dicentric chromosomes (DC) from peripheral blood lymphocytes of exposed individuals is the gold standard and is recommended for all biodosimetry laboratories (Bender et al. 1988). While biodosimetry based cytogenetic assays like the cytokinesis-blocked micronucleus (CBMN) technique (Prosser et al. 1988), fluorescence in-situ hybridization (FISH) (Pinkel et al. 1986), and premature chromosome condensation (PCC) (IAEA 1986) have advantages as well as limitations, measurement of DC is both specific and sensitive to the absorbed radiation dose. Further, this technique also allows differentiation between whole and partial body exposures (IAEA 1986). To quantify the absorbed radiation dose, DC frequency obtained from exposed individuals is then inferred from a standard in vitro reference doseresponse curve of any biodosimetry lab validated by the IAEA or by the authorized regulatory agency of the respective country. Many countries have developed capabilities in biological dosimetry that form an integral part of national emergency response systems in the event of individual or mass radiation casualties (Roy et al. 2004). Further, to meet the requirement for triage dosimetry, the inter-lab comparison exercise has been conducted in many parts of the world to investigate the sensitivity of the methodology employed for absorbed dose quantification and reduce the uncertainties related to dose estimation (Garcia et al. 1995). Therefore, the IAEA recommends that each service lab develop a quality programme to ensure the robustness, accuracy and reproducibility of its procedures in the form of inter- and intra-lab comparisons. International accreditation bodies like the International Organization for Standardization (ISO) have contributed immensely to this goal by providing standard performance criteria (pertaining to quality assurance and quality control) for cytogenetic service laboratories undertaking biodosimetry (Roy et al. 2004).
4
Lab inter-comparisons also help to harmonize protocols with regard to culture conditions, scoring criteria and statistical analysis. This harmonization is essential if laboratories are to set up networks to respond to a mass casualty event in which the number of potentially exposed individuals to be analysed exceeds the response capabilities of the local responders (Di Giorgio et al. 2011). The mutual assistance of several laboratories is required in such cases to increase the number of samples to be processed and achieve faster availability of results. To accomplish this, in 2007, the WHO initiated “BioDoseNet”, a network of more than 30 laboratories around the world and implemented revised regulations pertaining to human health, including the field of radio-nuclear incidents (Christie et al. 2010). The need for networking and quality assurance in biodosimetry is imperative and this is well established in the USA (Wilkins et al. 2011), Canada (Miller et al. 2007), Japan (Yoshida et al. 2007), Europe (Romm et al. 2008; Wojcik et al. 2010) and Portuguese (Martins et al. 2013). To our knowledge, no networking among biodosimetry laboratories exists in India. In this regard, the present study was attempted to compare doseresponse curves generated independently by two Indian laboratories so that a step towards networking be achieved. Further, the DC frequency obtained by two different techniques: Giemsa-stained metaphase preparations were confirmed with centromere-specific FISH for further reliability and validity. 2. Materials and methods 2.1. Participating Laboratories Lab-I: Department of Human Genetics, Sri Ramachandra University (SRU), Porur, Chennai, Tamilnadu, India. Lab-II: Chemical Radio Protection and Radiation Dosimetry Research Group, Institute of Nuclear Medicine and Allied Sciences (INMAS), Timarpur, New Delhi, India.
5
2.2. Blood sample collection All necessary human ethical clearance for this study was obtained prior to undertaking this study at the coordinating lab (IEC-NI/11/OCT/25/49). Informed consent of three healthy donors with no known history of smoking or genotoxic exposure was obtained. Blood samples from volunteers were collected by standard procedure in lithium-heparin containers. 2.3. In vitro irradiation of blood lymphocytes Blood samples from each volunteer were divided into aliquots and exposed to nine or ten different doses (0.05, 0.1, 0.25, 0.5, 0.75, 1.0, 2.0, 3.0, 4.0 and 5.0 Gy) of radiation at two different dose-rates.
Whole-blood irradiation was performed using a
60
Co teletherapy unit
(Theratron Phoenix Cobalt-60, 2.0 Gy/min) at the Bernard Institute of Radiology, Govt. Hospital, Chennai, India, and a teletherapy unit available at INMAS (Bhabhatron II, Panacea Med. Tech. Pvt. Ltd., Bangalore, India, 0.7Gy/min). A sham-irradiated aliquot of blood sample was used as control. After irradiation, blood samples were transported back to the lab, maintained at 37°C for an hour, after which the samples were cultured in duplicates for metaphase chromosome preparation. (The radiation sources of the teletherapy units are routinely calibrated by a team of radiation safety personnel using a 0.6cc ionization chamber (Scanditronix, Germany). 2.4. Cell culture and metaphase chromosome preparation Blood cultures were initiated using 1ml whole blood in 80% RPMI-1640 (Invitrogen, USA) medium, supplemented with 20% FBS (Invitrogen, USA), 40μg/ml phytohemagglutinin (Invitrogen, USA) and antibiotics (Penicillin and Streptomycin (Invitrogen, USA) at final concentrations of 100IU/mL and 100μg/mL respectively) at 37°C in a 5% CO2 incubator. At 24 hours post-incubation, colchicine (0.02μg/ml) was added aseptically and incubation was
6
continued for another 24 hours. The cells were harvested by treating with hypotonic solution (0.075M KCl) and fixing with Carnoy’s fixative (3:1 methanol and acetic acid). Finally, the cells were washed twice with fixative and dropped onto clean glass slides. The slides were air-dried and later coded used for analysis. 2.5. Giemsa staining and analysis of slides The slides were immersed in 4% Giemsa stain (pH 6.8 phosphate buffer for 5 min), rinsed in distilled water, dried and sealed under glass cover slips with DPX slide mountant. One set of coded slides was transported to the other lab participating in the inter-lab comparison exercise. Scoring was carried out at both the laboratories as per international recommended IAEA guidelines (IAEA 2011). CA was scored manually by two persons in each lab to construct dose-response curves. This data was also used for intra-lab comparison to determine scorer variation / efficiency. The frequency of aberrations and standard error were also calculated for the number of cells scored. 2.6. Fluorescence in-situ hybridization (FISH) and chromosome analysis Slides were prepared as detailed earlier. An area with good quality and quantity of metaphase chromosomes was marked on the slides prepared. In brief, prepared slides were treated in a series of ethanol concentrations (80%, 90% and 100%) and air dried. The pancentromeric probe (StarFISH, Cambio, Cambridge) employed for dicentric detection was diluted (1μl in 12.5 μl hybridization buffer) and loaded on the marked area which was then sealed with a glass cover slip. Each slide was placed in a moistened hybridization chamber (HyBriteTM, Vysis) which was programmed to perform co-denaturation at 73°C for 5 mins and hybridization at 37°C for 16-18 hrs. The slides were then washed in 0.4×SSC/0.3% IGEPAL solution at 68±1°C for 15-20 seconds and 2×SSC/0.1% IGEPAL solution at room temperature for 5-10 seconds. Cover
7
slips were then applied to the marked area with antifade-mountant containing DAPI (Vysis Inc, Downers Grove, USA) and the slides were stored at -20°C for an hour to enhance the probe signal. Chromosomal aberrations, in the form of dicentrics and rings, were scored manually to construct dose-response curves. For both the methods, only metaphase chromosomes with 46 centromeres were considered and each DC and ring chromosome (RC) scored was included only if accompanied by its respective acentric fragment. The slides were analyzed using a fluorescence microscope with appropriate filters. 2.7. Statistical analysis 2.7.1. Comparison of aberration yields The mean frequency of DC and RC scored at each dose by the two laboratories was compared to assess lab performance. The unpaired‘t’ test was used to compare mean aberration frequencies obtained at each dose by individual scorers for intra-lab comparison. The Kolmogorov-Smirnov and analysis of variance (ANOVA) tests were used to check scorer variation on the overall aberration frequencies obtained by each lab for inter-lab comparison. 2.7.2. Dose-response curve fitting From the aberration yields, the distribution pattern in cells was ascertained by the standard ‘u’ test as described in a recent IAEA publication (IAEA 2011). Curve fitting was done with the help of statistical software known as “Poly Fit,” developed by the National Radiation Protection Board (NRPB), UK (Edwards et al. 1973). The essence of this program is to fit the data points by a weighted least squares method, taking into account the scoring effort for each point. 2.7.3. Comparison of calibration curve co-efficient
8
Differences in calibration curves across the full dose range were analyzed using the t-test with Bonferroni correction to check whether , ß and c among the two laboratories fell within the respective 99.2% confidence interval (CI). 3 RESULTS 3.1. Intra- and inter-lab comparisons of DC frequency obtained from blood samples exposed to gamma radiation Scoring was carried out by two trained cytogeneticists, designated scorers A and B who scored 1 or more slides for each dose. In Lab-I, scorer A scored cells in excess of 500 for each dose, while scorer B scored 250 cells for low doses (upto 1Gy) and 100 cells for high doses (25Gy). In Lab-II, both scorers scored 1000 cells or more for each dose. The number of cells scored and the frequency of CA obtained by scorers in both labs are given in tables 1 and 2. Comparison of mean aberration frequencies obtained by scorers A and B at both labs using the ttest did not show any significant difference for most doses except at 1Gy and 2Gy in lab-I and 1Gy and 5Gy in lab-II (p< 0.05). However, when overall aberration frequencies were compared using the Kolmogorov-Smirnov test, the difference in the yield of aberrations was not statistically significant among scorers in both labs (p=0.96). As there was no significant difference in the aberration frequency obtained by scorers within the labs, the data on the aberration frequency obtained by the two scorers in each lab was pooled together and used for inter-lab comparison. Inter-lab comparison using the unpaired ‘t’ test revealed that aberration yields did not differ significantly between the two labs for doses up to 3Gy; however, at 4 and 5Gy doses, a significant difference was observed (p0.008). The coefficients from lab I and II are comparable with those obtained by other laboratories (Wilkins et al. 2011; Lloyd et al. 2006; Beinke et al. 2011), which shows the ability of the two participating laboratories in undertaking or sharing resources at
14
times of large-scale radionuclear emergencies. The overall results suggest that the obtained doseresponse curves can be readily used for dose estimation in triage scenarios to share the workload between laboratories. In a similar comparative exercise conducted by the Armed Forces Radiobiology Research Institute (AFRRI) where four other laboratories performed irradiation and analysis independently, dose-response coefficients of only two labs matched well with the AFRRI curves (Wilkins et al. 2011). The dose rate employed for irradiation was suggested as a contributing factor for the observed difference. It has been reported that dose-response curves for the same sample by different laboratories may vary considerably due to the contribution of external influencing factors such as culture conditions, slide preparation, metaphase cell selection and scoring (IAEA 2001). Therefore, we strongly feel that in scenarios where sample irradiation is performed by a single lab and scoring carried out by two laboratories, adherence to similar and consistent scoring criteria can reduce the observed variation to non-significant level for reliable dose estimation, which is critical in triage scenarios. Though many international networks exist for improvisation of dose estimation in Latin America (Garcia et al. 1995), Canada (Miller et al. 2007), Japan (Yoshida et al. 2007) and Europe (Romm et al. 2008; Wojcik et al. 2010) and inter-comparison exercises have been conducted in many countries, this is the first published report from India. It is hoped that the outcome from such studies and many others that foster collaborative research (Garcia et al. 1995, Miller et al. 2007; Yoshida et al. 2007) will ease cooperation among biological dosimetry laboratories, giving rise to more validating networks across the country and globally. 5 CONCLUSIONS The overall results suggest that despite minor variations in DC yield at higher doses (4 and 5 Gy), obtained from blood lymphocytes exposed to gamma radiation, the obtained co-
15
efficient, slope and intercept of dose-response curves do not show any significant difference between the labs. Further, as the frequency of DC did not show any significant difference when scored by conventional Giemsa-staining or centromere-specific FISH at any given dose, DC scoring can be readily used for dose estimation during triage scenarios to share workload among laboratories. FUNDING INFORMATION This work was supported by a grant from INMAS, DRDO to Sri Ramachandra University, Chennai (INM-311/1.7). ACKNOWLEDGEMENTS We wish to acknowledge Dr. K. Thayalan (Chief Medical Physicist, Dr. Kamakshi Memorial Hospital, Pallikaranai, Chennai) and radiotherapy unit staff (Bernard Institute of Radiology, Government General Hospital, Chennai) for permitting and helping us to use the teletherapy unit. We thank Dr. R.C. Wilkins and co-workers (AFRRI, USA) for helping us understand the Bonferroni correction. The authors also express their thanks to Dr. Solomon F.D. Paul, Professor and Head, Department of Human Genetics, College of Biomedical Sciences, Technology and Research, Sri Ramachandra University, Chennai, for his constructive inputs on the data analysis and interpretation. CONFLICT OF INTEREST STATEMENT The authors declare no conflict of interest.
Figure legends
16
Figure 1: Dose-response calibration curve obtained from two labs for
60
Co gamma rays (dose
rate: 2 Gy/min; Dose: 0-5Gy) using dicentric and centric ring yields as fitted by a linear quadratic model, Y = c + D + D2. (Lab-I is SRU and Lab-II is INMAS) Figure 2: Comparison of the dicentric and centric ring yields in peripheral blood lymphocytes exposed to
60
Co gamma radiation (Dose rate: 0.7 Gy/min; Dose: 0-5Gy) as obtained by two
methods (conventional and FISH). Each bar represents the mean ± SE of CA frequency obtained by two methods (Insert: Bar graph represents the mean ± SE of CA frequency obtained for 00.75 Gy by two methods). Significance levels of p
