TY - JOUR
AU - Dertinger, Stephen, D
AB - Abstract The in vivo Pig-a assay is being used in safety studies to evaluate the potential of chemicals to induce somatic cell gene mutations. Ongoing work is aimed at developing an Organisation for Economic Cooperation and Development (OECD) test guideline to support routine use for regulatory purposes (OECD project number 4.93). Among the details that will need to be articulated in an eventual guideline are recommended treatment and harvest schedules. With this in mind, experiments reported herein were performed with Wistar Han rats exposed to aristolochic acid I (AA), 1,3-propane sultone, chlorambucil, thiotepa or melphalan using each of two commonly used treatment schedules: 3 or 28 consecutive days. In the case of the 3-day studies, blood was collected for Pig-a analysis on days 15 or 16 and 29 or 30. For the 28-day studies blood was collected on day 29 or 30. The effect of treatment on mutant reticulocytes and mutant erythrocytes was evaluated with parametric pair-wise tests. While each of the five mutagens increased mutant phenotype cell frequencies irrespective of study design, statistical significance was consistently achieved at lower dose levels when the 28-day format was used (e.g. 2.75 vs 20 mg/kg/bw for AA). To more thoroughly investigate the dose–response relationships, benchmark dose (BMD) analyses were performed with PROAST software. These results corroborate the pair-wise testing results in that lower BMD values were obtained with the 28-day design. Finally, mutagenic potency, as measured by BMD analyses, most consistently correlated with the mutagens’ tumorigenic dose 50 values when the lengthier treatment schedule was used. Collectively, these results suggest that both 3- and 28-day treatment schedules have merit in hazard identification-type studies. That being said, for the purpose of regulatory safety assessments, there are clear advantages to study designs that utilise protracted exposures. Introduction The phosphatidylinositol glycan-class A (Pig-a) gene codes an enzyme that is essential for glycosylphosphatidylinositol (GPI) anchor synthesis (1,2). Inactivating Pig-a mutations therefore result in cells devoid of cell surface GPI-anchored protein(s), and this phenotype represents a reliable reporter of Pig-a mutation (3–,8). The analytical approach used to study this phenotype most often utilises flow cytometry, in conjunction with fluorescent antibodies against GPI-anchored cell surface epitopes, to enumerate non-fluorescent Pig-a mutant cells relative to fluorescent wild-type cells (9). Rodent studies have focussed on erythrocytes, as these cells are easily obtained in sufficient quantity via small volume blood draws (9). The low blood volume requirement, the Pig-a assay’s compatibility with commonly used rodent strains, and the relatively low cost of these studies relative to other test systems makes this assay particularly attractive for studies of somatic cell mutations (10,11). The erythrocyte-based Pig-a assay is considered useful for certain regulatory safety assessment programs as a means for investigating the potential of chemicals to induce somatic cell gene mutations in vivo. For example, as described by the ICH M7 impurities guideline, the rodent Pig-a assay is one of the recommended follow-up tests to a positive bacterial mutagenicity finding (12). This and other possible use cases have led to efforts to develop an Organisation for Economic Cooperation and Development (OECD) test guideline to support regulatory safety assessments (OECD project number 4.93). Among the details that will need to be articulated in any eventual guideline are recommendations for appropriate and preferred treatment and harvest schedules. With this in mind, experiments were performed with Wistar Han rats exposed to aristolochic acid I (AA), 1,3-propane sultone (PS), chlorambucil (Chlor), thiotepa (Thio) or melphalan (Mel) using each of two treatment schedules that have been most commonly described in peer-reviewed publications: 3 and 28 consecutive days of administration. According to the OECD Pig-a Detailed Review Paper (13), there is a recommendation to harvest blood at a minimum of two times points in the case of acute treatment schedules, and within days of final exposure in the case of a 28-day repeat-dose study. This advice maximises the likelihood of observing bona fide mutagen-induced effects in both the immature portion of circulating erythrocytes (i.e. reticulocytes or RETs) and the later-induced total erythrocyte pool (i.e. red blood cells or RBCs). We followed these Detailed Review Paper recommendations and expect our qualitative and quantitative comparisons across study designs to be a valuable addition to the OECD effort that will need to articulate recommended treatment/harvest schedule(s). Materials and Methods Reagents PS (CAS No. 1120-71-4), Chlor (CAS No. 305-03-3), Thio (CAS No. 52-24-4), Mel (CAS No. 148-82-3) and absolute ethanol were purchased from Sigma-Aldrich, Schnelldorf, Germany. AA (CAS No. 313-67-7) was purchased from Acros Organics, Geel, Belgium. Reagents used for flow cytometric enumeration of mutant erythrocytes (mutant RBC) and mutant reticulocytes (mutant RET) were from Rat MutaFlow® Kits (Litron Laboratories, Rochester, NY) and included Anticoagulant Solution, Buffer Solution, Nucleic Acid Dye Solution (contains SYTO® 13), Anti-CD59-PE and Anti-CD61-PE. Additional supplies included Lympholyte®-Mammal cell separation reagent from CedarLane, Burlington, NC; Anti-PE MicroBeads, LS Columns and a QuadroMACS™ Separator from Miltenyi Biotec, Bergisch Gladbach, Germany; and CountBright™ Absolute Count Beads and foetal bovine serum (FBS) from Invitrogen, Carlsbad, CA. Animals, treatments, blood harvests Experiments were conducted in compliance with all applicable Swiss and Novartis regulations that cover laboratory rodent studies. Male Wistar Han rats were purchased from Charles River Laboratories, Germany. Rodents were allowed to acclimate for approximately 1 week, and the age at start of treatment was approximately 7 weeks. Water and food were available ad libitum throughout the acclimation and experimental periods. Additional details about the eight studies are provided in Table 1, including the number of animals per group, exposure and blood harvest schedules, dose levels, etc. The dose levels were chosen based on dose range-finding experiments and/or the literature (14,15). In all cases, rats were either exposed to a given test agent at approximately 24 h intervals on days 1–3 or 1–28 via oral gavage administration in a volume of 10 ml/kg/body weight. Table 1. Study designs Study No. . Chemical . Vehicle, route . Dose groups (mg/kg/day) . No. rats per treatment group . Treatment schedulea . Blood samplinga . 1 AA PBS, pH 5.5–6, oral gavage 0, 10, 20, 30 7 Days 1–3 Days 15 and 30 2 AA PBS, pH 5.5–6, oral gavage 0, 2.75, 5.5, 11 7 Days 1–28 Day 29 3 PS Distilled water, oral gavage 0, 15, 30, 60 6 Days 1–3 Days 16 and 30 4 PS Distilled water, oral gavage 0, 12.5, 25, 50 6 Days 1–28 Day 29 5 Chlor 10% ethanol, 90% water, oral gavage 0, 3, 6, 12 7 Days 1–3 Days 15 and 30 6 Chlor 10% ethanol, 90% water, oral gavage 0, 1.5, 3, 6 7 Days 1–28 Day 29 7 Thio Distilled water, oral gavage 0, 3.75, 7.5 6 Days 1–3 Days 15 and 29 8 Thio Distilled water, oral gavage 0, 2, 4, 8 6 Days 1–28 Day 30 9 Mel Distilled water, oral gavage 0, 0.75, 1.5, 3 6 Days 1–3 Days 15 and 30 10 Mel Distilled water, oral gavage 0, 0.1875, 0.375, 0.75 6 Days 1–28 Day 30 Study No. . Chemical . Vehicle, route . Dose groups (mg/kg/day) . No. rats per treatment group . Treatment schedulea . Blood samplinga . 1 AA PBS, pH 5.5–6, oral gavage 0, 10, 20, 30 7 Days 1–3 Days 15 and 30 2 AA PBS, pH 5.5–6, oral gavage 0, 2.75, 5.5, 11 7 Days 1–28 Day 29 3 PS Distilled water, oral gavage 0, 15, 30, 60 6 Days 1–3 Days 16 and 30 4 PS Distilled water, oral gavage 0, 12.5, 25, 50 6 Days 1–28 Day 29 5 Chlor 10% ethanol, 90% water, oral gavage 0, 3, 6, 12 7 Days 1–3 Days 15 and 30 6 Chlor 10% ethanol, 90% water, oral gavage 0, 1.5, 3, 6 7 Days 1–28 Day 29 7 Thio Distilled water, oral gavage 0, 3.75, 7.5 6 Days 1–3 Days 15 and 29 8 Thio Distilled water, oral gavage 0, 2, 4, 8 6 Days 1–28 Day 30 9 Mel Distilled water, oral gavage 0, 0.75, 1.5, 3 6 Days 1–3 Days 15 and 30 10 Mel Distilled water, oral gavage 0, 0.1875, 0.375, 0.75 6 Days 1–28 Day 30 aAll time points are relative to start of treatment, which is defined as day 1. See Introduction section for chemical abbreviations. PBS = Phosphate buffered saline. Open in new tab Table 1. Study designs Study No. . Chemical . Vehicle, route . Dose groups (mg/kg/day) . No. rats per treatment group . Treatment schedulea . Blood samplinga . 1 AA PBS, pH 5.5–6, oral gavage 0, 10, 20, 30 7 Days 1–3 Days 15 and 30 2 AA PBS, pH 5.5–6, oral gavage 0, 2.75, 5.5, 11 7 Days 1–28 Day 29 3 PS Distilled water, oral gavage 0, 15, 30, 60 6 Days 1–3 Days 16 and 30 4 PS Distilled water, oral gavage 0, 12.5, 25, 50 6 Days 1–28 Day 29 5 Chlor 10% ethanol, 90% water, oral gavage 0, 3, 6, 12 7 Days 1–3 Days 15 and 30 6 Chlor 10% ethanol, 90% water, oral gavage 0, 1.5, 3, 6 7 Days 1–28 Day 29 7 Thio Distilled water, oral gavage 0, 3.75, 7.5 6 Days 1–3 Days 15 and 29 8 Thio Distilled water, oral gavage 0, 2, 4, 8 6 Days 1–28 Day 30 9 Mel Distilled water, oral gavage 0, 0.75, 1.5, 3 6 Days 1–3 Days 15 and 30 10 Mel Distilled water, oral gavage 0, 0.1875, 0.375, 0.75 6 Days 1–28 Day 30 Study No. . Chemical . Vehicle, route . Dose groups (mg/kg/day) . No. rats per treatment group . Treatment schedulea . Blood samplinga . 1 AA PBS, pH 5.5–6, oral gavage 0, 10, 20, 30 7 Days 1–3 Days 15 and 30 2 AA PBS, pH 5.5–6, oral gavage 0, 2.75, 5.5, 11 7 Days 1–28 Day 29 3 PS Distilled water, oral gavage 0, 15, 30, 60 6 Days 1–3 Days 16 and 30 4 PS Distilled water, oral gavage 0, 12.5, 25, 50 6 Days 1–28 Day 29 5 Chlor 10% ethanol, 90% water, oral gavage 0, 3, 6, 12 7 Days 1–3 Days 15 and 30 6 Chlor 10% ethanol, 90% water, oral gavage 0, 1.5, 3, 6 7 Days 1–28 Day 29 7 Thio Distilled water, oral gavage 0, 3.75, 7.5 6 Days 1–3 Days 15 and 29 8 Thio Distilled water, oral gavage 0, 2, 4, 8 6 Days 1–28 Day 30 9 Mel Distilled water, oral gavage 0, 0.75, 1.5, 3 6 Days 1–3 Days 15 and 30 10 Mel Distilled water, oral gavage 0, 0.1875, 0.375, 0.75 6 Days 1–28 Day 30 aAll time points are relative to start of treatment, which is defined as day 1. See Introduction section for chemical abbreviations. PBS = Phosphate buffered saline. Open in new tab As described in above and in Table 1, serial blood samples were collected at two time points in the case of the 3-day treatment studies, and at one time point in the case of the 28-day repeat-dose experiments. Blood collection occurred on isoflurane-anaesthetised rats via puncture of the sublingual vein. Whole blood (80 µl per rat) was transferred to tubes containing 100 µl kit-supplied heparin solution where they remained at room temperature for less than 2 h until leukodepletion as described previously (14). Pig-a sample preparation, data acquisition Mutant RET and mutant RBC frequencies were determined via immunomagnetic depletion of wild-type erythrocytes and flow cytometric analysis, as described previously (14,16,17). In addition to reducing analysis times to approximately 4 min per sample, immunomagnetic depletion made it practical to evaluate many times more cells than is otherwise feasible. Data acquisition was accomplished with a LSR II flow cytometer running FACSDiva software (v8.1). As described by the MutaFlow Instruction Manual, v140403 (www.litronlabs.com), an Instrument Calibration Standard was generated on each day of data acquisition. These samples contained approximately 50% wild-type and 50% mutant-mimic erythrocytes, and provided a means to rationally and consistently define the location of CD59-negative cells (18). Calculations, statistical analyses The formulas used to calculate mutant RBC and mutant RET frequencies based on pre- and post-immunomagnetic column data are described in the MutaFlow manual (www.litronlabs.com). Throughout this report the incidence of mutant phenotype cells is expressed as number per 106 cells. All mutant cell frequencies and averages were calculated in Microsoft® Excel for Mac® (v16.16.14). For statistical evaluations, mutant RBC and mutant RET frequencies were log(10) transformed in order to satisfy the homogeneity of variance assumption associated with parametric tests. Since zero mutant RET values were occasionally observed, a 0.1 offset was added to each mutant RET per 106 number prior to log transformation. The effect of treatment on transformed mutant RBC and mutant RET frequencies data was compared with concurrent vehicle control rats using Dunnett’s multiple comparisons test (JMP, v12.0.1, SAS Institute Inc., Cary, NC). Significance was evaluated using an alpha of 0.05. Strict reliance on pair-wise testing and comparisons across studies’ no observed effect levels/lowest observed effect levels is problematic when assessing the sensitivity of different experimental designs and/or chemicals’ potencies. For one, it is overdependent on the dose levels chosen. The interested reader is directed to a report by MacGregor et al. (19). We therefore used those authors’ recommendation and conducted benchmark dose (BMD) analyses as well (20). BMD analyses were performed for four chemicals with established tumorigenic dose 50 (TD50) values (AA is the exception). The endpoints evaluated were mutant RBC and mutant RET frequencies, and we used PROAST v67.0, accessed through the European Food Safety Authority online tool (see https://efsa.onlinelibrary.wiley.com/doi/pdf/10.2903/sp.efsa.2019.EN-1489). Results for one endpoint, one treatment duration and one blood sampling time point were analysed at a time, with chemical as the covariate. Note that for these analyses blood collection times days 15 and 16 were pooled, and the same is true for blood collection days 29 and 30. This so-called BMD covariate approach takes advantage of the software’s ability to analyse multiple dose responses at the same time using conserved shape parameters. This has been shown to improve the precision of BMD analyses (21). The critical effect size was set to 1.0, a small effect that in the authors’ expert judgment is nonetheless above the noise inherent to the assay. Model averaging was used, and the number of bootstraps was set to 50. In this manner, 90% BMD confidence intervals (CIs) were made for each study design, endpoint and time point combination. As described by Wills et al., BMDL (lower) and BMDU (upper) CIs provide an indication of potency and data quality. PS, Chlor, Thio and Mel’s relative potencies to induce Pig-a mutation, as ranked by BMD analyses, were compared with their carcinogenic potency. This was accomplished by graphing log(10) transformed TD50 values against Pig-a BMD CIs for each Pig-a study design, endpoint and time point combination. Note that Microsoft Excel files every study animal’s mutant RET, mutant RBC and RET frequencies will be uploaded and made available at the Pig-a In Vivo Gene Mutation Assay Database: www.pharmacy.umaryland.edu/centres/cersi-files. Results and Discussion Aristolochic acid I A dose-dependent reduction was observed in the body weight gain data following treatment with AA in the 3- and 28-day studies. Mean changes to weight gain over the exposure periods were 6.6, 1.1, −2.7 and −5.1 g for successively higher dose levels in the case of the 3-day study, and 66.4, 76.9, 45.6 and 35.9 g in the case of the 28-day study. Mutant cell frequencies for the 3-day study are shown in the top two panels of Figure 1. Whereas only the top AA dose group exhibited statistically significant increases in mutant RET and mutant RBC frequencies at the first time point (day 15), both the mid and top dose groups achieved statistical significance at day 30. Fig. 1. Open in new tabDownload slide Mutant RET and mutant erythrocyte (mutant RBC) frequencies are shown for individual rats exposed to various dose levels of AA for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Fig. 1. Open in new tabDownload slide Mutant RET and mutant erythrocyte (mutant RBC) frequencies are shown for individual rats exposed to various dose levels of AA for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Results from the 28-day AA study are shown in Figure 1, bottom panel. In this case, mutant RET and mutant RBC frequencies were significantly elevated in day 29 blood samples at each of the dose levels tested, including the lowest—2.75 mg/kg/day. This is 7.3-fold less than the lowest observed effect level in the short-term study. 1,3-Propane sultone While the highest dose levels of PS were well tolerated in the 3- and 28-day studies, dose-dependent reductions to weight gain were observed. For instance, in the case of the 28-day study, mean weight gain over the exposure period were 124, 92.2, 39.0 and 41.3 g with increasing dose levels. Mutant cell frequencies for the 3-day study are shown in the top two panels of Figure 2. Only the top dose group (60 mg/kg/day) exhibited statistically significant increases in mutant RET and mutant RBC frequencies at the first time point (day 16). On the other hand, there was no evidence of PS-induced mutagenic effects on day 30. Fig. 2. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of PS for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Fig. 2. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of PS for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Results from the 28-day PS study are shown in the bottom panel of Figure 2. In this case, mutant RET and mutant RBC frequencies showed dose-related increases on day 29 and pair-wise testing demonstrated statistical significance at each of the dose levels tested, including the lowest—12.5 mg/kg/day. This is 4.8-fold less than the lowest observed effect level in the short-term study. Chlorambucil Chlor was well tolerated in the 3- and 28-day studies. Treatment-related differences in mean weight gain over the exposure periods were 9.4, 7.9, −0.7 and −6.6 g with increasing dose levels in the case of the 3-day study, and 83.4, 76.1, 63.1 and 20 g in the case of the 28-day study. Mutant cell frequencies for the 3-day study are shown in the top two panels of Figure 3. At both time points, Chlor caused dose-related increases in mutant RET and mutant RBC frequencies. Pair-wise testing showed that the mid and high dose groups achieved statistical significance. Fig. 3. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of Chlor for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Fig. 3. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of Chlor for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Results from the 28-day Chlor study are shown in Figure 3, bottom panel. Mutant RET and mutant RBC frequencies were significantly elevated in day 29 blood samples at each of the dose levels tested, including the lowest—1.5 mg/kg/day. This is 4-fold less than the lowest observed effect level in the short-term study. Thiotepa Treatment-related differences in mean weight gain over the exposure periods were 9.8, 0.5, 1.7 and −4.2 g with increasing dose levels in the case of the 3-day study, and 104.5, 76.7, 66.3 and 50.8 g in the case of the 28-day study. Note that the intended high dose group in the 3-day study (15 mg/kg/day) was discontinued on day 10 because the animals continued to lose weight (−49.6 g on average). Mutant cell frequencies for the 3-day study are shown in the top two panels of Figure 4. At the first time point (day 15), only the 3.75 mg/kg/day group resulted in a statistically significant effect, and this was only evident for mutant RET. Conversely, on day 29, only the mutant RBC cohort exhibited a significant increase. Fig. 4. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of Thio for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Fig. 4. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of Thio for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Results from the 28-day Thio study are shown in the bottom panel of Figure 4. Mutant RET and mutant RBC frequencies were significantly elevated in day 29 blood samples at each of the dose levels tested, including the lowest—2 mg/kg/day. Melphalan Acute treatment of Mel for 3 consecutive days led to toxicity as demonstrated by the dose dependant reduced weight loss exhibited by the treated animals. Treatment-related differences in mean weight gain over the exposure periods were 13.2, 10.0, 8.8 and 7.5 g with increasing dose levels in the case of the 3-day study, and 124, 115, 109.6 and 69.5 g in the case of the 28-day study. Mutant cell frequencies for the 3-day study are shown in the top two panels of Figure 5. At the first time point (day 15), every Mel-exposed group showed elevated mutant RET frequencies, whereas this was limited to the high dose group in the case of mutant RBC. On the other hand, at day 30, each of the Mel-exposed groups exhibited statistically significant increases in mutant RBC, whereas no significant effects were observed for mutant RET. Fig. 5. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of Mel for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Fig. 5. Open in new tabDownload slide Mutant RET and mutant RBC frequencies are shown for individual rats exposed to various dose levels of Mel for 3 days (top two panels) and for 28 days (bottom panel). Note that mutant cell frequencies have been log(10) transformed, each dot represents an individual rat, and horizontal lines are mean values. Dunnett’s test results are shown, where statistically significant increases relative to the concurrent vehicle control group are indicated by dose levels written in italicised text and asterisks that appear next to significant dose level(s). Results from the 28-day Mel study are shown in the bottom panel of Figure 5. Mutant RET and mutant RBC frequencies were significantly elevated in day 30 blood samples at each of the dose levels tested, including the lowest—0.1875 mg/kg/day. This is 4-fold less than the lowest observed effect level in the short-term study. BMD analyses The x-axis of Figure 6 displays BMD 90% CIs for PS, Chlor, Thio and Mel. To assess whether this rank order is related to carcinogenic potency, the CIs have been graphed against rat TD50 point estimates on the y-axis. When viewed across all six plots, it is apparent that the chemicals’ Pig-a assay responses generally correspond to TD50 ranking. That being said, the plots associated with the 28-day study stand out in several ways. First, they show lower BMD CIs relative to the shorter-term treatment results, corroborating the results described above whereby lower dose levels (per day) attained statistical significance with pair-wise testing in the case of protracting dosing. Second, the CI is generally tighter, implying higher data quality. Third, Pig-a mutation potency generated by the lengthier repeat-dose studies more consistently exhibited the same rank order as tumorigenic potency. Fig. 6. Open in new tabDownload slide Mutant RET and mutant RBC dose–response data from PS, Chlor, Thio and Mel studies were analysed using the BMD covariate approach (critical effect size 1.0). Model averaging was used (50 bootstraps), with chemical as the covariate. The x-axis horizontal lines represent 90% BMDL–BMDU intervals. Intervals represented as dashed lines go off scale, and span several logs which is suggestive of unreliable estimates. Potency decreases from the left to the right side of the x-axis. The y-axis shows TD50 point estimates. Carcinogenic potency decreases from the bottom to the top of the y-axis. Fig. 6. Open in new tabDownload slide Mutant RET and mutant RBC dose–response data from PS, Chlor, Thio and Mel studies were analysed using the BMD covariate approach (critical effect size 1.0). Model averaging was used (50 bootstraps), with chemical as the covariate. The x-axis horizontal lines represent 90% BMDL–BMDU intervals. Intervals represented as dashed lines go off scale, and span several logs which is suggestive of unreliable estimates. Potency decreases from the left to the right side of the x-axis. The y-axis shows TD50 point estimates. Carcinogenic potency decreases from the bottom to the top of the y-axis. Conclusions Pig-a mutant cell frequencies were elevated in rats exposed to each of five diverse reference mutagens, and this occurred in both 3- and 28-day treatment schedules. Each of the 10 studies showed increased mutant cell frequencies in both mutant RET and mutant RBC populations. In the case of the 3-day Thio study, this required analyses at two time points, days 15 and 29. Furthermore, whereas PS showed the clearest induction of mutant cells at the first blood harvest time of the 3-day study, others required the later time point to exhibit the most pronounced effect(s) (e.g. Mel-induced mutant RBC). Together, these results support the OECD Detailed Review Paper’s recommendation to evaluate Pig-a mutant cell frequencies at both an early and later time point in the case of acute treatments. While the two study designs yielded qualitatively similar results for these five chemicals, several characteristics of the 28-day study results were found to be advantageous. In every instance, lower dose levels (per day) were effective at inducing Pig-a mutant cells, and greater magnitude effects were observed. Second, a single time point was sufficient to observe increases in both RET and RBC populations. Third, tighter BMD CIs were observed and they exhibited the same rank order as TD50 values. This implies that besides hazard identification, repeat-dose Pig-a studies may represent a valuable addition to risk assessments, for instance in the absence of 2-year bioassay results. Taken together, these characteristics combine to make protracted repeat-dose studies a preferred experimental design for regulatory safety assessments. Acknowledgements The authors would like to thank Prof. George Johnson, Swansea University, for advice about BMD analyses, and for sharing an Excel template that generates BMD confidence interval graphs. Funding This work was funded in part by grants from the National Institutes of Health/National Institute of Environmental Health Sciences (NIEHS; grant no. R44ES018017 and R44ES021973). The contents are solely the responsibility of the authors, and do not necessarily represent the official views of the NIEHS. Conflict of interest statement: SDD is an employee of Litron Laboratories. Litron holds patents covering flow cytometric methods for scoring GPI anchor-deficient erythrocytes and sells kits based on this technology (In Vivo MutaFlow®). References 1. Iida , Y. , Takeda , J., Miyata , T., Inoue , N., Nishimura , J., Kitani , T., Maeda , K. and Kinoshita , T. ( 1994 ) Characterization of genomic PIG-A gene: a gene for glycosylphosphatidylinositol-anchor biosynthesis and paroxysmal nocturnal hemoglobinuria . Blood , 83 , 3126 – 3131 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Kawagoe , K. , Takeda , J., Endo , Y. and Kinoshita , T. ( 1994 ) Molecular cloning of murine pig-a, a gene for GPI-anchor biosynthesis, and demonstration of interspecies conservation of its structure, function, and genetic locus . Genomics , 23 , 566 – 574 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Araten , D. J. , Nafa , K., Pakdeesuwan , K. and Luzzatto , L. ( 1999 ) Clonal populations of hematopoietic cells with paroxysmal nocturnal hemoglobinuria genotype and phenotype are present in normal individuals . Proc. Natl. Acad. Sci. U. S. A. , 96 , 5209 – 5214 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Chen , R. , Eshleman , J. R., Brodsky , R. A. and Medof , M. E. ( 2001 ) Glycophosphatidylinositol-anchored protein deficiency as a marker of mutator phenotypes in cancer . Cancer Res. , 61 , 654 – 658 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 5. Miura , D. , Dobrovolsky , V. N., Kasahara , Y., Katsuura , Y. and Heflich , R. H. ( 2008 ) Development of an in vivo gene mutation assay using the endogenous Pig-A gene: I. Flow cytometric detection of CD59-negative peripheral red blood cells and CD48-negative spleen T-cells from the rat . Environ. Mol. Mutagen. , 49 , 614 – 621 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Rondelli , T. , Berardi , M., Peruzzi , B., Boni , L., Caporale , R., Dolara , P., Notaro , R. and Luzzatto , L. ( 2013 ) The frequency of granulocytes with spontaneous somatic mutations: a wide distribution in a normal human population . PLoS One , 8 , e54046 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Revollo , J. , Pearce , M. G., Petibone , D. M., Mittelstaedt , R. A. and Dobrovolsky , V. N. ( 2015 ) Confirmation of Pig-a mutation in flow cytometry-identified CD48-deficient T-lymphocytes from F344 rats . Mutagenesis , 30 , 315 – 324 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Bemis , J. C. , Avlasevich , S. L., Labash , C., McKinzie , P., Revollo , J., Dobrovolsky , V. N. and Dertinger , S. D. ( 2018 ) Glycosylphosphatidylinositol (GPI) anchored protein deficiency serves as a reliable reporter of Pig-a gene mutation: support from an in vitro assay based on L5178Y/Tk+/− cells and the CD90.2 antigen . Environ. Mol. Mutagen. , 59 , 18 – 29 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Gollapudi , B. B. , Lynch , A. M., Heflich , R. H., et al. ( 2015 ) The in vivo pig-a assay: a report of the International Workshop On Genotoxicity Testing (IWGT) Workgroup . Mutat. Res. Genet. Toxicol. Environ. Mutagen. , 783 , 23 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Dobrovolsky , V. N. , Miura , D., Heflich , R. H. and Dertinger , S. D. ( 2010 ) The in vivo Pig-a gene mutation assay, a potential tool for regulatory safety assessment . Environ. Mol. Mutagen. , 51 , 825 – 835 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Schuler , M. , Gollapudi , B. B., Thybaud , V. and Kim , J. H. ( 2011 ) Need and potential value of the Pig-a in vivo mutation assay—a HESI perspective . Environ. Mol. Mutagen. , 52 , 685 – 689 . Google Scholar Crossref Search ADS PubMed WorldCat 12. ICH Guideline M7(R1) on Assessment and Control of DNA Reactive (Mutagenic) Impurities in Pharmaceuticals to Limit Potential Carcinogenic Risk. https://www.ema.europa.eu/en/ich-m7-assessment-control-dna-reactive-mutagenic-impurities-pharmaceuticals-limit-potential (accessed April 7, 2020 ). 13. OECD ( 2019 ) The In Vivo Erythrocyte Pig-a Gene Mutation Assay. Part 1: Detailed Review Paper and Performance Assessment. http://www.oecd.org/chemicalsafety/testing/pig-a-gene-mutation-assay-detailed-review-paper.pdf (accessed April 7, 2020 ). 14. Dertinger , S. D. , Phonethepswath , S., Avlasevich , S. L., et al. ( 2012 ) Efficient monitoring of in vivo pig-a gene mutation and chromosomal damage: summary of 7 published studies and results from 11 new reference compounds . Toxicol. Sci. , 130 , 328 – 348 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Bhalli , J. A. , Ding , W., Shaddock , J. G., Pearce , M. G., Dobrovolsky , V. N. and Heflich , R. H. ( 2013 ) Evaluating the weak in vivo micronucleus response of a genotoxic carcinogen, aristolochic acids . Mutat. Res. , 753 , 82 – 92 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Dertinger , S. D. , Bryce , S. M., Phonethepswath , S. and Avlasevich , S. L. ( 2011 ) When pigs fly: immunomagnetic separation facilitates rapid determination of Pig-a mutant frequency by flow cytometric analysis . Mutat. Res. , 721 , 163 – 170 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Dertinger , S. D. , Phonethepswath , S., Weller , P., et al. ( 2011 ) Interlaboratory Pig-a gene mutation assay trial: studies of 1,3-propane sultone with immunomagnetic enrichment of mutant erythrocytes . Environ. Mol. Mutagen. , 52 , 748 – 755 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Phonethepswath , S. , Franklin , D., Torous , D. K., et al. ( 2010 ) Pig-a mutation: kinetics in rat erythrocytes following exposure to five prototypical mutagens . Toxicol. Sci. , 114 , 59 – 70 . Google Scholar Crossref Search ADS PubMed WorldCat 19. MacGregor , J. T. , Frötschl , R., White , P. A., et al. ( 2015 ) IWGT report on quantitative approaches to genotoxicity risk assessment I. Methods and metrics for defining exposure-response relationships and points of departure (PoDs) . Mutat. Res. Genet. Toxicol. Environ. Mutagen. , 783 , 55 – 65 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Wills , J. W. , Long , A. S., Johnson , G. E., Bemis , J. C., Dertinger , S. D., Slob , W. and White , P. A. ( 2016 ) Empirical analysis of BMD metrics in genetic toxicology part II: in vivo potency comparisons to promote reductions in the use of experimental animals for genetic toxicity assessment . Mutagenesis , 31 , 265 – 275 . Google Scholar Crossref Search ADS PubMed WorldCat 21. Slob , W. and Setzer , R. W. ( 2014 ) Shape and steepness of toxicological dose-response relationships of continuous endpoints . Crit. Rev. Toxicol. , 44 , 270 – 297 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of the UK Environmental Mutagen Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Pig-a gene mutation assay study design: critical assessment of 3- versus 28-day repeat-dose treatment schedules
JF - Mutagenesis
DO - 10.1093/mutage/geaa014
DA - 2020-09-12
UR - https://www.deepdyve.com/lp/oxford-university-press/pig-a-gene-mutation-assay-study-design-critical-assessment-of-3-versus-iIhlFWsc1T
SP - 349
EP - 358
VL - 35
IS - 4
DP - DeepDyve
ER -