Erythrocyte-based Pig-a gene mutation assay: demonstration of cross-species potential

Abstract

Glycosylphosphatidylinositol (GPI) anchors attach specific proteins to the cell surface of hematopoietic cells. Of the genes required to form GPI anchors, only Pig-a is located on the X-chromosome. Prior work with rats suggests that the GPI anchor deficient phenotype is a reliable indicator of Pig-a mutation [Bryce et al., Environ. Mol. Mutagen., 49 (2008) 256-264]. The current report extends this line of investigation by describing simplified blood handling procedures, and by testing the assay principle in a second species, Mus musculus. With this method, erythrocytes are isolated, incubated with anti-CD24-PE, and stained with SYTO 13. Flow cytometric analyses quantify GPI anchor-deficient erythrocytes and reticulocytes. After reconstruction experiments with mutant-mimicking cells demonstrated that the analytical performance of the method is high, CD-1 mice were treated on three occasions with 7,12-dimethyl-1,2-benz[a]anthracene (DMBA, 75 mg/kg/day) or ethyl-N-nitrosourea (ENU, 40 mg/kg/day). Two weeks after the final treatment, DMBA-treated mice were found to exhibit markedly elevated frequencies of GPI anchor deficient erythrocytes and reticulocytes. For the ENU experiment, blood specimens were collected at weekly intervals over a 5-week period. Whereas the frequencies of mutant reticulocytes were significantly elevated 1 week after the last administration, the erythrocyte population was unchanged until the second week. Thereafter, both populations exhibited persistently elevated frequencies for the duration of the experiment (mean frequency at termination=310x10(-6) and 523x10(-6) for erythrocyte and reticulocyte populations, respectively). These data provide evidence that Pig-a mutation does not convey an appreciable positive or negative cell survival advantage to affected erythroid progenitors, although they do suggest that affected erythrocytes have a reduced lifespan in circulation. Collectively, accumulated data support the hypothesis that flow cytometric enumeration of GPI anchor deficient erythrocytes and/or reticulocytes represents an effective in vivo mutation assay that is applicable across species of toxicological interest.

Figure 1

Three bivariate graphs illustrate the gating logic used for the mutant scoring application described herein. Only RBCs are plotted in the bivariate graph to the right. Other events are excluded based on their failure to exhibit light scatter characteristics of cells (left panel), or else their high SYTO 13 fluorescence intensity (middle panel, “Leukocytes”). Thus, while the Lympholyte reagent physically removed the bulk of platelets and leukocytes, this gating strategy further ensures that mutant RBC frequencies generated by the right plot are not affected by these spurious events.

Figure 2. Gated events (RBCs) are plotted on SYTO 13 versus anti-CD24-PE bivariate plots

Left: Instrument calibration standard; mutant-mimicking cells have been spiked into blood that was processed according to the standard protocol. This specimen provides enough events with a full range of fluorescence intensities to optimize PMT voltages and compensation settings. This calibration standard also represents a means for rationally and consistently setting the position of the vertical line that defines mutant versus non-mutant cells. Middle: Vehicle control mouse blood; note the very low incidence of cells that appear in the upper left and lower left quadrants. These anti-CD24-negative events are presumably Pig-a mutant reticulocytes and RNA-negative RBCs, respectively. Right: Blood from a mutagen (DMBA) treated mouse; note the elevated numbers of events that appear in the upper left and lower left quadrants, i.e., presumably Pig-a mutant reticulocytes and RNA-negative RBCs, respectively.

Figure 3

Bivariate graphs illustrate analyses based on Forward Scatter thresholding (left panel), and SYTO 13 fluorescence thresholding (right panel). Forward Scatter thresholding is capable of efficiently evaluating 10⁶ or more RBCs for the mutant phenotype, whereas SYTO 13 thresholding facilitates interrogation of 10⁶ RETs.

Figure 4

Expected number of CD24-negative RBCs per 10⁶ total are plotted against observed frequencies measured by flow cytometry. The high correlation coefficient, coupled with the system’s ability to differentiate control (unspiked) specimens from those with low numbers of mutant cells spiked in (< 100) suggest that analytical performance is sufficient for an in vivo mutagenesis assay.

Figure 5

The number of presumed mutant cells are graphed for each of three vehicle (V) and DMBA (D75) treated mice. These two week post-dosing blood specimens exhibit higher mutant frequencies for the RET population (black bars) compared to the total RBC pool (white).

Figure 6

Average mutant frequencies are graphed versus time post-ENU treatment. Whereas the mutant RET frequency is elevated as soon as week 1, the response in the total RBC cohort is delayed. Note that there were actually no “time zero” bloods, rather this data point was generated by pooling vehicle control specimens that were obtained throughout the experimental time-frame (one per time point, for a total of 5 observations).