Background mutant frequency in healthy populations
Somatic PIG-A mutations can occur at low frequency through spontaneous mutational processes during normal haematopoiesis without inducing PNH-like clinical symptoms. This is perhaps because these mutated, GPI-negative blood cells occur at a lower frequency than is required to activate mass complement mediated lysis (Araten, Nafa, Pakdeesuwan, & Luzzatto, 1999; Hu et al., 2005). These PIG-Amutated erythrocytes are present in healthy individuals at low numbers (Cao et al., 2016; Dertinger, Avlasevich, Bemis, Chen, & MacGregor, 2015; Dobrovolsky, Elespuru, Bigger, Robison, & Heflich, 2011; Lawrence, Haboubi, Williams, Doak, & Jenkins, 2020), allowing for easy observation of any increase in mutant cell levels and association with individual mutagenic exposures. At low levels, such mutant mammalian cells are growth neutral, viable and are easily detectable using flow cytometry (David et al., 2018) (Figure 1).
PIG-A mutations can be measured in all blood cell types and have been reported for erythrocytes, reticulocytes, granulocytes, and lymphocytes, with the majority of the reports using human erythrocytes. A number of publications have measured PIG-A mutant erythrocyte levels in healthy populations with values ranging from 2.9 - 5.56 x 10-6 mutants per million erythrocytes (Figure 2). These 7 studies all measured PIG-A status through flow cytometric analysis, using antibodies that target the GPI-linked protein CD59 (Cao, Wang, Xi, et al., 2020; Cao et al., 2021; Cao et al., 2016; Dobrovolsky et al., 2011), or CD55 and CD59 in combination (Dertinger et al., 2015; Haboubi et al., 2019; Lawrence et al., 2020). Using both CD55 and CD59 in combination may reduce the number of falsely classified mutant cells by eliminating any effect of reduced CD59 expression (Peruzzi, Araten, Notaro, & Luzzatto, 2010). However, as seen in Figure 2, there is no significant difference in mutant cell levels between studies that applied either CD55 alone (blue bars) or in combination with CD59 (red bars) (p=0.057). The green bar on the graph represents the weighted average from these 7 publications which is 4.58 \(\pm\) 4.2 x 10-6 mutants per million erythrocytes. The consistency across studies is remarkable, considering that mutant frequency can be easily inflated by any slight technological variation. The number of mutant cells is subject to technical variation and can be artificially altered in a number of ways. This includes antibody batch to batch variation and supplier differences, sample preparation resulting in shearing of anchored proteins, flow cytometer choice, flow rate, laser strength and gating procedure. Test reproducibility is essential for successful biomarker implementation. It is particularly important if such a test were to be used in the assessment of human genotoxic exposures where a subtle change in mutant frequency, may be indicative of a positive genotoxic response.