LAUSR

April 23, 2019 2077 Huaizhu Wu, MD, MS Christie M. Ballantyne, MD To the Editor: We read with great interest the article by Raghavan et al,1 who reported that thrombin plays a critical role in atherogenesis via the Par1-Gα12 signaling pathway that targets Pyk2-Gab1–protein kinase C theta (PKCθ)-ATF2–dependent CD36 expression and mediates foam cell formation in monocytes/macrophages. This study is important in that it revealed a novel role of thrombin in atherogenesis and provided the potential mechanisms by which thrombin-Par1-Gα12 signaling accelerates foam cell formation, a key step for atherogenesis. We have some concerns with some of the observations as reported, mainly related to monocytes in the mouse models. First, the authors reported that deletion of PKCθ, which blocked the thrombin-mediated ATF2 phosphorylation and CD36 expression, reduced atherosclerosis and favored monocyte differentiation to Ly-6Clow monocytes in the circulation of apolipoprotein E–deficient (apoE–/–) mice on a Western diet.1 However, the data as reported raised serious questions about the gating strategy for mouse monocytes. Although the authors stated that they followed the protocol by Swirski et al2 and defined monocytes as CD11b+CD90– B220–CD49b–NK1.1–Ly6G–, their data on the percentage of CD90–B220–CD49b– NK1.1–Ly6G– cells accounting for ≈85% to 95% of total white blood cells (Figure 7A, meaning that the combination of T cells, B cells, NK cells, and neutrophils accounted for only 5%–15% of total white blood cells) are not consistent with the data shown by Swirski et al2 (≈25%–30%). Furthermore, several research laboratories showed that the majority of Ly-6Clow monocytes in apoE–/– mice on a Western diet were CD11c+,2,3 whereas the authors showed that ≈6.5% of Ly-6Clow monocytes were positive for CD11c, F4/80, or I-Ab. In addition, the example in Figure 7A shows a ratio of ≈46% Ly-6Clow to ≈54% of Ly-6Chigh monocytes in apoE–/– mice. However, quantification data in Figure 7C show ≈1.2×105 Ly-6Clow and ≈3.7×105 Ly-6Chigh monocytes, ≈3-fold difference, which is not consistent with the example. Second, the authors reported that, in comparison with Ly-6Clow monocytes, Ly6Chigh monocytes expressed higher levels of CD36 with enhanced capacity to form foam cells (Figure 7E), thereby concluding that PKCθ contributes to atherogenesis via activating ATF2-mediated CD36 expression and foam cell formation of Ly-6Chigh cells.1 It has been well established and demonstrated by several research laboratories that Ly-6Chigh monocytes express no or low levels of CD36, whereas Ly-6Clow monocytes express high levels of CD36.3–5 Previous data including ours indicated that in apoE–/– or LDLR–/– mice on a Western diet, monocytes in the circulation take up lipid, and that Ly-6Clow monocytes have a higher capacity of taking up lipid3,5 and becoming foamy monocytes.3 In summary, this work by Raghavan et al1 provides important information on the atherogenic role of thrombin. However, some of the observations, specifically related to the identification of mouse monocytes and subsets, higher expression of © 2019 American Heart Association, Inc. LETTER TO THE EDITOR