Exposure to ambient fine particulate matter (PM_2.5) is associated with increased risk for cardiovascular events and accelerated progression of cardiovascular disease.¹ Though substantive advancements have identified plausible mechanistic pathways, critical questions remain unanswered. Previously, we found that exposure to concentrated ambient PM_2.5 (CAP) for 30 days increased red blood cell distribution width,² which predicts multiple manifestations of cardiovascular disease, including heart failure.³ To probe the impact of PM_2.5 exposure on erythrocyte homeostasis, we performed daily (6 h/day) whole-body exposures using a versatile aerosol concentration enrichment system on 12–16-week-old C57BL6/J male mice for up to 30 consecutive days, as described previously.² This system concentrates ambient PM_2.5 without changing the chemical composition or physical properties of the particles achieving a daily CAP concentration (64 µg/m³) within the range relevant to human ambient PM_2.5 exposures (5–100 µg/m³). A single 6-h CAP exposure significantly increased the expression of stress-associated markers on circulating erythrocytes including heightened production of reactive oxygen species and externalized phosphatidyl serine residues (Figure [Ai]). These changes were largely maintained in mice that inhaled filtered air or CAP daily for up to 30 days (Figure [Aii]).

Figure. Concentrated ambient PM_2.5 (CAP) exposure alters erythrocyte homeostasis. A, Expression of stress markers on circulating red blood cells (RBCs; Ter119⁺) was determined by flow cytometry. Reactive oxygen species (CellROX) and phosphatidyl serine residue externalization (annexin V) were measured following a single 6-h CAP exposure (i, n=4–5 mice/group) or with cluster of differentiation 47 (CD47) during a 30-d exposure (ii, n=4–14 mice/group; P value is shown as determined by 2-way ANOVA). iii, Erythrocyte turnover was determined by administering biotin (intravenous) before exposure and then serially measuring the abundance of Biotin⁺RBC via tail bleed. P value is shown as determined by 2-way ANOVA. n=5 mice/group. B, Erythrocytes isolated from mice inhaling filtered air or CAP for 30 d were labeled (carboxyfluorescein diacetate succinimidyl ester [CFSE]) and administered (intravenous) to naive mice. After 1 h, uptake by red pulp macrophages (RPMs; defined as CD11b, F4/80, and vascular cell adhesion molecule 1 [VCAM-1] positive) was determined. C, RNA-sequencing of spleens from mice after 30-d air or CAP exposure. i, Gene ontology biological process (GO:BP) analysis. ii, Expression of genes in highlighted GO:BP pathways shown as fold change vs air. iii, Number of upregulated differentially expressed genes with the highest expression in each splenic cell type as reported by Gene Skyline (www.immgen.org). n=4 mice/group. D, GO:BP analysis from RNA-sequencing of isolated F4/80⁺ splenocytes after 30-d air or CAP exposure. n=5 mice/group. E, Quantification of RPM abundance and expression of SIRPα (signal regulatory protein α). n=5 mice/group. F, Erythrocytes isolated from naïve mice were labeled (CFSE) and administered (intravenous) to mice exposed to air or CAP for 30 d. Spleens were harvested after 1 h, and uptake by RPM was determined by flow cytometry. n=4 to 5 mice/group. Data are mean±SEM. P values are shown as determined by a 2-tailed unpaired Student t test, unless indicated otherwise.

Exposure to CAP for 14 and 30 days also decreased the surface expression of cluster of differentiation 47 (CD47; Figure [Aii]), a prominent marker of self and don’t eat me signal that is reduced by erythrocyte age and stress. To evaluate the functional impact of these CAP-induced changes, we assessed erythrocyte turnover in vivo using biotin labeling and found a significant decrease in CAP-exposed mice compared with air-inhaling mice (Figure [Aiii]). The increased stress markers and decreased turnover of erythrocytes of CAP-exposed mice indicate that PM_2.5 exposure decreases the capacity for recognition and removal of erythrocytes from circulation. A task achieved primarily by red pulp macrophages (RPMs), a specialized splenic macrophage population.⁴

To determine whether CAP-induced erythrocyte modifications trigger their clearance by RPM, we performed an ex vivo disposal assay (Figure [B]). Isolated erythrocytes (30-day CAP) were fluorescently labeled and administered to naïve mice. One hour later, we noted a significant increase in the uptake of erythrocytes obtained from CAP-exposed mice in RPM compared with erythrocytes isolated from air-inhaling mice. Although erythrocytes from CAP-exposed mice are primed for uptake by splenic RPM, it is unclear whether CAP exposure impacts the spleen. To test for PM_2.5-induced changes in the spleen, we performed global transcriptomics using spleens from mice exposed for 30 days to CAP or air and found a total of 561 differentially expressed genes (DEGs) with 481 upregulated and 80 downregulated. The gene ontology biological process analysis (Figure [Ci]) showed that CAP exposure significantly impacted several pathways related to erythrocyte turnover, such as iron handling, erythrocyte metabolism, and the metabolism of heme and porphyrin-related groups. Pathways related to myeloid cell homeostasis were also affected, suggesting a macrophage-mediated response.

Next, we compiled the constituent genes of the affected pathways and found that 49 of the 53 DEG were upregulated by CAP and are shown in Figure Cii. These include heme oxygenase 1 (Hmox1), the iron transporter, ferroportin (Slc40a1), and biliverdin reductase B (Blvrb), which are critical mediators of macrophage erythrophagocytosis. Given that most of these pathway DEG were upregulated, we cross-referenced all the upregulated DEG in our analysis (481 of 561 total DEG) with the immunologic Genome Project’s Gene Skyline database (www.immgen.org) to screen for a cell-specific fingerprint (Figure [Ciii]). Strikingly, this analysis revealed that 115 of 481 upregulated genes had the highest reported expression in RPM compared with other splenic immune cell types. Because CAP exposure appeared to target RPM, we isolated F4/80⁺-splenocytes following a 30-day CAP exposure and performed transcriptomics (Figure [D]). The gene ontology biological process analysis of the 330 total DEGs (262 upregulated and 68 downregulated) revealed changes in multiple pathways related to erythrocyte processing including several associated with iron handling, which largely recapitulated results of the whole spleen gene ontology biological process analysis.

With multiple indications that CAP exposure targets RPM, we next sought to examine this cell population (Figure [E]). Surprisingly, 30-day CAP significantly increased the abundance of RPM yet decreased the expression of SIRPα (signal regulatory protein α). The binding of SIRPα to CD47 is critical for the progression of phagocytosis and efferocytosis.⁵ Decreased expression of SIRPα on RPM suggests that PM_2.5 likely induced a phenotypic alteration that impairs their function despite increasing abundance. To test the direct erythrophagocytotic capacity of RPM of CAP-exposed mice, we again performed an ex vivo erythrocyte disposal assay (Figure [F]) and found that CAP decreased the uptake of erythrocytes by RPM, further supporting the functional impairment of these cells by CAP.

In summary, this study shows that PM_2.5 inhalation lowers rates of erythrocyte clearance despite an increase in the surface markers of stress, which appears to be a direct result of a functional decrement in splenic RPM. Even modest perturbations in erythrocyte turnover can have dramatic impacts on the quality and cumulative function of the cell population, inducing a right shift in the age distribution, which is reflected by an increase in red blood cell distribution width. Importantly, in human studies, it has been determined that for every 1-unit increase in red blood cell distribution width, there is an increased risk of all-cause mortality and major adverse cardiac events.³ Further studies focused on erythrocytes as novel targets and intermediaries of PM_2.5-induced pathology may reveal an overlooked, yet fundamental, contributor to cardiovascular disease exacerbation.

All animal procedures were approved by the University of Louisville Institutional Animal Care and Use Committee (#21953). The authors thank A. Ribble and the members of the Animal and Inhalation Core of the Center for Cardiometabolic Sciences for their technical assistance.

The methods, data, and materials that support the findings of this study are available from the corresponding author upon reasonable request. Results of the RNA-sequencing analyses have been deposited in the Gene Expression Omnibus database (accession number: GSE262906).

This work was supported by the National Institutes of Health grants ES034389 (to B.E. Sansbury), GM127607 (to B.E. Sansbury and D.J. Conklin), ES030283 (to D.J. Conklin), ES023716 (to D.J. Conklin), and GM103436 (E.C. Rouchka) and the Jewish Heritage Fund for Excellence (to B.E. Sansbury and D.J. Conklin).

Nonstandard Abbreviations and Acronyms

CAP	concentrated ambient PM2.5
DEG	differentially expressed gene
PM2.5	fine particulate matter
RPM	red pulp macrophage
SIRPα	signal regulatory protein α

concentrated ambient PM_2.5

differentially expressed gene

fine particulate matter

red pulp macrophage

signal regulatory protein α

Disclosures None.

For Sources of Funding and Disclosures, see page 1226.

中文翻译：

---

接触细颗粒物空气污染会扰乱红细胞周转

暴露于环境细颗粒物 (PM _2.5 ) 会增加心血管事件的风险并加速心血管疾病的进展。¹尽管实质性进展已经确定了看似合理的机制途径，但关键问题仍未得到解答。此前，我们发现暴露于浓环境 PM _2.5 (CAP) 30 天会增加红细胞分布宽度，²这预示着心血管疾病的多种表现，包括心力衰竭。^{3为了探究 PM} _2.5暴露对红细胞稳态的影响，我们使用多功能气溶胶浓缩系统对 12-16 周龄的 C57BL6/J 雄性小鼠进行每日（6 小时/天）全身暴露，最多 30 只连续几天，如前所述。²该系统在不改变颗粒化学成分或物理性质的情况下浓缩环境 PM _{2.5 ，实现每日 CAP 浓度 (64 µg/m} ^{3 )，处于与人类环境 PM} _2.5暴露相关的范围内(5–100 µg/m ³ )。单次 6 小时 CAP 暴露显着增加了循环红细胞上应激相关标记物的表达，包括活性氧和外化磷脂酰丝氨酸残基产量的增加（图 [Ai]）。这些变化在每天吸入过滤空气或 CAP 长达 30 天的小鼠中基本得以维持（图 [Aii]）。

数字。 浓环境 PM _2.5 (CAP) 暴露会改变红细胞稳态。 A ，通过流式细胞术测定循环红细胞（RBC；Ter119 ⁺ ）上应激标记物的表达。活性氧 (CellROX) 和磷脂酰丝氨酸残基外化 (膜联蛋白 V) 在单次 6 小时 CAP 暴露（i，n=4-5 只小鼠/组）后或在 30 小时内使用分化簇 47 (CD47) 进行测量。 d 暴露（ii，n=4-14 只小鼠/组；P值显示为通过双向方差分析确定）。iii，通过在暴露前施用生物素（静脉内）然后通过尾部放血连续测量生物素^{+ RBC的丰度来确定红细胞周转率。} P值显示为通过 2 路方差分析确定。 n=5只小鼠/组。B，从吸入过滤空气或 CAP 30 天的小鼠中分离出的红细胞被标记（羧基荧光素二乙酸琥珀酰亚胺酯 [CFSE]）并给予（静脉内）初始小鼠。 1 小时后，测定红髓巨噬细胞（RPM；定义为 CD11b、F4/80 和血管细胞粘附分子 1 [VCAM-1] 阳性）的摄取。C，暴露于空气或 CAP 30 天后小鼠脾脏的 RNA 测序。i，基因本体生物过程（GO：BP）分析。ii，突出显示的 GO:BP 途径中的基因表达，显示为相对于空气的倍数变化。iii，根据 Gene Skyline (www.immgen.org) 报告，每种脾细胞类型中表达量最高的上调差异表达基因的数量。 n=4只小鼠/组。D 、GO：对空气或 CAP 暴露 30 天后分离的 F4/80 ⁺脾细胞进行 RNA 测序进行 BP 分析。 n=5只小鼠/组。E，RPM 丰度和 SIRPα（信号调节蛋白 α）表达的定量。 n=5只小鼠/组。F，从初始小鼠中分离的红细胞被标记（CFSE）并给予（静脉内）暴露于空气或CAP 30天的小鼠。 1小时后收获脾脏，并通过流式细胞术测定RPM的摄取。 n=4至5只小鼠/组。数据为平均值±SEM。除非另有说明，P值显示为通过 2 尾未配对学生t检验确定。

暴露于 CAP 14 和 30 天还降低了分化簇 47（CD47；图 [Aii]）的表面表达，分化簇 47 是自我的显着标记，并且“不要吃我”信号会因红细胞年龄和压力而减弱。为了评估这些 CAP 引起的变化的功能影响，我们使用生物素标记评估了体内红细胞周转率，发现与吸入空气的小鼠相比，暴露于 CAP 的小鼠红细胞周转率显着下降（图 [Aiii]）。暴露于 CAP 的小鼠的应激标记物增加和红细胞更新减少表明，PM _2.5暴露降低了识别和清除循环中红细胞的能力。该任务主要由红髓巨噬细胞（RPM）（一种特殊的脾巨噬细胞群）完成。⁴

为了确定 CAP 诱导的红细胞修饰是否会通过 RPM 触发其清除，我们进行了离体处置测定（图 [B]）。对分离的红细胞（30 天 CAP）进行荧光标记并给予初始小鼠。一小时后，我们注意到与从吸入空气的小鼠中分离的红细胞相比，在 RPM 中从暴露于 CAP 的小鼠获得的红细胞的摄取显着增加。尽管暴露于 CAP 的小鼠的红细胞已准备好被脾 RPM 摄取，但尚不清楚 CAP 暴露是否会影响脾脏。为了测试 PM _2.5诱导的脾脏变化，我们使用暴露于 CAP 或空气 30 天的小鼠脾脏进行了全局转录组学，发现总共 561 个差异表达基因 (DEG)，其中 481 个上调，80 个下调。基因本体生物过程分析（图[Ci]）表明，CAP暴露显着影响与红细胞周转相关的多个途径，例如铁处理、红细胞代谢以及血红素和卟啉相关群体的代谢。与骨髓细胞稳态相关的通路也受到影响，表明巨噬细胞介导的反应。

接下来，我们编译了受影响途径的组成基因，发现 53 个 DEG 中的 49 个被 CAP 上调，如图 Cii 所示。这些包括血红素加氧酶 1 ( Hmox1 )、铁转运蛋白、铁转运蛋白 ( Slc40a1 ) 和胆绿素还原酶 B ( Blvrb )，它们是巨噬细胞吞噬红细胞作用的关键介质。鉴于大多数这些途径 DEG 均上调，我们将分析中所有上调的 DEG（561 个总 DEG 中的 481 个）与免疫基因组计划的 Gene Skyline 数据库 (www.immgen.org) 交叉引用，以筛选细胞特异性的 DEG。指纹（图[Ciii]）。引人注目的是，该分析表明，与其他脾免疫细胞类型相比，481 个上调基因中有 115 个在 RPM 中具有最高的报告表达。由于 CAP 暴露似乎针对 RPM，因此我们在 30 天的 CAP 暴露后分离了 F4/80 ⁺ -脾细胞并进行转录组学（图 [D]）。对330个DEG（262个上调和68个下调）的基因本体生物过程分析揭示了与红细胞加工相关的多个途径的变化，包括与铁处理相关的多个途径，这在很大程度上概括了整个脾基因本体生物过程分析的结果。

多种迹象表明 CAP 暴露以 RPM 为目标，我们接下来试图检查该细胞群（图 [E]）。令人惊讶的是，30 天的 CAP 显着增加了 RPM 的丰度，但降低了 SIRPα（信号调节蛋白 α）的表达。 SIRPα 与 CD47 的结合对于吞噬作用和胞吞作用的进展至关重要。⁵ RPM 上 SIRPα 表达的减少表明，PM _2.5可能会诱导表型改变，尽管丰度增加，但会损害其功能。为了测试暴露于 CAP 的小鼠 RPM 的直接吞噬红细胞的能力，我们再次进行了离体红细胞处置测定（图 [F]），发现 CAP 减少了 RPM 对红细胞的摄取，进一步支持了这些细胞的功能损伤帽。

总之，这项研究表明，尽管表面应激标志物有所增加，但吸入 PM _2.5仍会降低红细胞清除率，这似乎是脾 RPM 功能下降的直接结果。即使红细胞周转的轻微扰动也会对细胞群的质量和累积功能产生巨大影响，导致年龄分布右移，这通过红细胞分布宽度的增加反映出来。重要的是，在人类研究中已经确定，红细胞分布宽度每增加 1 个单位，全因死亡率和主要不良心脏事件的风险就会增加。³进一步研究将红细胞作为 PM _2.5诱导病理学的新靶点和中介，可能会揭示心血管疾病恶化的一个被忽视但根本性的因素。

所有动物程序均经路易斯维尔大学动物护理和使用机构委员会 (#21953) 批准。作者感谢 A. Ribble 和心脏代谢科学中心动物和吸入核心成员的技术援助。

支持本研究结果的方法、数据和材料可根据合理要求从通讯作者处获得。 RNA测序分析的结果已存入Gene Expression Omnibus数据库（登录号：GSE262906）。

这项工作得到了美国国立卫生研究院授予 ES034389（给 BE Sansbury）、GM127607（给 BE Sansbury 和 DJ Conklin）、ES030283（给 DJ Conklin）、ES023716（给 DJ Conklin）和 GM103436（EC Rouchka）以及犹太遗产卓越基金（致 BE Sansbury 和 DJ Conklin）。

非标准缩写词和首字母缩略词

帽	环境 PM 2.5浓度较高
二甘醇	差异表达基因
下午2.5	细颗粒物
转速	红髓巨噬细胞
SIRPα	信号调节蛋白α

环境 PM _{2.5浓度较高}

差异表达基因

细颗粒物

红髓巨噬细胞

信号调节蛋白α

披露无。

有关资金来源和披露信息，请参阅第 1226 页。