Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Oct 8.
Published in final edited form as: Blood Red Cells Iron. 2026 Jul 8;2(3):100100. doi: 10.1016/j.brci.2026.100100

Characterization of erythroblastic island and tissue macrophages using a novel CD169-tdTomato-Cre mouse model and scRNA-seq

Donghao Liu 1, Huan Zhang 1, Qianqian Yang 1, Anqi Guo 2, Huizhi Zhao 1, Xinhua Guo 1, Karina Yazdanbakhsh 2, Xiuli An 1
PMCID: PMC13645301  NIHMSID: NIHMS2211284  PMID: 42846664

Abstract

Macrophages play diverse roles, including supporting erythropoiesis. CD169 is a specific marker for macrophages. However, recent studies showed that anti-CD169 staining detected nonmacrophages, raising the concern for using anti-CD169 antibody to correctly assign macrophages. Genetically engineered mouse models provide another means of monitoring gene expression. However, no mouse model is currently available to directly monitor CD169 expression. Erythropoiesis relies on macrophage-mediated iron recycling for iron supply. Although it is established that spleen (SP) macrophages play critical roles in erythrophagocytosis and iron recycling, whether and how other tissue macrophages are involved remain incompletely understood. To address these issues, we generated CD169-tdTomato-Cre mice. We also compared various tissue macrophages by single-cell RNA sequencing, flow cytometry, and cytospin analyses. We showed that CD169-tdTomato was detected and that CD169-Cre–mediated recombination occurred only in macrophages. Notably, bone marrow (BM) F4/80+CD169+ macrophages resemble erythroblastic island macrophages, as demonstrated by the expression of Epor along with other macrophage feature genes. A comparison of tissue macrophages revealed significant differences, including erythrophagocytosis by SP and BM macrophages, but not liver macrophages, at steady state; differential expression of the macrophage surface markers CD169, Vcam1, Cx3cr1, and Csf1r and the iron recycling genes Hmox1, Spic, and CD91; and selective expression of globin genes in lung macrophages. Our findings have uncovered previously unknown differences among tissue macrophages and provide novel insights into the mechanisms by which tissue macrophages participate in iron recycling to support erythropoiesis. The novel dual-functional CD169-tdTomato-Cre mouse line will have broad applications for studies of gene function in macrophages and their role in modulating erythropoiesis.

Introduction

Macrophages are present in essentially every tissue in the body and perform highly versatile functions, including host defense,1 tissue repair and homeostasis,2 inflammatory mediation,3 and phagocytosis of apoptotic cells,4 as well as senescent red blood cells.5 Beyond their well-established immune roles, growing evidence has demonstrated the role of macrophages in hematopoiesis.6,7 Notably, erythroblastic island (EBI) macrophages play important roles in erythropoiesis, establishing the microenvironment through direct physical interactions with developing erythroid cells and the secretion of supportive cytokines.8 A recent study suggested that EBI macrophages may additionally support myelopoiesis, expanding their hematopoietic roles beyond erythroid development.9

Despite extensive studies on macrophages, there are still important technical issues that need to be addressed and significant knowledge gaps that need to be filled. First, recent studies have shown that anti-CD169 staining detected multiple nonmacrophage cell types because of the fragmentation of macrophages and the attachment of fragmented macrophage remnants to neighboring cells, leading to the misassignment of macrophages.10,11 These findings raise concerns about using anti-CD169 antibodies to assess macrophages and underscore the need to develop a CD169 reporter mouse model for monitoring CD169 expression. Second, although several Cre mouse models, such as Lyz2-Cre,12 F4/80-Cre,13 Csf1r-Cre,14 Cx3cr1-Cre,15 and CD169-Cre,16 have been developed to manipulate gene expression in macrophages, none can specifically and efficiently manipulate gene expression in macrophages. Third, although it is generally accepted that macrophages from each tissue exhibit different functions, the differences among tissue macrophages remain largely unknown. In the context of erythrophagocytosis and erythropoiesis, it has been well documented that spleen (SP) red pulp macrophages play essential roles in erythrophagocytosis and recycling iron to support erythropoiesis.17 However, whether and how macrophages in other tissues such as bone marrow (BM) or liver are engaged in this process remain unclear and somewhat controversial.18

This study aims to address these issues. We generated a CD169-tdTomato-Cre mouse model and documented that the novel dual-functional CD169-tdTomato-Cre mouse line can be used in monitoring CD169 expression and selective gene manipulation in macrophages. We compared tissue macrophages and found that, at steady state, SP and BM macrophages are engaged in erythrophagocytosis. In contrast, no erythrophagocytosis was detected in liver and lung macrophages, challenging the traditional view that liver Kupffer cells are involved in erythrophagocytosis.19,20 Interestingly, liver macrophages expressed the highest levels of the heme/hemopexin complex receptor CD91,21 along with the expression of CD163, the receptor for the hemoglobin/haptoglobin complex,22 suggesting that liver macrophages are involved in iron recycling via uptake of the heme/hemopexin complex and the hemoglobin/haptoglobin complex. A comparison of tissue macrophages also revealed different expression patterns of macrophage surface markers, such as CD169, Vcam1, Cx3cr1, and Csf1r, and iron recycling genes, such as Hmox1 and Spic. Our findings have uncovered previously unknown differences among tissue macrophages and suggest that different tissue macrophages contribute to iron recycling via shared as well as distinct mechanisms, providing novel insights into the mechanisms by which macrophages support erythropoiesis.

Materials and methods

Mice

CD169-tdTomato-Cre mice were generated using CRISPR/Cas9 technology at Biocytogen. The detailed procedures are described in supplemental Methods. Gt(ROSA)26Sortm1(EYFP)Cos/J and C57BL/6J mice were purchased from The Jackson Laboratory. Mice aged 10 to 13 weeks were used for the experiments. All animal protocols were reviewed and approved by the institutional animal care and use committee of the New York Blood Center.

Flow cytometric analysis

Single cells derived from mouse BM, SP, liver, lung, and peripheral blood were used to detect the expression of CD169 and other parameters. The detailed methods for the preparation of single-cell suspensions and flow cytometry staining are provided in Supplemental Methods.

Enrichment and imaging flow cytometry analysis of EBIs

Native mouse BM EBIs were enriched as previously described.8 After fixation, enriched EBIs were resuspended in 1× phosphate-buffered saline/2% bovine serum albumin to block nonspecific binding and then stained with anti-mouse Ter119-fluorescein isothiocyanate, F4/80-AF647 antibodies, and Hoechst33342. The detailed procedures for EBI enrichment, staining, and analysis are provided in Supplemental Methods.

Single-cell RNA sequencing (scRNA-seq) analysis

For fastq files, the Cell Ranger version 8.0 pipeline was used for genome alignment to obtain the gene expression matrix. R software version 4.2.2 was then used for downstream analysis of single-cell sequencing matrices. The detailed methods of single-cell sequencing analyses are provided in the supplemental Materials.

Statistics

FlowJo version 10.8.0 was used to analyze flow cytometry standard data. ImageJ version 1.54J was used to quantify western blots. GraphPad Prism version 9.0 was used for statistical analyses. All data are reported as mean ± standard error of the mean. Differences between 2 groups were calculated using an unpaired Student t test. Significance was set at P < .05.

The methods for other relevant experiments are provided in Supplemental Methods. The antibodies and primers used in this study are listed in supplemental Tables 1 and 2, respectively.

Results

Anti-CD169 antibody staining detected multiple nonmacrophage cell types in BM

CD169 is a macrophage surface marker, and macrophages have been assessed by anti-CD169 antibody staining. However, it has been reported recently that fragmented macrophage remnants attach to other cell types, leading to misassignment of macrophage identity.10,11 To examine the extent of macrophage misassignment by anti-CD169 antibody staining in BM, we stained BM cells using 2 monoclonal anti-CD169 antibodies. The gating strategy for F4/80+CD169+ cells is shown in supplemental Figure 1A. Quantification of the CD169+ cells revealed that both antibodies detected ~40% CD169+ cells within F4/80+ cells (Figure 1A). Importantly, cytospin images of the sorted F4/80+CD169+ cells showed that most of the sorted F4/80+CD169+ cells did not exhibit macrophage morphology (Figure 1B). To define which cell types are contained in F4/80+CD169+ cells, we stained BM cells using the anti-CD169 antibody along with a mixture of lineage-specific antibodies. We used dimensionality reduction analysis to identify populations in the CD169+ cells. The gating strategy for the CD169+ cells used for t-distributed stochastic neighbor embedding is shown in supplemental Figure 1B. Figure 1C shows that there were 7 clusters within CD169+ cells. Quantification of each cluster revealed that Ly6G+ granulocytes, SiglecF+SSChi eosinophils, Ly6C+ monocytes, CD19+ B cells, and CD3e+ T cells accounted for ~37.8%, ~11.6%, ~24.9%, ~7.9%, and ~2.0%, respectively (Figure 1D), demonstrating that most cells positive for anti-CD169 staining in BM were not macrophages. Notably, there was no Ter119+ erythroid cell population within CD169+ cells. We then directly examined CD169 staining on these cells. The gating strategy for various cell types is shown in supplemental Figure 1C. Figure 1E shows that ~78% granulocytes, ~92% eosinophils, ~41% monocytes, 23% T cells, and 13% B cells stained positive for the anti-CD169 antibody. Interestingly, no Ter119+ erythroblasts stained positive (Figure 1F). To examine whether staining patterns change under stress erythropoiesis, we performed similar staining using BM cells from mice injected with erythropoietin (EPO), as previously described.8 Supplemental Figure 2 shows that EPO injection had no effect on CD169 staining patterns. Together, these results indicate that CD169+ macrophage fragments attach to multiple nonmacrophage cell types and that anti-CD169 antibody staining cannot specifically label macrophages in BM.

Figure 1. Misidentification of mouse BM macrophages by anti-CD169 antibodies.

Figure 1.

Figure 1.

Open in a new tab

(A) Flow cytometry analysis of CD169 expression on mouse BM F4/80+ cells using 2 different antibody clones. Ter119−Ly6G−CD45+ mouse BM cells were analyzed for F4/80 expression by plotting against forward scatter height (FSC-H; left). F4/80+ cells were further assessed for CD169 expression by plotting against FSC-H (right). FMO controls were used to establish gating boundaries. (B) Representative cytospin images of sorted BM F4/80+CD169+ cells and the quantitative analysis of the percentage of typical macrophages among them. Scale bar, 10 μm. The black arrow indicates a representative macrophage. (C) Flow cytometry dimensionality reduction analysis showing the heterogeneity of the sorted BM F4/80+CD169+ cells. Beyond F4/80 and CD169, antibodies including Ly6G, SiglecF, Ly6C, CD19, CD3e, and Ter119 were used for lineage identification in this assay. (D) Quantification of cell subpopulations within the sorted BM F4/80+CD169+ cells. (E-F) Flow cytometry analysis showing the CD169+ percentage in granulocytes (CD45+Ly6G+ cells), eosinophils (CD45+SiglecF+ cells), monocytes (CD45+Ly6C+ cells), T cells (CD45+CD3e+ cells), B cells (CD45+CD19+ cells; E), and erythroid cells (CD45−CD11b−Gr1−Ter119+ cells; F). FMO, fluorescence-minus-one; t-SNE, t-distributed stochastic neighbor embedding.

Generation and characterization of CD169-tdTomato-Cre mice

In addition to antibody staining, another means of monitoring gene expression is the use of genetically engineered mouse models with knockin of a fluorescent protein under the promoter of the target gene. Currently, there is no mouse model to directly monitor the expression of CD169. To address this, we generated a CD169-tdTomato-Cre mouse line using the same strategy for generating the Epor-tdTomato-Cre mouse line as described in our previous study.23 As shown in the schematic model (Figure 2A), tdTomato and Cre genes were knocked in at the end of the CD169 coding sequence, and the P2A sequence was inserted between CD169 and tdTomato as well as between tdTomato and Cre to ensure independent expression of CD169, tdTomato, and Cre all under the control of the CD169 promoter. The genotype of the mice was confirmed by polymerase chain reaction (Figure 2B). Western blots showed that tdTomato and Cre recombinase were expressed in enriched BM F4/80+ cells from CD169-tdTomato-Cre+/− and CD169-tdTomato-Cre+/+ mice but not wild-type mice; unexpectedly, CD169 was decreased by ~50% in the sorted macrophages from CD169-tdTomato-Cre+/− mice and was not detected in the sorted macrophages from CD169-tdTomato-Cre+/+ mice (Figure 2C). These findings indicate that the insertion of tdTomato-P2A-iCre at the end of the CD169 coding sequence affected the expression of CD169 while allowing the expression of tdTomato and Cre. Nevertheless, both CD169-tdTomato-Cre+/− and CD169-tdTomato-Cre+/+ mice were born at the expected Mendelian ratio, were viable, and did not exhibit any gross defects. Analyses of peripheral blood (Figure 2D) revealed no differences in blood cells between CD169-tdTomato-Cre and littermate control mice. No differences in total cell numbers (Figure 2E), burst-forming unit–erythroid colonies (Figure 2F), colony-forming unit–erythroid colonies (Figure 2G), and erythroblasts (Figure 2H) in BM were detected between CD169-tdTomato-Cre and littermate control mice. Our results are consistent with a previous finding that deficiency of CD169 did not affect steady-state erythropoiesis in vivo,24 indicating a dispensable role of CD169 in erythropoiesis. As no differences were noted between CD169-tdTomato-Cre+/− and CD169-tdTomato-Cre+/+ mice, CD169-tdTomato-Cre+/− mice were used for subsequent studies.

Figure 2. Generation and characterization of CD169-tdTomato-Cre mice.

Figure 2.

Open in a new tab

(A) Schematic model of the CD169-tdTomato-Cre mouse model. (B) PCR analysis showing the genotypes of CD169-tdTomato-Cre mice. (C) Western blot and quantitative analyses showing the expression of tdTomato, Cre, and CD169 in enriched BM F4/80+ cells of CD169-tdTomato-Cre mice and WT controls. (D) Peripheral blood parameters, including red blood cell (RBC), HGB, HCT, Retic, MCV, MCH, MCHC, and WBC, of CD169-tdTomato-Cre mice and littermate controls (N = 10). (E-G) Quantitative analysis showing no differences in the number of total BM cells (E), BFU-E colonies (F), CFU-E colonies (G), and erythroblasts (H) between CD169-tdTomato-Cre mice and littermate controls (n = 3). BFU-E, burst-forming unit-erythroid; bp, base pairs; CFU-E, colony-forming unit-erythroid; HCT, hematocrit; HGB, hemoglobin concentration; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume; ns, not significant; PCR, polymerase chain reaction; Retic, reticulocyte percentage; WBC, white blood cell count; WT, wild-type.

CD169-tdTomato was only detected in macrophages in the BM

Next, we examined the expression of CD169-tdTomato in BM cells using CD169-tdTomato-Cre mice. In marked contrast to anti-CD169 antibody staining, which detected ~40% CD169+ cells within F4/80+ myeloid cells, CD169-tdTomato was detected in ~2% of F4/80+ myeloid cells (Figure 3A). Notably, ~90% of F4/80hi cells expressed CD169-tdTomato (Figure 3B), and the sorted F4/80+CD169-tdTomato+ cells exhibited typical macrophage morphology (Figure 3C), indicating that F4/80hi cells are macrophages. Furthermore, in marked contrast to the finding that a very high percentage of granulocytes, eosinophils, monocytes, T cells, and B cells stained positive for the anti-CD169 antibody, CD169-tdTomato was detected in a very low percentage in these cells (Figure 3D–I). Moreover, CD169-tdTomato was not detected in hematopoietic stem and progenitor cells (supplemental Figure 3). Finally, t-distributed stochastic neighbor embedding analyses of BM cells using CD169-tdTomato-Cre+/− revealed that, in contrast to the earlier finding that only ~4% of cells positive for anti-CD169 staining were F4/80hi macrophages, ~90% of CD169-tdTomato+ cells were F4/80hi macrophages (Figure 3J). These results demonstrate that the CD169-tdTomato-Cre mouse line can be used to define macrophages.

Figure 3. Expression of CD169-tdTomato in mouse BM cells.

Figure 3.

Open in a new tab

Representative flow cytometry analyses showing the expression of CD169-tdTomato in mouse BM F4/80+ (A) and F4/80high (B) cells. Ter119−Ly6G−CD45+ mouse BM cells were analyzed for F4/80 expression by plotting against FSC-H (left). Note that the same plot of F4/80 vs FSC-H of the Ter119−Ly6G−CD45+ cells was used for both panels A-B. F4/80+ (A) or F4/80high (B) cells were further assessed for CD169-tdTomato expression by plotting against FSC-H (right). (C) Composite representative cytospin images of the sorted BM F4/80+CD169-tdTomato+ macrophages. Scale bar, 10 μm. (D-I) Representative flow cytometry analyses showing the expression of CD169-tdTomato in mouse BM granulocytes (D), eosinophils (E), monocytes (F), T cells (G), B cells (H), and erythroblasts (I). For panels D-I, the left panels show gating strategies for distinct populations: granulocytes (Ly6G+), eosinophils (SiglecF+; analyzed via SiglecF vs side scatter height [SSC-H]), monocytes (Ly6C+), T cells (CD3e+), and B cells (CD19+), all analyzed against FSC-H, except eosinophils, which were analyzed via SiglecF vs SSC-H, whereas the right panels display CD169-tdTomato expression, analyzed against FSC-H for each population. (J) Flow cytometry dimensionality reduction analyses showing the heterogeneity of the sorted BM F4/80+CD169-tdTomato+ cells. Beyond F4/80 and CD169-tdTomato, antibodies including Ly6G, SiglecF, Ly6C, CD19, CD3e, and Ter119 were used for lineage identification in this assay. FSC-A, forward scatter area; t-SNE, t-distributed stochastic neighbor embedding; WT, wild-type.

F4/80+CD169-tdTomato+ macrophages resemble F4/80+Epor-tdTomato+ EBI macrophages

We have recently documented that EBI macrophages are characterized by the expression of Epor and that they express CD169.8,23 We hypothesize that F4/80+CD169-tdTomato+ macrophages and F4/80+Epor-tdTomato+ EBI macrophages are the same macrophage population. To test this, we first examined the expression of well-known EBI macrophage surface markers. Figure 4A shows that F4/80+CD169-tdTomato+ cells expressed Vcam1 and Timd4, whereas ~70% of them expressed CD163. Cx3cr1 and Csf1r are known markers for both monocytes and macrophages.25,26 We found that although Cx3cr1 and Csf1r were clearly expressed in BM monocytes (supplemental Figure 4), BM F4/80+CD169-tdTomato+ cells unexpectedly expressed low levels of Cx3cr1 and no Csf1r (Figure 4A). Furthermore, BM F4/80+CD169-tdTomato+ cells did not express monocyte-specific markers Ly6C and Ccr2 (Figure 4A), further demonstrating that the F4/80+CD169-tdTomato+ cells are macrophages. Next, we examined whether F4/80+CD169-tdTomato+ macrophages express Epor. Western blot analyses revealed the expression of Epor in the sorted F4/80+CD169-tdTomato+ macrophages (Figure 4B) and that EPO stimulation led to Stat5 phosphorylation in the sorted F4/80+CD169-tdTomato+ macrophages (Figure 4C). Moreover, EPO injection led to approximately threefold increases in F4/80+CD169-tdTomato+ macrophages (Figure 4D). Furthermore, ImageStream analyses revealed that the central macrophage of the EBI was CD169-tdTomato+ (Figure 4E). Quantitative analyses showed that >80% of EBIs were composed of F4/80+CD169-tdTomato+ macrophages (Figure 4F), which was similar to our previous finding that EBIs were composed of F4/80+Epor+ macrophages.8,23 These results strongly suggest that F4/80+CD169-tdTomato+ and F4/80+Epor-tdTomato+ macrophages are the same macrophage population.

Figure 4. F4/80+CD169-tdTomato+ macrophages resemble F4/80+Epor-tdTomato+ EBI macrophages.

Figure 4.

Open in a new tab

(A) Representative flow cytometry analyses exemplifying the expression of quintessential EBI macrophage surface markers, including CD169-tdTomato, Vcam1, Timd4, CD163, Csf1r, Cx3cr1, Ly6C, and Ccr2, on F4/80+CD169-tdTomato+ macrophages. (B) Western blot analysis showing the expression of Epor in the sorted F4/80+CD169-tdTomato+ macrophages. (C) Western blot analysis showing the phosphorylation of Stat5 in the sorted F4/80+CD169-tdTomato+ macrophages after an 8-hour starvation period and subsequent stimulation with EPO at 5 U/mL. (D) Quantitative analysis showing the number of F4/80+CD169-tdTomato+ macrophages in mouse BM after EPO injection. (E) Representative imaging flow analysis showing the expression of CD169-tdTomato in F4/80+Ter119+ EBIs. (F) Quantification analysis of the CD169-tdTomato+ percentage in EBI macrophages.

Differential expression of CD169-tdTomato in tissue macrophages

We next examined the expression of CD169-tdTomato in tissue macrophages beyond BM. As we and others previously reported, SP red pulp macrophages and liver Kupffer cells are defined as CD45+F4/80HighCD11blow.23,27 Figure 5A shows that CD169-tdTomato+ was detected in ~80% of SP red pulp macrophages but not in other cell populations (supplemental Figure 5A). Similarly, CD169-tdTomato was detected in >90% of liver Kupffer cells (Figure 5B) but not in other cell populations (supplemental Figure 5B). We also examined the expression of CD169-tdTomato in the lung and found that CD169-tdTomato expression was detected in >90% of the CD45+F4/80HighCD11blow population (Figure 5C) but not in other populations (supplemental Figure 5C). Notably, the expression levels of CD169-tdTomato were significantly different, with the highest expression in liver Kupffer cells, followed by BM, along with lower expression levels in lung and SP macrophages (Figure 5D). Given the critical role of CD169+ macrophages in pathogen recognition,28 these findings suggest that liver macrophages may play a more important role in pathogen recognition than macrophages from BM, SP, and lung.

Figure 5. Expression of CD169-tdTomato in tissue macrophages and differences among tissue macrophages.

Figure 5.

Open in a new tab

(A-C) Representative flow cytometry analyses showing the expression of CD169-tdTomato in tissue macrophages, including SP (A), liver (B), and lung (C). CD45+ cells from indicated tissues were analyzed using F4/80 vs CD11b (left). F4/80+CD11blow cells were assessed for CD169- tdTomato expression (vs FSC-H) (right). (D) Quantitative analyses showing the MFI of CD169-tdTomato in indicated tissue macrophages. (E) Composite representative cytospin images of enriched F4/80+ macrophages from the indicated tissues. (F) Quantitative analyses of the percentage of enriched tissue F4/80+ macrophages that had engulfed erythrocytes. (G) Quantitative analyses displaying the number of phagocytosed erythrocytes in the enriched SP and BM F4/80+ macrophages. The RBC number represents the number of RBCs engulfed by each macrophage. (H-I) Quantitative reverse transcription polymerase chain reaction (qRT-PCR; H) and western blot analysis (I) showing the expression of erythrophagocytosis and genes involved in iron recycling, including Hmox1, Spic, and Slc40a1, in enriched F4/80+ macrophages from indicated tissues. (J-K) qRT-PCR (J) and western blot analysis (K) showing the expression of CD163 and CD91 in enriched F4/80+ macrophages from indicated tissues. MFI, mean fluorescence intensity; WT, wild-type.

Engulfment of erythrocytes by SP and BM macrophages but not liver and lung macrophages at steady state

It is generally expected that significant differences exist among tissue macrophages. But it remains largely unclear what the differences are. As an initial step to compare the differences, we enriched macrophages using F4/80 magnetic beads and performed cytospin analyses. As expected, the enriched cells from SP (supplemental Figure 6A), BM (supplemental Figure 6B), liver (supplemental Figure 6C), and lung (supplemental Figure 6D) were CD45+ and expressed F4/80 and CD169-tdTomato, indicating that they are highly enriched macrophages. Representative cytospin images of highly enriched CD169-tdTomato+ macrophages are shown in Figure 5E, which shows that although macrophages from these tissues exhibited macrophage morphology, there were differences. One striking difference was that erythrocytes were present in macrophages from SP and BM but not from the liver and lung. Quantitative analyses revealed that ~20% of the enriched BM macrophages and ~25% of the enriched SP macrophages contained erythrocytes (Figure 5F). Surprisingly, although most SP macrophages engulfed 1 erythrocyte, most BM macrophages engulfed multiple erythrocytes (Figure 5G). More cytospin images of the enriched SP and BM macrophages with engulfed erythrocytes are shown in supplemental Figure 6E–F, respectively.

Examination of genes involved in iron recycling suggests that liver macrophages are involved in iron recycling via an erythrophagocytosis-independent manner

Our conclusion that SP and BM macrophages are involved in iron recycling is further supported by the finding that genes involved in iron recycling, such as Hmox1, Spic, and Slc40a1, were abundantly expressed in SP and BM macrophages, but not in lung macrophages, at both the mRNA (Figure 5H) and protein levels (Figure 5I). Notably, Hmox1 and Spic were also expressed in liver macrophages, although at a relatively lower level than that in SP and BM macrophages. Additionally, Slc40a1 was highly expressed in liver macrophages. Together with the finding that no erythrocytes were detected in liver macrophages, these findings strongly suggest that liver macrophages are involved in iron recycling in an erythrophagocytosis-independent manner. In addition to erythrophagocytosis, macrophages can also take up the hemoglobin/haptoglobin complex via CD16322 and the heme/hemopexin complex via CD91.21 We therefore compared the expression of CD163 and CD91 in various macrophages and found that CD91 was highly expressed in liver macrophages at both the mRNA (Figure 5J) and protein levels (Figure 5K), although they were also expressed in SP and BM macrophages. Altogether, our findings indicate that SP, BM, and liver macrophages are involved in iron recycling via shared (CD163-mediated hemoglobin/haptoglobin complex uptake and CD91-mediated heme/hemopexin uptake) and distinct mechanisms (phagocytosis of senescent/defective erythrocytes).

scRNA-seq analyses revealed the differential expression of iron metabolism genes, macrophage markers, and hemoglobin genes among tissue macrophages

To further compare the differences among tissue macrophages, we performed analyses on scRNA-seq data on cells from SP,29 BM,30 liver,31 and lung.32 We used Harmony to remove batch effects and to integrate the samples. Supplemental Figure 7A shows the cell clusters from each of the 4 tissues. In total, there were 33 clusters. Among them, 5 clusters (C8, C23, C28, C31, and C32) expressed macrophage marker genes, including C1qa, Adgre1, Vcam1, and Hmox1 (supplemental Figure 7B). The 5 clusters from each tissue were extracted and merged into a macrophage cluster (supplemental Figure 7C). As expected, these macrophage clusters expressed macrophage feature genes (supplemental Figure 7D). We first compared the expression of genes involved in iron metabolism and found that consistent with our real-time polymerase chain reaction and western blot analyses on enriched macrophages from different tissues, Hmox1 (Figure 6A) and Spic (Figure 6B) were most abundantly expressed in SP and BM macrophages, with Slc40a1 (Fpn) abundantly expressed in SP, BM, and liver macrophages (Figure 6C).

Figure 6. scRNA-seq analyses revealing the novel differences among tissue macrophages.

Figure 6.

Open in a new tab

(A-F) Violin plots showing the expression of Hmox1 (A), Spic (B), Slc40a1 (C), Csf1r (D), Cx3cr1 (E), and Vcam1 (F) in indicated tissue macrophages according to scRNA-seq data. (G-H) Violin plots showing the expression of Hba-a1 (G) and Hbb-bt (H) in indicated tissue macrophages according to scRNA-seq data. (I) Violin plots depicting the top 10 highly expressed genes in mouse reticulocytes identified by scRNA-seq. *P < .05; **P < .01; ***P < .001.

We were also interested in comparing the expression of macrophage marker genes because although Cx3cr1 and Csf1r are generally considered macrophage markers,25,26 our flow cytometry analyses showed that they were not expressed or were expressed at very low levels in BM macrophages. Consistent with our flow cytometry results, Figure 6D shows that Cx3cr1 was not detected in BM macrophages. Moreover, Cx3cr1 was not detected in SP and liver macrophages either, although it was detected in <50% of lung macrophages. Similarly, a low level of Csf1r was detected in only ~35% of BM macrophages and ~60% of lung macrophages, although a high level of Csf1r was detected in the majority of SP (~80%) and liver (~90%) macrophages (Figure 6E). In contrast, Vcam1 was highly expressed in almost all SP (~95%) and ~62% of BM macrophages, but low-level expression was detected in ~28% of liver macrophages and ~10% of lung macrophages (Figure 6F). These findings demonstrate a tissue-specific expression of previously well-known macrophage markers.

Finally, we examined the expression of globin genes and found that transcripts of Hba-a1 (Figure 6G) and Hbb-bt (Figure 6H) were detected in >80% of BM macrophages, whereas they were not detected in SP and liver macrophages. Surprisingly, Hba-a1 and Hbb-bt were also expressed in almost all lung macrophages. Given that reticulocytes, but not senescent red cells, contain RNA, we reasoned that the detection of globin transcripts in BM macrophages is associated with the phagocytosis of reticulocytes. To test this, we analyzed the scRNA-seq data of reticulocytes from a previous study30 and found that only α- and β-globin genes were detected, whereas transcripts of Gypa, Gata1, and Epor were not detected in mouse reticulocytes at the single-cell resolution (Figure 6I). This finding strongly suggests the engulfment of reticulocytes by BM macrophages. In contrast, the detection of globin genes in lung macrophages is cell intrinsic because no reticulocytes were present in lung macrophages.

CD169-tdTomato-cre–mediated efficient recombination in various tissue macrophages

In addition to using tdTomato as a reporter for CD169 expression, another purpose of generating the CD169-tdTomato-Cre mouse line is to use the Cre recombinase for selective gene manipulation in macrophages. To examine the recombination efficiency and specificity of CD169-Cre, we crossed CD169-tdTomato-Cre mice with R26R-EYFP mice, which have a loxP-flanked STOP sequence, followed by the EYFP inserted into the Gt(ROSA)26-Sor locus.33 The expression of the Cre recombinase will lead to cleavage of the STOP sequence, and EYFP expression will be observed in the Cre-expressing cells of the double mutant offspring. As shown in Figure 7, EYFP was detected in almost all BM CD169-tdTomato+ macrophages (Figure 7A), SP red pulp macrophages (Figure 7B), liver Kupffer cells (Figure 7C), and lung macrophages (Figure 7D). In contrast, EYFP was not detected in any other cell types in the BM (supplemental Figure 8A). Furthermore, EYFP was not detected in any cell populations in SP (supplemental Figure 8B), liver (supplemental Figure 8C), and lung (supplemental Figure 8D). These results demonstrate specific and efficient CD169-Cre–mediated recombination in macrophages in various tissues. Thus, CD169-tdTomato-Cre mice can be used to selectively manipulate gene expression in macrophages.

Figure 7. Cre-mediated efficient recombination in various tissue macrophages of CD169-tdTomato-Cre mice.

Figure 7.

Open in a new tab

(A) BM. (B) SP. (C) Liver. (D) Lung. Representative plots displaying F4/80 vs FSC-H (left). For BM, Ly6G-Ter119− cells were shown to highlight F4/80+ macrophages (A). For SP, liver, and lung, CD45+ cells were used to reveal RPM in the SP (B), Kupffer cells in the liver (C), and tissue macrophages in the lung (D), respectively. Representative plots of CD169-tdTomato vs FSC-H of tissue macrophages revealing the CD169-tdTomato+ macrophage population (middle). Representative plots of R26R-EYFP vs FSC-H of the CD169-tdTomato+ macrophages revealing Cre-mediated recombination efficiency in indicated tissue macrophages (right; n = 3). RPM, red pulp macrophage.

Discussion

In this study, we have sought to address the technical issues and knowledge gaps in the study of macrophages. We have generated a CD169-tdTomato-Cre mouse line that allowed not only the correct assignment of macrophages but also specific and efficient Cre-mediated recombination in macrophages. We anticipate that this dual-functional mouse line will be widely used for the study of gene function in macrophages and their subsequent effects on erythropoiesis. Through comparison of macrophages from different tissues using multiple approaches, we found that tissue macrophages differed in erythrophagocytosis at steady state and in the expression patterns of macrophage surface markers and iron recycling genes among tissue macrophages. Our findings provide novel insights into the mechanisms by which tissue macrophages contribute to the iron supply for erythropoiesis.

CD169 is a member of the sialic acid–binding immunoglobulin-like lectin (Siglec) family, which mediates cell-cell interaction via sialylated glycoconjugates. It is well accepted that CD169 is a specific marker for macrophages. However, recent studies have shown that anti-CD169 antibody staining led to the misassignment of macrophages because of the fragmentation of macrophages and the attachment of the fragmented macrophage remnants to surrounding cells.10,11 Here, we found that anti-CD169 antibody staining detected multiple nonmacrophages in the BM. A recent study reported that in addition to supporting erythropoiesis, the central macrophage may also support granulopoiesis.9 Given that the remnant attachment is not random but rather reflects possible interactions, our findings suggest that within the EBI, CD169 may mediate the interaction of EBI macrophages with granulocytic precursors. Future studies should explore whether there is a specific interaction between EBI macrophages and granulocytes. Although it has been reported that the deletion of macrophages led to impaired erythropoiesis in the BM,34,35 whether depletion of macrophages affects granulopoiesis has yet to be defined. Additionally, our findings that anti-CD169 antibody staining detected eosinophils, monocytes, B cells, and T cells suggest the potential interactions of macrophages with these cells. Together with the finding that the deletion of CD169+ macrophages led to egress of hematopoietic stem cells to peripheral blood,7 it is reasonable to propose that macrophages play a broad role in hematopoiesis, which warrants further investigation.

Several Cre mouse models have been used to manipulate gene expression in macrophages. These include Lyz2-Cre,12 F4/80-Cre,13 Csf1r-Cre,14 and Cx3cr1-Cre.15 However, it should be noted that these genes are not specifically expressed in macrophages. Both Lyz2 and F4/80 are widely expressed in myeloid cells. It has been reported that Csf1r-Cre mediated efficient cleavage in hematopoietic stem cells.36 Cx3cr1 is widely expressed in hematopoietic and nonhematopoietic cells.37,38 A CD169-Cre mouse line was generated16 and used for gene manipulation in macrophages.36,39 Although this CD169-Cre mouse line is macrophage specific, it has 2 limitations. First, a recent study reported that the recombination efficiency of the CD169 recombinase in macrophages was only ~50%.36 Second, because of the lack of a fluorescent reporter, this mouse line does not allow the simultaneous detection and characterization of the macrophages in which the gene is manipulated. In this study, we have generated a novel dual-functional CD169-tdTomato-Cre mouse line that enabled the accurate identification of macrophages as well as specific and efficient Cre-mediated cleavage in the macrophages.

Macrophages in various tissues exhibit different functions. In the context of erythrophagocytosis and erythropoiesis, it has been well established that SP red pulp macrophages play an essential role in recycling iron for erythropoiesis by phagocytosing senescent red cells.40 We recently documented that EBI macrophages in BM also possess the iron recycling machinery.8 Here, we found that erythrocytes were observed not only in the enriched SP macrophages but also in BM macrophages. Moreover, the finding that α- and β-globin gene transcripts, which are the only genes that can be detected in reticulocytes at the single-cell level by 10x Genomics,30 were found in BM but not SP macrophages indicate that BM macrophages engulf not only senescent red cells but also newly generated defective reticulocytes in BM. It should be noted that, different from enriched BM macrophages, sorted F4/80+Epor+ macrophages8,23 and F4/80+CD169-tdTomato+macrophages (Figure 3B) from BM did not contain erythrocytes in their cytoplasm, indicating that the macrophages that have engulfed erythrocytes are prone to the damage induced by the sorting process.

It has been reported that liver macrophages (Kupffer cells) are involved in the phagocytosis of damaged red cells and in iron recycling.41–43 However, we did not find erythrocytes in enriched liver macrophages at steady state. Nevertheless, we found that liver macrophages also expressed Hmox1, Spic, and Slc40a1, as well as the hemoglobin/haptoglobin complex receptor CD163 and the heme/hemopexin complex receptor CD91, which suggests that liver macrophages are involved in iron recycling but via an erythrophagocytosis-independent manner at steady state. Together, our findings strongly suggest that SP, BM, and liver macrophages participate in iron recycling via shared as well as distinct mechanisms.

Globin genes were initially thought to be expressed only in erythroid cells. However, there is increasing evidence that globin genes are also expressed in nonerythroid cells, including alveolar epithelial cells,44 arterial endothelial cells,45 neurons and glial cells,46 and chondrocytes.47 It has also been reported that the βminor globin gene can be induced in macrophages by lipopolysaccharide and interferon gamma.48 Here, we found the expression of both α- and β-globin genes in lung macrophages at the single-cell resolution. Interestingly, the function of hemoglobin appears to be different from that in erythrocytes, where it serves as an oxygen carrier, transporting oxygen to tissues throughout the body. Although it has been reported that hemoglobin regulates nitric oxide signaling in arterial endothelial cells45 and that hemoglobin plays an essential role in the survival of chondrocytes under a hypoxic environment,47 the function of the monomeric βminor globin in activated macrophages remains unknown. Together, these findings challenge the conventional perception that hemoglobin is solely expressed by erythroid cells. It will be interesting to investigate whether hemoglobin in lung macrophages contributes to gas exchange by selectively deleting Hba1 or Hbb1 using the novel CD169-tdTomato-Cre mouse line we generated.

Supplementary Material

1

The full-text version of this article contains a data supplement.

Key Points.

  • CD169-tdTomato-Cre mice enable the accurate identification of macrophages and selective gene manipulation in macrophages.

  • SP and BM but not liver macrophages are involved in erythrophagocytosis at steady state.

Acknowledgments

This work was supported in part by National Heart, Lung, and Blood Institute, National Institutes of Health grants R01HL181370, P01HL149626, and R01HL165202.

Footnotes

Conflict-of-interest disclosure: The authors declare no competing financial interests.

References

  • 1.Taylor PR, Martinez-Pomares L, Stacey M, Lin HH, Brown GD, Gordon S. Macrophage receptors and immune recognition. Annu Rev Immunol. 2005;23:901–944. [DOI] [PubMed] [Google Scholar]
  • 2.Wynn TA, Chawla A, Pollard JW. Macrophage biology in development, homeostasis and disease. Nature. 2013;496(7446):445–455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Shapouri-Moghaddam A, Mohammadian S, Vazini H, et al. Macrophage plasticity, polarization, and function in health and disease. J Cell Physiol. 2018;233(9):6425–6440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Luo B, Gan W, Liu Z, et al. Erythropoeitin signaling in macrophages promotes dying cell clearance and immune tolerance. Immunity. 2016;44(2):287–302. [DOI] [PubMed] [Google Scholar]
  • 5.Bratosin D, Mazurier J, Tissier JP, et al. Cellular and molecular mechanisms of senescent erythrocyte phagocytosis by macrophages. A review. Biochimie. 1998;80(2):173–195. [DOI] [PubMed] [Google Scholar]
  • 6.Winkler IG, Sims NA, Pettit AR, et al. Bone marrow macrophages maintain hematopoietic stem cell (HSC) niches and their depletion mobilizes HSCs. Blood. 2010;116(23):4815–4828. [DOI] [PubMed] [Google Scholar]
  • 7.Chow A, Lucas D, Hidalgo A, et al. Bone marrow CD169+ macrophages promote the retention of hematopoietic stem and progenitor cells in the mesenchymal stem cell niche. J Exp Med. 2011;208(2):261–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Li W, Wang Y, Zhao H, et al. Identification and transcriptome analysis of erythroblastic island macrophages. Blood. 2019;134(5):480–491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Romano L, Seu KG, Papoin J, et al. Erythroblastic islands foster granulopoiesis in parallel to terminal erythropoiesis. Blood. 2022;140(14):1621–1634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Millard SM, Heng O, Opperman KS, et al. Fragmentation of tissue-resident macrophages during isolation confounds analysis of single-cell preparations from mouse hematopoietic tissues. Cell Rep. 2021;37(8):110058. [DOI] [PubMed] [Google Scholar]
  • 11.Hume DA, Millard SM, Pettit AR. Macrophage heterogeneity in the single-cell era: facts and artifacts. Blood. 2023;142(16):1339–1347. [DOI] [PubMed] [Google Scholar]
  • 12.Clausen BE, Burkhardt C, Reith W, Renkawitz R, Förster I. Conditional gene targeting in macrophages and granulocytes using LysMcre mice. Transgenic Res. 1999;8(4):265–277. [DOI] [PubMed] [Google Scholar]
  • 13.Schaller E, Macfarlane AJ, Rupec RA, Gordon S, McKnight AJ, Pfeffer K. Inactivation of the F4/80 glycoprotein in the mouse germ line. Mol Cell Biol. 2002;22(22):8035–8043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Loschko J, Rieke GJ, Schreiber HA, et al. Inducible targeting of cDCs and their subsets in vivo. J Immunol Methods. 2016;434:32–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Yona S, Kim KW, Wolf Y, et al. Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis. Immunity. 2013;38(1):79–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Karasawa K, Asano K, Moriyama S, et al. Vascular-resident CD169-positive monocytes and macrophages control neutrophil accumulation in the kidney with ischemia-reperfusion injury. J Am Soc Nephrol. 2015;26(4):896–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Youssef LA, Rebbaa A, Pampou S, et al. Increased erythrophagocytosis induces ferroptosis in red pulp macrophages in a mouse model of transfusion. Blood. 2018;131(23):2581–2593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Klei TR, Meinderts SM, van den Berg TK, van Bruggen R. From the cradle to the grave: the role of macrophages in erythropoiesis and erythrophagocytosis. Front Immunol. 2017;8:73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Macrophages Ganz T. and iron metabolism. Microbiol Spectr. 2016;4(5). [DOI] [PubMed] [Google Scholar]
  • 20.Rahimy C. Acute leukemias in children: prognostic factors. Rev Infirm. 1990;40(14):49–50. [PubMed] [Google Scholar]
  • 21.Hvidberg V, Maniecki MB, Jacobsen C, Højrup P, Møller HJ, Moestrup SK. Identification of the receptor scavenging hemopexin-heme complexes. Blood. 2005;106(7):2572–2579. [DOI] [PubMed] [Google Scholar]
  • 22.Schaer DJ, Schaer CA, Buehler PW, et al. CD163 is the macrophage scavenger receptor for native and chemically modified hemoglobins in the absence of haptoglobin. Blood. 2006;107(1):373–380. [DOI] [PubMed] [Google Scholar]
  • 23.Zhang H, Wang S, Liu D, et al. EpoR-tdTomato-Cre mice enable identification of EpoR expression in subsets of tissue macrophages and hematopoietic cells. Blood. 2021;138(20):1986–1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bai J, Fan F, Gao C, et al. CD169-CD43 interaction is involved in erythroblastic island formation and erythroid differentiation. Haematologica. 2023;108(8):2205–2217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Imai T, Hieshima K, Haskell C, et al. Identification and molecular characterization of fractalkine receptor CX3CR1, which mediates both leukocyte migration and adhesion. Cell. 1997;91(4):521–530. [DOI] [PubMed] [Google Scholar]
  • 26.Pixley FJ, Stanley ER. CSF-1 regulation of the wandering macrophage: complexity in action. Trends Cell Biol. 2004;14(11):628–638. [DOI] [PubMed] [Google Scholar]
  • 27.Sheng J, Ruedl C, Karjalainen K. Most tissue-resident macrophages except microglia are derived from fetal hematopoietic stem cells. Immunity. 2015;43(2):382–393. [DOI] [PubMed] [Google Scholar]
  • 28.Perez OA, Yeung ST, Vera-Licona P, et al. CD169+ macrophages orchestrate innate immune responses by regulating bacterial localization in the spleen. Sci Immunol. 2017;2(16):eaah5520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Chen S, Gao Y, Fan Y, et al. The dynamic change of immune responses between acute and recurrence stages of rodent malaria infection. Front Microbiol. 2022;13:844975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Peng Y, Liang L, Zhang H, et al. Single-cell profiling of ineffective erythropoiesis in a mouse model of β-thalassaemia intermedia. Br J Haematol. 2023;201(5):982–994. [DOI] [PubMed] [Google Scholar]
  • 31.Xiong X, Kuang H, Ansari S, et al. Landscape of intercellular crosstalk in healthy and NASH liver revealed by single-cell secretome gene analysis. Mol Cell. 2019;75(3):644–660.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Wu G, Lee YY, Gulla EM, et al. Short-term exposure to intermittent hypoxia leads to changes in gene expression seen in chronic pulmonary disease. Elife. 2021;10:e63003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Srinivas S, Watanabe T, Lin CS, et al. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol. 2001;1(4):4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Chow A, Huggins M, Ahmed J, et al. CD169(+) macrophages provide a niche promoting erythropoiesis under homeostasis and stress. Nat Med. 2013;19(4):429–436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ramos P, Casu C, Gardenghi S, et al. Macrophages support pathological erythropoiesis in polycythemia vera and beta-thalassemia. Nat Med. 2013;19(4):437–445. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Wei Q, Boulais PE, Zhang D, Pinho S, Tanaka M, Frenette PS. Maea expressed by macrophages, but not erythroblasts, maintains postnatal murine bone marrow erythroblastic islands. Blood. 2019;133(11):1222–1232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Jung S, Aliberti J, Graemmel P, et al. Analysis of fractalkine receptor CX(3)CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol. 2000;20(11):4106–4114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Kim KW, Vallon-Eberhard A, Zigmond E, et al. In vivo structure/function and expression analysis of the CX3C chemokine fractalkine. Blood. 2011;118(22):e156–e167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Dalla E, Papanicolaou M, Park MD, et al. Lung-resident alveolar macrophages regulate the timing of breast cancer metastasis. Cell. 2024;187(23):6631–6648.e20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Gottlieb Y, Topaz O, Cohen LA, et al. Physiologically aged red blood cells undergo erythrophagocytosis in vivo but not in vitro. Haematologica. 2012;97(7):994–1002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Kondo H, Saito K, Grasso JP, Aisen P. Iron metabolism in the erythrophagocytosing Kupffer cell. Hepatology. 1988;8(1):32–38. [DOI] [PubMed] [Google Scholar]
  • 42.Theurl I, Hilgendorf I, Nairz M, et al. On-demand erythrocyte disposal and iron recycling requires transient macrophages in the liver. Nat Med. 2016;22(8):945–951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Terpstra V, van Berkel TJ. Scavenger receptors on liver Kupffer cells mediate the in vivo uptake of oxidatively damaged red blood cells in mice. Blood. 2000;95(6):2157–2163. [PubMed] [Google Scholar]
  • 44.Archimbaud E, Anglaret B, Thomas X, et al. Continuous-infusion daunorubicin and carboplatin for high-risk acute myeloid leukemia in the elderly. Leukemia. 1992;6(8):776–779. [PubMed] [Google Scholar]
  • 45.Straub AC, Lohman AW, Billaud M, et al. Endothelial cell expression of haemoglobin alpha regulates nitric oxide signalling. Nature. 2012;491(7424):473–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Biagioli M, Pinto M, Cesselli D, et al. Unexpected expression of alpha- and beta-globin in mesencephalic dopaminergic neurons and glial cells. Proc Natl Acad Sci U S A. 2009;106(36):15454–15459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Zhang F, Zhang B, Wang Y, et al. An extra-erythrocyte role of haemoglobin body in chondrocyte hypoxia adaption. Nature. 2023;622(7984):834–841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Liu L, Zeng M, Stamler JS. Hemoglobin induction in mouse macrophages. Proc Natl Acad Sci U S A. 1999;96(12):6643–6647. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2211284-supplement-1.pdf (11.6MB, pdf)

RESOURCES