Fig 1.
Induction of autoimmune hemolytic anemia (AIHA) in mice.
Mice were injected weekly 2x108 rat erythrocytes intraperitoneally. Blood samples from mice in control and rat-erythrocyte-administered (REA) mice were collected at different time points and analyzed on an automated cell counter. Relative changes in the erythrocyte count and hemoglobin content in the REA mice over a period of 6 weeks are given in panels A and B respectively (p = 0.001 for erythrocytes, and p = 0.013 for hemoglobin after 5–6 injections, ANOVA). Presence of anti-mouse erythrocyte autoantibody was estimated by staining the erythrocytes with anti-mouse IgG/IgM-FITC following flow cytometry. Relative binding of autoantibodies (MFI) to erythrocytes over a period of 4 weeks, from 3rd-6th week, (p<0.001, ANOVA) is given in panel C. Each point on the graph represents mean ± SEM of observations. n = 10 control and 15 REA mice. *p<0.05, **p<0.01 and ****p<0.001 for comparison of the groups (Student t-test).
Fig 2.
Fluorescence imaging of erythrocytes showing the presence of membrane-bound autoantibodies.
Mice were injected weekly with 2x108 rat erythrocytes intraperitoneally for 5 weeks. Blood samples from mice in control and REA groups were collected after 5 injections. Presence of membrane-bound autoantibodies in erythrocytes isolated from REA mice was confirmed by fluorescence microscopy after staining the erythrocytes with F(ab’)2 anti-mouse IgG-PE. Panel A shows the erythrocyte images from control and panel B the same from REA mice. The upper panels display fluorescence images and lower panels the DIC overlay (magnification 100X).
Fig 3.
Erythrocyte turnover in the blood of control and AIHA-induced mice.
Mouse erythrocytes were labeled with biotin in vivo by the two step biotinylation procedure. DIB labeled mice were given intraperitoneal injections of 2x108 rat erythrocytes weekly for 5–6 weeks to induce AIHA. The experimental protocol is given in panel A. Blood samples were collected from control and AIHA-induced mice 3 days after the 5th and 6th injections (6 and 13 days after 2nd biotinylation step). Erythrocytes were stained ex vivo with streptavidin-APC and anti-mouse CD71-PE, and proportions of the different age cohorts were determined. Representative flow histograms showing the proportion of different age groups of erythrocytes after 5 doses of injections are shown in panels B (control) and C (AIHA-induced). Turnover profile of reticulocytes (panel D; p = 0.02, ANOVA), biotinnegative young erythrocytes (panel E; p = 0.046, ANOVA), biotinlow intermediate age group of erythrocytes (panel F; p = 0.010, ANOVA) and biotinhigh old erythrocytes (panels G; p<0.001, ANOVA) in control and AIHA-induced mice have been shown. Each bar on the graph represents mean ± SEM of observations. n = 8 control and 12 AIHA-induced mice. *p<0.05, **p<0.01 and ***p<0.005 for comparison of the groups (Student t-test).
Fig 4.
Autoantibody binding in erythrocytes of different age groups.
DIB labeled mice were given weekly i.p. injections of 2x108 rat erythrocytes for 5–6 weeks to induce AIHA. Blood samples were collected from control and AIHA mice 3 days after the 5th and 6th injection. Erythrocytes were stained ex vivo with anti-mouse IgG-PE, streptavidin-APC and anti-mouse CD71-FITC. Erythrocytes of different age groups were gated and autoantibody level was analyzed in each of them. Each value represents mean ± SEM of data. n = 8 control and 12 AIHA mice. *p<0.05, **p<0.01, ***p<0.005 and ****p<0.001 for comparison of the groups (Student t-test). ANOVA test for autoantibody binding on different age groups of erythrocytes significant (p = 0.001 in terms of MFI and p = 0.021 in terms of erythrocyte proportion with membrane-bound autoantibody).
Fig 5.
Generation of Reactive Oxygen Species (ROS) in control and AIHA-induced mice.
Mice were given intraperitoneal injections of 2x108 rat erythrocytes weekly for 5–6 weeks to induce AIHA. Blood samples were collected from control and AIHA-induced mice 3 days after 5th and 6th doses of injection. Erythrocytes were stained with CM-H2DCFDA and intracellular ROS generation was determined by flow cytometry. Representative histograms showing ROS generation in control and AIHA-induced mice are shown in panel A, and the mean ROS level in whole erythrocyte population is shown in panel B (p = 0.010, ANOVA). Mouse erythrocytes were labeled with biotin in vivo by the two step biotinylation procedure (as per schedule given in Fig 3, panel A). Erythrocytes from DIB stained mice were incubated with CM-H2DCFDA and stained ex vivo with streptavidin-APC and anti-mouse CD71-PE. Erythrocytes of different age groups (reticulocytes, biotinnegative, biotinlow and biotinhigh) were gated (as in Fig 3, panels B and C) and ROS level was analyzed in each of them. ROS level in the different age cohorts of erythrocytes are given in panels C-F. ANOVA tests for each of the subgroups reveal p = 0.741 for reticulocytes (panel C), p = 0.009 for biotinnegative (panel D), p = 0.022 for biotinlow (panel E) and p = 0.005 for biotinhigh (panel F). Each bar on the graph represents mean ± SEM of observations. n = 8 control and 12 AIHA-induced mice. *p<0.05, **p<0.01 and ****p<0.001 for comparison of the groups (Student t-test).
Fig 6.
Total erythroid cells in control and AIHA induced mice.
Mice were given i.p. injections of 2x108 rat erythrocytes weekly for 5–6 weeks to induce AIHA. Mice were sacrificed 3 days after 5th and 6th doses of injection and their BM and spleen cells were harvested. Cells isolated were stained with anti-mouse CD71-PE, anti-mouse Ter119-APC and 7AAD, after blocking with anti-mouse CD16/32, and the proportions of erythroid cells (as described in S3 Fig) were determined. Proportions of erythroid cells in BM and spleen of mice are given above (BM p<0.001 and spleen p = 0.061, ANOVA). Each value represents mean ± SEM of observations. n = 4 control and 6 AIHA mice. *p<0.05, ***p<0.005 and ****p<0.001 for comparison of the groups (Student t-test).
Fig 7.
Changes in relative proportions of erythroid cells in different stages of differentiation in AIHA.
Mice were given weekly i.p. injections of rat erythrocytes to induce AIHA. Mice were sacrificed 3 days after 6th injection and their BM and spleen cells were harvested. Cells isolated were stained with anti-mouse CD71-PE, anti-mouse Ter119-APC and 7AAD, and proportion of erythroid cells were determined as described in S3 Fig. Proportion of erythroid cells in different maturational stages of development in BM and spleen in control and AIHA mice are given along with percent changes in AIHA (ANOVA, p<0.001 for both BM and spleen). Each value represents mean ± SEM of observations. n = 4 control and 6 AIHA-induced mice. *p<0.05, ***p<0.005 and ****p<0.001 for comparison of the groups (Student t-test).
Fig 8.
Generation of anti-mouse autoantibody in erythroid cells of bone marrow and spleen in mice with induced AIHA.
Mice were given intraperitoneal injections of 2x108 rat erythrocytes weekly for 5–6 weeks to induce AIHA. Mice were sacrificed and their bone marrow and spleen cells harvested. Cells isolated were stained with anti-mouse CD71-PE, anti-mouse Ter119-APC and 7AAD, after blocking with anti-mouse CD16/32, to determine the proportion of live erythroid cells as described in S3 Fig. The erythroid cells were co-stained with F(ab’)2 anti-mouse IgG-PE to detect the presence of autoantibodies. Erythroid cells in different stages of maturation (proerythroblasts, erythroblast A, B and C) were gated and autoantibody binding was analyzed in each of them. Presence of membrane-bound autoantibody in the total erythroid populations of bone marrow and spleen after 6 injections is shown in panel A (ANOVA test for bone marrow, p<0.001 and spleen, p = 0.001). Panels B and C show the binding of autoantibody in erythroid cells at various stages of differentiation in bone marrow and spleen (ANOVA test, p<0.001 for both bone marrow and spleen,) respectively. Each bar on the graph represents mean ± SEM of observations. n = 4 control and 6 AIHA-induced mice. *p<0.05, **p<0.01 and ****p<0.001 for comparison of groups (Student t-test).
Fig 9.
ROS generation in the erythroid cells of bone marrow and spleen in AIHA.
Mice were given i.p. injections of 2x108 rat erythrocytes weekly for 5–6 weeks to induce AIHA. Mice were sacrificed 3 days after 5th and 6th injections and their BM and spleen cells were harvested. Cells isolated were stained with anti-mouse CD71-PE, anti-mouse Ter119-APC and 7AAD, and incubated with CM-H2DCFDA. The erythroid cells at different stages of maturation were determined and ROS generation was estimated in each of them. ROS in erythroid cells at different stages of maturation are given above. Each value represents mean ± SEM of observations. n = 4 control and 6 AIHA-induced mice. *p<0.05, **p<0.01 and ****p<0.001 for comparison of the groups (Student t-test).
Fig 10.
A model to depict the stages in the life cycle of erythroid cells that are preferentially eliminated in AIHA is given above. Stages 1 to 4 depict erythroid stages in bone marrow and stages 5–8 depict different age groups of erythrocytes in blood circulation. A summary of results on binding of autoantibodies and ROS generation in different stages and changes in relative proportions of cells in various stages are shown in the table within the figure. Shaded area covering stages 4 till 7 are preferentially eliminated in AIHA mouse model.