Regular Articles
Differential Regulation of Lympho-Myelopoiesis by Stromal Cells in the Early and Late Phases in BALB/c Mice Repeatedly Exposed to Lipopolysaccharide
2016 年 39 巻 12 号 p. 1939-1947
詳細
2016 年 39 巻 12 号 p. 1939-1947
Chronic lipopolysaccharide (LPS) exposure to mice reduces the lymphoid compartment and skews the hematopoietic cell compartment toward myeloid-cells, which is considered to be a direct effect of LPS on hematopoietic stem cells. However, the effect of chronic LPS exposure on stromal-cells, which compose the hematopoietic microenvironment, has not been elucidated. Here, we investigated early- and late-phase effects of repeated LPS exposure on stromal-cells. During the early phase, when mice were treated with 5 or 25 µg LPS three times at weekly intervals, the numbers of myeloid-progenitor (colony forming unit-granulocyte macrophage (CFU-GM)) cells and B lymphoid-progenitor (CFU-preB) cells in the bone-marrow (BM) rapidly decreased after each treatment. The number of CFU-GM cells recovered from the initial decrease and then increased to levels higher than pretreatment levels, whereas the number of CFU-preB cells remained lower than pretreatment levels. In the BM, expression of genes for positive-regulators of myelopoiesis including granulocyte colony-stimulating factor (G-CSF), granulocyte macrophage colony-stimulating factor (GM-CSF), and interleukin (IL)-6 and negative-regulators of B lymphopoiesis including tumor necrosis factor (TNF)-α was up-regulated, whereas expression of positive-regulators of B lymphopoiesis including stromal cell-derived factor (SDF)-1, IL-7, and stem cell factor (SCF) was down-regulated. During the late phase, the number of CFU-preB cells remained lower than pretreatment levels 70 d after the first treatments with 5 and 25 µg LPS, whereas the number of CFU-GM cells returned to pretreatment levels. IL-7 gene expression in the BM remained down-regulated, whereas gene-expression levels of SDF-1 and SCF were restored. Thus, chronic LPS exposure may impair stromal-cell function, resulting in prolonged suppression of B lymphopoiesis, which may appear to be senescence similar to the hematological phenotype.
Inflammation and infection alter hematopoiesis output of bone marrow (BM) by favoring myelopoiesis, especially granulopoiesis, over B lymphopoiesis.1,2) Lipopolysaccharide (LPS), a major cell wall component of Gram-negative bacteria, induces inflammation via Toll-like receptor (TLR) 4 expressed on the cell surface. In the BM, TLR4 is expressed not only on hematopoietic cells, but also on non-hematopoietic cells such as stromal cells. Thus, LPS is a useful agent to study how hematopoiesis is regulated by stromal cells during inflammation and infection.
A single injection of LPS in mice augments myelopoiesis and suppresses B lymphopoiesis, which is similar to the phenomenon observed in mice during inflammation and infection. In the BM, hematopoietic stem cells and hematopoietic progenitor cells are regulated by positive and negative factors produced by stromal cells.3,4) This reciprocal regulation of myelopoiesis and B lymphopoiesis after LPS treatment is likely an indirect effect via stromal cells rather than a direct effect of LPS on hematopoietic cells.5,6)
Several groups recently used the mouse model of repeated low-dose LPS exposure and reported that chronic TLR signaling alters hematopoietic components.7–9) Chronic (repeated) exposure of mice to LPS reduces lymphoid cells and skews the hematopoietic cell compartment toward myeloid cells, circumstances that resemble the phenotype of aged mice.7–9) This lymphoid vs. myeloid bias is considered to be due to a direct effect of LPS on hematopoietic stem cells.7,9) However, several reports showed that age-related deterioration of B lymphopoiesis is in part due to functional impairment of stromal cells.10–13) The effect of chronic exposure of hematopoietic cells to LPS has been extensively studied. However, little is known about the effect of chronic exposure to LPS on stromal cell function. Thus, we addressed the effect of chronic exposure of stromal cells to LPS.
In the present study, mice were treated with 5 or 25 µg LPS three times at weekly intervals. We evaluated the hematological parameters in mice once a week after the first, second, and third treatments with LPS, which is designated as the early phase. The late phase was designated as 42 and 70 d after the first treatment with LPS.
The present results shed light on differences in the regulatory mechanisms for myelopoiesis and B lymphopoiesis between the early and late phases after repeated treatment with LPS.
Eight-week old BALB/c male mice purchased from CLEA Japan Inc. (Fuji, Japan) were used. All animals were maintained in pathogen-free conditions under a 12-h light/12-h dark cycle. Sterilized water and food pellets were provided ad libitum. All experimental protocols involving the laboratory mice used in this series of studies were reviewed and approved by the Interdisciplinary Monitoring Committee for the Proper Use and Welfare of Experimental Animals, a peer review panel established at Nihon University School of Medicine, with the experimental code AP14M026 and AP14M051.
ReagentsRecombinant mouse (rm) granulocyte macrophage colony-stimulating factor (GM-CSF), rm tumor necrosis factor (TNF)-α, and recombinant human (rhu) interleukin (IL)-6 were purchased from R&D Systems (Minneapolis, MN, U.S.A.).
LPS TreatmentFor this study, LPS from Escherichia coli (LPS055:B5, Sigma Chemical Co., St. Louis, MO, U.S.A.) was diluted in pyrogen-free saline to a final concentration of 25 or 125 µg/mL, and mice were injected intravenously with a single 5 or 25-µg dose three times at weekly intervals.6,14,15) A control group of mice was treated with the same volume of pyrogen-free saline (0.2 mL per mouse). In the early-phase experiment, three mice from each treatment group were evaluated 1, 3, and 6 h and 1, 3, 5, and 7 d after the first, second, and third treatments with LPS or saline (Fig. 1). In the late-phase experiment, three mice from each treatment group were evaluated 42 and 70 d after the first treatment with LPS or saline (Fig. 1).
Mice were treated with 5 or 25 µg LPS three times at weekly intervals. In the early-phase experiment, three mice from each treatment group were evaluated 1, 3, and 6 h and 1, 3, 5, and 7 d after the first, second, and third treatments with LPS or saline. In the late-phase experiment, three mice from each treatment group were evaluated 42 and 70 d after the first treatment with LPS or saline.
Peripheral blood was collected from the retro-orbital plexus of mice under isoflurane anesthesia. The number of peripheral blood cells was determined using a Sysmex PocH-100 iV Diff hematology analyzer (Sysmex Co., Kobe, Japan). Peripheral blood smeared on glass slides was stained with Wright-Giemsa reagent and then counted differentially according to the type of white blood cells (WBCs) based on 100 cells.16) BM cells were harvested from the femur of each mouse as described elsewhere.12)
Progenitor Cell Colony AssayMyeloid-progenitor (colony forming unit (CFU)-GM) cells were assayed using MethoCult M3231 (Stem Cell Technologies Inc., Vancouver, BC, Canada) supplemented with 10 ng/mL rmGM-CSF. B lymphoid-progenitor (CFU-preB) cells were assayed using MethoCult M3630 according to the manufacturer’s protocol.12,13) Cells were cultured in a humidified incubator at 37°C and 5% CO2. CFU-GM and CFU-preB cells were counted 7 d after plating the cells.
Preparation of Cultured BM Stromal CellsFor preparation of the stromal monolayer, BM cells were cultured at 1×106 cells/mL in a 6-well flat bottom plate (Falcon 3046 Becton Dickinson Labware) in 4 mL α-MEM supplemented with 10% fetal bovine serum (FBS).17) A subconfluent adherent layer was obtained after 14 d of culture. Then the supernatant was replaced with fresh medium. For LPS treatment, after 2 d of culture, 100 ng/mL LPS was added to the culture dish. For the control, the same volume of pyrogen-free saline was added to the culture. After 1, 3, or 6 h of culture, the culture medium was removed completely, and stromal cells were used for RNA extraction. For treatment with cytokines, 10 ng/mL rmTNF-α, rhuG-CSF, rmGM-CSF, or rhuIL-6 instead of LPS was added to the culture dish. After 4, 8, or 24 h of culture, the culture medium was removed completely, and stromal cells were used for RNA extraction.
Gene Expression AssayThe gene expression levels of cytokines were determined by real-time PCR using the Applied Biosystems 7900 Sequence Detection System. Briefly, total RNA from total BM cells and cultured stromal cells was isolated using ISOGEN reagent (Nippongene Corp., Toyama, Japan). mRNA was reverse transcribed using Superscript III (Life Technologies) and Oligo-dT (Promega Corp., Madison, WI, U.S.A.). The expression levels of genes were determined by real-time PCR using TaqMan™ Universal Fast PCR master mix (Applied Biosystems, Foster City, CA, U.S.A.) and specific primers. Specific primers and probes for murine G-CSF, GM-CSF, IL-6, stromal cell-derived factor (SDF)-1, IL-7, stem cell factor (SCF), TNF-α, transforming growth factor (TGF)-β, and glyceraldehyde phosphate dehydrogenase (GAPDH) genes were purchased from Applied Biosystems as described elsewhere.6,13) PCR conditions and data analysis were performed in accordance with the instructions provided with the Sequence Detection System (ver. 2.0).
Statistical AnalysisData are expressed as the mean±standard deviation (S.D.). Data sets were compared using the two-tailed unpaired Student’s t-test. Differences were considered statistically significant at p<0.05.
The WBCs in mice treated with the first, second, and third treatments of 5 µg LPS rapidly decreased to about 70% of the pretreatment level on day 1, followed by an increase above the pretreatment level (Fig. 2A). Changes in the number of WBCs in mice treated with 25 µg LPS were similar to those in mice treated with 5 µg LPS (Fig. 2A).
Numerical changes in the number of peripheral white blood cells (WBCs) (A), CFU-GM cells (B), and CFU-preB cells (C) in mice after repeated LPS treatment are shown. The samples of peripheral blood cells and femoral BM were obtained from three mice each at 1, 3, 5, and 7 d after the first, second, and third injections of 5 and 25 µg LPS. Each bar represents the mean±S.D. obtained from three mice.
Repeated treatment with 5 and 25 µg LPS did not change the number of red blood cells, whereas it decreased the number of platelets, followed by an increase above the pretreatment levels (Supplementary Fig. S1).
The number of CFU-GM cells after the first and second treatments with 5 µg LPS rapidly decreased to about 70% of the pretreatment level on day 1, followed by a prompt recovery and increase above the pretreatment level (Fig. 2B). However, after the third treatment with 5 µg LPS, no decrease in the number of CFU-GM cells on day 1 was observed. The number of CFU-GM cells increased to 138% of pretreatment levels on day 3 and returned to pretreatment levels on day 7. Changes in the number of CFU-GM cells in mice treated with 25 µg LPS were similar to those in mice treated with 5 µg LPS.
The number of CFU-preB cells after the first treatment with 5 µg LPS rapidly decreased on day 1, followed by a gradual recovery to 70% of the pretreatment level on day 7 (Fig. 2C). After the second and third treatment with 5 µg LPS, the number of CFU-preB cells decreased to 34% of the pretreatment level on day 1, followed by oscillation between 40 and 70% of pretreatment levels.
Changes in the number of CFU-preB cells after the first LPS treatment between the 5 and 25-µg LPS treatment groups were different, although changes in the number of CFU-preB cells after the second and third LPS treatments between the 5 and 25-µg LPS treatment groups were similar. After the first LPS treatment, the maximum decrease in the 25-µg LPS treatment group was more pronounced than that in the 5-µg LPS treatment group (16% of the pretreatment level in the 25-µg LPS treatment group vs. 30% of the pretreatment level in the 5-µg LPS treatment group). Furthermore, the recovery rate in the 25-µg LPS treatment group on day 7 was lower than that in the 5-µg LPS treatment group on day 7 (38% of the pretreatment level in the 25-µg LPS treatment group vs. 70% of the pretreatment level in the 5-µg LPS treatment group).
Changes in the Expression Levels of Regulatory Cytokine Genes (G-CSF, GM-CSF, IL-6, SDF-1, IL-7, SCF, TNF-α, and TGF-β) in the BM during the Early Phase after LPS TreatmentTo examine the role of stromal cells in myelopoiesis and B lymphopoiesis in mice after LPS treatment, gene expression levels of cytokines that regulate hematopoiesis and that are produced by stromal cells were evaluated.
The gene expression levels of positive regulators of myelopoiesis (G-CSF, GM-CSF, and IL-6) in the BM were markedly increased during the first 6 h after the first, second, and third treatments with 5 µg LPS (Figs. 3A–C) and 25 µg LPS (Figs. 4A–C), followed by a rapid decrease by day 1. The magnitude of the increase in the gene expression level after treatment was more pronounced for G-CSF than the other cytokines.
The expression levels of positive regulators of myelopoiesis such as G-CSF (A), GM-CSF (B), and IL-6 (C), positive regulators of B lymphopoiesis such as SDF-1 (D), IL-7 (E), and SCF (F), and negative regulators of B lymphopoiesis such as TNF-α (G) and TGF-β (H) were evaluated 1, 3, and 6 h and 1, 3, 5, and 7 d after the first, second, and third treatments with 5 µg LPS. Each bar represents the mean±S.D. obtained from three mice.
The gene expression levels of positive regulators of B lymphopoiesis (SDF-1, IL-7, and SCF) in the BM after the first, second, and third treatments with 5 µg LPS (Figs. 3D–F) and 25 µg LPS (Figs. 4D–F) were all decreased. Interestingly, although the gene expression levels of SDF-1 and IL-7 in the BM of mice after the first treatment with 25 µg LPS did not return to pretreatment levels and remained lower than pretreatment levels on day 7, those in mice after the second and third treatments with 25 µg LPS nearly returned to pretreatment levels on day 7 (Figs. 4D, E).
The expression levels of positive regulators of myelopoiesis such as G-CSF (A), GM-CSF (B), and IL-6 (C), positive regulators of B lymphopoiesis such as SDF-1 (D), IL-7 (E), and SCF (F), and negative regulators of B lymphopoiesis such as TNF-α (G) and TGF-β (H) were evaluated 1, 3, and 6 h and 1, 3, 5, and 7 d after the first, second, and third treatments with 25 µg LPS. Each bar represents the mean±S.D. obtained from three mice.
Next, we examined negative regulators of B lymphopoiesis such as TNF-α and TGF-β after treatment with both 5 µg LPS (Figs. 3G, H) and 25 µg LPS (Figs. 4G, H). The gene expression levels of TNF-α in the BM after the first, second, and third treatments with LPS were increased, whereas the gene expression levels of TGF-β in the BM were not increased except for after the second treatment with 25 µg LPS.
Changes in the Gene Expression Levels of Cytokines (G-CSF, GM-CSF, IL-6, SDF-1, IL-7, SCF, TNF-α, and TGF-β) in Cultured BM Stromal Cells after LPS TreatmentThe direct effect of LPS on stromal cells was investigated using cultured stromal cells. The expression levels of G-CSF, GM-CSF, and IL-6 increased to 55460, 65186, and 112878%, respectively, of the pretreatment levels by 6 h after LPS treatment (Fig. 5A). Gene expression levels of IL-7 and SCF after LPS treatment were decreased to 45 and 32% of pretreatment levels by 6 h, respectively (Fig. 5B). However, the gene expression of SDF-1 after LPS treatment oscillated within 77 and 167% of the pretreatment level (Fig. 5B). The gene expression of TNF-α after LPS treatment increased to 20045% of the pretreatment level and remained high thereafter, and the gene expression of TGF-β did not change (Fig. 5C).
Expression levels of positive regulators of myelopoiesis such as G-CSF, IL-6, and GM-CSF (A), positive regulators of B lymphopoiesis such as SDF-1, IL-7, and SCF (B), and negative regulators of B lymphopoiesis such as TNF-α and TGF-β (C) in BM stromal cells cultured with LPS (100 ng/mL) were evaluated after 4, 8, and 24 h of culture. The results are expressed as a ratio of each cultured stromal cell fraction without LPS. Each bar represents the mean±S.D. obtained from three cultures.
Direct effects of TNF-α, G-CSF, GM-CSF, and IL-6, which were increased by LPS in the in vivo study, on stromal cells were investigated by using cultured stromal cells (Figs. 6A–C).
Expression levels of positive regulators of B lymphopoiesis such as SDF-1, IL-7, and SCF in BM stromal cells cultured with 10 ng/mL TNF-α (Α), G-CSF (B), GM-CSF (C), or IL-6 (D) were evaluated after 4, 8, and 24 h of culture. The results are expressed as a ratio of each cultured stromal cell fraction without cytokine. Each bar represents the mean±S.D. obtained from three cultures.
When cultured stromal cells were treated with TNF-α (Fig. 6A), the gene expression of SDF-1, IL-7, and SCF decreased by 8 h (SDF-1: 13% of the untreated control group, IL-7: 7% of the untreated control group, SCF: 29% of the untreated control group), followed by an increase for each cytokine.
When treated with G-CSF (Fig. 6B), the gene expression of SDF-1 and IL-7 was temporarily increased by 4 h (SDF-1: 285% of the untreated control group, IL-7: 197% of the untreated control group), followed by a decrease for each cytokine (SDF-1: 71% of the untreated control group, IL-7: 49% of the untreated control group). The gene expression of SCF decreased to 72% of the untreated control group by 4 h and remained unchanged thereafter.
When treated with GM-CSF (Fig. 6C), the gene expression of SDF-1, IL-7, and SCF gradually decreased by 24 h (SDF-1: 11% of the untreated control group, IL-7: 6% of the untreated control group, SCF: 27% of the untreated control group).
When treated with IL-6 (Fig. 6D), the gene expression of SDF-1, IL-7, and SCF decreased by 8 h (SDF-1: 31% of the untreated control group, IL-7: 21% of the untreated control group, SCF: 79% of the untreated control group), followed by an increase for each cytokine.
Numerical Changes in the Numbers of Peripheral Blood Cell Types and CFU-GM and CFU-PreB Cells in the BM in Mice during the Late Phase after LPS TreatmentAt 42 d after the first LPS treatment, the numbers of WBCs in mice treated with 5 or 25 µg LPS were 63% (p<0.005) and 70% (p<0.05) of those in control mice without LPS treatment, respectively (Fig. 7A). At 70 d after the first LPS treatment, the numbers of WBCs in mice treated with 5 or 25 µg LPS were 47% (p<0.005) and 67% (p<0.05), respectively, of those in control mice without LPS treatment (Fig. 7A).
Changes in the number of peripheral white blood cells (WBCs) (A), CFU-GM (B) cells, and CFU-preB (C) cells in the BM of mice after the third treatment with 5 and 25 µg LPS are shown. In panel A, the gray and black columns show the lymphocyte component and granulocyte macrophage component, respectively. The columns on the left and right sides show data from 42 and 70 d after the first LPS treatment, respectively. In panels B and C, the gray columns on the left side and the black columns on the right side show data from 42 and 70 d after the first LPS treatment, respectively. The samples were obtained from three mice 42 and 70 d after the first treatment with 5 and 25 µg LPS. Each bar represents the mean±S.D. obtained from three mice. * p<0.05, †p<0.005 vs. control.
The numbers of femoral CFU-GM cells in mice 42 d after the first treatment with 5 or 25 µg LPS increased by 121 and 129% (p<0.05) of those in control mice without LPS treatment, respectively (Fig. 7B, gray columns on the left side). The number of CFU-GM cells in mice 70 d after the first treatment with 5 or 25 µg LPS was almost same as that in control mice without LPS treatment (Fig. 7B, black columns on the right side).
The numbers of femoral CFU-preB cells in mice 42 d after the first treatment with 5 or 25 µg LPS were 85 and 68% (p<0.05), respectively, of those in control mice without LPS treatment (Fig. 7C, gray columns on the left side). The numbers of CFU-preB cells in mice 70 d after the first treatment with 5 or 25 µg LPS were significantly decreased to 65% (p<0.05) and 44% (p<0.005), respectively, of the pretreatment level (Fig. 7C, black columns on the right side).
Changes in the Expression Levels of Regulatory Cytokine Genes (G-CSF, GM-CSF, IL-6, SDF-1, IL-7, SCF, TNF-α, and TGF-β) in the BM during the Late Phase after LPS TreatmentRegarding positive regulators of myelopoiesis such as G-CSF, GM-CSF, and IL-6 (Figs. 8A–C), the gene expression levels of these cytokines in mice 70 d after the first treatment with 5 or 25 µg LPS were all decreased compared with those in mice 42 d after the first treatment with 5 or 25 µg LPS and were also significantly lower than levels in control mice without LPS treatment.
The expression levels of positive regulators of myelopoiesis such as G-CSF (A), GM-CSF (B), and IL-6 (C), positive regulators of B lymphopoiesis such as SDF-1 (D), IL-7 (E), and SCF (F), and negative regulators of B lymphopoiesis such as TNF-α (G) and TGF-β (H) were evaluated 42 and 70 d after the first treatment with 5 and 25 µg LPS. In each panel, the gray columns on the left side and the black columns on the right side show data for treatment with 5 µg LPS and 25 µg LPS, respectively. Each bar represents the mean±S.D. obtained from three mice. * p<0.05, †p<0.005, ‡p<0.001 vs. control.
Regarding positive regulators of B lymphopoiesis such as SDF-1, IL-7, and SCF (Figs. 8D–F), the gene expression levels of SDF-1 and SCF in mice 70 d after 5 and 25 µg LPS treatment were increased compared with those in mice 42 d after 5 and 25 µg LPS treatment. In contrast, the gene expression levels of IL-7 in mice 70 d after the first treatment with 5 and 25 µg LPS were further decreased compared with those in mice 42 d after the first treatment with 5 and 25 µg LPS and were also significantly lower than that in control mice without LPS treatment (p<0.005).
Regarding negative regulators of B lymphopoiesis such as TNF-α and TGF-β (Figs. 8G, H), the gene expression levels of these cytokines in mice 70 d after the first treatment with 5 and 25 µg LPS were almost the same as those in mice 42 d after the first treatment with 5 and 25 µg LPS.
In this study, we focused on the mechanisms of the regulatory functions of stromal cells for myelopoiesis and B lymphopoiesis in the early and late phases after repeated LPS exposure.
During the early phase, first, second, and third LPS treatments all resulted in rapid decreases in the numbers of femoral CFU-GM and CFU-preB cells (Figs. 2B, C), which may be due to mobilization of myeloid- and B lymphoid-progenitor cells from the BM into the peripheral circulation.2,17) Myelopoiesis was subsequently accelerated to replenish the mature neutrophils and myeloid progenitor cells in the BM, whereas B lymphopoiesis remained suppressed (Figs. 2B, C). Local regulation by stromal cells in the BM during the early phase likely facilitates emergency myelopoiesis during inflammation by suppressing B lymphopoiesis (Figs. 3, 4).
The fluctuation of hematopoietic progenitor cells in BALB/c mice after 5 µg LPS treatment is compatible with that in C57BL/6 mice treated with 5 µg LPS, as we previously reported.6) However, the down-regulation of SDF-1, IL-7, and SCF in BALB/c mice treated with 5 µg LPS was more prolonged and profound than the decreases in C57BL/6 mice treated with 5 µg LPS. These different responses to LPS between BALB/c mice and C57BL/6 mice may be due to strain-dependent differences in stromal cell sensitivity to LPS and cytokines such as TNF-α, GM-CSF, and IL-6. The number of fibroblast colony-forming units, which are progenitors of mesenchymal cells, in the BM of C57BL/6 mice is lower compared with that in BALB/c mice.18,19) Thus, the different response to LPS between BALB/c mice and C57BL/6 mice may be due to not only qualitative but also quantitative differences in stromal cells.
After the first LPS treatment, the degree of recovery in the number of CFU-preB cells and in the gene expression levels of SDF-1, IL-7, and SCF from the minimum on day 7 after LPS treatment were lower in the 25-µg LPS treatment group than in the 5-µg LPS treatment group (Figs. 2C, 3D–F, 4D–F). Interestingly, after the second and third LPS treatments, the degree of recovery in the number of CFU-preB cells in the group treated with 25 µg LPS was similar to that in the group treated with 5 µg LPS (Fig. 2C). Furthermore, the degree of recovery in the gene expression of SDF-1, IL-7, and SCF after the second and third treatments was better than that after the first treatment (Figs. 4D–F). Endotoxin tolerance is defined as reduced responsiveness to LPS challenge following a first encounter with endotoxin.20) These results may suggest that the group treated with 25 µg may have become slightly LPS tolerant. However, further investigation is necessary to clarify the mechanism of these phenomena.
During the late phase (day 70 after the first treatment), myelopoiesis was no longer activated, but B lymphopoiesis remained suppressed (Figs. 7B, C). SDF-1 and SCF promote IL-7-dependen colony-formation by CFU-preB cells. In contrast, TNF-α suppresses IL-7-dependent colony-formation by CFU-preB cell (Supplementary Fig. S2). Although the gene expression levels of TNF-α, SDF-1, and SCF returned to pretreatment levels, IL-7 remained down-regulated (Figs. 8D–F).
IL-7 is a crucial factor for B cell development and nonreductant cytokines. B lymphopoiesis in the BM of IL-7 gene-deleted mice is blocked at the transition from the pro-B to the pre-B cell stage.21) IL-7 production by stromal cells decreases with aging, resulting in decreased B lymphopoiesis.10–12) Taken together, the decrease in the number of CFU-preB cells on day 70 after the first treatment with 5 and 25 µg LPS is likely due to decreased production of IL-7 by stromal cells despite restoration of the production of SDF-1 and SCF by stromal cells. Furthermore, these data suggested that the mechanism of suppression of B lymphopoiesis by stromal cells during the late phase after LPS treatment is different from that during the early phase after LPS treatment.
Irradiation induces long-term residual damage to BM stromal cells, resulting in the expression of senescence markers, and repeated LPS treatment promotes cellular senescence in mesenchymal stem cells in organs other than BM.22,23) Thus, chronic inflammation probably impairs not only hematopoietic stem cells7–9) but also stromal cells that compose the hematopoietic microenvironment, and appears to induce senescence similar to the hematological phenotype.
In this study, we observed that BM stromal cells responded well to repeated LPS treatment, showing a repeated and sustained inflammatory hematopoietic response. Although repeated LPS treatment did not cause acute failure of stromal cell function, it induced residual damage to stromal cell function such as a reduction in IL-7 production, resulting in suppression of B lymphopoiesis in the late phase. Thus, further study is necessary to clarify the mechanisms of residual damage to stromal cells caused by LPS.
We thank Sonoko Araki and Miyuki Yuda for their technical assistance. This work was supported in part by a Grant-in-Aid for Science Research C from the Japan Society for the Promotion of Science.
The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
Supplementary Fig. S1
Numerical changes in the number of red blood cells and platelets after LPS treatment during the early phase. Numerical changes in the number of peripheral red blood cells (A) and platelets (B) in mice after repeated LPS treatment are shown. The samples of peripheral blood cells were obtained from three mice each at 1, 3, 5, and 7 d after the first, second, and third injections of 5 and 25 µg LPS. Each bar represents the mean±S.D. obtained from three mice.
Supplementary Fig. S2
Effects of TNF-α, SDF-1, and SCF on colony formation by CFU-preB cells. Effect of TNF-α, SDF-1 and SCF on the proliferation or differentiation of CFU-preB cells was evaluated. Whole BM cells were cultured in a semi-solid medium system containing 10 ng/mL IL-7 in the presence of 10 ng/mL TNF-α, SDF-1, and SCF for 7 d, and the number of CFU-preB colonies was measured. Each bar represents the mean±S.D. obtained from three culture dishes.
* p<0.05, †p<0.001 vs. control.
