The target product D-dencichine (98% purity) was from the Institute of Medicinal Plant Development, Peking Union Medical College and Chinese Academy of Medical Sciences (Beijing, China). Recombinant human (rh) IL-11 was purchased from Qilu Pharmaceutical Co., Ltd. (Jinan, China). Recombinant murine (rm) TPO and rhTPO were obtained from PeproTech Inc. (Rocky Hill, NJ, United States). Stem cell factor (rhSCF) was from R&D Systems (Minneapolis, MN, United States). Primary monoclonal antibodies against phospho-JAK2 (Y1007+Y1008), JAK2, phospho-STAT5 (Y694), STAT5α+STAT5b, phospho-STAT3 (Y705), STAT3, phospho-Akt (S473), Akt, c-Mpl, and Bcl-2 were purchased from Abcam (Cambridge, MA, United States). Primary monoclonal antibodies against phospho-ERK1/2, ERK1/2, Bax polyclonal primary antibody, goat anti-rabbit horseradish peroxidase (HRP) and rabbit anti-mouse HRP were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, United States). APC-labeled anti-mouse CD41 was from eBioscience (San Diego, CA, United States). FITC-labeled anti-human CD41α, APC-labeled anti-human CD42b, PI/RNase staining buffer were purchased from BD Biosciences (La Jolla, CA, United States). The cell culture materials were acquired from Gibco (Grand Island, NY, United States). Plasma fibronectin and fluorescent phallotoxins were from Invitrogen (Eugene, OR, United States). All reagents were purchased from Sigma-Aldrich (St. Louis, MO, United States) unless otherwise stated.

Healthy male Balb/c mice (7 or 8 weeks old and 18–22 g in weight) were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). The mice were maintained under a 12 h/12 h light/dark cycle and given free access to sterilized food and purified water for 1 week in room temperature (RT) at 25 ± 1°C and humidity of 60%. The mice were randomly divided into six groups. (1) The mice in the control group were injected intraperitoneally with normal saline (NS). (2) The mice in the model group were treated with NS for 7 days and injected intravenously with 80 mg/kg carboplatin on day 7. (3) The mice in the positive group were injected subcutaneously with 0.37 mg/kg IL-11. (4) The mice in D-dencichine treatment groups were intraperitoneally injected with 0.75, 1.5, and 3.0 mg/kg D-dencichine. As shown in Figure 1B, these mice in the experimental groups received NS, IL-11, or D-dencichine for 7 days and then intravenously injected with carboplatin, whereas the control group was intravenously injected with volumetric NS. Each group was given the same treatments for another 7 days. After the 14-day treatment, the mice were deprived of food overnight and euthanized. All of the animal experiments were approved by the Research Ethics Committee of the Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China (SCXK 2014-0001).

Murine fetal livers from 14.5 days postcoital were prepared into single-cell suspension by successively passing through 18- and 22-gauge needles and filtering with a 70 μm cell strainer. The cells were cultured in Dulbecco modified Eagle medium (Gibco Life Technologies, United States) supplemented with 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin, and 50 ng/mL rmTPO for 5 days. Changing the medium or supplementing with additional rmTPO during the culture period was unnecessary.

Platelets in the peripheral blood from the retro-orbital plexus of the mice were bled through heparinized capillaries into EDTA-coated tubes; the samples were collected on the indicated days and counted automatically in a hematology analyzer (Sysmex XT-1800i/2000IV, Kobe, Japan). Hematopoietic factors in serum samples were determined using the commercially available murine TPO quantikine ELISA kit (R&D Systems, Minneapolis, MN, United States) and murine IL-6 ELISA kit (Abcam, Cambridge, MA, United States) according to the manufacturer’s protocol.

To assess platelet production (reticulated platelets) in vivo, a small amount of whole blood was collected from tail vein and placed in anticoagulant. One hundred microliter thiazole orange (TO, 0.1 μg/mL in PBS), 18 μL antibody solution (CD41-APC 1/40 in PBS), and 2 μL blood were mixed. The mixture was pipetted gently and incubated for 30 min at RT. Light exposure was avoided. Each sample was fixed in 1% paraformaldehyde for 15 min and analyzed immediately on a FACSCalibur (BD Biosciences, San Jose, CA, United States). TO-positive platelets were considered reticulated (Prislovsky et al., 2008).

In vivo biotinylation approach (double labeling platelets) was used to determine platelet clearance as reported in the literature (Prislovsky et al., 2008). Briefly, mice were injected via tail vein with 35 μg/g body weight sulfo-NHS-biotin (BioVision, United States). The blood was collected from the tail vein at the indicated time points and placed directly in microfuge tubes. The blood samples were then diluted 20× in anticoagulant mix, and centrifuged at 125 × g at RT. One hundred microliter supernatant was collected. CD41-APC antibody and streptavidin-PE (BD Biosciences) were added to label the biotinylated platelets for 40 min at RT in the dark. After fixation in 1% paraformaldehyde, the samples were analyzed via flow cytometry.

Murine femoral and tibia were dissected out and the marrow was flushed out with buffer solution. Mouse spleen capsule was cut at one end, and the cells were gently squeezed out from the cut end using blunt forceps. A single-cell suspension was prepared with three or four passages through a 3-ml syringe and 23-gauge needle, and red cells were lysed on ice using the lysate buffer (Beyotime, China). For the baseline ploidy analysis, cells were analyzed as described previously with minor modifications (Fuhrken et al., 2008). BM and spleen cells were harvested and labeled with an APC-conjugated CD41 antibody for 30 min at 4°C. The cells were then washed twice in 2 mM EDTA in PBS and fixed in 0.5% paraformaldehyde for 10 min at RT. The fixed cells were permeabilized with 70% methanol for 1 h at 4°C, then stained with propidium iodide (PI) solution (0.5 mL/test) containing RNase for 15 min at RT, and analyzed through flow cytometry.

Hematopoietic stem and progenitor Cells (CD34⁺) were purchased from Nuo Wei Biotechnology Co., Ltd. (Beijing, China). The purity of CD34⁺ cells is >90% by flow cytometry. CD34⁺ cells were cultured in serum-free medium (StemSpan SFEM, Stem Cell Technologies, Canada) supplemented with different growth factors, including 20 ng/mL rhSCF and 10 ng/ml rhTPO. Cells were seeded in 24-well plates at a density of 4 × 10⁴/mL, and were cultured for different days. A total of 1 × 10⁵ cultured cells were collected at the indicated time points, and were labeled with FITC-conjugated mouse anti-human CD41α and APC-conjugated mouse anti-human CD42b antibodies for 30 min at RT in the dark. The cells were then fixed in 1% paraformaldehyde and analyzed through flow cytometry.

The M07e and Meg-01 cells were purchased from the American Type Culture Collection (Manassa, VA, United States). The cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Hyclone, United States) containing 10% heat-inactivated FBS and kept in a humidified incubator with 5% CO₂ at 37°C. These cells were seeded into 96-well plates at a density of 5 × 10⁴/well in the presence of various D-dencichine concentrations and grown for 24, 48, and 72 h. Ten microliter CCK-8 regent (Dojindo, Japan) was then added to each well and incubated for 5 h. The absorbance was measured on a microplate reader at 450 nm (Infinite M1000, Tecan, Sunrise, Austria).

Bone marrow cells and platelets were stained with a biotinylated rat anti-mouse c-Mpl mAb (clone AMM2, Immuno-Biological Laboratories, Minneapolis, MN, United States) for 30 min, followed by streptavidin-APC (BD Biosciences) for 40 min, and was determined by flow cytometry.

Dissected tissues (sternum, liver and spleen) from four mice selected randomly from each group were removed and fixed overnight in 10% formaldehyde. Sternum was decalcified in Morse’s solution (10% sodium citrate and 22.5% formic acid) after fixing. Specimens were embedded in paraffin, cut into 5 μm thick sections, and stained with hematoxylin and eosin (H&E) using standard methods. Then we took photos under a microscope and the numbers of megakaryocytes in three microscopy fields per slide were counted under the microscope.

Platelet pellets were prepared as previously described with minor modifications (Lebois et al., 2016). Platelet-rich plasma was obtained by centrifugation at 125 × g for 8 min, followed by centrifugation of the supernatant buffy coat at 125 × g for 8 min. Platelets were washed by two sequential centrifugations at 860 × g in specific buffer. Western blotting analysis was performed as described (Yu et al., 2014). Cells were lysed in cold lysis buffer (Solarbio, China), and proteins were separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane (Bio-Rad, United States). The membranes were washed, blocked, and incubated with the primary antibody and with an appropriate HRP-conjugated secondary antibody afterward. Images were scanned using Image Lab software (Bio-Rad, Hercules, CA, United States) and the relative density of immunoreactive bands was determined using the Quantity One software.

Meg-01 cells were loaded with 5 μM BCECF-AM (Beyotime, China) for 1 h at 37°C. The labeled cells (1 × 10⁵ cells/well) were seeded in a 24-well culture plate precoated with 100 μg/mL fibrinogen and cultured for 3 h at 37°C in serum-free RPMI 1640 medium supplemented with or without D-dencichine. The cells were then washed three times with PBS, and the fluorescence of adhesive cells was determined on a microplate reader.

To assess megakaryocyte migration, 24-transwell chambers with 8-mm pore size polycarbonate membranes (Corning, United States) were coated with 20 μg/mL fibronectin overnight at 4°C. Meg-01 cells were serum starved overnight, and 2 × 10⁵ cells were placed in the upper chamber containing serum-free RPMI 1640 medium with 0.2% bovine serum albumin (BSA). The lower chamber comprised serum-free RPMI 1640 medium with or without D-dencichine (25, 50, and 100 μM). The cells were allowed to migrate for 5 h at 37°C, and the unmigrated cells were removed. The membrane was stained with crystal violet (Beyotime, China) to visualize the cells. The number of migrated cells was examined under an inverted microscope (Molecular Devices, United States). The dye on the membrane was eluted with 33% acetic acid, and crystal violet absorbance was measured with a microplate reader at 570 nm.

Proplatelet formation was performed as previously described with minor modifications (Larson, 2006). Briefly, the fetal liver-derived primary megakaryocyte cells were expanded with 50 ng/mL rmTPO for 5 days and purified over a discontinuous BSA density gradient (1.5%/3%). The purified megakaryocytes were seeded in 200 μg/mL fibrinogen-precoated plates for 5 h at 37°C. The cells were fixed with 10% formalin, permeabilized with 0.25% Triton X-100, and stained with fluorescent phallotoxins, the anti-β1-tubulin antibody and DAPI (Beyotime, China). Images were captured using a fluorescence microscope, and the number of megakaryocyte-producing proplatelets was enumerated.

Experimental data in this report are expressed as the mean + standard deviations (SDs) of at least three independent experiments. Statistical analyses were performed by Student’s t-test to assess the differences between two groups. Comparisons among the multiple groups were performed using the two-way analysis of variance (ANOVA). P < 0.05 was considered statistically significant.

Carboplatin is a second generation platinum drug that is highly effective in the treatment of malignant tumors. Among carboplatin’s toxicities are myelosuppression and thrombocytopenia (Ulich et al., 1995). Therefore, a mouse model of thrombocytopenia induced by carboplatin was applied in this study. To examine whether D-dencichine treatment could ameliorate CIT, we pretreated the randomized mice with different D-dencichine doses (0.75, 1.5, and 3.0 mg/kg) for 7 days and used rhIL-11 as a positive control. When the platelets in a whole blood sample were enumerated using a hematology analyzer on day 7, we observed that D-dencichine could not elevate the platelet level in the peripheral blood, whereas rhIL-11 increased the platelet counts by 29% more than that of the control group (Figure 1C). Carboplatin was then intravenously injected to the mice to induce severe thrombocytopenia on day 7. D-dencichine and rhIL-11 were administered for another 7 days after carboplatin stimulated myelosuppression. Both D-dencichine and rhIL-11 were effective in promoting platelet counts compared with the model control. The platelet counts in the D-dencichine treatment mice (0.75, 1.5, and 3.0 mg/kg) and the rhIL-11 group were approximately 18, 29, 32, and 44% compared with that in the model group, respectively. Differences between the D-dencichine treatment group (1.5 and 3.0 mg/kg) and the model group were statistically significant. The results indicated that D-dencichine had a capacity to heighten the platelets counts of mice suffering from thrombocytopenia induced by carboplatin.

The increased platelet number in mice could arise from increased production or decreased clearance. To clarify the causes of increased platelet counts with D-dencichine treatment, we first detected platelet production by TO staining because it specifically combines with RNA in newly formed platelets (reticulated platelets). As shown in Figure 1D, an increase of newly formed platelets in D-dencichine treatment groups was observed compared with that in the model group. To further investigation the effects of D-dencichine-heightened platelet number, we then assessed the platelet clearance in groups using an in vivo biotinylation method. In contrast, the percentage of platelets cleared was unaltered each group (Figure 1E). These data suggested that the increased platelet count with D-dencichine treatment probably resulted from promoted platelet production but not platelet clearance.

Platelets are released from megakaryocytes residing in BM, which has been proven as a major site of platelet production in mice. Using flow cytometry, we found that D-dencichine treatment group had no more megakaryocytes per nucleated cells in BM than in the model group (Figure 2A). In addition to the BM, the liver and spleen are also important sites of hematopoiesis in mice. Therefore, we evaluated the pathological changes in liver, spleen, and sternum to verify the role of D-dencichine in thrombopoiesis facilitation in vivo. H&E staining revealed that the numbers of megakaryocytes in the BM of D-dencichine treatment group did not increase compared with that in the model group (Figures 2B,C), which further supported the aforementioned observation. Through histologic observations, the spleen megakaryocyte numbers were lower in the model group compared with the other groups, but getting an accurate megakaryocyte counts was not easy because the distribution of megakaryocytes was particularly uneven in spleen. In contrast to the BM and spleen, the liver in the model group was infiltrated by focal necrosis and inflammatory cells. The liver cells of D-dencichine treatment groups were swollen, but the lesions were relieved. These results suggested that D-dencichine treatment promoting thrombopoiesis in vivo must be caused by other mechanisms.

It is known that the process of megakaryocyte maturation is driven by several factors and chemokines. TPO is the primary regulator of platelet production and supports survival, proliferation, and differentiation of megakaryocytes in BM (Szalai et al., 2006; Kaushansky, 2009; Kuter, 2009). We hypothesized that the elevated platelet counts in mice because of the increased serum TPO lever in the D-dencichine treatment group. To confirm this speculation, we tested serum TPO concentrations in groups. Surprisingly, serum TPO levers in D-dencichine treatment groups were lower than those in the model group (Figure 3A). TPO acts by binding itself to a specific cell surface receptor (c-Mpl), which is required for rapid response to platelet loss. Therefore, we analyzed the surface expression of c-Mpl protein in BM cells and platelets through flow cytometry. We found that the c-Mpl expression in platelets of D-dencichine treatment groups increased compared with that of the model group (Figure 3C). Furthermore, western blot analysis of c-Mpl protein expression in platelets increased in D-dencichine treatment group at a dose of 3.0 mg/kg, which further supported the results (Figure 3D). However, c-Mpl expression in BM cells using flow cytometric analysis was unaltered (Figure 3B). We concluded that circulating TPO concentrations might be inversely proportional to the “c-Mpl mass,” which was contributed by platelet counts. Therefore, our speculation about the possibility that altered expression levels of TPO and c-Mpl being directly responsible for the increased platelets could be ruled out. Megakaryopoiesis and platelet production are tightly regulated by additional growth factors and cytokines to maintain a normal number of circulating platelets, such as IL-3, SCF, IL-6, and IL-11. Among these growth factors, IL-6 has been shown to act directly on megakaryocytes to increase platelet production. To learn more about how D-dencichine affects IL-6 expression in mice, we tested serum IL-6 concentrations. Correspondingly, IL-6 concentrations in D-dencichine treatment group were lower compared with that in the model group in a manner that was similar to the effect of TPO (Figure 3E). The data suggested that IL-6 concentrations in the model group were high and its concentrations in the D-dencichine treatment group were in the low lever, which must have been caused by another mechanism.

We performed a series of experiments in vitro to investigate the direct effects of D-dencichine treatment on megakaryocytes proliferation and/or differentiation. First, we chose the megakaryoblastic cell line M07e, the mature megakaryocyte cell line Meg-01, and human primary megakaryocytes derived from umbilical cord blood CD34⁺ cells that were cultured with or without D-dencichine at different doses (25, 50, and 100 μM) for different days. As shown in Figures 4A,B, we found that D-dencichine treatment could not promote the proliferation in both M07e and Meg-01 cells. During development, megakaryocytes undergo a series of transformations that can be identified by expression specific surface proteins, including GPIIb (also known as the integrin subunit aIIb or CD41) and GPIb (GPIb-V-IX complex, CD42b), in association with nuclear maturation and subsequent cytoplasmic maturation (Mazharian et al., 2009). Megakaryocytes development starts expressing integrin CD41α and maturation marker CD42b on their surface upon differentiation from HSCs. The well-known megakaryocyte cell lines, such as M07e, Meg-01, and K562, produced mostly CD41α⁺, but CD42b^- (Nakamura et al., 2014). To detect whether D-dencichine has a direct effect on megakaryocyte differentiation, cord blood-derived CD34⁺ cells were cultured with or without different doses of D-dencichine supplemented by adding rhSCF and rhTPO for 7, 10, and 13 days. Afterwards, we detected the surface expression levels of CD41α⁺CD42b⁺ via flow cytometry. However, our data showed that D-dencichine had no significant effect on megakaryocyte differentiation (Figures 4C,D). Megakaryopoiesis is an especial process that involves gradual differentiation of immature megakaryocyte progenitors into diploid megakaryocytes, which undergoes a progressive megakaryocytic polyploidization and a subsequent process of cytoplasmic maturation leading to platelet release. Platelet production was considered to correlate positively with megakaryocyte polyploidy (several rounds of chromosomal duplication without cell division) to increase DNA contents. Therefore, we determined whether BM and spleen megakaryocytes DNA contents were altered with D-dencichine treatment using flow cytometry. Particularly, we found that DNA contents (2N-4N) of immature megakaryocyte progenitors in model group were significantly lower compared to the control group, but they were not altered in spleen cells. The observed phenomenon might be related to the ability of carboplatin to cross-link DNA in megakaryocyte progenitors in BM (Zeuner et al., 2007). Our data showed that D-dencichine was unable to promote megakaryocyte polyploidy in BM and spleen cells (Figures 5A,B). These data suggested that D-dencichine was unable to promote megakaryocyte proliferation, differentiation, and polyploidy (Figures 5A–D).

Elevated platelet numbers in murine models of carboplatin-induced thrombocytopenia suggested that JAK2-STATs and/or TPO-dependent signaling might be involved in D-dencichine treatment of thrombocytopenia. Based on the TPO lever that was dramatically altered in D-dencichine treatment groups, we concluded that appropriate signaling pathways may become active in hepatocytes. Hence, we determined the known cellular signal pathways of JAK2-STATs (Majka et al., 2002) and TPO-dependent signaling, including ERK1/2 and Akt, which are two key regulators (Nakao et al., 2008; Mazharian et al., 2009). In liver and platelet cells, protein investigation revealed that D-dencichine treatment increased the phosphorylation of JAK2 compared with that in the model group (Figures 6A,B). In contrast, the phosphorylation states of STAT3 and STAT5 (downstream effectors of JAK2 signaling) were not altered. The data suggested that D-dencichine was capable of increasing the phosphorylation of JAK2 but not completely facilitating the JAK2-STATs signaling pathways. In addition to the liver and platelets, spleen signaling proteins should be paid attention to. On the other hand, we examined whether key regulators of TPO-dependent signaling were altered in liver, platelet and spleen cells with D-dencichine treatment. The examination (Figures 7A–C) showed that D-dencichine treatment was capable of increasing the phosphorylation of Akt in platelet cells compared with that in the model group. The increase in Akt phosphorylation was not significant in the liver of D-dencichine treatment and model groups, whereas Akt phosphorylation was unaltered in spleen of each group. Similarly, we found that ERK1/2 phosphorylation in D-dencichine treatment group was enhanced compared with that in the model group in terms of the liver and spleen. Increase in Akt phosphorylation was significant, but the increase in ERK1/2 phosphorylation was more significant. Intriguingly, ERK1/2 phosphorylation was significantly attenuated in D-dencichine treatment group compared to the model group, which was in contrary to the results in liver and spleen samples.

Platelets undergo a specialized form of apoptosis to prolong its survival (Debrincat et al., 2015). The opposite result of ERK1/2 phosphorylation could be related with the platelet survival or apoptosis pathway (Katz et al., 2009; Zhang et al., 2011). Platelet apoptosis is triggered by pharmacological inhibition or conditional loss of Bcl-xL; in addition to thrombocytopenia, a small but significant decrease existed in Bak/Bax (Lebois and Josefsson, 2016). To further reveal the underlying mechanisms, we investigated the effects of D-dencichine treatment on the activation of Bcl-2 and Bax, which regulate the intrinsic pathway to apoptosis (Youle and Strasser, 2008). Western blot analysis showed that D-dencichine treatment caused a stronger activation of Bcl-2/Bax compared with that in the model group (Figure 7D).

D-dencichine treatment displayed a strong effect on the activation of ERK1/2 and Akt in Meg-01 cells, and their phosphorylation states were rapid and transient (Figure 8A). The results strongly revealed that D-dencichine was a positive regulator of TPO-dependent signaling, especially in increasing ERK1/2 phosphorylation. Western blot results indicated that D-dencichine might exert its effects in the terminal stage of thrombocytopoiesis. This process is highly dependent upon a complex network of protein filaments that represent the molecular struts and girders of the cell. Tubulin and actin are both major components of this cytoskeletal network (Italiano et al., 1999). Megakaryocytes migration from a proliferative osteoblastic niche within the BM environment to a capillary-rich vascular niche is an important step for platelet production. Cellular behaviors within adhesion, migration, and PPF, are crucial for megakaryocyte maturation and platelet production (Avecilla et al., 2004). Therefore, we performed mature Meg-01 cells to attach to fibrinogen for 3 h, and the number of attached cells was determined. The attachment to fibrinogen was significantly increased with D-dencichine treatment in a dose-dependent manner compared with that in the control group (Figure 8B). Similarly, we also found that D-dencichine had a strong ability to promote megakaryocyte migration compared with that in the control group (Figure 8C). We then chose fetal liver-derived primary megakaryocytes to investigate the effects of D-dencichine stimulation on PPF by immunofluorescence. Fetal liver-derived primary megakaryocyte cells were obtained from whole livers recovered from mouse fetuses between embryonic days 13 and 15. As shown in Figure 8D, after 5 days of culture in the presence of rmTPO, cell population was purified and enriched on mature megakaryocytes using a 2-step of 1.5%/3% BSA density gradient under gravity (1g) for 45 min at RT. The proportion of mature megakaryocytes in the enriched population was estimated to >60% (data not shown). Afterwards, we analyzed PPF in D-dencichine-stimulated fetal liver-derived primary megakaryocytes after their attachment to immobilized fibrinogen. We found that D-dencichine treatment distinctly promoted PPF characterized by generating large pseudopod-like structures that were elongated, thin, and branched to yield slender tubular projections in a dose-dependent manner compared with that in the control group (Figure 8E). This result suggested that D-dencichine regulated PPF, and the further analysis could be done to investigate this complicated process. Taken together, these results indicated that D-dencichine treatment could enhance megakaryocyte adhesion, migration, and PPF through ERK1/2 and Akt pathways.