Proteasome function is required for platelet production

The proteasome inhibiter bortezomib has been successfully used to treat patients with relapsed multiple myeloma; however, many of these patients become thrombocytopenic, and it is not clear how the proteasome influences platelet production. Here we determined that pharmacologic inhibition of proteasome activity blocks proplatelet formation in human and mouse megakaryocytes. We also found that megakaryocytes isolated from mice deficient for PSMC1, an essential subunit of the 26S proteasome, fail to produce proplatelets. Consistent with decreased proplatelet formation, mice lacking PSMC1 in platelets (Psmc1^fl/fl Pf4-Cre mice) exhibited severe thrombocytopenia and died shortly after birth. The failure to produce proplatelets in proteasome-inhibited megakaryocytes was due to upregulation and hyperactivation of the small GTPase, RhoA, rather than NF-κB, as has been previously suggested. Inhibition of RhoA or its downstream target, Rho-associated protein kinase (ROCK), restored megakaryocyte proplatelet formation in the setting of proteasome inhibition in vitro. Similarly, fasudil, a ROCK inhibitor used clinically to treat cerebral vasospasm, restored platelet counts in adult mice that were made thrombocytopenic by tamoxifen-induced suppression of proteasome activity in megakaryocytes and platelets (Psmc1^fl/fl Pdgf-Cre-ER mice). These results indicate that proteasome function is critical for thrombopoiesis, and suggest inhibition of RhoA signaling as a potential strategy to treat thrombocytopenia in bortezomib-treated multiple myeloma patients.

Thrombocytopenia (low platelet count) is observed in numerous diseases and can be life threatening due to bleeding complications. Bortezomib, a reversible chemotherapeutic inhibitor used to treat patients with relapsed multiple myeloma, often induces thrombocytopenia within a few days of therapy initiation (1–3). Bortezomib-induced thrombocytopenia is dose-limiting, and if severe, bortezomib is withheld (2, 3). Although the mechanisms by which bortezomib induces thrombocytopenia is not clear, its primary mode of action is inhibition of the proteasome. The clinical observation that platelet counts rise above pretherapy levels upon cessation of bortezomib treatment suggests that bortezomib affects thrombopoiesis (1, 3).

Like other cells, megakaryocytes and anucleate platelets possess proteasome activity (4, 5). While the specific functions of the proteasome in platelet precursors (e.g., megakaryocytes) is relatively unknown (6), there is evidence that bortezomib alters the function of platelets (7–11). It has also been hypothesized, but not proven, that bortezomib inhibits megakaryocyte development via nuclear factor κB (NF-κB) (12). The aim of the present work was to precisely define the roles of the proteasome in thrombopoiesis and, in doing so, determine whether bortezomib-induced thrombocytopenia can be reversed. Using a combination of pharmacologic and genetic tools, we showed that inhibition of proteasome activity in megakaryocytes blocks proplatelet formation. In addition, conditional deletion of proteasome activity in mouse megakaryocytes led to severe thrombocytopenia and postnatal death. Decreased thrombopoiesis in proteasome-inhibited mice was caused by accumulation and increased activity of RhoA, and inhibitors of the RhoA signaling pathway restored platelet production. These findings demonstrated that the megakaryocyte proteasome controls the final stages of platelet production and also provided a potential option for restoring platelet counts in thrombocytopenic patients treated with bortezomib.

Pharmacologic inhibition of the proteasome blocks platelet production. Due to its thrombocytopenic side effects, bortezomib is typically administered as a bolus twice weekly for 2 weeks (days 1, 4, 8, and 11), followed by a 10-day rest period (3). To ascertain the immediate effects of this inhibitor on platelets, mice were administered a clinically relevant dose (2 mg/kg body weight) of bortezomib, and platelet counts and proteasome activity were measured. Consistent with its well-known effect on platelet counts in patients, bortezomib induced a mild thrombocytopenia within 24 hours of treatment (Figure 1A). The thrombocytopenia was transient and temporally correlated with inhibition of proteasome activity in circulating platelets (Figure 1A).

Figure 1

Pharmacologic inhibition of the proteasome induces thrombocytopenia in mice by decreasing platelet production. (A) Mice were treated with a bolus of bortezomib (Bort) or vehicle (Veh), and platelet counts and platelet proteasome activity were measured at the indicated times. Data are mean ± SEM of 6 experiments. (B) Mouse platelets were labeled in vivo with Dylight 488, as described in Methods. In parallel, the mice were treated with a bolus of bortezomib or vehicle, and the percentage of labeled platelets was determined at 24, 48, and 96 hours after treatment. Data are mean ± SEM of 6 independent experiments. (C) Mouse platelets were depleted in the presence of bortezomib or its vehicle, as described in Methods. The percentage of platelets relative to baseline control (0 hours) is shown. Data are mean ± SEM of 5 independent experiments. Note that A–C are derived from separate experiments. *P < 0.05 vs. vehicle; ^#P < 0.05 vs. 0 hours.

To assess whether bortezomib-induced thrombocytopenia was due to accelerated clearance, we injected Dylight 488 conjugated to GPIBβ into the bloodstream to track the lifespan of platelets in vivo in the presence of bortezomib or its vehicle. As expected, the number of labeled platelets decreased over 96 hours as platelets were cleared from the circulation (13). However, the number of labeled platelets was similar between bortezomib and vehicle treatment groups at every time point tested (Figure 1B). This suggested that bortezomib did not induce platelet activation in the bloodstream, which would facilitate platelet clearance. Consistent with this notion, we found that bortezomib did not directly induce activation of integrin α_IIbβ₃ (Supplemental Figure 1; supplemental material available online with this article; doi:10.1172/JCI75247DS1), nor did it alter agonist-induced activation of integrin α_IIbβ₃ or surface expression of P-selectin in mouse platelets (Supplemental Figures 1 and 2). Similarly, bortezomib did not influence PAC-1 binding to human platelets in the presence or absence of agonist stimulation (Supplemental Figure 3 and data not shown).

Since acute administration of bortezomib did not shorten the lifespan of circulating platelets, we hypothesized that the bortezomib-induced thrombocytopenia was due to a decrease in platelet production. To test this, we depleted platelets with an antibody against GPIBα and then treated mice with bortezomib or its vehicle to determine whether bortezomib prevented platelet counts from rebounding. Platelet counts rebounded at a slower pace with bortezomib treatment than with the vehicle control (Figure 1C).

The data in Figure 1 suggested that bortezomib-induced thrombocytopenia was due to a defect in the formation of platelets from megakaryocytes. To examine this further, we determined whether inhibition of the proteasome with bortezomib blocked proplatelet formation in murine megakaryocytes. Bortezomib significantly decreased proplatelet formation in fetal liver–derived megakaryocytes (Figure 2A). Similar responses were observed in human megakaryocytes, and removal of bortezomib from the incubation media restored proplatelet formation (Figure 2B and data not shown). To confirm that this effect was specific to proteasome inhibition, megakaryocytes were treated with MG132 or lactacystin. Both proteasome inhibitors phenocopied the effects of bortezomib (Supplemental Figure 4 and data not shown). The inability to form proplatelets was accompanied by a notable increase in cell spreading on immobilized fibrinogen (Figure 2B), which indicates that the proteasome regulates key cytoskeletal proteins in megakaryocytes.

Figure 2

Pharmacologic inhibition of the proteasome blocks proplatelet formation in murine and human megakaryocytes. Mouse fetal liver–derived megakaryocytes (A) and human megakaryocytes (B) were pretreated with vehicle or bortezomib, and megakaryocytes producing proplatelets (PP) were examined. Shown are (A) representative transmission images and (B) representative confocal images with wheat germ agglutinin (WGA; red) and phalloidin (green) staining. Arrows denote proplatelet extensions. Also shown for each is the number of proplatelet-producing megakaryocytes relative to vehicle control. Data are mean ± SEM of 3 independent experiments. *P < 0.05 vs. vehicle. Scale bars: 100 μm (A); 50 μm (B).

Phenotypic consequences of proteasome inhibition are independent of NF-κB and integrin α_IIb. Bortezomib’s antitumor activity in multiple myeloma has been attributed to inhibition of NF-κB in plasma cells (14). Therefore, others have speculated that proteasome inhibitors may induce thrombocytopenia via the NF-κB signaling pathway (12). To test this hypothesis, we first treated megakaryocytes with bortezomib and examined the expression of nuclear factor of κ light polypeptide gene enhancer in B cells inhibitor, α (IκBα), which sequesters NF-κB in the cytoplasm until it is phosphorylated, ubiquitinated, and degraded (15). As expected, bortezomib increased the expression of IκBα in megakaryocytes (Figure 3A), which demonstrated that inhibition of the proteasome blocks the NF-κB signaling pathway. We also found that inhibition of IκB kinase with SC-514, which induces IκBα phosphorylation, increased IκBα protein levels in megakaryocytes (Figure 3A). However, unlike bortezomib, SC-514 did not halt proplatelet production (Figure 3B).

Figure 3

Proteasome-dependent formation of proplatelets in human megakaryocytes occurs independently of NF-κB. (A) Human megakaryocytes were treated with vehicle, bortezomib, or the NF-κB inhibitor SC-514. Shown are a representative Western blot for IκBα as well as IκBα expression levels, as measured by densitometry, relative to vehicle control. Data are mean ± SEM (n = 3). (B) Morphology of megakaryocytes treated with vehicle, bortezomib, or SC-514. Megakaryocytes were stained with WGA (red), phalloidin (green), and DAPI (blue). Arrows denote proplatelets. Images are representative of 3 independent experiments. Also shown is the number of proplatelet-producing megakaryocytes relative to vehicle control. Data are mean ± SEM of 3 independent experiments. Scale bars: 25 μm. *P < 0.05 vs. vehicle.

In addition to regulating NF-κB activity, Mitchell and colleagues previously demonstrated that the proteasome is capable of degrading pro–integrin α_IIb (5). Based on this published work, and our present finding that inhibition of the proteasome decreased the formation of proplatelets when megakaryocytes adhere to fibrinogen, we sought to determine whether bortezomib regulated the activity of integrin α_IIbβ₃ in megakaryocytes. Bortezomib did not alter the expression of mature integrin α_IIb protein, nor did it increase binding of soluble fibrinogen or PAC-1 to human megakaryocytes (Supplemental Figure 5 and data not shown). Bortezomib also had no effect on adherence of human megakaryocytes to fibrinogen (Supplemental Figure 6). Together, these data indicate that bortezomib does not directly block proplatelet formation through NF-κB– or integrin α_IIbβ₃–dependent mechanisms.

Phenotypic consequences of proteasome inhibition require RhoA. The changes in actin polymerization observed in megakaryocytes treated with proteasome inhibitors were reminiscent of cytoskeletal changes in endothelial cells that rely on the small GTPase RhoA (16). Indeed, we found that bortezomib increased total RhoA protein expression (Figure 4A). Bortezomib also increased RhoA-GTP activity and phosphorylation of myosin light chain (MLC) kinase, which is downstream of RhoA (Figure 4, A and B).

Figure 4

The proteasome regulates proplatelet formation through the RhoA signaling pathway. (A) Human megakaryocytes were treated with vehicle or bortezomib, and total RhoA and GTP-bound RhoA were measured. Shown are representative Western blots and expression of RhoA or RhoA-GTP, as measured by densitometry, relative to vehicle control. Data are mean ± SEM of 4 independent experiments. (B and C) Human megakaryocytes were treated with vehicle, bortezomib, or bortezomib plus Y27632. (B) Western blot for phospho-MLC. (C) Representative confocal images of human megakaryocytes stained with WGA (red) and phalloidin (green). Arrows denote proplatelets. Scale bar: 50 μm. Also shown is the number of proplatelet-producing megakaryocytes relative to vehicle control. Data are mean ± SEM of 3 independent experiments. (D) Representative transmission images of mouse bone marrow–derived megakaryocytes treated with vehicle, bortezomib, bortezomib plus Y27632, or bortezomib plus fasudil. Scale bar: 100 μm. Also shown is the number of proplatelet-producing megakaryocytes relative to vehicle control. Data are mean ± SEM of 3 independent experiments. *P < 0.05 vs. vehicle; ^#P < 0.05 vs. bortezomib alone.

RhoA-dependent signaling has been linked to the production of proplatelets (17, 18). Therefore, we treated human megakaryocytes with Y27632, a selective inhibitor of the Rho-associated protein kinase p160ROCK, or with C3 transferase, a direct RhoA inhibitor. Y27632 and C3 transferase rescued proplatelet formation in bortezomib-treated cells (Figure 4C and Supplemental Figure 7). This response was likely due to inhibition of downstream RhoA effectors, because Y27632 decreased phosphorylation of MLC kinase in the presence of bortezomib (Figure 4B). In agreement with the rescue of proplatelet formation observed in bortezomib-treated human megakaryocytes, mouse megakaryocytes treated with bortezomib plus Y27632 or with bortezomib plus fasudil, a more clinically relevant p160ROCK inhibitor, formed proplatelets (Figure 4D). These results in mouse megakaryocytes were similar to a recent report by Murai et al. (19).

Genetic deletion of the proteasome results in severe thrombocytopenia and death. To further dissect the role of the proteasome in thrombopoiesis, we focused on protease (prosome, macropain) 26S subunit, ATPase 1 (Psmc1; gene ID 19179) in mouse megakaryocytes and platelets. Psmc1 is an essential subunit of the 19S regulatory particle that is critical for ubiquitin-mediated protein degradation by the 26S proteasome complex (20–22). It is conserved at the protein level in human and mouse megakaryocytes (Supplemental Figure 8). mRNA for Psmc1 was also expressed in both species, although human megakaryocytes had lower levels of the transcript compared with mouse megakaryocytes (Supplemental Table 1).

Psmc1^fl/fl mice were crossed with platelet factor 4 Cre recombinase (Pf4-Cre) mice to disrupt proteasome activity in megakaryocytes and platelets. Psmc1^fl/fl Pf4-Cre mice had significantly reduced protein for PSMC1 in megakaryocytes, but not other tissues (Supplemental Figure 9). Ubiquinated proteins also accumulated in megakaryocytes from Psmc1^fl/fl Pf4-Cre mice (Supplemental Figure 10).

Despite a marked reduction in PSMC1 protein, the number of megakaryocytes from Psmc1^fl/fl Pf4-Cre mice in bone marrow or spleen was not reduced compared with Psmc1^fl/wt mice (Supplemental Figure 11). Unlike their littermate controls, however, Psmc1^fl/fl Pf4-Cre mice had severe thrombocytopenia at postnatal day 1 (P1), and the majority of Psmc1^fl/fl Pf4-Cre mice died before weaning (Figure 5, A and B). The reduction in platelet counts was more severe than in c-Mpl knockout pups at the same age (Supplemental Figure 12). In addition to reduced numbers of platelets, Psmc1^fl/fl Pf4-Cre mice had lower hematocrits than Psmc1^fl/wt mice, and bleeding was seen in the abdomen and limbs (Figure 5, C and D). Pathological signs of hemorrhage were also present in the bladder and testes of all animals and occasionally observed in the brain, lymph nodes, and intestines (Figure 5E and data not shown).

Figure 5

Genetic ablation of proteasome activity in megakaryocytes causes severe thrombocytopenia and postnatal death. (A) Platelet counts at P1 in Psmc1^fl/fl Pf4-Cre and Psmc1^fl/wt mice, expressed relative to Psmc1^fl/fl mice. Bars show mean ± SEM of 6 independent experiments. *P < 0.05 vs. Psmc1^fl/fl. (B) Mortality rates in Psmc1^fl/fl, Psmc1^fl/wt, and Psmc1^fl/fl Pf4-Cre mice at P1 and P21. Shown are ratios of expected versus observed genotypes, determined by χ² analysis, at P1 and P21 (n = 88). *P < 0.05 vs. P1, determined by χ² distribution table. (C) Hematocrits in Psmc1^fl/fl Pf4-Cre relative to Psmc1^fl/wt mice at P1. Data are mean ± SEM of 6 independent experiments. *P < 0.05 vs. Psmc1^fl/wt. (D) Left: Images of Psmc1^fl/wt and Psmc1^fl/fl Pf4-Cre mice at P1. Evidence of bleeding was observed in the abdominal region (arrow). Middle and right: Limbs of Psmc1^fl/wt and Psmc1^fl/fl Pf4-Cre mice. Hemorrhaging was observed in the limb of the Psmc1^fl/fl Pf4-Cre mouse (arrow). (E) Whereas P1 histological sections of Psmc1^fl/wt mice demonstrated normal histology, bleeding was observed in the bladder and testis of a Psmc1^fl/fl Pf4-Cre mouse. Boxed regions are shown at higher magnification at right. Scale bars: 100 μm.

Ultrastructure examination of megakaryocytes from Psmc1^fl/fl Pf4-Cre mice revealed less cytoplasm compared with megakaryocytes from Psmc1^fl/wt mice (Figure 6A). In addition, Psmc1^fl/fl Pf4-Cre megakaryocytes lacked demarcation membranes, which were readily visible in Psmc1^fl/wt megakaryocytes (Figure 6A). Similar to mouse megakaryocytes treated with bortezomib (Figure 2A), megakaryocytes from Psmc1^fl/fl Pf4-Cre mice failed to produce proplatelets (Figure 6B).

Figure 6

Platelet territories and proplatelets fail to form in PSMC1-deficient megakaryocytes. (A) Whereas Psmc1^fl/wt mouse megakaryocytes showed a large cytoplasmic region compared with the nucleus, those from a Psmc1^fl/fl Pf4-Cre mouse had less cytoplasm compared with the multilobed nucleus. Boxed regions are shown at higher magnification below, in which the demarcation membrane exhibited in the Psmc1^fl/wt megakaryocytes (arrowheads) was not observed in the Psmc1^fl/fl Pf4-Cre megakaryocyte. (B) Transmission images of megakaryocytes derived from Psmc1^fl/wt and Psmc1^fl/fl Pf4-Cre mice at P1. Proplatelet formation (arrows) was absent in Psmc1^fl/fl Pf4-Cre mice. Scale bars: 2 μm (A); 100 μm (B).

Inhibition of RhoA-dependent signaling prevents thrombocytopenia induced by genetic disruption of proteasome activity. As predicted from the pharmacological data, megakaryocytes from Psmc1^fl/fl Pf4-Cre mice expressed higher levels of total RhoA protein and RhoA-GTP (Figure 7A). Fasudil also rescued proplatelet formation in bone marrow–derived megakaryocytes from Psmc1^fl/fl Pf4-Cre mice (Figure 7B).

Figure 7

Genetic deletion of Psmc1 in megakaryocytes is associated with increased RhoA protein and activity. (A) Representative Western blot of total RhoA and RhoA-GTP in megakaryocytes derived from Psmc1^fl/wt and Psmc1^fl/fl Pf4-Cre mice at P1. Also shown is densitometry quantification relative to Psmc1^fl/wt control; for RhoA-GTP, megakaryocytes were isolated from 10 P1 mice, lysed, and then the lysates were pooled together for each pulldown experiment (see Methods). Data are mean ± SEM of 3 (total RhoA) and 2 (RhoA-GTP) experiments. *P < 0.05 vs. Psmc1^fl/wt. (B) Bone marrow–derived megakaryocytes from Psmc1^fl/wt and Psmc1^fl/fl Pf4-Cre mice were treated with vehicle or fasudil, and the number of proplatelet-producing megakaryocytes was quantified and expressed relative to Psmc1^fl/wt controls. Data are mean ± SEM of 3 independent experiments. *P < 0.05 vs. vehicle-treated Psmc1^fl/wt.

Next we generated inducible conditional knockouts by crossing Psmc1^fl/fl mice with platelet-derived growth factor–Cre estrogen receptor (Pdgf-Cre-ER) mice, which allowed for time-restricted deletion of Psmc1 in megakaryocytes and platelets after administration of the competitive estrogen receptor ligand tamoxifen. Although Pdgf is expressed by other cells besides megakaryocytes, Pdgf-Cre-ER mice were used because Pf4-Cre-ER mouse lines are not currently available. Like Psmc1^fl/fl Pf4-Cre mice (Figure 5, A and B), administration of tamoxifen to Psmc1^fl/fl Pdgf-Cre-ER mice at P1 resulted in thrombocytopenia and early postnatal mortality (Supplemental Figure 13, A and B). When tamoxifen was administered to adult Psmc1^fl/fl Pdgf-Cre-ER mice, platelet counts were reduced by approximately 50% after 6 days compared with Psmc1^fl/wt mice (Figure 8A). In the presence of fasudil, however, tamoxifen did not significantly decrease platelet counts in Psmc1^fl/fl Pdgf-Cre-ER mice (Figure 8A). Consistent with these rescue experiments, staining of megakaryocytes in crude bone marrow showed that the in vivo fasudil treatment rescued proplatelet formation (Figure 8B). These results are consistent with our in vitro findings that fasudil maintained proplatelet formation in bortezomib-treated megakaryocytes (Figure 4D).

Figure 8

Inhibition of RhoA signaling rescues platelet counts in adult mice in which proteasome activity is conditionally deleted. (A) Tamoxifen was administered to adult Psmc1^fl/wt and Psmc1^fl/fl Pdgf-Cre-ER mice, followed by treatment with fasudil or saline control (4 and 48 hours after tamoxifen). Shown are platelet counts at day 6 after tamoxifen administration relative to Psmc1^fl/wt controls treated with tamoxifen alone. Data are mean ± SEM of 9 experiments performed on independent mice. *P < 0.05 as indicated. No significant difference was observed between groups treated with tamoxifen plus fasudil. (B) Representative images of Dylight 488–positive megakaryocytes present in crude bone marrow isolated from Psmc1^fl/wt and Psmc1^fl/fl Pdgf-Cre-ER mice immediately after euthanasia. Bone marrow for these studies was isolated from a subset of the mice in A (n = 2 per treatment group). Scale bars: 100 μm.

In this study, we found that pharmacologic or genetic disruption of proteasome activity in megakaryocytes inhibits proplatelet formation. Pharmacologic inhibition was reversible in megakaryocytes treated in vitro with bortezomib, and thrombocytopenia was transient when bortezomib was administered as a bolus in vivo. When inhibition of proteasome activity was sustained, as was the case with genetic deletion of Psmc1 in megakaryocytes and platelets, megakaryocytes did not form proplatelets, and Psmc1^fl/fl Pf4-Cre mice had severe thrombocytopenia. Mice with genetic ablation of Psmc1 in megakaryocytes and platelets also died shortly after birth. Taken together, these data provide compelling evidence that the proteasome is critically involved in thrombopoiesis. The data also offer a strong explanation as to why multiple myeloma patients require cyclic treatment regimes of bortezomib in order to tolerate the drug (3, 12).

Our findings provide definitive proof that the megakaryocyte proteasome is required for the final stages of platelet production. Evidence for this is 2-fold: first, pharmacologic inhibition of proteasome activity in late-stage human or mouse megakaryocytes significantly blunted proplatelet formation; and second, platelet production was significantly reduced in Psmc1^fl/fl Pf4-Cre mice, in which genetic deletion of Psmc1 does not occur until megakaryocytes express platelet factor 4, which activates the Cre recombinase (23). In addition to regulating thrombopoiesis, others have shown that the proteasome is important for the proliferation of megakaryocyte precursors (24) and the degradation of cyclin B and pro–integrin α_IIb in megakaryocytes (5, 25).

Like their parent megakaryocytes, anucleate platelets also possess proteasome activity (4, 10), and several groups have demonstrated that pharmacologic inhibition of the proteasome regulates platelet function (8–11, 26). Under the conditions of our experiments, bortezomib did not affect indices of platelet activation in mouse or human platelets that included activation of integrin α_IIbβ₃ and translocation of P-selectin to the surface of platelets. However, similar to Gupta and coworkers (8), we observed that bortezomib reduced the aggregation of human platelets when low concentrations of thrombin were used as the agonist (Supplemental Figure 14). Although more work is needed, results generated by multiple independent groups strongly indicate that protein degradation systems regulate platelet function (6).

Other groups have shown that pharmacologic inhibition regulates the function of platelets ex vivo (8–11, 26), but bortezomib did not accelerate the clearance of labeled platelets under the conditions of our present studies. Our results contrasted those of Nayak and colleagues (7), who showed that pharmacologic inhibition of the proteasome reduced the half-life of platelets in mice. One potential explanation for these discordant findings is that the bolus dose of bortezomib used in our studies only produced a mild thrombocytopenia and did not completely abolish platelet proteasome activity (Figure 1A). Although Nayak’s group did not measure cellular proteasome activity (7), it is possible that they achieved more efficient pharmacologic inhibition of the proteasome in platelets and other vascular cells. Different routes of drug administration and types/concentrations of proteasome inhibitors between the studies may also explain the divergent results. Further studies are needed to resolve the in vivo pharmacology of proteasome inhibition and its relation to thrombocytopenia. However, our present studies clearly showed that platelet counts rebounded at a slower pace in mice subjected to platelet depletion in the presence of bortezomib. These data, in combination with the severe thrombocytopenia we observed in Psmc1^fl/fl Pf4-Cre mice, demonstrated that the proteasome directly modulates platelet production.

We found that thrombocytopenia was more severe in Psmc1^fl/fl Pf4-Cre mice compared with c-Mpl knockout mice (Supplemental Figure 12), which have normal life expectancies (27). Consistent with a marked reduction in platelet counts, Psmc1^fl/fl Pf4-Cre mice had low hematocrits and hemorrhaging in the abdominal region. Occasional hemorrhaging was also observed in the brain, lymph nodes, and intestines. This suggests that severe thrombocytopenia is the primary driver of postnatal death in Psmc1^fl/fl Pf4-Cre mice. Hemorrhaging in Psmc1^fl/fl Pf4-Cre mice may occur because platelet numbers are simply too low to prevent bleeding. Alternatively, insufficient platelet counts in Psmc1^fl/fl Pf4-Cre mice may result in abnormal vascular development or blood/lymphatic vessel separation, which could lead to excessive bleeding (28). In this regard, several groups have shown that platelet C-type lectin-like receptor 2 (CLEC-2) receptors regulate lymphatic vascular development, and, like Psmc1^fl/fl Pf4-Cre mice (29–33), platelet-specific knockout of CLEC-2 results in postnatal lethality (32). It should also be noted that a very low threshold of platelet function sufficiently maintains vascular function (27, 33, 34), raising the possibility that Psmc1^fl/fl Pf4-Cre mice produce dysfunctional platelets that are incapable of maintaining vascular integrity. Indeed, recent studies have demonstrated that immune-type receptors in platelets are critical for the prevention of inflammation-induced hemorrhage (35). Thus, it is entirely possible that in addition to being reduced in number, platelets from Psmc1^fl/fl Pf4-Cre mice express an abnormal repertoire of proteins resulting in platelet dysfunction.

Studies in megakaryocytes revealed that genetic or pharmacologic interruption of proteasome activity led to accumulation of IκBα and RhoA. Although both proteins were upregulated, we found that the final stages of proplatelet formation required RhoA signaling rather than inhibition of NF-κB, as previously suggested (12). The inability to sprout proplatelets resembled studies in neurons, in which acute inhibition of the proteasome blocks activity-dependent growth of new dendritic spines. It is not known what proteins are degraded by the proteasome in order to stimulate new spine growth; however, inactivation of RhoA leads to neurite outgrowth (36, 37). This suggests that, similar to megakaryocyte proplatelet formation, the proteasome may control neuronal outgrowth by degrading RhoA. Moreover, RhoA signaling has been shown to maintain normal megakaryocyte development, which is critical for platelet production (18).

Malfunction of the proteasome in human diseases may lead to aberrant platelet production or abnormal platelet generation. Disruption of proteasome activity could occur at multiple checkpoints, since human megakaryocytes expressed the full repertoire of proteasome components at the mRNA level (Supplemental Table 1). Identifying the complete portfolio of target proteins degraded by the proteasome in megakaryocytes will shed additional light on the mechanisms that control thrombopoiesis and the phenotype of platelets as they enter the circulation. Understanding the functions of the proteasome in platelets, which is active and capable of degrading proteins (7–9), also requires further investigation. From an immediate perspective, our present findings demonstrated that bortezomib directly inhibits proteasome activity in megakaryocytes and thereby decreases platelet production. Our findings also established fasudil as a potential treatment for preventing and/or reversing bortezomib-induced thrombocytopenia in multiple myeloma patients. Additionally, inhibitors of the RhoA signaling pathway may have efficacy in the treatment of other thrombocytopenic disorders caused by abnormal platelet production, especially if the disease is driven by proteasome-dependent mechanisms.

View Supplemental data

We thank Diana Lim for preparation of the figures, critical comments, and consultation regarding responsible and effective display of the images. We also thank Guy Zimmerman for helpful discussions regarding the roles of the proteasome in platelets and megakaryocytes. This work was funded by NIH grants HL066277 and HL112311 (to A.S. Weyrich), HL112311, HL084516, and AR064788 (to D.Y. Li), GM103806 (to J.W. Rowley), and HL68130 (to J.E. Italiano Jr.); by the American Heart Association (13POST13930019 to K.R. Machlus; 11POST7290019 to R.A. Campbell; 0625098Y and 09BG1A 2250381 to H. Schwertz); and by the Canadian Institutes of Health Research (MOP-259952 to W.H.A. Kahr). In addition, H. Schwertz was funded by a Lichtenberg-Professorship from the Volkswagen Foundation.

Address correspondence to: Dean Li or Andrew S. Weyrich, Eccles Institute of Human Genetics, 15 North 2030 East, Bldg. 533, Rm. 4150, Salt Lake City, Utah 84112, USA. Phone: 801.585.0950; E-mail: dean.li@u2m2.utah.edu (D. Li), andy.weyrich@u2m2.utah.edu (A.S. Weyrich).

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J Clin Invest. 2014;124(9):3757–3766. doi:10.1172/JCI75247.

1. Adams J, et al. Proteasome inhibitors: a novel class of potent and effective antitumor agents. Cancer Res. 1999;59(11):2615–2622.
   View this article via: PubMed Google Scholar
2. Richardson PG, et al. A phase 2 study of bortezomib in relapsed, refractory myeloma. N Engl J Med. 2003;348(26):2609–2617.
   View this article via: PubMed CrossRef Google Scholar
3. Field-Smith A, Morgan GJ, Davies FE. Bortezomib (Velcadetrade mark) in the treatment of multiple myeloma. Ther Clin Risk Manag. 2006;2(3):271–279.
   View this article via: PubMed CrossRef Google Scholar
4. Ostrowska H, Ostrowska JK, Worowski K, Radziwon P. Human platelet 20S proteasome: inhibition of its chymotrypsin-like activity identification of the proteasome activator PA28. A preliminary report. Platelets. 2003;14(3):151–157.
   View this article via: PubMed CrossRef Google Scholar
5. Mitchell WB, Li J, French DL, Coller BS. αIIbβ3 biogenesis is controlled by engagement of αIIb in the calnexin cycle via the N15-linked glycan. Blood. 2006;107(7):2713–2719.
   View this article via: PubMed CrossRef Google Scholar
6. Kraemer BF, Weyrich AS, Lindemann S. Protein degradation systems in platelets. Thromb Haemost. 2013;110(5):920–924.
   View this article via: PubMed CrossRef Google Scholar
7. Nayak MK, Kulkarni PP, Dash D. Regulatory role of proteasome in determination of platelet life span. J Biol Chem. 2013;288(10):6826–6834.
   View this article via: PubMed CrossRef Google Scholar
8. Gupta N, Li W, Willard B, Silverstein RL, McIntyre TM. Proteasome proteolysis supports stimulated platelet function and thrombosis. Arterioscler Thromb Vasc Biol. 2014;34(1):160–168.
   View this article via: PubMed CrossRef Google Scholar
9. Kumari S, Dash D. Regulation of β-catenin stabilization in human platelets. Biochimie. 2013;95(6):1252–1257.
   View this article via: PubMed CrossRef Google Scholar
10. Nayak MK, Kumar K, Dash D. Regulation of proteasome activity in activated human platelets. Cell Calcium. 2011;49(4):226–232.
    View this article via: PubMed CrossRef Google Scholar
11. Avcu F, Ural AU, Cetin T, Nevruz O. Effects of bortezomib on platelet aggregation and ATP release in human platelets, in vitro. Thromb Res. 2008;121(4):567–571.
    View this article via: PubMed CrossRef Google Scholar
12. Lonial S, et al. Risk factors and kinetics of thrombocytopenia associated with bortezomib for relapsed, refractory multiple myeloma. Blood. 2005;106(12):3777–3784.
    View this article via: PubMed CrossRef Google Scholar
13. Mason KD, et al. Programmed anuclear cell death delimits platelet life span. Cell. 2007;128(6):1173–1186.
    View this article via: PubMed CrossRef Google Scholar
14. McConkey DJ. Bortezomib paradigm shift in myeloma. Blood. 2009;114(5):931–932.
    View this article via: PubMed CrossRef Google Scholar
15. Spinelli SL, et al. Platelets and megakaryocytes contain functional nuclear factor-kappaB. Arterioscler Thromb Vasc Biol. 2010;30(3):591–598.
    View this article via: PubMed CrossRef Google Scholar
16. Whitehead KJ, et al. The cerebral cavernous malformation signaling pathway promotes vascular integrity via Rho GTPases. Nat Med. 2009;15(2):177–184.
    View this article via: PubMed CrossRef Google Scholar
17. Gobbi G, et al. Proplatelet generation in the mouse requires PKCepsilon-dependent RhoA inhibition. Blood. 2013;122(7):1305–1311.
    View this article via: PubMed CrossRef Google Scholar
18. Suzuki A, et al. RhoA is essential for maintaining normal megakaryocyte ploidy and platelet generation. PLoS One. 2013;8(7):e69315.
    View this article via: PubMed CrossRef Google Scholar
19. Murai K, et al. Bortezomib induces thrombocytopenia by the inhibition of proplatelet formation of megakaryocytes [published online ahead of print April 21, 2014]. Eur J Haematol. doi:10.1111/ejh.12342.
    View this article via: PubMed CrossRef Google Scholar
20. Rubin DM, Glickman MH, Larsen CN, Dhruvakumar S, Finley D. Active site mutants in the six regulatory particle ATPases reveal multiple roles for ATP in the proteasome. EMBO J. 1998;17(17):4909–4919.
    View this article via: PubMed CrossRef Google Scholar
21. Kohler A, Cascio P, Leggett DS, Woo KM, Goldberg AL, Finley D. The axial channel of the proteasome core particle is gated by the Rpt2 ATPase controls both substrate entry product release. Mol Cell. 2001;7(6):1143–1152.
    View this article via: PubMed CrossRef Google Scholar
22. Smith DM, Fraga H, Reis C, Kafri G, Goldberg AL. ATP binds to proteasomal ATPases in pairs with distinct functional effects, implying an ordered reaction cycle. Cell. 2011;144(4):526–538.
    View this article via: PubMed CrossRef Google Scholar
23. Tiedt R, Schomber T, Hao-Shen H, Skoda RC. Pf4-Cre transgenic mice allow the generation of lineage-restricted gene knockouts for studying megakaryocyte platelet function in vivo. Blood. 2007;109(4):1503–1506.
    View this article via: PubMed CrossRef Google Scholar
24. Galimberti S, et al. PS-341 (Bortezomib) inhibits proliferation induces apoptosis of megakaryoblastic MO7-e cells. Leuk Res. 2008;32(1):103–112.
    View this article via: PubMed CrossRef Google Scholar
25. Zhang Y, Wang Z, Liu DX, Pagano M, Ravid K. Ubiquitin-dependent degradation of cyclin B is accelerated in polyploid megakaryocytes. J Biol Chem. 1998;273(3):1387–1392.
    View this article via: PubMed CrossRef Google Scholar
26. Necchi V, et al. Ubiquitin/proteasome-rich particulate cytoplasmic structures (PaCSs) in the platelets and megakaryocytes of ANKRD26-related thrombo-cytopenia. Thromb Haemost. 2012;109(2):263–271.
    View this article via: PubMed CrossRef Google Scholar
27. Murone M, Carpenter DA, de Sauvage FJ. Hematopoietic deficiencies in c-mpl and TPO knockout mice. Stem Cells. 1998;16(1):1–6.
    View this article via: PubMed CrossRef Google Scholar
28. Bertozzi CC, Hess PR, Kahn ML. Platelets: covert regulators of lymphatic development. Arterioscler Thromb Vasc Biol. 2010;30(12):2368–2371.
    View this article via: PubMed CrossRef Google Scholar
29. Benezech C, et al. CLEC-2 is required for development maintenance of lymph nodes. Blood. 2014;123(20):3200–3207.
    View this article via: PubMed CrossRef Google Scholar
30. Herzog BH, et al. Podoplanin maintains high endothelial venule integrity by interacting with platelet CLEC-2. Nature. 2013;502(7469):105–109.
    View this article via: PubMed CrossRef Google Scholar
31. Osada M, et al. Platelet activation receptor CLEC-2 regulates blood/lymphatic vessel separation by inhibiting proliferation, migration, tube formation of lymphatic endothelial cells. J Biol Chem. 2012;287(26):22241–22252.
    View this article via: PubMed CrossRef Google Scholar
32. Finney BA, et al. CLEC-2 and Syk in the megakaryocytic/platelet lineage are essential for development. Blood. 2012;119(7):1747–1756.
    View this article via: PubMed CrossRef Google Scholar
33. Bertozzi CC, et al. Platelets regulate lymphatic vascular development through CLEC-2-SLP-76 signaling. Blood. 2010;116(4):661–670.
    View this article via: PubMed CrossRef Google Scholar
34. Levin J, et al. Pathophysiology of thrombocytopenia and anemia in mice lacking transcription factor NF-E2. Blood. 1999;94(9):3037–3047.
    View this article via: PubMed Google Scholar
35. Boulaftali Y, et al. Platelet ITAM signaling is critical for vascular integrity in inflammation. J Clin Invest. 2013;123(2):908–916.
    View this article via: JCI PubMed CrossRef Google Scholar
36. Hamilton AM, et al. Activity-dependent growth of new dendritic spines is regulated by the proteasome. Neuron. 2012;74(6):1023–1030.
    View this article via: PubMed CrossRef Google Scholar
37. Jeon CY, et al. Control of neurite outgrowth by RhoA inactivation. J Neurochem. 2012;120(5):684–698.
    View this article via: PubMed CrossRef Google Scholar
38. Foulks JM, et al. PAF-acetylhydrolase expressed during megakaryocyte differentiation inactivates PAF-like lipids. Blood. 2009;113(26):6699–6706.
    View this article via: PubMed CrossRef Google Scholar
39. Denis MM, et al. Escaping the nuclear confines: signal-dependent pre-mRNA splicing in anucleate platelets. Cell. 2005;122(3):379–391.
    View this article via: PubMed CrossRef Google Scholar
40. Thon JN, Italiano JE. Visualization and manipulation of the platelet and megakaryocyte cytoskeleton. Methods Mol Biol. 2012;788:109–125.
    View this article via: PubMed CrossRef Google Scholar
41. Mazharian A, Watson SP, Severin S. Critical role for ERK1/2 in bone marrow fetal liver-derived primary megakaryocyte differentiation, motility, proplatelet formation. Exp Hematol. 2009;37(10):1238–1249.
    View this article via: PubMed CrossRef Google Scholar
42. Thon JN, et al. Cytoskeletal mechanics of proplatelet maturation and platelet release. J Cell Biol. 2010;191(4):861–874.
    View this article via: PubMed CrossRef Google Scholar
43. Cecchetti L, Tolley ND, Michetti N, Bury L, Weyrich AS, Gresele P. Megakaryocytes differentially sort mRNAs for matrix metalloproteinases and their inhibitors into platelets: a mechanism for regulating synthetic events. Blood. 2011;118(7):1903–1911.
    View this article via: PubMed CrossRef Google Scholar
44. Kahr WH, et al. Mutations in NBEAL2, encoding a BEACH protein, cause gray platelet syndrome. Nat Genet. 2011;43(8):738–740.
    View this article via: PubMed CrossRef Google Scholar
45. Rowley JW, et al. Genome-wide RNA-seq analysis of human and mouse platelet transcriptomes. Blood. 2011;118(14):e101–e111.
    View this article via: PubMed CrossRef Google Scholar
46. Nix DA, Courdy SJ, Boucher KM. Empirical methods for controlling false positives and estimating confidence in ChIP-Seq peaks. BMC Bioinformatics. 2008;9.
    View this article via: PubMed Google Scholar
47. Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 2002;30(1):207–210.
    View this article via: PubMed CrossRef Google Scholar
48. Josefsson EC, White MJ, Dowling MR, Kile BT. Platelet life span and apoptosis. Methods Mol Biol. 2012;788:59–71.
    View this article via: PubMed CrossRef Google Scholar
49. Liu ZJ, et al. Expansion of the neonatal platelet mass is achieved via an extension of platelet lifespan. Blood. 2014;123(22):3381–3389.
    View this article via: PubMed CrossRef Google Scholar
50. Kahr WH, et al. Abnormal megakaryocyte development and platelet function in Nbeal2(–/–) mice. Blood. 2013;122(19):3349–3358.
    View this article via: PubMed CrossRef Google Scholar
51. Holly SP, et al. Chemoproteomic discovery of AADACL1 as a regulator of human platelet activation. Chem Biol. 2013;20(9):1125–1134.
    View this article via: PubMed CrossRef Google Scholar
52. Ramskold D, Wang ET, Burge CB, Sandberg R. An abundance of ubiquitously expressed genes revealed by tissue transcriptome sequence data. PLoS Comput Biol. 2009;5(12):e1000598.
    View this article via: PubMed CrossRef Google Scholar