SEARCH

SEARCH BY CITATION

Keywords:

  • adipocyte;
  • HIV protease inhibitors;
  • oxidative stress;
  • osteoblast;
  • senescence;
  • statin

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

HIV-infected patients receiving antiretroviral therapy present an increased prevalence of age-related comorbidities, including osteoporosis. HIV protease inhibitors (PIs) have been suspected to participate to bone loss, but the mechanisms involved are unknown. In endothelial cells, some PIs have been shown to induce the accumulation of farnesylated prelamin-A, a biomarker of cell aging leading to cell senescence. Herein, we hypothesized that these PIs could induce premature aging of osteoblast precursors, human bone marrow mesenchymal stem cells (MSCs), and affect their capacity to differentiate into osteoblasts. Senescence was studied in proliferating human MSCs after a 30-day exposure to atazanavir and lopinavir with or without ritonavir. When compared to untreated cells, PI-treated MSCs had a reduced proliferative capacity that worsened with increasing passages. PI treatment led to increased oxidative stress and expression of senescence markers, including prelamin-A. Pravastatin, which blocks prelamin-A farnesylation, prevented PI-induced senescence and oxidative stress, while treatment with antioxidants partly reversed these effects. Moreover, senescent MSCs presented a decreased osteoblastic potential, which was restored by pravastatin treatment. Because age-related bone loss is associated with increased bone marrow fat, we also evaluated the capacity of PI-treated MSCs to differentiate into adipocyte. We observed an altered adipocyte differentiation in PI-treated MSCs that was reverted by pravastatin. We have shown that some PIs alter osteoblast formation by affecting their differentiation potential in association with altered senescence in MSCs, with a beneficial effect of statin. These data corroborate the clinical observations and allow new insight into pathophysiological mechanisms of PI-induced bone loss in HIV-infected patients.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

Treatment of HIV-infected patients with combination antiretroviral therapy (cART) has led to a significant reduction in morbidity and mortality. However, despite unquestioned success, these patients present an increased risk of a number of ‘non-AIDS’-related complications (Capeau, 2011). These complications, which include cardiovascular, liver, brain and bone diseases, are similar to those observed among the elderly. These observations led to propose that HIV-infected persons suffer from accelerated or amplified aging (Capeau, 2011). The occurrence of these complications was proposed to result from both HIV infection and antiretroviral treatment, in addition to aging and immune dysfunction. Regarding adipose tissue, alterations are characterized by a lipodystrophy syndrome with subcutaneous lipoatrophy and visceral fat hypertrophy (Mallon et al., 2003; Brown et al., 2004; Caron-Debarle et al., 2010). Bone disorders are mainly characterized by reduced bone mineral density (BMD), leading to the development of osteopenia and osteoporosis (Stone et al., 2010).

Osteoporosis is a bone disorder characterized by a severe loss of bone mass and consequent reduction in bone strength, which in turn leads to an increased risk of fractures. Osteopenia is defined by a lesser reduction in bone mass and precedes osteoporosis (Mazziotti et al., 2010). Although a certain degree of bone weakening with age is considered normal, several clinical studies have shown that there is an increased incidence of reduced BMD, osteopenia and osteoporosis in HIV patients in general and in those receiving cART in particular (Brown & Qaqish, 2006; Powderly, 2012). This increased risk of premature osteoporosis seems to be linked to both cART and HIV infection itself (Stone et al., 2010). Brown's meta-analysis showed that among HIV-infected patients, 67% had osteopenia, and 15%, osteoporosis. The effect on bone was striking because a vast majority were young men under the age of 40 (Brown & Qaqish, 2006). Low body weight is a major contributor to low BMD in HIV infection. Moreover, cART initiation has been associated with decreased BMD (Duvivier et al., 2009). Several studies suggest that the prevalence of reduced BMD in cross-sectional studies of patients receiving cART ranges from 20% to 50%, with rates of osteoporosis up to 20% (Stone et al., 2010; Powderly, 2012). Another meta-analysis showed that in HIV-infected patients on cART at baseline, BMD remained stable, whereas in patients initiating cART, BMD declined by 2–4% regardless of ART regimens after 2 years (Bolland et al., 2011).

The exact role of specific agents, including the HIV aspartyl protease inhibitors (PIs), in the pathogenesis of bone loss remains unknown. PIs belong to a potent class of antiretrovirals and are commonly used as first-line therapy in addition to nucleoside analogue reverse transcriptase inhibitors (NRTIs). Ritonavir is not used for its own antiviral activity, but as a booster of other PIs. Existing data demonstrate significantly lower BMD with PI initiation (Duvivier et al., 2009), suggesting that PIs are indeed important players in decreased BMD. More specifically, atazanavir (ATV) and lopinavir (LPV) boosted with ritonavir (LPV/r and ATV/r), which are frequently used PIs in clinical practice, have been associated with bone loss. This has been shown in numerous studies for LPV/r (Grund et al., 2009; Hansen et al., 2011) and also for ATV/r (Grund et al., 2009).

Under physiological conditions, the balance between osteoclast-mediated bone resorption and osteoblast-mediated bone deposition controls bone remodelling. Within bone marrow, mesenchymal stem cells (MSCs) are pluripotent precursors, capable of differentiating into several cell types, including osteoblasts and adipocytes (Kim et al., 2012). To achieve osteogenic differentiation, MSCs proceed through a number of functional stages including proliferation, matrix maturation and mineralization. It has been hypothesized that a reciprocal relationship exists in the bone marrow cavity, where production of adipocytes from MSCs is at the expense of osteoblasts (Verma et al., 2002; Kim et al., 2012). In vitro studies have demonstrated the balance between the transcription factors RUNX2 (Runt-related transcription factor 2), driving the pro-osteogenic phenotype, and peroxisome proliferator-activated receptor (PPARγ), the pro-adipogenic one (Nuttall & Gimble, 2004). Therefore, the respective activities of PPARγ and RUNX2 are key determinants of the relationship between fat and bone. Numerous clinical data support this idea, such as increased adipocyte content of aging bone (Verma et al., 2002; Nuttall & Gimble, 2004). Initiation of cART affects whole body composition of HIV-infected patients with increased fat mass and decreased BMD. The role of some PIs in cART-induced lipodystrophy has been proposed according to in vitro studies and in vivo observations (Caron-Debarle et al., 2010). Among HIV-infected patients, lipodystrophy, and in particular increased visceral abdominal fat, is significantly and independently associated with reduced BMD (Huang et al., 2001). Importantly, treatment with ATV/r and LPV/r is associated with decreased BMD and alterations in adipose tissue storage/accumulation (Carr et al., 1999; McComsey et al., 2011).

We have previously shown that treatment of endothelial cells with RTV or LPV induces cellular senescence, characterized by a decreased cell proliferation and the accumulation of the pro-senescence protein prelamin-A (Lefevre et al., 2010). The post-translational processing of prelamin-A to mature lamin A requires a farnesylation step and a cleavage by the zinc metallopeptidase ZMP-STE24. The inhibition of ZMP-STE24 by RTV and LPV results in permanently farnesylated prelamin-A, which has been shown to induce cell senescence (Caron et al., 2007; Coffinier et al., 2007). LPV and, to a lesser extent, ATV can inhibit prelamin-A maturation (Liu et al., 2010). Moreover, blocking the farnesylation of prelamin-A, using statins, has been shown to ameliorate cellular senescence phenotype of accelerated aging (Capell et al., 2005). Few studies have examined the direct effect of PIs on osteoblasts and more specifically on MSCs. In vitro experiments demonstrated that a subset of PIs could inhibit both osteogenic and adipogenic differentiation (Jain & Lenhard, 2002).

In the present study, we hypothesized that PIs could induce premature senescence of osteoblast precursor stem cells, namely MSCs, which in turn could be involved in osteoblast depletion and bone loss. We investigated whether frequently used PIs, atazanavir (ATV), lopinavir (LPV) alone or associated with ritonavir (ATV/r and LPV/r), and ritonavir alone (RTV), were capable of inducing premature senescence in MSCs. We then examined the consequences of PI-induced senescence on MSCs fate evaluated by their osteoblastic and adipogenic potential. The data presented in this study show that PIs can induce cellular senescence of MSCs and alter both osteoblastic and adipogenic potential. Moreover, while antioxidant molecules only partially reversed the effect of PIs, pravastatin, used as a farnesyl synthesis inhibitor, prevented PI-induced senescence and restored adipocyte and osteoblast differentiation capacities. These findings offer new insights on the role of MSCs in PI-associated bone loss.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

Chronic exposure to ATV or LPV induces senescence and oxidative stress in proliferating MSCs

We first evaluated whether chronic PI exposure affected in vitro cell proliferation capacities. Human bone marrow mesenchymal stem cells (MSCs) treated with LPV and ATV alone or in association with RTV (LPV/r and ATV/r) displayed a reduced proliferative activity that worsened with increasing cellular passages, but cell survival remained unaffected (data not shown). We observed a progressive decline in the population doubling level (PDL) between passages 3 and 9 (30 days of treatment) in ATV- and LPV-treated cells, whereas PDL did not vary in control cells (DMSO-treated) and RTV-treated cells (Fig. 1A). Although RTV did not exert an effect on cell proliferation on its own, we observed an additive effect of RTV on ATV- and LPV-treated cells on day 30 (Fig. 1B).

image

Figure 1. Protease inhibitors (PIs) trigger senescence and oxidative stress in MSCs. The population doubling level (PDL) was calculated as stated in Experimental procedures. (A) Mean PDL values (± SEM) were determined at the indicated days post-PI treatment and (B) at day 30 post-PI treatment. SA-β-galactosidase activity was assessed by (C) the total % of SA-β-galactosidase cells at pH6 or by (D) the ratio of pH6- to pH4-positive staining at the indicated days post-PI treatment and expressed as % of control cells. (E) Representative micrographs of SA-β-galactosidase positive cells are shown. Whole cell lysates were extracted from MSCs at day 30 of PI treatment and analysed by immunoblotting. Representative immunoblots of cell cycle arrest markers P21, P16 and ERK1/2 (loading control) (F) and of prelamin-A, Lamin A/C and ERK1/2 (G) are shown. Reactive oxygen species (ROS) production was assessed at the indicated days post-PI treatment by (H) the reduction in nitroblue tetrazolium (NBT) and by (I) the oxidation of CM-H2DCFDA and expressed as % of control cells. All experiments were performed in duplicate or triplicate in MSCs isolated from 3 to 4 different bone marrow donors. *P < 0.05, ** P < 0.01*** P < 0.001 vs. control cells.

Download figure to PowerPoint

In control cells, the percentage of senescent cells (senescence-associated β-galactosidase positive) was about 2–3%. The percentage of senescent cells was similar in RTV-treated cells. Conversely, it reached 8.8% in ATV/r-treated and 16.3% in LPV/r-treated cells after 30 days (Fig. 1C–D). LPV/r and, to a lesser extent, ATV/r chronic exposure led to a progressive increase in SA-β-galactosidase activity (Fig. 1D–E). Thus, PI-treated MSCs presented an early onset on cell senescence (2- to 4-fold increase on day 30) in both ATV- and LPV-treated cells (Fig. 1E). In accordance with decreased cell proliferation and cellular senescence, long-term ATV and LPV exposure also increased the protein expression of cell cycle arrest markers p16INK4 and p21WAF1 upon 30 days of treatment (Fig. 1F) when compared to control and RTV-treated cells. LPV and ATV with or without RTV induced the accumulation of prelamin-A, on day 30 (Fig. 1G), while prelamin-A expression was almost undetectable in control and RTV-treated cells. In parallel, we observed a decreased protein expression of mature Lamin A/C, as an increased amount of unprocessed prelamin-A is often balanced by a decreased amount of matured Lamin A (Liu et al., 2010).

Finally, long-term ATV and LPV exposure markedly increased reactive oxygen species (ROS) production when compared to control and RTV-treated cells. Starting from day 10 post-PI treatment for LPV/r and from day 20 for ATV/r, we observed a significant increase in ROS production (up to 4-fold in LPV/r-treated cells on day 30), as quantified by the reduction in nitroblue tetrazolium (oxidase substrate) (Fig. 1H) and by the oxidation of CM-H2DCFDA by ROS (Fig. 1I). Parallel to this increase, we observed a decreased superoxide dismutase (SOD) activity by up to 45% in ATV/r-treated and 75% in LPV/r-treated cells (Fig. S1A).

Potent antioxidant treatment partially prevents PI-induced decline of MSCs proliferation

We performed experiments in which MSCs were treated with different antioxidant molecules. We first used NAC (N-acetyl cysteine) over a short period of time (72 h) on MSCs treated for 15 days with LPV/r and ATV/r. As shown in Fig. 2A, on day 20, NAC was able to significantly decrease ROS production, but this effect was transient and no longer seen upon day 30. Under these short-term conditions, NAC treatment failed to improve cell proliferation (Fig. 2B). We compared the differences in PDL between day 20 and day 30 by the means of PDL ratios (Fig. 2C). This representation allows to evaluate the impact of antioxidants and to show that PDL ratio remained unchanged when MSCs were treated with NAC (Fig. 2C). We then studied the effect of antioxidant molecules on a longer term, and cells were incubated for the last 15 days with a combination of Trolox, an antioxidant derivative of vitamin E and reduced glutathione (GSH). This association did not affect MTT survival test, but increased ROS production in control cells (Fig. 2D). However, importantly, the combination of Trolox and GSH did decrease ROS production in ATV/r and LPV/r cells at days 20 and 30 (Fig. 2D) although it failed to increase SOD activity (Fig. S1B). Even if this antioxidant treatment induced a slight decrease in cell proliferation on day 20 for all conditions, thereafter, it prevented the ongoing deleterious effect of PIs on the decline of cell proliferation, stressing the role of oxidative stress in the PI-induced reduced cell proliferation (Fig. 2E). Indeed, when comparing the differences in PDL between day 20 and day 30, PDL ratio became positive when treated by GSH/Trolox (Fig. 2F). Thus, GSH/Trolox, despite an early deleterious effect, was able to partially rescue PDL levels in PI-treated MSCs. These results indicate that a strong and prolonged antioxidant treatment was able to partially prevent the effect of PIs. However, none of these treatments were able to normalize SOD activity in PI-treated cells (Fig. S1B–C). Overall, these results suggest that increased oxidative stress during the onset of cellular senescence participated to PI-induced altered cell proliferation.

image

Figure 2. Antioxidant treatment prevents oxidative stress and partially prevents PI-induced senescence of MSCs. NAC was added upon day 15, for 72 h, and a combination of GSH and Trolox was added for the last 15 days of the 30-day PI treatment. Reactive oxygen species (ROS) production was assessed at the indicated days post-PI treatment with or without NAC for 72 h, by the reduction in nitroblue tetrazolium (NBT) (A). Mean population doubling level (PDL) values (± SEM) were determined at the indicated days post-PI treatment (B), and the PDL ratio between day 20 and day 30 was determined (C). Reactive oxygen species (ROS) production was assessed at the indicated days post-PI treatment with or without GHS/Trolox for 15 days (D). The population doubling level (PDL) was calculated as stated in Experimental procedures. Mean PDL values (± SEM) were determined at the indicated days post-PI treatment (E), and the PDL ratio between day 20 and day 30 was determined (F). * P < 0.05, ** P < 0.01 *** P < 0.001 vs. control cells, and # P < 0.05, ## P < 0.01 ### P < 0.001 vs. antioxidant-treated cells.

Download figure to PowerPoint

Pravastatin prevents PI-induced senescence of MSCs

To assess the implication of prelamin-A accumulation in the deleterious effect of PIs, we used pravastatin. We evaluated the ability of pravastatin to prevent or reverse PI-induced senescence by decreasing the amount of farnesylated prelamin-A and hence its cellular toxicity (Capell et al., 2005; Lefevre et al., 2010). Cells were treated with PIs up to 15 days, and pravastatin was added for the last 15 days. Upon day 30, pravastatin induced prelamin-A accumulation in control cells, as nonfarnesylated prelamin-A cannot be cleaved, and did not affect the prelamin-A accumulation in PI-treated cells, as it was already presented (Fig. 3A). Pravastatin markedly improved cell proliferation of ATV/r- and LPV/r-treated MSCs (Fig. 3B). After only 5 days of pravastatin treatment, PDL of PI-treated cells increased, whereas it continued to decrease in cells not treated with pravastatin. After 15 days, the reversion by pravastatin was complete. To quantify the impact of pravastatin, we used the same representation as above, determining the means of PDL ratios between day 20 and day 30 (Fig. 3C). Here, MSCs cotreated with PIs and pravastatin display an increased proliferation (positive PDL ratios), suggesting that proliferating cells overtook nonproliferating cells with a positive cumulative effect. Pravastatin also improved senescence markers in PI-exposed MSCs. It partly or totally reversed the expression of p16INK4 and p21WAF1 (Fig. 3D) and the SA-β-galactosidase activity (Fig. 3E). Pravastatin treatment normalized PI-induced overproduction of ROS (Fig. 3F–G) and SOD activity (Fig. S1D), by day 30, when PI-and pravastatin-treated cells have reached the level of proliferation achieved by control cells.

image

Figure 3. Pravastatin prevents protease inhibitors (PI)-induced senescence and oxidative stress of MSCs. Pravastatin was added for the last 15 days of the 30-day PI treatment. Whole cell lysates were extracted from MSCs treated or not with pravastatin at day 30 post-PI treatment and analysed by immunoblotting. (A) Whole cell lysates were extracted from MSCs at day 30 of PI treatment and analysed by immunoblotting. Representative immunoblots of prelamin-A, and ERK1/2 are shown. (B) Mean population doubling level (PDL) values (± SEM) were determined at the indicated days post-PI treatment, and (C) the PDL ratio between day 20 and day 30 was determined. (D) SA-β-galactosidase activity was assessed by the ratio of pH6- to pH4-positive blue staining at the indicated days post-PI treatment and expressed as % of control cells. (E) Representative immunoblots of P21, P16 and ERK1/2 from three separate experiments are shown. Reactive oxygen species (ROS) production was assessed at the indicated days post-PI treatment with or without pravastatin by (F) the reduction in NBT and by (G) the oxidation of CM-H2DCFDA and expressed as % of control cells. *P < 0.05, **P < 0.01 ***P < 0.001 vs. control cells, and ##P < 0.01, ###P < 0.001 vs. pravastatin-treated cells.

Download figure to PowerPoint

PI-induced senescence alters the osteoblastic differentiation potential of MSCs, and this potential is restored by pravastatin

As age-related loss of bone mass is associated with decreased osteoblast differentiation capacities, we undertook a study to evaluate the impact of PIs on the ability of MSCs to differentiate into mature osteoblasts. After a 30-day PI treatment, PIs were removed, and osteoblastic differentiation was induced. The ability of MSCs that displayed senescence makers at day 30 to differentiate into osteoblasts was then evaluated. MSCs pretreated with ATV/r and LPV/r displayed a severe decrease in calcium deposition, visualized by alizarin red staining (Fig. 4A–B), and in alkaline phosphatase activity (Fig. 4C–D). In accordance, we found that the mRNA expression of the osteoblast-specific markers COL1A1 and COL1A2 (Fig. 4E) and the protein expression of RUNX2 were decreased in ATV- and LPV-treated MSCs compared with control and RTV-treated cells (Fig. 4F). These results indicate that MSCs, in which cellular senescence was triggered by PIs, have a reduced osteoblastic potential.

image

Figure 4. PI-induced senescence is associated with altered osteoblastic potential of MSCs. MSCs were differentiated into osteoblasts or adipocytes after 30 days of treatment with PIs. To evaluate the osteoblastic potential of MSC, PIs were removed on the day of induction of differentiation, and osteoblastic differentiation media were added. (A) Cells were stained with alizarin red 15 days postinduction, and representative micrographs and scans are shown. (B) Quantification of alizarin red is expressed as percentage of control cells. (C) ALP activity was assessed using BCIP/NBT staining, and representative micrographs and scans are shown. (D) Quantification ALP is expressed as percentage of control cells. (E) Human COL1A1 and COL1A2 mRNA levels were measured using real-time RT–PCR. * P < 0.05, ** P < 0.01, *** P < 0.001, vs. control cells, = 4 experiments performed in duplicate. (F) Whole cell lysates were extracted at day 15 postinduction of differentiation from MSCs and analysed by immunoblotting. Representative immunoblots of RUNX2 and ERK1/2 are shown.

Download figure to PowerPoint

We performed additional experiments regarding the impact of PIs during the osteoblast differentiation process itself. MSCs were treated from day 0 to day 15 of osteoblast differentiation. As shown in Fig. S2 (Supporting information), PI treatment affected osteoblast differentiation of MSCs (Fig. S2A–D) and increased oxidative stress (Fig. S2F–G). Thus, MSCs treated with PIs when induced to differentiate towards osteoblasts also present impaired differentiation.

We then tested the effect of pravastatin on the capacity of PI-exposed MSCs to differentiate towards the osteoblastic lineage. Similarly to the previous experiment, PIs and pravastatin were removed from the differentiation media, on the day of induction. As shown by calcium accumulation visualized by alizarin red staining (Fig. 5 A–B) and by the evaluation of ALP activity (Fig. 5C–D), cells treated for 15 days with pravastatin recovered their ability to differentiate into osteoblasts. This effect cannot be attributed to cell proliferation, as in all conditions, cells were confluent when osteoblast differentiation was induced.

image

Figure 5. Pravastatin restores MSCs osteoblast differentiation potential. MSCs were differentiated into osteoblasts or adipocytes after a 30-day PI treatment with or without a 15-day pravastatin treatment. (A) Cells were stained with alizarin red to visualize calcium deposition 15 days postinduction, and representative micrographs and scans are shown. (B) Quantification of alizarin red is expressed as % of control cells. (C) ALP activity was assessed, and representative micrographs and scans are shown. (D) Quantification ALP is expressed as percentage of control cells. *** P < 0.001 vs. control cells and ## P < 0.01, ### P < 0.001 vs. pravastatin-treated cells.

Download figure to PowerPoint

PIs modulate the adipogenic differentiation potential of MSC: beneficial effect of pravastatin

As age-related loss of bone mass is associated with increased marrow fat, we hypothesized that PI-induced senescence of MSCs would favour the adipocyte differentiation, probably by altering the delicate balance between osteoblast and adipocyte differentiation. We therefore assessed whether cell fate could be affected by a pretreatment of MSCs with PIs.

As for osteoblast differentiation, upon 30 days of PI exposure, MSCs were induced to differentiate into adipocytes in the absence of PIs. The adipogenic potential of PI-exposed MSCs was evaluated by lipid accumulation by Oil-Red-O staining and by the expression of the adipocyte-specific transcription factor PPARγ. As shown in Figs 6A,B, lipid accumulation in MSCs after a 30-day exposure to LPV and LPV/r was significantly decreased by 50–65%. Conversely, ATV-treated MSCs showed an enhanced lipid accumulation (120%), and ATV/r-treated cells showed the same tendency although not significant. Relatedly and in contrast with LPV and LPV/r, ATV and ATV/r increased the protein expression of PPARγ (Fig. 6C) and the mRNA expression of the adipocyte-specific marker, fatty acid binding protein 4 (FABP4) (Fig. 6D), in differentiated MSCs.

image

Figure 6. PI-induced senescence is associated with altered adipogenic differentiation potential in MSCs and restored by pravastatin. MSCs were differentiated into adipocytes after a 30-day PI treatment. To evaluate the adipogenic potential of MSC, on the day of induction, PIs were removed, and pro-adipogenic differentiation media were added. (A) Cells were stained with Oil-Red-O to visualize lipid droplets 15 days postinduction, and representative micrographs and scans are shown. (B) Quantification of Oil-Red-O staining is expressed as percentage of control cells. (C) Whole cell lysates were extracted at day 15 postinduction from differentiated MSCs and analysed by immunoblotting. Representative immunoblots of PPARγ and ERK1/2 are shown. (D) Human FABP4 mRNA levels were measured using real-time RT–PCR. MSCs were differentiated into adipocytes after a 30-day PI treatment with or without pravastatin. (E) Representative micrographs and scans of cells stained with Oil-Red-O are shown. (F) Oil-Red-O staining quantification is expressed as percentage of control cells. * P < 0.05, ** P < 0.01 *** P < 0.001 vs. control cell, and ### P < 0.001 vs. pravastatin-treated cells. = 3 experiments performed in duplicate or triplicate.

Download figure to PowerPoint

Finally, a 15-day treatment with pravastatin before the induction of adipocyte differentiation improved the adipogenic potential of LPV/r-exposed MSCs (Fig. 6E–F). Therefore, pravastatin rescued the adipocyte differentiation potential of LPV/r-treated MSCs.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

We show here that some PIs, frequently used to control HIV infection, can induce cellular senescence and oxidative stress in bone marrow MSCs, which is associated with a reduced capacity of differentiation towards the osteoblastic lineage. Importantly, for the first time, we show that a treatment with pravastatin was able to reverse these alterations and restore osteoblast differentiation.

We evaluated the impact of widely used PIs, ATV/r and LPV/r, previously shown to increase the prevalence of osteopenia and osteoporosis in HIV-infected patients (Stone et al., 2010). In an initial experiment, we demonstrated that long-term exposure of MSCs to these PIs triggered the senescence programme. Senescent MSCs are present in human bone marrow of aged patients (Zhou et al., 2008) and in patients with severe progeroid syndromes (Scaffidi & Misteli, 2008), in which bone loss is a main feature (Rivas et al., 2009). It has been shown that senescent MSCs presented increased oxidative stress and decreased cell plasticity, affecting more particularly the osteoblast differentiation (Zhou et al., 2008).

Many studies have shown that oxidative stress plays a major role in MSCs alterations. Moreover, ROS are key mediators of signalling pathways that underlie cellular senescence. In other cell types, PI-induced ROS production via mitochondrial dysfunction was shown to be, at least in part, responsible for triggering senescence (Caron et al., 2007). Moreover, antioxidants improved PI-induced senescence in endothelial cells (Lefevre et al., 2010). Here, antioxidant molecules, Trolox/GSH, decreased ROS production and prevented further decline in cell proliferation of PI-treated MSCs, suggesting that increased oxidative stress participated to PI-induced altered cell proliferation. Additionally, using a statin, which exerts pleiotropic effects, including anti-oxidant effects (Iwasaki et al., 2011), we prevented PI-induced toxicity. The beneficial effect of pravastatin, which inhibits farnesyl synthesis in PI-treated MSCs, is in favour of the toxicity of farnesylated prelamin-A. It has been already shown that genetically determined accumulation of permanently farnesylated prelamin-A was responsible for severe syndromes of accelerated aging with early osteoporosis (Varela et al., 2008). Our present study favours the hypothesis that prelamin-A accumulation is the initial toxic event in MSCs and that ROS production is the secondary event responsible for the decrease in cell proliferation.

Increased adipogenesis and accumulation of bone marrow fat are features of aging bone (Verma et al., 2002). Early senescence induced by ATV was associated with the loss of the ability of MSCs to differentiate into osteoblasts but, in contrast, with an increase in their adipogenic capacities. The effect of ATV on adipogenesis is minor and does not really mirror the negative effect of ATV observed upon osteoblast differentiation. This observation is in agreement with recent clinical data revealing that ATV/r exerts a dual effect on bone and adipose tissue in HIV-infected patients with decreased BMD and increased central adipose tissue accumulation (McComsey et al., 2011). The ability of LPV/r to inhibit both osteoblast and adipocyte differentiation could be related to the fact that LPV accumulates into MSCs (Vernochet et al., 2005). Thus, it is possible that LPV accumulated prior to the induction of differentiation has an impact afterwards on the differentiation process itself. These results are in keeping with clinical data. Indeed, reduced intervertebral bone marrow fat was found in HIV-infected men treated by LPV/r (Huang et al., 2002), and the use of LPV/r was not associated with an increased amount of fat in HIV-infected patients (Carr et al., 2008). It could be hypothesized that in PI-treated HIV-infected patients, increased adipogenic potential of MSCs could lead to an increased fat mass and a reduced availability of osteoblast precursors. In conjunction with the reported increased incidence of fractures in HIV-1-infected patients, the management of these fractures has been shown to be complicated by poor regrowth of bone, suggesting reduced functional regenerative capacity of osteoblast precursors (Triant et al., 2008). These observations fit with the data indicating that increased abdominal visceral fat is associated with reduced bone density in HIV-infected patients receiving antiretroviral therapy (Huang et al., 2001; Brown et al., 2004).

Moreover, as shown here and in previous studies, some PIs can directly alter osteoblast and adipocyte differentiation (Lagathu et al., 2007; Caron-Debarle et al., 2010). We show here that long-term treatment with some PIs induces senescence and thereafter alters osteoblast and adipocyte differentiation. Possibly, this senescence phenotype could impede the differentiation of MSCs into other lineages. In particular, it could impact on myoblast differentiation, which could participate to the increased prevalence of frailty, muscle atrophy and sarcopenia observed in aging HIV-infected patients (Capeau, 2011).

We have shown here the beneficial effect of pravastatin on PI-induced senescence and osteoblast differentiation of MSCs. Accordingly, in the general population, statins were shown to moderately but significantly improve BMD and have been suggested to have a positive effect on bone turnover markers (Tang et al., 2008). It has been clearly shown that as the level of senescence of MSCs increases, adipogenic capacities increase and osteogenic differentiation decreases as a result of increased oxidative stress (Kim et al., 2012).

First, pravastatin could restore MSCs differentiation capacities by preventing severe oxidative stress (Iwasaki et al., 2011). MSCs shift from osteogenic to adipogenic lineage differentiation has been attributed to oxidative stress during the aging process (Kim et al., 2012). Oxidative stress is a pivotal pathogenic factor in age-related bone loss in mice, leading to decreased osteoblast number and bone formation (Almeida et al., 2007). Age-related shifts in stem cell commitment may be secondary to lineage-specific differences in susceptibility to oxidative stress and apoptosis (Bruedigam et al., 2010). Second, pravastatin could prevent the accumulation of farnesylated prelamin-A induced by PIs. Indeed, lamin A/C, which plays an important role in age-related bone loss, is required for osteoblast differentiation in MSCs. In contrast, lamin A/C inhibition in MSCs enhanced adipocyte differentiation (Akter et al., 2009; Naito et al., 2012), while the accumulation of farnesylated prelamin-A in MSCs results in premature cellular senescence and impaired adipocyte lineage (Ruiz de Eguino et al., 2012). Although a role of pravastatin on oxidative stress cannot be excluded, our results suggest that pravastatin counteracts all of the effect of PIs, including on cell differentiation, by preventing farnesylated prelamin-A accumulation.

Our study has some limitations. It may not account for many factors involved in the accelerated pro-osteoporotic process observed in HIV-infected patients, including HIV infection itself, immune balance, environmental factors and genetic predispositions. As cellular senescence is a common pathway of many cell lineages other than MSCs, it is possible that PIs could induce senescence of several cell types other than MSCs, including hematopoietic stem cells. The impact of PIs on osteoclasts, derived from the hematopoietic lineage, remains to be evaluated. Indeed, reduced BMD induced by cART is postulated to result from an imbalance between osteoclast and osteoblast activity. In clinical studies, bone turnover markers are often used to evaluate the balance between bone resorption and formation. Several studies indicate an increased level of resorption markers in HIV-infected patients, emphasizing the role of ART and more specifically of the NRTI class, on bone destruction rather than bone formation itself (Haskelberg et al., 2011). Our data suggest that some PIs could induce premature senescence of MSCs, therefore precluding the regeneration process normally associated with bone loss.

In conclusion, these data provide the first experimental support that RTV-boosted ATV (ATV/r) and LPV (LPV/r) could induce senescence of MSCs. This suggests that treatment with some PIs might participate to the early development of osteoporosis in HIV-infected patients. In several clinical studies, data on the benefit of pravastatin treatment on the fracture risk are controversial (Rejnmark et al., 2006; Tang et al., 2008). However, they suggest that pravastatin exerts a direct effect on osteoblast function and/or differentiation in vivo (Rejnmark et al., 2006; Tang et al., 2008). Thus, the ability of statins to improve PI-induced MSCs senescence could be beneficial in HIV-infected patients.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

Cell culture and treatment

Experimental procedures, with human bone marrow, have been approved by the Saint Louis Hospital Ethical Committees for human research (Paris, France), in accordance with the European Union guidelines and the Declaration of Helsinki. MSCs were isolated from washed filters used during bone marrow graft processing. MSCs were obtained and cultured as previously described (Arnulf et al., 2007). Briefly, healthy donors' bone marrow cells obtained after Ficoll (Gibco, Invitrogen Corporation, San Diego, CA, USA) were cultured at the initial density of 5 × 104 cells cm2 in alpha-minimum essential medium, supplemented with 10% foetal bovine serum (Gibco; Invitrogen Corporation), 2 mm glutamine, 2.5 ng mL−1 basic fibroblast growth factor (PeproTech, Rocky Hill, NJ, USA), penicillin and streptomycin (Gibco; Invitrogen Corporation). After 24–48 h, nonadherent cells were removed, and medium was changed. Adherent cells were then trypsinized, harvested and cultured by seeding 5 × 103 cells cm2. Cultures were fed every 2–3 days and trypsinized every 5 days.

All experiments were performed on MSCs isolated from at least 3 different bone marrow donors. Cells were exposed to 5 μm atazanavir (ATV) (kindly provided by Dr S. Azoulay, CNRS UMR 6001, Nice, France), 10 μm lopinavir (LPV) (Sigma-Aldrich, St Louis, MO, USA), alone or in association with 2 μm ritonavir (ATV/r and LPV/r), or to ritonavir alone (RTV) (Sigma-Aldrich), at clinically relevant concentrations near Cmax (Lagathu et al., 2007), or to the solvent (0.1% DMSO) for 30 days, from passages 2–9. The cells were also incubated with 25 μm pravastatin or a combination of 100 μm Trolox and 100 μm reduced glutathione during the last 15 days concomitantly with PI treatment, or with 1 mm N-Acetyl Cysteine, for 72 h from day 15 (All from Sigma-Aldrich).

Cellular proliferation and senescence

Cellular senescence was evaluated by the cell population doubling level (PDL) calculated as described previously (Caron et al., 2007), as log2 (D5/D0), where D0 and D5 are the number of cells at seeding and harvesting, respectively. The positive blue staining of β-galactosidase has been used as a biomarker of cellular senescence (Dimri et al., 1995). To detect senescence-associated β-galactosidase activity, cells were incubated with appropriate buffer solution containing X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) as described previously (Caron et al., 2007; Lefevre et al., 2010). The blue-stained cells observed at pH 6 and pH 4 were counted in 8 fields at 20× magnification, and the ratio of pH6- to pH4-positive blue cells, which specifically represents SA-β-galactosidase activity, was calculated. X-gal cellular staining was also dissolved in water, sonicated and quantified at 630 nm.

Cell differentiation

Adipogenic and osteogenic MSCs differentiation assays were performed in specific media. Differentiation was triggered on MSCs in the absence of PIs and/or pravastatin. Cells were differentiated into osteoblasts by culturing cells in osteoblast differentiation media for 15 days (10 mm β-glycerophosphate, 50 μg mL−1 ascorbate). MSCs were then stained for alizarin red (Sigma-Aldrich) as described previously (Cotter et al., 2011). Alkaline phosphatase (ALP) activity was determined using the BCIP/NBT substrate staining method (Sigma-Aldrich) described previously (Cotter et al., 2011). For adipogenic differentiation, MSCs were maintained for 3 days in an adipogenic induction media (1 μm dexamethasone, 500 μm IBMX, 1 μm insulin) followed by an adipogenic maintenance medium (1 μm insulin). Cells were then stained for Oil-red-O (Sigma-Aldrich) as described previously (Lagathu et al., 2007). All quantifications were normalized to protein content.

Oxidative stress

The production of ROS was assessed by the oxidation of 5-6-chloromethyl-2,7-dichlorodihydrofluorescein diacetate (CM-H2DCFDA) (Invitrogen) or the reduction in nitroblue tetrazolium (NBT) (Sigma-Aldrich) as previously described (Lefevre et al., 2010). Results were normalized to protein content. Superoxide dismutase (SOD) activity was evaluated using a commercially available kit (Sigma-Aldrich), according to the manufacturer's instructions.

Protein extraction and Western blotting

Proteins were extracted from cell monolayers as previously described (Caron et al., 2007). Proteins were electroblotted on nitrocellulose membrane (Amersham Biosciences, GE Healthcare Europe, Velizy Villacoublay, France). Specific proteins were detected by incubation with the appropriate primary and horseradish-peroxidase-conjugated secondary antibodies. Immune complexes were detected by enhanced chemiluminescence (Amersham Biosciences).

RNA isolation and quantitative RT–PCR

Total RNA was isolated from cultured cells using an RNeasy kit (Qiagen, Valencia, CA, USA), and mRNA expression was analysed by RT–PCR as described previously (Lagathu et al., 2007). The sequences of the oligonucleotides used as primers are available upon request.

Statistical analysis

All experiments were performed at least three times on triplicate samples. All results are expressed as means ± SEM. Statistical significance, between PI-treated cells vs. control (DMSO-treated) and pravastatin- or antioxidant-treated cells, was determined with parametric (Student's t-test) or nonparametric (Mann–Whitney U-test) tests, as appropriate.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

We Thank Dr Stephane Azoulay (CNRS, UMR 6001, Nice, France) for providing atazanavir. Research is funded by UPMC, INSERM and ANRS. Personal support is acknowledged as follows: ANRS postdoctoral fellowship (CL and SHV) and DIM-STEMPOLE doctoral fellowship (CB).

Author contributions

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information

CL and SHV conceived and designed the experiments. SHV, CL and CB performed the experiments. CL, SHV and JC analysed the data. JL contributed reagents/materials/analysis tools. CL and JC wrote the manuscript.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information
  • Akter R, Rivas D, Geneau G, Drissi H, Duque G (2009) Effect of lamin A/C knockdown on osteoblast differentiation and function. J. Bone Miner. Res. 24, 283293.
  • Almeida M, Han L, Martin-Millan M, O'Brien CA, Manolagas SC (2007) Oxidative stress antagonizes Wnt signaling in osteoblast precursors by diverting β-catenin from T cell factor to FOXO. J. Biol. Chem. 282, 2729827305.
  • Arnulf B, Lecourt S, Soulier J, Ternaux B, Lacassagne MN, Crinquette A, Dessoly J, Sciaini AK, Benbunan M, Chomienne C, Fermand JP, Marolleau JP, Larghero J (2007) Phenotypic and functional characterization of bone marrow mesenchymal stem cells derived from patients with multiple myeloma. Leukemia 21, 158163.
  • Bolland MJ, Wang TK, Grey A, Gamble GD, Reid IR (2011) Stable bone density in HAART-treated individuals with HIV: a meta-analysis. J. Clin. Endocrinol. Metab. 96, 27212731.
  • Brown TT, Qaqish RB (2006) Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review. AIDS 20, 21652174.
  • Brown TT, Ruppe MD, Kassner R, Kumar P, Kehoe T, Dobs AS, Timpone J (2004) Reduced bone mineral density in human immunodeficiency virus-infected patients and its association with increased central adiposity and postload hyperglycemia. J. Clin. Endocrinol. Metab. 89, 12001206.
  • Bruedigam C, Eijken M, Koedam M, van de Peppel J, Drabek K, Chiba H, van Leeuwen JP (2010) A new concept underlying stem cell lineage skewing that explains the detrimental effects of thiazolidinediones on bone. Stem Cells 28, 916927.
  • Capeau J (2011) Premature aging and premature age-related comorbidities in HIV-infected patients: facts and hypotheses. Clin. Infect. Dis. 53, 11271129.
  • Capell BC, Erdos MR, Madigan JP, Fiordalisi JJ, Varga R, Conneely KN, Gordon LB, Der CJ, Cox AD, Collins FS (2005) Inhibiting farnesylation of progerin prevents the characteristic nuclear blebbing of Hutchinson-Gilford progeria syndrome. Proc. Natl Acad. Sci. U.S.A. 102, 1287912884.
  • Caron M, Auclair M, Donadille B, Bereziat V, Guerci B, Laville M, Narbonne H, Bodemer C, Lascols O, Capeau J, Vigouroux C (2007) Human lipodystrophies linked to mutations in A-type lamins and to HIV protease inhibitor therapy are both associated with prelamin-A accumulation, oxidative stress and premature cellular senescence. Cell Death Differ. 14, 17591767.
  • Caron-Debarle M, Lagathu C, Boccara F, Vigouroux C, Capeau J (2010) HIV-associated lipodystrophy: from fat injury to premature aging. Trends Mol. Med. 16, 218229.
  • Carr A, Samaras K, Thorisdottir A, Kaufmann GR, Chisholm DJ, Cooper DA (1999) Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor-associated lipodystrophy, hyperlipidaemia, and diabetes mellitus: a cohort study. Lancet 353, 20932099.
  • Carr A, Ritzhaupt A, Zhang W, Zajdenverg R, Workman C, Gatell JM, Cahn P, Chaves R (2008) Effects of boosted tipranavir and lopinavir on body composition, insulin sensitivity and adipocytokines in antiretroviral-naive adults. AIDS 22, 23132321.
  • Coffinier C, Hudon SE, Farber EA, Chang SY, Hrycyna CA, Young SG, Fong LG (2007) HIV protease inhibitors block the zinc metalloproteinase ZMPSTE24 and lead to an accumulation of prelamin-A in cells. Proc. Natl Acad. Sci. U.S.A. 104, 1343213437.
  • Cotter EJ, Chew N, Powderly WG, Doran PP (2011) HIV type 1 alters mesenchymal stem cell differentiation potential and cell phenotype ex vivo. AIDS Res. Hum. Retroviruses 27, 187199.
  • Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano EE, Linskens M, Rubelj I, Pereira-Smith O, Peacocke M, Campisi J (1995) A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc. Natl Acad. Sci. U.S.A. 92, 93639367.
  • Duvivier C, Kolta S, Assoumou L, Ghosn J, Rozenberg S, Murphy RL, Katlama C, Costagliola D (2009) Greater decrease in bone mineral density with protease inhibitor regimens compared with NNRTI regimens in HIV-1 infected naive patients. AIDS 23, 817824.
  • Grund B, Peng G, Gibert CL, Hoy JF, Isaksson RL, Shlay JC, Martinez E, Reiss P, Visnegarwala F, Carr AD (2009) Continuous antiretroviral therapy decreases bone mineral density. AIDS 23, 15191529.
  • Hansen AB, Obel N, Nielsen H, Pedersen C, Gerstoft J (2011) Bone mineral density changes in PI-sparing vs. nucleoside reverse transcriptase inhibitor-sparing highly active antiretroviral therapy: data from a randomized trial. HIV Med. 12, 157165.
  • Haskelberg H, Carr A, Emery S (2011) Bone turnover markers in HIV disease. AIDS Rev. 13, 240250.
  • Huang JS, Rietschel P, Hadigan CM, Rosenthal DI, Grinspoon S (2001) Increased abdominal visceral fat is associated with reduced bone density in HIV-infected men with lipodystrophy. AIDS 15, 975982.
  • Huang JS, Mulkern RV, Grinspoon S (2002) Reduced intravertebral bone marrow fat in HIV-infected men. AIDS 16, 12651269.
  • Iwasaki Y, Yamato H, Fukagawa M (2011) Treatment with pravastatin attenuates oxidative stress and protects osteoblast cell viability from indoxyl sulfate. Ther. Apher. Dial. 15, 151155.
  • Jain RG, Lenhard JM (2002) Select HIV protease inhibitors alter bone and fat metabolism ex vivo. J. Biol. Chem. 277, 1924719250.
  • Kim M, Kim C, Choi YS, Kim M, Park C, Suh Y (2012) Age-related alterations in mesenchymal stem cells related to shift in differentiation from osteogenic to adipogenic potential: implication to age-associated bone diseases and defects. Mech. Ageing Dev. 133, 215225.
  • Lagathu C, Eustace B, Prot M, Frantz D, Gu Y, Bastard JP, Maachi M, Azoulay S, Briggs M, Caron M, Capeau J (2007) Some HIV antiretrovirals increase oxidative stress and alter chemokine, cytokine or adiponectin production in human adipocytes and macrophages. Antivir. Ther. 12, 489500.
  • Lefevre C, Auclair M, Boccara F, Bastard JP, Capeau J, Vigouroux C, Caron-Debarle M (2010) Premature senescence of vascular cells is induced by HIV protease inhibitors: implication of prelamin-A and reversion by statin. Arterioscler. Thromb. Vasc. Biol. 30, 26112620.
  • Liu Q, Kim DI, Syme J, LuValle P, Burke B, Roux KJ (2010) Dynamics of lamin-A processing following precursor accumulation. PLoS ONE 5, e10874.
  • Mallon PW, Miller J, Cooper DA, Carr A (2003) Prospective evaluation of the effects of antiretroviral therapy on body composition in HIV-1-infected men starting therapy. AIDS 17, 971979.
  • Mazziotti G, Canalis E, Giustina A (2010) Drug-induced osteoporosis: mechanisms and clinical implications. Am. J. Med. 123, 877884.
  • McComsey GA, Kitch D, Sax PE, Tebas P, Tierney C, Jahed NC, Myers L, Melbourne K, Ha B, Daar ES (2011) Peripheral and central fat changes in subjects randomized to abacavir-lamivudine or tenofovir-emtricitabine with atazanavir-ritonavir or efavirenz. Clin. Infect. Dis. 53, 185196.
  • Naito M, Omoteyama K, Mikami Y, Takagi M, Takahashi T (2012) Suppression of lamin A/C by short hairpin RNAs promotes adipocyte lineage commitment in mesenchymal progenitor cell line, ROB-C26. Histochem. Cell Biol. 137, 235247.
  • Nuttall ME, Gimble JM (2004) Controlling the balance between osteoblastogenesis and adipogenesis and the consequent therapeutic implications. Curr. Opin. Pharmacol. 4, 290294.
  • Powderly WG (2012) Osteoporosis and bone health in HIV. Curr. HIV/AIDS Rep. 9, 218222.
  • Rejnmark L, Vestergaard P, Mosekilde L (2006) Statin but not non-statin lipid-lowering drugs decrease fracture risk: a nation-wide case-control study. Calcif. Tissue Int. 79, 2736.
  • Rivas D, Li W, Akter R, Henderson JE, Duque G (2009) Accelerated features of age-related bone loss in zmpste24 metalloproteinase-deficient mice. J. Gerontol. A Biol. Sci. Med. Sci. 64, 10151024.
  • Ruiz de Eguino G, Infante A, Schlangen K, Aransay AM, Fullaondo A, Soriano M, García-Verdugo JM, Martín AG, Rodríguez CI (2012) Sp1 transcription factor interaction with accumulated prelamin-A impairs adipose lineage differentiation in human mesenchymal stem cells. Stem Cells Transl. Med. 1, 309321.
  • Scaffidi P, Misteli T (2008) Lamin A-dependent misregulation of adult stem cells associated with accelerated ageing. Nat. Cell Biol. 10, 452459.
  • Stone B, Dockrell D, Bowman C, McCloskey E (2010) HIV and bone disease. Arch. Biochem. Biophys. 503, 6677.
  • Tang QO, Tran GT, Gamie Z, Graham S, Tsialogiannis E, Tsiridis E, Linder T (2008) Statins: under investigation for increasing bone mineral density and augmenting fracture healing. Expert Opin. Investig. Drugs 17, 14351463.
  • Triant VA, Brown TT, Lee H, Grinspoon SK (2008) Fracture prevalence among human immunodeficiency virus (HIV)-infected vs non-HIV-infected patients in a large U.S. healthcare system. J. Clin. Endocrinol. Metab. 93, 34993504.
  • Varela I, Pereira S, Ugalde AP, Navarro CL, Suarez MF, Cau P, Cadinanos J, Osorio FG, Foray N, Cobo J, de Carlos F, Levy N, Freije JM, Lopez-Otin C (2008) Combined treatment with statins and aminobisphosphonates extends longevity in a mouse model of human premature aging. Nat. Med. 14, 767772.
  • Verma S, Rajaratnam JH, Denton J, Hoyland JA, Byers RJ (2002) Adipocytic proportion of bone marrow is inversely related to bone formation in osteoporosis. J. Clin. Pathol. 55, 693698.
  • Vernochet C, Azoulay S, Duval D, Guedj R, Cottrez F, Vidal H, Ailhaud G, Dani C (2005) Human immunodeficiency virus protease inhibitors accumulate into cultured human adipocytes and alter expression of adipocytokines. J. Biol. Chem. 280, 22382243.
  • Zhou S, Greenberger JS, Epperly MW, Goff JP, Adler C, Leboff MS, Glowacki J (2008) Age-related intrinsic changes in human bone-marrow-derived mesenchymal stem cells and their differentiation to osteoblasts. Aging Cell 7, 335343.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Conflict of interest
  9. Author contributions
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
acel12119-sup-0001-FigureS1.epsimage/eps888KFig. S1. Effect of PIs associated or not with pravastatin or antioxidant treatment on SOD Activity in MSCs.
acel12119-sup-0002-FigureS2.epsimage/eps2591KFig. S2. Direct effect of PIs on osteoblast differentiation of MSCs.
acel12119-sup-0003-legends.docxWord document14K 

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.