|Home | About | Journals | Submit | Contact Us | Français|
Rapamycin, a potent immunomodulatory drug, has shown promise in the amelioration of numerous age-associated diseases including cancer, Alzheimer’s disease and cardiac hypertrophy. Yet the elderly, the population most likely to receive therapeutic rapamycin, are already at increased risk for infectious disease; thus concern exists that rapamycin may exacerbate age-associated immune dysfunctions and worsen infection outcomes. Herein, we examined the impact of enteric delivered rapamycin monotherapy (eRapa) on susceptibility of aged (22–24 month) C57BL/6 mice to Streptococcus pneumoniae, the leading bacterial cause of community-acquired pneumonia. Following challenge with S. pneumoniae, administration of eRapa conferred modest protection against mortality. Reduced mortality was the result of diminished lung damage rather than reduced bacterial burden. eRapa had no effect on basal levels of Interleukin (IL)-1α, IL-6, IL-10, IL-12p70, KC, Interferon-γ, Tumor necrosis factor α and Monocyte chemotactic protein-1 in whole lung homogenates or during pneumococcal pneumonia. Previously we have demonstrated that cellular senescence enhances permissiveness for bacterial pneumonia through increased expression of the bacterial ligands Laminin receptor (LR), Platelet-activating factor receptor (PAFr) and Cytokeratin 10 (K10). These proteins are co-opted by S. pneumoniae and other respiratory tract pathogens for host cell attachment during lung infection. UM-HET3 mice on eRapa had reduced lung cellular senescence as determined by levels of the senescence markers p21 and pRB, but not mH2A.1. Mice on eRapa also had marked reductions in PAFr, LR, and K10. We conclude that eRapa protected aged mice against pneumonia through reduced lung cellular senescence, which in turn, lowered bacterial ligand expression.
Rapamycin, an mTOR Complex 1 inhibitor, when fed to mice via enteric-coated microcapsules in rodent chow, substantially extends lifespan (Harrison et al., 2009; Miller et al., 2011). In addition to this, rapamycin has shown promise in the amelioration of cognitive deficits in mouse models of Alzheimer’s disease (Caccamo et al., 2010; Spilman et al., 2010), reduced the formation of protein inclusions found in numerous neurodegenerative diseases (Caccamo et al., 2009), prevented cardiac hypertrophy (Finckenberg and Mervaala 2010), and has well established anti-tumor properties (Albert et al., 2010). Less recognized, rapamycin has also been shown to prevent development of cellular senescence (Demidenko and Blagosklonny 2008; Demidenko et al., 2009; Leontieva and Blagosklonny 2010), a phenomenon whereby cells with shortened telomeres or those that have undergone DNA damage (i.e., genotoxic stress) lose the capacity to replicate without undergoing apoptosis and adopt a pro-inflammatory secretory phenotype (Campisi and d’Adda di Fagagna 2007). Thus, rapamycin has manifold benefits for a wide range of age-associated conditions and diseases.
Pneumonia, due both to viral and bacterial infection, is the fourth leading cause of death for those >65 years of age (Lopez et al., 2006; Minino et al., 2006). Numerous age-related changes in lung function, a decrease in innate and adaptive immunity, and the presence of age-associated co-morbidities such as cardiovascular disease, collectively contribute to the enhanced occurrence and severity of pneumonia among the elderly (Lexau et al., 2005). Most recently, we have demonstrated that senescent cells, which accumulate within the lungs and other aged tissues (Kreiling et al., 2011), enhance permissiveness for bacterial pneumonia through increased expression of inflammation- (i.e. Nuclear-factor kappa B [NFκB]-) and Retinoblastoma protein (pRB)-regulated proteins that pathogenic bacteria co-opt for lung cell adhesion and blood stream invasion. Briefly, this includes Laminin Receptor (LR), Platelet activating factor receptor (PAFr), and Cytokeratin 10 (K10); which in turn are bound to by Streptococcus pneumoniae, Haemophilus influenzae, and Staphylococcus aureus among other bacterial pathogens (Shivshankar et al., 2011). Thus, the individuals that might benefit the most from therapeutic rapamycin, the elderly, are also at increased risk for bacterial pneumonia due in part to senescence-associated changes in lung protein expression.
Given rapamycin’s well-established immunomodulatory properties, which include altered T-cell development and dampened antigen presenting cell function (Araki et al., 2009; Battaglia et al., 2005; Hackstein et al., 2003), enthusiasm for the drugs potential as a broad range therapeutic has thus far been modest. Concerns about immunosuppressive consequences of rapamycin are supported by considerable anecdotal evidence for opportunistic infections in transplant recipients receiving rapamycin to prevent organ rejection (Groth et al., 1999; Husain et al., 2003; Webster et al., 2006). Nonetheless, the effect of prolonged rapamycin monotherapy on the ability of otherwise healthy aged animals to resist pneumonia is not known. Furthermore, as rapamycin has been reported to inhibit cellular senescence, it may instead protect aged animals through reduced bacterial ligand expression in the lungs. Thus, the impact of rapamycin on bacterial pneumonia remains an open question.
Herein, we examined the impact of enteric delivered rapamycin monotherapy (eRapa) on susceptibility of aged mice to Streptococcus pneumoniae (the pneumococcus), a Gram-positive bacterium and the leading bacterial cause of community-acquired pneumonia (Johansson et al., 2010). We tested the impact of eRapa on disease severity, the acute pro-inflammatory cytokine response, cellular senescence in the lungs, and bacterial ligand expression. Against expectation, we show that eRapa conferred modest protection against death; possibly through reduced cellular senescence and bacterial ligand expression.
For experiments that examined senescent markers and bacterial ligand levels within the lungs, UM-HET3 mice, the progeny of CB6F1 mothers and C3D2F1 fathers, were used. These mice, housed at The University of Michigan, received rodent chow ad libitum containing polymer-encapsulated rapamycin (i.e. eRapa) at 4.7- (low-), 14- (mid-), or 42-ppm (high-dose) rapamycin for 13 months from 9 to 22 months of age. Control mice received normal chow. For experiments involving bacterial challenge, male and female C57BL/6 mice were purchased from The Jackson Laboratories (Bar Harbor, Maine) and housed at The University of Texas Health Science Center. Mice were fed chow containing 14-ppm rapamycin such that they received 2.2 mg of rapamycin/kg body weight per day. This is the same dose and drug delivery formulation previously found to be neuroprotective and life extending (Caccamo et al., 2009; Caccamo et al., 2010; Harrison et al., 2009). Control mice received chow containing an equivalent amount of the encapsulating agent (Eudragit® Darmstadt, Germany) alone (Harrison et al., 2009). From 4 months of age C57Bl/6 mice received either eRapa long-term for 20 months (LT-eRapa), the control diet for 20 months (control), or the control diet for 16 months then eRapa for 4 months (ST-eRapa). For both sets of mice, activity of rapamycin was confirmed periodically by testing for phosphorylation of ribosomal protein S6 (at Ser 240 and Ser 244), a substrate of S6 kinase 1, in visceral adipose tissue of sentinel mice that received the diet (data not shown) (Miller et al., 2011). All mice were housed in specific pathogen free conditions.
Whole lung homogenates were prepared in lysis buffer and protein concentrations were determined via bicinchoninic acid assay. For each biological sample, 30 μg of sample was separated by SDS-PAGE under reducing conditions and proteins transferred onto nitrocellulose using a Semi-Dry Electrophoretic system (Bio-Rad, Hercules CA). Antibodies targeting PAFr (Cat# 160602, Cayman Chemical Co., Ann Arbor, MI), LR (Cat# MS-259-PIABX, NeoMarkers, Fremont, CA), K10 (Cat# Ab9026, Abcam, Cambridge, MA), pRB (Cat# 9309, Cell Signaling, Danvers, MA), p21 (Cat# SC-6246, Santa Cruz Antibody, Santa Cruz, CA), and histone macro H2A.1 (mH2A; Cat# 07-219, Millipore, Billerica, MA) were all obtained from commercial sources and incubated according to the manufacturers’ specifications. Membranes were subsequently probed using horseradish peroxidase conjugated secondary antibodies at 1:10,000 dilution using standard chemiluminescent methods. To confirm protein load, membranes were stripped and probed with antibody against actin (Cat# A300-48485A, Bethyl Laboratories, Montgomery, TX). Band intensities on autoradiographs were determined using Quantity One software (Bio-Rad).
Mice were challenged by forced inhalation of S. pneumoniae serotype 4 strain TIGR4 in 100 μl saline (Orihuela et al., 2004; Tettelin et al., 2001). Aged mice received 1.0 × 103 cfu for mortality and subsequent pathological and lung burden studies. All cohorts were divided equally between male and female mice and the infectious dose administered was verified. Disease severity was assessed by survival over time, quantitation of bacterial burden in the lungs and blood at fixed time points, and pathological examination of paraffin-embedded lung sections (Shivshankar et al., 2011). For pathological examination, two non-adjacent lung sections from each mouse were Hematoxylin and Eosin (H&E) stained and examined microscopically. These were scored blindly on a scale of 0–5 on the basis of peribronchial and perivascular inflammation, neutrophil infiltration, and alveolar consolidation. Levels of [Interleukin (IL)-6, IL-10, IL-12p70, Interferon (IFN)-γ, tumor necrosis factor (TNF)-α] and [Monocyte chemotactic protein (MCP)-1] in whole lung homogenates of the right inferior lobe were determined using a BD™ Cytometric Bead Array (CBA) - Mouse Inflammation Kit (BD Biosciences, San Jose, CA). Levels of IL-1α and KC were determined using mouse ELISA kits also from BD Biosciences.
Statistical analysis on survival following pathogen challenge was performed using a Kaplan-Meier Log Rank Test using the Holm-Sidak method. A One-Way ANOVA was used to compare bacterial titers, cytokine levels, and densitometric immunoblot values from eRapa fed groups versus the control for aged mice.
Herein, we evaluated whether monotherapy with eRapa altered the susceptibility of aged mice to Streptococcus pneumoniae. Briefly, S. pneumoniae is a Gram-positive bacterium and a leading cause of community-acquired pneumonia (Johansson et al., 2010).
Elevated lung levels of host proteins that are co-opted by respiratory tract pathogens for attachment and cell invasion are one reason for the enhanced susceptibility of the elderly to pneumonia (Hinojosa et al., 2009). Previously, we have shown that cellular senescence contributes to this phenomena through increased levels of LR, PAFr, and K10 on lung epithelial cells (Shivshankar et al., 2011). Strikingly, we observed that healthy aged UM-HET3 mice receiving low-, mid-, and high-dose eRapa for 13 months had dramatic and significant reductions in basal lung levels of all three ligands versus the controls (Fig 1A, C). Moreover, we observed a strong reduction in levels of the senescent marker p21, a cyclin-dependent kinase inhibitor that checks the progression of the cell cycle at G1 (Child and Mann 2006), and to a lesser extent a dose-dependent reduction for retinoblastoma protein (pRB), a second senescent marker and tumor suppressor protein that binds and inhibits transcription factors of the E2F family (Fig 1B, C) (Manning and Dyson 2011). In contrast, we observed a modest but significant increase for mH2A.1, a third senescence marker, for mice on the low-eRapa diet. mH2A.1 is an indicator of accumulated DNA damage/heterochromatin complexes within the nuclei (Kreiling et al., 2011). Nonetheless, no changes were apparent for mH2A.1 in mice receiving the mid- or high-eRapa diet (Fig 1). Thus, eRapa was associated with a sharp reduction in bacterial ligand expression that has previously been shown to enhance permissiveness for bacterial attachment to lung cells and was concomitant with decreased levels of two well-established senescence associated proteins.
We infected aged C57BL/6 mice on short-term (ST) and long-term (LT) eRapa to determine its impact on severity of pneumococcal pneumonia. Figure 2 outlines the administered eRapa diets, which contained rapamycin at the dose previously documented to extend lifespan (Harrison et al., 2009). Aged mice on ST-eRapa had improved survival versus controls following intratracheal challenge with S. pneumoniae strain TIGR4 (ST-eRapa n=14, control n=15, P=0.025). A non-significant but similar positive trend was observed for aged mice on LT-eRapa (n=10, P=0.098) (Fig 3A). Of note, no differences in bacterial titers were observed in the lungs of aged mice after 24 hours versus control (Fig 3B). Nor was there a difference in the blood at 24 and 48 hours post infection when comparing aged ST-eRapa or LT-eRapa to control mice (Fig 3C).
Consistent with the delayed mortality observed for aged mice on eRapa, pathological examination of H&E stained lung sections from infected mice after 24 hours revealed that infected C57BL/6 mice on ST-eRapa had a significantly reduced peribronchial and perivascular inflammation, decreased immune-cell infiltration, and less alveolar thickening than mice on the Eudragit controls, whereas mice on LT-eRapa trended towards the same (P=0.13) (Fig 4). No obvious differences in lung pathology were apparent in the sham-infected controls or between ST-eRapa or LT-eRapa infected mice. Given that rapamycin and inhibition of mTOR Complex 1 has been shown to inhibit de novo cytokine production by antigen presenting cells (Hackstein et al., 2003), we also examined whole lung homogenates for eRapa-induced alterations in the acute pro-inflammatory cytokine (i.e. IL-1α, IL-6, IL-12p70, IFN-γ, and TNFα), anti-inflammatory cytokine (i.e. IL-10), and pro-inflammatory chemokine (i.e. KC, MCP-1) response. No physiologically relevant differences were observed for mice on eRapa (ST-eRapa n=8, LT-eRapa n=5) in the production of these cytokines versus controls (n=6) during infection, nor were differences in basal levels of these inflammatory molecules observed in lung samples collected from uninfected aged C57BL/6 mice (control n=3, ST-Rapa n=5, LT-eRapa n=5) (Fig 5). Thus, eRapa did not alter the resting or acute cytokine response during infection, yet was associated with decreased tissue damage and vascular leakage, and reduced mortality.
While rapamycin and possibly other mTOR Complex I inhibitors, hold considerable promise as therapeutic agents to treat or delay the onset of Alzheimer’s disease and other age-associated diseases (Caccamo et al., 2009; Caccamo et al., 2010), there is a valid concern that the health benefits observed for rapamycin might be offset in the pathogen-rich real world by enhanced susceptibility to infectious diseases. This issue is particularly pressing for the elderly, the patient population most likely to use and benefit from rapamycin, as they are already at increased risk for pneumonia (Lexau et al., 2005; Lopez et al., 2006). As such, studies similar to this one, with aged animals on prolonged rapamycin monotherapy, are necessary to examine the impact of immune modulatory drugs on susceptibility to infectious diseases. One important caveat to this study is that the analyses of senescent markers and bacterial ligands were done using lung samples from heterogeneous UM-HET3 mice, whereas the susceptibility to pneumococcal pneumonia studies were done using inbred C57BL/6 mice. Thus, and despite the fact that rapamycin reduced cellular senescence and pneumococcal ligand expression in the lungs of genetically diverse animals, it is possible that the protection conferred by eRapa against pneumonia in the C57BL/6 mice was for other reasons and caution should be taken in the interpretation of our results.
We have previously demonstrated that cellular senescence results in the increased expression of bacterial ligands that respiratory tract pathogens co-opt for lung infection (Shivshankar et al., 2011). LR serves as a ligand for S. pneumoniae, H. influenzae and Neisseria meningitidis (Orihuela et al., 2009). K10 is a ligand for S. pneumoniae and S. aureus (O’Brien et al., 2002; Shivshankar et al., 2009). PAFr serves as a ligand for S. pneumoniae, H. influenzae, and Pseudomonas aeruginosa as well as many other bacteria that have surface-exposed phosphorylcholine (Barbier et al., 2008; Cundell et al., 1995; Swords et al., 2000). Respiratory tract pathogens also bind to other NFkB-regulated proteins such as intercellular adhesion molecule 1 (ICAM-1) and carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) that were not examined, but are also most likely up regulated as a result of the pro-inflammatory phenotype of senescent cells (Avadhanula et al., 2006; Conners et al., 2008). Thus, the decreased expression of bacterial ligands in the lungs of mice treated with eRapa suggests mTOR inhibition as a potential mechanism to decrease susceptibility to a broad range of bacterial pathogens. What is more, the concomitant reduction in senescent markers along with pneumococcal ligands in the lungs reinforces the notion that cellular senescence directly contributes to the susceptibility of the elderly to pneumonia.
Rapamycin’s broad effects are due to the fact that mTOR is a key component of multiple and overlapping cell-signaling pathways including those that affect de novo protein production, autophagy, T-cell differentiation, growth, and cell proliferation. Indeed, rapamycin has been demonstrated to have a potent anti-inflammatory effect due to its selective expansion of immunosuppressive T-regulatory cells (Araki et al., 2009; Battaglia et al., 2005), and its ability reduce cytokine production by antigen presenting cells following stimulation (Hackstein et al., 2003). It is for these reasons, the Food and Drug Administration has approved rapamycin along with calcineruin inhibitors to prevent organ rejection following transplantation. Unexpectedly, we observed no effect of prolonged low-dose enteric rapamycin delivery on basal cytokine levels or the ability of mice to respond to an infection despite the fact that eRapa suppressed LR and PAFr protein levels, which are encoded by NFkB-regulated genes. Reasons for the discrepancy in cytokine production between our study and others may include the strain of mouse tested, dose and duration of rapamycin, and route of drug administration. Importantly, it remains unclear at the molecular level how rapamycin suppresses pneumococcal ligand expression or that of the senescent markers in the lungs.
Several studies by Blagosklonny and colleagues have examined the effect of rapamycin on cellular senescence (Demidenko and Blagosklonny 2008; Demidenko et al., 2009; Leontieva and Blagosklonny 2010). These studies demonstrated that inhibition of mTOR in vitro either by serum starvation or rapamycin had a protective effect against stress-induced senescence. Blagosklonny and colleagues also showed that DNA damaging agents did not cause senescence in quiescent cells, while they do so in cells where mTOR is active (Leontieva and Blagosklonny 2010). Our experimental results support these observations and indicate that rapamycin reduces cellular senescence in vivo as measured by pRB and p21 levels. These markers are elevated in aged mice and were detectable in the control diet mice most likely as a result of normal aging processes and the accumulation of senescent cells.
Importantly, the findings published by Blagosklonny and colleagues does not suggest that rapamycin protect cells against DNA damage, but instead that rapamycin (i.e. mTOR inhibition) prevents activation of the cell-signaling pathways that leads to permanent cell cycle arrest. Along this line, we speculate that eRapa does not inhibit the damage or the signals that are responsible for enhanced NFkB-mediated pneumococcal ligand expression or pRB-mediated K10 expression but instead dampens the cellular response to these signals; possibly at the translational level. In support of this notion, we did not observe a decrease in the senescence marker mH2A.1, which is associated with heterochromatin remodeling following DNA damage (Kreiling et al., 2011).
Pneumococcal disease is characterized by intense inflammation at the affected site. Excessive inflammation results in vascular leakage, alveolar consolidation, and a loss of gas-exchange capability that often leads to death (Orihuela and Tuomanen 2006). Indeed this process was observed in all the experimentally infected mice and could be correlated with the overall mortality rates for the eRapa and control cohorts. Surprisingly, despite reduced lung pathology, which can in part be explained by reduced bacterial adhesion and invasion, we observed no differences in cytokine levels for the ST-eRapa or LT-eRapa cohorts versus controls. One possible explanation is that rapamycin promotes lung cell apoptosis over necrosis, the former that is immunoquiescent and anti-inflammatory in comparison to necrosis (Zhang et al., 2010). For example, rapamycin has been demonstrated to induce apoptosis in tumor cells by reducing the expression of Bcl-XL an inhibitor of apoptosis (Tirado et al., 2005). Rapamycin also expands the numbers of CD4+CD25+Foxp3+ T-regulatory cells (Battaglia et al., 2005). T-regulatory cells are present in humans with acute lung injury and have been shown to help resolve experimental lung injury in mice (D’Alessio et al., 2009). Importantly, we only tested a small number of cytokine/chemokine/immune modulators. Further experimentation is required to resolve this speculation and assess the impact of eRapa on apoptosis and T-regulatory cells following cellular insult.
Our study is not the first to examine the effects of rapamycin on infection, although it is the first to test the effects of prolonged monotherapy. Recently, Fielhaber et al. showed that intraperitoneal administration of rapamycin at 1.5 mg/kg six hours before challenge with lipopolysaccharide enhanced lung injury. This occurred as a result of inhibited LPS-induced apoptosis, yet was associated with reduced NFkB activation, reduced neutrophil infiltration, and diminished pro-inflammatory cytokine production (Fielhaber et al., 2012). In addition to differences in dose and drug schedule, our study is different in that S. pneumoniae is a Gram-positive bacterium and thereby lacks lipopolysaccharide, a highly potent pro-inflammatory and cytotoxic molecule. Other investigators report that rapamycin induced changes in autophagy following influenza infection that enhanced apoptosis within infected cells and worsened tissue damage (Sun et al., 2012). In contrast, rapamycin inhibited the development of Chlamydia pneumoniae, an obligate intracellular pathogen (Yan et al., 2010). Thus the impact of rapamycin on susceptibility to infection most likely depends on a wide range of factors that includes Gram-positive versus Gram-negative, intracellular versus extracellular, and host mechanisms of clearance. Importantly, these etiological agents must also be tested in context of an aged mouse in order to have some clinical applicability. Finally, the impact of short versus long-term rapamycin administration is also an important issue, as long-term rapamycin is now known to inhibit of mTOR Complex 2 which has other effects including development of glucose intolerance (Hughes and Kennedy 2012). Perhaps inhibition of mTOR complex 2 explains why our aged mice on LT-eRapa were slightly more susceptible to infection than those on ST-eRapa? Of note, the effects of rapamycin are complex and most likely impact infectious disease susceptibility/resistance in a variety of ways simultaneously; some of which are indirect. For example Chen et al. showed that that 8 weeks of rapamycin administration increased bone marrow production of B-cells in aged mice and enhanced their ability to respond to influenza vaccination. This improved aged mouse survival following influenza challenge (Chen et al., 2009).
In summary, we have determined that prolonged eRapa has no detrimental effect for aged mice against S. pneumoniae. This was a somewhat a surprising result given that rapamycin is generally considered to be immunosuppressive and aged mice are exquisitely susceptible to pneumococcal pneumonia. Our findings suggest that the acute antibacterial immune response to this pathogen is not seriously compromised and the beneficial effects of therapeutic rapamycin, seem to come from reduced cellular senescence in vivo and reduced lung pathology. Although our studies used a small number of mice and extrapolating mice data to humans is risky, they do support the possibility that rapamycin can be used as a human therapeutic. Importantly, the underlying mechanisms responsible for the differential effect of eRapa with other pathogens need to be evaluated empirically.
Support for this project was from NIH AI083387 to SB, AG0123823 to RM, and AG033274 and AG036613 - Project 5 to CJO.
AUTHOR CONTRIBUTIONSCAH and VEM together completed the presented work. SVR and RM at University of Michigan provided material support in regards to lung samples from mice administered low-, mid- and high- eRapa. SNA, SB and CJO were involved in experimental design and oversaw the completion of the research project.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.