Complement anaphylatoxins C3a and C5a induce a failing regenerative program in cardiac resident cells. Evidence of a role for cardiac resident stem cells other than cardiomyocyte renewal
- David Lara-Astiaso†1,
- Alberto Izarra†1,
- Juan Camilo Estrada1,
- Carmen Albo1,
- Isabel Moscoso1,
- Enrique Samper1,
- Javier Moncayo2,
- Abelardo Solano2,
- Antonio Bernad1Email author and
- Antonio Díez-Juan1, 2Email author
© Lara-Astiaso et al; licensee Springer. 2012
Received: 18 October 2012
Accepted: 30 November 2012
Published: 12 December 2012
Cardiac healing, which follows myocardial infarction, is a complex process guided by intricate interactions among different components. Some resident cell populations with a potential role in cardiac healing have already been described in cardiac tissues. These non-cardiomyocyte cell subsets, globally described as cardiac pluripotent/progenitor cells (CPCs), are able to differentiate into all three major cardiac cell lineages (endothelial, smooth muscle and cardiomyocyte cells) in experimental settings. Nevertheless, physiological cardiac healing results in a fibrous scar, which remains to be fully modelled experimentally. Since a role for complement anaphylatoxins (C3a and C5a) has been described in several regeneration/repair processes, we examined the effects that C3a and C5a exert on a defined population of CPCs. We found that C3a and C5a are able to enhance CPC migration and proliferation. In vitro studies showed that this effect is linked to activation of telomerase mRNA and partial preservation of telomere length, in an NFκB-dependent manner. In addition, anaphylatoxin signalling modulates the CPC phenotype, increasing myofibroblast differentiation and reducing endothelial and cardiac gene expression. These findings may denote that C3a and C5a are able to maintain/increase the cardiac stem cell pool within the heart, whilst simultaneously facilitating and modulating resident cell differentiation. We found that this modulation was directed towards scar forming cells, which increased fibroblast/myofibroblast generation and suggests that both these anaphylatoxins could play a relevant role in the damage-coupled activation of resident cells, and regulation of the cardiac healing process after injury.
The cardiac healing process is guided by intricate interactions between different components; following myocardial infarction (MI), injury, inflammation, regeneration, and repair are all interconnected processes. It is known that these processes are inadequate and over-complicated, since certain pathological or damaging factors, such as cardiomyocyte replacement cannot be repaired. The inflammatory response represents one critical element of the cardiac healing process and is triggered by cell stress and death caused during cardiac injury. The abolition of the inflammatory response using corticosteroids not only decreases the number of infiltrating leukocytes, but also delays healing and collagen deposition (Kloner et al. 1978). Moreover it has been shown that cardiac tissue reperfusion improves overall tissue repair and that this outcome is mediated by improving the inflammatory reaction (Frangogiannis 2012).
Over the last 30 years, the complement system has been shown to play a major role in myocardial inflammation and tissue injury following MI (Hill and Ward 1971; Walport 2001) has been experimentally inhibited at different levels, including that of anaphylatoxin (C3a, C5a) signalling, to reduce ischemic injury. Experiments using animal models have shown that anaphylatoxin C5a inhibition protects against ischemia reperfusion (I/R) injury in diverse organs including myocardium (Vakeva et al. 1998). While anaphylatoxin C5a has been shown to be a potent activator of inflammation (Walport 2001), the upstream role of anaphylatoxin C3a has shown mixed results in animal models of I/R injury. In mouse models of renal and cerebral I/R injury, C3a appears to play a key role in mediating inflammation (Mocco et al. 2006; Thurman et al. 2007), whereas other studies suggest it is less important (Busche and Stahl 2010; Proctor et al. 2004). Moreover, in mammals, anaphylatoxins are critical for hepatocyte proliferation and liver regeneration (Strey et al. 2003). C3a promotes homing, chemotaxis, and retention of hematopoietic stem and progenitor cells in the bone marrow (Ratajczak et al. 2004); the anaphylatoxin receptors (C3aR, C5aR) positively regulate adult neurogenesis as well as the quantity of replacement neurons produced following cerebral ischemia (Rahpeymai et al. 2006). Recently, a novel and unexpected function for C3a and its receptor C3aR has also been revealed in the mutual cell-cell attraction (named co-attraction), required for maintaining cohesive clusters of migrating mesenchymal cells during early development (Carmona-Fontaine et al. 2011). The authors proposed that co-attraction and contact inhibition must act in concert to allow cell clusters to self-organise and respond efficiently to external signals, such as chemoattractants and repellents. Therefore anaphylatoxins seem to play both positive and negative roles depending on the physiopathological context; hence their artificial modulation to improve healing still requires further research.
The adult myocardium has recently been shown to harbour multipotent progenitor cells that can give rise to both myogenic and vasculogenic lineages, which can both contribute to myocardial repair (reviewed in (Barile et al. 2007; Laflamme and Murry 2011)). On the other hand, since the myocardium has a low endogenous regenerative competence, loss of a substantial amount of cardiac muscle ultimately results in scar formation. Inflammatory signals are required to guarantee the optimal creation of a supportive scar in the injured tissue, modulating phenotype function and gene expression in fibroblasts, endothelial cells, and leukocytes which together control collagen, fibroblast/myofibroblast deposition and vascular network formation. An optimal inflammatory reaction leads to stable scar formation, and the low regenerative potential of cardiac tissue indicates that, at least in a physiological situation, CPCs should behave preferentially as a healing cell population that participates in scar formation, since they have a very low potential for cardiomyocyte replacement. Telomerase-competent CPCs with long telomeres are present in the atria and apex storage regions of the heart; following activation by growth factors they migrate to damaged areas, where they although have the potential to create a population of young myocytes (Itzhaki-Alfia et al. 2009) their contribution to regenerated cardiac tissue is very low. Experimental data suggest that the regenerative potential of endogenous adult stem cells is low, and multiple reports show that defects in telomere maintenance impairs organ regeneration like liver (Hartmann et al. 2011) or in haematopoietic stem cell maintenance (Calado and Young 2008). In addition, telomere shortening during progenitor cell proliferation affects the function of the brain, pancreas, bone marrow, and heart, pointing to stem cell dysfunction as a critical determinant of organ aging/regeneration (Beausejour and Campisi 2006; Harrington and Greider 1991). Therefore efficient pro-regenerative signalling should modulate telomerase activation, allowing efficient cell proliferation, and also modulation of cell differentiation towards a healing phenotype.
In this work we describe the role of complement anaphylatoxins in CPC biology and prove that C3a and C5a are able to enhance cell migration and proliferation. This effect is linked to telomerase mRNA activity and activation, and a partial preservation of telomere length. Conversely anaphylatoxin stimulation influences CPC fate, pushing them towards myofibroblast differentiation, reduces endothelial gene expression, and increases collagen and smooth muscle gene expression, thus supporting a role for them in cardiac scar formation. Hence, enhanced proliferation ability and telomere length maintenance could denote that both C3a and C5a anaphylatoxins can help to maintain the cardiac resident cell pool within the heart during injury, and facilitate their function as scar forming cells.
CPC culture and characterisation
CPCs express functional anaphylatoxin receptors
C3a and C5a promote CPC proliferation and migration
In some models cell proliferation is linked to cell migration (Diez-Juan and Andres 2003). Interestingly, in addition to the modulation of the chemotactic activity of immune cells, C3a and C5a has also been shown to induce migration of non-immune cells such as neural stem cells (Shinjyo et al. 2009) or mesenchymal stem cells (Schraufstatter et al. 2009), between others. Therefore we next explored the potential of C3a and C5a as chemotactic factors in CPCs. Using a Boyden chamber, CPCs were placed in the upper transwell compartment and stimulated with C3a or C5a in the lower compartment (Figure 4B). Both anaphylatoxins were able to induce cell migration when present at the same concentration that induced cell proliferation in the previous assay. C5a was more efficient at inducing cell migration at lower doses; at higher concentrations the stimulation was reduced, resulting in a bell-shaped response curve (Figure 4B), which may be explained by the fact that this agent primarily stimulates chemotaxis. In fact, at higher concentrations, diffusion of C5a from the lower to the upper compartment of the Boyden chamber could disrupt the C5a gradient and thus prevent chemotactic migration. In contrast, the use of a chemoattractant that stimulates chemokinesis would not result in a bell-shaped dose–response, since such a reaction is independent of chemical gradients.
Anaphylatoxins C3a and C5a induce telomerase activity and telomere maintenance in CPCs
It is well known that cell proliferation is dependent on telomere maintenance, which is mainly achieved through the action of telomerase (Tert) (Choudhary et al. 2012). Moreover, multiple experimental data show that telomere length maintenance is crucial to the preservation of the regenerative potential of adult endogenous stem cells (reviewed in (Choudhary et al. 2012)). Thus, we assessed if C3a and C5a could be considered as a pro-healing stimuli by activating CPC proliferation in combination with telomerase induction.
The induction of TERT expression by anaphylatoxins C3a and C5a is dependent of NFκB activation
To gain insight into the molecular mechanisms responsible for C3a and C5a anaphylatoxin mediated Tert induction we examined the Tert promoter and found two NFκB binding sites. Since C3a and C5a are able to induce NFκB signalling in CPCs, this molecular pathway appeared to be a good candidate for the mediation of C3a and C5a dependent Tert mRNA expression.
Anaphylatoxins C3a and C5a abolish the cardiac and endothelial potential of CPCs and promote myofibroblastic differentiation mediated by an endothelial to mesenchymal transition-like process
Although some experimental results have shown a limited regenerative response after heart injury, cardiac wound healing in mammals is hampered by the fact that the regeneration of heart muscle is virtually absent, and that damaged myocardium is replaced by scar tissue. The above results point towards pro-regenerative activity for complement anaphylatoxins: both of them promote CPC migration, proliferation, telomerase activity, and telomere maintenance. All of these anaphylatoxin-mediated processes can be envisioned to activate and expand the resident CPC pool within the heart. However, to be effective in regenerating damaged tissue, a stem cell pool must preserve its differentiation potential when it proliferates. A pro-regenerative event would mean that CPCs maintain their ability to differentiate into the main cardiogenic lineages (smooth muscle/myofibroblast, endothelium and cardiomyocytes), or are even able to enhance their ability to differentiate towards a cardiomyocyte fate, the main cell type that is not replenished after MI. Therefore we examined the effect of CPC exposure to complement anaphylatoxins.
Immunofluorescence analysis showed that when CPCs are in their multipotential state, they express moderate levels of both myofibroblast and endothelial markers (Figure 7) and so they can be considered as being in a state of balance between these two fates. Complement anaphylatoxin stimulation seems to break this balance, pushing CPCs towards the myofibroblast lineage. We then hypothesized about the mechanism underlying this C3a/C5a-dependent balance shift and noticed that the decrease in the endothelial markers could point to an endothelial to mesenchymal like transition (EndTM), which would be capable of altering this equilibrium. To test this hypothesis, we treated CPCs with the optimal C3a and C5a doses and measured the mRNA levels of two key transcription factors involved in EndMT, Snail1 and Twist1 (Huber et al. 2005). Figure 7A shows that cells treated with C3a and C5a presented higher levels of Twist1 (almost 2-fold) and Snail (> 5-fold) mRNAs. We also assessed the expression of the known EndMT marker Fsp1 and confirmed the C3a/C5a-dependent induction of an EndMT-like process; CPCs treated with these anaphylatoxins showed a 5-fold increase in the Fsp1 mRNA levels (Figure 7A).
Therefore a likely explanation for the observed results is that C3a and C5a promote an EndMT-like process in CPCs that restricts their endothelial potential and pushes them towards the mesenchymal fate inducing the expression of the myofibroblastic lineage promoters and thus committing them to differentiate towards myofibroblasts that express the typical markers Desmin, α-SMA and, SM22α. In addition, the cardiogenic potential of CPCs is abolished upon C3a and C5a stimulation, since this negates C3a and C5a dependent commitment of CPCs into myofibroblasts. Expression of the key pioneer transcription factor in cardiogenesis, Gata4, was clearly repressed upon anaphylatoxin treatment (Figure 7A); however the expression of other cardiogenic factors did not follow this pattern. When CPCs were treated with C3a and C5a, we found that while Gata4 mRNA decreased 5-fold, the mRNA levels of its downstream factors Nkx2.5, Tbx5 and Tbx3 markedly increased (2.6 to 17-fold), along with an increase in mRNA from the early cardiac marker Actc1, which rose by 3 - 4-fold.
Gata4 is known to be crucial for cardiomyocyte differentiation; therefore, the repression of Gata4 expression in CPCs stimulated by C3a and C5a clearly points toward a role for these factors in the compromise of cardiogenic differentiation potential in these CPCs. A role for C3a and C5a in the compromise of cardiogenic differentiation is further supported by the appearance α-SMA positive cells and by the induction of myofibroblast markers in C3a/C5a treated CPCs, since this process is obstructing cardiomyocyte differentiation (Wu et al. 2006). Therefore the C3a and C5a mediated transcriptional activation of Tbx3, Tbx5 and Nkx2.5 might suggest an alternative role for these transcription factors in the maturation of and/or functional properties of myofibroblasts. Another interesting possibility is the implication of these mesodermal factors in the C3a/C5a mediated proliferation of CPCs.
Thus, we conclude that C3a and C5a promote the differentiation and commitment of CPCs towards myofibroblast lineage, blocking their cardiac and endothelial potential.
Cell culture and stimulation
Sca1+ CPCs were grown on gelatin at 0.1% (v/v) in CSPH Media, composed of 50% Neurobasal Medium, 50% DMEM/F12, 1% 100 μg/ml penicillin and streptomycin, 1% 2 mM L-Glutamine, 0.1% 250 ng/ml Fungizone, 0.1% 50 mg/ml Gentamicin and 10% Stem Cell Quality (ESQ)-FSB, supplemented with LIF 10 ng/ml, 1% ITS, FGF 10 ng/ml, EGF 20 ng/ml, 1% N2 and 2% B27. C3a (FIDCCNH ITKLREQHRR DHVLGLAR) and C5a (AFNECCTIAN KIRKESPHKP VQLGR) synthetic peptides were synthetised by GeneScript USA, NJ. Inmortalized endothelial cell line was a gift of Dr. Alfonso Luque and has been previously described (Hortelano et al. 2010). Mouse embryonic fibroblasts (MEFs) were generated from mice littermates in C57BL/6 J background harvested from embryonic day12-13.5. Briefly, separated embryos from pregnant mice had their heads and livers removed and incubated at 37°C in the presence of 0.25% Trypsin/EDTA for 5 minutes in a humidified incubator. Trypsinized embryos were homogenized by pipetting up and down and allowed to sediment for 5 minutes, and cells sedimenting in the lower phase were collected and plated in fresh DMEM containing 10% FBS. If not specified culture mediums and supplements were obtained from Invitrogen, (Carlsbad, CA). Growth factors were purchased from Peprotech (Oak Park, CA)
CPCs were exposed to differentiation medium for 7 and 14 days, composed of DMEM, 1% 100 μg/ml penicillin and streptomycin, 1% 2 mM L-Glutamine, 10% Stem Cell Quality (ESQ)-FSB and 10nM Dexamethasone. Neonatal rat cardiomyocytes were isolated from P2 Wistar rat hearts using the Worthington Neonatal Cardiomyocyte Isolation System (Worthington, Lakewood, NJ). CPCs for co-culture experiments were transduced with pRRL.CMV.GFP lentiviral vector supernatants provided by CNIC viral vector facility, and cells with high GFP expression were selected by FACS. For the co-culture, NRCMs were seeded at 7.5 × 104 cells/cm2 on 75% M199, 25% DMEM, 5% FBS, 10% HS, 1% 100 μg/ml penicillin and streptomycin and 1% 2 mM L-Glutamine, and on the next day GFP+ CPCs were added at 3 × 103 cells/cm2.
Proliferation was measured by [3H]thymidine incorporation performed in triplicate as previously described (Diez-Juan and Andres 2003). Briefly cell cultures were plated on 24-well culture dishes at a density of 5 × 103 cells per well and incubated for 4 hours with [3H]thymidine (1 μCi/mL, Amersham) after which cells were washed with PBS and incubated for 1 h with ice-cold 5% trichloroacetic acid. Labeled DNA was extracted with 0.5 N NaOH and counted in a scintillation counter (Wallac, Turku, Finland).
CPCs were plated at 1 × 105 cells/well in the upper well of a transwell (8 μm pore size, Millipore) in 250 μl of serum free DMEM/F12 media supplemented with ITS (Invitrogen, CA, USA). Cells were allowed to migrate towards the lower transwell compartment (500 μl chemotactic gradient (C3a 50 nM/C5a 15 nM) for 3 hours at 37°C. CPCs were mechanically removed from the top side of the transwells and the membranes were fixed for 15 min with 4% PFA, stained with Prolong DAPI (Invitrogen, CA, USA) and counted. Experiments were performed in triplicate.
Cells were cultured in plastic tissue cover slips and fixed with 2% PFA. Cells were then washed twice with PBS, blocked with 10% horse serum and incubated with antibodies overnight. Specific fluorescent secondary antibodies were use to visualise immunolabelling. We used anti- Tropomyosin (Sigma), anti-C5aR/cd88 (AM01269PU-S, Acris Gmbh), anti-C3aR (sc-14624, Santa Cruz), anti-α-SMA (1A4, A2547, Sigma) and CD31 (MEC13.3, BD Pharmigen). Biotinilated anti goat IgG (Invitrogen) was added after 3 washes with PBS-T. Finally anti mouse FITC or Cy3 streptavidin (Invitrogen) was added.
Cells were lysated with RIPA buffer. The protein concentration was quantified by Protein detection kit "DC protein assay" (BIO-RAD, Hercules, CA, US), and 10 μg of protein was loaded into a 12% acrilamide gel. The proteins were transferred with iBlot Dry Blotting System (Invitrogen), and the membranes were blocked with 10% dry milk in PBS-T. The membranes were then incubated overnight at 4°C with the primary antibody in blocking buffer. This was followed by washing and 1 hour incubation with an appropriate peroxidase-conjugated secondary antibody. Protein bands were detected by chemiluminiscence using a commercial kit (Pierce) according to the manufacturer’s instructions. Primary antibodies were: anti p65, anti iKKa, anti p-IKKa anti pERK, anti ERK1/2, anti pPKC, anti PKC, anti p-AKT anti and AKT from Cell Signaling; rat IgG (Millipore), anti α-Actinin murine IgG (Sigma), anti Tubulin murine IgG (BD biosciences), anti C3aR Antibody (H-300) sc-20138 and anti C5aR(CD88) (H-100) sc-25774 (Santa Cruz Biotechnology). Secondary antibodies were HRP-anti rat IgG (Dako) and HRP-anti mouse IgG (Dako). If stated, densitometric analysis of the blots was performed with ImageJ software.
RNA extraction and nonQuantitative and quantitative real-time polymerase chain reaction (RTq-PCR)
SM 22 alpha Fw
SM 22 alpha Rv
TCC AAA CCC ACT CGG ATG TGA AGA
TTG GTG CTT GTG GAG CAA GGA CAT
Troponin T Fw
Troponin T Rev
AAT TCA CAA GAA TCA GGG CGT GGG
TCT ATC AGA ATG CAG AGG TGT GGG
Telomere length quantification by Q-FISH
FISH was carried out as described above, using a Cy3- labelled LL(CCCTAA)3 peptide nucleic acid (PNA) telomeric probe (Euregentec) and a FITC labelled LL(ATTCGTTGGAAACGGGA) PNA alpha satellite probe (Eurogentec, Liège) as previously described (Samper et al. 2001) with the following modifications. After hybridisation slides were washed three times with PBS supplemented with 0.1% Tween for 10 min at 60°C and dehydrated through an ethanol series (70%, 90%, and 100%) for 5 min each. Slides were then counterstained and mounted in Vectashield H-1200 mounting medium. Digital images were acquired as described above. Telomere signals were captured with the same exposure time in all samples and the telomere length was represented in relative fluorescence units. Telomere signals from at least 20–30 nuclei per group were quantified using the TFL-Telo (version 2) kindly provided by Dr Peter Lansdorp (British Columbia Cancer Centre, Vancouver). All images were captured and analysed in parallel on the same day by an experimenter blinded to the treatment groups.
Measurement of telomerase activity by TRAP assay
Telomerase assays were performed on 5000 CPCs as described (Banerjee and Jagadeesh 2009), with the following modifications. Protein extracts were made using lysis buffer NP40 (10 mM Tris–HCl, 1 mM MgCl2, 1 mM EDTA, 1%NP40, 0.25 mM Sodium deoxycholate, 10% glycerol, 150 mM NaCl, 5 mM β-mercaptoethanol (all from Sigma-Aldrich) and protease inhibitor 1x (Roche, Basel, Switzerland) at pH 8.0. The protein concentration was quantified by Protein detection kit "DC protein assay" (BIO-RAD, Hercules, CA, US). Were incubated 5 and 1 μg of protein per sample with telomerase extension buffer (500 mM Tris-AcH, 500 Mm AcK. 30 mM MgCl2, 10 mM spermin, 10 mM EGTA, 50 mM β-mercaptoethanol, 2 mM dAGT (all from Sigma-Aldrich) and 1 mM of Oligo TS (5´ - AAT CCG TCG AGC AGA GTT −3`. Extension reaction of telomerase (2 μl) was added to 23 μl PCR reaction mix containing 1 x Power SYBR Green PCR Master Mix, 5 mM EGTA, 2 ng/μl Oligo TS and 4 ng/μl Oligo ACX (5´ GCG CGG C(TTACCC)4. PCR was carried out at 94°C for 10 min followed by 40 cycles of, 94°C for 15 s and 60°C for 1 min. PCR products were monitored with a ABI PRISM 7700 sequence detection apparatus (Applied Biosystem, Foster City, CA, USA) and analyzed with SDS v2.3 software (Applied Biosystem, CA).
Chromatin immunoprecipitation assays
Chromatin immunoprecipitation (ChIP) experiments were performed using the EZ-Magna ChIP G kit (Millipore Corporation), according to the manufacturer's protocol, with slight modifications. Approximately 107 CPCs were used for each experiment. Two millilitres of cell lysate were sonicated using a Microson Ultrasonic Cell Disruptor. Samples were kept cool in an ethanol-ice-water bath and sonicated for four 30 s pulses with 1 min pauses between them. Fragmented chromatin was diluted in ChIP dilution buffer (Millipore EZ-Magna ChIP kit), distributed in 300 μl aliquots, and used directly or stored at −80°C. The anti-p65 (ab31481) and anti-histone H3 (ab1791) antibodies were purchased from Abcam. Following immunoprecipitation, DNA was analysed by real-time PCR. To identify NFκB binding sites, 3Kb upstream of murine Tert gene were analyzed using PROMO (Messeguer et al. 2002). One site was found at position chr13:73764072–73764172 at -322 bp of mTert star site within dissimilarity margin less or equal than 5%. The following primers were used TertFw1: GCCCAAACCTCGCCCCAGTC, TertFw2: CCCAAACCTCGCCCCAGTCT, TertRw1: TGCTTCTCGGGCTTCCTTTTCGC, TertRev2: ACAGAATGCTTCTCGGGCTTCCT, ActinFw: CCCAAGGGGCGCATTGGCAT, ActinRw: GGGAGGGGGCAGTCAGAGCA.
Results are reported as mean ± SD. Data were normalized versus control. Three independent experiments were analyzed. All values are expressed as mean ± SD. Statistical analysis were performed with Student's T-test for unpaired samples and normally distributed data sets. Statistically significant differences were considered at P < 0.05. (Statview, SAS Institute).
In this study we have demonstrated that (a) CPCs express functional anaphylatoxin receptors, (b) stimulation with C3a or C5a induces the expression of myofibroblast differentiation markers and EndMT-like gene expression, (c) C3a and C5a both increase CPC proliferation and migration and (d) C3a and C5a also induce telomerase activity (NFκB-dependent) and increase telomere maintenance, although not sufficiently to fully preserve telomere length. Taken together these data support a role for anaphylatoxins in CPC regulated cardiac healing and scar formation.
The discovery of endogenous c-kit–positive CPCs has driven a shift in cardiac biology. Work from multiple laboratories (revised in Barile et al. 2007; Laflamme and Murry 2011) have documented the heart’s ability to replace old and dying cells, and propose that this capacity depends on the persistence of a stem cell compartment. The initial interest generated by the identification and isolation of c-kit-positive CPCs was followed by a period of doubt, reflected in studies from different laboratories and accompanying editorials that questioned or neglected issues related to CPC function. These reservations regarding the role of CPCs in cardiac cell turnover are mainly related to the limited nature of myocyte renewal in the human heart (Bergmann et al. 2009). Although multiple studies in animal models have suggested that the adult heart is capable of some cellular turnover (reviewed in Choi et al. 2012), clinical evidence clearly shows that regeneration is inadequate after injury. This evidence supports the hypothesis that endogenous CPC populations probably have a role in cardiac homeostasis other than cardiomyocyte turnover, since the cardiac healing response is very poor. In adult mammals, cardiomyocyte regeneration is insufficient to functionally renew severely injured myocardium and consequently, scar tissue forms. Thus, the evolutionary adapted healing processes that initially benefited cardiac function finally results in a potentially maladaptive response in the long term.
After a myocardial infarction, massive cell infiltration into the myocardium results in fibrosis. MI injury is considered an inflammatory state, characterised by innate immune responses. Whereas it was initially thought that the immune system protects the body against foreign or non-self signals, there is now a body of evidence to suggest that the innate immune system is activated following tissue injury, triggering the release of endogenous ligands (Arslan et al. 2011). Participation of the complement system in myocardial ischemia was first demonstrated in 1971 in a rat MI model (Hill and Ward 1971). Administration of monoclonal antibodies against C5 and C5a as well as the C5a receptor reduced myocardial infarct size in pigs and rodents (Amsterdam et al. 1995; Vakeva et al. 1998; van der Pals et al. 2010; Zhang et al. 2007). However, a study concerning patients either undergoing percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) with pexelizumab (a single-chain fragment of a humanised monoclonal antibody against complement component C5) administration showed a significant reduction in mortality in the former case, but no benefit in patients with acute MI in the latter case (Testa et al. 2008). It is important to note that anaphylatoxin inhibition in these models has been performed in an acute MI phase but that the principal role of anaphylatoxins and their receptors here is to increase leukocyte recruitment to the reperfused myocardium following MI.
Since the complement system is activated in the milieu of tissue injury, the effect of C3a and C5a on CSC migration and proliferation is of specific interest, because these anaphylatoxins may contribute to the mobilisation, recruitment, and proliferation of cells at an injury site, thus facilitating wound healing. Our data support the role of anaphylatoxins in recruiting cells involved in tissue healing: C3a and C5a are able to induce CPC proliferation and migration, which supports the role of anaphylatoxins in the control of tissue healing. Indeed, apart from the known chemotactic role that C3a and C5a play with regard to leukocytes, anaphylatoxins are also chemotactic factors for other cells like mesenchymal stem cells (Schraufstatter et al. 2009) or neural stem cells (Shinjyo et al. 2009). Our data also shows that in addition to this enhancement in proliferation and migration, anaphylatoxins also induce telomerase mRNA and activity. Although addition of C5a significantly increases telomere length compared to controls, and to a lesser degree C3a, telomerase activity is not enough to maintain telomere length compared to the initial state. Interestingly a large body of data has accumulated regarding the capacity of telomerase to support roles other than telomere preservation (reviewed in Martinez and Blasco 2011). One of these roles is the facilitation of cell growth and proliferation. For instance, hTERT overexpression lengthens the proliferative lifespan of human bone marrow stromal cells (Simonsen et al. 2002) being also able to induce hyperplasia and hypertrophy in murine cardiac myocytes (Kohl 2001). Moreover, ectopically expressed hTERT confers resistance to the anti-proliferative effect of transforming growth factor β (TGF-β) in p16-null human mammary epithelial cells (Stampfer et al. 2001). In addition, telomerase activation in human mammary epithelial cells coincides with the stimulation of a cellular mitogenic program (Smith et al. 2003), indicating that telomerase may affect epithelial cell proliferation not only by stabilising telomeres, but also by affecting the expression of growth-promoting genes. There is also data that supports an extra-telomeric role for telomerase in protection against oxidative stress; in this regard Schraufstatter et al. (2009) demonstrated that C3a and C5a protect MSCs from oxidative damage. This can be translated into an in vivo scenario, where one would expect that the anaphylatoxins would recruit healing cells to areas of tissue injury, where these cells would encounter the production of oxidants due to neutrophil recruitment and reperfusion injury. Therefore all these data support a role for anaphylatoxins in promoting the recruitment of healing cells to areas of injury in order to restore tissue homeostasis. Telomerase activity would not only improve cell proliferation and survival but also protect cells against the hostile environment – providing protection from apoptosis, oxidative stress, and DNA damage.
Injured ischemic myocardium progresses to necrosis and subsequent healing; the local response to irreversible injury is the formation of granulation tissue, the accumulation of collagen, and resultant replacement fibrous scar (Sun et al. 2002). Although clinical and biological data shows that the heart has a very low endogenous capacity for regeneration when severe damage is inflicted, multiple endogenous putative cardiac stem cell or progenitor cell populations have been identified and isolated. Markers, traditionally associated with blood, bone marrow or pluripotent stem cells, have been used by several independent groups to identify these cells in adult or postnatal hearts in humans and other mammalian species (Smith et al. 2008). We have worked with a population (CPCs), as described by Messina et al. (Messina et al. 2004), this population expresses the cardiac markers, Nkx2.5, Gata4, Cx43 and Mef2c (Figure 1B) and therefore had a putative cardiogenic potential. Thus our first task was to understand why, if these cells have the potential to differentiate in cardiomyocytes, physiological cardiac healing usually results as a fibrotic scar and shows very low levels of cardiomyocyte replacement. We investigated the potential of these cells to differentiate to all three lineages described, and facilitated differentiation by using 10% FBS to improve myofibroblast differentiation, and EGM on Matrigel to increase endothelial lineage differentiation. Interestingly anaphylatoxins repress Gata4 expression and increase the expression of some early cardiac markers (Nkx2.5, Tbx3, Tbx5, Actc1) in CPCs. Endothelial cell (EC) lineage differentiation was clearly improved by culture on Matrigel using EGM media and in these conditions, stimulation with C3a and C5a anaphylatoxins reduced endothelial gene expression. In agreement with this data, immunofluorescence also showed the absence of ECs in the presence of anaphylatoxins. Inhibition of EC differentiation is paralleled by induction of Twist and Snail genes together with Fsp1 expression. Fsp1 is one of the markers of fibroblast formation and is a cytoskeletal protein belonging to the calmodulin-S100-troponin C superfamily of intracellular calcium binding proteins associated with cytoskeletal fibres, cell motility, and mesenchymal phenotype (Zimmer et al. 1995). Therefore the expression of Snail1 and Twist1 genes, together with the significant increase in Fsp1 expression, indicated that an endMT-like differentiation event might be occurring in CPCs. In addition, anaphylatoxins induce the expression of an array of genes associated with myofibroblasts that probably abolishes their role in regeneration but hints at their relevance in the cardiac healing process that follows MI. In this scenario, complement anaphylatoxins would activate migration of CPCs to the damaged area, induce their proliferation and finally, upon persistence of local signalling, would promote their differentiation into myofibroblast cells that would be able to fill and quickly repair the damaged heart area, preventing further cardiac complications.
Gata4 has been proved to be a critical early factor in cardiogenesis, it lays upstream of Tbx5, Nkx2.5, and Act1 in the cardiogenic transcription network, and acts together with Baf60c to generate a chromatin state competent for cardiogenic differentiation (Takeuchi and Bruneau 2009). Therefore C3a and C5a may block cardiomyocyte differentiation by repressing Gata4 expression. The cardiogenic factors Nkx2.5, Actc1 and specially Tbx3 and Tbx5 may have roles other than their cardiogenic ones, including modulating the response of CPCs to complement anaphylatoxins, for instance they might be involved in the C3a/C5a dependent myofibroblast differentiation process. It is known that several mesodermal cells (cardiomyocytes and myofibroblasts) share parts of their transcriptional differentiation networks, supporting the existence of a common myocardial and smooth muscle cell precursor in the developing embryo. This is consistent with in vivo studies, which show the co-expression of numerous smooth muscle genes in myocardial progenitor cells (Wu et al. 2006). Therefore Tbx3 and Tbx5 would be common to both cell types, and Gata4 levels would be critical in modulating the fate decision of progenitor cells. Interestingly it has been proposed that Gata4 expression can play a role in the developmental regulation of cardiac fibroblasts and has a function in the maintenance of cardiac-resident progenitors (Jankowski 2009). This raises the hypothesis that C3a and C5a are factors that affect CPC fate by switching the balance of different lineage specific factors: they shut down the expression of endothelial and cardiac factors and induce the expression of myofibroblast factors (TCF21, SM22α and Myocardin), irreversibly pushing CPCs towards the myofibroblast fate.
Taken together, these findings shed some light onto what have been described as endogenous cardiac stem cells and some mechanistic insight about the limited potential of CPCs to generate new cardiomyocytes after cardiac injury. We can hypothesize that anaphylatoxin release at early time after cardiac injury is able to increase CPC numbers and to promote migration towards the site of injury. This signal also increases CPC potential to differentiate into myofibroblast lineages that would participate in scar formation. This situation gives an advantage to cardiac tissue promoting a fast healing in MI. Differentiation toward myofibroblast in CPCs would make available a higher number of myofibroblasts to contribute to other sources of myofribroblasts (Krenning et al. 2010) during the resolution stage of MI. This situation is different from physiological cardiac cell turnover. During life possibly other signals in absence of inflammatory signals will permit cardiomyocyte differentiation to replace exhausted cells. Interestingly recent evidence reported by Jianqin Ye et al. (2012) have shown that there is a significant increase in the proliferative capacity of CS-forming cells isolated from the “middle aged” heart following acute MI resulting in a significant rise in the number of CSs in vitro and this increase is most pronounced within the first week post-MI. In addition they show that show for the first time that the CS cells obtained from 1-week post-MI hearts engraft in ischemic myocardium and restore cardiac function at 25 days post-injection in vivo. However, we did not find evidence for differentiation of these cells into mature cardiomyocytes or new vessels althouhgt promoted angiogenesis in vivo. Their data suggest that early signals that happens after MI would commit these cells to a more healing phenotype instead of regenerative fate.
To heal the heart adding new cardiomyocyte probably is a more delicate and time-consuming situation that mammalian wounded heart cannot go through in a wild environment. Thus the signals involved in cardiac healing have been selected to permit a fast healing situation increasing myofibroblast differentiation but in other hand this impairs cardiomyocyte renewal. Deeper understanding of these mechanisms would help to improve wounded heart regeneration.
We thank Dr Alfonso Luque for immortalized mouse endothelial cells. All members of Diez-Juan laboratory for their helpful discussions and technical support, specialy to Virginia Zorita, Nuria Martí and Vanessa Blanca, Microscopy and Cellomics Units, at CIPF and Maria Ledran at EFL for Scientific editing.
This work was supported by grant to ADJ from the Ministry of Science and Innovation (SAF 2009–1000) and a MICCIN "Ramon y Cajal" contract and by a grant to ABM from GENOMA ESPAÑA (Spain) part of the MACIA research project.
- Amsterdam EA, Stahl GL, Pan HL, Rendig SV, Fletcher MP, Longhurst JC: Limitation of reperfusion injury by a monoclonal antibody to C5a during myocardial infarction in pigs. Am J Physiol 1995, 268: H448-H457.
- Arslan F, de Kleijn DP, Pasterkamp G: Innate immune signaling in cardiac ischemia. Nat Rev Cardiol 2011, 8: 292-300. 10.1038/nrcardio.2011.38View Article
- Banerjee PP, Jagadeesh S: Non-radioactive assay methods for the assessment of telomerase activity and telomere length. Methods Mol Biol 2009, 523: 383-394. 10.1007/978-1-59745-190-1_25View Article
- Barile L, Messina E, Giacomello A, Marban E: Endogenous cardiac stem cells. Prog Cardiovasc Dis 2007, 50: 31-48. 10.1016/j.pcad.2007.03.005View Article
- Beausejour CM, Campisi J: Ageing: balancing regeneration and cancer. Nature 2006, 443: 404-405. 10.1038/nature05221View Article
- Bergmann O, Bhardwaj RD, Bernard S, Zdunek S, Barnabe-Heider F, Walsh S, Zupicich J, Alkass K, Buchholz BA, Druid H, et al.: Evidence for cardiomyocyte renewal in humans. Science 2009, 324: 98-102. 10.1126/science.1164680View Article
- Brown K, Gerstberger S, Carlson L, Franzoso G, Siebenlist U: Control of I kappa B-alpha proteolysis by site-specific, signal-induced phosphorylation. Science 1995, 267: 1485-1488. 10.1126/science.7878466View Article
- Busche MN, Stahl GL: Role of the complement components C5 and C3a in a mouse model of myocardial ischemia and reperfusion injury. Ger Med Sci 2010., 8: doi:pii: Doc20. PubMed PMID: 20930931; PubMed Central PMCID: PMC2940219. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2940219/
- Calado RT, Young NS: Telomere maintenance and human bone marrow failure. Blood 2008, 111: 4446-4455. 10.1182/blood-2007-08-019729View Article
- Carmona-Fontaine C, Theveneau E, Tzekou A, Tada M, Woods M, Page KM, Parsons M, Lambris JD, Mayor R: Complement fragment C3a controls mutual cell attraction during collective cell migration. Dev Cell 2011, 21: 1026-1037. 10.1016/j.devcel.2011.10.012View Article
- Choi SH, Jung SY, Kwon SM, Baek SH: Perspectives on stem cell therapy for cardiac regeneration. Advances and challenges. Circ J 2012, 76: 1307-1312. 10.1253/circj.CJ-11-1479View Article
- Choudhary B, Karande AA, Raghavan SC: Telomere and telomerase in stem cells: relevance in ageing and disease. Front Biosci (Schol Ed) 2012, 4: 16-30.View Article
- Diez-Juan A, Andres V: Coordinate control of proliferation and migration by the p27Kip1/cyclin-dependent kinase/retinoblastoma pathway in vascular smooth muscle cells and fibroblasts. Circ Res 2003, 92: 402-410. 10.1161/01.RES.0000059306.71961.EDView Article
- Frangogiannis NG: Regulation of the inflammatory response in cardiac repair. Circ Res 2012, 110: 159-173. 10.1161/CIRCRESAHA.111.243162View Article
- Harrington LA, Greider CW: Telomerase primer specificity and chromosome healing. Nature 1991, 353: 451-454. 10.1038/353451a0View Article
- Hartmann D, Srivastava U, Thaler M, Kleinhans KN, N'Kontchou G, Scheffold A, Bauer K, Kratzer RF, Kloos N, Katz SF, et al.: Telomerase gene mutations are associated with cirrhosis formation. Hepatology 2011, 53: 1608-1617. 10.1002/hep.24217View Article
- Hill JH, Ward PA: The phlogistic role of C3 leukotactic fragments in myocardial infarcts of rats. J Exp Med 1971, 133: 885-900. 10.1084/jem.133.4.885View Article
- Hortelano S, Lopez-Fontal R, Traves PG, Villa N, Grashoff C, Bosca L, Luque A: ILK mediates LPS-induced vascular adhesion receptor expression and subsequent leucocyte trans-endothelial migration. Cardiovasc Res 2010, 86: 283-292. 10.1093/cvr/cvq050View Article
- Huber MA, Kraut N, Beug H: Molecular requirements for epithelial-mesenchymal transition during tumor progression. Curr Opin Cell Biol 2005, 17: 548-558. 10.1016/j.ceb.2005.08.001View Article
- Itzhaki-Alfia A, Leor J, Raanani E, Sternik L, Spiegelstein D, Netser S, Holbova R, Pevsner-Fischer M, Lavee J, Barbash IM: Patient characteristics and cell source determine the number of isolated human cardiac progenitor cells. Circulation 2009, 120: 2559-2566. 10.1161/CIRCULATIONAHA.109.849588View Article
- Jankowski M: GATA4, a new regulator of cardiac fibroblasts, is sensitive to natriuretic peptides. Cardiovasc Res 2009, 84: 176-177. 10.1093/cvr/cvp297View Article
- Kloner RA, Fishbein MC, Lew H, Maroko PR, Braunwald E: Mummification of the infarcted myocardium by high dose corticosteroids. Circulation 1978, 57: 56-63. 10.1161/01.CIR.57.1.56View Article
- Kohl J: Anaphylatoxins and infectious and non-infectious inflammatory diseases. Mol Immunol 2001, 38: 175-187. 10.1016/S0161-5890(01)00041-4View Article
- Krenning G, Zeisberg EM, Kalluri R: The origin of fibroblasts and mechanism of cardiac fibrosis. J Cell Physiol 2010, 225: 631-637. 10.1002/jcp.22322View Article
- Laflamme MA, Murry CE: Heart regeneration. Nature 2011, 473: 326-335. 10.1038/nature10147View Article
- Martinez P, Blasco MA: Telomeric and extra-telomeric roles for telomerase and the telomere-binding proteins. Nat Rev Cancer 2011, 11: 161-176.View Article
- Messeguer X, Escudero R, Farre D, Nunez O, Martinez J, Alba MM: PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics 2002, 18: 333-334. 10.1093/bioinformatics/18.2.333View Article
- Messina E, De Angelis L, Frati G, Morrone S, Chimenti S, Fiordaliso F, Salio M, Battaglia M, Latronico MV, Coletta M, et al.: Isolation and expansion of adult cardiac stem cells from human and murine heart. Circ Res 2004, 95: 911-921. 10.1161/01.RES.0000147315.71699.51View Article
- Mocco J, Mack WJ, Ducruet AF, Sosunov SA, Sughrue ME, Hassid BG, Nair MN, Laufer I, Komotar RJ, Claire M, et al.: Complement component C3 mediates inflammatory injury following focal cerebral ischemia. Circ Res 2006, 99: 209-217. 10.1161/01.RES.0000232544.90675.42View Article
- Monsinjon T, Gasque P, Chan P, Ischenko A, Brady JJ, Fontaine MC: Regulation by complement C3a and C5a anaphylatoxins of cytokine production in human umbilical vein endothelial cells. FASEB J 2003, 17: 1003-1014. 10.1096/fj.02-0737comView Article
- Pan ZK: Anaphylatoxins C5a and C3a induce nuclear factor kappaB activation in human peripheral blood monocytes. Biochim Biophys Acta 1998, 1443: 90-98. 10.1016/S0167-4781(98)00198-5View Article
- Proctor LM, Arumugam TV, Shiels I, Reid RC, Fairlie DP, Taylor SM: Comparative anti-inflammatory activities of antagonists to C3a and C5a receptors in a rat model of intestinal ischaemia/reperfusion injury. Br J Pharmacol 2004, 142: 756-764. 10.1038/sj.bjp.0705819View Article
- Rahpeymai Y, Hietala MA, Wilhelmsson U, Fotheringham A, Davies I, Nilsson AK, Zwirner J, Wetsel RA, Gerard C, Pekny M, Pekna M: Complement: a novel factor in basal and ischemia-induced neurogenesis. EMBO J 2006, 25: 1364-1374. 10.1038/sj.emboj.7601004View Article
- Ratajczak J, Reca R, Kucia M, Majka M, Allendorf DJ, Baran JT, Janowska-Wieczorek A, Wetsel RA, Ross GD, Ratajczak MZ: Mobilization studies in mice deficient in either C3 or C3a receptor (C3aR) reveal a novel role for complement in retention of hematopoietic stem/progenitor cells in bone marrow. Blood 2004, 103: 2071-2078. 10.1182/blood-2003-06-2099View Article
- Samper E, Flores JM, Blasco MA: Restoration of telomerase activity rescues chromosomal instability and premature aging in Terc−/− mice with short telomeres. EMBO Rep 2001, 2: 800-807. 10.1093/embo-reports/kve174View Article
- Schraufstatter IU, Discipio RG, Zhao M, Khaldoyanidi SK: C3a and C5a are chemotactic factors for human mesenchymal stem cells, which cause prolonged ERK1/2 phosphorylation. J Immunol 2009, 182: 3827-3836. 10.4049/jimmunol.0803055View Article
- Shinjyo N, Stahlberg A, Dragunow M, Pekny M, Pekna M: Complement-derived anaphylatoxin C3a regulates in vitro differentiation and migration of neural progenitor cells. Stem Cells 2009, 27: 2824-2832. 10.1002/stem.225View Article
- Simonsen JL, Rosada C, Serakinci N, Justesen J, Stenderup K, Rattan SI, Jensen TG, Kassem M: Telomerase expression extends the proliferative life-span and maintains the osteogenic potential of human bone marrow stromal cells. Nat Biotechnol 2002, 20: 592-596. 10.1038/nbt0602-592View Article
- Smith LL, Coller HA, Roberts JM: Telomerase modulates expression of growth-controlling genes and enhances cell proliferation. Nat Cell Biol 2003, 5: 474-479. 10.1038/ncb985View Article
- Smith RR, Barile L, Messina E, Marban E: Stem cells in the heart: what's the buzz all about?–Part 1: preclinical considerations. Hear Rhythm 2008, 5: 749-757. 10.1016/j.hrthm.2008.02.010View Article
- Stampfer MR, Garbe J, Levine G, Lichtsteiner S, Vasserot AP, Yaswen P: Expression of the telomerase catalytic subunit, hTERT, induces resistance to transforming growth factor beta growth inhibition in p16INK4A(−) human mammary epithelial cells. Proc Natl Acad Sci U S A 2001, 98: 4498-4503. 10.1073/pnas.071483998View Article
- Strey CW, Markiewski M, Mastellos D, Tudoran R, Spruce LA, Greenbaum LE, Lambris JD: The proinflammatory mediators C3a and C5a are essential for liver regeneration. J Exp Med 2003, 198: 913-923. 10.1084/jem.20030374View Article
- Sun Y, Kiani MF, Postlethwaite AE, Weber KT: Infarct scar as living tissue. Basic Res Cardiol 2002, 97: 343-347. 10.1007/s00395-002-0365-8View Article
- Takeuchi JK, Bruneau BG: Directed transdifferentiation of mouse mesoderm to heart tissue by defined factors. Nature 2009, 459: 708-711. 10.1038/nature08039View Article
- Testa L, Van Gaal WJ, Bhindi R, Biondi-Zoccai GG, Abbate A, Agostoni P, Porto I, Andreotti F, Crea F, Banning AP: Pexelizumab in ischemic heart disease: a systematic review and meta-analysis on 15,196 patients. J Thorac Cardiovasc Surg 2008, 136: 884-893. 10.1016/j.jtcvs.2007.12.062View Article
- Thurman JM, Lenderink AM, Royer PA, Coleman KE, Zhou J, Lambris JD, Nemenoff RA, Quigg RJ, Holers VM: C3a is required for the production of CXC chemokines by tubular epithelial cells after renal ishemia/reperfusion. J Immunol 2007, 178: 1819-1828.View Article
- Vakeva AP, Agah A, Rollins SA, Matis LA, Li L, Stahl GL: Myocardial infarction and apoptosis after myocardial ischemia and reperfusion: role of the terminal complement components and inhibition by anti-C5 therapy. Circulation 1998, 97: 2259-2267. 10.1161/01.CIR.97.22.2259View Article
- Van Beek J, Bernaudin M, Petit E, Gasque P, Nouvelot A, MacKenzie ET, Fontaine M: Expression of receptors for complement anaphylatoxins C3a and C5a following permanent focal cerebral ischemia in the mouse. Exp Neurol 2000, 161: 373-382. 10.1006/exnr.1999.7273View Article
- van der Pals J, Koul S, Andersson P, Gotberg M, Ubachs JF, Kanski M, Arheden H, Olivecrona GK, Larsson B, Erlinge D: Treatment with the C5a receptor antagonist ADC-1004 reduces myocardial infarction in a porcine ischemia-reperfusion model. BMC Cardiovasc Disord 2010, 10: 45. 10.1186/1471-2261-10-45View Article
- Walport MJ: Complement second of two parts. N Engl J Med 2001, 344: 1140-1144. 10.1056/NEJM200104123441506View Article
- Wu SM, Fujiwara Y, Cibulsky SM, Clapham DE, Lien CL, Schultheiss TM, Orkin SH: Developmental origin of a bipotential myocardial and smooth muscle cell precursor in the mammalian heart. Cell 2006, 127: 1137-1150. 10.1016/j.cell.2006.10.028View Article
- Ye J, Boyle A, Shih H, Sievers RE, Zhang Y, Prasad M, Su H, Zhou Y, Grossman W, Bernstein HS, Yeghiazarians Y: Sca-1+ cardiosphere-derived cells are enriched for Isl1-expressing cardiac precursors and improve cardiac function after myocardial injury. PLoS One 2012, 7: e30329. 10.1371/journal.pone.0030329View Article
- Zhang H, Qin G, Liang G, Li J, Barrington RA, Liu DX: C5aR-mediated myocardial ischemia/reperfusion injury. Biochem Biophys Res Commun 2007, 357: 446-452. 10.1016/j.bbrc.2007.03.152View Article
- Zimmer DB, Cornwall EH, Landar A, Song W: The S100 protein family: history, function, and expression. Brain Res Bull 1995, 37: 417-429. 10.1016/0361-9230(95)00040-2View Article
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.