Myocardial infarction (MI) induces the release of pro-inflammatory substances that may affect the function of other tissues(Reference Francis, Chu and Johnson1, Reference Francis, Zhang and Weiss2). For example, we have observed that 3 d after MI, different structures of the limbic system such as the amygdala, the hippocampus or the hypothalamus present an increase in apoptosis(Reference Wann, Bah and Kaloustian3, Reference Wann, Boucher and Kaloustian4). Although the link is not clearly established, this cell death may account for post-MI depression that we have documented in the present experimental model(Reference Wann, Bah and Kaloustian3, Reference Wann, Bah and Boucher5). Reduction in this early apoptosis by pharmacological interventions results in an attenuation of the depressive behaviour(Reference Wann, Bah and Kaloustian3), and thus seems to be beneficial since post-MI depression patients present a three to four times increase in mortality as compared with non-depressive patients(Reference Frasure-Smith, Lesperance and Talajic6, Reference Lesperance, Frasure-Smith and Talajic7).
Pro-inflammatory cytokines are among the different substances that may explain the presence of apoptosis in the limbic system after MI(Reference Wann, Boucher and Kaloustian4). Inhibition of the synthesis of pro-inflammatory cytokines by pentoxifylline is sufficient to prevent apoptosis in the limbic system(Reference Wann, Boucher and Kaloustian4). This observation leads us to predict that interventions that induce a shift in the anti-/pro-inflammatory cytokine ratio must reduce the apoptosis tendency in the limbic system after MI.
Stress conditions, such as depression, may affect other organs that could perpetuate this condition. For example, it has been reported that stress predisposes individuals to develop functional bowel disorders or exacerbate symptoms of irritable bowel syndrome by decreasing mucosal barrier function(Reference Mawdsley and Rampton8, Reference Monnikes, Tebbe and Hildebrandt9) and thus increasing translocation of bacterial lipopolysaccharide from gram negative bacteria(Reference Gareau, Silva and Perdue10, Reference Maes, Kubera and Leunis11). Increased bacterial lipopolysaccharide translocation may result in maintenance of the activation of the inflammatory response system and elevated pro-inflammatory cytokines.
To prevent this gastrointestinal problems, probiotics defined as live micro-organisms, which, when consumed in adequate amounts confer a health benefit on the host, have been applied as an alternative approach of prevention and therapy. Probiotics may exert beneficial antibacterial effect on pathogens through the production of antibacterial substances, decrease adhesion of both pathogens and their toxins, increase barrier functions and inhibit pro-inflammatory cytokine production(Reference Lammers, Brigidi and Vitali12, Reference Yan and Polk13). It has been reported that a probiotic formulation beneficially affects the human stress response and its impact may be mediated through the gut–brain axis. In healthy volunteers suffering from stress-induced gastrointestinal symptoms in which the combination of probiotics Lactobacillus helveticus and Bifidobacterium longum was given showed a significant reduction in gastrointestinal symptoms(Reference Diop, Guillou and Durand14). Other studies with L. helveticus have shown that it can reduce Escherichia coli-induced lesions(Reference Sherman, Johnson-Henry and Yeung15) and modulate motility (in stress studies(Reference Gareau, Jury and MacQueen16)), whereas B. longum has been shown to downregulate TNF-α(Reference Wallace, Bradley and Buckley17) and maintain remission in ulcerative colitis patients(Reference Haskey and Dahl18) indicating a potential anti-inflammatory action. Therefore, we hypothesise that the regular intake of two biotherapeutic microbes, L. helveticus and B. longum, in combination in a probiotic formulation as a prophylactic agent, may diminish the apoptosis propensity induced by the inflammatory condition observed after MI in different brain regions.
Materials and methods
Experimental groups (animals and housing)
A total of thirty-five rats were used in the present experiment. They were 10-weeks-old adult male Sprague–Dawley rats (Charles River Canada, St-Constant, Que., Canada) weighing between 325 and 350 g (at the beginning of the experiment). The rats were housed individually under constant conditions (temperature of 21–22°C and humidity of 40–50 %). The animals were maintained on a 12-h dark–light cycle which began at 08.00 hours. Chow pellets (5075-US Charles River Rodent) and tap water were available ad libitum throughout the study. An acclimatisation period of 5 d after delivery by the supplier was allowed before the rats were randomly distributed to one of two groups, probiotics (n 18) or placebo (n 17). Both of the groups underwent a 40 min occlusion of the left anterior descending coronary artery. The animals were fed over a 4-week period and were killed following 3 d of reperfusion.
Probiotic treatment
The commercial probiotic given was a combination of two genus; L. helveticus R0052 and B. longum R0175 (Probio' Stick™ provided by Institut Rosell, Inc., Montreal, Que., Canada). The probiotics were administered by dissolving the freeze-dried culture or the vehicle only (maltodextrin) in 200 ml of drinking tap water. Each rat in the probiotic group received a daily dose of 109 colony forming units/ml. The drinking solution was newly prepared every second or third day for four consecutive weeks. Water intake was monitored throughout the entire investigation to ensure enough bacteria were administered. The body weight of each rat was also monitored every second or third day.
In vivo surgical procedure
Animals underwent anaesthesia following a ketamine/xylazine (50 and 5 mg/kg, respectively) intramuscular injection. Subsequently, the rats were intubated and anaesthesia was maintained using isoflurane (1·5 %) ventilation. Electrocardiogram and heart rate were monitored throughout the procedure using electrodes placed on their paws. A left thoracotomy was carried out at the fifth intercostal space permitting the occlusion of the left anterior descending coronary artery using a 4–0 silk suture (Syneture; Covidien, Mansfield, MA, USA). Ischaemia was confirmed by alterations of the ST segment and myocardial surface cyanosis. After 40 min of ischaemia, the thread was removed permitting reperfusion of the myocardial tissue. After the thorax was sutured by means of a 2–0 silk suture (Syneture; Covidien), the animals were given an antibiotic injection (15 000 IU penicillin G; Duplocillin LA, Intervet Canada Ltd, Whitby, Ont., Canada) as well as an analgesic injection (2 mg/kg butorphanol) before being returned to their respective cages.
Decapitation measurements, the area at risk of the heart and myocardial infarction size and tissue dissection
After 3 d of reperfusion, the rats were restrained in a cone bag and rapidly decapitated. Decapitation was preferred as the killing method to avoid any alteration of biochemical pathways that could arise ensuing anaesthesia or CO2 exposure. The heart was taken and the brain was placed on a dish positioned on ice. Brain regions were identified according to the atlas of Paxinos & Watson(Reference Paxinos and Watson19); frontal cortex, prefrontal cortex, hippocampus (Ca1, Ca3 and dentate gyrus), amygdala (medial and lateral parts) and hypothalamus (anterior and posterior parts). Tissues were frozen in liquid nitrogen and kept at − 80°C until needed.
The heart was removed and the left anterior descending coronary artery was occluded at the same site to establish the area at risk (AR) with infusion of Evans Blue (0·5 %) by retrograde perfusion into aorta. The heart was then placed at − 80°C for 5 min and sliced in four to five transverse sections of 2 mm. Each section was incubated 5 min at 37°C in a triphenyltetrazolium chloride solution (1 %, pH 7·4) to better distinguish the area of necrosis (I) from AR. MI was expressed as a percentage of necrosis (I) of the AR ((I/AR) × 100). Additionally, AR was expressed as a percentage of left ventricle area.
Caspase-3 activity
Cytosolic proteins were extracted in lysis buffer (1 % Triton X-100, 0·32 m sucrose, 10 mol/ml 2-amino-2-(hydroxymethyl)-1,3-propanediol (Tris; pH 8·0), 5 mmol/ml ethylenediamine tetra-acetate, 2 mmol/ml dl-1,4-dithiothreitol, 1 mmol/ml phenylmethanesulphonyl fluoride, 10 mg/ml leupeptin, 10 mg/ml pepstatin A and 10 g/ml aprotinin). Enzymatic reactions were carried out in a reaction buffer (50 mmol/ml Tris (pH 7·5), 5 mmol/ml MgCl2, 1 mmol/ml ethylene glycol-bis(2-aminoethyl ether)-N, N, N′, N′-tetraacetic acid, 0·1 % 3-[cholamidopropyl dimethylammonio]-1-propanesulphonate and 1 mmol/ml dithiothreitol), with 25 mg of proteins and a fluorogenic substrate, N-acetyl-asp-glu-val-asp-7-amido-4 methylcoumarin (40 μmol/ml). Reactions were incubated at 37°C for 3 h and stopped with the addition of 0·4 m NaOH and 0·4 m glycine buffer. Fluorescence was quantified using a spectrofluorometer (Photon Technology International, Lawrenceville, NJ, USA) at an excitation wavelength of 365 nm and an emission wavelength of 465 nm.
Western blot
Brain tissue samples were lysed in a buffer containing protease and phosphatase inhibitors (leupeptin, microcystine and benzamidine). After solubilisation, equal amounts of proteins (60 μg) in each line were loaded on a 10 % SDS-PAGE, and after migration, proteins were transferred onto a nitrocellulose membrane. Primary antibody directed against Akt (1:2000), phospho-Akt (1:1000; NEB Biolabs, Knowl Piece, Hitchin, UK), Bax or Bcl-2 (1:500; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) was incubated overnight at 4°C. After three washes, a secondary peroxidase-coupled antibody (1:5000) was added for 1 h at room temperature (anti-rabbit Ig (IgG)–horseradish peroxidase from Santa Cruz Biotechnology). A Renaissance Chemiluminescence kit (Perkin Elmer, Mississauga, Ont., Canada) was used to visualise the bands, and the quantitative analysis was conducted with a Kodak Image Station. After quantification, membranes were placed in stripping buffer (0·1 m glycine, 1 % SDS, pH 2·0, 1 h at room temperature). The same procedures were repeated with the other antibody (Akt or phospho-Akt) to obtain the phospho-Akt:Akt ratio.
Statistics
Results are expressed as means with their standard errors. Differences between groups were evaluated using Student's t test. P < 0·05 was considered significant. All the variances were equal and the data normally distributed.
Results
Infarct size
AR, expressed as a percentage of left ventricle area, was similar for both groups (placebo 53·0 (sem 4·3) %; probiotics 62·1 (sem 8·3) % of left ventricle; P = 0·36). Following 40 min of ischaemia and 72 h of reperfusion, the myocardial infarct size (I/AR) represented 44·1 (sem 2·5) % of the AR for the placebo group when compared with 45·2 (sem 3·3) % for the probiotics group (P = 0·80). There were no significant differences between the probiotic and the placebo groups.
Bax/Bcl-2 content
Bax/Bcl-2 ratio was measured in five different regions (Fig. 1). A significant decrease in Bax/Bcl-2 ratio was observed in the dentate gyrus, medial and lateral amygdala in the probiotic group when compared with the placebo group. In contrast, Ca1 and Ca3 regions indicate no significant difference in the ratio between groups.
Bax/Bcl-2 ratio in the different regions assessed by western blot (n 6–8/group; * P < 0·05 indicating a significant difference between the placebo (□) group and the probiotic (
) group: Ca1, P = 0·26; Ca3, P = 0·86; dentate gyrus, P = 0·007; medial amygdala, P = 0·034; and lateral amygdala, P = 0·01).
Caspase-3 activation
After 72 h of reperfusion, caspase-3 activity was significantly reduced in the probiotic group in the lateral amygdala, medial amygdala and dentate gyrus when compared with the placebo group (Fig. 2). No significant difference was observed between the groups in the Ca1 and Ca3 regions.
Caspase-3 activity in the different regions assessed by in vitro spectrofluorescence (n 6–8/group; * P < 0·05 indicating a significant difference between the placebo (□) group and the probiotic () group: Ca1, P = 0·67; Ca3, P = 0·09, dentate gyrus, P = 0·017; medial amygdala, P = 0·026; lateral amygdala, P = 0·015).
Phospho-Akt:Akt content
Phospho-Akt:Akt ratios were significantly different between the placebo and probiotic groups in the three different brain regions, lateral amygdala, medial amygdala and dentate gyrus (Fig. 3). No significant difference was observed between groups in the Ca1 and Ca3 regions.
Akt activity in the different regions assessed by the ratio of phospho-Akt on total Akt evaluated by western blot (n 6–8/group; * P < 0·05 indicating a significant difference between the placebo (□) group and the probiotic () group: Ca1, P = 0·80; Ca3, P = 0·69; dentate gyrus, P = 0·032, medial amygdala, P = 0·004; lateral amygdala, P = 0·01).
Discussion
The data obtained in the present study are the first reported observation that the gut–brain axis can modulate the apoptosis propensity observed in the limbic system after MI. However, there are numerous observations in the literature suggesting a link between depression and gastrointestinal diseases(Reference Maes, Kubera and Leunis11). It has been observed that intestinal mucosal dysfunction, characterised by an increased translocation of gram-negative bacteria, plays a role in the inflammatory pathophysiology of depression inducing the sickness behaviour(Reference Maes, Kubera and Leunis11). In a mouse model, depression increases the sensitivity to experimental colitis, which can be reversed by antidepressants(Reference Varghese, Verdu and Bercik20). It is also reported that patients with irritable bowel syndrome have a high prevalence of psychiatric disorders suggesting a link between brain and gut. The mechanisms are not clearly established, but it has been suggested that prolonged exposure to stress can induce low-grade inflammation, causing ultrastructural epithelial abnormalities, alter bacterial–host interactions allowing an increase in bacteria translocation(Reference Gareau, Silva and Perdue10), which in turn affects the brain. For example, treatment of mice with lipopolysaccharide increases insulin transport across the blood brain barrier by about threefold. The brain endothelial cells, which comprise the blood brain barrier, secrete many substances including cytokines and such secretion can be stimulated from one side of the blood brain barrier with release into the other side(Reference Banks21). However, in contradiction to this hypothesis, Verdu et al. (Reference Verdu, Bercik and Huang22) showed that the bacterial content of the gut influenced the rate of recovery of host pathophysiology induced by chronic Helicobacter pylori infection, including behavioural changes, and that these changes were not associated with modulation of intestinal permeability.
Probiotics have multiple and diverse influences on the host, which may include antimicrobial activity, enhancement of barrier function and immunomodulation. Recent investigations have suggested specific probiotics and gut microbiota can impact the immune system at both a systemic and a mucosal level(Reference Ng, Hart and Kamm23). Studies show that certain probiotic bacteria may increase the production of anti-inflammatory cytokines including IL-10(Reference Lammers, Brigidi and Vitali12). Alternately, probiotics could inhibit pro-inflammatory cytokine production such as TNF-α and IL-8, two pro-inflammatory cytokines(Reference Bai, Ouyang and Xiao24). The alteration of the pro/anti-inflammatory cytokine balance may explain the reduction in apoptosis that we observed in the different limbic regions. While the mechanisms have not been elucidated, it has been suggested that pro-inflammatory cytokines participate in the limbic cell death after MI(Reference Wann, Boucher and Kaloustian4).
The effect of probiotics on the susceptibility of apoptosis observed in the amygdala and in the dentate gyrus is similar to drugs with anti-inflammatory properties tested in our MI model such as cytokine inhibitor(Reference Wann, Boucher and Kaloustian4), anti-depressant(Reference Wann, Bah and Kaloustian3) or A2A adenosine receptor agonist(Reference Boucher, Wann and Kaloustian25).
In the present study, we observed that probiotics are unable to reduce the apoptosis observed in the Ca1, suggesting that different mechanisms are involved in this region. Apoptosis can be induced by at least two different pathways, extrinsic and intrinsic, although there are some proteins that are common to both such as caspase-3(Reference Adams26). Our data from amygdala and dentate gyrus suggest that the extrinsic signalling pathway is involved in the apoptosis in these regions as shown by the change in the Bax/Bcl-2 ratio. In contrast, intrinsic signalling pathway may be involved in the apoptosis observed in Ca1. It has been shown that the Ca1 region of the brain is the most sensitive to hypoxia(Reference Sugawara, Fujimura and Noshita27). It is well known that hypoxia can activate the intrinsic signalling pathway(Reference Reyland and Srivastava28), and thus can explain why probiotics are unable to reduce the apoptosis tendency in Ca1. However, previous data indicate that Ca1 apoptosis can be attenuated in the presence of pentoxifylline. This observation leads us to formulate the hypothesis that in addition to hypoxia, cytokines must also be involved in the apoptosis observed in Ca1. This can be verified by the effect of pentoxifylline(Reference Wann, Boucher and Kaloustian4) on the Bax/Bcl-2 ratio. The effect of L. helveticus and B. longum in combination on the extrinsic apoptotic pathway is probably not sufficient to attenuate apoptosis susceptibility in this region.
We also observe a higher level of activation of Akt in presence of the combination of L. helveticus and B. longum when compared to the placebo group. Akt plays a critical role in the proliferation, differentiation and apoptosis; and inhibition of the PI3K/Akt invariably leads to cell cycle arrest and/or apoptosis(Reference Datta, Bellacosa and Chan29–Reference Morley, Wagner and Kauppinen32). One possibility that explains the present result is a low molecular weight soluble factor released from the bacteria that stimulates Akt activation(Reference Yan and Polk13). Alternatively, it has been shown that some proteins isolated from bacteria may activate Akt(Reference Yan, Cao and Cover33). However, this is less probable since this protein needs to be present in the limbic system to activate Akt.
Myocardial reperfusion is associated with an important inflammatory response that can modulate the infarct size(Reference de Lorgeril, Rousseau and Basmadjian34, Reference Vinten-Johansen35). Since specific probiotic strains are capable of acting on the balance between pro/anti-inflammatory cytokines, the probiotics used in our model had the potential of attenuating the inflammatory process and the infarct size. However, the absence of the effect of probiotics on MI size is in accordance with our previous results indicating that pentoxifylline, a cytokine synthesis inhibitor, did not affect the infarct size(Reference Wann, Boucher and Kaloustian4). Although many studies seem to indicate that the reduction in pro-inflammatory cytokines has a beneficial effect on infarct size(Reference Gu, Yang and Bonde36, Reference Maekawa, Wada and Kanda37), other studies show that diminution of pro-inflammatory cytokines such as TNF-α has no beneficial effect(Reference Deuchar, Opie and Lecour38–Reference McVey, Perrone and Clark40). Overall, in our experimental model, the administration of L. helveticus and B. longum in combination as a prophylactic agent seems to have only a minor effect on myocardial infarct size. However, this lack of effect eliminates the infarct size as a possible explanation for the reduction of apoptosis susceptibility in the limbic system.
In conclusion, the probiotic preparation containing both L. helveticus and B. longum in combination reduced the apoptosis susceptibility observed in the limbic system after MI, but did not have any significant effect on myocardial infarct size.
Ethics statement
The investigation conformed to the Animal Care guidelines published by the Canadian Council and the procedures were approved by the Local Animal Care Committee.
Acknowledgements
The authors thank Caroline Bouchard for her skilful assistance and technical expertise. Conflicts of interest. G. R. is a scholar of ‘Fonds de la recherche en santé du Québec’. T. M. B. and S. K. hold a studentship from the Fonds de la recherche en santé du Québec. The present study was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC; # 250234-07) and Institut Rosell, Inc. All the authors declare that they have no conflict of interest with respect to the present study or its publication. S.-A. G. contributes to experiments, data analysis and writing. T. M. B., S. K., L. L.-M. and I. R. contribute to experiments, data analysis and writing. T. A. T., R. G., and G. R. contribute to the conception of the experiments, data analysis and writing.
