Integratedhealth.com.au

Neuroendocrinology Letters Volume 29 No. 1 2008
The gut-brain barrier in major depression: Intestinal mucosal dysfunction with an increased translocation of LPS from gram negative enterobacteria (leaky gut) plays a role in the inflammatory pathophysiology of depression Michael Maes 1, Marta Kubera 2 and Jean-Claude Leunis 3
1. MCare4U Outpatient Clinics, Belgium; 2. Department of Experimental Neuroendocrinology, Institute of Pharmacology, Polish Academy of Sciences, Krakow, Poland; 3. Laboratory Ategis, Waver, Belgium.
Correspondence to: Prof. Dr. M.Maes, M.D., Ph.D. Director: M-Care4U Outpatient Clinics, Olmenlaan 9, 2610 Antwerp, Belgium. tel: +32-3-4809282 fax: +32-3-2889185 www.michaelmaes.com e-mail: [email protected] Submitted: 2008-01-06 Accepted: 2008-01-27 Published online: 2008-02-22 Key words: major depression; chronic fatigue syndrome; inflammation; enterobacteria;
leaky gut; gut permeability; cytokines; LPS; oxidative stress
Neuroendocrinol Lett 2008; 29(1):117–124 PMID: 18283240 NEL290108A12 2008 Neuroendocrinology Letters • www.nel.edu
There is now evidence that major depression (MDD) is accompanied by an ac- tivation of the inflammatory response system (IRS) and that pro-inflammatory cytokines and lipopolysacharide (LPS) may induce depressive symptoms.
The aim of the present study was to examine whether an increased gastrointestinal permeability with an increased translocation of LPS from gram negative bacteria may play a role in the pathophysiology of MDD. Toward this end, the present study examines the serum concentrations of IgM and IgA against LPS of the gram-nega- tive enterobacteria, Hafnia Alvei, Pseudomonas Aeruginosa, Morganel a Morganii, Pseudomonas Putida, Citrobacter Koseri, and Klebsielle Pneumoniae in MDD patients and normal controls.
We found that the prevalences and median values for serum IgM and IgA against LPS of enterobacteria are significantly greater in patients with MDD than in nor- mal volunteers. These differences are significant to the extent that a significant diagnostic performance is obtained, i.e. the area under the ROC curve is 90.1%. The symptom profiles of increased IgM and IgA levels are fatigue, autonomic and gastro-intestinal symptoms and a subjective feeling of infection. The results show that intestinal mucosal dysfunction characterized by an increased translocation of gram-negative bacteria (leaky gut) plays a role in the inflammatory pathophysiol- ogy of depression. It is suggested that the increased LPS translocation may mount an immune response and thus IRS activation in some patients with MDD and may induce specific "sickness behaviour" symptoms. It is suggested that patients with MDD should be checked for leaky gut by means of the IgM and IgA panel used in the present study and accordingly should be treated for leaky gut.
To cite this article: Neuroendocrinol Lett 2008; 29(1): 117–124
Michael Maes, Marta Kubera and Jean-Claude Leunis creased IL-1β and IL-6 levels have been detected in the blood and various brain regions [19]. The IRS activa- There is now evidence that activation of the inflamma- tion theory of MDD is fueled by the high comorbidity tory response system (IRS) plays a role in the patho- of MDD with inflammatory disorders such as multiple physiology of major depression (MDD). This theory sclerosis (MS), coronary-heart disorder, HIV-infection, has been described as the "cytokine" or "monocyte-T inflammatory bowel disease and rheumatoid arthritis lymphocyte hypothesis" of depression, although recent- [1,20]. For example, in MS, the depressive episodes are ly we described this newly discovered pathway as the preceded by increased IFNγ production [21], suggesting "IRS activation" theory [1–4] and now as the "inflam- that IRS activation may underpin depression in MS.
matory and neurodegenerative (I&ND) hypothesis" of The aim of the present study is to examine anoth- depression [5].
er immune pathway which may underpin MDD, i.e. an The IRS findings in MDD show: a) an increased pro- immune response mounted against an increased trans- duction of pro-inflammatory cytokines, such as inter- location of LPS from gram-negative enterobacteria. Re- leukin-6 (IL-6), IL-1β and tumor necrosis factor alpha cently, we published that the median values for serum (TNFα); b) an increased expression of T lymphocyte IgA and IgM against LPS of enterobacteria are signifi- activation markers, such as HLA-DR+ and CD25+; c) cantly greater in patients with chronic fatigue syndrome the presence of an acute phase response, with increased (CFS) than in normal volunteers, suggesting that an in- serum levels of haptoglobin, decreased serum zinc levels creased translocation of LPS from enterobacteria is a and increased serum concentrations of the alpha2 glob- new pathway underpinning CFS [22,23]. This condi- ulin fraction obtained by electrophoresis; d) and signs tion can also be described as increased gut permeability of poor cel ular immunity, such as lowered natural kil - or leaky gut and indicates intestinal mucosal dysfunc- er cell cytotoxity and decreased mitogen–induced lym- tion (IMD). Since there is a strong degree of comorbid- phocyte responses; and e) an increased induction of in- ity between MDD and CFS, and since fatigue is one of doleamine oxidase (IDO) with consequent tryptophan the key symptoms of major depression and gastrointes- depletion and formation of TRYCATs (tryptophan cat- tinal symptoms frequently occur in MDD [24,25], we abolites along the IDO pathway) [6].
hypothesized that also MDD might be accompanied by Activation of the IRS is found in animal models of an increased translocation of LPS.
MDD, such as the chronic mild stress and the olfac- The specific aims of the present study are to examine tory bulbectomized rat models of depression [7–10]. whether MDD is accompanied by increased serum lev- Newly generated rat models of depression are based els of IgM and IgA against the LPS of 6 enterobacteria, on induced inflammation, e.g. the LPS-induced model i.e. Hafnei Alvei, Pseudomanes Aeruginosa, Morganel- [11] and sustained administration of IL-6 by infecting la Morganii, Pseudomanus Putida, Citrobacter Koseri, healthy MRL +/+, C3H.SW and Balb/C mice with ad- and Klebsiel a Pneumoniae indicating an immune re- enovirus vector carrying cDNA for murine IL-6 or mice sponse directed to endotoxins secreted by gram-nega- infected with Ad5mIL6 adenovirus [12,13].
tive enterobacteria and which cannot be detected when It is known that systemic LPS and administration of the gut-intestinal lining is intact.
pro-inflammatory cytokines cause chronic central neu- roinflammation. For example, systemic LPS results in SubJeCtS And MethodS SubJeCtS
rapid increases in brain TNFα levels, which may remain elevated for 10 months [14]. Moreover, brain microglia Fifty-one subjects participated in the present study, 23 are activated to produce chronical y elevated pro-in- controls (staff or their family members), and 28 MDD flammatory factors in the brain [14]. Central neuroin- patients admitted to the M-Care4U Outpatient Clinics, flammation and an increased production of pro-inflam- Belgium. The patients were classified according to the matory cytokines, such as TNFα, IL-1β and IL-6, may Diagnostic and Statistical Manual of Mental Disorders, induce a behaviour complex, i.e. sickness behaviour, 4th edition [26]. We have excluded: characterized by the appearance of symptoms such as anorexia, psychomotor retardation, malaise, loss of in- a) subjects with life-time diagnosis of other psychiat- terest, etc. [14]. This symptom complex is quite similar ric DSM-IV-TR disorders, such as anxiety disor- to the symptoms of MDD [15].
ders, schizophrenia, substance use disorders and External (psychological) as well as internal (organic organic mental disorders; stressors) stressors, which induce IRS activation, are re- b) subjects with CFS as diagnosed by the CDC criteria lated to the appearance of depressive episodes. Exam- [27]; c) subjects with other medical il ness, such as ples are severe negative life events, some inflammato- other inflammatory or autoimmune disorders; ry and autoimmune illnesses, the postnatal period, etc. d) subjects who ever had been treated with anti-psy- Maes et al. [16–18] were the first to show that – in hu- chotic drugs or anticonvulsants and subjects who mans – psychological stress induces IRS activation with had been taking psychotropic drugs during the last increased production of pro-inflammatory cytokines, year prior to the studies; e) subjects with abnor- such as IFNγ and TNFα. In experimental animals in- mal values for routine blood tests, such as alanine Copyright 2008 Neuroendocrinology Letters ISSN 0172–780X • www.nel.edu
The gut-brain barrier in major depression: Intestinal mucosal dysfunction . aminotransferase (ALT), alkaline phosphatase the 12 Ig values. The same indices were also employed (ALP), blood urea nitrogen (BUN), calcium, cre- to assess the IgM- versus the IgA-related translocation atinine, electrolytes, thyroid stimulating hormone loads. Toward this end, we computed the first PCs of (TSH), total protein, and iron or transferrin satu- the 6 IgM and 6 IgA data; the peak IgM and peak IgA data; and the total sum of the 6 IgM and 6 IgA data. We f) subjects with acute inflammatory and al ergic reac- employed Pearson's product moment correlation coef- tions for at least 1 month prior to the study. ficients, Spearman's rank order correlations, and multi- ple regression analyses in order to examine the relation- The total sum of the FibroFatigue (FF) scale, i.e. the ships between variables. The relationships between the Fibromyalgia and Chronic Fatigue Syndrome Rating IgM / IgA values and the symptoms of the FF scale were Scale [28,29], was used in the MDD patients to com- assessed by means of canonical correlation analyses and pute the severity of specific symptoms, some of which by means of stepwise (F-to-enter p=0.05) linear discri- have been shown to be correlated to the IgM and IgA minant analysis (LDA) with the 12 FF scale items as the levels against LPS in CFS [22]. This scale measures 12 discriminatory variables and the dichotomized peak Ig items reminiscent for CFS and fibromyalgia: pain, mus- (IgM and IgA) data as grouping variable. The signifi- cular tension, fatigue, concentration difficulties, failing cance was set at α=0.05 (two tailed).
memory, irritability, sadness, sleep disturbances, auto- nomic disturbances, irritable bowel, headache, and sub- jective experience of infection. Patients and controls gave written informed consent after the study protocol There were no significant differences (F=1.6, df=4/49, was ful y explained. The study has been approved by p=0.2) in age between normal controls (mean ±SD = the local ethical committee.
40.0 ± 12.9 years) and MDD patients (44.3 ±11.2 years). There were no significant differences (χ2 Yates = 0.1, df=1, p=0.7) in the female / male ratio between normal controls (16 / 7) and MDD patients (17 / 11). There Blood was taken during the morning hours for the de- were no significant correlations between age and any termination of IgM and IgA against the LPS of 6 differ- of the serum IgM or IgA levels directed against the LPS ent enterobacteria, i.e. Hafnei Alvei, Pseudomanes Aer- of the different enterobacteria and no significant differ- uginosa, Morganel a Morganii, Pseud omanus Putida, ences in the serum IgM or IgA values between males Citrobacter Koseri and Klebsiel a Pneumoniae. Serum and females. The FF score was not significantly differ- IgM and IgA levels were measured by means of an in- ent between men and women (F=0.06, df=1/26, p=0.8) direct ELISA method according to the methods out- and there was no significant correlation between age lined by the manufacturer (Gemacbio, The Ultimate and the FF score (r=–0.33, p=0.08).
Biopharmaceuticals, France). Each serum sample was Table 1 shows the serum IgM values in the MDD measured in duplicate and tested simultaneously with patients and controls. ANOVAs showed that the IgM three standard solutions. The optical densities (OD) levels directed against the LPS from Pseudomonas Aer- of the three standards are expressed as Z values and uginosa and Putida were significantly greater in MDD from this the reference linear curve is calculated as Z that in the normal controls. Table 1 shows that there = f(OD) with Z = a OD + b. Thus, the Z value of the were significant differences between the study groups lowest standard can be negative. This curve allows to in the 6 IgA levels against LPS except against Pseu- deduce the mean values of the duplicate measurements domonas Aeruginosa. Covarying for age and sex did of the OD values. The biological interassay CV values not change any of the above results. By means of Fish- were < 10%.
er's exact probability test we found a significantly (ψ = 0.50, p=0.0002) greater number of MDD patients (12 / 28) with abnormal y increased IgM levels (i.e. anyone Group mean differences were assessed by means of of the 6 IgM values > 2 Z values) than in controls (0 / analysis of variance (ANOVA) or analysis of covariance 23). Also, the prevalences of MDD patients (11 / 28) (ANCOVA). The diagnostic performance was checked with abnormal y increased IgA levels (i.e. anyone of the by means of ROC (receiver operating characteristics) 6 IgA > 2 Z values) were significantly (ψ = 0.35, p=0.01) analysis with computation of the area under the ROC higher than in normal controls (2 / 23).
curve, sensitivity, specificity and predictive value of a Table 2 gives the peak serum Ig values (IgM or IgA, positive test result (PV+) and with kappa statistics. Data alone and together), the total sum of the Ig data (IgM or reduction was obtained by means of principal compo- IgA, alone and together) and the first PC of the IgM and nent (PC) analyses. In order to assess the "total LPS the IgA data. However, we were unable to find that one translocation load" we have employed three different PC could reflect the 12 Ig (IgM and IgA) data: indeed, indices: a) the first PC of the 12 Ig (IgM and IgA) val- the first PC of the 12 Ig data explained only 38.2% of ues; b) the total sum of the 12 Ig (IgM and IgA) levels; the variance, while the 6 IgM variables loaded highly on and c) the peak Ig (IgM or IgA) levels, i.e. the highest of the first PC and the 6 IgA data on the second PC even Neuroendocrinology Letters Vol. 29 No. 1 2008 • Article available online: http://node.nel.edu
Michael Maes, Marta Kubera and Jean-Claude Leunis without performing a rotation. Thus, there is no com- The results of our study show that there is an IgM mon PC which can reflect the 12 Ig variables. The peak and IgA-related immune response raised to the LPS IgM and IgA values and the total sum of all 6 IgM and of enterobacteria in MDD. Increments in serum IgM IgA values as well as the first PC subtracted from the 6 levels can be seen in immune activation and mucosal IgM and IgA values were significantly higher in MDD immunity. B1 lymphocytes are a significant source of patients than in normal controls. The peak Ig (either natural serum IgM and they constitute a first line of IgM or IgA) values and the sum of the 12 Ig values were defence against systemic viral and bacterial infections both significantly higher in MDD patients than in nor- [30]. Moreover, B1 cel s migrate to the intestinal lamina mal controls. Covarying for age and sex in ANCOVAs propria to differentiate into IgA-producing serum cel s, did not change any of these results. ROC analysis per- which in turn play a role in mucosal immunity [30].
formed on the peak IgM or IgA values showed that the The increased serum IgM and IgA levels against LPS area under the ROC curve (AUC) was highly significant in MDD indicate that MDD is accompanied by an in- (AUC=90.8%). Accordingly, the diagnostic perform- creased gut permeability and that there is an immune ance computed was highly significant; at a cut off value response directed against LPS of enterobacteria. Indeed, for the peak IgM or IgA values > 2.2 Z, the diagnostic the intestinal epithelial barrier has critical functions, performance was: sensitivity=67.9%, specificity=95.6%, such as a) the formation of a barrier, which separates and PV+=95.0% (κ=0.62, t=5.66, p=0.00002).
the luminal contents from the interstitium, and which By means of Spearman's rank order correlation coef- protects against micro-organisms including gram neg- ficients we found a significant positive correlation be- ative bacteria, larger toxic and antigenic molecules; b) tween the peak IgM-IgA data and the total sum on the the transportation of fluids, electrolytes and nutrients FF scale (r=0.49, p=0.007). Two items of the FF scale across the intestinal wall; and c) the secretion of IgA were significantly related to the peak IgM-IgA data, i.e to bind to bacteria thus preventing their attachment to fatigue (r=0.54, p=0.003) and irritable bowel (r=0.56, epithelial cel s.
p=0.002). Table 3 shows the results of a canonical cor- The function of the intestinal barrier may be com- relation analysis with peak IgM and peak IgA values as promised by IRS activation. The latter may cause a explanatory variables and the 12 items of the FF scale as loss of the protective barrier function [31–33], which dependent variables. We found that the symptom pro- in turn causes enlarged spaces between the cel s of the files of increased IgM and IgA were fatigue, autonomic gut wall [31–33]. In this respect, the important inflam- disturbances, irritable bowel and a subjective feeling of matory mediators which induce "leaky gut", IFNγ and infection. In order to further examine the relationships IL-6, are both significantly increased in MDD [34,35]. between serum IgM and IgA levels and the severity of These disruptions of the intestinal epithelium allow the FF scale, we carried out stepwise LDA with the 12 normal y poorly invasive enterobacteria to exploit lipid FF scale symptoms as the explanatory variables and the raft-mediated transcytotic pathways or the enlarged dichotomized peak IgM (higher versus lower than 2 Z spaces to cross the gut wall [31–33]. Thus, IRS activa- values) and peak IgA (higher versus lower 2 Z values) tion – through an increased production of IFNγ and as groups. We found that sadness (loading=0.71), irri- IL-6 – is an essential factor in the loss of the epithelial table bowel (loading=0.58) and muscular tension (load- barrier function [31,33]. The former may induce an in- ing=0.40) were significant discriminatory variables for creased translocation of LPS and thus cause increased the dichotomized peak IgM values (F=18.5, df=1/26, serum concentrations of LPS which, in turn, may trig- p=0.0004). The dichotomized peak IgA groups were ger an IgM or IgA-mediated immune response to LPS best discriminated (F=5.7, df=1/26, p=0.02) using fa- [22,23].
tigue (loading = 0.86) and irritable bowel (loading = Systemic increases in LPS not only cause a system- 0.77) as discriminatory variables. Final y, the dichot- ic inflammation, but also a central neuroinflammation; omized peak IgM or IgA values were best predicted by increased brain tumor necrosis factor-α (TNFα) activi- fatigue (loading = 0.87), irritable bowel (loading = 0.75) ties, which may remain elevated for 10 months; and ac- and a subjective feeling of infection (loading = 0.65) tivation of brain microglia with a chronical y elevated (F=18.2, df=1/26, p=0.0004).
production of pro-inflammatory mediators [14]. It is well-known that an increased production of pro-in- flammatory cytokines, such as IL-1, IL-6 and TNFα, ei- ther peripheral or central, and brain neuroinflammation The findings of the present study show that MDD is ac- may induce the sickness behaviour complex [14]. Also companied by increased serum levels of IgM and IgA systemic LPS may provoke sickness behaviour [36,37]. directed against LPS of gram-negative enterobacteria As explained previously, symptoms of sickness behav- and that the IgM-IgA values are related to symptoms iour, such as anorexia, soporific effects, disturbances of reminiscent of MDD and CFS, e.g. fatigue, autonomic locomotor activity and exploration, and anhedonia bear and gastro-intestinal symptoms, and a subjective feel- a strong similarity with those of MDD [15].
ing of infection.
Increased LPS translocation may play a role in the O&NS, which occurs in MDD. Indeed, MDD is ac- Copyright 2008 Neuroendocrinology Letters ISSN 0172–780X • www.nel.edu
The gut-brain barrier in major depression: Intestinal mucosal dysfunction . Table 1. Measurements of serum IgM levels against the LPS of Hafnia Alvei, Pseudomonas Aeruginosa, Morganel a
Morgani , Pseudomonas Putida, Citrobacter Koseri and Klebsiel a Pneumoniae in normal controls and MDD patients.
Pseudomonas Aeruginosa Morganel a Morgani Pseudomonas Putida Citrobacter Koseri Klebsiel a Pneumoniae Al results are shown as mean (±SD). Al results of ANOVAs (df=1/49) Table 2. Measurements of peak IgM or IgA values, total sum of the 6 IgM or 6 IgA data, the first principal component
(PC) of the 6 IgM and 6 IgA data, as wel as the peak, total sum and the first PC of the 12 Ig (IgM and IgA) data.
First PC of the 6 IgM data First PC of the 6 IgA data Peak Ig (IgM and IgA) data Sum of the 12 Ig (IgM and IgA) data Al results are shown as mean (±SD). Al results of ANOVAs (df=1/49) Table 3. Results of canonical correlation analysis with the regression of the first principal
component (PC) of the IgM and the first PC of the IgA data directed against the LPS of Hafnia Alvei,
Pseudomonas Aeruginosa, Morganel a Morgani , Pseudomonas Putida, Citrobacter Koseri, and
Klebsiel a Pneumoniae on the one hand and the symptoms of the FibroFatigue scale, on the other.
Sleep disturbances Autonomic disturbances Subjective experience of infection Canonical correlation coefficient
Shown are the significant loadings (> 0.35) of the canonical regressions of the IgM / IgA data on the different symptoms of the FibroFatigue scale.
Neuroendocrinology Letters Vol. 29 No. 1 2008 • Article available online: http://node.nel.edu
Michael Maes, Marta Kubera and Jean-Claude Leunis companied by O&NS as indicated by increased levels of ic and induce sickness behaviour, e.g. fatigue, anorex- malondialdehyde (MDA), a byproduct of polyunsatu- ia, weight loss, sleep disorders, psychomoror retarda- rated fatty acid peroxidation and arachidonic acid, 8- tion, etc. Also, O&NS is implicated in the production hydroxy-2-deoxyguanosine, indicating damage to DNA of musceloskeletal pain, muscle fatigue resistance, re- by oxygen radicals [38–40]; increased catalase and duced responses to aerobic exercise and reduced maxi- MDA levels [41,42], increased peroxidase and catalase mal exercise time [55–59]. The significant correlation activities in blood and saliva [43]; and increased IgM between gastro-intestinal symptoms and the Ig-me- levels directed against nitro-BSA (bovine serum albu- diated immune response against LPS in MDD and in min), indicating increased nitrosative stress [44]. It is CFS indicates that the gastro-intestinal symptoms in well known that LPS causes induction of nuclear factor some patients with MDD (and in CFS) reflect – in part kappa beta (NFκβ), the major upstream, intracel ular – disorders in gut-intestinal permeability and cannot be mechanism which regulates inflammatory and O&NS considered as a symptom of mental stress as most psy- mediators, such as cyclo-oxygenase (COX-2) and in- chiatrists tend to confirm. The significant positive cor- ducible NO synthase (iNOS) [45,46].
relation between the IgM and IgA-mediated immune Based on the above, we may conclude that an in- response against LPS and the subjective experience of creased translocation of LPS is another pathway which infection in MDD indicates that the latter is an index of may explain the inflammatory pathophysiology of the inflammation from which the patients are suffering. MDD and which underpins the link between IMD and The canonical correlation analysis which was carried MDD. For example, external stressors (psychological out showed also a significant correlation between the stress) may compromise the intestinal barrier [47,48] Ig-mediated immune response against LPS and auto- as well as inducing the cytokine network causing in- nomic disturbances, such as gastroparesis, neuropathy, creased IFNγ and IL-6 production [16], which in turn vascular neuropathy, and dysautonomia. This may be may further endanger the mucosal gut barrier.
explained since LPS as wel as IRS activation and O&NS Increased LPS translocation offers also an explana- have multivarious and profound effects on the autono- tion for the increased incidence of MDD during alco- mous nervous system [60–66].
hol dependence. Thus, alcoholism is known to induce Increased gut permeability may be another factor an increased translocation of LPS [49] and is accom- explaining the occurrence of autoimmunity in MDD. panied by activation of the IRS with an increased ca- There are many reports that MDD is accompanied tabolism of tryptophan [50,51]. Internal (organic) stres- by autoimmune responses, such as against phopholi- sors, such as infections (AIDS/HIV) [52], autoimmune pids [67]. Indeed, enterobacteria may act as superan- disorders [53] and inflammatory bowel disease [54] tigens for T lymphocytes or may induce autoimmu- – which all show some degree of comorbidity with de- nity through a mechanism called molecular mimicry pression – may endanger gut mucosal function and may [68,69]. This may be explained since enterobacteria increase gut barrier permeability. Thus, the LPS path- have antigenic sites which are very similar to those of way may either primarily (an increased translocation of the lipid structures of neuronal tissue. These antigens LPS inducing inflammation) or secondarily (a primary will go into various tissues and trigger inflammation inflammation may induce LPS translocation through and once autoantibodies are formed the inflammation increased IL-6 and IFNγ production) be involved in may become more chronic. Thus, systemic LPS caused the inflammatory pathophysiology of MDD. Conse- by an increased translocation not only induces periph- quently, the intertwined systemic inflammation and the eral inflammation and O&NS, but also induces a cen- IgM and IgA-mediated immune response fol owing LPS tral neuroinflammation and eventual y an autoimmune translocation may further aggravate the depressogenic responses directed against neuronal tissues.
effects of IRS activation.
The results of the present study show that patients The symptom profiles of increased IgM and IgA lev- with MDD should be checked for the presence of leaky els to LPS of enterobacteria in MDD are fatigue, sad- gut by the measurements of IgM and IgA against the ness, autonomic and gastro-intestinal symptoms, mus- LPS of gram-negative bacteria. The results of the present cular tension, and a subjective feeling of infection. study suggest that MDD patients who have a leaky gut These findings corroborate our previous report that, in should be treated with specific antioxidants with an ef- CFS, there are significant positive correlations between ficacity for leaky-gut and a leaky gut diet [23].
the IgA responses to LPS of enterobacteria and symp- toms of the FibroFatigue scale, such as muscular ten- sion, fatigue, concentration difficulties, failing memory, The research reported was supported by a NARSAD autonomic disturbances, irritable bowel and the subjec- Distinguished researcher award to M.Maes and by M- tive experience of infection [22]. These correlations CARE4U and CRC-MH, Antwerp, Belgium. The secre- probably reflect causal relationships between IRS acti- tarial assistance of Indra Corten is greatly appreciated.
vation and O&NS inducing the abovementioned symp- toms occurring in CFS and MDD. As discussed above, LPS and pro-inflammatory cytokines are depressogen- Copyright 2008 Neuroendocrinology Letters ISSN 0172–780X • www.nel.edu
The gut-brain barrier in major depression: Intestinal mucosal dysfunction . 19 Ishikawa I, Kitamura H, Kimura K, Saito M: Brain interleukin-1 is involved in blood interleukin-6 response to immobilization 1 Maes M: A review on the acute phase response in major depres- stress in rats. Jpn J Vet Res 2001; 49(1): 19–25.
sion. Rev Neurosci 1993; 4(4): 407–416. Review.
20 Yirmiya R, Weidenfeld J, Pol ak Y, Morag M, Morag A, Avitsur R, 2 Maes M: Evidence for an immune response in major depression: Barak O, Reichenberg A, Cohen E, Shavit Y, Ovadia H: Cytokines, a review and hypothesis. Prog Neuropsychopharmacol Biol Psy- "depression due to a general medical condition," and antidepres- chiatry 1995; 19(1): 11–38. Review.
sant drugs. Adv Exp Med Biol 1999; 461: 283–316. Review.
3 Maes M: Major depression and activation of the inflammatory 21 Mohr DC, Goodkin DE, Islar J, Hauser SL, Genain CP: Treatment response system. Adv Exp Med Biol 1999; 461: 25–46. Review.
of depression is associated with suppression of nonspecific 4 Schiepers OJ, Wichers MC, Maes M: Cytokines and major depression. and antigen-specific T(H)1 responses in multiple sclerosis. Arch Prog Neuropsychopharmacol Biol Psychiatry 2005; 29(2): 201–
Neurol 2001; 58(7): 1081–1086.
217. Review. Erratum in: Prog Neuropsychopharmacol Biol Psy- 22 Maes M, Mihaylova I, Leunis JC: Increased serum IgA and IgM chiatry 2005; 29(4): 637–638.
against LPS of enterobacteria in chronic fatigue syndrome (CFS): 5 Maes M, Brene S, Hibbeln J, etc.: An Experimental Medicine indication for the involvement of gram-negative enterobacteria Approach to examine the Inflammatory & Neurodegenerative in the etiology of CFS and for the presence of an increased gut- hypothesis in Depression: leads for future research and drug intestinal permeability. J Affect Disord. 2007; 99(1–3): 237–240.
development in depression. Submitted.
23 Maes M, Coucke F, Leunis JC: Normalization of the increased 6 Maes M, Mihaylova I, Ruyter MD, Kubera M, Bosmans E: The translocation of endotoxin from gram negative enterobacteria immune effects of TRYCATs (tryptophan catabolites along the (leaky gut) is accompanied by a remission of chronic fatigue syn- ido pathway): Relevance for depression – and other conditions drome. Neuro Endocrinol Lett 2007; 28(6): 739–744.
characterized by tryptophan depletion induced by inflamma- 24 Maes M: From Freud to Omega-3. 2005, Brussels, Standaard Uit- tion. Neuro Endocrinol Lett 2007; 28(6): 826–831.
7 Kubera M, Basta-Kaim A, Holan V, Simbirtsev A, Roman A, Piga- 25 Maes M, Schotte C, Maes L, Cosyns P: Clinical subtypes of uni- reva N, Prokopieva E, Sham J: Effect of mild chronic stress, as a polar depression: Part I . Quantitative and qualitative clinical model of depression, on the immunoreactivity of C57BL/6 mice. differences between the vital and nonvital depression groups. Int J Immunopharmacol 1998; 20(12): 781–789.
Psychiatr Res 1990; 34(1): 43–57.
8 Kubera M, Symbirtsev A, Basta-Kaim A, Borycz J, Roman A, Papp 26 American Psychiatric Association: Diagnostic and Statistical M, Claesson M: Effect of chronic treatment with imipramine Manual of Mental Disorders, Fourth Edition. Text Revision (DSM- on interleukin 1 and interleukin 2 production by splenocytes IV-TR), 2000; Washington DC.
obtained from rats subjected to a chronic mild stress model of 27 Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff depression. Pol J Pharmacol 1996; 48(5): 503–506.
A: The chronic fatigue syndrome: a comprehensive approach to 9 Goshen I, Kreisel T, Ben-Menachem-Zidon O, Licht T, Weiden- its definition and study. International Chronic Fatigue Syndrome feld J, Ben-Hur T, Yirmiya R: Brain interleukin-1 mediates chronic Study Group. Ann Int Med 1994; 121(12): 953–959.
stress-induced depression in mice via adrenocortical activation 28 Zachrisson O, Regland B, Jahreskog M, Kron M, Gottfries CG: A rating and hippocampal neurogenesis suppression. Mol Psychiatry scale for fibromyalgia and chronic fatigue syndrome (the FibroFa- 2007 [Epub ahead of print] tigue scale). J Psychosom Res 2002; 52(6): 501–509.
10 Song C, Leonard BE: Interleukin-2-induced changes in behav- 29 Shapiro CM, Mol er HJ: Chronic fatigue: listen and measure. J ioural, neurotransmitter, and immunological parameters in the Psychosom Res 2002; 52(6): 427–436.
olfactory bulbectomized rat. Neuroimmunomodulation 1995; 30 Thurnheer MC, Zuercher AW, Cebra JJ, Bos NA: B1 cel s contrib- ute to serum IgM, but not to intestinal IgA, production in gno- 11 Yirmiya R: Endotoxin produces a depressive-like episode in rats. tobiotic Ig al otype chimeric mice. J Immunol 2003; 170(9):
Brain Res 1996; 711(1–2): 163–174.
12 Sakiƒc B, Szechtman H, Braciak T, Richards C, Gauldie J, Denburg 31 Clark E, Hoare C, Tanianis-Hughes J, Carlson GL, Warhurst G: JA: Reduced preference for sucrose in autoimmune mice: a pos- Interferon gamma induces translocation of commensal Esche- sible role of interleukin-6. Brain Res Bul 1997; 44(2): 155–165.
richia coli across gut epithelial cel s via a lipid raft-mediated pro- 13 Sakiƒc B, Gauldie J, Denburg JA, Szechtman H: Behavioral effects cess. Gastroenterology 2005; 128(5): 1258–1267.
of infection with IL-6 adenovector. Brain Behav Immun 2001; 32 Chavez AM, Menconi MJ, Hodin RA, Fink MP: Cytokine-induced intestinal epithelial hyperpermeability: role of nitric oxide. Crit 14 Qin L, Wu X, Block ML, Liu Y, Breese GR, Hong JS, Knapp DJ, Crews Care Med 1999; 27(10): 2246–2251.
FT: Systemic LPS causes chronic neuroinflammation and pro- 33 Yang R, Han X, Uchiyama T, Watkins SK, Yaguchi A, Delude RL, gressive neurodegeneration. Glia 2007; 55(5): 453–462.
Fink MP: IL-6 is essential for development of gut barrier dysfunc- 15 Maes M, Meltzer HY, Scharpé S, Cooreman W, Uyttenbroeck W, Suy tion after hemorrhagic shock and resuscitation in mice. Am J E, Vandervorst C, Calabrese J, Raus J, Cosyns P: Psychomotor retar- Physiol Gastrointest Liver Physiol 2003; 285(3): G621–629.
dation, anorexia, weight loss, sleep disturbances, and loss of 34 Maes M, Scharpé S, Meltzer HY, Okayli G, Bosmans E, D'Hondt energy: psychopathological correlates of hyperhaptoglobinemia P, Vanden Bossche BV, Cosyns P: Increased neopterin and inter- during major depression. Psychiatry Res 1993; 47(3): 229–241.
feron-gamma secretion and lower availability of L-tryptophan 16 Maes M, Song C, Lin A, De Jongh R, Van Gastel A, Kenis G, Bos- in major depression: further evidence for an immune response. mans E, De Meester I, Benoy I, Neels H, Demedts P, Janca A, Schar- Psychiatry Res 1994; 54(2): 143–160.
pé S, Smith RS: The effects of psychological stress on humans: 35 Maes M, Scharpé S, Meltzer HY, Bosmans E, Suy E, Calabrese J, increased production of pro-inflammatory cytokines and a Th1- Cosyns P: Relationships between interleukin-6 activity, acute like response in stress-induced anxiety. Cytokine 1998; 10(4):
phase proteins, and function of the hypothalamic-pituitary-adre- nal axis in severe depression. Psychiatry Res 1993; 49(1): 11–27.
17 Maes M, Lin AH, Delmeire L, Van Gastel A, Kenis G, De Jongh R, 36 Borowski T, Kokkinidis L, Merali Z, Anisman H: Lipopolysaccha- Bosmans E: Elevated serum interleukin-6 (IL-6) and IL-6 receptor ride, central in vivo biogenic amine variations, and anhedonia. concentrations in posttraumatic stress disorder fol owing acci- Neuroreport 1998; 9(17): 3797–3802.
dental man-made traumatic events. Biol Psychiatry 1999; 45(7):
37 Lacosta S, Merali Z, Anisman H: Behavioral and neurochemical consequences of lipopolysaccharide in mice: anxiogenic–like 18 Song C, Kenis G, van Gastel A, Bosmans E, Lin A, de Jong R, Neels effects. Brain Res 1999; 818(2): 291–303.
H, Scharpé S, Janca A, Yasukawa K, Maes M: Influence of psycholog- 38 Sarandol A, Sarandol E, Eker SS, Erdinc S, Vatansever E, Kirli S: ical stress on immune-inflammatory variables in normal humans. Major depressive disorder is accompanied with oxidative stress: Part I . Altered serum concentrations of natural anti-inflamma- short-term antidepressant treatment does not alter oxidative- tory agents and soluble membrane antigens of monocytes and T antioxidative systems. Hum Psychopharmacol 2007; 22(2): 67– lymphocytes. Psychiatry Res 1999; 85(3): 293–303.
Neuroendocrinology Letters Vol. 29 No. 1 2008 • Article available online: http://node.nel.edu
Michael Maes, Marta Kubera and Jean-Claude Leunis 39 Forlenza MJ, Mil er GE: Increased serum levels of 8-hydroxy-2'- 54 Shanahan F: Current concepts of the pathogenesis of inflamma- deoxyguanosine in clinical depression. Psychosom Med 2006; tory bowel disease. Ir J Med Sci 1994; 163: 544–549.
55 Vecchiet J, Cipol one F, Falasca K, Mezzetti A, Pizzigal o E, Buc- 40 Irie M, Miyata M, Kasai H: Depression and possible cancer risk due ciarel i T, De Laurentis S, Affaitati G, De Cesare D, Giamberardino to oxidative DNA damage. J Psychiatr Res 2005; 39(6): 553–560.
MA: Relationship between musculoskeletal symptoms and blood 41 Khanzode SD, Dakhale GN, Khanzode SS, Saoji A, Palasodkar R: markers of oxidative stress in patients with chronic fatigue syn- Oxidative damage and major depression: the potential antioxi- drome. Neurosci Lett 2003; 335(3): 151–154.
dant action of selective serotonin re-uptake inhibitors. Redox 56 Matuszczak Y, Farid M, Jones J, Lansdowne S, Smith MA, Taylor Report 2003; 8(6): 365–370.
AA, Reid MB: Effects of N-acetylcysteine on glutathione oxida- 42 Ozcan ME, Gulec M, Ozerol E, Polat R, Akyol O: Antioxidant tion and fatigue during handgrip exercise. Muscle Nerve 2005; enzyme activities and oxidative stress in affective disorders. Int Clin Psychopharmacol 2004; 19(2): 89–95.
57 Luan Y, Xu W: The function of the selective inhibitors of cyclooxy- 43 Lukash AI, Zaika VG, Kucherenko AO, Miliutina NP: Free radical genase 2. Mini Rev Med Chem 2006; 6(12): 1375–1381.
processes and antioxidant system in depression and treatment 58 Mets T, Bautmans I, Njemini R, Lambert M, Demanet C: The influ- efficiency. Zh Nevrol Psikhiatr Im S S Korsakova 2002; 102(9): 41–
ence of celecoxib on muscle fatigue resistance and mobility in elderly patients with inflammation. Am J Geriatr Pharmacother 44 Maes M, Mihaylova I, Kubera M, Leunis J-C: An IgM-mediated 2004; 2(4): 230–238.
immune response directed against nitro-bovine serum albumin 59 McIver KL, Evans C, Kraus RM, Ispas L, Sciotti VM, Hickner RC: NO- (nitro-BSA) in chronic fatigue syndrome (CFS) and major depres- mediated alterations in skeletal muscle nutritive blood flow and sion (MDD): evidence that nitrosative stress is another factor lactate metabolism in fibromyalgia. Pain 2005; 120(1–2): 161–
underpinning the comorbidity between CFS and MDD. Neuro Endocrinol Lett 2008 (in press).
60 Hartrick CT: Increased production of nitric oxide stimulated by 45 Maes M, Mihaylova I, Bosmans E: Not in the mind of neurasthenic interferon-gamma from peripheral blood monocytes in patients lazybones but in the cel nucleus: patients with chronic fatigue with complex regional pain syndrome. Neurosci Lett 2002; syndrome have increased production of nuclear factor kappa beta. Neuro Endocrinol Lett 2007; 28(4): 456–462.
61 Stoyneva Z, Lyapina M, Tzvetkov D, Vodenicharov E: Current 46 Maes M, Mihaylova I, Kubera M, Bosmans E: Not in the mind but in pathophysiological views on vibration-induced Raynaud's phe- the cel : increased production of cyclo-oxygenase-2 and inducible nomenon. Cardiovasc Res 2003; 57(3): 615–624.
NO synthase in chronic fatigue syndrome. Neuro Endocrinol Lett 62 Alexander GM, Perreault MJ, Reichenberger ER, Schwartzman RJ: 2007; 28(4): 463–469.
Changes in immune and glial markers in the CSF of patients with 47 Meddings JB, Swain MG: Environmental stress-induced gastroin- Complex Regional Pain Syndrome. Brain Behav Immun 2007; testinal permeability is mediated by endogenous glucocorticoids in the rat. Gastroenterology 2000; 119(4): 1019–1028.
63 Zhao J, Wecht JM, Zhang Y, Wen X, Zeman R, Bauman WA, Car- 48 Cameron HL, Perdue MH: Stress impairs murine intestinal barrier dozo C: iNOS expression in rat aorta is increased after spinal cord function: improvement by glucagon-like peptide-2. J Pharmacol transection: a possible cause of orthostatic hypotension in man. Exp Ther 2005; 314(1): 214–220.
Neurosci Lett 2007; 415(3): 210–214.
49 Napolitano LM, Koruda MJ, Zimmerman K, McCowan K, Chang 64 Delrue-Perol et C, Li KS, Vitiel o S, Neveu PJ: Peripheral catechola- J, Meyer AA: Chronic ethanol intake and burn injury: evidence mines are involved in the neuroendocrine and immune effects of for synergistic alteration in gut and immune integrity. J Trauma LPS. Brain Behav Immun 1995; 9(2): 149–162.
1995; 38(2): 198–207.
65 Mailman D: A role for abdominal vagal afferents in lipopolysac- 50 Maes M, Lin A, Bosmans E, Vandoolaeghe E, Bonaccorso S, Kenis charide-induced hypotension. Shock 2002; 18(2): 177–181.
G, De Jongh R, Verkerk R, Song C, Scharpé S, Neels H: Serotonin- 66 Rogausch H, Vo NT, Del Rey A, Besedovsky HO: Increased sensi- immune interactions in detoxified chronic alcoholic patients tivity of the baroreceptor reflex after bacterial endotoxin. Ann N without apparent liver disease: activation of the inflammatory Y Acad Sci 2000; 917: 165–168.
response system and lower plasma total tryptophan. Psychiatry 67 Maes M, Meltzer H, Jacobs J, Suy E, Calabrese J, Minner B, Raus J: Res 1998; 78(3): 151–161.
Autoimmunity in depression: increased antiphospholipid auto- 51 Song C, Lin A, De Jong R, Vandoolaeghe E, Kenis G, Bosmans E, antibodies. Acta Psychiatr Scand 1993; 87(3): 160–166.
Whelan A, Scharpe S, Maes M: Cytokines in detoxified patients with 68 Levin MC, Lee SM, Kalume F, Morcos Y, Dohan FC Jr, Hasty KA, chronic alcoholism without liver disease: increased monocytic Cal away JC, Zunt J, Desiderio D, Stuart JM: Autoimmunity due to cytokine production. Biol Psychiatry 1999; 45(9): 1212–1216.
molecular mimicry as a cause of neurological disease. Nat Med 52 Lim SG, Menzies IS, Lee CA, et al: Intestinal permeability and function in patients infected with human immunodeficiency 69 Wucherpfennig KW: Structural basis of molecular mimicry. J virus. A comparison with coeliac disease. Scand J Gastroenterol Autoimmun 2001;16(3): 293–302.
1993; 28: 573–580.
53 Sundqvist T, Lindstrom F, Magnusson KE, et al: Influence of fasting on intestinal permeability and disease activity in patients with rheu- matoid arthritis. Scand J Rheumatol 1982; 11: 33–38.
Copyright 2008 Neuroendocrinology Letters ISSN 0172–780X • www.nel.edu

Source: http://www.integratedhealth.com.au/wp-content/uploads/2014/04/MaesGutBrainDepression2-Copy1.pdf