According to our previous reports [9], [10] circulating Ms isolated from CF patients fail to mount an appropriate inflammatory response in the presence of Gram-negative endotoxin. The reported “ET-signature” (IL10^high/IL12^low/IL23^low) was corroborated in this new cohort of CF patients (n = 14) (Fig. 1A–1C) as well as a low TNFα production after ex vivo LPS challenge (Fig. 1D). However, endotoxin tolerance was detected neither in healthy controls (n = 11) nor in patients suffering from the respiratory disease COPD (n = 8). A group of septic patients used as positive control was included in our study (n = 8). Additionally, cells from CF as well as sepsis patients showed high phagocytosis ability and impaired antigen presentation (Fig. 1E and 1F), both are main features of an endotoxin tolerant phenotype [1], [10]. Note that a set of other cytokines and chemokines (IL6, CCL3, CCL4, CCL20 and CCL22) were checked to verify the ET status of CF patients (data not shown).

Having established that CF patients suffer from an ET status, no strong evidence has reported the cause of this phenomenon. In the case of sepsis the presence of circulating traces of endotoxin or viable pathogen itself provides an explanation for their endotoxin tolerance. However, in general terms CF patients suffer a local rather than a systemic infection [20]. Others authors have reported noteworthy levels of microbial products in the circulation of HIV patients, a fact that correlates to systemic immune deregulation in chronic HIV infection [17], [21]–[23]. In those patients detected endotoxins probably derived from the gastrointestinal tract and have an impact on the innate immune control. These reports prompted us to study the presence of LPS in the plasma of CF patients. As shown in figure 2A, significant levels of endotoxin were detected in CF plasma (∼0.3 ng/ml). In contrast, low quantities of endotoxin were reported in both healthy controls and COPD patients (∼0.05 ng/ml). These data suggested a role for endotoxin in the generation of an ET status in these patients. In fact, in a global analysis including all groups, we found that endotoxin concentrations inversely correlated to the inflammatory response after an ex vivo LPS challenge (Fig. 2B–2C). Those Ms from patients with high levels of LPS (CF and sepsis) were unable to produce significant TNFα quantities after ex vivo LPS challenge (Fig. 2B). Similar results were obtained when the expression of IL23p19 mRNA was analysed (Fig. 2C). In contrast, LPS concentration correlates with high IL10 mRNA expression after ex vivo endotoxin stimulation (Fig. 2D). In addition, a significant correlation (R²∼0.7) was obtained when CF patients were analyzed (Fig. 2B–2D, insets). Note that two CF-patients were not included in the analysis (insets) because their LPS concentration was very low (<0.1 ng/mL) in comparison to the rest of the patients enrolled. However, their innate response after ex vivo LPS-challenge were consistent the observed tendency.

The detected levels of LPS in plasma of CF patients suggested the presence of microbes in this fluid. Similarly to the case of HIV patients, the source of plasma endotoxin could be bacteria translocation from the gastrointestinal tract [22], [23]. In addition, due to chronic airways colonization that these patients suffer, their lungs could be an alternative source of microbial translocation. To test this hypothesis we studied the putative presence of bacteria in CF plasma using a universal 16S rRNA PCR (seeMaterial and Methods). Consistent negative PCR results were obtained for the CF plasma since no bands were visualized in an Acrylamide Denaturing Gradient Gel Electrophoresis (Fig. 3). In addition, negative standard blood cultures were observed in duplicate fresh plasmas from these patients (data not shown).

In order to demonstrate that levels of endotoxin detected in plasma from CF patients could be a cause of the observed ET, we cultured human Ms isolated from healthy controls in presence of supplemental sera from CF patients (50% of total volume). After 48 hours, dishes were washed twice and fresh medium was added. Then, cells were challenged with LPS for 24 hours and the production of TNFα and IL6 was evaluated at the supernatant of each culture (see the experimental design in figure 4A). In those cultures “pre-treated” with sera from CF patients but no with controls both, TNFα and IL6 exhibited a marked down regulation of its expression, after LPS challenge (Fig. 4B and 4C). Moreover, we have also proven that LPS doses found in CF sera induce ET in healthy monocytes. To test that Ms were pre-stimulated with different doses of LPS, including those detected in CF patients, for six hours. Then, cells were washed and kept in completed medium for 16 hours. Next, a standard LPS dose (10 ng/ml) was added and production of TNFα and IL6 were evaluated 24 hours after in the supernants of these cultures (see the scheme in figure 4 D). Note that this experimental design was previously established by us as a solid human model for endotoxin tolerance [1], [10]. Data shown in figure 4E and 4F demonstrate that the endotoxin concentrations detected in the plasma of CF patients are sufficient to induce an ET state in human circulating monocytes (∼0.3 ng/ml, figure 2A). In this way, we noticed a down regulation of both TNFα and IL6 generation when cells were re-challenged with a standard dose of LPS. Additionally, IL23p19 mRNA expression is also down regulated and IL10mRNA shows a patent increase. Others markers of ET phenotype were also checked and showed a tolerant behavior (data not shown).

The following antibodies were used: anti-CD3-PE (Becton Dickinson; CA, USA); anti-CD14-APC (Miltenyi Biotec; CA, USA). The medium used for cell culture was Dulbecco's MEM from Invitrogen (Paisley, UK). LPS from Salmonella abortus was a kind gift from Dr. Galanos (Max-Planck-Insitut für Immunobiologie, Freiburg, Germany). All other reagents were obtained from Sigma-Aldrich (Saint Louis, MI, USA), unless otherwise stated.

Starting from peripheral blood from CF, COPD, septic patients or healthy volunteers, adherent cells were purified following the same protocol employed for buffy coats in previous reports from our laboratory [5], [9], [10], [40]–[42]. Note that plasmas from these patients were collected for other analysis. The composition of the adherent population was analyzed by FACS (∼92% of CD14). Once plated, cells were exposed or not to LPS for various times according to particular experiments. All reagents used for cell culture were endotoxin-free, as assayed with the Limulus Amebocyte test (Cambrex; North Brunswick, NJ, USA).

Once seeded, adherent cells were treated or not with indicated concentrations of LPS during the “time of tolerization” (t_tol = 8 h). After that, cells were washed three times with PBS and kept in complete medium for different times through the phase called “time of recovery” (t_rec = 16 h). Then cells were re-stimulated or not with 10 ng/ml of LPS for 24 hours. Note that this model was established by our group in a previous study [10].

Cells were washed once with PBS and their RNA was isolated using the High Pure RNA Isolation Kit from Roche Diagnostics (Mannheim, Germany). cDNA was obtained by reverse transcription of 1 µg of RNA using the High Capacity cDNA Reverse Transcription kit from Applied Biosystems (Foster City, CA, USA).

Gene expression levels were analyzed by real-time quantitative PCR (Q-PCR) using the LightCycler system from Roche Diagnostics and cDNA obtained as described above. Q-PCR was performed using a QuantiMix Easy SYG kit from Biotools (Madrid, Spain) and specific primers. Results were normalized to the expression of the β-actin, and the cDNA copy number of each gene of interest was determined using a seven-point standard curve as we described before [5], [9], [10], [40]–[42].

The sequences of oligonucleotides used and their annealing temperatures are:

IL-10: sense 5′-ATG CCC CAA GCT GAG AAC CA-3′, antisense 5′-TCT CAA GGG GCT GGG TCA GC-3′ (58°C);

IL12p40: sense 5′-GAC ATT CAG TGT CAA AGC AGC A-3′, antisense 5′-CCT TGT TGT CCC CTC TGA CTC T-3′ (64°C);

IL23p19: sense 5′-GTT CCC CAT ATC CAG TGT GG-3′, antisense 5′-GAG GCT TGG AAT CTG CTG AG-3′ (60°C);

β-actin; sense 5′-GTG GGG CGC CCC AGG CAC CA-3′, antisense 5′-CTC CTT AAT GTC ACG CAC GAT TTC-3′ (60°C).

All primers were synthesized, desalted, and purified by Bonsai Biotech (Madrid, Spain).

Concentrations of TNF-α in supernatants were determined using the ELISA development kit supplied by PeproTech (Rocky Hill, NJ, USA). IL-6 levels in supernatants were determined with a commercial ELISAs purchased from Bender MedSystem (Burlingame, CA, USA).

We followed protocols previously described by de las Heras and co-workers [43] or da Silva and co-workers [44].

We followed a protocol previously described by Hernández-Fuentes and co-workers and Adam and co-workers [45], [46].

The LPS concentrations were determined in 200 µl of plasma using a kit based on a Limulus amaebocyte extract (LAL kit endpoint-QCL1000, Cambrex BioScience, Walkersville, MD). Determinations were done five times per sample.

DNA was extracted from 200 µl of plasma using a QIAamp tissue kit (Qiagen, Hilden, Germany), and a universal bacteria set of primers for 16S rRNA (sense 5′-ATT AGA TAC CCT GGT AGT CCA-3′ and antisense 5′-AGG CCC GGG AAC GTA TTC AC-3′) were used yielding an amplicon size of ca. 550 bp. All PCRs were carried out in a final volume of 50 µl containing 100 ng of DNA, 0.5 µM of each primer, 0.2 mM of dNTPs, 100 ng/ml of BSA, 3 mM of MgCl2, and 2 U of FastStart Taq polymerase (Roche Diagnostics, Indianapolis, IN, USA). The thermal cycling conditions used were as follows: an initial DNA denaturation step at 95°C for 7 min, followed by 40 cycles of denaturation at 95°C for 30 s, primer annealing at 52°C for 45 s, and a final extension at 72°C for 45 s. Positive amplicons were visualized in both 0.8% agarose gels and separated in vertical electrophoresis polyacrylamide gels (8%) at 60°C; the urea-formamide denaturating gel gradient (33–43%) was submitted to 130 V during 330 min. Gels were visualized with ethidium bromide.

The number of patients or experiments analyzed is indicated in each figure. In the case of in vitro assays, data were collected from a minimum of three experiments to calculate the mean ± SD and the statistical significance was calculated using the unpaired Student's test and differences were considered significant at p values<0.05 using Prism 5.0 software (GraphPad, San Diego, CA, USA). Also ANOVA analysis following of a Turkey test was performed.