Background
Mortality induced by sepsis is highly associated with secondary acute lung injury. Systemic inflammation during sepsis leads to acute respiratory distress syndrome (ARDS) caused by an exacerbated response of the immune system to bacteria and their products [
1‐
4]. Indeed, mice subjected to experimental model of sepsis induced by cecal ligation and puncture (CLP) show deregulation in pulmonary immune response, marked by cytokine storm and intense accumulation of activated leukocytes in lung tissue, including T lymphocytes [
5‐
8].
γδ T lymphocytes are unconventional lymphocytes that have antigen recognition properties fundamentally different from those of αβ T lymphocytes, and are comprised by distinct functional subsets, defined by the differential usage of Vγ and Vδ gene repertoire [
9,
10]. The Vγ4 T lymphocyte subset is highly associated with lung immune surveillance and increases in number in mouse lungs at early time points during bacterial infections [
10‐
13]. Increased migration of γδ T cells into the lungs has been previously demonstrated during experimental sepsis; however, the identification of γδ T cell subtypes has not been previously described [
7,
8,
14].
The migration of γδ T lymphocytes is largely dictated by the activation of chemokine receptors by their counterpart ligands, among which members from both CC and CXC subfamilies play compelling roles [
15‐
17]. Once at the infection site, these cells can rapidly respond to microbial antigens via innate surface receptors [
18‐
21], producing high amounts of interferon (IFN)-γ and interleukin (IL)-17, which are signature cytokines produced by specific subsets of γδ T cells [
22‐
26]. Vγ4 T lymphocytes represent one of the major subsets that produce IL-17 in different experimental models [
27‐
30].
γδ T lymphocytes have been shown to play divergent roles in mouse models of sepsis [
8,
14,
31‐
34]. The protective role of γδ T lymphocytes during experimental sepsis has been attributed to the production of IL-17, a cytokine that triggers neutrophil recruitment and improves bacterial clearance [
33,
35‐
37]. Furthermore, the accumulation of activated γδ T lymphocytes in the lungs of CLP mice has been correlated with beneficial outcome of septic mice [
8,
14]. Here we show that during the course of experimental severe sepsis, Vγ4 T lymphocytes migrate into injured lungs of CLP mice and exert a protective role via the production of IL-17.
Discussion
Sepsis triggers a complex immune response that involves both innate and adaptive systems. γδ T lymphocytes represent a link between these two branches of the immune system, by coordinating the activation of different cell populations via cytokine production [
27]. γδ T lymphocytes have been described as a major source of IL-17 in peritoneum and lymphoid organs during experimental sepsis, a phenomenon shown to present either beneficial or deleterious effects, depending on the experimental model [
33‐
35,
38]. The data presented here identifies the Vγ4 subset as a dominant producer of IL-17 in the lungs of septic mice and as a central T cell population involved in host defense against sepsis.
The experimental model of severe sepsis used in the present work resulted in the accumulation of T lymphocytes in lung tissue, which were likely originated from lymphoid organs. Increased numbers of both γδ and αβ T lymphocyte subsets were detected in the lungs; however it is noteworthy that, differently from αβ T lymphocytes, γδ T cell numbers continually increased up to day 10 after CLP, mainly due to the accumulation of Vγ4 subset. The progressive accumulation of γδ T cells in the lungs of CLP-operated mice has been previously demonstrated by Hirsh and coworkers [
7]; however, the presence of γδ T cell subtypes has not been described. The decrease in αβ T cell numbers observed at day 3 after CLP is in accordance with several reports in mice and humans that demonstrate a reduction in circulating CD3
+ T lymphocytes during sepsis [
39]. This reduction is explained by a massive apoptotic event of T lymphocytes, which is correlated with severity and mortality in experimental animals and patients [
2,
39‐
41]. The fact that the percentage of CD25
+ T lymphocytes increased among γδ T lymphocytes in the lungs at early time points after CLP (day 1) suggests that γδ (but not αβ) T lymphocytes are constantly activated in lymphoid tissues during the course of sepsis and continuously migrate towards inflamed lungs. In accordance with our data, Matsushima and co-workers [
40] demonstrated the early activation of γδ T lymphocytes from peripheral blood of patients with sepsis and systemic inflammatory response syndrome. These patients presented increased percentages of peripheral CD69
+ γδ T cells at acute time points after injuries, whereas CD69 expression by αβ T cells did not increase during a 2-week period [
40]. It is noteworthy that, in our study, such early activation was also evident for Vγ4 T cell population, as observed in mouse spleen and lungs 1 day after CLP.
The selective migration of γδ T lymphocyte subsets into the tissue during inflammation is dictated by elevated levels of chemoattractant mediators in the tissue and by the expression pattern of chemokine receptors on cell surface [
9,
10,
15,
16,
42,
43]. Our results suggest that γδ and Vγ4 T cell migration into the lungs of CLP mice is likely accounted by the combined
in situ accumulation of multiple chemokines. CLP-induced lung inflammation increased tissue levels of CCL2, CCL3 and CCL5, chemokines that are known to mediate γδ T lymphocyte migration
in vivo and
in vitro [
44‐
47]. Consistently, here we show that adoptively transferred γδ T cells preferentially accumulated in the lungs (rather than in blood or spleen) of recipient CLP mice. Even though CCL2, CCL3 and CCL5 are also chemoattractant for αβ T lymphocytes [
48], the neutralization of these chemokines in CLP lung homogenates did not impair the chemotaxis of γδ
- T lymphocytes (data not shown), suggesting that these chemokines selectively dictate the migration of γδ T cells into the lungs in our model. The involvement of CCL25 in γδ T cell migration towards inflamed lungs during sepsis was also investigated by us, since CCL25 has been shown to attract IL-17
+ γδ T cells into inflamed airways [
15]. However, CCL25 was not enhanced in CLP mouse lungs (data not shown).
Our data demonstrate that the Vγ4 T lymphocyte subset predominated among the IL-17
+ cell populations in CLP mouse lungs. In line with this, it has been demonstrated that mice lacking γδ T cells (but not αβ T cells) subjected to CLP failed to present elevated IL-17 levels in the plasma and peritoneal lavage, showing that γδ T cells are the major producers of IL-17 during experimental sepsis [
33‐
35,
38]. It has been established that, among murine γδ T lymphocytes, IL-17 production is restricted to Vγ4 and Vγ6 subtypes [
27,
49]. Consistently with our data, Vγ4 T lymphocytes comprise the major subset that migrates into the lungs and have been shown to produce IL-17 in different experimental models [
10,
13,
24‐
30]. It is noteworthy that our supplemental data (Additional file
1: Figure S1A) demonstrate that IL-17 production by γδ T cells from CLP-mouse lungs predominated over the expression of other cytokines, including IFN-γ. These data are reinforced by the increase in the percentage of CD27
- γδ
+ (and Vγ4
+) population in CLP-mouse spleen and by the fact that, upon α-CD3 mAb
in vitro stimulation, these cells were enriched for IL-17 but not for IFN-γ (Additional file
1: Figure S1B–C).
Increased numbers of γδ T lymphocytes in the blood, peritoneum and lungs have been correlated with sepsis positive outcome in patients and experimental animals [
8,
14,
32,
35,
50]. Indeed, mice lacking γδ T lymphocytes and subjected to CLP presented increased mortality rate and decreased survival periods [
14,
32]. The protective role of γδ T lymphocytes during sepsis results from the ability of these cells to produce inflammatory mediators capable to modulate other leukocyte populations, among which IL-17 is of particular importance [
35,
51,
52]. Here we show that IL-17 production in the lungs of CLP mice depends on infiltrated Vγ4 γδ T cell subset, which likely contributes to host protective immune response. Since adverse roles have been proposed for IL-17 during experimental sepsis, the effect of IL-17 in the lungs of CLP mice needs further investigation. It has been described that IL-17 derived from γδ T cells promotes epithelial repair in different tissues [
53‐
55], suggesting that IL-17 produced by Vγ4 T cells might act on lung epithelium, promoting tissue repair and ameliorating mouse illness after CLP [
33‐
35,
56]. IL-17 has also been associated with neutrophil influx into inflamed tissue, which can lead to either protective or harmful outcomes [
35,
37,
57,
58]. Concerning lung immune response, effective bacteria clearance by neutrophils reduces the risk of lung failure [
36]; however, it is well known that excessive neutrophil activation and production of myeloperoxidase (MPO) can cause tissue damage [
57]. In our study, we observed increased neutrophil numbers in the lungs of CLP mice, which was significantly reduced after anti-Vγ4 mAb treatment (data not shown), suggesting neutrophil involvement in the resolutive response. The involvement of tissue-recruited neutrophils coordinated by IL-17
+ γδ T cells in tissue repair has been demonstrated in different experimental models. γδ T cell knockout (KO) mice submitted to inflammatory insults are shown to present reduced neutrophil and MPO accumulation in the lungs, liver and cornea, which correlated with increased lesions and delayed epithelial regeneration [
53‐
55]. Moreover, in a model of corneal epithelial abrasion, it was demonstrated that γδ T cells induced, via IL-17, the production of vascular endothelial growth factor (VEGF) by neutrophils, promoting corneal nerve regeneration [
59]. Our study evidenced that IL-17
+ Vγ4 T lymphocytes migrate into injured lungs of CLP mice, presenting a beneficial role during the course of sepsis.
Methods
Cecal ligation and puncture
Polymicrobial sepsis was induced by cecal ligation and puncture (CLP) in normal fed and anesthetized (112.5 mg/kg of ketamin and 7.5 mg/kg of xylazine, i.p. Rhobifarma, Brazil) male C57BL/6 mice (18 to 20 g) provided by Oswaldo Cruz Foundation breeding unit (Rio de Janeiro, Brazil). After laparotomy (incision of 0.5–1 cm), the cecum was ligated with a cotton suture distal to the ileocecal valve to avoid bowel obstruction, and punctured nine times with a 21-gauge needle. The cecum was placed back into the abdomen and the incision was closed by a 4–0 polyamide suture. Sham-operated animals received midline laparotomies, exteriorization of the cecum with its immediate return and closure of incisions. Mice were resuscitated by a subcutaneous injection of 1 ml sterile saline solution. Mice were treated with ertapenem (Merck, Germany; 75 mg/kg, i.p.) 6, 24 and 48 h after surgery. For lung analysis, mice were euthanized in a CO2 chamber 1, 3, 7 and 10 days after CLP operation. For the assessment of survival rate, mice were evaluated every 12 h following CLP until death. During all experimental procedures, mice were monitored daily and those that presented impaired locomotor activity and no struggle response to sequential handling were euthanized. All experimental procedures were performed according to the Committee on Ethical Use of Laboratory Animals of Oswaldo Cruz Foundation (Fiocruz, Brazil, #L62/12).
Antibody treatment
Hamster anti-TCR γδ (3A10, anti-pan-δ, described by Itohara
et al. [
60]) and anti-Vγ4 (UC3-10A6, described by Dent
et al. [
61]) monoclonal antibodies (mAb) were obtained from SCID mice (Oswaldo Cruz Foundation breeding unit, Rio de Janeiro, Brazil) ascitic fluid. 3A10 preparation was further purified/concentrated by Protein G (GE Healthcare, USA) affinity chromatography while UC3 was concentrated by ammonium sulfate precipitation. Both antibody preparations were dialyzed against saline solution before use. mAbs were i.p. administered (500 μg/mice every other day for 7 days, starting 1 day before CLP). Control mice were similarly sham-treated with normal hamster serum IgG.
Recovery of leukocytes from lung and spleen
Lung tissue samples were obtained from euthanized C57BL/6 mice at 1, 3, 7 and 10 days after CLP, macerated in RPMI 1640 medium containing collagenase type IV (250 IU/ml, 37 °C, 30 min) and centrifuged (400 g, 10 min). Spleens were dissected, macerated in PBS containing EDTA (10 mM, pH 7.4), and centrifuged (420 g for 10 min at 20 °C). Cell pellets from lung and spleen were re-suspended in 3 ml of PBS/EDTA and subjected to centrifugation on a Histopaque 1083 gradient (400 g for 30 min) for mononuclear cell separation.
Flow cytometric analysis
Leukocytes were stained with the appropriate concentration of the following antibodies: PE/FITC CD3 (145–2C11), PE/FITC TCR δ chain (GL3), PE TCR β chain (H57–597), FITC Vγ4 TCR (UC3-10A6), FITC Vδ4 TCR (GL2), FITC CD25 (7D4), PE/FITC IgG1 and IgG2 isotypes (BD Pharmingen, USA) and APC Vγ1 TCR (2.11) (Biolegend, USA). For intracellular cytokine staining, cells were pre-incubated for 4 h with PMA (20 ng/ml), ionomycin (500 ng/ml) and brefeldin A (10 μg/ml) at 37 °C and 5 % CO
2. After surface marker staining, cells were fixed, permeabilized and stained with anti-IFN-γ, anti-TNF-α, anti-IL-4, anti-IL-10, anti-IL-12 and anti-IL-17 antibodies (BD Pharmingen, USA). IgG isotypes were used as irrelevant antibodies. Cells were acquired by FACScalibur flow cytometer (Becton Dickinson, USA) and analyzed either by Cell Quest or FlowJo softwares. Counts are reported as percentage and as numbers of cells after the multiplication of the percentage of T lymphocyte population by the total number of leukocytes. Gating strategies are shown in additional files (Additional file
3: Figure S3 and Additional file
4: Figure S4).
Adoptive transfer assay
Naïve C57BL/6 splenocytes were labeled with CFSE (Invitrogen USA, 1 μM/8×106 cells) and i.v. injected (4 × 107 cells, ≥ 90 % viability) into recipient mice 3 and 8 days after CLP or sham operations. Recipient mice were euthanized 10 days after adoptive transfer and their lungs were recovered for leukocyte analysis.
Preparation of lung homogenates
Lung homogenates were prepared by homogenizing perfused whole lung tissue using a glass potter homogenizer (Kontes Glass Company, USA) in 2 ml of PBS containing cell lysis buffer (Sigma Aldrich, USA) and protease inhibitor (1 μl/ml; Sigma Aldrich, USA), at 4 °C. The homogenates were centrifuged (8400 g for 30 min, 4 °C) and the supernatants were filtered (0.2 μm). For chemotaxis assays, lungs were homogenized using PBS only.
Cytokine quantification
Levels of chemokines were evaluated in lung homogenates from lungs recovered 7 days after CLP surgery by sandwich enzyme-linked immunosorbent assay (ELISA) by using matched antibody pairs from R&D (Minneapolis, MN), according to manufacturer’s instructions. IL-17 quantification was performed using the BD™ Cytometric Bead Array (CBA) mouse Th1/Th2/Th17 kit (BD Biosciences, USA), and samples were analyzed using a FACScalibur flow cytometer.
Transwell migration assay
Spleen T lymphocytes (3 × 106 in HBSS without Ca2+/ Mg2+) were placed in the upper chamber of 5.0 μm pore diameter transwell tissue culture inserts (BD Falcon, USA). Transwell inserts were placed in the individual wells of a 24-well cell culture plate containing assay buffer or lung homogenates from naïve, sham-operated and CLP-operated mice, neutralized (30 min, 37 °C) with anti-CCL2 mAb (2.5 ng/well), anti-CCL3 mAb (200 ng/well) or anti-CCL5 mAb (50 ng/well). The recombinant chemokines rmCCL2 (2.5 ng/well), rmCCL3 (4 ng/well) and rmCCL5 (4 ng/well) (R&D Systems, USA) were used as positive controls. After 2 h, the migrated cells were counted, labeled as described above, and analyzed by FACScalibur. Results are expressed as chemotactic index, generated by using the number of cells that migrated towards buffer as comparison.
Statistical analysis
Data are reported as the mean ± SEM and were statistically evaluated by analysis of variance (ANOVA) followed by Newman-Keuls-Student test or Student’s t test. Values of p ≤ 0.05 were regarded as significant.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contribution
MFSC, CBTN and VUB performed the experiments and analyzed the data; RHV, JM, and MGH contributed with reagents and analysis tools; MFSC, VUB, RHV, JM, CFB and CP conceived and designed the experiments; RHV and CFB critically reviewed the manuscript; MFSC and CP wrote the manuscript. All authors read and approved the final manuscript.
Claudia Farias Benjamim, Carmen Penido are considered last authors.