British Journal of Nutrition (2022), 127, 384–397 doi:10.1017/S0007114521001124 © The Author(s), 2021. Published by Cambridge University Press on behalf of The Nutrition Society. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited. n-3 long-chain PUFA promote antibacterial and inflammation-resolving effects in Mycobacterium tuberculosis-infected C3HeB/FeJ mice, dependent on fatty acid status Arista Nienaber1*, Mumin Ozturk2,3, Robin Dolman1, Renee Blaauw4, Lizelle L. Zandberg1, Simone van Rensburg1, Melinda Britz1, Frank E. A. Hayford1,5, Frank Brombacher2,3,6, Du Toit Loots7, Cornelius M. Smuts1, Suraj P. Parihar2,3,6,8 and Linda Malan1 1Centre of Excellence for Nutrition, North-West University, Potchefstroom, South Africa 2International Centre for Genetic Engineering and Biotechnology (ICGEB), Cape Town-Component, University of Cape Town, Cape Town, Western Cape, South Africa 3Institute of Infectious Diseases and Molecular Medicine (IDM), Division of Immunology and South African Medical Research Council (SAMRC) Immunology of Infectious Diseases, University of Cape Town, Cape Town, Western Cape, South Africa 4Division of Human Nutrition, Stellenbosch University, Tygerberg, Cape Town, Western Cape, South Africa 5Department of Nutrition and Dietetics, University of Ghana, Accra, Ghana 6Wellcome Centre for Infectious Diseases Research in Africa (CIDRI-Africa), Institute of Infectious Diseases and Molecular Medicine (IDM), University of Cape Town, Cape Town, Western Cape, South Africa 7Human Metabolomics, Faculty of Natural and Agricultural Sciences, North-West University, Potchefstroom, South Africa 8Division of Medical Microbiology, Department of Pathology, Faculty of Health Sciences, University of Cape Town, Cape Town, Western Cape, South Africa (Submitted 17 February 2020 – Final revision received 29 January 2021 – Accepted 9 March 2021 – First published online 5 April 2021) Abstract Non-resolving inflammation is characteristic of tuberculosis (TB). Given their inflammation-resolving properties, n-3 long-chain PUFA (n-3 LCPUFA) may support TB treatment. This research aimed to investigate the effects of n-3 LCPUFA on clinical and inflammatory outcomes ofMycobacterium tuberculosis-infected C3HeB/FeJmicewith either normal or lown-3 PUFA status before infection. Using a two-by-two design, uninfected mice were conditioned on either an n-3 PUFA-sufficient (n-3FAS) or -deficient (n-3FAD) diet for 6 weeks. One week post-infection, mice were randomised to either n-3 LCPUFA supplemented (n-3FAS/n-3þ and n-3FAD/n-3þ) or continued on n-3FAS or n-3FAD diets for 3 weeks. Mice were euthanised and fatty acid status, lung bacterial load and pathology, cytokine, lipid mediator and immune cell phenotype analysed. n-3 LCPUFA supplementation in n-3FAS mice lowered lung bacterial loads (P= 0·003), T cells (P= 0·019), CD4þ T cells (P= 0·014) and interferon (IFN)-γ (P< 0·001) and promoted a pro-resolving lung lipidmediator profile. Comparedwithn-3FASmice, then-3FAD group had lower bacterial loads (P= 0·037), significantly higher immune cell recruitment and a more pro-inflammatory lipid mediator profile, however, significantly lower lung IFN-γ, IL-1α, IL-1β and IL-17, and supplementation in the n-3FAD group provided no beneficial effect on lung bacterial load or inflammation. Our study provides the first evidence that n-3 LCPUFA supplementation has antibacterial and inflammation-resolving benefits in TB when provided 1 week after infection in the context of a sufficient n-3 PUFA status, whilst a low n-3 PUFA status may promote better bacterial control and lower lung inflammation not benefiting from n-3 LCPUFA supplementation. Key words: Host-directed therapy: Inflammation: n-3 long-chain PUFA: Tuberculosis The bacterial manipulation of host responses in tuberculosis patients(1,2). In addition, TB patients endure drug side effects (TB) favours bacterial growth and excessive inflammation, with and toxicity, long treatment periods and poor cure rates(3). the resultant lung tissue damage that persists in some TB Host-directed therapy (HDT), aimed at enhancing the host’s Abbreviations: AA, arachidonic acid; FA, fatty acid; HDT, host-directed therapy; HEPE, hydroxyeicosapentaenoic acid; HETE, hydroxyeicosatetraenoic acid; IFN-γ, interferon-γ; Mtb, Mycobacterium tuberculosis; PBMC, peripheral blood mononuclear cell; SPM, specialised pro-resolving mediator; TB, tuberculosis; n-3FAS; n-3 PUFA-sufficient diet; n3FAD, n-3 PUFA-deficient diet; n-3 LCPUFA, n-3 long-chain PUFA.. * Corresponding author: Arista Nienaber, email arista.nienaber@nwu.ac.za Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 385 response to infection, rather than treatment strategies directed at cagewith filter tops (type 2 long), as well as driedwood shavings bacterial killing, has lately been suggested for improving current and shredded filter paper as floor coverings. The temperature TB treatment regimens(3). Since TB is characterised by excessive, rangewas set at 22– 24 °C and 12-to-12 h light cycles. The experi- non-resolving inflammation, various anti-inflammatory drugs ments were performed in accordance with the South African have been investigated for use as possible HDT options(4,5). National Guidelines and University of Cape Town practice These medications have been shown to reduce lung lesions guidelines for laboratory animal procedures. The protocol was and bacillary load, favouring host survival(4,6,7). However, they approved by the Animal Ethics Committee, Faculty of Health are not without side effects and, therefore, a nutritional approach Sciences, University of Cape Town (AEC 015/040) and the may be considered a safer alternative(8). AnimCare Animal Research Ethics Committee of the North- Dietary n-3 long-chain PUFA (n-3 LCPUFA) consumption West University (NWU-00260-16-A5). alters membrane phospholipid fatty acid (FA) composition of blood and tissue cells that play a role in immune and inflamma- Experimental design and animal diets tory responses(9–11). It is well known that various lipid mediators, Mice had ad libitum access to food and water. The experimental synthesised from n-3 LCPUFA, contribute to inflammation design of the present study is illustrated in Fig. 1. Mice were ran- resolution. Eicosapentaenoic acid (EPA) and docosahexaenoic domly allocated to an n-3 PUFA-deficient (n-3FAD) (n 20) or - acid (DHA) serve as precursors for specialised pro-resolving sufficient diet (n-3FAS) (n 20) and kept on these diets for 6 mediators (SPM), including resolvins, protectins and maresins. weeks prior to infection, in order to establish a sufficient or a These SPMplay a role in significantly reducing pro-inflammatory low n-3 PUFA status. The n-3FAS diet contained the essential lipid mediator, chemokine and cytokine production and altering n-3 PUFA α-linolenic acid. Mice were then infected via the aero- immune cell recruitment, whilst promoting anti-inflammatory (12) sol route (described below) and their respective dietsmaintainedcytokine release . The incorporation of dietary EPA and for an additional week. One week post-infection (week 7), mice DHA into cell membranes has also been found to enhance the that were conditioned on the n-3 PUFA-sufficient diet (n-3FAS) phagocytosis of apoptotic cells and bacteria, whilst SPMpromote were randomised to continue on this diet (n-3FAS) (n 10) or bacterial killing(12,13). Although these functions have not been were switched to the same diet supplemented with n-3 proven in TB specifically, n-3 LCPUFA have been successfully LCPUFA (EPA plus DHA) (n-3FAS/n-3þ group, n 10) (Fig. 1). used as anti-inflammatory and inflammation-resolving agents Similarly, the mice in the n-3FAD group either continued on in other conditions driven by inflammation(9). the n-3FAD diet (n 10) or were switched to the n-3 LCPUFA-sup- Considering this, it is reasonable to hypothesise that n-3 plemented diet (n-3FAD/n-3þ group, n 10). The mice received LCPUFA supplementation would benefit TB patients, but these diets for an additional 3 weeks until euthanasia at 28 d after research on the application of n-3 LCPUFA as HDT in TB is lim- infection (as described below). The welfare of the mice was ited at present. Moreover, the effects of n-3 LCPUFA supplemen- assessed daily and body weight and food intake were measured tation after the acute inflammatory response in Mycobacterium weekly. The daily food intake per mouse was calculated by tuberculosis (Mtb) infection have not yet been investigated. The dividing the weekly food intake by seven (days) and then by five aim of the present study is, therefore, to determine the effects of (five mice per cage). The results of this experiment were repro- EPA and DHA supplementation, administered 1 week after Mtb duced in a second experiment (resulting in ten mice per treat- infection for 28 d, on inflammatory, immune and clinical out- ment group). The data of one experiment (five mice per comes in C3HeB/FeJ mice. The well-established C3HeB/FeJ group) are presented in this article. mouse model has been reported to be the closest representative (14) All the purified experimental diets were obtained commer-murine model of human pulmonary TB lung histopathology . cially (Dyets) and were based on the AIN-93G(17) formulation, Furthermore, the n-3 LCPUFA status of the general human adult all containing 10 % fat, but with modifications in the fat source population is not considered optimal, owing to insufficient (Table 1). All the diets were isoenergetic with identical macronu- dietary n-3 PUFA consumption and high dietary n-6 (n-6)/n-3 (15,16) trient contents. The mice in the n-3FAS group received thePUFA ratios, often resulting in low n-3 PUFA status . We fur- AIN-93Gdiet, which provides bothn-3 andn-6 PUFA at amounts ther aim to mimic this scenario of possible suboptimal n-3 PUFA found to induce optimal tissue saturation of DHA and intakes among TB patients to determine whether supplementa- arachidonic acid (AA), in rodents(17). The EPA- and DHA- tion outcomes depend on n-3 PUFA status before Mtb infection supplemented diets (n-3þ) contained commercially obtained (interaction effects between n-3 PUFA status and n-3 LCPUFA Incromega TG4030 oil (Croda Chemicals) supplemented at supplementation). amounts that could reasonably be achieved in humans. GC-MS analysis was performed by the manufacturer to confirm the FA composition of the diets (Table 1). From this composition, Materials and methods the actual EPA and DHA intake could be calculated and was Animals and ethics statement expressed as percentage of total energy intake. Male C3HeB/FeJ mice (Jackson Laboratory), aged 10–12 weeks, Aerosol infection were bred and housed at the Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, Cape Town, A virulent Mtb H37Rv strain was cultured and stocks were pre- South Africa. Following infection, mice were housed in a bio- pared and stored at −80°C, as described elsewhere(18). Mice safety level 3 containment facility, five per individually ventilated were exposed to aerosol infection for 40 min by nebulising Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 386 A. Nienaber et al. Week 1 n-3FAS diet n-3FAD diet Week 2 (n 22) (n 22) Week 3 Week 4 Week 5 Week 6 Week 7 Blood sampling - baseline FA status and H37Rv strain infection Euthanasia of two mice per group to confirm infection dose (1 day post-infection) Week 8 Diet switch Week 9 Week 10 Week 11 Euthanasia Fig. 1. The study design of this research. Animals were fed an n-3 fatty acid-deficient diet (n-3FAD) or n-3 fatty acid-sufficient diet (n-3FAS) for 6 weeks. Baseline blood samples were collected to determine fatty acid status. Mice were then aerogenically infected with Mtb and after 1 week some animals were switched to n-3 long-chain PUFA-supplemented diets (n-3þ) for 3 weeks. Mice were then euthanised for end-point analysis. FA, fatty acid; n-3FAD, n-3 fatty acid-deficient diet; n-3FAS, n-3 fatty acid sufficient diet; n-3þ, n-3 long-chain PUFA-supplemented diet; /, switched to. Table 1. Fat source and fatty acid content of experimental diets* LA ALA AA DHA EPA Diet Fat source g/100g g/100g g/100g g/100g g/100g n-3FAS 70 g/kg Soyabean oil 3·54 0·44 < 0·01 < 0·01 < 0·01 30 g/kg Coconut oil n-3FAD 81 g/kg Coconut oil 1·30 0·01 < 0·01 < 0·01 < 0·01 19 g/kg Safflower oil n-3þ 70 g/kg Soyabean oil 3·44 0·43 < 0·01 0·06 0·09 27 g/kg Coconut oil 28% of total FA† 44% of total FA† 3 g/kg Incromega TG4030 ALA, α-linolenic acid; FA, fatty acids; LA, linoleic acid; n-3FAD, n-3 fatty acid-deficient; n-3FAS, n-3 fatty acid-sufficient; n-3þ, n-3 long-chain PUFA-supplemented diet. * Based on GC-MS analysis of diets. Values expressed as g/100 g of diet. † Indicates which percentage of the total FA in the diet is comprised of DHA or EPA. 6 ml of a suspension that contained 2·4 × 107 live bacteria in an Tween-80 for the analysis of the bacillary load and lung cyto- inhalation exposure system (model A4224, Glas-Col). One day kines. The right superior and post-caval lung lobes were snap- following infection, four mice were euthanised to confirm the frozen in liquid N2 and stored at −80 °C for lung FA and lipid infection dose, which was 500 colony-forming units/mouse. mediator analysis. The right middle lobe was submerged in 10 % neutral buffered formalin for histology analysis and the right inferior lobe prepared for flow cytometry. End point blood and tissue collection At the end of the 3 weeks of receiving intervention diets, mice Total phospholipid fatty acid composition analysis were euthanised by halothane exposure, followed by trunk blood collection by heart puncture. The bloodwas collected into FA were extracted from ∼20 mg lung tissue, homogenised in EDTA-coated Microtainer® tubes (K2EDTA, 1000 μl, BD), and 10 μl PBS with protease inhibitor (homogenisation buffer) per then centrifuged. The plasma and buffy coat were removed 1 mg tissue, or from ∼200 μl erythrocytes or peripheral blood for FA analysis. The erythrocytes were washed twice with saline mononuclear cells (PBMC) collected as buffy coat. Lipids were before storage at −80 °C and subsequent FA analysis. The lung extracted from each lipid pool with chloroform–methanol (2:1, lobes were removed aseptically and weighed prior to prepara- v:v; containing 0·01 % butylated hydroxytoluene) by a modifica- tion. The left lung lobe was homogenised in saline and 0·04 % tion of the method of Folch et al.(19) The lipid extracts were n-3FAS (n 10) n-3FAS/n-3+ (n 10) n-3FAD (n 10) n-3FAD/n-3+ (n 10) Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 387 concentrated and the neutral lipids separated from the phospho- A700 (Clone 500A2), CD62L-V450 (Clone MEL-14), CD19- lipids by TLC (silica gel 60 plates, Merck) and eluted with diethyl PerCPCy5.5 (Clone 1D3), CD8-APC (Clone 53-6.7) and ether–petroleum ether–acetic acid (30:90:1, v:v:v). The lipid KLRG1-BV786 (Clone 2F1) purchased from BD (Biosciences) band containing phospholipids was removed from the TLC plate and eBioscience (ThermoFisher)(20,21). and transmethylated with methanol–sulphuric acid (95:5, v:v) at 70°C for 2 h to form FAmethyl esters. FAmethyl esters were ana- Lipid mediator analysis lysed with an Agilent Technologies 7890A GC system equipped with an Agilent Technologies 7000B triple quad mass selective Lipid mediators in crude lung homogenates were analysed detector (Agilent Technologies) and quantification performed with liquid chromatography-tandem mass spectrometry. with Masshunter (B.06.00). Relative percentages of FA (% w/ 17-Hydroxydocosahexaenoic acid (17-HDHA); 5-, 11-, 12-, w) were calculated by taking the concentration of a given FA 15- and 18-hydroxyeicosapentaenoic acid (HEPE); 5-, 8-, 9-, as a percentage of the total concentration of all FA identified 11-, 12- and 15-hydroxyeicosatetraenoic acid (HETE); prosta- in the sample. glandin (PG)D1; PGE2; PGE3 and PGD2 concentrations were measured. Lipid mediators were extracted from ∼50 mg lung tis- Bacterial load determination sue, in 10 μl/mg homogenisation buffer, with solid-phase extrac- tion using Strata-X (Phenomenex). Themethodwasmodified for The bacterial loads of lungs were determined at euthanasia (28 d Strata-XSPE columns from a previously described method(22). after infection). The left lung of each mouse was aseptically Data were quantified with Masshunter B0502, using external cal- removed, weighed, homogenised and serial dilutions were ibration for each compound and internal standards (PGD2-d4, plated onto DifcoTM Middlebrooks 7H10 Agar (BD PGE2-d4, PGF2-d4 and 5-and 12-HETE-d8; 1000 pg of each Biosciences) medium with oleic acid–albumin–dextrose– (Cayman Chemicals)) to correct for losses and matrix effects. catalase supplementation and 0·005 % glycerol. The colony- forming units were determined 21 d following incubation at 37°C. Data are expressed as log colony-forming units. Cytokine analysis10 The left lung lobe homogenates leftover from determining bac- Histopathology analysis terial load were centrifuged at 2000 g for 5 min and the superna- Right middle lobes of the lungs were dissected out and fixed in tant was frozen at −80 °C until analysis. The cytokines were 10 % neutral buffered formalin. The tissue was processed using measured in cell-free lung homogenates, using the Quansys the Leica TP 1020 Processor for 24 h and subsequently Biosciences Q-Plex™ Mouse Cytokine Screen (West Logan, embedded in paraffin wax. The Leica Sliding Microtome WV) Q-Plex Array 16 plex (IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, 2000R was used to cut 2-μm thick sections of the embedded IL-6, IL-10, IL-12p70, IL-17, monocyte chemoattractant tissues. Three sections with 30 μm distance apart per section protein-1, interferon-γ (IFN-γ), TNF-α, chemokine ligand 3 were cut, deparaffinised and subsequently stained with the (CCL3), granulocyte-macrophage colony-stimulating factor, haematoxylin/eosin stain. The images were acquired in Nikon RANTES) according to manufacturer instructions, using the Eclipse 90i microscopes and analysed with NIS-Elements AR soft- QView Imager Pro, Q-View Software. ware (Nikon Corporation) to determine the granulomatous area and alveolar space as a percentage of the total lung tissue(20). Statistical analysis Using the G*Power statistical package version 3.1.9.7, a two-way Flow cytometry ANOVA power analysis was done. A total sample size of 34 was Briefly, single-cell suspensions from the lung tissues were calculated for an α of 0·05, a power of 80 % and an effect size prepared by chopping them into small pieces followed by estimated at 0.5. Therefore, a total sample size of 40 mice was incubation in Dulbecco’s Modified Eagle Media containing included in this research in two experiments (n 20 each) of five 0·18 mg/ml collagenase type I (Sigma), 0·02 mg/ml DNase I mice per group. Data are presented asmeans and standard errors (Sigma) for 1 h at 37°C under constant rotation, followed by of the means. Statistical analyses were performed using IBM being mechanically passed through a 100 μm and 70 μm cell SPSS statistics software (version 25; IBM Corporation). To deter- strainer sequentially. Erythrocytes were lysed using RBC lysis mine the differences between FA composition at baseline in the buffer (155 mM NH4Cl, 12 mM NaHCO3, 0·1 mM EDTA). n-3FADandn-3FAS group, the Student Fischer t test for indepen- Cells were then counted and subjected to flow cytometry. dent variableswas used. Themain effects ofn-3 LCPUFA supple- Lymphoid and myeloid compartments were investigated in mentation (n-3FAS/n-3þ and n-3FAD/n-3þ v. n-3FAS and n- the lung samples of mice on various intervention diets. 3FAD) and a low pre-infection n-3 PUFA status (n-3FAD and Antibodies used for flow cytometry analysis were as follows: n-3FAD/n-3þ v. n-3FAS and n-3FAS/n-3þ), and their interac- CD64-PeCy7 (Clone X54-5/7.1), Ly6C-PerCPCy5.5 (Clone tion (pre-infection status × n-3þ), on all outcome variables, AL-21), CD11b-V450 (Clone M1/70), MHCII-APC (Clone M5/ were analysed by using two-way ANOVA. Significant treatment 114.15.2), CD103-PE (Clone M290), CD11c-A700 (Clone effects in the absence of a significant interaction effect indicate HL3), SiglecF-APCCy7 (Clone E5-2440), Ly6G-FITC (Clone additive effects of the treatments, whereas a significant interac- 1A8), PD-1-FITC (Clone 29F.1A12), CD4-BV510 (Clone RM4-5), tion implies synergism or antagonism. In the presence of a sig- CD44-PE (Clone IM7), NK1.1-APCCy7 (Clone PK136), CD3- nificant main effect or interaction, between-group differences Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 388 A. Nienaber et al. were examined using the Bonferroni correction for multiple of lung tissue and immune cells may exert local immune- and comparisons. inflammation-modulatory effects(11,24). There were antagonistic pre-infection status × n-3þ interactions for DHA, total n-3 LCPUFA, osbond acid, total n-6 LCPUFA and n-6/ n-3 Results LCPUFA ratios in erythrocytes, PBMC and lung homogenates (P< 0·001 for all) and AA in erythrocytes and PBMC Body weight gain and food intake (P< 0·001 and P= 0·001) (Table 3). n-3 LCPUFA supplementa- There were no significant differences in the pre-infection tion resulted in higher phospholipid EPA, DHA and total n-3 weight (33 (SE 0·47) g) and daily food intake per mouse LCPUFA (P< 0·001 for all), whilst there was an effect of a low (3·30 (SE 0·25) g). There was a trend towards a main effect of n-3 PUFA pre-infection status for lower EPA, DHA and total n-3 LCPUFA supplementation for a higher percentage weight n-3 LCPUFA in erythrocytes, PBMC and lung homogenates gain (n-3FAS, 6·65 (SE 0·57) %; n-3FAS/n-3þ, 8·11 (SE 0·89) %; (P< 0·001 for all, except for EPA in lung homogenates P= 0·82). n-3FAD, 3·23 (SE 1·67) %; n-3FAD/n-3þ, 6·98 (SE 0·60) %, With regard to n-6 PUFA, n-3 LCPUFA supplementation low- P= 0·07). The mice in the n-3 LCPUFA supplemented groups ered AA, osbond acid, total n-6 LCPUFA and total n-6/n-3 (n-3FAS/n-3þ and n-3FAD/n-3þ) consumed approximately LCPUFA ratios in erythrocytes, PBMC and crude lung homoge- 1·98 mgDHA and 2·94mg EPA daily or 1 % of total energy intake nates (P< 0·001 for all). In contrast, there was an effect of a when calculated on average daily food consumption. low n-3 PUFA pre-infection status for higher AA, osbond acid, total n-6 LCPUFA and n-6/n-3 LCPUFA ratios (P< 0·001 for The total phospholipid fatty acid composition of all, except for AA in lung homogenates P= 0·27). Respective erythrocytes, peripheral blood mononuclear cells differences between groups are shown in Table 3. and crude lung homogenates Bacterial load and lung pathology Table 2 presents the phospholipid FA composition of erythro- cytes following the 6-week dietary conditioning period on either Fig. 2 shows the lung bacterial loads, percentage of free alveolar n-3FAS or n-3FAD diets. Erythrocyte FA composition has been space and lung histology images. There was an antagonistic pre- reported to be representative of the FA content of other infection status × n-3þ interaction on lung bacterial load tissues(23). Following the conditioning period, the n-3FAD group (P= 0·006, Fig. 2(a)). Within the n-3 PUFA-sufficient arm, the had lower EPA, DHA and total n-3 LCPUFA, and higher AA, n-3FAS/n-3þ group had a lower lung bacterial load when com- osbond acid and total n-6 LCPUFA compositions, as well as a pared with the n-3FAS group (P= 0·003). However, this lower- higher total n-6/n-3 LCPUFA ratio, in comparison with the ing effect was attenuated by a low n-3 PUFA status (in the n-3FAS group (P< 0·001 for all). There was no significant n-3FAD/n-3þ group). The n-3FAD group had a lower bacterial difference between the n-3FAS and n-3FAD groups in terms load compared with the n-3FAS group (P= 0·037). The quanti- of erythrocyte saturated fatty acid composition following the fication of the percentage of free alveolar space revealed no conditioning period of 6 weeks (n-3FAS, 34·97 ( 2·71); significant main effects for neither n-3 PUFA pre-infection statusSE n-3FAD, 34·62 ( 2·53)). nor n-3 LCPUFA supplementation (Fig. 2(b) and (c)).SE The phospholipid FA composition of erythrocytes, PBMC and crude lung homogenates of Mtb-infected mice after 3 weeks of Immune cell phenotyping dietary intervention is presented in Table 3. In addition to We also compared lung immune cell phenotypes from a single- recruited immune cells, lung epithelium also synthesises lipid cell suspension of the lungs as determined by flow cytometry, mediators, and therefore, themodification of the FA composition presented as percentages of total cells (Fig. 3). We found antago- nistic pre-infection status × n-3þ interactions in interstitial and CD11bDC percentages (P= 0·045 and 0·014) and trends Table 2 towards interactions for T cells, CD4 þ T cells and natural killer Phospholipid fatty acid composition of erythrocytes in mice receiving n-3FAS or n-3FAD diets for 6 weeks* cells (P= 0·08, 0·06 and 0·05, Fig. 3(a)–(e)). n-3 LCPUFA supple- (Percentages and standard errors) mentation resulted in a reduced percentage of T cells, CD4þ T cells and natural killer cells (P= 0·009, 0·026 and 0·005, n-3FAS n-3FAD Fig. 3(a)–(c)), with the percentage T cells (P= 0·019, Fig. 3(a)) Fatty acids % of FA SE % of FA SE P and CD4þ T cells (P= 0·014, Fig. 3(b)) lower in the n-3FAS/ 20:5n-3 (EPA) 0·20 0·01 0·04 0·01 < 0·001 n-3þ group when compared with the n-3FAS group. On the 22:6n-3 (DHA) 7·84 0·26 3·92 0·22 < 0·001 other hand, the n-3FAD group presented with a higher percent- Total n-3 LCPUFA 8·70 0·20 4·12 0·22 < 0·001 age of natural killer cells (n-3FAS v. n-3FAD: P= 0·017; n-3FAS/ 20:4n-6 (AA) 17·95 0·38 19·80 0·40 < 0·001 22:5n-6 (Osbond) 1·11 0·05 4·06 0·33 < 0·001 n-3þ v. n-3FAD: P= 0·004; n-3FAD v. n-3FAD/n-3þ: P= 0·010, Total n-6 LCPUFA 22·86 0·28 28·60 0·48 < 0·001 Fig. 3(c)) compared with other groups, whilst interstitial macro- n-6/n-3 LCPUFA 2·63 0·04 7·04 0·38 < 0·001 phages (n-3FAS v. n-3FAD: P< 0·001, n-3FAS/n-3þ v. n-3FAD: AA, arachidonic acid; LCPUFA, long-chain PUFA; n-3FAD, n-3 fatty acid-deficient diet; P= 0·001) and CD11bDC percentages (n-3FAS v. n-3FAD: n-3FAS, n-3 fatty acid-sufficient diet. P= 0·002; n-3FAS/n-3þ v. n-3FAD: P= 0·014) were higher in * Values are reported as means and standard errors of the means percentage of total fatty acids. Intervention effects were estimated using the independent Student the n-3FAD than in n-3FAS and n-3FAS/n-3þ groups Fischer t test (n 6 per group). (Fig. 3(d) and (e)). The aforementioned effects induced by a Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 389 Table 3. Phospholipid fatty acid composition of erythrocytes, PBMC and crude lung homogenates in Mtb-infected mice receiving n-3FAS, n-3FAS/n-3þ, n-3FAD or n-3FAD/n-3þ diets for 3 weeks* (Mean values with their standard errors) n-3FAS n-3FAS/n-3þ n-3FAD n-3FAD/n-3þ Pre-infection Pre-infection status × Mean % Mean % Mean % Mean % status main n-3þ main n-3þ interaction FA SE FA SE FA SE FA SE effect effect effect 18:53n-3 (ALA) Erythrocyte 0·05 0·00b 0·05 0·00a 0·00 0·00d 0·04 0·00c < 0·001 < 0·001 < 0·001 PBMC 0·02 0·00a 0·01 0·00b 0·00 0·00d 0·01 0·00c < 0·001 0·97 0·003 Lung 0·09 0·00a 0·01 0·00c 0·01 0·00d 0·07 0·00b < 0·001 0·001 < 0·001 20:5n-3 (EPA) c Erythrocyte 0·13 0·00 0·51 0·04a 0·02 0·00d 0·39 0·02b < 0·001 < 0·001 0·79 PBMC 0·21 0·01c 0·89 0·04a 0·05 0·11d 0·70 0·04b < 0·001 < 0·001 0·68 Lung 0·17 0·01 0·40 0·01 0·23 0·15 0·38 0·01 0·82 0·021 0·59 22:6n-3 (DHA) Erythrocyte 6·09 0·21b 7·48 0·41a 1·72 0·15d 5·41 0·06c < 0·001 < 0·001 < 0·001 PBMC 9·04 0·20b 9·82 0·22a 3·21 0·19c 9·49 0·24a < 0·001 < 0·001 < 0·001 Lung 8·08 0·15b 9·68 0·28a 2·39 0·26c 9·56 0·11a < 0·001 < 0·001 < 0·001 Total n-3 LCPUFA Erythrocyte 6·63 0·20b 8·59 0·48a 1·81 0·15d 6·19 0·05c < 0·001 < 0·001 < 0·001 PBMC 10·60 0·21c 12·48 0·33a 3·48 0·21d 11·59 0·31b < 0·001 < 0·001 < 0·001 Lung 10·03 0·14b 12·62 0·26a 2·91 0·24c 12·33 0·10a < 0·001 < 0·001 < 0·001 20:4n-6 (AA) Erythrocyte 17·71 0·27b 16·42 0·30c 20·45 0·25a 16·35 0·30c < 0·001 < 0·001 < 0·001 PBMC 16·36 0·39b 14·76 0·31c 21·48 0·47a 16·85 0·27b < 0·001 < 0·001 0·001 Lung 14·40 0·15ab 13·32 0·42b 15·19 0·58a 13·30 0·19b 0·28 0·001 0·20 22:5n-6 (Osbond) Erythrocyte 0·87 0·10c 0·37 0·01c 3·37 0·32a 0·84 0·07c < 0·001 < 0·001 < 0·001 PBMC 1·42 0·04b 0·74 0·05c 5·68 0·30a 0·98 0·05c < 0·001 < 0·001 < 0·001 Lung 1·13 0·05b 0·54 0·01c 5·16 0·27a 0·85 0·03bc < 0·001 < 0·001 < 0·001 Total n-6 LCPUFA Erythrocyte 20·17 0·32b 18·66 0·44c 25·47 0·39a 18·68 0·32c < 0·001 < 0·001 < 0·001 PBMC 21·83 0·41b 19·36 0·54c 32·75 0·69a 21·48 0·27b < 0·001 < 0·001 < 0·001 Lung 21·24 0·19b 18·27 0·50c 26·58 0·65a 18·81 0·26c < 0·001 < 0·001 < 0·001 n-6/n-3 LCPUFA Erythrocyte 3·05 0·05b 2·19 0·07b 14·38 1·19a 3·02 0·05b < 0·001 < 0·001 < 0·001 PBMC 2·06 0·02b 1·55 0·01b 9·60 0·82a 1·86 0·05b < 0·001 < 0·001 < 0·001 Lung 2·11 0·03b 1·45 0·06b 9·29 0·65a 1·52 0·02b < 0·001 < 0·001 < 0·001 AA, arachidonic acid; FA, fatty acids; LCPUFA, long-chain PUFA; n-3FAD, n-3 fatty acid-deficient group; n-3FAS, n-3 fatty acid-sufficient group; n-3þ, n-3 long-chain PUFA-supplemented group; PBMC, peripheral blood mononuclear cell. * Values are reported as means and the standard errors of the means percentage of total fatty acids. Results repeated in two experiments, data shown for one experiment (n 5 per group). A two-wayANOVAwas used to test effects of n-3þ (n-3FAS/n-3þ plus n-3FAD/n-3þ v. n-3FADplus n-3FAS), pre-infection status (n-3FAS plus n-3FAS/n-3þ v. n-3FADplus n-3FAD/n-3þ) and pre-infection status × n-3þ interactions. Bonferroni correction for multiple comparisons was used. Means in a row without common superscript letters differ sig- nificantly, P< 0·05. lown-3 PUFA statuswere attenuated in then-3FAD/n-3þ group. IFN-γ, IL-1α, IL-1β and IL-17 compared with the n-3FAS group In addition, neutrophils appeared to remain unaffected by n-3 (P< 0·001, 0·002, 0·009 and 0·006, Fig. 4(a), (c), (d) and (e)). LCPUFA supplementation and pre-infection status in n-3FAS These individual lowering effects of a low pre-infection n-3 and n-3FAD groups (Fig. 3(f)). PUFA status and n-3 LCPUFA supplementation were attenuated in the n-3FAD/n-3þ mice which instead presented with higher Lung cytokines concentrations of lung IL-6 (P= 0·001, Fig. 4(b)) and IL-1α (P= 0·043, Fig. 4(c)) compared with the n-3FAD group. The lung cytokine responses measured in cell-free lung homog- There was also a trend towards a main effect of n-3 LCPUFA enates are presented in Fig. 4. We observed antagonistic pre- × þ supplementation for higher lung IL-10 (P= 0·07, Fig. 4(f)).infection status n-3 interactions in lung IFN-γ, IL-6 and IL-1α (P< 0·001, 0·005 and 0·011) and a trend towards antago- Lung lipid mediators nistic interactions for IL-1β and IL-17 concentrations (P= 0·06 and 0·05) (Fig. 4(a)–(e)). The n-3FAS/n-3þ group had signifi- Fig. 5 presents the less inflammatory and pro-resolving lipid cantly lower lung IFN-γ (P< 0·001, Fig. 4(a)) and tended to have mediators of crude lung homogenates. There were pre-infection lower IL-1α (P= 0·07, Fig. 4(c)) compared with the n-3FAS status × n-3þ interactions for PGE3 and 5-HEPE (P= 0·049 and group. A low n-3 PUFA status had an effect for lower lung 0·027), where a combination of a low n-3 PUFA status (n-3FAD) IL-1β and IL-17 concentrations (P= 0·044 and 0·026, Fig. 4(d) and n-3 LCPUFA supplementation (n-3þ) resulted in higher and (e)). The n-3FAD group presented with lower levels of PGE3 and 5-HEPE concentrations (P< 0·001 and 0·003, Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 390 A. Nienaber et al. (a) Pre-infection status: P = 0·98 Pre-infection status: P = 0·57n-3+: P = 0·08 (b) n-3+: P = 0·79 Pre-infection status x n-3+: P = 0·006 Pre-infection status x n-3+: P = 0·37 6.9 50 6.8 40 6.7 30 6.6 20 6.5 . 106 4 6.3 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ (c) n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Fig. 2. (a) Lung bacterial loads, (b) percentage free alveolar space and (c) representative haematoxylin–eosin stained sections of the lungs after providing Mtb-infected mice with n-3FAS, n-3FAS/n-3þ, n-3FAD and n-3FAD/n-3þ diets for 3 weeks (scale bar= 1000 μm). The values represent means and standard errors of the means. Results repeated in two experiments, data shown for one experiment (n 5 per group). A two-way ANOVA was used to test effects of n-3þ (n-3FAS/n-3þ plus n-3FAD/n-3þ v. n-3FAD plus n-3FAS), pre-infection status (n-3FAS plus n-3FAS/n-3þ v. n-3FAD plus n-3FAD/n-3þ) and pre-infection status × n-3þ interactions. Bonferroni correction for multiple comparisons was used, *P< 0·05, **P< 0·01. CFU, colony-forming units; n-3FAD, n-3 fatty acid-deficient diet; n-3FAS, n-3 fatty acid sufficient diet; n-3þ, n-3 long-chain PUFA-supplemented diet; /, switched to. Fig. 5(a) and (b)). There were also trends towards pre-infection Discussion status × n-3þ interactions on 9-HEPE and 17-HDHA (P= 0·08 The present study provides evidence that n-3 LCPUFA supple- and 0·07, Fig. 5(c) and (e)). n-3 LCPUFA resulted in higher con- mentation, commenced 1week post-infection, reduced bacterial centrations of the less inflammatory EPA-derived PGE3, as well burden, altered the local lung immune response and assisted in as the pro-resolving EPA-derived intermediates 5-, 9-, 11-, 12-, weight gain in a C3HeB/FeJ mouse model of TB. Importantly, 15-, 18-HEPE and the DHA-derived 17-HDHA (P< 0·001 for these findings applied only to mice conditioned to have an n- all except 9-HEPE, P= 0·002, Fig. 5(a)–(f), results not shown 3 PUFA-sufficient status before infection, whereas the low n-3 for 12- and 15-HEPE). On the other hand, a low pre-infection PUFA status mice also showed a lower bacterial load compared status (n-3FAD) had a significant effect towards lowering with the sufficient n-3 PUFA status group and did not benefit 9-HEPE and 18-HEPE (P< 0·001 and 0·005) and also reduced from n-3 LCPUFA supplementation. 11-HEPE (P= 0·06) (Fig. 5c–e). The other respective between- The finding that n-3 LCPUFA supplementation lowered group differences are shown in Fig. 5. bacterial burden in n-3 PUFA sufficient mice is similar to that With regard to the more pro-inflammatory AA-derived lipid published by Jordao et al., who found lower bacterial loads in mediators, there were pre-infection status × n-3þ interactions the lungs and spleens of BALB/c Mtb-infected mice fed n-3 for PGD2, PGF2α, 9-, 11- and 15-HETE (P= 0·001, P= 0·008, PUFA-rich diets, compared with mice that were fed a fat-free P< 0·001, P= 0·012 and P= 0·034) and trends towards inter- diet(25). The incorporation of n-3 LCPUFA into phagocytic cell actions on PGE2 and 8-HETE (P= 0·09 and 0·08, Fig. 6(a)–(d), membranes changes membrane fluidity, in addition to recep- data not shown for PGF2α, 9- and 15-HETE). The n-3FAD group tor expression, thereby enhancing bacterial phagocytosis, had higher PGE2, PGD2 and 11-HETE comparedwith the n-3FAS P= · · · which has also been shown in TB (26,27). This is confirmed by group ( 0 010, 0 013 and 0 002, Fig. 6(a), (b) and (d)). The n-3FAD/n-3þ group had lower PGF2 , PGD , 8-, 9- and the higher n-3 LCPUFA composition found in crude lungα 2 11-HETE compared with the n-3FAD group (P= 0·002, homogenates and PBMC in our study, and subsequently, P< 0·001, P= 0·011, P= 0·001 and P= 0·043, Fig. 6(a) (d)). higher EPA incorporation would be expected in the macro-– However, n-3 LCPUFA supplementation did not significantly phage and neutrophil phospholipid bilayers as well. This lower pro-inflammatory lipid mediators in the n-3FAS/ may partly explain the lower lung bacterial loads of the n-3þ group, with only a trend towards lower 9-HETE in the n-3FAS/n-3þ group. Additionally, the changes in FA composi- n-3FAS/n-3þ compared with the n-3FAS group (P= 0·08). tion resulted in a more pro-resolving lipid mediator profile. Lung LOG10 CFU Free alveolar space (%) Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 391 (a) Pre-infection status: P = 0∙39 (b) Pre-infection status: P = 0∙49 30 n-3+: P = 0∙009 20 n-3+: P = 0∙026 Pre-infection status x n-3+: P = 0∙08 Pre-infection status x n-3+: P = 0∙06 * 15 20 * 10 10 5 0 0 (c) Pre-infection status: P = 0∙011 (d) Pre-infection status: P = 0∙001 n-3+: P = 0∙005 n-3+: P = 0∙13 20 Pre-infection status x n-3+: P = 0∙05 8 Pre-infection status x n-3+: P = 0∙045 * *** 15 ** * 6 ** 10 4 5 2 0 0 Pre-infection status: P = 0∙002 Pre-infection status: P = 0∙49 n-3+: P = 0∙23 n-3+: P = 0∙35 (e) Pre-infection status x n-3+: P = 0∙014 (f) Pre-infection status x n-3+: P = 0∙31 1.2 ** 20 1. * 0 15 0.8 0.6 10 0.4 5 0.2 0.0 0 Fig. 3. Immune cell phenotyping of (a) T cells (CD3þ CD19-), (b) CD4þ T cells (CD3þ CD4þ), (c) natural killer cells (CD3- NK1·1þ), (d) interstitial macrophages (CD64þ CD11bþCD11c- SiglecF-), (e) CD11bþ dendritic cells (CD11bþCD11cþMHCIIþCD64-) and (f) neutrophils (CD11bþ Ly6Gþ), in crude lung homogenates after providing Mtb-infected mice with n-3FAS, n-3FAS/n-3þ, n-3FAD or n-3FAD/n-3þ diets for 3 weeks. The values represent means and standard errors of themeans% of total cells. Results repeated in two experiments, data shown for one experiment (n 5 per group). A two-way ANOVA was used to test effects of n-3þ (n-3FAS/n-3þ plus n-3FAD/n-3þ v. n-3FAD plus n-3FAS), pre-infection status (n-3FAS plus n-3FAS/n-3þ v. n-3FAD plus n-3FAD/n-3þ) and pre-infection status × n-3þ interactions. Bonferroni correction for multiple comparisons was used, *P< 0·05, **P< 0·01, ***P< 0·001. DC, dendritic cells; Macs, macrophages; NK, natural killer; n-3þ, n-3 long-chain PUFA-supplemented diet; n-3FAD, n-3 fatty acid-deficient diet; n-3FAS, n-3 fatty acid-sufficient diet; /, switched to. , n-3FAS; , n-3FAS/n-3þ; , n-3FAD; , n-3FAD/n-3þ. The n-3FAS/n-3þ group presented with higher lung concen- supplementation. Previous experiments were focused on the trations of the pro-resolving 18-HEPE, which is an intermediate conditioning of the animals with n-3 LCPUFA before infection of the E-series resolvins (SPM) synthesised from EPA(28,29). or upon infection(27,31–34). However, the timing of immunonutri- Since SPM aid in the differentiation and activation of tion in any HDT approach for TB is critical and an early strong macrophages and neutrophils for phagocytosis and bacterial inflammatory response is essential(4). In the present study, we killing(12,13,30), this may further explain the bactericidal effects aimed to provide n-3 LCPUFA supplementation as therapy after of n-3 LCPUFA supplementation observed in the present study. the initial acute inflammatory response, by initiating the dietary Our findings are different from those previously published, intervention which showed that n-3 LCPUFA inhibits immune responses 1 week post-infection. Early ingestion of n-3 LCPUFA, or and worsen TB outcomes(27,31–34). We hypothesise that the main upon infection initiation, has been shown to inhibit phagosome reason for these discrepancies may be the timing of and phagolysosome maturation, which causes higher initial CD11b DCs (%Total cells) NK cells (%Total cells) T cells (%Total cells) Neutrophils (%Total cells) Interstitial Macs (%Total cells) CD4 T cells (%Total cells) Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 - 392 A. Nienaber et al. Pre-infection status: P = 0·001 Pre-infection status: P = 0·59 n-3+: P = 0·017 n-3+: P = 0·036 γ Pre-infection status x n-3+: P < 0·001 Pre-infection status x n-3+: P = 0·005(a) IFN- (b) IL-6 1500 2000 1500 1000 1000 500 500 0 0 -3FAS -3FAS/ -3+ -3FAD -3FAD/ -3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+n n n n n n Pre-infection status: P = 0·035 Pre-infection status: P = 0·044 n-3+: P = 0·84 n-3+: P = 0·39 (c) IL-1α Pre-infection status x n-3+: P = 0·011 Pre-infection status x n-3+: P = 0·06(d) 3000 800 IL-1β 600 2000 400 1000 200 0 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Pre-infection status: P = 0·026 Pre-infection status: P = 0·08 (e) n-3+: P = 0·45 n-3+: P = 0·07 IL-17 Pre-infection status x n-3+: P = 0·05 (f) Pre-infection status x n-3+: P = 0·31 100 IL -10 80 25 20 60 15 40 10 20 5 0 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Fig. 4. Cytokine concentrations, including (a) IFN-γ, b) IL-6, c) IL-1α, d) IL-1β, e) IL-17 and f) IL-10 in crude lung homogenates after providing Mtb-infected mice with n-3FAS, n-3FAS/n-3þ, n-3FAD or n-3FAD/n-3þ diets for 3 weeks. The values represent means and standard errors of the means (pg/mL). Results repeated in two experiments, data shown for one experiment (n 5 per group). A two-way ANOVA was used to test effects of n-3þ (n-3FAS/n-3þ plus n-3FAD/n-3þ v. n-3FAD plus n-3FAS), pre-infection status (n-3FAS plus n-3FAS/n-3þ v. n-3FAD plus n-3FAD/n-3þ) and pre-infection status × n-3þ interactions. Bonferroni correction for multiple comparisons was used, *P< 0·05, **P< 0·01, ***P< 0·001. IFN-γ, interferon-γ; n-3þ, n-3 long-chain PUFA-supplemented diet; n-3FAD, n-3 fatty acid-deficient diet; n-3FAS, n-3 fatty acid-sufficient diet; /, switched to. pg/mL pg/mL pg/mL pg/mL pg/mL pg/mL Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 393 Pre-infection status: P = 0·82 Pre-infection status: P = 0·76 -3+: < 0·001 Pre-infection status: P < 0·001 n-3+: P < 0·001 n P n-3+: P = 0·002 Pre-infection status x n-3+: P = 0·049 Pre-infection status x n-3+: P = 0·027 PGE 5-HEPE Pre-infection status x n-3+: P = 0·083 9-HEPE (a) (b) 20 (c) 250 60 200 15 40 150 10 100 20 5 50 0 0 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Pre-infection status: P = 0·06 Pre-infection status: P = 0·005 Pre-infection status: P = 0·10 n-3+: P < 0·001 n-3+: P < 0·001 n-3+: P < 0·001 Pre-infection status x n-3+: P = 0·83 11-HEPE Pre-infection status x n-3+: P = 0·94 Pre-infection status x n-3+: P = 0·07 18-HEPE 17-HDHA (d) (e) (f) 200 50 400 40 150 300 30 100 200 20 50 10010 0 0 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Fig. 5. Pro-resolving lipid mediator concentrations of (a) PGE3, (b) 5-HEPE, (c) 9-HEPE, (d) 11-HEPE, (e) 18-HEPE and (f) 17-HDHA in crude lung homogenates after providing Mtb-infected mice with n-3FAS, n-3FAS/n-3þ, n-3FAD or n-3FAD/n-3þ diets for 3 weeks. The values represent the means. Results repeated in two experiments, data shown for one experiment (n 5 per group). A two-way ANOVA was used to test effects of n-3þ (n-3FAS/n-3þ plus n-3FAD/n-3þ v. n-3FAD plus n-3FAS), pre-infection status (n-3FAS plus n-3FAS/n-3þ v. n-3FAD plus n-3FAD/n-3þ) and pre-infection status × n-3þ interactions. Bonferroni correction for multiple comparisons was used, *P< 0·05, **P< 0·01, ***P< 0·001. HDHA, hydroxydocosahexaenoic acid; HEPE, hydroxyeicosapentaenoic acid; n-3þ, n-3 long-chain PUFA-supplemented diet; n-3FAD, n-3 fatty acid-deficient diet; n-3FAS, n-3 fatty acid-sufficient diet; /, switched to. bacterial loads(27,35). Therefore, the timely initiation of n-3 and lipid mediator synthesis(29). These changes, together with LCPUFA supplementation was an important contributor to pos- the lower bacterial burden in this group, may explain the lower itive outcomes. T cell percentages in the n-3FAS/n-3þ mice. Furthermore, the dietary composition provided in previous Concerning lung cytokines, IFN-γ is important in the studies differed from that which we used. Whilst the EPA/ protection against TB; however, higher concentrations have DHA ratio in the n-3þ diet groups was comparable to that of been correlated with cavitary TB, higher bacterial loads and Jordao et al., who also found antibacterial effects of n-3 delayed culture conversion(2,39). We found that IFN-γ concentra- LCPUFA supplementation in TB, other studies that found nega- tions were lower in the n-3FAS/n-3þ group, which is consistent tive effects provided either higher DHA concentrations or DHA with the findings of others in TB(32). Similarly, n-3 LCPUFA sup- only(25,27,32,33,36). Previous studies also used in vitro cell culture plementation reduced lung IL-6 and IL-1α tended to be lowered. models(27) or endogenously enriched mice (fat-1 mice)(31), This complements our findings on T cell numbers mentioned and differences in the genetic backgrounds of the mice may also above and confirms previous findings(40). As expected, there have contributed. was also a trend towardsn-3 LCPUFA supplementation elevating As lung inflammation is central in lesion formation, granu- the concentrations of the anti-inflammatory IL-10, therefore loma liquefaction, cavity formation and clinical outcomes, we promoting inflammation resolution(12). hypothesised that the resolution of inflammation would also Supplementation of n-3 LCPUFA was successfully confirmed improve lung pathology(2,37). However, confirming previous by elevated cell membrane compositions and a pro-resolving evidence, no effect of n-3 LCPUFA supplementation could be lung lipid mediator profile of the n-3 PUFA sufficient status found in terms of percentage of free alveolar space in the present arm of the study. This translated into the lowering of some study(32). On the other hand, n-3 LCPUFA supplementation has pro-inflammatory lung cytokines and lipid mediators, but not previously been found to inhibit T cell proliferation, elsewhere in all markers. A similar result to ours was found in a rat model and in TB, specifically(32,38). Consistent with this, we also found a injected with Salmonella enteritidis endotoxin, where the lower percentage of lung T cells and CD4þ T cells in the n-3FAS/ administration of fish oil altered pro-resolving lipid mediators n-3þ group, which may have been driven by the effects of n-3 without significantly changing the cytokine concentrations in LCPUFA supplementation causing structural changes to cell bronchoalveolar lavage fluid(41). The fact that n-3 LCPUFA membranes, producing subsequent alterations in cell signalling have been reported to affect the Th1/Th2 balance mainly by pg/l pg/l pg/l pg/l pg/l pg/l Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 394 A. Nienaber et al. Pre-infection status: P = 0·010 Pre-infection status: P = 0·33 PGE n-3+: P = 0·43 n-3+: P = 0·021 Pre-infection status x n-3+: P = 0·09 PGD2 Pre-infection status x n-3+: P = 0·001 (a) 2 (b) 15000 200 150 10000 100 5000 50 0 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Pre-infection status: P = 0·80 Pre-infection status: P = 0·10 8-HETE n-3+: P = 0·043 n-3+: P = 0·78 (c) Pre-infection status x n-3+: P = 0·08 (d) 11-HETE Pre-infection status x n-3+: P = 0·012 500 5000 400 4000 300 3000 200 2000 100 1000 0 0 n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ n-3FAS n-3FAS/n-3+ n-3FAD n-3FAD/n-3+ Fig. 6. Pro-inflammatory lipid mediator concentrations of (a) 8-HETE, (b) PGD2, (c) PGE2 and (d) 11-HETE in crude lung homogenates after providingMtb-infectedmice with n-3FAS, n-3FAS/n-3þ, n-3FAD or n-3FAD/n-3þ diets for 3 weeks. The values represent the means. Results repeated in two experiments, data shown for one experiment (n 5 per group). A two-way ANOVA was used to test effects of n-3þ (n-3FAS/n-3þ plus n-3FAD/n-3þ v. n-3FAD plus n-3FAS), pre-infection status (n-3FAS plus n-3FAS/n-3þ v. n-3FAD plus n-3FAD/n-3þ) and pre-infection status × n-3þ interactions. Bonferroni correction for multiple comparisons was used, inhibiting the production of Th1 type cytokines (including IFN-γ) the n-3 PUFA sufficient group, supplemented with n-3 may serve as an explanation for the current findings(42). LCPUFA. Bonilla et al. also reported that n-3 PUFA-deficient Furthermore, Kroesen and colleagues found amore pronounced mice had a lower susceptibility to TB when compared with effect on systemic (serum) cytokine concentrations as compared fat-1 transgenic mice, with an endogenous abundance of n-3 with lung cytokines when administering aspirin in the same PUFA(31). This may indicate that n-3 PUFA deficiency is protec- animal TB model as in our study(4). In contrast with our results, tive against TB. Nevertheless, the clinical relevance of these find- previous studies on n-3 LCPUFA treatment in Mtb-infected ings for humans is questionable. It would be unrealistic to animals, macrophages and peritoneal cells showed reduced promote low n-3 PUFA consumption in TB infection as a PGE2, leukotriene B4, TNF-α, IL-6, IL-1β and monocyte chemo- protective measure, considering the other important biological tactic protein-1 synthesis(25,27,31,33). Nevertheless, irrespective of functions that n-3 PUFA would have in these individuals. the fact that some of the pro-inflammatory lipid mediators and However, considering that there may be populations with cytokines were not significantly altered in the n-3FAS/n-3þ a low n-3 PUFA status at risk for TB, the interaction between group, the higher pro-resolving lipid mediator concentrations a low n-3 PUFA status, TB medication and treatment outcomes were a positive finding, demonstrating the pro-resolving proper- require further investigation, before continuing human trials. ties of n-3 LCPUFA. Therefore, our results suggest that n-3 As expected, the lipid mediator profile of the low n-3 PUFA LCPUFA supplementation does not inhibit the host’s natural status group was in congruence with their FA status. A low n-3 immune and inflammatory responses necessary to protect PUFA status promoted lower concentrations of n-3 PUFA- and against bacteria. This supports the notion that SPM are not higher n-6 PUFA-derived lung lipid mediators. However, the immunosuppressive and do not block inflammation, but instead n-3FAD group presented with lower lung concentrations of elicit pro-resolving effects(12). IFN-γ, IL-1α, IL-1β and IL-17 compared with the n-3FAS group, On the other hand, the low n-3 PUFA status mice also which is conflicting with the FA status results and the less presented with lower bacterial loads, similar to that seen in pro-resolving lipid mediator profiles found in these mice. The pg/l pg/l pg/l pg/l Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 395 reasons why the low n-3 PUFA status mice presented with lower group, specifically, the dose of n-3 LCPUFA supplementation levels of some of the inflammatory cytokines may be related to may have been too low and/or the duration too short. Future the timing of the cytokine measurement. An initially higher prospects would be to perform the present studywith euthanasia inflammatory response due to a higher n-6 PUFA status and time points at the different phases of the inflammatory and pro-inflammatory lipid mediator profile may have resulted in immune response, also including systemic markers of inflamma- lower cytokine concentrations by the time assessed (4 weeks tion. Additionally, the possible beneficial effects of n-3 LCPUFA, after infection). Another plausible explanation is that the lower when administered in combination with standard TB treatment, bacterial loads of these mice likely provoked a lower inflamma- are yet to be determined. tory response. Seemingly, in contrast, the low n-3 PUFA status in our study promoted higher percentages of certain immune cells, Conclusions including the natural killer cells, interstitial macrophages and dendritic cells which were higher in the n-3FAD group com- In conclusion, the present study showed that n-3 LCPUFA pared with the n-3FAS group. This could have contributed to supplementation, administered after the initial inflammatory bacterial control of the n-3 PUFA low-status group via cell-intrin- response in Mtb-infected mice, lowered the bacterial burden sic killing functions independent of cytokine levels. The higher in n-3 PUFA-sufficient mice, but not in mice with a low n-3 percentages of dendritic cells and macrophages can be PUFA status. It further promoted a more pro-resolving explained by the fact that PGE concentrations were higher in lipid mediator profile, lower production of inflammatory cyto-2 the n-3FAD group, which have been implicated to induce kines and tended to enhance weight gain. Considering this, human DC and mice macrophage recruitment, whilst in a peri- n-3 LCPUFA supplementation in the context of a sufficient n-3 tonitis mouse model COX-2 deficient mice presented with PUFA status may be a promising approach as an HDT in TB. reduced macrophage recruitment(43–45). The present study emphasises, however, that the timing, n-3 LCPUFA supplementation of the low n-3 PUFA status the EPA/DHA ratio administered and n-3 PUFA status before group (n-3FAD/n-3þ) did not have the same beneficial effects supplementation are critical considerations. It further shows that as in the n-3FAS/n-3þ group. Our results show that both a a low n-3 PUFA status before TB infection may be protective, low n-3 PUFA status and n-3 LCPUFA supplementation had low- which requires further investigation. ering effects on pro-inflammatory lung cytokines, but combining a low status, and supplementation attenuated these lowering effects. This was despite the successful alteration of the n-3 Acknowledgements LCPUFA cell membrane composition and lipid mediators The authors thank Rodney Lucas (UCT, Cape Town, SA) and towards a more pro-resolving lung profile in the n-3FAD/ þ Kobus Venter (North-West University, SA) for their technicaln-3 group. Moreover, n-3 LCPUFA supplementation in the assistance with animals and Adriaan Jacobs, Cecile Cooke and low n-3 PUFA status mice (n-3FAD/n-3þ) led to a more pro- Marike Cockeran (North-West University, SA) for their assistance nounced increase in PGE3 and 5-HEPE than supplementation with laboratory and statistical analyses. in n-3 PUFA sufficient mice. Also, different from the n-3FAS/ þ This research was supported by the South African Medicaln-3 group, the n-3FAD/n-3þ group showed significantly Research Council under a Self-Initiated Research Grant (L.M., lower lung concentrations of the pro-inflammatory lipid media- MRC-SIR) and by the Nutricia Research Foundation (A.N.), but tors PGF2α, PGD2, 8-, 9- and 11-HETE. Still, n-3 LPCUFA supple- the views and opinions expressed are those of the authors mentation in n-3FAD mice resulted in higher lung IL-6 and IL-1α and not of the SAMRC or the Nutricia Research Foundation. concentrations. Possible reasons why n-3 LCPUFA supplemen- The research conducted at the UCT was supported by core tation did not exert the same beneficial effects in the n-3FAD/ þ funding from the Wellcome Trust (203135/Z/16/Z).n-3 group may be related, firstly, to the dosage and duration A. N., L. M., S. P. P., R.D., C.M.S., R.B. and D.L. contributed to of supplementation and secondly, to possible epigenetic adap- the study conception and design. Material preparation, data col- tation to deficiency. As discussed previously, the n-3FAD group lection and analysis were performed by A. N., L. M., S. K., F. E. A. itself also presented with low lung cytokine concentrations and H., M. B., M. O. and S. P. P. The first draft of the manuscript was possible clinical benefit to start with, which may be the reason written by A. N. and all authors commented on previous versions why n-3 LCPUFA supplementation in this group did not improve of the manuscript. All authors read and approved the final cytokine concentrations or bacterial load. Nevertheless, with this manuscript. in mind, it cannot be said with certainty that a low n-3 PUFA sta- The authors declare that they have no conflict of interest. tus improves TB outcomes due to the inconsistent immune and inflammatory findings of this group, or that n-3 LCPUFA should not be supplemented under conditions of a low n-3 PUFA status. References Further investigation into these findings is warranted. One of the strengths of the present study was that we used a 1. Meghji J, Simpson H, Squire SB, et al. (2016) A systematic murine model that is well-established and reflective of human review of the prevalence and pattern of imaging defined post-TB lung disease. PLoS One 11, e0161176. pulmonary TB. Furthermore, our experimental design, including 2. Kumar NP, Moideen K, Banurekha VV, et al. (2019) Plasma the timing of supplementation, comparison of n-3 PUFA pro-inflammatory cytokines are markers of disease severity sufficiency and low status and the EPA/DHA ratio of our supple- and bacterial burden in pulmonary tuberculosis. Open Forum ment, also strengthens our findings. However, in the n-3FAD Infect 6, ofz257. Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 396 A. Nienaber et al. 3. Stek C, Allwood B, Walker NF, et al. (2018) The immune mech- 22. Malan L, Baumgartner J, Zandberg L, et al. (2016) Iron and anisms of lung parenchymal damage in tuberculosis and the a mixture of DHA and EPA supplementation, alone and in role of host-directed therapy. Front Micro 9, 2603. combination, affect bioactive lipid signalling and morbidity of 4. Kroesen VM, Rodríguez-Martínez P, García E, et al. (2018) iron deficient South African school children in a two-by-two A beneficial effect of low-dose aspirin in a murine model of randomised controlled trial. Prostaglandins Leukot Essent active tuberculosis. Front Immunol 9, 798. Fatty Acids 105, 15–25. 5. Critchley JA, Young F, Orton L, et al. (2013) Corticosteroids 23. Brenna JT, Plourde M, Stark KD, et al. (2018) Best practices for for prevention of mortality in people with tuberculosis: a the design, laboratory analysis, and reporting of trials involving systematic review and meta-analysis. Lancet Infect Dis 13, fatty acids. Am J Clin Nutr 108, 211–227. 223–237. 24. Sanak M (2016) Eicosanoid mediators in the airway inflamma- 6. Marzo E, Vilaplana C, Tapia G, et al. (2014) Damaging role tion of asthmatic patients: what is new? Allergy Asthma of neutrophilic infiltration in a mouse model of progressive Immunol Res 8, 481–490. tuberculosis. Tuberculosis 94, 55–64. 25. Jordao L, Lengeling A, Bordat Y, et al. (2008) Effects of n-3 7. Kroesen VM, Gröschel MI, Martinson N, et al. (2017) Non- and-6 fatty acids on Mycobacterium tuberculosis in macro- Steroidal anti-inflammatory drugs as host-directed therapy for phages and in mice. Microbe Infect 10, 1379–1386. tuberculosis: a systematic review. Front Immunol 8, 772. 26. Calder PC, Bond JA, Harvey DJ, et al. (1990) Uptake and 8. Ivanyi J & Zumla A (2013) Nonsteroidal anti-inflammatory incorporation of saturated and unsaturated fatty acids into drugs for adjunctive tuberculosis treatment. J Infect Dis 208, macrophage lipids and their effect upon macrophage adhesion 185–188. and phagocytosis. Biochem J 269, 807–814. 9. Calder PC (2017) n-3 fatty acids and inflammatory 27. Bonilla DL, Ly LH, Fan Y-Y, et al. (2010a) Incorporation of a processes: from molecules to man. Biochem Soc Trans 45, dietary n 3 fatty acid impairs murine macrophage responses 1105–1115. to mycobacterium. PLoS One 5, e10878. 10. Browning LM, Walker CG, Mander AP, et al. (2012) 28. Oh SF, Pillai PS, Recchiuti A, et al. (2011) Pro-resolving Incorporation of eicosapentaenoic and docosahexaenoic acids actions and stereoselective biosynthesis of 18S E-series into lipid pools when given as supplements providing doses resolvins in human leukocytes and murine inflammation. equivalent to typical intakes of oily fish. Am J Clin Nutr 96, J Clin Invest 121, 569–581. 748–758. 29. Calder PC (2015b) Marine n-3 fatty acids and inflammatory 11. Jakiela B, Gielicz A, Plutecka H, et al. (2013) Eicosanoid processes: effects, mechanisms and clinical relevance. biosynthesis during mucociliary and mucous metaplastic Biochim Biophys Acta Mol Cell Biol Lipids 1851, 469–484. differentiation of bronchial epithelial cells. Prostaglandins 30. CodagnoneM, Cianci E, Lamolinara A, et al. (2018) Resolvin D1 Other Lipid Mediat 106, 116–123. enhances the resolution of lung inflammation caused by 12. Serhan CN, Chiang N & Dalli J (2017) New pro-resolving n-3 long-term Pseudomonas aeruginosa infection. Muc Immunol mediators bridge resolution of infectious inflammation to tissue 11, 35. regeneration. Mol Aspects Med 64, 1–17. 31. BonillaDL, FanY-Y, Chapkin RS, et al. (2010b) Transgenicmice 13. Chiang N, Fredman G, Bäckhed F, et al. (2012) Infection enriched in n-3 fatty acids are more susceptible to pulmonary regulates pro-resolving mediators that lower antibiotic tuberculosis: impaired resistance to tuberculosis in fat-1 mice. requirements. Nature 484, 524. J Infect Dis 201, 399–408. 14. Lenaerts A, Barry CE & Dartois V (2015) Heterogeneity in 32. McFarland CT, Fan Y-Y, Chapkin RS, et al. (2008) Dietary poly- tuberculosis pathology, microenvironments and therapeutic unsaturated fatty acids modulate resistance to Mycobacterium responses. Immunol Rev 264, 288–307. tuberculosis in guinea pigs. J Nutr 138, 2123–2128. 15. Stark KD, Van Elswyk ME, Higgins MR, et al. (2016) Global 33. Paul KP, Leichsenring M, Pfisterer M, et al. (1997) survey of the n-3 fatty acids, docosahexaenoic acid and Influence of n-6 and n-3 polyunsaturated fatty acids on eicosapentaenoic acid in the blood stream of healthy adults. the resistance to experimental tuberculosis. Metabolism 46, Prog Lip Res 63, 132–152. 619–624. 16. Baker EJ, Miles EA, Burdge GC, et al. (2016) Metabolism and 34. Mayatepek E, Paul K, Leichsenring M, et al. (1994) Influence functional effects of plant-derived n-3 fatty acids in humans. of dietary (n-3)-polyunsaturated fatty acids on leukotriene Prog Lip Res 64, 30–56. B 4 and prostaglandin E 2 synthesis and course of experimental 17. Reeves PG, Nielsen FH & Fahey GC (1993) AIN-93 purified tuberculosis in guinea pigs. Infection 22, 106–112. diets for laboratory rodents: final report of the American 35. Anes E, Kühnel MP, Bos E, et al. (2003) Selected lipids activate Institute of Nutrition ad hoc writing committee on the reformu- phagosome actin assembly and maturation resulting in killing lation of the AIN-76A rodent diet. J Nutr 123, 1939–1951. of pathogenic mycobacteria. Nat Cell Biol 5, 793–800. 18. Guler R, Parihar SP, Spohn G, et al. (2011) Blocking IL-1α but 36. Bhattacharya A, Sun D, Rahman M, et al. (2007) Different not IL-1β increases susceptibility to chronic Mycobacterium ratios of eicosapentaenoic and docosahexaenoic n-3 fatty acids tuberculosis infection in mice. Vaccine 29, 1339–1346. in commercial fish oils differentially alter pro-inflammatory 19. Folch J, Lees M & Stanley GS (1957) A simple method for the cytokines in peritoneal macrophages from C57BL/6 female isolation and purification of total lipides from animal tissues. mice. J Nutr Biochem 18, 23–30. J Biol Chem 226, 497–509. 37. Vilaplana C, Marzo E, Tapia G, et al. (2013) Ibuprofen therapy 20. Parihar S, Ozturk M, Marakalala M, et al. (2018) Protein kinase resulted in significantly decreased tissue bacillary loads and C-delta (PKC delta), a marker of inflammation and tuberculosis increased survival in a new murine experimental model of disease progression in humans, is important for optimal active tuberculosis. J Infect Dis 208, 199–202. macrophage killing effector functions and survival in mice. 38. Yaqoob P, Newsholme E &Calder P (1994) The effect of dietary Muc Immunol 11, 578–579. lipid manipulation on rat lymphocyte subsets and proliferation. 21. Parihar SP, Guler R, Khutlang R, et al. (2014) Statin therapy Immunol 82, 603. reduces the mycobacterium tuberculosis burden in human 39. Mayer-Barber KD, Andrade BB, Oland SD, et al. (2014) macrophages and in mice by enhancing autophagy and Host-directed therapy of tuberculosis based on interleukin-1 phagosome maturation. J Infect Dis 209, 754–763. and type I interferon crosstalk. Nature 511, 99. Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124 n-3 improves tuberculosis outcomes 397 40. Desvignes L, Wolf AJ & Ernst JD (2012) Dynamic roles of type I 43. Díaz-Muñoz MD, Osma-García IC, Íñiguez MA, et al. (2013) and type II IFNs in early infection with Mycobacterium Cyclooxygenase-2 deficiency in macrophages leads to tuberculosis. J Immunol 188, 6205–6215. defective p110γ PI3K signaling and impairs cell adhesion and 41. Mancuso P, Whelan J, DeMichele SJ, et al. (1997) Dietary migration. J Immunol 191, 395–406. fish oil and fish and borage oil suppress intrapulmonary proin- 44. Luft T, Jefford M, Luetjens P, et al. (2002) Functionally distinct flammatory eicosanoid biosynthesis and attenuate pulmonary dendritic cell (DC) populations induced by physiologic stimuli: neutrophil accumulation in endotoxic rats. Crit Care Med 25, prostaglandin E2 regulates the migratory capacity of specific 1198–1206. DC subsets. Blood J Am Soc Hematol 100, 1362–1372. 42. Wallace FA, Miles EA, Evans C, et al. (2001) Dietary fatty acids 45. Osma-Garcia IC, Punzon C, Fresno M, et al. (2016) influence the production of Th1-but not Th2-type cytokines. Dose-dependent effects of prostaglandin E2 in macrophage J Leukoc Biol 69, 449–457. adhesion and migration. Eur J Immunol 46, 677–688. Downloaded from https://www.cambridge.org/core, on subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0007114521001124