key: cord-0277809-z07hum7g authors: Arnardottir, H.; Pawelzik, S.-C.; Sarajlic, P.; Quaranta, A.; Kolmert, J.; Religa, D.; Wheelock, C. E.; Back, M. title: Immunomodulation by intravenous omega-3 fatty acid treatment in older subjects hospitalized for COVID-19: a single-blind randomized controlled trial date: 2021-12-31 journal: nan DOI: 10.1101/2021.12.27.21268264 sha: 0e05cbd96bb073753bb38d9b03a8aef0f2bce1a6 doc_id: 277809 cord_uid: z07hum7g Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes coronavirus disease 2019 (COVID-19) with respiratory distress and systemic hyperinflammation. The primary objective of this single-blind randomized controlled proof-of-concept clinical trial was to establish the effects of intravenous (i.v.) omega-3 (n-3) polyunstaurated fatty acid (PUFA) treatment compared to placebo on inflammatory markers in COVID-19, represented by leukocytes as well as inflammatory protein and lipid mediators. Here we also present an exploratory analysis of the mechanisms of action to elucidate the potential to resolve the COVID-19 hyperinflammation through interfering with lipid mediators. Inclusion criteria were COVID-19 diagnosis and clinical status requiring hospitalization. After randomized 1:1 to a once daily i.v. infusion (2 mL/kg) of either placebo (NaCl) or n-3 PUFA emulsion containing 10g of fish oil per 100 mL, results from 22 older subjects (mean age 81+/-6.1 years) were analyzed. The neutrophil to lymphocyte ratio was significantly decreased after n-3 PUFA administration. Changes in the PUFA metabolome assessed by LC-MS/MS-based lipid metabolite analysis established increased proresolving lipid mediator precursor levels and decreased formation of leukotoxin and isoleukotoxin diols by n-3 PUFA treatment. The mechanistic exploration revealed decreased immunothrombosis and preserved interferon-response. Finally, n-3 PUFA treatment may serve to limit cortisone-induced immunosuppression, including preserving leukocyte phagocytic capacity. In conclusion, i.v. n-3 PUFA administration was safe and feasible during hospitalization of multimorbid older subjects for COVID-19. The results identified a n-3 PUFA treatment mediated lipid signature of increased proresolving precursor levels and decreased leukotoxin diols in parallel to beneficial immune responses. EudraCT: 2020-002293-28; clinicaltrials.gov: NCT04647604. The uncontrolled inflammatory response in coronavirus disease 2019 (COVID-19) is characterized by a high neutrophil to lymphocyte ratio (NLR) and increased circulating levels of inflammatory mediators, represented by a monocytic cytokine release syndrome (1) and increased proinflammatory lipid mediators (2) . Systemic uncontrolled inflammation is common in severe SARS-CoV2 infection and a potential therapeutic target to improve prognosis in COVID-19 (3, 4) . Older subjects represent a particularly vulnerable patient population in COVID- 19 . The introduction of anti-inflammatory cortisone treatment (5) has confirmed the benefit of immunomodulation, although with the possible caveat of also causing immunosuppression (6) . In addition to inhibiting proinflammatory signaling, acute inflammation could potentially be disrupted by a functional and effective resolution to limit and eventually turn off inflammation while retaining an intact host defense. Fatty acids are crucial in controlling inflammation. A disturbed balance in endogenous PUFA metabolism increases proinflammatory and leukotoxic lipid mediators with concomitant decrease in proresolving lipid mediators (7) . The omega-3 (n-3) polyunstaurated fatty acids (PUFA) eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) decrease proinflammatory lipid mediators and serve as precursors for lipid mediators of the resolution of inflammation that dampen systemic inflammation while also improving healing and microbial elimination (7) . The recent VASCEPA-COVID-19 CardioLink-9 trial showed that oral n-3 PUFA treatment with the EPA ethyl ester icosapent ethyl for 14 days improves symptoms and reduces C-reactive protein (CRP) in ambulatory COVID-19 patients (8) . Similar results have been reported in a smaller study of critically ill COVID-19 patients (9) , however excluding patients not receiving enteral nutrition (10) . Intravenous (i.v.) n-3 PUFA treatment reduces hyperinflammation in other critical infectious conditions (11) but has not previously been studied in COVID-19. The primary objective of this randomized controlled proof-of-concept clinical trial was to establish the effects of i.v. n-3 PUFA compared to placebo on inflammatory markers in COVID- 19 , represented by leukocytes as well as inflammatory protein and lipid mediators (12) . Here we also present an exploratory analysis of the mechanisms of action to elucidate the potential to resolve the inflammatory storm in COVID-19 through interfering with lipid mediators. Lipid mediator quantification was performed as previously described (13) with minor modifications to the sample preparation to account for the solvent composition. One mL MeOH was added to 250 µL plasma and samples were vortexed and centrifuged. The supernatant was transferred to glass tubes and evaporated under gentle N 2 stream to remove the MeOH before reconstitution in 1 mL of extraction solution (0.2 M Na 2 HPO 4 / 0.1 M citric acid, pH 5.6) followed by solid phase extraction as previously reported (13) . Twelve additional compounds were included in the current study that were not reported in the previous methods description, including epoxides and vicinal diols from AA and DHA (Supplementary Table 1 Whole blood was collected into an 8 mL sodium heparinized CPT vacutainer and processed within 2 h of collection. Peripheral blood mononuclear cells (PBMC) were separated from erythrocytes and neutrophils following centrifugation at 1800 x g for 15 min at room temperature. Erythrocytes were isolated by 3% dextran sedimentation for 20 minutes at room temperature and then washed 3 times with PBS -/-(followed each time by centrifugation at 1000 x g, 10 min, 4°C). PBMC were collected as described above, washed with PBS -/followed by centrifugation for 10 min at 250 x g, 4°C, and then resuspended in RPMI media containing 1% heat-inactivated (56°C for 30 min) autologous serum at a concentration of 2.5x10 6 cells/mL. 0.5x10 6 cells/well were seeded into a round bottom 96-well plate and incubated with or without 1 µg/mL LPS (serotype O111:B4, Sigma-Aldrich, Merck KGaA, St. Louis, MO, USA) for 6 h. Following centrifugation for 5 min at 300 x g, 4°C, the supernatants were collected and stored at -80°C. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021. 12.27.21268264 doi: medRxiv preprint Waltham, MA, USA), were seeded into a 96-well plate and cultured over night at 37°C in a humified 5% CO 2 atmosphere. At the day of the experiment, the medium was aspirated and replaced with 200 µl MEM supplemented with 50% patient sera. Control cells were incubated with serum-free medium containing 0-200 IU/ml recombinant human IFN-α, IFN-β, and IFN-γ, respectively (RnD Systems, Minneapolis, MO, USA). After incubation for 6 h at 37°C in a humified 5% CO 2 atmosphere, the media were aspirated, the cells were lyzed in 350 µl RLT buffer (Qiagen, Hilden, Germany) supplemented with 10 µl/ml β -mercaptoethanol and frozen at -80°C until further analysis. Total RNA was extracted using the RNeasy Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instruction and subsequently reverse transcribed into cDNA using the High-Capacity RNA-to-cDNA Kit (Thermo Fisher Scientific, Waltham, MA, USA). Expression levels of the type I IFN-induced response gene MX1 and the type II IFN-induced response gene CXCL9 were analyzed by quantitative real time PCR (qPCR) on a QuantStudio 7 Flex Real-Time PCR System using specific TaqMan Gene expression assays (MX1, Hs00895608_m1; CXCL9, Hs00171065_m1, all from Thermo Fisher Scientific, Waltham, MA, USA) and quantified by comparison to the expression levels in the respective standard curves obtained from cells that were incubated with recombinant IFN protein. The procedure was replicated with WISH cells that were cultured for three consecutive passages prior to the experiment in MEM supplemented with 0.01% n-3 PUFA emulsion (v/v) to enrich n-3 PUFA in the cell membrane. Outcome measures are expressed as means with standard errors. Key results are presented as separate line plots for the placebo and n-3 PUFA groups showing changes within subjects between baseline and the endpoint. Paired t-tests were used to determine significance in changes over time. To determine the effect of time and the n-3 PUFA intervention, a mixed model analysis of variance was implemented in which time was defined as a within-subjects effect and the intervention as a between-subjects effect. Moreover, the interaction between these effects was measured. Due to the limited statistical power of the study, descriptive outcomes were not tested for significance as defined in the prespecified analysis plan in the study protocol, which is provided as an Appendix online. To gain insight into the key lipid metabolites separating the All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All subjects were included between from June to December 2020, before COVID-19 vaccination was available, at a first COVID-19 infection. The flow chart of patient inclusion is shown in Supplementary Fig 1. The study was completed for 22 older subjects (mean age 81±6.1 years). There were no significant differences in baseline characteristics between the groups ( Table 1 ). The number of administered doses were on average 4.3±1.5 in the intervention group (n=10) and 4.5±1.0 in the placebo group (n=12; P=0.66). Seven patients in each group received concomitant corticoids (P=0.68). There were 2 in-hospital deaths in each group. The n-3 PUFA treatment was administered in the form of triglycerides, and as a safety measure, fasting serum triglyceride levels were monitored daily during the study and not significantly increased. No serious adverse events were encountered. The neutrophil to lymphocyte ratio (NLR) was significantly decreased after n-3 PUFA administration (Fig 1) , while it was not significantly altered in the placebo group. The mean increase in lymphocyte count was 0.4±0.08x10 9 /L in the intervention group and 0.03±0.16x10 9 /L in the placebo group (P=0.03). There were no significant changes in total leukocyte, neutrophil or monocyte counts (data not shown). CRP decreased from 62 (20-75) to 19 (3.8-27) mg/L after n-3 PUFA (P=0.08) and from 65 (40-92) to 32 (24-40) mg/L after placebo treatment (P=0.08). Patients randomized to n-3 PUFA treatment exhibited significant increase in plasma concentrations of EPA and DHA compared with the placebo group (Fig 2A) . At the end of the study, EPA-and DHA concentrations were increased 3.7 (2.5-5.4) fold and 2.0 (1.5-2.5) fold from All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021.12.27.21268264 doi: medRxiv preprint baseline, respectively in these patients. In contrast, the EPA and DHA plasma levels were not significantly altered in the placebo group; EPA 1.0 (0.8-1.2) and DHA 1.0 (0.9-1.2) -fold at the end of the study as compared to baseline (Fig 2A) . The plasma concentrations of the n-6 PUFA arachidonic acid (AA) were not significantly altered from baseline and not significantly different between the n-3 PUFA and placebo group (Fig 2A) . The n-3 PUFA and placebo groups evidenced a separation in OPLS-DA based upon quantified levels of lipid metabolites, with a statistic fit (R2=0.75) higher that its approximation (Q2=0.38) to a testing dataset ( Fig 2B) . The DHA-and EPA-derived lipid metabolite profiles displayed in Table 2 revealed a significant increase in n-3 PUFA-derived metabolites, with EPA metabolites exhibiting the most pronounced increases in the n-3 PUFA compared with placebo groups ( Table 2 ). The arachidonic acid (AA) -derived lipid metabolite profile is shown in Supplementary Table 2 , and the octadecanoid pathways from linoleic acid (LA), -LA, -LA, in Supplementary Table 3 . The variable importance calculated from the OPLS-DA ( Fig 2B) identified 18-HEPE, 14,15-diHETE, and 9,10-diHOME among the molecules explaining the differences in plasma lipid mediator profile between n-3 PUFA and placebo treatment ( Fig 2C) . These metabolites ( Fig 2D) and the corresponding EPA and LA diols 11,12-diHETE and 12,13-diHOME (Fig 2F) , were significantly altered after n-3 PUFA but not placebo treatment. The levels of the plasma cytokines interleukin (IL)-1β, IL-6, and TNF-α were not significantly altered in either of the groups during the study duration ( Supplementary Fig 2) . Immunothrombosis has been implicated as a key pathophysiological consequence of SARS-CoV-2 infection (16) . To evaluate the n-3 PUFA effects on immunothrombosis in the present trial, ex vivo ADP-stimulation of whole blood was performed followed by flow cytometry analysis for All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021.12.27.21268264 doi: medRxiv preprint platelet-leukocyte aggregates. Blood from n-3 PUFA-treated COVID-19 patients displayed a significantly reduced platelet-leukocyte aggregates at end of treatment from baseline compared with blood from the placebo group ( Fig 3A) . This change in platelet-leukocyte aggregates was driven by a significant decrease in the formation of platelet-neutrophil aggregates in the n-3 PUFA group that was not observed in the placebo group (Fig 3B-C) . Since the platelet-leukocyte aggregation is mediated through monocyte-derived platelet activators, we next investigated the release of thrombogenic mediators from PBMC isolated from all trial participants. A significantly lower release of the immuno-thrombosis mediator Platelet-Derived Growth Factor (PDGF) and Regulated on Activation, Normal T Expressed and Secrete (RANTES) release, was observed in LPSstimulated PBMC isolated from the n-3 PUFA group compared with the placebo group ( Fig 3D) . The release of pro-inflammatory cytokines from PBMCs isolated from the trial patients was low under basal conditions measured in supernatants from unstimulated PBMC and was significantly increased with LPS stimulation ex vivo (Fig 4) . The ex vivo LPS-stimulated release of cytokines that have been most strongly associated with COVID-19 in previous reports, e.g. IL1-β, IL-6, IL-12p70 and TNF-α, was not significantly different between PBMC derived from the n-3 PUFA and placebo groups (Fig 4) . In contrast, the LPS-stimulated cytokine release from PBMC was blunted in cortisone-treated patients without apparent alterations by n-3 PUFA in either the absence or presence of cortisone treatment (Fig 4) . In COVID-19, increased cytokine release is accompanied by decreased interferon release (17) . The release of the type 1 interferons (IFN-1)  and  from PBMCs derived from COVID-19 patients in the present trial were low at baseline with a small but significant increase after LPS stimulation ( Supplementary Fig 3) . The type II interferon IFN- was secreted at higher levels and significantly increased after LPS stimulation. There were no significant differences between the treatment groups ( Supplementary Fig 3) . To increase the sensitivity and to assess the functional consequences of IFN signaling, the specific IFN-responsive WISH cell line was used to determine All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021.12.27.21268264 doi: medRxiv preprint IFN-1 responses in n-3 PUFA-and placebo-treated patients. Stimulating WISH cells with serum from trial patients revealed a time-dependent differential transcription of the IFN-1 response gene MX1 between the groups (Fig 5A) . While the IFN-1-response was preserved after stimulation with serum from n-PUFA treated patients, a time-dependent decrease was observed in cells stimulated with placebo serum (Fig 5A, left panel) . To exclude that the observed cellular effects were due to a direct effect of the higher n-3 PUFA concentrations in the active treatment compared with the placebo group (Fig 2) , the serum-stimulation experiments were repeated with WISH cells that were grown for three consecutive passages in the presence of n-3 PUFA Phagocytosis is one of the key events to permit an adequate resolution of inflammation. Therefore, treatment effects on phagocytic capacity were evaluated by ex vivo stimulation and flow cytometry analysis of blood leukocytes derived from COVID-19 patients in the present trial. Blood leukocytes derived from n-3 PUFA-treated patients exhibited a trend towards enhanced phagocytosis of Zymosan particles, which was driven by increased phagocytosis in the group with a combined cortisone and n-3 PUFA treatment ( Fig 5B) . Length of hospital stay is shown in Supplemental Fig 4. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. This single-blind randomized controlled proof-of-concept trial of i.v. n-3 PUFA in COVID-19 met its primary endpoint of changes in inflammatory biomarkers for white blood cell counts and lipid metabolites, whereas observed changes in CRP and cytokines were non-significant. We present here also the results of the exploratory analysis of the underlying mechanisms using multiple in vitro experimental approaches on the clinically obtained samples. The immunothrombosis interaction was significantly reduced by n-3 PUFA compared with placebo. In addition, n-3 PUFA treatment preserved IFN-1 signaling and phagocytosis. Taken together, these results point to a beneficial immunomodulation by n-3 PUFA in COVID-19 with implications for systemic inflammation, immunothrombosis, and counteraction of cortisone-induced immunosuppression. The present trial included multimorbid older subjects, which are at a particular high risk of adverse outcomes in COVID-19 and may not always benefit from advanced treatments. The results show beneficial effects of n-3 PUFA on NLR, which is a marker of poor prognosis in COVID- 19 (18) . This observation was driven by a significantly increased lymphocyte count after n-3 PUFA treatment compared with placebo. The study was not adequately powered to detect effects on clinical course of disease and therefore length of hospital stay is shown in a descriptive analysis. (19, 20) . 18-HEPE is the precursor for the short-lived, locally produced, specialized proresolving mediators (SPM) resolvins (Rv) E1 (RvE1) (21) and E3 (22) , which may contribute to control inflammation in COVID-19 (23, 24) . Few clinical studies have however previously addressed changes in SPM and their precursors after n-3 PUFA supplementation in All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (25, 26) , and for limited number of subjects from cardiovascular n-3 PUFA trials (27) . In addition to the increased SPM precursor, the present trial identified changes in PUFA metabolites derived from double bond epoxidation followed by conversion to the corresponding vicinal diols by the enzyme soluble epoxide hydrolase (sEH). Treatment with n-3 PUFA increased the circulatory levels of the EPA-derived diols 11,12-and 14,15-diHETE, while the corresponding LA-derived diols of leukotoxin 9,10-diHOME and isoleukotoxin 12,13-diHOME decreased. Leukotoxins and their diols have been identified as key effectors in adult respiratory distress syndrome (ARDS) (28, 29) and detected at increased levels in COVID-19 patients with severe pulmonary involvement (30) . The results of the present study extend the therapeutic implications of n-3 PUFA treatment to substrate competition for sEH-catalyzed formation of leukotoxins to support the emerging notion of sEH as a COVID-19 therapeutic target (2) . Beneficial effects of decreased leukotoxin diols by n-3 PUFA treatment are supported by the increased lymphocyte counts in n-3 PUFA treated COVID-19 patients in the present trial. Leukocyte-platelet aggregates are of importance for the severity of ARDS (31) and have emerged as of importance for pathophysiology and thrombotic complications of COVID-19 (32) through an immunothrombosis interaction (33, 34) . In support of representing an additional targeted pathway in the immunomodulation by n-3 PUFA, the present trial revealed decreased platelet leukocyte aggregates in the n-PUFA treatment group, which was driven by reduced formation of neutrophil platelet aggregates. Furthermore, PMBCs isolated during and after n-3 PUFA treatment released significantly lower levels of PDGF and RANTES, which are potent mediators of leukocyte-platelet aggregates (35) . Taken together, these results point to improved immunothrombosis in COVID-19 by n-3 PUFA. The subjects included in this trial required hospitalization but did not exhibit signs of ongoing cytokine storm given the moderate cytokine increase in plasma and low spontaneous release from isolated PBMCs. Nevertheless, when mimicking the cytokine release syndrome by ex vivo LPS stimulation, cytokine release was significantly increased, without significant attenuation in n-3 PUFA treated subjects. These findings contrast with sepsis patients, in which PBMCs isolated after intravenous n-3 PUFA treatment reduce the LPS-stimulated cytokine All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. release by approximately 30% (11) . The present study was performed during the introduction of cortisone treatment (5) , which was equally distributed within the placebo and active treatment groups. Lower stimulated cytokine release was observed in PBMCs isolated from cortisonetreated COVID-19 patients, confirming the anti-inflammatory effects of cortisone treatment on the release of the mediators identified as key in the cytokine release syndrome (6) . Concerns have been raised of cortisone-induced anti-inflammation also being indicative of an immunosuppression (36) . One important beneficial immune response to consider for an optimal resolution of COVID-19 inflammation is the anti-viral effects of IFN-1. Particularly the enhanced inflammation and more severe disease reported in older patients with comorbidities has been linked to suppressed IFN-1 (37) . The latter observation was extended in the present study to a low IFN-1 release from isolated PBMCs in older patients. Through a translational approach examining MX1 expression after stimulation of specific IFN-responsive WISH cell with serum from the trial participants, we show that IFN-1 responses were retained in patients treated with n-3 PUFA whereas a time-dependent decrease in the transcriptional IFN-1 response was observed in cells treated with placebo serum. Importantly, cortisone treatment suppressed IFN-1 responses to a higher degree in placebo-compared with n-3 PUFA treated patients. These results provide an initial suggestion for a retained IFN-1 response with concomitant n-3 PUFA administration with cortisone-treatment of COVID-19 In further support of n-3 PUFA as a promoter of the resolution of COVID-19-induced inflammation, the decreased phagocyte function in COVID-19 is restored in vitro after exogenous administration of n-3 PUFA-derived proresolving lipid mediators (38) . Interestingly, the present study extends those in vitro effects to a clinically relevant effect of i.v. n-3 PUFA administration in COVID-19 by showing that zymosan phagocytosis was enhanced in whole blood from n-3 PUFA compared with placebo-treated patients. The apparent most prominent phagocytosis increase was observed in the group with a combined cortisone and n-3 PUFA treatment, further supporting a beneficial synergetic effect of cortisone and n-3 PUFA treatment in COVID-19. The stimulated phagocytosis by n-PUFA suggest retaining of crucial immune functions for clearance mechanisms to sustain the resolution response and counteract against cortisone-induced immunosuppression. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The main limitation is the low number of participants. It is important to consider that the study was a proof-of-concept trial powered to detect significant effects on leukocyte, lipid, and protein inflammatory biomarkers and designed to allow mechanistic exploration of n-3 PUFA metabolism and cellular effects (12) . Larger studies are nevertheless needed to determine if the significantly improved NLR by n-3 PUFA translates into better clinical outcomes in COVID-19. The clinical outcomes were not statistically tested in the present trial according to the a priori study plan and were therefore only presented in descriptive analyses. The generalisability of identified All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The study received a research grant from King Gustaf V and Queen Victoria Freemason All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021.12.27.21268264 doi: medRxiv preprint All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Neutrophil-to-lymphocyte ratio (upper panels) and lymphocyte counts (lower panels) at baseline and after treatment (End) with intravenous infusion (2 mL/kg) of either placebo (NaCl; left panels, n=12) or n-3 PUFA emulsion containing 10 g of fish oil per 100 mL (right panels, n=10). Horizontal lines represent the median at each time point. Statistical analyses were performed with a Wilcoxon Signed Rank Test for repeated measures. NS, non-significant. Statistical significance was set at P=0.05. , and the LA metabolite 9,10-diHOME (right) among the critical molecules explaining the differences in plasma lipid mediator profile caused by n-3 PUFA intravenous infusion (blue= higher and red = lower in the n-3 PUFA group), which were (D) significantly altered in the n-PUFA (blue symbols; n=9) but not placebo (black symbols; n=12) groups between baseline and End. (F) Change in the corresponding EPA and LA diols, 11,12-diHETE (middle) and 12,13-diHOME (right). Results are presented as individual values and statistical analysis was performed by a paired Student's t test. Statistical significance was set at P=0.05. * p<0.05; **p<0.01; ***p<0.001 compared to baseline and # p<0.05; ## p<0.01 between placebo and n-3 PUFA treatment. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021.12.27.21268264 doi: medRxiv preprint placebo n=6, n-3 PUFA+LPS n=7 and n-3 PUFA NS n=6) concomitant cortisone treatment are shown in middle and right panels, respectively. Results are expressed as mean ± SEM. Statistical analyses were performed with 2-way ANOVA for repeated measures. RT-qPCR analysis of MX1 mRNA expression in WISH cells stimulated with patient serum in the absence (right panel) or presence (middle panel) of exogenous n-3 PUFA (placebo n=12; n-3 PUFA n=10). Subgroups with (dotted line; placebo n=6, n-3 PUFA n=7) and without (solid lines; placebo n=6, n-3 PUFA n=3) concomitant cortisone treatment are shown in right panel. (B) Fold change in phagocytosis of pHrodored labeled zymosan by human peripheral blood leukocytes (CD45 + ) after incubation for 45 min at 37°C (placebo n=11; n-3 PUFA n=9). Subgroups without (middle panel; placebo n=5 and n-3 PUFA n=3) or with (right panel; placebo n=6, n-3 PUFA n=6) concomitant cortisone treatment are shown in middle and right panels, respectively. (C) Representative histograms from each group and subgroup at End. Serum and whole blood were derived from patients at baseline, at 48 h (Early) and after treatment (End) with intravenous infusion (2 mL/kg) of either placebo (black symbols NaCl) or n-3 PUFA emulsion containing 10 g of fish oil per 100 mL (blue symbols). Subgroups with (dotted line) and without concomitant cortisone treatment are shown in A, right panel and B middle panel. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. ; https://doi.org/10.1101/2021.12.27.21268264 doi: medRxiv preprint All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Supplemental Figure 1 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted December 31, 2021. Cytokines were measured in plasma collected from patients at baseline, 48 h (Early) and after treatment (End) with intravenous infusion (2 mL/kg) of either placebo (NaCl; black n=12) or n-3 PUFA emulsion containing 10 g of fish oil per 100 mL (blue n=10). Results are expressed as mean ± SEM. Statistical analyses were performed with 2-way ANOVA for repeated measures and statistical significance was set at P=0.05. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Figure 3 Supplemental Figure 3 . Interferon (IFN) , , and g in supernatants from non-stimulated (NS, dotted lines) and, LPS-stimulated (LPS, solid lines) PBMC isolated from patients at baseline, at 48 h (Early), and after treatment (End) with intravenous infusion (2 mL/kg) of either placebo (black symbols NaCl); n=12) or n-3 PUFA emulsion containing 10 g of fish oil per 100 mL (blue symbols, LPS n=10; NS n=9) for 5 days. Results are expressed as mean ± SEM. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Figure 4 Supplemental Figure 4 : Length of hospital stay after start of treatment with intravenous infusion (2 mL/kg) of either placebo (NaCl; dotted blue line n=12) or n-3 PUFA emulsion containing 10 g of fish oil per 100 mL (solid red line n=10). All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 2 . Lipid mediator metabolites from the n-6 PUFA AA in patient plasma. Data is expressed as mean (SEM) in pg/mL of plasma. P-values were derived from the mixed effects ANOVA containing intervention (Interv; n-PUFA vs. placebo) and progress (early and end samples) as a within-subjects effect and as a between subjects effect. Intervention:progress (Interv:Prog) represents the significance of the interaction between these two terms. Monocyte-driven atypical cytokine storm and aberrant neutrophil activation as key mediators of COVID-19 disease severity Eicosanoids: The Overlooked Storm in Coronavirus Disease 2019 (COVID-19)? Calming the cytokine storm in COVID-19 Targeting cytokine storm in COVID-19: what have we learned? Dexamethasone in Hospitalized Patients with Covid-19 After 62 years of regulating immunity, dexamethasone meets COVID-19 Pro-resolving lipid mediators are leads for resolution physiology A randomized trial of icosapent ethyl in ambulatory patients with COVID-19 The effect of omega-3 fatty acid supplementation on clinical and biochemical parameters of critically ill patients with COVID-19: a randomized clinical trial The effect of omega-3 fatty acid supplementation on clinical and biochemical parameters of critically ill patients with COVID-19: a randomized clinical trial Parenteral nutrition with fish oil modulates cytokine response in patients with sepsis Stimulating the Resolution of Inflammation Through Omega-3 Polyunsaturated Fatty Acids in COVID-19: Rationale for the COVID-Omega-F Trial Lipid Mediator Quantification in Isolated Human and Guinea Pig Airways: An Expanded Approach for Respiratory Research All rights reserved. No reuse allowed without permission the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted Orchestrating high-throughput genomic analysis with Bioconductor Immunothrombosis in severe COVID-19 The cytokine storm in COVID-19: An overview of the involvement of the chemokine/chemokine-receptor system IL-6-based mortality risk model for hospitalized patients with COVID-19 ERV1/ChemR23 Signaling Protects Against Atherosclerosis by Modifying Oxidized Low-Density Lipoprotein Uptake and Phagocytosis in Macrophages 18-HEPE, an n-3 fatty acid metabolite released by macrophages, prevents pressure overload-induced maladaptive cardiac remodeling Resolvin E1 metabolome in local inactivation during inflammation-resolution Cutting Edge: Severe SARS-CoV-2 Infection in Humans Is Defined by a Shift in the Serum Lipidome, Resulting in Dysregulation of Eicosanoid Immune Mediators Acute severe SARS COVID-19 patients produce pro-resolving lipids mediators and eicosanoids Dysregulated plasma lipid mediator profiles in critically ill COVID-19 patients Identification of specialized pro-resolving mediator clusters from healthy adults after intravenous low-dose endotoxin and omega-3 supplementation: a methodological validation Bioactive products formed in humans from fish oils Specialized proresolving lipid mediators in patients with coronary artery disease and their potential for clot remodeling Bioactivation of leukotoxins to their toxic diols by epoxide hydrolase Leukotoxin-diol: a putative toxic mediator involved in acute respiratory distress syndrome Plasma Linoleate Diols Are Potential Biomarkers for Severe COVID-19 Infections Early Intravascular Events Are Associated with Development of Acute Respiratory Distress Syndrome. A Substudy of the LIPS-A Clinical Trial Neutrophil-Platelet and Monocyte-Platelet Aggregates in COVID-19 Patients The dual role of platelet-innate immune cell interactions in thrombo-inflammation Thrombocytopathy and endotheliopathy: crucial contributors to COVID-19 thromboinflammation Platelet-leukocyte interplay during vascular disease Caution against corticosteroid-based COVID-19 treatment Adaptive immunity to SARS-CoV-2 and COVID-19 Disrupted Resolution Mechanisms Favor Altered Phagocyte Responses in COVID-19 Table 3 . Lipid mediator metabolites from the n-6 PUFA LA in patient plasma. Data is expressed as mean (SEM) in pg/mL of plasma. P-values were derived from the mixed effects ANOVA containing intervention (Interv; n-PUFA vs. placebo) and progress (early and end samples) as a within-subjects effect and as a between subjects effect. Intervention:progress (Interv:Prog) represents the significance of the interaction between these two terms.