Differential Roles of Interleukin-6 in Severe Acute Respiratory Syndrome-Coronavirus-2 Infection and Cardiometabolic Diseases

Abstract

Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) infection can lead to a cytokine storm, unleashed in part by pyroptosis of virus-infected macrophages and monocytes. Interleukin-6 (IL-6) has emerged as a key participant in this ominous complication of COVID-19. IL-6 antagonists have improved outcomes in patients with COVID-19 in some, but not all, studies. IL-6 signaling involves at least 3 distinct pathways, including classic-signaling, trans-signaling, and trans-presentation depending on the localization of IL-6 receptor and its binding partner glycoprotein gp130. IL-6 has become a therapeutic target in COVID-19, cardiovascular diseases, and other inflammatory conditions. However, the efficacy of inhibition of IL-6 signaling in metabolic diseases, such as obesity and diabetes, may depend in part on cell type-dependent actions of IL-6 in controlling lipid metabolism, glucose uptake, and insulin sensitivity owing to complexities that remain to be elucidated. The present review sought to summarize and discuss the current understanding of how and whether targeting IL-6 signaling ameliorates outcomes following SARS-CoV-2 infection and associated clinical complications, focusing predominantly on metabolic and cardiovascular diseases.

Editor note:

Peter Libby is an Associate Editor of Cardiology Discovery. Guo-Ping Shi is an Editorial Board Member of Cardiology Discovery. The article was subject to the journal’s standard procedures, with peer review handled independently of these editors and their research groups.

1. Introduction

Interleukin-6 (IL-6) is a pro-inflammatory cytokine that plays pleiotropic roles in many human inflammatory diseases. Infection with severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) provides a timely example in which blood and alveolar lavage fluids contain elevated levels of IL-6 that correlate positively with disease stages and negatively with oxygen saturation and oxygen partial pressure, and predicts the mortality of patients with COVID-19.^[1–3] SARS-CoV-2 infection exacerbates the damage and dysfunction of not only the respiratory system, but many other organs.

The activity we now know as IL-6 emerged from several independent findings. Early reports of IL-6 described a soluble factor from T cells that induced B-cell antibody production.^[4,5] IL-6 plays a major role in CD4⁺ T-cell subset differentiation.^[6] In combination with transforming growth factor (TGF)-β, IL-6 promotes Th17 lineage differentiation, although TGF-β is an essential regulatory T-cell (Treg) differentiation factor.^[7] During inflammation, IL-6 levels in the circulation can increase from 1–5 pg/mL to several μg/mL.^[8] IL-6 controls chemokine-directed leukocyte trafficking and innate-to-adaptive immunity transition by regulating leukocyte activation, differentiation, and proliferation.^[9] Many cell types, in addition to T cells, express IL-6. Numerous studies have implicated IL-6 in regulating cardiovascular injury and metabolic disorders,^[10,11] including glucose metabolism,^[12,13] insulin sensitivity, energy expenditure, and lipid homeostasis, as well as endothelial function.^[14] Vascular smooth muscle cells (VSMC) elaborate IL-6 in atherosclerotic plaques.^[15,16] In aged atherogenic apolipoprotein E-deficient (Apoe^−/−) mice, IL-6 deficiency increased atherosclerotic lesion size and calcification, but did not affect hypercholesterolemia.^[17] However, administration of recombinant IL-6 to high-fat diet (HFD)-fed Apoe^−/− mice increased systemic inflammation and atherosclerotic lesion growth.^[18] In atherosclerosis-prone, low-density lipoprotein receptor-deficient mice, IL-6 inhibition with sgp130Fc also reduced atherosclerosis without affecting bodyweight or serum lipid profile.^[19] IL-6 associates positively with the incident or extent of coronary artery disease (CAD).^[20–22] Myeloid cell-derived IL-6 suppresses accumulation of macrophages in adipose tissue, while adipocyte- or skeletal muscle-derived IL-6 increases macrophage infiltration into adipose tissue.^[23] Therefore, IL-6 may have disparate functions in cardiovascular and metabolic diseases. Recent studies from patients with COVID-19 have also yielded mixed results. Furthermore, as with most interventions, global targeting of IL-6 may produce some adverse effects that could balance beneficial actions. Ultimately, properly powered and rigorous randomized clinical trials will be needed to ascertain the effects of numerous strategies of IL-6 inhibition on various clinical outcomes.

2. IL-6 receptors

IL-6 receptor (IL-6R, CD126) has 2 forms: an 80,000-Da transmembrane receptor (mIL-6R) and a 50,000 – 55,000-kDa soluble fragment generated by limited proteolysis (sIL-6R).^[24] Transmembrane IL-6R localizes in hepatocytes, leukocytes (including macrophages, neutrophils, and some T-cell subtypes), islet cells, podocytes, and microglia.^[25–27] The sIL-6R is generated from mIL-6 proteolytic shedding predominantly or from mRNA splicing to a much lesser extent.^[28,29] Membrane-bound proteases of the disintegrin and metalloproteinases (ADAMs) family (ADAM-10 and ADAM-17) affect sIL-6R shedding in humans [Figure 1A].^[27,30,31] Proteolysis of human mIL-6R was found to occur between Pro³⁵⁵ and Val^356[31] or between Gln³⁵⁷ and Asp^358[32] [Figure 1B], corresponding to murine Pro³⁵⁴ and Val³⁵⁵ or Gln³⁵⁶ and Glu³⁵⁷, respectively.^[33] Furthermore, mutation of Val³⁵⁶ abrogated human IL-6R proteolysis. Deletion of the 3 serine residues in mIL-6R (Ser³⁵⁹, Ser³⁶⁰, Ser³⁶¹) [Figure 1B] muted the activity of ADAM17 but did not affect the activity of ADAM10.^[32] The ADAM17 inhibitor GW280264X or the ADAM10 inhibitor GI254023X blocked mIL-6R proteolysis. However, none of these manipulations affected IL-6R cell surface targeting, IL-6 binding, or IL-6-induced cell proliferation.^[31,32] Atherosclerotic lesions exhibit increased ADAM17 expression.^[34] Tumor necrosis factor (TNF)-α converting enzyme (TACE) is the counterpart of ADAM17 that increases in HFD-fed obese mice.^[35] Mouse IL-6R may also undergo limited proteolysis to generate sIL-6R.^[36]

Figure 1:

IL-6R proteolytic processing. (A) Cell-membrane-bound metalloproteinases ADAM10 and ADAM17 are responsible for IL-6R proteolytic processing to generate sIL-6R; alternative splicing produces the same product but to a much lesser extent. (B) Human and mouse IL-6R extracellular domain and transmembrane domain junction sequences. PVQD in humans and PVQE in mice are the ADAM10/ADAM17 cleavage sites. The shaded areas belong to the transmembrane domain. IL-6R: Interleukin-6 receptor; mIL6-R: Transmembrane IL-6R; sIL6-R: Soluble IL-6R.

IL-6R mediates IL-6 signaling in 3 pathways: classic-signaling, trans-signaling, and trans-presentation (also called cluster-signaling) [Figure 2A–C]. Classic-signaling is mediated by mIL-6R and a 130,000-Da type-I transmembrane glycoprotein (gp130), also called CD130, that serves as a signal transducer of IL-6 [Figure 2A].^[37] In contrast to IL-6R, many cell types express gp130 and it serves as a co-receptor for all IL-6 family members, including IL-11, ciliary neurotrophic factor, oncostatin-M, and cardiotrophin-1.^[38] While gp130 is ubiquitously expressed, mIL-6R is only expressed with gp130 on limited cell types, such as immune cells, hepatocytes, and interstitial epithelial cells. Formation of the IL-6/IL-6R/gp130 complex activates downstream Janus kinases (JAKs ) that subsequently phosphorylate the docking site of signal transducer and activator of transcription (STAT)3 and STAT1 [Figure 2A]. IL-6-mediated STAT3 activation induces Th17 development and contributes to the pathobiology of CAD and metabolic diseases.^[39] Unlike classic-signaling, sIL-6R and gp130 mediate trans-signaling independent of mIL-6R. sIL-6R circulates in blood,^[27] where it forms IL-6/sIL-6R complexes that activate gp130 on target cells [Figure 2B].^[31,40] Surface expression of gp130 normally exceeds that of IL-6R. Trans-signaling amplifies the IL-6 signal by increasing gp130 engagement.^[41,42] Th17 differentiation requires IL-6 classic-signaling and trans-signaling to maintain Th17 cell function.^[43,44] IL-6 trans-presentation was first reported on Sirpα⁺ dendritic cells; mIL-6R on these cells initially binds to IL-6, then presents IL-6 to gp130 on gp130-expressing Th17 cells as a mechanism to prime pathogenic Th17 cells [Figure 2C].^[45]

Figure 2:

Three IL-6 signaling pathways. (A) Classic-signaling pathway: IL-6 binds to mIL-6R and this complex then binds to gp130 to stimulate downstream signaling. (B) Trans-signaling: ADAM10 and ADAM17 mediate mIL-6R proteolytic processing to generate sIL-6R. IL-6 then binds to sIL-6R and this IL-6-sIL-6R complex interacts with gp130 to stimulate IL-6 trans-signaling. (C) Trans-presentation (also called Cluster-signaling): IL-6 binds to mIL-6R on 1 cell (eg, dendritic cell) and mIL-6R then presents IL-6 to gp130 on a different cell (eg, Th17 cell). IL-6R: Interleukin-6 receptor; mIL-6R: Transmembrane IL-6R; sIL6-R: Soluble IL-6R.

Alternative splicing of gp130 produces 4 isoforms of soluble gp130 (sgp130), and to a lesser extent, ADAM17 and ADAM10 mediate gp130 proteolysis and produce sgp130.^[46,47] These sgp130 isoforms do not bind to IL-6 or IL-6R alone^[48] or to other IL-6 family members,^[19] but bind to the IL-6/sIL-6R complex as a mechanism to mute IL-6 trans-signaling [Figure 3A].^[48–51] The concentration of sgp130 in the circulation can reach 100–400 μg/L. However, if the plasma concentration of sgp130 is low, the counterbalance of IL-6 trans-signaling may be compromised, which occurred in CAD patients.^[19] Several mouse experiments have tested this activity of sgp130.^[52]

Figure 3:

Mechanisms of blocking IL-6 trans-signaling. (A) Alternative splicing of gp130 produces the majority of sgp130, while ADAM17 and ADAM10 also mediate gp130 proteolysis to produce sgp130. Subsequently, sgp130 then interacts with the sIL-6R-IL-6 complex to block IL-6 trans-signaling. (B) Like sgp130, the manipulated sgp130Fc (olamkicept) also binds to the sIL-6R-IL-6 complex to avoid the trans-signaling pathway. IL-6: Interleukin-6; mIL-6R: Transmembrane IL-6R; sIL6-R: Soluble IL-6R.

3. IL-6 signaling drug targeting

Direct targeting of IL-6 using antibodies against IL-6 has been tested in several human diseases, such as CAD, rheumatoid arthritis (RA), chronic kidney disease (CKD), organ transplantation rejection, and systemic lupus erythematosus (SLE). The antibodies include ziltivekimab, olokizumab, siltuximab, sirukumab, clazakizumab, and MEDI5117 [Table 1]. The RESCUE trial (Randomized Evaluation of Patients with Stable Angina Comparing Utilization of Noninvasive Examinations; ClinicalTrials.gov, NCT03926117) was a randomized, double-blind, phase II trial from 40 clinical sites in the USA, including 264 adult participants (≥18 years old) with elevated plasma high-sensitivity C-reactive protein (hsCRP≥2 mg/L) and moderate to severe CKD. Participants received the humanized anti-IL-6 antibody ziltivekimab (7.5, 15, or 30 mg) every 4 weeks for 24 weeks. At 12 weeks, the median changes of hsCRP relative to the baselines reduced by 77%, 88%, and 92%, respectively, compared with a 4% reduction from the placebo group. Fibronectin, serum amyloid A, haptoglobin, secretary phospholipase A2, and lipoprotein(a) also decreased dose-dependently. However, this antibody did not affect hemoglobin A1c,^[53] even though IL-6 plays a vital role in diabetes.

Table 1:

Selected anti-IL-6 antagonists.

Anti-IL-6 antibodies	Developers	Latest status†
Ziltivekimab*  COR 001  NN 6018	Novo Nordisk A/S (Bagsværd, Denmark)	Phase I, chronic kidney disease and systemic inflammation (NCT05379829) Phase M, heart failure (NCT05636176) Phase M, inflammation; cardiovascular disorders; kidney disorders (NCT05021835)
Olokizumab*  Anti-IL-6-R-Pharm  Artlegia  CDP-6038	R-Pharm International, LLC (Princeton, New Jersey, USA)	Phase III, SARS-CoV-2 acute respiratory disease (NCT05196477, NCT05187793) Phase III, rheumatoid arthritis (NCT02760433, NCT03120949, NCT02760368, NCT02760407)
Siltuximab*  Centocor  cCLB8  CNTO-328  Sylvant	H. Lee Moffitt Cancer Center and Research Institute (Tampa, Florida, USA) University of Alabama at Birmingham (Birmingham, Alabama, USA) Timothy Voorhees (Columbus, Ohio, USA) Emory University (Atlanta, Georgia, USA) Carla Greenbaum, MD (Seattle, WA, USA) Memorial Sloan Kettering Cancer Center (New York, New York, USA) Janssen Research & Development, LLC (Raritan, New Jersey, USA) A.O. Ospedale Papa Giovanni XXIII (Bergamo, Italy)	Phase I, large granular lymphocyte leukemia (NCT05316116) Phase II, cytokine release syndrome ICANS lymphoma, non-Hodgkin multiple myeloma acute lymphoblastic leukemia (NCT04975555) Phase I, non-Hodgkin lymphoma (NCT05665725) Phase I, metastatic pancreatic adenocarcinoma (NCT04191421) Phase I, type 1 diabetes (NCT02641522) Phase II, multiple myeloma (NCT03315026) Phase II, Multicentric Castleman’s Disease (NCT01400503) Phase III, SARS-CoV-2 acute respiratory disease (NCT04322188)
Sirukumab*  CNTO-136  PLIVENSIA™  Shirukumabu	Janssen Research & Development, LLC (Raritan, New Jersey, USA)	Phase II, major depressive disorder (NCT02473289) Phase II, COVID 2019 infections (NCT04380961)
Clazakizumab*  ALD 518  BMS-645429  BMS-945429  CSL-300	Stanley Jordan, MD Cedars-Sinai Medical Center (Los Angeles, California, USA) Medical University of Vienna (Vienna, Austria) NYU Langone Health (New York, New York, USA) CSL Behring (King of Prussia, Pennsylvania, USA) University of North Carolina, Chapel Hill (Chapel Hill, North Carolin, USA)	Phase I/II, chronic kidney failure; end-stage renal disease; transplant glomerulopathy; kidney transplant failure and rejection; antibody-mediated kidney transplant rejection; complications (NCT03380962, NCT03380377) Phase II, COVID-19 (NCT04348500, NCT04343989) Phase II, antibody-mediated rejection (NCT03444103) Phase III, antibody-mediated rejection (NCT03744910) Phase II/III, atherosclerotic cardiovascular disease; end-stage kidney disease (NCT05485961) Phase II, rheumatoid arthritis (NCT02015520) Phase II, asthma (NCT04129931) Phase II, arthritis; psoriatic (NCT01490450)
MEDI5117*  WBP216	Novo Nordisk A/S (Bagsværd, Denmark)	Discontinued, rheumatoid arthritis (NCT01559103)
PF-04236921	Pfizer (New York, New York, USA)	Phase II, Crohn’s disease (NCT01287897) Phase I, rheumatoid arthritis (NCT00838565) Phase II, systemic lupus erythematosus (NCT01405196) Phase II, Crohn’s disease (NCT01345318)

^*Below shows alternative names

^†Selected most recent trials.

IL-6: Interleukin-6.

A phase II randomized pilot trial of late antibody-mediated rejection (ABMR) in 20 kidney transplant recipients tested the effect of IL-6 inhibition with clazakizumab. Patients received subcutaneous injections of 25 mg clazakizumab or placebo every 4 weeks for 12 weeks, then underwent 40 weeks of open-label extension in which all participants received clazakizumab. Patients were assessed at ≥360 d post-transplantation. IL-6 inhibition with clazakizumab significantly reduced donor-specific antibody production and the expression of transplant rejection-related genes. Clazakizumab also reduced plasma hsCRP to 0.13 (0.04–0.26) compared with 0.42 (0.08–0.48) in the placebo group (mg/dL, median (Q1–Q3)).^[54] PF-04236921 is a humanized anti-IL-6 IgG2 monoclonal antibody. A phase I trial on PF-04236921 involved 183 patients with SLE receiving subcutaneous injection of PF-04236921 at 10, 50, and 200 mg every 8 weeks for 24 weeks. Patients who received 10 mg (n=0) or 50 mg (n=2) of PF-04236921 had a significantly reduced incidence of severe flares compared with those who received a placebo (n=8, P<0.01). In addition, patients who received PF-04236921 developed much lower SLE responder index-4 (P=0.004) and British Isles Lupus Assessment Group-based Composite Lupus Assessment (P=0.012) response rates compared with placebo-treated patients.^[55] Table 1 summarizes the most recent on-going and completed clinical trials of selected anti-IL-6 antibody drugs.

Targeting IL-6 and IL-6R might yield similar results.^[43,56] However, IL-6R (mIL-6R and sIL-6R) may vary less between patients compared with IL-6, suggesting that targeting IL-6R vs. its ligand IL-6 may differ due to different types of IL-6R.^[57,58] Antibodies used to neutralize IL-6R include tocilizumab, sarilumab, satralizumab, and vobarilizumab [Table 2]. The anti-IL-6R antibody tocilizumab binds to mIL-6R and sIL-6R to block IL-6 signaling.^[59,60] Tocilizumab reduces systemic inflammation after out-of-hospital cardiac arrest. Of 80 comatose patients with out-of-hospital cardiac arrest, one dose of tocilizumab in addition to standard care reduced plasma hsCRP (−84%, P<0.001) and leukocyte counts (−34%, P<0.001) compared with those of the placebo group at 24 h after treatment.^[61] However, IL-6R targeting with tocilizumab and sarilumab may have unwanted actions, including dyslipidemia, increased risk of infections, mild increase of blood lipids, increased risk of liver malfunction, bodyweight gain,^[62] and pancreatitis.^[63,64] Therefore, selective targeting of IL-6 trans-signaling may merit future consideration as it may help avoid the adverse effects associated with global IL-6 signaling blockade. Olamkicept (also called sgp130Fc) [Table 2] is a recombinant version of sgp130. This gp130 trans-signaling inhibitor comprises the 2 complete extracellular domains of gp130 dimerized by fusion to the fragment crystallizable region (Fc region) of human IgG1. Olamkicept acts as an IL-6/sIL-6R trap to target the trans-signaling pathway without interacting individually with IL-6 or IL-6R [Figure 3B].^[48] Table 2 summarizes the most recent on-going and completed clinical trials of selected anti-IL-6R antibody drugs.

Table 2:

Selected anti-IL-6 receptor antagonists.

Anti-IL-6R antibodies	Developer	Latest status†
Tocilizumab*  Actemra  ACTPen  Atlizumab  HPM-1  MRA  R-1569  RG-1569  rhPM-1  RO4877533  RoActemra	Karadeniz Technical University (Trabzon, Turkey) Hospital Italiano de Buenos Aires (Caba, Argentina) Assistance Publique - Hôpitaux de Paris (Paris, France) University of Colorado, Denver (Denver, Colorado, USA) Shaheed Zulfiqar Ali Bhutto Medical University (Islamabad, Pakistan) Hoffmann-La Roche (Berlin, Germany) Peking Union Medical College Hospital (Beijing, China) Tianjin Medical University General Hospital (Tianjin, China) Columbia University (New York, New York, USA) Nationwide Children’s Hospital (Aurora, Colorado, USA)	COVID-19 critical care mortality (NCT04893031, NCT04924829, NCT04873M1)‡ Phase III, giant cell arteritis neurovascular disorder (NCT04888221) Phase I, adamantinomatous craniopharyngioma (NCT03970226) Rheumatoid arthritis (NCT02809833, NCT02648035)‡ Chronic periaortitis (NCT05133895)‡ Phase I/II, neuromyelitis optica spectrum disorders; neuromyelitis optica; Devic’s disease (NCT03062579) Phase II, immune-related adverse events advanced solid tumor (NCT04375228) Phase II, adamantinomatous craniopharyngioma; recurrent adamantinomatous craniopharyngioma (NCT05233397)
Sarilumab*  Kevzara  REGN-88  SAR-153191	Westyn Branch-Elliman, VA Boston Healthcare System (Boston, Massachusetts, USA) Maria del Rosario Garcia de Vicuña Pinedo, Hospital Universitario de la Princesa (Madrid, Spain) Sanofi (Paris, France) National Institute of Allergy and Infectious Diseases (NIAID) (Bethesda, Maryland, USA) Stanford University (Palo Alto, California, USA)	Phase II, COVID-19 (NCT04359901, NCT04357808) Phase III, corona virus infection (NCT04327388) Phase III, rheumatoid arthritis (NCT02121210) Phase II, indolent systemic mastocytosis (NCT03770273) Phase II, juvenile idiopathic arthritis (NCT02991469) Phase II, sarcoidosis (NCT04008069)
Satralizumab*  ENSPRYNG  RG-6168  SA-237  Satralizumab-mwge	International University of Health and Welfare (Tokyo, Japan) Hoffmann-La Roche (Berlin, Germany) Dongzhimen Hospital, Beijing (Beijing, China)	Phase II, pulmonary arterial hypertension (NCT05679570) Phase III, neuromyelitis optica spectrum disorder (NCT05199688) Phase III, generalized myasthenia gravis (NCT04963270) Phase III/IV, neuromyelitis optica spectrum disorder (NCT04660539, NCT05269667) Phase III, myelin oligodendrocyte glycoprotein antibody-associated disease (NCT05271409) Phase III, neuromyelitis optica; spectrum disorder (NCT02028884) Phase III, NMDAR (n-methyl-D-aspartic acid receptor) autoimmune encephalitis LGI1 (leucine-rich glioma-inactivated 1) autoimmune encephalitis (NCT05503264) Demyelinating diseases of the central nervous system (NCT05415579)‡
Vobarilizumab*  ALX 0061	Ablynx, a Sanofi company (Ghent, Belgium)	Phase II, rheumatoid arthritis (NCT02518620) Phase II, systemic lupus erythematosus (NCT02437890)
Olamkicept*  sgp130Fc  FE-301  TJ-301  FE-999301	I-Mab Biopharma Hong Kong Limited (Hong Kong, China) Ferring Pharmaceuticals (Saint-Prex, Switzerland)	Phase II, active ulcerative colitis (NCT03235752) Blood sampling, inflammatory bowel disease (NCT02790281)‡

^*Below shows alternative names

^†Selected most recent trials

^‡No phase information available.

IL-6: Interleukin-6.

4. Inconsistent outcomes of IL-6 inhibition from COVID-19 patients

SARS-CoV-2 infection causes injury to the lungs and other organs in part by cytokine activation and downstream pro-inflammatory networks, so called cytokine storm.^[65] This process is mediated by SARS-CoV-2 infection-induced pyroptosis of virus targeting macrophages and monocytes as well as by release of damage-associated molecular patterns (DAMPs) and, in some cell types, direct consequences of viral infection and release of pathogen-associated molecular patterns.^[66,67] One of the prime candidates for mediating inflammation following SARS-CoV-2 infection is IL-6 from macrophages or monocytes. The prototypical pro-inflammatory cytokine IL-1β and the prominent DAMP IL-1α both induce IL-6, markedly boosting its production. ClinicalTrials.gov currently lists 104 registered clinical trials of tocilizumab and 20 registered trials of sarilumab in COVID-19 patients, while there were only approximately 10 randomized controlled trials of tocilizumab or sarilumab listed in March 2021.^[68] The rapid growth in the number of clinical trials supports the strong interest in targeting IL-6 to control the cytokine storm in COVID-19 patients. However, not all trials targeting IL-6 have yielded positive results. SARS-CoV-2 replication peaks in the early days after infection, while the peak of the inflammatory response to SARS-CoV-2 often coincides with or shortly precedes clinical deterioration.^[69] The severity of disease, the timing of IL-6R blockade in relation to the course of clinical deterioration, the concurrent use of other drugs such as corticosteroids and/or antiviral agents, and elevated inflammatory markers vary among studies and may explain disparate outcomes.

4.1. IL-6 signaling in COVID-19 patients

Based on the IL-6 signaling pathways outlined in Figures 2B and 3A, increased levels of IL-6 and sIL-6R lead to enhanced IL-6 trans-signaling and inflammation, and this is what occurs in COVID-19 patients. In a cohort of 366 symptomatic COVID-19 patients, including patients with moderate symptoms (n=257), severe survivors from the intensive care unit (ICU) (n=40), and severe non-survivors (exitus, n=69),^[70] serum levels of IL-6, sIL-6R, and sgp130, and binary (IL-6:sIL-6R) to ternary (IL-6:sIL-6R:sgp130) complex (B/T complex) ratio were all markedly increased in severe survivors compared with those with moderate symptoms, but fold of the molar excess (FME) of sgp130 over sIL-6R was greatly decreased in severe survivors. In the exitus group, IL-6 and FME were increased, but sIL-6R, sgp130, and B/T complex ratio were decreased. Serum levels of sIL-6R correlated positively with sgp130 in all 3 groups of patients. In severe survivors, but not in exitus patients, serum levels of IL-6 correlated positively with sIL-6R. Area under receiver operating characteristic curves and univariate logistic analyses demonstrated the significant predictive value of IL-6, sIL-6R, sgp130, B/T complex ratio, and FME for COVID-19 symptom severity. Patients with IL-6 signaling variables above the cutoffs presented a significantly increased risk of developing severe COVID-19. Univariate logistic regression analysis also showed that these IL-6 signaling components were significant individual predictors of death. B/T complex ratio, IL-6, and FME remained as significant predictors of death in a multivariate analysis. All patients with levels of IL-6 and B/T complex ratio above the cutoffs and sIL-6R, sgp130, and FME below the cutoffs died in the first week of hospitalization. The lowest survival time corresponded to patients with increased IL-6 and decreased sIL-6R and sgp130.^[70] Mechanistic studies revealed that the spike protein on the surface of SARS-CoV-2 promoted angiotensin-II type 1 (AT1) receptor-mediated signaling, activated nuclear factor (NF)-κB and activator protein-1/c-fos via the mitogen-activated protein kinases pathway, and increased IL-6 release from epithelial cells. Upregulation of AT-1 signaling increased ADAM-17 expression and sIL-6R release. Moreover, the AT1 receptor antagonist (Candesartan cilexetil) downregulated IL-6 and sIL-6R release in cells expressing SARS-CoV-2 spike protein.^[71] Culture supernatant from spike protein-transfected epithelial cells contains IL-6 and sIL-6R, and such culture supernatant promoted IL-6 trans-signaling in TMNK-1 liver endothelial cells that express gp130, but not mIL-6R.^[71] Therefore, IL-6 contributes to the cytokine storm in COVID-19 patients using the IL-6 trans-signaling pathway.

4.2. Targeting IL-6R in COVID-19 patients

Critically ill COVID-19 patients often receive intensive care. The REMAP-CAP (Randomized, Embedded, Multifactorial Adaptive Platform Trial for Community-Acquired Pneumonia) was an open-label trial involving approximately 800 patients in need of respiratory or blood-pressure support or both, and who were assigned to anti-IL-6R antibodies tocilizumab (n=353), sarilumab (n=48), or placebo (n=402). The trial used a 24-h window for randomization after the initiation of organ support irrespective of the time of symptom onset. The primary outcomes were in-hospital death and days free of support to day 21. The in-hospital death rate was 28% in tocilizumab-treated patients and 22% in sarilumab-treated patients, compared with 36% in the placebo group.^[72] The median numbers of organ support-free days were 10 (Q1–Q3, −1–16), 11 (Q1–Q3, 0–16), and 0 (Q1–Q3, −1–15), respectively, and the adjusted cumulative odds ratio (OR) of organ support-free days were 1.64 (95% confidence interval (CI): 1.25–2.14) for tocilizumab treatment and 1.76 (95% CI: 1.17–2.91) for sarilumab treatment compared with placebo treatment.^[72] The RECOVERY (Randomized Evaluation of COVID-19 Therapy) trial^[73] – probably the largest randomized trial assessing the effect of tocilizumab in hospitalized COVID-19 patients – enrolled a total of 21,550 patients. Of these patients, 4,116 adults were included to assess the efficacy of tocilizumab treatment by assigning 2,094 patients for usual care, including use of systemic corticosteroids, and 2,022 patients for usual care plus tocilizumab. One dose of tocilizumab and a second dose 12–24 h later if necessary reduced the rate of death (31% vs. 35%) (rate ratio (RR)=0.85, P=0.0028) and the use of invasive mechanical ventilation (IMV) (35% vs. 42%, RR=0.84, P<0.0001), and increased the rate of hospital discharge within 28 d (57% vs. 50%, RR=1.22, P<0.0001). The subgroup that received corticosteroids appeared to benefit the most from tocilizumab; tocilizumab reduced the death rate from 39% in the non-corticosteroid group to 29% in the corticosteroid group.^[73] In the REMAP-CAP and RECOVERY trials, up to 20% of COVID-19 patients showed beneficial effects from IL-6 blockade if combined therapy of tocilizumab and dexamethasone was administered early after hospitalization.^[72,73]

The World Health Organization (WHO) Rapid Evidence Appraisal for COVID-19 Therapies (REACT) working group meta-analysis of 27 randomized trials of IL-6R antagonists, which included 10,930 patients with COVID-19, also supported a pathogenic role of IL-6 in SARS-CoV-2 infection. By 28 days, IL-6R antagonists reduced all-cause death in hospitalized COVID-19 patients (OR=0.86, 95% CI: 0.79–0.95, P=0.003 based on a fixed-effects meta-analysis). The absolute mortality risk with IL-6R antagonists was reduced to 22% vs. an assumed mortality risk of 25% for usual care or placebo. The use of tocilizumab had an OR of death of 0.83 (95%CI: 0.74–0.92, P<0.001) vs. an OR of death of 1.08 (95% CI: 0.86–1.36, P=0.52) for sarilumab. As in the RECOVERY trial, the mortality rates of patients who received corticosteroids were further reduced by IL-6R antagonists. The OR for the tocilizumab group reduced to 0.77 (95% CI: 0.68–0.87), while that for the sarilumab group reduced to 0.92 (95% CI: 0.61–1.38). The rate of IMV or death OR was 0.77 (95% CI: 0.70–0.85) for all IL-6 antagonists, 0.74 (95% CI: 0.66–0.82) for tocilizumab, and 1.00 (95% CI: 0.74–1.38) for sarilumab. Secondary bacterial and fungal infections by 28 d occurred in 21.9% of patients in the IL-6 antagonist group vs. 17.6% in the usual care or placebo group. The OR benefit of all-cause mortality in those who received glucocorticoids reduced to 0.78 from 1.09 among those who received no glucocorticoids by day 28 after randomization, although the 90-d mortality reduction was not affected.^[74] A similar systematic review and meta-analysis of 763 studies provided further support for a beneficial role of targeting IL-6R. This meta-analysis included 15 randomized controlled trials of 9,320 patients. IL-6R antagonist (tocilizumab or sarilumab) treatment reduced all-cause mortality to 24.4% vs. 28.3% in the control group from 13 studies (risk ratio (RR) =0.90, 95% CI: 0.84–0.96, P=0.003). IL-6R antagonists reduced the 28/30-d mortality and intubation rate without significant increase of adverse events or secondary infection.^[75]

4.3. Use of corticosteroids and anti-IL-6R therapy efficacy

Not all studies of IL-6R antagonists yielded the same favorable results. The COVACTA trial (NCT04320615) was a randomized, double-blind, and placebo-controlled trial that enrolled 452 patients with COVID-19. Patients were assigned 2:1 to receive one dose of tocilizumab or placebo. The outcomes included day 28 clinical status and mortality. The use of tocilizumab did not affect patient mortality rates (19.7% on tocilizumab vs. 19.4% on placebo).^[76] One explanation for these negative results is the rate of glucocorticoid use. Only a minority of patients in the COVACTA trial received glucocorticoids: 19.4% in the tocilizumab group and 28.5% in the placebo group. In contrast, 93% of patients in the REMAP-CAP trial and 82% of patients in the RECOVERY trial received glucocorticoids.^[77] Other smaller randomized controlled trials also yielded negative results. In one report of 2 trials, one trial included 46 patients who received usual care and 51 patients who received usual care plus tocilizumab, while the second separate trial contained 41 patients who received usual care and 50 patients who received usual care plus sarilumab. At 14 days, these groups in both studies showed no differences in need for non-invasive ventilation, mechanical ventilation, or death.^[78] These negative results may again be due to the low rate of corticosteroid or anti-viral drug usage. In both trial protocols, only <5% of patients received corticosteroids or anti-viral drugs.^[78] Collectively, targeting IL-6 activity appears to reduce morbidity and mortality rates among COVID-19 patients. The use of corticosteroids may boost the therapeutic efficacy of IL-6R antagonists, a hypothesis that merits further exploration. Current guidance from the United States Food and Drug Administration (FDA) and the Centers for Disease Control and Prevention (CDC) supports the use of tocilizumab in selected patients with advanced COVID-19. To enhance the efficacy of anti-IL-6R therapy, a combined administration of corticosteroids and IL-6R antagonists may become routine in COVID-19 patients.

4.4. The role of corticosteroids in COVID-19 patients

Corticosteroids are important therapeutic agents to treat allergic and inflammatory disorders or to suppress undesirable or inappropriate immune responses. Corticosteroids affect numerous steps in the inflammatory pathway and are probably the most effective regimens to reduce morbidity and mortality in asthmatic patients.^[79] This is probably why corticosteroid use enhanced the therapeutic efficacies of IL-6R antagonists in reducing COVID-19 patient morbidity and mortality. In addition to their anti-inflammatory effects, corticosteroids are associated with beneficial effects on β₂-adrenergic receptors (β2-AR)^[80] by upregulating β2-AR expression and function and reversing β2-AR downregulation from chronic β₂-adrenergic therapies.^[81]

β2-AR are expressed on all SARS-CoV-2 infection-associated cells, including airway epithelial cells and immune cells such as macrophages, dendritic cells, and B and T lymphocytes that are located in the lungs, gastrointestinal tract, liver, uterus, and vascular smooth muscle.^[82–84] In asthmatic patients, the combined use of inhaled corticosteroids and long-acting β2-agonist showed synergistic benefit.^[85] It is possible that SARS-CoV-2 infection shares similar pathology to asthma and involves the function of β2-AR. After SARS-CoV-2 infection, Th1 cells are primary responders to clear the virus. An elevated cytokine storm triggers Th2 cells, leading to a poor prognosis.^[86,87] Patients with COVID-19 exhibited a significant reduction of Th1 and Th17 cells but elevated numbers of activated Th2 cells that respond to the virus.^[88] Deceased COVID-19 patients had more senescent Th2 cells compared with those of survivors. The percentage of such Th2 cells is a risk factor of lymphocyte loss.^[88] Although not tested and currently lacking clinical evidence, combined use of IL-6R antagonists, corticosteroids, and β2-AR agonists may yield unexpected synergistic benefits.

5. Differential roles of IL-6 in metabolic diseases

5.1. Metabolic diseases and morbidity and mortality in patients with COVID-19

Co-morbidities such as obesity, metabolic syndrome, CVD, and other inflammatory diseases are associated with worse outcomes following SARS-CoV-2 infection. Clinical and preclinical evidence suggest that targeting IL-6 or its receptor benefits COVID-19 patients and those with CVD. However, the same strategy yielded mixed results in patients or experimental models of metabolic diseases. Different IL-6 signaling pathways [Figure 2] and disease stages may affect the outcomes of IL-6 targeting therapy.

Obesity and diabetes aggravate outcomes in COVID-19 patients, even those infected with the less virulent Omicron strains. A recent study of 118,078 COVID-19 patients from the Kaiser Permanente Northern California cohort included 48,101 cases due to Omicron and 69,977 due to Delta infection.^[89] Although the adjusted hazard ratios (HRadj) of hospitalization (HRadj=0.55), low-flow oxygen support (HRadj=0.46), high-flow oxygen support (HRadj=0.47), IMV (HRadj=0.43), and death (HRadj=0.54) were much lower in Omicron cases than in Delta cases, diabetes increased the risk of hospitalization (HRadj=1.22), low-flow oxygen support (HRadj=1.30), high-flow oxygen support (HRadj=1.37), IMV (HRadj=1.48), and death (HRadj=1.15). Body mass index (BMI) changes from <25 kg/mm² to ≥40 kg/mm² increased the risk of hospitalization (HRadj from 1.0 to 2.54), low-flow oxygen support (HRadj from 1.0 to 2.96), high-flow oxygen support (HRadj from 1.0 to 4.53), IMV (HRadj from 1.0 to 4.11), and death (HRadj from 1.0 to 2.37).^[89] Patients with diabetes or obesity showed higher susceptibility to SARS-CoV-2 infection compared with those without these conditions, and SARS-CoV-2 infection increased the risk of hospitalization, intensive care, and death among patients with type 1 diabetes (T1D), type 2 diabetes (T2D), and obesity. Earlier studies from more aggressive strains of SARS-CoV-2 variants revealed prolonged persistence of SARS-CoV-2 and a high mortality rate in T2D patients.^[90] Dysregulated glucose control was associated with poor outcomes for hospitalized COVID-19 patients in a Glytec (Waltham, Massachusetts, United States) database.^[91]

Observations from patients of multiple countries unequivocally support the adverse impacts of obesity and/or diabetes on the outcomes of COVID-19 patients. A retrospective study of 1,637 adult patients from Wuhan, China, between 4 February and 23 March 2020 and followed up until 31 March 2020, showed that obesity was associated with higher odds of severe pneumonia (adjusted odds ratio, ORadj=1.47, P=0.002) and oxygen therapy (ORadj=1.40, P=0.007) after adjusting for age, sex, and comorbidities. In those with diabetes, being overweight (ORadj=1.68, P=0.014) or obese (ORadj=2.06, P=0.028) increased the odds of in-hospital oxygen therapy.^[92] In a French study of 134,209 COVID-19 inpatients from February to September 2020, including 13,596 inpatients who underwent IMV and 19,969 inpatients who died, death occurred more frequently in patients with obesity (ORadj=1.2) and diabetes (ORadj=1.2), and IMV was also more frequently necessary in patients with obesity (ORadj=1.9), diabetes (ORadj=1.4), and hypertension (ORadj=1.7).^[93] In a prospective community-based cohort study of 6,910,965 individuals from the UK QResearch database between 24 January and 30 April 2020, 13,503 COVID-19 patients were admitted to hospital, among which 1,601 patients were admitted to ICU, and 5,479 patients died due to SARS-CoV-2 infection. BMI and T2D both associated with hospitalization (HRadj=1.05 and 1.72), ICU admission (HRadj=1.10 and 2.32), and death (HRadj=1.04 and 1.66) after adjusting for age, sex, demographic factors, smoking, non-obesity-related mortality, and obesity-related mortality.^[94] In a large-scale general population study of 334,329 community-dwelling samples in England from 16 March to 26 April 2020, a total of 640 patients warranted hospitalization. COVID-19 hospitalization correlated with increased BMI. The risk of COVID-19-associated hospitalization increased from overweight (OR=1.39) to obesity stage I (BMI 30–35 kg/mm², OR=1.70) and to obesity stage II (BMI >35 kg/mm², OR=3.38).^[95] Studies from the United States yielded the same conclusion. A cohort of 28,095 COVID-19 patients from the US COVID-19 Research Database from 1 January to 20 November 2020 included 11,294 obese patients and 4,445 T2D patients. Cox proportional hazard model showed that age greater than 65 years (HR=4.188, P<0.001), T2D (HR=2.378, P<0.001), and obesity (HR=1.412, P=0.001) each associated with the risk of hospitalization. Age greater than 65 years (HR=3.993, P<0.001), T2D (HR=2.873, P<0.001), and obesity (HR=1.871, P=0.002) were also risk factors for needing critical care. These risks were more prominent in White and Black patients than in Asians. However, age greater than 65 years was the only significant risk factor for hospitalization in Asian patients (HR=11.746, P=0.004).^[96] A meta-analysis of 75 studies selected from 1,733 studies also supports the deleterious impact of obesity on COVID-19-associated organ failure. This meta-analysis contained 399,461 COVID-19 patients from 10 countries in Asia, Europe, and North and South America between January and June 2022. Of the 75 studies, 20 assessed the association of obesity with the risk for SARS-CoV-2 infection and associated outcomes. Eighteen of these 20 studies reported that patients with obesity had an increased risk of SARS-CoV-2 infection (OR=1.46, P<0.0001), hospitalization (OR=2.13, P<0.0001), ICU admission (OR=1.68, P<0.0001), IMV use (OR=1.66, P<0.0001), or death (OR=1.48, P<0.001).^[97]

There may be various mechanisms underlying the greater susceptibility to and poorer outcomes from SARS-CoV-2 infection in obese and diabetic individuals. Angiotensin-converting enzyme 2 (ACE2) is the cell surface protein that binds the SARS-CoV-2 spike (S) protein, and transmembrane serine protease 2 (TMPRSS2), ADAM10, and ADAM17, together with cathepsin L are proteases that cleave the S protein for viral entry and replication.^[98,99] High expression of ACE2, TMPRSS2, cathepsin L, ADAM10, and ADAM17 in the nasal epithelium ciliated and mucus-secreting goblet cells,^[100] and all major epithelial cells in the small airway, large airway, and trachea^[101–103] makes these target tissues sensitive to SARS-CoV-2 infection. In the human pancreas, most ACE2 is localized in the microvasculature, CD31⁺ capillaries, and pericytes, whereas TMPRSS2 and ADAM17 are situated mainly in the exocrine pancreas within the duct. Single-molecule fluorescent in situ hybridization showed that pancreatic ACE2, ADAM17, and TMPRSS2 localized in acinar cells, ducts, and CD34⁺ endothelial cells, but not in islet endocrine cells in normal non-diabetic donors.^[104,105] Consistent with this observation, SARS-CoV-2 nucleocapsid protein was detected in ducts but not in islets from COVID-19 patients.^[104] ACE2 expression was increased in the bronchial washings from obese patients.^[106] Expression of ACE2, ADAM17, and TMPRSS2 to a lesser extent was elevated in the liver of patients with T2D and correlated with hepatic fat.^[107,108] These expression patterns of ACE2, ADAM17, and TMPRSS2 may increase sensitivity to SARS-CoV-2 infection that consequently damage these targeting cells and associated endocrine and hepatic systems. Minimal information regarding ACE2 and TMPRSS2 expression in adipocytes is available. Most studies show an increase of ACE2, ADAM17, and cathepsin L expression in white adipose tissues from obese patients, which might enhance susceptibility to SARS-CoV-2 infection,^[109–111] while one study reported reduced ACE2 expression in subcutaneous white adipose tissue from obese and diabetic patients compared with that in normal individuals.^[112]

5.2. IL-6 sources and targets in metabolic diseases

Distinct from the pathogenic roles of IL-6 in inflamed tissues from patients with CAD, RA, CKD, organ transplantation, and SLE, some functional studies of IL-6 in metabolic diseases suggest that IL-6 may exert beneficial effects. In patients with obesity and T2D, adipose tissues are a major source of elevated plasma IL-6 that rises as high as 2–3 pg/mL.^[113,114] Skeletal muscles also release IL-6 during exercise.^[115] Muscle-derived IL-6 inhibits TNF-α and IL-1β expression, and protects against the risk of TNF-α- and IL-1β-induced insulin resistance.^[116] Some researchers have attributed the benefit of exercise on obesity, diabetes, and CVD to such effects of IL-6 on metabolism.^[117] Mechanistically, IL-6 targets hepatocytes, skeletal muscle cells, and adipocytes to regulate peripheral insulin sensitivity and glucose homeostasis [Figure 4A]. High concentrations of IL-6 in muscle in postabsorptive healthy men infused with recombinant human IL-6 activated the pathways that facilitated energy turnover, enhanced insulin sensitivity, and promoted skeletal muscle lipolysis and systemic fatty acid oxidation.^[118] Intravenous infusion of IL-6 into healthy volunteers increased adipose tissue and splanchnic tissue uptakes of fatty acid, glycerol, and lactate and promoted adipose tissue lipolysis. IL-6 infusion also enhanced splanchnic lipid and carbohydrate oxidation rates.^[119] In cultured primary mouse hepatocytes and human hepatocarcinoma cell HepG2, IL-6 inhibited insulin receptor signaling with decreased insulin receptor substrate 1 (IRS-1) total Tyr phosphorylation and decreased association between phosphoinositide 3-kinases and IRS-1 in response to physiological insulin. Moreover, IL-6 inhibited insulin-dependent AKT (protein kinase B) activation.^[120] Intracerebroventricular injection of IL-6 reduced energy intake and glucose homeostasis in obese mice via enhanced IL-6 trans-signaling in the hypothalamus or the forebrain.^[121] IL-6 infusion in healthy humans increased skeletal muscle fatty acid oxidation, basal and insulin-stimulated glucose uptake, and glucose transporter GLUT4 plasma membrane translocation.^[122]

Figure 4:

Differential roles of IL-6 in pancreatic islet, skeletal muscle, and liver. (A) Adipocyte- and skeletal muscle cell-derived IL-6 plays a beneficial role in pancreatic α cells and β cells, skeletal muscle cells, and hepatocytes. (B) Adipocyte-derived IL-6 may also have a detrimental role in hepatocytes. AKT: Protein kinase B; AMPK: AMP-activated protein kinase; GIP1: Gastric inhibitory polypeptide; GLP1: Glucagon-like peptide-1; GLUT4: Glucose transporter type-4; IL-6: Interleukin-6; IRS-1: Insulin receptor substrate-1; PC1/3: Prohormone convertase 1/3; PPAR-γ: Peroxisome proliferator-activated receptor-γ; SOCS: Suppressor of cytokine signaling; UCP1: Uncoupling protein-1.

5.3. IL-6 activity may mute some adverse effects of obesity and diabetes

Some results from humans, animal experiments, and cultured cells support a beneficial role of IL-6 in obesity- and diabetes-associated events. Consequently, use of anti-IL-6 antibodies increased bodyweight gain.^[123] Similarly, IL-6R inhibition with tocilizumab increased bodyweight gain and serum triglyceride and total cholesterol levels in humans.^[124] In a multicenter, double-blind trial, T2D and obese patients received intravenous tocilizumab (n=19), oral dipeptidyl peptidase-4 inhibitor stiaglipin (n=17), or placebo (n=16) during a 12-week training intervention. While stiaglipin elevated glucagon-like peptide 1 (GLP1) by 53% (P=0.004), tocilizumab reduced GLP1 by 26% (P=0.034).^[125] Agents such as ziltivekimab that target IL-6 ligand rather than the receptor appear to have less adverse metabolic effects. In mice, IL-6 improved β-cell function and glucose homeostasis by augmenting muscle IL-6 and GLP1 release during exercise. In mice or ex vivo human islets, IL-6 reduced oxidative stress,^[126] which is responsible for the demise of pancreatic β cells in T1D patients. Expression of IL-6 from pancreatic α and β cells^[127] diminished in insulin-deficient donor islets.^[128]

In humans, IL-6 increases fatty acid oxidation and GLUT4 translocation to the plasma membrane to promote glucose uptake in skeletal muscle via AMP-activated protein kinase activation.^[122,129] In rats, infusion of IL-6 increased basal insulin sensitivity, improved glucose tolerance, and enhanced fat oxidation through the activation of peroxisome proliferator-activated receptor-γ and uncoupling protein in skeletal muscle, accounting for 90% of insulin-stimulated glucose uptake.^[130] In the pancreas, IL-6R is expressed in α cells as well as in β cells, although to a lesser extent.^[26] Elevated plasma IL-6 from contracting skeletal muscle or white adipose tissues increases α-cell prohormone convertase 1/3 expression, leading to peptide production shifting from glucagon to GLP1. Like in the brain,^[121] IL-6 uses the classic-signaling pathway to induce insulin secretion from the β cells, resulting in improved glucose homeostasis.^[26,131] Therefore, IL-6-deficient Il6^−/− mice developed obesity with interrupted glucose and lipid metabolism^[10] and reduced pancreatic GLP1 production.^[26] Mice with hepatocyte-selective depletion of IL-6Rα showed reduced insulin sensitivity and glucose tolerance, which can be stored by TNF-α neutralization or Kupffer cell depletion, suggesting a protective role of hepatic IL-6 signaling.^[12] IL-6 overexpression in obese mice reduced fat mass and fat pad size.^[132] In these obese mice, IL-6 controls glucose concentration by enhancing pancreatic α-cell activity to assist islet insulin secretion via production of GLP1 and glucose-dependent insulinotropic peptide [Figure 4A].^[131,133] Neutralization of IL-6 in wild-type mice fed a HFD or in diabetic db/db mice impaired α-cell function and glucose homeostasis. Increased IL-6 in the circulation promoted pancreatic α-cell proliferation in response to a HFD.^[26] Subcutaneous injection of IL-6 improved several measures of nerve functions in a rat diabetic peripheral neuropathy streptozotocin model.^[134] All these observations from humans, animal experiments, and cultured cells point to a potentially beneficial role for IL-6 in metabolic diseases [Figure 4A].

5.4. IL-6 activity promotes obesity, diabetes, and associated organ dysfunction

Not all studies support a potential beneficial role for IL-6 in obesity and diabetes. Patients with T1D may have elevated^[135,136] or normal^[137–139] circulating IL-6. In fasted healthy individuals, adipose tissue contributes to circulating IL-6 concentrations that are correlated with BMI.^[140] Adipose tissue IL-6 levels but not TNF-α or leptin levels correlate with insulin resistance in obese subjects.^[141] The liver is a key target of IL-6. In the liver, IL-6 reduces hepatocyte insulin sensitivity by interfering with insulin signaling,^[142] inhibits insulin-dependent glycogen synthesis, and promotes glucose release from hepatocytes.^{[120,143,144]} IL-6 regulates auto-aggressive T effector cells and Th17 cells, and inhibits Tregs [Figure 4B].^[137] Monocytes from T1D patients exhibit increased IL-6 expression^[145] and trans-signaling.^[137] Systemic chronic inflammation causes insulin resistance and contributes pivotally to the development of T2D.^[146–149] T2D patients and obese people without diabetes show elevated plasma hsCRP and IL-6,^[150,151] in part due to excess adipose tissue, although some studies suggest that the relationship between hsCRP and T2D is independent of insulin resistance and BMI.^[152,153] These observations do not support a beneficial role of IL-6 in obesity or diabetes. Indeed, HFD caused more weight loss in Il6^−/− mice than in wild-type mice.^[154] IL-6 injection increased plasma glucose and insulin levels and decreased liver glycogen.^[155] In addition, mice exposed to IL-6 for 90 min showed reduced insulin signaling in the liver.^[156] Minipump-mediated IL-6 infusion for 5–7 d induced hepatic insulin resistance with impaired insulin receptor signaling without affecting skeletal muscle insulin responses.^[157] Regarding liver regeneration, short-term (1–2 d) IL-6 exposure was initially protective, but long-term (5–7 d) IL-6 exposure proved injurious to the liver.^[158] In RA patients, IL-6 inhibition with tocilizumab improved insulin resistance.^[159] Furthermore, in a multicenter, randomized, placebo-controlled, double-blinded trial of 52 T1D patients (screened within 100 d), 7 months of IL-6R inhibition with tocilizumab (tocilizumab vs. placebo 2:1) reduced T-cell IL-6R downstream signaling as expected, but did not change the numbers of Th17 cells, Treg cells, and CD4⁺ T effector cells, or slow residual β-cell function.^[160] These studies from rodents and humans support that IL-6 has a detrimental role or no role in obesity and diabetes. Clinical trials of IL-6 inhibition should monitor metabolic variables carefully to assess the net in vivo significance in humans of these various observations.

The mechanisms of aforesaid differences remain unknown. Tissue type difference and disease stages may explain some of these observations. IL-6 classic-signaling in T cells is instrumental in the early phase of obesity in HFD-fed mice, but trans-signaling seems more important in the late phase.^[161] IL-6 trans-signaling in hypothalamic neurons controls the feeding and systemic glucose homeostasis in HFD-fed mice.^[121] Therefore, the time of anti-IL-6 intervention is crucial. Blocking trans-signaling with sgp130 [Figure 3A] in HFD-fed mice reduced adipose tissue macrophage accumulation and improved insulin resistance.^[162] IL-6 may directly interact with IRS-1, which is important in insulin signaling, and this activity of IL-6 is tissue-specific. IL-6 induced immunocomplex formation between IRS-1 with IL-6R or gp130, leading to p-Tyr⁷⁰⁵ phosphorylation of IRS-1 in C2C12 human myoblast cells. This IL-6 activity induced a rapid p-Ser³¹⁸ phosphorylation of IRS-1 in skeletal muscle cells and muscle tissue, but not in the livers of IL-6-treated mice. Use of a p-Ser³¹⁸ antibody demonstrated that p-Ser³¹⁸-IRS-1 phosphorylation improved insulin-stimulated AKT phosphorylation and glucose uptake [Figure 4A]. In contrast, use of a p-Ser³⁰⁷ antibody showed that p-Ser³⁰⁷-IRS-1 phosphorylation and induction of suppressor of cytokine signaling-3 (SOCS-3) expression only occurred in the liver and not in the muscle of IL-6-treated mice.^[163] Increased expression of SOCS-3 is known to block skeletal muscle activation of IRS-1 and AKT and glucose uptake, and causes insulin resistance [Figure 4B].^[164,165] Therefore, IL-6 activity in regulating glucose metabolism and insulin signaling can be cell type-specific.

5.5. Inconsistent clinical evidence

Multiple studies have established that blood IL-6 concentration is an independent predictor of T2D and CVD and correlates with insulin resistance. Circulating levels of IL-6 are 2–4-fold higher in obese or T2D patients compared with those in non-obese controls.^[166–168] From a mixed study of obese (n=12), diabetic (n=7), and non-diabetic patients (n=5), adipose tissue IL-6 levels inversely correlated with the maximal insulin glucose transport rate (P<0.02) and glucose infusion rate (P<0.02) during the hyperinsulinemic normoglycemic clamp and in isolated adipocytes (P<0.02).^[141] In the Women’s Health Study of a nationwide cohort of 27,628 women without diabetes mellitus, CVD, or cancer, 188 women developed diabetes mellitus after 4 years of follow-up. Baseline levels of IL-6 (P<0.001) and hsCRP (P<0.001) were both higher in the cases than in 362 age-matched controls. A conditional logistic regression test showed that the relative risk of developing future diabetes for women in the highest quartile vs. the lowest quartile of IL-6 and hsCRP was 7.5 (P<0.001) and 15.7 (P<0.001). These associations persisted after adjustment for BMI (relative risk=2.9, P=0.008), but not after adjustment for BMI, family history of diabetes, smoking, exercise, alcohol use, and hormone replacement therapy (relative risk =2.3, P=0.07).^[113] A study of 54 patients with autoimmune diabetes and T1D, 70 first-degree relatives, and 60 healthy controls showed that diabetic patients had significantly higher plasma IL-6 and homeostatic model assessment-insulin resistance and lower estimated glucose disposal rate compared with those of first-degree relatives and healthy controls (all P<0.001).^[169] A recent meta-analysis of 15 prospective studies, including 5,421 T2D patients and 31,562 non-T2D controls showed that high blood IL-6 levels were associated with a higher risk of T2D (HR=1.24, 95%CI 1.17–1.32, P=1×10⁻¹²), suggesting that IL-6-mediated inflammation was involved in the etiology of T2D.^[170]

As with the results from preclinical studies, clinical investigations also produced inconsistent results regarding the role of IL-6 in metabolic diseases. Acute administration of IL-6 (2 h) in 8 healthy fasting males increased systemic fatty acid oxidation, which was followed by an increase in systemic lipolysis with elevated unidirectional fatty acid and glycerol release from the skeletal muscle, although acute IL-6 administration did not affect glucose metabolism.^[118] Similarly, in healthy volunteers or T2D patients, IL-6 infusion increased lipolysis and reduced circulating insulin without affecting glycemia or insulin-stimulated glucose metabolism.^[171–173] IL-6 signaling following exercise increased insulin sensitivity.^[174] However, some studies support a detrimental role for IL-6 in diabetes. In a placebo-controlled phase III study of 184 diabetic patients and 1,928 non-diabetic patients, IL-6 signaling inhibition by subcutaneous administration of the anti-IL-6R antibody sarilimab alone (150–200 mg) or together with the conventional synthetic disease-modifying anti-rheumatic drugs markedly reduced Hb1Ac in patients with diabetes or baseline Hb1Ac >7% compared with those who received placebo or anti-rheumatic drug alone (methotrexate or adalimumab).^[175] Inconsistent results were also obtained from new-onset T1D patients treated with tocilizumab. In a multicenter, randomized, placebo-controlled, double-blind trial involving 81 children and 55 adults (NCT02293837), 7 monthly doses of tocilizumab did not slow the loss of β cells in these patients.^[176] Overall, low levels of circulating IL-6 (2–3 pg/mL) are associated with an increased risk of developing diabetes.^[113,177] Thus, in some circumstances, IL-6 seems to mitigate diabetes and obesity, although there are few in-depth mechanistic studies.

6. Targeting IL-6 in cardiovascular diseases

We have previously reviewed the interactions between SARS-CoV-2 infection and the risk of CVD.^[178] As in metabolic diseases, IL-6 served as a biomarker in patients with CVD. A study of 4,939 patients from the SOLID-TIMI 52 (Stabilization of Plaque Using Darapladib-Thrombolysis in Myocardial Infarction 52) trial with a median of 2.5 years of follow-up for cardiovascular events showed that an increase in plasma IL-6 concentration was a significant risk for major adverse cardiovascular events (MACE) (HRadj=1.10) and cardiovascular death or heart failure (HRadj=1.22) after adjusting for baseline characteristics and clinical predictors. Thus, serum IL-6 associated with poor cardiovascular outcomes.^[22,179,180] A similar study was conducted on 3,489 patients with unstable CAD, of which half did not undergo an early invasive treatment strategy. In the non-invasive group of patients, high plasma IL-6 levels greatly increased patient mortality, compared with those who received invasive treatment (7.9% vs. 2.3%, relative risk=3.47).^[181] Human abdominal aortic aneurysm (AAA) studies revealed that these patients also had high levels of blood IL-6.^[182,183] A long-term prospective study showed that elevated blood IL-6 levels independently associated with future AAA.^[184] Our early study of 113 patients without aortic dilation indicated that indexed aortic diameter was positively associated with serum IL-6 levels (r=0.285, P=0.002) in a linear regression analysis, but not with any of the serum lipid levels. In a multivariate regression analysis adjusted for age, hypertension, diabetes, smoking, history of myocardial infarction or angina, and lipid and non-lipid serum measurements, plasma IL-6 levels remained significantly correlated with indexed aortic diameter (P=0.02).^[185] These clinical studies suggest a pathogenic role of IL-6 in human CVD.

As noted above, in the RESCUE trial, the anti-IL-6 antibody ziltivekimab showed dose-dependent reductions of plasma hsCRP in patients with moderate to severe CKD.^[53] In the ASSAIL-MI (ASSessing the effect of Anti-IL-6 treatment in Myocardial Infarction) study involving 199 patients with ST segment elevation myocardial infarction, the anti-IL-6R antibody tocilizumab (280 mg intravenous vs. placebo) given within 6 h of hospital admission modestly but significantly increased the myocardial salvage index on magnetic resonance imaging at 3–6 d after the events, and reduced patient microvascular obstruction (P=0.03) and blood hsCRP levels (P<0.001) at 6 months after the events. However, tocilizumab did not affect plasma N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels, infarct sizes, or end-diastolic volume at 6 months after the events.^[186] In a small, two-center, double-blind, placebo-controlled phase II trial of 117 non-ST-elevation myocardial infarction patients, tocilizumab treatment at a median of 2 d after symptom onset significantly reduced plasma hsCRP, high-sensitivity troponin T, NT-proBNP, and cholesterol levels within 2–3 d.^[187] The large-scale ZEUS-CVOT (Effects of Ziltivekimab versus Placebo on Cardiovascular Outcomes in Participants with Established Atherosclerotic Cardiovascular Disease, Chronic Kidney Disease and Systemic Inflammation) trial, including 6,200 patients with CKD and elevated hsCRP was initiated in 2021.^[188] This trial is testing whether the anti-IL-6 antibody ziltivekimab reduces the risk of MACE in patients at high cardiovascular risk, the risk of heart failure that leads to hospitalization or urgent care, and CKD-related outcomes.

7. Conclusions

More than 3 years after SARS-CoV-2 was first reported, we still do not completely understand how this virus provokes organ damage, and to what extent and why patients with various immunological diseases respond differently from healthy individuals. This review briefly considered a few common SARS-CoV-2 infection-associated diseases and focused on the role of IL-6 in the pathogenesis and protection of metabolic and cardiovascular diseases. IL-6 is only one member of the large pool of cytokines elicited by SARS-CoV-2 infection, but this molecule has already produced many puzzles for us to resolve. First, there is an apparent additional beneficial effect of corticosteroids when co-administered with IL-6 antagonist therapy in COVID-19 patients. The mechanism of this interaction remains to be elucidated. Glucocorticoids may favor IL-6 toward one of the signaling pathways to block SARS-CoV-2 entry or replication. Prior studies showed that glucocorticoids directly inhibit IL-6 signaling^[189] or interfere with IL-6-induced expression of suppressor of cytokine signaling 3, an IL-6 signaling feedback inhibitor that regulates the pro-inflammatory and anti-inflammatory activities of IL-6.^[190] Second, increased expression of genes that mediate SARS-CoV-2 binding to and entry into target organs, such as the respiratory epithelium, pancreas, and liver, may partially why COVID-19 patients with co-morbid metabolic and cardiovascular diseases experienced an elevated risk of hospitalization and death. Moreover, IL-6 functions in these diseases may extend beyond those of a simple pro-inflammatory cytokine. IL-6 functions can be organ-specific or cell type-specific. The role of IL-6 in IRS-1 activation in the pancreas, muscle, and liver is more like that of a hormone. How IL-6 signaling impacts IRS-1 activation remains unclear. Third, most preclinical studies and clinical trials suggest a detrimental role of IL-6 in CVD. Anti-IL-6 or anti-IL-6R antibodies showed therapeutic efficacy in these patients. Targeting IL-6 in many other diseases, such as cancers, inflammatory bowel disease (IBD), and autoimmune diseases, also confirmed a pro-inflammatory role of this cytokine. For example, in mice, IL-6 blockade improved tumor control, increased the density of CD4⁺ and CD8⁺ T effector cells, and reduced the Th17 population, macrophages, and myeloid cells. In mouse experimental autoimmune encephalomyelitis (EAE) with tumors, IL-6 blockade together with immune checkpoint blockade (ICB) enhanced tumor rejection and improved EAE symptoms compared with ICB alone.^[191] In a 12-week open-label, prospective phase IIa trial (the FUTURE trial), 16 patients with active IBD received the trans-signaling inhibitor olamkicept (sgp130Fc). This drug induced a clinical response in 44% of these patients and clinical remission in 19% of these patients, with reduced IL-6R downstream STAT3 phosphorylation and transcriptional change of the inflamed mucosa.^[192] In a study of 2,869 patients with RA and COVID-19 from the Global Rheumatology Alliance physician registry, patients treated with the anti-CD20 antibody rituximab or JAK inhibitors (tofacitinib or upadacitinib) were more likely to have had a severe illness requiring hospitalization compared with those taking TNF inhibitors (infliximab, etanercept, adalimumab, or golimumab) or IL-6R antagonist (tocilizumab). Rituximab, tofacitinib, and upadacitinib may impair the ability of the immune system to fight SARS-CoV-2 by lowering B cell functions.^[193] Additional information is required to determine whether patients with CVD, RA, cancer, IBD, or other autoimmune diseases may develop unwanted actions on the endocrine system, such as changes in blood insulin level, insulin signaling, insulin sensitivity, etc. Fourth, IL-6 has at least 3 signaling pathways, and it remains unknown why IL-6-induced gp130 engagement on the plasma membrane from different target cells showed different actions on downstream IRS-1 activation (Tyr and Ser phosphorylation). However, this finding may explain why IL-6 effects on fatty acid oxidation, glucose uptake, insulin signaling and sensitivity, and cell proliferation in skeletal muscles differed from those in the liver, and it may be due to more than the simple explanation of different IL-6 signaling pathways. The range of anti-IL-6 strategies including anti-IL-6, anti-IL-6R, and the selective trans-signaling blocker sgp130Fc merits continued exploration in different human diseases.