Interleukin-1β, heat shock protein 90α, and hypocretin-1 in chronic fatigue

Abstract

Background: Fatigue, defined as an overwhelming sense of tiredness, lack of energy, and feeling of exhaustion, is a phenomenon many people have experienced in connection with infections such as influenza, Epstein-Barr virus, etc. Fatigue is also common in cancer, neurological conditions like multiple sclerosis, Parkinson’s disease, and in chronic inflammatory and autoimmune diseases such as rheumatoid arthritis, psoriasis, and others. “Sickness behavior” observed in animals is a conceptual model for understanding fatigue. In this model, infection or tissue damage is followed by behavioral changes like social withdrawal, inactivity, sleepiness, fatigue, and reduced food and water intake. The proinflammatory cytokine interleukin (IL)-1β produced during activation of innate immune cells has a prominent role in mediating this behavior. IL-1β crosses the blood-brain barrier and in the brain IL-1β amplifies its own signaling by inducing microglia to produce IL-1β. In cerebral neurons IL-1β signals through a receptor complex including interleukin-1 receptor I (IL-1RI) and an alternative IL-1 receptor accessory protein that does not mediate inflammation but induce neuronal activation and sickness behavior. The inflammatory response needs to be controlled and is therefore downregulated in a timely manner not to run rampant. In addition, cellular protection mechanisms are activated during inflammation and tissue damage to preserve cellular life from reactive molecules that kill pathogens. Some variants of heat shock proteins (HSPs) released into the extracellular space could represent a defense mechanism of cellular life that also influence fatigue mechanisms. In addition, sleepiness and weariness are closely related to fatigue and part of the sickness behavior response. Inflammation can alter sleep patterns. The master regulator of sleep- and wakefulness, neuropeptide hypocretin-1 (Hcrt1), could therefore have a role in fatigue generation. Objectives: I) Investigate if mechanisms that protect cellular life and homeostasis are involved in the generation of fatigue. II) Develop a non-radioactive, sensitive and selective method for measurement of Hcrt1 in cerebrospinal fluid (CSF). III) Explore how IL-1β and other selected molecules interact in generation of fatigue, and to investigate a possible link between the neuropeptide Hcrt1 and fatigue. Methods: To explore mechanism of fatigue, a cohort of 71 patients with primary Sjögren’s syndrome were investigated. CSF samples where available from 49 patients. A method based on liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) was developed for measurements of Hcrt1. Hcrt1 was measured in CSF samples from 22 healthy subjects and 9 patients with narcolepsy type 1. The clinical variables fatigue, depression and pain were scored using the fatigue Visual Analogue Scale (fVAS), Beck Depression Inventory, and the pain item of the Medical Outcome Survey short form 36, respectively. ELISAs were used to measure HSP32, -60, -72, and -90α in plasma, and in CSF to measure concentrations of IL-1Ra, IL-1RII, IL-6, and the calcium binding protein S100B. Hcrt1 in CSF was measured using a radioimmunoassay (RIA) method, in addition to a non-radioactive method based on liquid chromatography coupled with tandem mass spectrometry. Results were analyzed by non-parametric group comparisons, logistic regression, univariate- and multiple regression, and principal component analysis (PCA). Results: Measures of HSP32, -60, -72, and -90α in plasma revealed that the concentrations of HSP90α were significantly higher in pSS patients with high fatigue versus low fatigue. A tendency toward higher concentrations of HSP72 was observed in patients with high fatigue compared to patients with low fatigue. The LC-MS/MS method for Hcrt1 in CSF revealed much lower concentrations in healthy subjects than what has previously been published. Patients with narcolepsy type 1, a sleep disorder characterized by low levels of Hcrt1 in CSF, also had lower levels of Hcrt1 in CSF compared to previous published studies. The LC-MS/MS method was compared to the commonly used RIA method. A Bland-Altman plot showed agreement between the two methods. Analysis of IL-1β related proteins (IL-1Ra, IL-1RII, and S100B), IL-6, and Hcrt1 in CSF demonstrated that IL-1Ra showed significant association with fVAS scores together with the clinical variables BDI scores and pain scores. The relationship of the biochemical variables was explored in PCA, and two significant components appeared: Variables related to IL-1β activity dominated the first component while in the second component there was a negative association between IL-6 and Hcrt1. Fatigue was introduced as an additional variable in a second model. In this PCA, fVAS scores were associated with the first component as was the IL-1β related variables. In addition, the second PCA model revealed a third component that showed a negative relationship between Hcrt1 and fatigue. Conclusions: I) HSP90α and to a lesser degree HSP72 in blood may possibly be parts of a fatigue inducing mechanism. II) The LC-MS/MS method with high selectivity and accuracy revealed considerably lower levels of Hcrt1 in CSF than previously reported. III) IL-1β signaling is a primary driver in fatigue. Several other proteins and molecules interact with IL-1β in a complex network, in which several cell types (neurons, microglia, and astrocytes) probably participate. IV) Hcrt1 also influences fatigue, but probably through another pathway than the IL-1���� route.