Innate immune mechanisms of infection: what we know and potential conserved mechanisms affecting sleep during infection

Abstract

Evidence indicates relationships between sleep and the innate immune system during homeostatic sleep and sleep responses after infection. The innate immune system and sleep-like states are highly conserved between simple species and more complex species such as humans. A wide variety of bacteria, viruses, and parasites change sleep patterns in the host during infection. The effects of infection on sleep can occur, in part, due to the bolus and route of infection, prior exposure, immune status of the individual/organism, and the type of pathogen. In addition, elements of circadian patterns and sleep prior to and after infection can modulate the infection pathology and resolution. Innate immune molecules, such as the cytokines interleukin-1 beta and tumor necrosis factor-alpha, fluctuate with the time of day of increased activity and sleep propensity, increase in response to increased waking activity from sleep loss, and are altered from infection by bacteria and viruses to alter sleep and the electroencephalogram. This review focuses innate immune mechanisms of how pathogen recognition receptors, pathogen-associated molecular patterns and danger-associated molecular patterns, energy-related molecules, oxidative stress, and inflammasomes are activated with infection to potentially affect sleep.

Highlights

• •Mechanisms that regulate sleep responses to pathogens are highly conserved from single cell organisms to humans.
• •Pathogens including bacteria, viruses, and parasites can alter sleep and slow-wave activity.
• •Innate immune molecules are activated by infectious stimuli to alter sleep and the electroencephalogram.
• •Sleep loss or dysregulated sleep can impair proper immune responses to infection and immune molecules that regulate sleep.
• •Pattern recognition receptors, inflammasomes, toll-like receptors, complement recognize pathogens to dysregulate sleep.

1. Introduction

Sleep or periods of reduced and increased activity and immune functions appears to exist to across species (Ungurean et al., 2020). Indeed, sleep and immunity are highly conserved between species (Ungurean et al., 2020; Bernheim et al., 2024). Sleep has several functions including being vital for proper immune responses to pathogens (Zielinski et al., 2016). The relationship between sleep and the immune system are known to individuals anecdotally in that often one sleeps more or has difficulty sleeping when sick. The understanding of the relationship between sleep and infection has evolved with the work of several influential researchers in the field as well as well-developed and founded novel approaches to elucidate the mechanisms behind this relationship. These studies revealed that key immune-related mechanisms including cytokines, pattern recognition receptors, energy-related molecules, and oxidative stress are vital for sleep regulation and sleep responses to infection. The majority of immune mechanisms known to be involved in homeostatic and sleep responses to infection involve the innate immune system, which we focus on in this review; however, the adaptive immune system is vital to responses to infection and is an additional area needed to be explored in future research. In this review, we discuss innate immune mechanisms of how pathogens can alter sleep and the specificity and redundancy of these mechanisms so future areas of research can be studied to improve our understanding and for development of better treatments.

2. James Krueger and colleagues groundbreaking findings linking sleep and innate immunity

In 1975, Pappenheimer, Koski, Fencl, Karnovsky, and Krueger extracted a substance from the cerebral spinal fluid of sleep deprived goats and sheep that was intraventricularly infused into rats and rabbits and found reduced nocturnal locomotor activity in rats and the duration and increased the amplitude of slow wave cortical EEG in rabbits, that they termed “sleep-promoting factor S” (Pappenheimer et al., 1975). The sleep factor contained muramic acid, alanine, glutamic acid, and diaminopimelic acid that are components of muramyl dipeptide—a Gram-positive bacterial cell wall component (Krueger et al., 1982). Infusion of synthetic muramyl peptides intravenously into rats, rabbits, and cats induced slow-wave sleep (Krueger et al., 1982). These extraordinary studies describe two understandings about sleep including linking a molecule found on pathogens (i.e., muramyl dipeptide) with increasing sleep and that a bacterial component in a non-infected animal increases in the cerebrospinal fluid to enhance sleep. Thereafter, Krueger and colleagues including Charles Dinarello found that the cell signaling molecule interleukin (IL)-1 beta (IL-1β), a pyrogen (fever-causing) molecule that is activated in most nucleated cells in the brain and periphery body, increases NREM sleep, which was a breakthrough in understanding immunologic mechanisms of cell-signaling molecules in regulating homeostatic sleep and sleep responses to infection (Krueger et al., 1984). Intriguingly, Krueger and colleagues produced evidence that in a mixed neuron and glia cultures recorded in a microelectrode array exhibit increased EEG delta power oscillations with the application of inflammatory mediators that stimulate NREM sleep delta power in animals suggesting that local groupings of cells can produce EEG delta power changes independent of brain regional circuitry (Jewett et al., 2015).

3. Cytokines

Cytokines include chemokines, interleukins, interferons, lymphokines, and tumor necrosis factors and are small protein, peptide or polypeptide cell signaling molecules that allows cells to communicate with themselves, cells next to them, or more distally (i.e., autocrine, paracrine, or endocrine mechanisms). Cytokines are highly potent with effects occurring at as little as picomolar concentrations. Cytokines are highly conserved between mammalian species and analogs are found in invertebrates and even bacteria and archaea (Wein and Sorek, 2022). Many cytokines including interferons (IFNs), IL-1β, IL-2, IL-4, IL-6, IL-10, IL-13, IL-15, IL-18, and tumor necrosis factor-alpha (TNFα) can alter sleep and/or slow wave activity (SWA) (Krueger, 2008). Interestingly, many of these cytokines that alter sleep are altered by infection to pathogens. Pro-inflammatory cytokines tend to increase NREM sleep while anti-inflammatory cytokines, such as IL-4, IL-10, and IL-13 tend to inhibit sleep after sleep promoting stimuli such as influenza infection, bacterial components, or sleep deprivation (Fig. 1) (Zielinski and Krueger, 2011). Nevertheless, larger dosages of cytokines can induce fragmented sleep and wakefulness (Opp et al., 1991; Krueger and Majde, 2003). Further, larger dosages of cytokines including TNFα can reduce sleep by causing sleep fragmentation from increased waking episodes (Zielinski and Krueger, 2011).

Fig. 1.

A visual representation of how pathogens have the capacity to facilitate dysregulated sleep. Pathogens can interact with both the innate and adaptive immune systems to facilitate the production of cytokines which have regulatory effects on both sleep and slow-wave activity. Created using Biorender.com.

IL-1β and TNFα are the two best characterized cytokines that regulate sleep and they are known to increase after bacterial cell wall components are applied to the periphery, in the circulation, or directly to the central nervous system (CNS) (Zielinski and Gibbons, 2022). James Kreuger, Mark Opp, and colleagues well characterized IL-1β and TNFα to be sleep regulatory using pharmacology, transgenic mice, siRNA, and other cutting-edge technologies targeting these molecules or their receptors over the years (Zielinski and Gibbons, 2022). In rodents, IL-1β and TNFα have diurnal variations in the brain that correspond with sleep propensity and are upregulated in the cortex after sleep loss and pathogens or their components are administered to the periphery or centrally (Zielinski and Gibbons, 2022). When IL-1β, TNFα, or the gram-negative bacterial wall component lipopolysaccharide (LPS) are applied to the peritoneum, they can enter the circulation to enter the brain through leaky areas or stimulate the vagal afferent nerves to stimulate the nucleus tractus solitarii region of the brainstem leading to increased IL-1β and TNFα expression in other brain areas including the somatosensory cortex to enhance NREM sleep which suggests that pathogens that induce these inflammatory cytokines in the periphery can induce inflammatory sleep regulatory molecules to alter sleep during infection (Zielinski and Gibbons, 2022). Administration of the anti-inflammatory drug anakinra that is an IL-1 receptor antagonist increased SWA. Notwithstanding, there are discrepancies in SWA and NREMs in response to peripheral versus central administration of inflammatory stimuli including administration of TNFα, IL-1β, or LPS in mice (Zielinski and Gibbons, 2022). Furthermore, additional circumstances exist that strongly suggest that NREM sleep amounts and SWA are regulated by different mechanisms yet altered by sleep loss and pathogens and their components (Davis et al., 2011a). Cytokines including IL-1β and TNFα are activated by transcription factors and cell signaling pathways including nuclear factor- κB (NF-κB), cyclooxygenase (COX), prostaglandins (PG), oxidative stress, energy-related molecules and their receptors, as well as pattern recognition receptors (PRRs) that recognize unique pathogen associated molecular patterns (PAMPs) and danger associated molecular patterns (DAMPs) (Zielinski and Gibbons, 2022).

4. NF-κB

Nuclear factor (NF), NF-κB is a transcription factor in the nucleus and mitochondria intermembrane space of cells that functions to transcribe molecules that regulate sleep and immune function including IL-1β and TNFα (Zielinski et al., 2016). For example, LPS activates toll-like receptor (TLR) 4 (TLR4) to induce NF-κB activation and the transcription of IL-1β and TNFα (Zielinski et al., 2016). NF-κB translocation to the nucleus involves the phosphorylation, ubiquitination, and degradation of the inhibitory-kappa B (I-κB) kinase complex. Subunits of NF-κB including p50, p52, p65, ribonucleic acid editing ligase A (RelA), ribonucleic acid editing ligase B (RelB), and c-terminal ribonucleic acid editing ligase (c-Rel). In the cortex of rodents, NF-κB expression variation exhibits a diurnal effect and is increased from sleep deprivation in the lateral hypothalamus, basal forebrain, and cortex (Chen et al., 1999; Brandt et al., 2004; Ramesh et al., 2007). NF-κB activity shows similar increased effects with waking activity in peripheral blood mononuclear cells (Irwin et al., 2008). Peptidergic inhibition of NF-κB and NF-κB inhibition with transgenic mice lacking NF-κB p50 subunit had reduced spontaneous sleep or sleep responses to LPS and mouse-adapted influenza suggesting that NF-κB is a major pathway in sleep responses to infection (Kubota et al., 2001; Jhaveri et al., 2006; Ramesh et al., 2007).

5. Cyclooxygenase-prostaglandin-nitric oxide pathway

The cyclooxygenase-prostaglandin-nitric oxide pathway is activated by sleep regulatory cytokines IL-1β and TNFα and can alter their activation (Zielinski and Gibbons, 2022). Prostaglandins (PG) are a family of structurally related bioactive lipid mediators that are activated after many types of infections. Interestingly, cyclooxygenase and prostaglandins appear to be highly conserved evolutionary with prostaglandins and similar molecules existing in bacteria and plants (Järving et al., 2004; Kubata et al., 2007). The cyclooxygenase-prostaglandin pathway is activated by infectious pathogens and is involved in altering sleep and SWA responses to pathogens and/or their components and homeostatic sleep (Zielinski and Gibbons, 2022). Arachidonic acid is converted to prostaglandin H2 by the rate limiting enzyme cyclooxygenase. Cyclooxygenase (COX) exists in two major forms, COX-1 which is constitutively active, and COX-2 that is induced by inflammatory producing stimuli including infectious pathogens and their components, growth hormones, and physiological stimuli (Sheremeta et al., 2024). Non-steroidal anti-inflammatory agents that attenuated COX and prostaglandins have significant but small effects on sleep in humans (Murphy et al., 1994). COX-2 inhibition in rabbits reduced spontaneous sleep and sleep responses to TNFα infused into the basal forebrain (Yoshida et al., 2003). COX-2 inhibitors reduced spontaneous sleep and sleep after TNFα was given in rats and rabbits (Terao et al., 1998; Yoshida et al., 2003). Evidence in COX-2 KO mice suggest that homeostatic sleep responses to sleep loss prostaglandins are independent in homeostatic responses but are dependent, in part, on responses to LPS (Szentirmai et al., 2024). Prostaglandin D2 synthase converts prostaglandin H2 to prostaglandin D2—a molecule that can activate adenosine A2a receptors and inhibit histaminergic arousal system. PGD2 acts on DP1 and DP2 receptors and can enhance vasoconstriction, inflammation, and stimulate adenosine to promote sleep through the adenosine A2a receptors. After sleep deprivation, PGD2 is increased in the brain (Huang et al., 2007). In rats, PGD2 applied centrally increased NREM sleep (Matsumura et al., 1994). Prostaglandin H2 is converted to prostaglandin E2 by prostaglandin E2 synthase. PGE2 has multiple functions including increasing body temperature, leukocyte infiltration, vasodilation and vasoconstriction, and vascular permeability, regulating lymphocyte functions, and modulating sleep. PGE2 acts on 4 receptors EP1, EP2, EP3, and EP4 (Andreasson, 2010). Neuronal and glial cell activation by IL-1β can lead to an increased expression of COX-2 (Neeb et al., 2011). Evidence indicates that in astrocytes, IL-1β stimulates PGE2 secretion, in part, via the protein kinase C and mitogen-activated protein kinases (MAPK) (Molina-Holgado et al., 2000). PGE2 is shown to reduce wakefulness, which likely occurs, in part, form the activation of prostaglandin EP1 and EP2 receptors that are found in the tuberomammillary nucleus within the posterior hypothalamus (Onoe et al., 1992; Yoshida et al., 2000). Sleep promoting actions of PGD2 can occur, in part, through the EP4 receptor (Yoshida et al., 2000). Mice lacking EP4 receptors have partly attenuated increased NREM sleep and reduced SWA responses to intraperitoneally given LPS (Oishi et al., 2015).

Nitric oxide is a free radical signaling molecule produced by most organisms including bacteria, plants, fungi, mammals, and is found in individual cells (Feelisch and Martin, 1995). Nitric oxide (NO) functions include relaxing smooth muscle to induce vasodilation and increase blood flow, acting as a neurotransmitter, signal transduction, apoptosis, gene transcription, mRNA translation, post-translational modification of proteins, immune and inflammatory responses to pathogens, and altering sleep and SWA. NO is produced from NO synthase (NOS) catalyzing arginine and nicotinamide adenine dinucleotide phosphate (NADPH) and dioxygen (Zielinski and Gibbons, 2022; Andrabi et al., 2023). Prostaglandins produced by COX can alter the activity of NOS to alter nitric oxide (Salvemini et al., 2013). Three types of NOS exist including inducible NOS (iNOS), endothelial NOS (eNOS) and neuronal NOS (nNOS). The NOS designations describe their expression activity pattern, where iNOS is inducible while eNOS and nNOS are constitutively active, or cells that produce them, although nNOS is also produced by astrocytes. Evidence using transgenic mice, pharmacology, and optogenetics indicates that iNOS and nNOS can enhance sleep and SWA. Inhibition of NO with L-nitro-arginine methyl ester attenuates enhanced NREM sleep and REM sleep responses to sleep deprivation in rats and spontaneous sleep in both rats and rabbits (Kapás et al., 1994a, 1994b; Ribeiro et al., 2000). Intracerebroventricular administration of NO donors 3-morpholinosydnonimine (SIN-1) and S-nitroso-N- acetyl -D, L-penicillamine (SNAP) enhanced NREM sleep in rats (Kapás and Krueger, 1996). iNOS and NO levels assessed by microdialysis and dual stained iNOS and c-Fos immunoreactivity are enhanced in sleep deprived rat basal forebrain and frontal cortex (Kalinchuk et al., 2010, 2011). iNOS KO mice have lower spontaneous NREM sleep and NREM sleep responses to TNFα and influenza infection than wild-type controls (Chen et al., 2003, 2004). Mice lacking nNOS have lower SWA responses to sleep deprivation (Morairty et al., 2013). A unique group of sleep-active nNOS expressing interneurons in the cortex, caudate-putamen, olfactory bulb, corpus callosum and amygdala GABAergic interneurons that also co-express neuropeptide Y, somatostatin, and the neurokinin-1 receptor are activated during sleep occurring after sleep loss (Gerashchenko et al., 2008). Evidence implicates somatostatin, and neurokinin 1 receptor (NK1) receptor in enhancing SWA (Zielinski et al., 2015). Inhibiting somatostatin nNOS expressing cells in mice using a cross-sectional breeding strategy found that these cells are involved with homeostatic responses to sleep loss in the <1.5 Hz frequency range during NREM sleep (Zielinski et al., 2019). nNOS KO mice also have reduced NREM sleep responses influenza infection (Chen et al., 2004).

6. Pattern recognition receptors

Innate immune responses respond within minutes to hours after an exposure to a pathogen (Marshall et al., 2018). Proteins and peptides, and polypeptides are made by both immune cells and a multitude of other cells in the periphery or central nervous system to function in innate immunity against pathogens. Cells in humans and other species have developed the ability to recognize pathogens or their components by pattern recognition receptors. PRRs recognize PAMPs and DAMPs including Toll-like receptors (TLRs), nucleotide-binding oligomerization domain (NOD) like receptors (NLRs), Retinoic acid-inducible gene-I (RIG-I) like receptors, absent in melanoma 2 (AIM2)-like receptors, and cyclic GMP-AMP synthase (cGAS)/simulator of interferon genes (STING) (Wicherska-Pawłowska et al., 2021).

6.1. Toll-like receptors

Toll protein was first discovered in the fly Drosophila melanogaster (Anderson et al., 1985). Toll-like proteins, receptors, and related pathways are found in many basic life forms including Porifera, Mollusca, Cnidaria to mammals including mice and humans (Brennan and Gilmore, 2018). Metazoan TLR proteins typically have a hydrophobic tandem leucine-rich repeat extracellular domain that recognizes PAMPS, a short transmembrane domain and a toll-interleukin-1 receptor signaling domain that activates downstream pathways including myeloid differentiation primary response 88 (MyD88), toll/interleukin-1 receptor/resistance protein (TIR)-domain-containing adapter-inducing interferon-β (TRIF) that activate three transcription factors NF-κB, activator protein-1 (AP-1), and interferon regulatory factors (IRFs) to induce inflammatory responses. Toll-like receptor with extracellular leucin-rich repeat (LRR) motifs and intracellular TIR in the activation of the transcription factor NF-κB genes are seen in phylogenetically distant lineages and in simpler organisms than humans such as Drosophila melanogaster (Leulier and Lemaitre, 2008). Interestingly, placozoans that are some of the simple metazoans that do not have true Toll- and NOD-like receptors exhibit many genes downstream of signaling cascades implicating that primordial toll- and NOD-like pathways exist in simple species (Kamm et al., 2019). Indeed, placozoans that are eukaryotes have altered circadian-dependent behavior and are suggested to have sleep/wake like difference states. Moreover, evidence suggests that cytokines are also conserved to some extent evolutionarily in that the marine protozoan Euplotes raikovi has structural similarities with IL-2 (Luporini et al., 1994). Plants also express toll-like proteins, LRRs and TIRs that are pathogen sensors such as the tobacco N gene product that shows resistance to the tobacco mosaic virus (Dembic, 2000–2013). Additionally, there are similar pathways to mammals including plants have a NLR protein, ZAR1, that when fused to green fluorescent protein (GFP) assembles into a pentamer at the plasma membrane and promotes calcium influx that likely results in direct or indirect activation of the cell-death pathways which is similar to inflammasome activation (Bi et al., 2021). Indeed, prokaryotic cells possess innate and adaptive immune systems that parallel ones seen in eukaryotes although lack organelles that are present in eukaryotes. NLRs of the signal transduction ATPases with numerous domains (STAND) superfamily detect molecular patterns of pathogens are prevalent in bacteria and archaea (Gao et al., 2022). TLR4 is a major activator of the innate immune response against bacterial infections, such as IL-1β responses to LPS, and TLR4 diverged from TLR2 approximately 700 million years ago during vertebrate evolution (Beutler and Rehli, 2002). Intriguingly, TLR4 knockout mice exhibit reduced enhancements in NREMS episode durations and episode frequencies and SWA after 24 continuous hours of sleep restriction compared to wild-type mice (Wisor et al., 2011). Modestly reduced amounts of time spent in NREMS was observed during spontaneous sleep in TLR4 KO mice when compared with wild type mice (Wisor et al., 2011). Mice lacking both TLR2 and TLR4 are reported to spend 41 % more time awake during the dark period, which is consistent with TLR4 being involved in homeostatic sleep regulation (Sartorius et al., 2012). Further, the role of TLR4 in sleep regulation is supported by findings in MYD88 deficient mice that have increased wakefulness compared to wild-type mice (Choudhury et al., 2022).

6.2. NOD-like receptors

NLRs are conserved to all life forms as simple as prokaryotes (Gao et al., 2022). NLR family of receptors are found in the cytosol of most cells including immune cells and non-immune cells (Almeida-da-Silva et al., 2023). NLRs are sensing entities that determine changes in infections, oxidative stress, and energy-related molecular changes (Almeida-da-Silva et al., 2023). NLRs induce three signaling pathways including NF-κB, MAPK, and inflammasome activation. Interestingly, the NLR family share similarities with nucleotide-binding site (NB) NB-LRR subgroup of disease-resistance genes in plants. Currently, there 34 NLR members have been identified with 23 of them being human that are highly conserved between species (Allen, 2014). NLRs are organized in a three-part formation. First, there is a central nucleotide-binding domain (NACHT) that allows self-oligomerization for activation. Second, a C-terminal leucine-rich repeats region that senses the binding ligand. Thirdly, a variable N-terminal homotypic protein-protein interaction domain. Five subfamily members of the NLR family exist with different N-terminal effector domains and primary functions including NLRA, NLRB, NLRC, NLRP, and NLRX. Distinct NLRs exist that recognize different components to form unique inflammasomes (Allen, 2014). Muramyl peptide are recognized by NOD-2 receptors and oligomerization to signal via RIP2 kinases leading to NF-κB and activating protein-1 transcription, NLRP1 and/or NLRP3 inflammasome activation, and pro-inflammatory cytokine activation (Hsu et al., 2008; Strober and Watanabe, 2011). Muramyl peptide induces NREMS in rats, rabbits, and cats and muramyl dipeptide, lysyl derivative acetylmuramyl-L-alanyl-D-isoglutaminyllsine, and peptidoglycans demonstrate similar somnogenic effects when applied to animals (Krueger et al., 1982, 1987). Therefore, since muramyl peptide is present on most bacteria, it is plausible that NOD-like receptors and inflammasomes are involved in these somnogenic effects.

7. Inflammasomes

Inflammasomes are a protein complex that forms in response to pathogens by their PRRs to mount inflammatory and immune responses to infection (Fig. 2) (Zheng et al., 2020). Inflammasomes are involved with other functions as well including cognition, alteration of mood, and sleep (Zielinski and Gibbons, 2022). Four classes of NLR N-terminal assembly domains have been identified including the acidic transactivation domain, pyrin domain, caspase recruitment domain (CARD), and baculoviral inhibitory repeat (BIR)-like domains (Zheng et al., 2020). Another class of inflammasome assembling domains include the PYHIN protein family members including AIM2 that have (hematopoietic interferon-inducible nuclear with 200 amino acid repeats) HIN200 and pyrin (PY) domains (Zheng et al., 2020). Inflammasomes have specificity to certain pathogenic stimuli as well as multiple types of inflammasomes can be activated by different components or downstream effects from the particular infectious agent. Inflammasome complexes form with the recruitment of the “adapter and effectors to affiliate the apoptosis-associated speck-like protein containing a CARD” (ASC) that is found in most inflammasome complexes to activate caspase-1 (Zheng et al., 2020). Caspase-1 functions to cleave the pro-forms of IL-1β and/or IL-18. Canonical inflammasomes include NLRP3 and CARD-absent NLRs such as AIM2, IFI16, NLRP3, NLRP6, and NLRP7, PYD–PYD interactions between NLRs and ASC nucleate the PYD filaments of ASC to nucleate CARD filaments of caspase 1 by CARD-CARD interactions for self-activation (Zheng et al., 2020). However, NLRC4 contains a card and the nucleation and activation occur downstream of caspase-1 by homotypic CARD-CARD binding independent of ASC (Zheng et al., 2020). Noncanonical inflammasome recruit caspase-1 in a caspase-1 independent fashion, in part by the recognition of cytosolic LPS by caspase activation and recruitment domains (CARDs) of caspase-4 and caspase-5 (in humans and caspase-11 (in mice to in induce caspase-dimerization and activation to cleave gasdermin D (GSDMD). GSDM induces proptosis, which is a lytic form of programmed cell death in response to pathogens and danger signals. This occurs, in part to pore induced intracellular traps arrests of bacteria to recruit other cells to kill the pathogen. Apoptosis is another primary function of inflammasomes that is categorized as noncanonical and different than pyroptosis and canonical inflammasomes that utilize caspase-8 independent apoptosis is activated by AIM2, NLRP3, NLR family of apoptosis inhibitory proteins (NAIP)-NLRC4 inflammasomes (Zheng et al., 2020). Currently the role of inflammasomes in sleep responses related to pathogenic-related stimuli is limited to NLRP3 inflammasomes, although the unique pathogen and components of pathogens that activate specific inflammasomes are likely to be involved in sleep and SWA responses found from most infections.

Fig. 2.

A visual representation of various inflammatory stimuli and resulting downstream inflammasome pathways activated. Created using Biorender.com.

7.1. NLRP3 inflammasomes

NLRP3 inflammasome activation typically occurs in a two-step process involving a priming step that involves the activation of NF-κB transcription of NLRP3 inflammasome components and an activation phase that induces the activation of the NLRP3 inflammasome formation (Zheng et al., 2020). NLRP3 inflammasomes are primed by several mechanisms including LPS activating TLR4, IL-1β activating IL-1R1, or TNF-α activating the TNFR1 (Zheng et al., 2020). Interestingly, LPS activation of TLR4 to activate MyD88 signaling can induce deubiquitylation in a relatively short amount of time such as 10 min in macrophages, although it is unknown what occurs in microglia, astrocytes, and neurons (Juliana et al., 2012). TLR4 can also stimulate Interleukin-1 receptor-associated kinase 1 (IRAK1) and IRAK4 dependent NLRP3 inflammasome components for rapid assembly without the need for priming (Lin et al., 2014). A secondary step involved in activation generally involves energy-related molecules and oxidative stress—two processes that are altered by pathogenic infection and increased waking activity (Ai et al., 2023; Di Virgilio et al., 2023). Oxidative stress plays a role in inflammation, responses to pathogenic infection, and sleep (Atrooz and Salim, 2020). Thioredoxin (TRX) activity in the mitochondria and cytosol sustains the activity of the TRX system by a redox mechanism (Choi and Park, 2023). The TRX system includes TRX reductase, nicotinamide adenine dinucleotide phosphate, and TRX interacting protein (TXNIP). TXNIP functions as a negative regulator to activate TRX to maintain redox homeostasis and is involved in apoptosis, inflammation, and is a critical component of NLRP3 inflammasome activation. Consequently, TXNIP functions as a pro-oxidant to enhance ROS generation and induce oxidative stress. Bacteria, archaea, eukaryotes, and cells in higher level vertebrates including humans respond to extracellular adenosine triphosphate (ATP) (Tronnet et al., 2024). Extracellular ATP binds to purine type 2 receptors (P2Rs). In the brain, all cells express P2Rs and extracellular ATP is involved in fine tuning microglia and astrocyte responses (Di Virgilio et al., 2023). It is postulated that calcium content in water in the earth contributed to depletion of phosphate compounds in an extracellular environment of single celled organisms (Williams, 2006). Thus, phosphate generated by ATP hydrolysis precipitates into calcium phosphate when increased calcium concentrations are present. The relationship between the calcium and ATP gradients thus appears to be conserved. The controlled fluctuations in cytoplasmic calcium are important to a multitude of functions including cell proliferation, motility, trophism, expression of differentiated responses, neurotransmitter, cytokine, inflammation, cell death, necrosis, and apoptosis (Di Virgilio et al., 2023). ATP is believed to exist early in molecular evolution as it is versatile as a phosphate donor under thermodynamic or kinetic control. P2X receptors are thought to detect extracellular pyrophosphates and were environmental sensors such as primeval DAMPS driven by dying organisms such as amino acids or nucleotides (Di Virgilio et al., 2023). Interestingly, tetrameric glutamate and the trimeric ATP ionotropic receptors are conserved and involved in sleep and responses to infection (Zielinski et al., 2016). Many controlled release pathways exist from plasma membrane channels and transporters to stimulator or constitutive exocytosis that allow the release of ATP into the extracellular space (Di Virgilio et al., 2023).

Ectonucleoside triphosphate phosphohydrolases, ectonucleotide pyrophosphatase phosphodiesterase, 5′-ectonucleotidase (also known as CD73), and ectoalkaline phosphatases convert extracellular ATP into molecules that effect the somnogenic molecule adenosine (Zielinski et al., 2016). Consequently, a large literature indicates that adenosine and its receptors adenosine A1 receptor and A2a receptor affect sleep. Moreover, mice that lack CD73 exhibit reduced sleep responses to sleep deprivation further suggesting adenosine and extracellular ATP in sleep regulation (Zielinski et al., 2012). An important study by Krueger and colleagues using pharmacology and mice lacking the P2X7 receptor that ATP binds to suggests that extracellular ATP acts, in part, through the P2X7 receptor to induce NREM sleep and SWA (Krueger et al., 2010). Evidence suggests that extracellular ATP is increased in the pericellular space locally and is restricted to the release site when a buildup occurs and upon a strong stimulation or reduction of ectonucleotidase activity, allowing extracellular ATP to diffuse further away from the area of release (Dale, 2021). Indeed, extracellular ATP is secreted from Enterococcus gallinarum bacteria itself (Iwase et al., 2010). Consequently, infection or enteric bacteria could provide extracellular ATP to activate inflammatory mechanisms locally or in more distal areas including the CNS. Pathogens can also trigger ATP release from cells (Spari and Beldi, 2020). LPS can induce the release of extracellular ATP from microglia; extracellular ATP can act as a neurotransmitter (Ferrari et al., 1997). P2X and P2Y receptors are widely expressed in the central nervous system. In addition to synapses, neurons, microglia, and astrocytes release substantial amounts of ATP via pannexin-1, connexins, secretory granules, and the P2X7 receptor—molecular entities that are involved in NLRP3 inflammasome activation. Intracerebroventricularly applied LPS increased extracellular ATP release throughout much of the periphery and the brain when luminescence was shown to increase in pmeLUC-TG mice (Di Virgilio et al., 2023). Studies using a energetically-encoded G protein-coupled receptor (GPCR) activation-based (GRAB) sensors coupled to a P2Y receptor, as a specific ATP-binding moiety, demonstrate increased extracellular ATP in the CNS after LPS (Wu et al., 2022). Intriguingly, astrocyte calcium activity in the basal forebrain was highly correlated with extracellular ATP and adenosine levels but mice lacking CD73 did not alter adenosine level dynamics or calcium levels during the sleep-wake cycle suggesting basal forebrain adenosine might not contribute to astrocyte SWA and sleep regulation (Peng et al., 2023). Thus, this finding supports the role of NLRP3 inflammasomes activation in sleep regulation by increased extracellular ATP and calcium efflux.

NLRP3 inflammasomes can be regulated by post-translation modification such as phosphorylation and deubiquitylation (Zheng et al., 2020). MicroRNAs, such as microRNA-132 and long non-coding RNAs can promote or inhibit inflammasome activation post-transcriptionally. LPS stimulates microRNA-132 in THP-1 cells—a monocyte isolated from peripheral blood of a leukemia patient (Taganov et al., 2006). MicroRNA-132 has been found to reduce NLRP3 inflammasome activation in THP-1 cells stimulated by palmitate (Byeon et al., 2017). Indeed, Kreuger and colleagues found that microRNAs, such as miRNA-132, applied intracerebroventricularly decreased NREMS and SWA (Davis et al., 2011b). MicroRNAs can also regulate sleep and sleep homeostasis in Drosophila (Goodwin et al., 2018). Further, miRNAs, such as miRNA-132, are associated with changes in circadian rhythms and sleep disorders (Kinoshita et al., 2020). Nitric oxide can also inhibit NLRP3 inflammasome activity post-translationally by direct thiol nitrosylation to control immunopathology of Mycobacterium tuberculosis (Mishra et al., 2013). Additional negative activators of NLRP3 inflammasomes exist including B-cell adapter for phosphoinositide 3-kinase (BCAP) that interacts with caspase −1 pseudosubstrate inhibitor flightless-1 to inhibit the recruitment of pro-caspase-1 and reduce pathogen clearance, although the negative inhibitors in the CNS remain poorly understood (Carpentier et al., 2019). Pyrin-only protein (POP) members 1 and 2 can inhibit inflammasome assembly by binding to ASC and inhibiting the formation of the NLRP3 inflammasome in stressed cells (Samir et al., 2019). Heat shock protein 70 and NLRC3 can also inhibit ASC from being recruited to NLRP3 (Youm et al., 2015; Martine et al., 2019). Human recombinant heat shock protein 70 given intranasally to mice reduced sleep deprived-induced anxiety and cognitive impairment and reduced sleep deprivation-induced enhancements in IL-1β, TNFα, and monocyte chemotactic protein (MCP-1) protein upregulation (Kang et al., 2023). Additionally, molecules that affect ion efflux, mitochondrial function, and ROS signaling can inhibit NLRP3 inflammasome activation. G-protein coupled receptors can both enhance and suppress NLRP3 inflammasomes by altering ion fluxes and mitochondrial ROS (Tang et al., 2018). In Mycobacterium tuberculosis infections, NO can inhibit NLRP3 inflammasome activation by stabilizing mitochondria (Mishra et al., 2013). Interestingly, IL-10 that attenuates sleep responses to sleep promoting stimuli and can inhibit NLRP3 inflammasome activation, in part, by attenuating mitochondrial ROS production (Ip et al., 2017). Therefore, it is plausible that anti-inflammatory cytokines that attenuate sleep responses to pathogens could act by inhibiting inflammasomes.

The complement system was discovered in the late 19th century as a heat labile lytic factor in the serum that aided in killing bacteria and complemented antibodies (Nesargikar et al., 2012). Three pathways of complement activation are known including the classical pathway with complement component C1q as a primary pattern recognition sensor, the lectin pathway with mannose-binding lectin and ficolins that are sensors, and the alternative pathway where C3b and properdin are the main sensors (Nesargikar et al., 2012). The three pathways of the complement system all induce C3 activation leading to downstream activation of pathways involved in C3a, C3b, and C5 activation by C4 and C5 convertases (Nesargikar et al., 2012). Additionally, C5 cleavage ends with the deposition of an oligomeric structure of complement components in the cell membrane that creates a pore called the membrane attack complex (MAC), which includes C5b-9 that lysis infected and damaged cells (Nesargikar et al., 2012). MAC activation can induce NLRP3 inflammasome and caspase-1 activation (Suresh et al., 2016). C3a and C5a are inflammatory mediators that are also involved in non-immune functions including vasodilation, pro-inflammatory cytokine production, histamine release, altered TLR4, extracellular ATP, and P2X7 receptor responses (Suresh et al., 2016). A study in human, found that plasma C3 and C5 were increased after sleep deprivation. Another study found that plasma C3a levels were higher during sleep than during wakefulness (Hui et al., 2007). A study in mice indicated chronic sleep restriction increased hippocampal astrocyte derived complement 3 secretion and microglial C3a receptor expression (Wadhwa et al., 2019). In rats, hippocampal C3, C5, C3a, and C5a and receptors C3aR and C5aR were upregulated after sleep deprivation (Wadhwa et al., 2019). Increased cortical C3 expression in microglia was also reported in mice after acute sleep deprivation or chronic sleep restriction (Bellesi et al., 2017). Nevertheless, the specific effects of how the complement pathway activates inflammasomes and sleep remain poorly understood.

7.2. RIG-I

RIG-I recognizes RNA in viruses (Thoresen et al., 2021). RIG-1 is highly conserved between vertebrates. RIG-I is a primary defense against paramyxoviruses, coronaviruses, orthomyxoviruses, flaviviruses, rotaviruses, filoviruses, reoviruses, herpesviruses, alphaviruses, west Nile virus, Japanese encephalitis virus, influenza A, Sendai virus, HIV-1, and arena viruses (Thoresen et al., 2021). Additionally, RIG-I recognizes dsRNA viral genomes, defective interfering particles, and mini viral RNA (Thoresen et al., 2021). Evidence also indicates that RNA components of Listeria, Salmonella enterica, and Mycobacterium tuberculosis are sensed by RIG-I (Thoresen et al., 2021). RIGI- can mediate type 1 interferon responses to pathogens (Thoresen et al., 2021). RIG-I is an ATP-dependent DExD/H box RNA helices protein that upon activation, the N-terminus CARDs migrate and bind with CARDs attached to mitochondrial antiviral signaling proteins (MAVS) to activate signaling pathway for IFN-1. IFN-1 activates the Janus kinase-signal transducer and activator of transcription (JAK-STAT) pathway to activate IFN-stimulated genes (Thoresen et al., 2021). Stimulator of interferon genes (STING) antagonizes RIG by binding to the N-terminus to limit activation. Interestingly, evidence from James Krueger group found that sleep and body temperature responses in an acute viral infection model are altered in interferon type 1 receptor knockout mice and this mouse genotype also have a 30 % reduction in spontaneous REM sleep (Bohnet et al., 2004; Traynor et al., 2006). While evidence indicates that IFN-beta does not have somnogenic activity, IFN-alpha has somnogenic effects particularly for NREM sleep (De Sarro et al., 1990; Kimura et al., 1994; Majde and Krueger, 2005). Findings also indicates that RIG-1 can activate and release the somnogenic cytokine IL-1β, which is likely due to RIG-I close associates with caspase-1, 3, 8 and 9 (Ma et al., 2023). RIG-I is expressed on human and murine microglia and murine astrocytes and neurons (Furr et al., 2010; Nazmi et al., 2011; Johnson et al., 2020).

7.3. AIM2

AIM2 recognizes double stranded DNA from microbial or host cellular origin including bacterial and viral DNA (Kumari et al., 2020). AIM2 can recognize bacteria including Listeria monocytogenes, Streptococcus pneumonia, mycobacterium, Porphyromonas gingivalis, Staphylococcus aureus, Brucella abortus, Chlamydia muridarum, and Francisella tularensis (Kumari et al., 2020). There appears to be limitations to AIM2 in the ability to recognize viruses with cytomegalovirus, vaccinia virus, and human papillomaviruses being recognized by AIM2 (Kumari et al., 2020). AIM2 can also recognize fungal pathogen Aspergillus fumigatus and the protozoan Plasmodium berghei (Kumari et al., 2020). AIM2 recruits ASC which results in caspase-1 binding to form the AIM2 inflammasome and downstream activation of IL-1β and IL-18 (Kumari et al., 2020). Additionally, caspase-1 and IL-1β transformation into their mature forms occurs during Streptococcus pneumonia and is dependent on AIM2 inflammasome activation in macrophages (Fang et al., 2011). The literature also describes that AIM2 has protective effects from Staphylococcus aureus infection in the central nervous system, in part, demonstrated by AIM2 knockout mice having reduced levels of IL-1β, CXCL1, CXCL10, CCL2, and IL-6 (Kumari et al., 2020). Intriguingly, many of these inflammatory molecules altered by AIM2 can modulate sleep. IL-6, in particular, is shown to enhance NREM and suppress REM sleep when recombinant rat IL-6 was administered intracerebroventricularly in a dose-dependent manner (Hogan et al., 2003). AIM2 is expressed in neurons, microglia, and astrocytes (Li et al., 2021).

7.4. NLRP1

Several bacterial exotoxins including diphtheria toxin from Corynebacterium diphtheriae, exotoxin A from Pseudomonas aeruginosa, and sidl from Legionella pneumophila can activate NLRP1 inflammasomes (Barry et al., 2023). NLRP1 inflammasomes can also sense nucleic acid from viral products including double stranded RNA poly (I:C) and poly (dA:dT) (Barry et al., 2023). NLRP1 in humans and NLRP1b in mice have a CARD domain at the C-terminus and a unique function to find domain (FIIND) (Barry et al., 2023). Release of a C-terminal FIND-conserved in UNC5, PIDD, and ankyrins-(UPA)-CARD fragment is required to recruit caspase-1 to start NLRP1b inflammasome assembly (Sandstrom et al., 2019). Bacillus. anthracis can activate NLRP1b in rodents. NLRP1b is activated by Listeria monocytogenes, the intracellular parasite Toxoplasma gondii and the depletion of cytosolic ATP—a mechanism that has been proposed to occur from increased waking activity (Liao and Mogridge, 2013; Neiman-Zenevich et al., 2017). Nevertheless, it remains unknown if sleep loss affects NLRP1. NLRP1 expression found in the cerebral cortex in humans especially in neurons and oligodendrocytes although it is also found in glia (Kummer et al., 2007). Nevertheless, it remains unknown if NLRP1 activation during infection affects sleep.

7.5. NLRP6

NLRP6 is structurally similar to NLRP3, although differences in the PYD component that allow for NLRP6 PYD to nucleate ASC by itself, which NLRP3 PYD cannot (Ghimire et al., 2020). NLRP6 is involved in the intestinal microbiome and against enteric viral infections and bacterial infections (Ghimire et al., 2020). NLRP6 inflammasomes are involved in control of microbiota, regulation of metabolic disease and modulation of host defense during microbial infection, and the regulation of neuroinflammation (Ghimire et al., 2020). Staphylococcus aureus, Listeria monocytogenes, lipoteichoic acid from Listeria monocytogenes activate NLRP6 signaling (Ghimire et al., 2020). NLRP6 activates caspase and NF-κB and MAPK pathways that induce the transcription or activation of cytokines that modulate sleep, although NLRP6 was shown to attenuate NF-κB in mice. Increasing evidence indicates that the MAPK/ERK pathway can modulate the circadian system (Wang et al., 2020b). A study in Drosophila demonstrated that ERK phosphorylation regulates sleep (Vanderheyden et al., 2013). Moreover, genes of the MAPK/ERK pathway are linked to sleep shortening in Drosophila (Allebrandt et al., 2017). Listeria monocytogenes also induces type 1 IFN signaling in macrophages to upregulate caspase 11 and NLRP6. Although the role of NLRP6 in specific brain cells and sleep has not been determined, NLRP6 expression has been found in neurons, microglia, and astrocytes (Zhang et al., 2020; He et al., 2024).

7.6. NLRP9

NLRP9 inflammasomes are activated by rotavirus and Helicobacter pylori infection (Mullins and Chen, 2021). Mice have three isoforms including mNLRP9a, mNLRP9b and mNLRP9c while humans only have one hNLRP9 gene (Mullins and Chen, 2021). NLRP4, NLRP5, NLRP8, NLRP9 and NLRP14 share similarities in that they are produced in reproductive organs, although they are also involved in infection related functions (Mullins and Chen, 2021). NLRP9 differs from NLRP3, NLRP6, and AIM2 in that the PYD domain is not oligomerized to nucleate the ASC filament assembly and instead acts as a monomer compatible with higher oligomeric filament formation (Mullins and Chen, 2021). Sleepiness and dysregulated sleep are common occurrences in rotavirus infection, although the specific mechanisms governing these relationships and the impact of NLRP9 inflammasomes have yet -to-be determined (Hellysaz and Hagbom, 2024). While the expression of NLRP9 appears to be highly expressed in the reproductive tract and intestinal and not the brain, studies in rodents indicate that inflammatory cytokines such as IL-1β or TNFα or LPS can increase pro-inflammatory somnogenic molecules in the brain to increase sleep, in part, through the stimulation of the vagal afferents (Zielinski et al., 2013; Zielinski and Gibbons, 2022).

7.7. NLRP12

NLRP12 has a role in Yersinia pestis, Salmonella enterica, Burkholderia pseudomallei, influenza A, Francisella tularensis, Pseudomonas aeruginosa, Staphylococcus aureus, and Leishmania major, Brucella abortus, P aeruginosa, and COVID-19 infection (Huang et al., 2023). NLRP12 mediates inflammatory responses and innate immunity (Huang et al., 2023). NLRP12 can also function as a negative regulator of inflammation. NLRP12 contains an N-terminal pyrin domain, a central nucleotide-binding domain, and a C-terminal leucine-rich repeat region (Huang et al., 2023). NLRP12 is shown to induce NF-κB, caspase 1 and IL-1β in non-immune cells (Huang et al., 2023). However, other studies report that proinflammatory cytokine and chemokines are suppressed by downregulation of canonical and non-conical NF-κB signaling (Allen et al., 2012). NLRP12 also has anti-inflammatory properties as activation of NLRP12 from B. pseudomallei leads to a reduction in TNFα production in macrophages (Pudla et al., 2022). NLRP12 can also inhibit inflammatory signaling, in part, through modulating canonical and noncanonical NF-κB and MAPK/ERK pathways (Huang et al., 2023). Further, NLRP12 attenuated IκBα, and ERK to inhibit ERK and NF-κB activation after Salmonella enterica serovar Typhimurium infection which are mechanisms that lead to pro-inflammatory somnogenic cytokine activation. Additionally, NLRP12 is shown to suppresses TNF-α and Type 1 interferons after Burkholderia pseudomallei and Dengue Virus infections respectively (Huang et al., 2023). Thus, there are multiple somnogenic molecules and mechanisms that are induced by NLRP12 inflammasome activation from infection that can alter sleep.

7.8. NLRC4-NAIP

Bacterial flagellin, rod protein, and secretion system components are found in most gram-negative pathogens and can activate NLRC4 inflammasomes (Vance, 2015). NLRC4-NAIP is involved in Salmonella and Legionella pneumophila infections (Vance, 2015). NLRC4 inflammasome activation is tightly regulated by transcriptional and post-transcriptional mechanisms to prevent unbridled inflammation (Duncan and Canna, 2018). NLRC4 expression is upregulated by somnogenic cytokines including TNFα (Duncan and Canna, 2018). NLRC4 utilizes NAIP binding to bacterial ligands and then binds to NLRC4 to induce inflammasome formation (Kofoed and Vance, 2011). Bacterial effector proteins including type III and IV secretory systems to trigger NLRC4 inflammasome activation. Nucleotide binding domain-associated helical domains of NAIPS are involved in specificity of NLRC4 activity (Tenthorey et al., 2014). NLRC4 inflammasomes also activate caspase-7 and GSDMD downstream (Gonçalves et al., 2019). Notwithstanding the broad responses of NLRC4 to NAIP protein activation from cytosolic flagellin or type III secretary systems, certain pathogens, such as Listeria monocytogenes require priming by LPS for NLRC4 inflammasome formation (Wu et al., 2010). Also, Anaplasma phagocytophilum can induce NLRC4 activation in leu of PAMPs and utilize mechanisms dependent on the cyclogenesis-prostaglandin pathway—a pathway we described in this review that is involved in regulating sleep (Wang et al., 2016).

7.9. IFI16

Interferon inducible protein 16 (IFI16) is a member of the pyrin hematopoietic interferon-inducible nuclear antigen with 200 amino acid repeats (PYHIN) protein family that forms an atypical inflammasome (Yang et al., 2017). IFI16 is comprised of a pyrin domain, HIN-A and HIN-B domain and is currently found in humans but not mice (Zhao et al., 2015). IFI16 Is found in the nucleus of human cells and is a nucleus associated inflammasome sensor that recognizes viruses replicating with in the nucleus (Zhao et al., 2015). IFI16 inflammasomes are involved in viral infections including human cytomegalovirus, herpes simplex virus-1 (HSV-1) (Zhao et al., 2015). IFI16 inflammasomes activate pro-inflammatory interleukins and Type 1 interferons that modulate or regulate sleep (Zhao et al., 2015). It is hypothesized that IFI16 inflammasome cytokine activation occurs primarily from STING activation of TANK-binding kinase 1 phosphorylation of interferon regulatory factor 3 to induce transcription and expression of IFN genes and cytokines (Liu et al., 2023). Nevertheless, currently little is understood about the IFI16 inflammasome pathway and sleep during infection.

7.10. Pyrin

Pyrin inflammasomes are found in immune cells and recognizes inactivation modification of the Ras homolog family member A guanosine triphosphatase (RhoA GTPase) by pathogens (Schnappauf et al., 2019). Interestingly, similarities exist between pyrin inflammasomes and TLR4 and the guard mechanisms in plants (Kagan, 2014; Schnappauf et al., 2019). Pyrin senses bacterial virulence through cytoskeletal remolding. When stimulated by the toxin or bacterial infection Rho modification Ser-205 and Ser-241 are dephosphorylated and induce 14-3-3 disassociation to stimulate pyrin activation and pyrin-ASC-inflammasome complex (Kagan, 2014). Pyrin can associate with cytoskeletal microtubules and actin filaments suggesting that innate immune component can interact with the cytoskeleton (Kagan, 2014). REM sleep deprivation shows the importance of organization of neural cytoskeletons in that increase in tubulin associated unit (TAU) and microtubule-associated protein (MAP2) cytoskeleton after 24 h of sleep deprivation in the frontal cortex (Rodríguez-Vázquez et al., 2012). Pyrin expression is upregulated by sleep promoting mediators including LPS, IFN-gamma, TNFα and cytokines that have anti-inflammatory properties that also function to attenuate sleep responses to somnogenic stimuli such as IL-10 and IL-4 (Zielinski and Gibbons, 2022; Wouters et al., 2023).

7.11. CARD8

CARD8 inflammasomes contain a FIIND domain that auto-cleaves for inflammasome activation and are involved in HIV infection (Gong et al., 2021; Clark et al., 2023). CARD8 is not found in rodents, but present in humans and bypasses ASC to activate caspase-1 directly (Karakaya et al., 2024). CARD8 is ubiquitously expressed in many tissues and cell types including the brain (Sjöstedt et al., 2020; Karakaya et al., 2024). CARD8 possesses pro- and anti-inflammatory activities but are not well understood (Karakaya et al., 2024). For example, proteasome 20S can degrade the N-terminal fragment enabling zonula occludens-1 and uncoordinated-5 (ZU5), UPA, and CARD domains to inhibit inflammasome assembly (Hsiao et al., 2022). Furthermore, CARD8 N-terminal site stimuli can function to attenuate caspase-1 (Karakaya et al., 2024). Dipeptidyl peptidases inhibitors DPP8 and DPP9 inhibitors can enhance the degradation of CARD8 to activate CARD8 inflammasomes (Chui et al., 2020). CARD8 can inhibit NLRP3 inflammasome activation and pro-inflammatory signaling, partially, through the FIIND of CARD8 and the NACHT domain of NLRP3, CARD8 interaction with caspase-1 and caspase-1-inhibiding CARD-only proteins, and CARD8 inhibition of NF-κB by several mechanisms such as the interaction with NEMO (Karakaya et al., 2024). Consequently, targeting CARD8 could be a potential strategy for sleep dysregulation from HIV infection since profound persistent sleep dysregulation occurs after HIV infection and the downstream targets of CARD inflammasome activation, such as IL-1β, affect sleep (Lee et al., 2012; Zielinski and Gibbons, 2022).

7.12. Cyclic GMP-AMP synthase (cGAS)/STING

Bacterial DNA is recognized by the CpG motif or viral DNA and RNA are recognized as foreign to induce innate immune responses (Liu et al., 2024). cGAS/STING is also conserved to simple life-forms including prokaryotes (Liu et al., 2024). The cGAS-STING pathway recognizes cytosolic DNA from intracellular bacterial and viral infection by sensing double stranded DNA viruses (Cai and Imler, 2021). cGAS is activated by guanosine triphosphate (GTP) and ATP to form cGMP-AMP that binds to STING (Liu et al., 2024). This process induces the phosphorylation of IRF2 by TANK-binding kinase-1 (TBK1) that then triggers the transcription of inflammatory genes including IFN-beta. STING can activate NF-κB (Liu et al., 2024). cGAS-STING signaling also induces inflammasome activation and pyroptosis (Liu et al., 2024). Furthermore, caspase-1, Gasdermin D, ASC, and the potassium channel regulate the cGAS-STING pathway (Liu et al., 2024). Although cGAS-STING, inflammasomes, and pyroptosis are independent signaling pathways, crosstalk between these pathways is important in mounting proper immune responses to pathogens (Liu et al., 2024). In AIM2 KO mice, mycobacterial infection reduced IFN gamma, in part, by reducing STING to increase more severe infection (Yan et al., 2018). AIM2-like receptors (ALRs), which have a pyrin domain that mediates protein-protein interactions and a HIN domain that directly binds to DNA are also involved in the cGAS-STING pathway (Liu et al., 2024). NLRP3 deficiency increases the production of type 1 IFN in response to Zika virus in vitro and in vivo (Zheng et al., 2018). Canonical and noncanonical inflammasomes caspase-1 activation interacts with cGAS to inhibit cGAS-STING-mediated type I IFN production (Liu et al., 2024). This is seen with cGAS cleavage after Zika virus infection by caspase-1 inhibiting the phosphorylation of TBK1 and IRF2 and reduced type 1 IFN production to reduce responses to the viral infection (Wang et al., 2017). Evidence also suggest that ASC is required for the regulation of the cGAS-STING signaling pathway (Wang et al., 2017). NLRs are also involved in the cGAS-STING pathway. ATP binding by NLRC3 reduces the interaction with STING to reduce IFN and IL-6 (Li et al., 2019b). The cGAS-STING pathway is required for cytoplasmic DNA-induced NLRP3 activation from viral and bacterial infections (Gaidt et al., 2017). STING-IRF3 is activated by LPS to induce inflammation by activating NLRP3 in mice (Li et al., 2019a). Studies report that STING can recruit NLRP3 to the endoplasmic to promote NLRP3 inflammasome formation after HSV-1 infection (Wang et al., 2020a). The transmembrane region 5 (TM5) location of STING interacts with NACHT and LRR domain in NLRP3 to inhibit NLRP3 polyubiquitination, which induces STING deubiquitinated of NLRP3 inflammasome formation. Microglia express cGAS-STING as evident by microglia cGAS-STING activation enhances neuroinflammatory responses after TBI through the NLRP3 inflammasome (Zhang et al., 2022). That cGAS-STING inflammasomes interact with other inflammasomes or component, such as AIM2 and NLRP3, and modulate type 1 IFNs, IL-1β, and IL-6, there exists a potential for cGAS-STING inflammasomes activation by pathogens is involved in dysregulating sleep.

8. Bacteria

In multiple species including rats, rabbits, dogs, and goats, the gram-positive bacterial cell wall component muramyl dipeptide increases NREM sleep (Zielinski and Krueger, 2011). Similar enhancements in NREM sleep are found in mice, rats, rabbits, goats after central or peripheral administration of the gram-negative bacterial cell wall component LPS is applied (Zielinski and Krueger, 2011). Interestingly, SWA tends to increase when LPS is applied centrally but is reduced when LPS is applied to the periphery—a finding that is consistent when IL-1β or TNFα proteins are applied (Zielinski and Gibbons, 2022). In humans, NREM sleep increases when Salmonella abortus equi endotoxin (0.4 ng/kg) was given, although SWA was not changed (Pollmächer et al., 1993). Using a larger dosage (0.8 ng/kg) of Salmonella abortus equi endotoxin, NREM sleep was attenuated and waking and daytime sleepiness increased (Hermann et al., 1998). Interestingly, when LPS is applied there is a preferential increase in NREMS at the expense of REMS in most species. Mice lacking NLRP3 have greatly attenuated NREM sleep and SWA responses to LPS applied intraventricularly (Zielinski et al., 2017). Rats given an intracerebroventricular pretreated with the caspase-1 inhibitor Ac-Tyr-Val-Ala-Asp chloromethyl ketone exhibited reduced NREM sleep responses to intraperitoneal injected LPS (Imeri et al., 2006). Transgenic mice expressing the human IL-37b have attenuated NREM sleep enhancements after LPS compared to WT control mice (Davis et al., 2017). Staphylococcus aureus is a gram-positive coccus that can activate NLRP3 inflammasomes (Melehani and Duncan, 2016). Inoculating rabbits intravenously with Staphylococcus aureus infection increased NREM sleep and amplification of slow waves in the proceeding 10 h with attenuations in REM sleep (Toth and Krueger, 1988). However, sleep responses were attenuated below baseline values thereafter until the end of the 48-h recording (Toth and Krueger, 1988). Rabbits intravenously inoculated with Streptococcus pyogenes or C. albicans exhibited increased NREM sleep and slow wave activity with reductions occurring 24–38 h post-infection (Toth and Krueger, 1989). A similar pattern of increased NREM sleep and SWA responses followed by reductions in these sleep-related variables were found after inoculation of E. coli was given intravenously to rabbits, although the timing of these changes differed (Toth and Krueger, 1989). The gram-negative bacteria Pasteurella multocida inoculation in rabbits resulted in increased NREM sleep followed by reductions in NREM sleep compared to baseline measures after different routes of administration including intravenous, intramuscular, subcutaneous, or intranasal inoculation in rabbits suggesting that there are multiple peripheral routes that bacterial infection can affect sleep (Toth and Krueger, 1990). In humans, Borrelia burgdorferi is a gram-negative bacterial species of spirochete that causes Lyme disease, which can induce meningitis infections can induce sleep apnea and increased waking arousals and increase sleep amounts, with more daytime sleep (Milhaud et al., 1999). Pneumococcal meningitis also can affect respiration during both REM sleep and NREM sleep (Hasegawa et al., 1995).

9. Viruses

Viruses contain RNA or DNA that is covered by a protein layer capsid that is not encompassing the infected cell during replication (Louten, 2016). Viruses containing the dsDNA, ssDNA, dsRNA, ssRNA, or nucleic acid genome activate multiple inflammasome types and can affect sleep and SWA (Zielinski and Krueger, 2011; Jacobs and Damania, 2012). Mice inoculated with influenza typically tend to have increased NREM sleep and SWA and reductions in REM sleep. Nevertheless, differences in sleep responses to influenza are seen between genotypes where C57BL6 mice have greater NREM sleep responses to infection than BALB/c mice (Fang et al., 1995, 1996; Toth et al., 1995; Toth and Williams, 1999; Toth and Verhulst, 2003). Interestingly, while C57BL6 mice tend to have a greater T helper (Th1) lymphocyte responses and IL-6, MCP-1, and IL-1β to infectious stimuli such as influenza, BALB/C mice tend to have greater Th2 responses which suggest that genetic differences in genotypes contributes to the sleep differences from infection that produce cytokines that are involved in inflammasome activation (Zhao et al., 2014). Swiss-webster showed enhancements in NREM sleep and suppressions of REM sleep after a lethal (H1N1) and nonlethal strain (H3N2) intranasal infections (Fang et al., 1995). Sleep is enhanced for multiple days and IL-1β, TNFα, IL-6, and interferon are enhanced in the lung and brain after influenza infection (Zielinski and Krueger, 2011). TNF double receptor KO mice have reduced NREM sleep after influenza infection (Kapás et al., 2008). Mice lacking macrophage inflammatory protein1α that is involved in microglia regulation have reduce increased sleep during the dark phase after influenza infection (Toth and Hughes, 2004). The neuron specific interleukin-1 receptor accessory proton is required for hemostatic sleep and sleep responses to H1N1 influenza in mice (Davis et al., 2015). Transgenic mice expressing the human IL-37b have slower and reduce NREM and REM sleep enhancements from a mouse adapted H1N1 influenza virus compared to wild-type mice (Davis et al., 2017). Mice deficient in TLR3 have attenuated influenza virus sickness behavior in response to a mouse-adapted X-31 influenza virus including attenuated increases in TLR3 KO mice and faster returns to baseline values (Majde et al., 2010). Mice infected in the respiratory tract with influenza that were sleep deprived, exhibited a lack of viral clearance difference between unimmunized (Brown et al., 1989). Total lung RNA from mice infected with influenza virus infused into rabbit brain induced increased sleep and slow wave amplitude (Majde et al., 1991). NLRP3 has been shown to sense viral RNA during influenza A viral infection in mice (Allen et al., 2009). Influenza A viral infection increased the expression of IL-1β in the bronchoalveolar lavage fluid of wild-type mice, but not NLRP3 KO mice, caspase-1 KO mice, or ASC KO mice, suggesting the importance of inflammasome proteins in the antiviral response (Allen et al., 2009; Malinczak et al., 2021). Along with this, in wild-type mice infected with intranasal H1N1 influenza virus, an expected enhanced sleep response was observed. In contrast, IL1 receptor accessory protein, alternative splice variant (AcPb) knockout mice had less sleep after influenza challenge compared to their own baseline values and compared to wild-type mice (Davis et al., 2015). In addition to NLRP3, AIM2 inflammasome expression and activation have also been described for the influenza virus (Zhang et al., 2017). Experimentally induced influenza A and B and rhinovirus-induced common colds in healthy participants led to an increase of self-reported sleep duration in the symptomatic phase of the illness (Besedovsky et al., 2019). Rhinovirus-type-23-induced common colds disrupted sleep in symptomatic individuals, with reduced total sleep time and reduced sleep efficiency in the active phase of the illness (Drake et al., 2000). Additionally, a study in humans showed that individuals with short sleep duration had higher incidence of symptoms of cold and flu implicating sufficient sleep with infection, although another study did not see this relationship (Orzech et al., 2014; Ghilotti et al., 2018).

Sleep disturbances with various types of hepatitis are found in some but not all studies (Ghilotti et al., 2018). Based on objective sleep assessment using actigraphy, increased nocturnal wake time and reduced sleep efficiency have been found in women infected with hepatitis C virus HCV (Heeren et al., 2014). Notably, sleep alterations in this study were independent of the viremic state, suggesting that the sleep-immune relationship in hepaticus C virus is complex. It has been reported that over half of patients with chronic hepatitis C report to have sleep disturbances and daytime sleepiness (Heeren et al., 2014). Individuals with hepatitis B also report greater incidence of sleep disorder compared to healthy subjects (Wu et al., 2023). The mechanisms behind the various hepatitis effects on sleep remain unknown, although evidence suggests NLRP3 inflammasomes are activated by hepatitis (Wan et al., 2024).

SARS-COV and COVID 19 cause increased daytime sleepiness and insomnia (Datta and Tripathi, 2021). Additionally, approximately 40 % of people with long COVID report sleep issues such as insomnia, daytime sleepiness, waking up multiple times during the night and not feeling refreshed in the morning. SARS-COV and COVID 19 can activate NLRP3 inflammasomes with substantial upregulation of caspase-1, ASC in microglia (Albornoz et al., 2023; Wang et al., 2023). RS-CoV-2 infection triggered NLRP1 inflammasome response in human lung epithelial cells (Planès et al., 2022). SARS-CoV-2 is an enveloped ssRNA virus with a genome of approximately 30 kb. Established studies have proven that SARS-CoV-2 represents a PAMP able to trigger inflammasome activation (Albornoz et al., 2023). SARS-CoV-2 infection initiated robust NLRP3 inflammasome activation in microglia (Albornoz et al., 2023). Moreover, SARS-CoV-2 S protein primed inflammasome assembly in response to diverse stimuli (e.g., ATP and nigericin) via the NF-κB signaling pathway. Astrocytes and microglia are two main sources of proinflammatory cytokines and are critical for SARS-CoV-2-induced neuroinflammation (Kwon and Koh, 2020).

Individuals with human immunodeficiency virus (HIV) infections report changes in sleep including, insomnia, and increased wakefulness, reduction in slow-wave activity, and changes in sleep-spindles and K-complex density reductions (Kubicki et al., 1989). Sleep changes from HIV infections often occur early after infection. However, other studies report enhancements in NREM sleep (Norman et al., 1992). Notwithstanding, sleep assessment studies in HIV-infected individuals is often difficult due to potential confounds related to medications and stress. NLRP3, IFI16, CARD8, and NLRC4 are activated by HIV infection by various mechanisms including HIV RNA, ROS, potassium efflux, HIV envelope glycoprotein gp120, HIV-1 trans-activator of transcription (Tat), HIV-1 incomplete reverse transcripts, HIV protease, and HIV gp41 in various peripheral and central cell types including microglia (Wang and Shan, 2022). HIV polyprotein precursor of 160 kDa (gp160) is cleaved into a surface exposed amino terminus subunit gp120 and a carboxyl transmembrane subunit gp41. The envelope protein gp120 is a glycoprotein that is part of the capsid protein and is exposed on the surface of the HIV envelope and is essential for virus entry into cells. Evidence suggests that NLRP3 is required for gp120-induced neuroinflammation in the pathogenesis of HIV (He et al., 2020). HIV also triggers the TNF signaling pathways (Pasquereau et al., 2017). In rats, intraventricular administration of recombinant HIV-1IIIB gp120 increased NREMS and REM sleep after a lower dosage (100 ng) but NREMs was further enhanced for a longer period after a 500-ng dosage (Opp et al., 1996). Administering recombinant HIV gp160 or gp41 increased NREM sleep, fragmented sleep, and altered slow-wave frequencies. Interestingly, gp160 can activate NLRP3 inflammasomes while gp41 can activate NLRC4 inflammasomes (Gemma and Opp, 1999). Feline immunodeficiency virus- (FIV) is a lentivirus that exhibits similarities in etiology to HIV infection in humans. Interestingly, FIV infection in cats increased NLRP3, caspase-1, and IL-1β expression in the cortex (Walsh et al., 2014). FIV infected cats had increased time spent awake and reductions in REM sleep, and more transitions between sleep and wake than controls. Additionally, the FIV infected cats had increased spindle activity during NREM sleep, which is similarly found in humans infected with HIV (Prospéro-García et al., 1994). EcoHIV is an HIV-1 target in mice with a coding region of gp120 in HIV-1/NL4-3 replaced by gp80 from an ectopic murine leukemia virus that only infects rodents, increases MCP-1, STAT1, IL-1beta, and complement component C3 in brain tissue (Potash et al., 2005). EcoHIV mice assessed with a piezoelectric sleep-like state sensing system found reductions in sleep-like state amounts in EcoHIV mice (Bell et al., 2022).

10. Parasites

Parasites are organisms that live in or on another organisms, require another organism, and can be detrimental or not to the host. Human African trypanosomiasis such as the hemoflagellate Trypanosoma brucei, Trypanosoma cruzi, and Trypanosoma brucei gambiense infect mammals including humans by flies. Sleeping sickness is found with human African trypanosomiasis and is associated with substantial increases in sleep, narcoleptic-like events, and dysregulated circadian rhythms (Buguet et al., 2005). Interestingly, IL-1β and TNFα gene expression is increased in rats infected with Trypanosoma brucei (Quan et al., 1999). Further, intracerebroventricular infusion of IL-1β, and TNFα antagonist attenuates neurodegeneration from Trypanosoma brucei (Quan et al., 2003). Plasmodium are unicellular eukaryotes that are parasites of vertebrates and insects such as mosquitos (Sato, 2021). Malaria is caused by a Plasmodium that causes sleep disturbances including sleepiness (Sato, 2021). Plasmodium can alter many mechanisms that affect sleep including disrupting the blood brain barrier (BBB), enhancing ROS, activating NF-κB, prostaglandins, Toll-like receptors, IL-1β, and TNFα (Wang et al., 2024). In particular, NLRP3 and IL-1β are activated in the cortex after Plasmodium infection and a protective effect of ginsenoside Rh2 against Toxoplasma gondii occurred through inhibiting microglial NLRP3 inflammasome signaling (Wang et al., 2024). Interestingly, it has been proposed that increased sleep after parasite infection is adaptive for resistance, which is consistent with a study that showed impairments in immunity with sleep deprived rats infected with Trichinella spiralis and increases in NREM sleep after 15 days of infection (Preston et al., 2009; Ibarra-Coronado et al., 2015). Lyme disease is parasite disease caused by the bacteria Borrelia burgdorferi and induces insomnia, difficulty falling asleep, nocturnal awakenings, excessive daytime sleepiness, and sleep fragmentation (Greenberg et al., 1995). The mechanisms governing changes in sleep from Lyme disease are not known but evidence indicates that cytokines including IL-1β and TNFα and multiple signaling pathways inducing PRRs and the complement system are upregulated by Borrelia burgdorferi in astrocytes and neurons (Myers et al., 2009).

11. Conclusion

James Krueger and colleagues have provided great insights and a foundation of how the innate immune system is critical to sleep regulation and sleep responses to infection. However, currently, there remains much that is not understood about the specific mechanisms that regulate innate immunity and infection. Strides have been made involving specific cytokines, such as IL-1β and TNFα, and NLRP3 inflammasomes in sleep responses to pathogenic stimuli and their components. Intriguingly, these basic immune, oxidative stress, and energy-related pathways are largely conserved between complex organisms such as humans to single cell organisms and mediate immune responses to infection. These molecules also appear to regulate sleep responses to infection. Yet, the specific mechanisms responsible for altering sleep responses to particular pathogens remains unknown and understanding how the specific components and types of inflammasomes function can allow for targeted treatments to reduce off target effects, improve outcomes, and long-lasting damage from infections.

Funding

This research and preparation of the report was supported by the Department of Veterans Affairs grant I01BX002823 (MZ).

CRediT authorship contribution statement

Mark R. Zielinski: Writing – review & editing, Writing – original draft, Conceptualization. Sean D. Carey: Writing – review & editing, Writing – original draft, Conceptualization. John A. Craig: Writing – review & editing, Writing – original draft, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.