Authors: Bala S. C. Koritala (1Division of Pediatric Otolaryngology-Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 2Department of Otolaryngology-Head and Neck Surgery, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA), Laetitia S. Gaspar (1Division of Pediatric Otolaryngology-Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 3Center for Neuroscience and Cell Biology, University of Coimbra, 3004-504 Coimbra, Portugal), Shweta S. Bhadri (1Division of Pediatric Otolaryngology-Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA), Kyla S. Massie (1Division of Pediatric Otolaryngology-Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 4University of California San Diego, San Diego, California, 92093, USA), Yin Yeng Lee (5Department of Pediatrics, Division of Human Genetics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 6Department of Pharmacology and Systems Physiology, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA), Jiffin Paulose (5Department of Pediatrics, Division of Human Genetics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA), David F. Smith (1Division of Pediatric Otolaryngology-Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 2Department of Otolaryngology-Head and Neck Surgery, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA; 7Division of Pulmonary Medicine, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 8The Sleep Center, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA; 9The Center for Circadian Medicine, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH 45229, USA)
Categories: Article, Intermittent hypoxia, obstructive sleep apnea, immune response, pro-inflammatory mediators, animal models
Source: The Laryngoscope
Doi: 10.1002/lary.30915
Authors: Bala S. C. Koritala, Laetitia S. Gaspar, Shweta S. Bhadri, Kyla S. Massie, Yin Yeng Lee, Jiffin Paulose, David F. Smith
Obstructive sleep apnea (OSA) is characterized by chronic systemic inflammation; however, the mechanisms underlying these pathologic consequences are incompletely understood. Our objective was to determine the effects of short- versus long-term exposure to intermittent hypoxia (IH) on pro-inflammatory mediators within vulnerable organs impacted by OSA.
Experimental animal study
8–10 week old C57BL6/J mice were exposed to normoxic or IH conditions for seven days (short-term) or six weeks (long-term) under 12 h light, 12 h dark cycles. After exposure, multiple tissues were collected over a 24 h period. These tissues were processed and evaluated for gene expression and protein levels of pro-inflammatory mediators from peripheral tissues.
We observed a global decrease in immune response pathways in the heart, lung, and liver compared to other peripheral organs after short-term exposure to IH. While there were tissue-specific alterations in the gene expression of pro-inflammatory mediators, with down-regulation in the lung and up-regulation in the heart, we also observed reduced protein levels of pro-inflammatory mediators in the serum, lung, and heart following short-term exposure to IH. Long-term exposure to IH resulted in an overall increase in the levels of inflammatory mediators in the serum, lung, and heart.
We demonstrated novel, longitudinal changes in the inflammatory cascade in a mouse model of OSA. The duration of exposure to IH led to significant variability of inflammatory responses within blood and cardiopulmonary tissues. Our findings further elucidate how inflammatory responses change over the course of the disease in vulnerable organs.
Although the anatomic factors that are associated with obstructive sleep apnea (OSA) can vary, this disorder is universally characterized by episodes of airway obstruction during sleep, leading to periods of oxygen desaturation followed by recovery to normal levels (intermittent hypoxia). The worldwide prevalence of OSA is now estimated to be over one billion people; greater than 50% of the population in some countries.^1^ Untreated OSA is associated with a host of medical conditions, including cardiovascular disease,^2,3^ metabolic dysregulation,^4^ neurocognitive deficits,^5,6^ and even increased mortality.^7^ In 2015 alone, the estimated cost for diagnosing and treating OSA in the US was greater than 12 billion.^8^ However, the cost burden of undiagnosed OSA, including factors such as lost productivity and treatment of the complex comorbid diseases, reached almost 150 billion.^8^ As OSA becomes more prevalent in subsequent years,^9^ much effort has been aimed at better understanding this disease. Without a basic grasp of the molecular mechanisms that lead to the associated comorbidities, our ability to identify new diagnostic and therapeutic strategies is significantly hindered.
Recurrent episodes of obstruction and hypoxia, followed by reoxygenation during sleep, are thought to induce systemic inflammation,^10,11^ oxidative stress,^12,13^ and endothelial dysfunction.^14,15^ These low oxygen conditions within organs lead to changes in regulation of molecular pathways, including nuclear factor-kB (NF-kB) and hypoxia-inducible factor (HIF)-1-dependent signaling.^16–18^ The HIFs, specifically, are transcription factors that are stabilized under low oxygen conditions to activate transcriptional programs that operate at the cellular, tissue, and whole organism level. Through transcriptional regulation, changes in these pathways lead to upregulation of pro-inflammatory responses and altered inflammatory cytokine levels.^19^
These pro-inflammatory responses are thought to be regulated by the activation of immune cells, such as lymphocytes and monocytes, that then secrete a host of inflammatory mediators.^11^ Studies have demonstrated altered levels of a number of specific inflammatory mediators, including tumor necrosis factor (TNF)-α, interleukin (IL)-6, and C-reactive protein (CRP) in patients with OSA.^20–24^ Even young children with OSA demonstrate changes in these inflammatory biomarkers compared to healthy children.^10,25,26^ Interestingly, the clinical effects of these same inflammatory markers on blood pressure are dependent on the sleep state of children with OSA compared to healthy control.^25^ Specific inflammatory cytokines and acute phase reactants exhibit diurnal changes in serum levels in children with OSA.^25^
One of the major limitations in studying inflammatory responses in patients with OSA involves our inability to evaluate tissue-specific changes over time. Patients recruited to clinical studies have had months to years of exposure to sleep-dependent obstruction and intermittent hypoxic episodes prior to evaluation and treatment, making it difficult in humans to understand how inflammatory responses become dysregulated over the course of the disease. Animal models representing OSA have allowed researchers to study physiologic responses to IH at these organ and system levels. Models using IH in mice have even demonstrated cellular-specific consequences to these gene expression pathways, revealing potential candidates for cell-targeted therapy.^27^ In our study, we evaluated the tissue-specific responses of pro-inflammatory mediators over short- versus long-term / chronic exposure to IH in tissues that demonstrate negative sequelae in patients with OSA, including serum, lung, and heart. We hypothesized that pro-inflammatory responses would demonstrate tissue-specific variability over the course of the disease.
Prior to initiation of our study, Institutional Animal Care and Use Committee (IACUC) approval was obtained from our institution (#IACUC2019–0028). C57BL/6J male mice (#000664) were purchased from the Jackson Laboratory and entrained for 14 days under a 12 h light / 12 h dark cycle (LD 12 h). Mice aged between 8 and 10 weeks were housed under LD 12 h during exposure to normoxic (room air) or intermittent hypoxic (IH) conditions (~50 episodes / hour of oxygen desaturation ranging between 21% and 8%) for either seven days (short-term) or six weeks (long-term / chronic exposure). Using diagnostic criteria for patients with OSA,^28^ our model represents clinically severe disease. Given the nocturnal nature of mice (active during the night and inactive / rest during the day), IH exposure was given only during the light / inactive phase for mice using an OxyCycler A84XOV (Biospherix, Parish, NY, USA). The overview of the study design is presented in Fig. 1. During the experiment, mice were given ad libitum access to food and water. After exposure to normoxia or IH, the mice from both conditions were released to constant dark until end-organ collection, as a means to measure persistent changes in gene and protein expression levels independent of light-induced effects. On the second day of constant darkness, we collected liver, lung, kidney, muscle, heart, cerebellum, and blood. Given the changes that occur in inflammatory mediators throughout the day, we collected tissues over a period of 24 hours at three-hour intervals to calculate 24-hour mean expression. The tissues were snap frozen in liquid nitrogen, homogenized with mortar and pestle, and stored at −80°C for further assays. The blood was collected using cheek bleeds into a 1.5 ml Eppendorf tube containing 50 μl of 0.5 M EDTA. Blood was further centrifuged at 1500 × g for 30 mins at 4°C. The supernatant was collected and preserved at −80°C for further use. The protein levels of the pro-inflammatory mediators were measured from the tissues collected at ZT 30 (ZT 0 = start of light phase) following short-term vs long-term IH exposure. At least three mice were used for each condition across all time points in our short- and long-term experiments.
TRizol cocktail was prepared by mixing 100 mg of grounded tissue with 1000 μl of ice-cold TRIzol^™^ (Invitrogen^™^, #15596018, Carlsbad, CA, USA). The cocktail was then treated with 200 μl of chloroform (Fisher Chemical^™^, #C298–500, Fair Lawn, NJ, USA). The samples were mixed thoroughly and incubated for 3 minutes at room temperature, then centrifuged at 12,000 × g for 15 min at 4°C. A clear aqueous phase was collected and mixed with isopropanol (Sigma Aldrich, #437522–4L, Saint Louis, MO, USA). The samples were inverted to mix and then incubated at room temperature for 10 minutes, followed by centrifuging at 12,000 × g for 10 minutes at 4°C. After discarding the supernatant, the pellet was washed three times in 70% ice-cold ethanol at 12,000 × g for 10 min at 4°C. After washing, pellets were air dried for 10 min and dissolved in 25 μl of UltraPure^™^ DNase/RNase-Free Distilled Water (Invitrogen^™^, #10977015, Carlsbad, CA, USA).
RNA-Seq was performed for each tissue at the Genomics Center of Rutgers New Jersey Medical School with Illumina Novaseq 6000 S4 Reagent Kit v1.5 (Illumina, #20028312, California, United States). The RNA input was measured using a Qubit 2.0 instrument (Thermo Fisher Scientific, #Q10211, Waltham, MA, USA). Before library preparation, RNA integrity was measured with Agilent 2200 (Agilent Technologies, #5067–5576, Santa Clara, CA, USA). Samples with RINe values of 8 or higher were used to prepare libraries. A total of 1 μg of RNA was used for the preparation of the library using Poly (A) module of NEBNext^®^ UltraTM II Directional RNA Library Prep Kit for Illumina^®^ (NEBNext, #E7765L, Ipswich, MA). The Qubit was used to quantify libraries (Thermo Fisher Scientific, #Q33231, Waltham, MA, USA), and the TapeStation 2200 with D1000 tape (Agilent Technologies, #5067–5582, Santa Clara, CA, USA) to verify their quality. Data quality was assessed using Illumina’s SAV. The data was demultiplexed using Illumina Bcl2fastq2 Conversion Software v2.17. Sequence quality was assessed using FASTQC (v0.11.3). The raw FASTQ reads of each sample were aligned and quantified by Kallisto using the mouse reference genome, release-102, provided by ENSEMBL. We quantified the differences between normoxic versus IH conditions using estimated Kallisto counts. Differential 24-hour mean expression was quantified using the DESeq2 R package. The statistical cutoff of q ≤ 0.05 was used to evaluate significant differences between the two conditions. These samples were also screened for the protein array panel of cytokines, chemokines, and growth factors from Biotechne^®^ R&D Systems.
Pathway enrichment analysis was performed for genes significantly up- or down-regulated with at least a 1.5-fold difference upon IH exposure. In this study, we used DAVID Bioinformatics Resource to enrich tissue-specific differentially regulated genes for the Gene Ontology Biological Processes.
Approximately 100 mg of ground lung or heart tissue were further homogenized in 500 μl of tissue lysis buffer (1% v/v Triton X-100 in 1 × PBS protease inhibitor cocktail) using a dounce homogenizer. We then quantified proteins of each sample using the Pierce^™^ Rapid Gold BCA protein assay kit (Thermo Fisher Scientific, #A53226, Waltham, MA, USA), following the microplate procedure in the manufacturer’s protocol. Absorbance was measured at 480 nm using a Synergy Neo2 plate reader.
The protein expression of cytokines, chemokines, and growth factors were evaluated using the Mouse XL cytokine array kit (Biotechne^®^ R&D Systems, #ARY028, Minneapolis, MN). This kit was able to quantify 111 proteins in multiple tissues, including lung, heart, and serum, through dot blot. We used this proteome profiler to quantify proteins from the tissues (dark / active phase) after exposure to normoxic or IH conditions. Approximately 300 μg of protein from lung and heart and 100 μl of serum were used for this array. A standard manufacturer’s protocol from Biotechne^®^ R&D Systems was used to perform this array. To quantify protein expression, 200 μl of chemiluminescent reagent mixture was added to develop the array membrane and incubated for 60 sec. At 202.80 sec exposure, Bio-rad Chemidoc recorded the chemiluminescence signal from the membrane. The analysis was performed using Bio-Rad Image Lab Software v6.1 with three biological replicates of each condition for short- and long-term / chronic experiments.
To examine the impact of IH on tissue-specific biology, differential gene expression analysis was performed for multiple tissues including the organs associated with cardiovascular dysfunction. We further enriched the changes seen in gene expression through Gene Ontology (GO) biological process and assessed the impact of IH on each organ-level expression profile. We found that the most biological processes were dysregulated within cardiopulmonary tissues, lung more so than heart, as compared to other organs. 9 out of 14 biological processes in the lung and 5 of 14 in the heart were impacted after IH exposure. However, at least 3 of the 6 tissues in our experiment showed dysregulation of metabolic processes, immune responses, and transcriptional regulation. In particular, the liver, lung, and heart exhibited a down-regulation of immune response pathways (Fig. 2). Based on our findings, we then evaluated the impact of IH on pro-inflammatory mediators known to be dysregulated in patients with OSA.
We examined gene expression profiles for cytokines, chemokines, and growth factors following IH exposure. Gene expression was assessed for those inflammatory mediators covered by the Mouse XL Cytokine Panel of R&D systems. A total of 73 genes were detected in the lung and 59 genes in the heart. 60.2% of inflammatory mediators in the lung were differentially expressed after IH, whereas only 38.9% were impacted in the heart. Interestingly, a majority of these mediators were down-regulated in the lung. In contrast, more genes were up-regulated in the heart (Fig. 3A, 3B, and Suppl. Table 1).
Our analysis revealed that exposure to IH had varying effects on different pro-inflammatory mediators in serum which were dependent on the duration of exposure. Short-term exposure led to reduced levels of several pro-inflammatory mediators, while long-term exposure generally led to increased inflammatory mediator levels (as demonstrated in Fig. 4A–C and Suppl. Table 2). However, some targets showed increased levels of inflammatory mediators after both short- and long-term exposure, such as Pentraxin 2/SAP and vascular endothelial growth factor (VEGF), while others, including insulin-like growth factor-binding protein (IGFBP) 5, intercellular adhesion molecule (ICAM) 1 / cluster of differentiation (CD) 54, epidermal growth factor (EGF), and IGFBP1, were decreased in serum after both short- and long-term exposure to IH. Specifically, IGFBP5, VEGF, CD40, MIP-2, and thymidine phosphorylase showed significant changes in serum (as shown in Fig. 4A–C and Suppl. Table 2).
Following our observation that the lung displayed the most significant changes in gene expression profiles (Fig. 3), we further investigated the protein levels of inflammatory mediators in these tissues. Our results demonstrated that short-term exposure to IH generally led to a decrease in inflammatory mediator levels, except for EGF, adipocyte complement-related protein (ACRP) 30, and chitinase 3-like (CHI3L) 1. Specifically, the levels of interleukins (IL-1ra, IL-F2, IL-2, IL-3, IL-7, and IL-22), macrophage proteins (CD14), granulocyte-macrophage colony-stimulating factor (GM-CSF), and macrophage inflammatory protein (MIP) 1a/B and MIP-3B were significantly decreased after short-term exposure to IH. Other targets, including cellular communication network factor (CCN) 4, C-reactive protein (CRP), chemokine (C-X-C motif) ligand (CXCL) 10, complement (C) 5a, Dickkopf (DKK) 1, granulocyte-colony stimulating factor (G-CSF), human growth factor (HGF), resistin, and thrombopoietin were also significantly decreased after short-term IH exposure (Fig. 5A and B and Suppl. Table 2).
In contrast, long-term / chronic IH exposure generally increased the levels of inflammatory mediators, with some exceptions. For example, EGF, Pentraxin, and IGFBP6 levels increased, while receptor for advanced glycation end products (RAGE), regenerating family member (Reg) 3G, IGFBP-2, and CD105 protein levels decreased. Notably, the levels of several interleukins (IL-3, IL-5, IL-6), MIP-2, osteoprotegerin, and thymidine phosphorylases were significantly increased only after long-term exposure to IH (Fig. 5A and 5C and Suppl Table 2). The levels of Reg3G and CD105 constitutively decreased, while EGF was constitutively increased, after both short- and long-term exposure to IH (Fig. 5B and 5C and Suppl Table 2).
We also examined the protein levels of inflammatory mediators in the heart, which is another pertinent organ of the cardiopulmonary system. We found that protein levels of inflammatory mediators were generally lower after short-term exposure to IH but higher after long-term exposure, following the same trend as both serum and lung. Particularly, short-term exposure to IH led to a significant reduction of IL-11, IL-13, Pentraxin 3, Fibroblast growth factor (FGF) 21, CXCL10, CD160, and IGFBP5. On the other hand, prolonged exposure to IH resulted in a significant decrease of Endostatin protein levels. Among the proteins with extreme differences in the heart, including alpha 2-HS-glycoprotein (AHSG), ACRP30, CD105, and FGF acid, a decreased trend was observed with long-term exposure. Similar to the lung, most of the immune modulators (pro-inflammatory mediators) exhibited subtle changes in protein levels in the heart after long-term exposure to IH (Fig. 6A–C and Suppl Table 2).
OSA is highly prevalent^29^ and is associated with a host of significant comorbid conditions.^30–34^ Translational and clinical research has focused on delineating the basic mechanisms that lead to end-organ damage in at-risk patients with OSA. Much effort has specifically focused on evaluating the heightened inflammatory response in the blood samples of this patient population. However, there is a lack of data comparing inflammatory mediator protein levels in serum to those in vulnerable organs that are at risk for pathologic changes in the heart and lung among patients with OSA. To address this, we used an animal model of OSA to investigate the impact of IH on tissue-specific immune responses in both serum and peripheral organs. Our findings suggest that short-term IH resulted in differential regulation of inflammatory mediators in the heart and lung, with down-regulation of mRNA expression in the lung and up-regulation in the heart. At the protein level, inflammatory mediators showed a global decrease in serum, lung, and heart after short-term exposure, but a trend for global increases after long-term exposure. Dysregulation of immune responses and the inflammatory mediators may contribute to organ dysfunction in patients with OSA, and our animal model provides insight into the effects of short-term versus long-term IH on organs at risk from untreated disease. However, the underlying pathophysiological mechanisms of OSA-related damage remain unclear.
Numerous animal models have been employed in controlled laboratory settings to replicate OSA^35^. These models employ various techniques such as tracheostomy^36^, the use of inflated balloon in the trachea^37^, or a nasal mask to intermittently obstruct the airway^38^. By employing these methods, researchers can simulate the physiological responses observed in OSA, including hypercapnia and the sympathetic response. An alternative, less invasive approach involves manipulating the oxygen levels inhaled by mice within a sealed chamber over a period of weeks to months, exposing them to cycles of a hypoxic gas mixture and ambient air^39,40^. Although each method has its own advantages and disadvantages, the intermittent episodes of oxygen desaturation in the surrounding air offer the advantage of not requiring anesthesia and enabling longer duration of experiments. Typically these experiments involve exposing the mice to air with 21% oxygen followed by desaturation levels ranging from 5% to 15%^41–44^. This oxygen deprivation leads to the lowest levels of oxygen saturation (SaO2) ranging from 50% to 70%. This range is comparable to 70% to 90% SaO2 nadir levels observed in humans diagnosed with OSA^44^. These animal models have demonstrated phenotypic outcomes in animals consistent with clinical sequelae seen in patients with untreated OSA, including changes in sympathetic responses,^45^ altered blood pressure,^46^ and metabolic dysregulation.^47^ The IH models allow for the study of oxygen desaturations and pathways related to changes in the levels of HIFs without the risks for surgical intervention that can scar the airways.^27^
In our current study, we induced IH by reducing the fraction of inhaled oxygen from 21% to 8% over a period of approximately 30 seconds, followed by an immediate recovery period of 15 seconds up to 21% oxygen. This IH regimen was designed to simulate severe levels of OSA commonly observed in humans. While the human body is typically exposed to around 21% oxygen in its natural environment, the oxygen requirements and ability to tolerate periods of hypoxia vary significantly across different tissues. For example, the lung and liver typically receive higher levels of oxygen compared to the muscle^35^. Recent studies have also demonstrated that IH can lead to both time- and tissue-specific responses in a mouse model of OSA^48,49^. Therefore, in our study, we investigated the tissue-dependent changes in immune responses in short-term versus long-term IH exposures. Our findings indicate distinct mRNA expression levels of inflammatory responses to short-term IH, which differed between tissues. Specifically, a significant reduction in the expression of inflammatory mediators was observed in the lung tissue, while an up-regulation was noted in the heart tissue. However, studies suggest that protein levels do not necessarily reflect gene expression due to factors involved in translational processes.^50^ Interestingly, we did not detect any tissue-specific differences in inflammatory mediators at the protein level following short-term IH, as a consistent decrease was seen across all the tissues, including lung, heart, and serum. For example, pro-inflammatory markers, such as TNF- α, IL-6, and CRP, were decreased in serum, lung, and heart to varying degrees. It is important to mention that there are limited studies in the literature that shed light on the progression of inflammatory mediators over time during IH exposure in animal models. IH exposure in rodents demonstrates tissue-specific variations of pro-inflammatory mediators. In a study of rats exposed to IH for 4 weeks, serum pro-inflammatory markers and oxidative stress proteins were suppressed.^51^ Furthermore, female mice exposed to IH for 30 days demonstrate increased mRNA levels for IL-6 in both the brain and the heart.^52^ The findings, though, could be explained by the differences in IH exposure time as well as sex of the mice.
In contrast to what we observed after short-term exposure to IH, we found trends for global increases in many of pro-inflammatory mediators after long-term, chronic exposure to IH, although there was variability in the degree of change in cardiopulmonary tissues versus serum. Using a chronic IH model in rats, He et al. demonstrated increases in inflammatory mediators in rat plasma and adipocyte cell extracts,^53^ further implicating inflammation in insulin resistance that was seen in these animals. As chronic IH represents OSA, we demonstrated similarities between inflammatory responses in our model compared to findings from patients with OSA. For example, Pentraxin-3 protein levels were elevated in serum in mice exposed to chronic IH. These data correlate to elevated Pentraxin-3 levels also seen in humans with OSA.^54^ Cumulatively, our data demonstrate an evolution of the inflammatory responses in serum and peripheral tissues over longer periods of disease burden.
The longitudinal changes that occur to inflammatory mediators could be the result of several physiologic responses. The potential mechanisms contributing to comorbid conditions often seen in patients with OSA are thought to be driven by HIF-dependent signaling pathways. IH stabilizes HIFs, resulting in activation of different immune cells, including neutrophils and macrophages, that regulate production of inflammatory mediators.^55,56^ In a study comparing serum inflammatory mediator protein levels in children with OSA compared to healthy controls, certain pro-inflammatory cytokines, including IL-6 and 8, were significantly higher in children with OSA.^25^ Pro-inflammatory cytokines that play a role in the development of atherosclerosis in adults with OSA are also elevated in children with OSA, but their role in altering vascular stiffness is dampened by the protective effects of acute phase reactants.^25^ These groups of inflammatory mediators appear to orchestrate different physiologic effects in order to maintain balance within the cardiovascular system. If this homeostatic mechanism is overwhelmed, then the balance swings in the direction of a chronic pro-inflammatory state, potentially favoring increasing serum levels of pro-inflammatory mediators.
Recent studies have evaluated the use of inflammatory mediators as biomarkers to screen for OSA or to follow clinical treatment. For example, in a study of 80 patients with OSA, elevated levels of pro-inflammatory mediators, such as IL-6 and TNF-a, were associated with increased thickness of the carotid wall, an early change seen in atherosclerosis.^57^ Although multiple studies have demonstrated associations between a diagnosis of OSA and elevated inflammatory biomarkers,^58,59^ data are inconsistent and varied in how well they predict OSA or response to treatments.^60,61^ This could be due to the presence of other medical comorbidities, such as obesity.^62^ Alternatively, this could simply be due to the fact that the inflammatory responses change over time throughout the course of the disease. Certainly, our data would suggest that a single measurement of inflammatory biomarkers would yield very different results depending on how long exposure to disease has occurred. More research is needed to delineate the significance of these changes in inflammatory responses over time in patients with OSA, especially when considering their roles in diagnostic medicine.
Our study has several limitations that are worth considering. While IH can be used as surrogate for OSA, it does not fully replicate the complexity of OSA. Mice can exhibit central sleep apneic events that resemble aspects of OSA, but they do not naturally experience OSA. The lack of standardization of IH protocols further complicates the interpretation and comparison of research findings among different research groups. In the current study, our model is designed to study pathways both up- and down-regulated after exposure to IH during the sleep/inactive phase. While it is known that sleep fragmentation is highly prevalent among patients with OSA, further studies will be necessary to determine the role of sleep fragmentation in the inflammatory cascade highlighted here, independent of IH exposure. Despite this, our methodology uses a well-established model to evaluate pro-inflammatory responses,^63,64^ endothelial dysfunction,^65^ and oxidative stress pathway dysregulation^66^ as a representation of OSA. Although we identified longitudinal changes in inflammatory mediators of known importance to the development of atherosclerosis in adults with OSA, more work should be done to measure these changes in side-by-side studies in mice that demonstrate end-organ damage and other clear phenotypes associated with untreated OSA.
Further studies can be conducted to build upon these results. A longitudinal study can be performed to track the progression of inflammatory mediators over time during IH exposure, providing insights to the temporal changes associated with the disease. Another avenue for investigation is to explore the role of sleep fragmentation, independent of IH exposure, in the inflammatory cascade. By manipulating sleep patterns and assessing the impact on inflammatory responses, researchers can determine the specific contributions of sleep fragmentation as well. Additionally, a study can be conducted to compare inflammatory responses observed in animal models with end-organ damage in the IH model described. This would help establish the relevance and validity of the IH model in understanding the inflammatory pathways associated with untreated OSA. Mechanistic studies could be used to further evaluate the molecular pathways underlying tissue-specific regulation of inflammatory mediators in OSA, providing a better understanding of the involved mechanisms. Clinical correlation studies can analyze inflammatory mediator levels in patients with OSA and their correlation with disease severity, comorbidities, and treatment responses. Finally, intervention studies can explore the effects of treatments targeting inflammatory responses in OSA, such as anti-inflammatory agents or lifestyle modifications, to assess their efficacy in mitigating OSA-related complications.
In the current study, we demonstrated novel, longitudinal changes in the inflammatory cascade in a mouse model of obstructive sleep apnea. Interestingly, the duration of exposure to intermittent hypoxia led to significant variability of pro-inflammatory responses within blood and cardiopulmonary tissues. Unlike what is seen with short-term exposure to intermittent hypoxia, chronic, long-term exposure resulted in increased pro-inflammatory mediator levels. Our findings further elucidate how the inflammatory responses change over the course of the disease in vulnerable organs.