Yogic breathing when compared to attention control reduces the levels of pro-inflammatory biomarkers in saliva: a pilot randomized controlled trial

Pro-inflammatory cytokines and chemokines are implicated in several diseases and therefore are potential targets for molecular therapy. While non-pharmacological methods of stress reduction have long been in practice among several cultures, recently such practices are gaining acceptance worldwide. Translating information on stress reduction from ancient cultures into modern scientific literature is important to preserve and propagate such wellness practices [1]. Thirumanthiram is an ancient Tamil text written by Saint Thirumoolar who was contemporary to Saint Patanjali [2, 3]. Thirumanthiram (Thiru means holy or sacred or noble, manthiram means chant or mantra or hymn or sutra) consists of over 3000 poems referring to various practices including Ethics, Devotion, and Yoga. In this anthology there are 14 poems (Numbers 564 through 577) dedicated to the practice of Yogic Breathing (YB, Pranayamam or Pranayama). These 14 poems speak in depth about the benefits of YB and explain how breathing is an important connection between mind and body.

YB is one of the several practices within the broad field of Yoga and is known to cause key changes within mind and body including blood pressure reduction, heart rate variability changes, and breathing frequency reductions [4], predominant abdominal/diaphragmatic breathing [5‐7], improved cognitive functions (e.g., mental alertness and reduced cognitive failure) [8, 9], increased bimanual dexterity and visuo-motor co-ordination [10], stress and symptom reduction in diseases such as cancer [11]. Behavioral practices such as relaxation and exercise could bring about meaningful molecular changes among practitioners [12, 13]. Although Yoga practitioners from various schools widely practice YB, the techniques specified by Thirumoolar have not yet been studied for their physiological effects or changes in biomarkers. This study focused on the YB exercise specified in Poem 568 of Thirumanthiram (named Thirumoolar Pranayamam, TMP) that includes an inhalation (Purakam), breath-holding (Kumbakam), and exhalation (Resakam) for specified time periods [3]. To measure the changes in physiological responses to TMP, saliva was used to measure biomarkers. Saliva is an easily accessed non-invasive biologic sample rich in various biomarkers including proteins, peptides, metabolites, mRNA, DNA, lipids, and miRNA [14‐18]. We recently reported that the practice of TMP increases salivary nerve growth factor [19], and caused wide range alteration to salivary proteome [20].

Based on the ability of YB to reduce stress and because salivary biomarkers could be utilized to assess stress, it was hypothesized that the practice of TMP would reduce the salivary expression of pro-inflammatory biomarkers. The results indicate that YB practice as specified in Thirumanthiram in fact reduced some of the key biomarkers implicated in stress and inflammation. Identification of molecular changes in easily accessible saliva could increase the ability of correlating subjective responses with biochemical measures in clinical trials involving Yoga and other mind-body practices.

A total of 20 participants were enrolled in the study: 10 males and 10 females. There were 50% males and 50% females, and within each gender 50% each were randomized to AC and YB groups. The median age and interquartile range (IQR) for each group are presented in Table 1. There were no dropouts or adverse reactions reported due to the study exercises.

Salivary cytokine levels were analyzed over time (time points 0, 5, 10, 15 and 20 min) in YB and AC groups. The analysis indicated that the cytokines IL-1β, IL-8 and MCP-1 showed significant differences between groups and changed significantly over time. Overall, IL-1β, IL-8, and MCP-1 levels were significantly lower for YB than for AC participants. In general, the YB group also had more stable values over time than the AC group. All other cytokines analyzed (IL-1RA, IL-6, IL-10, IL-17, IP-10, MIP-1b, and TNF-alpha) did not show any significant differences between groups at all time points (Additional file 1: Table S1).

Figure 2 shows the heat-map at individual time points for IL-1β from all individuals in the AC and YB groups. At each time point the level of IL-1β was significantly reduced in YB group when compared with AC group (Table 2). It is interesting to note that the values of IL-1β were diminished in YB group at the 0 min mark, which is believed to be because the YB exercises were taught to the subjects just prior to taking the measurements. This suggests that mere practicing/learning for a brief period (typically around 5 min) could reduce the levels of IL-1β. It can also be observed from at least some of the individuals from AC group that there is a trend to reduce the IL-1β levels by mere quiet reading. However, these levels were significantly higher than those observed with YB group at each time point. These data indicate that IL-1β level could be significantly reduced by the YB exercise.

Changes in the level of IL-8 are shown in a heat-map in Fig. 3 and in Table 2. It is clear from these data that YB could reduce the level of IL-8 at all time points. While AC could also show a trend of reduction, the magnitude of IL-8 reduction in YB group was 3–6 orders lower than the AC group. This indicates that salivary expression of IL-8 could be significantly reduced by YB practice.

Apart from these two cytokines, there were marginally significant alterations in the chemokine monocyte chemotactic protein (MCP-1) also known as chemokine (C-C motif) ligand 2 (CCL2). While the AC group did not show any decline in salivary expression of MCP-1 as in the case of other cytokines (IL-1β and IL-8), the YB group showed a significant reduction in MCP-1 level in saliva (Fig. 4 and Table 2). The overall results for the main effects of each model are provided in Table 3, and the interaction plot showing changes in all the three cytokines with time in both YB and AC groups is shown in Additional file 2: Figure S1. The heat maps provide more information than the interaction plots for this study because they showed individual level and group level values for each of the cytokines while an interaction plot with means/SDs summarized the data per group and did not provide the individual level data. These results, in combination with the pooled data seen in Table 2 and the heat-maps for each of the significant cytokines, give a full picture on how these cytokines are changing between groups and over time.

This is the first time a multiplex assay format has been used for the determination of salivary cytokines from a Yoga based exercise. These results show a significant reduction in the levels of IL-1β, IL-8 and MCP-1 in the YB group when compared to the AC group. IL-1β is implicated in stress and inflammation in various diseases. Cleavage by caspases-1 and 8 activates IL-1β for subsequent induction of cyclooxygenase-2 in the central nervous system, which is involved in stress and pain perception [22]. Also, at the systemic level, the increased IL-1β level potentiates a variety of pro-inflammatory cellular processes such as proliferation, migration, apoptosis and differentiation. Although these processes are important for host defense mechanisms, higher levels of IL-1β is deleterious as shown in several diseases including cancer (reviewed in [23]). Therefore IL-1β is pursued as a clinical target (for example, Canakinumab [24], which is used for treating autoinflammatory syndromes and in clinical trial for chronic obstructive pulmonary disease).

IL-8 is a key mediator of inflammatory processes during various stress conditions involving oxidative stress. Elevated levels of IL-8 is associated with poor prognosis of breast cancer [25] and in psychosomatic disorders [26, 27]. Although targeting IL-8 pharmacologically is perceived to be important, there is no targeted therapy specifically directed against IL-8. Under these circumstances the current study shows that Yogic Breathing could be an effective non-pharmacological approach to regulate the level of this cytokine.

During tissue injury or infection, MCP-1 recruits key immune cells including monocytes, memory T cells, and dendritic cells; however, the level of MCP-1 is elevated in chronic pathological conditions. The level of salivary MCP-1 is used as a biomarker of various physiological dysfunctions including coronary heart disease and post-traumatic stress disorder (PTSD) where MCP-1 is secreted in large quantities by monocytes recruited in response to chronic endothelial inflammation [28]. These studies also indicate that there was a correlation between the salivary levels of MCP-1 and PTSD symptoms, depression, and anxiety. Interestingly, all the three biomarkers decreased by YB in the present study were reported to be significantly higher in asthmatic individuals when compared to controls [29]. It is important to note that Yoga practices including YB exercises are increasingly being used in the control of asthma [30, 31]. This is the first study ever to find a relationship between Yoga practice and MCP-1. Several studies found key association between elevated levels of MCP-1 and increased severity of pathogenesis including Alzheimer’s disease (AD) [32]. While MCP-1 is increased in AD patients, these observations showing a decrease in MCP-1 could be important for the management of AD. In addition to a reduction in MCP-1, an elevation in nerve growth factor [19] upon YB practice could be at least two key molecular targets elicited in response to YB towards the management of AD.

Identification of these salivary biomarkers in response to a mind-body practice may serve as an important point towards establishing molecular mechanisms associated with these practices. Recently it was shown that relaxation response could produce transcriptomic level changes in several molecules associated with inflammation, immune response, and energy metabolism [12], which correlates with previous studies showing proteomic level changes following YB [20], and changes in pro-inflammatory biomarkers in the present study. Although the mechanism by which breath regulation could reduce salivary pro-inflammatory biomarkers is not clear, it is speculated that the YB exercises administered in the present study might have elicited a mild hypercapnia similar to earlier observations following slow breathing [7], certain slow Yogic breathing exercises [33‐35], and breath holding [36]. Hypercapnia has been used as a therapeutic strategy in clinical settings such as lung injury [37] and hypothesized to be useful in cardiac arrest [38]. It is suggested that hypercapnia reduces inflammation through modulating NFkB mediated inflammatory signaling [39, 40]. This observation needs to be correlated with the studies of Bhasin et al. [12] where the transcriptome that showed significant reductions during relaxation response was associated with NFκB signaling node, and the present data showing a reduction in all three biomarker expressions is reported to be mediated through NFκB signaling in numerous cellular contexts [41, 42]. Therefore, it is hypothesized that hypercapnia could be a possible mediator of reducing inflammatory biomarker release following YB practice. Further studies are needed to understand this relationship. The immunological effects of observing the reduced levels of cytokines could be as a result of sequestration of cytokines in the salivary milieu. Our earlier published work shows that several proteins with broad agglutinating/sequestering functions are induced upon Yogic breathing exercise. These proteins include clusterin, CRISP3, DMBT1, and mucin 7. It has been proposed that such cytokine or growth factor sequestering function is important in cancer prevention and control [43]. This could be one of the ways by which a short term Yogic breathing training reduces cytokine levels in saliva, however requires further experimental support.

This study is innovative in the following respects: 1) Use of salivary pro-inflammatory biomarkers using multiplex assay format for the analysis of Yoga outcomes, and 2) This is the first time that Thirumoolar Pranayamam from an ancient Tamil literature was studied for its effects on the salivary expression of pro-inflammatory biomarkers. While this is the first study approaching Thirumoolar Pranayamam for its anti-inflammatory biomarkers, earlier studies have shown evidence for the stress reducing ability of Pranayama methods. The other Pranayama method used in the present study, Om chanting also called Pranava Pranayama, is an ancient cultural practice involving breath regulation thought to promote overall physical and emotional well-being [8, 44]. Previous studies show that this type of breath regulation could exert the following effects: caused significant reduction in heart rate and blood pressure in hypertensive patients within 5 min of practice [45], which incidentally correlates with the stress biomarker changes seen in 5 min of practice in the present study; and, increased cutaneous peripheral vascular resistance, which is a sign of increased mental alertness even while being physiologically relaxed [44]. Taken together, this data provides a biochemical support for the observed relaxation experience with the practices of regulated breathing.

Salivary cytokines have recently been tested in various stress related conditions including diabetes [46], and as a measure of immune function [47]. Yoga studies so far have relied mainly on subjective measures (e.g., self report of quality of life and stress) to determine the effectiveness of Yoga intervention [48]. Recently, gene expression studies in peripheral blood following 8 weeks of relaxation techniques (including meditation, Yoga, and breath regulation) identified a number of genes involved in metabolism, energy regulation, and inflammation [12]. We reported that salivary nerve growth factor could be induced [19], and salivary proteome could be altered [20] by these YB exercises. This further supports the hypothesis that YB could acutely alter the levels of salivary biomolecules with key biological functions immediately after practice, and this might be important not only in the oral cavity but might also have systemic effects. Some limitations of the present study include the small sample size, and the cytokine levels at Time 0 being taken after learning the breathing exercise in the YB group as opposed to prior to learning the technique. Because participants had already learned the technique, the cytokine levels may not represent a true baseline measurement.

For the first time, data showing that a single 20 min session of Yogic Breathing practice could reduce the levels of key pro-inflammatory biomarkers in saliva. Future studies using saliva as a source of biomarkers in individuals during the practice of various Yoga techniques could help deduce molecular mechanisms through which these Yoga techniques may work.