Nitric Oxide 44 (2015) 88–97

Contents lists available at ScienceDirect

Nitric Oxide j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y n i o x

Effects of hyperbaric oxygen on nitric oxide generation in humans Johan Uusijärvi a,b, Karin Eriksson a,b, Agneta C. Larsson a, Carina Nihlén a, Tomas Schiffer a, Peter Lindholm a,c, Eddie Weitzberg a,b,* a b c

Department of Physiology and Pharmacology, Karolinska Institutet, Stockholm, Sweden Deparment of Anesthesia & Intensive Care, Karolinska University Hospital, Stockholm, Sweden Department of Radiology, Karolinska University Hospital, Stockholm, Sweden

A R T I C L E

I N F O

Article history: Received 1 October 2014 Revised 12 November 2014 Available online 11 December 2014 Keywords: Exhaled Nitrate Nitrite Bacteria

A B S T R A C T

Background: Hyperbaric oxygen (HBO2) has been suggested to affect nitric oxide (NO) generation in humans. Specific NO synthases (NOSs) use L-arginine and molecular oxygen to produce NO but this signaling radical may also be formed by serial reduction of the inorganic anions nitrate and nitrite. Interestingly, commensal facultative anaerobic bacteria in the oral cavity are necessary for the first step to reduce nitrate to nitrite. The nitrate-nitrite-NO pathway is greatly potentiated by hypoxia and low pH in contrast to classical NOS-dependent NO generation. We investigated the effects of HBO2 on NO generation in healthy subjects including orally and nasally exhaled NO, plasma and salivary nitrate and nitrite as well as plasma cGMP and plasma citrulline/ arginine ratio. In addition, we also conducted in-vitro experiments in order to investigate the effects of hyperoxia on nitrate/nitrite metabolism and NO generation by oral bacteria. Methods: Two separate HBO2 experiments were performed. In a cross-over experiment (EXP1) subjects breathed air at 130 kPa (control) or oxygen at 250 kPa for 100 minutes and parameters were measured before and after exposure. In experiment 2 (EXP 2) measurements were performed also during HBO2 at 250 kPa for 110 minutes. Results: HBO2 acutely reduced orally and nasally exhaled NO by 30% and 16%, respectively. There was a marked decrease in salivary nitrite/nitrate ratio during and after HBO2, indicating a reduced bacterial conversion of nitrate to nitrite and NO. This was supported by in vitro experiments with oral bacteria showing that hyperoxia inhibited bacterial nitrate and nitrite reduction leading to reduced NO generation. Plasma nitrate was unaffected by HBO2 while plasma nitrite was reduced during HBO2 treatment. In contrast, plasma cGMP increased during HBO2 as did citrulline/arginine ratio after treatment and control. Conclusion: HBO2-exposure in humans affects NO generation in the airways and systemically differently. These data suggest that the individual NOSs as well as the nitrate-nitrite-NO pathway do not respond in a similar way to HBO2. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Hyperbaric oxygen (HBO2) therapy is used to treat various clinical conditions, including patients with severe tissue infections [1] but the underlying mechanisms are largely unknown. Apart from inhibiting growth of anaerobic and facultative anaerobic bacteria [2–5] and promoting leukocyte function [6], effects on endogenous nitric oxide (NO) production has been suggested [7,8]). NO is a central mediator in numerous physiological processes, including regulation of blood flow, nerve transmission, mitochondrial function and host defense [9]. Specific NO synthases (NOS) use the

* Corresponding author. Department of Physiology and Pharmacology, Section of Anesthesiology and Intensive Care, Nanna Svartz väg 2, Karolinska Institutet, 17177, Stockholm, Sweden. E-mail address: [email protected] (E. Weitzberg). http://dx.doi.org/10.1016/j.niox.2014.12.002 1089-8603/© 2014 Elsevier Inc. All rights reserved.

amino acid L-arginine and molecular oxygen to catalyze the generation of NO and equimolar amounts of L-citrulline, in a complex five-step oxidation process (see equation).

L-arginine + 3 2 NADPH + H+ + 2 O2 → citrulline + NO + 3 2 NADP + There are three different NOS isoforms: Neuronal NOS (nNOS, NOS I), inducible NOS (iNOS, NOS II) and endothelial NOS (eNOS, NOS III). nNOS and eNOS are calcium-dependent and constitutively expressed while iNOS is calcium-independent, induced by certain cytokines and bacterial toxins and generates much higher amounts of NO compared to the constitutive NOSs [10]. Many of the effects of NO including vasodilation, are propagated via activation of soluble guanylyl cyclase which leads to increased levels of cyclic guanosine monophosphate (cGMP) [11]. NO is extremely short-lived under biological circumstances due to its oxidation to nitrite (NO2−) and in the presence of hemoglobin to nitrate (NO3−).

J. Uusijärvi et al./Nitric Oxide 44 (2015) 88–97

Hence, hemoglobin is a potent scavenger of NO thereby regulating its activity. In addition to the classical NOS-dependent generation this mediator can also be produced by stepwise reduction of the inorganic anions nitrate and nitrite, thereby recycling these oxidation products of NO back to NO-like bioactivity [12–14]. Interestingly, nitrate from the diet (especially leafy green vegetables) can be used as a substrate for this type of NO production [15]. Circulating nitrate, both from endogenous NO generation or from the diet, is actively taken up by the salivary glands and excreted in saliva [16]. The first step of the nitrate-nitrite-NO pathway requires oral commensal facultative anaerobic bacteria to reduce nitrate to nitrite [17]. These bacteria contain potent nitrate and nitrite reductases that are efficiently inhibited by oxygen. In addition, in lack of oxygen oral bacteria use nitrate and nitrite as a terminal electron acceptor for respiration. Due to the fact that nitrate is concentrated in saliva and bacteria reduce nitrate to nitrite, the levels of these anions in saliva are 10–1000 fold higher compared to plasma [18]. In the body there are several pathways for further reduction of nitrite to NO, all of which are potentiated during hypoxia and low pH [19]. Interestingly, deoxyhemoglobin is a potent nitrite reductase in contrast to oxyhemoglobin and for this reason oxygen saturation and allosteric changes in hemoglobin will regulate NO generation from circulating nitrite [20]. As mentioned above, NO is highly unstable in biological fluids and surrogate markers are often used to indicate NO production such as nitrite and nitrate, citrulline/arginine ratio and cGMP. However, in the gas phase NO is much more stable and this fact has been used in measurements of NO in exhaled air in relation to inflammation in the airways, mostly in patients with asthma [21,22]. A major source of NO in orally exhaled air is the bronchial epithelium but as much as 50% can be contributed by the oral cavity due to bacterial reduction of nitrate and nitrite [23,24]. In the nasal airways high levels of NO are constantly produced in the paranasal sinuses by a constitutively expressed iNOS and this NO is responsible for the high levels found in nasally exhaled air [25]. Taken together, there are several sources contributing to the levels of NO found in exhaled air and this has to be taken into account when evaluating this parameter during different circumstances. Therefore consensus guidelines have been published on how to measure NO in exhaled air and how to interpret the results [22]. Only a few studies have examined the effect of HBO2 on exhaled NO in humans [26–30]. These studies have been performed in patients or divers and with different techniques for measurement of exhaled NO. A common finding is an acute as well as chronic reduction in orally exhaled NO after exposure to HBO2. However, these studies did not take into account the contribution of NOS-independent NO generation in the oral cavity or NO from the paranasal sinuses. The latter is interesting since nasally exhaled NO is iNOS dependent and therefore any change could indicate effects by HBO2 on this isoform. The effects of HBO2 on systemic NO generation has been investigated in vitro, in animals and in a few human studies [31]. In cell cultures data are somewhat contradictory regarding the effect on eNOS [32,33]). In animals HBO2 seems to enhance NO generation by activation of eNOS [34]. In animal models of inflammation HBO2 treatment seems to decrease iNOS expression [35]. Regarding systemic NO production after HBO2 in humans, scarce data exist and the effects of HBO2 on NOS-activity estimated from plasma nitrate and nitrite levels are lacking in healthy subjects. In a clinical trial in diabetic patients plasma levels of nitrate and nitrite increased after HBO2 therapy and correlated with successful wound healing [36]. Other indicators of systemic NO production such as plasma cGMP or citrulline/arginine ratio has, to our knowledge, not previously been documented after HBO2 in humans. We hypothesized that HBO2 could have differential effects on NO production in the airways and systemically. We therefore aimed at

89

a broader descriptive study on the NO system. In healthy subjects we investigated how HBO2 influenced orally and nasally exhaled NO levels as well as nitrate and nitrite reduction in the oral cavity. Moreover, we measured plasma nitrate, nitrite, citrulline/arginine ratio and cGMP, as measures of systemic NO production, during and after HBO2. 2. Material and methods Two separate experiments (EXP1 and EXP 2) were performed in healthy subjects. The mean age was 27 ± 4 and 30 ± 4 years in EXP1 and 2, respectively. All were non-smokers without any ongoing medication. None of the participants had a history of asthma or had performed strenuous exercise within 24 hours prior to experiments. They were told not to eat vegetables or other nitrite/ nitrate rich food the same day. Subjects were recreational divers which facilitated subject information and pressure equalization. All subjects gave their written, informed consent to participation prior to the tests. The study was conducted in conformity with Ethical Principles for Medical Research Involving Human Subjects (World Medical Association Declaration of Helsinki, 2013) and was approved by the Local Ethics Committee [52]. 2.1. Protocol 2.1.1. Experiment 1 in monoplace chambers These placebo-controlled, randomized, cross-over experiments were performed in Monoplace Chambers (Model 2500, Sechrist Industries, Anaheim, California). Subjects (n = 12, 5 males) served as their own controls in a randomized order with active HBO2 or air breathing (control). Each session was followed by a washout period of at least 7 days. In the chamber subjects rested supine with slightly elevated back and head and were allowed to watch video or television during exposure. Plasma C-reactive protein, glucose and hemoglobin levels were analyzed prior to first exposure and were normal in all subjects. The HBO2 protocol consisted of a 100-minute session (including 5 minute pressurization and 5 minute decompression) at 250 kPa breathing 100% oxygen (PiO2 = 250 kPa), interrupted by two 10minute air-breaks (PiO2 = 52.2 kPa) (Fig. 1A). Decompression to atmospheric pressure was performed with continued oxygen breathing. The control protocol was identical to HBO 2 except for pressurization with air to only 130 kPa (PiO2 = 27 kPa) throughout the session, with simulated switches for gas change. Continuous ventilation of the chamber was performed with 300–400 l/minute for optimal comfort. The temperature inside the chamber was not regulated or measured during exposure. Before and immediately after each exposure, fraction of exhaled NO (FENO) was measured during nasal and oral exhalations and venous blood and saliva samples were taken. 2.1.2. Experiment 2 in multiplace intensive care chamber These non-controlled experiments (n = 7, 3 males) were performed in a Multiplace Intensive Care Chamber (HAUX Quadro 3200, Haux Life-Support Inc, Karlsbad, Germany) with measurements before, during and after HBO2 exposure. Subjects rested supine with slightly elevated back and head and were allowed to watch video or television. They stayed in the same position during the measurements before and after exposure. Exposure consisted of pressurization to 250 kPa (10 minutes), O2-breathing on mask (75 minutes), interrupted by two 10 minute air brakes (mask off), 5 minutes decrease to 130 kPa, 5 minutes air breathing at 130 kPa and decrease to atmospheric pressure (open doors) with a total time 120 minutes (Fig. 1B). Demand-type built in breathing-systems (BIBS), connected to lightweight masks were used. A controlled temperature (21 °C) was maintained during the

90

J. Uusijärvi et al./Nitric Oxide 44 (2015) 88–97

Fig. 1. (A) Exposure table for EXP1 in a monoplace hyperbaric chamber. Air-breaks were included as a standard prophylaxis against oxygen induced seizures. Sampling of saliva and blood, and measurements of orally and nasally exhaled NO are indicated (1, 2). The control experiment was performed in the chamber breathing air throughout the exposure at a slightly elevated ambient pressure (130 kPa, PiO2 = 27 kPa). (B) Exposure table for EXP2 in a multiplace hyperbaric chamber. Air-breaks were included. Numbers indicate sampling of blood (1, 2, 3, 5, 6, 7) and saliva (1, 4, 6).

experiment using the environmental control system in the chamber. Oxygen enriched air (Nitrox 60) was breathed intermittently by research staff exposed to elevated ambient pressure [37]. Samples of venous blood and saliva were taken before, during and after exposure. 2.2. Measurements 2.2.1. Exhaled NO (EXP1) Fraction of exhaled NO (FENO) was measured in the sitting position before and after HBO 2 or control according to current international guidelines using a NIOX MINO® Airway Inflammation Monitor (Aerocrine AB, Solna, Sweden) [22]. Under guidance, all subjects inhaled NO-free air (built-in NO scrubber) close to total lung capacity and then exhaled during 10 s at a flow rate of 50 ml/s to provide three approved FENO measurements from the mouth and the nose, respectively. During nasal exhalations, the NIOX MINO inspiratory inlet was applied directly to one nostril. Subjects were instructed to keep the pressure between 12 and 18 cmH2O with the help of visual feedback provided by the NIOX MINO. At these pressures exhalation flow rate is kept at 50 ± 5 ml/s with calibrated dynamic flow resistors. Four subjects, with FENO values >25 ppb before experiments were excluded since a high FENO is suggested to signal an on-going airway inflammation [22]. 2.2.2. Blood and saliva sampling (EXP 1 and EXP 2) A catheter was inserted in the antecubital vein for collection of blood. In EXP1 blood was sampled before and within 2 minutes after each exposure. Saliva (3–4 ml) was sampled before and immediately after exposures. During HBO2 exposure in EXP2 blood sample tubes were locked out via the chamber medical locks (2 minutes duration). Sampling occurred after 20 minutes of supine rest before exposure, during exposure at the end of each breathing gas period (except for the second air period) and directly after hyperbaric exposure. A late sample was also collected during supine rest 60 minutes after exposure. Samples were kept on ice during lock-out and during a 5–7 min delay before centrifugation. Saliva (3–4 ml) was sampled before exposure, immediately after the second HBO2-period at 250 kPa and directly after exposure.

2.3. Laboratory methods 2.3.1. Nitrate and nitrite in plasma and saliva (EXP 1 and EXP 2) Blood and saliva samples were collected in prechilled vacuum tubes containing EDTA (250 mM, 40 μL per mL blood). The tubes were immediately centrifuged at 2000 × g for 10 minutes at +4 °C, and aliquots of plasma and saliva were frozen (−20 °C for 1–8 hrs and then at −80 °C) until analyzed. Nitrate and nitrite were determined by chemiluminescence after reductive cleavage and subsequent release of NO into the gas phase as described earlier in detail [18]. In brief, plasma or saliva samples were introduced into a reaction vessel purged with nitrogen containing reducing solutions (iodine-iodide) and release of NO was analyzed by chemiluminescence (NO analyzer CLD 77AM, Ecophysics, Dürnten, Switzerland). NO peaks were further analyzed with the Windows Azur platform and the levels of nitrite and nitrate were calculated and reported in M (mol/L) by comparing the areas under the curve to known concentration of nitrite or nitrate.

2.3.2. Plasma cGMP (EXP 1 and EXP 2) cGMP was assayed in plasma after ethanol extraction according to the manufacturers’ instructions (Assay Designs/Enzo Life Sciences, AH Diagnostics, Sweden) using cGMP EIA- kit, ADI900-013.

2.3.3. Plasma citrulline and arginine (EXP 1) Arginine and citrulline were measured with HPLC. The samples were mixed with an equal amount (10 μL) of an orthophthaldialdehyde/mercaptoethanol reagent by a CMA/200 autoinjector (CMA Microdialysis AB, Sweden). After 60 s reaction at 8 °C, 15 μL was injected on the HPLC column (60 × 4 mm Nucleosil 100 C18, 5 μm, Knauer GmbH, Germany). The elution was achieved with a Na acetate buffer (0.03 M), methanol 2.5% v/v, and tetrahydrofuran 2% v/v. A methanol 0–60% gradient was established between 4 and 28 min. The initial buffer was then used to regenerate the column. A fluorescence detector (CMA/280) with excitation and emission band around 350 and 495 nm respectively was used for the detection. The quantification was done with an integration program

J. Uusijärvi et al./Nitric Oxide 44 (2015) 88–97

(EZ Chrom Chromatography Data system, Scientific software Inc., CA, USA) comparing peak heights. 2.4. In vitro experiments 2.4.1. Oral bacteria and nitrate reduction Four healthy volunteers were asked to scrape the back of the tongue with a 10 μl inoculation loop and the content was immediately transferred into 1 ml Mueller Hinton containing 500 μM nitrate. These mixtures were divided in three tubes which were bubbled 2 minutes with 100% nitrogen, air or 100% oxygen, respectively. The tubes were sealed and incubated for 1 h in 37 °C. Nitrite and nitrate were measured before and after the incubation time using a high performance liquid chromatography (HPLC) system (ENO20; Eicom, Kyoto, Japan). 2.4.2. NO production from nitrate by oral bacteria Five healthy volunteers were asked to scrape their tongue with their teeth and 1 ml of the bacteria rich saliva was mixed with 1 ml of Mueller Hinton containing 10 mM nitrate. The samples were bubbled with oxygen or nitrogen and immediately injected into 500 ml gas tight bags (Infubags, S.E Nundel Kunstoff-Technik GmbH, Germany) prefilled with 100% nitrogen or oxygen. The bags were incubated in 37 °C. Gas samples (5 ml) were taken out at different time points during 5 h and NO was analyzed with chemiluminescence (77 AM, Eco Physics, Switzerland). 2.4.3. Oxidation of nitrite 1.5 ml Eppendorf tubes (n = 4) containing 500 μM nitrite in PBS were bubbled with 100% oxygen for two minutes, sealed and incubated for 1 h in 37 °C. Nitrite and nitrate were measured before and after the incubation using HPLC/ENO-20 as described earlier [38].

91

2.4.4. NO stability in high oxygen tension NO gas was adjusted to a concentration of 100 pbb in gas tight bags (Infubags, S.E Nundel Kunstoff-Technik GmbH, Germany) containing 100% oxygen. 20 ml gas samples were taken out at different time points during 5 h and directly analyzed with chemiluminescence. All in vitro experiments were performed during normobaric conditions. 2.5. Statistical methods Statistics were performed with GraphPad Prism 6 for Windows (GraphPad Software Inc., La Jolla, California, USA). Non-parametric testing was used for all parameters and conditions. In EXP1, Wilcoxon matched pairs signed rank test was used to compare before- vs. afterHBO2 or placebo conditions. Mann–Whitney’s test was used for baseline comparisons (before HBO2 vs. before Control). For serial measurements before, during and after HBO2 (EXP2), Friedman’s test and Dunn’s multiple comparison post hoc test were used. Data are presented as median and CI 25–75%. 3. Results 3.1. Exhaled NO Exhaled NO was measured in EXP1. Orally and nasally exhaled NO levels before exposure did not differ between control and HBO2. HBO2 induced a 30% decrease in oral FENO while there was no change after the control experiment (Table 1, Fig. 2A and B). HBO2 also led to a decrease by 16% in nasal FENO while control treatment had no effect (Table 1, Fig. 2C and D). When subtracting oral FENO from nasal FENO in order to get values more representative of the nasal airway NO production similar findings were observed; a 14% decrease in

Table 1 Data from EXP1 in the monoplace hyperbaric chamber. Results shown as median (25–75%). Parameter

n

Condition

Oral FENO (ppb)

12 12

HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO2 vs. Control HBO2 Control HBO vs. Control

- Baseline Nasal FENO(ppb) - Baseline Nasal-oral FENO (ppb) - Baseline Salivary nitrite (μM) - Baseline Salivary nitrate (μM) - Baseline Salivary nitrite/nitrate - Baseline Plasma nitrite (nM) - Baseline Plasma nitrate (μM) - Baseline Plasma cGMP (nM) - Baseline Plasma cit /arg ratio

12 12 12 12 9 11 6 8 6 8 12 11 12 11 10 10 11 11

- Baseline P values in bold denotes statistically significant changes.

Before

After

17 (14–21) 14 (12–16)

12 (10–18) 15 (14–19)

74 (54–89) 62 (46–83)

62 (50–76) 68 (49–88)

56 (38–69) 48 (32–66)

48 (38–63) 49 (35–66)

81 (46–123) 131 (71–271)

48 (36–117) 119 (95–196)

198 (95–585) 345 (106–767)

277(196–795) 204 (170–877)

0.28 (0.21–0.69) 0.75 (0.33–1.48)

0.16 (0.10–0.20) 0.50 (0.38–0.83)

162 (118–196) 161 (114–192)

164 (107–220) 156 (112–196)

18 (14–23) 21 (16–35)

20 (11–28) 31 (24–40)

4.0 (3.3–5.1) 3.8 (3.4–4.4)

3.8 (2.4–4.2) 3.9 (3.0–4.2)

0.49 (0.32–0.55) 0.39 (0.34–0.50)

0.54 (0.42–0.65) 0.46 (0.34–0.52)

p-value