A novel device for in-situ on-demand hydrogen sulfide generation and delivery to increase tissue perfusion to chronic wounds

Abstract

Background

There is significant evidence linking deficiency in endogenous hydrogen sulfide (H₂S) to delayed healing of wounds. Systemic administration of H₂S donors has been shown to markedly improve the healing rate of ischemic open wounds. This paper describes a novel H₂S therapy approach, namely H₂EALS™, as an exogenous supplement to accelerate healing. H₂EALS™ employs a controllable process for in-situ conversion of a metal sulfide coating to H₂S gas for safe and effective H₂S treatment.

Methods

The core technology topically releases a targeted quantity of H₂S molecules within the dressing for absorption into the open wound using control and delivery protocol specified by the end-user. Benchtop studies have evaluated precision in the rate and amount of H₂S released and the degree of control, all required for safe treatment. In preclinical studies using rodent and swine models, the potential widespread dispersion of the H₂S delivered was evaluated with an H₂S sensing approach, while the physiological dose-response in terms of absorption and perfusion within the wound and periwound regions were examined using both H₂S measurement and Laser Speckle Contrast Imaging (LSCI).

Results

Benchtop studies on H₂EALS™ demonstrate clinically relevant performance durability and control over H₂S dose administered. With low power requirement, high dosage control, and relatively low dose-to-dose variability, electroactive H₂S sources with capacity of > 2 µmol/application were fabricated to last beyond a typical 14-day wound treatment period. In-vivo studies on open wounds in Sprague-Dawley rats treated at 25 to 100 nmol/dose indicated rapid tissue absorption of H₂S gas. Real-time breath- and skin-emitted H₂S measured on swine revealed no evidence of widespread dispersion when up to 500 nmol/dose was delivered directly to an 8-mm open wound.

Conclusion

Our results emphasize the need for informed delivery and absorption of H₂S to the target tissue and we demonstrate the potential of the H₂EALS™ technology of safe and effective H₂S treatment of open wounds and in acceleration of wound healing. The results are a partial contribution to evaluating the impact of H₂EALS™ on wound healing acceleration and tissue repair.

Introduction

Chronic limb-threatening ischemia (CLTI), the most severe stage of PAD present in nearly 11% of cases, is projected to impact more than 4 million Americans by 2030 [1, 2]. The development of CLTI heralds a significant risk of complications ranging from partial limb loss to major amputation, severely impacting the patient’s functional status, quality of life and longevity. Chronic diabetic-related wounds represent a significant global cause of morbidity and public health care burden [3]. In 2007, more than 60% of non-traumatic limb amputations and 27% of the $116 billion diabetic health care costs in the US were attributed to chronic wounds [4, 5]. From 2007 to 2012, the prevalence of diabetes increased by 26%, while the liability costs related to diabetic complications jumped a hefty 51% to $176B [6]. Despite improvements in standard treatments and the emergence of new diabetic wound therapies [7, 8], the healing rates are still variable, often resulting in recurrence [9]. Delayed healing of these wounds is exacerbated by scarring with loss of epidermal and dermal architecture, and insufficient strength of the re-epithelialized tissue leading to wound recurrence and increased lower-limb amputations. In a recent review of 19 matching studies on incidence rates for ulcer recurrence, Armstrong and colleagues estimated that roughly 40% of patients have a recurrence within 1 year, 60% within 3 years, and 65% within 5 years [10], with 5-year mortality rate exceeded only by lung cancer [11].

The translational importance of hydrogen sulfide (H₂S) is highlighted by studies performed within leading academic institutions, which reveal decreased sulfide levels in the plasma of both humans and rodents afflicted with diabetes [12]. H₂S has been shown to promote angiogenesis behavior in cultured endothelial cells by activating pathways such as nitric oxide (NO) and the HIF-1α and VEGF-A-mediated angiogenesis cascade [13–15]. H₂S is endogenously produced in the endothelial cells of blood vessels [16] to regulate vascular function by stimulating endothelial growth, vessel formation [17], local dilation, blood flow increase, tissue ischemia recovery [18, 19], and inflammation reduction [20, 21]. In diabetes, the commonly occurring endothelial dysfunction decreases production leading to low bioavailability of H₂S [22]. This is apparent in pancreatic cells of rats [23], endothelial cells exposed to hyperglycemia [24], animal models of diabetes [25, 26] and diabetic human subjects [25, 27], thus contributing to vascular dysfunction [16] and poor wound healing [28].

In-vitro studies of exogenous H₂S supplements have been shown to stimulate endothelial cell growth, migration, and neovascularization through the action of VEGF leading to more rapid wound closure [13]. Recent small animal studies show that a two-week treatment with systemic delivery of H₂S can ameliorate hindlimb ischemia [29]. Other studies show that open wound healing time is reduced by more than 30% [30, 31]. However, systemic H₂S delivery comes with high risk of cytotoxicity [32] and off-target effects including hypotension and hepatotoxicity [33], an issue that can be avoided by local-regional delivery. As proposed by Jensen et al. [19], “…the ideal H₂S releasing ‘drug’ should work locally without systemic effects, have controlled and sustained release, and limited side effects…”. A comprehensive review of the H₂S delivery and therapy techniques being developed in various laboratories has been provided by Jin et al. [34] and Ge et al. [35] and provided in Table 1. Several limitations have been revealed on solution-based H₂S therapy delivered using microfluidic systems (i.e., microelectromechanical systems, MEMS), for delivery of small quantities of pre-mixed H₂S donor solutions to tissue, including system performance (e.g., electromechanical failure), system size, and chemical stability of the H₂S donor compounds. Similar problems plague in-situ mixing of the solution and/or passive H₂S delivery techniques such as H₂S-eluting dressings [36, 37] or GYY4137 impregnated hydrogels (e.g., gelatin methacryloyl, GelMA) and fibrous material [38, 39]. While the recently developed light-actuated or photoactivatable (alternatively, photoremovable, photo-releasable, or photocleavable) protecting group (PPG) donors show promise of better control, higher stability, and more targeted delivery, they are either too complex and cumbersome or are not suitable for broad applications in treating ischemic wounds [35, 40]. These issues highlight the need for a standardized approach that offers precision in delivery of H₂S to facilitate clinical translation of H₂S therapy [38].

Table 1.

Summary of Relevant state of the Art H₂S Donor/Source (LEGEND: X Poor; o Acceptable; ✔ Excellent)

Donor/Delivery Technique	Parameters (Critical to Adoption)
	Reliable, Sustained Delivery	Controllable (Active vs. Passive)	Stable Shelf-Life	Low Cost	Targeted	Scalability/Versatility	Delivery Record
Fast-acting Donors (e.g., NaHS/Na2S donors delivered via gels and Phase Change Media) [35–38]	X	X	X	✔	o	✔	X
Controlled Release Donors (e.g., GYY4137 Encapsulatedin Gels, PCM, Polymeric Coatings, etc.) [39, 40]	o	o	X	✔	✔	✔	X
Light Actuated and Photo-Controllable Donors (e.g., PPG such as o-nitrobenzyl caged geminal dithiol) [36, 41]	✔	✔	✔	X	✔	X	o
Microfluidic System (i.e., MEMS) Delivery of DonorSolutions (Any of the donors listed above) [38]	X	✔	X	X	✔	✔	✔
Electrolytic Solid-State Donor as H25 Source [42]	✔	✔	✔	o	✔	✔	✔

This paper summarizes the details of a new innovative technique and approach developed by the engineering team at Exhalix, namely H₂EALS™ (trademark of Exhalix LLC), for delivery and control of H₂S to wounded tissue. A summary of advantages and disadvantages of the various donors and delivery methods, including H₂EALS™, have been provided in Table 1. As it will become evident throughout the following sections, two elements of the innovative platform technology in H₂EALS™ address the safety and efficacy requirements with clear impact on commercial viability: (1) in-situ, real-time synthesis of H₂S, and (2) sustained, uniform, and local delivery of H₂S to the wounded tissue. Actively-controlled synthesis of H₂S from a chemically stable metal sulfide precursor is achieved by an add-on, application-specific module, which in addition to the topical chronic wound healing, can be used for tissue therapy (e.g., allograft/autograft conditioning) and smart bandages.

Materials and methods

Description of H₂EALS™ for in-situ generation and delivery of H₂S

At the core, the patent-protected technology is a viable alternative to solution-based precursors (e.g., Na₂S or NaHS) [30, 41], for in-situ, actively-controlled synthesis of H₂S from a chemically stable metal sulfide precursor [42]. We have previously shown that oxidation of silver nanoparticle-coated Nafion electrodes exposed to H₂S can generate electrochemical signals for high fidelity detection of H₂S concentration in air [43]. The innovative and patented technology developed at Exhalix, transdermal arterial gasotransmitter sensor, TAGS™ (a trademark of Exhalix LLC), is based on this approach and has been used as a diagnostic tool to measure the bioavailability of endogenously generated H₂S in an area that matters most for wound healing – in the microvascular capillary bed region of the skin [44]. As such, transdermal H₂S emissions are measured with a limit-of-detection of less than 3 ppb and signal-to-noise ratio (SNR) > 15 @10 ppb sufficient for detection of peripheral artery disease (PAD) [45, 46].

For treatment of an open wound in a system schematically depicted in Fig. 1(a), the H₂EALS™ module generates H₂S gas by electrochemical conversion of a silver sulfide coating to silver and H₂S. When released into the wound headspace, the H₂S gas molecules diffuse and migrate into the wound bed as described later in this paper. Because of the relatively low permeability of the skin membrane (i.e., epidermis), any H₂S gas present in the wound headspace is topically adsorbed through the open wound for diffusion and metabolic reaction. The open wound acts like a sink for H₂S molecules generated by H₂EALS™ if the headspace concentration, C_d, is higher than the gas-phase equilibrium concentration, C_t, of H₂S in the wound and surrounding tissue (i.e., within the region of influence).

Fig. 1.

(a) Model of H₂EALS™ H₂S generation and transport for chronic, open wound healing application; (b) The disposable H2S source cartridge; (c) exploded view of an autonomous H₂EALS™ module attachable to dressing or skin (insert picture of a device)

The H₂EALS™ module shown in Fig. 1(b) and (c) consists of a proprietary disposable H₂S generator cartridge, an electronic control module and battery, protective cover, and dressing attachment interface. The cartridge uses a 2032 battery that allows up to one week of operation at normal doses. The electronic control module is programmed, and data is retrieved through near-field communication (NFC) via a cell phone application software. The electronic control module executes the dosage protocol, communication, data storage, and power management to ensure higher battery life.

As shown in Fig. 2(a), the electrolytic method employs a controllable, in-situ, real-time conversion of a stable metal sulfide compound, in this case silver sulfide, to H₂S gas for a ‘true sustained delivery’. Figure 2(b) is a typical SEM image of the nanoporous coating deposited onto a silver substrate using a standard chemical deposition approach that generates sub-micron sized, high surface area features. By using a potentiometric process electrons are injected into the reaction at low voltage (< 1 VDC), instantaneously releasing H₂S gas from the surface of the Ag₂S coating:

1

Fig. 2.

(a) Electrolytic H₂S sensing and synthesis approach taking place in the disposable coupon; (b) SEM image of the in-house fabricated high porosity, high surface area coating used for H₂S synthesis

Although not the subject of the current study, it has been shown that reversal of the current between working electrode (WE) and counter-electrode (CE), from generation to sensing reaction, leads to sensing of the concentration of H₂S in the surrounding electrolyte. The H₂S released from the generation reaction initially enters the gel electrolyte before leaving the generator through a H₂S permeable membrane that seals the electrochemical cell. The dosage, or amount delivered to tissue, can be sustained at a low level within the therapeutic window for continuous delivery (e.g., 5 nmol/hour for two weeks) or equal incremental delivery (e.g., a discrete dose of 120 nmol per day for two weeks). The total dosage is a defined quantity, equal to the integral or sum of doses provided during the treatment period:

2

where, is total dose (nmol), is the controllable H₂S emanation rate (nmol/s), t is time (s), is the instantaneous current provided (nA), is generation or synthesis rate constant (nmol/nA), and N is number of dose segments for a treatment protocol. Embodied in the core technology are proprietary configurations (i.e., materials and methods) of the laminate package in a disposable cartridge which enables in-situ synthesis of H₂S, while sensing the concentration to monitor both H₂S synthesis and tissue adsorption rate in real-time, allowing better and safer dose management.

Our studies and others indicate that the target dosage required for healing of open wound is < 1 µmol/cm²/day depending on the condition treated. We’ve tested other candidates to electrochemically or thermally generate H₂S using metal sulfides such as ZnS, Cu₂S, and NiS₂. Shown in Table 2, H₂EALS™ compared to the other two techniques is shown to possess the best characteristics offering better controllability, donor stability, efficiency, safety in use, manufacturability, reversibility for in-situ sensing, and cost. Accommodating the full functionality of H₂EALS™, the disposable, scalable module for the prototype will thus be composed of two components: (a) Laminate Core: multiple layers of the electrodes, conductors, electrolyte, and diffusion barrier membrane; (b) Laminate Package: includes multilayer source/sensor and interface electronics.

Table 2.

Comparison of Various Electrochemical H₂S Generation Techniques. (LEGEND: X Poor; o Acceptable; ✔ Excellent)

Approach	Selection Criteria
	Control-ability	Donor Stability	Power Efficiency	Low Risk to Tissue	Manufactur-ability	Cost Effectiveness
Spontaneous Anodic Reaction of Al, Fe, etc. for Reduction of Ag2S (or ZnS) to H2S	o	o	✔	X	✔	✔
Zinc Sulfide (or Iron Sulfide, etc.) Thermal Actuated Reduction to H2S	o	o	X	o	✔	✔
Potentiometric Reduction of Ag2S to H2S (H2EALS™)	✔	✔	o	✔	✔	✔

Benchtop studies on H₂S generation, tissue delivery, and absorption

The accuracy and kinetics of release and absorption were measured in a series of benchtop studies described in this section. Figure 3 schematically shows the apparatus to measure the output of H₂EALS™ generator cartridges and to quantify generator-to-generator variation at different doses. A gas-tight enclosure was created by using an O-ring sealed 5.3-L polycarbonate container (Cambro 36CW136). Luer-lock gas ports and check valves were installed on the sides of the container so that sample gas could be drawn to determine the H₂S concentration without allowing dilution with outside air when container air is not sampled. H₂S concentration measurements were made with the H2S-1010 device (an Exhalix LLC developed system), which uses the same sensing platform as the TAGS™ diagnostic system described earlier. A magnetic stirrer (Lab Fish MS3) was placed under the container and a magnetic stirrer fan to ensure uniform mixing and seal integrity. Eight H₂EALS™ modules were sequentially characterized, by adhering them to the inner walls of the container using double-sided adhesive tape (Permanent Adhesive Dots, Gorilla Glue, Inc.). Initial apparatus calibration was performed using a material balance and multiple injections of National Institute of Standards and Technology (NIST)-certified calibration gas (GASCO, 34 L-98-10). Replacement or intake gas was scrubbed by using a potassium permanganate (KMnO₄) filter (United Filtration Systems Inc., DIA-NPP). In these benchtop studies, dose-to-dose stability was tested for the target doses of 1-250 nmol on a timescale relevant to the in-vivo studies described later. In each condition, a minimum of 3 generators were tested with 6 independent measurements to characterize the release dynamics and precision. In these measurements, the target dose was set and controlled by the digital electronic control circuit in terms of cumulative millicoulomb (or mC, which is equivalent to 1 µAh) of charge delivered to the working electrode per dose, with a typical conversion rate constant of 4 nmol/mC or nearly 20% below the theoretical maximum that assumes a stoichiometric conversion of Ag₂S to H₂S. Actual delivered doses were calculated from concentration measurements, sensor sample volume (200-ml), the gas-tight container volume (5.3-L), ideal gas law, Reimann sum of samples taken every minute, and a material balance from mass conservation equation.

Fig. 3.

Schematic diagram of the test setup for characterization of the H₂EALS™ modules

The apparatus for characterization of the tissue absorption and perfusion response to H₂S is schematically shown in Fig. 4(a). The H₂EALS™ module was placed adjacent to the wound to deliver H₂S through a series of channels, to the open wound headspace (~ 5.6 ml) below the laser speckle contrast imaging (LSCI) observation window. The gas within the cavity was sampled through a tailor-made apparatus shown in Fig. 4(b) that attenuated the concentration by 2400x using a combination of silicone diffusion membrane and a < 1-mm diameter aperture to further restrict mass transfer. This apparatus was calibrated with 5,000 ppm calibration gas (GASCO, 34 L-98-5000) using the sensor and calibration approach mentioned previously.

Fig. 4.

(a) Exploded view of apparatus used for real-time characterization of tissue absorption and tissue perfusion visualization with Laser Speckle Contrast Imaging (LSCI); (b) Top view of the adapter showing gas transport from H₂S delivery to open wound cavities through connecting channels sealed against the skin

In-vivo studies of tissue absorption and perfusion response in rodent model

Test subjects studied were adult male and female Sprague Dawley (SD) rats, weighing between 250 g and 450 g, purchased from Charles River Laboratory (USA). Animals were housed in climate-controlled rooms with food and water ad libitum. All handling and procedures were conducted in accordance with the National Research Council’s “Guide for the Care and Use of Laboratory Animals” as part of a protocol approved by the University of New Mexico Institutional Animal Care and Use Committee (IACUC) under the Animal Welfare Assurance # D16-00228 (A3350-01) and USDA Registration # 85-R-0014. After experimentation, animals were euthanized using pentobarbital sodium (FatalPlus 150 mg/kg IP).

For excisional wounds in rodents, animals were induced by inhalation anesthesia with 2.5% isoflurane, the dorsal side of SD rats were depilated and prepped with povidone-iodine and 70% ethanol. Withdraw reflex was tested to ensure proper level of anesthesia prior to conducting a full thickness excisional wound with an 8-mm biopsy punch along the midline between the gluteus maximus. Immediately following wounding, the H₂EALS™ device and the air-tight adapter described in Figs. 5(a) and 5(b) was attached to the dorsum using Duoderm^® and a medical adhesive tape (e.g., 3 M 1509). The rodents were then subjected to a 20-minute acclimation period prior to experiments.

Fig. 5.

(a) In-vivo H₂S measurement apparatus for combined absorption rate and tissue perfusion imaging; (b) complete apparatus of the real-time H₂S delivery, H₂S measurement, and LSCI perfusion characterization in rats; (c) LSCI images showing the ROIs used for perfusion quantification within the excisional open wound and in the periwound area

By using the apparatus described earlier in Fig. 4(a) and shown in Fig. 5, imaging the open wound perfusion was performed in real-time while topically delivering H₂S to the wound bed. Animals were assigned to either a sham (zero H₂S), 25, 50, or 100 nmol H₂S dose group (n = 6, evenly split between male and female rats). Blood perfusion of the open wound and peripheral skin was measured through a borosilicate glass window using LSCI with four distinct regions-of-interrogation or ROIs (Fig. 5(c)). Two-dimensional perfusion maps were recorded in real-time over a 60-min period using a 785-nm laser source at an orthogonal distance of 15 cm at 25 frames per second using a one second time constant and a 4-ms exposure time. After approximately 5 min of measurement (baseline condition), H₂S was administered to the open wound via the H₂EALS™ device. Once H₂S was generated, the H₂EALS™ device and adapter remained closed for the duration of each experiment to allow sufficient time for wound uptake of H₂S, during which time the maximum relative perfusion was achieved and quantified.

To determine H₂S tissue absorption, animals were assigned to groups that either had the peripheral skin available or masked for absorption of H₂S, denoted as S⁺ or S⁻, respectively. The group with blocked skin included a H₂S impermeable membrane between the air-tight adapter and the skin to prevent absorption by tissue. H₂S was delivered to the adapter and then measured pre- and post-wounding for both groups, denoted as S⁺W⁻, S⁺W⁺, S⁻W⁻, and S⁻W⁺. Measurements were taken with the H2S-1010 device described earlier to obtain time-dependent concentrations of H₂S in the adapter. The adapter was opened after the pre-wounding assessment to allow residual H₂S to dissipate and allow time to obtain a new baseline measurement for the post-wounding assessment.

In-vivo studies of tissue absorption, perfusion response, and spread of H₂S in a porcine model

These studies are considered as a pre-requisite for efficacy studies, which is the subject of a follow-on publication. The animal subjects were Yorkshire swine, 4–6 months of age weighing between 24 and 32 kg. The animal protocols were approved by the Texas A&M University IACUC under the OLAW Assurance D16-00511 / USDA Registration 74-R-0012, which conformed to the National Institutes of Health (NIH) Guide for Care and Use of Laboratory Animals (8th edition, revised 2011). For surgical creation of ischemic dorsal soft tissue wounds, swine were pre-anesthetized with glycopyrrolate (0.004 mg·kg-1, i.m.), midazolam (0.5 mg·kg^-1, i.m.), and ketamine (20 mg·kg^-1, i.m.). General anesthesia was induced with propofol bolus (4 mg.kg^-1, i.v.) and maintained with propofol (0.1–0.3 mg·kg·min^-1, i.v.) and fentanyl (1–3 µg·kg·hr^-1, i.v.) throughout. Animals were intubated and ventilated at a rate of 12 to 13 breaths/min. Shaved dorsal skin and ears were prepared with 70% ethanol and chlorohexidine wash. Indwelling 20-gauge intravenous catheters were placed in both ears and heparin locks were secured. Body temperature was maintained with a forced air warming unit (Bair Hugger™). ECG leads were secured to the animal, and an oral temperature probe was placed. Lactated ringers solution was delivered at 5 to 10 ml/kg/hour throughout the procedure. A surgical marker pen and sterile surgical ruler were used to mark the surgical sites of the skin flaps on the dorsal surface of the animal. Four full-thickness, longitudinal incisions parallel to the spine were made using a scalpel blade (#10) and the skin was separated from the underlying fascia to create four bipedicle flap wounds (12 cm x 4 cm), a model used previously to mimic chronic human skin pathology [47, 48]. Shown in Fig. 6, an 8-mm full-thickness punch biopsy excisional wound was then created in the direct center of the ischemic flap. The bipedicle cutaneous flaps were reapproximated and closed in a simple continuous pattern with an undyed monofilament suture (2 − 0).

Fig. 6.

(a) and (b) Images of a domestic male pig on sternal position instrumented with H₂S measurement of breath for detection of H₂S spread into the lungs during delivery, and TAGS™ dermal units placed on 4 positions to monitor potential spread to adjacent sites S3 and S4, before, during and after H₂S delivery with H2EALS™ placed directly over the excisional wound (see insert); (c) View of the wound pattern and size, while showing the placement of the LSCI adaptor (without H₂S source cartridge) for real-time H₂S delivery, H₂S measurement, and LSCI perfusion characterization in swine; (d) LSCI observation window showing the open, excisional wound, (e) Regions-of-Interrogation (ROI) within the wound (darker region on top image) and periwound areas (smaller ROIs around the wound); (f) Perfusion map of the wound and periwound area, where blue and red are the low and high perfusion extremes, respectively

Expired air was captured via a gas sampling elbow attached to the breathing tube and sampled in real-time prior to and during dosing at up to 500 nmol/dose. The purpose of monitoring the relative changes in breath H₂S content with the H2S-1010 device was to evaluate the potential systemic and broad spread of H₂S during H₂S delivery at the wound site (Fig. 6(a)). The TAGS™ device was used to measure the relative changes in H₂S emitted from the skin at the sites adjacent to the point of administration of H₂S (Fig. 6(b)). This was accomplished by simultaneously taking samples during a 10-minute accumulation period with different dermal modules placed on each site and subsequently measuring the H₂S content in the sampled gas. These measurements were made both at sites with incisional open wound (S1 and S2) and reference, unoperated sites with intact skin (S5 and S6). Immediately after completion of delivery of H₂S doses of up to 500 nmol/dose to wound sites S3 and S4, LSCI measures were taken at those sites to assess changes in perfusion in response to H₂S exposure. The Pericam LSCI system from Perimed (Las Vegas, NV) measures were collected from the center of the bipedicle flap (Fig. 6(c)) for baseline perfusion units prior to creation of wounds, after creation of the ischemic flap, after creation of the excisional wound, and after topical delivery of H₂S to excisional wound sites. Real-time perfusion monitoring was performed using the LSCI adaptor as discussed in the previous sections and shown in Fig. 6, in order to assess the percentage of H₂S doses of 100 nmol/dose absorbed by the tissue and wound as well as monitor the physiological response shortly after delivery of H₂S to the wound at several stages of wound healing (i.e., postoperative days 2, 4, and 6 – POD2, POD4, and POD6).

Statistical analysis

Statistical analysis was conducted using the GraphPad Prism 10 software. Data is presented as mean ± standard error of the mean unless otherwise specified. For datasets that followed a normal distribution, comparative analysis was conducted using a t-test between two groups or a two-way analysis of variance (ANOVA) for multiple groups. Datasets that did not show normal distributions were compared using Dunnett’s multiple comparisons test. Significance levels between groups were either indicated numerically or defined as follows: *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001.

Results

Benchtop studies of H₂S delivery

Figure 7 is the result of benchtop evaluation of the solid-state H₂S source cartridges using the experimental approach described in Section “Benchtop studies on H₂S generation, tissue delivery, and absorption” and Fig. 3. Figure 7(a) shows dose-to-dose consistency with a standard deviation of less than 10% in most cases and a correlation coefficient of 0.998 between target and mean average delivered dose. Figure 7(b) is a typical cumulative or total amount of H₂S released by the H₂EALS™ generator module into the 5.3-L container for a typical 200-nmol delivery episode. Note that the peak available H₂S (i.e., point A) is reached at approx. 15 min. In the absence of losses or degradation, instead of a peak, there would be an asymptotic limit representing the total generated and released by H₂EALS™. However, when delivered into a moist, oxygen containing environment that normally exists within a dressing, hydrogen sulfide oxidizes and degrades naturally. This is seen by the decline in the curve to a value (i.e., point B) more than 20% lower than the total generated.

Fig. 7.

(a) Target dose vs. measured mean average delivered dose (n generator per target dose > = 3; n dose per generator > = 6; error bars represent one standard deviation); (b) Illustrative plot showing total H₂S generated by H₂EALS™ module and is available for tissue adsorption for 200 nmol target delivery episode (n = 1)

Figure 8 shows the results of similar experiments performed in an apparatus described in Fig. 4 to evaluate the H₂S availability and the degree of H₂S absorption and degradation in a smaller volume representative of typical dressing. Figure 8(a) shows benchtop results for three different conditions tested (n = 3 per condition), namely, an empty dressing cavity below the H₂EALS™ generator or source, a dry Aquacel^®-filled cavity, and an exudate-wetted Aquacel^®-filled cavity (i.e., at ~ 50% saturation level). The concentration within the empty cavity reaches nearly 800 ppm peak for a 50-nmol delivery episode, resulting from superposition of an exponential increase or a filling process with that of an exponential decline or emptying process. In the absence of attenuation within the cavity (i.e., decay, decomposition, and/or absorption), the concentration would reach an asymptotic limit identified as the equilibrium concentration at which point no more H₂S leaves the H₂EALS™ H₂S source. It is important to note that for the same release amount and rate (i.e., 50 nmol/dose), the peak concentrations are nearly ½ and ¼ of the empty cavity for the dry and wet Aquacel^®, respectively, indicating that Aquacel^® degrades the gas phase H₂S generated by H₂EALS™. This is particularly true when there is moisture in the Aquacel^®. Another observation is that the rate of degradation of H₂S is a function of the size of the cavity in which H₂S is released and is directly a function of the maximum concentration reached within the cavity (i.e., the higher the concentration, the higher the rate of decay). In Fig. 8(b) the concentration within a wound headspace (5.6 mL) during open wound healing in two different swine at postoperative days 2 and 6 (i.e., POD2 and POD6). Note that the lower peak concentrations at POD2 indicate higher absorption by tissue and conversely, lower absorption and higher peak concentrations reached during a H₂S delivery episode. This observation is confirmed by visual evidence of wound closure and smaller wound area in POD6, thus leading to a slower rate of absorption by tissue.

Fig. 8.

(a) The average and standard deviation of estimated H₂S concentration reached above the skin during a 50 nmol dose or release episode from a H₂S cartridge (n = 3 per series, error bars represent on standard deviation); (b) measurement of the instantaneous concentration within the headspace of wound during healing for two different pigs (n = 1 for each series) and at two different post operative days (POD2 and POD6). Note that the timescale is referenced from the trigger point, at which a short pulse of current was supplied to the working electrode. These results indicate that the concentration peak is much higher when there is less absorption or decay (e.g., empty headspace versus when filled with dry or wet Aquacel^® dressing and during early stages of wound healing)

Note that total dosage or amount of H₂S delivered during a treatment period is expressed as follows:

3

D is the total dosage in nmol, is the dose per release event and is the number of release events with equal doses, although the delivery protocol can be changed to have a varying dose during each delivery event. Depending on the delivery protocol requirements, the successive doses can be spaced more closely (i.e., delivered more often) and at a smaller amount for a more sustained and uniform delivery. For example, to deliver 240 nmol per day, the options may range from a one-time 240-nmol dose per day to doses as frequently as every hour at 10 nmol/dose. To avoid cell toxicity, it is more desirable to lower the H₂S dose concentration during each delivery event to allow absorption and dispersion of H₂S within the tissue being treated without creating local concentration ‘hot spot’.

Kinetics of H₂S absorption by tissue

The results shown in Fig. 9 quantitatively depict the rate of tissue absorption of H₂S released within the wound headspace. Two pathways of tissue absorption were postulated: through intact skin and through open wound. In these graphs, baseline conditions were created by masking the skin with a H₂S impermeable layer, thus masking tissue absorption. The attenuated quantities shown at 30 and 50 min from initial H₂S exposure represent the release dynamics and losses to oxidation as discussed earlier. As mentioned before, the – and + indicate masked or exposed skin and/or wound condition by using a properly dimensioned impermeable layer. Indeed, these results show significant differences in H₂S availability between baseline or masked and exposed conditions attributable primarily to tissue absorption through the skin and via open wound in rodents (Fig. 9(a)): the lower the ratio of the average available H₂S to total H₂S delivered, the higher is the expected amount absorbed by wound and periwound tissue. Moreover, the longer exposure from 30 to 50 min appears to significantly increase the absorbed amount, particularly noteworthy in the skin-only exposure condition. In swine wound healing studies (Fig. 9(b)), data were collected for both skin and open wound exposure at various post-operative days (POD) covering up to POD6. Due to the much higher thickness of the skin in pigs versus rats, the contribution of skin permeation is considered insignificant, thus only open wound absorption is assumed to be the dominant mode through which tissue absorption takes place. Notably, there is a significant difference in tissue absorption between the baseline and POD2 to POD6, manifested by the amount of H₂S remaining in the wound headspace after 30 min and 50 min of exposure. Although not significant, the trend in POD2 versus POD4/POD6 suggests that closure of the wound reduces the propensity for tissue absorption.

Fig. 9.

(a) Tissue absorption kinetics in animal models of rodent for POD0 at different skin/wound conditions, where the lower the ratio shown, the higher the expected tissue absorption; and (b) Tissue absorption kinetics in swine at various stages of wound healing, between POD2 and POD6. S + or Skin + are exposed skin, and W + or Wound + are exposed wound conditions. Data points and error bars represent the mean ± standard error from six replicates (n = 6). Asterisks denote statistically significant differences between groups, as determined by a two-way ANOVA (**p < 0.01, ***p < 0.001, ****p < 0.0001)

Perfusion response to H₂S absorption

Figure 10 are the dose-response measurements showing perfusion changes within the excisional wound ROI in response to delivery and absorption of H₂S performed as described in Section “In-vivo studies of tissue absorption and perfusion response in rodent model” Figure 10(a) shows significant increase from baseline conditions and the maximum reached during the 60-minute observation period for Sham (i.e., no skin barrier and no H₂S exposure), 50-nmol H₂S administration over the wound covered with a membrane barrier, and after wound exposure to 50-nmol dose of H₂S. The perfusion response is further highlighted by the shorter response time realized in the condition in the wound clearly exposed to a 50-nmol dose of H₂S. Indeed, in Fig. 10(b), the dose-response time appears to shorten with increasing dosage, although the results do not show statistically significant differences and may require a much larger sample size to achieve lower standard error and deviations to yield conclusive results.

Fig. 10.

(a) Relative perfusion measured with LSCI within the wound for each treatment group tested at 50-nmol dosage condition showing significant perfusion increase when the wound is exposed (i.e., S-/W+ condition) within a 60-minute period after dosage compared to fully masked condition with and without H₂S exposure (S+/W + and Sham, respectively) ; (b) time for perfusion response to H₂S exposure at which point a boost in perfusion exceeding the standard deviation is realized; (c) Dose-response at various doses in male SD rats, showing an increase in the relative perfusion at higher dosages; (d) Relative perfusion increase over baseline for both male and female SD rats, indicating no statistically significant variation between the two sexes. LSCI measurements in the periwound ROIs showed small, non-significant changes, thus are not reported here. Data points and error bars represent the mean ± standard error from six replicates (n = 6) for panels (a) and (b) and eight replicates (n = 8) for panel (c). Asterisks denote statistically significant differences between groups, as determined by a two-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001)

Although baseline tissue perfusion in Fig. 10(c) was highly variable at the start of each experiment, after normalization the maximum perfusion was shown to increase in a dose dependent manner, with statistically significant differences between the baseline condition and those of 50-nmol and 100-nmol doses. The data obtained from the two sexes were separated and plotted in Fig. 10(d), which shows no evident correlation and differences between the two sexes.

Real-time perfusion measurements shown in Fig. 11 were made on two pigs of both sexes within a 60 min period, post H₂S dose of 100 nmol. These measurements were made within the excisional wound area, similar to what was shown in rats. Although the data is not statistically significant, the trend indicate higher perfusion at the early stages of wound healing (i.e., POD2 and POD4). This is believed to correspond to higher absorption of the doses provided, as discussed earlier in relation to Fig. 9, triggering a more pronounced perfusion response. Note that results associated with POD6 provide no indications that there was higher perfusion which is consistent with less absorption of H₂S by tissue. Although the results are not shown here, lower doses resulted in less to no impact on perfusion dose-response neither in the periwound, nor within the open wound areas.

Fig. 11.

(a) and (b) Perfusion dose-response for a male pig at post operative conditions POD2, POD4, and POD6, at high dose of 100 nmol within the wound and periwound areas; (c) and (d) perfusion dose-response experiments similar to (a) and (b) performed on a female swine. Note that despite some differences between female and male, they both respond to higher doses earlier in the wound healing stage, where there is higher absorption of H₂S by a more open wound

H₂S dispersion assessment in porcine animal models

The focus of this section is on the results of work to evaluate the possibility of systemic spread of topically delivered H₂S at the wound site. Although perfusion response is not a direct measure of the spread of H₂S, as discussed in the previous section, there was no measurable impact of H₂S absorption by the open wound on perfusion in the vicinity and in the periwound area, suggesting that the impact of H₂S absorbed by tissue is local. In terms of farther afield, two other measurements were made to assess systemic spread of delivered H₂S, namely measuring detectable levels of H₂S at adjacent sites to the area of delivery and within the expired breath of the animal by real-time sampling of the breath air taken from the anesthesia mask. Neither breath sampling (Fig. 12) nor measurements of the H₂S with the TAGS™ system in the wound sites adjacent to the points of delivery (results not shown) revealed no detectable changes in H₂S concentration between pre- and post- delivery conditions. In Fig. 12(a), the baseline expired air fluctuates between cleaner outside air supplied by the ventilator and the maximum H₂S present in the mask and demonstrated no statistically significant changes in H₂S measures between pre- and post- delivery (Fig. 12(b)) for the delivery amounts tested for at least 7 min post-delivery. Note that measurements at the highest dose concentrations of up to 500 nmol/dose provided no obvious evidence in changes in perfusion in the adjacent sites nor an increase in the baseline H₂S levels expired from the breath.

Fig. 12.

(a) Example of a real-time trace of the breath concentration for a delivery event. Note that the oscillations are the animal breath with the H₂S concentration varying between room air (zero) and maximum breath concentration measured in real-time within the anesthesia mask. (b) Table summary of breath analysis for 3 topical deliveries of H₂S at up to 500 nmol (mean average of n = 750 data points, ~ 6.4 min of data, for each condition, +/- indicates the variation of 1 standard deviation)

Discussion

During the past two decades, significant progress has been made toward understanding the role of H₂S as a signaling molecule. As an endogenous molecule produced within the body, it regulates a variety of biochemical and metabolic processes, such as angiogenesis and vasodilation [49]. The focus of the current research is on topically-administered H₂S, as an exogenous small-molecule supplement, that precipitates a sequence of physiologic responses to promote healing of chronic wounds. The diagram on Fig. 13 captures the pathway to realizing a long-term physiologic outcome. Indeed, while the innovation described herein as H₂EALS™ are primarily related to the first two steps, namely, how to generate H₂S and deliver to the tissue, the ensuing steps depend mostly on the recipient’s response to bioavailability of H₂S in the prescribed form. The results shown in this paper represent a partial progress toward addressing the need for precision delivery of H₂S while actively and quantitatively controlling H₂S bioavailability to tissue for wound treatment. This subject, while recognized in the literature, could benefit from receiving further attention to studies related to H₂S-based therapy, most of which relies on passive delivery methods. In most instances, researchers provide insufficient and unclear evidence that the amount synthesized, delivered to, and absorbed by tissue remained consistent, thus leaving a gap in interpretations and most often inconsistency in results.

Fig. 13.

Sequence of events taking place that cumulatively play a role in producing a long-term physiological rimpact from H₂S treatment

Synthesis

The first step and a crucial part of the H₂S treatment is producing the H₂S either in-situ, as is the case in the current program, or ex-situ by in-vitro production prior to delivery. One of the difficulties of in-vitro production is control over how much and the amount of bioavailable H₂S remaining in solution or is attached to binder/carrier media after conversion of the precursor substance (e.g., donors such as NaHS, Na₂S, or GYY4137). In fact, precise knowledge of conversion efficiency and control over the conversion reaction pathways may be controllable in a laboratory experiment, but difficult to achieve in real practical environments. Longevity of dissolved H₂S is also compromised by rapid decomposition within an aqueous environment and by volatilization loss due to high vapor pressure (i.e., low solubility). Post synthesis, the dissolved H₂S is dissipated within minutes to a fraction of its initially mixed concentration. While managing the volatilized amount is possible, self-decay and oxidation in an aqueous solution is a much more difficult problem to address. It is critical then to monitor the solution concentration during the experiment using a reliable method and report the initial and final, or an average value for each experimental condition rather than rely on an initial estimate based on the molarity of reactants during mixing. In the case of H₂EALS™, the donor itself maintains stability so long as there is no electrochemical motive force. Once a threshold electrical potential is reached and controlled within the production range, the charge transferred during dosing is proportional to the amount of H₂S gas released into a sealed container of electrolyte, raising its concentration to an amount close to the target dose. For example, for a 50 nmol dose, the electrolyte concentration of > 100 µM is produced, which is then released into the headspace through a diffusion barrier within a few minutes, as shown in Figs. 7 and 8. Decay within the solution does not occur in significant amounts and insolubility of H₂S in the H₂EALS™ module electrolyte becomes an asset that is exploited. Stability of the precursor therefore becomes an invaluable trait of H₂EALS™, providing long shelf-life needed for clinical and commercial applications.

Delivery

While this step is often combined with or made indistinguishable from synthesis, it is in practice a step that can be separately managed. In the current study, the in-situ synthesis is taking place at the point of delivery to avoid transfer losses if not in-situ synthesis. Fundamentally, delivering a donor is not the same as delivering bioavailable H₂S, the former leaving uncertainty in terms of how much bioavailable H₂S is present for tissue absorption – a point that cannot be over-emphasized. Figure 8 is a testament that indeed bioavailability of H₂S can vary, even in a gas environment due to oxidation. The 1/e time constant for decay in this case is approximately 35 min from start and 22 min from the peak, and both are strongly a function of the volume of the wound headspace and humidity, among other parameters like exudate management material, e.g., Aquacel^®. Under more attenuative conditions when there is both rapid absorption and decay losses, both the peak and the 1/e time constants shift to the left proportionate to the amount of concentration attenuation. Fully deconvoluting the several anticipated loss mechanisms was, unfortunately, outside of the scope of the current study. While the current study has this limitation, understanding the baseline attenuation behavior is a necessary first step that provides an opportunity to begin to estimate the absorptive properties of the wound during healing (e.g., wound size). It should be noted that while H₂S was delivered topically in the current study, the same approach has been used in separate studies for subcutaneous delivery and uptake of H₂S for treatment of surgical wounds.

Tissue absorption

Indisputably, this is the link between externally bioavailable H₂S and internal, free H₂S within the tissue (i.e., in the interstitial fluid and plasma). For wound healing, the question of tissue absorption of H₂S is broadened by the fact that the path to absorption can be direct into the open wound or indirect via the periwound and through the skin covered by the dressing. The total amount absorbed by both the open wound and the surrounding periwound area may contribute to chronic wound healing, as increased angiogenesis and vasodilation can result in greater perfusion and tissue oxygenation, which are known to enhance tissue remodeling and reconstruction rates in and around the wounded tissue [50]. Some of the experimental results of absorption by wound versus periwound tissue in rodents shown on Fig. 9 suggest that periwound tissue absorption does take place and constitutes as much as 1/3 of the total absorbed with a preferred absorption path being the open wound despite the lower surface area for exposure. Interestingly, when the absorption pathway to open wound tissue is masked (denoted as S+/W-), then periwound absorption becomes appreciable, with potential benefits to the microvascular health beneath the skin.

Wound perfusion

Augmentation of perfusion within and around the wounded tissue can be both a short-term response to H₂S exposure through vasodilation and long-term dose-response through angiogenesis and vascularization. The results provided in this paper cover only the former, initial and immediate dose-response, while the latter is the subject of a separate study and publication. Clearly, there appears to be a minimum exposure or dose threshold before a response is triggered, evident by the lack of perfusion response in the periwound area despite H₂S absorption by this region (see Fig. 11). Although the results seem to suggest that there is higher perfusion related to H₂S absorption associated with POD2 (vs. POD4 or POD6) in the porcine results, there is no clear evidence at this point that the long-term impact on perfusion is not responsible for suppressing the immediate dose-response in POD4 and POD6. Further work on this topic is in progress and will be published in subsequent papers.

Additionally, although the two animal models reflect different physiologies and cannot be compared directly, the porcine model mimicking ischemic wounds while the rodent model exhibiting non-ischemic wounds, this study shows the efficacy of tissue absorption and subsequent changes in tissue perfusion due to administration of topical H₂S from the H₂EALS™ module. In the results, the accumulation of moisture or condensate in the adapter during in-situ experiments were shown not to affect LSCI results, but as previously mentioned, could contribute to the attenuation of H₂S concentrations through oxidation. The contribution of water condensate to the total loss was estimated based on H₂S solubility in water to be less than 10% of the total absorbed thus believed to be within the measurement error but may further need to be studied in situ.

Conclusion

Within this paper, we’ve demonstrated an innovative technique for quantitative synthesis, release, and delivery of H₂S gas to wounded tissue. The technique proves to be a reliable and stable source of measurable and repeatable quantity of H₂S, making it attractive as a candidate for widespread use in various laboratories performing H₂S studies, as well as translatable for commercial use in wound healing. This technology has also the potential to be a platform for a broader set of applications involving H₂S as a healing substance. We have demonstrated that the H₂S gas generated in the headspace of open wounds is not only directly and rapidly absorbed into the wounded tissue, but also indirectly permeates into the periwound skin, thus promoting immediate perfusion augmentation in the short term and enhanced wound healing over the duration of healing period. Of important clinical relevance is our findings from breath analysis and measurement within the surrounding tissue that while local perfusion of H₂S affects exposed tissue, there was no measurable systemic exposure.

Abbreviations

CE

Counter-electrode

ECM

Electronic Control Module

ECG

Electrocardiogram

H₂S

Hydrogen Sulfide

IACUC

Institutional Animal Care and Use Committee

LSCI

Laser speckle contrast imaging

NIST

National Institute of Standards and Technology

PAD

Peripheral Artery Disease

ppb

Parts per billion

PTFE

poly(tetrafluoroethylene)

RE

Reference electrode

ROI

Region of Interrogation

SD

Sprague-Dawley (Rat)

SNR

Signal-to-Noise Ratio

VEGF

Vascular Endothelial Growth Factor

WE

Working electrode

Author contributions

The authors would like to thank a number of individuals who helped along the way with technical support and realization of the experiments, some of which were presented in this paper. These individuals include but are not limited to Benjamin T. Matheson, PhD, Joseph Giacolone, MD, PhD, and Gary C Cohn, PLLC.

Funding

Support for these studies was provided by grants provided by National General Medical Sciences SBIR Program under the Award GM144027 and National Institute of Aging SBIR under the Award AG076027. RMC is additionally funded by the National Center for Advancing Translational Sciences grant K12TR005467.

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

Animal studies were performed under the OLAW Assurances OLAW Assurance D16-00511 / USDA Registration 74-R-0012 (TAMU) and D16-00228/USDA Registration # 85-R-0014 (UNM).

Competing interests

Reza Shekarriz is the director of Exhalix LLC, a for-profit R&D organization and commercialization of the device hereto presented in this paper.