Research paperLocalized delivery of carbon monoxide
Graphical abstract
Introduction
The heme oxygenase (HO)/carbon monoxide (CO) system is an important cellular feedback-loop towards oxidative and inflammatory insults in man (Fig. 1) [3]. The central building block of this system is the HO-1, an inducible enzyme catalyzing heme degradation into CO, biliverdin, and Fe2+. Detailed reports on cell-protective, anti-oxidative, anti-apoptotic, and anti-inflammatory effects of the HO-1 and its effectors (including CO) have previously been outlined [3]. Polymorphisms impairing upregulation of this enzyme have been linked to various pathophysiological conditions including coronary artery disease [4], diabetes [5], or the outcome of organ transplantations [6]. The disease modifying impact of CO was demonstrated in disease models mainly by systemic application either directly applying CO mostly through inhalation or CO-Releasing Molecules (CORM) [3]. Both modalities face substantial challenges driving the need for novel approaches as discussed below. Inhaled CO leads to one principal challenge of this (and other systemic) administration routes as of the rather exhaustive binding of CO to hemoglobin (Hb), e.g. following diffusion through the alveolar membrane when pulmonary administered. Therefore, systemic trafficking of pulmonary delivered CO is mainly in a tightly bound form as the carboxyhemoglobin (CO-Hb) complex and very little ‘free’ CO is available to act pharmacologically (Fig. 2A). Noteworthy, CO also tightly binds to a variety of other heme-containing proteins including myoglobin [7], neuroglobin [8], and cytochrome C oxidase [9]. For simplification and comparability among various preclinical and clinical reports this review will focus on CO-Hb among these CO binding moieties. The extraordinary stability of CO-Hb requires quite high CO-Hb plasma concentrations such that therapeutically relevant levels of free CO are achieved – critically narrowing the therapeutic window [10]. This conundrum has to be solved for any systemic CO delivery attempt, as very high, perhaps toxic CO-Hb levels have to be condoned in order to achieve therapeutically relevant levels of free CO in the target tissues (Fig. 2A) [11]. This disadvantageous prerequisite is also true for the delivery of CO using systemically administered CORMs. On top, these CORMs aggravate the development risk in that they further hold potentially toxic transition metals required for binding CO (Fig. 2A) [3], [12].
Therefore, and in spite of the exciting therapeutic potential of CO, current systemic application strategies resulted in an uphill battle against both, the CO-Hb to free CO ratio and potential systemic safety concerns associated with the molecular scaffolds most CORMs deploy. Consequently, novel developments point to local delivery and targeting approaches addressing these concerns. These include CORMs that decarbonylate in diseased hepatic tissue (thereby, the CO is carried into the liver before it releases from the CORM – therapeutically relevant free CO is increased while CO-Hb percentages can be kept low) as well as Drug Delivery Systems (DDS) that locally control CO release in the gastrointestinal (GI) tract (thereby further reducing unfavorable CO-Hb binding while maximizing exposure of the target tissue to free CO) [13], [14]. This review will start off past and current CO delivery concepts and therapeutic evidence, define desirable pharmaceutical specifications turning CO into a druggable molecule and finally link to future delivery strategies aiming at translating CO into effective medication of tomorrow.
Nature’s equivalent for the therapeutic approaches discussed here within is HO-1 being central to the local feedback loop of the HO/CO system (Fig. 1). Technologies designed for mimicking this loop, therefore, preferably simulate (i) signaling and (ii) the special spatiotemporal characteristics of this system (vide infra). Resembling the function as a cellular sentinel, the enzyme is induced by an array of diverse imbalances including inflammation, radiation, and hypoxia [15]. Multiple transcriptional factors including activator protein 1 (AP-1), NF-κB, nuclear factor like 2 (NRF-2), cAMP response element-binding protein (CREB), and enhancer box (E-Box) induce HO-1 expression and are themselves controlled through a multitude of pathways including extracellular-signal regulated kinases (ERK), p38 mitogen-activated protein kinases (p38 MAPK), as well as c-Jun N-terminal kinases (JNK) activation [15]. Thereby, HO-1 orchestrates a network of para- and autocrine events providing a protective response (Fig. 1). Autocrine effects are mainly attributed to the high diffusivity and stability of CO, a special characteristic that differentiates CO from other signaling molecules [16]. Hemoglobin is tightly controlling this spatial distribution by instantaneously scavenging CO following passage to the capillary networks (Fig. 1) [17].
The mobility of CO as well as its ability to induce HO-1 has been discussed as important features for effective exogenous application of CO and CORMs [16], [18]. Likewise, genuine HO-1 inducers including cobalt protoporphyrin and hemin have been used as almost equivalent lab stage alternative [3], [19]. Induction is also observed in response to various other compounds including drugs like curcumin [20], acetylsalicylic acid [21], resveratrol [22], lansoprazole [23], [24], and 5-aminosalic acid [25]. The lack of selectivity of this mode of action leading into numerous downstream effects, however, challenged the development of HO-1 inducers for therapeutic purposes [26], [27]. Until today, induction of HO-1 in man was solely demonstrated for hemin [28], which failed in a trial profiling for gastroparesis both in prolonged HO-1 induction (>7 days) and ameliorating the disease activity [29] contrasting previous preclinical trials [30], [31]. Furthermore, hemin is a neurotoxic [32], fairly unselective molecule that besides HO-1 induction represses δ-aminolevulinic acid synthase, the rate-limiting enzyme for the synthesis of tetrapyrrols [15], [33]. This feature is clinically relevant for the treatment of acute porphyria – a rare disease characterized by porphyrin accumulation – for which hemin infusions are used [34]. Therefore, novel effectors of the HO-1 pathways are required and CO is a promising option discussed here within.
Section snippets
Transferability of preclinical results
The systemic application of CO by inhalation demonstrated efficacy in numerous preclinical models [3] contrasting clinical trials: Promising preclinical evidence on endotoxemia [35], [36] was not confirmed in a clinical trial exposing healthy volunteers to 500 ppm CO for one hour resulting in 7% CO-Hb [37]. Likewise, a safety and tolerability study using inhaled CO in kidney transplant patients motivated by promising preclinical results [38], [39], [40] was withdrawn for unknown reason [41].
CO-Releasing Molecules
Metal carbonyl complexes for CO delivery were introduced in 2002 [84]. Starting off these initial studies, various classes of CORMs have been developed with the ultimate goal to combine efficacy and safety as fundamental requirements for drug approvals (Fig. 3). Novel targeting approaches including CORMs for hepatic CO delivery significantly reduce safety issues associated with systemic CO toxicity. Metal toxicity, however, is a golden threat running through all metal based CORMs and has been
Pharmaceutical formulations of CORMs
Within the last years an emerging number of controlled drug delivery systems (DDS) have been introduced for CORMs. Some of these DDS feature first generation CORMS for stimulus driven CO generation, thereby forming an alternative approach to the exciting achievements reported for third generation CORMs (vide supra). Successes were reported for systemic as well as non-invasive use [92], [98].
Conclusion
The therapeutic potential of local CO delivery is enormous as physiological mechanisms have been elucidated and linked to various preclinical mechanistic as well as efficacy studies. Likewise, numerous approaches for the delivery of this challenging gaseous drug have been introduced, thereby providing a versatile toolbox for future translational approaches. Inspired by these, an emerging number of local and target delivery approaches have been introduced addressing the major issue of systemic
References (145)
- et al.
Association between the heme oxygenase-1 promoter polymorphism and renal transplantation outcome in Greece
Transplant. Proc.
(2010) - et al.
Determination of carbon monoxide (CO) in rodent tissue: effect of heme administration and environmental CO exposure
Anal. Biochem.
(2005) - et al.
Oral drug delivery of therapeutic gases – carbon monoxide release for gastrointestinal diseases
J. Control. Release
(2014) - et al.
The social network of carbon monoxide in medicine
Trends Mol. Med.
(2013) - et al.
Curcumin, an antioxidant and anti-inflammatory agent, induces heme oxygenase-1 and protects endothelial cells against oxidative stress
Free Radic. Biol. Med.
(2000) - et al.
Heme oxygenase-1 induction may explain the antioxidant profile of aspirin
Biochem. Biophys. Res. Commun.
(2003) - et al.
The involvement of heme oxygenase-1 activity in the therapeutic actions of 5-aminosalicylic acid in rat colitis
Eur. J. Pharmacol.
(2008) - et al.
Heme oxygenase-1 drives metaflammation and insulin resistance in mouse and man
Cell
(2014) - et al.
Gene regulation of heme oxygenase-1 as a therapeutic target
Biochem. Pharmacol.
(2000) - et al.
Factors predicting duration of delayed graft function in non-heart-beating donor kidney transplantation
Transplant. Proc.
(2005)