Carbon monoxide (CO) is endogenously produced in mammals, with blood concentrations in the high micromolar range in the hemoglobin-bound form. Further, CO has shown therapeutic effects in various animal models. Despite its reputation as a poisonous gas at high concentrations, we show that CO should have a wide enough safety margin for therapeutic applications. The analysis considers a large number of factors including levels of endogenous CO, its safety margin in comparison to commonly encountered biomolecules or drugs, anticipated enhanced safety profiles when delivered via a noninhalation mode, and the large amount of safety data from human clinical trials. It should be emphasized that having a wide enough safety margin for therapeutic use does not mean that it is benign or safe to the general public, even at low doses. We defer the latter to public health experts. Importantly, this Perspective is written for drug discovery professionals and not the general public.

Carbon monoxide (CO) is an endogenous signaling molecule with demonstrated pharmacological effects. For studying CO biology, there is a need for sensitive and selective fluorescent probes for CO as research tools. In developing such probes, CO gas and/or commercially available metal-carbonyl-based "CO-releasing molecules" (CORMs) have been used as CO sources. However, new findings are steadily emerging that some of these commonly used CORMs do not release CO reliably in buffers commonly used for studying such CO probes and have very pronounced chemical reactivities of their own, which could lead to the erroneous identification of "CO probes" that merely detect the CORM used, not CO. This is especially true when the CO-sensing mechanism relies on chemistry that is not firmly established otherwise. Cu²⁺ can quench the fluorescence of an imine-based fluorophore, DPHP, presumably through complexation. The Cu²⁺-quenched fluorescence was restored through the addition of CORM-3, a Ru-based CORM. This approach was reported as a new "strategy for detecting carbon monoxide" with the proposed mechanism being dependent on CO reduction of Cu²⁺ to Cu¹⁺ under near-physiological conditions ( Anal. Chem. 2022, 94, 11298-11306). The study only used CORM-3 as the source of CO. CORM-3 has been reported to have very pronounced redox reactivity and is known not to release CO in an aqueous solution unless in the presence of a strong nucleophile. To assess whether the fluorescent response of the DPHP-Cu(II) cocktail to CORM-3 was truly through detecting CO, we report experiments using both pure CO and CORM-3. We confirm the reported DPHP-Cu(II) response to CORM-3 but not pure CO gas. Further, we did not observe the stated selectivity of DPHP for CO over sulfide species. Along this line, we also found that a reducing agent such as ascorbate was able to induce the same fluorescent turn-on as CORM-3 did. As such, the DPHP-Cu(II) system is not a CO probe and cannot be used to study CO biology. Corollary to this finding, it is critical that future work in developing CO probes uses more than a chemically reactive "CO donor" as the CO source. Especially important will be to confirm the ability of the "CO probe" to detect CO using pure CO gas or another source of CO.

The dopaminergic system plays important roles in neuromodulation, including prominent roles in complex neurological functions such as cognition, reward, motivation, and memory. Understandably, the highly complex nature of such physiological functions means that their regulation is intertwined with other signaling pathways, as has been demonstrated by numerous studies. Contrary to its public perception of being poisonous at all concentrations, carbon monoxide (CO) is produced endogenously from heme degradation by heme oxygenase (HO) as part of the physiological process of red blood cell turnover. Physiological concentrations of CO can reach high micromolar ranges in the hemoglobin bound form. Low-dose CO has shown therapeutic effects in numerous animal models, including traumatic brain injury via engaging various hemoprotein targets. As such, the HO-CO axis has been shown to offer beneficial effects in organ protection, anti-inflammation, and neuroprotection, among many others. Further, a large number of publications have shown the interactions among CO, HO, and the dopaminergic system. In this review, we critically examine such experimental evidence in a holistic fashion and in the context of a possible dopamine-HO-CO signaling axis. We hope that this Perspective will stimulate additional investigations into the molecular connectivity related to this possible axis and open doors to the development of novel therapeutics that impact the dopaminergic system.

Carbon monoxide (CO) is endogenously produced with a range of pharmacological activities. Sensitive and selective detection of CO is critical to studying its biology. Since the first report of a CO fluorescent probe in 2012, more than 100 papers on this topic have appeared. Noteworthy in such work is the widespread use of two commercially available ruthenium-carbonyl complexes (CORM-2 and CORM-3) as CO surrogates. Unfortunately, these two CORMs are chemically very reactive and preferentially release CO₂ but not CO, unless in the presence of a nucleophile. As a result, there are "two tales" of the reported CO probes: those that detect CO and those that detect only the CORM used but not CO. In addition, because of their lack of reliable CO production and fast degradation in an aqueous solution, there is the question of what "detecting CORM-2 or CORM-3" really means in the context of CO research. Additionally, for applying fluorescent CO probes in detecting low levels (often nanomolar) of CO in vivo, fast reaction kinetics is a prerequisite for meaningful results. In this Perspective, we discuss in detail these issues with the understanding of the evolutionary nature of scientific discoveries and the aim of preventing further confusion.

Because of endogenous signaling roles of carbon monoxide (CO) and its demonstrated pharmacological effects, there has been extensive interests in developing fluorescent CO probes. Palladium-mediated CO insertion has been successfully used for such applications. However, recent years have seen many publications of using uncatalyzed CO insertion into a hydrazone double bond as a way to sense CO. Such chemistry has no precedents otherwise. Further, the rigor of the CO-sensing work was largely based on using ruthenium-carbonyl complexes such as CORM-3 as CO surrogates, which have been reported to have extensive chemical reactivity and to release largely CO₂ instead of CO unless in the presence of a strong nucleophile such as dithionite. For all of these, it is important to reassess the feasibility of such a CO-insertion reaction. By studying two of the reported "CO probes" using CO gas, this study finds no evidence of CO insertion into a hydrazone double bond. Further, the chemical reaction between CO gas and a series of eight hydrazone compounds was conducted, leading to the same conclusion. Such findings are consistent with the state-of-the-art knowledge of carbonylation chemistry and do not support uncatalyzed CO insertion as a mechanism for developing fluorescent CO probes.

Despite current vaccines and therapeutics targeting SARS-CoV-2, the causative agent of the COVID-19 pandemic, cases remain high causing a burden on health care systems. Spike-protein mediated membrane fusion of SARS-CoV-2 is a critical step in viral entry. Herein, we describe entry inhibitors identified by first screening a library of about 160 compounds and then analogue synthesis. Specifically, compound 261 was found to inhibit SARS-CoV-2 infection in a tissue model with IC₅₀ of 0.3 µM. Using NMR, we found that 261 interacts with key residues in the aromatic-rich region of the spike protein directly next to the transmembrane domain. Molecular dynamic simulations of the 261 binding pocket in the spike protein was also mapped to the transmembrane domain, consistent with NMR findings. The amino acids in the binding site are conserved among different coronaviruses known to infect humans; therefore, inhibitors targeting this conserved binding site could be a useful addition to current therapeutics and may have pan-coronavirus antiviral activities.