Breakthroughs in the development of antibiotics, antifungals and antiparasitics targeting the pathogens’ respiratory chain

Abstract

The aerobic respiratory chain is vital to bacterial and eukaryotic cell energy transformation. Embedded in the mitochondrial inner membrane and the bacterial plasma membrane, the respiratory chain couples sequential redox reactions with ion pumping, thereby generating the motive force that is used to drive ATP synthesis. Due to the essential role of oxidative phosphorylation in cellular life, the electron transport chain proteins, their cofactors, and ATP synthase components serve as a target for antibacterial, antifungal, and antiparasitic drugs. Whether by (1) inhibition of electron flow through transport chain complexes, (2) collapsing of the motive force, (3) competitive inhibition, or (4) blocking proton flow through the catalytic subunits of ATP synthase, small molecules can selectively inhibit bacterial, fungal, and parasitic life while not showing high toxicity in mammalian systems. Because of robust antimicrobial resistance against the traditional mechanisms of microbial control (cell wall integrity, protein synthesis, nucleotide and nucleic acid synthesis, etc.), the study of alternative targets, such as the respiratory chain, is prudent and timely. This review summarizes the current research on small molecule and peptide inhibition of the aerobic respiratory chain complexes, electron flow, and ion translocation in a series of human and plant pathogens.

Introduction

The mitochondrial respiratory chain and its role in cell energetics

The respiratory chain, also known as the electron transport chain, plays a crucial role in aerobic organisms by serving as the final stage of cellular respiration, supporting the majority of ATP production. This metabolic pathway is ubiquitous, consisting of a series of protein complexes and electron carriers that transfer electrons supplied by NADH, succinate, UQH₂, and other sources that reduce flavoenzymes (FADH₂/FMNH₂), which are derived from nutrients like glucose, amino acids, and lipids (Lehninger 1971). As electrons move through the chain and are delivered to the final electron acceptor(s), energy is released in the form of Δp, which is used to synthesize ATP. In eukaryotes, the respiratory chain is located in the inner mitochondrial membrane and consists of four multi-subunit redox protein complexes (I-IV) and two mobile electron carriers: ubiquinone and cytochrome c (Nicholls and Ferguson 2002; Wikström 2005) (Figure 1A). These components work concertedly to oxidize NADH and other electron carriers generated during glycolytic, tricarboxylic acid cycle and fatty acid metabolism, via flavoenzymes, and reduce molecular oxygen. This reaction represents a physiological step that distinguishes aerobic respiration from anaerobic counterparts. As electrons flow through the transport chain, protons are pumped from the mitochondrial matrix into the intermembrane space (Nicholls and Ferguson 2002; Wikström 2005). The coupling of these reactions thereby creates an electrochemical gradient that drives ATP synthesis via complex V, completing the process of oxidative phosphorylation (Mitchell 2011).

Figure 1.

Human (A), bacterial (B) and parasitic/fungal (C) respiratory chains.

The mitochondrial respiratory complexes play distinct roles in electron transport chain physiology. Complex I (NADH:ubiquinone oxidoreductase) transfers electrons from NADH to ubiquinone, and pumps protons from the matrix to the intermembrane space, with a H⁺/e⁻ stoichiometry of 2:1. Complex II (succinate: ubiquinone oxidoreductase) or succinate dehydrogenase (SDH), forms part of the Krebs cycle and the respiratory chain, and oxidizes succinate to fumarate while delivering electrons to ubiquinone. Complex III (ubiquinol:cytochrome c oxidoreductase) or cytochrome bc₁ complex, oxidizes ubiquinol to ubiquinone and transfers electrons to cytochrome c, translocating protons from the matrix to the intermembrane space, with a H⁺/e⁻ of 2:1. Complex IV (cytochrome c oxidase) oxidizes cytochrome c, reduces molecular oxygen to water, and pumps protons from the matrix to the intermembrane space, with a H⁺/e⁻ ratio of 1:1. Complex V (ATP synthase) facilitates the movement of protons along their concentration gradient back into the matrix through its two ion hemichannels, utilizing the energy released by the process to synthesize ATP from ADP and inorganic phosphate (Roberts GCK 2013; Cook et al. 2014). These five complexes work together to establish the proton motive force across the inner mitochondrial membrane, whose energy is harnessed and transduced to more usable cellular energy currency in the form of ATP.

The bacterial respiratory chain

The bacterial respiratory chain is located in the cell membrane and differs from the relatively simple mammalian respiratory chain, as it is able to obtain electrons through various routes due to its modular and branched architecture (Richardson 2000; Cook et al. 2014; Kaila and Wikström 2021) (Figure 1B). Branched respiratory chains containing “alternative” complexes provide flexibility to use respiratory components in different patterns, allowing them to thrive in diverse niches and under adverse conditions (Roberts GCK 2013; Cook et al. 2014; Kaila and Wikström 2021). For instance, many types of microorganisms can readily switch from aerobic to anaerobic environments by expressing terminal oxidases which have adapted to use electron acceptors other than oxygen, like nitrate or fumarate (Unden and Bongaerts 1997; Richardson 2000). Bacteria share the same, albeit structurally different, respiratory enzymes with mitochondria, including complexes I-IV. Bacterial complex I comprises 13–14 subunits, compared to >45 in mitochondria (Yagi 1991; Schägger 2002; Nolfi-Donegan et al. 2020). On the other hand, SDH is highly similar to mitochondrial complex II (Hägerhäll 1997; Schägger 2002; Nolfi-Donegan et al. 2020). Complex III is composed of only 3–4 subunits in comparison with its eukaryotic counterpart, which comprises 8–11 subunits (Schäfer and Penefsky 2008; Xia et al. 2013; Nolfi-Donegan et al. 2020). Aa₃-type cytochrome c oxidases are found in the mitochondrial (Complex IV) and bacterial respiratory chains (Schägger 2002; Nolfi-Donegan et al. 2020; Borisov et al. 2021). The respiratory chain is coupled to ATP production via F-ATPases (Complex V), found in all domains of life (Schäfer and Penefsky 2008; Cook et al. 2014; Ahmad M et al. 2023) and conserved between aerobic and anaerobic organisms. The mitochondrial complex V contains non-catalytic supernumerary subunits, absent in bacterial counterparts (Vaillier et al. 1999; Lee J et al. 2015).

In addition to these “core” enzymes, bacterial respiratory chains contain accessory enzymes, which at first sight appear redundant, but play essential roles in different conditions. In addition to complex I, bacteria contain two other NADH dehydrogenases that transfer electrons from NADH to ubiquinone, including NDH-2 and NQR (Yagi 1991). NDH-2 is a ubiquitous single subunit enzyme that lacks proton pumping activity and contains FAD as the sole cofactor (Grandier-Vazeille et al. 2001; Melo et al. 2004; Saleh et al. 2007; Marreiros et al. 2016). NQR is a six subunit respiratory complex found mostly in proteobacteria, Chlamydiae, Chlorobi and Bacteroides (Reyes-Prieto et al. 2014). This is the only respiratory complex that is able to pump sodium, using a unique ion transport mechanism, which drives processes related to homeostasis and is also involved in pathogenesis (Juárez et al. 2010). In addition to several NADH dehydrogenases, the bacterial respiratory chains use a wide variety of other dehydrogenases that directly feed the respiratory chain, allowing the operation of a flexible metabolism. These dehydrogenases include malate, ethanol, glucose, and lactate, among others (Liang P et al. 2020; González-Montalvo et al. 2024). In contrast with eukaryotes, which have Complex IV as the only terminal oxidase, bacteria have a wider array of terminal oxidases that support aerobic metabolism (Richardson 2000,Nolfi-Donegan et al. 2020; Roberts GCK 2013; Kaila and Wikström 2021). Bacteria carry several types of heme-copper oxidases that differ from the eukaryotic Complex IV in structure, subunit composition, electron donors, and stoichiometric reactions. These oxidases include aa₃ and ccb₃ cytochrome c oxidases, as well as bo₃ ubiquinol oxidases (Roberts GCK 2013; Borisov et al. 2021; Kaila and Wikström 2021; Siletsky and Borisov 2021). Furthermore, bacteria also carry non-copper bd-type ubiquinol oxidases (Roberts GCK 2013; Borisov et al. 2021; Kaila and Wikström 2021; Siletsky and Borisov 2021). In addition to the respiratory enzymes, the quinone carrier preference in bacteria differs from eukaryotes. While mitochondria exclusively use ubiquinone, anaerobic and microaerophilic bacteria use menaquinone (Søballe and Poole 1999; Franza and Gaudu 2022). The great flexibility in the components of bacterial respiratory chains allows quick adaptation and metabolic switching depending on the growth conditions or environmental cues, making bacteria highly adaptable. Due to this complexity, the study of the respiratory chain is critical, as several of its components are emerging as potential targets for drug development.

Fungal and parasitic respiratory chains

Fungal and several types of eukaryotes, including parasites, carry the core components described in the mammalian respiratory chain, complexes I-V. In addition, these eukaryotes contain accessory (alternative) respiratory complexes (Figure 1C), including NDH-2, which can be internal (NDH-2i), facing the mitochondrial matrix, or external (NDH-2e), facing the periplasmic space (Kerscher et al. 2008). Moreover, these organisms also carry an alternative oxidase, AOX, which bypasses complexes III and IV, and transfers electrons directly from ubiquinol to oxygen (Elthon and McIntosh 1987; Juárez et al. 2006), similar to bo₃ and bd oxidases, but without pumping protons in the process. These complexes appear to be important for pathogenicity in some species, offering flexibility to the respiratory chain that allows for redox pool regulation and operation of the respiratory chain under stress conditions (Kerscher et al. 2008; Moore et al. 2013; Shiba et al. 2013).

The respiratory chain as a target for antibiotic development

Antimicrobial resistance (AMR) is among the most pressing public health challenges of our time. It causes approximately 1.7 million deaths annually and is a contributing factor in nearly 5 million deaths worldwide (Murray et al. 2022; Naghavi et al. 2024), with associated economic costs of billions of dollars per year. It is projected that by 2050, AMR will produce 10 million deaths yearly (Naghavi et al. 2024; O’Neill 2016), with one trillion dollars in additional healthcare costs in the US alone (Jonas et al. 2017; Dadgostar 2019). The extensive use of antibiotics has produced the emergence of multidrug-resistant (MDR) strains for most human pathogens (Putman et al. 2000; Alanis 2005; Davies and Davies 2010; Andersson and Hughes 2011), which greatly complicates the treatment of infectious diseases. AMR resistance now includes all classes of commercially available antibiotics (Alanis 2005; Hurdle et al. 2011; Avner et al. 2012), and will likely expand to newer generations of antibiotics, since in most cases they are directed against the same molecular targets and are based on the same general structures (Shahid et al. 2009; Sykes 2010; Bassetti et al. 2013; Bassetti and Righi 2015). The overexploitation of certain antibiotic scaffolds and lack of new suitable targets pose a major challenge in drug discovery (Overington et al. 2006; Hurdle et al. 2011; Lewis 2013; Santos and Torres 2013). Current classes of antibiotics, to which pathogens have already acquired resistance, all target the same biological processes: DNA replication, transcription, protein biosynthesis, and cell wall synthesis. Therefore, it is crucial to identify novel molecular targets for the development of new classes of antibiotics against essential cellular processes for innovative antimicrobial treatments.

Bacterial energy metabolism, and in particular pathogenic respiratory chains, represent interesting molecular targets for the development of new antibiotics. The respiratory chains of pathogenic organisms play essential roles in cell physiology, sustaining the entire metabolic network and all transport activities throughout the membrane, which are crucial to maintaining homeostatic processes and growth. Notably, pathogenic respiratory chains are critical in disease pathogenesis, making them highly attractive pharmacologic targets. Recently, the respiratory chain has been the target for novel antimicrobials against different bacteria. For instance, targeting the respiratory chain has been successful for developing new antibiotics against Mycobacterium tuberculosis, which produces one of the deadliest infections in the world. Tuberculosis treatment is complex, spanning 6–9 months with multiple antibiotics, and can exceed two years in MDR cases (CDC 2024). M. tuberculosis is notorious for producing dormant forms that cause persistent infections and treatment failure. However, recent developments show that inhibition of the M. tuberculosis respiratory chain represents a successful pathway for infection elimination (Sindhu and Debnath 2022). M. tuberculosis ATP synthase (Mahajan 2013), NDH-2, and Complex III are predicted to be essential for optimal growth in this pathogen, and are being pursued as drug targets (Pethe et al. 2013; Sellamuthu et al. 2017; Beites et al. 2019; Kim J et al. 2022; Capela et al. 2023). Thus, understanding the operation of bacterial, fungal, and parasitic respiratory chains, and how to disrupt them, can provide critical insights for the development of novel and more effective antimicrobials. By targeting specific components and mechanisms within these respiratory chains, compounds that are more selective, stable, and less prone to resistance can be selected or synthesized, ultimately enhancing their therapeutic potential. This work describes recent advances in the development of new antibiotics, antifungals and antiparasitics that target the respiratory chain and ATP synthase, as well as membrane-based bioenergetics.

Respiratory enzymes as metabolic targets for antibiotic, antifungal, and antiparasitic development

NDH-2

The non-proton pumping Type II NADH: ubiquinone oxidoreductases (NDH-2) comprise a group of five different isotypes (A-D) (Marreiros et al. 2016) that are found in all domains of life, but are absent in mammals (Melo et al. 2004). NDH-2s are widely distributed among bacteria and serve as the primary NADH dehydrogenase in several pathogens including Staphylococcus aureus, Streptococcus agalactiae, M. tuberculosis, M. leprae, Klebsiella aerogenes, Escherichia coli, and Listeria monocytogenes (Schurig-Briccio et al. 2014; Yano et al. 2014; Lencina et al. 2018; Beites et al. 2019; Smith et al. 2023; González-Montalvo et al. 2024). NDH-2 regulates the NADH/NAD⁺ ratio by uncoupling NADH oxidation from the generation of the ion gradient, as this enzyme lacks H⁺ pumping activity (Gyan et al. 2006; Vamshi Krishna and Venkata Mohan 2019). Since NDH-2 is not inhibited by ionic gradients, it allows for an increase in biosynthetic pathway involvement and ATP synthesis while simultaneously reducing ROS formation (Cook et al. 2014). NDH-2 is a 40–70 kDa single subunit (Heikal et al. 2014; Nakatani et al. 2017) homodimeric protein (Heikal et al. 2014; Blaza et al. 2017; Nakatani et al. 2017) that belongs to the two-dinucleotide binding domains flavoprotein superfamily (Ojha et al. 2007; Marreiros et al. 2017), which has two Rossmann folds (Blaza et al. 2017; Marreiros et al. 2017; Nakatani et al. 2017) forming the non-covalently bound FAD cofactor site and the NADH/NADPH binding site (Blaza et al. 2017; Marreiros et al. 2017) (Figure 2A,B). In bacteria, NDH-2 is a monotopic enzyme bound to the cytoplasmic side of the innermost membrane by two amphipathic helices (Marreiros et al. 2017; Nakatani et al. 2017). In fungi and parasites, NDH-2 can be found on the external or internal sides of the mitochondrial inner membrane (Feng Y et al. 2012; Heikal et al. 2014).

Figure 2.

Structure of NDH-2 accessed from PDB ID: 6BDO (Berman et al. 2000; Petri et al. 2018). (A) Tri-dimensional structure with inhibitor HQNO and cofactor FAD shown as spheres, light gray: FAD, dark gray: HQNO. (B) Close view of FAD binding. (C) Close view of HQNO binding. (D) Two-dimensional structure of NDH-2 inhibitors.

NDH-2 dehydrogenases are suitable drug targets due to their absence in human mitochondria. Indeed, several compounds have been identified as NDH-2 inhibitors (Cook et al. 2014; Heikal et al. 2014), such as phenothiazines, quinolinyl pyrimidines (Shirude et al. 2012), and quinoline quinones (Heikal et al. 2016) (Figure 2D; Table 1). M. tuberculosis studies show that NDH-2 is essential for growth in laboratory media (Sassetti et al. 2003; Griffin et al. 2011; Nakatani et al. 2020), and cells lacking this protein exhibit reduced virulence in vivo (Beites et al. 2019). Mycobacterial NDH-2 can be effectively inhibited using phenothiazines such as chlorpromazine and trifluoperazine (Amaral et al. 2001; Weinstein et al. 2005; Yano et al. 2006) (Figure 2D). Phenothiazine analogues have shown promising results in a mouse infection model, where it was demonstrated to reduce bacterial load by 90% over the course of an 11-day treatment (Weinstein et al. 2005). Moreover, these molecules are effective in the treatment of MDR-tuberculosis (Sellamuthu et al. 2017). In addition to phenothiazines, phenazines such as clofazimine have also shown efficacy as antibiotics against mycobacteria. Clofazimine, an FDA-approved drug against leprosy, was recently repurposed to treat MDR-tuberculosis, targeting NDH-2 (Lechartier and Cole 2015). The action mechanism of clofazimine is interesting as it is reduced by NDH-2, producing ROS, which dissipate the membrane potential (Feng X et al. 2015) and are bactericidal (Yano et al. 2011).

Table 1.

Reported NDH-2 inhibitors.

Compound	Microorganism targeted	IC50	MIC50	Toxicity
Trifluoperazine	M. tuberculosis	12 μM (Yano et al. 2006)	MIC 9 μg/mL (Yano et al. 2006)	Hepatotoxic (Long term therapy) (National Institute of Diabetes and Digestive and Kidney Diseases 2012a)
Chlorpromazine	M. tuberculosis	10 μM (Weinstein et al. 2005)	MIC 19 μg/mL (Weinstein et al. 2005)	Hepatotoxic (Long term therapy) (National Institute of Diabetes and Digestive and Kidney Diseases 2012b)
Trifluoperazine	S. aureus	19 μM (Schurig-Briccio et al. 2014)	70 μM (Schurig-Briccio et al. 2014)	Hepatotoxic (Long term therapy) (National Institute of Diabetes and Digestive and Kidney Diseases 2012a)
Thioridazine	S. aureus	7 μM (Schurig-Briccio et al. 2014)	70 μM (Schurig-Briccio et al. 2014)	Retinal (Rare), Hepatotoxic (Feinberg et al. 2025)
Triclabendazole	S. agalactiae	0.14 μM (Lencina et al. 2018)	EC50: 15 μM (Lencina et al. 2018)	Rarely shown liver injury, associated with fluke clearing (National Institute of Diabetes and Digestive and Kidney Diseases 2012c)
Zafirlukast	S. agalactiae	0.83 μM (Lencina et al. 2018)	EC50: 6 μM (Lencina et al. 2018)	Hepatotoxic (High dose) (Dhaliwal and Bajaj 2025)
Quinlinyl pyrimidine Compound-1	M. tuberculosis	0.03 μM (Lu L et al. 2022)	MIC 3 μg/mL (Lu L et al. 2022)	HEPG2 TD50 0.2 μM (Lu L et al. 2022) MRC-5 IC50 1.3 μM (Lu L et al. 2022)
Quinlinyl pyrimidine Compound-1	M. smegmatis	2 μM (Lu L et al. 2022)
Quinlinyl pyrimidine Compound-6	M. tuberculosis	1 μM (Lu L et al. 2022)	MIC 13 μg/mL (Lu L et al. 2022)	TD50 17 μM (Lu L et al. 2022)
Compound 13a	M. tuberculosis	0.04 μM (Shirude et al. 2012)	MIC 2 μg/mL (Shirude et al. 2012)
Diphenylene iodonium chloride (DPI)	P. falciparum	16 μM (Biagini et al. 2006) (DB) 25 (CoQ)	EC50 0.24 μM (Biagini et al. 2006)	Non-specific toxic effects. Acute gastrointestinal
Phenyl iodonium chloride (IDP)	P. falciparum	66 μM (Biagini et al. 2006) (DB)	EC50 6 μM (Biagini et al. 2006)	Mitotoxic via ROS production and apoptosis induction (Li N et al. 2003)
HDQ (1-hydroxy-2-alkyl-4(1) quinolone)	T. gondii		EC50 4 nM (Saleh et al. 2007)
HDQ	P. falciparum		EC50 15 nM (Saleh et al. 2007)

In S. aureus, NDH-2 is the sole NADH dehydrogenase of the respiratory chain (Schurig-Briccio et al. 2014) and a suitable drug target (Zhou J-L et al. 2024). Phenothiazines have been found to be effective inhibitors of S. aureus NDH-2 with strong antibiotic properties (Schurig-Briccio et al. 2014). In S. agalactiae, the elimination of NDH-2 does not affect growth in culture media but it attenuates organ colonization in a mouse systemic model (Lencina et al. 2018). FDA-approved drugs, such as zafirlukast (Figure 2D) and triclabendazole have been shown to inhibit NDH-2 activity and bacterial growth in the micromolar range (Lencina et al. 2018) (Table 1). Additionally, quinolinyl pyrimidines have been proposed as alternative options for phenothiazines, which have significant effects on the central nervous system (Shirude et al. 2012). This family of compounds inhibits NDH-2 and bacterial growth in the sub μM range (Shirude et al. 2012; Lu L et al. 2022) (Table 1). Interestingly, these molecules increase NADH oxidation and ROS overproduction, which is both toxic and starves the respiratory chain of redox equivalents, decreasing ATP synthesis (Heikal et al. 2016). In addition to its essential role in bacterial metabolism, NDH-2 also plays a relevant role in eukaryotic parasites like Plasmodium falciparum (Biagini et al. 2006; Saleh et al. 2007) and Toxoplasma gondii (Saleh et al. 2007), in which this enzyme is the only NADH dehydrogenase. Growth of T. gondii and P. falciparum is inhibited by 1-hydroxy-2-dodecyl-4(1H)quinolone (HDQ), which appears to inhibit NDH-2 (Table 1) (Saleh et al. 2007).

NQR

The proton (Raba et al. 2018)/sodium (Juárez et al. 2009) pumping NADH:ubiquinone oxidoreductase (NQR complex) is an essential membrane-associated protein found in numerous bacterial species (Häse and Mekalanos 1998; Stephens et al. 1998; Verkhovsky and Bogachev 2010; Reyes-Prieto et al. 2014; Tuz et al. 2015; Kamath et al. 2016; Liang P et al. 2018; González-Montalvo et al. 2024). NQR plays a critical role in the bacterial respiratory chain, contributing to energy conservation by generating either a proton or sodium motive force across the membrane (Dimroth 1994; Barquera et al. 2002; Raba et al. 2018). The resulting ion gradient supplies energy for essential cellular processes such as ATP synthesis, flagellar rotation, drug extrusion, and nutrient transport (Kojima et al. 1999; Häse and Barquera 2001; Juárez and Barquera 2012; Reyes-Prieto et al. 2014).

NQR is a transmembrane protein complex composed of six subunits, NqrA-F, and a unique collection of cofactors that facilitate the transfer of electrons through the enzyme (Verkhovsky and Bogachev 2010; Juárez and Barquera 2012; Steuber et al. 2015). NQR contains six redox cofactors: two [2Fe-2S] centers and four flavins (Juárez et al. 2009; 2010) (Figure 3). One of the hallmark features of the NQR family is the use of riboflavin as a redox-active molecule (Barquera et al. 2002; Juárez et al. 2008). NQR, and its close relative ferredoxin: NAD⁺ oxidoreductase (RNF) (Vitt et al. 2022), are the only enzymes reported that directly use this molecule as a cofactor (Barquera et al. 2002; Juárez et al. 2008). NQR is widely distributed among pathogenic bacteria and parasites, prominently in Gram-negative bacteria like Vibrio cholerae (Häse and Mekalanos 1998), Pseudomonas aeruginosa (Liang P et al. 2020), and Haemophilus influenzae (Hayashi et al. 1996). In V. cholerae, the NQR complex is crucial not only for energy production but also for the regulation of virulence. It influences the transcription of toxT, a key regulator that activates the expression of cholera toxin (ctxAB), toxin-coregulated pilus (tcpA), and accessory colonization factors (Häse and Mekalanos 1999). Elimination of NQR disrupts this regulatory pathway, impairing the pathogen’s ability to express virulence genes and effectively colonize the host. NQR plays a central role in the aerobic respiratory chain of P. aeruginosa, where it serves as the main NADH dehydrogenase. Over 70% of the oxygen consumption linked to NADH dehydrogenase activity in this organism is sensitive to the NQR inhibitor HQNO, demonstrating the dominant role of this enzyme in energy metabolism (Raba et al. 2018). Additionally, for the obligate intracellular parasite Chlamydia trachomatis, NQR plays a vital role during infection and intracellular multiplication (Liang P et al. 2018), underscoring the enzyme’s evolutionary conservation and essentiality across diverse pathogenic organisms (Reyes-Prieto et al. 2014). Given its crucial metabolic role, NQR serves as an attractive target for developing antimicrobial strategies (Dibrov et al. 2017).

Figure 3.

Structure of NQR accessed from PDB ID: 7XK7 (Berman et al. 2000; Kishikawa et al. 2022). (A) Tri-dimensional structure with cofactors/ligands show as spheres. Dark gray: Korormicin, yellow: 2Fe-2S clusters, teal: FMNB, lavender: FMNC, pink: FAD. (B) Close view of tri-dimensional structure of NQR chain B. Dark gray: Korormicin, light gray: Riboflavin, teal: FMNB, lavender: FMNC. (C) Close view of the Koror.

Recent studies highlight NQR as a critical target in pathogens, exemplified by the promising repurposing of FDA-approved drugs. For instance, the antibiotic clofazimine, a phenothiazine derivative, demonstrates potent inhibitory effects against NQR in V. cholerae. Clofazimine targets the catalytic ubiquinone-binding site, inhibiting bacterial growth with an IC₅₀ of approximately 3 μM and significantly reducing production of the cholera toxin virulence factor. Its effective therapeutic concentration lies within the clinical dosage range, suggesting minimal toxicity risks for human hosts (Table 2). This dual antimicrobial and antivirulence activity makes clofazimine an attractive candidate for treating cholera and potentially other bacterial infections reliant on NQR-mediated metabolism (Yuan et al. 2025).

Table 2.

Reported NQR inhibitors.

Compound	Microorganism	IC50	MIC
PEG-2S	C. trachomatis	2 nM (Dibrov et al. 2017)	0.7 μM (Dibrov et al. 2017)
Korormicin A	V. cholerae	5 nM (Ito et al. 2017)	20 μM (Ito et al. 2017)
Aurachin D-42	V. cholerae	2 nM (Ito et al. 2017)	–
Clofazimine	V. cholerae	3 μM (Yuan et al. 2025)	5 μM (Yuan et al. 2025)
Thioridazine	V. cholerae	27 μM (Yuan et al. 2025)	–

Additionally, quinone-like inhibitors, including korormicin and aurachin, effectively target bacterial NQR (Figure 3). Aurachin D-42 and korormicin A are strong inhibitors, with IC₅₀s of 2 and 5 nM, respectively (Table 2) (Ito T et al. 2017; Masuya et al. 2020). These low IC₅₀ values highlight their potential for antimicrobial drug development. Although korormicin was effective against C. trachomatis, it also had significant cytotoxic effects (Dibrov et al. 2017). The newly developed koromicin-like furanone-based NQR inhibitor PEG-2S effectively prevented C. trachomatis-induced alterations in intracellular H⁺ and Na⁺ concentrations (Table 2). It also demonstrated exceptional potency and specificity by inhibiting NQR activity in V. cholerae membrane vesicles, with an IC₅₀ of 2 nM. Notably, PEG-2S exhibits minimal cytotoxicity in mammalian cells at therapeutically relevant concentrations (Dibrov et al. 2017), consistent with the low host toxicity seen in other NQR-targeting agents, which is an advantage stemming from the absence of NQR homologs in human cells. This high degree of specificity reinforces the promise of these compounds as safe and selective candidates for antimicrobial drug development (Dibrov et al. 2017).

SDH

Succinate dehydrogenase (SDH) is a membrane-bound protein involved in both the Krebs cycle and electron transport chain (McNeil and Fineran 2013). In bacteria, SDH couples the oxidation of succinate to fumarate in the cytoplasm with quinone reduction in the cell membrane (Yankovskaya et al. 2003). Succinate oxidation by SDH is carried out by four cofactors: FAD and three iron-sulfur clusters (2Fe-2S, 4Fe-4S, and 3Fe-4S) (Cheng et al. 2015). Prokaryotic SDH comprises 3–4 subunits (Figure 4A), while the mitochondrial counterpart is composed of four subunits (SdhA-D). Subunit SdhA is a flavoprotein that harbors a covalently-bound FAD factor (Figure 4A,C). Subunit SdhB contains the three iron-sulfur cofactors that mediate the electron transfer to quinone. In prokaryotes, the hydrophobic membrane anchor of the protein, which is least conserved amongst organisms, either consists of one larger SdhC subunit or two smaller SdhC and SdhD subunits, with the quinone binding site positioned at the interface of the hydrophobic membrane subunit(s) (Figure 4A,B) (Hägerhäll 1997). Eukaryotic SDH contains both SdhC and SdhD subunits and two quinone binding sites, a QP site toward the cytosol, and a QD site near the periplasmic side of the membrane (Oyedotun and Lemire 2001). In addition to these catalytically relevant cofactors, SDH also contains a heme b in the membrane domain that is evolutionarily conserved. Remarkably, heme b stabilizes SDH structure but does not play a role in quinone reduction, as E. coli SDH mutants lacking heme b maintain catalytic activity (Tran QM et al. 2007). In addition to the canonical catalytic subunits, subunit SdhE plays a role in FAD attachment (McNeil and Fineran 2013).

Figure 4.

Structure of SDH accessed from PDB ID: 7D6V (Berman et al. 2000; Zhou X et al. 2021). (A) Tri-dimensional structure of SDH with cofactors/ligand as spheres. Light gray: FAD, yellow: Fe-S clusters, dark gray: Ubiquinone-1. (B) Close view of the ubiquinone-1 (UQ1) site. (C) Close view of FAD binding. (D) Two dimensional structures of SDH inhibitors.

SDH is highly conserved and found in most organisms, including mammals, fungi, and pathogenic bacteria. SDH is present in oxygen-consuming bacteria, including strict aerobes and facultative anaerobes (Hederstedt and Rutberg 1981). SDH plays a crucial role in the metabolism of many clinically relevant, antibiotic-resistant bacterial pathogens, including Acinetobacter baumannii, Neisseria spp., Shigella spp., and P. aeruginosa, all of which have been designated by the World Health Organization as priority pathogens due to their multidrug resistance profiles (Kanehisa and Goto 2000; Kanehisa 2019; André et al. 2021; Kanehisa et al. 2023). In addition, SDH is equally critical in the metabolism of numerous fungal plant pathogens, including Botrytis cinerea and Rhinzoctonia solani (Scalliet et al. 2012). Due to its essential metabolic role, SDH has been recognized as a compelling target for antibacterial and fungicide development, offering broad-spectrum efficacy across diverse applications.

Promysalin is a secondary metabolite produced by P. putida, which exhibits antibiotic activity and specifically targets P. aeruginosa SDH by binding to the ubiquinone site, with an in vitro IC₅₀ of 2.5 μM, and an MIC₅₀ of 1.75 μM (Table 3) (Li W et al. 2011; Keohane et al. 2018). Notably, promysalin preferentially targets the pathogenic enzyme by a 50-fold factor compared to its mitochondrial counterpart, demonstrating the compound’s selectivity for the bacterial enzyme (Keohane et al. 2018). Other studies have highlighted the significance of SDH as a drug target against P. aeruginosa, with siccanin, derived from the fungus Helminthosposium siccans, targeting P. aeruginosa SDH with an IC₅₀ of 0.9 μM, thus exhibiting more potency for the bacterial enzyme compared to porcine heart mitochondria (Table 3) (Mogi et al. 2009). Notably, siccanin, a dual SDH and cytochrome bc₁ inhibitor in the malaria-causing agent P. falciparium, is extremely potent against SDH—critical for parasitic survival during the mosquito stage—exhibiting biphasic inhibition with an IC₅₀ of 16 nM and 8.93 μM (Table 3) (Komatsuya et al. 2022). Additionally, in M. tuberculosis, SDH inhibitors have been shown effective against both the enzyme and bacterial growth (Adolph et al. 2024; 2024).

Table 3.

Reported SDH inhibitors.

Compound	Microorganism targeted	IC50	MIC50 or EC50	Toxicity	Inhibition mechanism
Boscalid	Botrytis cinerea	0.8 μM (Bénit et al. 2019)	2.7 μM (Wang et al. 2016)	Hepatotoxic, thyrotoxicosis (Federal Register 2025)	QP (Sang and Lee 2020)
Fluopyram	Botrytis cinerea	0.2 μM (Bénit et al. 2019)	75.6–731 nM (Veloukas and Karaoglanidis 2012)	Hepatotoxic, thyrotoxicosis (Federal Register, 2019)	QP (Sang and Lee 2020)
Bixafen	Rhinzoctonia solani	3 μM (Zhang A et al. 2019)	104 nM (Zhang A et al. 2019)	Hepatotoxic, thyrotoxicosis (solely in the presence of the former) (Federal Register 2018)	QP (Sang and Lee 2020)
Promysalin	P. aeruginosa	2.5 μM (Keohane et al. 2018)	1.8 μM (Li W et al. 2011)		Ubiquinone-binding site
Siccanin	P. aeruginosa	0.9 μM (Mogi et al. 2009)	150 μM (Mogi et al. 2009)	DLD-1 EC50 34 μM HDF EC50 16 μM (Komatsuya et al. 2022)	Ubiquinone-binding site
Siccanin	P. falciparum	16 nM and 8.9 μM (biphasic) (Komatsuya et al. 2022)	8.4 μM (Komatsuya et al. 2022)	DLD-1 EC50 34 μM HDF EC50 16 μM (Komatsuya et al. 2022)	Ubiquinone-binding site

SDH has also been extensively exploited as a molecular target for antifungal agents, primarily in agricultural applications. SDH inhibitors act by binding to the Q_P site of the enzyme, disrupting mitochondrial respiration and impairing fungal viability (Scalliet et al. 2012). These inhibitors encompass diverse drug classes, with prominent groups including carboxamides, benzamines, and pyrazole-carboxamides, such as Boscalid, Fluopyram (Figure 4D), and Bixafen, respectively (Table 3) (Bénit et al. 2019), which are approved by the United States Environmental Protection Agency (EPA) for use as agricultural pesticides. The inhibitory potency of these compounds is species-dependent, with reported IC₅₀ values typically in the low micromolar range (Table 3) (Bénit et al. 2019). Against B. cinerea, Boscalid exhibits an IC₅₀ of 0.8 μM (Bénit et al. 2019), while Fluopyram shows enhanced potency with an IC₅₀ of 0.2 μM (Table 3) (Bénit et al. 2019). Bixafen is considerably more effective, with its IC₅₀ of 0.07 μM (Table 3) (Bénit et al. 2019) highlighting its superior activity against this fungal pathogen (Bénit et al. 2019). However, against the fungal pathogen R. solani, Bixafen is less potent, demonstrating an IC₅₀ of 2.95 μM (Table 3) (Zhang A et al. 2019). Although these molecules were designed to selectively inhibit fungal SDH, off-target activity against the mammalian homolog has been observed. Boscalid, Fluopyram, and Bixafen inhibit human SDH with IC₅₀ values of 4.8 μM, 160 μM, and 0.34 μM, respectively (Bénit et al. 2019). These interactions can impair mitochondrial function in human cells, raising concerns about potential cytotoxic and off-target effects, particularly with high-level exposure. Thus, while these inhibitors are effective and selective antifungals, their use requires careful environmental and toxicological monitoring to mitigate potential risks. Resistance to these molecules has emerged in several important crop pathogens. Amino acid and genetic Q_P site mutations in the Sdh B, C, and D subunits have been identified as major mechanisms of resistance, reducing inhibitor binding affinity in pathogens such as B. cinerea and Alternaria alternata, underscoring the need for resistance management (Leroux et al. 2010; Scalliet et al. 2012).

Cytochrome bc₁

Cytochrome bc₁ (complex III) is a critical enzyme essential for energy generation in aerobic prokaryotes and eukaryotes. Cytochrome bc₁ is an oligomer spanning the plasma membrane of prokaryotes and the inner mitochondrial membrane of eukaryotes. The enzyme is composed of three distinct electron-transferring proteins: cytochrome b, cytochrome c₁, and a Rieske iron-sulfur protein containing an 2Fe-2S cluster. Cytochrome b is a hydrophobic integral membrane protein and houses two noncovalently bound b-type hemes (hemes b_L and b_H), while cytochrome c₁ on the periplasmic side of the membrane contains a single covalently bound c-type heme (heme c₁) (Figure 5) (Trumpower 1990). In prokaryotes, the cytochrome bc₁ complex typically consists of three or four subunits per monomer, whereas in eukaryotes, each monomer can comprise up to eleven subunits (Berry et al. 2000).

Figure 5.

Structure of cytochrome bc1 accessed from PDB ID: 7TCE (Berman et al. 2000). (A) Tri-dimensional structure with cofactors/ligands shown as spheres. Light gray: Atovaquone, beige: Heme c, pink: Heme, yellow: Fe-S clusters. (B) Close view of atovaquone (AOQ) binding site. (C) Two-dimensional structure of bc1 inhibitors.

Cytochrome bc₁ is essential for energy production via the respiratory chain in many pathogenic organisms, including bacteria, fungi, and protozoa. Notably, the enzyme is significant and has been therapeutically targeted in Pneumocystis jirovecii, a fungus which causes life-threatening pneumonia among immunosuppressed patients, P. falciparium, the protozoa responsible for malaria, T. gondii, which causes toxoplasmosis, and Mycobacterium spp., which includes the causative agent of tuberculosis (Kessl et al. 2003; Nixon et al. 2013; Pethe et al. 2013). Cytochrome bc₁ is also a critical enzyme in the opportunistic pathogen P. aeruginosa, responsible for respiratory and systemic infections in immunocompromised individuals (Liang P et al. 2020). Within P. aeruginosa, cytochrome bc₁ has been demonstrated essential for energy production via its significant contribution to respiration and ion pumping during both logarithmic and stationary growth phases (Hu Y et al. 2024). Furthermore, deletion of the operon encoding for cytochrome bc₁ in P. aeruginosa PAO1 severely decreased virulence factor production and pathogenicity (Shen et al. 2021). Despite the fact that some pathogens expressing cytochrome bc₁ possess alternative pathways for ubiquinol oxidation, these routes are less efficient for energy production, as they do not couple electron transfer with proton translocation, thus resulting in less ATP production. Furthermore, the widespread presence of this complex across diverse organisms highlights its evolutionary conservation and functional importance (Trumpower 1990).

Several clinically relevant antiparasitic and antifungal agents exert their effects by targeting cytochrome bc₁. Fungicides such as Azoxystrobin, Ametoctradin, and Cyazofamid, which are approved by the EPA for use on crops, target the enzyme and disrupt respiration (Figure 5C) (Table 4). The action mechanism of these molecules involves inhibition of the quinone binding sites. For instance, Azoxystrobin is a Q_O site inhibitor, while Cyazofamid binds to the Q_i site. Additionally, it has recently been suggested that Ametoctradin is both a Q_O and Q_i site inhibitor (Table 4). However, resistant strains against these pesticides have emerged, with numerous crop pathogens developing mutations in the cytochrome b gene that alter the binding sites (Dreinert et al. 2018).

Table 4.

Reported complex III inhibitors.

Compound	Microorganism targeted	IC50	MIC50	Toxicity	Inhibition mechanism
Azoxystrobin	Pythium spp.	27 nM (Dreinert et al. 2018)	0.25–6 μM (Matić et al. 2019)	Acute oral, dermal, inhalation (Rodrigues et al. 2013)	Qo site
Ametoctradin	Pythium spp.	90 nM		Acute oral, dermal, inhalation (European Food Safety Authority 2012)	Qo/Qi site
Cyazofamid	P. insidiosum	>10 mM (Dreinert et al. 2018)	≤3 μM (Yolanda et al. 2024)	Nephrotoxic (Federal Register 2020)	Qi site
Atovaquone	P. falciparum	3 nM (Biagini et al. 2008)	1 nM (Biagini et al. 2008)	Hepatotoxic (rare) (National Institute of Diabetes and Digestive and Kidney Diseases 2012d)	Qo site
Telacebec	M. tuberculosis	1 nM (Pethe et al. 2013)	3 nM (Pethe et al. 2013)	None observed (Kim et al. 2022)	Qo site
Lansoprazole	M. tuberculosis	0.5 μM ((Kovalova et al. 2024)	2 μM (Rybniker et al. 2015)	Hepatotoxic (rare) (National Institute of Diabetes and Digestive and Kidney Diseases, 2012e)	Qo site
Siccanin	P. falciparum	8 μM (Komatsuya et al. 2022)	8 μM (Komatsuya et al. 2022)	DLD-1, 34 μM HDF 16 μM (Komatsuya et al. 2022)	Qi site

Atovaquone is a naphthoquinone (Figure 5C) approved by the FDA as an antiprotozoal and antifungal agent that is employed against T. gondii and P. jirovecii, and also utilized in combination with proguanil to treat malaria caused by P. falciparum (Atovaquone/Toxoplasmic Encephalitis Study Group 1997; Walker et al. 1998; Baggish and Hill 2002). Atovaquone inhibits the Q_O site, blocking the transfer of electrons from ubiquinol to the Rieske iron-sulfur protein (Figure 5B). Atovaquone is extremely potent against P. falciparum with MIC₅₀ values of 3 nM against the enzyme and 1 nM against bacterial growth (Table 4) (Fry and Pudney 1992; Biagini et al. 2008; Khositnithikul et al. 2008). However, single nucleotide substitutions in the P. falciparum cytochrome b gene following preventive administration of atovaquone have been demonstrated to result in amino acid changes at the drug’s binding site, enabling the parasite to maintain respiratory activity at drug concentrations that would normally limit this function (Korsinczky et al. 2000). In addition, acridinedione compounds have also been shown to inhibit cytochrome bc₁ with higher selectivity compared to atovaquone, also binding the Qo site (Biagini et al. 2008).

Regarding the role of cytochrome bc₁ as an antibacterial target, Telacebec (Q203) is a promising candidate targeting the enzyme in M. tuberculosis, binding to the Q_O site (Table 4) (Pethe et al. 2013) and inhibiting ATP synthesis and bacterial growth (Pethe et al. 2013; Kim J et al. 2022; Capela et al. 2023). Telacebec has been granted Orphan Drug Designation from the FDA and in a Phase 2 clinical trial demonstrated a reduction in M. tuberculosis load in sputum (Janssen et al. 2025). Telacebec exhibits nanomolar-range potency, reducing intracellular ATP levels in M. tuberculosis with an IC₅₀ of 1 nM, while inhibiting bacterial growth in macrophages with an MIC₅₀ of 0.3 nM and in culture medium with an MIC₅₀ of 3 nM (Table 4) (Pethe et al. 2013). Resistance to the drug has been linked to polymorphisms that cause mutations in the target binding site (Sorayah et al. 2019).

Lansoprazole, an FDA approved gastric proton-pump inhibitor which targets H⁺/K⁺-ATPase, has also been demonstrated effective against M. tuberculosis, interfering with ATP synthesis by inhibiting cytochrome bc₁. Preliminary investigations indicate that the drug inhibits growth of intracellular M. tuberculosis within macrophages with an IC₅₀ of 2.2 μM (Table 4) (Rybniker et al. 2015). Intracellular lansoprazole subsequently undergoes a sulfoxide reaction where it is converted to lansoprazole sulfide prior to binding to the enzymatic Q_O site (Rybniker et al. 2015; Kovalova et al. 2024). Lansoprazole’s off-target efficacy against cytochrome bc₁ offers promising potential for drug repurposing, particularly for drugs with established strong safety profiles.

bd oxidase

bd-type oxidases are absent in human mitochondria, but are widely distributed among bacterial and archaeal species (Borisov et al. 2011; Siletsky and Borisov 2021). This enzyme transfers electrons from reduced quinones to oxygen, pumping two H⁺ per electron (Borisov et al. 2011; Siletsky and Borisov 2021). bd oxidases comprise a family of enzymes that carry two heme b cofactors and a heme d (Figure 6A) (Iwata et al. 1995; D’mello et al. 1996; Borisov et al. 2021; Siletsky and Borisov 2021), which are grouped based on their genetic organization into bd-I and bd-II type enzymes and are encoded in the cydABX and appCBXA operons, respectively. In E. coli, these operons are expressed differentially depending on environmental conditions. For instance, bd-I is expressed during microaerophilic growth and in oxygen-rich conditions (Brøndsted and Atlung 1996), and bd-type II is expressed during anaerobic growth, phosphate starvation, and at the beginning of the stationary phase (Brøndsted and Atlung 1996; Grauel et al. 2021).

Figure 6.

Structure of bd oxidase accessed from PDB ID: 5DOQ (Berman et al. 2000; Safarian et al. 2016). (A) Tri-dimensional structure with heme groups presented as spheres. Light gray: Heme b/c, dark gray: Heme d. (B) Two-dimensional structures of bd oxidase inhibitors, 1: N-(4-(4-(trifluoromethyl) phenoxy) phenyl) quinazolin-4-amine, 2: 3- methyl-2-[4-[4-(trifluoromethoxy) phenoxy] phenyl]-2,3,4a,5,6,7,8,8a-octahydro-1H-quinolin-4-one.

bd oxidases use different type of quinones, including ubiquinone and menaquinone (Jünemann et al. 1995; Yang et al. 2008; Borisov et al. 2011). In aerobic bacteria such as K. aerogenes (González-Montalvo et al. 2024), M. tuberculosis (Van Der Velden et al. 2025), and E. coli (Van Beilen and Hellingwerf 2016), bd oxidases use ubiquinol. bd oxidases have higher affinity for oxygen than other terminal oxidases (D’mello et al. 1996; Poole and Cook 2000; Forte et al. 2017), which allows them to function in microaerophilic conditions. In addition, these enzymes are highly resistant to cyanide compared to heme-copper oxidases (Sakamoto et al. 1999; Quesada et al. 2007; Forte et al. 2017; Siletsky and Borisov 2021). Organisms like P. aeruginosa actively produce cyanide to eliminate competing bacteria and switch their metabolism to bd oxidase to survive auto poisoning and colonize new environments (Blumer and Haas 2000; Yan et al. 2019). bd-type enzymes have also been linked to ROS protection. In E. coli, bd oxidase has catalase activity, converting H₂O₂ into O₂ (Borisov et al. 2013). Moreover, bd oxidases detoxify other radical species, like peroxynitrite, produced by macrophages phagosomes (Borisov et al. 2015). bd oxidases have also been proposed as electron sinks during anaerobic growth of E. coli, protecting from ROS production upon sudden exposure to oxygen (Korshunov and Imlay 2010). Finally, bd oxidases have also been demonstrated to provide advantages during exposure to adverse conditions like hypoxia or the presence of H₂S (Forte et al. 2016), CO (Nastasi et al. 2023), and NO (Mason et al. 2009).

Due to their importance in biological process and their absence in mammalian mitochondria, bd oxidases have been an attractive target for drug development. One of the first compounds found to have inhibitory activity against these enzymes was the Aurachin family of compounds (Meunier et al. 1995), which includes aurachin D and C, which preferentially inhibit bd oxidases and bo₃ oxidases, respectively (Figure 6B; Table 5) (Meunier et al. 1995; Kruth and Nett 2023). Due to aurachin D’s significant cytotoxicity (Dejon and Speicher 2013), alternative compounds have been evaluated against the enzyme. The quinazoline ND-011992 was identified to inhibit mycobacterial (Lee BS et al. 2021) and E. coli (Kägi et al. 2023) bd oxidases (Table 5). However, this compound depleted ATP while inhibiting oxygen consumption and growth only in the presence of Bedaquiline (Lee BS et al. 2021), indicating that other terminal oxidases can be used to overcome bd oxidase inhibition. Further studies show that this molecule inhibits mitochondrial and bacterial complex I activities. However, brominated derivatives did not affect NADH oxidase activity of E. coli membranes at micromolar ranges (Kägi et al. 2023).

Table 5.

Reported bd oxidase inhibitors.

Compound	Microorganism targeted	IC50	Bacterial inhibition	Toxicity	Inhibition mechanism
Aurachin D	E. coli	18 nM (Radloff et al. 2021)	>128 μg/mL (MIC) (Dejon and Speicher 2013)	HCT-116 1 μg/mL (GI50) U-2 OS 1 μg/mL (GI50) U937 1 μg/mL (GI50) (Dejon and Speicher 2013)	Blocking of quinone binding site (Miyoshi et al. 1999; Grauel et al. 2021)
ND-01192	E..coli	0.6–1.3 μM (Kägi et al. 2023)		Mt Complex I IC50 3 μM (Kägi et al. 2023)
CK-2–63	M. tuberculosis	3 nM (Jeffreys et al. 2023)	37.8% at 5 μM (Jeffreys et al. 2023)	HepG2 IC50 85 μM 181	Blocking of quinone binding site (Jeffreys et al. 2023)

More recently, quinolone-based compounds, specifically a 2-aryl-quinolone backbone, was found to have inhibitory activity on mycobacterial bd oxidases and on bacterial growth (Jeffreys et al. 2023). Experimental compounds CK-2–63, SL-2–25, MTD-403 and PG-203 had sub micromolar inhibitory concentrations, with CK-2–63 as the most potent compound (Table 5). While this compound showed inhibitory activity against bovine cytochrome bc₁, it showed no toxicity in a HePG2 cell line. This aryl-quinolone synergized with other respiratory chain inhibitors like Bedaquline and ND-011992 to reduce ATP production, oxygen consumption rate, and growth (Jeffreys et al. 2023). The development of drugs against bd oxidases has shown promising results, with some patients already receiving such treatments. Thus, bd oxidases continue to be attractive antibiotic targets since these enzymes are limited to the prokaryotic domains. However, special care is needed to ensure their specificity.

bo₃ oxidase

Cytochrome bo₃ is a multimeric complex composed of subunits I (cyoB), II (cyoA), III (cyoC), and IV (cyoD) (Chepuri et al. 1990; Minghetti et al. 1992; Nakamura et al. 1997; Richhardt et al. 2013; Choi SK et al. 2017). Subunit I possesses three active redox centers: a high-spin heme o₃, a low-spin heme b, and a Cu_B site (Saiki et al. 1992; 1993; Mogi et al. 1994; Choi SK et al. 2017) and contains all residues required for catalysis and proton translocation (Figure 7A) (Garcia-Horsman 1995; Schäfer and Penefsky 2008). This enzyme belongs to the superfamily of heme-copper oxygen reductases, which is characterized by the presence of subunit I. Heme-copper terminal oxidases can be classified into families A, B, and C (Schäfer and Penefsky 2008; Li J et al. 2021). Cytochrome bo₃ oxidase belongs to family A, which is the most extensively distributed and also includes cytochrome c oxidases (Schäfer and Penefsky 2008), such as the mammalian aa₃-type cytochrome c oxidases (Chepuri et al. 1990). This complex oxidizes ubiquinol on the periplasmic side and uses four protons from the cytoplasm to reduce dioxygen while pumping four protons to generate a proton motive force across the membrane (Puustinen et al. 1989; Siletsky and Borisov 2021), with a stoichiometry of 2 H⁺ pumped per e⁻ (Musser and Chan 1998; Siletsky and Borisov 2021). Bo₃ oxidase is expressed preferentially under high oxygen concentrations and is under the control of regulators Fnr and ArcAB (Rice and Hempfling 1978; Cotter et al. 1990; Tseng et al. 1996).

Figure 7.

Structure of cytochrome bo3 accessed from PDB ID: 7CUW (Berman et al. 2000; Li J et al. 2021). (A) Tri-dimensional structure with cofactors/ligands shown as spheres, light gray: Ubiquinone-8, dark gray: Heme, slate blue: Cu-heme. (B) Close view of ubiquinone-8 (UQ8) binding. (C) Two-dimensional structures of bo3 inhibitors.

Studies on bo₃ oxidase have been mostly focused on elucidating its functions and role in the respiratory chain. This enzyme is widely distributed among bacterial species, and has been identified as an important component of respiration in E. coli (Cheesman et al. 1993), P. aeruginosa (Liang P et al. 2020), Gluconobacter oxydans (Richhardt et al. 2013), K. aerogenes (González-Montalvo et al. 2024), K. pneumoniae (Juty and Hill 1997), Proteus mirabilis (Armbruster et al. 2018), Asaia bogorensis (Kawai et al. 2015), V. alginolyticus (Miyoshi-Akiyama et al. 1993), and Salmonella (Jones-Carson et al. 2016). Although widely distributed, there are only a few studies regarding its metabolic role. In E. coli, bo₃ is the favored terminal oxidase in high aeration conditions (Puustinen et al. 1989; Tseng et al. 1996; Weiss et al. 2010; Borisov et al. 2011). This same pattern has been observed in other bacteria, including G. oxydans (Richhardt et al. 2013) and P. aeruginosa (Liang P et al. 2020). On the other hand, in the opportunistic pathogen K. aerogenes, bo₃ is not the main terminal oxidase (González-Montalvo et al. 2024). Regarding its role in pathogenicity, it has been shown that bo₃ expression is increased in the pathogen P. mirabilis in an urinary tract infection model in mice (Pearson et al. 2011; Armbruster et al. 2018). Moreover, this enzyme is critical during catheter-associated urinary tract infections (Armbruster et al. 2017, 2018).

Regarding drug development against this enzyme, there are major gaps. Previous studies indicate that aurachin C analogs almost completely inhibited bo₃ activity. Inhibitor screening showed that aurachin C derivates also showed strong inhibition but were unspecific, as bd-oxidases were also inhibited by these compounds (Table 6) (Radloff et al. 2021). Other studies showed that aurachin D derivatives could be used to develop new bo₃ inhibitors, specifically those that contain double bonds in the R1 position, with strong dependence on the length of side chains in position R2 (Radloff et al. 2021).

Table 6.

Reported bo₃ oxidase inhibitors.

Compound	Microorganism targeted	IC50	MIC	Toxicity	Inhibition mechanism
Aurachin C	E. coli		>50 μg/mL (MIC) (Kunze et al. 1987)	50 nM (IC50) (Mitochondria) (Kunze et al. 1987)	Blocking of quinone binding site (Miyoshi et al. 1999; Grauel et al. 2021)
CORM-3	E. coli		4 to >512 μg/mL (MIC) (Southam et al. 2018)	>100 μM for RAW264.7 (Song et al. 2009)	Binding of CO to heme o3 (Jesse et al. 2013; Nastasi et al. 2023)
AC1–10	E. coli	10–80 nM (Radloff et al. 2021)			Block quinone binding site (Miyoshi et al. 1999; Grauel et al. 2021)
AC4–11	E. coli	0.1–0.14 μM (Radloff et al. 2021)			Block quinone binding site (Miyoshi et al. 1999; Grauel et al. 2021)
AC2–11	E. coli	30 nM (Radloff et al. 2021)			Block quinone binding site (Miyoshi et al. 1999; Grauel et al. 2021)

AOX

One key difference in the respiratory chain of mammalian mitochondria and many human pathogens is the presence of a cyanide-insensitive alternative oxidase (AOX) (Figure 8). Although AOX was first identified in plants in the 1920s (Schmucker 1925; Moore and Siedow 1991) and later described in Neurospora in the 1950s (Tissieres et al. 1953), its presence in bacteria was not determined until bacterial genomic sequences became available in the early 2000s. The eukaryotic AOX is encoded by the nuclear genome and is a homodimer (32–36 kDa per monomer) associated with the matrix leaf of the inner mitochondrial membrane as a monotopic membrane protein (Elthon and McIntosh 1986; 1987; May et al. 2017). AOX is a terminal diiron quinol oxidase with its active site oriented to the matrix, catalyzing the cyanide and antimycin A-resistant oxidation of ubiquinol and the reduction of oxygen to water (Berthold and Stenmark 2003; Moore et al. 2013; Shiba et al. 2013). This enzyme branches the electron flux of the respiratory chain prior to proton translocation by complexes III and IV (Medentsev et al. 2002; Kern et al. 2007). As AOX is not a proton pump, its activity is not directly linked to the generation of the proton motive force (Berthold et al. 2000). Instead, the energy liberated during ubiquinol oxidation is released as heat (Albury et al. 2009; Vanlerberghe et al. 2009).

Figure 8.

Structure of alternative oxidase accessed form PDB ID: 5ZDR (Berman et al. 2000; Shiba et al. 2019). (A) Tri-dimensional structure with small molecules presented as spheres. (B) Close view of the binding site, FE: Iron, CHW: 3-chloro-4,6-dihydroxy-5-[(2E,6E,8S)-8-hydroxy-3,7-dimethylnona-2,6-dien-1-yl]-2-methylbenzaldehyde.

AOX is ubiquitous in plants (Sharpless and Butow 1970; Dinant et al. 2001; Gérin et al. 2010; Mathy et al. 2010) and many unicellular eukaryotes (Juárez et al. 2004; Sierra-Campos et al. 2009) including pathogenic organisms such as Candida parapsilosis (Guérin and Camougrand 1986; Milani et al. 2001; Ruy et al. 2006), C. albicans (Huh and Kang 2001), P. falciparum (Murphy and Lang-Unnasch 1999), Philasterides dicentrarchi (Mallo et al. 2013), Trypanosoma congolense, T. evansi (Barrett et al. 2003; Roberts CW et al. 2004; Suzuki et al. 2005; Williams et al. 2010), T. brucei (Roberts CW et al. 2004), Cryptococcus neoformans (Akhter et al. 2003), Cryptosporidium parvum, Blastocystis hominis (Moore and Albury 2008; Young et al. 2013; May et al. 2017), and other microsporidia (Williams et al. 2010). Depending on the organism, AOX function can be vastly different, from its thermogenic activity in plants (Albury et al. 2009; Ito K et al. 2011), to the regulation of ubiquinol pools and reduction of ROS production. It is commonly accepted that AOX promotes the metabolic plasticity of cells for rapid adaptation to nutrient sources or to biotic and abiotic stress factors (Albury et al. 2009; Rasmusson et al. 2009; Chai et al. 2010; Hanqing et al. 2010). Moreover, AOX seems to be involved in pathogenicity and virulence, survival of parasites, and regulation of host–pathogen interactions (Black et al. 2021).

Recently, efforts have been made to understand the role of AOX in pathogenic metabolisms, allowing for the discovery of specific AOX inhibitors that could serve as novel antifungals and antiparasitics that do not affect human mitochondrial function. For example, nitroprusside (NO donor) and the AOX inhibitor salicylhydroxamic acid (SHAM) reduce C. albicans viability and increase phagocytic uptake by macrophages (Table 7) (Duvenage et al. 2019). In addition to SHAM, other fungal AOX inhibitors have been identified, including 2,5-dibromo-3-methyl-6-isopropyl-p-benzo quinone, disulfiram, 5-decyl-6-hydroxy-4,7-dioxobenzothiazole, propyl gallate, and various hydroxamic acids (Table 7) (Veiga et al. 2003). Recent findings indicate that P. falciparum AOX can be inhibited by propyl gallate, 8-hydroxyquinoline, and SHAM (Murphy et al. 1997; Murphy and Lang-Unnasch 1999). Moreover, T. brucei AOX, which is essential for the respiratory activity of the bloodstream form of the parasite (Clarkson et al. 1989; Chaudhuri et al. 2006), can be inhibited by ascofuranone (Kido et al. 2010; Menzies et al. 2018).

Table 7.

Reported AOX inhibitors.

Compound	Microorganism targeted	IC50
Ascochlorin	T. brucei brucei	3.7 nM (Young et al. 2020; Xu et al. 2021)
Ascofuranone	T. brucei brucei	0.03 nM (Shiba et al. 2013)
Octyl gallate	T. brucei brucei	0.23 μM (Young et al. 2020)
Salicylhydroxamic acid (SHAM)	T. brucei brucei	6 μM (Meco-Navas et al. 2018)
Salicylic hydroxamic acid	T. brucei brucei	4 nM (Young et al. 2020)
8-Hydroxyquinoline	P. falciparum	11 μM (Roberts et al. 2004)
8-Hydroxyquinoline	C. parvum	0.7 μM (Roberts et al. 2004)

ATP synthase

ATP synthases convert the transmembrane electrochemical gradient potential into chemical bond energy in the form of adenosine triphosphate (ATP) (Junge and Nelson 2015). Although ATP synthases are ubiquitously distributed in all life domains, they appear to play different roles. ATP synthases are multicomponent membrane enzymes that are highly conserved across species in the phylogenetic tree (von Ballmoos et al. 2009; Walker 2013; Lu P et al. 2014). ATP synthases have two major functional domains, a membrane motor domain (F₀) and a cytoplasmic catalytic domain (F₁), which are joined together by the central γ and ε-subunits of F₁ and the peripheral stalk of F₀ (b-subunit). The F₀ domain is composed of a, b and c subunits. C-subunits form a ring embedded in the membrane with rotor function. The F₁ domain is composed of α, β, γ, δ, and ε subunits. F₀ is a motor that rotates, driven by the controlled dissipation of the transmembrane electrochemical gradient (Figure 9A) (Kühlbrandt 2019). Protons flow through a- and c-subunits, which allows the rotation of the γ-subunit, ε-subunits, and c-subunits (Yoshida et al. 2001; Nesci et al. 2016). The rotational energy is transmitted to the catalytic domain in F₁ through the γ-subunit, which is then used to synthesize ATP from adenosine diphosphate and inorganic phosphate (Figure 9B,C) (Okuno et al. 2011).

Figure 9.

Structure of ATP synthase accessed from PDB ID: 8J0S (Berman et al. 2000; Zhang Y et al. 2024). (A) Tri-dimensional structure with inhibitors/cofactors visible as spheres. (B–D) Close view of the binding of ADP, ATP, and bedaquiline (BQ1). (E) Two-dimensional structures of ATP synthase inhibitors, upper: Venturicidin A, lower: Bedaquiline.

ATP synthase has been validated as a druggable target against M. tuberculosis. The diarylquinoline bedaquiline (formerly TMC207) (Figure 9D,E), which targets ATP synthase, has potent inhibitory activity against Mycobacterium spp. (Biagini et al. 2008; Mahajan 2013), and it became the first drug approved for tuberculosis treatment with a novel mechanism of action in 40 years (Mahajan 2013). Regimens that included bedaquiline reduced the risk of death of patients with MDR or rifampicin-resistant tuberculosis (Schnippel et al. 2018). Moreover, bedaquiline has strong activity on the growth of a large number of mycobacteria species (Andries et al. 2005). Although the activity of bedaquiline is restricted to mycobacteria, with no activity on Gram-positive of Gram-negative bacteria (Andries et al. 2005), diarylquinolines with novel lateral chains show activity against the ATP synthase and growth of S. aureus (Balemans et al. 2012). Bedaquiline targets the c-ring and the interface of aand c-subunits with high affinity in the lower nanomolar range (Table 8) (Preiss et al. 2015; Guo et al. 2021). Several reports indicate that bedaquiline has little inhibitory activity on human ATP synthase (Haagsma et al. 2009; Balemans et al. 2012). However, a recent study suggests that it may have some toxicity (Luo et al. 2020). There are reported cardiac and liver adverse events, namely QT prolongation and elevated liver enzymes (Table 8) (Kim JH et al. 2023). However, it has been determined that the cardiac effect is due to its metabolites, which inhibit the cardiac delayed rectifier potassium channels (Kakkar and Dahiya 2014). Venturicidin A is an ATP synthase inhibitor that binds to subunit c of fungal ATP synthase, blocking proton translocation and inhibiting ATP synthesis (Figure 9E) (Galanis et al. 1989). Although it has also been shown to deplete ATP in bacteria (Yarlagadda et al. 2020), it has no antibacterial activity on its own. However, it potentiates aminoglycosides against resistant clinical isolates of P. aeruginosa, A. baumannii, methicillin-resistant S. aureus, and vancomycin-resistant enterococci (Yarlagadda et al. 2020), serving as an antibiotic adjuvant. Therefore, ATP synthase inhibitors can prolong the clinical life of certain classes of antibiotics. Finally, in facultative anaerobes, ATP synthase is not essential for growth in aerobic conditions in Gram-negative and Gram-positive bacteria, with the exception of P. aeruginosa. In this bacteria, ATP synthase mutants are avirulent in infection models in vivo (Jensen and Michelsen 1992; Turner et al. 2003; PLOS Pathogens 2025; Skurni et al. 2013; Fuller et al. 2000; Sheehan et al. 2003; Fey et al. 2013; Lee E-J et al. 2013; Koo et al. 2017; Poulsen et al. 2019). Therefore, this is an excellent target for drug development in Pseudomonas.

Table 8.

Reported ATP synthase inhibitors.

Compound	Microorganism targeted	IC50	MIC50	Inhibition mechanism	Toxicity
Bedaquiline	M. tuberculosis	20–25 nM (Preiss et al. 2015)	54 nM (Matteelli et al. 2010) (drug-resistant strain)	Blocks proton flow	Hepatotoxicity

Quinone synthesis

Ubiquinone (UQ), also known as coenzyme Q, is a critical molecule in cellular energy production and antioxidant defense. It acts as an electron carrier in the mitochondrial and bacterial electron transport chains, enabling ATP generation, and protection from reactive oxygen species. Ubiquinone biosynthesis in humans and microorganisms shares core pathways but differs in some aspects, including isoprenoid chain length and aromatic ring precursor origin. Humans synthesize UQ₁₀, while bacteria produce UQ₈, and fungi produce UQ₆ (Meganathan 2001). Humans derive 4-hydroxybenzoate (4-HB) from dietary aromatic amino acids, whereas microorganisms synthesize 4-HB via the shikimate pathway (Figure 10). This pathway is also found in plants and apicomplexan parasites (T. gondii, P. falciparum), but is absent in humans (Meganathan 2001), making it a promising antimicrobial target. The shikimate pathway converts phosphoenolpyruvate (PEP) and erythrose-4-phosphate (E4P) into chorismate through seven enzymatic steps. The first step, catalyzed by DAHP synthase (Figure 11B), forms 3-deoxy-D-arabino-heptulosonate-7-phosphate (DAHP) from PEP and E4P (Figures 11B and 12B) (Shende et al. 2024). Inhibiting DAHP synthase could halt the pathway, depriving microbes of aromatic compounds, though redundancy (e.g. three DAHP synthase isoenzymes in E. coli) poses challenges. DAHP oxime (Figure 12E), a product analog, is a potent inhibitor of the aroG encoded E. coli DAHP synthase (Balachandran et al. 2016). Another key target, EPSP synthase (Figure 11E), forms 5-enolpyruvylshikimate-3-phosphate (EPSP) from shikimate-3-phosphate and PEP. Targeted by glyphosate (Pöppe et al. 2019), EPSP synthase is essential for E. coli in minimal medium (Joyce et al. 2006). In bacterial ubiquinone biosynthesis, the enzymes (UbiA through UbiX in E. coli) synthesize and modify the quinone head and isoprenoid tail (Figure 11A,C,D) (Aussel et al. 2014). While some enzymes have human homologs, complicating inhibitor design, others, like UbiC (chorismate pyruvate-lyase), which converts chorismate to 4-HB, lack human equivalents (Figures 11C and 12C) (Mitchell 2011). Humans produce 4-HB from tyrosine or phenylalanine via a distinct pathway. UbiC, a small monomeric enzyme with a unique chorismate-binding site, is a prime target for antibiotic design. Synthetic chorismate analogs could block bacterial 4-HB production without affecting humans, as can 4-HB analogs like vanillate inhibit UbiC (Figure 12D) (Holden et al. 2002). Similarly, UbiD, which decarboxylates 3-polyprenyl-4-hydroxybenzoate to 2-polyprenylphenol, and UbiX, which produces the prenylated Flavin mononucleotide (prFMN) cofactor (Figure 12A) for UbiD, have no human homologs (Aussel et al. 2014). In humans, UQ₂ bypasses this decarboxylation, yielding hydroxylated intermediates directly. Inhibitors targeting UbiX (e.g. prFMN synthesis blockers) or UbiD active sites could be suitable inhibitors, for example substrate analogs like vanillate, ferulic and cinnamic acid inhibit P. aeruginosa UbiD (Jacewicz et al. 2013) (Table 9).

Figure 10.

Ubiquinone biosynthesis pathway.

Figure 11.

Tridimensional structure of ubiquinone biosynthesis pathway enzymes. (A) UbiD accessed from PDB ID: 2IDB (Berman et al. 2000). (B) DAHP synthase (ribbons) accessed from PDB ID: 1RZM in complex with erythrose-4-phosphate, phospho-enol-pyruvate, and cadmium (spheres) (Berman et al. 2000; Shumilin et al. 2004). (C) UbiC (ribbons) accessed from PDB ID: 1G1B in complex with Para-hydroxy benzoic acid (spheres) (Berman et al. 2000; Gallagher et al. 2001). (D) UbiX (ribbons) accessed from PDB ID: 6QLK in complex with prenylated flavin-mononucleotide (spheres) (Berman et al. 2000; Marshall et al. 2019). (E) EPSP synthase accessed from PDB ID: 5BUF (Berman et al. 2000; Sutton et al. 2016).

Figure 12.

Close view of ubiquinone synthesis enzyme binding pockets. (A) UbiX in complex with prenylated flavin-mononucleotid1971e (4 LU) (Berman 2000; Marshall et al. 2019). (B) DAHP in complex with erythrose-4-phosphate (E4P), phospho-enolpyruvate (PEP), and cadmium (CD) (Berman 2000; Shumilin et al. 2004). (C) UbiC in complex with Para-hydroxy benzoic acid (PHB) (Berman et al. 2000; Gallagher et al. 2001). (D) UbiC in complex with inhibitor vanillate (VNL) accessed from PDB ID 1XLR (Berman et al. 2000; Smith et al. 2006). (E) Two dimensional structure of DAHP inhibitor, DAHP oxime.

Table 9.

Reported ubiquinone biosynthesis inhibitors.

Compound	Enzyme target	Microorganism targeted	IC50	MIC50	Inhibition mechanism
DAHP oxime	DAHP synthase	E.coli		Ki = 1.5 ± 0.4 μM	Product analogue, binds the active site competitive inhibition (Balachandran et al. 2016)
Glyphosate	EPSP synthase	Broad-spectrum herbicide. Bacteria having shikimate pathway	Not cytotoxic up to 1 mM for Human mononuclear white blood cells (Nagy et al. 2019)	MIC 100 10–80 mg/mL Salmonella spp.	Competitive for PEP (Poppe et al. 2019)
Vanillate	chorismate pyruvate-lyase (UbiC)	E. coli	Not cytotoxic up to 100 μM in human monocytes (Zhao D et al. 2019)	Ki 280 ± 50 uM	Likely competitive inhibitor (Holden et al. 2002)
Vanillate	UbiD	P. aeruginosa	Not cytotoxic up to 100 μM in human monocytes (Zhao D et al. 2019)		Likely competitive inhibitor (Jacewicz et al. 2013)
Ferulic acid	UbiD	P. aeruginosa	Nontoxic, considered GRAS (Ou and Kwok 2004)		Likely competitive inhibitor (Jacewicz et al. 2013)
Cinnamic acid	UbiD	P. aeruginosa	Nontoxic, considered GRAS (Eilerman 2000)		Likely competitive inhibitor (Jacewicz et al. 2013)

Humans and some bacteria (E. coli) use the mevalonate pathway to produce the ubiquinone isoprenoid tail, forming isopentenyl pyrophosphate (IPP) from acetyl-CoA. Others, like M. tuberculosis, use the non-mevalonate (MEP/DOXP) pathway (Meganathan 2001). MEP inhibitors like fosmidomycin target enzymes in bacteria and protozoa (Plasmodium), disrupting isoprenoid and ubiquinone synthesis without affecting human cells (Knak et al. 2022). Additionally, bacteria produce shorter polyisoprenoid chains, and enzymes controlling tail length (e.g. bacterial IspB vs. human UQ₁) could be targeted to impair microbial energy metabolism selectively.

Antimicrobial peptides

In addition to traditional drug-like compounds, current research is showing that peptides can be used as antibiotics by disrupting the pathogen’s metabolism. Structurally diverse classes of antimicrobial peptides (AMPs) such as Defensins (Doss et al. 2010), Daptomycin (Alborn et al. 1991; Tally and DeBruin 2000; Nguyen et al. 2010; Tran TT et al. 2015), Lanthipeptides (Breukink and De Kruijff 1999; Dickman et al. 2019; Zhao X and Kuipers 2021), Gramicidin (Hu C et al. 2021), Polymyxins (Slingerland et al. 2022), Cathelicidins (Henzler Wildman et al. 2003; Ridyard and Overhage 2021), Histatin-5 (Puri and Edgerton 2014), Magainins (Kim M et al. 2018; Rashid et al. 2020), and Cecropins (Pillai et al. 2005) interrupt the respiratory chains of pathogens by targeting key membrane components (Table 10). These molecules disrupt membrane integrity, ion gradients, or the proton motive force, which are essential for maintaining cellular respiration and energy production. For instance, the host-produced peptide cathelicidin permeabilizes and depolarizes membranes (Henzler Wildman et al. 2003), which halts ATP synthesis. The antifungal Histatin-5 binds to surface proteins in C. albicans, accumulates in mitochondria, and disrupts the electrochemical gradient (Puri and Edgerton 2014). Another antifungal, HsAFP1, isolated from Hemerocallis pigmenti seeds, binds to fungal cell walls, penetrates the cytoplasm, targets mitochondria, and inhibits the respiratory chain, leading to cell death (Aerts et al. 2011). Magainins also affect fungal membranes by forming transient pores that compromise mitochondrial function (Rashid et al. 2020). Cecropins interact with parasite membrane lipids, compromising integrity and potentially impairing mitochondrial function (Pillai et al. 2005).

Table 10.

Antibacterial, antifungal and antiparasitic peptides targeting respiratory enzymes.

Type	Target	Compound	Target	Spectrum (microorganisms affected)	Toxicity	Inhibition mechanism
AMPs	Antibacterial	Defensins	Unspecific, target the membrane.	Broad	In vitro cytolytic effects on human lymphocytes, endothelial cells, and murine lymphoid cells.	Pore formation in the membrane (Aerts et al. 2011; Doss et al. 2010; Lehrer et al. 1993).
		Gramicidins	Unspecific, target the membrane.	Broad	Hemolytic	Increases membrane permeability to monovalent cations (Hu et al. 2021).
		Polymyxins	Lipid A	Gram negative	Nephrotoxicity, neurotoxicity, and allergic reactions	Binds to the negatively charged lipid A component of LPS, displacing divalent cations and disorganizing the membrane (Slingerland et al. 2022).
		Lanthipeptides (Nisin)	• Membrane • Lipid II	Gram positive	None	Creates membrane pores and disrupts cell wall formation (Li et al. 2023; Liang et al. 2023; Zhao and Kuipers 2021).
		Bacteriocins (e.g. Acidocin J1132β and analogues)	Unspecific, target the membrane.	Narrow, non-typhoidal Salmonella	None
		Cathionic peptides (ascaphin-8, XT-7, citropin 1.1, aurein, and dermaseptin)	• ATP synthase • peripheral membrane proteins displacement • phosphatidylglycerol (PG)	Broad, including antifungal and antiparasitic	None or very low	Inhibit ATP synthase activity (Complex V). Remove cytochrome c and MurG from respiratory chain (Wenzel et al. 2014).
		Daptomycin	Phosphatidylglycerol (PG)	Gram positive	Myopathy and rhabdomyolysis	Cell membrane depolarization (Alborn et al. 1991; Nguyen et al. 2010; Tally and DeBruin 2000; Tran et al. 2015)
		Cathelicidins (LL-37)	Unspecific, target the membrane	Broad	Hemolytic	Pore formation (Choi et al. 2022)
	Antifungal	Histatin-5 and derivatives	A direct mitochondrial target was hypothesized, but the mechanism has not been elucidated.	Broad, including antibacterial	None	Pore formation and mitochondrial disruption (Puri and Edgerton 2014).
		Magainins	The mechanism has not been elucidated.	Broad, including antifungal and antiparasitic	None	Permeabilize the inner mitochondrial membrane through pore formation, leading to uncoupling of electron transfer and ATP production (Kim et al. 2018; Rashid et al. 2020).
	Antiparasitic	Cecropins, melittin	Phosphatidylglycerol (PG)	Broad	None	Cell membrane depolarization (Pillai et al. 2005).
		Synthetic/Modified Peptides (RP-1, AA-RP-1, (Fxr)3)	A direct mitochondrial target was hypothesized, but the mechanism has not been elucidated.	Unknown	Low	Cell membrane depolarization.

Cationic antimicrobial peptides (CAMPs) like ascaphin-8, XT-7, citropin 1.1, aurein, and dermaseptin have been reported to inhibit E. coli’s ATP synthase (Ahmad Z et al. 2013), likely through membrane depolarization. Another proposed mechanism for small cationic AMPs involves the displacement of peripheral membrane proteins, such as cytochrome c and the lipid II biosynthesis protein MurG, from the membrane (Wenzel et al. 2014). This displacement disrupts the respiratory chain, damaging the electron transfer chain and breaking down the proton gradient. Consequently, ATP synthesis is impaired, leading to altered energy metabolism and reduced energy production for all other macromolecule biosynthesis, including that of the cell wall.

Among the dozens of respiratory chain-affecting AMPs discovered so far, some are of particular interest, especially those that are now in clinical use or preclinical development. Daptomycin, a broad range AMP, binds to bacterial membranes enriched in phosphatidylglycerol, causing pore formation and collapsing the proton motive force essential for respiratory complexes (Rashid et al. 2020). Many AMPs that exhibit membrane-disrupting properties, such as nisin, LL-37, melittin and cecropin, also efficiently bind to teichoic acids, disrupting cell wall biosynthesis and membrane potential of most Gram-positive bacteria (Moreno-Vargas and Prada-Gracia 2024). Gramicidin forms dimeric channels in the lipid bilayer, leading to unregulated cation flux and dissipation of the ion gradient. Polymyxins bind to lipopolysaccharides and inner membrane phospholipids, disrupting membrane integrity and affecting the electron transport chain.

By leveraging these mechanisms, AMPs can effectively impair respiratory function and overall viability of the target cells, making them powerful tools in the fight against resistant pathogens. While most AMPs, such as defensins and cathelicidins, exhibit a broad spectrum of activity against a wide range of pathogens (Li L et al. 2023; Liang H et al. 2023), others, including bacteriocins (microcins) (Telhig et al. 2022; Kim S-Y et al. 2024) and certain lanthipeptides (Lehrer et al. 1993; Choi M et al. 2022), demonstrate a narrow spectrum of activity, targeting specific microorganisms. These variations are closely linked to the distinct molecular interactions and mechanisms of action inherent to each molecule. Notably, molecules that possess multiple mechanisms of action are of particular interest due to their potential to offer both a narrow spectrum of activity and a reduced likelihood of resistance development.

As is evident from the examples presented, most naturally occurring antimicrobial peptides indirectly affect the respiratory chain because the peptides target key components that are essential for maintaining the proton motive force or membrane integrity, and only in a few engineered examples is there evidence for binding to specific proteins (such as cationic AMPs inhibiting ATP synthase). Continued high-resolution structural and biochemical studies are needed to further delineate precise protein targets within pathogen respiratory chains. Although antimicrobial peptides represent a new alternative to target the respiratory activity of pathogens, their clinical implementation faces challenges due to their cytotoxicity, susceptibility to protease degradation, poor pharmacokinetics, and the potential for resistance development (Table 10). Effective delivery systems are needed to target AMPs precisely to infection sites. Despite these challenges, successful AMP therapies like Polymyxins, Daptomycin, LL-37, and Nisin are in use or under development. These therapies demonstrate the clinical potential of AMPs in combating resistant pathogens.

Conclusions

Antimicrobial resistance is a growing problem in bacterial, fungal, and parasitic disease treatment. Annually, millions of deaths are linked to antimicrobial resistance, and the annual economic impact is substantial. Due to these figures, maintaining the status quo regarding microbial disease treatment options is not possible, and would only exacerbate this public health crisis. The respiratory chain complexes and ATP synthase physiology are just as vital to cellular life as other mechanisms that have been therapeutically targeted (cell wall, ribosome, enzymatic pathways, etc.) with success and high chemotherapeutic index. Using aerobic respiratory chain inhibition as a sole disease intervention may be feasible, or, using respiratory chain inhibition synergistically with traditional antimicrobial drugs may increase their efficacy and diminish resistance levels of these traditional drugs. It is important to note that the respiratory chain of pathogens may not all share the same components, and the effect of drugs may vary, which allows for the creation of selective compounds that may have no effect on beneficial bacteria and no impact on the mammalian host.