Decoding Urease Inhibition: A Comprehensive Review of Inhibitor Scaffolds

Abstract

Urease is a metalloenzyme produced by a wide range of organisms and plays a critical role in nitrogen microbial metabolism by catalyzing the hydrolysis of urea into ammonia and carbamic acid. High urease activity can lead to excessive ammonia levels, causing nitrogen loss from the soil, and also contributes as a virulent factor in some human pathogenic infections. Given its impact, urease inhibition has garnered significant attention in agriculture, environmental sciences, and medicine. Despite efforts to develop potent and selective inhibitors, discovering new agrochemicals and drugs faces significant challenges due to chemical and metabolic stability, lack of selectivity and toxicity, typically seen in urease inhibitors. As a result, extensive chemical diversity has been reported for urease inhibition and can lay the foundation for designing new inhibitors. Previous structure–activity relationship analyses have mainly focused on small subsets of compounds, leaving a broader comprehensive understanding across all known classes yet to be described. In an effort to support new exploratory strategies, this review provides a complete overview of the chemical landscape in urease inhibitors, covering more than 8000 compounds. We discuss current challenges, the biological and practical implications of urease inhibition and highlight potential scaffolds associated with activity. By exploring the diversity of reported inhibitors, we aim to identify broad activity patterns and gain deeper insights into activity relationships that govern inhibitory efficacy, paving the way for the future directions of finding new classes of urease inhibitors.

Representative functional groups known for urease inhibition are reviewed within diverse scaffolds using structure–activity analyses to identify features associated with high inhibitory activity.

1. Brief History of Urease

Urease has historically been a pivotal enzyme that paved the way for numerous advancements in scientific discovery in topics of enzymatic structure and function. This enzyme was the first to be crystallized and solved by X‐ray crystallography, and it was also the first enzyme to be shown to require nickel ions as essential cofactors for its catalytic activity.

The first ureolytic microorganism, Micrococcus ureae, was isolated in 1864, and years later an isolated fraction referred to as a “soluble ferment” was shown to be able to produce volatile ammonia in putrid urine.^{[ 1 ]} It was subsequently observed that many organisms demonstrated this ability and in 1890 Pierre Miquel coined the term urease, refering to the substance responsible for this activity.^{[ 2 ]} However, it was only much later that, in 1926, James B. Summer isolated crystals of urease while working with seeds from Canavalia ensiformis and showed for the first time that proteins could function as enzymes which ultimately led to Summer receiving the Nobel Prize in Chemistry.^{[ 1 , 3 ]} In 1975, it was demonstrated that urease required two nickel atoms for its catalytic functions^{[ 1 , 2 ]} and then another 20 years, for the first X‐ray crystal of urease isolated from K. aerogenes to be obtained recombinantly from E. coli.^{[ 4 ]} Since then, multiple other ureases have been described but the crystal structure of jack bean urease (JBU) was only reported in 2010, 85 years after being first crystallized by James Summer.

A bacterium deeply connected to urease is Helicobacter pylori (originally termed Campylobacter pylori), first identified in 1984 by Marshall and Warren^{[ 5 ]} as the agent responsible for gastritis and peptic ulcers. Urease is essential for H. pylori to survive the acidic environment of the stomach and in 1994 the World Health Organization classified it as the first class‐I carcinogen bacterium. Also related to urease, its substrate urea was the first organic compound to be synthesized in laboratory from an inorganic starting material in 1828.^{[ 2 ]}

2. The Importance of Urease

The seemingly simple catalytic breakdown of urea plays a vital role in our society as it is critical in nitrogen recycling, metabolism and chemotactic communication across ecosystems.^{[ 6 , 7 ]} Fixation of nitrogen using urea‐based fertilizers thus helps enhance soil fertility and health as well as improve crop growth.^{[ 1 , 8 ]} However, excessive urease activity can lead to rapid nitrogen loss via ammonia volatilization, damaging soil health (leaching and pH changes) and contributing to particulate pollution.^{[ 9 , 10 ]} Thus, proper management of nitrogen by modulation of urease activity is seen as a way to efficiently control the nitrogen cycle^{[ 11 ]} and nutrient uptake.^{[ 12 ]} Beyond agricultural applications, urease is also important in bacterial infections. In fact, the WHO's “priority pathogen” list includes multiple urease‐dependent bacteria,^{[ 13 , 14 ]} among which Helicobacter pylori and bacteria that cause persistent urinary tract infections use urease as a crucial feature of pathogenesis.

The reliance of H. pylori on urease to survive in the acidic environment of the stomach is so critical that it expresses the enzyme in significantly higher levels than any other known ureolytic bacteria (≈10% of its total protein content)^{[ 15 , 16 ]} and urease appears to be readily available when the organism needs to survive without the need for new protein synthesis.^{[ 17 ]} Additionally, the observation that H. pylori urease is not found exclusively in the cytoplasm has been hypothesized to be a survival mechanism from lysed bacteria.^{[ 15 , 18} , ^{19 – 20 ]} Therefore, urease is essential for this bacterium's survival in the gastric environment, enabling colonization in nearly half of the global population and, in some cases, leading to a high risk of gastritis, ulcers and gastric cancer.^{[ 7 , 21 , 22 ]} While conventional antibiotic therapy is typically effective, it still fails in 5%–20% of cases due to resistance and compliance issues,^{[ 23 , 24 ]} making urease inhibition a key target for new therapies.

The excessive ammonia production by ureolytic bacteria also leads to raised pH levels, promoting the development of various pathologies. For example, urine alkalinization leads to the formation of urinary stones and creates an ideal environment for bacterial invasion of Staphylococcus saprophyticus and Proteus spp., resulting in urolithiasis and pyelonephritis.^{[ 7 , 25 ]} This issue is especially common in catheterized patients with urinary infections leading to encrustation blockages,^{[ 7 , 26 ]} and it has been shown to also provide protection from neutrophil recognition.^{[ 27 ]} While several clinical trials have shown the effectiveness of urease inhibitors, their use remains limited^{[ 28 ]} highlighting the need for new treatments.

Beyond these two pathologies, urease is also produced by other organisms implicated in human infections. In Klebsiella pneumoniae, urease is linked to stress resistance, while in Cryptococcus neoformans, it plays a key role in the fungus’ ability to invade and colonize the central nervous system, causing meningitis.^{[ 29 ]} Urease's role in dental health has also been studied where Streptococcus salivarius reduces dental caries but contributes to plaque and calculus deposition.^{[ 7 ]} Furthermore, the enzyme can modulate the immune response, aiding bacterial survival from macrophages and phagocytes.^{[ 30 , 31 ]}

Additionally, urease's specificity and rapid catalytic rate make it valuable in various applications, from sensor technology for environmental monitoring to medical diagnostics. It also shows potential in artificial kidneys,^{[ 32 ]} the food industry^{[ 33 ]} and in chemical catalysis for organic reactions.^{[ 34 , 35 ]} In addition, urease displays insecticide and antifungal activity^{[ 36 ]} and various urease fragments have been shown to disrupt cell membranes.^{[ 37 ]}

3. Urease Structure, Regulation, and Function

Structurally, urease belongs to the superfamily of amidohydrolases and catalyzes the hydrolysis of urea into ammonia (NH₃) and carbamic acid (H₂NCOOH) which subsequently decomposes to carbon dioxide (CO₂). Despite urea being inherently stable due to resonance stabilization of its amides,^{[ 38 ]} urease is able to accelerate its hydrolysis by several orders of magnitude.^{[ 9 , 39 ]} Crystallographic studies revealed this efficiency comes from the conserved active site^{[ 40 , 41 ]} featuring a di‐nickel metallocenter, coordinated by histidine, aspartic acid and a nonstandard carbamylated lysine,^{[ 2 , 42 ]} and an adjacent flap helix‐turn‐helix region with high mobility that allows an open/close state^{[ 43} , ^{44 – 45 ]} which shifts up to 5 Å.^{[ 42 , 46 ]} (Figure  1C). The latter acts as a gate to regulate substrate influx and release^{[ 41 , 47 ]} and is involved in the catalytic events.^{[ 48} , ^{49 – 50 ]} The amino acid sequence of this flap region is highly conserved^{[ 18 ]} and many inhibitors aim to disrupt its normal dynamics.^{[ 44 , 51 ]} A tetrahedral cluster of water molecules fills the active site, and a hydroxide ion bridging both nickel ions is responsible for the urea hydrolysis (Figure 1C,D).

Figure 1.

A) Surface and ribbon representation of the monomeric subunits showcasing the α, β, and γ subunits as well as the location of the active site. B) Structural comparisons of urease isoforms in different species with front and side view of the different multimeric assemblies bearing the common trimeric substructure highlighted.^{[ 2 ]} C) Close‐up view of urease active site illustrating the conformational change of the “mobile flap” (orange) observed in different crystals highlighting its high mobility. When in open conformation, it is possible to observe the entrance of the active site including the conserved water molecules (represented by red spheres). D) Schematic representation of the proposed catalytic mechanism by ureases leading to the formation of the bidentate interactions that is susceptible to the nucleophilic attack by hydroxyl forming the tetrahedral intermediate that then splits into ammonia and carbonate.^{[ 2 ]}

The monomeric units of urease vary among species from a single chain (α) for JBU and Pigeon Pea urease (PPU), an αβ heterodimer for H. pylori urease (HPU) and αβγ in Klebsiella aerogenes urease (KAU), Sporosarcina pasteurii urease (SPU) and Yersinia enterocolitica urease (YEU) (Figure 1A). Despite the variations, urease still maintains a conserved trimeric quaternary structure (≈90 kDa), although the number of subunits within this assembly is species‐dependent.^{[ 52 ]} In the case of HPU, the enzyme forms a spherical dodecameric assembly ((αβ)₃)₄ about 13 nm in diameter with 12 catalytic units^{[ 18 , 53 , 54 ]} This particular architecture has been hypothesized to provide a self‐supporting buffer system^{[ 18 , 55 ]} that effectively maintains activity even at a low pH,^{[ 16 , 53 ]} which in turn allows H. pylori to survive.^{[ 54 ]}

The exact catalytic mechanism has been a matter of constant debate, but a detailed discussion is beyond the scope of this review. Overall, it involves a concerted mechanism by which the urea's carbon becomes more susceptible to a nucleophilic attack by the hydroxide that splits the molecule (Figure 1D). Urea binds to the active site in the open conformation, displacing water molecules and stabilized by hydrogen bonding from histidine residues^{[ 44 , 56 ]} that coordinate the carbonyl oxygen to the electrophilic Ni(1) ion,^{[ 1 ]} while the amine group interacts with the nearby charged carboxylate groups. Closing of the flap forces interactions withing the metallic cluster, stabilizing the tetrahedral transition state and enabling the attack by the hydroxyl^{[ 1 , 2 , 50 , 57 ]} to form the hydrated urea.^{[ 42 , 58 ]} The molecule then splits into ammonia and carbamate which are then released with the assistance of residues in the mobile flap which, upon opening, also allows water to enter and regenerates the enzyme.^{[ 44 , 59 ]} The tight binding mechanism has been attributed to the high reaction rate compared with other hydrolases.^{[ 44 , 60 ]}

4. Urease Inhibition

Enzymes have long been prime targets for modulation by small molecules and urease inhibition has been extensively explored for agricultural and medicinal purposes. Despite significant advancements, current urease inhibitors present clear trade‐offs in potency, stability, specificity and toxicity.^{[ 61} , ^{62 – 63 ]} A critical limitation has also been using potency observed in isolated enzymes and translate it into whole biological cells, where factors like permeation, efflux and chemical and metabolic stability are crucial. Additionally, many of these inhibitors may only provide short‐term efficacy allowing for microbial adaptation, and their long‐term effects are often still largely unknown.^{[ 64 , 65 ]} Cumulative evidence also suggests that even inhibitors with moderate activity may provide better practical performance in the soil by being more stable and having longer duration of activity.^{[ 66 ]} Furthermore, clinical success as a standalone therapy has yet to be achieved, even though urease inhibition may be beneficial in combination with antibiotics for both treatment of H. pylori infections and to reduce bacterial growth and catheter encrustation.^{[ 11 ]} An example is acetohydroxamic acid, approved by the FDA in 1983 to prevent urinary stones. Although it completely inhibits bacterial ureolysis, it does not inhibit bacterial growth,^{[ 67 ]} and its use is limited by severe side effects.^{[ 26 , 68 ]}

Thus, substantial opportunities are available to develop new inhibitors that meet the unique challenges of urease inhibition. Continuous screening of new molecular scaffolds is ongoing,^{[ 69 ]} and design of new inhibitors has greatly benefited from the vast wealth of structure–activity relationship data.^{[ 69 ]} The two main drug development strategies either target the catalytic center or the key cysteine in the flap region.^{[ 70 ]}

To date, over 8000 reported compounds have been tested against ureases from different species; showcasing a remarkable chemical diversity^{[ 69 ]} and growing interest in this topic is evidenced by numerous reviews on urease inhibition,^{[ 71} , ^{72 – 73 ]} including a review of patented inhibitors.^{[ 74 ]} The main aim of this review is to build upon the state‐of‐the‐art developments and provide an in‐depth analysis of the various small‐molecule scaffolds reported against urease. We present the scaffold activity of each compound class, illustrated by a color‐coded bar beneath each structure representing the distribution of the potency (reported as compound‐to‐control activity ratio, where controls are either thiourea or acetohydroxamic acid), as a way to provide insights into the structure–activity relationships.

4.1. Crystal Structures of Urease in the Presence of Inhibitors

Given the high similarity of the active site across different species, crystallographic studies (Table  1 ) have been pivotal in elucidating the structure and function of urease. The rapid rate of urea hydrolysis (adduct duration of ≈20 μs) by urease poses a challenge for crystallographic characterization. However, using fluoride alongside urea hampers the hydrolysis by replacing the bridging hydroxide. This approach has confirmed that urea binds via chelation of the carbonyl oxygen to Ni(1) and an amide nitrogen to Ni(2), supported by hydrogen bonds with residues in the flap region.^{[ 75 ]} Fluoride itself has long been recognized as an effective inhibitor against both microbiota of the bovine rumen^{[ 76 , 77 ]} and isolated ureases.^{[ 78} , ⁷⁹ , ⁸⁰ , ⁸¹ , ^{82 – 83 ]} Crystal structures revealed two fluoride anions coordinated to the nickel ions, supporting the kinetic studies of mixed‐type inhibition against SPU.^{[ 78 ]} However, the mode of inhibition varies among species and experimental conditions.^{[ 83 ]}

Table 1.

Crystallographic data of structures of urease with ligands including activity data, key interactions, and distance of the mobile flap region as measured by the distance of cysteine on the flap and the metal center. W: water.

PDB code	Species	Ligand	Ki/IC50	Interaction residues	Distance flap‐Ni [Å]	Ref.
6QDY	SPU	Urea	Substrate	αHIS‐222, αALA‐366, αALA‐170, αGLY‐280, Ni(1), Ni(2)	7.79	[75]
4CEX	SPU	Fluoride	0.28–3.00 mM	Ni(1), Ni(2), ASP‐363, W3	10.96	[78]
4GOA	JBU	Fluoride	0.28–3.00 mM (SPU)	Ni(1), Ni(2), ASP‐363, W3	11.85	[500]
6G48	SPU	Silver [AgNO3]	8.5 nM (JBU)	αCYS‐322; αHIS‐323; αMET‐367	10.24	[95]
7B58	SPU	[Ag(PET3)Cl]4	55 nM (JBU)	αCYS‐322; αHIS‐323; αMET‐367	10.32	[96]
7B59	SPU	[Ag(PET3)Br]4	45 nM (JBU)	αCYS‐322; αHIS‐323; αMET‐367	10.33	[96]
7B5A	SPU	[Ag(PET3)2NO3	>22 nM (JBU)	αCYS‐322; αHIS‐323; αMET‐367	10.33	[96]
6I9Y	SPU	(Au(Pblm)Cl2)	9 nM (JBU)	αCYS‐322; αHIS‐323; αMET‐367	10.50	[92]
7P7N	SPU	Au(PEt3)I	38 nM (JBU)	αCYS‐322; W; MET‐367	12.82	[501]
7P7O	SPU	Au(PEt3)2Cl	7452 nM (JBU)	αCYS‐322	12.67	[501]
1S3T	SPU	Borate	0.08–0.35 mM	ALA‐170, HIS‐222, W	11.31	[89]
4UBP	SPU	Acetohydroxamic acid	2.6 μM (KAU)[ 83 ]	Ni; HIS‐222; ASP‐363; W	12.94	[224]
1E9Y	HPU	Acetohydroxamic acid	2 mM	Ni, HIS‐221; ASP‐362	10.89	[18]
1FWE	KAU C319A	Acetohydroxamic acid	1.4 mM	Ni, HIS‐219; ALA‐363; ASP‐360	–	[502]
5A6T	SPU	Sulfite	0.19–1.17 mM	HIS‐222, W, W, Ni(1), Ni(2) ALA‐170, ASP‐363	11.89	[85]
5FSD	SPU	2,5‐dihydroxybenzensulfonate	–	CYS‐322, W, W, LYS‐169	13.63	[94]
5FSE	SPU	1,4‐benzoquinone	–	CYS‐322, W	13.65	[94]
1IE7	SPU	Phosphate	0.1–50 mM (KAU)	Ni(1), Ni(2), ALA‐170, ALA‐366, W, ASP‐363	11.37	[86]
7KNS	JBU	Phosphate	0.1–50 mM (KAU)	NI(1), ALA‐636, ASP‐633, ALA‐440	12.06	[503]
3UBP	SPU	Diamidophosphate (DAP)	–	Ni(1), N(2), HIS‐222, ALA‐366, HIS‐323, ASP‐363, ALA‐170	7.85	[87]
5OL4	SPU	NBPT → MATP	6.4 μM	Ni(1), Ni(2), HIS‐222, CYS‐322, ASP‐363	7.85	[504]
6H8J	SPU	NBPTO → DAP	0.62 nM	αALA‐170, αALA‐366, αGLY‐280, αHIS‐222, αHIS‐323	7.82–12.08	[218]
6RKG	SPU	NBPTO → DAP	0.58 nM	Ni(1), N(2), HIS‐222, ALA‐366, HIS‐323, ASP‐363, ALA‐170	6.88	[42]
6RP1	SPU	NBPTO → DAP	1.09 nM	Ni(1), N(2), HIS‐222, ALA‐366, HIS‐323, ASP‐363, ALA‐170	6.86–11.36	[42]
4AC7	SPU	Citrate	1.180 mM	Ni(1), Ni(2), GLY‐280, AG‐339, HIS‐222, ALA‐170, ASP‐224	12.56	[505]
1UBP	SPU	Beta‐mercaptoethanol	0.435 mM (HPU)	Ni(1), Ni(2), GLY‐280, ASP‐363, CYS‐322, ALA‐366	11.42	[97]
6QSU	HPU	Beta‐mercaptoethanol	0.435 mM	Ni(1), Ni(2), ALA‐169, ASP‐632, GLY‐279	12.27	[70]
4GY7	JBU	Beta‐mercaptoethanol + Phosphate	–	–	11.91	[506]
5G4H	SPU	Catechol	–	WA, WB, CYS‐322	13.19	[507]
6ZO0	SPU	3,4‐Dimethylcatechol	–	WA, WB, CYS‐322	13.12	[46]
6ZO1	SPU	3,5‐Dimethylcatechol	–	WA, WB, CYS‐322	13.48	[46]
6ZO2	SPU	4,5‐Dimethylcatechol	–	WA, WB, CYS‐322	13.42	[46]
6ZO3	SPU	3,6‐Dimethylcatechol	–	WA, CYS‐322	13.37	[46]
6ZNY	SPU	3‐Methylcatechol	–	WA, WB, CYS‐322	13.22	[46]
6ZNZ	SPU	4‐Methylcatechol	–	WA, WB, CYS‐322	13.19	[46]
8A18	SPU	Hydroquinone	–	WA, CYS‐322	13.69
7ZCY	SPU	N‐(2‐chloranyl‐4‐fluoranyl‐phenyl)‐2‐selanyl‐benzamide > DiSelenium	0.0058 nM	CYS‐322	12.51	[91]
8Q2E	SPU	Thiram and dimethylditiocarbamate	13 μM	CYS‐322	13.23	[90]
6ZJA	HPU	2‐{[1‐(3,5‐dimethylphenyl)‐1H‐imidazol‐2‐yl]sulfanyl}‐N‐hydroxyacetamide	0.630 μM	Ni(1), Ni(2), HIS‐221, CYS‐321, HIS‐322	13.31	[70]

Urease inhibition is complex, and it is frequently observed that inhibitors have mixed‐type inhibition mechanisms as well as time‐dependent inhibition. For example, while acetohydroxamic acid (AHA) achieves maximum activity only after 2 h, the bulkier biscoumarins (which display uncompetitive inhibition) require prolonged pre‐incubation to form stable complexes.^{[ 84 ]} Comparative studies of crystallographic data have however identified two primary modes of inhibitor binding (Figure  2 ): 1) targeting the di‐nickel ions at the catalytic centrr and 2) binding to critical cysteine residues in the flap region (αCYS‐319 for KAU, αCYS‐322 for SPU, αCYS‐321 for HPU, αCYS‐592 for JBU). These studies also highlight the flexibility of the flap region which can block urea access to the active site or expand to accommodate bulky inhibitor scaffolds.

Figure 2.

Crystal structures of different urease structures bound to inhibitors, highlighting the key residues of interaction of each inhibitor the enzyme as well as the distance of the mobile flap region measured from the corresponding cysteine to the carbamylated lysine.

Among the compounds that bind to the nickel ions, various binding modes have been reported and their potency has been rationalized by the hydrogen bond networks that play a crucial role in their interactions. For example, both competitive inhibitors phosphate and sulfite are able to bind to the nickel metal center in the form of monoanion H₂PO₄ ⁻ and [SO₂(OH)]⁻ and are supported by a diverse range of H‐bonding groups from various nearby residues (Figure 2). However, for the doubly charged HPO₄ ²⁻ and sulfuric acid the loss of activity comes from losing important hydrogen bonding to residues such as αALA‐170 and αALA‐366.^{[ 85 , 86 ]} On the other hand, phosphate binds less strongly than the diamido phosphate (DAP) as the latter is able to establish a larger number of H‐bonds via its amines. Indeed, phosphoramidates are among the most potent urease inhibitors as they closely mimic the transition state of urea‐urease. Interestingly, phosphoramidates are known to be hydrolyzed in situ by urease through a nucleophilic attack on the phosphate by urease's hydroxide and this is clearly observed in crystal structures where the actual inhibitor monoamido thiophosphate (MATP) and DAP are observed to be the molecules bound after treatment with N‐(n‐butyl)‐thiophosphoric triamide) (NBPT) and N‐(n‐butyl)‐phosphoric triamide (NBPTO) respectively. The binding mode of these analogs is found in a tetrahedral shape with a symmetric bridging to the nickel ions that closely mimics the cluster of water/hydroxide molecules of the active site that they replace. In both DAP and MATP crystal structures, the flap region is observed to be in the “closed state,”^{[ 87 ]} but notably for NBPTO, the flap was found to be in a 60% closed and 40% open state. This change in dynamics also results in protection from titration of the cysteine residue of the flap.^{[ 88 ]} Similarly, both anionic forms of AHA (competitive inhibitor) and citrate (uncompetitive inhibitor and activator at low concentration) chelate the nickel ions and rely on additional hydrogen bonds for assisting their interaction that contribute to tight complexes (Figure 2). Additionally, the analog 2‐{[1‐(3,5‐dimethylphenyl)‐1H‐imidazol‐2‐yl]sulfanyl}‐N‐hydroxyacetamide demonstrates the importance of extending interactions toward the flap region.^{[ 70 ]} In contrast, boric acid, another well‐characterized competitive inhibitor of urease (K _i of 0.1–0.33 mM), shows a different binding mode. The form of boric acid that shows inhibitory activity appears to do so via chelation of the nickel ions by the neutral trigonal form B(OH)₃ rather than anionic B(OH)₂ ⁻. This neutral form replaces a cluster of three water molecules in the nickel coordination sphere but does not displace the bridging hydroxide.^{[ 89 ]}

In addition to targeting the nickel center, some inhibitors target the cysteine in the flap region. From three cysteines available for conjugation, only two are readily accessible. Crystal structures reveal that the mobile flap cysteine is the most reactive, likely due to the presence of an adjacent neighboring histidine that facilitates acid–base proton transfer.^{[ 90 ]} This has been hypothesized as molecules such as N‐(2‐chloranyl‐4‐fluoranyl‐phenyl)‐2‐selanyl‐benzamide resulted in the formation of a CYS–S–Se–Se adduct with αCYS‐322 (SPU) that suggests hydrolysis of the molecule, whereas at αCYS‐555 (SPU), the molecule was not hydrolyzed but solely bound to cysteine via CYS–S–Se‐Molecule.^{[ 91 ]} Similarly, hydrolysis of the ligand at this site has been observed for Au(III) ligands,^{[ 92 ]} and, in the case of Thiram, the adduct was only observed at αCYS‐322 and not αCYS‐555.^{[ 90 ]} This is also observed for other cysteine‐targeting compounds, as quinones were able to react with both cysteines (CYS‐322, CYS‐555), whereas catechol‐based inhibitors bonded only to the cysteine of the flap (CYS‐322).^{[ 93 ]} In the case of quinones, it was observed that, beyond the presence of the thiol‐substituted benzene‐1,4‐diol, when in presence of sulfite, the latter also reacted with the quinone resulting in the observation of a 2,5‐dihydroxy‐benzenesulphonate moiety bound to αCYS‐322.^{[ 94 ]} From these results, it was also suggested that the inactivation of urease was not by blockade of the substrate transport but rather by preventing the flap from closing.

For catechol analogs it was suggested, based on kinetics, that catechol was first converted to the active molecule by the oxidation with molecular oxygen dissolved which then forms a covalent bond with αCYS‐322 as seen in the crystal structure.^{[ 46 ]} Targeting the mobile flap cysteine changes the dynamics of urease and can also disrupt proper binding of the substrate.^{[ 95 , 96 ]} Silver (Ag(1)), known for its affinity toward sulfhydryl and thioether, can be observed in the crystal structure interacting with αCYS‐322, αHIS‐323 and αMET‐367, keeping the flap region in an open state. Another known inhibitor, β‐mercaptoethanol, displays a dual mechanism by which it chelates nickel supported by an H‐bond from αGLY‐280,^{[ 97 ]} but in the crystal structure of SPU, it is also possible to observe a second molecule of β‐mercaptoethanol bridged to αCYS‐322 via disulfide bonding and establishing a hydrogen‐bond with αALA‐366. On the other hand, β‐mercaptoethanol was an activator of Citrullus vulgaris urease, highlighting the complexity of inhibition and possible differences between species.^{[ 81 ]}

Despite the number of available crystal structures of urease‐inhibitor complexes, these represent only a small fraction of the vast chemical space that has been screened. A comprehensive understanding of these is therefore instrumental in guiding the design of new inhibitors. In terms of mechanism, these follow the same observation drawn from crystalized inhibitors and can be broadly categorized into: 1) substrate‐like analogs, 2) sulfonamide analogs, 3) transition‐state analogs, 4) hydroxamic acid analogs, 5) hydrazones and Schiff bases, 6) natural products, 7) quinolones, quinolines and quinazolinones, 8) aliphatic chain compounds, 9) boronic acids, 10) metal ions and organometallic compounds, 11) heterocycles, and 12) covalent binders to the flap motif.

4.2. Substrate‐Like Analogs

4.2.1. Small Urea Analogs are also Urease Substrates

Designing compounds that mimic natural substrates or enzyme reaction intermediates has long been a successful strategy for urease inhibition. Even though this approach has led to the identification and development of several potent inhibitors, it has also been observed throughout the years that some of these structural analogs are hydrolyzed by urease (Figure  3 , M1–M15). These include the larger phenylurea (M12) which underscores the tolerance of the catalytic center for larger compounds.^{[ 59 , 98} , ^{99 – 100 ]} While it is expected that these analogs compete with urea, some that share the substructure of urea, such as hydroxyurea (M6)^{[ 101 ]} and dihydroxyurea (M7),^{[ 102 ]} are in fact both noncompetitive inhibitors while still slowly hydrolyzed by urease. On the other hand, acetohydroxamic acid (M18), a known competitive inhibitor, is not a substrate^{[ 101 ]} and similarly the cyclic 2‐imidazlidinone (M19) has shown no evidence of being hydrolyzed.^{[ 98 ]} Interestingly semicarbazide (M13) has been reported as a substrate whereas its closely related analog carbazide (M16) was shown not to be a substrate.^{[ 98 ]} Furthermore, having two amide groups does not seem to be a requirement as both formamide (M14), acetamide (M3), and 4‐nitrophenyl carbamate (M15) have undergo hydrolysis by urease, with the latter being reported to have even higher affinity than urea for urease.^{[ 99 , 103 ]} This is in agreement with the first step of the binding mechanism where the carbonyl binds to the metal nickel ion (Figure 1D).

Figure 3.

Chemical structure of small substrate‐like analogs known to interact with urease. These compounds share structural features with urea, making some of them also susceptible to hydrolysis by urease.

Thiourea (M2), a sulfur analog of urea and a “soft base” relative to oxygen,^{[ 99 ]} is a known competitive inhibitor (K _i : 70 mM) commonly used as a control in screening assays against urease.^{[ 104 ]} Despite alterations in the transition state and the fact that is binds less tightly than urea (with a diference of around 2 kcal mol⁻¹), thiourea still slowly hydrolyzes as a substrate.^{[ 99 , 103 ]} Similarly, thioacetamide (M5, K _i : 83 mM) is also a substrate of JBU,^{[ 59 , 99 ]} while the more potent 2‐mercaptoacetamide (M17, K _i : 2.95 mM) has not been tested as a substrate.^{[ 26 ]} Interestingly, for reasons not yet clear, thiourea (M2), thioacetamide (M5), and acetamide (M3) showed neither inhibitory activity nor to be substrates of KAU despite the similarity of the active site.^{[ 105 ]} Finally, the substitution of urea oxygen by a nitrogen as found in guanidine chloride (M20, K _i : 29.8 mM)^{[ 99 ]} also results in a competitive inhibitor but yet was not a substrate, whereas the sulfamide (M21) analog was neither hydrolyzed nor showed inhibitory activity.^{[ 104 ]}

4.2.2. Urea‐ and Thiourea‐Bearing Compounds: More than a Substrate

Incorporating both urea and thiourea moieties into compounds is a very intuitive strategy for urease inhibition, as these groups are found in many biologically active molecules, including antimicrobial, antiviral, and antitumor agents. In our dataset, over 400 and 1300 unique compounds bearing urea and thiourea, respectively, have been tested for their urease inhibitory activity. While they generally show moderate activity, thiourea compounds tend to be more potent than their urea counterparts. However, only half of those compounds outperform thiourea itself. Although these substrate analogs are expected to act competitively, many exhibit mixed‐type^{[ 106 , 107 ]} or noncompetitive inhibition^{[ 108} , ¹⁰⁹ , ¹¹⁰ , ¹¹¹ , ^{112 – 113 ]} (Figure  4 , M25–M34). Moreover, among the most potent inhibitors (activity < 100 nM), the importance of the urea and thiourea groups is unclear, as they often appear in cyclic forms (e.g., dihydropyrimidinones) or as part of larger chemical groups (e.g., carbamoylthioureas). Indeed, the simple 1‐phenylureas (Figure  5 , C1) exhibit very weak activity and, as seen with phenylurea, are likely to hydrolyze.^{[ 62 , 114 ]} Even modification to a hydroxamic acid only improves the activity to the same level of thiourea.^{[ 115 ]} Surprisingly, while 3,3‐dimethyl‐1‐phenylurea is almost devoid of activity, the closely related thiourea analog 1‐(4‐bromophenyl)‐3,3‐dimethylthiourea was found to be a potent inhibitor of JBU (K _i : 255 nM).^{[ 116 ]} As for the di‐substituted 1,3‐diphenyl ureas (C2)^{[ 62 , 114 , 117 , 118 ]} and 1,3‐diphenyl thioureas (C3)^{[ 117 , 119 ]} analogs, these have shown similar moderate to weak inhibitory activity and the type of substitutions was determinant for both activity and mechanism of action as competitive, mixed‐type, and noncompetitive inhibitory kinetics have been reported.^{[ 119 ]} For example, 1,3‐diphenyl urea itself is inactive but 1,3‐bis(4‐nitrophenyl)urea has an IC₅₀ of 1.25 μM and several 1‐nitrophenylurea‐phenylhydrazones analogs showed moderate activity (IC₅₀: 10–36 μM). Similarly, while the 3‐phenyl‐1‐(pyridinyl)urea analogs were shown to be inactive,^{[ 114 ]} their corresponding sulfur analogs were shown to be moderate inhibitors (IC₅₀: 8–24 μM).^{[ 119 ]} For comparison, the structurally similar N‐phenylbenzenesulfonamides (C4) were all shown to have weak activity,^{[ 120 ]} but further lengthening of the molecule^{[ 107 , 121 , 122 ]} resulted in increased potency particularly for the sulfadiazine substitution (IC50: 2–233 nM)^{[ 107 ]} alluding to the possibility that further extending the urea‐mimetic scaffolds may increase binding. Indeed, such extension from analogs of diphenylureas with bridged disulfides showed strong to moderate activity with the best compound exhibiting an IC₅₀ of 400 nM.^{[ 123 ]}

Figure 4.

Several examples of potent analog series of urea‐ and thuiourea‐bearing inhibitors. Despite some class diversity can be observed, it appears that for the most potent classes some structural elements are common.

Figure 5.

Comparison of inhibitory activity between classes of substrate mimetics compounds and similar scaffolds. Green bars represent the distribution of the ratio of activities using thiourea as reference (activity = 1). A darker color indicates higher frequency of compounds in each activity bin.

The substitution of the phenyl by adamantane in 1,3‐diphenylthioureas typically results in scaffolds with similar activity (C5),^{[ 123 , 124 ]} but analogs with strong activity were found for the 4‐[3‐(1‐adamantyl)ureido]phenyl sulfamate scaffold (Figure 4‐M29).^{[ 125 ]} However, improvements in activity for analogs with phenyl‐to‐adamantane replacement were also observed in a series of 4‐[(adamantan‐1‐yl)carbamoyl]phenyl sulfamate analogs (C6) where the phenyl analogs had ≈11‐fold loss in activity.^{[ 125 ]} Surprisingly, the most potent compound of this series was found to be a competitive inhibitor despite having an amide group instead of thiourea (C6), indicating that the urea group may not be crucial for the activity of these similar scaffolds. This is also supported by incorporation of thiourea into cyclic analogs^{[ 126 , 127 ]} that maintain similar levels of activity (C7 and C8). Additionally, the N‐(adamantan‐1‐yl)benzenesulfonamide analogs (C9) were shown to be potent inhibitors (IC₅₀: 17–288 nM) further emphasizing an increase in potency when compared with the diphenyl substitution (C4). However, the most potent compound was shown to inhibit via a noncompetitive mechanism.^{[ 107 ]}

Increasing the flexibility of N,N′‐substituted ureas as in the case of the series of 1‐benzyl‐3‐phenylureas (C10), did not result in improved activity.^{[ 62 , 117 ]} However, replacing the benzyl ring with a benzylpyridine‐1‐ium and further extending the scaffold (C11) resulted in molecules with strong activity (IC₅₀: 4.08–6.20 μM) with the most potent analog showing mixed‐type inhibition.^{[ 128 ]} The pyridium seems to be as important for the activity as the pyridine analog counterparts,^{[ 119 ]} and changing to a benzyl‐triazole produces only moderate to weak activity.^{[ 129 ]} Similarly, changing to an indole group resulted in moderate to weak inhibitors that were shown to inhibit either via mixed‐type or noncompetitive mechanisms.^{[ 130 ]}

The potency of substrate mimetics is however significantly increased when combining thiourea with other groups such as the case of carbamoylthioureas. Even though benzoylthiourea is itself a weak inhibitor (32% at 0.5 mM)^{[ 131 ]} and the majority of these compounds only demonstrates moderate to weak activity (C12), there are several modifications on the phenyl ring that result in potent inhibitory activity (K _i : 47–931 nM), with these compounds mostly exhibiting mixed‐type inhibition.^{[ 132 ]} Similarly, (2‐phenylacetyl)thiourea is itself a relatively weak inhibitor (C13) but its p‐chloro analog (Figure 4‐M33) resulted in a potent mixed‐type inhibitor with a K _i of 40 nM.^{[ 133 ]} In fact, the chloride substitution placed at the right location seems to greatly contribute to enhance the inhibitory activity of this class of inhibitors. In the case of benzoylthioureas (C12), the most potent analog (IC₅₀: 60 nM) features an o‐chloro substitution while its p‐chloro analog only shows moderate activity (IC₅₀: 66.5 μM).^{[ 132 ]} For phenylacetylthioureas, the ortho substitution also showed similar moderate activity (IC₅₀: 63.6 μM).^{[ 133 ]} These results emphasize the importance of the chloro position in determining the inhibitory potency across these families of compounds. The increased spacing in phenylacetylthioureas may also explain why these are on average more potent than their corresponding benzoylthioureas, as well as the differences in potency between similar analogs. This is further supported by a series of Schiff bases of N‐phenylthiosemicarbazones (C14), which have shown potent activity,^{[ 134} , ¹³⁵ , ^{136 – 136 ]} and similarly the o‐chloro analog (K _i : 170 nM) was more potent than the para‐ and meta‐analogs. However, the most potent compound of this class was an o‐nitro substituted (Figure 4‐M34, K _i : 90 nM) that was shown to display a competitive mechanism of inhibition.^{[ 134 ]}

As observed with other thiourea analogs, the pyridine ring substitution led to only moderate activity in the low micromolar range.^{[ 136 ]} Other ring substitutions generally resulted in weak activity, except for furan^{[ 136 ]} and substituted benzofurans, which showed potent inhibition, with 11 compounds in the nanomolar range (IC₅₀: 77–592 nM).^{[ 58 , 135 ]} Replacing the phenyl ring with cyclohexyl yielded similar potency.^{[ 135 ]} However, substitutions with 1,4‐benzodioxane^{[ 137 ]} and thiosemicarbazones with extended coumarin showed moderate activity.^{[ 138 ]} Nevertheless, the thiosemicarbazone substructure also appears in a series of dihydropyrazolecarbothiamide (C15)^{[ 111 , 139 ]} which exhibited potent to moderate inhibitory activity. Notably, the substitution with a coumarin group resulted in particularly strong inhibition, with an IC₅₀ ranging from 0.358 to 128.7 nM and the most potent inhibitor in this series was found to be noncompetitive (Figure 4‐M31).

Another class of structurally similar analogs are the series of phenylacetylureas and thioureas in conjugation with an extended alkyl chain (C16). Smaller chains, up to six carbons, generally result in weak inhibitory activity.^{[ 131 ]} However, for longer 1‐acyl‐3‐arylthioureas, moderate^{[ 140 , 141 ]} to potent compounds (IC₅₀: 21–75 nM) have been reported.^{[ 112 , 113 ]} Alkyl chains with eight carbons have been shown to inhibit via competitive mechanisms^{[ 140 ]} while larger alkyl chains typically act through a noncompetitive mechanism.^{[ 112 , 113 ]} Similar to observations with phenylthioureas, amantadine analogs with varying chain lengths exhibit potent inhibition (IC₅₀: 8.5–30 nM), with the most potent also demonstrating a noncompetitive inhibition mechanism.^{[ 142 ]} These results highlight the impact of the chain length on the inhibitory activity and mechanism of this class of compounds.

Benzoylphenylthiourea analogs have also been extensively explored against different types of ureases.^{[ 108 ]} Although some compounds demonstrated weak^{[ 131 ]} to moderate^{[ 143 ]} activity, this scaffold has also yielded potent molecules^{[ 141 , 144} , ^{145 – 146 ]} (C17). As a common theme in the discussion of thiourea analogs, the most potent compound described so far is the o‐chloro analog M33 (IC₅₀: 1.9 nM, K _i : 3 nM), showcasing once again the importance of this substitution.^{[ 108 , 132 ]} Despite the fact that the acylthiourea group is able to complex with nickel ions^{[ 138 , 147 ]} and some of these compounds having shown to be competitive^{[ 146 ]} and mixed‐type inhibitors,^{[ 131 ]} this analog in particular was shown to be a noncompetitive inhibitor.^{[ 108 ]} As with previous phenylthioureas, the substitution of the phenyl ring with pyridine resulted in loss of activity^{[ 131 ]} while the adamantane substitution (C18) resulted in compounds with potent activity (IC₅₀: 8.7–30 nM), with the most potent being a o‐chloro analog as well.^{[ 142 ]} Other saturated ring substitutions adjacent to the carbonyl instead, such as in 3‐benzoyl‐1‐cyclohexylthiourea resulted however in compounds with only moderate to weak activity.^{[ 143 , 148 , 149 ]} On the other hand, by far the most active substitution was observed to be with 4‐aminocoumarin (C19) in a similar manner to the dihydropyrazolecarbothiamides. Like this class, compound M30 (Figure 4, K _i : 0.32 nM) was shown to be a noncompetitive inhibitor.^{[ 110 ]}

Further expanding the distance between the two phenyl groups adjacent to the thiourea substructure resulted in compounds with very weak activity^{[ 131 , 148 ]} whereas when these were adjacent to the carbonyl resulted in a compound with moderate‐to‐potent activity (C20).^{[ 150 , 151 ]} In the case of 1‐(2‐(4‐isobutylphenyl)propanoyl)‐3‐arylthioureas, very potent inhibitors were reported,^{[ 151 ]} and the kinetics of three compounds showed that two were mixed‐type inhibitors while the most potent of the series (a 2,3‐dichloro analog) was a competitive inhibitor. The isobutylphenylpropanoyl part seems important for the activity since its corresponding analog 1‐(2‐phenylacetyl)‐3‐(4‐sulfamoylphenyl)thiourea showed lower levels of activity in comparison (K _i : 1.6 μM vs. IC₅₀: 34 nM).^{[ 141 , 151 ]} A slightly different group but with similar spacing has been reported for N‐[(phenylcarbamothioyl)amino]benzamides with more than 70 compounds tested showing overall potent to moderate levels of inhibition (C21).^{[ 152} , ¹⁵³ , ¹⁵⁴ , ¹⁵⁵ , ¹⁵⁶ , ^{157 – 158 ]} Again replacement with pyridine resulted in loss of activity,^{[ 159 ]} whereas replacement with 4‐hydroxyquinoline resulted in strong to moderate inhibition (IC₅₀: 0.6–24.1 μM).^{[ 160 ]}

Despite these classes of substrate mimetics displaying quite potent activity against urease, there are still concerns over their potential lack of selectivity and potential side effects. Their potential to coordinate metals as bidentate ligands can lead to inhibition of other important enzymes, and some of these compounds have also shown ability to bind to DNA.^{[ 58 , 138 , 161 ]}

4.2.3. Cyclic Ureas and Thioureas

Urea and thiourea substructures are also present in cyclic scaffolds (C22–C25). While hydantoin and thiohydantoins show weak activity,^{[ 162 ]} barbiturates^{[ 106 , 163} , ¹⁶⁴ , ¹⁶⁵ , ¹⁶⁶ , ¹⁶⁷ , ¹⁶⁸ , ¹⁶⁹ , ¹⁷⁰ , ¹⁷¹ , ¹⁷² , ^{173 – 174 ]} (C22) and thiobarbiturates^{[ 39 , 163 , 164 , 169 , 172} , ¹⁷³ , ¹⁷⁴ , ¹⁷⁵ , ¹⁷⁶ , ^{177 – 178 ]} (C23) have a wide range of activities among the many tested compounds against urease (>100 molecules each class). These have generally exhibited moderate to strong activity, with thiourea analogs consistently demonstrating slightly higher potency, in line with previous observed trends for linear analogs. Barbituric and thiobarbituric acids are known for their antibacterial, sedative, fungicidal, and antiviral activity,^{[ 175 ]} and this class has been proposed to inhibit via chelation of the nickel ions, while establishing hydrogen bonding with key residues.^{[ 164 , 166 , 175 , 179 ]} Nevertheless, a diverse array of types of inhibition has been described ‐ from competitive^{[ 164 ]} and mixed‐type^{[ 39 , 106 ]} to uncompetitive and noncompetitive.^{[ 39 ]} However, even among the most potent described compounds for this class (Figure 4‐M27 and M28), these are still in the same range of activity of other urea‐based compounds but significantly less active than carbamoylthioureas. Furthermore, whether the thiobarbituric part of these analogs is the factor contributing to the activity still remains unclear as some closely similar compounds without this group have shown potent activity such as the case of the mixed‐type inhibitor 2‐((4‐isopropylphenyl)amino)‐2‐oxoethyl 3‐chlorobenzoate (K _i : 0.7 nM).^{[ 180 ]} Closely related dihydropyrimidinones^{[ 109 , 181 ]} (C24) and dihydropyridine‐thiones^{[ 182} , ¹⁸³ , ¹⁸⁴ , ¹⁸⁵ , ^{186 – 187 ]} (C25) have similar levels of activity and both mixed‐type,^{[ 183} , ¹⁸⁴ , ¹⁸⁵ , ¹⁸⁶ , ^{187 – 188 ]} and noncompetitive inhibition have been reported. Interestingly, in this class of compounds, both thiophene (M25) and benzopyranone (M26) analogs showed potent inhibition whereas the thiophene analog of dihydropyridine‐thiones does not have inhibitory activity alluding to the importance of the oxygen for the activity.

4.3. Sulfonamide Analogs

Despite the structural similarity to substrate and transition state analogs, sulfamide has shown not to be an inhibitor nor a substrate of urease,^{[ 104 ]} and sulfanilamide showed no inhibitory activity against soil urease as well.^{[ 189 ]} Sulfonamides however, have been extensively screened against urease (>500 unique compounds) with a diverse range of scaffolds that in general display strong activity (C26). This class of compounds is also well‐known for its antibacterial properties and structural versatility, having shown to inhibit several enzymes, notoriously carbonic anhydrase (by displacing a water molecule bound to Zn(II)) and thus the lack of potential specificity is to be expected.^{[ 63 ]}

Among the scaffolds with the most potent activity, the sulfonamide group is observed to be both incorporated into the core scaffold of the molecule or positioned at the periphery where it remains as the free amine (Figure  6 ). In contrast, cyclic sulfonamides, such as those observed in saccharin analogs, have generally shown only moderate to weak inhibition.^{[ 190 , 191 ]} As previously highlighted, some of the potent sulfonamide analogs (C6) include amantadine analogs (IC₅₀: 17.76–288.80 nM) via noncompetitive inhibition.^{[ 107 ]} Other notorious compounds of high potency bearing the sulfonamide group include: hybrids of sulfadiazine (C27, most potent IC₅₀: 2 nM), pyrazolotriazine (C28, most potent K _i : 10 nM), iminothiazoline (C29, most potent K _i : 20 nM), thiosemicarbazide (C30, most potent with IC₅₀: 26 nM), arylthioureas (C31, most potent IC₅₀: 0.034 nM), hydroxamic acid (C32, most potent IC₅₀: 38 nM), and ciprofloxacin (C33 most potent IC₅₀: 45 nM), all of which have shown mixed‐type inhibition.^{[ 107 , 151 , 163 , 172 , 179 , 192} , ¹⁹³ , ¹⁹⁴ , ¹⁹⁵ , ¹⁹⁶ , ¹⁹⁷ , ¹⁹⁸ , ¹⁹⁹ , ^{200 – 201 ]} Similar to substrate analogs, the contribution of the sulfonamide group itself for the activity remains uncertain. For example, for the thiosemicarbazides (IC₅₀: 26–255 nM)^{[ 107 ]} the activity seems to be driven by this group as the replacement with other groups such as isonicotinoyl, diphenylmethanimine, alkyl chains, naproxen, or isatin results in compounds with only moderate or weak activity^{[ 188 , 202} , ²⁰³ , ²⁰⁴ , ^{205 – 206 ]} and, as discussed previously, this group is well‐known to have potent activity. Likewise, there are known arylthioureas without sulfonamide that are more potent.^{[ 141 , 151 ]} Furthermore, in the case of enamine barbiturate and thiobarbiturates the sulfonamide analogs are relatively weak compared to other substitutions.^{[ 163 , 172 ]} Sulfamethoxypyridazine analogs, despite their similarity to the sulfadiazine, only display moderate activity.^{[ 207 ]} Conversely, incorporation of sulfadiazine and other groups into levofloxacin increased the potency in comparison to the reference antibiotic up to sevenfold.^{[ 208 ]} However, for cephalosporanic acid the hybridization with sulfonamides did not improve activity.^{[ 117 ]}

Figure 6.

Overview of sulfonamide‐containing scaffolds and their range of urease inhibitory activities. Examples include both core‐integrated sulfonamide and those positioned peripherally. A darker color indicates higher frequency of compounds in each activity bin.

Sulfamate analogs of substituted carbamolphenyl (C34) show similar SAR to the substrate mimetics previously described. Again, the amantadine analogs showed higher potency (IC₅₀: 62–9730 nM) in comparison to their phenyl analog (IC₅₀: 1052–4391 nM) and substitutions with saturated rings resulted in inactive compounds. The most potent inhibitor 4‐[(adamantan‐1‐yl)carbamoyl]‐2‐bromophenyl sulfamate was also shown to be a competitive inhibitor.^{[ 125 ]} Furthermore, the removal of bromo showed little loss in activity but more electronegative halogens (F, Cl) resulted in a significant decrease in the inhibition potency.

4.4. Transition State Analogs

Because enzymes stabilize the transition state of substrates, designing inhibitors that mimic this intermediate state is usually an effective strategy that leads to potent inhibition. In this context, organophosphorus compounds resemble the tetrahedral intermediate formed during the hydrolysis of urea (Figure  7 ). As early as 1934, phosphate was recognized as a pH‐dependent weak competitive inhibitor of urease. Later, crystal structures (PDB 1IE7 and 7KNS) confirmed its interactions with the metal center, revealing four coordinated bonds with nickel ions.^{[ 105 ]} Furthermore, organophosphate‐based insecticides were known to inhibit soil urease. Since then, various organophosphorus compounds have been screened and emerged as one of the most successful classes in reducing urease activity in terms of potency.^{[ 209} , ²¹⁰ , ²¹¹ , ^{212 – 213 ]} These are particularly valuable for agricultural purposes to control urea hydrolysis and reduce ammonia loss from the soil, thus allowing better yields of urea‐based fertilizers^{[ 214 , 215 ]} where, for certain conditions, ammonia volatilization can be reduced as much as 50%.^{[ 72 ]} The potency from these compounds comes from their ability to bind tightly to the active metallocentre and act as a classical transition state analog of the enzymatic reaction (Figure 7). This can clearly be observed in the crystal structure of JBU with DAP with the formation of a tetrahedral conformation with both nickel ions supported by hydrogen‐bond stabilization from other residues in the active site.

Figure 7.

Schematic representation of the transition state conformation of urea in the active site of urease and different organophosphorus analogs that mimic this transition.

Nearly 200 unique compounds from this class have been screened. While this class includes some of the most potent inhibitors, only about 50% of these compounds are more active than thiourea. However, around 20% exhibit activity below 100 nM, showcasing their high potency in inhibiting urease. Overall, these inhibitors tend to be slow binders^{[ 216 ]} and their efficacy is also strongly limited by their hydrolytic stability. For instance, in the various crystal structures it is often observed that the bound compound is actually the result of the catalytic reaction with urease. For example, the commercial N‐(n‐butyl)thiophosphoric triamide (NBPT) used in agriculture worldwide is converted to the intermediate diamido compound (N‐(n‐butyl)thiophosphoric diamide, NBDP) that then breaks down into monoamidothiophosphoric acid (MATP). This then binds to urease in the tridentate conformation found in the crystal (PDB 5OL4).^{[ 217 ]} This is also observed with other tri‐and diamides such as NBPTO, NBDP and PPDA that bind to the active site and are enzymatically hydrolyzed to form DAP or monoamidophosphoric acid.^{[ 8 ]} Furthermore, as these compounds mimic the transition state by complexing with the nickel ions in the active site, they impose steric requirements and can be seen as more difficult to derivatize for highly substituted molecules.

Although tested to a larger extent, phosphoryl analogs on average seem to be more active than thiophosphoryl analogs (Figure  8 ). Indeed, when comparing the class of triaminothiophosphoryl (Figure 8, C35) to triaminophosphoryl (C36) the corresponding analogs are orders of magnitude more potent. Moreover, NBPT which is the most potent within the class of triaminothiophosphoryl sharply decreases activity with other substitutions whereas the same substitutions do not decrease significantly the activity for triaminophosphoryl analogs.^{[ 216 , 218 , 219 ]} Amides of phosphonic acid bonds (P–N bond) are however not very stable in aqueous solutions making them less appealing for therapeutic applications, such as treating H. pylori infections. At pH 2, their half‐life can be as short as 5 min, leading to a rapid loss of activity.^{[ 220 ]} To overcome such hydrolytic lability, phosphonates (C37–C41) and bisphosphoramides (C42) have been tested as alternative and are among the most potent urease inhibitors to be described. Surprisingly, despite their bulky structure, which includes four phenyl rings, that might raise concerns about steric hindrance, this class of compounds has demonstrated mixed‐type inhibition with remarkably high potency (K _i of 0.057 nM).^{[ 221 ]} These bisphosphoramides are also highly resistant to hydrolysis. On the contrary, when increasing the steric hindrance in N‐alkyl‐substituted phosphoramides it generally results in loss of potency.

Figure 8.

Overview of organophosphorous analogs and their range of urease inhibitory activities. A darker color indicates higher frequency of compounds in each activity bin.

The small 4‐chloro‐N‐diaminophosphinyl)benzamide was shown to be one of the most potent compound described, with a K _i of 0.041 and 0.350 nM for JBU and SPU respectively. Again, in agreement with observations for substrate mimetics, the chlorine atom appears to be fundamental in promoting the right interactions since its analog phenylphosphorodiamidate results in 15‐fold loss in activity. The importance of the positioning of the chlorine is also apparent when comparing to 4‐chloro‐N‐(diaminophosphinyl)benzamide which has a K _i of 1.38 nM and its analog without chloride, N‐(diaminophosphinyl)benzamide (K _i of 1.62 nM). Moreover substitution with a fluorine as seen in fluorofamide results in significant loss of activity (IC₅₀ of 500 nM).^{[ 216 ]}

Phosphinic analogs have been identified as competitive inhibitors, ranging from strong to weak activity. Notably, these inhibitors have shown to exhibit significantly different potencies against ureases from P. vulgaris and P. pasteurii.^{[ 210 ]} This difference between species has also been observed with phenylphosphoriamidate (PDD), a known mixed‐type inhibitor of soil ureases.^{[ 222 ]} Interestingly, PDD demonstrated slow‐binding kinetics with a K _i of 95 nM against KAU,^{[ 105 ]} and a K _i of 1.9 mM against JBU.^{[ 79 ]}

Despite the potency of this class of compounds, their translation into therapeutic applications still needs to be addressed, in particular the need for increased specificity as these compounds are known to have other biological targets. Furthermore, for compounds such as PPDA and NBPT that have shown to inhibit ureolysis of K. pneumoniae, they only reduced their growth marginally.^{[ 67 ]} Similarly, fluorofamide was not able to translate the in vitro activity into in vivo efficacy even when in combination with an antibiotic.^{[ 223 ]} In part this is a result of these compounds being highly hydrophilic as well as their ability to bind ions resulting in an expected limited permeation. Furthermore, for agricultural purposes it has since been reported that even compounds with potency orders of magnitude lower than that of organophosphates may behave very similarly in practice. Some of the main reasons for this are due to the low stability and fast degradation in the soil.^{[ 219 ]}

4.5. Hydroxamic Acid Analogs

Hydroxamic acids are structurally very similar to substrate mimetics and bind to the urease active site in a bidentate manner, chelating the nickel ions similarly to urea.^{[ 224 ]} This class of compounds has been extensively screened against different ureases, with over 400 unique molecules tested. Hydroxamic acids are however inherently nonselective to metal groups, which can lead to off‐target inhibition. For this reason, among approved drugs, only a few utilize the hydroxamic acid functional group.^{[ 225 ]} A prominent example is acetohydroxamic acid (AHA; lithostat, uronefrex), the first compound approved for urease inhibition. Its use however has been limited by severe side effects, such as bone marrow suppression. AHA, a slow‐binding inhibitor with slow dissociation kinetics, is stable under acidic conditions and can permeate bacterial membranes.^{[ 67 ]} However, while AHA suppresses urease activity, it does not inhibit bacterial growth, and even more potent analogs seem to behave in a similar manner.^{[ 121 ]} The hydroxamic acids AHA and benzohydroxamic acid were also shown not to be effective at inhibiting soil ureases.^{[ 189 ]} Indeed, hydroxyurea acts as a substrate, and once depleted, urease regains its full activity.^{[ 226 ]}

Despite these limitations, there is still significant interest in hydroxamic acids due to their ability to reversibly bind to the metal cluster and act as the driving factor of the binding conformation. Similar to observations with organophosphorous compounds, only about half of the tested compounds achieve activity superior to thiourea,^{[ 199 , 227 , 228 ]} alluding to the need for the rest of the molecule to establish proper interactions (C43–C47)^{[ 199 , 227 , 229} , ^{230 – 231 ]} (Figure  9 ). Moreover, these compounds are comparatively less potent than other structural counterparts. Nevertheless, potent hydroxamic acid compounds against urease have been reported, including two drugs (dacinostat and panobinostat).^{[ 121 ]} As observed for other classes, small compounds bearing chloride‐substituted phenyl rings are among the most potent exhibiting IC₅₀ values ranging from 18 to 82 nM.^{[ 227 ]}

Figure 9.

Overview of hydroxamic acid analogs and their range of urease inhibitory activities. A darker color indicates higher frequency of compounds in each activity bin.

In the case of the arylamino (C43) and sulfonylamino acetohydroxamic acids (C46), the most potent compounds were shown to be a mixed‐type inhibitors. Notably, the former still remained significantly active against H. pylori (≈1000 fold more potent in comparison to AHA) and was able to significantly reduce gastric mucosa injury in mice.^{[ 227 ]} Both amino acids and small peptides have also been functionalized with an acetohydroxamic acid moiety resulting in compounds with strong activity.^{[ 232 , 233 ]} Interestingly, across the board these analogs were shown to be more potent against HPU than JBU^{[ 233 ]} highlighting the need for fine tuning of the interactions and conformations allowed.

4.6. Hydrazones and Schiff Bases

The azo group itself (C48) has not been extensively explored in urease inhibition and is generally associated with a wide range of activities, from weak to strong. In contrast, disubstituted hydrazines (C49) have been thoroughly investigated with 380 compounds featuring the R–NH–NH–R group. However, the most active compounds of this class are mostly part of larger chemical groups such as sulfonylhydrazine‐carbothioamide (C51) and thiosemicarbazides (M34, C14, C52). Similar findings are observed for hydrazones (over 680 unique structures) where it is also part of larger chemical groups (C50) and have activities similar to those of hydrazines.

In the case of thiosemicarbazides, the benzofuran analogs^{[ 58 ]} (C51) appear to be more potent than phenyl analogs (C53), while further extending the molecule tends to be detrimental for activity (C54). Although not extensively explored, benzofuran also seems to have a positive impact on the activity of hydrazones^{[ 234 ]} (C55). Additionally, introducing phenolic hydroxyl groups into diazene compounds has a positive impact on activity, with catechol derivatives identified as the most potent analogs in a series of scaffolds.^{[ 152 , 235 , 236 ]}

Hydrazones and hydrazides are known to form stable chelating complexes with many transition metals including nickel.^{[ 237 ]} Inhibition kinetics of this class of compounds have reported both competitive^{[ 134 , 238 ]} and mixed‐type inhibitors^{[ 106 ]} likely via this mechanism. Benzohydrazide analogs (C56) with strong potency showed to be competitive inhibitors as well but the most potent of the series showed to be noncompetitive.^{[ 239 ]} Interestingly, however, in a series of acyl hydrazones with low activity, some of the compounds were able to significantly activate urease.^{[ 240 ]} An overview of the activity profile of azo, hydrazone, and hydrazide analogs is provided in Figure  10 .

Figure 10.

Overview of chemical structures of azo, hydrazone, and hydrazide analogs and their range of urease inhibitory activities. A darker color indicates higher frequency of compounds in each activity bin.

However, this group shows low potential for biological applications. In comparison to previous classes, they are not as potent and this type of compounds is prone to hydrolysis and to targeting other enzymes.^{[ 241 ]} Nevertheless, even for moderate inhibitors, they can show antibacterial activity at concentrations close to known antibiotics.^{[ 206 , 242 , 243 ]}

4.7. Natural Products

Natural product‐derived compounds have long been a pivotal source of inspiration in medicinal chemistry due to their unique structural diversity paired with the fact that compounds of interest are obtained from natural sources. Numerous plant extracts and isolated bioactive phytochemicals have been tested against ureases, offering a wealth of potential lead compounds. Among these, several classes of scaffolds have been tested including alkaloids,^{[ 244 , 245 ]} (e.g., epiglucoisatisin), terpenes and terpenoids,^{[ 246} , ²⁴⁷ , ²⁴⁸ , ²⁴⁹ , ²⁵⁰ , ²⁵¹ , ²⁵² , ^{253 – 254 ]} flavonoids,^{[ 255} , ²⁵⁶ , ²⁵⁷ , ^{258 – 259 ]} stilbenes, xanthones, sphingolipids, and steroids^{[ 260 ]} (some compounds can be observed in Figure  11 ). While detailed characterization of all these natural products is beyond the scope of this review, comprehensive analyses are available elsewhere.^{[ 261 , 262 ]}

Figure 11.

Examples of isolated natural products and mechanism of inhibition. A darker color indicates higher frequency of compounds in each activity bin.

Many different extracts from plants and fruits have been evaluated for urease inhibition, exhibiting a wide range of responses.^{[ 263} , ²⁶⁴ , ^{265 – 266 ]} Some extracts demonstrate promising activity comparable to known inhibitors such as thiourea^{[ 252 , 267} , ^{268 – 269 ]} and acetohydroxamic acid.^{[ 270 ]} Due to their complex mixtures, a single extract can contain compounds that act both as competitive, noncompetitive, and uncompetitive inhibitors.^{[ 271} , ^{272 – 273 ]} Often an advantage of extracts may be the multitude of compounds that can act in multiple pathways since they can inhibit urease but can also inhibit bacteria via other means.^{[ 274} , ²⁷⁵ , ^{276 – 277 ]} For example, isolated abruquinone A and B exhibited moderate activity against urease but also exerted antimicrobial activity via induction of morphological damages and membrane segmentation in bacteria.^{[ 278 ]} However, this property can also cause unwanted effects as they can also inhibit other essential enzymes.^{[ 279} , ^{280 – 281 ]} This lack of specificity greatly limits their therapeutic potential. For example, gossypol, a natural polyphenol from Gossypium species, exhibited inhibitory activity against urease comparable to acetohydroxamic acid and showed MIC values similar to metronidazole and levofloxacin. However, gossypol is known for its toxicity, which poses significant limitations.^{[ 282 ]}

The diversity and variation in the concentration of compounds within extracts make it challenging to attribute observed activities to specific constituents.^{[ 283 ]} Additionally, as an example, extracts from Zanthoxylum nitidum, traditionally used to treat traumatic injuries and gastrointestinal disease, exhibited mixed‐type inhibition against H. pylori urease but noncompetitive inhibition against jack bean urease, with thiol groups playing a significant role in the inhibitory activity.^{[ 284 , 285 ]} Similar observations were also obtained from isolated compounds from such extracts.^{[ 286 ]}

Generally, described isolated natural products exhibit moderate to weak urease inhibition activity compared with thiourea.^{[ 253 , 279 , 287} , ²⁸⁸ , ²⁸⁹ , ^{290 – 291 ]} These compounds are often chiral, and chirality greatly influences inhibitory activity.^{[ 292 , 293 ]} As mentioned, they often also inhibit a broad range of other proteins and enzymes such as acetylcholinesterase, tyrosinase, α‐chymotrypsin^{[ 245 ]} and phosphodiesterase are often screened as well.^{[ 294 ]} Moreover, significant concentrations are typically required to have an impact on MIC against bacteria.^{[ 295} , ^{296 – 297 ]} Some compounds have also been described to target urease by covalently binding to cysteine residues. Sulforaphanes (M60) and isothiocyanates, commonly found in broccoli, are examples of such compounds.^{[ 298 ]} Conversely, allicin, commonly found in garlic, readily diffused into bacteria and was able to inhibit urease and to inhibit biofilm development in P. mirabilis at sub‐MIC concentrations.^{[ 299 ]} Baicalin (M49), a flavone glucuronide, was found to be a competitive, slow‐binding, and concentration‐dependent inhibitor of both JBU and HPU,^{[ 258 ]} and has also shown to target the thiol group in the active site.^{[ 261 ]}

An effective approach for improving the potency of natural compounds is structural modification. For example, 11 natural compounds and 19 synthetic de rivatives (modified from the first set) were screened against H. pylori urease and resulted in one compound exhibiting nanomolar range activity with over 160‐fold more potency than its parent compound,^{[ 300 ]} highlighting the potential of natural products as scaffolds for further development. Modification of flavonoids has also been shown to improve activity.^{[ 301 ]} Synthetic xanthene‐based analogs exhibited potent inhibitory activity with the most potent acting as a noncompetitive inhibitor (K _i : 40 nM).^{[ 302 ]} Similarly, curcumin (M58) showed similar activity to thiourea but analogs containing a diazine ring showed improved urease inhibition overall and the most potent compounds were found to be competitive or mixed‐type inhibitors.^{[ 196 ]}

4.7.1. Phenolic Compounds

Phenolic compounds are a diverse class of naturally occurring molecules characterized by the presence of one or more hydroxyl groups attached to an aromatic ring. These compounds are widely distributed in natural products and have a range of biological properties including chelation of metals and as such have garnered significant attention as urease inhibitors.

Phenolic substitutions are extensively represented in several scaffolds of urease inhibitors. The pattern of activity across the phenolic substitutions seems to be however similar (Figure  12 ) apart from the ortho‐substituted catechol (C60) that tends to yield more potent analogs, on average. In contrast, pyragallol substituents seem to be weaker in direct comparison (C62, C63),^{[ 170 , 178 , 303} , ³⁰⁴ , ³⁰⁵ , ^{306 ]} with only a few examples showing improved potency.^{[ 307 , 308 ]}

Figure 12.

General patterns of activity associated with phenolic substitutions. A darker color indicates higher frequency of compounds in each activity bin.

Phenolic substitutions across various chemical scaffolds are in general associated with enhanced inhibitory activity against urease. For instance, this type of substitution resulted in several examples where the most potent analogs were obtained, which include benzofuran‐based hydrazones (K _i : 200 nM),^{[ 234 ]} thiosemicarbazine (IC₅₀: 127 nM),^{[ 134 ]} phosphonates (K _i 320 nM),^{[ 27 ]} indoyl‐based hydrazone (IC₅₀: 600 nM), benzohydrazines carbothioamide (IC₅₀: 600 and 800 nM),^{[ 152 ]} triazunoindole,^{[ 235 ]} and other hydrazones.^{[ 308 ]} Conversely, methylation does not generally appear to affect inhibitory activity overall.

As discussed previously, catechol is found in many natural products and is a known mixed‐type urease inhibitor since the 1970s.^{[ 187 ]} It is able to coordinate with transition metals but it also forms a covalent bound with cysteine residues in the flap region believed to occur through a multistep, time‐dependent oxidation to ortho‐benzoquinone via dissolved oxygen that further reacts to benzoquinone and then cysteine.^{[ 309 ]} The formation of this covalent bond was used to design a hybridization strategy with a phosphonate analog to further extend the interactions and fill the pocket.^{[ 27 ]} However, this compound was found to be a reversible inhibitor (even if durable complexes are formed) and showed limited activity against P. mirabilis.

Similarly, other polyphenolic compounds such as epigallocatechin^{[ 303 ]} and quercetin have been suggested to inactivate urease by reacting with cysteine,^{[ 51 ]} while kaempferol analogs seem to complex with Ni(II) ions. Another example were two aliphatic esters of 3,4‐dihydroxyphenylacetic acid analogs connected to catechol with submicromolar activity (IC₅₀: 518 and 667 nM).^{[ 305 ]} They exhibited irreversible inhibition, but the majority of compounds only demonstrated moderate to weak inhibition. Interestingly, the kinetics of chemical reactivity towards thiols was similar between different analogs which suggests that the higher potency may also be related to specific interactions.

Polyphenols have also shown to remain active against H. pylori. ^{[ 304 , 307 , 310 , 311 ]} Despite being weak inhibitors as they were only half as effective as NBPT against soil urease,^{[ 187 ]} among various phenolic compounds catechol showed significant inhibition of soil ureases.^{[ 189 , 305 ]}

4.7.2. Flavonoids

Flavonoids are a diverse group of polyphenolic compounds ubiquitously found in nature and are common in many medicinal chemistry campaigns. Flavonoids are divided into several subclasses including flavones, flavonols, flavanones, flavanols, isoflavones, neoflavonoids, and flavenes, each differing in oxidation and pattern substitutions. These compounds are well‐known for their broad spectrum of biological activities and have been extensively tested against ureases^{[ 255 , 294 , 300 , 303 , 312} , ³¹³ , ³¹⁴ , ³¹⁵ , ^{316 – 317 ]} and have been extensively reviewed by others.^{[ 318 ]} Flavonoids in general exhibit moderate to weak activity and have broad activity against urease (Figure  13 ). However, a major limitation may be their promiscuity to other targets.^{[ 319 , 320 ]} Among the different classes of natural and synthetic flavonoids, flavones (C73), and flavonols (C74) seem to include more potent analogs.^{[ 276 , 287 , 300 , 321 , 322 ]} However, the most active flavones are hybrids of other known active chemical groups such as hydrazide^{[ 301 ]} and hydroxamic acids,^{[ 323 ]} and thus the activity is likely due to these instead.

Figure 13.

General patterns of activity associated with flavonoids. A darker color indicates higher frequency of compounds in each activity bin.

Among the natural flavonoids, both kaempferol and quercetin have shown to be noncompetitive urease inhibitors and demonstrate some antibacterial activity against H. pylori despite being much less potent against HPU.^{[ 294 , 317 , 324 ]} However, both of these molecules have toxicity issues related to carcinogenicity and mutagenicity.^{[ 318 ]} Similar to these flavonoids, morin was shown to be a moderate noncompetitive inhibitor,^{[ 325 ]} while 7,8‐dihydroxyflavone was shown to be competitive^{[ 326 ]} and some catechin analogs to be mixed‐type inhibitors.^{[ 327 ]} It is clear that the pattern of hydrogen‐bonding in the A and C rings dictates the activity,^{[ 255 , 300 , 317 ]} but for kaempferol, measurements by isothermal titration calorimetry show that the hydrophobic forces were dominant in driving the interaction with the protein target.^{[ 324 ]}

Glycosylflavonoids have also been described (part of C75), but these are typically weak inhibitors^{[ 259 , 317 , 326 , 328 ]} with some analogs having been described as competitive inhibitors.^{[ 258 , 315 , 326 , 329 ]} For example, glycosyl analogs of quercetin are 1.5–2.5‐fold less potent.^{[ 330 ]} Interestingly, hesperetin‐7‐rhamnoglucoside, a competitive inhibitor (K _i : 27.8 μM), was shown to also induce membrane disruptions in H. pylori showing the potential synergism of a single compound targeting two different biological targets.^{[ 315 ]}

Isoflavonoids are also weak in activity (C76) but the very structurally similar 4′,7,8‐trihydroxyl‐2‐isoflavene was shown to be a strong inhibitor of HPU (K _i : 641 nM) and to be a competitive type inhibitor.^{[ 300 ]} This was a surprising level of activity as the corresponding flavone (4′,7,8‐Trihydroxyisoflavone) exhibited an IC₅₀ of 140 μM suggesting that the absence of the hydroxyl in the B ring might be a key to improve activity.^{[ 310 ]}

Chalcones (C80), abundantly found in natural products, are a prominent group of natural compounds belonging to the flavonoid family and this scaffold is also implicated in a plethora of biological activities. Against urease, however, these compounds show only moderate to weak activity^{[ 331} , ³³² , ^{333 – 334 ]} and, among those tested, showed to be competitive inhibitors.^{[ 122 ]}

4.7.3. Coumarins

Coumarins are an important scaffold in medicinal chemistry, playing an important role in urease inhibition,^{[ 291 , 297 ]} and have been extensively reviewed.^{[ 71 , 335 ]} Naturally occurring coumarins exhibit a wide range of biological activity.^{[ 336 ]} Over 300 unique molecules containing the coumarin group have been tested, ranging from very potent urease inhibitors to very weak ones (C81, including single and bis‐coumarins) (Figure  14 ). These compounds have also shown inhibitory effects against H. pylori.^{[ 71 , 337 ]} While coumarin itself exhibits minimal MIC inhibitory activity against H. pylori, its derivatives 3‐hydroxy and 4‐hydroxy showed inhibitory activity.^{[ 314 ]} Conversely, its thiol analogs based on 1‐thioisocumarin did not exhibit urease inhibitory activity.^{[ 338 ]}

Figure 14.

General patterns of activity associated with coumarins. A darker color indicates higher frequency of compounds in each activity bin.

Similar to the observations with flavonoids, the oxidation pattern significantly influences their urease inhibitory activity. For example, 4‐hydroxycoumarin exhibited moderate inhibition with an IC₅₀ of 58.94 μM comparable to the control acetohydroxamic acid (IC₅₀: 44.64 μM) against HPU^{[ 314 ]} but not against JBU.^{[ 339 ]} In contrast, 7‐hydroxycoumarin (umbelliferone) showed lower activity than thiourea.^{[ 254 ]} Furthermore, small substitutions of coumarins both at the 3′ and 4′ position as well as the phenyl ring generally resulted in moderate to negligible activity.^{[ 291 , 314 , 337 , 340} , ³⁴¹ , ^{342 – 343 ]}

Among the most potent analogs containing coumarin are the previously discussed hybrids of aroylthioureas (C83)^{[ 110 ]} and the structurally similar pyrazolinyl analogs (C84).^{[ 111 ]} For the former, the most potent inhibitor (K _i of 0.32 nM) was shown to be a noncompetitive inhibitor,^{[ 110 ]} and the coumarin fragment seems to contribute significantly to the potency of aroylthioureas. In a similar manner, the pyrazolinyl analog (K _i : 0.26 nM) showed to be a noncompetitive inhibitor, but any substitution of the phenyl ring resulted in loss of activity.^{[ 111 ]} Closely related to these compounds, the thiosemicarbazones and hydrazide analogs showed to be strong to moderate inhibitors (C85)^{[ 344} , ³⁴⁵ , ^{346 – 347 ]} and one analog, ((E)‐N′‐(4‐hydroxybenzylidene)‐2‐oxo‐2H‐chromene‐3‐carbohydrazide), showed to activate urease instead.^{[ 346 ]} Other structurally similar analogs featuring thiazolo[3,2‐b[1,2,4] triazole bridges also showed moderate to weak activity,^{[ 348 ]} and a series of iminothiazolidinone analogs exhibited moderate activity against urease.^{[ 349 ]}

Bis‐coumarins (C82) which are also naturally occurring compounds have demonstrated extensive biological activity, including urease inhibition. Several glycosides of bis‐coumarins extracted from Daphne oleoids showed moderate activity against both BPU and JBU.^{[ 350 ]} Additionally, synthetic bis‐4‐hydroxycoumarins have also been tested against urease. The compound 3,3^′‐methylenebis‐(4‐hydroxycoumarin), also known as dicoumarol, was shown to be a moderate inhibitor with IC₅₀ ranging from 13.30–15.01 μM and UV–vis spectroscopy provided evidence of coordination with the nickel cluster.^{[ 339 ]} This is in agreement with observations that several analogs were shown to coordinate with nickel in solution.^{[ 84 ]} Over 90 analogs modified at the C‐11 carbon that bridges both coumarins have been reported. Simple modifications with various alkyl or aryl substitutions generally led to inhibitors of moderate to weak activity^{[ 84 , 351} , ^{352 – 353 ]} with only the 2′,3′,4‐trihydroxyphenyl showing improved activity (IC₅₀: 4.4 μM).^{[ 351 ]} Several of these analogs were studied for their mode of action and were all competitive inhibitors against JBU but about half were noncompetitive and uncompetitive against BPU. The analysis of these results suggests that it is likely that the C‐11 substitutions may result in steric hindrance. However, further expansion of the phenyl ring as in substituted phenyl‐1,3,4‐thiadiazoles led to some increase in potent inhibition with the most potent analog exhibiting an IC₅₀ of 120 nM. This enhancement indicates that appropriate extensions and substitutions can overcome the probable steric challenges by improving interactions.

4.8. Quinolones, Quinolines, and Quinazolinones

Quinolones, and particularly fluoroquinolones (C88), are a class that have emerged as a promising candidate for urease inhibition, offering potential expansion of the therapeutic application beyond their well‐established antimicrobial activity of inhibition of DNA synthesis.

It is evident that extending the fluoroquinolone scaffold significantly improves activity^{[ 354} , ³⁵⁵ , ^{356 – 357 ]} (Figure  15 ). For instance, an N‐acetyl analog of ciprofloxacin slightly improved the activity compared with ciprofloxacin itself (IC₅₀: 3.50 μM),^{[ 331 ]} and oxadiazole^{[ 358 ]} and N‐thioacetylated^{[ 359 ]} analogs exhibited similar activity to that of the first analog. On the other hand, expanding the piperazine ring with benzenesulfonamide resulted in a potent series of analogs, with the most potent showing mixed‐type inhibition and a K _i of 16 nM.^{[ 107 ]} Conversely, expanding the piperazine ring of sparfloxacin with a benzoyl led to weak inhibitors, although the inhibitory activity against gram‐positive and gram‐negative bacteria was maintained.^{[ 360 ]} Similar observations were made when the piperazine ring was removed.^{[ 361 ]}

Figure 15.

General patterns of activity associated with quinolones, quinolines and quinazolinones. A darker color indicates higher frequency of compounds in each activity bin.

The structurally similar 2‐hydroxyquinoline (C89) has also shown strong to moderate inhibitory activity.^{[ 362} , ³⁶³ , ^{364 – 365 ]} However, the core structure may not significantly contribute to the activity, as the most potent analogs are hybrids incorporating known active groups such as thiosemicarbazide and 4‐thiazolidinone. Similarly, the structurally related quinoline‐4‐ol (C90) exhibits activity patterns similar to those of 2‐hydroxyquinoline (C89).^{[ 366 ]}

A less frequently discussed scaffold for urease inhibition, yet structurally similar to this class, is quinazolinones (C91).^{[ 45 , 345 , 367} , ^{368 – 369 ]} Although the tested compounds are also hybrids of hydrazinecarbothioamide, they still display strong to moderate activity across a considerable number of compounds. However, hybrids of nitrothioazolacetamide examined for the mechanism of action were found to be uncompetitive inhibitors.^{[ 45 ]}

4.9. Aliphatic Chain‐Bearing Compounds

Compounds with an aliphatic chain have emerged as an interesting class of urease inhibitors, offering a unique structural feature that contributes to inhibitory activity. Acyl chains are important elements in drug discovery and cosmetics and in particular for their inherent antimicrobial activity. It has been reported that various free fatty acids including lauric acid, myristoleic acid, linoleic acid, and linolenic acid have broad antibacterial activity that includes H. pylori.^{[ 370 ]} A major interest is that these compounds are commonly found in natural products. For example, the sphingolipids from Heliotropium ophioglossum (Ophiamides A–B, cinnamic acid tetracosyl) showed to have activity similar to that of thiourea^{[ 371 ]} and palmitoyl sitosterol occurring naturally in a plant exhibited moderate activity.^{[ 372 ]}

Over 500 unique compounds with varying chain length have been reported and, while they generally show moderate‐to‐strong activity (Figure  16 ) and are usually noncompetitive inhibitors,^{[ 142 , 373} , ^{374 – 375 ]} some hybrid analogs have shown potent activity. After analysing the many different scaffolds within this subset, it appears that, generally, potency is strongly favored with acyl chains of 13 or more carbons. Nevertheless, it is clear that beyond the chain length, specific features of the rest of a given scaffold also determine the activity.^{[ 376 , 377 ]}

Figure 16.

General patterns of activity associated with aliphatic chains of different size. A darker color indicates higher frequency of compounds in each activity bin.

Indeed, for the thiazolidine scaffold, the C7 length was found to be best for activity which significantly dropped for higher lengths.^{[ 378 ]} On the contrary, small chains had a negative impact in bis‐coumarins,^{[ 84 , 339 ]} hydroxycoumarins,^{[ 340 ]} benzophenone sulfonamides,^{[ 205 ]} and hydroxamic acids.^{[ 233 ]}

For hydroxy fatty acids, the position of the hydroxyl groups strongly correlates with activity as exemplified by a series of monohydroxyeicosanoic acid isomers where the location of the hydroxyl group in the middle of the chain showed the best inhibitory activity and the most potent compound also showed stronger ability to chelate Cu²⁺.^{[ 379 ]} Similar observations for the hydroxyl location within the alkyl chain were also observed for a series of monohydroxy tetradecanoic acid isomers.^{[ 380 ]}

Notably, as expected among the potent inhibitors are the acylthiourea analogs comprising long acyl chains (13C), and the most potent compound was found to be a noncompetitive inhibitor (K _i : 21 nM).^{[ 375 ]} These are however not the most active in the class of acylthioureas and even other hydrophobic substitutions have shown to be more potent.^{[ 108 , 110 , 151 ]} Nevertheless, for this type of inhibitors, both the chain length and substitutions of the phenyl ring dictated the activity and bigger chain lengths are associated with more potent inhibitory activity.^{[ 112 , 113 , 140 , 141 ]}

In the case of organophosphorous however, the addition of acyl chains also does not seem to improve activity, and despite some potent analogs have been described, there are several that significantly show a loss in activity.^{[ 209 , 212 , 219 , 220 ]}

4.10. Boronic Acids

Boronic acids have played a crucial role in advancing our understanding of urease inhibition mechanisms. Boric acid itself is a rapid, reversible and competitive inhibitor of urease.^{[ 381 ]} Its inhibitory activity is highly pH‐dependent, correlating with its pKa values around 6.2 and 9.3. Studies suggest that the neutral trigonal form of boric acid B(OH)₃ is the active inhibitory species, rather than the tetrahedral B(OH)₄ ⁻ anion.^{[ 382 ]} This is supported by observations that urease inhibition by boric acid is maximal at acidic pH and minimal under alkaline contitions. Both boric acid (K _i : 0.33 mM) and phenyl borate (K _i : 11.5 mM) were found to be competitive inhibitors of KAU, but boric acid was much more active against JBU (K _i : 0.19 mM) and PMU (K _i : 0.099 mM).^{[ 382 ]} Besides urease inhibition, 4‐bromobenzeneboronic acid (K _i :0.124 mM) was also shown to be able to inhibit the growth of K. pneumoniae.^{[ 67 ]}

4.11. Metal Ions and Organometallic Compounds

Replacement of the metal ions of the active site has been a viable strategy to modify urease activity, as early studies have shown that exchanging with other metals resulted in complete loss of activity.^{[ 383 ]} This simple strategy can be effective and urea fertilizers coated with Cu and Zn have shown to improve nitrogen uptake of the soil.^{[ 72 ]} Additionally, several metals have shown to react with sulfhydryl groups, thus interfering with the mobile flap region of the active site as in the case of silver (PDB 6G48) and gold (PDBs 7B58, 7B59, 7B5A, 6I9Y, 7P7N, 7P7O). Among the various metals shown to have inhibitory activity against urease these include: Ag⁺, Cu⁺, Pb⁺, Na⁺, K⁺, Ca⁺, Ba²⁺, Hg²⁺, Cu²⁺, Ni²⁺, Cd²⁺, Zn²⁺, Co²⁺, Fe²⁺, Pb²⁺, Mn²⁺, Cu²⁺, Bi³⁺, Au³⁺, As³⁺, Cr³⁺, Al³⁺, and Fe³⁺.^{[ 189 , 384 , 385 ]}

However, a major limitation of metals is their toxicity which limits their use. A common strategy to improve selectivity is complexation of these metals with organic molecules. It has been proposed that the selectivity coming from the scaffold itself is improved by pre‐organizing the scaffold with the metal to fit more effectively in the pocket,^{[ 52 , 386 ]} in a way acting as a drug delivery system.^{[ 387 ]} This is observed with ligands that have no inhibitory effect by themselves but become potent inhibitors once complexed with metals, improving the potency of the metal itself too. Common ligands include classes of metal chelator groups such as the known carboxylates,^{[ 386 ]} hydrazones, Schiff bases,^{[ 384 , 388} , ^{389 – 390 ]} benzohydrazides,^{[ 391 , 392 ]} semicarbazide, and thiosemicarbazones, and often these complexes display a wide range of activities.^{[ 138 , 386 , 393} , ³⁹⁴ , ³⁹⁵ , ³⁹⁶ , ^{397 – 398 ]} Moreover, it is generally observed that metal complexes consistently outperform the free ligand in terms of inhibitory activity.^{[ 391 , 399} , ⁴⁰⁰ , ^{401 – 402 ]} However, despite this increased activity^{[ 138 ]} and considering the fact that potent groups such as coumarins and thiosemicarbazones are present, the corresponding metal complexes for each scaffold are not among the most potent analogs.

Among the different metals, copper‐based organic metal complexes by far seem to be consistently the most potent combination^{[ 394 , 403 , 404 ]} with very potent analogs having been described,^{[ 393 ]} while iron analogs have shown low potency.^{[ 384 , 388 ]} The inhibitory mechanism of these complexes is also dependent on the ligand. In cage‐like triazole‐5‐thione hybrids of hydrazones complexed with different metal ions (Cu²⁺, Ni²⁺, Pd²⁺, Co²⁺, Fe²⁺, and Zn²⁺), the most potent inhibitor was the Cu²⁺ (noncompetitive inhibition)^{[ 405 ]} while heterocyclic amino pyridine complexed with copper resulted in a strong inhibitor (IC₅₀: 340 nM, mixed‐type inhibition).^{[ 406 ]} In another example, thiophene displayed more potency than its oxygen counterpart but the activity was only weak.^{[ 407 ]} Finally, Schiff bases complexed with Cu²⁺ showed only strong activity in the mononuclear form but lost activity in the dimer form.^{[ 408 ]}

4.12. Heterocycles

Heterocycles encompass a wide range of families of scaffolds (Figure  17 ) that are often less discussed as they are usually hybrids and therefore their individual activity is hard to evaluate. Unpacking this information however provides very useful insights. As an example, when analyzed in isolation, pyridines (C117) seem to be associated with strong activity but when analyzed for each individual scaffold, this group seems to decrease the activity in the majority of the cases. Some others, on the other hand, have been previously discussed as exhibiting notorious activity such as is the case of conjugated coumarins.

Figure 17.

General patterns of activity associated with scaffolds containing different heterocycle rings. A darker color indicates higher frequency of compounds in each activity bin.

Aza‐heterocycles include a large set of classes such as diazoles, triazoles, tetrazoles, oxazoles, thiazoles, and pyridines.^{[ 409 ]} These are known to complex with metal ions and imidazoles are interesting moieties as they behave as borderline bases (HSAB theory) and therefore should bind to borderline acids such as Ni²⁺. Indeed, imidazoles and benzimidazoles (C108) are scaffolds positively associated with activity against urease. Benzimidazoles are important heterocycles in drug discovery and have biological activity against H. pylori.^{[ 410 ]} Several potent inhibitors have been reported including bulky compounds such as tri‐ and tetra‐arylimidazole.^{[ 154 , 235 , 376 , 411} , ^{412 – 413 ]} However, as with many of the scaffolds discussed here, its contribution to the overall activity is still difficult to assess.^{[ 411 , 414 , 415 ]} For example, in the case of thiosemicarbazides, carbothioamides, and urea analogs, the corresponding benzimidazole exhibited only moderate to weak activity.^{[ 134 , 186 , 416 ]} Benzimidazoles also have the potential to complex with metal ions with high inhibitory potency.^{[ 393 ]} For instance, gold complexes of 1‐methy‐2‐(pyridine‐2‐yl)‐benzimidazole were effective (IC₅₀: 9 nM) at covalently modifying the cysteine in the mobile flap region (PDB 6I9Y).^{[ 92 ]} Comparatively indole (C109) exhibits a similar range of activities,^{[ 123 , 186 , 203 , 230 , 344 , 417} , ⁴¹⁸ , ^{419 – 420 ]} but the thiobarbituric acid analogs exhibited lower potency.^{[ 39 , 176 ]}

Similarly, for oxindole (C112), a similar distribution pattern of activity^{[ 137 , 186 , 420} , ⁴²¹ , ⁴²² , ^{423 – 424 ]} is observed but the scaffold itself seems to be essential as 3‐substitutions that include Schiff bases analogs showed potent inhibitory activity,^{[ 422 ]} and these analogs showed antimicrobial activity against H. pylori as well.^{[ 420 ]} Benzothiazoles and benzothiazolones (C110) display similar levels of activity to benzimidazoles and indoles^{[ 66 ]} with potent inhibitors of (diaminophosphoryl)amines having been described (IC₅₀: 2–10 nM), and being the most potent of this class.^{[ 219 ]} Conversely, benzothiazoles of benzoylthioureas resulted in loss of activity compared to the phenyl analogs.^{[ 143 ]} However, 1,2‐benzisothiazo‐3‐one exhibits potent inhibitory potency against both HPU (IC₅₀: 55 nM) and JBU (IC₅₀: 132 nM),^{[ 425 ]} but further N‐phenyl substitution resulted in loss of potency.^{[ 121 ]} On the other hand, the structurally similar 1,2‐benzoselenazol‐3‐one behaves in an opposite way to the N‐phenyl analog, exhibiting higher potency (K _i : 2.11 nM vs. 24.70 nM). This class (C111) has exhibited potent levels of activity and its activity is related to the formation a covalent bond with cysteine. Interestingly, even though dibenzyldiselenide itself does not have inhibitory activity,^{[ 121 ]} various diselenide analogs have shown potent activity.^{[ 426 ]}

Among triazoles, 4‐amino‐4H‐1,2,4‐triazole‐3‐thiol (C113) shows moderate activity overall,^{[ 348 ]} but a series of potent analogs (K _i : 1.8–24 nM) bearing this scaffold have been reported to inhibit via noncompetitive and mixed‐type inhibition.^{[ 111 ]} The amino‐triazole substructure is also observed in 1,2,4‐triazolo[3,4‐a]phthalazine that was shown to be a competitive inhibitor with an IC₅₀ of 320 nM.^{[ 427 ]} The fused 3,6‐disubstituted 1,2,4‐triazolo[3,4‐b]1,2,4‐thiadizole analogs exhibited consistently strong to moderate activity (C114)^{[ 428} , ^{429 – 430 ]} but were also able to inhibit acetylcholine and alkaline phosphatase, which raises concerns of low selectivity. The substructure 1,3,4‐thiadiazole (C115) showed similar levels of activity,^{[ 336 ]} and thiophene also exhibits moderate activity.^{[ 109 , 229 ]} The other structurally similar analog ring 1,3,4‐oxadiazole (C116) including 1,3,4‐oxadiazole‐2‐thione shows a broader response with strong to weak compounds having been reported.^{[ 345 , 377 , 431} , ⁴³² , ⁴³³ , ^{434 – 435 ]}

The structurally similar quinazolines (C118),^{[ 45 , 345 , 365 , 368 , 369 ]} pyrimidin‐4‐ol (C119),^{[ 166 , 196 , 208 , 436 ]} quinolin‐4‐ol (C120),^{[ 366 ]} and 1,2‐dihydroquinolin‐2‐one (C124)^{[ 362} , ⁴⁶³ , ^{464 – 365 ]} all exhibit similar levels of activity. On the other hand, 1,2,4‐triazine (C123)^{[ 200 , 235 , 437 ]} exhibits a broader range of activities with some potent analogs described, particularly when hybridized with sulfonamides.^{[ 200 ]} Tiophenes (C121) also displays strong to moderate activity, and several analogs of thioacetohydrazide and 1,3,4‐oxadiazole‐2‐thiol and 1,3,4‐oxadiazole‐2(3H)‐thione exhibit strong activity.^{[ 438 ]} Finally, carbazole (C125) despite present in several structures does not contribute significantly to activity as it is only associated with moderate to weak activity and its activity generally comes from association with hybridization with other groups (e.g., coumarin).^{[ 111 , 439} , ^{440 – 441 ]}

4.13. Covalent Binders to the Flap Motif

Covalent inhibition still remains controversial in drug development. In this approach molecules form covalent bonds with the target protein and lead to irreversible suppression of its biological activity with potential long‐lasting effects (sometimes with complete inactivation of the target). They have the potential to use lower doses but often are considered less attractive due to potential drawbacks from off‐target effects due to the unpredictability of modification of other biomolecules. Indeed, lower selectivity can lead to general toxicity and immunogenicity which are great concerns.

Urease has several cysteine residues per subunit but only a few are accessible to thiol reagents.^{[ 442} , ^{443 – 444 ]} As previously mentioned, ureases have at least 3 reactive cysteines but the one in the mobile flap is associated with most of the inhibitory capability of ligands and has been shown to be the most reactive. An ever‐growing number of potent molecules have emerged that target this cysteine in particular and block the mobility of the active site flap. This can be clearly observed in crystal structures of ureases and includes catechol, 2‐mercaptoethanol and Ag(1) and Au(1) (Figure 2).

A multitude of the classes have shown covalent binding to urease, including metals, quinones (C126), catechols (Figure 12), thiol‐containing compounds (C127 and C128),^{[ 67 ]} α,β‐unsaturated ketones (C129) aldehydes (C130), isothiocyanates (C131), diketones (C132), acetylene (C133),^{[ 445 ]} and selenoorganic compounds (C134). Alternatively, cysteine modification can be achieved through oxidation with potassium permanganate (KMnO₄).^{[ 67 ]}

Quinones (C126) are a class of organic compounds that also occurs naturally^{[ 278 ]} and the vast majority has only demonstrated weak activity at a single point concentration. Among those further explored, most have moderate activity,^{[ 446 , 447 ]} and only tetrachloro‐p‐benzoquinone and some diaminobenzoquinones have shown potent inhibitory activity.^{[ 448 ]} Benzoquinones however seem to have strong activity against soil ureases^{[ 449 ]} and are able to significantly reduce ammonia emissions from crops.^{[ 72 ]} They react with cysteine^{[ 450 , 451 ]} via the formation of a semiquinone radical.^{[ 93 ]} However, their lack of specificity still remains challenging as they can equally react with nucleophiles such as glutathione.

Several thiol‐based compounds (C127) including cysteamine, cysteine, and β‐mercaptoethanol have exhibited competitive inhibition via thiolate anion formation.^{[ 105 , 452 ]} However, the response to β‐mercaptoethanol seems to be species‐specific alluding to the importance of the small differences in the active site.^{[ 81 , 453 ]} Besides competing to the nickel ions forming a bridge via charge–transfer complex, β‐mercaptoethanol also binds to thiol groups. Moreover, N‐ethylmaleimidine commonly used as a titrating agent for determination of protein sulfhydryl groups was used to show that the increasing incubation times resulted in the decrease of urease activity until the total loss of catalytic activity.^{[ 443 ]} One of the interesting aspects about this type of compounds is that they are present in edible natural sources, with compounds such as iberin, sulforaphane, and allicin, showing moderate to weak activity.^{[ 381 , 454 ]} Disulfides also seem to retain the activity associated with thiol‐bearing compounds (C128). A series of 38 dithiobisacetamides tested against HPU showed nanomolar‐to‐micromolar range activity^{[ 455 ]} and were also effective in intact cells of H. pylori. The tested compounds were shown to be mixed‐type inhibitors, but the most potent inhibitor did not act through the cysteine modification.

Selenoorganic compounds (C134) with potent inhibitory activity are known to covalently bind to urease via cysteine.^{[ 116 , 426 ]} One of these analogs has been co‐crystallized with urease (Figure 2, PDB 7ZCY) where di‐selenium is found bound to the flap region, whereas the original ligand is covalently bound to another cysteine that is not in the flap region of urease.

Based on this approach, an interesting way to maximize the potential of inhibition has been the dual targeting of both cysteine and the nickel ions. This has been achieved by modifying organophosphates with catechol in a way to fill the pocket.^{[ 27 ]} This resulted in potent inhibitors (K _i : 130 nM) that were shown to work against P. mirabilis urease in cell assays. A similar strategy resulted in a potent compound (K _i : 509 nM) by conjugating phosphonic acid (binds to the nickel ions) with cinnamate (covalently binds to cysteine, assists hydrogen bonding via a carboxylic acid).^{[ 456 ]} This mediation of the carboxylic acid is also observed in acetylenedicarboxylic acid (IC₅₀: 42.5 μM),^{[ 445 ]} and it is thought to be mediated in a similar manner to what is observed for citrate–urease complex (PDB 4AC7).^{[ 445 ]}

4.14. Approved Drugs and Repurposed Compounds

When it comes to developing therapeutic applications, repurposing already approved drugs has become a very desirable route, as it facilitates the process of developing and approval a new drug product or agricultural compound.^{[ 457 ]} Drug repurposing has become very sought after, whereby new applications are identified for already established drugs, and it would be particularly advantageous to find molecules with known antibacterial activity that can also inhibit urease. Several approved drugs target different metalloenzymes, lending themselves to the expectation that they may inhibit urease due to similarities in metal‐binding mechanisms. However, many of these drugs are primarily used in anticancer therapies, which limits their applicability in the context of urease inhibition. Examples of drugs that bind to other binuclear metal centers include Baloxavir marboxil, NOHA, and 2(S)‐amino‐6‐norleucanoic acid that interact with manganese ions. Crisaborole targets zinc and magnesium, while kojic acid (M64) and hydroquinone bind to copper ions. However, despite their ability to chelate metal ions, it does not guarantee translation to urease as seen in the case of kojic acid (M64) that has no urease inhibitory activity.

Recently, a screen of 3,904 FDA‐approved drugs at 100 μM concentration revealed that multiple drugs could inhibit urease activity but only 5 were significantly more potent (4 to 800‐fold) than AHA and were further selected to further study.^{[ 121 ]} Among them, Panobinostat (M65, IC₅₀: 0.2 μM) and Dacinostat (M66, IC₅₀: 1.1 μM) which are metalloenzyme inhibitors of histone deacetylase (HDAC) were found to be competitive inhibitors of HPU. On the other hand, both ebeselen (M67, IC₅₀ of 0.4 μM) and captan (M68, IC₅₀: 2.3 μM) were found to be noncompetitive by covalent inhibition. Finally, disulfiram (M69, IC₅₀: 38.9 μM), a drug used as alcohol‐abuse deterrent, was found to be a reversible urease inhibitor that was dependent on the concentration of Ni²⁺ ions and likely acted by forming a complex with the catalytic metal center. However, others have reported the formation of diethylthiocarbamate and formation of covalent bonding similar to inhibition observed for acetaldehyde dehydrogenase (ALDH).^{[ 458 ]} In addition, these inhibitors demonstrated the ability to prevent H. pylori infections in a gastric cell‐based model.

Other drugs have also been tested against urease (Figure  18 ). Ropinirole (M70),^{[ 417 ]} a dopamine agonist, showed moderate mixed‐type inhibition. Captopril (M71),^{[ 459 ]} commonly used for hypertension, demonstrated moderate mixed‐type inhibition.^{[ 458 ]} Ethacrynic acid (M72), used for chronic heart failure, exhibits weak inhibitory activity, likely through binding to cysteine due to its α,β‐unsaturated carbonyl structure.^{[ 460 ]} Chloroxine (M73), used for seborreheic dermatitis, displayed moderate activity (IC₅₀: 6.18 μM).^{[ 461 ]} Finally, 4‐butyresorcinol (M74), a molecule used to treat hyperpigmentation, was identified as weak inhibitor of urease.^{[ 462 ]}

Figure 18.

Examples of drugs tested against urease and their mechanism of action.

Several studies have shown that urease inhibitors by themselves may not be able to inhibit bacterial growth. Therefore, the idea that an antibiotic could itself also be active against urease challenges this notion and broadens the horizon for urease inhibitors. The limitation of broad‐spectrum antibiotics is always the possible dysbiosis particularly in the treatment of H. pylori where a combination of multiple antibiotics is required. Therefore, if this effect can be limited with antibiotics that can also synergistically have urease activity, this could improve the therapeutic profile in urea‐dependent infections.

Cephalosporins are common antibiotics with a β‐lactam core and show strong activity against gram‐positive and gram‐negative bacteria. Different analogs were tested against urease (C135) but only displayed weak to moderate inhibitory activity of urease^{[ 437 ]} (Figure  19 ). For example, both ceftriaxone and cefotaxime were shown to be moderate irreversible competitive inhibitors but not as potent as thiourea.^{[ 437 ]}

Figure 19.

Activity against urease of several analogs of drugs. A darker color indicates higher frequency of compounds in each activity bin.

Metronidazole and secnidazole are antibiotics that have been tested for their urease activity but showed weak to no inhibitory activity.^{[ 463 , 464 ]} However, several extensions of this scaffold have been shown to yield analogs with potent to moderate activity (C137).^{[ 464 ]} As previously discussed, quinoline antibiotics (Figure 15) such as levofloxacin and ciprofloxacin have shown promising activity against urease.^{[ 331 , 354 , 361 , 465} , ⁴⁶⁶ , ^{467 – 468 ]} Schiff bases of amikacin have also been tested and showed moderate activity against urease.^{[ 469 ]}

Another class of drugs that would be useful as they are related to H. pylori treatment is the gastric proton pump inhibitors (PPIs). These are used as antiulcer agents by inhibition of acid secretion and are known to also inhibit H. pylori's growth. Several PPIs have shown strong to moderate activity (C136) against HPU both against the isolated enzyme and the whole bacterium.^{[ 470 , 471 ]} The activity of these compounds was dependent on the experimental conditions (e.g., more potent under acidic pH) and were species‐dependent. It is proposed that these compounds are transformed in the acidic pH and become active.^{[ 472 ]} In the case of omeprazole, it was found that it is a noncompetitive inhibitor by reacting with the cysteine of JBU, and similarly rabeprazole activity was completely reversed by addition of sulfhydryl‐containing compounds.^{[ 471 ]}

Similarly, bismuth is also used in quadruple therapy regimen for the treatment of H. pylori.^{[ 473 ]} Being a borderline metal ion, Bi(III) is known to have affinity to thiols. Several bismuth complexes have shown antibacterial activity that has been associated to urease inhibition.^{[ 474 ]} For example, bismuth complexes of β‐mercaptoethanol were more potent than β‐mercaptoethanol.^{[ 475 ]} Similar to observations with other metal complexes, the ligand dictates the mechanism of inhibition as weak inhibitors Bi(EDTA) and Bi(Cys3) inhibited via competitive mechanism while ranitidine bismuth citrate was shown to be a noncompetitive inhibitor with likely binding to the cysteine.^{[ 474 ]}

Another interesting class of drugs that has been tested against urease is non‐steroidal anti‐inflammatory drugs (NSAIDs). While the activity of ibuprofen is limited,^{[ 160 , 476 ]} there are various hybrid analogs (C138) that have shown potent to strong activity^{[ 160 , 477 ]} with arylthiourea analogs being among the most potent urease inhibitors described.^{[ 151 ]} Analogs of other NSAIDs such as flurbiprofen,^{[ 306 ]} S‐naproxen,^{[ 478 ]} and indomethacin^{[ 479 ]} have also been tested and showed moderate to weak inhibitory activity.

Atenolol, a beta blocker medication, and atenolol analogs have been tested with the analogs showning moderately improved inhibitory activity when compared to the weakly active parent compound (C139).^{[ 480 ]}

5. Screening Campaigns, Docking, Molecular Dynamics, and Predictive Machine Learning

Screening is critical in the discovery process and significantly enhances the identification of novel inhibitors. As early as 1971, Brenner and Doubles screened 100 compounds against soil ureases.^{[ 189 ]} The ongoing need for new urease inhibitors for the different applications means that the continuous exploration through screening methods remains vital. Screening of random scaffolds, though inefficient, still remains an approach used to find interesting leads. However, virtual screening approaches have become a key component of many discoveries particularly saving time and resources. Both ligand‐based and structure‐based virtual screening campaigns have been popular to find urease inhibitors. Urease has been a good model for screening since urease structures from different species are well characterized, and there is an abundant amount of data including the catalytic mechanism and a diverse chemical diversity of inhibitors. Moreover, urease has been shown to be able to accommodate a wide variety of chemical structures making it amenable to screening libraries that include a diverse chemical space.^{[ 69 ]} A summary of several screening campaigns is described in Table  2 .

Table 2.

Summary of virtual screening campaigns.

Dataset	Technique	Notes	Ref.
Dataset in house library of compounds	Pharmacophore modeling using GASP. Docking for screening (PDB 4UBP)	Screening of 3000 compounds. 2‐Aminothiophenes were screened against JBU and BPU with best two 770 and 830 nM IC50. 11/13 urease inhibitors and 9/13 better activity than control acetohydroxamic acid (IC50 24.1 μM)	[170]
Dataset 647 compounds from literature to create predictive models. Screening with 59 in‐house derivatives	kNN, ASNN, XGBOOST, WEKA‐RF	Moderate inhibitors of thiazole‐2‐imine with moderate activity (IC50 44.9–49.4 μM) with accurate predicted activity (48–52 μM). Classification models: Balanced accuracy of 77–89%. Regression models: R 2 of 0.5–0.7	[486]
In house library to screening with 10,000 compounds	Docking with Fitness score from GOLD and MOE (PDB 4UBP)	Docking validated with bis‐coumarin, imidazole and hydroxamic acid. The best hit was monasterol (IC50 11.76 μM) and further analogs were synthesized with the best IC5 0 being 5.36 μM. Optimized to H‐bond with amino acids of the active site. 7/20 hits with moderate activity similar to thiourea.	[338]
ZINC dataset of 180,313	Docking Glide score	10 highest docking score were selected from different classes (ZBC, Drug, Natural product, in man, biogenic) and compared with literature. 21 compounds were already reported in the literature against HPU. Four selected compounds for experiments were inactive.	[51]
In house library of 90,000 compounds	Docking comparative interactions	Compounds were selected based on interacting with Nickel. Strong to moderate compounds were obtained 3.03–37.29 μM). Docking was compared with known inhibitors. 10 out of 10 compounds selected were inhibitors	[430]
63 N‐acylglycino‐ and hippurohydroxamic acids phamacophore; subset of ZINC (3000 drugs approved)	Pharmacophore and 3D‐QSAR with k‐means clustering. Docking (1E9Y)	Threshold of activity model of different classes of inactive, moderate and active. From screening one inhibitor (ascorbic acid) was found that was already reported as inhibitor. 3D QSAR model: From 86 molecules from the literature 21 were found as matches. Regression of 0.704 R 2	[491]
24 potential dipeptide hydroxamic acid inhibitors	Comparative molecular field analysis (CoMFA); Docking (HPU homology model)	24 dipeptides were used to construct a model and effectively predict the activity of 5 others. Cross validation correlation coefficient q2(100) f 0.610 and R 2 of 0.988	[228]
1253 phytochemicals screening NPACT	Monte‐Carlo method‐based QSAR model using SMILES and GRAPH descriptors. Docking (1E9Y)	436 molecules distributed for model creation (BindingDB). Inophyllum E, curcumin and (2S)‐2′‐methoxykurarinone were predicted as leads. r2 0.577–0.790. Predicted pIC50 close to a compound that was described in the literature.	[492]
870k compounds	Docking score (Ruminal metagenomic urease homology model)	20 were selected and 20 were weak inhibitors. 1 selected compound was mixed‐type inhibitor	[12]

Both pharmacophore modeling^{[ 481 ]} and docking^{[ 481 ]} have been used to a successful degree. Typically, docking score does not correlate with activity. For example, in a study of naproxen analogs, the R ² was found to be only 0.5847,^{[ 479 ]} but, for a series of benzimidazole, a correlation of 0.9079 was found.^{[ 482 ]} Recently, large‐scale docking of urease inhibitors against JBU also did not have a significant correlation with docking score, but it was observed that compounds with high docking score were likely to be inhibitors more potent than thiourea.^{[ 483 ]} Moreover, the docking poses were used to create a machine learning model that was able to classify if compounds were at least as active as thiourea. Other than docking score, the predicted H‐binding energy showed a correlation of R ² of 0.9481 with activity.^{[ 484 ]}

Large scale virtual screening has also led to finding novel compounds. Screening over 700k compounds from the ZINC database using docking against HPU led to finding some barbituric acid analogs with moderate activity. Despite being a well‐known class, it still exemplifies how screening was able to find inhibitors.^{[ 165 ]} In another example, 1776 compounds from plants were screened by molecular docking and resulted in the discovery coptisine as a new hit compound, which was confirmed to have strong activity (IC₅₀: 2.45 μM, K _i of 0.68 μM) and mixed‐type inhibition.^{[ 485} , ⁴⁸⁶ , ⁴⁸⁷ , ⁴⁸⁸ , ⁴⁸⁹ , ⁴⁹⁰ , ⁴⁹¹ , ⁴⁹² , ^{493 – 494 ]}

6. Functional Inhibition of Urea Conversion: Alternatives to Urease

Despite the significant interest in directly targeting urease, novel approaches have emerged that explore inhibitors.^{[ 495 ]} One promising alternative approach involves evodiamine,^{[ 496 ]} which downregulates urease subunit expression, thereby hindering initial bacterial colonization.^{[ 497 ]} Another strategy targets accessory proteins related to urease function, such as UreI, an inner membrane protein that acts as a H⁺ gated urea channel, regulated by external pH. Inhibiting UreI has been shown to disrupt urease activation by preventing urea from entering the bacteria,^{[ 16 , 55 ]} since urea has poor permeation across lipid membrane and requires transporters.

Considering urease is a nickel‐dependent metalloenzyme, regulating nickel availability and trafficking is tightly controlled and is essential for the survival of several pathogenic bacteria including H. pylori, Klebsiella, and Pseudomonas.^{[ 498 ]} Therefore, chelators like EDTA can sequester nickel, preventing its integration into urease active site. For instance, deferoxamine, an iron chelator, has been shown to limit nickel availability. Similarly, vitamin C (ascorbic acid) increases gastric acidity, potentially influencing nickel homeostasis by reacting with nickel ions and facilitating thiol oxidation in urease. Interestingly, citric and malic acids, though also acids, display different effects compared to ascorbic acid, suggesting a more complex mechanism involving the intracellular trafficking of nickel rather than simply opening urea channels.^{[ 499 ]} In addition, ascorbic acid's mechanism is thought to involve urease denaturation at low pH, while in neutral environments, it may target metal oxidation via thiol groups in the presence of Fe³⁺ and H₂O₂.^{[ 499 ]}

7. Lessons Learned: General SAR of Urease Inhibitors

Despite the wealth of crystal structure data for urease‐inhibitors complexes, extraction of meaningful SAR is limited as most reported complexes involve compounds with only moderate potency rather than the most active inhibitors. Moreover, while many inhibitors were designed to act as competitive agents, kinetics studies show that the most potent compounds frequently display mixed or noncompetitive inhibition. Importantly, potency alone does not fully determine the effectiveness of compounds, as stability is also a crucial factor.

Despite these challenges, several common SAR trends emerge across chemically diverse scaffolds, highlighting some converging pharmacophoric features. A consistent determinant of activity is the presence of functional groups capable of coordinating with the catalytic metal center. Thioureas (and carbonylthioureas), selenoureas, hydroxamic acids, and organophosphorous derivatives exemplify this principle, with the latter showing enhanced potency as they mimic the transition state.

Beyond metal chelation, the hydrogen‐bonding network is a critical contributor of potency. Activity depends not only on the presence of H‐bond donor/acceptors but also on their spatial orientation. This is particularly evident in the flavonoid class, where phenolic substitutions exert scaffold‐specific influences on activity (Figure 12). Substituent effects also play a role, and as also mentioned throughout this review, the chlorine substitution for a variety of scaffolds is not only crucial for potency, but its position/orientation was dependent on the scaffold itself. Moreover, a recent analysis further showed that activity correlates with an optimal number of hydrogen bonds (around 2–4), while both the absence of H‐bonds and excessive H‐bonding are detrimental to activity,^{[ 483 ]} emphasizing that geometry and balanced interactions are crucial.

Some of these hydrogen‐bonding and chelating interactions also extend to residues in the mobile flap region, where they can stabilize the inhibitor–enzyme complex and modulate flap flexibility, a key determinant of catalytic regulation. Strategies that exploit this include combining metal chelation with covalent modification of the flap cysteine. Although covalent modification typically yields inhibitors of moderate potency (Figure  20 ), recent studies with selenoorganic compounds have shown that this approach can produce highly potent inhibitors.

Figure 20.

General patterns of activity associated with compounds that react with cysteine. A darker color indicates higher frequency of compounds in each activity bin.

Finally, hydrophobic aliphatic chains and aromatic substituents adjacent to the chelating groups are a recurrent feature of potent inhibitors, particularly evident in the thioureas (Figure 5) and organosphosporous class (Figure 8). Interestingly, the most active compounds also tend to be relatively small, with an average molecular weight of ≈350 g mol⁻¹. Despite this suggesting that excessive bulk compounds may hinder optimal position, two of the most potent inhibitors are bulky bisdiphenylphosphoramides (C42) which achieve high activity through a mixed‐type inhibition mechanism. Taken together, these observations support a minimal pharmacophore model for urease inhibitors.

8. Outlook

The modulation of urease activity remains a critical focus of research due to its wide‐ranging application in agriculture, medicine, and environmental management. While urease inhibitors hold significant promise, the development of new compounds faces several challenges, particularly in balancing efficacy, safety, and minimizing environmental impact. Despite extensive efforts, only a limited number of inhibitors have demonstrated substantial potential in the agriculture and pharmaceutical field.

Current data from screening of various compounds provide valuable insights into the mechanism of urease inhibition. This review has systematically highlighted and summarized the diverse set of classes of urease inhibitors, emphasizing important structure–activity relationships in guiding the design of more effective and safer inhibitors. It is clear that different chemical groups and scaffolds exhibit various levels of efficacy, and even well‐established classes of compounds do not guarantee potent inhibition. This variability underscores the need for a more refined understanding of how structural features influence urease activity. Looking forward, future in research should focus not just on potency but more in selectivity and stability of compounds. Additionally, the integration of computational approaches such as machine learning is bringing significantly ability to predict and optimize inhibitors.

Major findings indicate there is a fine line between potency and promiscuity particularly when using groups that rely on metal binding motifs. The compounds should have an appropriate number and specific network of hydrogen bonds, and it appears that chloride analogs seem promising in increasing the potency. Another major pursuit may be the modification of current antibiotics to also display potent activity against urease.

Even though we do not address in depth QSAR of scaffolds, we focus on stating emerging patterns of activity by looking at the overall activity of scaffolds in a way to facilitate broad design ideas of urease inhibitors.

Conflict of Interest

The authors declare no conflict of interest.

Biographies

Nuno Martinho has a degree in Pharmaceutical Sciences and a Ph.D. in pharmaceutical technology at the University of Lisbon (Portugal) from which he graduated in 2019. He then joined the Institute for Bioengineering and Biosciences (iBB) as a postdoc working on drug development and polymer functionalization for therapeutic purposes. He is now a senior researcher at the Faculty of Pharmacy, University of Lisbon, currently working on in silico approaches for drug discovery. Dr. Martinho's main areas of expertise are chemical synthesis, molecular docking, molecular dynamics, and drug discovery.

Natália Aniceto has a degree in Pharmaceutical Sciences and a Ph.D. in cheminformatics and in silico modeling, at the University of Kent (UK) from which she graduated in 2018. She then joined the Centre for Molecular Informatics (University of Cambridge, UK) for a year as a post doc where she worked on modeling compound synergy. She then joined the Faculty of Pharmacy at the University of Lisbon, Portugal, as a Junior Professor of Pharmacokinetics and Biopharmaceutics. Dr. Aniceto's main areas of expertise are cheminformatics, machine learning and molecular docking for drug discovery, data analysis and curation.

Martinho Nuno, Aniceto Natália, ChemMedChem 2026, 21, e202500423. 10.1002/cmdc.202500423