Drug discovery targeting heme-based sensors and their coupled activities

Journal of Inorganic Biochemistry 167 (2017) 12–20

Contents lists available at ScienceDirect

Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio

Focused review

Drug discovery targeting heme-based sensors and their coupled activities Eduardo Henrique Silva Sousa a,⁎, Luiz Gonzaga de França Lopes a, Gonzalo Gonzalez b, Marie-Alda Gilles-Gonzalez b a b

Laboratory of Bioinorganic Chemistry, Department of Organic and Inorganic Chemistry, Federal University of Ceara, Center for Sciences, Fortaleza, Ceara 60440-900, Brazil Department of Biochemistry, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-9038, USA

a r t i c l e

i n f o

Article history: Received 10 August 2016 Received in revised form 8 November 2016 Accepted 16 November 2016 Available online 20 November 2016 Keywords: Drug discovery Heme-based sensor Nucleotide cyclase Phosphodiesterase Histidine protein kinase Nuclear receptor

a b s t r a c t Heme-based sensors have emerged during the last 20 years as being a large family of proteins that occur in all kingdoms of life. A myriad of biological adaptations are associated with these sensors, which include vasodilation, bacterial virulence, dormancy, chemotaxis, biofilm formation, among others. Due to the key activities regulated by these proteins along with many other systems that use similar output domains, there is a growing interest in developing small molecules as their regulators. Here, we review the development of potential activators and inhibitors for many of these systems, including human soluble guanylate cyclase, c-di-GMP-related enzymes, Mycobacterium tuberculosis DevR/DevS/DosT (differentially expressed in virulent strain response regulator/sensor/ dormancy survival sensor T), the Rev-erb-α and β nuclear receptor, among others. The possible roles of these molecules as biochemical tools, therapeutic agents, and novel antibiotics are critically examined. © 2016 Elsevier Inc. All rights reserved.

1. Introduction A myriad of heme-based sensor proteins have emerged during the last two decades [2,3]. These biological sensors are modular: they contain at least one input domain that harbors a heme cofactor, and output domains that are involved in the biological response. So far, all domains that are capable of stably binding heme have been demonstrated to be capable of sensing, with examples known for PAS (Period-Sim-ARNT) [4], GAF (cGMP-binding PDE, adenylyl cyclase, FhlA) [5], globin [6], HNOB (heme NO binding) [7], CooA (CO oxidation Activator) [8], LBD (ligand binding domain of nuclear receptor) [9], and SCHIC (sensor containing heme instead of cobalamin) [10] domains. These input domains' detection of heme ligands can be coupled to a wide range of output domains and signal transduction modes, such as histidine protein kinase, mono- and di-nucleotide cyclases, phosphodiesterases, DNA-binding and protein-binding domains. It is therefore unsurprising that the physiological adaptations controlled by these sensors are very broad. These adaptations include the regulation of blood pressure, circadian clock, CO metabolism, chemotaxis, dormancy/pathogenicity, metamorphosis, symbiosis, biofilm formation or dispersal, among others (Fig. 1). The modularity of heme-based sensors provides an opportunity to target synthetic small-molecule activators (agonists) and inhibitors (antagonists) to distinct sites within them. These target sites include ⁎ Corresponding author. E-mail addresses: [email protected] (E.H.S. Sousa), [email protected] (M.-A. Gilles-Gonzalez).

http://dx.doi.org/10.1016/j.jinorgbio.2016.11.022 0162-0134/© 2016 Elsevier Inc. All rights reserved.

the heme domain, the output domain, and additional putative regulatory domains (Fig. 2). The latter occur in many sensors, where they are likely to be involved in protein oligomerization and mediation of signal transduction, though their specific regulatory roles are unclear [11–13]. The heme domain is one potential target site for exogenous regulation of heme-based sensors. There are already interesting examples of this kind of regulation for the soluble guanylate cyclase (sGC), which will be further reviewed. One caveat, however, is that any species that acts directly on the heme cofactor of a sensor could potentially act on the heme in hemoglobin and myoglobin, found in high concentration in humans. However, the variety of folds harboring heme allow for the design of agents more selective for these heme domains (e.g. HNOB, PAS, GAF), able to minimize cross-reactions with globin-based proteins, e.g. myoglobin, hemoglobin. This approach has made the heme domain of these systems a valid and suitable target. Targeting the output domains, rather than the input domains, has been a more common strategy for development of small molecule modulators (Fig. 2A). As we will discuss farther, several interesting molecules are being developed for the regulation of output domains of heme-based sensors. Selective inhibitors have begun to emerge for diguanylate cyclase and histidine kinase: two enzymatic activities that have been shown to have important roles in bacteria [14,15]. There is a clear opportunity to develop novel small molecules for these enzymatic activities, with a much wider application than just heme-based sensors. One additional and no less interesting opportunity might come from targeting the “extra” putative regulatory domains found in many hemebased sensors that we also call “non-functional” domains (see Fig. 2).

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

Proteinprotein binding (Circadian clock)

c-di-GMP (Biofilm formation/ dispersal and RNA regulation)

HEME-BASED SENSORS / OUTPUT DOMAINS Methylaccepting (Bacterial swimming)

Adenylate Guanylate Cyclase (Cardiovascular regulation, Virulence)

Histidine kinases (Symbiosis, Dormancy)

DNA binding (Metamorphosis, CO metabolism) Fig. 1. Heme-based sensors and their associated output domains are involved in key biological activities and are potential targets for the development of small molecule regulators.

Here we review some of these systems for which regulators have been developed. 1.1. Regulators of soluble guanylate cyclase Soluble guanylate cyclase is a nitric-oxide (NO) regulated hemebased sensor involved in a variety of important biological processes, including smooth muscle relaxation, platelet aggregation, and neurotransmission. This heterodimeric protein contains a larger α-subunit and a smaller β-subunit. The latter harbors the iron(II)-protoporphyrin IX in a HNOB domain (also known as HNOX), where iron is axially

13

bound to a histidine (H105) and in the unliganded state is pentacoordinated. The human sGC's cyclase activity increases 200 fold upon binding of NO to the heme. The activation of sGC following ligation of NO was once thought to be only due to the rupture of the bond between the heme iron and H105, resulting in pentacoordinate heme iron. However, hemeless sGC was shown to be activated upon binding to organic molecules mimicking heme, whose carboxylate groups are critical for activation, suggesting specific side chains interaction are very important [16]. Additionally, it is now known that pentacoordination is not required for full sGC activation, since the activity of the hexacoordinate carbon monoxide (CO) bound form of sGC in the presence of the synergistic activator YC-1 (5-[1-(phenylmethyl)1H-indazol-3-yl]-2-furanmethanol) is the same as that of the pentacoordinate NO-bound sGC. Nevertheless, there is evidence, based on resonance Raman spectroscopy, of sGC iron-histidine bond weakening or the formation of a fraction of heme pentacoordinated with CO induced by YC-1 or BAY-412272 [17–19]. Either YC-1 by itself, or CO by itself, enhances cyclase activity (12-fold and 5-fold, respectively), but these effect are much smaller than that of the two effectors combined (about 200-fold) [20]. YC-1 greatly enhances binding of NO due to deceleration of the NO dissociation rate, resulting in sGC that is sensitive to subnanomolar concentrations of NO. Interestingly, the binding affinity and kinetics of CO are unaffected by YC-1, even though it tremendously increases the effect of CO on the cyclase activity of sGC [20,21]. These results have opened new opportunities to develop and apply not only NO donors but also CO donors and small-molecule regulators to modulate sGC. 1.1.1. Nitric oxide donors NO donors have been investigated as sGC activation agents, showing important pharmacological applications for the control of blood pressure, angina pectoris, ischemia, hypertension, among other disorders

Fig. 2. Potential target sites for the development of small-molecule regulators (A); examples of some domain organizations of heme-based sensors (B); domain organization of sGC (C). The sensory domain where the heme group is bound is in red and stippled; the other sensory domains are putative and are of unknown function. The output domains are underlined in black. Output domains include GGDEF-EAL, which is involved in cyclase and phosphodiesterase activity for c-di-GMP; HisKA-HATPase_c, which has histidine kinase activity; guanylate cyc usually has a GTP cyclase activity that generates cGMP, although in the HemAC protein it has an ATP cyclase activity that generates cAMP; ReFixL also possesses a receiver domain of a response regulator, called a REC domain, in which the aspartate 573 residue can be phosphorylated.

14

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

[22]. Currently, under physiological conditions, upon thiol stimulation, or even light activation, a series of organic-based synthetic molecules and metal-based compounds can generate or release NO (or HNO) [22–26]. HNO donors have also emerged as potential modulators of sGC, although this molecule in vitro cannot fully activate this sensor to the same extent as does NO [27]. Indeed there are conflicting results that put on the stage the actual capacity of HNO for direct activation of sGC. Nonetheless, HNO has quite distinct reactivity, particularly toward thiol-contaning proteins and ferric hemeproteins, which might account for some other unique biological targets, e.g. aldehyde dehydrogenase, glyceraldehyde 3-phosphate dehydrogenase [27]. An impressive number of the metal-based NO donors emerged during the last 15 years [24,28,29]. Unfortunately, there are few examples in the literature regarding the direct activation of sGC by inorganic NO donors. A good example of an inorganic NO donor that activates sGC is the trans-[Cr(cyclam)(ONO)2]+ [23]. In this study, Ford and co-workers observed that this compound, when irradiated with low-intensity light, could promote an increase in the activity of sGC. In separate experiments, they demonstrated under the same conditions that there was production of NO. Additionally, this complex showed vasodilation properties under blue LED light irradiation, reinforcing its physiological potential [23]. This latter approach can indeed be an indirect measurement of sGC activation, which has been performed by many other groups along with selective sGC inhibitors (e.g. ODQ). In this way, our group and others [29] have prepared several ruthenium nitrosyl complexes, which also exhibited vasodilation properties among other biological activities that can be related to sGC activation as a biological indication of NO release. In this context, the complex [Ru(bpy)(SO3)NO]PF6 showed several interesting biological activities that are related to sGC activation: i.e. inhibition of inflammatory pain by involvement of TRPV1 and cGMP/PKG/ATP [30]; neuroprotection by inhibiting NF-kB signaling and helping to stabilize the blood pressure during the transition from ischemia to reperfusion [31]; gastric protection by activation of sGC and KATP channels [32], among others [33,34]. The development of NO donors has offered a wide range of tools to modulate not only the activity of sGC, but also other heme-based sensors responsive to NO, which deserve further study. However, the lack of selectivity of NO, along with drug tolerance, has stimulated the search for other agents to control sGC-mediated processes. The development of non-NO based synthetic molecules targeting sGC was the first case reported for a modulator of a heme-based sensor [35]. Currently, organic-based molecules that enhance sGC activity are separated in two classes of compounds based on their mechanism of action on sGC. The first one requires the presence of NO or CO along with the reduced heme group to work on sGC and is called a stimulator. The second class of compounds works independent of NO and works on oxidized or heme-depleted sGC, and they are called activators. The first generation of YC-1 related compounds resulted in a selection of promising sGC stimulators that were heme-dependent, e.g. BAY 41-8543 [36]. Our group recently proposed a new strategy of metal-based agents as potential modulators of sGC. In these compounds, a common organic skeleton of YC-1 was modified with a metal-complex moiety. This relatively simple approach to generate structural diversity produced compounds with promising vasodilation activity, e.g. FOR005 (Table 1) [37]. Unfortunately, we had only modest evidence that these compounds worked by stimulating sGC. Other metal modifications have been conducted using iron-based complexes, [Fe(CN)5L]3−, whose good vasodilation activity assayed in rat aorta was completely abolished upon treatment with ODQ or l-NAME, indicating a mechanism of stimulation of sGC (submitted). Besides the stimulators of sGC that require a reduced iron heme cofactor, novel molecules, such as ataciguat and cinaguat, have been developed that can activate even oxidized or hemeless sGC (Table 1). BAY 63-2521 (riociguat) completed phase III clinical trials and was approved by the FDA for the treatment of pulmonary hypertension under the market name of Adempas, sold by Bayer. These molecules

have been called sGC activators. They represent a second generation of sGC agents with important cardiovascular potential for treatment of decompensated heart failure, liver fibrosis and also lowering both pulmonary and systemic arterial pressures [35]. How these compounds promote sGC activation is not well understood yet, but it is believed that they might simply replace the heme cofactor in sGC [38]. Indeed, X-ray structures that have been solved for a homologous sGC heme domain, Nostoc H-NOX, with cinaciguat and BAY 60-2770 replacing the heme group, reinforce those assumptions [16,39]. Cinaciguat was found to adopt a conformation very much like a heme group inside the conserved YxSxR motif of the heme domain of H-NOX, while the proximal histidine was maintained in a position expected for an NO-activated sGC [16]. This knowledge has contributed to further development and investigation of activators of sGC [35]. Recently, Martin's lab showed that a fibrate drug, gemfibrozil, was also an sGC activator, whose activity was improved against oxidized sGC [40]. This compound worked presumably by binding to the heme domain, similarly to other heme-independent activators, and removal of the carboxylic group abolished its activity.

1.1.2. Porphyrins and related compounds Beyond the compounds described so far, there are other efforts including A-350619 [41], IWP-051 [42], and BI 703704 [43,44] (Table 1). Martin and Gryko's labs have prepared a series of new compounds aiming to activate sGC, based on a macrocyclic scaffold. It has been known for a long time that metal-free porphyrin IX can activate sGC in an NO-independent manner. This is most likely due to replacement of the heme by the porphyrin IX, which would assume a position similar to pentacoordinate heme-NO, where no proximal attachment to histidine occurs but new interactions with side chains would be essential for sGC activation. In a recent paper, these labs prepared porphyrin derivatives with modifications at the meso position to investigate their possible activation of sGC [1]. Unfortunately, all of those compounds were less efficient than porphyrin IX itself at activation, but they showed measurable activity, which deserves further study. Other macrocyclic derivatives were previously prepared by these labs based on vitamin B12 derivatives, cobinamides, whose enhancement of sGC activity has been promising, with activation up to 60-fold [45]. Interestingly, these compounds do not seem to activate sGC through heme replacement but at the catalytic site [46]. Additionally, they were shown to potentiate the effect of BAY41-2272 and BAY582667 by a synergy. Furthermore, at higher concentrations some of these molecules worked also as inhibitors, making their mechanism of action more complex [47].

1.1.3. Possible allosteric activators Besides the important medical applications of these agents, these studies also contributed to our emerging understanding of the signal transduction mechanism of sGC and its relatives [16,38,39,48,49]. It is important to note, regarding the work in vitro, that there is an ongoing debate about the possible allosteric site(s) for binding of these stimulators and how they might promote signal transduction. Interestingly, in sGC the HNOBA domain of the α subunit is a PAS-like region of unknown function. The PAS structure is a well-known regulatory domain found in many signal transduction systems. In sGC, that region is thought to be responsible for dimerization, but it could be also a point of fine regulation. Most heme-based sensors share this feature of containing an additional domain of unknown function, but with a regulatory-domain fold (e.g. PAS, GAF, HNOB), as we noted earlier [11]. So, it would be quite reasonable that these “extra domains” might be employed in further regulation of these proteins [11]. Particularly in sGC, where the heme is bound in an HNOB of the β-subunit, there is an HNOB without any bound cofactor in the α-subunit that could be used in vivo for regulatory purposes as yet undiscovered (Fig. 2C).

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20 Table 1 Examples of sGC activators and stimulators.

15

Table 2 Examples of some sGC inhibitors.

Activators

O

O N

O

N

O

O N

OH

S

S+

N

N

O

O Cl

OH

N N O

Methylene blue

ODQ

LY 83583

Na+

N-

N

N

Cl-

O N H

O

O

H N

O S

O

S Cl

O

N

O Br

Ataciguat

O

O

O

O

N

N

N

N

N

Cinaciguat

O

O O

NS 2028

O

F

O

2 O2N

F

N

H N

N

NH

O

O

S O

1

F

N N

O

NH

N

O N

OH

N

Br

N

OH

N

O

O OH

Gemfibrozil

O

BI 703704

OH

3

4 (lack of activity)

5

Other regulator CO2Me

N

MeO2C

NC N

MeO2C

Co

N CN CO2Me

N H

Co2M2 CO2Me

Stimulators N

N

N

N

F

N

F

N

N N

N

N

N

N NH2

NH2

O H2N

O

H2N

N

N

O

BAY 63-2521 (Riociguat/Adempas)

O

YC-1

BAY 41-8543 F

O

N O

H N

N

N

N

NH N

N

N Ru Cl

N

S N

N N

OH Cl

F

FOR005

IWP-051

A-350619

1.1.4. Inhibitors of sGC Despite the greater interest in sGC stimulators and activators, there is also a particular effort to develop sGC inhibitors, which might be useful for a series of medical conditions such as septic shock [50,51], migraine [52], Parkinson's disease [53], control of intraocular pressure (glaucoma) [54,55], and the prevention and treatment of dental disorders [56] among others [57,58]. Unfortunately, there is even a smaller number of compounds developed for those tasks, but new candidate molecules have been found. The first generation of sGC inhibitors features methylene blue and LY83583, one of the oldest reported inhibitors (Table 2). These inhibitors, however, are non-specific to sGC and work by generating superoxide, which causes deactivation of NO [59,60]. So, a second generation of selective inhibitors was developed, where some molecules have been known for almost 20 years already. The two closely related molecules, ODQ (1H-[1,2,4]-oxadiazole[4,3-a]quinoxalin-1-one) and NS2028 (8-

Bromo-4H-[1,2,4]oxadiazolo[3,4-c][1,4]-benzoxazin-1-one), were reported early to inhibit selectively NO-activated sGC (Table 2) [61,62]. Due to these properties, these molecules have been widely used as important biological tools for investigations defining the role of sGC [63, 64]. The mechanism of action of ODQ and NS2028 is associated with heme-iron oxidation, which disrupts NO activation and leaves only basal sGC activity [61,65]. Indeed, these inhibitors were showed also to promote iron oxidation of hemoglobin, but only at much higher concentrations, which might not be a real concern [65]. Interestingly, Gerber's lab suggested that sGC might have an allosteric binding site for ODQ, based on circular dichroism studies. Her lab suggested even a possible overlap of binding sites with the sGC-stimulator YC-1 [66], which has not been followed up yet. It is still unclear if ODQ causes sGC oxidation through intramolecular transfer or via an outer sphere mechanism by binding to a site remote from the heme iron. In 2009 a class of compounds, based on tricyclic indole and dihydroindole derivatives, were prepared and their inhibitory activity on sGC assessed [67]. These compounds inhibited sGC by disabling NO activation, while they maintained basal sGC activity. Despite this behavior mirroring ODQ and NS2028 activity, there was no study investigating heme oxidation. Unfortunately, the maximum inhibition of cGMP accumulation measured using NO-activated aorta tissue was only moderately decreased. The best compound 1, at 100 μM, reduced cGMP level to half (Table 2) [67]. Nevertheless, this class of compounds deserves further investigation and might lead to promising inhibitors. Rekowski and colleagues have reported syntheses of ODQ and NS2028 derivatives, which has allowed them to identify important elements for the structure-activity of these compounds toward sGC inhibition [68]. Compounds 2 and 3 exhibited the best ability to disrupt cGMP accumulation on NO-activated aorta cells, while compound 4 did not show any activity, despite its close similarity to NS2028 (Table 2) [68]. These results indicated a key role of the oxadiazole ring for the inhibitory activity, where a triazole ring led to inactivity. Another lab in 2011 followed up on the synthesis of NS2028 analogs and prepared a series of sulfur-based compounds [69]. None of those compounds, however, have had their potential sGC inhibitory activity assessed so far. Another interesting old molecule, thalidomide, a well-known FDAapproved drug, currently used mainly for cancer treatment, showed inhibition of sGC, which might be responsible for its anti-angiogenesis effects [58]. Chatterjee's lab provided data indicating that thalidomide binds to sGC using a pull-down assay, followed by fluorescence aiming to measure thalidomide carried over with sGC. Other measurements using cells treated with NO showed a decrease of cGMP levels upon thalidomide treatment, which was not due to any decrease in sGC

16

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

expression. Besides this, molecular modeling and docking supported binding of the thalidomide to the catalytic domain of sGC [58]. These interesting results should be further studied, and they might open new opportunities for the development of sGC inhibitors as anti-angiogenesis drugs, based on thalidomide's structure. In 2015, Mota and colleagues identified new modulators of sGC when investigating the action of lamotrigine (anti-epileptic drug), sipatrigine and another analog [70]. They noticed that lamotrigine caused sGC activation, while the others functioned as sGC inhibitors at low millimolar concentrations. These structures were employed as a start point for in silico screening, where 500 potential candidates were identified and 16 were selected based on structural diversity, molecular size, and availability. These selected compounds were assayed at 100 μM against purified bovine lung NO-activated sGC. Compound 5, a quinoxaline derivative, showed promising sGC inhibitory activity in vitro, which motivated the authors to prepare other structural modifications (Table 2). These compounds also were investigated using surface plasmon resonance (SPR) employing the catalytic domain and fulllength sGC, where it was possible to measure the affinity of these compounds to sGC. Interestingly, many of them did not show any inhibitory activity, even though they could bind to sGC. Nevertheless, these results shed light on the key structural modifications required to have good sGC binding along with inhibitory activity. Compound 5 was one of the best sGC inhibitors identified, yielding Kd values of 11.4 μM and 19.4 μM for binding to the catalytic domain and full-length sGC, respectively [70]. Assays conducted with atrial natriuretic peptide-stimulated particulate guanylate cyclase (pGC) showed that this compound could also inhibit a similar guanylate cyclase enzymatic site. However, it could not inhibit a closely related site of adenylate cyclase or forskolin-stimulated adenylate cyclase at all, indicating a certain level of selectivity. Biological measurement of cGMP accumulation was carried out, where compound 5 at 100 μM caused a decrease in cGMP accumulation of about 20-fold [70]. Nonetheless, we should note that this compound shares some structural features with OQD, although the authors argue that it is a novel allosteric inhibitor selective for a catalytic site distinct of ODQ action. Altogether, the development of sGC modulators is evidently a hot spot and may lead to many great opportunities in the near future.

1.2. Modulators of nuclear receptor heme-based sensors The E75 protein from Drosophila melanogaster is a nuclear receptor protein, also identified as a heme-based sensor [9]. This protein is involved in an ecdysone-triggered cascade that controls the molting, pupation, and eclosion processes of larval–pupal metamorphosis, which could require NO or CO as a signal [71]. E75 contains a heme group bound to a LBD domain, which has been found in other nuclear receptor proteins. Rev-erbα and β are homologous nuclear receptor heme-based sensors found in humans [72]. These proteins are involved in circadian biology. They play a role in the homeostasis of lipids and several other processes, including circadian rhythm, both in the brain and in peripheral tissues. Rev-erbα and β have also been claimed to respond to gaseous signal (NO and CO), and to free heme [73], but the latter has been refuted [74]. Recently, these proteins were subjected to a search for activators or inhibitors that might be useful for controlling a series of physiological disturbances, including obesity, sleep-associated and emotional disorders (e.g. anxiety). GSK4112, initially named SR6452, was the first non-porphyrinic activator found for Rev-erbα (Table 3) [75,76]. This compound was identified from a biochemical assay that measured the interaction between the LBD domain and a cofactor partner as a screening strategy. Interestingly, interaction suppression was noticed in an assay with GSK4112 upon addition of hemin, suggesting that GSK4112 might compete with heme for the LDB binding site [76]. Later a series of small molecules were identified as activators of Rev-erbα and β; in particular, SR9011 [77], SR9009 [77], GSK2945 [78], GSK0999 [78],

Table 3 Examples of small molecule modulators of Rev-erbα and β.

Cl

O

N O N

O2N

O2N

N

S

O2N

GSK4112 (activator)

S

SR9011 (activator)

Cl

N

S

O

N NH

GSK0999 (activator)

F

Cl

N

O

O

S

N

S

N O

S

Cl

NO2

SR10067 (activator)

O

O

SR8278 (inhibitor)

GSK2945 (activator)

N Zn2+

N

N

O N OH

Zn protoporphyrin IX (inhibitor) O OH

among others [79,80], showed IC50 values in the nanomolar range (Table 3). Based on the structure of SR9011, new structural modifications were further prepared in 2014; this effort identified SR10067, which showed greater activation, with an IC50 ~ 160 nM, better pharmacokinetics profile and high target selectivity [81]. These studies indicated a potential application of these compounds in treating sleep disorders, disorders associated with anxiety, as well as addiction, upon selective modulation of these heme-based sensor nuclear receptors [81]. Many descriptions of the relevance of developing small molecules for proteins involved in circadian clock have been thoroughly reviewed in the literature [82]. In addition to those compounds, inhibitors have been prepared, such as SR8278 [83] and porphyrin derivatives (Co and Zn protoporphyrin IX). This inhibitory behavior observed for Co and Zn protoporphyrin is particularly exciting since heme itself has a stimulatory activity [84]. When X-ray and NMR studies were carried out on Rev-erbβ bound to co-protoporphyrin or heme, however, only modest structural differences were observed despite these two molecules' opposing activities, highlighting that much more must be learned [84]. It is also true that signal transduction investigations might advance further if full-length proteins are employed, rather than the isolated heme domains, since domain-domain interaction might be a key event. Nonetheless, the development of these modulators is important, and this might also open opportunities for the development of agents for insect control targeting E75. Similarly to these examples, many other heme-based sensors are ripe for a search for modulators that could provide tools for our understanding of their signal transduction processes.

1.3. DevS/DosT/DevR and histidine kinase regulators Two-component regulatory systems have recently emerged as promising antibiotic drug targets [14]. This is partially due to the fact that their absence from mammals allows them to serve as unique targets in microorganisms. These systems regulate many important genes in microbes involved in sporulation, respiration, metabolism,

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

biofilm formation, antibiotic resistance, bacterial pathogenicity, among others [14,85]. In 2007, DevS (differentially expressed in virulent strain sensor, also known as DosS, dormancy survival sensor) and DosT (dormancy survival sensor T) were showed to be O2-regulated heme-based histidine kinases that, along with the response regulator DevR (differentially expressed in virulent strain response regulator), constituted an important two-component systems for Mycobacterium tuberculosis (Mtb) adaptation [5]. The Mtb DevS/DosT/DevR system is involved in the persistence and/or latency of Mtb, a strategy developed by this bacterium for adaptation and survival within diverse microenvironments found in humans. During the last few years, Mtb DevS/DosT/DevR inhibitors have been sought, because they might help to prevent latency and shorten the treatment for Mtb infections, which now requires a minimum of six months of an antibiotic cocktail. Tyagi and colleagues targeted the response regulator DevR in their studies. They used in silico screening of 2.5 million compounds to identify some candidates, such as a phenylcoumarin derivative 6 (Table 4). This compound was able to disrupt DevR binding to its DNA target, and they suggested that this inhibitor had helped to lock DevR in an inactive conformation [86]. In addition, Tyagi's lab prepared a combinatorial peptide phage display library, using DevR as bait, and identified one peptide candidate with the TLHLHHL sequence [87]. This peptide, at 2.5 to 5 mM, reduced the viability of Mtb more than 80% under hypoxia, but it did not prevent DevR from getting phosphorylated or from binding to its DNA target, which makes its mechanism of action rather puzzling. More recently, Tyagi and colleagues repeated the peptide phage display approach using DevS as bait, and they identified four new peptide candidates Table 4 Examples of inhibitors for two-component systems.

O

PDVAVLDVRLPD peptide

H N O

Cl

O N H

6

TLHLHHL peptide

O

Other histidine kinases

O

O S N H

S N H

NH2 HN

N

N H

H2N

N

S

Br

O

TEP

LED209 HO O

O

S

S

H2N

N

NH2

N

N

N

N

S

7

CAA cyanoacetoacetamide O

N

N

O O2N

N N H

NH2 F

F F

9

8 OH

HO O N

NSC4836

[88]. These peptides resembled the N-terminal domain of DevR in sequence, suggesting that they might disrupt DevR/DevS interactions; however, they showed little anti-DevR activity. Tyagi and colleagues persevered and looked for peptide sequences in DevR that were analogous to those selected by the phage display [88]. These new peptides disabled DevS autophosphorylation and disrupted DevS/DevR binding, while showing anti-tuberculosis activity in a hypoxic Mtb culture. The authors did not check whether the peptides could also inhibit DosT, but they surmised that DosT would also have to be inhibited, since it failed to rescue the cells from the lack of activity of the inhibited DevS. The cytotoxicity of these peptides in mammalian cells was very low (20–40% cytotoxicity using 2.5 to 5 mM of peptide) [88]. These interesting data would benefit very much from in vitro assays with full-length DevS and DosT. During the last few years several groups have worked to identify inhibitors for two-component systems such as LED209, TEP, CAA, 6, 7, 8, 9, NSC4836 (Table 4) [89–94]. These compounds might very well be starting points also for modulators of DevS-DosT-DevR, or even SenX3-RegX3 (sensor X3 and response regulator X3), another putative heme-based two-component system of Mtb, which deserves further investigations [14]. Some strategies have been proposed for screening larger libraries for potential modulators, but many compounds are still subject to limitations; in some cases they cause non-specific protein aggregation and in others they are not effectively transported across cell walls and membranes [14,95–97]. Despite this, some new proposals have emerged, such as the investigations of the two-component system CpxA/CpxR using an E. coli strain containing a CpxR-responsive lacZ reporter [93]. These types of new efforts might help to fulfill the urgent need for novel antibiotics and lead to big leaps in the field. 1.4. Diguanylate cyclase regulators

DevS/DevR O

17

NH2 N

A diguanylate cyclase activity (DGC) of some sensors that produce cdi-GMP under specific conditions has received a great deal of attention [98,99]. The role of c-di-GMP in bacteria is quite extensive, but a regulation of polysaccharide biosynthesis and motility has been particularly interesting, due to the implications for bacterial biofilm formation [100]. Bacteria in biofilms benefit from a huge increase in antibiotic tolerance that can reach up to 1000-fold the common dosage [101,102]. Additionally, biofilm formation is a key factor leading to persistence of bacterial infections in some chronic conditions like pneumonia in cystic fibrosis patients, otitis media, and non-healing wounds. In addition to this effect of biofilm on persistence, along with the rise in multi-drug resistant bacterial infections all over the world, the scientific community has started to seek new targets for antibiotics. The DGC and PDE (phosphodiesterase) enzymes that catalyze the synthesis and degradation of c-di-GMP are among those targets [15,103,104]. The number of heme-based-sensor proteins that regulate a DGC or PDE activity has increased and may illustrate the exciting opportunities ahead [3]. Indeed, soon after the discovery of c-di-GMP as second messager in G. xylinus bacteria, there followed the discovery of an O2 heme-based sensor, AxPDEA1, which controls c-di-GMP levels. G. xylinus is a bacterium that produces a cellulose biofilm, which is directly under regulation of c-di-GMP and ultimately O2 through AxPDEA1 [105]. The pathogenic bacterium Bordetella pertussis, i.e. the agent of whooping cough, controls biofilm formation at least partly with an O2regulated heme-based sensor that contains a DGC activity [106]. In E. coli, the O2 sensors DosC (direct oxygen sensor cyclase) and DosP (direct oxygen sensor phosphodiesterase) are involved in control of c-diGMP and RNA degradation, and they also seem to have an important role in biofilm formation [107]. Similarly to O2, NO seems to have a role in inducing or dispersing biofilms, depending on the microorganism. In certain cases, a hemebased sensor regulates the DGC or PDE activity. Nitric oxide responsive heme-based sensors that contain DGC or PDE activities and that are implicated in the regulation of biofilm formation, have been described for

18

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

bacteria such as Legionella pneumophila and Shewanella woodyi (reviewed elsewhere) [108]. These sensing proteins present many opportunities for chemists to develop molecules to modulate these processes either at the sensing domain (e.g. heme domain), allosteric sites, or even at the level of the catalytic domains. Compounds such as NO donors have great potential to interrupt and disperse the biofilms of several microorganisms like P. aeruginosa, E. coli, V. cholerae, Staphylococcus aureus, and Nitrosomonas europaea [108]. Despite the fact the mechanism of NO action on many of these bacteria is still unknown, some pathways have been linked to the disruption of cdi-GMP biosynthesis and to heme-based sensor systems [108]. Some interesting molecules have been identified as inhibitors of DGC with major roles in biofilm disruption as described below. By employing a sequence of screening strategies, starting with congo red dye staining of exopolysaccharides produced by E. coli harboring a plasmid encoding the AdrA DGC, Landini and colleagues screened 1120 compounds and identified a sulfathiazole that could affect biofilm formation by blocking synthesis of c-di-GMP. They suggested, however, that this compound most likely works by indirectly affecting c-di-GMP biosynthesis rather than by directly binding to the DGC [109]. On the other hand, Sambanthamoorthy and colleagues reported several new compounds that block biofilm formation and can inhibit multiple DCGs in V. cholerae and P. aeruginosa [104]. Those molecules were screened in a bioassay and also assayed using purified DCGs from V. cholerae and P. aeruginosa. One of these compounds, DI-3, was selected as a promising candidate for biofilm inhibition on V. cholerae with IC50 of 26 μM, showing inhibitory activity on two DGCs from these bacteria (Table 5) [104]. More recently, the same group used an in silico screening method to identify some drug candidates. Compound LP3134 was found, which inhibited purified DGC enzymes from P. aeruginosa and Thermotoga maritima. Importantly, this compound also blocked the initial attachment of the bacteria as well as development of biofilm, and it promoted biofilm dispersion (Table 5) [110]. Lieberman and colleagues have developed a high-throughput screen for drugs that target DGCs, and they identified a candidate called ebselen (Table 5). This promising compound was able to revert c-diGMP regulated phenotypes, particularly motility and biofilm formation [111]. Hits against PleD were also identified in silico; these were validated in vitro, and a compound with a 3-nitro-benzenesulfonamide scaffold (10) and IC50 ~ 11 μM was found among them (Table 5) [112]. Analogs of c-di-GMP have also been investigated as potential modulators of the DGCs, but these compounds can exhibit quite different effects, either activation or inhibition, depending on the DGC [113,114]. Nevertheless, there is a huge potential for discoveries that should also bring about a better understanding of the mechanisms of function of these systems.

1.5. Molecules to target “non-functional” domains A large fraction of heme-based sensors have domains besides the heme-binding and output domains, whose functions are unknown (Fig. 2). Many of these domains belong to families such as PAS, GAF, HNOBA, whose members often have signaling functions. These extra domains might potentially interact with some endogenous small molecules such as cGMP, bilin, 4-hydroxycinnamic acid, malonate, succinate that have already been found to bind to structurally analogous domains [115,116]. Alternatively, they might be involved in regulatory interactions with macromolecules. Our labs conducted a study on a hybrid FixL (nitrogen fixation gene L) from Rhizobium etli (ReFixL) to shed some light on the roles of some of these “non-functional” domains. The ReFixL protein possesses two PAS domains in tandem. The second PAS contains the O2 sensing heme group, but the first PAS has no known function (Fig. 2B). A deletion of the “non-functional” first PAS domain enhanced O2 binding but disrupted all signal transduction, leaving the kinase inactive regardless of O2 ligation state of the heme [11]. More recently, an adenylate-cyclase heme-based sensor (HemAC) from Leishmania major showed a somewhat similar behavior. In that case, the removal of a hemeless globin domain (Globin-B) of unknown function did not change the ligand binding behavior of the hemebound globin domain but led to a ligand-independent loss of the adenylate cyclase activity [12]. The authors expressed the adenylate cyclase domain by itself, which maintained an impressive activity, at least 250fold higher than the Globin-B deletion mutant but still 10-fold lower than the full-length HemAC. These results supported the role of the Globin-B domain in signal transduction [12]. Altogether, it looks quite plausible that, by targeting these extra domains with small molecules, one could regulate the activities of the sensing and output domains. Indeed, a similar approach of targeting extra putative regulatory domains has been conducted with human phosphodiesterases (PDE). PDE inhibitors are a new class of therapeutic agents that have been applied to a variety of diseases. The best known of these is sildenafil, which targets the enzymatic site of PDE5 and is used to treat erectile dysfunction as well as reduce pulmonary arterial pressure. A nonselective inhibition of other PDEs is thought to be the cause of some of sildenafil's side effects; therefore, some researchers attempted to target regulatory cGMP-binding GAF domains on PDE5 rather than the enzymatic region itself [117]. In another example, the phosphodiesterase activities of PDE10 and PDE11 were shown to be enhanced by binding of compounds to GAF domains with as yet unknown functions [118]. These examples support the potential for “extra” domains to serve as therapeutic targets, and also to shed light on the mechanisms of function of these proteins. Although there is not yet a conclusive example of this type of regulator for heme-based sensors, YC-1 and its derivatives might work by targeting the HNOBA (coiled-coil and PAS) domain of soluble guanylate cyclase [119,120]. 2. Final considerations

Table 5 Examples of inhibitors for di-c-GMP enzymes. H N

O

O N H

N Se

Ebselen

DI-3 OH

N O O N

OH

H N

H3C

OH O

O

OH

O2N O

N

LP 3134

N

S

10

N H

OH

We have pointed out that a significant number of heme-based sensor proteins can respond to small molecule modulators, and that this is resulting in potential new drugs, including a case of an FDA approved drug (riociguat, Adempas®). Despite this, there remain other hemebased sensors, like NPAS2 (neuronal Per-Arnt-Sim 2), which are potential drug targets and for which the work on modulators has not taken off yet, but for which there are surely great prospects. Modulators for NPAS2, a protein involved in circadian processes, could lead to potential agents for treatment of bipolar disorder, sleep disturbances, drug addiction, cancer, and other illnesses [121–123]. Along these lines, SenX3, a new putative heme-based sensor from M. tuberculosis, also might be an interesting target. This sensor is part of a two-component system, SenX3-RegX3, that shares similarities with DevS/DosT/DevR and that has been implicated in the recovery of Mtb from a latent state [124].

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

Additionally, the important roles identified for c-di-GMP in Mtb pathogenicity and dormancy make the corresponding cyclase and phosphodiesterase enzymes potential drug targets [125]. Other putative heme-based sensors in the mosquitoes Aedes aegypti (HNOB, LDB) and Anopheles gambiae (HNOB), identified only by bioinformatics (unpublished data), if proven to play essential roles in these organisms, could open further opportunities for the development of novel modulators for the control of these insect vectors of pathogenic microorganisms. The emerging work in designing modulators for these systems can lead to the development of unmatched novel drugs and pest-control agents. Furthermore, these small molecules will likely assist our understanding of the molecular functions of these proteins and suggest possible endogenous regulators.

Acknowledgements We acknowledge FUNCAP (PPSUS 12535691-9), CAPES (PVE – 09/ 2014 process 88881.068193/2014-01) and CNPq (E.H.S.S. # 312030/ 2015-0; L.G.F.L. 303732/2014-8) for providing financial support. References [1] S. Smolen, D.J. Walaszek, M. Karczewski, E. Martin, D. Gryko, Isr. J. Chem. 56 (2016) 156–168. [2] M.A. Gilles-Gonzalez, G. Gonzalez, in: A. Ghosh (Ed.), The Smallest Biomolecules: diatomics and their Interactions with Heme Proteins, Elsevier 2008, pp. 18–65. [3] T. Shimizu, D. Huang, F. Yan, M. Stranava, M. Bartosova, V. Fojtikova, M. Martinkova, Chem. Rev. 115 (2015) 6491–6533. [4] M.A. Gilles-Gonzalez, G.S. Ditta, D.R. Helinski, Nature 350 (1991) 170–172. [5] E.H.S. Sousa, J.R. Tuckerman, G. Gonzalez, M.A. Gilles-Gonzalez, Protein Sci. 16 (2007) 1708–1719. [6] S. Hou, R.W. Larsen, D. Boudko, C.W. Riley, E. Karatan, M. Zimmer, G.W. Ordal, M. Alam, Nature 403 (2000) 540–544. [7] L.J. Ignarro, K.S. Wood, M.S. Wolin, Proc. Natl. Acad. Sci. U. S. A. 79 (1982) 2870–2873. [8] D. Shelver, R.L. Kerby, Y. He, G.P. Roberts, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 11216–11220. [9] J. Reinking, M.M.S. Lam, K. Pardee, H.M. Sampson, S.Y. Liu, P. Yang, S. Williams, W. White, G. Lajoie, A. Edwards, H.M. Krause, Cell 122 (2005) 195–207. [10] O.V. Moskvin, S. Kaplan, M.A. Gilles-Gonzalez, M. Gomelsky, J. Biol. Chem. 282 (2007) 28740–28748. [11] E.H.S. Sousa, J.R. Tuckerman, A.C.S. Gondim, G. Gonzalez, M.A. Gilles-Gonzalez, Biochemistry 52 (2013) 456–465. [12] J. Roy, S. Sen Santara, A. Adhikari, A. Mukherjee, S. Adak, Arch. Biochem. Biophys. 579 (2015) 85–90. [13] R. Purohit, A. Weichsel, W.R. Montfort, Protein Sci. 22 (2013) 1439–1444. [14] A.E. Bem, N. Velikova, M.T. Pellicer, P. Baarlen, A. Marina, J.M. Wells, ACS Chem. Biol. 10 (2015) 213–224. [15] D.L. Caly, D. Bellini, M.A. Walsh, J.M. Dow, R.P. Ryan, Curr. Pharm. Des. 21 (2015) 12–24. [16] F. Martin, P. Baskaran, X. Ma, P.W. Dunten, M. Schaefer, J.P. Stasch, A. Beuve, F. van den Akker, J. Biol. Chem. 285 (2010) 22651–22657. [17] E. Martin, K. Czarnecki, V. Jayaraman, F. Murad, J. Kincaid, J. Am. Chem. Soc. 127 (2005) 4625–4631. [18] B. Pal, Z.Q. Li, T. Ohta, S. Takenaka, S. Tsuyama, T. Kitagawa, J. Inorg. Biochem. 98 (2004) 824–832. [19] R. Makino, E. Obayashi, N. Homma, Y. Shiro, H. Hori, J. Biol. Chem. 278 (2003) 11130–11137. [20] A. Friebe, G. Schultz, D. Koesling, EMBO J. 15 (1996) 6863–6868. [21] A. Friebe, D. Koesling, Mol. Pharmacol. 53 (1998) 123–127. [22] T. Munzel, S. Steven, A. Daiber, Vasc. Pharmacol. 63 (2014) 105–113. [23] A.D. Ostrowski, S.J. Deakin, B. Azhar, T.W. Miller, N. Franco, M.M. Cherney, A.J. Lee, J.N. Burstyn, J.M. Fukuto, I.L. Megson, P.C. Ford, J. Med. Chem. 53 (2010) 715–722. [24] E. Tfouni, F.G. Doro, L.E. Figueiredo, J.C.M. Pereira, G. Metzker, D.W. Franco, Curr. Med. Chem. 17 (2010) 3643–3657. [25] F.O.N. Silva, M.C.L. Candido, A.K.M. Holanda, I.C.N. Diogenes, E.H.S. Sousa, L.G.F. Lopes, J. Inorg. Biochem. 105 (2011) 624–629. [26] E.H. Silva Sousa, L.A. Ridnour, F.S. Gouveia Jr., C.D. Silva da Silva, D.A. Wink, L.G. de Franca Lopes, P.J. Sadler, ACS Chem. Biol. 11 (2016) 2057–2065. [27] C.L. Bianco, J.P. Toscano, M.D. Bartberger, J.M. Fukuto, Arch. Biochem. Biophys. (2016), http://dx.doi.org/10.1016/j.abb.2016.08.014. [28] N.L. Fry, P.K. Mascharak, Acc. Chem. Res. 44 (2011) 289–298. [29] R.G. de Lima, B.R. Silva, R.S. da Silva, L.M. Bendhack, Molecules 19 (2014) 9628–9654. [30] L. Staurengo-Ferrari, S.S. Mizokami, J.J. Silva, F.O.N. da Silva, E.H.S. Sousa, L.G. da Franca, M.L. Matuoka, S.R. Georgetti, M.M. Baracat, R. Casagrande, W.R. Pavanelli, W.A. Verri, Pharmacol. Biochem. Behav. 105 (2013) 157–165.

19

[31] M.W. Campelo, R.B. Oria, L.G. Lopes, G.A. Brito, A.A. Santos, R.C. Vasconcelos, F.O. Silva, B.N. Nobrega, M.T. Bento-Silva, P.R. Vasconcelos, Neurochem. Res. 37 (2012) 749–758. [32] A.P. Santana, B.M. Tavares, L.T. Lucetti, F.S. Gouveia Jr., R.A. Ribeiro, P.M. Soares, E.H. Sousa, L.G. Lopes, J.V. Medeiros, M.H. Souza, Nitric Oxide 45 (2015) 35–42. [33] M.W. Campelo, A.P. Campelo, L.G. Lopes, A.A. Santos, S.B. Guimaraes, P.R. Vasconcelos, Acta Cir. Bras. 26 (Suppl. 1) (2011) 57–59. [34] J.J. Silva, P.M. Guedes, A. Zottis, T.L. Balliano, F.O. Nascimento Silva, L.G. Franca Lopes, J. Ellena, G. Oliva, A.D. Andricopulo, D.W. Franco, J.S. Silva, Br. J. Pharmacol. 160 (2010) 260–269. [35] M. Follmann, N. Griebenow, M.G. Hahn, I. Hartung, F.J. Mais, J. Mittendorf, M. Schafer, H. Schirok, J.P. Stasch, F. Stoll, A. Straub, Angew. Chem. Int. Ed. Eng. 52 (2013) 9442–9462. [36] J.P. Stasch, C. Alonso-Alija, H. Apeler, K. Dembowsky, A. Feurer, T. Minuth, E. Perzborn, M. Schramm, A. Straub, Br. J. Pharmacol. 135 (2002) 333–343. [37] D.S. Sa, A.F. Fernandes, C.D. Silva, P.P. Costa, M.C. Fonteles, N.R. Nascimento, L.G. Lopes, E.H. Sousa, Dalton Trans. 44 (2015) 13633–13640. [38] W.R. Montfort, J.A. Wales, A. Weichsel, Antioxid. Redox Signal. (2016), http://dx. doi.org/10.1089/ars.2016.6693. [39] V. Kumar, F. Martin, M.G. Hahn, M. Schaefer, J.S. Stamler, J.P. Stasch, F. van den Akker, Biochemistry 52 (2013) 3601–3608. [40] I.G. Sharina, M. Sobolevsky, A. Papakyriakou, N. Rukoyatkina, G.A. Spyroulias, S. Gambaryan, E. Martin, Br. J. Pharmacol. 172 (2015) 2316–2329. [41] L.N. Miller, M. Nakane, G.C. Hsieh, R. Chang, T. Kolasa, R.B. Moreland, J.D. Brioni, Life Sci. 72 (2003) 1015–1025. [42] T. Nakai, N.R. Perl, T.C. Barden, A. Carvalho, A. Fretzen, P. Germano, G.Y. Im, H. Jin, C. Kim, T.W. Lee, K. Long, J. Moore, J.M. Rohde, R. Sarno, C. Segal, E.O. Solberg, J. Tobin, D.P. Zimmer, M.G. Currie, ACS Med. Chem. Lett. 7 (2016) 465–469. [43] A. Berry, T. Bosanac, J.D. Ginn, T.D. Hopkins, S. Schlyer, F. Soleymanzadeh, J. Westbrook, M. Yu, Z. Zhang, in: U.S. patent (Ed.), USA, WO2013/0203729 A1, 2013. [44] C.M. Boustany-Kari, P.C. Harrison, H. Chen, K.A. Lincoln, H.S. Qian, H. Clifford, H. Wang, X. Zhang, K. Gueneva-Boucheva, T. Bosanac, D. Wong, R.M. Fryer, J.G. Richman, C. Sarko, S.S. Pullen, J. Pharmacol. Exp. Ther. 356 (2016) 712–719. [45] M. Chrominski, L. Banach, M. Karczewski, K.O. Proinsias, I. Sharina, D. Gryko, E. Martin, J. Med. Chem. 56 (2013) 7260–7277. [46] I. Sharina, M. Sobolevsky, M.F. Doursout, D. Gryko, E. Martin, J. Pharmacol. Exp. Ther. 340 (2012) 723–732. [47] M. Giedyk, K.O. Proinsias, S. Kurcon, I. Sharina, E. Martin, D. Gryko, ChemMedChem 9 (2014) 2344–2350. [48] P.M. Schmidt, M. Schramm, H. Schroder, F. Wunder, J.P. Stasch, J. Biol. Chem. 279 (2004) 3025–3032. [49] R. Purohit, B.G. Fritz, J. The, A. Issaian, A. Weichsel, C.L. David, E. Campbell, A.C. Hausrath, L. Rassouli-Taylor, E.D. Garcin, M.J. Gage, W.R. Montfort, Biochemistry 53 (2014) 101–114. [50] D. Fernandes, R. Sordi, L.K. Pacheco, G.M. Nardi, B.T. Heckert, C.G. Villela, A.R. Lobo, C. Barja-Fidalgo, J. Assreuy, J. Pharmacol. Exp. Ther. 328 (2009) 991–999. [51] J. Loscalzo, J. Stamler, Use of guanylate cyclase inhibitors in the treatment of shock, Brigham And Women's Hospital, Patent EP0639076 A1, 1995. [52] J. Olesen, Neurotherapeutics 7 (2010) 183–190. [53] K.Y. Tseng, A. Caballero, A. Dec, D.K. Cass, N. Simak, E. Sunu, M.J. Park, S.R. Blume, S. Sammut, D.J. Park, A.R. West, PLoS One 6 (2011), e27187. [54] E.S. Buys, L.R. Potter, L.R. Pasquale, B.R. Ksander, Front. Mol. Neurosci. 7 (2014). [55] A.H. Krauss, Use of guanylate cyclase inhibitors for reducing intraocular pressure, Allergan, Patent PCT/US1995/015869, 1995 link: http://www.google.com.na/patents/WO1996017608A1?cl=da. [56] Z.-T. Xia, Combination products of a guanylate cyclase inhibitor and a local anesthetic for pain relief, Linden Biotechnology Inc, Xia Zhi Tao, Patent PCT/US1999/ 027498, 2000. [57] K. Kajiya, M. Detmar, Method for inhibiting lymphangionesis and inflammation, Patent US 11/953,626, Shiseido Company, Ltd., Eth Zurich, 2008 link: https://www. google.com.na/patents/US20080176851?dq=US+11/953,626&hl=pt-BR&sa= X&ved=0ahUKEwj929D3_7XQAhUGiJAKHaX0CKkQ6AEIHDAA. [58] S. Majumder, M. Rajaram, A. Muley, H.S. Reddy, K.P. Tamilarasan, G.K. Kolluru, S. Sinha, J.H. Siamwala, R. Gupta, R. Ilavarasan, S. Venkataraman, K.C. Sivakumar, S. Anishetty, P.G. Kumar, S. Chatterjee, Br. J. Pharmacol. 158 (2009) 1720–1734. [59] A. Mulsch, R. Busse, S. Liebau, U. Forstermann, J. Pharmacol. Exp. Ther. 247 (1988) 283–288. [60] E.A. Dierks, J.N. Burstyn, Arch. Biochem. Biophys. 351 (1998) 1–7. [61] S.P. Olesen, J. Drejer, O. Axelsson, P. Moldt, L. Bang, J.E. Nielsen-Kudsk, R. Busse, A. Mulsch, Br. J. Pharmacol. 123 (1998) 299–309. [62] A. Schrammel, S. Behrends, K. Schmidt, D. Koesling, B. Mayer, Mol. Pharmacol. 50 (1996) 1–5. [63] J. Garthwaite, E. Southam, C.L. Boulton, E.B. Nielsen, K. Schmidt, B. Mayer, Mol. Pharmacol. 48 (1995) 184–188. [64] L.S. Hoffmann, P.M. Schmidt, Y. Keim, C. Hoffmann, H.H. Schmidt, J.P. Stasch, PLoS One 6 (2011), e23596. [65] Y. Zhao, P.E. Brandish, M. Di Valentin, J.P. Schelvis, G.T. Babcock, M.A. Marletta, Biochemistry 39 (2000) 10848–10854. [66] D.N. Kosarikov, J.M. Lee, V.N. Uversky, N. Counts Gerber, J. Inorg. Biochem. 87 (2001) 267–276. [67] K. Spyridonidou, M. Fousteris, M. Antonia, A. Chatzianastasiou, A. Papapetropoulos, S. Nikolaropoulos, Bioorg. Med. Chem. Lett. 19 (2009) 4810–4813. [68] M. von Wantoch Rekowski, A. Pyriochou, N. Papapetropoulos, A. Stossel, A. Papapetropoulos, A. Giannis, Bioorg. Med. Chem. 18 (2010) 1288–1296. [69] A.A. Berezin, P.A. Koutentis, M.J. Manos, Tetrahedron 67 (2011) 5437–5443.

20

E.H.S. Sousa et al. / Journal of Inorganic Biochemistry 167 (2017) 12–20

[70] F. Mota, P. Gane, K. Hampden-Smith, C.K. Allerston, J. Garthwaite, D.L. Selwood, Bioorg. Med. Chem. 23 (2015) 5303–5310. [71] K.P. White, P. Hurban, T. Watanabe, D.S. Hogness, Science 276 (1997) 114–117. [72] L. Yin, N. Wu, J.C. Curtin, M. Qatanani, N.R. Szwergold, R.A. Reid, G.M. Waitt, D.J. Parks, K.H. Pearce, G.B. Wisely, M.A. Lazar, Science 318 (2007) 1786–1789. [73] S. Raghuram, K.R. Stayrook, P.X. Huang, P.M. Rogers, A.K. Nosie, D.B. McClure, L.L. Burris, S. Khorasanizadeh, T.P. Burris, F. Rastinejad, Nat. Struct. Mol. Biol. 14 (2007) 1207–1213. [74] E.L. Carter, N. Gupta, S.W. Ragsdale, J. Biol. Chem. 291 (2016) 2196–2222. [75] Q.J. Meng, A. McMaster, S. Beesley, W.Q. Lu, J. Gibbs, D. Parks, J. Collins, S. Farrow, R. Donn, D. Ray, A. Loudon, J. Cell Sci. 121 (2008) 3629–3635. [76] D. Grant, L. Yin, J.L. Collins, D.J. Parks, L.A. Orband-Miller, G.B. Wisely, S. Joshi, M.A. Lazar, T.M. Willson, W.J. Zuercher, ACS Chem. Biol. 5 (2010) 925–932. [77] L.A. Solt, Y. Wang, S. Banerjee, T. Hughes, D.J. Kojetin, T. Lundasen, Y. Shin, J. Liu, M.D. Cameron, R. Noel, S.H. Yoo, J.S. Takahashi, A.A. Butler, T.M. Kamenecka, T.P. Burris, Nature 485 (2012) 62–68. [78] R.P. Trump, S. Bresciani, A.W. Cooper, J.P. Tellam, J. Wojno, J. Blaikley, L.A. OrbandMiller, J.A. Kashatus, M. Boudjelal, H.C. Dawson, A. Loudon, D. Ray, D. Grant, S.N. Farrow, T.M. Willson, N.C. Tomkinson, J. Med. Chem. 56 (2013) 4729–4737. [79] Y. Shin, R. Noel, S. Banerjee, D. Kojetin, X.Y. Song, Y.J. He, L. Lin, M.D. Cameron, T.P. Burris, T.M. Kamenecka, Bioorg. Med. Chem. Lett. 22 (2012) 4413–4417. [80] R. Noel, X.Y. Song, Y. Shin, S. Banerjee, D. Kojetin, L. Lin, C.H. Ruiz, M.D. Cameron, T.P. Burris, T.M. Kamenecka, Bioorg. Med. Chem. Lett. 22 (2012) 3739–3742. [81] S. Banerjee, Y. Wang, L.A. Solt, K. Griffett, M. Kazantzis, A. Amador, B.M. El-Gendy, S. Huitron-Resendiz, A.J. Roberts, Y. Shin, T.M. Kamenecka, T.P. Burris, Nat. Commun. 5 (2014) 5759. [82] B.K. He, Z. Chen, Curr. Drug Metab. 17 (2016) 503–512. [83] D. Kojetin, Y. Wang, T.M. Kamenecka, T.P. Burris, ACS Chem. Biol. 6 (2011) 131–134. [84] E. Matta-Camacho, S. Banerjee, T.S. Hughes, L.A. Solt, Y. Wang, T.P. Burris, D.J. Kojetin, J. Biol. Chem. 289 (2014) 20054–20066. [85] E.J. Capra, M.T. Laub, Annu. Rev. Microbiol. 66 (2012) 325–347. [86] R.K. Gupta, T.S. Thakur, G.R. Desiraju, J.S. Tyagi, J. Med. Chem. 52 (2009) 6324–6334. [87] S. Dhingra, K. Kaur, N.K. Taneja, J.S. Tyagi, FEMS Microbiol. Lett. 330 (2012) 66–71. [88] K. Kaur, N.K. Taneja, S. Dhingra, J.S. Tyagi, BMC Microbiol. 14 (2014) 195. [89] D.A. Rasko, C.G. Moreira, D.R. Li, N.C. Reading, J.M. Ritchie, M.K. Waldor, N. Williams, R. Taussig, S.G. Wei, M. Roth, D.T. Hughes, J.F. Huntley, M.W. Fina, J.R. Falck, V. Sperandio, Science 321 (2008) 1078–1080. [90] K.E. Wilke, S. Francis, E.E. Carlson, ACS Chem. Biol. 10 (2015) 328–335. [91] J.E. Foster, Q. Sheng, J.R. McClain, M. Bures, T.I. Nicas, K. Henry, M.E. Winkler, R. Gilmour, Microbiology-Sgm 150 (2004) 885–896. [92] R. Gilmour, J.E. Foster, Q. Sheng, J.R. McClain, A. Riley, P.M. Sun, W.L. Ng, D.L. Yan, T.I. Nicas, K. Henry, M.E. Winkler, J. Bacteriol. 187 (2005) 8196–8200. [93] J.J. van Rensburg, K.R. Fortney, L. Chen, A.J. Krieger, B.P. Lima, A.J. Wolfe, B.P. Katz, Z.Y. Zhang, S.M. Spinola, Antimicrob. Agents Chemother. 59 (2015) 3789–3799. [94] Y.T. Tang, R. Gao, J.J. Havranek, E.A. Groisman, A.M. Stock, G.R. Marshall, Chem. Biol. Drug Des. 79 (2012) 1007–1017. [95] K.E. Wilke, S. Francis, E.E. Carlson, J. Am. Chem. Soc. 134 (2012) 9150–9153.

[96] H.K. Carlson, L. Plate, M.S. Price, J.J. Allen, K.M. Shokat, M.A. Marletta, Anal. Biochem. 397 (2010) 139–143. [97] S. Yamada, H. Nakamura, E. Kinoshita, E. Kinoshita-Kikuta, T. Koike, Y. Shiro, Anal. Biochem. 360 (2007) 160–162. [98] U. Romling, M.Y. Galperin, M. Gomelsky, Microbiol. Mol. Biol. Rev. 77 (2013) 1–52. [99] R. Hengge, A. Grundling, U. Jenal, R. Ryan, F. Yildiz, J. Bacteriol. 198 (2016) 15–26. [100] D.G. Ha, G.A. O'Toole, Microbiol. Spectr. 3 (2015) (MB-0003-2014). [101] L. Hall-Stoodley, P. Stoodley, Cell. Microbiol. 11 (2009) 1034–1043. [102] H.C. Flemming, J. Wingender, U. Szewzyk, P. Steinberg, S.A. Rice, S. Kjelleberg, Nat. Rev. Microbiol. 14 (2016) 563–575. [103] D. Kalia, G. Merey, S. Nakayama, Y. Zheng, J. Zhou, Y. Luo, M. Guo, B.T. Roembke, H.O. Sintim, Chem. Soc. Rev. 42 (2013) 305–341. [104] K. Sambanthamoorthy, R.E. Sloup, V. Parashar, J.M. Smith, E.E. Kim, M.F. Semmelhack, M.B. Neiditch, C.M. Waters, Antimicrob. Agents Chemother. 56 (2012) 5202–5211. [105] A.L. Chang, J.R. Tuckerman, G. Gonzalez, R. Mayer, H. Weinhouse, G. Volman, D. Amikam, M. Benziman, M.A. Gilles-Gonzalez, Biochemistry 40 (2001) 3420–3426. [106] X. Wan, J.R. Tuckerman, J.A. Saito, T.A. Freitas, J.S. Newhouse, J.R. Denery, M.Y. Galperin, G. Gonzalez, M.A. Gilles-Gonzalez, M. Alam, J. Mol. Biol. 388 (2009) 262–270. [107] J.R. Tuckerman, G. Gonzalez, M.A. Gilles-Gonzalez, J. Mol. Biol. 407 (2011) 633–639. [108] D.P. Arora, S. Hossain, Y. Xu, E.M. Boon, Biochemistry 54 (2015) 3717–3728. [109] D. Antoniani, P. Bocci, A. Maciag, N. Raffaelli, P. Landini, Appl. Microbiol. Biotechnol. 85 (2010) 1095–1104. [110] K. Sambanthamoorthy, C.Y. Luo, N. Pattabiraman, X.R. Feng, B. Koestler, C.M. Waters, T.J. Palys, Biofouling 30 (2014) 17–28. [111] O.J. Lieberman, M.W. Orr, Y. Wang, V.T. Lee, ACS Chem. Biol. 9 (2014) 183–192. [112] S. Fernicola, A. Paiardini, G. Giardina, G. Rampioni, L. Leoni, F. Cutruzzola, S. Rinaldo, J. Bacteriol. 198 (1) (2015) 147–156. [113] S.M. Ching, W.J. Tan, K.L. Chua, Y.L. Lam, Bioorg. Med. Chem. 18 (2010) 6657–6665. [114] J. Zhou, S. Watt, J.X. Wang, S. Nakayama, D.A. Sayre, Y.F. Lam, V.T. Lee, H.O. Sintim, Bioorg. Med. Chem. 21 (2013) 4396–4404. [115] J.T. Henry, S. Crosson, Annu. Rev. Microbiol. 65 (2011) 261–286. [116] R. Zoraghi, J.D. Corbin, S.H. Francis, Mol. Pharmacol. 65 (2004) 267–278. [117] J.E. Schultz, T. Dunkern, E. Gawlitta-Gorka, G. Sorg, Handb. Exp. Pharmacol. (2011) 151–166. [118] R. Jager, C. Russwurm, F. Schwede, H.G. Genieser, D. Koesling, M. Russwurm, J. Biol. Chem. 287 (2012) 1210–1219. [119] M. Lamothe, F.J. Chang, N. Balashova, R. Shirokov, A. Beuve, Biochemistry 43 (2004) 3039–3048. [120] B. Pal, T. Kitagawa, Biochem. Biophys. Res. Commun. 397 (2010) 375–379. [121] B.G. Bunney, W.E. Bunney, Biol. Psychiatry 73 (2013) 1164–1171. [122] F.C. Kelleher, A. Rao, A. Maguire, Cancer Lett. 342 (2014) 9–18. [123] A.R. Ozburn, E. Falcon, A. Twaddle, A.L. Nugent, A.G. Gillman, S.M. Spencer, R.N. Arey, S. Mukherjee, J. Lyons-Weiler, D.W. Self, C.A. McClung, Biol. Psychiatry 77 (2015) 425–433. [124] N. Singh, A. Kumar, Antioxid. Redox Signal. 22 (2015) 603–613. [125] Y.Z. Hong, X.D. Zhou, H.H. Fang, D. Yu, C.Y. Li, B.L. Sun, Tuberculosis 93 (2013) 625–634.