RNA polymerase

Resistance to rifampicin: a review

Posted by

0


Abstract

Resistance to rifampicin (RIF) is a broad subject covering not just the mechanism of clinical resistance, nearly always due to a genetic change in the β subunit of bacterial RNA polymerase (RNAP), but also how studies of resistant polymerases have helped us understand the structure of the enzyme, the intricacies of the transcription process and its role in complex physiological pathways. This review can only scratch the surface of these phenomena. The identification, in strains of Escherichia coli, of the positions within β of the mutations determining resistance is discussed in some detail, as are mutations in organisms that are therapeutic targets of RIF, in particular Mycobacterium tuberculosis. Interestingly, changes in the same three codons of the consensus sequence occur repeatedly in unrelated RIF-resistant (RIFr) clinical isolates of several different bacterial species, and a single mutation predominates in mycobacteria. The utilization of our knowledge of these mutations to develop rapid screening tests for detecting resistance is briefly discussed. Cross-resistance among rifamycins has been a topic of controversy; current thinking is that there is no difference in the susceptibility of RNAP mutants to RIF, rifapentine and rifabutin. Also summarized are intrinsic RIF resistance and other resistance mechanisms.

Introduction

In celebrating the life of Professor Piero Sensi and his discovery of rifampicin (RIF), also known as rifampin, we have recognized the importance of this drug in treating infectious disease, in particular tuberculosis. In thinking about resistance, we should keep in mind that it is not just an inconvenient clinical phenomenon. The study of RIF and of resistant mutants in different bacterial species has had a key role in the elucidation of the structure and function of bacterial DNA-dependent RNA polymerase (RNAP) and its involvement in the modulation of complex physiological pathways.

Early in the study of the rifamycins, the occurrence, in cultures of susceptible organisms, of spontaneous one-step mutations to high-level resistance became apparent, initially as a ‘skipped tube’ phenomenon in MIC determinations. The literature dealing with resistance to RIF is extensive and some of the early publications are not readily available. Areas of study include: resistance mechanisms (primarily acquired resistance because of mutation in the rpoB gene, which encodes the β subunit of RNAP); identification of the amino-acid changes in β associated with resistance in laboratory strains and clinical isolates; the practice of combined antimicrobial therapy to limit the emergence of resistance; cross-resistance of RIF with other RNAP inhibitors; pleiotropic effects of RIF-resistant (RIFr) enzymes on gene expression; and development of rapid tests for the detection of resistance in Mycobacterium tuberculosis. Some of these aspects will be touched upon only briefly.

Primary mechanism of resistance to RIF: mutations affecting RNAP subunit β

Clinical and laboratory studies of RIF initially targeted a broad spectrum of susceptible bacteria, and resistance was reported in laboratory studies and emerging in patients who received monotherapy with RIF. Resistance rates to rifamycins, determined in the laboratory, have ranged from 10−10 to 10−7, depending on the organism and the methodology used.1, 2, 3, 4 RIF resistance was reported in different Gram-negative urinary tract pathogens, in vitro and in treated patients;5 in gonococci and meningococci in the laboratory and the clinic;2, 6, 7 and in tuberculosis patients who failed therapy when RIF was the only active drug administered.8 When treating tuberculosis and other diseases, RIF is almost always combined with other active antimicrobials to prevent the emergence of resistance.

Shortly after RIF was shown to inhibit transcription, cell-free assays demonstrated that resistance, at least in laboratory strains, was related to a change in the properties of the polymerase: RNAP from resistant bacteria was itself resistant in these assays, and it did not bind RIF.9, 10, 11, 12 That the target within the enzyme was the β subunit, one of its two largest polypeptides, was first suggested by the observation that in a RIFr Escherichia coli strain the migration of this subunit in polyacrylamide gel electrophoresis was altered.13 Subsequently, separation and mixed reconstitution of enzymatically active core enzyme, using subunits from susceptible and RIFr strains of E. coli14 and Bacillus subtilis,15 provided a more direct demonstration that resistance was determined by a change in the β subunit.

The spectrum and localization within rpoB of RIFr mutations, in both clinical isolates and strains selected in the laboratory, has been studied in a number of species, including E. coli, M. tuberculosis and Staphylococcus aureus. Although E. coli is generally not a therapeutic target for RIF, it is a model species for genetic and physiological studies, and there were detailed investigations of transcription initiation and termination in this organism. Nearly complete saturation of the RIFr mutational spectrum in E. coli and mapping and sequencing of the mutations in rpoB was achieved in the 1980s, largely by the efforts of Jin and Gross.16 The mapping of RIFr mutations in other organisms has most often been reported with alignment to the consensus numbering scheme of E. coli RNAP, facilitating comparison across species. As RNAP is highly conserved among eubacteria, it is not surprising that the sites of RIFr mutations are also conserved. Mutations affecting residues 516, 526 and 531 of the β consensus sequence predominate in resistant clinical isolates of a number of bacterial species. In the discussion that follows, only strains with a single mutation in rpoB that determines an amino-acid change and a RIFr phenotype are considered; for this reason, conclusions about mutation sites are not always identical to those of the authors.

RIF resistance in E. coli

Complete sequencing of rpoB posed a challenge as, with 1342 amino acids in E. coli, β is the second largest polypeptide in the bacterial cell. In the 1970s, refinements in the cloning and sequencing of overlapping DNA restriction fragments enabled the complete determination and alignment of the nucleotide sequence of rpoB with the amino-acid sequence of the β subunit from a RIFr strain of E. coli.17 Ovchinnikov et al.18 then utilized a susceptible strain to sequence the region of rpoB to which the RIFr mutation had been localized genetically, and identified it as an aspartic acid to valine change at residue 516 of the polypeptide, corresponding to an A:T to T:A transversion in the corresponding codon.

Jin and Gross16 constructed an isogenic set of mutants in E. coli K-12 derived from 42 RIFr strains from their own laboratory (both spontaneous and UV induced) and other sources. As the goal of generating and studying RIFr mutations in this organism was to understand the structure of the β subunit and its functional interaction within RNAP, a broad array of RIFr mutations were included. Thus, although they would likely have no relationship to those emerging in the clinic, the E. coli RIFr mutations included some that determined temperature-dependent and dominant phenotypes, as well as defects in transcription. Mapping was achieved by transformation with plasmids from a susceptible strain having various length deletions of rpoB; the region to which each mutation mapped was sequenced, identifying 17 unique alleles (excluding mutants with more than one change), a few of which had also been described by others, as had two additional unique mutations. Most were point mutations, although there were also three deletions of one to five codons and one insertion of two codons. Mutations specifying two different amino-acid changes were found in each of three codons. A number of the unique alleles were isolated several times, both by Jin and Gross and others. In cell-free RNAP assays, RIF 50% inhibition concentrations (IC50s) for the mutant enzymes ranged from 10 to >10 000 times the IC50 for the enzyme from the isogenic susceptible strain. These values roughly paralleled the concentrations that inhibited the growth of the mutant strains. In a later study, the extent of binding of RIF to the RNAPs from 12 of the mutants was also shown to correlate with the IC50s and with growth inhibition.19 The mutations mapped in the center of the rpoB gene, in three clusters: cluster I (covering amino acids 507–533) included 13 of the 17 RIFr alleles from this study, as well as a deletion mutant mapped by others; three of the mutations were in cluster II (amino acids 563–572); and one was at amino-acid 687 (cluster III). The segment of rpoB encompassing these clusters was initially called the ‘RIF region’, but is also known as the RIF resistance-determining region (RRDR). Another point mutation identified by others mapped outside of the RRDR at amino-acid 146. A few other unique RIFr mutations have since been described in E. coli. Landick et al.20 utilized bisulfite-induced cytosine deamination to mutagenize selected regions of rpoB and screened for termination-altering mutations. A number of the selected mutants were RIFr, and one had a single-amino-acid change within the RRDR that had not been previously reported. Another RIFr mutation within this region is cited by Severinov et al.21

As discussed by Jin and Gross,16 it appeared likely that the different regions of the β subunit in which RIFr mutations occur cooperate, within the core enzyme, to form the RIF-binding site. Approaches to the topology of the active center of the E. coli enzyme and its interaction with the RIF-binding site have been largely indirect; for example, utilizing the binding of different rifamycin derivatives, or of antibodies raised to rifamycin–albumin adducts, or of RIF–nucleotide adducts,22, 23 as well as molecular modeling based on the amino-acid sequence. By cross-linking the polymerase–promoter complex to β, followed by limited proteolysis and chemical degradation, it was demonstrated that cluster I of the RRDR forms part of the active center of the enzyme.24 E. coli RNAP has been crystallized only very recently.25 Although the enzyme is naturally partially resistant to RIF and has limitations in the co-structure of its binding site, studies with crystallized RNAP from Thermus aquaticus showed that the RIF-binding pocket is in the fork domain, part of the active center, and established that most RIFr mutations map to this region.26

RIF resistance in M. tuberculosis and other mycobacteria

Mapping the mutations found in clinical isolates has been critical to the development of rapid methods to detect resistance in patients. Standard susceptibility testing of slow-growing species generally requires 4 weeks of culture (M. tuberculosis) and as long as 1 year in an animal infection model (Mycobacterium leprae). An important finding was the predominance of a single mutation, Ser531Leu, in different studies. Only a selection of the many publications describing RIFr mutations in mycobacteria will be discussed.

Using cell-free RNAP assays, Yamada et al.27 had demonstrated that the RIFr phenotype of two clinical isolates of M. tuberculosis was determined by resistance of their enzymes. Telenti et al.28 determined the amino-acid changes in a collection of 66 RIFr clinical isolates from different geographical areas. They identified 15 distinct mutations in 8 codons within a segment of rpoB that aligned with the RRDR region of E. coli. In a set of 128 isolates from the United States, Kapur et al.29 identified a number of additional RIFr mutations in this region; interestingly, some of the mutations were identified in both studies but at different frequencies, suggesting geographic variation. Data from these and other studies, for a total of 307 RIFr isolates, were compiled by Musser.30 Twenty-eight unique amino-acid changes corresponded to point mutations in 12 different codons, 2 insertions of 1–2 codons and 7 deletions of 1–3 codons; all of them mapped to the region corresponding to cluster I of the E. coli RRDR. In later studies, four additional amino-acid changes and a three-codon deletion, all but one mapping to codons previously identified, were sequenced in clinical isolates from Japan and China.31, 32, 33 Another study of Japanese isolates identified two mutations affecting the N-terminal region of β (one of them corresponding to residue 146, previously identified in E. coli) and two mutations specifying low-level RIF resistance (MIC=12.5 μg ml–1) in RRDR cluster III;34 one of the latter corresponded to the previously identified E. coli mutation in this cluster. Among 63 rpoB clinical isolates from Germany, Heep et al.35 described an additional amino-acid change at residue 526.

Among the point mutations reviewed by Musser,30 there were three major hotspots, each with multiple amino-acid changes: residue 516 (four different amino acids in 25 clinical isolates); residue 526 (eight amino acids in 111 strains); and residue 531 (four amino acids in 132 strains). At each of these loci, a single-amino-acid change predominated, with Ser531Leu (TCG to TTG) alone occurring in 128 isolates. Although all but two of the RIFr mutations identified in M. tuberculosis clinical isolates mapped to codons that aligned with E. coli mutations, Ser531Leu, the single most frequently identified amino-acid change in mycobacteria, was not, although other amino-acid changes occurred at the same residue. This is not of biological significance, however, as this amino-acid replacement in E. coli would require two nucleotide changes (TCT to TTG or TTA) as compared with the single TCG to TTG transition in the M. tuberculosis codon.

In a laboratory study, using the Luria-Delbrück fluctuation test, Morlock et al.36 generated 64 spontaneous, independent RIFr mutations in M. tuberculosis H37Rv and identified eight different point mutations, one insertion and one deletion. All of them mapped to codons identified by mutations in clinical isolates, 20 of them in consensus residue 526 and 41 in residue 531 (39 of them Ser to Leu, TCG to TTG). In a selection experiment in which a few cultures were grown in the presence of RIF, all (non-independent) mutations occurred at residues 526 and 531, with Ser531Leu again predominating.37

In other mycobacterial species smaller numbers of RIFr isolates have been available for sequencing. Honore and Cole38 mapped eight of nine RIFr mutations in clinical isolates of Mycobacterium leprae to the residue corresponding to 531; six of them were Ser531Leu. Williams et al.39 also identified the Ser531Leu mutation in four strains of M. leprae analyzed, and in one Mycobacterium africanum strain and one Mycobacterium avium. Another M. avium isolate had a Ser531Trp mutation. In Mycobacterium kansasii, five RIFr clinical isolates and one laboratory mutant had mutations in codons 513, 526 or 531.40

Knowing the prevalence of different RIFr mutations in M. tuberculosis made it possible to design rapid nucleic acid amplification based molecular tests to detect the organism in patients with suspected infection and to identify resistance in patient isolates. A large number of methods were explored and tested for correlation of the results with those of standard susceptibility tests. In 2013, the Food and Drug Administration approved a commercial PCR-based test to detect the DNA of M. tuberculosis, as well as RIFr mutations, in sputum.41

RIF resistance in S. aureus

Most reports of rpoB mutations in this organism have used S. aureus numbering, in some cases with the consensus codon numbers provided; however, the present discussion will be based only on the E. coli numbering system. In 1979, Morrow and Harmon42 demonstrated that laboratory-generated rifamycin-resistant mutations in S. aureus were chromosomal and affected the ability of the antibiotics to inhibit RNAP activity in cell-free transcription assays. RIFr mutations in paired clinical isolates (susceptible and resistant strains from the same patients) and in laboratory strains of S. aureus were mapped by Aubry-Damon et al.43 A MIC histogram divided these strains into three categories: susceptible, low-level resistant and high-level resistant (MICs of ⩽0.5, 1–4 and ⩾8 μg ml–1, respectively). All 17 RIFr strains sequenced had single mutations, which included 8 distinct changes at 7 sites; 6 of the sites were within the consensus cluster I of the RRDR and corresponded to mutant codons identified in E. coli. Several additional sites of single mutations, two of them in cluster II and the others in cluster I, were reported by Wichelhaus et al.44; all of them corresponded to mutational sites in E. coli and one was the Ser531Leu mutation that predominates in M. tuberculosis. Ser531Leu was also identified by others in S. aureus RIFr clinical and laboratory strains, as was His526Asn; a few additional amino-acid changes have been described, most of them in the same codons identified previously in both E. coli and S. aureus.45, 46, 47, 48

RIF resistance in other species

There is extensive literature on RIF resistance in a number of bacterial species. Mutational changes in the β subunit of RNAP will be briefly discussed only in organisms for which RIF is commonly prescribed. Resistance in these species is frequently associated with mutations in rpoB codons 516, 526 or 531 of the consensus sequence.

RIF has been utilized clinically for prophylaxis in individuals exposed to Neisseria meningitidis, for prophylaxis and treatment of invasive Haemophilus influenzae, and for treatment of β-lactam-resistant Streptococcus pneumoniae infections. Fourteen RIFr strains of N. meningitidis from Italy and the United Kingdom had rpoB mutations in cluster I of the RRDR corresponding to Asp516Val, His526Tyr or His526Asp.49, 50 In clinical isolates of H. influenzae, RIFr single mutations mapped to cluster I of the RRDR, with changes at Asp516 predominating; a strain with intermediate susceptibility had a mutation in cluster II.51 Single-site mutations in invasive isolates of S. pneumoniae from Taiwan were aligned as Asp516Val and His526Tyr.52 Different amino-acid changes at the same two residues, Asp516Glu and His526Asn, were found in RIFr pneumococcal isolates from South Africa.53

Rhodococcus equi, an intracellular organism that causes life-threatening infections in young foals and opportunistic infections in immunocompromised humans, is often treated with RIF combination therapy. RIFr mutations in this species have included Ser531Leu/Trp and several different amino-acids substitutions at consensus codon 526.54, 55

Intrinsic and polymerase-independent resistance to RIF

Low-level RIF resistance in various organisms, including mycobacteria, has been suggested to involve permeability or efflux/influx56 and a plasmid-mediated efflux mechanism has been reported in a strain of Pseudomonas fluorescens.57 These will not be reviewed as their clinical implications are not evident. Relatively high RIF MICs in Enterobacteriaceae and other non-fastidious Gram-negative bacteria, determined by long or abundant outer membrane lipopolysaccharide chains, will also not be discussed.

A few species of bacteria are intrinsically non-susceptible to RIF because of a refractory RNAP. Treponema spp. and other spirochetes, including members of the genera Borrelia and Leptospira, are in this category, as are many strains of soil actinomycetes; resistance in these organisms correlates with the amino-acid naturally present at consensus codon 531 in rpoB: Asn substituting for the Ser of susceptible bacteria.58, 59 Another group of bacteria that are intrinsically non-susceptible to RIF are the mollicutes, which include Mycoplasma, Ureaplasma and Spiroplasma species. Sequencing of the rpoB gene of Spiroplasma citri indicated that the presence of Asp at the consensus residue 526, instead of the His present in susceptible species, is the determinant of resistance; Asp is also present at this residue in various Mycoplasma species.60 Among Rickettsia there is a cluster of naturally RIFr spotted fever group species; the relationship between the rpoB mutation identified and the consensus sequence is not evident.61 A variation on this theme has been reported in the opportunistic pathogen, Nocardia farcinica: the presence of a second gene, homologous to rpoB, termed rpoB2 or rpoBR, which encodes a RIF-refractory β subunit.62 An rpoB2 paralog reportedly occurs in a number of actinomycetes, and in a Nonomuraea sp. the two paralogs are expressed under different physiological conditions; in the latter organism the expression of rpoB2, in stationary phase, is associated with secondary metabolism.63

There are a few examples of inactivation of RIF, mainly in bacterial species that are not its therapeutic targets and associated with low-level resistance These include glucosylation, ribosylation, phosphorylation and decolorization, the latter because of a monooxygenase.64, 65 A monooxygenase has been identified as a secondary resistance mechanism in N. farcinica, revealed when its rpoB2 was deleted.66 The gene for an enzyme (termed ARR-2) capable of ADP-ribosylating RIF has been identified on an integron in a Pseudomonas aeruginosa strain.67

Cross-resistance among RNAP inhibitors

Rifamycins

There are currently five rifamycins marketed in various countries: rifamycin SV (with limited availability); RIF, rifapentine and rifabutin (all three mainly utilized for the systemic treatment of mycobacterial infections) and rifaximin (indicated for travelers’ diarrhea). Rifalazil, a sixth rifamycin has been in development for a number of years for various indications. Although there is clinical evidence that the relapse and acquired resistance rates in tuberculosis patients treated with RIF or rifabutin are similar,68 there have been a number of reports suggesting incomplete cross-resistance between rifabutin, RIF and rifapentine. For example, M. tuberculosis RIFr clinical isolates with the rpoB mutations Asp516Val, Asp516Tyr and Leu533Pro69 and Ser522Leu70 were reported to be susceptible to rifabutin. However, in E. coli, mutant RNAPs with Asp516Val (as well as Asp516Asn), Leu533Pro and a different mutation at residue 522 (Ser to Phe) were resistant to all three rifamycins.19 The Asp516Asn RNAP from E. coli was also resistant to all three compounds in a cell-free transcription assay.19 Complete cross-resistance among RIF, rifapentine and rifabutin was also reported in sequenced rpoB mutants of S. aureus.44 It should be noted that the MICs of rifabutin for the M. tuberculosis isolates in question are at the upper end (0.5 μg ml–1) of what has been considered its susceptibility limit. Current thinking is that this breakpoint is too high and that, in the absence of clinical evidence to the contrary, the isolates in question should be considered resistant to rifabutin.71, 72

Other inhibitors of RNAP

Fidaxomicin (lipiarmycin), a macrocyclic antibiotic, is the only marketed non-rifamycin that inhibits bacterial RNAP; it is not cross-resistant with RIF. It acts at the initiation step of transcription but, unlike RIF, it requires core enzyme plus σ factor for optimal inhibitory activity in cell-free transcription assays.73, 74 Kurabachew et al.75 sequenced rpoB and rpoC from a set of lipiarmycin-resistant strains of M. tuberculosis. They identified two codons in rpoB distal to the RRDR that specified various amino-acid changes in the β subunit and two mutations in rpoC specifying changes in β′.

Sorangicin A, a macrolide polyether, also inhibits the initiation step of transcription. In E. coli, it is partially cross-resistant with RIF.19, 76, 77 Two studies19, 76 examined the effects of known amino-acid changes in the β subunit on the extent to which bacterial growth was inhibited by RIF and sorangicin. There is some difference between the two studies regarding the effect of certain mutations on cross-resistance; however, the two sets of mutants were not identical. The study by Xu, et al.19 also compared the extent of inhibition of transcription by both antibiotics in a cell-free assay; inhibition of the mutant RNAPs correlated with growth inhibition of the mutants.

Streptolydigin, a tetramic acid antibiotic that inhibits elongation of transcripts, is not cross-resistant with RIF.19, 21 Although most mutations to streptolydigin resistance map in rpoB, they are found mainly in the spacer region between clusters I and II of the RRDR.21, 78 Mutations in rpoC, encoding the β′ subunit, have also been identified.79, 80

Pleiotropic effects of RIFr mutations

Transcription is a complex, intricately regulated process in which initiation at specific promoters, pausing and termination involve the transient interaction of RNAP core enzyme with other subunits such as σ factors, small molecules, DNA sequences and the transcript RNA itself. RIF has been an important tool in probing bacterial physiology because RIFr mutations affect a number of phenotypes. As summarized by Jin and Gross,81 some of the mutations in E. coli determine temperature-dependent growth; affect the stability of plasmids, the growth of bacteriophages and susceptibility to other inhibitors; and affect phenotypes associated with mutations in other subunits or enzymes. One important effect of some RIFr mutations is on the expression of termination/anti-termination including: the bacteriophage λ N anti-termination function involved in the transcription of late genes; and the cell’s stringent response that regulates transcription of stable RNAs and other stringently controlled genes and is normally controlled by the ρ termination factor and ppGpp.82, 83, 84

Other examples of effects associated with RIFr mutations are: control of sporulation, germination, cell shape and metabolism in B. subtilis;85, 86, 87 abnormal termination at the tryptophan operon attenuator in E. coli;88 alteration of nutritional requirements in Lactobacillus casei;89 activation of silent genes and upregulation of antibiotic production in some actinomycetes.90 There have also been numerous studies in various organisms of the fitness and virulence of RIFr mutants and of adaptation by means of secondary mutations.

Conclusions

RIF is a valuable antibiotic for the treatment of mycobacterial and other diseases; the emergence of resistance during therapy can generally be avoided with the use of adequate combination therapy. The clinically significant resistance mechanism is mutation within a defined region of the rpoB gene, which encodes the target of RIF, the β subunit of bacterial RNAP. The portion of this sequence defined as cluster I is particularly important for high-level resistance. As a result of the high degree of conservation of RNAP, including this region of β, the mutations that determine resistance are also conserved across species. In RIFr clinical isolates of various organisms, mutations are most often found in three specific codons of the consensus sequence, and the Ser531Leu substitution in β predominates in mycobacteria and some other species. Cross-resistance appears to be complete among the rifamycins currently used to treat mycobacterial diseases. For these reasons, it has been possible to develop PCR-based tests to rapidly identify resistant M. tuberculosis. Intrinsic resistance in a few bacterial genera is also determined by the amino acids naturally present at residues 531 or 526 of the consensus sequence. Partial cross-resistance has been reported to another class of RNAP inhibitor, sorangicin A. RIFr mutations can have profound effects on transcription rate, initiation, pausing and termination. As a result of the central role played by RNAP in the bacterial cell, RIFr mutations often affect a number of physiological processes. RIF itself and RIFr mutations have been of great importance in elucidating the structure and function of RNAP and its complex role in the regulation of gene expression.

Uncovering the Resistance Mechanism of Mycobacterium tuberculosis to Rifampicin Due to RNA Polymerase H451D/Y/R Mutations From Computational Perspective

Posted by

0


Tuberculosis is still one of the top 10 causes of deaths worldwide, especially with the emergence of multidrug-resistant tuberculosis. Rifampicin, as the most effective first-line antituberculosis drug, also develops resistance due to the mutation on Mycobacterium tuberculosis (Mtb) RNA polymerase. Among these mutations, three mutations at position 451 (H451D, H451Y, H451R) are associated with high-level resistance to rifampicin. However, the resistance mechanism of Mtb to rifampicin is still unclear. In this work, to explore the resistance mechanism of Mtb to rifampicin due to H451D/Y/R mutations, we combined the molecular dynamics simulation, molecular mechanics generalized-Born surface area calculation, dynamic network analysis, and residue interactions network analysis to compare the interaction change of rifampicin with wild-type RNA polymerase and three mutants. The results of molecular mechanics generalized-Born surface area calculations indicate that the binding free energy of rifampicin with three mutants decreases. In addition, the dynamic network analysis and residue interaction network analysis show that when H451 was mutated, the interactions of residue 451 with its adjacent residues such as Q438, F439, M440, D441, and S447 disappeared or weakened, increasing the flexibility of binding pocket. At the same time, the disappearance of hydrogen bonds between R613 and rifampicin caused by the flipping of R613 is another important reason for the reduction of binding ability of rifampicin in H451D/Y mutants. In H451R mutant, the mutation causes the binding pocket change too much so that the position of rifampicin has a large movement in the binding pocket. In this study, the resistance mechanism of rifampicin at the atomic level is proposed. The proposed drug-resistance mechanism will provide the valuable guidance for the design of antituberculosis drugs.

Introduction

Tuberculosis (TB), an infectious disease caused by Mycobacterium tuberculosis (Mtb), is the leading cause from a single infectious agent worldwide. Mtb, a pathogenic bacterium species of the family Mycobacteriaceae, can attack the lung of people and spread in the population through the droplets from the throat of an active TB infection patient (Mishra and Surolia, 2018). Millions of people continue to fall sick with TB each year. The 2018 WHO Global Tuberculosis Report (Organization, 2018) estimated 10.0 million new cases of TB and 1.6 million deaths in 2017.

Although it spreads widely, TB is preventable and curable. The 6-month short-course regimen with a combination of four anti-TB drugs (rifampicin, isoniazid, pyrazinamide, and ethambutol) has been used as a standard treatment for active, drug-susceptible TB patients over the past decades (Service and Council, 1981; Chang et al., 2018; Seid et al., 2018; Tiberi et al., 2018). Furthermore, the risk of relapse is generally below 5% reported among drug-susceptible TB patients after treated with standard 6-month regimens in clinical trials (Chang et al., 2006). However, the multidrug-resistant tuberculosis (MDR-TB), which at least resists to both rifampicin and isoniazid, had been emerged in the early 1990s due to multiple factors (He et al., 2008; Sandgren et al., 2009; Ahmad et al., 2018).

Rifampicin, one of the most effective anti-TB drugs, has been used as the first-line treatment in drug-susceptible TB patients, and it is also effective against initial isoniazid resistance (Mitchison and Nunn, 1986). Unfortunately, rifampicin resistance in Mtb arises due to the residues’ mutations on its molecular target, Mycobacterium tuberculosis RNA polymerase (Mtb-RNAP). More than 95% of the rifampin-resistant strains have mutations in a small region defined “rifampicin resistance-determining region” in Mtb-RNAP (Morlock et al., 2000; Zaw et al., 2018). The most common mutation in rifampicin resistance-determining region are S456, H451, and D441, corresponding to S531, H526, and D516 in Escherichia coli, respectively. Studies have shown that 70% of rifampicin-resistant clinical isolates have point mutation in two residues (S456 and H451) (Morlock et al., 2000), and H451 is most usually substituted for Asp (D), Tyr (Y), and Arg (R) (Telenti et al., 1993; Caws et al., 2006; Ma et al., 2006; Chikaonda et al., 2017; Wu and Hilliker, 2017). As early as 1995, the in vitro activity experiment of rifampicin by Bodmer et al. (1995) had been demonstrated that H451D/Y/R mutations could cause high-level resistance to rifampicin. After more than 20 years, the level and frequency of resistance to rifampicin are also increasing.

In 2017, there is about 558,000 new cases of rifampicin-resistant tuberculosis (RR-TB), of which 82% are MDR-TB and about 230,000 deaths from MDR/RR-TB (Organization, 2018). Currently, although MDR/RR-TB can be cured with the second-line drugs (e.g., fluoroquinolone and an injectable aminoglycoside), poor efficiency, high toxicity, and expensive price of these drugs make it still difficult for many MDR-TB patients. In some cases, more severe extensively drug-resistant TB may occur, and it will not respond to the most effective second-line anti-TB drugs (Sotgiu et al., 2015; Jeon, 2017; Tiberi et al., 2018). Obviously, the development of new anti-TB drugs is urgent, and exploring the resistance mechanism of rifampicin is of great significance for the discovery of effective drugs.

In this work, in order to uncover the resistance mechanism of Mtb to rifampicin due to the mutation of Mtb-RNAP at position 451, three independent molecular dynamics (MD) simulations for the wild-type Mtb-RNAP and H451D/Y/R mutants were carried out. Based on the obtained trajectories, the molecular mechanics generalized-Born surface area (MM-GBSA) method was applied to calculate the binding free energy of rifampicin with Mtb-RNAP. Furthermore, dynamic network analysis combined with residue interaction network (RIN) analysis was used to show the detailed changes of interactions among the residues surrounding the binding pocket. With the structural and energy analysis, a possible rifampicin-resistant mechanism was also proposed. Compared with the traditional experimental method, MD simulations can show the intuitive and dynamics interaction change process between rifampicin and Mtb-RNAP due to the point mutation. Together with the energy analysis and the dynamics network analysis, the present study show the essential reason of Mtb-RNAP resistant to rifampicin, which can provide the useful guidance for the further drug design against drug resistance.

Materials and Methods

Systems Preparation

The initial atomic coordinate of the wild-type Mtb-RNAP with rifampicin was obtained from Protein Data Bank (Protein Data Bank ID: 5UHB). The crystal structure of Mtb-RNAP reported by Lin et al. (2017) reveals that Mtb-RNAP is composed of six chains, for the A, B chains encoded by the rpoA gene, the C chain encoded by the rpoB gene (Miller et al., 1994), and the D, E, F chains encoded by rpoC, rpoZ, and rpoD, respectively. Rifampicin binds at the active site of the C chain (shown in Figure 1) and inhibits the DNA-directed RNA synthesis of Mtb (McClure and Cech, 1978; Campbell et al., 2001; Somoskovi et al., 2001). Considering that the speed to simulate the whole Mtb-RNAP (~3,826 residues) is too slow, only the C chain complexed with rifampicin was extracted and used as the initial structure of simulations. Furthermore, the deletion of other chains will make the residues of the interface between the two chains unstable, which is inconsistent with that in the multimer. Thus, to simulate the state of interface in the multimer, some relatively flexible and far from the active site amino acid residues were deleted. The three-dimensional structures of three mutants (H451D/Y/R) were obtained by mutating H451 residue in wild type.

FIGURE 1

Figure 1. (A) The structural overview for Mycobacterium tuberculosis RNA polymerase bound with rifampicin (the protein is shown as cartoon, rifampicin is shown as a stick, the binding pocket of rifampicin is shown as surface). (B) The molecular scheme of the rifampicin.

To generate the force field parameters for the ligand, the Gaussian 09 program (Frisch et al., 2009) was used to optimize the structure of rifampicin and calculate the electrostatic potential at the Hartree–Fock level with 6-31G* basis set. Then, the restraint electrostatic potential protocol (Bayly et al., 1993; Cieplak et al., 1995; Fox and Kollman, 1998) was employed to fit the atomic partial charges. The general amber force field (gaff) (Wang et al., 2004) generated by the antechamber program in the Amber14 package (Case et al., 2014) was applied to describe the ligand. The standard ff99SB force field (Hornak et al., 2006) was used to describe the protein. Then, the LEaP module was applied to add all missing hydrogen atoms and a certain amount of sodium counter-ions to neutralize the unbalanced charges and maintain the systems electro-neutrality. Finally, a rectangular periodic water box of TIP3P (Jorgensen et al., 1983) was added to each system with the water molecules extended 10-Å distance around the complex. The size of the periodic boundary box is 92.6 × 114.4 × 114.9 Å. The whole system has a total of ~100,000 atoms per periodic cell.

Molecular Dynamics Simulations

All MD simulations were performed with Amber14 package (Case et al., 2014). The process of energy minimization, heating, and equilibration was carried out with the Particle Mesh Ewald Molecular Dynamics module. Initially, the energy minimization of each solvated complex includes three steps. For each step, energy minimization was carried out by the steepest descent method for the first 2,500 steps and conjugated gradient method for the subsequent 2,500 steps. In the first step, all the atoms of the complex were restrained with a force constant of 2.0 kcal/(mol·Å2) to only minimize the solvent and ion molecules. After that, the protein backbone atoms were fixed with a restraint force of 2.0 kcal/(mol·Å2) in the second step. Finally, all atoms in the system were minimized without any restraint. After energy minimization, all systems were heated up from 0 to 310 K in the canonical (NVT) ensemble over 100 ps by restraining the protein and the ligand with a 2.0 kcal/(mol·Å2) force constant and using a Langevin thermostat with a coupling coefficient of 2.0/ps. After heating, five steps MD pre-equilibration at 310 K were performed in the NPT ensemble by restraining all the atoms of the complex with decreasing restraints from 2.0 to 1.5, to 1.0, to 0.5, to 0.1. Then, 50-ns equilibration MD simulation without any restraints was performed to eliminate collisions between atoms. Finally, 500-ns production MD simulation was carried out without any restraints on each system in the NPT ensemble at the temperature of 310.0 K and pressure of 1 atm. During the simulations, all the bonds involving hydrogen atoms were restrained with SHAKE algorithm (Ryckaert et al., 1977) to avoid too fast vibration of the hydrogen atoms. In addition, periodic boundary conditions were employed, and the long-range Coulombic interactions were treated using the Particle Mesh Ewald (Darden et al., 1993). The time step was set to 2 fs.

Molecular Mechanics Generalized-Born Surface Area Calculation

To uncover the effects of mutations on the binding affinity of rifampicin to Mtb-RNAP, the MM-GBSA method was applied to estimate the binding free energy of each complex, which has been successfully used in a lot of researches (Pan et al., 2011; Yang et al., 2012). Here, we extracted 1,000 snapshots at 100-ps interval from the last 100-ns trajectory for each system. The binding free energy was calculated from the equation:

ΔGbind=Gcomplex−Greceptor−Gligand    (1)ΔGbind=Gcomplex-Greceptor-Gligand    (1)

where Gcomplex, Greceptor, and Gligand are the free energy of complex, protein, and ligand, respectively. The free energy for each molecular species was calculated based on an average over the extracted snapshots. Each of them can be estimate with the following equations:

G=EMM+Gsol−TS    (2)G=EMM+Gsol-TS    (2)

EMM=Eint+Eele+Evdw    (3)EMM=Eint+Eele+Evdw    (3)

Eint=Ebond+Eangle+Etorsion    (4)Eint=Ebond+Eangle+Etorsion    (4)

Gsol=GGB+GSA    (5)Gsol=GGB+GSA    (5)

GSA=γ∗SASA+β    (6)GSA=γ*SASA+β    (6)

where EMM is the gas-phase energy calculated using the Amber ff03 molecular mechanics force field. Eint is the internal energy, including the energy of bond (Ebond), angle (Eangle), and torsion (Etorsion). Eele and Evdw are the Coulomb and van der Waals energy, respectively. Gsol is the solvation free energy and can be decomposed into polar solvation free energy (GGB) and non-polar solvation free energy (GSA). GGB was calculated by solving the GB equation and the dielectric constants for solute as well as solvent were set to 1.0 and 80.0, respectively (Rocchia et al., 2001). GSA was estimated by the solvent accessible surface area determined using a water probe radius of 1.4 Å. The surface tension constant γ was set to 0.0072 kcal/(mol·Å2), and the non-polar contribution to the solvation free energy term β was set to 0 (Sitkoff et al., 1994). T and S are the temperature and the total solute entropy, respectively. The entropy contributions can be estimated by normal mode analysis (Pearlman et al., 1995). However, here, we did not calculate the entropy contributions since our aim is not to obtain the absolute Gibbs energy but to identify the key residues of binding pocket and the detailed interaction features. In addition, previous studies have proven that it is sufficient to compare the binding ability of receptors and ligands based on the values of enthalpy changes (ΔHbind) (Aruksakunwong et al., 2006; Xue et al., 2012).

Moreover, in order to identify the key residues responsible for the binding of rifampicin, the MM-GBSA binding free energy decomposition process was used to decompose the interaction energy to each residue by considering molecular mechanics and solvation energy without considering the contribution of entropy.

Dynamic Network Analysis

Dynamic network analysis, as an effective method to extract information from the obtained molecular dynamics trajectories, has been successfully applied in protein misfolding (Zhou et al., 2019) and protein–protein interaction analysis (Sethi et al., 2009; Bai et al., 2014). Here, in order to observe the dynamic changes of the residues interaction network, 2,000 snapshots were extracted from the last 50-ns trajectory for each system. In the network, one node represents one residue, and the position of each node is defined at the center of Cα atom of residue. The edge represents the interactions of two residues. Furthermore, the edge weight (Wij) between two nodes (i, j) was defined with the following equation:

Wij=−log ( |Cij| )    (7)Wij=-log ( |Cij| )    (7)

where Cij represents the pairwise correlations, which is calculated by Carma program (Glykos, 2006), a plugin in VMD (Humphrey et al., 1996). Finally, the NetworkView module in VMD was used to visualize the residue interaction network.

Residue Interaction Network Analysis

RIN uses a network diagram to simplify the inter-residue interaction, which considers the residues as nodes and physico–chemical interactions such as covalent and non-covalent bonds as edges. RIN method has been successfully used to analyze the effects of mutations on drug resistance (Xue et al., 2012, 2013, 2014). In this work, the residue interaction network generator 2.0 (RING-2.0) (Piovesan et al., 2016) software was applied to generate the network for the representative structures. The calculation process of RING-2.0 is described as follows: (i) the calculation of the secondary structure elements by incorporating the DSSP algorithm (Kabsch and Sander, 1983); (ii) hydrogen atom placement based on geometric criteria; (iii) hydrogen bond calculation; and (iv) the calculation of van der Waals interactions. Moreover, Cytoscape (Shannon et al., 2003) and the plugin RINalyzer (Doncheva et al., 2011) were used to visualize the residue interaction network.

Results

H451D/Y/R Mutations Increased the Flexibility of the Active Pocket

Firstly, the root-mean-square-deviations (RMSD) value for the protein backbone atoms, the active pocket, and the heavy atoms of rifampicin relative to the initial structure were calculated to monitor the equilibrium of each system. As shown in Figure 2 and Supplementary Figures 1–3, three parallel MD simulations have similar fluctuations, suggesting each parallel trajectory can produce reproducible results. Thus, the following analysis was based on one of three parallel MD simulations. As can be seen from the RMSDs of wild-type Mtb-RNAP and three mutants, each system achieves equilibrium after 100 ns. Therefore, the last 100-ns trajectory for each system was used for the following structural and energetic analysis. Additionally, from the monitoring of the RMSD value of the heavy atoms of ligand, we can justify roughly if the ligand can bind to the target stably. From Figure 2C and Supplementary Figures 1–3 the RMSDs of rifampicin in mutants were larger than that in wild type, indicating that rifampicin had a large fluctuation in mutants.

FIGURE 2

Figure 2. Root-mean-square-deviations (RMSDs) for the wild-type system of three independent molecular dynamics simulations: Sim1 (black), Sim2 (green), Sim3 (orange): (A) RMSDs for the backbone atoms of protein versus time. (B) RMSDs for the backbone atoms of active pocket vs. time. (C) RMSDs for the heavy atoms of rifampicin vs. time. (D) Root-mean-square-fluctuation for the backbone atoms of active pocket vs. residue number of wild type (black), H451D (red), H451Y (green), and H451R (purple) in Sim1.

The root-mean-square-fluctuation (RMSF) of each residue was calculated based on the last 100-ns trajectory for each system, and the corresponding results were shown in Figure 2D. It can be seen from Figure 2D that the RMSF values of the residues have similar trends for all systems. However, H451D/Y/R mutations will increase the RMSF values relative to the wild-type Mtb–RNAP, which indicates that these three mutations increased the flexibility of the binding pocket and weakened the interaction of rifampicin with Mtb-RNAP.

H451D/Y/R Mutations Weaken the Binding Ability of Mycobacterium tuberculosis RNA Polymerase With Rifampicin

To explore the effects of three mutations on the binding of Mtb-RNAP with rifampicin, the binding free energy calculation was performed based on MM-GBSA method. As shown in Table 1, the enthalpy changes (ΔHbind) of the wild-type Mtb–RNAP and H451D/Y/R mutants with rifampicin are −43.89, −31.20, −35.55, and −28.58 kcal/mol, respectively. As expected, the binding affinity of Mtb–RNAP to rifampicin reduced obviously due to H451D/Y/R mutations.

TABLE 1

Table 1. The calculated binding free energy and the detailed contribution of different energy terms (kcal/mol).

By assessing the contributions of individual energy terms, we found that the non-polar interactions (sum of van der Waals interaction energy ΔEvdw and non-polar interaction energy in solvation free energy ΔGSA) are the driving force for the binding of rifampicin. However, the energy contributions of Evdw and ΔGSA decrease in the mutants. Relative to non-polar interactions, the polar interaction (sum of electrostatic interaction energy ΔEele and polar interaction energy in solvation free energy ΔGGB) seems like unfavorable for the binding of rifampicin and more apparent in the mutants (15.22, 19.21, 20.70, and 23.15 kcal/mol for the wild type, H451D, H451Y, and H451R mutants, respectively). Although the contributions of intermolecular electrostatic interactions (ΔEele) are very favorable, their contributions are compensated by the large desolvation penalties.

Key Residues Responsible for the Reduced Binding Ability of H451D/Y/R Mutants

By decomposing the binding free energy of the wild-type Mtb-RNAP with rifampicin, 10 key residues such as Q438, F439, D441, R454, P489, E490, N493, I497, R613, and Q614 (Figure 3A) with energy contributions over 2 kcal/mol are identified. It can be seen that relative to the wild-type Mtb-RNAP, the contributions of Q438, F439, D441, E490, N493, R613, and Q614 have an obvious decrease in H451D mutant. For H451Y mutant, the reduced energy contribution of residues Q438, D441, R454, N493, R613, and Q614 should be responsible for the reduced binding affinity of rifampicin to Mtb-RNAP. Moreover, the energy reduction in H451R mutant is more obvious relative to that in H451D/Y mutants. Actually, the profile of each residue’s energy contribution in three mutants shares some similar features. For example, in all three mutants, the residues Q438, D441, N493, R613, and Q614 have obvious reduced contribution for the binding of rifampicin, suggesting that the drug-resistance mechanisms due to three mutations have some similarities.

FIGURE 3

Figure 3. (A) The total energy contributions of the key residues for binding of rifampicin: wild type (black), H451D (red), H451Y (green), H451R (purple). (B) The electrostatic energy contributions of the key residues for binding of rifampicin. (C) The van der Waals energy contributions of the key residues for binding of rifampicin.

To further explore the origin of the reduced residues’ contribution, we compared the electrostatic contribution and van der Waals contribution of the key residues (shown in Figures 3B,C, respectively). From Figure 3B, we can obtain when H451 is mutated to D451, the electrostatic energy contribution of D451 (4.80 kca/mol) is detrimental for rifampicin binding. Moreover, the contributions of electrostatic energy of Q438, F439, E490, and R613 decrease after H451D mutation. For H451Y mutant, the electrostatic energy contribution of three residues (Q438, R454, and R613) also have an obvious reduction. In H451R mutant, the contributions of electrostatic interactions of almost all key residues reduced. In addition, Figure 3C shows that the reduction of energy contribution of D441, P489, E490, N493, and Q614 is mainly from the loss of the van der Waals interaction contribution in H451D/Y/R mutants.

The Dynamic Network Analysis and Residue Interaction Network Analysis Reveal That H451D/Y/R Mutations Weaken the Interaction of Mutated Residue With Its Adjacent Residues

To investigate how H451D/Y/R mutations change the binding pocket and further lead to the resistance of Mtb–RNAP to rifampicin, the dynamic network analysis was further carried out based on the 2,000 snapshots extracted from the equilibrium phase for each system. The obtained results are shown in Figure 4. The strength of the total interactions (including van der Waals interaction and hydrogen bond interaction and so on) is indicated by the edge thickness. Moreover, the type of interactions is shown by the two-dimensional RIN interactively based on the representative structure of each system (Figure 5). Here, the representative structures were extracted by clustering analysis, and the conformation with the lowest RMSD to the cluster center was selected.

FIGURE 4

Figure 4. The picture of dynamic network analysis for the active pocket of rifampicin binding: (A) WT; (B) H451D; (C) H451Y; (D) H451R; The purple spheres represent the residues and the sticks represent the total interactions. The strength of interactions of two residues is indicted by the thickness degree of stick, the thicker of the stick, and the stronger interaction of two residues.

FIGURE 5

Figure 5. RIN for the active pocket of rifampicin binding (A) wild type; (B) H451D; (C) H451Y; (D) H451R. The pink octagons represent residues, and the yellow octagons represent the mutated residues; the edges represent van der Waals (blue) and hydrogen bond (red) interactions and salt bridge interaction (cyan).

In the wild-type Mtb–RNAP, H451 residue can be identified as the central node with six first neighbors in the interaction network according to Figures 4A, 5A. H451 not only can form strong van der Waals interactions with R454, Q438, F439, M440, and D441 but also form hydrogen bonds with D441, S447. Moreover, S447 acts as a key node in the network; in addition to forming hydrogen bond with H451, it also can form strong interactions with D441 (van der Waals interaction), R613 (hydrogen bond interaction), and Q614. I497, as another key node in the other side of the network, can form van der Waals interactions with R454, P489, and N493. Finally, R454 as a joint node can connect the two sides of the networks together by forming the van der Waals interactions with I497, H451, and the hydrogen bond with Q438. In addition, the weak “triangular shape” interaction network formed by R454, R613, and P489 residues that make the binding pocket more compact and coherent.

However, in H451D mutant, it is evident that the mutation causes the disappearance of the interactions between D451 and Q438, F439, M440, D441, and S447 mentioned previously (Figures 4B, 5B). For H451Y/R mutants, the corresponding interactions are also obviously weakened (Figures 4C,D). Overall, in H451D/Y/R mutants, the “triangular shape” interaction network is broken. Based on these results, it can be concluded that the mutations on 451 indeed reduced the interaction connection of the residues in the binding pocket and then result in the active pocket more flexible and open.

The Comparison of Binding Modes of Rifampicin With the Wild-Type Mycobacterium tuberculosis RNA Polymerase and Three Mutants

To show the detailed rifampicin-resistance mechanism to Mtb-RNAP H451D/Y/R mutants intuitively, further structural analysis was performed. The representative structures are depicted in Figure 6. As shown in Figures 3C, 6A, the van der Waals interactions of Q438, D441, H451, R454, P489, E490, N493, I497, and Q614 with rifampicin are pivotal for the binding of rifampicin to the wild-type Mtb-RNAP. In addition, the calculation of hydrogen bond occupancy was carried out to monitor the formation of hydrogen bonds between rifampicin and Mtb-RNAP over the whole MD simulations. Based on the result in Table 2, we can see that the O8, O11, O9/O2, and O12 atoms of rifampicin can form stable hydrogen bonds with the side chains of Q438, F439, and R613 with high occupancy rate. The formation of these hydrogen bonds makes F439, R613, and Q438 have large electrostatic energy contributions (Figure 3B).

FIGURE 6

Figure 6. Three-dimensional representation for the binding mode of rifampicin with wild-type Mycobacterium tuberculosis RNA polymerase and three mutants based on the obtained representative structure: (A) wild type; (B) H451D; (C) H451Y; (D) H451R. The protein is shown as a cyan cartoon; rifampicin and the key residues are represented as hotpink and cyan carbon sticks, respectively; the intermolecular hydrogen bonds are indicated as dashed yellow lines.

TABLE 2

Table 2. The occupancy (%) of hydrogen bonds between rifampicin and Mycobacterium tuberculosis RNA polymerase.

By comparing the binding modes of wild-type and H451D mutant (Figures 6A,B), the position of rifampicin has a clear movement. Combined with the results of RIN analysis (Figures 5A,B), H451D mutation causes the interactions of residue 451 between with some residues of binding pocket disappear, further causing the conformations of some amino acids (such as D441, P489, E490, N493, and Q614) that changed a lot. Finally, the residues D441, P489, E490, N493, and Q614 are away from rifampicin, causing the weakened van der Waals interaction of these residues. From Table 2, some hydrogen bonds between rifampicin and Q438, F439, and R613 disappeared in H451D mutant, which causes the reduction of the electrostatic energy contribution (Figure 3B). Moreover, the H451D mutation leads to the electrostatic repulsion between carboxyl group of D451 and O10 atom of rifampicin and is unfavorable for the binding of rifampicin.

For H451Y mutant, the binding mode of rifampicin with Mtb-RNAP is more similar to that in the wild-type Mtb-RNAP, and the position of rifampicin just has a slight movement (Figures 6A,C). Despite this, the mutation still causes some residues (such as N493, I497, and Q614) away from rifampicin, reducing the nonpolar contribution of these residues. In addition, the result in Table 2 shows that the disappearance of the hydrogen bond between R613 and rifampicin is responsible for the reduction of electrostatic contributions of R613. The reduction of hydrogen bond occupancy rate formed between Q438 and rifampicin causes the loss of electrostatic contribution of Q438.

From the energetic and structural analysis discussed previously, it can be seen that the flipping of R613 greatly affects the binding of rifampicin in H451D/Y mutants. Therefore, it is worthy to explore how the mutation of H451 affects the flipping of R613. For this aim, we superimpose the binding pocket of the wild-type Mtb-RNAP with those of H451D/Y mutants. Figure 7 shows the superimposition results. From Figure 7A, it can be seen that the position of H451 is on H1. When H451 is mutated to D451, no hydrogen bond can be formed between D451 and S447 (Figure 5B), which causes the helix structure of S447 to transform into disordered loops. At the same time, the fluctuation of this loop further causes the disappearance of the hydrogen bond between R613 (located on H2) and S447 (located on H1). Therefore, the R613 residue flips with H2 due to the weaker interaction between H1 and H2. However, the mutation of H451Y enhances the van der Waals interaction of Y451 and S447 (Figure 5C), which increases the distance of S447 and R613 and further to interfere the formation of hydrogen bond. Moreover, the rotation of H2 makes the flipping of R613 more obvious. The flipping of R613 directly causes the disappearance of the hydrogen bonds with rifampicin, further leading that the electrostatic contribution of R163 reduced.

FIGURE 7

Figure 7. (A) The superimposition of wild-type Mycobacterium tuberculosis RNA polymerase and H451D mutant. (B) The superimposition of wild-type Mycobacterium tuberculosis RNA polymerase and H451Y mutant; the protein is shown as cartoon with cylindrical helices colored with cyan (wild type) and magenta (mutant); rifampicin and R613, S447, H451, D451, and Y451 are shown as stick colored with cyan (wild type) and magenta (mutant); the intermolecular hydrogen bond between S447 and R613 is indicated as dashed yellow lines.

For H451R mutant, there was a large conformational transition of the binding pocket and rifampicin as shown in Figure 6D. Moreover, R451 has a steric clash with rifampicin and so that rifampicin moves out of the binding pocket. Thereby, the movement of rifampicin causes the energy contribution of some key residues such as R613, E490, S447, D441, M440, and F439 reduce (Figure 3). In addition, some hydrogen bonds between rifampicin and F439/R613 also disappear (Table 2). In order to observe the changes of the binding pocket more intuitively, the surface map for the binding pocket of the wild-type and H451R mutant is depicted in Figure 8. It can be seen that when H451 is replaced by R451, the long side chain of R451 made the binding pocket smaller and cannot accommodate the relatively rigid rifampicin, which leads to the movement of rifampicin.

FIGURE 8

Figure 8. The binding pocket of rifampicin: (A) wild type; (B) H451R mutant; the pocket is shown as surface, and rifampicin is shown as stick colored with cyan.

In summary, there are three main binding sites between rifampicin and the active pocket in wild type (Figure 9A). The polar pocket formed by the residues N610, R613, Q614, etc. (S1) can act as the hydrogen bond acceptor to form the hydrogen bond with the oxygen atom of rifampicin. The other polar pocket (S2) consists of residues Q438, M440, D441, etc., which also forms stable hydrogen bonds with rifampicin. In addition, the hydrophobic pocket composed of residues such as P489, N493, and I497 just accommodates the naphthalene ring of rifampicin. However, H451D/Y/R mutants changed the initial binding mode for rifampicin resulting from the change of the side chain size of 451. The N atom from imidazole ring of H451 can form the hydrogen bond interaction with rifampicin in wild-type RNA polymerase, which is interfered by these mutations. Furthermore, the mutations in 451 position impaired the interaction between the residue 451 and other key residues in the active pocket, interfering with the specific orientation of the key residues’ side chains, thereby increasing the flexibility of the key residues such as R613. As a result, the binding affinity of rifampicin reduced, and the important interactions between rifampicin and active pocket were disturbed. One of the most apparent changes from the results of binding free energy decomposition was that the reduction of energy contribution from R613, which was mainly because the disappearance of the hydrogen bond between R613 and O atom at the C15 position of rifampicin (Figure 9B). Therefore, according to the obtained mechanism, to overcome the drug resistance induced by H451D/Y/R, one possible strategy is to enhance the interaction of the inhibitor with R613 by replacing the carbonyl group at the C15 position in rifampicin with a longer and negatively charged group (R). Such group may recover the hydrogen bond interaction between R613 and inhibitor, which could stabilize the binding of rifampicin.

FIGURE 9

Figure 9. (A) Three main binding sites between rifampicin and the active pocket. Rifampicin is shown as stick colored with magenta, and the residues are shown as lines colored with cyan. The yellow dotted lines represent the hydrogen bond interactions. (B) Structural optimization of rifampicin. R represents a longer negatively charged group.

Conclusion

In this study, MD simulations together with MM-GBSA, dynamic network analysis, and RIN analysis were carried out on the complexes of rifampicin with the wild-type Mtb-RNAP and H451D/Y/R mutants to explore the resistant mechanism of rifampicin. The results from the MM-GBSA calculations are well-consistent with the experimental result. The reduced binding affinity for the studied mutants mainly comes from the loss of van der Waals contribution of D441, P489, E490, N493, I497, and Q614 and electrostatic contribution of Q438, F439, and R613 in mutants. The electrostatic energy contribution of R613 decreases obviously with the disappearance of the hydrogen bonds between R613 and rifampicin, which caused by the conformation flipping of R613 in H451D/Y mutants. The binding modes and dynamic network analysis show that the weakened interactions among D/Y/R451 with Q438, F439, M440, D441, and S447 increase the flexibility of the binding pocket, thereby reducing the binding affinity of rifampicin to Mtb-RNAP. In addition, these mutations caused the key hydrogen bond interactions between residue 451 and rifampicin disappear. Finally, the position of rifampicin had a clear movement, which changed the stable binding mode of rifampicin. We firstly proposed the atomic level resistance mechanism of Mtb to rifampicin due to H451D/Y/R mutations on Mtb-RNAP. In addition, we also proposed some guidance for the alteration of the rifampicin and the development of new drugs in the future. Though the hypothesis is still unvalidated on the traditional experimental, it can provide some theoretical underpinnings for the design of new anti-TB drugs to some extent, or a particular aspect.

Biological computer that ‘lives’ inside the body comes one step closer as scientists make transistor out of DNA and RNA.

Posted by

0


Molecular computer graphic of DNA double helixMolecular computer graphic of DNA double helixMolecular computer graphic of DNA double helixmolecular_computer3Finding could lead to new biodegradable devices based on living cells that are capable of detecting changes in the environment

Scientists believe they are close to building the first truly biological computer made from the organic molecules of life and capable of working within the living cells of organisms ranging from microbes to man.

The researchers said that they have made a transistor – the critical switch at the heart of all computers – from DNA and RNA, the two biological molecules that store the information necessary for living things to replicate and grow.

Silicon transistors control the direction of flow of electrical impulses within computer chips, but the biological transistor controls the movement of an enzyme called RNA polymerase along a strand of the DNA molecule, the scientists said.

Ultimately, the aim is to use the biological transistors – called transcriptors – to make simple but extremely small biological computers that could be programmed to monitor and perhaps affect the functioning of the living cells in which they operate, researchers said.

It could lead to new biodegradable devices based on living cells that are capable of detecting changes in the environment, or intelligent microscopic vehicles for delivering drugs within the body, or a biological monitor for counting number of times a human cell divides so that the device could destroy the cell if it became cancerous, the scientists said.

“Biological computers can be used to study and reprogram living systems, monitor environments and improve cellular therapeutics,” said Drew Endy, assistant professor of bioengineering at Stanford University in California, who led the study published in the journal Science.

Last year, Professor Endy announced new ways of using biological molecules to store information and to transmit data from one cell to another. The latest study adds the third critical component of computing – a biological transistor that acts as a “logic gate” to determine whether a biochemical question is true or false.

Logic gates are critical for a computer to function properly. In a biological setting the use of logical data processing is almost as limitless as its use in conventional electronic computing, said Jerome Bonnet, a bioengineer within the Endy laboratory, and the lead author of the study.

“You could test whether a given cell had been exposed to any number of external stimuli – the presence of glucose and caffeine for instance. [Logic] gates would allow you to make the determination and store that information so you could easily identify those which had been exposed and which had not,” Dr Bonnet said.

Biological computers have been the dream of electronic engineers for decades because they open the possibility of a new generation of ultra-small, ultra-fast devices that could be incorporated into the machinery of living organisms.

“For example, suppose we could partner with microbes and plants to record events, natural or otherwise, and convert this information into easily observed signals. That would greatly expand our ability to monitor the environment,” Professor Endy said.

“So the future of computing need not only be a question of putting people and things together with ubiquitous silicon computers. The future will be much richer if we can imagine new modes of computing in new places and with new materials – and then find ways to bring those new modes to life,” he said.

Source: http://www.independent.co.uk