Shigella: Antibiotic-Resistance Mechanisms And New Horizons For Treatment

Abstract

Shigella spp. are a common cause of diarrheal disease and have remained an important pathogen responsible for increased rates of morbidity and mortality caused by dysentery each year around the globe. Antibiotic treatment of Shigella infections plays an essential role in reducing prevalence and death rates of the disease. However, treatment of these infections remains a challenge, due to the global rise in broad-spectrum resistance to many antibiotics. Drug resistance in Shigella spp. can result from many mechanisms, such as decrease in cellular permeability, extrusion of drugs by active efflux pumps, and overexpression of drug-modifying and -inactivating enzymes or target modification by mutation. Therefore, there is an increasing need for identification and evolution of alternative therapeutic strategies presenting innovative avenues against Shigella infections, as well as paying further attention to this infection. The current review focuses on various antibiotic-resistance mechanisms of Shigella spp. with a particular emphasis on epidemiology and new mechanisms of resistance and their acquisition, and also discusses the status of novel strategies for treatment of Shigella infection and vaccine candidates currently under evaluation in preclinical or clinical phases.

Keywords:

• Shigella
• antibiotics
• resistance
• drug resistance
• mechanism
• treatment
• biofilm
• efflux pumps
• prevention
• vaccine

Introduction

Shigella spp. are a Gram-negative, rod-shaped, immotile, and non-spore–forming bacteria and a causative agent of acute diarrhea that may progress to bloody mucoid diarrhea, also known as bacillary dysentery (or shigellosis).¹ Shigella is the most common cause of diarrheal disease and has remained a major pathogen responsible for increased rates of morbidity and mortality caused by dysentery each year around the globe, particularly affecting children aged <5 years in developing countries.² The four types of Shigella spp. comprise subgroup A (S. dysenteriae), subgroup B (S. flexneri), subgroup C (S. boydii), and subgroup D (S. sonnei). Each subgroup contains several serotypes. Shigellosis can occur in pandemic, epidemic, and sporadic forms. Epidemiological reports have shown that the epidemic subgroup of diarrhea typically occurs as a result of infection with S. flexneri in developing countries and S. sonnei in industrialized countries.^1,3 Shigella spp. are categorized by the World Health Organization (WHO) as bacteria mainly causing infections in the community.⁴ Shigellosis is a great public health threat, because its infective dose is on the order of 10–100 organisms compared to other enteric pathogens (usually it is 105–108 for Salmonella and Vibrio, respectively).^5,6

Antibiotic treatment of common bacterial infections plays an important role in reducing prevalence and death rates of the disease. However, incorrect antibiotic use or overuse in treating diarrhea increases antibiotic resistance. Shigella spp. are resistant to most antibiotics, and drug treatment related to these bacteria is costly, time-consuming, and sometimes problematic, particularly in areas with limited medical care.^7,8 About half the strains of Shigella in many parts of the world are now resistant to multiple drugs. Recently, various antibiotic-resistance mechanisms have been described by researchers, and these antibiotic-resistance mechanisms limit therapeutic options for treatment of Shigella infections.^8,9

Drug resistance in Shigella spp. can result from many mechanisms, such as extrusion of drugs by active efflux pumps, decrease in cellular permeability, and overexpression of drug-modifying and -inactivating enzymes or target modification by mutation.^6,10,11 The current study was done to review various antibiotic-resistance mechanisms of Shigella spp., with a particular focus on epidemiology and new mechanisms of resistance and their acquisition, and also to discuss treatment and prevention measures for diseases caused by these organisms.

Search Strategy

We searched the biomedical electronic databases Web of Science, Scopus, PubMed (Medline), Embase, Cochrane Library, and Google Scholar for articles on Shigella published in English between 1990 and May 2019, using the key terms (alone and in combination) “Shigella”, “drug resistance”, “mechanism”, “biofilm”, “efflux pump”, “vaccine”, and “treatment”. We excluded case reports and some studies on the implementation of established techniques. Articles published before 1990 were also excluded, except when necessary. A total of 193 relevant articles were identified from the databases and included in this review.

Drug-Resistance Mechanisms In Shigella Spp

Role Of Outer-Membrane Permeability

Natural resistance to antimicrobial drugs by various mechanisms preventing the drug from being absorbed is capable of transforming the drug, its biotransformation into the cell, or reducing affinity with the drugs’ target.¹² Cell walls of microorganisms are the first barrier against penetration of the drug. Some modifications of membrane permeability or changes in the membrane lead to porin loss, which can result in an increase in minimum inhibitory concentration(MIC) for antimicrobial agents.¹³ Most antibiotics used in treatment of Shigella infection should be able to penetrate the cell membrane to reach intracellular accumulation and target sites. For example, quinolone antibacterial agents, such as nalidixic acid, ofloxacin, and ciprofloxacin, interfere with DNA replication by inhibiting DNA topoisomerase IV and gyrase. Aminoglycoside antibiotics, such as streptomycin and spectinomycin, mediate inhibition of protein synthesis by binding to ribosomal subunits and reaching intracellular targets. β-Lactam antibiotics, eg, penicillin and cephalosporin, are a class of antibiotics containing a β-lactam ring in their molecular structures and inhibit cell-wall biosynthesis by targeting penicillin-binding proteins. Mutation or absence of ∼39 kDa porin in the membrane of such Gram-negative bacteria as Shigella spp. mainly influences susceptibility to slow penetration of β-lactams, such as aztreonam and dianionic moxalactam, and also low permeability of hydrophilic antibiotics, such as penicillin and piperacillin.^6,13 Indeed, resistance toward β-lactam antibiotics is associated with modification of the outer-membrane porins OmpF (∼38 kDa) and OmpC (∼42 kDa) and cytosolic proteins of ∼26 kDa, OmpR as a transcriptional regulator.⁶ In a study, three imipenem-resistant mutants of S. dysenteriae were obtained from India and showed lower levels of both major OMPs (∼38 and ∼43 kDa). Increasing imipenem resistance in mutants was associated with permeability of outer-membrane proteins.¹⁴ Lipopolysaccharides (LPSs) have been recognized as an essential outer-membrane component needed for assembly of trimeric PhoE porin and confer colicin E₂ resistance in S. flexneri strains,¹⁵ and have also been reported to be linked with the rise in resistance toward imipenem in S. dysenteriae.¹⁴ Some outer-membrane components, such as IcsA molecules, are not only associated with bile salts resistance but are also related to promotion in development of biofilm by mediating bacterial cell–cell interactions. Consequently, they produce resistant phenotypes.¹⁶

Efflux Systems

Active efflux pumps play a significant role in antibiotic-resistance phenotypes of Gram-negative bacteria and expelling toxic compounds from their cells. Efflux systems are grouped into five families: the major facilitator superfamily (MFS), resistance–nodulation–division family, small multidrug resistance (MDR) family, ATP-binding cassette superfamily, and multidrug and toxic compound extrusion family.¹⁷ AcrAB–TolC pump is involved in antibiotic resistance phenotype of Escherichia spp., Enterobacter spp., Salmonella spp., and Shigella spp isolates. The AcrAB–TolC system is a tripartite complex comprising TolC (outer-membrane channel), AcrB (inner-membrane transporter protein), and periplasmic AcrA involved in assembly and maintenance of these two integral membrane proteins. AcrAB–TolC belongs to resistance–nodulation–division family of efflux pumps, associated with efflux of quinolones, and one of factors responsible for development of resistance among Shigella isolates. Indeed, overexpression of AcrAB–TolC results in overall decreased accumulation of quinolones inside bacterial cells, also resulting in reduced susceptibility to them.¹⁸ AcrAB-associated to bile-salt resistance has been found in some strains of S. flexneri. Their expression has been shown to increase after exposure to bile salts, and enabled Shigella to resist bactericidal effects of bile. Researchers believe that this phenomenon may confer resistance to other antimicrobial agents. Furthermore, overexpression of AcrB has been found to be linked to multiple drug-resistance phenotypes in some Gram-negative bacteria.¹⁹

Synergistic action regarding activation of acrAB–tolC efflux pumps has been shown to decrease expression of outer-membrane porins, gyrase, and topoisomerase target–gene mutations toward fluoroquinolone resistance in Shigella isolates.^20,21 Drug-efflux pumps, such as marA, tolC, ydhE, and mdfA, confer quinolone resistance. Kim et al demonstrated that resistance to fluoroquinolone is due to increased expression level of MdfA efflux pump in Shigella spp. This efflux is a member of the MFS antibiotic-efflux system, and MdfA efflux pump–mediated fluoroquinolone resistance was first identified among MDR Escherichia coli.²² Tetracycline efflux and resistance is associated with the MFS antibiotic–efflux system encoded by various tet genes in Gram-negative bacteria, such as Shigella spp. and Klebsiella spp. Among tet efflux systems, it seems that tetA and tetB are mediated by resistance to tetracycline in S. sonnei and S. flexneri, respectively (for more details, refer to “Tetracycline Resistance” section).¹¹

Resistance To β-Lactam Antibiotics

Class A β-Lactamases

Class A β-lactamases can hydrolyze narrow-spectrum penicillin, but not carbapenems or cephalosporins, and are inhibited by tazobactam and clavulanic acid. Extended-spectrum β-lactamases (ESBLs) belong to Ambler class A. ESBLs conferring resistance to third-generation cephalosporins have been found in Shigella isolates. The first report documenting identification of ESBL-producing Shigella strains was from Bangladesh in 2004.²³ Emergence of ESBLs in Shigella spp. is a global major health threat affecting both developed and developing countries. Different β-lactamases belonging to Ambler class A have been reported among Shigella isolates, such as TEM, SHV, and CTX-M enzymes. The first isolate of S. flexneri producing an ESBL and harboring a plasmid encoding the bla_SHV-2 gene was reported from France in 1995.²⁴

To date, several reports from Argentina, Israel, Canada, Turkey, Lebanon, Japan, Iran, South Korea, China, and other various regions in Asia have identified Shigella spp. harboring different types of ESBL genes (Table 1).^23,25–32 Although most ESBLs are derivatives of TEM and SHV β-lactamase families, which were first identified, Shigella spp. can also express the CTX-M family, among which CTX-M-15 is one of the most relevant findings associated with the current epidemiology of ESBLs, which has been predominantly identified in commensal and pathogenic ESBL-producing Shigella isolates around the world. Despite many surveillance studies and investigations, the reason for this epidemiological shift remains unknown.^24,33 These enzymes are responsible for selective hydrolysis of ceftriaxone and cefotaxime and even more distinctly for ceftazidime, although some types of CTX-M, such as CTX-M-15, may hydrolyze ceftazidime.³³ In general, CTX-M-15 has been found to have great catalytic efficiency (high k_cat/K_m) against piperacillin, benzylpenicillin, ceftriaxone, and cefotaxime, as reported for other of CTX-M-types, such as CTX-M-3, CTX-M-16, and CTX-M-18.

Table 1 Prevalence Of Antimicrobial Resistance Genes In Shigella Spp. Isolated From Different Regions Of The World

Antimicrobial Class	Resistance Mechanism	Genes Mediating Antimicrobial Resistance	Origin	Geographic Origin	Reference
β-Lactams	Class A β-lactamases	blaSHV-2	P	Argentina, France	^Citation32
		blaSHV-11	C, P	India	^Citation191
		blaSHV-12	P	China, Turkey	^Citation40
		blaPER-2	—	Argentina	^Citation32
		blaTEM-1	I, P	Lebanon, Chile, China, India, Iran, US, Djibouti, Denmark, France, Greece, Brazil, South Korea, UK, Romania	^Citation25,Citation26,Citation31,Citation40,Citation120,Citation192
		blaTEM-1b	I, P	China, South Korea	^Citation34,Citation40
		blaTEM-15	–	South Korea	^Citation28
		blaTEM-17		South Korea	^Citation28
		blaTEM-19	–	South Korea	^Citation28
		blaTEM-20	–	South Korea	^Citation28
		blaTEM-52	P	South Korea	^Citation28
		blaCTX-M-1	P	China	^Citation40
		blaCTX-M-2	P	Argentina, Turkey, South Korea, Israel	^Citation30,Citation32,Citation39
		blaCTX-M-3	P	China, Turkey, Argentina, South Korea, India, Israel	^Citation30,Citation39,Citation40
		blaCTX-M-14	P	China, Turkey, Argentina, South Korea, US, Japan	^Citation39,Citation40,Citation193
		blaCTX-M-15	C, P	China, Spain, Iran, South Korea, India, US, Lebanon, Japan, Poland, New Zealand, France, Romania	^Citation26,Citation31,Citation33,Citation34,Citation40,Citation194
		blaCTX-M-22	P	China	^Citation51
		blaCTX-M-24	P	China	^Citation40
		blaCTX-M-27	P	China	^Citation40
		blaCTX-M-28	–	China	^Citation36,Citation51
		blaCTX-M-39	–	Israel	^Citation30
		blaCTX-M-55	P	China, South Korea	^Citation35,Citation40
		blaCTX-M-57	–	China	^Citation51
		blaCTX-M-64	P	Japan	^Citation195
		blaCTX-M-65	P	China	^Citation192
		blaCTX-M-79	P	China	^Citation40
		blaCTX-M-123	P	China	^Citation40
	Class B β-lactamases	blaIMP-like	P	India, France	^Citation43,Citation46
		blaKPC	–	Senegal, France	^Citation46
		blaIMP-3	P	Japan	^Citation44
		blaVIM-like	–	India	^Citation43
	Class C β-lactamases	blaCMY-2	C, P	China, Mexico, India, Iran, Taiwan, Costa Rica, Romania	^Citation48,Citation51
		blaCMY-59	C	Iran	^Citation31
		blaDHA-1	C, I, P	China, India, Israel	^Citation30,Citation51
	Class D β-lactamases	blaoxa-1-like	I, P	Mozambique, Chile, China, India, US, Egypt, Djibouti, Spain, Greece, Denmark, Peru, Iran	^Citation105,Citation193,Citation196
		blaoxa-2-like	I	Mozambique, Spain, Israel	^Citation30,Citation105
		blaoxa-5-like	–	Mozambique, Spain	^Citation105
		blaoxa-30-like	I, P	Senegal, China, France, Japan, Spain, Brazil	^Citation51,Citation91,Citation120
Quinolones	Plasmid-borne resistance	qnrA	P	Iran	^Citation50,Citation75
		qnrB	P	China, India, Iran	^Citation64,Citation65,Citation75
		qnrB4	P	Switzerland	^Citation61
		qnrB19	P	Switzerland	^Citation61
		qnrC	P	India	^Citation65
		qnrS	P	Iran, China, India, Pakistan	^Citation53,Citation66,Citation75,Citation94
		qnrS1	P	Iran, China, Switzerland, US	^Citation48,Citation61,Citation64
		aac-(60)-Ib-cr	P	US, India, Japan, China, Iran	^Citation50,Citation64^–^Citation66
	Efflux pumps	qepA	P	China, Switzerland	^Citation61,Citation64
Fosfomycin	Fosfomycin resistance enzymes	fosA3	P	China	^Citation38
Aminoglycosides: streptomycin		strA	MGE	India, Australia, Chile, Pakistan, South Korea	^Citation86,Citation93,Citation94,Citation97,Citation122
		strB	MGE	India, Chile, Pakistan, South Korea	^Citation86,Citation94,Citation97,Citation122
	Adenyltransferase	aadA1	I, P	Senegal, Bhutan, India, Taiwan, Spain, China, Iran, France, Australia, Brazil, Pakistan, South Korea	^Citation83,Citation86,Citation88,Citation91,Citation94,Citation97,Citation98,Citation120
		aadA2	I, P	Taiwan, Spain, China, South Korea, France, Australia, Korea	^Citation46,Citation83,Citation88,Citation97,Citation98
		aadA5	I, P	Taiwan, China, Iran	^Citation88,Citation98
Tetracycline	Efflux pumps	tetA	C, P	Mozambique, Taiwan, Chile, Peru, Brazil, Iran, Spain, Pakistan, South Korea	^Citation11,Citation88,Citation94,Citation97,Citation104^–^Citation106,Citation120
		tetB	C, P	Mozambique, Taiwan, Peru, France, Brazil, Iran, Spain, Pakistan	^Citation11,Citation46,Citation88,Citation94,Citation104,Citation106,Citation120
		tetG	C, P	Mozambique, Spain	^Citation104,Citation105
Trimethoprim	Dihydrofolate reductases	dfrA1	I, P	Spain, Taiwan, Senegal, Mozambique, India, Bhutan, China, Iran, South Korea, Peru, France, Chile, Australia	^Citation46,Citation83,Citation86,Citation88,Citation91,Citation98,Citation105,Citation122
		dfrA5	I, P	Spain, Senegal, China	^Citation91,Citation98
		dfrA7	I, P	Spain, Iran, France	^Citation46,Citation106
		dfrA8	P	Mozambique, Chile	^Citation105,Citation122
		dfrA12	I, P	Spain, Taiwan, Korea, Australia, South Korea	^Citation11,Citation83,Citation88,Citation97
		dfrA13	P	South Korea	^Citation97
		dfrA14	P	Mozambique, Chile	^Citation105,Citation122
		dfrA14-like	P	Mozambique	^Citation105
		dfrA15	I, P	Spain, Senegal	^Citation91
		dfrA16	P	Spain	^Citation91
		dfrA17	I, P	Taiwan, China, Iran, Brazil	^Citation88,Citation98,Citation106,Citation120
		dfrV	I, P	China, India	^Citation98
Sulfonamides	Plasmid-borne resistance	sul1	I, P	Australia, South Korea, Taiwan	^Citation83,Citation88,Citation97
		Sul2	I, P	Taiwan, India, Peru, Chile, Bangladesh, South Korea	^Citation86,Citation88,Citation97,Citation122,Citation126
		Sul3	I, P	Taiwan	^Citation88
Phenicols	Chloramphenicol acetyltransferase genes	CatA-like	P	Taiwan, Mozambique, India, Peru, France, Brazil, Pakistan	^Citation46,Citation86,Citation88,Citation94,Citation120
		catP	P	Pakistan	^Citation94
	Efflux pumps	cmlA1	I, P	Taiwan, Mozambique	^Citation88,Citation105
Colistin	Plasmid-borne resistance	mcr-1	P	China, Vietnam	^Citation112,Citation116
Macrolide	Enzymatic inactivation	mphA	P	Palestine, Switzerland, Vietnam, China, Canada, UK, Peru	^Citation61,Citation116,Citation127,Citation128,Citation133
	rRNA methylase	ermB	P	Vietnam, Canada, UK	^Citation128,Citation133

Abbreviations: P, plasmid; C, chromosome; I, integron; —, unknown; MGE, mobile genetic element.

To date, CTX-M-15 has been detected in Shigella isolates from various countries across the world, including Canada, Russia, Poland, the UK, France, Bulgaria, Turkey, and Iran.^33,34 CTX-M-type β-lactamases contain at least 40 enzymes, and these can be readily transferred among Shigella isolates by conjugative plasmids belonging to IncF, IncZ, and Incl groups.^34,35 Li et al revealed that ISEcp1 was present adjacently to all bla_CTX-M genes in Shigella strains, meaning that it plays an important role in mobilizing bla_CTX-M genes (Figure 1).³⁶ A large study conducted in Vietnam analyzed IncI1 plasmid pKHSB1 carrying bla_CTX-M-15 collected from a clonal population of S. sonnei.³⁷ Another study reported full sequence of IncI1 plasmid pSH4469 carrying bla_CTX-M-15 in a clinical isolate of S. sonnei isolated from an outbreak in South Korea.³⁴ The overall structure of plasmid pSH4469 was nearly identical to pKHSB1, and it seemed that both were responsible for dissemination of bla_CTX-M-15 among Shigella isolates (Figure 1).

Figure 1 Schematic representation of bla_CTX-M-55, bla_CTX-M-15, and bla_TEM-1 genes in different types of plasmid. Arrows indicate positions and directions of different genes and IRL, terminal inverted repeats at the left, IRR, terminal inverted repeats at the right.

Recently, a new hybrid of CTX-M-9 and CTX-M-1 β-lactamases named CTX-M-123 was identified among S. flexneri isolates from patients in China. In this study, bla_CTX-M-123 was carried by two conjugative plasmids named IncHI2 and IncF, and these conjugatable plasmids were responsible for dissemination of bla_CTX-M-123 in S. flexneri isolates.³⁸ Moreover, several new β-lactamase subtypes — CTX-M-79, CTX-M-27, CTX-M-24, CTX-M-15, CTX-M-14, CTX-M-64, CTX-M-65, CTX-M-55, and CTX-M-3 — have been found in clinical Shigella strains isolated from different provinces in China.^36,39,40 ESBL genes among Shigella isolates might be transferred from E. coli isolates to Shigella spp., especially S. sonnei isolates, through conjugation in human gut.^41,42 The increase in MDR and emergence of ESBL in Shigella spp. may be the cause of treatment failures and accordingly limitation in therapeutic options.⁴¹

Class B β-Lactamases

Class B β-lactamase enzyme can hydrolyze carbapenem and other β-lactams, except for aztreonam, and classical β-lactamase inhibitors, such as tazobactam and clavulanic acid, do not inhibit them. Metallo-β-Lactamase was first detected in a transferable plasmid from Pseudomonas aeruginosa, and also IMP-1 was first identified from many kinds of Gram-negative rods in Japan.^43,44 O’Hara et al reported a novel type of metallo-β-lactamase named MET-1 mediated by a S. flexneri plasmid. They believed that MET-1 was a derivative of IMP-1 β-lactamase.⁴⁴ This plasmid conferred resistance against sulfonamide and kanamycin, in addition to β-lactamase.⁴⁴ Lyobe et al found that MET-1 experienced two amino-acid changes from IMP-1. The gene was renamed IMP-3, and thus IMP-3 could be considered an ancestor of IMP-1 β-lactamase. They also showed that this gene was located on a cassette inserted in a class I integron, widely disseminated among other species of Shigella, and conferred resistance to almost all β-lactam antibiotics.⁴⁵

Carbapenem resistance conferred by bla_VIM and bla_IMP genes has recently been detected in S. sonnei and S. flexneri isolates from pediatric patients with diarrhea in the Andaman and Nicobar islands in India.⁴³ In this study, after analysis of nucleotide sequencing of bla_IMP and bla_VIM genes by BLAST, 100% similarity with sequences of these genes isolated from Acinetobacter baumannii and P. aeruginosa available at the NCBI database was confirmed. After spread of carbapenem resistance to Shigella in another part of the world, an indication of a potential public health challenge, treatment options will be limited, and infection-control measures remain of high importance.⁴³ Finally, although class A β-lactamase KPC is one of the most commonly identified carbapenemases among other Gram-negative bacteria in some parts of the world, it has not yet been identified in Shigella isolates, except for a single bla_KPC-carrying S. flexneri strain isolated from the National Senegalese Enterobacteriaceae Center located at the Pasteur Institute in Dakar.⁴⁶

Class C β-Lactamases

Ceftriaxone is recommended for treatment of ciprofloxacin-resistant Shigella isolates. However, today some strains of Shigella spp. have a resistance gene to cephalosporins. Class C β-lactamases, also known as AmpC-type enzymes, confer high-level resistance against cephalosporins. AmpC β-lactamase is encoded by both plasmid and chromosomal genes, and the first report of plasmid-encoded CMY-2-type AmpC β-lactamase was detected among ceftriaxone-resistant S. sonnei isolates obtained from an outbreak of bacillary dysentery in Taiwan.⁴⁷ CMY-2 enzymes have been reported in China, Taiwan, Costa Rica, Iran, and India from several epidemic strains.^47–50 Zhang et al found two AmpC β-lactamase producers with bla_CMY-2 and bla_DHA-1 in Shigella strains recovered from diarrhea patients in China.⁵¹

Tajbakhsh et al reported on the first AmpC β-lactamase (bla_CMY-2) producers in S. sonnei isolated from patients in Tehran, Iran.⁴⁸ In recent years, other studies conducted in Iran have indicated spread of resistance to extended-spectrum cephalosporins among Shigella isolates.^31,50 In a similar study carried out in Iran, studying cephalosporin-resistant Shigella isolates, the researchers identified gene CMY-59 in one S. sonnei isolate from pediatric patients aged <12 years.³¹ However, the majority of AmpC-positive isolates studied in other parts of the world belonging to the CMY-2 genotype and other AmpC genes (bla_MOX, bla_FOX, bla_MIR(ACT-1), bla_CIT, and bla_ACC) have been identified in Shigella isolates; however, so far there has not a study on them.^52,53 Indeed, few reports have described the presence of AmpC β-lactamases among Shigella isolates worldwide (see Table 1 for more details).^47,48

Class D β-Lactamases

Class D β-lactamases or OXA-type β-lactamases confer resistance to ampicillin and cephalothin and can hydrolyze oxacillin and cloxacillin, as well as benzylpenicillin, but they are not inhibited by tazobactam or other inhibitors.^52,54 Initially, bla_OXA β-lactamases were reported among P. aeruginosa isolates, although now bla_OXA genes have been identified in integrons and plasmids in many Gram-negative bacteria.⁵⁵ In Shigella spp., resistance to ampicillin is mainly associated with an OXA-type β-lactamase.^54,56 bla_OXA-30 was initially described in ampicillin-resistant S. flexneri strains from China in 2000.⁵⁷ Results of some studies have shown that S. flexneri isolates are a probable host specific for bla_OXA-type β-lactamase.^52,58 Another study from Iran showed that all bla_OXA-positive isolates carried bla_OXA-1 and many of them were present in S. flexneri, suggesting an individual host preference of these enzymes in S. flexneri isolates.⁵⁰

bla_OXA-1 and bla_OXA-30 genes, containing Tn2603 and Tn1409 transposons, respectively, differ from each other by having one mutation at codon 131. A gene encoding bla_OXAβ-lactamases is carried on integron.⁵⁸ Furthermore, other studies have shown that bla_OXA-30 and aadA1 are located in the gene cassettes of class 1 integrons (Figure 2). Therefore, class 1 integrons carry resistance traits for β-lactams (bla_OXA) and trimethoprim (dfrA1).⁵²

Figure 2 Physical map of Shigella atypical class 1 integron and locations of bla_OXA-30 and aadA genes.

Quinolone And Fluoroquinolone Resistance

Resistance To (Fluoro)quinolones Due To Chromosomal Target–Site Mutations

Corresponding subunits for DNA gyrase and topoisomerase IV are gyrA, gyrB, parC, and parE, encoded by genes gyrA, gyrB, parC, and parE genes, respectively. DNA gyrase consists of two gyrA subunits and two gyrB subunits, and topoisomerase IV contains two parC and two parE subunits. The most mutations have been found in a small region near the start of the gyrA gene termed a quinolone resistance–determining region (QRDR), between Ala67 and Gln107, and as reported in several studies (Table 2), most frequently mutations occur at codons 83, 87, and 211, while mutations in gyrB were detected with lower frequencies in different studies.^59,60 Some researchers believe that when a single mutation occurs in gyrA, it may confer resistance to quinolones, but for decreased susceptibility to fluoroquinolones, a number of further mutations in parC and gyrA regions are needed.^61,62 parC gene mutations most frequently occur at codons 80 among Shigella isolates. gyrA gene mutations have been confirmed to be much more prevalent than mutations in the gyrB gene.^63,64 Most nucleotide and amino-acid changes in QRDRs of gyrA, gyrB, parC, and parE among Shigella spp. are shown in Table 2. Novel mutations in QRDRs have also been identified from different regions of the world. Two novel mutations at codons 86 and 129 in parC and a mutation in codon 211 of gyrA were first reported in S. sonnei strains recovered from China in 2009.⁶⁵

Table 2 Frequency Of Amino-Acid And Nucleotide Changes In The Quinolone Resistance–Determining Regions Of ShigellaIsolates In Different Parts Of The World

Target Site Mutations	Codon	Amino-Acid Changes	Nucleotide Mutation	Shigella spp.	Country Of Detection	Reference(s)
gyrA	57	Asn→Lys	AAT→AAA	S. flexneri	China	^Citation197
	69	Gln→Trp	—	S. sonnei	India	^Citation65
	71	Phe→Ser	—	S. sonnei	India	^Citation65
	72	Ser→Pro	—	S. sonnei	India	^Citation65
	75	Met→Leu	—	S. sonnei	India	^Citation65
	80	His→Pro	CAT→CCT	S. flexneri	China	^Citation197
	80	His→ Gly	CAT→GGT	S. dysenteriae	Belgium	^Citation197
	83	Ser→Leu	TCG →TTG	S. sonnei, S. flexneri, S. dysenteriae, S. boydii	China, Bangladesh, Switzerland, Thailand, India	^Citation21,Citation58,Citation61,Citation64
	87	Asp→Asn	GAC→AAC	S. sonnei, S. flexneri, S. dysenteriae	China, Bangladesh, Switzerland, India	^Citation21,Citation52,Citation64
	87	Asp→Gly	GAC→GGC	S. sonnei, S. flexneri, S. dysenteriae	China, Switzerland, Thailand, India	^Citation21,Citation61,Citation64
	87	Asp→Tyr	GAC→TAC	S. sonnei, S. flexneri	Switzerland	^Citation61
	90	Ser→Cys	—	S. sonnei	India	^Citation65
	94	Met→Leu	—	S. sonnei	India	^Citation65
	106	His→Pro	—	S. sonnei	India	^Citation65
	161	Asn→His	—	S. sonnei	India	^Citation65
	163	Thr→Ala	—	S. sonnei	India	^Citation65
	196	Val→Ala	—	S. flexneri, S. dysenteriae	India	^Citation21
	211	His→Tyr	CAC→TAC	S. sonnei, S. flexneri	China, Bangladesh	^Citation8,Citation62,Citation64
gyrB	517	Gln→Arg	CAG→CGA	S. flexneri	China	^Citation62,Citation71
parC	64	Ala→Asp	GCC→GAC	S. sonnei, S. flexneri,	India, China	^Citation65
	64	Ala→Cys	GCC→TGC	S. sonnei	China	^Citation65
	80	Ser→Ile	AGC→ATC	S. sonnei, S. flexneri, S. dysenteriae,	China, Bangladesh, Switzerland, India	^Citation21,Citation40,Citation57,Citation61,Citation62,Citation64
	81	Ala→Pho	—	S. flexneri	China	^Citation71
	83	Ser→Leu	—	S. flexneri	China	^Citation30
	85	Ala→Thr	GCG→ACG	S. flexneri	China	^Citation71
	85	Ala→Ser	GCG→TCG	S. boydii	Switzerland	^Citation61
	86	Met→Trp	ATG→TGG	S. sonnei	China	^Citation64
	91	Gln→His	—	S. flexneri	China	^Citation71
	93	Phe→Val	—	S. flexneri, S. dysenteriae	India	^Citation21
	101	Asp→Glu	—	S. flexneri, S. dysenteriae,	India	^Citation21
	110	Asp→Glu	—	S. flexneri, S. dysenteriae	India	^Citation21
	111	Asp→His	GAT→CAT	S. flexneri	China	^Citation62,Citation64
	129	Ser→Pro	TCC→CCC	S. sonnei	China	^Citation62,Citation64
parE	408	Gly→Asp	GGC→GAC	S. flexneri	China	^Citation62
	458	Ser→Leu	TCG→TTG	S. sonnei, S. flexneri, S. dysenteriae	China, India	^Citation62
	458	Ser→Ala	TCG→GCG	S. sonnei, S. flexneri, S. dysenteriae	India	^Citation62

Note: —, unknown.

Finally, two novel mutations at codons 408 and 458 in parE have recently been discovered among Shigellaspp., isolated in India in 2013 and Jiangsu Province in China in 2016.^10,63 Mutation in codon 458 is believed to result in resistance to ciprofloxacin and nalidixic acid, while a single isolate with a mutation at codon 408 in parE is related to resistance to nalidixic acid, but susceptible to ciprofloxacin. It seems that both the novel mutations in parE of S. flexneri isolates may be correlated with the increased MIC for ciprofloxacin and mediate fluoroquinolone resistance. Also, in neither study were parE mutations identified among quinolone-sensitive isolates.^10,63

Data presented in Table 2 show that mutation patterns in gyrA and parC genes, particularly common mutations, are similar to those reported by other studies, reflecting a universal pattern among Shigella spp. A direct contribution to quinolone and fluoroquinolone resistance by each of these mutations in chromosomal target–site mutations (QRDRs) remains unknown. However, other mechanisms may be present in Shigella isolates, and further investigations are needed.

Resistance To (Fluoro)quinolones Due To Plasmid-Mediated Resistance Mechanisms

Distribution of plasmid genes called plasmid-mediated quinolone-resistance regions (PMQRs) namely qnr (qnrA, qnrB, qnrC, qnrD, qnrS, qep, aac[6ʹ]-lb-cr) genes is the main reason for resistance to quinolones among Shigella isolates, and they are usually associated with transposable or mobile elements on plasmids.^64,66 qnr genes, which are often incorporated into integrons, may allow for dissemination among Shigella and possibly other members of the Enterobacteriaceae family, and then quinolone-resistance isolates may spread across geographic regions and even across countries with population mobility.^65,67 The aac(6′)-Ib-cr gene encodes an acetyltransferase associated with reduced quinolone activity, and is identified in many members of the Enterobacteriaceae family.⁶⁸ PMQRs have been identified widely among human and animal isolates, and have become a pressing issue worldwide. In a study conducted in China, fluoroquinolone-resistance rates in animal isolates of S. flexneri were reported to be higher than those in human strains.⁶⁸

In the US, the Shigella resistance rate to fluoroquinolones reached 87% during 2014–2015.⁶⁹ Resistance of Shigella isolates to fluoroquinolone is mainly due to mutational alterations in QRDRs of DNA gyrase and topoisomerase IV genes, but PMQRs may facilitate in selection of isolates exhibiting higher levels of resistance through extrachromosomally encoded mechanisms and confer reduced susceptibility to quinolones (or fluoroquinolones).⁶⁸ aac(6′)-Ib-cr and qnrS genes were first identified in isolates of S. flexneri 2a in 1998 and S. flexneri serotype 1a in 2002, respectively.^70,71 Furthermore, the qnrS gene was identified in isolates of Shigella flexneri 2b in 2005 in Japan.⁷⁰ Also, as reported in two previous studies, aac(6′)-Ib-cr and qnrS were predominant PMQR determinants across two provinces in China, conferring high levels of fluoroquinolone resistance.^72,73 These studies indicated that aac(6′)-Ib-cr–positive Shigella isolates have been present in China for many years.^70,72,73 S. flexneri serotypes 1a, 2a, 2b, 4c, and S. sonnei carrying the qnrS gene have been reported worldwide.⁶⁷

Importantly, qnrS-positive isolates of Shigella, especially S. flexneri strains, show high-level resistance to fluoroquinolones, and many researchers from different part ofs the world suggest that the plasmid-mediated quinolone-resistance gene qnrS plays an essential role in reduced susceptibility of Shigella strains to fluoroquinolones.^67,72,73 Indeed, qnrS plasmid could change fluoroquinolone susceptibility of S. flexneri isolates containing both gyrA83 and parC80 mutations into ciprofloxacin-resistant isolates.⁷⁴ Overall, aac(6ʹ)-Ib-cr is the most prevalent gene, followed by qnrS detected in isolates of S. flexneri from the US, India, Japan, China, and Iran, and also most studies have highlighted an increased prevalence of PMQR determinants through the years.^{50,70,75–77} A recent study conducted in China reported that aac(6)-Ib-cr–positive isolates and qepA-positive isolates expressed high levels of quinolone resistance. This finding indicates that other mechanisms, such as reduced outer-membrane permeability, active efflux pumps, and harboring of different resistance genes, may be responsible for resistance to quinolones.^18,77

Fosfomycin Resistance

Fosfomycin (Fom) is a broad-spectrum antibiotic inhibiting bacterial cell-wall biogenesis by inactivating the MurA enzyme.^38,78 Despite the use of Fom in treatment of microbial infections for four decades, Fom has remained effective against common uropathogens, and Fom resistance has remained rare throughout the world.⁷⁹ However, Fom resistance was observed among E. coli strains to harbor novel transferable fosfomycin-resistance determinants named FosC2 and FosA3.⁸⁰ Two primary resistance mechanisms have been described for fosfomycin resistance: mutations in uhpA/T and glpT genes encoding proteins for two carrier-dependent systems responsible for fosfomycin uptake, and attainment of fosfomycin-modifying enzymes containing two kinases, FomA and FomB, and three types of metalloenzymes: FosX, FosA, and FosB.⁸¹

Fosfomycin-modifying enzymes were discovered for the first time among S. flexneri strains isolated from patients in China.³⁸ Some studies have suggested that increasing prevalence of fosA3 was due to dissemination of IncN and IncI plasmids, facilitating its quick dispersal.^38,80 Indeed, ESBL (bla_CTX-M-123, bla_CTX-M-55, or bla_CTX-M-15) and fosA3 genes were cocarried by transconjugant plasmids from diverse incompatibility groups, and all of them contained determinants encoding resistance to cefotaxime, ceftriaxone, and fosfomycin.^38,80 In this regard, conjugatable plasmids are likely to play an essential role in dissemination of fosA3 and ESBL genes among Shigella isolates with high clonal diversity, and they should be closely monitored.³⁸

Aminoglycoside Resistance

Aminoglycosides are used to treat a wide range of infections. Aminoglycosides mediate inhibition of protein synthesis.⁸² Resistance to aminoglycosides is associated with enzymatic inactivation, ribosomal modification, and active efflux pumps. Among these mechanisms, aminoglycoside-modifying enzymes are the most common in the clinical setting.^83–85 These enzymes are activated through three general reactions, resulting in adenylation, acetylation, or phosphorylation. Aminoglycoside adenyltransferase (aadA gene cassettes) are very common in Enterobacteriaceae, especially among Salmonella and Shigella isolates, conferring resistance to streptomycin and spectinomycin.^84,86 Indeed, streptomycin resistance is strongly associated with integrons because of the high prevalence of aadA gene cassettes within class 1 and 2 integrons. A typical class 2 integron has a gene cassette of 2.2 kb with a resistance-gene arrangement (dfrA1–sat–aadA1) conferring resistance to trimethoprim, streptothricin, and spectinomycin/streptomycin, respectively, while aadA1 was absent in atypical class 2 integrons.^87,88 Class 2 integrons have been identified in transposon Tn7 and predominantly inserted into chromosomes with high frequency.⁸⁹ An atypical class 1 integron with an unusual 3ʹ conserved sequence carrying a estX–psp–aadA2–cmlA-aadA1–qacH cassette array has been detected among different Gram-negative species (E. coli, Shigella, and Salmonella) from different hosts (human, animal, and food), periods and geographical regions. Accordingly, horizontal transfer of these integrons by plasmids promotes spread of multiple-resistance genes in sporadic and outbreak isolates of Shigella.^89–91

Many types of aadA gene cassettes have been identified among Enterobacteriaceae, but types aadA1 and aadA2 have high prevalence among Shigella isolates.^84,89,92 Aminoglycoside phosphotransferases encoded by strA and strB are the most common genes dispersed among Shigella isolates by plasmids, such as IncFII and pNV-Y394.^83,87,93,94 The gene encoding strA has been identified in 42.1% of Shigella isolates recovered from diarrheal patients in Pakistan.⁹⁵ In a study conducted in India, 100% and 88% of S. dysenteriae type 1 and S. sonnei strains harbored strA genes, encoding resistance to streptomycin.⁹⁶ The majority of Shigella isolates harbored strA and strB, along with unrelated resistance determinants, which are coded by bla_TEM, bla_CTX-M, qnrS, aadA1, tet(A), tet(B), catA, and catP.^95,96 A 6.3 kb plasmid has been detected in the S. flexneri 3a strain and is involved in a streptomycin-resistance phenotype. This plasmid is likely to cause acquired resistance to streptomycin.⁹⁷ Also, a study conducted in South Korea showed that resistance to streptomycin was mediated by strA or strB among S. sonnei isolates obtained there and revealed that tetA, strA–strB, and sul1 were encoded and present in 8.4 kb of untransferable R plasmid.⁹⁸

Occurrence of aminoglycoside-resistance genes among Shigella isolates does not occur, because these drugs have long been excluded for treatment of shigellosis in different geographical areas.⁸⁵ However, there is still a concern, because class 1 integrons containing the gene-cassette array of bla_OXA-30 + aadA1 with complete 3′CS have been reported on plasmids in Shigella spp. isolates, Salmonella enterica serovar Typhimurium and E. coli strains, and transferable plasmids may enhance spread of resistance genes within integrin, establishing arole of plasmids in horizontal transfer of resistance genes.⁹⁹

Tetracycline Resistance

Tetracyclines are used against a wide variety of diseases in humans and animals.¹⁰⁰ Tetracycline-resistant bacteria are found in opportunistic pathogens and normal flora species. According to the study by Roberts¹⁰¹ and nomenclature for tetracycline resistance genes(https://faculty.washington.edu/marilynr), five tetracycline-efflux genes — tet(A), tet(B), tet(C), tet(D), and tet(G) — and one ribosomal protection protein encoded by tet(M) have been identified among Shigella isolates, most of which are encoded in transmittable elements, with extensive dissemination in different groups of bacteria. In a study, a 20.4 kb genomic island was identified encoding MDR genes, such as a wide variety of tet genes flanked by transposases.¹⁰² This identical MDR cassette was first identified in S. flexneri serotype 2a strain YSH6000 and was referred to as Shigella resistance locus–pathogenicity island.¹⁰³ Interestingly, these MDR genes have recently been found in an E. coli plasmid, pRSB225, with a similar arrangement.104 The IncB/O/K/Z-type plasmid, termed p866, carrying resistance genes tet(A) and tet(B) have been identified in S. sonnei strains.105 These findings suggest that tet genes might be dispersed among other species by horizontal gene transfer.

Among 154 tetracycline-resistant isolates recovered as confirmed causes of traveler’s diarrhea in Spain, 79.2% (n=122) harbored at least tet(A) or tet(B). Combinations of tet(A) + tet(B), tet(A) + tet(G), and tet(B) + tet(G) were found in five, one, and seven isolates, respectively.¹⁰⁶ Results of two studies revealed that tet(A) was more frequent among S. sonnei strains, whereas tet(B) was more frequent among S. flexneri strains.^106,107 Also, S. sonnei and S. flexneri differed from each other in terms of prevalence of plasmid Inc groups.¹⁰⁶ In 50 isolates of Shigella spp. identified from stool samples collected from children with diarrhea in Iran, 90% and 18% of isolates carried tetA and tetB, respectively, and no positive results were identified for tet(C) or tet(D) in this study.¹¹ Results of the same study conducted in Iran revealed that tet(A) and tet(B) were present in 75.7% and 21.42% of Shigella spp. and that tet(A) was more frequent in S. flexneri and S. sonnei populations.¹⁰⁸

A study carried out in Mexico identified genes tet(A), tet(B), and tet(C) in 1, 6, and 18 S. sonnei isolates, and 2, 7, and 1 S. flexneri isolates, respectively, whereas tet(D) was observed only in S. sonnei isolates (8%).¹⁰⁹ Among 20 S. dysenteriae isolates from dysentery outbreaks obtained from different parts of India, tet(B) was more common (90%) than tet(A) (10%). In the same study, genes tet(A) and tet(B) were detected in 15% and 79% of Shigella isolates, respectively, samples of which were obtained from children with diarrhea in southern Mozambique.¹⁰⁷ Based on recent studies,^107,108,110 it seems that efflux-mediated tetracycline resistance to tetracycline in S. sonnei and S. flexneri strains may be related to expression of tet(A) and tet(B), respectively. According to tetracycline resistance–gene nomenclature (https://faculty.washington.edu/marilynr), tet(B) is able to confer resistance to minocycline, while other efflux pumps encoded in a transferable tet gene do not have such a property and may cause clonal dissemination of tet(B)-positive Shigella isolates worldwide. In general, tet(A) and tet(B) are the most prevalent tetracycline-resistance genes in Shigella spp. Other tetracycline-efflux genes, such as tet(C) or tet(D), are rarely detected alone.

Phenicol Resistance

Phenicols have used to treat Shigella infections during the past few years around the world, and application of them led to strong selective pressure for resistance to these antibiotics.¹¹¹ Resistance in Shigella is associated with enzymatic inactivation of unfluorinated phenicols by chloramphenicol acetyltransferase genes encoded by variants of catA (catA1, catA2, catA3) and catB (catB2, catB3, catB7, catB8), as well as active efflux by cmlA (cmlA1, cmlA4, cmlA9) and/or fluorinated and unfluorinated phenicols (floR) by major facilitator–superfamily proteins.¹¹²

Chloramphenicol resistance in Shigella isolates is mainly associated with the presence of cat genes. Among 95 Shigella isolates collected from diarrheal patients in Pakistan, at least 69 (72.6%) were resistant to chloramphenicol. catA and catP were detected in 32 (33.68%) and 24 (25.26%) isolates, respectively, from chloramphenicol-resistant Shigella isolates.⁹⁵ Among 103 S. sonnei isolates associated with several waterborne outbreaks in Taiwan, in 84% of sporadic isolates, tetB and catA genes were transferred along with 130 kb plasmids.⁸⁹ The presence of catA encoding chloramphenicol O-acetyltransferase was detected and confirmed in S. flexneri strains with strA or aadA1 genes or both.⁹⁶

Colistin Resistance

Colistin (polymyxin E) is a polypeptide antimicrobial agent interacting with outer membranes of Gram-negative bacteria. Since the first report of the plasmid-mediated polymyxin-resistance gene mcr-1 was published in an E. coli isolate in November 2016 in China,¹¹³ this gene has also been identified in Salmonella enterica and Klebsiella pneumoniae, but is rarely reported in other Enterobacteriaceae members here.¹¹⁴ This gene has now been identified in other Enterobacteriaceae genera, such as Shigella, Cronobacter, Kluyvera, and Enterobacter isolated from vegetables, the environment, food, animals, and human beings.^115,116 The mechanism of resistance of mcr-1 is produced by a phosphatidylethanolamine transferase, leading to modification of lipid A present as a result of addition of phosphoethanolamine to lipid A in cell membranes of Gram-negative bacteria, resulting in more cationic LPSs and lower affinity for colistin and related polymyxins and consequently reduced antimicrobial activity.¹¹⁷ The presence of mcr-1 in Shigella isolates leads to a four- to eightfold increase in the MIC of polymyxin B.¹¹⁸ mcr-1 is located on various plasmid backbones, including IncI2, IncFI, IncHI2, IncFIB, IncP, IncY, and IncX4, sized 58–252 kb (Figure 3).¹¹⁴ ESBL-encoding genes or other resistance genes might coexist with it. Also, plasmid-mediated colistin-resistant S. flexneri isolates recovered from animal feces on a farm showed that it might be circulated via the fecal–oral route, at least between animals on that farm, and possibly distributed via the food-production network.¹¹⁸ In most reports, mcr-1 is known to be the only resistance gene for related plasmids, indicating that selective pressure associated with polymyxin is responsible for mcr-1 acquisition.^114,118 It means that other plasmids conferring MDR phenotypes can be achieved from resistant S. flexneri strains. Mobile elements, such as IS and integrons, could also help isolates acquiring other resistance elements from the environment.¹¹⁸ A novel transposon, Tn6390 has been detected in S. flexneri C960, in which two copies of ISApl1 bracketed to the mcr-1 gene play a pivotal role in transposition of mcr-1.¹¹⁸ Recently, mcr-1 was identified in S. sonnei isolates from Shanghai (2010–2012) with polymyxin B resistance (MIC 4–8 μg/mL).¹¹⁸

Figure 3 Structure of genes surrounding mcr-1 in S. flexneri.

Sulfonamide And Trimethoprim Resistance

After the spread of trimethoprim–sulfonamide resistance in different parts of the world, these agents are currently considered ineffective for empirical therapy of shigellosis.¹¹⁹ Acquired resistance mechanisms have frequently been identified, mostly due to mutational or recombinational changes in target enzymes (dihydropteroate synthase and dihydrofolate reductase, respectively) or acquired resistance by drug-resistant target enzymes, such as acquired sul genes coding for drug-resistant dihydropteroate synthases or dfr genes coding for drug-resistant dihydrofolate reductases.^120,121

Resistance To Trimethoprim

At least 42 dfr genes conferring trimethoprim resistance have been detected in different groups of bacteria worldwide, 12 of which have been identified in Shigella spp. Resistance to trimethoprim might be explained by the presence of integron-borne dfr genes (Figure 4). Gene cassettes within class 1 integrons detected on plasmids or chromosomes in Shigella isolates often encode resistance to trimethoprim (dfrA), streptomycin (aadA), and ampicillin (oxa-1).^58,85 Class 2 integrons borne on Tn7 have often been found in Shigella spp., and gene-cassette arrays of them usually contain dfrA1, sat1, and aadA1.¹²²

Figure 4 Variable regions of class 1 (A) and class 2 (B) integrons reported in different geographic area. Horizontal arrows indicate transcriptional orientation of genes.

The presence of dfrA1 genes among Shigella isolates is the main mechanism of trimethoprim resistance, occurring in a cassette in both class 1 and class 2 integrons (Figure 4). Two types of class 2 integrons among S. sonnei strains have been identified in Japan. One of them was typical type of class 2 integrons (2,158 bp) with dfrA1, sat1, and aadA1 cassette arrays, and the other was an atypical type of class 2 integrons (1,313 bp) carried only two gene-cassette arrays with dfrA1 and sat1.¹²³ This integron‐associated antibiotic resistance can be transferred to other species via plasmid conjugation. dfr12–orfF–aadA2, dfr17–aadA5, and aadA1 cassette arrays carried by class 1 integrons have been recognized in S. sonnei isolates recovered from South Korea, China, Vietnam, and Australia.^84,85,99 Also, dfrA1–sat1–aadA1–orfX, free aadA1, or free orfX cassette arrays carried by class 2 integrons have been detected in Senegalese Shigella spp. isolates.⁹² In contrast to dfrA genes, dfrB genes have not been identified among Shigella isolates. Recently, trimethoprim-resistance clones were sequenced by primer walking, and a native 6,779 bp plasmid was identified with presence of the dfrA14 gene in a sul2–strA’–dfrA14–‘strA–strB gene arrangement in S. sonnei strains, suggesting that dfrA14 was associated with a small nonconjugative plasmid.¹²⁴ Class 2 integrons within a dfr1–sat2–aadA1 cassette array were predominant in S. sonnei isolates from outbreaks cases in Taiwan, while class 2 integrons were absent in sporadic cases.⁸⁹

A large study conducted in South Korea analyzed 122 S. sonnei isolates collected from stool samples in different parts of the country from 1991 to 2000. Resistance to trimethoprim was associated with dfrA1 and dfrA12. dfrA1 was found as a gene cassette of Tn7 located in chromosomes, while dfrA12 was located in conjugative R plasmids as a gene cassette of class 1 integrons. Tn7 was not detected in S. sonnei isolates recovered from the 1980s, while in this study Tn7 was found in all S. sonnei isolates. The authors proposed that S. sonnei isolates carrying Tn7 were responsible for outbreaks of shigellosis in different parts of South Korea in the 2000s.⁹⁸

Resistance To Sulfonamides

Since the first report of resistance to sulfonamides was found in both S. sonnei and S. flexneri isolates recovered from the early 1970s in South Korea,^98,125 resistance to this antibiotic has been identified in 94 % of S. sonnei strains from the 1980s and 100% of isolates from the 2000s and 2010s in different parts of the world.^2,50,118 Sulfonamide resistance is mediated by sul1, sul2, and sul3 genes, and they are common in Shigella.⁸⁹ The sul1 gene is highly frequent among Shigella isolates, because it is part of the 3ʹ-conserved sequence region of class 1 integrons. Also, sul3, linked to an unusual 3ʹ-conserved segment, is associated with class 1 integrons.^89,126 According to one study, sul2 is one of the three sulfonamide-resistance genes, and is usually located on large transmissible plasmids or small non-conjugative plasmids and was first detected on a small nonconjugative plasmid of E. coli.⁹⁰ Since then, this gene has been found mostly on plasmids in Shigella isolates from humans in South Korea,⁹⁸ Taiwan,⁸⁹ Australia,¹²⁷ and Bangladesh.¹²⁸

Several studies have suggested that sul genes are linked to other resistance genes. The sul1 gene is often identified together with other antimicrobial-resistance genes located on gene-cassette arrays in variable regions of class 1 integrons. Class 1 integrons differ from class 2 in their excise gene cassettes, capacity to integrate, and presence of sul1 in 3′ conserved region (3′CS). In Brazil, sul1 (sulfonamide-resistant) was identified in two (3%) MDR Shigella samples, which were also positive for class 1 integrons.¹²² In previous studies on MDR S. sonnei isolates obtained in South Korea, resistance to sulfamethoxazole was mainly associated with sul1, located in 8.4 kb of nonconjugative R plasmid.⁹⁸

Antibiotic-resistance gene clusters containing strA, strB, and sul2 are widespread among Gram-negative bacteria, particularly in Shigella isolates.¹²⁴ Iqbal et al¹²⁸ found sul1, sul2, sul3, integron 1, and integron 2 genes in all MDR S. flexneri 2a strains, and also found that sul2 was absent in all sulfamethoxazole-sensitive strains (n= 54), while it was present in all sulfamethoxazole-resistant strains (n=146). However, in this study, no change was observed in expressions of sul1, sul3, integron 1, or integron 2 genes in sulfamethoxazole-resistant and -sensitive S. flexneri 2a strains. Interestingly, curing of this 4.3 MDa plasmid resulted in loss of sul2 and susceptibility to sulfamethoxazole in paired strains, suggesting involvement of sul2 and this plasmid in resistance to sulfamethoxazole. In the same study, 24 S. sonnei isolates were detected to carry an atypical class 1 integron associated with sul3. The estX–psp–aadA2–cmlA1–aadA1–qacH cassette array of sul3-associated class 1 integron has been found to encode an esterase, a lipase, putative phosphoserine phosphatase/resistance to streptomycin, chloramphenicol, and quaternary amines.²³

Macrolide Resistance

The American Academy of Pediatrics and the Infectious Diseases Society of America have recommended azithromycin as a medication for treatment of shigellosis in children, and also the WHO introduced it as second-line treatment in adults.¹²⁹ The Centers for Disease Control and Prevention has observed resistance to azithromycin in approximately 3% of Shigella cases tested. Resistant outbreaks involving Shigella spp. isolates with reduced susceptibility to azithromycin (RSA) are more recent phenomena and continually detected in Asia, North America, the US, Australia^130,131 and other geographic regions.^132–134 According to updated Clinical and Laboratory Standards Institute guidelines, if MICmeasured by broth microdilution is ≤16 and ≤8 µg/mL, then epidemiological cutoff values denote susceptibility for S. sonnei and S. flexneri wild-type, and if MIC is ≥32 and ≥16 µg/mL, then susceptibility is confirmed for non–wild-type, respectively.¹³⁵ Recently, several reports have suggested that resistance to azithromycin in Shigella spp. isolates is associated with presence of mphA or ermB plasmid-mediated genes or by both genes.^133,136 Macrolide resistance is mediated by four main mechanisms: enzymatic inactivation by phosphotransferases encoded by mph genes or esterases encoded by ere determinants; target-site modification by an rRNA methylase encoded by erm genes; punctual mutations in rplV encoding L22 ribosomal protein, rplD encoding L4 ribosomal protein, and rrlH (23S rRNA); and drug-resistance mediated by efflux pumps, such as OmpA, OmpW, mefA, and msrA.^137,138 All macrolide-resistance mechanisms can mediate resistance to azithromycin and erythromycin. The mph(A) gene was first identified in an E. coli isolate from Japan.¹³⁹ Since then, this gene has been recognized among Pseudomonas spp., Aeromonas spp., Stenotrophomonas spp., Shigella spp., and other enteropathogens.¹³⁷ Dissemination and acquisition of macrolide mphA resistance mechanisms in Shigella spp. has been shown to be mainly due to spread of plasmids from E. coli.¹⁴⁰ All the discovered mphA-associated plasmids have been identified in E. coli isolates, indicating their role as a repository from which antimicrobial resistance to Shigella spp. may appear.¹³⁵ This phenomenon has been previously been described in E. coli donating mphA to S. sonnei.¹⁴¹

Additionally, Shigella strains with RSA have been found mostly in strains recovered from men who have sex with men (MSM; 68.8% or higher) from the Montreal region. In this study, complete sequence analysis of six selected plasmids from different serotypes of S. flexneri and S. sonnei emphasized the role of IS26 in dispersal of RSA.¹³⁰ Also, in a study conducted in Taiwan, a series of clonally related azithromycin insusceptible Shigella spp. isolates was reported in relation to MSM.¹³⁶ Various gene-transfer systems (mobile genetic element acquisitions) are involved in acquiring antibiotic-resistance genes, such as transposons, integrons, and conjugative plasmids. The IS26–mphA–mrx–mphR (A)–IS6100 gene cassette has been characterized in a clinical strain of S. boydii carrying the p2246–CTXM plasmid. Insertion sequences of IS6100 and IS26 have been found in the neighborhood of mphA in an S. sonnei strain recovered from France. Indeed, in addition to plasmid mobilization, dissemination, and acquisition of RSA among Shigella spp. and serotypes has also potentially occurred through IS26 mobilization.^130,137

IncFI and IncFII plasmids have been shown to carry the azithromycin-resistance gene erm.¹³⁵ The IncFI plasmid containing an ISCR3 insertion sequence surrounds both ermC and ermB genes and carries a bla_CTX-M-24 gene downstream of an ISEcp1 element. The IncFII plasmid carries ermB and ermC genes downstream of an IS6 transposase. Both plasmids share significant DNA homology with other previously sequenced plasmids found in S. sonnei and S. flexneri serotype 3a isolates associated with development of the disease in MSM.¹³⁵

Recently, a plasmid carrying azithromycin-resistance genes, namely pKSR100 (conjugative R-plasmid) in S. flexneri serotype 3a has been described to be involved in intercontinental spread of RSA among MSM-associated outbreak lineage.¹⁴² pKSR100-like plasmids have been found to be predominantly related to MSM-associated outbreak lineage in Australia and elsewhere.¹³¹ There are a considerable number of studies on the prevalence of RSA among Shigella isolates throughout the world, particularly in MSM, and demonstrated global dissemination of a multi-resistant plasmid, highly associated with MSM, which is present across different continents.^131,142

Biofilm Formation–Mediated Resistance

Recently, much attention has been given to biofilm formation in bacteria, because microbial cells grown in biofilms are less sensitive to antimicrobial agents and more resistant to environmental stress such as dehydration and oxidation. Microbial infections caused by biofilm-associated Shigella spp. are global health challenges.¹⁴³ Biofilm formation is regulated by a multifactorial process, with cellular adherence, exopolysaccharide secretion, and numerous gene regulations controlling detachment of bacteria from mature biofilm, and is mainly related to quorum sensing and social networking in the microbial world.^144,145 Ellafi et al investigated biofilm formation by Shigella strains grown in different NaCl concentrations, and they showed that all the isolates produced biofilm. According to their study, biofilm formation is a protective system under different environmental stress conditions.¹⁴⁶ Recent studies have demonstrated that bile salts increase the capacity of S. flexneri strains to adhere to and penetrate epithelial cells.^147,148 Indeed, extended exposure of Shigella to bile salts occurred in cases of increased biofilm formation, and thus it is an important resistance mechanism for Shigella sp. Similar biofilm phenotypes have been observed for Campylobacter, Listeria, and Vibrio, demonstrating that bile salt–induced biofilm production is conserved among members of the Enterobacteriaceae family.^19,148 Also, biofilm formation has been shown to require the presence of glucose, whereas biofilm diffusion requires the elimination of bile salts from the medium.¹⁹ Bacteria in the form of biofilm can be 100,000 times more resistant to antimicrobial agents than planktonic forms of bacteria in the same species.¹⁴³ During biofilm formation, the effect of shf, mdoH, VpsT, and LuxR-like genes and OpgH protein expression has been confirmed among enteric bacteria, as well as Shigella.^143,145 Another study described biofilm-formation potentials and pathological behaviors of various mutants S. flexneri strains with an incomplete inner core of LPS containing only Kdo moieties.

Interestingly, 1rfaC (also called waaC) mutant, with an incomplete inner core of LPS due to deficiency in Hep biosynthesis, shows strong biofilm-formation ability and considerably high invasiveness and adhesiveness to human epithelial cells compared LPS-mutant strains. However, this strategy is successful in conferring high-level resistance only in bacterial species with a deficiency in Hep synthesis of LPS.¹⁴⁹ The relationship between biofilm formation and pathogenicity, as well as virulence factors and antimicrobial properties, has not been thoroughly studied in Shigella spp., and further studies are needed.

Therapeutics

Treatment for Shigella infections is recommended to prevent spread of infection to others and to shorten disease duration. According to current WHO guidelines and a systematic review, the use of fluoroquinolones (first-line, preferably ciprofloxacin), cephalosporins (second-line), and β-lactams (second-line) for 7–10 days is recommended for treatment of shigellosis.^4,111 In regions known to have high rates of resistance to ciprofloxacin, azithromycin may be considered appropriate second-line therapy. Cefixime is also a good alternative, although its use should be balanced with respect to risk of developing antimicrobial resistance and spread of ESBL.¹¹¹ Conventional antibiotic therapeutics against shigellosis have become increasingly inefficient, due to the increase in number of MDR strains. However, no well-designed in vitro or in vivo combinations of antimicrobial agents have been performed to evaluate different antibiotic-class regimens for treating infections caused by Shigella. There has only been a related study on an antimicrobial-sensitivity case series reporting resistance in different regions and treatment outcomes for infections caused by Shigella.^4,111 In general, it has been observed that there is a decrease in susceptibility to first- and second-line agents. As such, there is an urgent need for development of novel therapeutic strategies for treatment of MDR Shigella infections.¹¹¹

Alternative Therapeutics

Natural And Organic Products

Natural products are small molecules produced naturally by microbial agents, plants, and animals that have been demonstrated to be useful in treating Shigella infections. Biotherapeutic agents (preferably probiotics) have been suggested in prevention of antibiotic-induced diarrhea, and are also an alternative therapeutic choice for treatment of gastroenteritis infectious.¹⁵⁰ Bacteria and yeast are the most frequent microorganisms used as probiotics. Several mechanisms have been suggested for antimicrobial activity of bacteria toward enteric bacterial pathogens, including production of undissociated organic acids, organic acid molecules, and bacteriocin, competition for adhesion sites, and coaggregation with pathogens.¹⁵¹

Pretreatment of cells with bacterial components and products obtained from Lactobacillus rhamnosus and L. acidophilus results in interference with Shigella adherence and internalization into host cells and leads to an absence of IL8 expression, substantiating attenuation of inflammatory response during aggregate pretreatment. Lactobacilli have been shown to regulate cytokine production and stimulate the immune system.¹⁵² Time-kill methodology has shown that viability of S. sonnei decreased after contact with cell-free culture supernatants of lactobacilli, which could be potentially used as probiotic strains in food industry.¹⁵¹ Zhang et al selected a total of 91 lactobacilli for antimicrobial activity against Shigella isolates, among which 16 lactobacilli displayed potent antibacterial activity against S. sonnei strains. The nature of these antimicrobial agents was studied and found to be dependent on production of organic acids.¹⁵³ Also, other studies have indicated that lactobacilli and lactic and acetic acid bacteria possess high activity against MDR Shigella pathogenic strains, and that they can be the best candidate for probiotics.^152,154,155 Saccharomyces boulardii is a thermophilic and nonpathogenic yeast showing antagonistic activity against several bacterial pathogens, such as enterohemorrhagic and enteropathogenic E. coli, Vibrio cholera, Salmonella typhimurium, and S. flexneri.¹⁵⁶

In one study, the effect of aqueous ethanol extract of Euphorbia prostrata was investigated in vitro and in vivo on bacterial growth of S. dysenteriae type 1 and found to be effective against Shigella isolates, with MIC and minimal bactericidal concentration of 3,500–12,000 µg/mL.¹⁵⁷ The preventive role of orally administered Aloe barbadensis Miller (Aloe vera)–supplemented probiotic lassi (APL) was determined for S. dysenteriae infection in mice, with a significant (P<0.05) decrease found in Shigella counts (log CFU/mL) and immunoprotective effects of APL against S. dysenteriae.¹⁵⁸ Antivirulence activity of a boiling black tea (Camellia sinensis) extract was shown to reduce expression of virulence traits by S. dysenteriae, as shown by decreased bacterium-survival strategies, and also an enhancement was found in innate immunoresponse against Shigella isolates.¹⁵⁹ Antishigellosis activity of Picralima nitida Stapf (Apocynaceae) extract has been found to be effective against S. dysenteriae type I strains, and MIC and minimal bactericidal concentration were 800 and 6,400 μg/mL, respectively.¹⁶⁰ In vitro antibacterial activity of methyl gallate isolated from Terminalia chebula has been shown to cause total disintegration of outer and inner membranes and leakage of cytoplasmic contents of MDR S. dysenteriae. Viable intracellular S. dysenteriae reduced in a time-dependent manner in the presence of methyl gallate and decreased to zero within 20 hours.¹⁶¹ In another study, antimicrobial activity of thyme oil and ciprofloxacin and their synergistic effects were evaluated, and the combination displayed differing degrees of effects on microbial cell formation based on results obtained from scanning electron microscopy and transmission electron microscopy. In vitro and in vivo synergy between them showed maximum growth inhibition in S. flexneri. Bacterial loads in infected colons reduced as a result of treatment with thyme oil, while the conventional drug failed to heal colon ulcers. Also, it decreased penetration of lamina propria by inflammatory cells.¹⁶²

Antishigellosis activity of a Crinum jagus water–ethanol extract was found to be effective against S. flexneri, with inhibition of diameter by 18.90 (0.39 mg/mL) and 25.36 (200 mg/mL) mm, respectively. Indeed, Crinum jagus extract drastically decreased (P<0.01) diarrheal stool emission and microbial load and also reduced IFNγ, IL2, IgM, IgA, and motilin blood levels in S. flexneri–induced diarrheic rats.¹⁶³ An in vitro study on bovine lactoferrin recognized it as a coadministered adjuvant therapeutic in antibiotic therapy against Shigella isolates. Some strains of Shigella show a twofold or more decrease in their ampicillin MIC values in the presence of bovine lactoferrin.¹⁶⁴ In vitro data showed that antibacterial activity of gallic acid inhibited the effect on biofilm formation and reduced the number of viable S. flexneri strains. Indeed, gallic acid inhibited biofilm formation in S. flexneri by regulating expression of the mdoH gene. mdoH is essential for glucosyltransferase activity and osmoregulated periplasmic glucans synthesis, as they both contribute to biofilm formation and develop antibiotic resistance in pathogenesis. Inhibition of mdoH can help in treatment of S. flexneri biofilm.¹⁴³ Organic acids, such as citric, acetic, lactic, and malic acid, are natural substances categorized as “generally recognized as safe” according to the US Food and Drug Administration. They have antimicrobial activity and are widely used to inactivate bacteria. Results of a study showed that organic acids, carvacrol, and their combination were useful against S. sonnei. However, S. sonnei was shown to decrease to 4.53 and 3.25 log CFU/mL using 0.5% w:v malic, lactic acid, respectively, indicating the synergistic effect of combination therapy.¹⁶⁵ A previous study reported the effects of dithiocarbamate transition-metal complexes on survival and recovery of pathogenic bacteria, and they are very attractive and novel pharmaceutical targets for control and management of antibiotic-resistant bacteria, as well as Shigella isolates.¹⁶⁶

Novel Therapeutic Strategies For Shigella Treatment

Nanoparticles

Nanoparticles (NPs) have gained growing importance in recent years and shown broad-spectrum antibacterial activity against pathogenic bacteria, due to their bactericidal characteristics.¹⁶⁷ NPs usually destroy bacterial targets with a damaging effect on membrane load cells and their integrity, along with generation of free oxygen radicals. Commonly, they can be delivered efficiently as antimicrobial agents. Recently, copper oxide NPs have been recognized as an antimicrobial agent for treatment of Shigella. MIC and minimum bactericidal concentration of copper oxide NPs were 2,500 μg/mL and ≤5,000 IU/mL, respectively, in treatment of S. sonnei using 33 nm NPs. The study also showed that smaller copper oxide NPs had stronger antibacterial effects than larger NPs at a specific time and concentration.¹⁶⁸ Iron is a biocompatible element, and can be directly used for treatment of many types of microbial pathogens. In one study, antimicrobial properties of Fe₂O₃ and Ag–Fe₂O₃ NPs against S. dysenteriae strains was evaluated, and both Fe₂O₃ and Ag–Fe₂O₃ NPs were shown to have antimicrobial effects, with antimicrobial activity of Ag–Fe₂O₃ NPs much more than that of Fe₂O₃ NPs alone.¹⁶⁹ The bactericidal effect of iron oxide NPs has been determined, with values of 50–100 μg/mL against Gram-positive and Gram-negative bacteria, as well as S. dysenteriae and E. coli.¹⁶⁷ The use of nanoantibiotic formulations is another strategy for treatment of drug-resistant Shigella. Mukherjee et al reported on the synthesis of a nanosized form of tetracycline by loading it in calcium phosphate NPs and showed that this treatment significantly decreased incidence of colon-length shortening, mushy-stool excretion, weight loss, and microbial colonization in gastrointestinal tracts of Shigella-infected mice. Immunohistological research has shown that as a result of tetracycline–calcium phosphate NP treatment, changes in morphology and level of inflammatory cytokines IL1β, IFNγ, and TNFα in intestinal tissue of mice caused by shigellosis were reverted to almost normal characteristics.¹⁷⁰ Silver NPs (AgNPs) are characterized by their broad-spectrum bactericidal toxic effects against broad-spectrum bacterial pathogens. Omara et al tested AgNPs against pathogenic Salmonella and Shigella strains recovered from layer-poultry farms. AgNPs at a concentration of 16 μg/mL were found to have both bacteriostatic and bactericidal effects against Salmonella and Shigella isolates.¹⁷¹ Although NPs have shown high antibacterial activity during in -vitro and in -vivo experiments, future studies and judiciously performed clinical trials are required to achieve a better understanding of their potential side effects and clear regulatory guidelines.

Phage Therapy

Bacteriophages (phages) are the most abundant organisms killing bacteria a through lysis mechanism. They can be recovered from various environments. Phage therapy has earned increasing attention due to several advantages, including high specificity to target bacteria without effects on normal microflora of the human body, replication at infection site, bactericidal activity against antibiotic-resistant bacteria, and fewer side effects than other therapies. Phages are self-limiting, because phages remain at a shallow level on target sites after killing bacterial targets.¹⁷² Phage therapy of shigellosis was discovered by a French microbiologist named d'Herelle, who described efficacy of phages in curing symptoms of dysentery.¹⁷³ Use of phages for treatment of MDR S. dysenteriae recovered from wastewater has been investigated as an alternative to antibiotics.¹⁷² Another study showed that a virulent phage named pSb1 was able to infect all the S. boydii strains and had productive lytic activity against them. Also, results indicated that pSb1 might be a member of an N4-like phage group and might have potential applications as an alternative option for treatment of shigellosis.¹⁷⁴ Regular targeting of only a subgroup of strains within one bacterial species or closely related species without causing distortions in the gut microbiota is one of the benefits of phage treatment over antibiotic therapy in treatment of shigellosis.¹⁷⁵ Numerous animal have studies demonstrated that phages are able to survive in experimental animals with dysentery. Despite no reports on significant undesirable reactions during the long history of phage therapy in humans,¹⁷⁶ phage treatments still need to overcome admission constraints in the main medical repertoire.

Vaccine Strategies

Varieties of candidate vaccines have been developed to prevent infection by Shigella spp., most of which are currently under evaluation for safety and immunogenicity. However, there is no licensed vaccine available against this pathogen. At present, studies in humans and animals have shown that protection by vaccination is possible. Potential candidates for Shigella vaccines include glycoconjugate vaccines, such as recombinant glycoconjugate, synthetic glycoconjugate, O-polysaccharide covalently linked to immunogenic carrier proteins,^177,178 virG-based live attenuated (WRSS1, WRSs3, WRSf3, WRSf2G12, WRSf2G15and WRSd1),^179,180 recombinant outer-membrane proteins,¹⁸¹ live attenuated vaccines,^182,183 invasion-plasmid antigens B, C, and D,¹⁸⁴ DNA-based vaccines, Ty21a typhoid vaccine expressing Shigella LPS,¹⁸⁵ recombinant probiotic–based candidates,¹⁸⁶ and whole-cell-killed and Shigella trivalent inactivated whole-cell and heat-killed multiserotype Shigella,^187,188 as well as novel antigen candidates, such as triacylated S-LPS, subcellular complexes purified from virulent cultures (Invaplex),¹⁸⁹ GMMA protein particles,¹⁹⁰ and an OMV-NP vaccine (Table 3).¹⁹¹

Table 3 Overview Of Shigella Vaccines In A Aifferent Phase

Class/type	Investigational Vaccines	Delivery Systems	Target Antigen (Type)	Valued Immunoresponse	Status And Results	Limitations	Development Phase	Reference
Live vaccine
Live-vaccine candidate strain	Escherichia albertii strain DM104	Intranasal	Whole-cell E. albertii	Humoral	Completed, efficacy, protection in guinea pigs	Unknown	Preclinical	^Citation198
Attenuated vaccines
Live attenuated strains	S. flexneri 2a WRSf2G12, WRSf2G15	Ocular	virG, set, senA, sbB2 genes	Humoral	Ongoing, protection in guinea pigs	Needs balancing immunogenicity with reactogenicity	Preclinical	^Citation176
	S. dysenteriae type 1 WRSd1	Oral	virG gene	Humoral	Efficacy when given orally	Transient diarrhea	Phase I	^Citation199
	S. flexneri 2a (SC602)	oral	virG, iuc genes	Humoral	Completed, efficacy, protection	ND	Phas IIB	^Citation200
	S. dysenteriae type 1 SC599	Oral	virG, ent, fep, stxA genes	Humoral	Completed, efficacy, protection	Unknown	Phase II	^Citation201
	S. sonnei WRSS1	Oral	virG gene	Humoral	Phase I, ongoing, protection in adults	Children elicited lower mucosal immunoresponses than adults	Phase I	^Citation177
	S. sonnei WRSs2, WRSs3	Oral	virG, senA, senB, msbB2 genes	Humoral	Ongoing, need for human challenge models for the efficacy of the vaccine	Fever, transient diarrhea	Phase I	^Citation176
	Trivalent of S. flexneri serotypes	Intranasal	guaBA gene	Humoral	Ongoing, prevention in guinea pigs and needs further development	Unknown	Preclinical	^Citation179
	S. flexneri 2a CVD 1208S	Oral	guaBA, set, sen genes	Humoral and cellular	Completed, efficacy, prevention and induces diverse T-CMI responses in human volunteers	ND	Phase II	^Citation180
	RNA-binding protein mutants of S. dysenteriae type 1 and S. sonnei	Ocular and oral	hfq gene and ipaBCDA plasmid	Humoral	Ongoing, induced protective immunity	Animals vaccinated in the eye showed fewer symptoms	Preclinical	^Citation181
Hybrid and live attenuated vectors	Salmonella enterica serovar Typhi strain, Ty21a	Oral	O-antigen biosynthesis gene	Humoral	Completed, efficacy, protection in mice	Unknown	Preclinical	^Citation182
	S. flexneri 2a	Intranasal and oral	virG, aroA genes	Humoral	Ongoing, efficacy and future investigation need for more attenuated recombinant mutant strains	Fever and diarrhea	Phase II	^Citation202
	Live transconjugant Shigella hybrid (LTSHΔstx) strain	Oral	stx gene	Humoral and cellular	Completed, efficacy and good protection in mice	Unknown	Preclinical	^Citation203
	Escherichia coli K12	Oral	pWR110–R64drdll invasiveness plasmid	Humoral	Stopped, causes mild diarrhea in human primates	Fever, mild diarrhea to frank dysentery	Preclinical	^Citation204
	E. coli K12, EcSf2a-3, EcSf2a-5	Oral	virG, aroD genes	Humoral	Stopped, EcSf2a was immunogenic but also reactogenic and thus not sufficiently attenuated in the guinea pig	Not sufficiently attenuated	Preclinical	^Citation204
Inactivated vaccines
Whole cell–killed vaccines	Shigella with a truncated O-SP	Intranasal	wzy gene	Humoral	Completed, efficacy, protection	ND	Preclinical	^Citation205
	S. flexneri 2a	Oral	Inactivated whole cells	Humoral and cellular	Completed, safety, robust immunoresponse	ND	Phase I	^Citation206
	S. sonnei	Oral	Inactivated whole cells	Humoral	Completed, efficacy, protection	ND	Phase I	^Citation205
	Trivalent of Shigella whole cells	Intranasal	Formalin inactivation of S. flexneri 2a, S. sonnei, and S. flexneri 3a	Humoral	Completed, efficacy, protection from lethality in guinea pig infection model	Unknown	Preclinical	^Citation184
	Heat-killed multiple serogroups/serotypes of Shigella (HKMS)	Oral	S. dysenteriae 1, S. flexneri 2a, S. flexneri 3a, S. flexneri 6, S. boydii 4 and S. sonnei	Humoral and cellular	Completed, efficacy, protection in rabbit infection model	ND	Preclinical	^Citation185
Subunit vaccines	IpaDB fusion proteins	Intradermal	Invasion plasmid antigens B and D (IpaB and IpaD)	Humoral	Ongoing, protective efficacy using a mouse pulmonary infection model	ND	Preclinical	^Citation178
	rIpaB domain rGroEL	Lung	S. flexneri IpaB, S. Typhi GroEL	Humoral	Immunogenic and protective efficacy against S. flexneri, S. boydii and S. sonnei in BALB/c mouse infection model	ND	Preclinical	^Citation207
Glycoconjugate candidates	Bioglycoconjugates	Intramuscular	S. dysenteriae 1 LPS exoprotein A	Humoral	Ongoing, protection, additional serotypes will be tested shortly	ND	Phase I	^Citation208
	Lipid-linked S. dysenteriae–type 1 O-polysaccharide	Intramuscular	O-SP Ag	Humoral	Ongoing, need future studies using synthetic saccharides of different size	Low level of antibodies	Preclinical	^Citation209
	O-polysaccharide covalently linked to immunogenic carrier proteins	Subcutaneous	O-SP Ag	Humoral	Ongoing studies	Unknown	Preclinical	^Citation174
	Synthetic oligosaccharides	Intramuscular	S. flexneri 2a O-SP Ag tetanus toxoid	Humoral	Ongoing, need future studies for the development of multivalent glycoconjugate vaccines	Unknown	Preclinical	^Citation209
Novel antigen candidates	Artificial Invaplex (recombinant IpaB and IpaC proteins with purified Shigella LPS)	Intranasal	LPS, IpaB, and IpaC Ag	Humoral	Efficacy, protection in mice, need future testes for determination of safety and immune response in humans	ND	Preclinical	^Citation186
	Triacylated S-LPS	Parenteral	Partial alkaline deacylation of S-LPS	Humoral	Completed, efficacy, protection with robust humoral immunoresponse	Unknown	Phase I	^Citation108
	GMMA protein particles	-	Outer-membrane particles from S. sonnei and S. flexneri 2a	Humoral	Completed, good safety and immunogenicity profiles in healthy adults	ND	Phase IIa	^Citation187
	ΔtolA-OMV vaccine	Intramuscular	Outer-membrane vesicles, disruption of tolA gene from S. boydii 4	Humoral	An ongoing, potentially cost-effective vaccine in the mouse infection model	ND	Preclinical	^Citation210
	OMV-nanoparticle vaccine	Intranasal	Shigella outer-membrane vesicles	Humoral	Ongoing, need future studies for the development of a multivalent vaccine	ND	Preclinical	^Citation188
Recombinant probiotic–based candidates	Lactococcus lactis bacterium like particles	Intranasal	IpaB and IpaD	Humoral	Ongoing, prevention in adult and infant mice	Unknown	Preclinical	^Citation183
	L. lactis	Subcutaneous	ompA of S. dystenteriae type 1	Humoral	Ongoing, efficacy of provoked immunoresponses in affording protection from Shigella needs to be evaluated	Unknown	Preclinical	^Citation183

Live attenuated strains such as S. dysenteriae type 1 WRSd1, S. sonnei WRSS1, WRSs2, WRSs3, and some whole-cell-killed and novel antigen candidates are now developing, and were safe and immunogenic in a phase I trial (more details are presented in Table 3). Among possible Shigella candidate vaccines, GMMA protein particles, live attenuated Shigella flexneri 2a SC602, and S. dysenteriae type 1 SC599 strains have entered phase II clinical evaluation, and only glycoconjugate candidates have already undergone Phase III trials, with other formulations still under development in patients with shigellosis.^192,193 As shown in Table 3, many studies have demonstrated humoral response as a main value of an immunoresponse to vaccination, and also fever and transient diarrhea have been repaorted as the most frequent complications in relation to some vaccine candidates in clinical investigations (Table 3). Finally, it seems that development and evaluation of multivalent candidates may provide a means for protection against serogroups/serotypes of Shigella in future.

Conclusion

In this review, antibiotic-resistance mechanisms and therapeutic strategies have been summarized regarding Shigella infection. Antimicrobial resistance in Shigella spp. is multifactorial in that it can occur through innate, acquired, or adaptive mechanisms, and infections resulting from it are exceedingly difficult to treat. Drug resistance among Shigella spp. occurs as a result of selective pressure and horizontal resistance-gene transmission. Accordingly, there is an urgent need to comprehensively learn and understand mechanisms of drug resistance among Shigella isolates, in order to develop antishigellosis drugs. The multifarious nature of antibiotic-resistance mechanisms contributes in an increase in the number of MDR strains and causes conventional antibiotic therapeutics to be highly inefficient against shigellosis. Despite intensive research efforts in the last few decades related to determination of antimicrobial resistance, researchers have not been able to find the best solution to control MDR isolates. Also, direct contributions to antibiotic resistance by many antimicrobial-resistant mechanisms remains unknown, showing a need for continuous monitoring of a broader range of associated mechanisms contributing to development of MDR strains. Most candidate vaccines cause an improvement in host immunity along with prevention of infection, but some have low efficiency and host challenge, which must be thoroughly overcome before being used against Shigella infection. More understanding on the host–microbe relationship is needed to develop novel therapeutic strategies to fight shigellosis. Development of innovative therapeutic and alternative strategies is also required for prevention and treatment of Shigella infections.

Author Contributions

Both authors conceptualized and designed the review, contributed towards data analysis, drafting and critically revising the paper, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.