Acquired Resistance to Macrolide–Lincosamide–Streptogramin Antibiotics in Lactic Acid Bacteria of Food Origin

Abstract

Antibiotic resistance is a growing problem in clinical settings as well as in food industry. Lactic acid bacteria (LAB) commercially used as starter cultures and probiotic supplements are considered as reservoirs of several antibiotic resistance genes. Macrolide–lincosamide–streptogramin (MLS) antibiotics have a proven record of excellence in clinical settings. However, the intensive use of tylosin, lincomysin and virginamycin antibiotics of this group as growth promoters in animal husbandry and poultry has resulted in development of resistance in LAB of animal origin. Among the three different mechanisms of MLS resistance, the most commonly observed in LAB are the methylase and efflux mediated resistance. This review summarizes the updated information on MLS resistance genes detected and how resistance to these antibiotics poses a threat when present in food grade LAB.

Introduction

Lactic acid bacteria (LAB) are a taxonomically diverse group of microorganisms that can convert fermentative carbohydrates into lactic acid [1]. The most typical LAB members are organisms with low G+C content, belonging to the genera Lactobacillus (L), Lactococcus (Lc), Leuconostoc (Le) and Pediococcus (P) [2]. LAB are ubiquitous in nature and important microorganisms in the gastro intestinal tract (GIT) of humans and animals [3]. In fermented foods, they are present as contaminants or deliberately added as starter cultures for preparation and preservation purposes. [4]. Owing to their long history of consumption, LAB are considered to be non pathogenic and given the status “Generally Regarded As Safe” (GRAS) [4]. For a better understanding of their safety for human consumption, European Food Safety Authority (EFSA) [5] has outlined a scheme based on qualified presumption of safety (QPS) set on establishment of identity, body of knowledge, possible pathogenicity and end use of a taxonomic group [6]. The Gram-positive bacteria considered for QPS assessment require only qualification in the assessment of susceptibility to antibiotics except for Enterococcus species as they are associated with human infections, virulence factors, transferable antibiotic resistance (AR) and lack of information on safety [6].

Antibiotic Resistance in Food LAB and its Significance

The extension of clinical use of antibiotics to non-human applications (companion animals, aquaculture and horticulture) has exerted a very strong selective pressure resulting in the appearance of resistant strains [7]. As LAB may acquire AR and play a role in their transfer to pathogenic bacteria, the food chain has been considered as the main route for the introduction of AR bacteria into the GIT [4]. A number of initiatives have been recently launched across the globe to address the biosafety concerns of starter cultures and probiotic microorganisms. In order to check for signs of transferable AR in starter cultures, EFSA [5] has proposed “microbiological breakpoints” for several genera of LAB, that have also been updated [8]. The phenotypic analysis is now accompanied by molecular tests that detect specific AR genes using single or multiplex PCR, real time PCR and/or DNA microarrays [4]. In this review, the distribution, phenotypic and genotypic resistance to MLS antibiotics, association of MLS resistance with other AR genes, transposons and their mechanism of transfer in LAB has been reviewed.

MLS Resistance in LAB

Erythromycin, produced by Saccharopolyspora eryhthraea was the first macrolide introduced in 1952 as an antibiotic against a wide range of clinical pathogens. Unfortunately, within a year, erythromycin resistant (ER^r) staphylococci from US, Europe and Japan were discovered [9]. This has led to the development and increased use of newer macrolides, and thus enhanced the exposure of clinical bacteria to macrolide group of antibiotics (9, 10). Macrolides, Lincosamides, Streptogramins, Ketolides (semi-synthetic derivatives of erythromycin A) and Oxazolidinones (MLSKO) though chemically distinct, are usually grouped together that inhibit protein synthesis [11]. Currently, there are 66 MLSKO resistance genes identified in multiple genera that fall into three major headings; (1) Modification of the target site (2) Efflux pumps and (3) Inactivation of the antibiotic(s) [10, 11] (Fig. 1). Among the resistance genes, rRNA methylase(s) (erm) are the best studied that confer resistance to macrolides, lincosamides and streptogramin B (MLS_B) group antibiotics. As the MLSK antibiotics share overlapping binding sites on the 50S ribosomal subunit, modification of the ribosomal structure by methylases reduces the binding of this group of antibiotics to their targets [9].

Fig. 1.

Mechanisms of bacterial resistance to macrolide–lincosamide–streptogramin antibiotics

The occurrence of MLS resistance in bacteria of animal origin is unlikely as these antibiotics are used mainly for clinical infections. However, administration of certain MLS antibiotics (tylosin, tilmicosin, lincomycin and virginamycin) as growth promoters and/or therapeutic agents in animal husbandry and poultry has imposed selective pressures on the development of MLS resistance in commensal bacteria [12]. There is now a growing concern regarding the food grade bacteria with frequent detection of MLS resistance genes in LAB isolated from animals, their products, fermented dairy products, starter cultures and also naturally fermented traditional foods.

MLS Resistant LAB from Farm Animals

The evidence of animals carrying MLS resistant LAB comes with the detection of erm(B) gene from diverse LAB isolated from different organs of swine [4]. A high degree of macrolide resistance was observed among Lactobacillus strains isolated from gastro intestinal tract (GIT) of chicken, pig and human and were found harboring erm genes [4, 13] (Table 1).

Table 1.

MLS resistance genes identified in lactic acid bacterial species from diverse sources

Isolate	Source	Resistance gene	Localization	Reference
L. reuteri
1044, N16,L1	Pig and chicken intestine	erm(B)		[4]
100-63	Poultry	erm(T)		[4]
8557-1, 1068, LMG-18391, 1048	Human and pig intestine	erm(B)	Plasmid	[13]
PA-16	Pig	erm(C)	Plasmid	[13]
100-67	Chicken intestine	erm(T)	Plasmid	[13]
1 strain	Fermented dry sausage	erm(B)		[19]
11 and 14	Beef	erm(B), msr(A/B)		[22]
SD 2112	Probiotic strain	lnu(A)		[43]
ATCC 55730	Commercial probiotic strain	lnu(A)	Plasmid	[44]
CH2-2	Fermented dry sausage	erm(B)		[20]
1 strain	Poultry and pork meat	erm(B)		[23]
L. sakei
5 strains	Fermented dry sausage	erm(B)		[19]
L. plantarum
3 strains	Fermented dry sausage	erm(B)		[19]
2 strains	Fermented dry sausage	erm(C)		[19]
DG507	Fermented dry sausage	erm(B)	Plasmid	[21]
80 isolates	Human origin and dairy products	erm(B)		[4]
6 strains	Poultry and pork meat	erm(B)		[23]
NWL22	Yogurt	erm(B)		[38]
L. curvatus
10 strains	Fermented dry sausage	erm(B)		[19]
26	Beef	erm(B), msr(A/B)		[19]
L. paracasei
1 strain	Fermented dry sausage	erm(B)		[19]
20	Pork	erm(B), msr(A/B)		[19]
LMG 23371 and  23372		erm(B)		[40]
L. brevis
2 strains	Fermented dry sausage	erm(B)		[19]
1 strain	Poultry and pork meat	erm(C)		[23]
L. rhamnosus
1 strain	Fermented dry sausage	erm(B)		[19]
43 isolates	Human origin	erm(B)		[4]
L. animalis
NA	Pig tonsils and nasal cavities	erm(B)		[4]
NWL39	Fermented vegetable	erm(B)		[38]
L. Johnsonii
NA	Pig tonsils and nasal cavities	erm(B)		[4]
49 isolates	Human origin	erm(B)		[4]
4 strains	Poultry and pork meat	erm(B), erm(C)		[23]
L. salivarius
NA	Pig tonsils and nasal cavities	erm(B)		[4]
3 strains	Poultry and pork meat	erm(B)		[23]
CHS1-E, CH7-1E	Fermented dry sausage	erm(B)		[20]
NWL33	Pickle	erm(B)		[38]
L. crispatus
CHCC3692	Human origin	erm(B)		[4]
L-295, L-296	Probiotic isolate	erm(B)		[42]
2 strains	Poultry and pork meat	erm(B), erm(C)		[23]
L. fermentum
LEM89	Pig faeces	erm(B)		[4]
NWL24, NWL26	Yogurt	erm(B)		[38]
ROTI	Raw milk dairy product	erm(LF), vat(E)		[53]
L. gasseri
49 isolates	Human origin and dairy products	erm(B)		[4]
E. faecium
21, 25, 27, 30	Pork	erm(B), msr(A/B)		[22]
~10 isolates	Beef processing plant	erm(B)		[17]
17 strains	Chicken, pork, meat and faecal samples	erm(B)		[30]
8 strains	Cheese and pharmaceutical product	erm(B), msr(A/B)		[51]
9 strains	Traditional fermented foods	erm(B), msr(C)		[20]
E. faecalis
78 isolates	Different organs of Swine	erm(A), erm(B), erm(C), msr(C) and mef(A/E)		[15]
~20 isolates	Beef processing plant	erm(B), vat(E)		[17]
6 strains	Chicken, pork, meat and faecal samples	erm(B)		[30]
6 strains	Milk and cheese	erm(B)		[51]
E. mundtii
1 strain	Chicken, pork, meat and faecal samples	erm(B)		[30]
E. gallinarum
1 strain	Chicken, pork, meat and faecal samples	erm(B)		[30]
E. durans
11 strains	Chicken, pork, meat and faecal samples	erm(B)		[30]
21 strains	Traditional fermented foods	erm(B), msr(C)		[20]
P. acidilactici
6990	Traditional cheese	erm(B)	Plasmid	[41]
J83	Wine	erm(B		[42]
HM3020	Stools (Clinical samples)	erm(B		[4]
AR-63	Pig or pet faeces	erm(B		[4]
P. pentosaceus
15 strains	Traditional fermented foods and curd	erm(B), msr(C)		[20]
S. agalactiae
10	Pork	erm(B), msr(A/B)		[22]
S. sanguinis
18	Pork	erm(B), msr(A/B)		[22]
Lc. Lactis
17 strains	Dairy product	erm(B)	Plasmid	[46]
CWM2143, CWM286	Bovine milk	erm(B)		[47]
3 Isolates	Poultry and pork meat	erm(B), erm(C)		[23]
L. garvieae
20 Isolates	Poultry and pork meat	erm(B), erm(C)		[23]
L. acidophilus NWL23	Yogurt	erm(B)		[38]
L. vaginalis NWL35	Dairy	erm(B)		[38]

MLS macrolide–lincosamide–streptogramin, NA Not available

In the recent time, a lot of attention has focused on enterococci as reservoirs and vehicles of AR as they readily develop resistance in response to antibiotic selective pressure [14]. Tylosin, lincomycin and neomycin are the prime antibiotics that are commonly used in animal husbandry [15]. Resistance to such antibiotics is observed in a large number of Enterococcus spp. carrying erm(B) and streptogramin A modifying enzyme, virginamycin acetyltransferase [vat(E)] genes isolated from US dairy cattle operations and commercial beef processing plant [16, 17]. Similarly, a recent work carried out by Zou et al. [15] on clinical isolates of Enterococcus faecalis (n = 78), erm(B) gene was the most common followed by erm(A), erm(C), macrolide efflux [mef(A/E)] and macrolide–streptogramin B resistant [msr(C)] efflux genes, displaying higher level of ER^r.

MLS Resistant LAB from Animal Products

As LAB are natural inhabitants of the gastrointestinal tracts of many food animals, and present in high numbers, it is often unavoidable that these organisms enter the food chain [18]. This has been substantiated with the detection of resistance genes among Lactobacillus and Lactococcus spp. isolated from meat products such as fermented dry sausage [19–21], pork, poultry and beef samples [22, 23]. Further, macrolide resistance genes detected in specimens of chicken and pork meat was comparable to that of the faecal samples raising major concerns of raw meat and fermented foods as potential vehicles for antibiotic resistance dissemination [24].

Enterococci are commonly found in the intestine of farm animals and humans. In food microbiology, they have been—like E. coli—regarded as indicators of fecal contamination [3]. The high prevalence of multiple drug resistant (MDR) enterococci in farm animals and their meat is confirmed with the detection of Enterococcus isolates resistant to tetracycline, erythromycin and vancomycin from chicken samples [25]. Similar results were also obtained by others in Enterococcus species isolated from dairy cattle, poultry and animal meat [26–30] where most of the isolates resistant to erythromycin carried erm(B) gene (Table 1).

MLS Resistant LAB from Fermented Foods

Large numbers of LAB are consumed through fermented foods to maintain microbial balance in the intestines and for their beneficial attributes [5, 31]. Although Lactobacillus, Lactococcus, Leuconostoc and Streptococcus spp. are sensitive to erythromycin, clindamycin and quinipristin/dalfopristin [32–34], resistance to these antibiotics was observed among LAB strains from cheese production environment [35] and commercial products [36–38]. Among the 473 isolates of LAB (Lactobacillus, Pediococcus and Lactococcus) analyzed by Klare et al. [39], majority of the isolates were susceptible to quinipristin/dalfopristin. However, 17 Lactobacillus isolates were resistant to one or more of the antibiotics and eight of them, including six probiotic cultures possessed erm(B) gene. This erm(B) gene was also detected among food isolates of L. paracasei, [40] and P. acidilactici isolated from traditional cheese and wine [41, 42]. In the study of Kastner et al. [43], L. reuteri SD 2112 was found to harbor lincosamide resistance gene, lincomycin nucleotidyltransferase [lnu(A)] and the same gene was found on two plasmids from a commercial strain of L. reuteri ATCC 55730 [44]. Of the several probiotic LAB of Africa and European origin, L. reuteri strain LY:12002 [45] and Lc. lactis strains from an Italian dairy product, bovine milk and meat products, high level of macrolide resistance was observed and erm(B) gene was found to be the resistance determinant [23, 46, 47].

Regarding the prevalence of AR in enterococcal strains from different environments, the frequency of MLS resistance was much lower in food isolates in comparison to clinical strains [37, 48] where Vankerkhoven et al. [49], could detect erm(B) only in one strain among the 128 E. faecium isolates. However, the studies carried out on the Moroccan food isolates [50] and probiotic strains [51] showed a high frequency of macrolide resistance in E. faecium and E. faecalis that harbored erm(B) and msr(C) genes. Such reports were also made in Enterococcus species (n = 150) isolated from raw milk cheese [52] and in our recent studies on naturally fermented foods (idli and dosa batter) and commercial dairy products [20] documenting the presence of erm(B), erm(C), msr(A/B), msr(C) and macrolide phosphotransferase (mph) encoding genes.

These observations raise the question of AR among desired food-borne bacteria with the food chain being the main route of transmission of AR bacteria between the animal and human populations [4, 41]. More specifically, fermented dairy products and fermented meats that are not heat-treated prior to consumption provide a vehicle for AR bacteria with a direct link between the animal’s indigenous flora and the human gastrointestinal tract [53].

Transfer of Conjugative Plasmids and Transposons Associated MLS Resistance

The abuse of antibiotics, a major cause of accumulation and dissemination of AR is now complicated by LAB that may act as reservoirs and transfer such resistance to pathogens [54]. The prerequisites for AR transfer from LAB to other bacteria are conjugative plasmids and transposons [53]. Lc. lactis was the first LAB in which conjugative plasmids were discovered [4]. R-plasmids encoding resistance to MLS antibiotics have been reported in Lactobacillus and Enterococcus species isolated from raw meat, silage and faeces [53, 55]. Resistance to MLS antibiotics has also been reported to be encoded by certain well characterized plasmids such as pAMb1 and RE25 where the latter encodes resistance to five macrolides and two lincosamides [4]. Using molecular techniques, erm(B), erm(C) and erm(T) genes, localized on plasmids, were detected in P. acidilactici, L. reuteri,L. plantarum and L. acidophilus [13, 43].

Conjugative transposons are the main type of vehicle for antibiotic resistance transfer and have been discovered in LAB such as E. faecalis (Tn916, Tn920, Tn925, Tn2702), E. faecium (Tn5233), Streptococcus pyogenes (Tn3701), Streptococcus agalactiae (Tn93951) and Lc. lactis (Tn5276 and Tn5301) that determine resistance to erythromycin along with chloramphenicol, tetracycline and kanamycin [53]. Due to association of macrolide resistance with conjugative plasmid and transposons, they are often found linked with other antibiotic resistance genes such as tetracycline and are also detected in LAB (11, 39, 40, 46, 51). Additionally, a recent study carried out by Vignaroli et al. [30] described the linkage of erm(B) with tetracycline [tet(M) and tet(O)], vancomycin (vanA), aminoglycoside (aac(6′)-Ie-aph(2″)-Ia) and ampicillin (blaZ) resistance genes in enterococcal isolates that could be co-transferred to the recipient through conjugation.

Among the three mechanisms (transformation, transduction and conjugation) of antibiotic resistance transfer, the impact of conjugation is significant on the global spread of antibiotic resistance mediated through conjugative plasmids and transposons [56]. Although there are few reports on conjugative transfer of MLS resistance from LAB, successful transfer of macrolide resistance from food LAB to pathogens and intra-and inter-generic LAB has been demonstrated. As depicted in Table 2, several workers have reported the role of LAB in the transfer of macrolide resistance to other bacteria under in-vitro conditions [18, 21, 38]. These studies are now extended to animal and plant models also to understand their prospective role in dissemination of resistance genes under natural conditions [54, 57–59]. Considering the evidences such as the prevalence of resistance genes and their potential to act as donor and recipient, it can be suggested that LAB act as reservoirs of MLS resistance genes that can be disseminated to pathogens. This present situation of food-grade LAB pose a threat to a variety of antibiotics especially the MLS, that have a proven record of excellence to cure illness and that are still currently used in human and veterinary medicine [20].

Table 2.

Conjugal transfer of MLS resistance from LAB

Donor	Recipient	Conjugal mating method	Transfer frequency	Mode of transmission	MLS resistance gene transferred	Reference
L. reuteri L4:12002	E. faecalis JH2-2	In vitro			erm(B)	[45]
L. plantarum pLFE1	E. faecalis	In vitro	5.7 × 10−8		erm(B)	[57]
			3 × 10−9		erm(B)
L. plantarum DG 507	E. faecalis	In vitro	3.3 × 10−7	Plasmid	erm(B)	[21]
L. plantarum DG 522	E. faecalis	In vitro	1 × 10−3	Plasmid	erm(B)	[54]
E. faecalis	E. faecalis	Sausage	10−6	pAMβ-1 Plasmid	erm(B)	[58]
	P. acidilactici UC 8840	Sausage	10−4	pAMβ-1 Plasmid	erm(B)
	S. vitulinus UC 8837	Sausage	10−3	pAMβ-1 Plasmid	erm(B)
L. fermentum NWL24	E. faecalis 181		2.6 × 10−5		erm(B)	[38]
L. lactis SH4174	L. lactis BU2-60	In vitro	2.6 × 10−2	pAMβ-1 Plasmid	erm(B)	[56]
		Animal rumen model	3.3 × 10−8	pAMβ-1 Plasmid	erm(B)
		Plant model	3.9 × 10−1	pAMβ-1 Plasmid	erm(B)
S. thermophilus	E. faecalis JH2-2	In vitro	4.1 × 10−4	Plasmid	erm(B)	[56]
		Animal rumen model	4 × 10−8	Plasmid	erm(B)
L. salivarius NWL33	E. faecalis 181	In vitro	2.9 × 10−6		erm(B)	[38]
E. faecalis CM5 V	E. faecalis OG1RF	In vitro	3 × 10−8	Plasmid	erm(B)	[30]
E. faecalis CM6 V	E. faecalis OG1RF	In vitro	1 × 10−8	Plasmid	erm(B)	[30]
E. durans PF1 V	E. faecalis 64/3	In vitro	1 × 10−7	Plasmid	erm(B)	[30]
E. durans PF3 V	E. faecalis OG1RF	In vitro	7 × 10−8	Plasmid	erm(B)	[30]
	E. faecalis 64/3	In vitro	3 × 10−6	Plasmid	erm(B)
E. faecalis LMG20790	E. faecalis JH2-2	In vitro		Tn916-Tn1545	erm(B)	[18]
E. faecalis LMG20927	E. faecalis JH2-2	In vitro		Plasmid/Tn916-Tn1545	erm(B)	[18]
Lc. lactis SH4174	Listeria monocytogenes (H7)	In vitro	5.1 × 10−4	pAMβ-1 Plasmid	erm(B)	[59]
S. thermophilus	Listeria monocytogenes (H7)	In vitro	3.1 × 10−6	Plasmid	erm(B)	[59]

Conclusions

Besides the beneficial properties of LAB as starters or probiotics, there is a great concern that these bacteria may serve as reservoirs of antibiotic resistance. This concern is strengthened due to the increasing number of strains displaying atypical resistance to antibiotics especially erythromycin and tetracycline. Macrolide antibiotics such as erythromycin and its successors were introduced to contend with the problem of methicillin resistance. Although MLS antibiotics are not used for animal therapeutic purposes, the exploitation of the macrolide tylosin in animals has resulted in cross resistance to these antibiotics. All these facts persuade undoubtedly that resistance is selected in man and animals by the use of antibiotics in organisms that are part of the normal flora. Because of their broad environmental distribution, LAB may function as reservoirs of antibiotic resistance genes that can be disseminated via the food chain or within the GIT to other bacteria including pathogens. For the food microbiologists, it is essential to avoid the distribution of bacteria with mobilizable antibiotic resistance. Therefore, strains intended to be used in feed and food systems should be systematically monitored for resistance in order to avoid their inclusion as starters and probiotics. Above all, the biosafety of the probiotic LAB for human consumption must be assessed by proposing criteria, standards, guidelines and regulations.