Relationship Between Different Resistance Mechanisms and Virulence in Acinetobacter baumannii

Abstract

Aim:

This study analyzed the virulence of several Acinetobacter baumannii strains expressing different resistance mechanisms using the Caenorhabditis elegans infection model.

Results:

Strains susceptible/resistant to carbapenems (presenting class D (OXA-23, OXA-24), class B metallo-β-lactamase (MBL) (NDM-1), penicillin binding protein (PBP) altered and decreased expression of Omp 33–36 kDa) and isogenic A. baumannii strains susceptible/resistant to colistin (presenting loss of lipopolysaccharide (LPS) and pmrA mutations) were included to evaluate the virulence using the C. elegans infection model. The nematode killing assay, bacterial ingestion in worms, and bacterial lawn avoidance assay were performed with the Fer-15 mutant line. A. baumannii strains generally presented low virulence, showing no difference between carbapenem-resistant strains (expressing class D, MBLs, or altered PBP) and their isogenic susceptible strains. In contrast, the absence of the Omp 33–36 kDa protein in the knockout was associated with a decrease of virulence, and a significant difference was observed between colistin-resistant mutants and their susceptible counterpart when the mechanism of resistance was associated with the loss of LPS but not with its modification.

Conclusions:

Resistance to carbapenems in A. baumannii associated with the production of OXA-type or NDM-type enzymes does not seem to affect their virulence in the C. elegans infection model. In contrast, the presence of Omp 33–36 kDa, and high level resistance to colistin related with the loss of LPS, might contribute with the virulence profile in A. baumannii.

Introduction

Acinetobacter baumannii has emerged as an important cause of nosocomial infections, most notably ventilator-associated pneumonia and bacteremia, which are associated with a high mortality, meningitis, skin and soft tissue infections, urinary tract infections, and endocarditis.^1–3 However, members of the A. calcoaceticus–A. baumannii (ACB) complex such as A. pittii and A. nosocomialis are also increasingly recognized as clinically important within the complex and have also been associated with outbreaks and infections.⁴ In addition, the increasing reports of multi extended- and pan-drug resistance in A. baumannii also represent an important concern for the treatment of the infections caused by these microorganisms which have the capability to acquire a plethora of different mechanisms of resistance.⁵ Carbapenems have been the treatment of choice for multidrug resistant (MDR) A. baumannii. However, mechanisms of carbapenem resistance such as oxacillinases (OXAs), mainly OXA-23-like, OXA-24-like, and OXA-58-like, are the most common cause of resistance to carbapenems.⁶ One of the last-line treatment options with activity against MDR strains are the polymyxins, especially colistin (polymyxin E). However, pan-drug resistant isolates, including those resistant to colistin, have already been observed.⁷

A. baumannii is not known to be particularly virulent despite having innate resistance to desiccation that contributes to survival and persistence under selective environmental pressure. However, several studies concerning the virulence factors in this pathogen have been conducted.^2,5,8–11 Russo et al. established that the K1 capsular polysaccharide from the A. baumannii strain is a clear virulence factor.¹² Kim et al. stand out the AbOmpA as an important virulence factor and could be used as a target in the development of antibiotics and vaccines against this microorganism.¹³

It is thought that the genetic and phenotypic changes that confer resistance also result in concomitant reductions in in vivo fitness, virulence, and transmission.¹⁴ However, experimental validation of this accepted paradigm is modest. Different resistance determinants lead to a modification of pathogenesis of A. baumannii (e.g., Omps and modification or loss of lipopolysaccharide [LPS]).^15–18 The pivotal role of Omp33–36 and loss of LPS have been suggested as important factors in the virulence of A. baumannii.^18–21

The infections caused by MDR organisms always result in complex interaction between the bacterial pathogenicity and the fitness costs of resistance in the human host.^22,23 The link between antimicrobial resistance and virulence traits has been previously studied, some of them using animal models. In experimental practice, the in vivo virulence of A. baumannii is usually assessed using a mouse model that is technically unsuitable to evaluate the virulence of large numbers of clinical isolates; alternative models, such as Galleria mellonella and the Caenorhabditis elegans model, have been reported in the last years.^7,23–26 The nematode C. elegans has been widely used as a nonmammalian model to study the host–pathogen interactions and virulence factors of some pathogenic bacteria.^27–31 Some of the intestinal pathogens in C. elegans are also pathogenic to humans, with many virulence factors affecting both nematodes and mammals. Hence, C. elegans model has been considered useful for screening studies of virulence factors.^32–34 With the aim to a best understanding of the pathogenesis of MDR A. baumannii and to evaluate the interest of the C. elegans model, we evaluated the virulence potential of a panel of these isolates expressing different resistance mechanisms.

Materials and Methods

Bacterial strains and growth

Experiments were performed with A. baumannii strains from the collection of the Microbiology Laboratory of the Hospital Clinic (Barcelona, Spain), including carbapenem, and colistin susceptible and resistant strains. The characteristics of the strains are presented in Table 1: four A. baumannii clinical isolates expressing OXA-23, one A. baumannii expressing OXA-24, three isogenic pair strains susceptible and resistant to carbapenems, recovered from patients pre- and post-treatment with imipenem [these strains were previously classified by pulsed-field gel electrophoresis and multilocus sequence typing³⁵ into three clones (A, B, and C)], one isolate expressing NDM-2, and one isolate expressing the Omp33–36 kDa (Omp33–36_wt) and their corresponding knockout strain constructed by insertional mutagenesis (Omp33–36_ko).²⁰ A colistin-susceptible clinical isolate (77778-ColS) and its colistin-resistant mutant strain derived in vitro (77778-ColR) after serial passages of the wild strain on increasing antibiotic concentration media; A. baumannii ATCC-19606 colistin-susceptible wild type strain and its colistin-resistant mutant strain derived in vitro (ATCC 19606-ColR) (after serial passages of the wild strain on increasing antibiotic concentration media), a pair of previously described isogenic strains recovered from a patient pre- and post-treatment with colistin (CS01 and CR17, respectively³⁶), and the spontaneous reversion strain after 15 serial passages on plates in the absence of colistin (CDR17-D15). Escherichia coli strain OP50, a nonvirulent strain corresponding to the usual worm food in the laboratory, was used as the negative control in the C. elegans model.

Table 1.

Lethal Time Fifty (LT50) of Caenorhabditis elegans Infected with Acinetobacter baumannii Strains

A. Carbapenem-resistant strains
Strain	Characteristics of the strain	Country	ST	LT50 in days (Mean ± SE)	p value (OP50 vs. others)	p-value (Between each pair)	Reference
A-CS	Carbapenem susceptible	Spain	181	8.0 ± 0.5	NS	0.5641	This study
A-CR	Carbapenem resistant: ISAba-OXA-51, OXA-24, and altered PBP	Spain	181	8.0 ± 0.5	NS		This study
B-CS	Carbapenem susceptible	Spain	154	8.0 ± 0.5	NS	0.4895	This study
B-CR	Carbapenem resistant: OXA-24 and altered PBP	Spain	154	8.0 ± 0.5	NS		This study
C-CS	Carbapenem susceptible	Spain	2	8.0 ± 0.5	NS	0.9185	This study
C-CR	Carbapenem resistant: ISAba-OXA-51 and ISAba-AmpC	Spain	2	8.0 ± 0.5	NS		This study
OXA-a	Carbapenem resistant: OXA-23	Algeria	ND	8.5 ± 0.5	NS	—	This study
OXA-b	Carbapenem resistant: OXA-23	Asia	ND	8.5 ± 0.5	NS	—	This study
OXA-c	Carbapenem resistant: OXA-23	France	ND	8.5 ± 0.5	NS	—	This study
OXA-d	Carbapenem resistant: OXA-23	Spain	ND	8.5 ± 0.5	NS	—	This study
OXA-e	Carbapenem resistant: OXA-24	France	ND	8.5 ± 0.5	NS	—	This study
NDM-2	Carbapenem resistant: A. baumannii NDM-2	Israel	103	7.0 ± 0.5	<0.0001		⁵⁵
Omp33–36_wt	A. baumannii ATCC 17978 wild type strain positive 33–36 kDa Omp	Spain	ND	7.0 ± 0.5	<0.0001	0.0337	²⁰
Omp33–36_ko	Knockout ATCC 17978 strain (A. baumannii omp33: pGEM-T)	Spain		8.0 ± 0.5	NS		²⁰

B. Colistin-resistant strains
Strains	Characteristics of the strain	Country	MIC (mg/L)	LT50 in days (Mean ± SE)	p value (OP50 vs. others)	p value (wt vs. mutant)	Reference
77778-ColS	Clinical strain	Spain	0.125	6.5 ± 0.5	<0.0001	<0.0001	This study
77778-ColR	Derivative mutant (^LPS loss: E120K mutation in lpxD*)	Spain	>256	10 ± 0.5	NS	<0.0001	This study
ATCC 19606-wt	Wild type strain	Spain	0.125	8.5 ± 0.5	NS	0.0002	This study
ATCC 19606-ColR	In vitro derivative mutant (LPS loss: G275E mutation in lpxD)	Spain	>256	9.5 ± 0.5	0.0165	0.0002	This study
CS01	Clinical strain	Spain	0.125	7.8 ± 0.5	NS	0.1695	⁵¹
CR17	Clinical derivative mutant (M12K mutation in pmrA)		64	6.6 ± 0.5	0.0041
CR17-D15	Spontaneous reversion (after 15 serial passages on plates in the absence of colistin)		0.125	7.8 ± 0.5	NS
E. coli OP50	Nonvirulent E. coli control strain		—	9.0 ± 0.5	—	—	CGC

The results are representative of at least five independent assays for each group of strains.

P, pairwise comparison between LT50s using a log rank test; LT50, 50% lethal time; SE, standard error; CS, carbapenem susceptible; CR, carbapenem resistant; Col, colistin; ST, sequence type; ND, not determined; NS, not significant; CGC, Caenorhabditis genetics center; PBP, penicillin binding protein.

Characterized by PCR and sequencing.¹⁸

Bacterial strains were cultured on Luria Bertani (LB) agar plates at 37°C.

Growth curves were performed for all the strains. Bacteria at a concentration of 5 × 10⁵ CFU/mL were grown in 5 mL of LB broth; 200 μL from a 1:1000 dilution of each cultured strain were deposited in a 96-well plate and incubated at 37°C during 48 hr in iEMS Reader MF (Labsystems) equipment. Measurements were registered every 15 min. Growth curves were performed in triplicate for each strain.

Nematode killing assay

The C. elegans infection assay was carried out using the Fer-15 mutant line, which has a temperature sensitive fertility defect. Fer-15 was provided by the Caenorhabditis Genetics Center. The method described by Lavigne et al. was applied to synchronize the growth of worms.³⁷ Eggs were collected using the hypochlorite method. Fifty microliters of Acinetobacter cultures at a concentration of 10⁵ CFU/mL were inoculated in the center of 6 cm Nematode Growth Medium (NGM) agar plates and incubated at 37°C for 2 hr. The plates were allowed to cool to room temperature and were seeded with L4 stage worms (27–30 per plate). The plates were incubated at 25°C and scored each day for live worms under a stereomicroscope Leica MS5 (Leica). A worm was considered dead when it no longer responded to touch. Five replicates were performed at independent times for each strain. The time required by A. baumannii strains to kill the worms compared with the duration observed when the worms were fed with E. coli OP50 strain was an indirect marker for the potential virulence of the studied strains. The lethal time (LT50) corresponding to the time (in days) required to kill 50% of the nematode population was registered.

Bacterial ingestion in worms

Measurement of the number of bacteria within the C. elegans digestive tract was performed by in vivo experiments according to Garsin et al.³⁸ Five C. elegans were picked at 72 hr, and the surface bacteria were removed by washing the worms twice on a BHI agar plate with 4 μL drops of M9 medium containing 50 μg/mL amikacin to remove bacteria present on the worm cuticle. The nematodes were placed in a 1.5 mL Eppendorf tube containing 20 μL of M9 medium with 1% Triton X-100 and were mechanically disrupted using a motorized pestle. The volume was adjusted to 50 μL with M9 medium containing 1% Triton X-100. The lysates were serially diluted and plated on LB agar containing 50 mg/L of ampicillin. After an overnight incubation at 37°C, colonies were counted and the colony forming units (CFU) counts per worm were determined. Three replicates were performed for each bacterial combination.

Bacterial lawn avoidance assay

It has been shown that C. elegans can discriminate between different species of bacteria and modify its olfactory preferences after exposure to pathogenic bacteria. To know if the nematodes present a specific behavior with the studied strains, the lawn avoidance assay was performed as described previously.³⁹ Small lawns of the studied strains were cultured on NGM plates overnight at 25°C. Approximately 30 young adult worms grown on E. coli OP50 were placed in the center of each bacteria lawn. The number of worms on each lawn was counted after 4, 8, and 16 hr. The results are expressed in percent occupancy corresponding to the number of worms in the bacterial lawn/the total number of worms. Three replicates were performed for each strain. A. pittii Ap41 strain was used as a positive control of the assay.

Statistical analysis

Kaplan–Meier (KM) survival curves were constructed to analyze the virulence data for each group of strains. Pairwise comparison between two different strains was carried out with a log rank test. The Kruskal–Wallis test was used to compare the in vitro bacterial growth of the different strains. The bacterial ingestion was compared using the unpaired, two-tailed Student's test. A p-value ≤0.05 obtained by GraphPad Prism 5.01 (GraphPad Software) was considered statistically significant.

Results

Virulence of A. baumannii strains

The results of the nematode killing assays are presented in Table 1.

The life span of C. elegans feeding on carbapenem-resistant strains showed different results: (1)

The majority of carbapenem-resistant strains harbored no virulence, while their LT50 were similar to LT50 of the avirulent control strain E. coli OP50. Only statistically significant differences were observed between E. coli OP50 and isolates expressing NDM-2 (p < 0.0001) and Omp 33–36 (p < 0.0001).

(2)

The four OXA-23-producing isolates (OXA-a, -b, -c, -d) showed no difference in virulence behavior. Generally, they harbored a low virulence with a LT50 = 8.5 ± 0.5 days.

(3)

Isogenic strains of A. baumannii (clones A, B, and C) showed an identical LT50 (8 ± 0.5 days). Kaplan–Meier curves did not show any difference between carbapenem-susceptible and carbapenem-resistant strains (Fig. 1a).

(4)

Isolate expressing NDM-2 was significantly more virulent than the OP50 strain (p < 0.001) and showed a LT50 = 7 ± 0.5 days.

FIG. 1.

Kaplan–Meier survival analysis illustrating the infection of Caenorhabditis elegans by some Acinetobacter baumannii strains expressing resistance mechanisms to carbapenems and colistin. (a) Isogenic pair strains susceptible and resistant to carbapenems (clone A). (b) Omp 33–36 kDa knockout strain (Omp33–36_ko) and the original strain (Omp33–36_wt). (c) Colistin-resistant strains: clinical isolate (7777-8-ColS) and its colistin-resistant mutant strain derived in vitro (77778-ColR).

(5)

The Omp33–36 kDa positive strain (Omp33–36_wt) elucidated a slight difference in comparison with the knockout counterpart (Omp33–36_ko), LT50 7 ± 0.5 days versus 8 ± 0.5 days, respectively, (p = 0.0337) (Fig. 1b).

Regarding colistin-resistant strains, the life span of C. elegans showed: (1)

The colistin-resistant strains had a very low virulence profile comparable to the E. coli OP50 strain (Table 1); notably, the ATCC 19606-ColR strain was statistically less virulent than the control E. coli OP50 strain (p = 0.0165). On the other side, a statistically significant difference was observed between E. coli OP50 and the derivative mutant of CS01 induced by treatment with colistin (CR17) (p = 0.0041).

(2)

The LT50 for the ATCC-19606 wild type was statistically significant compared to the derived in vitro colistin-resistant mutant ATCC-19606-ColR (LT50 = 8.5 vs. 9.5 ± 0.5 days, respectively) (p = 0.0002).

(3)

The LT50 of the clinical isolate 77778-ColS was statistically significant compared to the in vitro derived colistin-resistant mutant 77778-ColR (LT50 = 6.5 vs. 10 ± 0.5 days, respectively) (p < 0.0001) (Fig. 1c).

(4)

Finally, the clinical strain CS01 and its derivative mutants (derivative mutant post-treatment with colistin (strain CR17) and a spontaneous reverted strain (CR17-D15)) harbored a similar virulence (LT50 = 7.8 vs. 6.6 vs. 7.8 ± 0.5 days, respectively) (p = not significant).

Bacterial ingestion and avoidance assay

Determination of the number of bacteria within the C. elegans digestive tract confirmed the presence of A. baumannii in the nematodes' intestines and demonstrated that the differences in virulence were not due to differences in the survival and proliferation of the strains in the nematode gut, since the bacterial colonization with the strains studied in the digestive tract of nematodes measured 72 hr after ingestion was approximately 10⁵ CFU/worm without statistical difference between studied strains (Table 2).

Table 2.

Evaluation of the Colony Forming Units of the Panel of Acinetobacter baumannii Strains in the Caenorhabditis elegans Digestive Tract and Evaluation of Feeding Behavior by Measuring the Pathogen Avoidance

Strains	Median CFU (range)/nematode after 72 hr	Occupancy test after 16 hr (%, SD)	CFU p value (wt vs. mutant)	Occupancy p value (wt vs. mutant)
A-CS	4.4 × 10⁵ [(4.1–4.7) × 10⁵]	97 (±3)	NS	NS
A-CR	3.3 × 10⁵ [(3.0–3.6) × 10⁵]	98 (±2)	NS	NS
B-CS	5.8 × 10⁵ [(5.5–6.0) × 10⁵]	95 (±5)	NS	NS
B-CR	5.9 × 10⁵ [(4.4–7.2) × 10⁵]	97 (±3)	NS	NS
C-CS	3.8 × 10⁵ [(3.3–4.3) × 10⁵]	99 (±1)	NS	NS
C-CR	3.3 × 10⁵ [(2.5–4.1) × 10⁵]	98 (±2)	NS	NS
Omp33–36_wt	4.0 × 10⁵ [(3.5–4.3) × 10⁵]	99 (±1)	NS	NS
Omp33–36_ko	4.1 × 10⁵ [(3.5–4.4) × 10⁵]	99 (±1)	NS	NS
77778-ColS	5.2 × 10⁵ [(4.6–5.6) × 10⁵]	88 (±5)	NS	NS
77778-ColR	4.4 × 10⁵ [(3.7–5.1) × 10⁵]	90 (±4)	NS	NS
ATCC 19606-wt	5.0 × 10⁵ [(4.6–5.4) × 10⁵]	97 (±3)	NS	NS
ATCC 19606-ColR	3.6 × 10⁵ [(2.8–4.5) × 10⁵]	96 (±3)	NS	NS

The results are representative of at least three independent assays for each group of strains. p: The Kruskal–Wallis test was used to compare the in vitro bacterial growth of the different strains. The bacterial ingestion was compared using the unpaired, two-tailed Student's test.

CFU, colony forming units; NS, not significant; SD, standard deviation.

In addition, the results of the avoidance assay determining the occupancy of the worms on bacterial lawns were similar at around 90% among the studied strains, although without significant differences (Table 2). When larvae entered in the E. coli OP50 lawn, they remained on or near the inoculum until all the bacteria had been finished. However, when larvae were fed with Acinetobacter spp., distinct behavioral responses were noted: larvae fed with A. baumannii strains acted in the same way as with E. coli OP50. Nematodes were attracted by active bacteria that were likely to be a good food source; they initially entered the bacterial lawn and remained around the plate. In contrast, larvae fed with the control A. pitti Ap41 strain were attracted and entered the bacterial lawn but were trapped and died within 24 hr of infection. Almost 100% of the worms were enclosed and dead inside the lawn with no signs of migration or avoidance around the plate.

Bacterial growth

Bacterial growth curves determined in LB medium showed differences among the studied strains. The carbapenem-resistant strains showed a much slower growth curve than the carbapenem-susceptible strains (p < 0.001) (Supplementary Fig. S1A; Supplementary Data are available online at www.liebertpub.com/mdr). Likewise, colistin-resistant strains showed slower bacterial growth than their counterparts, particularly 77778-ColR and ATCC 19606-ColR strains with MICs >256 mg/L (p < 0.001); although CR17 (MIC 64 mg/L) also reflects the same behavior as the other two strains, it showed a less pronounced curve (Supplementary Fig. S1B, C).

Discussion

In vivo model systems are essential for our understanding of infectious diseases in human host. While mammalian models such as rodents are the first choice, nonmammalian models are useful and ethical surrogate hosts.^29,40,41 These models have been used to develop an easy model system of host–pathogen interactions to identify basic evolutionarily conserved pathways associated with microbial pathogenesis. The C. elegans infection assay is based on the capacity of pathogens ingested by C. elegans to infect and ultimately kill the worms.^27–31,34

The results of this study demonstrate that A. baumannii can kill C. elegans when presented to the nematodes as a source of nutrient, similar to what was observed in other Gram-negative microorganisms, including Pseudomonas aeruginosa, Salmonella enterica, and extraintestinal pathogenic E. coli.^31,42–45 The life span of the nematodes fed varied depending on the A. baumannii strain, which generally displayed low virulence. During the infection of C. elegans by A. baumannii, the digestive tracts of the worms were colonized in the same way as in other human pathogens. However, although accumulation of bacteria in the intestinal tract has been correlated with worm killing,⁴⁶ this apparently differed in most of the strains in the present study (approximately 10⁵ CFU/worm), suggesting that bacteria possibly adhere to the intestinal receptors of the worms and might establish persistent infection.⁴⁶ In addition, no response to avoid the infectious agent was observed with the strains studied, indicating that A. baumannii is an attractive food to the nematodes. Our results agree with those reported by Grewal and Wright who found A. calcoaceticus to be attractive to C. elegans even when dead.⁴⁷ Interestingly, it was observed that larvae fed with the A. pitti Ap41 strain were attracted and entered the bacterial lawn but were trapped and died in few hours, which could indicate that bacteria release toxic components difficult to avoid by the larvae. Pradel et al. showed that C. elegans is attracted by many bacterial metabolites, specifically avoids certain strains of Serratia based on their production of the cyclic lipodepsipentapeptide serrawettin W2, and revealed sensory mechanisms for pathogen recognition in C. elegans.³⁹

A. baumannii strains can accumulate a great variety of resistance mechanisms against most antibiotics, which together with its innate resistance to desiccation and disinfection help them to survive longer in the environment and clinical settings.^5,22 Concerning carbapenem resistance, the main mechanism in A. baumannii is the production of acquired class D β-lactamases or oxacillinases (OXA-23, OXA-24/40, OXA- 58, and OXA-143) and class B metallo-β-lactamases or MBLs (IMP, VIM, NDM, and SIM-1).⁵ In the present study, the C. elegans killing assays showed no significant difference among the isolates expressing the class D enzymes OXA-23 or OXA-24, within a single strain or isogenic pairs, and neither in the strain expressing the metallo-β-lactamase NDM-2. Hence, the influence of the carbapenem resistance mechanisms, such as the production of OXAs, MBLs, or changes in the affinity of the penicillin binding proteins (PBPs) on the ability of A. baumannii to infect the nematodes, is weak. In addition, the growth curves suggest that the expression of antimicrobial resistance influences the in vitro bacterial growth associated with lower in vitro bacterial fitness.

Other nonenzymatic mechanisms which include alteration in the permeability of bacterial membranes had been described in A. baumannii strains, among them, the outer membrane proteins (OMPs) were mainly involved in carbapenem resistance upon OMP-loss or reduced expression (47-, 44-, 37-, 33- to 36-kDa OMPs).⁵ In contrast to the enzymatic results above mentioned, the 33–36 kDa Omp positive strain (Omp33–36_wt) revealed difference in the killing assays with the knockout counterpart (Omp33–36_ko), suggesting a possible role of this OMP that might contribute with the virulence profile. Smani et al. identified 33 kDa Omp as a fibronectin binding protein, found to be an important adhesin involved in the binding of A. baumannii to human lung epithelial cells through interaction with fibronectin.²¹ In a further study, the same and other authors demonstrated that Omp33 is essential for in vivo infection of A. baumannii, since they found that the loss of Omp33 reduced both, fitness and virulence. In vitro and in a murine peritoneal sepsis model, the complemented mutant Omp33–36 strain by the 33–36 kDa omp restored the virulence of A. baumannii as observed with the wild type strain.^19,20 Vallejo et al. obtained similar results using the fertility assay in C. elegans.⁴⁸ Mutants ATCC 17978 omp33–36 were less virulent than their wild type parent strain ATCC 17978. These results, which agree with those obtained in our study using the killing assay in C. elegans, reinforce the importance of the OMPs as a virulence factor in the genera Acinetobacter. It is worth mentioning that OmpA, the main OMP in A. baumannii, is well known to play a role in the pathogenesis of this microorganism, and it has also been described in A. nosocomialis.^5,13

As mentioned previously, A. baumannii has an outstanding ability to accumulate different mechanisms of resistance against all antimicrobial agents, including polymyxins; in some cases, the only effective therapeutic alternative for the treatment of A. baumannii infections. Resistance to colistin in A. baumannii has been associated with a complete loss of LPS production through mutation in one of the genes, lpxA, lpxC, or lpxD, or by modification of lipid A components of LPS through mutations in the PmrAB two-component system.^49,50 In the present study, C. elegans killing assays showed that in vitro derived colistin-resistant strains (77778-ColR and ATCC 19606-ColR) were statistically significantly less virulent than their wild type strains (p < 0.001), and the results derived from the bacterial growth demonstrated that these A. baumannii strains highly resistant to colistin (MIC >256 mg/L, Table 1) presented a lower bacterial growth rate with respect to the susceptible counterparts. Our results agree with previous reports by López-Rojas et al. which showed that the mean survival time for an in vitro colistin mutant with an MIC of 64 mg/L in a mouse model was twice that for wild type A. baumannii ATCC-19606.^36,51 In a recent study, similar results were found in the colistin-resistant mutants of A. nosocomialis, which showed a significant decrease in virulence compared to the wild type strain.¹⁸ In this case, mutants presented several mutations in the lpxD gene, leading to the loss of LPS in the colistin-resistant strains. In our study, several mutations were detected in lpxA, lpxC, and lpxD, but only amino acid changes were detected in lpxD. Beceiro et al. agree that there is a clear fitness and virulence loss associated with colistin resistance in mutants that lack LPS compared to those that have phosphoethanolamine attached to lipid A, taking into account that acquisition of high-level colistin resistance requires genetic and structural changes that affect fitness and, hence, virulence.⁷

In contrast, the colistin-resistant clinical isolate of A. baumannii (CR17), which developed resistance after colistin treatment due to changes in the pmrA gene, showed a slight difference (not significant) in virulence in comparison with the susceptible strain (CS01) using the C. elegans model. As previously published by López-Rojas et al., CS01 was more virulent than CR17 with respect to mortality and time to death in a murine sepsis model, and the MIC to colistin was stable in the CR17 strain without spontaneous reversion.⁵¹ In our study, a spontaneous reversion was achieved after 15 serial passages on plates in the absence of colistin, recovering the susceptible phenotype, contrary to the hypothesis of stability previously described.⁵¹ Although colistin resistance may suggest a fitness and virulence cost for bacteria,⁵² our results further support the hypothesis by Anderson and Wand et al. about how resistant bacteria may improve the fitness costs by acquisition of additional fitness-compensatory mutations that may lead to more virulent strains under certain environmental conditions.^53,54 Wand et al. recently described that acquisition of colistin resistance does not always involve loss of virulence.⁵⁴ Using the Galleria mellonella killing assays, they showed that some virulent strains of A. baumannii retained virulence despite an increase in colistin resistance, even though mutations in the pmrB gene were present. In the same way, Beceiro et al. observed no difference in the virulence in the colistin-resistant pmrB mutant strain in a mouse systemic-infection model or in the C. elegans fertility model.⁷ Considering reports described above, we could suggest that mutations in the pmrA gene presented in the CR17 strain might contribute to a fitness loss but retain virulence in the same way as pmrB. In addition, it is important to highlight that the studies referred to have been carried out in several different models and the differences in the results could be influenced by this.

This work provides evidence that the acquisition of different mechanisms of resistance to carbapenems in A. baumannii, such as OXA- and NDM-type carbapenemases or altered PBPs, does not seem to affect the virulence of these strains, while in strains highly resistant to colistin due to loss of LPS, the virulence is indeed affected. The killing assays with the C. elegans model host used in this work has provided a variety of features for identifying relationships between resistance and virulence and could be a useful model for the screening of novel virulence factors in A. baumannii as the most clinically relevant within the genera, but also in other members of the ACB complex which are becoming increasingly important as pathogens that may represent a public health problem.

Footnotes

Acknowledgments

Fer-15 was provided by the Caenorhabditis Genetics Center, which is funded by the NIH National Center for Research Resources (NCRR).

This work was supported by INSERM, the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) under Research Grant (2013) to PE, the Spanish Society of Clinical Microbiology and Infectious Diseases (SEIMC) under travel grant to PE, and the Instituto de Salud Carlos III (ISCIII), Madrid, Spain under “Sara Borrell” Postdoctoral Contract CD12/00482 to DR and CD15/00017 to PE.

Disclosure Statement

None to declare. All authors have read and approved this article.

References

García-Quintanilla

, Pulido

M.R.

, López-Rojas

, Pachón

, and McConnell

M.J.

. 2013. Emerging therapies for multidrug resistant Acinetobacter baumannii. Trends Microbiol. 21:157–163.

Y.J.

, Pan

C.Z.

, Fang

C.Q.

, Zhao

Z.X.

, Chen

H.L.

, Guo

P.H.

, and Zhao

Z.W.

. 2017. Pneumonia caused by extensive drug-resistant Acinetobacter baumannii among hospitalized patients: genetic relationships, risk factors and mortality. BMC Infect. Dis. 17:371.

McConnell

M.J.

, Actis

, and Pachón

. 2012. Acinetobacter baumannii: human infections, factors contributing to pathogenesis and animal models. FEMS Microbiol. Rev. 8:1574–6976.

Espinal

, Roca

, and Vila

. 2011. Clinical impact and molecular basis of antimicrobial resistance in non-baumannii Acinetobacter. Future Microbiol. 6:495–511.

Roca

, Espinal

, Vila-Farrés

, and Vila

. 2012. The Acinetobacter baumannii Oxymoron: commensal Hospital Dweller Turned Pan-Drug-Resistant Menace. Front. Microbiol. 3:148.

Dahdouh

, Gómez-Gil

, Pacho

, Mingorance

, Daoud

, and Suarez

. 2017. Clonality, virulence determinants, and profiles of resistance of clinical Acinetobacter baumannii isolates obtained from a Spanish hospital. PLoS One, 12:e0176824.

Beceiro

, Moreno

, Fernández

, Vallejo

J.A.

, Aranda

, Adler

, Harper

, Boyce

J.D.

, and Bou

. 2014. Biological cost of different mechanisms of colistin resistance and their impact on virulence in Acinetobacter baumannii. Antimicrob. Agents Chemother. 58:518–526.

Cerqueira

G.M.

, and Peleg

A.Y.

. 2011. Insights into Acinetobacter baumannii pathogenicity. IUBMB Life, 63:1055–1060.

Espinal

, Marti

, and Vila

. 2012. Effect of biofilm formation on the survival of Acinetobacter baumannii on dry surfaces. J. Hosp. Infect. 80:56–60.

10.

Rosales-Reyes

, Gayosso-Vazquez

, Fernández-Vazquez

J.L.

, Jarillo-Quijada

M.D.

, Rivera-Benitez

, Santos-Preciado

J.I.

, and Alcantar-Curiel

M.D.

. 2017. Virulence profiles and innate immune responses against highly lethal, multidrug-resistant nosocomial isolates of Acinetobacter baumannii from a tertiary care hospital in Mexico. PLoS One, 12:e0182899.

11.

Vallenet

, Nordmann

, Barbe

, Poirel

, Mangenot

, Bataille

, Dossat

, Gas

, Kreimeyer

, Lenoble

, Oztas

, Poulain

, Segurens

, Robert

, Abergel

, Claverie

J.M.

, Raoult

, Médigue

, Weissenbach

, and Cruveiller

. 2008. Comparative analysis of Acinetobacters: three genomes for three lifestyles. PLoS One, 3:e1805.

12.

Russo

T.A.

, Beanan

J.M.

, Olson

, MacDonald

, Cox

A.D.

, St Michael

, Vinogradov

E.V.

, Spellberg

, Luke-Marshall

N.R.

, and Campagnari

A.A.

. 2013. The K1 capsular polysaccharide from Acinetobacter baumannii is a potential therapeutic target via passive immunization. Infect. Immun. 81:915–922.

13.

Kim

S.W.

, Oh

M.H.

, Jun

S.H.

, Jeon

, Kim

S.I.

, Kim

, Lee

Y.C.

, and Lee

J.C.

. 2016. Outer membrane Protein A plays a role in pathogenesis of Acinetobacter nosocomialis. Virulence. 7:413–426.

14.

Roux

, Danilchanka

, Guillard

, Cattoir

, Aschard

, Fu

, Angoulvant

, Messika

, Ricard

J.D.

, Mekalanos

J.J.

, Lory

, Pier

G.B.

, and Skurnik

. 2015. Fitness cost of antibiotic susceptibility during bacterial infection. Sci. Transl. Med. 7:297ra114.

15.

Dafopoulou

, Xavier

B.B.

, Hotterbeekx

, Janssens

, Lammens

, De

, Goossens

, Tsakris

, Malhotra-Kumar

, and Pournaras

. 2015. Colistin-Resistant Acinetobacter baumannii Clinical Strains with Deficient Biofilm Formation. Antimicrob. Agents Chemother. 60:1892–1895.

16.

Imperi

, Antunes

L.C.

, Blom

, Villa

, Iacono

, Visca

, and Carattoli

. 2011. The genomics of Acinetobacter baumannii: insights into genome plasticity, antimicrobial resistance and pathogenicity. IUBMB Life, 63:1068–1074.

17.

, Wang

, Li

, Shi

, Zhou

, Yu

, and Hua

. 2016. The Effect of Colistin Resistance-Associated Mutations on the Fitness of Acinetobacter baumannii. Front. Microbiol. 7:1715.

18.

Vila-Farrés

, Ferrer-Navarro

, Callarisa

A.E.

, Martí

, Espinal

, Gupta

, Rolain

J.M.

, Giralt

, and Vila

. 2015. Loss of LPS is involved in the virulence and resistance to colistin of colistin-resistant Acinetobacter nosocomialis mutants selected in vitro. J. Antimicrob. Chemother. 70:2981–2986.

19.

Rumbo

, Tomás

, Fernández Moreira

, Soares

N.C.

, Carvajal

, Santillana

, Beceiro

, Romero

, and Bou

. 2014. The Acinetobacter baumannii Omp33–Omp36 porin is a virulence factor that induces apoptosis and modulates autophagy in human cells. Infect. Immun. 82:4666–4680.

20.

Smani

, Dominguez-Herrera

, and Pachón

. 2013. Association of the outer membrane protein Omp33 with fitness and virulence of Acinetobacter baumannii. J. Infect. Dis. 208:1561–1570.

21.

Smani

, McConnell

M.J.

, and Pachón

. 2012. Role of fibronectin in the adhesion of Acinetobacter baumannii to host cells. PLoS One, 7:e33073.

22.

Geisinger

, and Isberg

R.R.

. 2017. Interplay between antibiotic resistance and virulence during disease promoted by multidrug-resistant bacteria. J. Infect. Dis. 215:S9–S17.

23.

Lavigne

J.P.

, Sotto

, Nicolas-Chanoine

M.H.

, Bouziges

, Bourg

, Davin-Regli

, and Pages

J.M.

. 2012. Membrane permeability, a pivotal function involved in antibiotic resistance and virulence in Enterobacter aerogenes clinical isolates. Clin. Microbiol. Infect. 18:539–545.

24.

Da Silva

G.J.

, and Mendonca

N..

2012. Association between antimicrobial resistance and virulence in Escherichia coli. Virulence. 3:18–28.

25.

Larcher

, Pantel

, Arnaud

, Sotto

, and Lavigne

J.P.

. 2017. First report of cavitary pneumonia due to community-acquired Acinetobacter pittii, study of virulence and overview of pathogenesis and treatment. BMC Infect. Dis. 17:477.

26.

Richmond

G.E.

, Evans

L.P.

, Anderson

M.J.

, Wand

M.E.

, Bonney

L.C.

, Ivens

, Chua

K.L.

, Webber

M.A.

, Sutton

J.M.

, Peterson

M.L.

, and Piddock

L.J.

. 2016. The Acinetobacter baumannii Two-Component System AdeRS regulates genes required for multidrug efflux, biofilm formation, and virulence in a strain-specific manner. MBio. 7:e00430–16.

27.

Kurz

C.L.

, and Ewbank

J.J.

. 2000. Caenorhabditis elegans for the study of host-pathogen interactions. Trends Microbiol. 8:142–144.

28.

Kurz

C.L.

, and Ewbank

J.J.

. 2007. Infection in a dish: high-throughput analyses of bacterial pathogenesis. Curr. Opin. Microbiol. 10:10–16.

29.

O'Callaghan

, and Vergunst

. 2010. Non-mammalian animal models to study infectious disease: worms or fly fishing? Curr. Opin. Microbiol. 13:79–85.

30.

Pujol

, Cypowyj

, Ziegler

, Millet

, Astrain

, Goncharov

, Jin

, Chisholm

A.D.

, and Ewbank

J.J.

. 2008. Distinct innate immune responses to infection and wounding in the C. elegans epidermis. Curr. Biol. 18:481–489.

31.

Tan

M.W.

, Rahme

L.G.

, Sternberg

J.A.

, Tompkins

R.G.

, and Ausubel

F.M.

. 1999. Pseudomonas aeruginosa killing of Caenorhabditis elegans used to identify P.aeruginosa virulence factors. Proc. Natl. Acad. Sci. USA. 96:2408–2413.

32.

Bialek

, Lavigne

J.P.

, Chevalier

, Marcon

, Leflon-Guibout

, Davin

, Moreau

, Pages

J.M.

, and Nicolas-Chanoine

M.H.

. 2010. Membrane efflux and influx modulate both multidrug resistance and virulence of Klebsiella pneumoniae in a Caenorhabditis elegans model. Antimicrob. Agents Chemother. 54:4373–4378.

33.

Ermolaeva

M.A.

, and Schumacher

. 2014. Insights from the worm: the C. elegans model for innate immunity. Semin. Immunol. 26:303–309.

34.

Ewbank

J.J.

, and Zugasti

. 2011. C. elegans: model host and tool for antimicrobial drug discovery. Dis. Models Mech. 4:300–304.

35.

Nemec

, Krizova

, Maixnerova

, Diancourt

, van der Reijden

T.J.

, Brisse

, van den Broek

, and Dijkshoorn

. 2008. Emergence of carbapenem resistance in Acinetobacter baumannii in the Czech Republic is associated with the spread of multidrug-resistant strains of European clone II. J. Antimicrob. Chemother. 62:484–489.

36.

López-Rojas

, Jimenez-Mejias

M.E.

, Lepe

J.A.

, and Pachón

. 2011. Acinetobacter baumannii resistant to colistin alters its antibiotic resistance profile: a case report from Spain. J. Infect. Dis. 204:1147–1148.

37.

Lavigne

J.-P.

, Nicolas-Chanoine

M.-H.

, Bourg

, Moreau

, and Sotto

. 2008. Virulent synergistic effect between Enterococcus faecalis and Escherichia coli assayed by using the Caenorhabditis elegans model. PLos One, 3:e3370.

38.

Garsin

D.A.

, Sifri

C.D.

, Mylonakis

, Qin

, Singh

K.V.

, Murray

B.E.

, Calderwood

S.B.

, and Ausubel

F.M.

. 2001. A simple model host for identifying Gram-positive virulence factors. Proc. Natl. Acad. Sci. USA. 98:10892–10897.

39.

Pradel

, Zhang

, Pujol

, Matsuyama

, Bargmann

C.I.

, and Ewbank

J.J.

. 2007. Detection and avoidance of a natural product from the pathogenic bacterium Serratia marcescens by Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA. 104:2295–2300.

40.

Glavis-Bloom

, Muhammed

, and Mylonakis

. 2012. Of model hosts and man: using Caenorhabditis elegans, Drosophila melanogaster and Galleria mellonella as model hosts for infectious disease research. Adv. Exp. Med. Biol. 710:11–17.

41.

Tsai

C.J.

, Loh

J.M.

, and Proft

. 2016. Galleria mellonella infection models for the study of bacterial diseases and for antimicrobial drug testing. Virulence. 7:214–229.

42.

Lavigne

J.P.

, Blanc-Potard

A.B.

, Bourg

, Moreau

, Chanal

, Bouziges

, O'Callaghan

, and Sotto

. 2006. Virulence genotype and nematode-killing properties of extra-intestinal Escherichia coli producing CTX-M beta-lactamases. Clin. Microbiol. Infect. 12:1199–1206.

43.

Nicholas

H.R.

, and Hodgkin

. 2004. Responses to infection and possible recognition strategies in the innate immune system of Caenorhabditis elegans. Mol. Immunol. 41:479–493.

44.

Pantel

, Dunyach-Remy

, Ngba Essebe

, Mesureur

, Sotto

, Pages

J.M.

, Nicolas-Chanoine

M.H.

, and Lavigne

J.P.

. 2016. Modulation of Membrane Influx and Efflux in Escherichia coli Sequence Type 131 Has an Impact on Bacterial Motility, Biofilm Formation, and Virulence in a Caenorhabditis elegans Model. Antimicrob. Agents Chemother. 60:2901–2911.

45.

Peleg

A.Y.

, Seifert

, and Paterson

D.L.

. 2008. Acinetobacter baumannii: emergence of a successful pathogen. Clin. Microbiol. Rev. 21:538–582.

46.

Sifri

C.D.

, Begun

, and Ausubel

F.M.

. 2005. The worm has turned—microbial virulence modeled in Caenorhabditis elegans. Trends Microbiol. 13:119–127.

47.

Grewal

, and Wright

. 1992. Migration of Caenorhabditis elegans (Nematoda: Rhabditidae) larvae towards bacteria and the nature of the bacterial stimulus. Fundam. Appl. Nematod. 15:159–166.

48.

Vallejo

J.A.

, Beceiro

, Rumbo-Feal

, Rodriguez-Palero

M.J.

, Russo

T.A.

, and Bou

. 2015. Optimisation of the Caenorhabditis elegans model for studying the pathogenesis of opportunistic Acinetobacter baumannii. Int. J. Antimicrob. Agents. 9:00241–1.

49.

Adams

M.D.

, Nickel

G.C.

, Bajaksouzian

, Lavender

, Murthy

A.R.

, Jacobs

M.R.

, and Bonomo

R.A.

. 2009. Resistance to colistin in Acinetobacter baumannii associated with mutations in the PmrAB two-component system. Antimicrob. Agents Chemother. 53:3628–3634.

50.

Moffatt

J.H.

, Harper

, Harrison

, Hale

J.D.

, Vinogradov

, Seemann

, Henry

, Crane

, Michael St

, Cox

A.D.

, Adler

, Nation

R.L.

, Li

, and Boyce

J.D.

. 2010. Colistin resistance in Acinetobacter baumannii is mediated by complete loss of lipopolysaccharide production. Antimicrob. Agents Chemother. 54:4971–4977.

51.

López-Rojas

, McConnell

M.J.

, Jimenez-Mejias

M.E.

, Dominguez-Herrera

, Fernández-Cuenca

, and Pachon

. 2013. Colistin resistance in a clinical Acinetobacter baumannii strain appearing after colistin treatment: effect on virulence and bacterial fitness. Antimicrob. Agents Chemother. 57:4587–4589.

52.

López-Rojas

, Dominguez-Herrera

, McConnell

M.J.

, Docobo-Perez

, Smani

, Fernandez-Reyes

, Rivas

, and Pachón

. 2011. Impaired virulence and in vivo fitness of colistin-resistant Acinetobacter baumannii. J. Infect. Dis. 203:545–548.

53.

Andersson

D.I.

, and Hughes

. 2010. Antibiotic resistance and its cost: is it possible to reverse resistance? Nat. Rev. Microbiol. 8:260–271.

54.

Wand

M.E.

, Bock

L.J.

, Bonney

L.C.

, and Sutton

J.M.

. 2015. Retention of virulence following adaptation to colistin in Acinetobacter baumannii reflects the mechanism of resistance. J. Antimicrob. Chemother. 70:2209–2216.

55.

Espinal

, Fugazza

, López

, Kasma

, Lerman

, Malhotra-Kumar

, Goossens

, Carmeli

, and Vila

. 2011. Dissemination of an NDM-2-producing Acinetobacter baumannii clone in an Israeli rehabilitation center. Antimicrob. Agents Chemother. 55:5396–5398.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.34 MB