Rapid Emergence of Daptomycin Resistance in Clinical Isolates of Corynebacterium striatum… A Cautionary Tale

Abstract

Purpose

The objective of this study was to investigate the observation of daptomycin resistance in Corynebacterium striatum, both in vivo and in vitro. We describe a case of C. striatum bacteremia in a patient with a left ventricular assist device (LVAD); the initial isolate recovered was daptomycin susceptible with a minimum inhibitory concentration (MIC) of 0.125 μg/ml. Two months later, and after daptomycin therapy, the individual became bacteremic with an isolate of C. striatum with a daptomycin MIC of >256 μg/ml.

Methods

To study the prevalence of daptomycin resistance in C. striatum, clinical isolates of C. striatum were grown in broth culture containing daptomycin to investigate emergence of resistance to this antimicrobial. Molecular typing was used to evaluate serial isolates from the index patient and the clinical isolates of C. striatum we assayed.

Results

In vitro analysis of isolates from the index patient and seven of eleven additional C. striatum isolates exhibited emergence of high level daptomycin resistance, despite initially demonstrating low MICs to this antimicrobial agent. This phenotype was persistent even after serial subculture in the absence of daptomycin.

Conclusions

Together, these data demonstrate that caution should be taken when using daptomycin to treat high-inoculum infections and/or infections of indwelling medical devices with C. striatum. To our knowledge, this is the first report characterizing emergence of daptomycin resistance in C. striatum.

Introduction

Daptomycin is a lipopeptide antibiotic used to treat infections caused by Gram-positive bacteria; it functions by inserting its lipophilic tail into the bacterial cell membrane in a calcium-dependent manner. After insertion, daptomycin molecules aggregate, cause rapid depolarization and loss of membrane potential, leakage of cellular contents, and result in inhibition of RNA, DNA, and protein synthesis, ultimately leading to cell death. It is often considered an antibiotic of last resort and reserved to treat resistant Gram-positive pathogens such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococci (VRE) [1-5].

Historically, few Corynebacterium spp. were considered to be disease-causing pathogens and when isolated in clinical cultures, they were frequently attributed to normal skin flora and/or culture contaminants. Recent advances in bone marrow and solid organ transplantation and an increasing life expectancy for patients with chronic diseases has led to a rise in the number of immunosuppressed patients. The use of mechanical left ventricular assist devices (LVAD) in patients with advanced heart failure also continues to increase in frequency. LVADs are used as a “bridge to transplant” (weeks-years) and as long term “destination therapy” (years) and over 11,500 patients have been implanted with these devices in the United States [6]. These successes have led to a concomitant increase in serious infectious attributed to organisms that were traditionally considered to be of low virulence, such as Corynebacterium spp. and other Gram-positive flora commonly found on the skin [7,8]. Infections of indwelling hardware are difficult to treat because of the patient's lack of a functional immune response, reduced drug bioavailability at the site of infection due to poor vascularization, drug interactions, and often the hardware cannot be removed if infected. Patients with these infections require long courses of antibiotic treatment with a limited armamentarium of antibiotic options [9,10].

Daptomycin resistance has been previously documented for S. aureus and enterococci; the mechanisms of resistance are complex and multifactorial. In S. aureus, daptomycin resistance has been attributed to single nucleotide polymorphisms (SNPs) in mprF, yycG, yycH, ropB, and ropC genes as well as upregulation of the dltABCD operon [11-16]. These are genes involved in cell membrane synthesis and protein translocation across the cell membrane as well as the generalized bacterial stress response. An accumulation of SNPs in these genes results in depolarization and permeabilization of the cell membrane, increased cell wall thickening, and reduced affinity for daptomycin binding which renders the antibiotic less effective for killing these bacteria [17]. Daptomycin resistance in enterococci has been linked with mutations in genes liaF, yycG, and gdpD, which regulate cell membrane homeostasis, cell membrane phospholipid metabolism, and bacterial stress response [18-21]. Homologues of genes associated with daptomycin resistance in S. aureus and enterococci have not yet been identified in Corynebacterium spp.

Herein, we describe a patient who developed bacteremia with an isolate of C. striatum which was initially susceptible to daptomycin but became resistant while the patient was receiving daptomycin therapy. There are only two case reports in the literature of daptomycin-resistant Corynebacterium spp.; an isolate of C.jeikeium from a patient who received a hematopoietic stem cell transplant and an isolate of C. striatum from a patient on long term daptomycin therapy for native valve endocarditis [22,23]. The objective of our study was to demonstrate that daptomycin resistance can be induced in a high percentage of Corynebacterium striatum isolates in vivo and in vitro.

Case report

A 61-year-old male with a history of ischemic cardiomyopathy underwent LVAD (Heart Mate II, Thoratec, Pleasanton, CA) implantation as a bridge to heart transplantation. Two months after implantation, the patient developed an LVAD pump pocket abscess from which Staphylococcus epidermidis was isolated. After incision and drainage of the abscess, the patient received nine weeks of intravenous (IV) ceftriaxone followed by oral cephalexin for suppression; however, three weeks after finishing the IV antibiotic course, the patient presented with fever and fatigue without pain, tenderness, erythema, or purulent drainage around the LVAD driveline exit site. A computed tomography (CT) scan demonstrated a 6.5 × 3.8 × 4.7 cm abscess inferior to the LVAD pump and a smaller fluid collection adjacent to the exit site of the left ventricle conduit. At this time, C. striatum was isolated from two sets of blood cultures (VersaTREK, Thermofisher Scientific) obtained by venipuncture from two sites (susceptibility results shown in Table 1, Isolate 1). The patient was started on IV vancomycin and two days later intraoperative cultures obtained during debridement of the pump pocket abscess were negative. A transthoracic echocardiogram (TTE) did not reveal valvular or pacemaker lead vegetations. After six-weeks of vancomycin therapy, the patient developed a petechial rash with biopsy results showing superficial lymphocytic vasculitis that was attributed to vancomycin. Subsequently therapy was changed to IV daptomycin (8 mg/kg/day). Seventeen days into the new regimen, the patient presented with fevers and hypotension. Two out of two blood culture sets grew C. striatum that was non-susceptible to daptomycin (Table 1, Isolate 2). The antimicrobial regimen was changed to linezolid and prior to completion of this course the patient received an orthotopic heart transplant. Intra-operative cultures obtained from the LVAD pocket were negative; however, cultures from an implantable cardioverter-defibrillator (ICD) lead vegetation grew daptomycin-susceptible C. striatum (Table 1, Isolate 3). Six weeks of IV linezolid were completed after the heart transplant and the patient recovered well as of one year post transplantation.

Table 1.

Antibiotic susceptibility profile of three C. striatum isolates from the index patient.

	Isolate 1		Isolate 2		Isolate 3
Antibiotic	MIC (μg/ml)	Interpertation	MIC (μg/ml)	Interpertation	MIC (μg/ml)	Interpertation
Daptomycin	0.125	S	>256	R	0.125	S
Ceftriaxone	>256	R	>256	R	>256	R
Clindamycin	>256	R	>256	R	>256	R
Ciprofloxacin	>32	R	>32	R	>32	R
Linezolid	n/a	n/a	0.25	S	0.5	S
Tetracycline	1	S	0.5	S	1	S
Vancomycin	1	S	2	S	1	S

S=susceptible; R=resistant; n/a=antimicrobial susceptibility testing not performed

Isolate 1 was isolated from blood (day 1); isolate 2 was isolated from blood 66 days later; isolate 3 was isolated from tissue 78 days after the initial C. striatum isolate was isolated from blood.

Methods

Bacterial Isolates

This study characterizes isolates of C. striatum that were recovered from clinical cultures at Barnes-Jewish Hospital from February 2012 to April 2013. All isolates were recovered using standard primary culture media including sheep blood agar (Remel) and/or chocolate agar (Remel). Per standard laboratory procedures for identification of Corynebacterium spp., all isolates were catalase positive, Gram-positive rods that were identified as C. striatum (≥99% confidence) using the RapID CB Plus System (Remel, Lenexa, KS). The identity of all isolates in our study was confirmed using Bruker Biotyper MALDI-TOF MS (software version 3.0) with all isolates receiving identification scores of ≥2.0 corresponding to “acceptable for species-level identification” [24].

Strain Typing

To determine the relatedness of the C. striatum isolates, molecular typing was performed by repetitive-sequence-based PCR (repPCR) with primer RW3A, resolved by the Agilent 2100 analyzer (Agilent Technologies, Santa Clara, CA) and analyzed using Diversilab software (bioMérieux Clinical Diagnostics, Durham, NC) as previously described [25,26]. Isolates with a similarity index of <95% are considered distinct.

Antimicrobial Susceptibility Testing

Antibiotic susceptibility profiles for the case patient's three C. striatum isolates (Table 1) were determined using the gradient diffusion method Etest (bioMerieux, Durham, NC) on standard Mueller-Hinton agar containing 5% sheep blood (Remel) incubated in ambient air for 24 to 48 h. Quality control of reagents and medium and interpretation of MIC values were performed according to Clinical and Laboratory Standards Institute (CLSI) standards. For daptomycin and Corynebacterium spp., an MIC of ≤ 1 is susceptible and an MIC of >1 is non-susceptible [27], For ease of discussion, those isolates categorized as daptomycin non-susceptible will be referred to as daptomycin resistant throughout the manuscript.

Assay for Selection of Daptomycin Resistant Isolates

The three isolates recovered from the index patient in addition to twelve clinical isolates of C. striatum from unique patient specimens were each inoculated into two five ml tubes of tryptic soy broth at an optical density of a 0.5 McFarland standard. A daptomycin Etest (bioMerieux), was cut in half and fully submerged into one of the broth cultures while the other broth culture served as a growth control. Broth cultures were vortexed for five seconds to facilitate release of daptomycin from the Etest strip. The strip was left in the broth and the cultures were then incubated with constant agitation at 35°C in room air. After 24 hours of incubation, antimicrobial susceptibility testing for daptomycin was performed on the bacteria grown in both broth tubes as described above.

Results

C. striatum isolates recovered from index patient

The index patient developed blood stream infection with a C. striatum isolate that was highly resistant to daptomycin subsequent to daptomycin therapy. All isolates from our index patient were susceptible to vancomycin with MICs ranging from 1-2 μg/ml (Table 1). We performed repPCR to assess clonality of the isolates; all three isolates demonstrated a similarity index (SI) of >98.5% to one another (Figure 1A) [26,25].

Figure 1. Profile of three C. striatum isolates from our index patient.

A. Dendogram illustrating the results of repPCR strain typing analysis of the three C. striatum isolates recovered from the index patient. All isolates have a SI ≥95%, illustrating that the infections were caused by a single strain type. B. Daptomycin MIC results as evaluated by Etest for the index patient's initial C. striatum bloodstream infection (top, left), C. striatum isolated from the blood stream 66 days later (top, center), and C. striatum isolated from tissue 78 days after the initial blood stream infection (top, right). Each isolate was grown overnight in broth culture in the absence (top row) and presence (bottom row) of daptomycin.

In vitro selection for daptomycin resistant C. striatum

Since the index patient developed infection with a daptomycin-resistant C. striatum while on daptomycin therapy, we hypothesized that a large burden of organism and/or prolonged exposure to daptomycin may have caused the organism to become resistant. To test our hypothesis, we designed an in vitro assay to expose isolates to daptomycin and query for resistance. Our patient's three C. striatum isolates were grown overnight in broth culture that did not contain daptomycin resulted in MICs of 0.125, >256, and 0.125 μg/ml, respectively for patient isolates 1, 2, and 3 (Figure 1B, top row). The same isolates all had MICs of >256 μg/ml after overnight incubation in the presence of daptomycin (Figure 1B, bottom row). After selection for high-level daptomycin resistance in our patient's C. striatum isolates, they were each subcultured daily for ten days on non-selective, 5% sheep blood agar and incubated at 35°C in 5% C0₂. The daptomycin MICs continued to be >256 μg/ml for all three isolates after ten rounds of subculture, suggesting that the resistant phenotype is sustained even when daptomycin selective pressure is removed.

To determine if rapid emergence of daptomycin resistance is common among Corynebacterium, we collected twelve additional clinical isolates of C. striatum originating from a variety of specimen types (Table 2). We performed the same in vitro experiment, and of the twelve isolates, one was initially resistant to daptomycin and seven became resistant after overnight incubation in broth containing daptomycin. Four isolates remained susceptible to daptomycin (i.e. either they did not grow in the presence of daptomycin or their MIC did not change) (Table 2). All resistant isolates retained high-level resistance to daptomycin after ten rounds of subculture on non-selective media.

Table 2.

In vitro emergence of daptomycin resistance in clinical isolates of C. striatum.

Isolate #	Specimen Source	Initial Daptomycin MIC (μg/ml)	Interpretation	MIC after Incubation with Daptomycin (μg/ml)	Interpretation
1	bone, hip	0.064	S	>256	R
2	sputum#	0.125	S	0.125	S
3	bone, tibia	0.125	S	>256	R
4	bone, hip	0.064	S	>256	R
5	blood	0.125	S	8	R
6	blood	0.125	S	>256	R
7	blood	0.25	S	NG	S
8	blood	0.125	S	>256	R
9	blood	>256	R	>256	R
10	wound, ear	0.064	S	NG	S
11	blood	0.25	S	NG	S
12	blood	0.125	S	>256	R

S=susceptible; R=resistant; NG=no growth

^#Daptomycin would not be used to treat infections of the respiratory tract.

All C. striatum isolates were isolated from unique patients.

There was some variability in the C. striatum daptomycin susceptibility testing following our in vitro experiment. Although most isolates showed complete resistance to daptomycin (no zone around Etest strip), one isolate from our index patient had a zone of inhibition with pinpoint colonies within the zone (Figure 1B; bottom, right). A second clinical isolate, exhibited an MIC of 8 μg/ml after exposure to daptomycin in broth culture.

Strain Typing

RepPCR analysis of the three isolates from our index patient and the twelve clinical isolates of C. striatum showed that all but one were >95% similar, suggesting that they are the same strain type. One isolate of C. straitum obtained from a blood culture had an SI of <80% to the other isolates tested, and is a distinct strain type.

Discussion

An increasing number of immunosuppressed patients and those with long term indwelling hardware have led to an increase in the number of patients with serious, invasive Corynebacterium infections that require prolonged treatment [8,28,7]. Successful treatment of Corynebacterium infections with daptomycin has been reported in the literature, and a number of studies have demonstrated Corynebacterium spp. possessing low MICs to daptomycin in vitro [29-32]. In this report, the index patient developed a daptomycin-resistant isolate of C. striatum during treatment with daptomycin. There are only two case reports of daptomycin-resistant Corynebacterium in the literature, one isolate of C. jeikeium and a second of C. striatum [23,22], although since daptomycin susceptibility testing is rarely performed on Corynebacterium spp., it is possible that occurrence of this phenomenon is underestimated. Similar to our index patient, in both of the prior case reports, daptomycin-resistant Corynebacterium spp. were isolated from blood specimens of patients who had received daptomycin treatment. However, in contrast to our study, there was no recovery of daptomycin-susceptible Corynebacterium prior to the recovery of daptomycin-resistant Corynebacterium in either case report. Although our index patient received daptomycin treatment specifically for his C. striatum infection, both patients reported in the literature were initially treated with daptomycin for infections with other organisms (MRSA; Enterococcus faecium and Staphylococcus haemolyticus), after which they became infected with Corynebacterium spp. This highlights the role of antibiotic selection for acquisition of daptomycin-resistant Corynebacterium spp. infections in patients receiving daptomycin treatment.

Daptomycin resistance is tightly correlated with increased vancomycin MICs in isolates of S. aureus [33]. This is thought to be due to the phenotypic phenomenon of cell wall thickening found in vancomycin-intermediate S. aureus, and that the thickened cell wall may alter the net charge of the outer membrane rendering daptomycin insertion into the cell membrane less effective. The only report of daptomycin-resistant Corynebacterium that assessed cell wall thickness found no difference between daptomycin-sensitive and resistant Corynebacterium isolates [34,23]. Daptomycin-resistant Corynebacterium isolates from both case reports and our index patient were all susceptible to vancomycin [23,22].

Our in vitro experiments demonstrated that a large number of C. striatum isolates that are initially susceptible to daptomycin can rapidly acquire high-level resistance when incubated in the presence of daptomycin. The majority of C. striatum clinical isolates in which we induced daptomycin resistance in vitro were resistant to the highest concentration of daptomycin present on the Etest strip (>256 μg/ml). Of note, our in vitro experiment demonstrated decreased turbidity in broth cultures (determined by visual inspection of the broth) containing daptomycin compared to control cultures at four hours of incubation, but this lag in growth disappeared by 24 hours of incubation. The initial delay of growth in broth of the daptomycin-susceptible isolates suggests that one of two mechanisms of resistance may be occurring: a heteroresistant population may be present which is selected as the predominant population in the presence of daptomycin, or a phenotype of inducible resistance may exist. Reports of daptomycin resistance in S. aureus and enterococci show an increase in MIC, but not the extreme change demonstrated herein over a very short time interval. For example, reports of enterococci illustrate that mutations of two genes known to be involved in daptomycin resistance increased the MIC from one to twelve μg/ml [18]. Future studies are necessary to determine the mechanism of daptomycin resistance in Corynebacterium.

In the current study, daptomycin-resistant C. striatum isolates retained their resistant phenotype after subculture on non-selective agar for ten days, indicating that once acquired, daptomycin resistance in C. straitum is relatively stable. This differs from reports in S. aureus, which demonstrated that to maintain daptomycin resistance, isolates must be subcultured on media containing daptomycin [14].

Many clinical laboratories do not routinely perform antimicrobial susceptibility testing for daptomycin, especially for organisms such as Corynebacterium spp. Controversy exists surrounding the optimal method for daptomycin susceptibility testing, and some evaluations have shown variability in results for isolates that are very near to the “resistant” breakpoint [2,35-37]. However, all but one of the isolates in this study that initially demonstrated very low daptomycin MICs (0.25 μg/ml or less) subsequently demonstrated MICs of >256 μg/ml, the highest dilution resolvable using Etest, after exposure to daptomycin. It is highly unlikely that the demonstration of resistance is an artifact of our testing methodology due to the dramatic change in MIC. Additionally, the in vivo evidence, treatment failure, supports the result of daptomycin resistance in the Corynebacterium strain described herein.

In summary, our data demonstrate just how rapidly Corynebacterium striatum can develop daptomycin resistance both in vivo and in vitro. Since there are several differences between what is known about daptomycin resistance in S. aureus and enterococci compared to what we and others have observed in Corynebacterium spp., it is possible daptomycin resistance is acquired by a novel mechanism not previously described. This is a serious concern for patients receiving long term daptomycin therapy as these patients are at increased risk for developing serious, opportunistic Corynebacterium infections.