sign in sign up
<<
Home , Buruli ulcer, Mycobacterium, Mycobacterium abscessus, Mycobacterium bolletii, Mycobacterium chelonae, Mycobacterium fortuitum

Current Infectious Disease Reports (2018) 20:6 https://doi.org/10.1007/s11908-018-0611-3

SKIN, SOFT TISSUE, BONE AND JOINT (N SAFDAR AND A POP-VICAS, SECTION EDITORS)

Skin and Soft Tissue Infections Due to Nontuberculous Mycobacteria

Elizabeth Ann Misch1 & Christopher Saddler2 & James Muse Davis3

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract Purpose of Review This review describes recent trends in the epidemiology of nontuberculous mycobacteria (NTM), emerging , new insights into NTM pathogenesis, and advances in diagnosis and treatment. Recent Findings Emerging pathogens include chimaera and drug-resistant subspecies of . Important mechanisms of pathogenic NTM include the ability to alter the ’spermis- siveness to intracellular bacterial growth. New diagnostic tools consist of DNA probes, gene sequencing, and matrix- assisted laser desorption ionization-time of flight. These methods allow rapid speciation of NTM , in some cases directly from patient samples. There are few novel agents available to treat NTM, although some repurposed drugs show excellent activity. Summary The incidence of NTM infections appears to be increasing in a number of regions around the world. Molecular methods are now the diagnostic tools of choice. Discovery of novel effective agents and/or drug combinations with greater likelihood of cure, shorter treatment duration, and fewer side effects are research priorities.

Keywords Nontuberculousmycobacteria .M.chimaera .Skinandsofttissue .Rapidlygrowingmycobacteria(RGM) .Molecular diagnostics

Introduction and Background complex1 and , the agent of . More than 190 species have been identified Nontuberculous mycobacteria (NTM) encompass all (http://www.bacterio.net/mycobacterium.html) [1], of mycobacteria species other than those in the Mycobacterium which approximately 40 species are considered patho- genic [2]. Mycobacteria are ubiquitous in the natural and constructed environment. They can be found in all This article is part of the Topical Collection on Skin, Soft Tissue, Bone and regions of the world in , natural or treated water Joint Infections (including tap and shower water), and in association • * Elizabeth Ann Misch with plants, birds, fish, and other animals [3 , 4, 5, 6]. [email protected] A few species, including Mycobacterium haemophilum and , are rarely isolated from the environment [7], but this fact may reflect special 1 Division of Allergy and Infectious Disease, Department of Medicine, School of Medicine and Public Health, University of Wisconsin, growth requirements (M. haemophilum), or the extended in- Medical Foundation Centennial Building, 1685 Highland Avenue, cubation periods (M. ulcerans) needed for successful culture. 5th floor, Madison, WI 53705-2281, USA In humans, NTM are facultative intracellular pathogens. 2 Division of Allergy and Infectious Disease, Department of Medicine, They may be present as colonizers or pathogens in the respi- School of Medicine and Public Health, University of Wisconsin, ratory tract. Asymptomatic is probably far more Madison, WI, USA 3 Division of Infectious Disease, Department of Pediatrics, School of Medicine and Public Health, University of Wisconsin, Madison, WI, 1 M. tuberculosis complex members: M. africanum, M. bovis, M. cannetti, USA M. caprae, M. microti, M. pinnepedii,andM. tuberculosis. 6 Page 2 of 17 Curr Infect Dis Rep (2018) 20:6 frequent than disease. Antibodies against , dissemination in immune-compromised hosts [27]. There is a glycolipid abundant in the of mycobacteria, are little evidence of person-to-person transmission of NTM, with detected in increasing numbers of healthy children after the thepossibleexceptionofM. abscessus,whichhasbeenim- age of 1 [8]. In the USA, skin test reactivity to Mycobacterium plicated in outbreaks of clonally related strains in cystic fibro- avium sensitin has been reported in 16% of adults who other- sis patients at two different centers [3•, 28, 29]. wise would have been misclassified as having latent tubercu- Mycobacteria elaborate a lipid-rich, waxy cell wall that losis infection (due to cross-reactivity of the immune response functions as a hydrophobic and allows organisms to to environmental mycobacteria and purified protein derivative adhere to hard surfaces, including pipes, drains, and tubing. (PPD)) [9]. In one survey, 33% of adults showed evidence of inside pipes resist being physically dislodged despite prior, presumably subclinical, infection with M. avium [10]. high flow rates. Because of their biofilm-forming ability, en- In adults, NTM are most frequently associated with lung dis- vironmental mycobacteria are highly resistant to decontami- ease. In children, lymphadenitis is the most common disease nation with standard antiseptics and biocides, including chlor- manifestation and comprises 75 to 85% percent of total infections hexidine and glutaraldehyde [30–34]. As a result, mycobacte- [11, 12]. Other disease presentations include skin and soft tissue rial species are frequently found even in hospital environ- infection, tenosynovitis, , , or keratitis ments and have been associated with outbreaks of nosocomial [3•, 13, 14]. Disseminated disease involving the blood, central infections following or cosmetic procedures [35, 36, nervous system, or other dispersed sites can occur, particularly 37•, 38, 39, 40]. Clustered cases have also been reported from for more virulent species, such as M. avium and M. abscessus [5]. non-hospital settings, including parlors (M. chelonae, Immune-compromised hosts, including patients with ad- M. fortuitum); fish markets (M. marinum); and nail salons vanced HIV infection, organ or stem cell transplantation, and (M. chelonae, M. fortuitum, M. abscessus)[41–46]. individuals with Mendelian susceptibility to mycobacterial In contrast to M. tuberculosis, treatment regimens for NTM disease (MSMD), are also at greater risk of disseminated dis- infections have not been extensively studied in clinical trials. ease due to NTM [5, 15]. The latter individuals carry autoso- For most species, there are no standard regimens and our mal or X-linked mutations in genes within the interferon-gam- understanding of the role of susceptibility testing ma/IL-12 immune response pathway [15]. Several naturally is incomplete [47]. Treatment decisions are thus largely based occurring polymorphism in TLR1 and TLR2, invariant recep- on case studies and expert opinion published in guidelines [3•, tors on the surface of immune cells that recognize bacterial 48–52]. Some emerging pathogens, such as M. abscessus,are lipopeptides, have also been associated with altered suscepti- highly drug-resistant and cannot be treated with typical bility to M. tuberculosis or M. leprae in human populations agents, which include , fluoroquinolones, amino- [16–20]. These data suggest that there are likely other com- glycosides, cefoxitin, imipenem, sulfamethoxazole, and tetra- mon variants in immune response genes that modulate human cyclines (other than tigecycline) [53]. Table 1 summarizes the susceptibility to clinically important NTMs via complex in- clinical presentation, notable features, and treatment of select- heritance (non-Mendelian) patterns. ed NTMs that cause SSTIs. Skin and soft tissue infections are the most common pre- sentation for the rapid-growing species , M. abscessus,andM. chelonae [3•]. Certain slow- Recent Epidemiology growing species of mycobacteria, namely , M. ulcerans, M. chimaera,andM. haemophilum, An accurate estimate of the burden of NTM disease in the are also more frequently associated with skin disease [3•, general population is limited by the fact that infections are 21–23]. However, virtually any species can localize to the skin not reportable in most countries, including the USA [3•, 51]. or soft tissues. Clinically, lesions may appear as papules, pus- Available evidence suggests, however, that NTM pulmonary tules, nodules, , panniculitis, , or plaques. disease is increasing in North America, much of Europe, New They are typically erythematous or violaceous, but not warm. Zealand, Australia, and tertiary centers in East Asia [81•, 82]. Lesions may later become ulcerated [7]. Infection may present Data from China, India, and Africa are far from complete [81•, with a “sporotrichoid” pattern of spread, tracking along sub- 83, 84]. In the USA, the annual rate of pulmonary disease is cutaneous lymphatics from the site of inoculation proximally. currently 5–10/100,000 [81•]. Several US studies have shown This pattern is particularly classical for M. marinum cutaneous increases in the frequency of NTM infection [85•, 86, 87•]. disease [21, 24]. Skin and soft tissue infection (SSTI) may In the past, 75 to 80% of reported NTM disease represented follow minor trauma and inadvertent inoculation, such as oc- pulmonary infection [3•]. It is unclear whether this distribution curs when the skin is punctured by wood splinters, fish spines, is stable or changing. One single center series from a university or needles [21]. Infections can also arise through accidental hospital in France, for example, reported that pulmonary infec- contamination of surgical or open wounds [24–26]. Finally, tions made up only 54% of 170 total NTM infections between skin involvement may result from hematogenous 2003 and 2010. Skin infections comprised 23% of the total urIfc i Rep Dis Infect Curr Table 1 Selected nontuberculous Mycobacteria associated with skin and soft tissue infections

Species Clinical presentation Distinguishing features Treatmentb

Rapidly growing • Variable lesion appearance: tender, • Optimal incubation: 28–30 °C [3•, 54] • Depends on species Mycobacteria erythematous nodules, papules, pustules, (2018)20:6 (RGM) plaques, or abscesses • Frequently reported as complication of surgery, cosmetic procedures, tattooing, pedicures • Disseminated disease in immune- compromised hosts can occur

M. abscessus • Skin infections less frequent than pulmonary • Presence of the erm(41) gene in M. abscessus subspp. • Not standardized; at least two agents, often for Subspecies disease, but occur in (non-HIV) immune- abscessus and bolletii confers resistance 4–6 months. Longer duration for disseminated disease abscessus compromised hosts • Macrolide-based therapy if susceptible Subspecies • Also reported in normal hosts in health care • In vitro activity: macrolides (, massiliense settings and after cosmetic procedures or ) (if erm(41)-negative), , Subspecies bolletii environmental exposure [25] cefoxitin, imipenem, tigecycline (a gylcylcycline) (rare) M. chelonae • Skin, bone, soft tissue, corneal or disseminated • Some strains grow between 28 and 33 °C [3•, 54] • Not standardized; often two agents for 4–6months. infection • Lacks erm(41) gene Longer duration for disseminated disease. • Contact lens keratitis • In vitro activity: macrolides, tetracyclines (doxycycline • Normal or in (non-HIV) immune-compromised or minocycline), , fluoroquinolones hosts ( or ), imipenem, tobramycin • Reported after cosmetic, ocular and other surgical • A single agent (fluoroquinolone or clarithromycin) may procedures or be used as for less severe disease, although resistance to clarithromycin has been reported when used as monotherapy [55]

M. fortuitum • Local infection after trauma or inadvertent • Contains erm(39) gene, which encodes inducible • Many options; at least two agents 4–6 months inoculation during surgery, tattooing, or macrolide resistance [56] • Often susceptible to quinolones, tetracyclines, pedicure sulfonamides, cefoxitin, imipenem, amikacin • Even with documented in vitro susceptibility, macrolides should be used with caution due to possible inducible resistance [57–59] • In vitro susceptibility predicts clinical response [3•]

Slowly growing • Variable lesion appearance: tender, erythematous • For most SGM species (exceptions noted below), • Depends on species Mycobacteria nodules, papules, pustules, plaques, or growth occurs at 35–37 °C [3•, 54] abscesses • Only M. chimaera reported in nosocomial outbreaks • Disseminated disease in immune-compromised

hosts can occur 17 of 3 Page

M. chimaera • Global outbreak linked to contaminated • Will grow on standard mycobacterial culture media • Not standardized; If clarithromycin-susceptible, may use heater-cooler units used in cardiac surgerya at 37 °C regimen of clarithromycin, and rifampin • Presentation can be delayed by months to years with or without an , as for MAC [60] • Surgical site infection, sometimes with 6 dissemination and endocarditis [60] 6 Table 1 (continued)

b Species Clinical presentation Distinguishing features Rep Dis Infect Curr Treatment 17 of 4 Page

M. haemophilium • Skin and soft tissue infection, lymphadenitis, • Optimal growth: 28–32 °C • No standardized susceptibility methods occasionally disseminated disease [61–63] • Culture growth requires media supplementation with iron • In vitro activity: amikacin, macrolides, quinolones, rifampin/rifabutin, • HIV or non-HIV immune compromise [62–64] (hemin, hemoglobin, or ferric ammonium citrate) [62] sometimes doxycycline and sulfonamides • Uniformly resistant to ethambutol • Treatment durations usually > 6 months [3•, 62]

M. marinum • Skin and soft tissue infection, tenosynovitis, • Optimal growth: 28–30 °C [3•, 54] • No standardized regimens synovitis; occasionally disseminates to bone • M. marinum infection can produce false-positive • Active agents: clarithromycin, tetracyclines, rifampin, •“Fish tank” or “swimming pool” granuloma with results on interferon-gamma release assays used to ethambutol, sulfonamides, amikacin [65–67] a sporotrichoid pattern of nodules [21, 24] detect infection (QuantiFERON®-TB • Recommended treatment: 1–2 agents for 1–2 months • Outbreaksreportedinfishmarketworkers[44] Gold or T-SPOT®.TB test) beyond clinical resolution [3•]

M. ulcerans • Initially, a nontender nodule, plaque or edema • PCR of IS2404 from direct clinical specimens (punch • WHO: (1) (10 mg/kg once daily) and • Subsequently, the original area ulcerates and for nonulcerated lesions; swabs for ulcerated (15 mg/kg once daily) or (2) rifampicin forms disfiguring scar [68–70] lesions) is diagnostic methodology of choice [70, 72–75] (10 mg/kg once daily) and clarithromycin (7.5 mg/kg • Seen predominantly in children • Diagnosis in resource-limited settings: a loop-mediated twice daily) for 8 weeksc • Some forms may have an association with HIV, isothermal amplification method that does not require a • Australia: rifampicin (10 mg/kg once daily) and but this is controversial (see references cited in cold chain has been reported [76]; a dry reagent-based moxifloxacin (400 mg once daily) for 8 weeksc [79] [71]) qPCR method can be used by unskilled staff [77] • Shorter regimens may be possible: rifampin/rifampicin with • Thin layer chromatography to detect mycolactone may ciprofloxacin or rifampin/rifampicin with clarithromycin for 29 days [80] be a future field diagnostic technique [78] • Laboratory cultivation: 9–12 weeks on Lowenstein-Jensen media incubated at temperatures of 28–32 °C [70] • Cultures used for epidemiological investigation, relapses, detection of drug resistance [70]

IS2404, insertion sequence 2404 MAC Mycobacterium avium complex, PCR polymerase chain reaction, qPCR quantitative polymerase chain reaction, WHO World Health Organization a See text for discussion and additional references b Treatment regimens for lung or disseminated infection are generally longer, rely on an initial phase of intravenous therapy, and may involve three or more agents c http://www.who.int/mediacentre/factsheets/fs199/en. (2018)20:6 Curr Infect Dis Rep (2018) 20:6 Page 5 of 17 6

[88]. In a Taiwanese university hospital, the annual incidence complex (MAC), not found in the North American study, were (per 100,000 patients) of overall NTM disease rose from 8.6 in also found in three SSTIs in these studies [12, 91]. Finally, a 1997 to 16.55 in 2003 [14].Therateofskinandsofttissue study from Singapore reported eight cases of skin and soft infection also rose, from 1.67 to 6.72 per 100,000 patients over tissue infection, five with M. abscessus, two with the same period. Approximately 14% of these patients had M. haemophilium, and one with M. kansasii, out of a total of SSTI or osteomyelitis. M. marinum was isolated in 28%, 67 positive cultures reviewed [94]. Together, these four re- M. abscessus in 26%, and M. kansasii in 12% of those individ- ports represent 50 years of mostly tertiary center inpatient uals [14]. In a report from Taiwan of 1105 NTM patients, a experience and yet describe only 59 cases of pediatric skin sharp rise in the rate of pulmonary disease was observed be- and soft tissue NTM infections, underscoring how little epi- tween 2000 and 2008. Extra pulmonary disease also increased demiological data is available [12, 91, 92, 94]. The incidence over this time period [89•]. Over the 9-year study period, 11.4% of NTM infections in children in the outpatient setting, for of patients had a skin and soft tissue infection. M. abscessus example, is unknown. was the most frequent cause of SSTI (37.7% of cases) [89•]. In the USA, a recent study of NTMs isolated in inpatient Emerging : M. chimaera and outpatient laboratories in three counties in North Carolina found that 0.2 isolates per 100,000 persons (15 in total) were Mycobacterium chimaera was once considered a rare patho- from dermal sites. The most frequent strains from dermal sites gen. However, since 2015, this species has been identified in were M. abscessus (10/15), M. marinum (1/15), nosocomial outbreaks of surgical site infections. The infec- M. immunogenum (1/15), and M. avium (1/15) [86]. A study tions have been epidemiologically linked to a specific brand in the state of Oregon, where reporting of NTM infection of heater-cooler units (HCUs), which are stand-alone devices became mandatory in 2005–2006, found that skin and soft used during requiring cardiopulmonary bypass. First tissues represented the second most common site of NTM reported in Switzerland by Sax et al., M. chimaera-contami- disease, with a reported incidence of 0.9 cases per 100,000 nated HCUs on three different continents have been associated persons [85•]. M. marinum, followed by M. abscessus and with surgical site infections that frequently disseminate M. chelonae, were the most often isolated [85•]. A retrospec- and cause significant morbidity and mortality [95]. HCUs regu- tive series of 40 patients who presented with cutaneous NTM late the temperature of patient blood and cardioplegia solution disease to a tertiary center in Minnesota between 1980 and via a water circuit without direct contact between fluids. The 2009 found that the incidence of skin infection had increased HCUs are not airtight and utilize fans to dissipate heat more from 0.7 per 100,000 person-years in the period 1980–1999 to efficiently. Fans within the LivaNova 3T HCUs aerosolize 2.0 per 100,000 person-years in the period 2000–2009. M. chimaera, which can then deposit on the operative field, Infections due to M. marinum declined from 64% of cases contaminating wounds and prosthetic material. The in the earlier era to 33% in the more recent era, while those due aerosol-dispersion mechanism has been proven using smoke to M. chelonae or M. abscessus rose from 7% to 46% of tests, sedimentation plates, and particle counters. Strains se- SSSTI cases [87•]. quenced from infected patients have also matched those isolated from the HCUs [95, 96•, 97, 98•]. Pediatric Population In most cases, contamination of the HCU seems to have occurred at the manufacturer, given the similarity of isolates There is no definite evidence of an increase in the incidence of found in the manufacturing plant to clinical isolates from NTM infections among children. However, the absence of this Australia, New Zealand, the USA, and UK [99, 100, 101•]. observed trend may reflect inadequate data collection [12, 90]. Nonetheless, the risk of local contamination also exists. Based on inpatient surveys, the incidence of total NTM infec- Studies have shown that a large variety of NTMs colonize the tions in children ranges from 0.6 to 1.6 cases per 100,000, of health care and home environment [101•, 102]. Van Ingen et al. which 2–18% involve skin and soft tissues [90–94]. Skin in- identified at least one clinical strain of M. chimaera whose gene fections comprise a higher proportion of cases in countries sequence was more homologous to local isolates than the out- such as Australia, where M. ulcerans is endemic. break strains [101•]. A separate single-center outbreak of In North America, three species of rapid growers, M. abscessus was felt to have arisen from two local strains M. chelonae, M. abscessus,andM. fortuitum accounted for spread via HCUs [103]. Extracorporeal membrane oxygenation 74% (17/23) of pediatric infections, according to one study of machines may also become colonized, although none so far hospitalized cancer patients. Most of the children in this study have been implicated in this outbreak, perhaps due to their had catheter-related infections [92]. In contrast, two other airtight, closed-system construction [104]. studies of 28 inpatient children in Australia reported that Despite the extensive use of the LivaNova 3T HCUs and M. marinum and M. ulcerans were the most common species, prevalence of cardiac surgery worldwide, rates of representing 16 out of 28 infections. Members of the M. avium M. chimaera infection have been low. In the UK, 0.17 to 3 6 Page 6 of 17 Curr Infect Dis Rep (2018) 20:6 cases per 1000 procedures (0.39 per 10,000 patient years) following the manufacturer’s evolving guidelines for cleaning were reported, similar to other cohorts [101•]. The majority and decontamination. Some published protocols have demon- of infections have involved chest surgery (requiring an HCU) strated clearance of cultures from HCUs for up to 6 months and either prosthetic material or heart transplant. For example, [111]. Other authors using a similar decontamination regimen of 24 cases identified in a US investigation, 46% had heart documented the re-emergence of Mchimaeragrowth; although surgery for a prosthetic valve or valve ring, 29% received a in this study, tubing was not replaced [38]. The manufacturer’s vascular graft, 21% had a left ventricular assist device im- most recent decontamination protocols include initial cleaning planted, 13% underwent heart transplant, and 4% had coro- with peracetic acid prior to clinical use, daily cleaning with nary artery bypass grafts [36]. Prosthetic material is a signif- (2 μm) filtered containing 3% hydrogen peroxide, icant risk factor for deep or disseminated infection and poor and regular (i.e., weekly) full decontamination with peracetic outcomes. In one series, endocarditis was seen in 28%, endo- acid. While there has been continued evolution of decontami- carditis with aortic root in 17%, disseminated infec- nation procedures, it remains unclear whether colonized HCUs tion in 17%, sternal osteomyelitis in 11%, anterior mediastinal can be decontaminated and remain sterile over the long-term. abscess in 6%, and spinal osteomyelitis or diskitis in 6% Despite these issues, the LivaNova 3T heater-cooler units are [101•]. Ophthalmologic involvement occurs and may reflect still in use in many hospitals. disease burden [106]. The time to presentation has varied widely, ranging from 3 months to 5.1 years. It has been rec- General Features of Mycobacterial Virulence ommended that a diagnosis of M. chimaera surgical site in- and the Immune Response fection be considered as late as 6 years following surgery [95, 101•]. Identification of cases requires a high index of suspi- As a class, NTM possess two features important for virulence: cion. are often vague, but may include the ability to elaborate an antibiotic-impervious biofilm and (90%), (75%), sweats (60%), dyspnea (60%), thepresenceofacellwallrich in glycolipids, such as weight loss (60%), and cough (50%) [36]. Laboratory findings lipoarabinomannan (LAM). Microbial biofilm may modify are nonspecific. Anemia, lymphopenia, and thrombocytope- immune responses, by sequestering microbes from immune nia are common. Most cases are diagnosed by positive blood cells and effector molecules, thereby preventing immune cell cultures. Thus, blood cultures for acid-fast organisms are rec- recognition [112]. Separately, biofilm allows the physical per- ommended in suspected cases. Bone marrow, tissue biopsy, or sistence of organisms on hardware and other materials, such as operative samples may also be cultured. Histopathological non-biologic vascular grafts. Environmental mycobacteria re- samples are frequently negative on acid-fast ; granu- siding in biofilms around faucets, drains, and pipes, or on lomatous changes in such samples have led to misdiagnosis medical equipment are resistant to most conventional disin- with . fectants [33, 112]. As discussed above, environmental There are no formal treatment guidelines for M. chimaera mycobacteria have been associated with disease outbreaks infections. Until 2004, M. chimaera was classified within the epidemiologically linked to water systems and medical sup- M. avium complex (MAC) [107]. Therefore, treatment regi- plies and equipment [37•, 102, 113]. The second virulence mens similar to those for MAC have been used. Macrolide, factor shared among mycobacteria is a thick and waxy cell , and ethambutol form the backbone of therapy, of- wall, frequently decorated with LAM and other closely related ten with the addition of moxifloxacin or amikacin. Linezolid glycolipids. LAM and similar molecular motifs have been or clofazimine has been added in rare cases [35, 95, 106]. associated with altered differentiation and cytokine expression Similar to MAC, there is no evidence that susceptibility test- of the cell (reviewed in [114, 115]). Bacterially mediated ing of drugs other than macrolides correlates with treatment immune modulation is both species- and strain-dependent. For response [108–110]. example, in M. marinum, LAM elongation and branching en- Unfortunately, treatment responses are often poor. ables virulence in and disruption of this process Mortality has reached 50–60% in the largest series, with more leads to attenuated disease in this host [116]. Whether similar than half of patients experiencing breakthrough on therapy. mechanisms extend to all pathogenic NTM is not yet known. The disease is very difficult to cure, and most patients who survive report feeling unwell on therapy. Extensive operative interventions, including valve replacement, endovascular graft New Insights into NTM Pathogenesis excision, and serial are often required, even for patients on appropriate therapy [95, 101•]. Rapidly Growing NTM To date, the best infection control measures involve removal of colonized HCUs from the operating theater or engineering an Rapidly growing NTM are species that grow on solid media airtight structure that surrounds the HCU and connects directly within 7 days. Overall, this group of mycobacteria is consid- to the ventilation system [38, 97]. The CDC recommends ered less pathogenic than the slow-growing species. Some Curr Infect Dis Rep (2018) 20:6 Page 7 of 17 6 recent data suggest, for example, that M. abscessus and phenotype. Epidemiological evidence for this mechanism, M. fortuitum are less efficient at infecting cells than however, is weak. Survival within free-living amoebae in M. tuberculosis or M. celatum. This reduced infection effi- drinking water has not been demonstrated for M. abscessus, ciency may allow the host to mount a more successful im- although it has for other mycobacteria [128]. This proposed mune response [117]. virulence mechanism also does not have relevance for skin and soft tissue infections due to M. abscessus. M. abscessus M. chelonae Mycobacterium abscessus is the most pathogenic and clinical- ly challenging of the rapid-growing mycobacteria. is more likely than M. fortuitum to M. abscessus causes entrenched pulmonary infections in pa- cause disseminated rather than localized skin infection. This tients with or lung transplantation, and is often species is also less closely related to M. fortuitum than it is to associated with broad resistance. There are three M. abscessus [129]. However, M. chelonae lacks the erm(41) subspecies in the M. abscessus group: M. abscessus subsp. gene that confers macrolide resistance in M. abscessus [120]. abscessus,M. abscessus subsp. massiliense,and It also differs from M. abscessus in having a greater predilec- M. abscessus subsp. bolletii [118, 119]. M. abscessus subsp. tion for skin than lung infections. In the environment, massiliense lacks the intrinsic (nota- M. chelonae has been found within free-living amoebae bly, the erm(41) gene conferring macrolide resistance) of the [128]. In zebrafish, experimental infectivity is enhanced when other members [120]. the are associated with the amoeba Paramecium During pulmonary infection, M. abscessus changes from a caudatum [130]. There is a single case report of M. chelonae smooth to a rough colony phenotype, through loss of surface living within a protozoan colonizing the nasal mucosa of an glycopeptolipids (GPLs). The smooth morphotype of this species HIV patient [131]. Other evidence that protozoa enhance the predominates in skin and soft tissue infections [121]. The smooth virulence of M. chelonae during natural infection, however, is variantislessvirulentandsurvives well in biofilms, while the so far lacking. rough variant is more virulent and pro-inflammatory and grows aggressively in the host by forming cords that impede phagocy- M. fortuitum tosis [122–124]. The surface GPLs of the smooth colony variant appear to shield other cell wall molecules from interactions with Mycobacterium fortuitum is a frequent cause of localized skin Toll-like receptor 2 [125]. Only the rough morphotype induces and soft tissue infections [132]. Relatively little is known and autophagy, events that have the potential to en- about the immune pathogenesis of infections due to this spe- hance virulence, in [124]. Absence of phenolic cies. M. fortuitum infection of murine macrophages induces glycolipid (PGL) on the surface of the rough variant appears both pro-and anti-inflammatory responses that are mediated central to this -inducing phenotype, since apopto- by Toll-like receptor 2 [133]. A 2016 study in zebrafish mac- sis is not induced when cultures of rough variant rophages demonstrated that M. fortuitum induces mycobacteria are supplemented with PGLs. Interestingly, only caspase-mediated apoptosis in macrophages via a calcium thelessvirulentsmoothformofM. abscessus induces breaks influx-dependent mechanism, thus enhancing infection [134]. in the phagosome membrane [124], a process associated with enhanced virulence in other mycobacteria. One proposed explanation for the particular virulence of Slowly Growing Nontuberculous M. abscessus in lung infections is the recently discovered Mycobacteria putative phospholipase C (PLC), a gene which is absent in other rapidly growing species [126]. This enzyme is cytotoxic M. marinum to mouse macrophages and shows similar activity to PLC-N of Pseudomonas aeruginosa, an organism that causes chronic In nature, M. marinum is found in freshwater, brackish water, infections in cystic fibrosis patients [127]. The PLC gene is and salt water. Infection with the organism produces a fatal expressed in smooth colony variants of M. abscessus when the tuberculosis-like, granulomatous disease in fish. In humans, bacteria are residing within amoebae. When pre-cultured in M. marinum causes “fish-tank” or “swimming pool” granuloma, amoebae, M. abscessus is more infective in mouse lungs producing single or multiple papulo-nodular skin lesions, often [127]. This finding suggests that environmental amoebae on the hands, which may spread proximately in a sporotrichoid may be important for pulmonary infection of mammalian pattern [21, 24]. M. marinum has been exploited experimentally hosts. In this model of virulence, smooth variant bacteria re- to model tuberculosis in zebrafish embryos, which are conve- siding within amoebae are primed to establish infection in the niently transparent. The zebrafish model has yielded major in- lungs. After infection, the bacteria transition to a rough colony sights into the earliest stages of mycobacterial pathogenesis 6 Page 8 of 17 Curr Infect Dis Rep (2018) 20:6

[135–137]. Two recent studies demonstrated that M. marinum been proposed, including superinfection of minor skin abra- uses specific membrane lipids (phenolic glycolipids) to recruit sions, inoculation by mosquitoes or specific species of wet- more permissive macrophage types to the site of infection land insects, and via amoebae [152], but there is no current [138•]. The bacilli then transfer from their original macrophage consensus. A common theme appears to be microbial access host cells, which restrict intracellular growth, to the more per- to the microvasculature, which is achieved through punctures missive new arrivals, thus surviving and disseminating [139]. or vector inoculation, rather than superficial abrasion. Like M. tuberculosis,theM. marinum contains an ESX-1 locus that encodes major virulence factors ([140]; reviewed in [141•]). The ESX-1 locus had previously been Newer Diagnostic Tools shown to contribute to lysis of the phagosome membrane (and the consequent escape of bacteria to the ) and gran- Traditionally, identification of NTMs in the laboratory relied uloma formation [137, 142]. More recent studies have elucidat- heavily upon phenotypic characteristics such as biochemical ed the mechanism of phagosome membrane lysis [143]and tests, growth rate, and colony pigmentation and morphology. identified ESX-1-encoded proteins that modulate macrophage More advanced recent diagnostic tools have included high- function to promote intracellular survival [144]. A paralogous performance liquid chromatography and molecular diagnostic locus, ESX-5, has been shown to influence autophagy (using tests [153, 154•, 155]. DNA hybridization, DNA sequencing simultaneous activation and repression of its mechanisms) to of 16S ribosomal RNA and other gene regions have permitted enhance intracellular bacterial persistence [145]. the identification of many new species. Molecular methods are now routinely used in reference labs and academic centers M. ulcerans Recently, Pranada and colleagues were successful in discrim- inating Mycobacterium intracellulare from M. chimaera,two Mycobacterium ulcerans causes an ulcerating skin infection, species whose 16S RNA gene sequences differ by a single known as Buruli or Bairnsdale , which may bore through base pair, using MALDI-TOF [156]. Table 2 summarizes overlying soft tissue to bone. Worldwide, it is the third most these important new diagnostic tools. commonly reported mycobacterial infection, after tuberculo- sis and leprosy [68, 146]. , a neglected , is encountered in Africa, Australia, Southeast Asia, Advances in Antibiotic Treatment and Central and South America. The majority of cases come of Nontuberculous Mycobacteria from West Africa and the Democratic Republic of Congo (90% of global cases reported in 2016) and Australia (10% Novel Anti-tuberculosis Agents of global cases) [147]. Mycobacterium ulcerans is most closely related to Most of the recent advances in the treatment of NTMs have M. marinum, and similarly prefers lower temperatures for come from repurposing used against other bacteria. growth. It is unique among the NTM species in being a pre- However, two new agents recently developed for tuberculosis, dominantly extracellular pathogen [69]. In 1999, the factor delamanid and bedaquiline, have some activity against responsible for the unique pathogenesis of M. ulcerans was nontuberculous mycobacteria. A recent study reported that found to be mycolactone, a highly diffusible, cytopathic lipid the mean inhibitory concentrations (MICs) of delamanid, a toxin [148]. In the last 5 years, two direct mechanisms of dihydro-nitroimidazooxazole derivative, were quite low for action for this toxin have been elucidated (reviewed in 20 strains of M. avium and M. intracellulare(MAC). [149]). First, mycolactone directly binds the Wiscott-Aldrich Nineteen of the 20 strains had MICs ≤ 0.2 ng/ml. Included syndrome protein (WASP), leading to uncontrolled actin as- among these strains were some with resistance to sembly in the cytoplasm and defective cell adhesion and mi- clarithromycin, the mainstay of MAC therapy [169]. The gration [150••]. Second, mycolactone directly inhibits the anti-tuberculosis drug bedaquiline is an ATP synthase inhibi- Sec61 translocon, which, under normal circumstances, facili- tor with bactericidal activity against M. tuberculosis [170]and tates transit of secretory proteins to the is FDA-approved for multi-drug-resistant and extremely drug [151••]. Mycolactone blockade thus broadly inhibits translo- resistant tuberculosis. Bedaquiline also has in vitro activity cation (and therefore secretion) of immune response proteins, against M. avium complex, M. abscessus, M. ulcerans, and including Cox-2, TNF and IL-6 [149, 151••]. These new find- M. smegmatis [171–176]. ings help to explain the pleiotropic effects of a toxin that Unlike delamanid, bedaquiline has been studied in several makes M. ulcerans so clinically distinct from other NTMs different animal models. In a mouse footpad model of pathogenic to humans. However, another fundamental aspect M. ulcerans infection, bedaquiline monotherapy has shown of M. ulcerans pathogenesis—how the infection is some bactericidal activity [175, 177]. In this model, a regimen transmitted—remains unsolved. Various mechanisms have containing rifamycin and bedaquiline was comparable to Curr Infect Dis Rep (2018) 20:6 Page 9 of 17 6 rifamycin and streptomycin or rifamycin and amikacin, as demonstrated that the majority of linkages in the cell wall measured by lesion size and bacteria recovered from treated of M. abscessus are synthesized by lesions [175]. In one study utilizing nude (-deficient) L,D-transpeptidases, rather than the traditional mice with disseminated M. abscessus, bedaquiline did not have D,D-transpeptidases [184••]. The authors cloned, expressed, significant activity, as assessed by mortality, lesion burden, and and purified two different L,D-transpeptidases and then mea- bacteria recovered from the spleen and lung [178]. In a separate sured the ability of various ß-lactam antibiotics to inhibit these study, in gamma interferon knockout (GKO) mice and B- and enzymes. Cephalosporins and carbapenems, but not , T-cell-deficient severe combined immune deficiency (SCID; were able to cause inhibition. While each of the two L,D- spontaneous mutation Prkdcscid) mice, bedaquiline alone and transpeptidases had a different inhibition profile, cefdinir- in combination with clofazimine reduced bacterial burden in doripenem was the most effective combination for both [184••]. the lung, spleen, and liver [179]. Experimental infections of Resistance to ß-lactams in M. abscessus is mediated in part mice with M. abscessus have generally utilized immune- through the chromosomally encoded Ambler class A ß- deficient mouse strains, without a clear consensus as to which lactamase and BlaMab, which hydrolyzes both ß-lactams and strain makes the ideal model. Thus, both methods and results the ß-lactamase inhibitors clavulanate, sulbactam, and tazo- have been conflicting. In zebrafish larvae, bedaquiline treat- bactam [185]. Deletion of the BlaMab gene in human macro- ment was associated with improved survival, fewer abscesses phages and zebrafish restores in vitro susceptibility to most and the inhibition of in vivo cord formation [174]. penicillins, cephalosporins, and carbapenems (with the excep- The human clinical experience with bedaquiline for NTM tion of aztreonam and ceftazidime) and rescues susceptibility disease so far involves two uncontrolled case studies of pa- to amoxicillin [186]. The novel ß-lactamase inhibitor, tients with lung infection [105, 180]. In the first, 9 of 10 avibactam, is not hydrolyzed by BlaMab and can restore sus- patients with refractory pulmonary mycobacterial disease (6 ceptibility to ß-lactams in BlaMab-positive M. abscessus with MAC, 4 with M. abscessus infection), experienced early strains, similar to strains with BlaMab gene deletion [186]. symptomatic improvement. Four out of 10 also had radio- The addition of avibactam to the carbapenems imipenem, graphic improvement at 4 months. None had sustained micro- faropenem, biapenem, doripenem, meropenem, panipenem, biologic clearance [180]. In the second, 13 patients received ertapenem, and tebipenem resulted in heightened activity for bedaquiline as part of multi-drug salvage regimen for pulmo- all penems against 28 clinical isolates of M. abscessus,some nary M. intracellulare, with 8 initially responding to treat- of which were highly resistant. This effect is likely due to both ment. However, all eight subsequently had microbiologic re- carbapenem inhibition of L,D-transpeptidase and efficacy lapse and developed resistant (mmpT5) variants with two- to against BlaMab [187]. Studies in human macrophages and eightfold higher bedaquiline MICs [105]. Of note, six of the zebrafish further support the efficacy of avibactam when used eight patients who relapsed were also receiving rifampin, with a carbapenem (imipenem) [188••]. which is known to decrease exposure to bedaquiline due to Although more data is needed, the combination of cytochrome P 3A4 induction [181]. Given the difficulty of avibactam and other carbapenems may allow clinicians addi- treating multi-drug-resistant strains of M. abscessus, tional options for the treatment of drug-resistant and refractory bedaquiline could prove useful, although more human clinical NTM infections, particularly M. abscessus. Currently, in the data are needed. USA, avibactam is only available in co-formulation with cef- tazidime, which has no activity against M. abscessus. Oxazolidinones Tebipenem also exists in an oral formulation, but is not yet available in the USA. Linezolid and tedizolid both have activity against NTMs, but tedizolid is emerging as a more tolerable alternative to linez- olid, with equivalent or lower MICs in vitro for both slow- and Conclusion rapid-growing mycobacteria [182]. There is at least one case report of the successful use of tedizolid, as part of a multi-drug Although systemic surveillance data are lacking, the incidence regimen for pulmonary for NTM disease, following linezolid of NTM infections, including those involving skin and soft toxicity [183]. Although more experience is needed to firmly tissues, is rising in many regions.M.chimaerais an emerging establish efficacy and tolerability, tedizolid may become a nosocomial pathogen associated with skin and soft tissue infec- very useful addition to the NTM armamentarium. tions following cardiac surgery. In some series, rapid- growing mycobacteria, including the multi-drug-resistant spe- ß-Lactams cies M. abscessus, have surpassed M. marinum as the most frequently isolated pathogen in patients with skin and soft tissue The role of ß-lactams in the treatment of NTMs is evolving, disease. M. ulcerans continues to cause major morbidity in especially in the case of M. abscessus. Kumar et al. recently Africa. The most reliable diagnostic approaches for NTM 6 Page 10 of 17 Curr Infect Dis Rep (2018) 20:6

Table 2 Molecular methods for the identification of nontuberculous mycobacteria

Assay name or type RNA/DNA target and methods Species identified Disadvantages/advantages

GenoType Mycobacterium CM Target: 230 bp fragment of 23S > 20 species of NTM and • Requires growth on culture (Hain Lifescience, Nehren, rRNA gene M. tuberculosis • 37 species identifiable. Germany) Method: PCR of a 23S rRNA gene • Tsukamurella may be region, followed by reverse misidentified as mycobacterial hybridization and line probe spp. [158] technology [157] • M. chimaera misidentified as GenoType Mycobacterium AS Same as Mycobacterium CM [157] 19 less common NTM species M. intracellulare [158] (Hain Lifescience, Nehren, • M. terrae and M. Germany) nonchromogenicum not covered • Commercially available • Highly sensitive and specific

GenoType NTM-DR Species identification: same as • Detects M. avium, • Requires growth on culture (Hain Lifescience, Nehren, Mycobacterium CM M. intracellulare, M. chimaera, • Distinguishes M. chimaera from Germany) drug resistance targets: erm41, rrl, M. chelonae and species and M. intracellulare and rrs M. abscessus subspecies • Detects drug resistance mutations: massiliense, bolletti, abscessus —macrolides (erm(41), rrl genes) —amikacin/tobramycin (rrs gene) • Sensitivity: 79% for clarithromycin resistance, 71% for aminoglycoside resistance [159]

Genotype CMdirect reverse hybridization line probe > 20 species of NTM and M. • Maybeuseddirectlyon (Hain Lifescience, Nehren, assay (DNA) tuberculosis decontaminated sputum Germany) • Commercially available

Gen-Probe AccuProbe® Target: rRNA Individual assays for each species: • Requires growth on culture (Hologic, Marlborough, MA, USA) Single-stranded DNA probe M. avium, M. intracellare, M. • Lower sensitivity for detection of hybridizes to a ribosomal RNA gordonae,andM. kansasi M. fortuitum strains than target GenoType [158] • Commercially available • Not multiplexed INNO LiPA® Mycobacteria Target: 16S–23S rRNA intergenic 16speciesofNTM • Requires growth on culture v2 (Fujirebio, Tokyo, Japan) spacer (ITS) • 87–100% sensitivity; 100% DNA hybridization line probe assay specificity [160, 161] • Commercially available

Restriction enzyme pattern analysis Target: segment of the 65-kDa heat Often used to compare a handful of • Requires growth on culture (hsp65 gene) shock protein gene (hsp 65). species • hsp gene encodes sequences that DNA segment is amplified by PCR are unique and common –some and the product is digested by species may not have unique restriction enzymes, with restriction patterns [161] fragments separated by agarose • Time-intensive, research or gel electrophoresis. The resulting reference lab method restriction patterns are analyzed by comparison to those of known reference strains [161]

16S rRNA, rpoB, or hsp65 gene Molecular target: hsp65, 16S rRNA • In theory, many species are • Maybedoneondirectpatient sequencing [158, 162, 163] genes individually identifiable sample/specimen, if sterile commercial platform: MicroSeq DNA segment is amplified and • In practice, closely related species • Available at reference centers 500 16S rDNA microbial sequenced using broad range may not be resolved (clearly identification system (Applied primersapplicabletoall distinguished) due to negligible Biosystems, Foster City, mycobacterial species. Nucleotide sequence differences California) sequences are compared to public database (GenBank or others*) [158, 162, 163] Curr Infect Dis Rep (2018) 20:6 Page 11 of 17 6

Table 2 (continued)

Assay name or type RNA/DNA target and methods Species identified Disadvantages/advantages

Multiplex SNaPshot method Target: polymorphic sites in 16S • Simultaneous analysis of 8 single • Requires growth on culture rRNA and hsp 65 genes nucleotide polymorphisms • Research method but utilizes -specific primers amplify (SNPs) in 16S rRNA and hsp 65 commercially available kit (ABI 354-bp fragment of hsp65 and a genes allowing discrimination of 5 Prism SNaPshot Multiplex Kit), 436-bp fragment of 16S rRNA. NTM species (M. avium, capillary electrophoresis on an Forward or reverse extension M. intracellulare, M. chelonae, ABI Prism 3130 × 1 genetic primers are designed to anneal M. kansasii, M. gordonae)and Analyzer (Applied Biosystems), adjacent to polymorphic sites in M. tuberculosis and genetic analysis using 16S rRNA gene (positions 125, commercial software 141, 231, 264, 471) or hsp65 gene (GeneMapper, version 4.0, (positions 163, 235, 265) present software, Applied Biosystems) in the PCR products. Single base [164] extension reactions with fluorescently labeled ddNTPs are performed for multiple SNP sites simultaneously. Resultant fragments are resolved by electrophoresis and SNP alleles are identified by fluorescence emission (red, green, blue, black) after laser excitation [164]

Denaturing high-performance liquid Target: 16S–23S rRNA ITS gene Many NTM species identifiable • Amplification of the 16S–23S chromatography (DHPLC) region or 16S rRNA region rRNA ITS region allows [165, 166] differentiation of M. chimaera Amplification of the target DNA, from M. intracellulare [165] followed by denaturing • Applicable to M. chimaera high-performance liquid outbreak settings chromatography • Requires growth on culture • Much more sensitive than culture [166] • Rapid, high through-put • Notwidelyavailable

Matrix-assisted laser desorption Mycobacterial cultures are treated • In theory, unlimited • Requires growth on culture ionization time-of-flight mass with formic acid and acetonitrile • In practice, limited by size of • Rapid method spectrometry (MALDI-TOF MS) to break down cell walls and spectral library • M. abscessus subspecies may not extract protein. The resultant be well separated samples (analytes) are applied to a • M. chimaera and M. steel target plate and coated with a intracellulare may be conflated. matrix solution (e.g., However, one report of a model α-cyano-4-hydroxycinnamic that distinguishes Mchimaera acid). The plate is irradiated with a and M. intracellulare [156] pulsed laser and energy is transferred from the matrix to the analyte, which undergoes desorption (removal into the gas phase) and ionization. The ionized particles are subjected to an electric potential that propels them through a flight tube into the mass spectrometer, where they are separated by mass and charge. Different bacterial species display distinct spectral fingerprints (stereotypical pattern of peaks), and are identified by reference to a large database [156, 167, 168].

*Other databases include the DNA Data Bank of Japan (DDBJ), the European Molecular Biology Laboratory (EMBL), and the Ribosomal Differentiation of Medical Microsystems database (RIDOM) 6 Page 12 of 17 Curr Infect Dis Rep (2018) 20:6 species are molecular, and involve DNA probes, gene sequenc- 10. Reed C, von Reyn CF, Chamblee S, Ellerbrock TV, Johnson JW, ing (of 16S ribosomal RNA, hsp65, rpoB, or the 16-23S Marsh BJ, et al. Environmental risk factors for infection with Mycobacterium avium complex. Am J Epidemiol. 2006;164(1): intergenic spacer region) and MALDI-TOF. Research into the 32–40. basic mechanisms of virulence of the more pathogenic NTMs, 11. Haverkamp MH, Arend SM, Lindeboom JA, Hartwig NG, van particularly M. marinum and M. abscessus, has the potential to Dissel JT. Nontuberculous mycobacterial infection in children: a offer biologically based therapeutics. There is an unmet demand 2-year prospective surveillance study in the Netherlands. Clin Infect Dis. 2004;39(4):450–6. for such novel agents, especially for drug-resistant species, such 12. Tebruegge M, Pantazidou A, MacGregor D, Gonis G, Leslie D, as M. abscessus, which causes significant morbidity and mor- Sedda L, et al. Nontuberculous mycobacterial disease in chil- tality in cystic fibrosis and lung transplant patients. Study of the dren—epidemiology, diagnosis & management at a tertiary center. activity of available and new antibiotics, both singly and in PLoS One. 2016;11(1):e0147513. combination, against NTM species lags behind the develop- 13. Piersimoni C, Scarparo C. Extrapulmonary infections associated with nontuberculous mycobacteria in immunocompetent persons. ment of new drugs for tuberculosis and is urgently needed. Emerg Infect Dis. 2009;15(9):1351–8. quiz 1544 14. Ding LW, et al. Disease caused by non-tuberculous mycobacteria Compliance with Ethical Standards in a university hospital in Taiwan, 1997–2003. Epidemiol Infect. 2006;134(5):1060–7. 15. Bustamante J, Boisson-Dupuis S, Abel L, Casanova JL. Conflict of Interest Each author reports no conflicts of interest. Mendelian susceptibility to mycobacterial disease: genetic, immu- nological, and clinical features of inborn errors of IFN-gamma Human and Animal Rights and Informed Consent This article does not immunity. Semin Immunol. 2014;26(6):454–70. contain any studies with human or animal subjects performed by any of 16. Hawn TR, et al. A common human TLR1 polymorphism regulates the authors. the innate immune response to lipopeptides. Eur J Immunol. 2007;37(8):2280–9. 17. Johnson CM, Lyle EA, Omueti KO, Stepensky VA, Yegin O, References Alpsoy E, et al. Cutting edge: a common polymorphism impairs cell surface trafficking and functional responses of TLR1 but pro- tects against leprosy. J Immunol. 2007;178(12):7520–4. Papers of particular interest, published recently, have been 18. Misch EA, Macdonald M, Ranjit C, Sapkota BR, Wells RD, highlighted as: Siddiqui MR, et al. Human TLR1 deficiency is associated with • Of importance impaired mycobacterial signaling and protection from leprosy re- •• versal reaction. PLoS Negl Trop Dis. 2008;2(5):e231. Of major importance 19. Ma M-j, Xie LP, Wu SC, Tang F, Li H, Zhang ZS, et al. Toll-like receptors, tumor factor-α, and interleukin-10 gene poly- 1. Rastogi N, Legrand E, Sola C. The mycobacteria: an introduction morphisms in risk of pulmonary tuberculosis and disease severity. to nomenclature and pathogenesis. Rev Sci Tech. 2001;20(1):21– Hum Immunol. 2010;71(10):1005–10. 54. 20. Thuong NT, et al. A polymorphism in human TLR2 is associated 2. Daley CL, Griffith DE. Pulmonary non-tuberculous mycobacterial with increased susceptibility to tuberculous . Genes infections. Int J Tuberc Lung Dis. 2010;14(6):665–71. Immun. 2007;8(5):422–8. 3.• Griffith DE, et al. An official ATS/IDSA statement: diagnosis, 21. Petrini B. Mycobacterium marinum: ubiquitous agent of water- treatment, and prevention of nontuberculous mycobacterial dis- borne granulomatous skin infections. Eur J Clin Microbiol Infect eases. Am J Respir Crit Care Med. 2007;175(4):367–416. Most Dis. 2006;25(10):609–13. recent US guidelines for treatment of nontuberculous 22. Jernigan JA, Farr BM. Incubation period and sources of exposure Mycobacteria. for cutaneous Mycobacterium marinum infection: case report and 4. van Ingen J, Blaak H, de Beer J, de Roda Husman AM, van review of the literature. Clin Infect Dis. 2000;31(2):439–43. Soolingen D. Rapidly growing nontuberculous mycobacteria cul- 23. Saubolle MA, Kiehn TE, White MH, Rudinsky MF, Armstrong D. tured from home tap and shower water. Appl Environ Microbiol. Mycobacterium haemophilum: and expanding clin- 2010;76(17):6017–9. ical and geographic spectra of disease in humans. Clin Microbiol 5. Simner, P.J., G.L. Woods, and N.L. Wengenack, Mycobacteria. Rev. 1996;9(4):435–47. Microbiol Spectr, 2016. 4(4). https://doi.org/10.1128/ 24. Fleming P, Keystone JS. Mycobacterium marinum infection with microbiolspec.DMIH2-0016-2015 sporotrichoid spread from fish tank exposure. CMAJ : Can Med 6. Falkinham JO III. Surrounded by mycobacteria: nontuberculous Assoc J. 2017;189(2):E76. mycobacteria in the human environment. J Appl Microbiol. 25. Cheng A, Liu YC, Chen ML, Hung CC, Tsai YT, Sheng WH, et al. 2009;107(2):356–67. Extrapulmonary infections caused by a dominant strain of 7. Gonzalez-Santiago TM, Drage LA. Nontuberculous Mycobacterium massiliense (Mycobacterium abscessus subspe- mycobacteria: skin and soft tissue infections. Dermatol Clin. cies bolletii). Clin Microbiol Infect. 2013;19(10):E473–82. 2015;33(3):563–77. 26. Leao SC, Tortoli E, Viana-Niero C, Ueki SYM, Lima KVB, Lopes 8. Fairchok MP, Rouse JH, Morris SL. Age-dependent humoral re- ML, et al. Characterization of mycobacteria from a major sponses of children to mycobacterial . Clin Diagn Lab Brazilian outbreak suggests that revision of the taxonomic status Immunol. 1995;2(4):443–7. of members of the mycobacterium chelonae-M. abscessus group is 9. Larson EM, O’Donnell M, Chamblee S, Horsburgh CR Jr, Marsh needed. J Clin Microbiol. 2009;47(9):2691–8. BJ, Moreland JD, et al. Dual skin tests with Mycobacterium avium 27. Ichihara A, Jinnin M, Fukushima S, Inoue Y, Ihn H. Case of sensitin and PPD to detect misdiagnosis of latent tuberculosis disseminated cutaneous mycobacterium chelonae infection mim- infection. Int J Tuberc Lung Dis. 2011;15(11):1504–9. i icking cutaneous vasculitis. J Dermatol. 2014;41(5):414–7. Curr Infect Dis Rep (2018) 20:6 Page 13 of 17 6

28. Bryant JM, et al. Whole-genome sequencing to identify transmis- 46. Vugia DJ, Jang Y, Zizek C, Ely J, Winthrop KL, Desmond E. sion of Mycobacterium abscessus between patients with cystic Mycobacteria in nail salon whirlpool footbaths, California. fibrosis: a retrospective cohort study. Lancet. 2013;381(9877): Emerg Infect Dis. 2005;11(4):616–8. 1551–60. 47. Li B, Yang S, Chu H, Zhang Z, Liu W, Luo L, et al. Relationship 29. Aitken ML, Limaye A, Pottinger P, Whimbey E, Goss CH, Tonelli between antibiotic susceptibility and genotype in Mycobacterium MR, et al. Respiratory outbreak of Mycobacterium abscessus sub- abscessus clinical isolates. Front Microbiol. 2017;8:1739. species massiliense in a lung transplant and cystic fibrosis center. 48. Haworth CS, Floto RA. Introducing the new BTS guideline: man- Am J Respir Crit Care Med. 2012;185(2):231–2. agement of non-tuberculous mycobacterial pulmonary disease 30. Russell AD, Hammond SA, Morgan JR. Bacterial resistance to (NTM-PD). Thorax. 2017;72(11):969–70. antiseptics and . J Hosp Infect. 1986;7(3):213–25. 49. Schoenfeld N, Haas W, Richter E, Bauer T, Boes L, Castell S, et al. 31. Russell AD. Activity of biocides against mycobacteria. Soc Appl Recommendations of the German central committee against tuber- Bacteriol Symp Ser. 1996;25:87S–101S. culosis (DZK) and the German respiratory society (DGP) for the 32. Selvaraju SB, Khan IU, Yadav JS. Biocidal activity of formalde- diagnosis and treatment of non-tuberculous Mycobacterioses. hyde and nonformaldehyde biocides toward Mycobacterium Pneumologie. 2016;70(4):250–76. immunogenum and Pseudomonas fluorescens in pure and mixed 50. Floto RA, Olivier KN, Saiman L, Daley CL, Herrmann JL, Nick suspensions in synthetic metalworking fluid and saline. Appl JA, et al. US Cystic Fibrosis Foundation and European cystic – Environ Microbiol. 2005;71(1):542 6. fibrosis society consensus recommendations for the management 33. Carson LA, Petersen NJ, Favero MS, Aguero SM. Growth char- of non-tuberculous mycobacteria in individuals with cystic fibro- acteristics of atypical mycobacteria in water and their comparative sis: executive summary. Thorax. 2016;71(1):88–90. resistance to disinfectants. Appl Environ Microbiol. 1978;36(6): 51. van der Werf MJ, Ködmön C, Katalinić-Janković V, Ku mmi k T, – 839 46. Soini H, Richter E, et al. Inventory study of non-tuberculous 34. Smithwick RW, Stratigos CB, David HL. Use of cetylpyridinium mycobacteria in the European Union. BMC Infect Dis. 2014;14: chloride and sodium chloride for the decontamination of sputum 62. specimens that are transported to the laboratory for the of 52. van Ingen J, Ferro BE, Hoefsloot W, Boeree MJ, van Soolingen D. – Mycobacterium tuberculosis. J Clin Microbiol. 1975;1(5):411 3. Drug treatment of pulmonary nontuberculous mycobacterial dis- 35. Balsam LB, Louie E, Hill F, Levine J, Phillips MS. ease in HIV-negative patients: the evidence. Expert Rev Anti- Mycobacterium chimaera left ventricular assist device infections. Infect Ther. 2013;11(10):1065–77. J Card Surg. 2017;32(6):402–4. 53. Nessar R, Cambau E, Reyrat JM, Murray A, Gicquel B. 36. Ben Appenheimer A, et al. Mycobacterium chimaera outbreak Mycobacterium abscessus: a new antibiotic nightmare. J response: experience from four United States healthcare systems. Antimicrob Chemother. 2012;67(4):810–8. Open Forum Infect Dis. 2016;3(suppl_1):2392. 54. Tortoli E. Impact of genotypic studies on mycobacterial taxono- 37.• Li T, et al. A of waterborne infections from my: the new mycobacteria of the 1990s. Clin Microbiol Rev. nontuberculous mycobacteria in health care facility water systems. 2003;16(2):319–54. Int J Hyg Environ Health. 2017;220(3):611–20. This comprehen- 55. Vemulapalli RK, et al. Emergence of resistance to clarithromycin sive review of 22 observational studies provides a thorough during treatment of disseminated cutaneous Mycobacterium analysis of NTM infection outbreaks associated with chelonae infection: case report and literature review. J Inf Secur. healthcare facility water systems. 2001;43(3):163–8. 38. Schreiber PW, Kuster SP, Hasse B, Bayard C, Rüegg C, Kohler P, et al. Reemergence of Mycobacterium chimaera in heater-cooler 56. Nash KA, Zhang Y, Brown-Elliott BA, Wallace RJ Jr. Molecular units despite intensified cleaning and disinfection protocol. Emerg basis of intrinsic macrolide resistance in clinical isolates of Infect Dis. 2016;22(10):1830–3. Mycobacterium fortuitum. J Antimicrob Chemother. 2005;55(2): – 39. Schnabel D, Esposito DH, Gaines J, Ridpath A, Barry MA, 170 7. Feldman KA, et al. Multistate US outbreak of rapidly growing 57. Swenson JM, Wallace RJ, Silcox VA, Thornsberry C. mycobacterial infections associated with medical tourism to the Antimicrobial susceptibility of five subgroups of Dominican Republic, 2013–2014(1). Emerg Infect Dis. Mycobacterium fortuitum and Mycobacterium chelonae. – 2016;22(8):1340–7. Antimicrob Agents Chemother. 1985;28(6):807 11. 40. Zosso C, Lienhard R, Siegrist HH, Malinverni R, Clerc O. Post 58. Brown BA, Wallace RJ, Onyi GO, de Rosas V, Wallace RJ. liposuction infections by rapidly growing mycobacteria. Infect Dis Activities of four macrolides, including clarithromycin, against (Lond). 2015;47(2):69–72. Mycobacterium fortuitum, Mycobacterium chelonae, and M. 41. Kennedy BS, et al. Outbreak of Mycobacterium chelonae infec- chelonae-like organisms. Antimicrob Agents Chemother. – tion associated with tattoo ink. N Engl J Med. 2012;367(11): 1992;36(1):180 4. 1020–4. 59. Wallace RJ Jr, Brown BA, Onyi GO. Susceptibilities of 42. LeBlanc PM, Hollinger KA, Klontz KC. Tattoo ink-related infec- Mycobacterium fortuitum biovar. fortuitum and the two sub- tions—awareness, diagnosis, reporting, and prevention. N Engl J groups of Mycobacterium chelonae to imipenem, cefmetazole, Med. 2012;367(11):985–7. cefoxitin, and amoxicillin-clavulanic acid. Antimicrob Agents 43. Centers for Disease, C. and Prevention. Tattoo-associated Chemother. 1991;35(4):773–5. nontuberculous mycobacterial skin infections—multiple states, 60. Chiesi S, Piacentini D, Salerno ND, Luise D, Peracchi M, Concia 2011–2012. MMWR Morb Mortal Wkly Rep. 2012;61(33): E, et al. Disseminated Mycobacterium chimaera infection after 653–6. open heart surgery in an Italian woman: a case report and a review 44. Yacisin K, et al. Outbreak of non-tuberculous mycobacteria skin of the literature. Infez Med. 2017;25(3):267–9. or soft tissue infections associated with handling fish—New York 61. Cross GB, le Q, Webb B, Jenkin GA, Korman TM, Francis M, City, 2013–2014. Epidemiol Infect. 2017;145(11):2269–79. et al. Mycobacterium haemophilum bone and joint infection in 45. Wertman R, Miller M, GrobenP,MorrellDS,CultonDA. HIV/AIDS: case report and literature review. Int J STD AIDS. /Mycobacterium massiliense furunculosis 2015;26(13):974–81. associated with pedicure footbaths: a report of 3 cases. Arch 62. Lindeboom JA, Bruijnesteijn van Coppenraet LES, van Soolingen Dermatol. 2011;147(4):454–8. D, Prins JM, Kuijper EJ. Clinical manifestations, diagnosis, and 6 Page 14 of 17 Curr Infect Dis Rep (2018) 20:6

treatment of Mycobacterium haemophilum infections. Clin duration be reduced in selected patients? PLoS Negl Trop Dis. Microbiol Rev. 2011;24(4):701–17. 2015;9(2):e0003503. 63. Stürenburg E, Horstkotte MA, Aberle J, Meyer K, Richter E, 81.• Prevots DR, Marras TK. Epidemiology of human pulmonary in- Laufs R, et al. Disseminated Mycobacterium haemophilum infec- fection with nontuberculous mycobacteria: a review. Clin Chest tion as initial manifestation of AIDS. Tuberculosis. 2004;84(6): Med. 2015;36(1):13–34. Valuable summary of known data on 341–5. NTM incidence trends in Europe, Asia, North and South 64. Takeo N, et al. Case of mycobacterium haemophilum infection in America, Africa, and Australia. a Japanese renal transplant patient and a review of Japanese cases. 82. Hermansen TS, Ravn P, Svensson E, Lillebaek T. Nontuberculous J Dermatol. 2012;39(11):968–9. mycobacteria in Denmark, incidence and clinical importance dur- 65. Aubry A, Jarlier V, Escolano S, Truffot-Pernot C, Cambau E. ing the last quarter-century. Sci Rep. 2017;7(1):6696. Antibiotic susceptibility pattern of Mycobacterium marinum. 83. Yu X, et al. The prevalence of non-tuberculous mycobacterial Antimicrob Agents Chemother. 2000;44(11):3133–6. infections in mainland China: systematic review and meta-analy- 66. Aubry, A., Mougari F., Reibel F., Cambau E., Mycobacterium sis. J Inf Secur. 2016;73(6):558–67. marinum. Microbiol Spectr, 2017. 5(2). https://doi.org/10.1128/ 84. Parija S, Verma A. Prevalence of non-tuberculous mycobacterium AEM.00843-10 in a tertiary care hospital from South India. J Commun Dis. 67. Braback M, Riesbeck K, Forsgren A. Susceptibilities of 2012;4(3):129–37. Mycobacterium marinum to gatifloxacin, gemifloxacin, 85.• Cassidy PM, Hedberg K, Saulson A, McNelly E, Winthrop KL. , linezolid, moxifloxacin, telithromycin, and Nontuberculous mycobacterial disease prevalence and risk fac- quinupristin-dalfopristin (Synercid) compared to its susceptibili- tors: a changing epidemiology. Clin Infect Dis. 2009;49(12): ties to reference macrolides and quinolones. Antimicrob Agents e124–9. A detailed prevalence study of pulmonary and Chemother. 2002;46(4):1114–6. extrapulmonary NTM infections in the state of Oregon, USA. 68. van der Werf TS, et al. Mycobacterium ulcerans disease. Bull 86. Smith GS, Ghio AJ, Stout JE, Messier KP, Hudgens EE, Murphy World Health Organ. 2005;83(10):785–91. MS, et al. Epidemiology of nontuberculous mycobacteria isola- 69. Forbes BR, Wannan JS, Kirkland WB. Indolent cutaneous ulcer- tions among central North Carolina residents, 2006–2010. J Infect. ation due to infection with Mycobacterium ulcerans. Med J Aust. 2016;72(6):678–86. 1954;41(1 13):475–9. 87.• Wentworth AB, et al. Increased incidence of cutaneous 70. Sakyi SA, et al. Clinical and laboratory diagnosis of buruli ulcer nontuberculous mycobacterial infection, 1980 to 2009: a disease: a systematic review. Can J Infect Dis Med Microbiol. population-based study. Mayo Clin Proc. 2013;88(1):38–45. 2016;2016:5310718. This study documents a rising incidence of NTM infections 71. Vincent QB, Ardant MF, Marsollier L, Chauty A, Alcaïs A, in Olmstead County, Minnesota, USA. Franco-Beninese Buruli Research Group (listed in appendix). 88. Blanc P, Dutronc H, Peuchant O, Dauchy FA, Cazanave C, Neau HIV infection and Buruli ulcer in Africa. Lancet Infect Dis. D, et al. Nontuberculous mycobacterial infections in a French 2014;14(9):796–7. hospital: a 12-year retrospective study. PLoS One. 2016;11(12): 72. Ross BC, Marino L, Oppedisano F, Edwards R, Robins-Browne e0168290. RM, Johnson PD. Development of a PCR assay for rapid diagno- 89.• Lai C-C, et al. Increasing incidence of nontuberculous sis of Mycobacterium ulcerans infection. J Clin Microbiol. Mycobacteria, Taiwan, 2000–2008. Emerg Infect Dis. 1997;35(7):1696–700. 2010;16(2):294–6. This study shows an increasing incidence 73. Bretzel G, Siegmund V, Nitschke J, Herbinger KH, Thompson W, of NTM isolation from both pulmonary and extrapulmonary Klutse E, et al. A stepwise approach to the laboratory diagnosis of sources using a national hospital laboratory database. It is not Buruli ulcer disease. Tropical Med Int Health. 2007;12(1):89–96. clear how persons from whom the isolates were derived were 74. Siegmund V, Adjei O, Nitschke J, Thompson W, Klutse E, defined as infected vs colonized. Herbinger KH, et al. Dry reagent-based polymerase chain reaction 90. Lopez-Varela E, et al. Non-tuberculous mycobacteria in children: compared with other laboratory methods available for the diagno- muddying the waters of . Lancet Respir sis of Buruli ulcer disease. Clin Infect Dis. 2007;45(1):68–75. Med. 2015;3(3):244–56. 75. Herbinger KH, Adjei O, Awua-Boateng NY,Nienhuis WA,Kunaa 91. Al Yazidi LS, et al. Nontuberculous mycobacteria in children: a L, Siegmund V, et al. Comparative study of the sensitivity of focus on . Pediatr Infect Dis J. 2017;36(4): different diagnostic methods for the laboratory diagnosis of 374–8. Buruli ulcer disease. Clin Infect Dis. 2009;48(8):1055–64. 92. Apiwattankul N, Flynn PM, Hayden RT, Adderson EE. Infections 76. Beissner M, Phillips RO, Battke F, Bauer M, Badziklou K, Sarfo caused by rapidly growing Mycobacteria spp in children and ad- FS, et al. Loop-mediated isothermal amplification for laboratory olescents with cancer. J Pediatric Infect Dis Soc. 2015;4(2):104– confirmation of Buruli ulcer disease—towards a point-of-care test. 13. PLoS Negl Trop Dis. 2015;9(11):e0004219. 93. Iroh Tam PY, et al. Non-tuberculous mycobacterial infection in 77. Babonneau J, Bernard C, Marion E, Chauty A, Kempf M, Robert hospitalized children: a case series. Epidemiol Infect. R, et al. Development of a dry-reagent-based qPCR to facilitate the 2015;143(15):3173–81. diagnosis of mycobacterium ulcerans infection in endemic coun- 94. Ng SS, et al. Pediatric cutaneous nontuberculous mycobacterium tries. PLoS Negl Trop Dis. 2015;9(4):e0003606. infections in Singapore. Pediatr Dermatol. 2015;32(4):488–94. 78. Converse PJ, Xing Y,Kim KH, Tyagi S, Li SY,Almeida DV,et al. 95. Sax H, Bloemberg G, Hasse B, Sommerstein R, Kohler P, Accelerated detection of mycolactone production and response to Achermann Y, et al. Prolonged outbreak of Mycobacterium chi- antibiotic treatment in a mouse model of Mycobacterium ulcerans maera infection after open-chest heart surgery. Clin Infect Dis. disease. PLoS Negl Trop Dis. 2014;8(1):e2618. 2015;61(1):67–75. 79. Friedman ND, Athan E, Walton AL, O'Brien DP. Increasing ex- 96.• Sommerstein R, et al. Transmission of Mycobacterium chimaera perience with primary oral medical therapy for Mycobacterium from heater-cooler units during cardiac surgery despite an ultra- ulcerans disease in an Australian cohort. Antimicrob Agents clean air ventilation system. Emerg Infect Dis. 2016;22(6):1008– Chemother. 2016;60(5):2692–5. 13. This experimental study used multiple methods, including 80. Cowan R, Athan E, Friedman ND, Hughes AJ, McDonald A, particle counter, smoke dispersal, and sedimentation plates, to Callan P, et al. Mycobacterium ulcerans treatment—can antibiotic demonstrate airborne transmission of M. chimaera. Curr Infect Dis Rep (2018) 20:6 Page 15 of 17 6

97. Gotting T, et al. Heater-cooler units: contamination of crucial de- 112. Bryers JD. Medical biofilms. Biotechnol Bioeng. 2008;100(1):1– vices in cardiothoracic surgery. J Hosp Infect. 2016;93(3):223–8. 18. 98.• Chand M, et al. Insidious risk of severe Mycobacterium chimaera 113. Walker J, Moore G, Collins S, Parks S, Garvey MI, Lamagni T, infection in cardiac surgery patients. Clin Infect Dis. 2017;64(3): et al. Microbiological problems and biofilms associated with 335–42. This 18-patient series is the result of a UK national Mycobacterium chimaera in heater-cooler units used for cardio- investigation and provides the best current estimate of inci- pulmonary bypass. J Hosp Infect. 2017;96(3):209–20. dence, timing, and outcome of M. chimaera infection. 114. Mishra AK, Driessen NN, Appelmelk BJ, Besra GS. 99. Perkins KM, Lawsin A, Hasan NA, Strong M, Halpin AL, Rodger Lipoarabinomannan and related glycoconjugates: structure, bio- RR, et al. Notes from the field: Mycobacterium chimaera contam- genesis and role in mycobacterium tuberculosis physiology and ination of heater-cooler devices used in cardiac surgery—United host-pathogen interaction. FEMS Microbiol Rev. 2011;35(6): States. MMWR Morb Mortal Wkly Rep. 2016;65(40):1117–8. 1126–57. 100. Svensson E, Jensen ET, Rasmussen EM, Folkvardsen DB, 115. Kallenius G, et al. Lipoarabinomannan, and its related glycolipids, Norman A, Lillebaek T. Mycobacterium chimaera in heater- induce divergent and opposing immune responses to cooler units in Denmark related to isolates from the United Mycobacterium tuberculosis depending on structural diversity States and United . Emerg Infect Dis. 2017;23(3):507–9. and experimental variations. Tuberculosis (Edinb). 2016;96:120– 101.• van Ingen, J., Kohl, TA., Kranzer, K., Hasse, B., Keller, PM., 30. Szafrańska, AK., et al., Global outbreak of severe Mycobacterium 116. Stoop EJ, et al. Mannan core branching of lipo(arabino)mannan is chimaera disease after cardiac surgery: a molecular epidemiological required for mycobacterial virulence in the context of innate im- – study. Lancet Infect Dis, 2017;17(10):1033–41. This molecular munity. Cell Microbiol. 2013;15(12):2093 108. epidemiologic study based on whole genome sequencing of 117. Helguera-Repetto AC, et al. Differential macrophage response to 250 isolates from patient, equipment and environmental slow- and fast-growing pathogenic mycobacteria. Biomed Res Int. sources confirmed the LivaNova factory as the likely source 2014;2014:916521. for the global Mycobacterium chimaera outbreak and is an 118. Griffith DE, Brown-Elliott BA, L. Benwill J, Wallace RJ Jr. “ excellent overview. Mycobacterium abscessus. Pleased to meet you, hope you guess ” – 102. Allen KB, Yuh DD, Schwartz SB, Lange RA, Hopkins R, Bauer my name ... . Ann Am Thorac Soc. 2015;12(3):436 9. K, et al. Nontuberculous mycobacterium infections associated 119. Griffith DE, Philley JV, Brown-Elliott BA, Benwill JL, Shepherd with heater-cooler devices. Ann Thorac Surg. 2017;104:1237–42. S, et al. The significance of Mycobacterium abscessus subspecies abscessus isolation during Mycobacterium avium complex lung 103. Baker AW, Lewis SS, Alexander BD, Chen LF, Wallace RJ Jr, disease therapy. Chest. 2015;147(5):1369–75. Brown-Elliott BA, et al. Two-phase hospital-associated outbreak 120. Nash KA, Brown-Elliott BA, Wallace RJ Jr. A novel gene, of Mycobacterium abscessus: investigation and mitigation. Clin Infect Dis. 2017;64(7):902–11. erm(41), confers inducible macrolide resistance to clinical isolates of Mycobacterium abscessus but is absent from Mycobacterium 104. Trudzinski, F.C., Schlotthauer U., Kamp A., Hennemann K., chelonae. Antimicrob Agents Chemother. 2009;53(4):1367–76. Muellenbach R.M., Reischl U., et al., Clinical implications of 121. Ridell M. Mycobacterium abscessus: an environmental Mycobacterium chimaera detection in thermoregulatory devices mycobacteria being a human pathogen. In: Farnia P, editor. 1st used for extracorporeal membrane oxygenation (ECMO), Asian-African Congress of International Journal of Germany, 2015 to 2016. Euro Surveill. 2016;21(46). https://doi. Mycobacteriology. Iran: Asian-African Society for org/10.2807/1560-7917.ES.2016.21.46.30398. Mycobacteriology; 2014. p. 41. 105. Alexander DC, Vasireddy R, Vasireddy S, Philley JV, Brown- 122. Bernut A, Herrmann JL, Kissa K, Dubremetz JF, Gaillard JL, Elliott BA, Perry BJ, et al. Emergence of mmpT5 variants during Lutfalla G, et al. Mycobacterium abscessus cording prevents bedaquiline treatment of mycobacterium intracellulare lung dis- phagocytosis and promotes abscess formation. Proc Natl Acad ease. J Clin Microbiol. 2017;55(2):574–84. Sci U S A. 2014;111(10):E943–52. 106. Zweifel SA, Mihic-Probst D, Curcio CA, Barthelmes D, Thielken 123. Jonsson B, Ridell M, Wold AE. Phagocytosis and cytokine re- A, Keller PM, et al. Clinical and histopathologic ocular findings in sponse to rough and smooth colony variants of Mycobacterium disseminated Mycobacterium chimaera infection after cardiotho- abscessus by human peripheral blood mononuclear cells. APMIS. – racic surgery. Ophthalmology. 2017;124(2):178 88. 2013;121(1):45–55. 107. Tortoli E, et al. Proposal to elevate the genetic variant MAC-A, 124. Roux AL, Viljoen A, Bah A, Simeone R, Bernut A, Laencina L, included in the Mycobacterium avium complex, to species rank as et al. The distinct fate of smooth and rough Mycobacterium Mycobacterium chimaera sp. nov. Int J Syst Evol Microbiol. abscessus variants inside macrophages. Open Biol. 2016;6(11). – 2004;54(Pt 4):1277 85. https://doi.org/10.1098/rsob.160185. 108. Chaisson RE, Benson CA, Dube MP, Heifets LB, Korvick JA, 125. Rhoades ER, Archambault AS, Greendyke R, Hsu FF, Streeter C, Elkin S, et al. Clarithromycin therapy for bacteremic Byrd TF. Mycobacterium abscessus glycopeptidolipids mask un- Mycobacterium avium complex disease. A randomized, double- derlying cell wall phosphatidyl-myo-inositol mannosides blocking blind, dose-ranging study in patients with AIDS. AIDS Clinical induction of human macrophage TNF-alpha by preventing inter- Trials Group protocol 157 study team. Ann Intern Med. action with TLR2. J Immunol. 2009;183(3):1997–2007. 1994;121(12):905–11. 126. Ripoll F, Pasek S, Schenowitz C, Dossat C, Barbe V, Rottman M, 109. Griffith DE, Brown BA, Girard WM, Murphy DT, Wallace RJ. et al. Non mycobacterial virulence genes in the genome of the Azithromycin activity against Mycobacterium avium complex emerging pathogen Mycobacterium abscessus. PLoS One. lung disease in patients who were not infected with human immu- 2009;4(6):e5660. nodeficiency virus. Clin Infect Dis. 1996;23(5):983–9. 127. Bakala N'Goma JC, le Moigne V, Soismier N, Laencina L, le 110. Wallace RJ Jr, et al. Initial clarithromycin monotherapy for Chevalier F, Roux AL, et al. Mycobacterium abscessus phospho- Mycobacterium avium-intracellulare complex lung disease. Am lipase C expression is induced during coculture within amoebae J Respir Crit Care Med. 1994;149(5):1335–41. and enhances M. abscessus virulence in mice. Infect Immun. 111. Nielsen C, Winther CL, Thomsen PK, Andreasen JJ. Elimination 2015;83(2):780–91. of Mycobacterium chimaera in a heater cooler device used during 128. Delafont V, Mougari F, Cambau E, Joyeux M, Bouchon D, on-pump cardiothoracic surgery. Perfusion. 2017;32(6):507–10. Héchard Y, et al. First evidence of amoebae-mycobacteria 6 Page 16 of 17 Curr Infect Dis Rep (2018) 20:6

association in drinking water network. Environ Sci Technol. 144. Lienard J, Movert E, Valfridsson C, Sturegård E, Carlsson F. ESX- 2014;48(20):11872–82. 1 exploits type I IFN-signalling to promote a regulatory macro- 129. Adekambi T, Drancourt M. Dissection of phylogenetic relation- phage phenotype refractory to IFN gamma-mediated autophagy ships among 19 rapidly growing Mycobacterium species by 16S and growth restriction of intracellular mycobacteria. Cell rRNA, hsp65, sodA, recA and rpoB gene sequencing. Int J Syst Microbiol. 2016;18(10):1471–85. Evol Microbiol. 2004;54(Pt 6):2095–105. 145. Abdallah AM, Bestebroer J, Savage NDL, de Punder K, van Zon 130. Peterson TS, Ferguson JA, Watral VG, Mutoji KN, Ennis DG, M, Wilson L, et al. Mycobacterial secretion systems ESX-1 and Kent ML. Paramecium caudatum enhances transmission and in- ESX-5 play distinct roles in host cell death and Inflammasome fectivity of Mycobacterium marinum and M. chelonae in zebrafish activation. J Immunol. 2011;187(9):4744–53. Danio rerio. Dis Aquat Org. 2013;106(3):229–39. 146. Sizaire V, Nackers F, Comte E, Portaels F. Mycobacterium 131. Cabello-Vilchez AM, et al. Endosymbiotic Mycobacterium ulcerans infection: control, diagnosis, and treatment. Lancet chelonae in a Vermamoeba vermiformis strain isolated from the Infect Dis. 2006;6(5):288–96. nasal mucosa of an HIV patient in lima, Peru. Exp Parasitol. 147. WHO, http://www.who.int/gho/neglected_diseases/buruli_ulcer/ 2014;145(Suppl):S127–30. en/. 2016. 132. Brown-Elliott BA, Philley JV, Rapidly growing mycobacteria. 148. George KM, Chatterjee D, Gunawardana G, Welty D, Hayman J, Microbiol Spectr. 2017;5(1). https://doi.org/10.1128/ Lee R, et al. Mycolactone: a polyketide toxin from microbiolspec.TNMI7-0027-2016. Mycobacterium ulcerans required for virulence. Science. 133. Lee GJ, Lee HM, Kim TS, Kim JK, Sohn KM, Jo EK. 1999;283(5403):854–7. Mycobacterium fortuitum induces A20 expression that impairs 149. Sarfo FS, Phillips R, Wansbrough-Jones M, Simmonds RE. macrophage inflammatory responses. Pathog Dis. 2016;74(3): Recent advances: role of mycolactone in the pathogenesis and ftw015. monitoring of Mycobacterium ulcerans infection/Buruli ulcer dis- 134. Datta D, Khatri P, Banerjee C, Singh A, Meena R, Saha DR, et al. ease. Cell Microbiol. 2016;18(1):17–29. Calcium and superoxide-mediated pathways converge to induce 150.•• Guenin-Mace L, et al. Mycolactone activation of Wiskott-Aldrich nitric oxide-dependent apoptosis in mycobacterium fortuitum- syndrome proteins underpins Buruli ulcer formation. J Clin Invest. infected fish macrophages. PLoS One. 2016;11(1):e0146554. 2013;123(4):1501–12. One of two recent studies to provide the 135. Davis JM, Clay H, Lewis JL, Ghori N, Herbomel P, Ramakrishnan first molecular explanation for the cytopathic effects of L. Real-time visualization of mycobacterium-macrophage interac- mycolactone, the key toxin produced by M. ulcerans. tions leading to initiation of granuloma formation in zebrafish 151.•• Hall BS, et al. The pathogenic mechanism of the Mycobacterium embryos. Immunity. 2002;17(6):693–702. ulcerans virulence factor, mycolactone, depends on blockade of 136. Davis JM, Ramakrishnan L. The role of the granuloma in expan- protein translocation into the ER. PLoS Pathog. 2014;10(4): sion and dissemination of early tuberculous infection. Cell. e1004061. One of two recent studies to provide the first molec- 2009;136(1):37–49. ular explanation for the cytopathic effects of mycolactone, the 137. Volkman HE, Pozos TC, Zheng J, Davis JM, Rawls JF, key toxin produced by M. ulcerans. Ramakrishnan L. Tuberculous granuloma induction via interac- 152. Merritt RW, Walker ED, Small PLC, Wallace JR, Johnson PDR, tion of a bacterial secreted protein with host epithelium. Science. Benbow ME, et al. Ecology and transmission of Buruli ulcer dis- 2010;327(5964):466–9. ease: a systematic review. PLoS Negl Trop Dis. 2010;4(12):e911. 138.• Cambier CJ, et al. Mycobacteria manipulate macrophage recruit- 153. Brown-Elliott BA, Wallace RJ Jr. Enhancement of conventional ment through coordinated use of membrane lipids. Nature. phenotypic methods with molecular-based methods for the more 2014;505(7482):218–22. This basic research article exemplifies definitive identification of nontuberculous Mycobacteria. Clin how the zebrafish/M. marinum model of tuberculosis has ad- Microbiol Newsl. 2012;34(14):109–15. vanced our understanding of mycobacterial pathogenesis. 154.• van Ingen J. Diagnosis of nontuberculous mycobacterial infec- 139. Cambier CJ, O’Leary SM, O’Sullivan MP, Keane J, tions. Semin Respir Crit Care Med. 2013;34(1):103–9. A nice Ramakrishnan L. Phenolic glycolipid facilitates mycobacterial overview of new molecular tools for the diagnosis of NTM. escape from Microbicidal tissue-resident macrophages. 155. Springer B, Stockman L, Teschner K, Roberts GD, Böttger EC. Immunity. 2017;47:552–565.e4. Two-laboratory collaborative study on identification of 140. Mortimer TD, Weber AM, Pepperell CS. Evolutionary thrift: mycobacteria: molecular versus phenotypic methods. J Clin mycobacteria repurpose plasmid diversity during adaptation of Microbiol. 1996;34(2):296–303. type VII secretion systems. Genome Biol Evol. 2017;9(3):398– 156. Pranada AB, Witt E, Bienia M, Kostrzewa M, Timke M. Accurate 413. differentiation of Mycobacterium chimaera from Mycobacterium 141.• Groschel MI, et al. ESX secretion systems: mycobacterial evolu- intracellulare by MALDI-TOF MS analysis. J Med Microbiol. tion to counter host immunity. Nat Rev Microbiol. 2016;14(11): 2017;66(5):670–7. 677–91. This review provides an introduction to the evolution 157. Richter E, Rüsch-Gerdes S, Hillemann D. Evaluation of the and functions of the family of type VII secretion systems and GenoType Mycobacterium assay for identification of mycobacte- how they function to promote mycobacterial survival and ex- rial species from cultures. J Clin Microbiol. 2006;44(5):1769–75. pansion in various hosts. 158. Russo C, Tortoli E, Menichella D. Evaluation of the new 142. de Jonge MI, Pehau-Arnaudet G, Fretz MM, Romain F, Bottai D, GenoType mycobacterium assay for identification of mycobacte- Brodin P, et al. ESAT-6 from mycobacterium tuberculosis disso- rial species. J Clin Microbiol. 2006;44(2):334–9. ciates from its putative chaperone CFP-10 under acidic conditions 159. Mougari F, Loiseau J, Veziris N, Bernard C, Bercot B, Sougakoff and exhibits membrane-lysing activity. J Bacteriol. 2007;189(16): W, et al. Evaluation of the new GenoType NTM-DR kit for the 6028–34. molecular detection of antimicrobial resistance in non-tuberculous 143. Conrad WH, Osman MM, Shanahan JK, Chu F, Takaki KK, mycobacteria. J Antimicrob Chemother. 2017;72(6):1669–77. Cameron J, et al. Mycobacterial ESX-1 secretion system mediates 160. Tortoli E, Mariottini A, Mazzarelli G. Evaluation of INNO-LiPA host cell lysis through bacterium contact-dependent gross mem- MYCOBACTERIA v2: improved reverse hybridization multiple brane disruptions. Proc Natl Acad Sci U S A. 2017;114(6): DNA probe assay for Mycobacterial identification. J Clin 1371–6. Microbiol. 2003;41(9):4418–20. Curr Infect Dis Rep (2018) 20:6 Page 17 of 17 6

161. Garcia-Agudo L, et al. Evaluation of INNO-LiPA mycobacteria 176. Hards K, Robson JR, Berney M, Shaw L, Bald D, Koul A, et al. v2 assay for identification of rapidly growing mycobacteria. Braz J Bactericidal mode of action of bedaquiline. J Antimicrob Microbiol. 2011;42(3):1220–6. Chemother. 2015;70(7):2028–37. 162. Hall L, Doerr KA, Wohlfiel SL, Roberts GD. Evaluation of the 177. Chauffour A, Robert J, Veziris N, Aubry A, Jarlier V. Sterilizing MicroSeq system for identification of mycobacteria by 16S ribo- activity of fully oral intermittent regimens against mycobacterium somal DNA sequencing and its integration into a routine clinical Ulcerans infection in mice. PLoS Negl Trop Dis. 2016;10(10): mycobacteriology laboratory. J Clin Microbiol. 2003;41(4):1447– e0005066. 53. 178. Lerat I, Cambau E, Roth dit Bettoni R, Gaillard JL, Jarlier V, 163. Arnold C, Barrett A, Cross L, Magee JG. The use of rpoB se- Truffot C, et al. In vivo evaluation of antibiotic activity against quence analysis in the differentiation of Mycobacterium abscessus Mycobacterium abscessus. J Infect Dis. 2014;209(6):905–12. and Mycobacterium chelonae: a critical judgement in cystic fibro- 179. Obregon-Henao A, et al. Susceptibility of Mycobacterium sis? Clin Microbiol Infect. 2012;18(5):E131–3. abscessus to antimycobacterial drugs in preclinical models. 164. Wang H, Yue J, Han M, Yang J, Zhao Y. Rapid method for iden- Antimicrob Agents Chemother. 2015;59(11):6904–12. tification of six common species of mycobacteria based on multi- 180. Philley JV, Wallace RJ Jr, Benwill JL, Taskar V, Brown-Elliott plex SNP analysis. J Clin Microbiol. 2010;48(1):247–50. BA, Thakkar F, et al. Preliminary results of bedaquiline as salvage 165. Schweickert B, Goldenberg O, Richter E, Göbel UB, Petrich A, therapy for patients with nontuberculous mycobacterial lung dis- – Buchholz P, et al. Occurrence and clinical relevance of ease. Chest. 2015;148(2):499 506. Mycobacterium chimaera sp. nov., Germany. Emerg Infect Dis. 181. Lounis N, Gevers T, van den Berg J, Andries K. Impact of the 2008;14(9):1443–6. interaction of R207910 with rifampin on the treatment of tubercu- 166. Chen R, Gao XB, Liu ZH, Shen XB, Guo AZ, Duan YY, et al. losis studied in the mouse model. Antimicrob Agents Chemother. – Combination of multiplex PCR with denaturing high-performance 2008;52(10):3568 72. liquid chromatography for rapid detection of Mycobacterium ge- 182. Brown-Elliott BA, Wallace RJ Jr. In vitro susceptibility testing of nus and simultaneous identification of the mycobacterium tuber- tedizolid against nontuberculous mycobacteria. J Clin Microbiol. – culosis complex. Diagn Microbiol Infect Dis. 2013;77(1):53–7. 2017;55(6):1747 54. 183. Yuste JR, Bertó J, del Pozo JL, Leiva J. Prolonged use of tedizolid 167. Singhal N, et al. MALDI-TOF mass spectrometry: an emerging in a pulmonary non-tuberculous mycobacterial infection after technology for microbial identification and diagnosis. Front linezolid-induced toxicity. J Antimicrob Chemother. 2017;72(2): Microbiol. 2015;6:791. 625–8. 168. Murray PR. What is new in clinical microbiology—microbial 184.•• Kumar P, Chauhan V, Silva JRA, Lameira J, d'Andrea FB, Li SG, identification by MALDI-TOF mass spectrometry: a paper from et al., Mycobacterium abscessus l,d-transpeptidases are suscepti- the 2011 William Beaumont Hospital symposium on molecular ble to inactivation by carbapenems and cephalosporins but not pathology. J Mol Diagn. 2012;14(5):419–23. penicillins. Antimicrob Agents Chemother. 2017;61(10). https:// 169. Krieger D, Schönfeld N, Vesenbeckh S, Bettermann G, Bauer TT, doi.org/10.1128/AAC.00866-17. This study examined the l,d- Rüssmann H, et al. Is delamanid a potential agent in the treatment transpeptidases of M. abscessus and cloned, expressed, and of diseases caused by Mycobacterium avium-intracellulare? Eur studied the interactions of β-lactams with two of the L,D- Respir J. 2016;48(6):1803–4. transpeptidases, and demonstrated inhibition with carbapen- 170. Andries K, Verhasselt P, Guillemont J, Göhlmann HW, Neefs JM, ems and cephalosporins, and potential role of dual-beta lac- Winkler H, et al. A diarylquinoline drug active on the ATP syn- tam therapy in treatment of M. abscessus. thase of Mycobacterium tuberculosis. Science. 2005;307(5707): 185. Soroka D, Dubee V, Soulier-Escrihuela O, Cuinet G, Hugonnet – 223 7. JE, Gutmann L, et al. Characterization of broad-spectrum 171. Vesenbeckh S, Schönfeld N, Roth A, Bettermann G, Krieger D, Mycobacterium abscessus class a beta-lactamase. J Antimicrob Bauer TT, et al. Bedaquiline as a potential agent in the treatment of Chemother. 2014;69(3):691–6. M. intracellulare and M. avium infections. Eur Respir J, 2017. 186. Dubee V, et al. Beta-lactamase inhibition by avibactam in 49(3). https://doi.org/10.1183/13993003.00083-2017. Mycobacterium abscessus. J Antimicrob Chemother. 172. Brown-Elliott BA, Philley JV, Griffith DE, Thakkar F, Wallace Jr 2015;70(4):1051–8. RJ, In vitro susceptibility testing of bedaquiline against 187. Kaushik A, Gupta C, Fisher S, Story-Roller E, Galanis C, Parrish Mycobacterium avium Complex. Antimicrob Agents N, et al. Combinations of avibactam and carbapenems exhibit Chemother. 2017;61(2). https://doi.org/10.1128/AAC.01798-16. enhanced potencies against drug-resistant Mycobacterium 173. Pang Y, Zheng H, Tan Y, Song Y, Zhao Y. In vitro activity of abscessus. Future Microbiol. 2017;12:473–80. bedaquiline against nontuberculous mycobacteria in China. 188.•• Lefebvre AL, Le Moigne V, Bernut A, Veckerlé C, Compain F, – Antimicrob Agents Chemother. 2017;61(5):e02627 16. Herrmann JL, et al. Inhibition of the β-lactamase BlaMab by 174. Dupont C, Viljoen A, Thomas S, Roquet-Banères F, Herrmann JL, avibactam improves the in vitro and in vivo efficacy of imipenem Pethe K, et al. Bedaquiline inhibits the ATP synthase in mycobac- against Mycobacterium abscessus. Antimicrob Agents terium abscessus and is effective in infected zebrafish. Antimicrob Chemother. 2017;61(4). https://doi.org/10.1128/AAC.02440-16. – Agents Chemother. 2017;61:e01225 17. The authors review the effects of BlaMab on M abscessus 175. Ji B, Lefrancois S, Robert J, Chauffour A, Truffot C, Jarlier V. susceptibility to β-lactams and demonstrate that the benefit In vitro and in vivo activities of rifampin, streptomycin, amikacin, of adding avibactam to imipenem is likely underestimated moxifloxacin, R207910, linezolid, and PA-824 against mycobac- in vitro. They show that in human macrophages and zebrafish terium ulcerans. Antimicrob Agents Chemother. 2006;50(6): β-lactamase is produced at a high level. The addition of 1921–6. avibactam increases the efficacy of imipenem.