Extended-Spectrum β-Lactamases in Escherichia coli and Klebsiella pneumoniae in Gulls, Alaska, USA

Previously reported babesiosis cases in children have been mostly acquired by blood transfusion (10). The patient had no history of transfusions with blood products and had never traveled outside his home town before disease onset. Although he and his parents did not recall any tick bites, he was at high risk for exposure to ticks because he often played with his dog, which frequently went outdoors in a tick-infested forested area. The dog may have transmitted a Babesia sp.–infected tick to the patient. However, ticks from the dog were not available for identification and testing. The patient in our study was presumed to be healthy and immunocompetent, which indicates that Babesia species can cause infections even in healthy persons. Babesiosis should be considered in the differential diagnosis of patients with a history of tick exposure and prolonged and irregular fever. Blood smear evaluation for intraerythrocytic parasites should be considered. The patient was treated with azithromycin and atovaquone and the parasites were cleared within 1 month. This combined treatment was well tolerated and effective, and it can be recommended as an alternative treatment to the commonly used therapy of quinine and clindamycin (1).

Although CTX-Ms are geographically widely distributed, reports of extendedspectrum β-lactamase (ESBL) dissemination are few from remote regions. In 2008, we reported phenotypic resistance traits in Escherichia coli isolates in 8.2% of wild birds sampled in the Arctic (2). We sampled approximately 260 wild birds, mainly gulls and geese, but found no ESBL-harboring isolates (J. Bonnedahl et al., unpub. data). Here we report results of our 2010 study at Barrow, Alaska, USA, a follow up to our 2005 study in which we found vancomycin-resistant enterococci (VRE) with clear traits of human origin in glaucous gulls (3). Our findings show a remarkable change, not in VRE dissemination, which is fairly unchanged, but in the emergence of ESBLs and general resistance of E. coli isolates.
We collected 150 fecal samples from a population of adult gulls residing close to a landfill site. For a description of general resistance levels (4,5), susceptibility of 1 randomly selected E. coli isolate per sample (137 isolated from 150 samples) was tested to a set of 10 antibacterial agents. Nearly half (48%) of the 137 E. coli isolates were resistant to at least 1 of the drugs tested. Resistance to 1 or 2 antimicrobial agents was found in 32% and 13% of the tested isolates, respectively, and resistance to >3 was found in 3% of isolates (online Technical Appendix Table, wwwnc. cdc.gov/EID/article/20/5/13-0325-Techapp1.pdf).

LETTERS
We analyzed samples for presence of VRE (3). Seven (4.7%) E. faecium isolates were found, all of which harbored both the vanA and the esp genes (found in isolates of the CC17 lineage) (3). No other VRE were found.
To investigate the presence of ESBL-producing bacteria, we conducted a selective screen as described (6). ESBL-producing bacteria were found (E. coli and K. pneumoniae), and ESBL genes (bla CTX-M , bla SHV , and bla TEM ) in ESBL-positive isolates were analyzed (6). We found 33 E. coli and 35 K. pneumoniae ESBLproducing isolates in 55 samples (12 samples had >1 unique isolate), a total of 37% of ESBL-harboring samples (Table).
In our 2005 study in Barrow, general resistance was relatively low, and no ESBL was found; surprisingly, however, 2 VRE isolates of a human clonal lineage were found (3; M. Drobni et al., unpub. data). Since then, resistance dissemination, particularly that of ESBLs, has exploded globally (1). In 2010, we found a high level of general resistance; 48% of randomly selected E. coli isolates displayed resistance toward >1 antibacterial drugs. This level is similar to the level we found in gulls in France in 2008, an area with high current and historical clinical antibacterial drug use and where birds have close contact with human activities (4).
We screened samples for VRE and ESBL-producing bacteria.  (7), more similar to results of our current study but in contrast also because they investigated gulls from a highly populated area.
E. coli isolates mainly carried bla CTX-M-14 or bla TEM-19 , whereas K. pneumoniae isolates mainly carried bla CTX-M-15 , bla SHV-12 , or bla SHV-102 . To our knowledge, ESBLs in E. coli and K. pneumoniae have not been reported from Alaska, but in two 10-year perspective reports from Canada (8,9), similar patterns and genotypes are reported in E. coli and K. pneumoniae in clinical isolates (mainly from samples of persons with urinary tract infections and urosepsis). Our MLST of E. coli indicated 4 known STs; ST10, ST38, ST131, and ST405, all very common in the material from Canada (8), and major STs responsible for CTX-M dissemination worldwide (1). Two novel STs were found; several isolates were designated to 1 of them. We conclude that the relatively limited variation in clonal variants (STs) and ESBL genotypes is a consequence of recent introduction from connecting areas, such as Canada, possibly directly by bird migration or human activities, of a few resistant clones, followed by a local clonal expansion. This conclusion is supported by our 2005 study showing no ESBLs and by studies showing where different clones might have been introduced continuously for long periods, such as our study in France (4), which display a much larger diversity.
The dissemination of ESBLs to Barrow is part of this global pattern, and it is safe to say that humans and wildlife share resistant E. coli flora. When areas such as remote parts of Alaska are affected, global coverage is imminent. 14

Staphylococcus aureus Carrying mecC Gene in Animals and Urban Wastewater, Spain
To the Editor: A new methicillin resistance mechanism gene, a divergent mecA homologue named mecC (formerly mecA LGA251 ), was recently described in Staphylococcus aureus (1). Methicillin-resistant S. aureus (MRSA) isolates carrying mecC have been recovered from humans, ruminants, pets, and other animals such as rats, seals, and guinea pigs (1-3). It has been suggested that mecC-carrying MRSA isolates might not be detected by using MRSA selective media (4 (5), and MICs can be affected by the drug-susceptibility testing method used (1,5).
mecC was detected in a total of 4 isolates from wild boar (n = 1), fallow deer (n = 2), and urban wastewater (n = 1); these isolates represent 1% of the 361 tested isolates. The 3 isolates recovered from animals were susceptible to all antimicrobial drugs tested other than β-lactams and to oxacillin (MICs 0.5-1 mg/L) but were resistant to penicillin (MICs 0.5-2 mg/L). Two of the isolates were resistant to cefoxitin (MICs 8 and 16 mg/L) and the third was susceptible (MIC 4 mg/L). The wastewater isolate was resistant to penicillin (MIC 2 mg/L) and erythromycin (MIC 16 mg/L) and susceptible to all other antimicrobial drugs tested, including cefoxitin (MIC 4 mg/L) and oxacillin (MIC ≤0.25 mg/L).