Fly Reservoir Associated with Wohlfahrtiimonas Bacteremia in a Human

Wohlfahrtiimonas species bacteria were isolated from the bloodstream of a patient with septicemia and wound myiasis. Environmental investigations identified a Wohlfahrtiimonas sp. among insects in the Americas and in a previously undescribed vector, the green bottle fly (Lucilia sericata). The isolates possibly represent a new species within the genus Wohlfahrtiimonas.

Billerica, MA, USA) in accordance with the manufacturer's recommendations (2). The generated spectra were compared with the spectra referenced in the MALDI-TOF MS database (PHL: RUO library MBT BD-5627 MSP List; CDC: Bruker 6903 database). For analysis, an isolate was considered as identified for log scores >2.0.

16S Ribosomal RNA Sequencing
Purification of whole-cell DNA and amplification and sequencing of the near full-length 16S ribosomal RNA (rRNA) gene were performed at CDC as previously described (3). The 16S rRNA gene sequence was assembled with Geneious 8.1.8 software (Biomatters, Auckland, New Zealand) by using de novo assembly. To identify related gene sequences in the GenBank database, consensus sequence was submitted to GenBank by using BLASTN software (https://www.ncbi.nlm.nih.gov/blast/). A multiple sequence alignment was created by using Clustal W (within Geneious 8.1.8), from which gaps and 5 and 3 ends were trimmed.
We inferred a phylogenetic tree by using the neighbor-joining method (4), and the topology was assessed by a bootstrap analysis of 1000 replicates (5). We computed the evolutionary distances by using the Tamura-Nei method (6); they are in the units of the number of base substitutions per site. Evolutionary analyses were conducted in MEGA6 (7).

Entomology
All larvae found on the patient were discarded before the investigation and could not be sampled, but live and dead insects were collected from the patient's home. Insects were identified to genus or species level by using standard entomological identification keys (8)(9)(10).
Live fly larvae were left to hatch at room temperature inside a mosquito-breeding container (BioQuip Products, Rancho Dominguez, CA, USA). Fly larvae were provided raw meat and freshly boiled chicken; adult flies were provided slices of fresh fruit.

Bacterial Isolation from Insects
Live fly larvae and adults were cultured for Wohlfahrtiimonas spp. Because the preferred growth medium of Wohlfahrtiimonas spp. is poorly described, all samples were inoculated on 3 replicates of blood agar, MacConkey agar, and nutrient agar. Phenylethyl alcohol agar was used for initial specimens (larvae) obtained from the patient's home. Plates were incubated at 25°C, 30°C, and 35°C in aerobic conditions over 2 days.
To remove surface contamination, we rinsed all insect specimens 5 times with sterile PBS in sterile tubes, as described previously (11). We homogenized the washed specimens with a sterile mortar (Fisher Scientific, Richardson, TX, USA) in 0.5 mL PBS and sequentially diluted to 1/1000. The first wash, fifth wash, undiluted homogenate, and diluted homogenate were inoculated in the same manner as fly larvae and adults. The objective was to verify whether surface contamination was present (first wash) and, if so, whether it was removed through rinsing (fifth wash). To verify that live insects were not inoculated with Wohlfahrtiimonas spp. through contaminated food, the meat and fruit (with the exception of boiled chicken) was homogenized and cultured in the same manner as fly larvae and adults.
To reduce the risk for agar plates being overgrown by other bacterial species and increase the sensitivity of isolating Wohlfahrtiimonas spp., we placed a sample of the diluted homogenate in nutrient broth made with 11.5 g of Difco nutrient agar (Becton Dickinson) and 500 mL distilled water and incubated at 25°C, 30°C, and 35°C over 2 days. We then inoculated this material on agar for all samples that did not initially grow Wohlfahrtiimonas spp.
Because PFGE has not been previously described for Wohlfahrtiimonas spp., we experimented with restriction enzymes commonly used on gram-negative organisms related to Wohlfahrtiimonas spp. (13,14), including NotI (Roche), XBaI (Roche) and SpeI (Roche) restriction enzymes. The standard lanes consisted of Salmonella ser. Braenderup H9812 cut with restriction enzyme XbaI. Gel electrophoresis was run by using CHEF Mapper XA system (Bio-Rad Laboratories, Hercules, CA, USA) for 18.5 hours with initial and final switch time at 2.2 and 54.2 seconds, respectively. Analysis of PFGE patterns was conducted with BioNumerics (version 6.6). Salmonella. ser. Braenderup H9812 was used as the reference standard.

Case Investigation and Entomological Identification
On September 15, 2016, we inspected the patient's home, which consisted of an urban, single-story detached building in insalubrious condition. The outdoor temperature range was 21°C-27°C. One window was ajar by 5cm, allowing insects to enter the home.
We found a substantial number of dead insects in a desiccated state inside the home. We  (Table). One L. sericata ( Figure 1) and one M. domestica were caught alive in the house and kept in separate sterile containers. Both died during transport, but the L. sericata specimen laid eggs inside the container before dying.

Diagnostic Identification
The patient isolate grew on blood agar, yielding colonies with a smooth center and rough edges and displaying α hemolysis. With time, the colonies spread and exhibited a distinct brown pigment underneath. A brown pigment was also exhibited on heart infusion tyrosine agar.
Moderate growth was recovered on MacConkey and phenylethyl alcohol agar. Optimal growth temperatures were 25°C and 35°C in aerobic conditions. No growth was recovered anaerobically or at 42°C. Colonies were strong oxidase positive, catalase positive. Acid was not produced from glucose, D-xylose, mannitol, lactose, sucrose, or maltose. Cells were nonmotile. Gram staining revealed large, pleomorphic, gram-negative rods, including curved and eye-shaped rods, some of which displayed vacuolation.
MALDI-TOF MS recorded a value of 2.40 for W. chitiniclastica, but the identification was presumptive because the MALDI-TOF MS database includes only W. chitiniclastica and the uniqueness of the isolate required confirmatory 16S rRNA sequencing.
Wohlfahrtiimonas sp. was isolated from 2 insect samples on blood agar plates incubated at 25°C.
The first isolate was obtained from the fifth wash and diluted homogenate of the larva that hatched from the eggs laid by the green bottle fly. MALDI-TOF MS recorded a value of 2.41 and 2.43, respectively, using RUO library (unvalidated result). The second isolate was obtained from diluted homogenate of the pooled larvae collected from the carpet with a MALDI-TOF MS score of 2.29 (unvalidated result). These larvae were not taxonomically identified. The negative washes from these same samples demonstrated considerable bacterial overgrowth from which Wohlfahrtiimonas sp. could not be identified. Wohlfahrtiimonas sp. was not isolated from any other samples, including adult flies that emerged from the positive batch of larvae and from the fruit and meat fed to larvae and adult flies.

Isolate Relatedness
All isolates were successfully cut with NotI restriction enzyme. The PFGE pattern indicated that all isolates from flies and fly larvae were indistinguishable and were 74% similar to that of the patient isolate (Technical Appendix Figure 1).