Characterization of presumptive vancomycin-resistant enterococci recovered during infection control surveillance in Dallas, Texas, USA

Enterococcus faecalis and E. faecium are Gram-positive bacteria that normally inhabit the human gastrointestinal tract. They are also opportunistic pathogens and can cause nosocomial infection outbreaks. To prevent the spread of nosocomial infections, hospitals may rely on screening methods to identify patients colonized with multidrug-resistant organisms including vancomycin-resistant enterococci (VRE). Spectra VRE agar (Remel) contains vancomycin and other medium components that select for VRE and phenotypically differentiate between E. faecalis and E. faecium by colony colour. We obtained 66 de-identified rectal swab cultures on Spectra VRE agar that were obtained during routine patient admission surveillance at a hospital system in Dallas, Texas, USA. We analysed 90 presumptive VRE from 61 of the Spectra VRE agar cultures using molecular and culture methods. Using ddl typing, 55 were found to be E. faecium and 32 were found to be E. faecalis . While most of the E. faecium were positive for the vanA gene by PCR (52 of 55 strains), few of the E. faecalis were positive for either vanA or vanB (five of 32 strains). The 27 E. faecalis vanA- and vanB-negative strains could not be recultured on Spectra VRE agar. Overall, we found that Spectra VRE agar performed robustly for the identification of vancomycin-resistant E. faecium , but presumptive false positives were obtained for vancomycin-resistant E. faecalis .


36
Vancomycin resistant enterococci (VRE), typically Enterococcus faecium and less commonly E. 37 faecalis, are hospital-associated pathogens of significant public health concern. VRE are among

51
In this study, we analyzed presumptive VRE obtained during routine patient admission screening 52 for multidrug-resistant organisms (MDRO) at a Dallas, Texas, hospital. Our goals were to validate 53 the isolates obtained from Spectra VRE agar cultures and to develop a strain collection to be used 54 in future studies of enterococcal biology.
for independent testing of presumptive VRE isolates were purchased from Fisher Scientific.

61
Genomic DNA was isolated using the DNeasy Blood and Tissue kit (Qiagen). Taq polymerase 62 (New England Biolabs) was used for PCR. PCR products used for sequencing analysis were 63 purified using the GeneJET PCR Purification kit (Thermo Fisher Scientific). Sequencing of PCR 64 products was performed at the Massachusetts General Hospital CCBI Core facility.

112
Cas, CRISPR2, and CRISPR3-Cas) (10, 11). Specifically, previously reported primer sets were 113 used to screen for the presence of CRISPR1 cas9 and CRISPR3 cas9 (10). Strains with negative 114 cas9 results were then confirmed to lack the entire CRISPR-Cas loci by using previously reported 115 primer sets that anneal outside the conserved chromosomal locations were the loci occur (10).

116
CRISPR2 arrays were amplified and sequenced as previously described (10,

128
Confirmation of species prediction. Of the 100 stocked strains, 10 could not be revived from 129 freezer stock (Table S2). Of these, 1 was a presumptive E. faecalis by colony color, and 9 were 130 presumptive E. faecium by colony color. The inability to revive these isolates reduced our isolate

138
Of the 32 presumptive E. faecalis VRE that were recovered from freezer stock, 25 were confirmed 139 to be E. faecalis by ddl PCR, and 7 were not confirmed by E. faecalis ddl PCR. Of these 7 strains, 140 6 were found to be E. faecium by ddl PCR, and 1 was found to be a presumptive E. raffinosus or 141 E. gilvus by 16S rRNA gene analysis (99.78% sequence identity). In total, 78% of the colonies 142 predicted to be E. faecalis based on Spectra VRE colony color were confirmed to be E. faecalis 143 using ddl typing (Table S2).

145
Of the 55 presumptive E. faecium VRE that were recovered from freezer stock, 47 were confirmed 146 to be E. faecium by ddl PCR, 1 was a mixed culture of E. faecium and E. faecalis, and 7 were not 147 confirmed by E. faecium ddl PCR. Of these, all 7 were found to be E. faecalis by ddl PCR. In total, 148 85% of the colonies predicted to be E. faecium based on Spectra VRE colony color were 149 confirmed to be E. faecium using ddl typing (Table S2).

151
Of the 3 atypical white colonies, two were found to be E. faecium by ddl PCR, and one was found 152 to be Staphylococcus epidermidis by 16S rRNA gene analysis (99.85% sequence identity) ( Table   153 S2).

155
In summary, of the 90 presumptive VRE isolates analyzed using ddl typing, 55 were found to be 156 E. faecium, 32 were found to be E. faecalis, 2 were found to be neither, and one was a mixed 157 culture.

159
Determination of vancomycin resistance type and re-testing for growth on Spectra VRE.

160
Vancomycin resistance in enterococci is conferred by the synthesis of peptidoglycan precursors 161 for which vancomycin has reduced binding affinity (13). Vancomycin resistance loci can be 162 classified by the gene sequences for the D-alanine-D-lactate ligases, VanA and VanB, which are 163 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

166
Of the 55 ddl-confirmed E. faecium, 52 were positive for vanA and negative for vanB, and 3 were 167 negative for both vanA and vanB (Table S3)  CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020. . https://doi.org/10.1101/2020.10.09.20209569 doi: medRxiv preprint required to determine the mechanism for incongruence between initial clinical culture on Spectra 188 VRE and subsequent molecular typing and Spectra VRE re-culture results.

190
Finally, the presumptive E. raffinosus/E. gilvus isolate was found to be vanA-positive and vanB-191 negative (Table S5). This isolate was assessed for growth on Spectra VRE, and it grew, forming 192 light purple colonies. The broth microdilution vancomycin MIC for this strain is 128 µg/mL.

208
The CRISPR2 arrays of 23 E. faecalis isolates were amplified, sequenced, and analyzed (Table   209 1). We compared the CRISPR2 spacer sequences obtained here with our previous analysis of 210 CRISPR2 loci in 228 E. faecalis genomes, wherein spacers of unique sequence were assigned 211 unique numerical identifiers (12). We identified 12 CRISPR2 types among the 23 isolates. Only 2 212 of the 23 isolates analyzed here possessed novel spacer sequences relative to our previous study 213 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

218
Both of these systems can protect E. faecalis from the acquisition of mobile genetic elements that 219 confer antibiotic resistance (16,17). No isolates possessed CRISPR3-Cas. Two of the 23 isolates 220 possessed CRISPR1-Cas (Table 1)

228
We report the collection and initial analysis of presumptive VRE obtained from hospitalized 229 patients in Dallas, Texas. Our short-term goal (this study) was to validate the isolates obtained 230 from Spectra VRE agar cultures. Our long-term goal is to use genome sequencing approaches to 231 study the phylogeny and antibiotic resistance of these isolates.

233
We determined that most (52 of 55; 94.5%) of ddl-confirmed E. faecium from Spectra VRE rectal CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020.

256
In conclusion, we present a collection of fecal enterococcal isolates from the Dallas, Texas, area.

257
Most VRE identified via Spectra VRE agar were confirmed to be VanA-type E. faecium.

259
Microbiology lab at Methodist Health System, without whom this work would not be possible.

265
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020.

291
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020.

316
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020. . https://doi.org/10.1101/2020

331
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 13, 2020. . https://doi.org/10.1101/2020.10.09.20209569 doi: medRxiv preprint Table S1. Primers used in this study. See materials and methods section for references for primers.

Primer name
Sequence

Vancomycin resistance typing
Species determination Table S2. ddl typing data for presumptive VRE obtained from Spectra VRE plates.

E. faecalis MLST
Notes: * For PCR analysis, an empty cell indicates a negative result for the PCR reaction. NT indicates Not Tested. n/a indicates that the PCR reaction was not performed because the strain could not be recovered from freezer stock. E. faecium not revived from stock n/a n/a n/a n/a 87−2 88 88 navy blue E. faecium not revived from stock n/a n/a n/a n/ E. faecium not revived from stock n/a n/a n/a n/a 144-2 145B 145B navy blue E. faecium not revived from stock n/a n/a n/a n/a 151-1 purple E. faecium not revived from stock n/a n/a n/a n/a 151-1 151-2 navy blue E. faecium not revived from stock n/a n/a n/a n/a  Notes: * For PCR analysis, an empty cell indicates a negative result for the PCR reaction. NT indicates Not Tested. n/a indicates that the PCR reaction was not performed because the strain could not be recovered from freezer stock. ** For Spectra VRE re-testing, a blank cell indicates that the experiment was not performed for that strain. *** Two trials of broth microdilution were performed. A blank cell indicates that the experiment was not performed for that strain.