Escherichia marmotae—a Human Pathogen Easily Misidentified as Escherichia coli

ABSTRACT We hereby present the first descriptions of human-invasive infections caused by Escherichia marmotae, a recently described species that encompasses the former “Escherichia cryptic clade V.” We describe four cases, one acute sepsis of unknown origin, one postoperative sepsis after cholecystectomy, one spondylodiscitis, and one upper urinary tract infection. Cases were identified through unsystematic queries in a single clinical lab over 6 months. Through genome sequencing of the causative strains combined with available genomes from elsewhere, we demonstrate Es. marmotae to be a likely ubiquitous species containing genotypic virulence traits associated with Escherichia pathogenicity. The invasive isolates were scattered among isolates from a range of nonhuman sources in the phylogenetic analyses, thus indicating inherent virulence in multiple lineages. Pan genome analyses indicate that Es. marmotae has a large accessory genome and is likely to obtain ecologically advantageous traits, such as genes encoding antimicrobial resistance. Reliable identification might be possible by matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS), but relevant spectra are missing in commercial databases. It can be identified through 16S rRNA gene sequencing. Escherichia marmotae could represent a relatively common human pathogen, and improved diagnostics will provide a better understanding of its clinical importance. IMPORTANCE Escherichia coli is the most common pathogen found in blood cultures and urine and among the most important pathogenic species in the realm of human health. The notion that some of these isolates are not Es. coli but rather another species within the same genus may have implications for what Es. coli constitutes. We only recently have obtained methods to separate the two species, which means that possible differences in important clinical aspects, such as antimicrobial resistance rates, virulence, and phylogenetic structure, may exist. We believe that Es. marmotae as a common pathogen is new merely because we have not looked or bothered to distinguish between the thousands of invasive Escherichia passing through microbiological laboratories each day.

RESULTS & DISCUSSION: L71 & L80: Please, rephrase or define the meaning of "two sets of blood cultures".
L113, "Six of the available E. marmotae whole genomes...": Is this an error? Both in Fig. 2 and in suppl. T1, there are 8 human isolates in addition to the 5 described in this study. Maybe you mean that 6 of those 8 available genomes were originally submitted as E. coli but later reassigned as E. marmotae? Please, clarify. Same in L122, if you consider all human isolates, it would be 13 rather than 11. L133-143: Did authors try to use other AMR databases for comparison purposes? Some of the strains downloaded from other studies do have AMR determinants as defined with other databases besides those mentioned here. Please check in respective publications. Once this is revised, re-evaluate the statement in L111 "The phylogeny was associated with (strain origin and) presence of antimicrobial resistance genes" just in case this wouldn't hold true. L146: how can authors state that these accessory genes are reminiscent from E. coli? Did authors conduct any genome comparative analyses such as mauve? L147: same here. Did authors try to predict mobile genetic elements from the isolates? What is a "visual correlation"? Correlations cannot be "visually" inferred.
CONCLUSIONS: Maybe the authors could briefly discuss on the clinical implications of misidentifying E. marmotae as E. coli.

METHODS:
L279: Can authors provide more details about the microbiological isolation of the E. marmotae strains? Also, how were the isolates stored since isolation and how were they recovered prior to sequencing? L304-305: No need to state here that sequences were deposited in ENA; already in "Data Availability". However, it would be reasonable to briefly describe how authors selected the other E. marmotae strains (source of information, key words, etc).
L309: can authors provide the summary of descriptive statistics regarding the sequencing output and assembly assessment (total reads per sample, Q score coverage, N50, number of contigs, GC content, size of draft genome, etc) to the results section?
Did authors pre-processed raw data before the Bactopia pipeline? Adapters or barcode trimming in case of multiplexing, etc?
I am not aware about Bactopia using abricate to screen for antimicrobial resistance. Did authors use abricate independently from Bactopia? If so, can authors please elaborate more? Which versions of blastn and abricate were used? Did authors screen for antimicrobial point mutations?
Please provide the versions of all bioinformatic tools used and the date of the last update of the databases. L319: use "default settings" rather than "standard settings".
Figure 2: If no country data is available for isolate E1118, authors could consider NA instead of white/blank. SupT2: Are there any differences between those cases designated as "s" and "S" for a particular antimicrobial? If so, please indicate in a footnote. Also, consider providing cut-off values rather than referring to the Nordicast guidelines of September 2021. Cell A13: typo in Trimethoprim REFERENCES: Please thoroughly check for formats, for example italics in species names. Some minor suggestions: -Use the past tense to describe the results.
-Abbreviate to E. marmotae after first description in full.
-16 rRNA should always be followed by "gene" as in "16 rRNA gene". -The authors seem to include in the term "environmental isolates" all non-human isolates. However, I would recommend splitting environmental isolates into isolates of animal and environmental origin.
Reviewer #2 (Comments for the Author): The paper of Sivertsen et al. describes four cases of extra-intestinal infections in human due to Escherichia marmotae. From the complete genome sequences, the authors suggest that E. marmotae has a high virulence potential. The topic is of interest, the methods are accurate but the conclusions are overstated and should be mitigated for the following reasons.
First, most of the cases occurred in immunosuppressed patients. Second, E. marmotae is not so frequent in bloodstream infections as large series for example from England (see Kallonen et al., Genome Res, 2017) and France (see Royer et al., Genom Med 2021) found it at a very low rate (less than 0.1%). Third, the list of the "virulence genes" presented is not convincing: ent, fep and fim operons as well as ompA are found in E. coli K-12, an archetypal non-virulent strain. Fourth, when tested in a mouse model of sepsis representative of the intrinsic virulence of the strains, E. marmotae do not kill mice (see ref 18).

Reviewer #3 (Comments for the Author):
The authors present an interesting study on the prevalence of the recently described species E. marmotae. It describes four different clinical manifestations. Here are some suggestions for an addendum to the manuscript. 1. As mentioned there is only one MALDI-TOF spectrum in the database. Did the authors try to add mean spectra (MSP) oftheir own strains to the database (normally the manufacturers help with that, either with a protocol or with actually providing the spectra from the strains)? It would be interesting what a re-analysis of the E. coli spectra would result in! 2. Apparently E. marmotae was already found in Norwegian sheep. Are there any information on what it causes there? 3. Where all patients asked for their interaction with sheep? 4. The last sentence of the introduction is irritating. It is too common place. Either omit it or explain in detail why you think it will add to the monitoring apart from "we have a new species". From what I read E. marmotae is a gram-negative bacterium with no overtly terrible resistance markers and was successfully treated 4 times?! 5. Did you look into the biochemistry of the bacteria? Would it have been possible to identify it with biochemical means like gramnegative cartridges e.g.? 6. supplementary data on susceptibility testing: what is the difference in meaning of s and S?
Staff Comments:

Preparing Revision Guidelines
To submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex. Go to Author Tasks and click the appropriate manuscript title to begin the revision process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Here are a few examples of required updates that authors must address: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER. • Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file. • Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. • Manuscript: A .DOC version of the revised manuscript • Figures: Editable, high-resolution, individual figure files are required at revision, TIFF or EPS files are preferred For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process. Submissions of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me. If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publication Fees, including supplemental material costs, please visit our website.
Corresponding authors may join or renew ASM membership to obtain discounts on publication fees. Need to upgrade your membership level? Please contact Customer Service at Service@asmusa.org.
Thank you for submitting your paper to Microbiology Spectrum.

Response to reviewers
To the editor We sincerely thank you and the reviewers for a thorough review process of our paper resulting in a more precise and cohesive work. We have implemented most of the suggestions as described in the point-by-point response.
We would also like to mention, that since we finished the first version of this manuscript, we have encountered an additional nine Escherichia marmotae strains, one from blood and eight from urine specimens. We run a medium sized lab, serving approximately 500.000 people. Although these latter strains were discovered too late to be included in this article, they serve to indicate that E. marmotae indeed is present if looked for.

Reviewer comments:
Reviewer #1 (Comments for the Author):

General comments:
This is a very interesting study which describes the potential clinical importance of Escherichia marmotae, as the strains studied were isolated from the lesion sites or blood from diseased patients. Prior to this, the potential pathogenicity or E. marmotae was presumed through genomic prediction and/or in-vitro assays. Nevertheless, the paper has not been organized properly and readers need to go back and forth through the MS to have a clear picture of the study. I understand that Microbiology Spectrum has a format-neutral submission policy, but I believe that including the Methods section before the Results and Discussion might ease comprehension. Furthermore, I consider that some parts need to be elaborated in more detail. For example, the M&M section lacks important details such as isolation methods, bioinformatic tools versions, etc. The results section needs to include bioinformatic summary of sequencing output and assembly performance, etc.
It really helps to have line and page numbers for reviewers to comment on typos or make specific comments, and this observation is clearly stated in the journal submissions guidelines (https://journals.asm.org/format-neutral-submissions). I added the line numbers to help with the reviewing process (L1 -We hereby present....) and will refer my comments to those line numbers.

Response: We thank you for your positive and constructive remarks. To accomodate a better flow in the text, we have now placed the Materials & Methods section between the Introduction and the Results/Discussion section. We have also included isolation methods (routine isolation from the clinical lab) (lines 62-70) and bioinformatic tool versions. We have also provided a summary of the Sequencing output data in the new supplementary table S4. Line numbers have been added -we will refer to line numbers below to track changes in the manuscript text.
Specific comments: ABSTRACT L9, "The invasive isolates were scattered among isolates from a range of non-human sources": The type of analysis used to reach this conclusion should be briefly stated, otherwise it is not understood here. Response: The sentence now reads "The invasive isolates were scattered among isolates from a range of non-human sources in the phylogenetic analyses, thus indicating inherent virulence in multiple lineages." (Lines 16 to 18)
L24: Clade 5 vs. clade V in abstract. Please check throughout the manuscript and use consistently one way or the other, clade V is the term most widely used. Response: Clade designation has been corrected to "Clade V" throughout the manuscript.
L33: authors claim to present 4 case reports. However, the manuscript does not actually fulfil a case report by definition. The journal has reporting guidelines for case reports in case authors need clarification: https://journals.asm.org/reporting-guidelines. Maybe authors can re-phrase "Here, we present four case reports where E. marmotae was isolated as the likely pathogen of invasive human infections" as follows: Here, we report four human clinical cases where E. marmotae was isolated as the likely case of invasive infections. Also, in the sub-heading in L55 "case reports" could be deleted. Response: We thank you for these suggestions, which have both been implemented (Lines 48 and 136-137). L113, "Six of the available E. marmotae whole genomes...": Is this an error? Both in Fig. 2 and in suppl. T1, there are 8 human isolates in addition to the 5 described in this study. Maybe you mean that 6 of those 8 available genomes were originally submitted as E. coli but later reassigned as E. marmotae? Please, clarify. Same in L122, if you consider all human isolates, it would be 13 rather than 11. Response: Thank you for pointing this out. This was an error from our side. The correct number is 8 human strains, including four faecal strains, two from blood, one from urine, and one (UMB2500_14) with unknown isolation source. We have corrected the numbers and modified the sentence regarding species reassignation by NCBI (Lines 187 to 190.

RESULTS
L133-143: Did authors try to use other AMR databases for comparison purposes? Some of the strains downloaded from other studies do have AMR determinants as defined with other databases besides those mentioned here. Please check in respective publications. Once this is revised, re-evaluate the statement in L111 "The phylogeny was associated with (strain origin and) presence of antimicrobial resistance genes" just in case this wouldn't hold true. Response: We expanded our search to include the following AMR databases which were downloaded with abricate: amrfinder, card, megares, argannot, in addition to AMRfinder+. Several of the databases reported the presence of 30-50 resistance associated genes, of which none were deemed as clinically relevant apart from the ones already mentioned in the text. The remaining suggested resistance determinants were chromosomal genes present in most/all E. marmotae genomes, and may represent resistance towards metals and pesticides not relevant in the clinical setting. We have therefore chosen not to include them in the text, but define our focus on line