<xhtml:span xmlns:xhtml="http://www.w3.org/1999/xhtml" xml:lang="en">Comparative analysis of virulence gene profiles of&#160;Escherichia coli&#160;from human and non-human sources&#160;in Rivers State, Nigeria </xhtml:span>

This article has been Reviewed by the following groups

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Traditionally, the presence of virulence features have been thought to be a key factor in differentiating pathogenic from commensal strains. An understanding of virulence potential of Escherichia coli isolates from various sources is essential to help shed light on potential contamination/transmission rates between the various sources. This study was therefore aimed at exploring the occurrence of specific virulence genes and gene profiles associated with Escherichia coli from human and non-human sources in Rivers State, Nigeria. Two hundred samples from human (urine and feces) and non-human (soil and poultry droppings) sources (50 each) were analyzed using standard microbiological procedures. DNA was extracted from isolates presumptively identified as Escherichia coli using PrestoTM Mini gDNA Bacteria-Kit Quick protocol following the manufacturer’s instructions. Isolate identities were confirmed using E. coli specific 16S rRNA primers and confirmed isolates screened for the presence of six virulence genes (Afimbriae binding adhesin (afa), type 1 fimbriae (fimH), P-fimbrial Usher Protein (papC)), iron acquisition systems: aerobactin (aer), Cytotoxic necrotizing factor I (cnf1) and alpha hemolysin (hly). Results showed that all isolates haboured at least one of the tested virulence genes, with fimH (97%) as the most prevalent virulence gene and papC the least commonly occurring (35%). A higher occurrence of virulence genes was noted in non-human isolates though hly and cnf were not detected at all in any of the isolates studied (0%). Ten different profiles were observed with the afaCc-aer-fimH profile the most commonly occurring virulence gene profile in general (33.3%). For non-human isolates however, the aer-afaCc-fimH-papC was the most commonly occurring profile (42.9%). This study shows that the test Escherichia coli from human and non-human sources do not carry distinct virulence gene profiles. Studies on a larger subset of isolates would however be necessary to determine if indeed the virulence genes tested for in this study really cannot be used to tell whether an isolate is from a human source or not in the South-South of Nigeria.

Article activity feed

  1. Figure 2 is not the correct graph. This is now the 4th round of revisions required. Please thoroughly check all aspects of your manuscript before resubmitting- if the manuscript isn't publishable after this round, I will issue a no longer under consideration decision.

  2. I am looking over your most recent version of this manuscript, and I can't see any point by point response to my last round of editorial comments (copy and pasted below for convenience). I can see that you have reconfigured the results presentation to streamline them, and added the P values to table 4, but you haven't made any comment or rebuttal on what was my major concern with the manuscript, the designation of strains as clinical if they were not associated with an infection. All of your results are still presented as clinical vs non-clinical strain comparisons which both myself and the peer review process has picked up as problematic. This needs to be addressed and the manuscript will not be accepted for publication until this is resolved. Please respond to my previous comments (repeated below) with either a convincing rebuttal as to why this is not a concern, or make the necessary changes to the designation of the strains. Editor comments from previous revision that remain unresolved: My main reservation with this manuscript remains the grouping of samples into clinical and non-clinical. I appreciate that you have altered the wording of this to indicate that these designations refer only to where you obtained the samples from. However, as the point of the manuscript is to look at difference in virulence gene profiles between the pathogenic and commensal organisms (“Traditionally, the presence of virulence features have been thought to be a key factor in differentiating pathogenic strains from commensals”), and to see if there is crossover between the pathogenic and commensal/environmental samples, it seems to me that the groupings need to be human isolates from disease associated samples, human isolates that are non-disease associated (human commensals), animal commensals, and soil. Referring to a strain as coming from a clinical source implies it was associated with an infection. In what context were the samples collected at the University of Port Harcourt Teaching Hospital? Which were from patients presenting with urinary/gastro symptoms and can be considered disease associated (pathogenic) vs from people presenting with no relevant symptoms and the strains can therefore be considered commensal? Whilst grouping them into clinical vs non-clinical slightly increases the resolution power I believe it undermines what you are trying to look for. At the very least I think you need to give the breakdown of how many strains within the clinical grouping were associated with an infection and could therefore be considered pathogenic vs how many are clinically sourced but not associated with an infection (and could therefore be grouped as a commensal for comparative purposes). Reviewer 2 asked for the source of the isolates to be broken down in the figures with the use of stacked bar charts, which you have responded to by trying to ignore the source of the samples and lump chicken strains and soil strains in together as non-clinical- is it reasonable to expect that samples from chickens and soils would be the same? And similarly, that human fecal and urinary isolates would be the same? Please go back and reconsider the comments of both reviewers about the designation of strains as clinical if they were not infection associated, and presenting the breakdown of sample source (urine, feces, chicken, soil) as stacked bar plots.

  3. My main reservation with this manuscript remains the grouping of samples into clinical and non-clinical. I appreciate that you have altered the wording of this to indicate that these designations refer only to where you obtained the samples from. However, as the point of the manuscript is to look at difference in virulence gene profiles between the pathogenic and commensal organisms (“Traditionally, the presence of virulence features have been thought to be a key factor in differentiating pathogenic strains from commensals”), and to see if there is crossover between the pathogenic and commensal/environmental samples it seems to me that the groupings need to be human isolates from disease associated samples, human isolates that are non-disease associated (human commensals), animal commensals, and soil. Additionally, referring to a strain as coming from a clinical source implies it was associated with an infection. In what context were the samples collected at the University of Port Harcourt Teaching Hospital? Which were from patients presenting with urinary/gastro symptoms and can be considered disease associated (pathogenic) vs from people presenting with no relevant symptoms and the strains can therefore be considered commensal? Whilst grouping them into clinical vs non-clinical slightly increases the resolution power I believe it undermines what you are trying to look for. At the very least I think you need to give the breakdown of how many strains within the clinical grouping were associated with an infection and could therefore be considered pathogenic vs how many are clinically sourced but not associated with an infection (and could therefore be grouped as a commensal for comparative purposes). Reviewer 2 asked for the source of the isolates to be broken down in the figures with the use of stacked bar charts, which you have responded to by trying to ignore the source of the samples and lump chicken strains and soil strains in together as non-clinical- is it reasonable to expect that samples from chickens and soils would be the same? And similarly, that human fecal and urinary isolates would be the same? Please go back and reconsider the comments of both reviewers about the designation of strains as clinical if they were not infection associated, and presenting the breakdown of sample source (urine, feces, chicken, soil) as stacked bar plots. Data presentation can be streamlined by combining the information in figure 1 and table 3, and combining information in figures 2 and 3 into one figure. Please add P values into table 4 (like in table 3)

  4. Thank you to both reviewers for their assessment of this manuscript. Please can the authors revise their manuscript in line with these comments, paying particular attention to the following; - concerns about the lack of strain typing raised by reviewer 1 - the classification of faecal strains as "clinical" without an associated infection. Are these samples clinical vs non-clinical? Or human vs animal vs environmental? - the suggestions from reviewer 2 for improving the figures I acknowledge concerns about the small sample size and restricted geographical study area. However, the remit of Access Microbiology is sound science, without selection for impact. The authors do need to comment on how a small sample size might limit their conclusions, and place their findings from this geographical area into the larger global context, but there is no need to address comments about why a larger audience would be interested in this study site.

  5. Comments to Author

    This is an interesting study and there's a few revisions to make for each section which I've made below: Abstract and introduction: Overall it is clear from the abstract what you are investigating and the results. Introduction has the relevant references included. Minor comments from both sections are written below. Line 17: What were numbers for clinical samples from urine and faeces? Would be good to include in the abstract as well, especially if the number of samples for the non-clinical has been categorised Line 62: Do you have references for the studies you mentioned? Line 64: Do you have any more recent references? Would be good to include these too Methods: The methods are divided into clear sections, however some sections of the methods are lacking in some detail and clarity. I have highlighted some of these below: Line 90 and line 91: It would be good to have a breakdown the time period of sample collection. Furthermore, it would be good to present this in in a table which details the sample date collection, how many urine samples were collected, how many faeces samples were collected, how many soil samples were collected and how many poultry droppings were collected. At the moment it is unclear how many you samples you have from urine and faeces Line 90: The authors mention samples were tested immediately. More clarity is required for this. What was the process of getting the urine/faecal samples? Were the samples tested immediately upon receipt to the university laboratories? Did the samples all come together in one batch or were they sent overtime? Line 96: Although a reference has been provided for the conventional biochemical tests used, it would be good to emphasise which of the conventional biochemical tests were used? Were all of them used or a selection of these? Line 101: What size DNA ladder was used and what was the voltage and time used for gel electrophoresis? Results: Overall the findings are described appropriately, however there are some key details missing in some of the figures especially. I've also included some further suggestions add to the results section (see below): Line 138, Table 2: It would be good to also have the percentages for urine and faeces as well/include all the sources Line 149, Figure 1: Whilst this is a good figure to have as an overview of the virulence genes in all isolates, regardless of source, it may be worth doing a stack bar chart to represent graphically the percentage of the test virulence genes in all sources e.g. for urine, faeces, poultry and soil Line 151, Table 3: It would also be good to split the clinical column into urine and faeces and the non-clinical column into soil and poultry - as these were mentioned within the paper, it would be worth adding the raw data into this table for these. Some of the p values for hly and cnf are missing. Please put these into the table. Line 158, Figure 2: Please add the labels to the X and Y axis on this bar chart as they are missing Line 159, Figure 3: Please add the labels to the Y axis on this bar chart as they are missing. Additionally, some of the percentages at the top of each bar overlap into the neighbouring bar - maybe a better way to present this would be in a table which could include the combination of virulence genes for clinical and non-clinical. Please also do a graph for urine, faeces, poultry and soil - is there a difference observed within each category in each class? Line 171 - please refer to table 4 Further clarity is required regarding the significance found in the virulence profiles between clinical and non-clinical. It is not clear to the reader whether there was significance or no significance found between virulence in clinical and non-clinical. Please add a statement to address this. Discussion: Appropriate references have been used for discussion with the authors work. Please include the limitations to this study within the discussion as well.

    Please rate the manuscript for methodological rigour

    Satisfactory

    Please rate the quality of the presentation and structure of the manuscript

    Poor

    To what extent are the conclusions supported by the data?

    Partially support

    Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?

    No

    Is there a potential financial or other conflict of interest between yourself and the author(s)?

    No

    If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?

    Yes

  6. Comments to Author

    The manuscript reports on the distribution of selected so-called virulence factors in E. coli strains recovered from clinical or environmental sources in a specific area of Nigeria. The conclusionis, that the detected virulence factors could not discern clinical from non-clinical strains. Comments: 1. The material reported is rather small, i.e., some 68 strains which proved to be E.coli by 16S rRNA studies. The clinical strains were from urine or feces; the latter would normally not be grouped as clinical strains, since they have not caused infection in contrast to strains recovered from urine (one guesses that this must be connected with UTI?). Therefore, major conclusions on distribution of virulence factors cannot be drawn; e.g., the lack of two selected virulence factors may simply be due to too low numbers. 2. It is sperhaps interesting for the authors, but why should a larger audience be interested in the virulence factors in E.coli from a very selected area in Nigeria, or any other small geographical area? E. coli is widespread in the World and we generally expect more or less same distribution of E. coli types ? 3. No typing of the strains was performed, there is therefore a risk, that there could be overrepresentation of certain clones ? Especially in the clinical strains. 4: The reference list is very narrow in citing work from Africa and MIddleeast; i could mention quite a few papers on the same subject from Europe and the US with whole genome sequencing showing more or less the same result, i.e., difficulty in discerning between clinical and environmental strains of E.coli. 5. line 49-50: It is not obvious that all virulence genes are taken up from the environment; some of the meiitoned genes are usually belonging to the core genome in E.coli. 6. Lines 207 ff: The absences of the hlyA gene may simply be due to low number of strains as mentioned above. 7. Linee 213-4: Virulence profiles of E. coli have previously been shown not to be related to e.g., UTI. 8. The referenced study no. 37 concerns diareal E.coli strains which belong to a separate virulence group of E.coli, and cannot be taken into account regarding UPEC virulence factors.

    Please rate the manuscript for methodological rigour

    Poor

    Please rate the quality of the presentation and structure of the manuscript

    Satisfactory

    To what extent are the conclusions supported by the data?

    Not at all

    Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?

    No

    Is there a potential financial or other conflict of interest between yourself and the author(s)?

    No

    If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?

    Yes