Antibiotic residues and antibiotic-resistant bacteria detected in milk marketed for human consumption in Kibera, Nairobi.

The use of veterinary antibiotics is largely unregulated in low-income countries. Consequently, food producers rarely observe drug withdrawal periods, contributing to drug residues in food products. Drug residues in milk can cause immunogenic reactions in people, and selectively favor antibiotic-resistant bacteria in unpasteurized products. We quantified the prevalence of antibiotic residues in pasteurized and unpasteurized milk, and antibiotic-resistant bacteria from unpasteurized milk sold within Kibera, an informal settlement in Nairobi, Kenya. Ninety-five milk samples (74 pasteurized and 21 unpasteurized) were collected from shops, street vendors or vending machines, and tested for the presence of β-lactam and tetracycline residues using IDEXX SNAP kits. MacConkey agar without- and with antibiotics (ampicillin, 32 μg/ml; tetracycline, 16 μg/ml) was used to enumerate presumptive E. coli based on colony morphology (colony forming units per ml, CFU/ml). β-lactam and tetracycline residues were found in 7.4% and 3.2% of all milk samples, respectively. Residues were more likely to be present in unpasteurized milk samples (5/21, 23.8%) compared to pasteurized samples (5/75, 6.8%); P = 0.039. Two thirds of unpasteurized samples (14/21, 66.7%) contained detectable numbers of presumptive E. coli (mean 3.5 Log10 CFU/ml) and of these, 92.8% (13/14) were positive for ampicillin- (mean 3.2 Log10 CFU/ml) and 50% (7/14) for tetracycline-resistant E. coli (mean 3.1 Log10 CFU/ml). We found no relationship between the presence of antibiotic residues and the presence of antibiotic-resistant E. coli in unpasteurized milk sold within Kibera (P > 0.2).

Introduction

Antibiotics can be used as “insurance” against livestock losses to disease, challenging the control of antibiotic use and antibiotic residues in food products. This situation is common in many low-income countries where the burden of infectious diseases drives the demand for antibiotics. In these settings, informal food markets are supplied with animals or animal products produced under limited antibiotic regulations, lack of enforcement of drug withdrawal periods, and absence of residue testing programs. For milk, depending on the drug formulation, the recommended withdrawal periods for ampicillin and oxytetracycline are 2 and 4 days [1], respectively. Adherence to these recommendations can be very expensive for persons living at the economic margins.

The presence of antibiotic residues in household and commercially available milk has been reported in East Africa [2-8]. β-lactams and oxytetracyclines, which are commonly used to treat mastitis and livestock respiratory diseases in this region, can trigger hyper-allergenic reactions in people if their residue concentration in consumed milk is sufficient [9-11]; maximum residue limits for amoxicillin and oxytetracycline are 4 ppb and 100 ppb, respectively [12]. Furthermore, for milk that is contaminated with pathogenic bacteria, antibiotic residues can favor the growth of antibiotic-resistant strains that may be directly ingested by the consumer. This is in addition to the risk posed when contaminated milk is exposed to temperatures that are optimal for bacterial growth (37–42°C) [13].

In densely populated urban settlements, poor environmental hygiene and improper milk storage can contribute to milk contamination and proliferation of bacteria within milk, respectively. We estimated the prevalence of antibiotic residues in milk sold in Kibera, an informal settlement located within Nairobi. Kibera is serviced by a formal market which supplies pasteurized milk in sealed plastic bags or through automated vending machines, and by an informal market (small-scale farms) which supplies unpasteurized milk [14]. Given that most households in Kibera have no means to refrigerate milk, they are likely to encounter conditions that are ideal for growth of high-density populations of antibiotic-resistant bacteria in stored milk. To assess the degree to which this problem may arise in communities like Kibera, we collected pasteurized and unpasteurized milk samples from local vendors and tested them for antibiotic residues and bacterial counts (colony forming units per ml; CFU/ml). Bacterial counts were log transformed (base 10).

Materials and methods

Sampling

During September 2015, milk samples were purchased from formal and informal vendors serving Soweto and Gatwekera villages in Kibera. Milk samples were purchased from vendors trading within a 200 m radius from households that were participating in a longitudinal study on antimicrobial resistance. Sample collection occurred over a 2-week period, primarily between 9 and 11 a.m. Once collected, samples were transported on ice to a microbiology laboratory located in Kibera within two hours of collection.

Residue testing

All samples were transferred into sterile 50-ml conical tubes and tested for the presence of β-lactam and tetracycline residues by using IDEXX SNAP kits (IDEXX Laboratories Inc., Maine, USA) following manufacturer instructions [15,16]. These commercial test kits provide rapid presence/absence results at a sensitivity approaching 50 ppb and cross-react with a variety of β-lactam and tetracycline analogues, respectively [17]. Residue testing was completed on the day of sample collection and results were recorded as “positive” or “negative” for presence of the respective antibiotic residue. Milk spiked with tetracycline and ampicillin at 20 μg/ml (20 ppm) was used as the positive control for the SNAP kits.

Bacteriology

The total number of presumptive E. coli and antibiotic-resistant E. coli was also determined for each sample on the day of sample collection. Unpasteurized milk samples were serially diluted (10-fold) with phosphate-buffered saline and 50 μl of the 100 to 10−3 dilution was plated onto MacConkey agar plates with no antibiotic, with ampicillin (32 μg/ml) and with tetracycline (16 μg/ml). The latter two plates selected for ampicillin-resistant (AmpR) or tetracycline-resistant (TetR) E. coli, respectively. Plates were incubated at 37°C for 18–24 hours and presumptive E. coli identified by colony morphology [18]. Plates with 10–100 colonies were selected for colony counts, and the colony-forming units (CFU) per mL recorded for each sample. When fewer than 10 colonies were observed at the 100 dilution, all visible colonies were counted. If colony density greatly exceeded 100 colonies at the 10−3 dilution, the refrigerated left-over sample was diluted further and re-plated. Prior to sample collection, five pasteurized, packaged milk samples were purchased and plated as described, and no bacteria were detected. Consequently, no additional pasteurized milk samples were tested for bacterial growth.

Minimum detection sensitivity

To determine the analytic detection sensitivity of the methods employed in this study, we serially diluted (10-fold) 2.5 x 109 CFU of E. coli with whole pasteurized milk. Four dilutions (100 to 10−3) were plated onto MacConkey agar using the spread plating technique and incubated at 37°C for 18 hours. The minimum number of detectable E. coli was determined from the plate containing the highest milk dilution with visible colonies.

Data analysis

A Wilcoxon rank-sum test was used to compare the number of unpasteurized samples relative to the presence of antibiotic residues and antibiotic-resistant E. coli. To compare the correlation between counts (CFU/ml) of AmpR and TetR E. coli, zero counts were transformed to a random number between 0 and 650 (uniform distribution) to account for detection sensitivity limits, and all values were log-transformed (base 10) before the analysis. Statistics were calculated by using Stata software (ver. 15.1, StatCorp LLC, College Station, TX).

Results

In total, 96 milk samples were collected, 75 of which were purchased from shops (pasteurized) and 21 from mobile vendors (unpasteurized). Pasteurized samples were mainly sold in 250–500 mL sealed plastic packages, while unpasteurized samples were measured in a 250 mL glass and transferred into thin plastic bags (Fig 1). One pasteurized milk sample was excluded from the analysis due to fermentation. Ten of the total 95 milk samples (10.5%) tested positive for antibiotic residues, including seven (7.4%) which were positive for β-lactam residues and three (3.2%) for tetracycline residues; none were positive for both. Residues were more likely to be present in unpasteurized samples (5/21, 23.8%) compared with pasteurized samples (5/74, 6.8%); P = 0.039. Among the 21 unpasteurized samples, 14 samples (66.7%) contained detectable numbers of presumptive E. coli colonies (mean 3.5 Log10 CFU/ml) out of which 92.8% (13/14) and 50% (7/14) were positive for AmpR–(mean 3.2 Log10 CFU/ml) and TetR–E. coli (mean 3.1 Log10 CFU/ml), respectively. No E. coli were recovered from seven of the unpasteurized samples (S1 and S2 Tables).

Fig 1

Examples of milk samples tested; (a) unpackaged/unpasteurized milk and (b) packaged/pasteurized milk (modified version of original packaging; used for illustrative purposes only).

The minimum detection sensitivity of our methods was 2.8 Log10 CFU/ml. Unpasteurized milk samples had E. coli counts ranging from 1.1–7.5 Log10 CFU/ml, while the counts of AmpR E. coli or TetR E. coli ranged from 1.3–6.9 and 2.0–6.7 Log10 CFU/ml, respectively (Fig 2). There was a significant correlation between the number of AmpR and TetR E. coli (r2 = 0.81, P = 0.001). The presence of antibiotic residues was not associated with the number of antibiotic-resistant E. coli (P > 0.5 for all comparisons).

Fig 2

Total, AmpR and TetR E. coli counts (Log10 CFU/mL) for individual unpasteurized milk samples (n = 21).

No bacterial growth was observed across the three media types for samples 2, 7, 17–21. An explanation for the variation shown below the detection limit is provided under the methods section.

Discussion

Unpasteurized milk has a potential role in disseminating both pathogens and antibiotic-resistant bacteria to people through several mechanisms. First, antibiotic-resistant bacteria can be directly acquired through ingestion of milk contaminated with these bacteria [19]. In this study, 67% of unpasteurized milk samples contained E. coli, most of which were resistant to ampicillin and/or tetracycline. Further, the strong correlation between the number of AmpR and TetR E. coli suggests that these were likely multi-drug resistant strains. Consuming just one cup of milk contaminated with 106 antibiotic-resistant bacteria per ml can result in inoculation with over 108 bacteria, a problem that can be prevented through pasteurization. Nevertheless, where storage is poor post-pasteurization hygiene problems (e.g., use of contaminated containers) can lead to re-contamination. Beyond transmission of antibiotic-resistant bacteria, livestock serve as reservoirs for multiple gastrointestinal pathogens of public health concern.22 Ingestion of these pathogens in unpasteurized milk can increase antibiotic use by the consumer, contributing to the emergence of AMR [20-22].

Antibiotic residues in milk can select for antibiotic-resistant bacteria within milk itself, which can then be transmitted directly to people through ingestion. It was unclear from this study if this mechanism is important since we found no correlation between the presence of antibiotic residues and that of antibiotic-resistant E. coli. This may be a limitation of the small sample size considered in this study. Additionally, being that the samples collected in this study were obtained from vendors rather than individual households, they unlikely represent the full range of storage conditions that may exist within Kibera. Fortunately, most of the milk samples (n = 75) collected in this study were pasteurized, perhaps reflecting a higher prevalence of vendors selling pasteurized than unpasteurized milk in Kibera, and a relative affordability of packaged pasteurized milk. Further, the relationship between antibiotic residues and antibiotic-resistant bacteria is dose-dependent [23]. The SNAP tests used in this study simply allowed a “positive” or “negative” classification of samples without quantifying the concentration of residues within milk samples classified as “positive”. Additional work is needed to quantify antibiotic residues within milk to re-examine their relationship with antibiotic-resistant bacteria.

The consumption of antibiotic residues in milk can potentially select for antibiotic-resistant bacteria within a consumer’s gut microflora, a mechanism that has yet to be fully investigated [24]. A study that administered a therapeutic dose (10 mg/kg) of oxytetracycline intramuscularly to groups of cows reported antibiotic residue concentrations in milk as high as 1.92 μg/ml. This concentration falls within a range that can selectively favor antibiotic-resistant E. coli [25] and is likely sufficient to do so after ingestion of contaminated milk depending on rates of absorption and dissipation [26]. In this study, the prevalence of β-lactam (7%) and tetracycline (3%) positive samples was 9 and 4 times higher than the prevalence of residues (0.8%) reported in U.S. milk in 2012 [27]. Residues were observed in both pasteurized and unpasteurized samples, indicating that residue control needs to be focused on all producers, although there is a clear trend towards a lower prevalence of contamination for pasteurized products. In Kenya, boiling is commonly used when consumers prepare milk for consumption, but this practice does not appear to affect presence of residues in milk [28].

Aside from the potential antibiotic-resistance consequences of having antibiotic residues in milk, ingestion of these residues can cause allergic reactions, carcinogenicity, hepatotoxicity, bone marrow toxicity, and reproductive disorders [9,11,28-31]. Limiting antibiotic residues in milk will require a multimodal approach including education of producers, stricter oversight of antibiotic sales and withdrawal times (in milk, ampicillin and oxytetracycline withdrawal times are 2 days and 4 days following injection, respectively [1]), stronger surveillance of residues and AMR in food animal products, and increased awareness and concern of AMR and its pathways of dissemination amongst policy makers and veterinary officials [3,32].

There were several limitations to this study. Firstly, the SNAP tests used to detect antibiotic residues (presence/absence) required a subjective interpretation of the results; we classified samples as negative unless the test was very clearly positive. Secondly, we cannot conclusively tell how the milk affects consumers, who are likely to process milk prior to consumption or consume it later after purchasing, given that milk was tested soon after its purchase from vendors. It is also possible that antibiotic residues may have a greater effect on microbial contaminants the longer the consumer stores milk. We purchased and tested milk directly from the vendor and did not consider consumer behaviors and practices. We also acknowledge that sample collection was opportunistic, rather than random, which could introduce bias to these findings.

Study data for all milk samples collected.

(XLSX)

Click here for additional data file.

Study data for unpasteurized milk samples only.

(XLSX)

Click here for additional data file.

13 Feb 2020

PONE-D-19-29752

Antibiotic Residues and Antibiotic-Resistant Bacteria in Milk Marketed for Human Consumption in Kibera, Nairobi

PLOS ONE

Dear Dr. Omulo,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Mar 29 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

James E. Wells, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

1. We understand that you purchased milk from local markets for this study. In your Methods section, please provide additional details regarding the source of this material. Please provide the geographic coordinates and names of the purchase locations (e.g., shops, vendors), if available, as well as any further details about the purchased items (e.g., quantity, source origin) to ensure reproducibility of the analyses.

2. Thank you for including the following funding information within the acknowledgements section of your manuscript; "This work was funded by the Paul G. Allen School for Global Animal Health, and by the Washington State University College of Veterinary Medicine Summer Research Fellowship Program. "

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

"The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

3.

We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Additional Editor Comments (if provided):

The reviewers observe merit in the study, but there are significant concerns relative to the methods presented and the statistical analyses conducted. The authors need to clearly state how the study was conducted and the analyses that were done. Furthermore, the data presentation could be improved and statistical analyses need to be conducted properly as noted by Reviewer 1.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript by Brown et al presents results of public health importance. They reported antibiotic residues in 10% of ready to consume milk, and majority of unpasteurized milk samples containing high bacterial load including antibiotic resistant bacteria. Although the study is of great interest, the methods are vaguely presented, results are inconsistent, poorly presented and discussed. Please see general and specific comments below.

Abstract

Line 32: please change “hypoallergenic” to “hyperallergenic” as used in the body of the manuscript (line 61). I suggest using the more general term (without the specific type) “allergic”.

Line 43-44 and elsewhere in the text: would you present the CFU/ml values in log10 scale?

Line 43-44 and elsewhere in the text: how these percentages (note that in the text it was mentioned 92% for ampicillin line 131) were calculated? Please clearly state this in the materials and methods. How is it possible to calculate resistant fraction obtained on media supplemented with antibiotics from total bacterial count obtained from plain media? This biases towards the numerator as we already select for the resistant population, simply the methods by which the two values were determined are different. For the stated proportions replica platting would be appropriate. See more comment on this below.

Line 45: How did you arrive at the conclusion of “no evidence”? Please show statistical analysis.

Introduction

Please add literature (perhaps between lines 61 and 62) on the maximum residue level (MRL) allowed in milk for the beta-lactam and oxytetracycline.

Line 71: replace” with” by “which”

Materials and methods

Line 89: Please add manufacturer’s information (company, city, State or Country)

Line 99-100: what dilution(s) were plated? This is important since you used direct plating, too numerous to count indicates high bacterial load; on the other hand, most samples can be BDL if contamination is low which should be expected in milk (and water) and normally filtration is also used for detection.

Line 113: please correct the degree sign as 37°C

Why your analytical sensitivity of 650 CFU/ml is high? How many colonies per plate was your cut-off to count? Did you count 1CFU/plate and consider such a sample as positive, or how did you deal with? Please clarify your methods.

Data analysis

Was the “random number” substitution a form of imputation? Please state. Also discuss if this method would introduce a bias, for not adding it to observations with enumerable values.

What log scale was used? If log10 state so.

Results

Line 129-130: please provide statistical significance for the difference.

Line 130-131: as commented above please clarify how the percentages were calculated.

“Among unpasteurized samples, 14 samples (67%) contained presumptive E. coli colonies out of which 92.8% and 50% were positive for AmpR– and TetR–E. coli, respectively.” This is inconsistent with what was presented in the abstract: Line 42-44: “One third of unpasteurized samples (8/21, 38%) contained detectable numbers of presumptive E. coli (mean 9.2 x 106 CFU/ml) and of these, 87% were positive for ampicillin- (3.7 x 106 CFU/ml) and 50% for tetracycline-resistant E. coli (1.4 x 106/ml).”

Line 130: what does the phrase “contained presumptive” indicate and how was it derived? Is it enumeration positive (i.e. above BDL)? Please clarify.

Please mention that E. coli (wild or resistant strains) were not observed on enumeration plate or are BDL.

Line 133-134: please also include the mean values (preferably log10) for each media type used.

Line 136: given you have only tested two antibiotic classes, I would replace “multidrug-resistance” with co-selection.

Line 137: change “residuals” to residues.

Fig: Please use log10 scale since most studies report that way. In the title please add “by media type” to avoid confusion with drug residues. For the milk samples with no plot observations (example #2, 7, 17-21), were the three types of bacteria (generic, Amp resistant and Tet resistant E. coli) absent across the three media types? Please state. What do ME and MD mean? Please state.

Discussion

Line 146: please insert “unpasteurized” before “milk” and elsewhere in the discussion. Pasteurization removes this issue.

Line 146-148: again, you did not test individual colonies of the total E. coli you obtained on plain media to give you proportion resistant, which would only be possible through replica platting. Rather you isolated the three bacterial strains independently. So, present all your results on sample basis throughout the manuscript.

Line 161: should read “represent”

Reviewer #2: This is a modest but well executed study that examines the presence of antibiotic residues in milk from vendors in Nairobi and seeks to determine if there are any correlations between presence of residues, and presence of resistant bacteria. Essentially looking to inform the question of whether the residues, if present, might enrich for antibiotic resistant bacteria in the milk. There are a number of experimental design limitations in the current study that impact the interpretation of the results. However the authors clearly acknowledge these, and provide informative discussion points so that readers who might not be familiar with the details of antibiotic resistance in this system will be able to understand the limitations. There is a need to better understand antibiotic resistance in African countries, and this study provides information that informs question related to food safety and antibiotic resistance in a culturally relevant setting. The data need to be made available in supporting information.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

1 Apr 2020

Comment 1

We understand that you purchased milk from local markets for this study. In your Methods section, please provide additional details regarding the source of this material.

• Please provide the geographic coordinates and names of the purchase locations (e.g., shops, vendors), if available.

>> We have provided a study description for this community, but we cannot supply the requested geographic coordinates. This is because such coordinates are not available for the mobile vendors who sell raw milk or even for the formal vendors that typically erect a shop that is later moved or abandoned within a matter of months. As such, reporting such coordinates would not help in the replication of this study

• Further details about the purchased items (e.g., quantity, source origin) to ensure reproducibility of the analyses.

>> These details have now been included. We have also included a figure (not Fig 1) to illustrate the packaging of these milk samples. We did not collect data regarding the source origin of the milk samples we purchased.

Comment 2

Thank you for including the following funding information within the acknowledgements section of your manuscript; "This work was funded by the Paul G. Allen School for Global Animal Health, and by the Washington State University College of Veterinary Medicine Summer Research Fellowship Program. "

• Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

>> Funding information deleted from the acknowledgement section

Comment 3

The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

>> The dataset has been provided as part of supporting information.

Comment 4

The reviewers observe merit in the study, but there are significant concerns relative to the methods presented and the statistical analyses conducted.

• The authors need to clearly state how the study was conducted and the analyses that were done.

>> Specific questions about the methods and analyses have been clarified in the revision as indicated under each specific comment below.

• Furthermore, the data presentation could be improved, and statistical analyses need to be conducted properly as noted by Reviewer 1.

>> We believe that we have addressed the clarity issues that were noted by the reviewer. Please note that the reviewer mistakenly concluded that we calculated the proportion of resistant bacteria by making a ratio of the CFU count from agar plates without antibiotic and agar plates with antibiotic. All our comparisons are on a sample basis (presence or absence of detectable antibiotic-resistant bacteria) as the reviewer suggests should have been done. We report total counts from the plates, but not a proportion of the plate without antibiotic. It is unclear to us why this confusion may have occurred.

Comment 5

Reviewer #1: The manuscript by Brown et al presents results of public health importance. They reported antibiotic residues in 10% of ready to consume milk, and majority of unpasteurized milk samples containing high bacterial load including antibiotic resistant bacteria.

• Although the study is of great interest, the methods are vaguely presented, results are inconsistent, poorly presented and discussed. Please see general and specific comments below.

• Abstract

o Line 32: please change “hypoallergenic” to “hyperallergenic” as used in the body of the manuscript (line 61). I suggest using the more general term (without the specific type) “allergic”.

>> we have replaced this term with “immunogenic”

o Line 43-44 and elsewhere in the text: would you present the CFU/ml values in log10 scale?

>> This has now been addressed.

o Line 43-44 and elsewhere in the text: how these percentages (note that in the text it was mentioned 92% for ampicillin line 131) were calculated? Please clearly state this in the materials and methods.

>> The calculation of all percentages has been clarified by indicating the specific counts that form the numerators and denominators.

o How is it possible to calculate resistant fraction obtained on media supplemented with antibiotics from total bacterial count obtained from plain media? This biases towards the numerator as we already select for the resistant population, simply the methods by which the two values were determined are different. For the stated proportions replica platting would be appropriate. See more comment on this below.

>>As above, we are not sure why the reviewer drew this conclusion about the methods. There is no text describing calculation of a resistant fraction in terms of bacterial numbers. All proportion values that we report are relative to the proportion of positive or negative samples, not bacterial counts. For bacterial counts, we calculated total bacterial counts based on the corresponding plates from which these counts were made.

o Line 45: How did you arrive at the conclusion of “no evidence”? Please show statistical analysis.

>> We have revised this statement to read: “We found no relationship between the presence of antibiotic residues and the abundance of antibiotic-resistant E. coli”.

• Introduction

o Please add literature (perhaps between lines 61 and 62) on the maximum residue level (MRL) allowed in milk for the beta-lactam and oxytetracycline.

>> This information has been added

o Line 71: replace” with” by “which”

>> Corrected. Thank you.

• Materials and methods

o Line 89: Please add manufacturer’s information (company, city, State or Country)

>> This information has been added

o Line 99-100: what dilution(s) were plated? This is important since you used direct plating, too numerous to count indicates high bacterial load; on the other hand, most samples can be BDL if contamination is low which should be expected in milk (and water) and normally filtration is also used for detection.

>> The dilutions that were plated have been included in the revision.

o Line 113: please correct the degree sign as 37°C

>> This has been corrected. Thank you.

o Why your analytical sensitivity of 650 CFU/ml is high?

>> We agree that this value is higher than what would be expected with, for example, water, but it is an empirically derived value. We surmise that the count seems high because bacteria clump with milk components and thus a single colony from a milk sample potentially represents more than one bacterium. For the same reason, filtration would not have remedied this issue. Please note that if such clumping occurred, we consider this effect to be similar across all samples

o How many colonies per plate was your cut-off to count?

>> We primarily chose the dilution that gave us at least 10 colonies per plate. Counts below this threshold were only included for undiluted/neat milk samples. These thresholds have been included in the methods.

o Did you count 1CFU/plate and consider such a sample as positive, or how did you deal with? Please clarify your methods.

>> There was only one instance where one colony was observed in plate containing an undiluted milk sample. This plate had 5 colonies in the Mac agar without antibiotics, 1 in the Amp+ plate and no colonies in the Tet+ plate.

• Data analysis

o Was the “random number” substitution a form of imputation? Please state.

>> This was based on a random number drawn from a uniform distribution. This has been added to the manuscript.

o Also discuss if this method would introduce a bias, for not adding it to observations with enumerable values.

>> We used this method to avoid issues that arise when transforming small numbers to a log scale (negative axis values) and because it acknowledges that we should not treat a negative value as truly negative given that there is a minimum analytic sensitivity of this assay. Not adding this value has no consequence to statistical comparisons that are made on a sample basis (positive vs. negative). Failure to add a similar randomly selected number between 0 and 650 would potentially bias the final counts to a lower value, but this is unlikely to be significant as the counts increase on a log scale.

o What log scale was used? If log10 state so.

>> This has now been addressed.

• Results

o Line 129-130: please provide statistical significance for the difference.

>> This information has been added

o Line 130-131: as commented above please clarify how the percentages were calculated.

“Among unpasteurized samples, 14 samples (67%) contained presumptive E. coli colonies out of which 92.8% and 50% were positive for AmpR– and TetR–E. coli, respectively.” This is inconsistent with what was presented in the abstract: Line 42-44: “One third of unpasteurized samples (8/21, 38%) contained detectable numbers of presumptive E. coli (mean 9.2 x 106 CFU/ml) and of these, 87% were positive for ampicillin- (3.7 x 106 CFU/ml) and 50% for tetracycline-resistant E. coli (1.4 x 106/ml).”

>>This was an error that has now been corrected.

o Line 130: what does the phrase “contained presumptive” indicate and how was it derived? Is it enumeration positive (i.e. above BDL)? Please clarify.

>>Please recall that we identified E. coli based only on colony morphology. While this works the majority of the time (cite your JMM paper), it is not perfect. We use “presumptive” as a qualifier to acknowledge this uncertainty.

o Please mention that E. coli (wild or resistant strains) were not observed on enumeration plate or are BDL.

>> This has now been addressed.

o Line 133-134: please also include the mean values (preferably log10) for each media type used.

>> This information has now been added.

o Line 136: given you have only tested two antibiotic classes, I would replace “multidrug-resistance” with co-selection.

>> This has been amended to make it clear that there is a probability that these were multi-drug resistant strains, even though we did not test for other resistance types. Additionally, we have moved this bit of statement to the discussion section to make it clear that this is our interpretation of this finding.

o Line 137: change “residuals” to residues.

>> This has been corrected. Thank you

o Fig: Please use log10 scale since most studies report that way.

>> This has now been addressed.

o In the title please add “by media type” to avoid confusion with drug residues.

>> We have changed values to log scale. We changed the title by inserting the word “detected” to eliminate confusion (although we are not certain that we follow the reviewer’s meaning with respect to the original title).

o For the milk samples with no plot observations (example #2, 7, 17-21), were the three types of bacteria (generic, Amp resistant and Tet resistant E. coli) absent across the three media types? Please state.

>> This information has been included in the figure caption.

o What do ME and MD mean? Please state.

>> This information has been included in the figure caption.

• Discussion

o Line 146: please insert “unpasteurized” before “milk” and elsewhere in the discussion. Pasteurization removes this issue.

>> This has been amended as suggested.

o Line 146-148: again, you did not test individual colonies of the total E. coli you obtained on plain media to give you proportion resistant, which would only be possible through replica platting. Rather you isolated the three bacterial strains independently. So, present all your results on sample basis throughout the manuscript.

>> we are uncertain why the reviewer is making this argument since our results are presented on a sample basis. We report total counts with and without antibiotic in the agar, but we did not calculate a proportion of resistant bacteria based on these values.

o Line 161: should read “represent”

>> This has been corrected. Thank you

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

6 May 2020

Antibiotic residues and antibiotic-resistant bacteria detected in milk marketed for human consumption in Kibera, Nairobi

PONE-D-19-29752R1

Dear Dr. Omulo,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

James E. Wells, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors addressed all of my concerns. Please correct the following few issues.

Line 34 delete the phrase "using IDEXX SNAP kits" from the abstract, not to make it appear endorsement of the kit.

A statement about E. coli, Ampicillin resistant E. coli and Tetracycline resistant E. coli results for the pasteurized milk may need to be mentioned in the abstract since pasteurized milk is also mentioned here.

Line 38: the numerator should be 74 (unpasteurized milk). Please make sure this has been corrected when results were reported as proportions of total samples analyzed (n=95 analyzed and not 96 collected) and when unpasteurized milk samples are considered separately (n=74 analyzed and not 75 collected)

Reviewer #2: This is a modest study that examines the presence of antibiotic residues in milk from vendors in Nairobi and seeks to determine if there are any correlations between presence of residues, and presence of resistant bacteria.There is a need to better understand antibiotic resistance in African countries, and this study provides information that informs question related to food safety and antibiotic resistance in a culturally relevant setting.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

13 May 2020

PONE-D-19-29752R1

Antibiotic residues and antibiotic-resistant bacteria detected in milk marketed for human consumption in Kibera, Nairobi

Dear Dr. Omulo:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. James E. Wells

Academic Editor

PLOS ONE