History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood.

BACKGROUND: The naïve neonatal gut is sensitive to early life experiences. Events during this critical developmental window may have life-long impacts on the gut microbiota. Two experiences that have been associated with variation in the gut microbiome in infancy are mode of delivery and feeding practices (eg, breastfeeding). It remains unclear whether these early experiences are responsible for microbial differences beyond toddlerhood. AIMS: Our study examined whether mode of delivery and infant feeding practices are associated with differences in the child and adolescent microbiome. DESIGN, SUBJECTS, MEASURES: We used an adoption-sibling design to compare genetically related siblings who were reared together or apart. Gut microbiome samples were collected from 73 children (M = 11 years, SD = 3 years, range = 3-18 years). Parents reported on child breastfeeding history, age, sex, height, and weight. Mode of delivery was collected through medical records and phone interviews.
RESULTS: Negative binomial mixed effects models were used to identify whether mode of delivery and feeding practices were related to differences in phylum and genus-level abundance of bacteria found in the gut of child participants. Covariates included age, sex, and body mass index. Genetic relatedness and rearing environment were accounted for as random effects. We observed a significant association between lack of breastfeeding during infancy and a greater number of the genus Bacteroides in stool in childhood and adolescence.
CONCLUSION: The absence of breastfeeding may impart lasting effects on the gut microbiome well into childhood.

Introduction

The gut microbiome contains a vast collection of microorganisms residing in the human gastrointestinal ecosystem. The microbial composition of the human gut microbiome has been implicated as part of the etiology of both healthy and diseased states [1-3]. In the past decade, research on the interaction between the host and its microbiota has flourished. Individual variation in the human microbiome has been attributed to a variety of factors [4]. Two such factors that have been shown to be salient for predicting gut-microbiome composition during infancy are mode of delivery (MOD), specifically, whether a child was born via vaginal delivery or cesarean section [5-7], and feeding practices (FP; ie, breastfed v. formula fed) [4,8]. Although these factors have been associated with variation in the gut microbiome in infancy and toddlerhood, it is unclear whether associations persist into childhood and adolescence. Whether or not these early life experiences predict microbial composition into childhood may impact the emphasis placed on interventions for cesarean delivered (e.g., vaginal swabbing) [9] and formula fed infants (e.g., probiotic supplementation in formula) [10]. For example, interventions to supplement alterations in the microbial environment occurring from exposure to cesarean delivery and formula feeding may be less relevant to the future composition of the gut microbiome if early life experiences are unrelated to microbial composition later in childhood.

Given that the microbiome interacts with host genetics, especially in the case of gut dysbiosis [11], and is highly influenced by the host environment [12], it is important to account for differences in host genetic background and shared rearing environment. The current adoption design accounts for genetic relatedness among siblings in shared versus separate home environments using a sample of adoptees, their adoptive siblings reared together in the adoptive home, and their biological siblings reared apart from the adoptee in the biological home [13]. Using this sibling-adoption design, we examined the abundance of bacteria at the phylum and genus levels of taxonomy while controlling for known influences on the gut microbiome including body mass index, age, sex, related pairs and households [11,14] to characterize the association between early life experiences and gut microbiome composition during childhood and adolescence.

Methods

Participants

Participants are a subset of children from the Early Growth and Development Study, which is a prospective, longitudinal adoption study [15], and their siblings. Adopted children and their genetically related siblings reared apart or together, as well as their genetically unrelated siblings reared in the same household, were part of the subset who participated (n = 73). Adopted children were placed in the home within approximately 90 days after birth. Fifty-one percent of children were female, and the average age at the time of the stool collection was 11 years old (SD = 3, range = 3–18 years). There were a total of 32 linked sibling constellations with two to six children per constellation. In terms of the rearing environment, 66% (n = 48) of children were reared in an adoptive home, and 34% (n = 25) were reared in the biological home. Table 1 provides information about the rearing environment for the adopted children, their genetically related siblings, unrelated siblings in the adoptive home, and other children in the birth parent home. The BMI in the sample was age corrected using the Centers for Disease Control and Prevention growth charts [16] and was, on average 20.5 (SD = 5.8). Research was approved by the University of Oregon institutional review board (protocol number: 09032013.002). Written consent was obtained for all participants. This study included children under the age of 18. Consent for child participation was obtained from the parent or guardian.

Table 1

Child rearing environment.

	Adoptive home	Biological home	Total
Adopted child	25	0	25
Sibling genetically related to adopted child	17	11	28
Child genetically unrelated to adopted child	6	14	20
Total	48	25

‘Siblings genetically related to the adopted child’ could include siblings with the same biological mother and father or just one biological parent in common. ‘Child genetically unrelated’ are children who may have also been adopted, but did not have the same biological parent as the focal adopted child from the larger study, or could be a biological child of one or both of the adoptive parents.

Microbiome collection and analyses

Samples were collected from July 2016 to September 2017, in the home, using the Omnigene fecal collection kit following kit instructions (Genotek OMR-200) and returned via standard mail. Upon receipt, fecal samples were frozen at -20 degrees Celsius until they could be resuspended in a PBS buffer solution as needed and frozen at -80 degrees Celsius until DNA extraction. Metadata were collected using a survey booklet returned with the samples. The survey booklet included information about the sampling date/time, child age, sex, height, weight, and feeding practices in infancy. MoBio PowerFecal DNA Isolation Kits were used to extract DNA from stool samples following the procedure outlined by the manufacturer. Samples and negative controls were sequenced on the Illumina HiSeq4000 sequencing platform using paired-end 150bp reads with a target sequencing depth of 50k reads per sample. Quality filtering was done in QIIME2 [17] using default settings, and the DAD2 pipeline was used to identify amplicon sequence variants (ASVs) at 100% sequence similarity from the 16S ribosomal RNA variable region V4 [18]. The sequencing depth of final, quality filtered libraries ranged from 39,523 to 84,296 reads with 143 to 469 unique ASVs identified. Alpha diversity metrics (Shannon’s H, Pielou’s evenness index, Faith’s phylogenetic diversity index) were calculated in QIIME2. Data were rarified to 39,500 reads for subsequent analyses comparing phylum and genus level abundances [19]. We observed no effect of transportation and freezing time on variation in alpha diversity (Pearson’s r = -0.04, p = 0.72) or sequencing depth (Pearson’s r = -0.07, p = 0.56).

Feeding practices

Parents were asked to report on whether their child was breastfed or formula fed. If parents indicated that their child was breastfed for any duration of time, they were classified as breastfed, whereas infants who were never breastfed were classified as formula fed. However, we acknowledge that infants who were not breastfed may not have consistently been formula fed. We use the term formula to include the wide variety of formula types, some of which may be created by the infant’s rearing parent, rather than purchased as marketed formula.

Mode of delivery

MOD was collected from all 25 adoptees and from 18 biological siblings reared in the biological home from medical records. Medical records were missing for 30 children, and was thus collected by phone interview from the parent. These data collection efforts were nested within data collection efforts for the larger study.

Covariates

Body mass index (BMI), age, and biological sex were collected in the booklet at the time of microbiome sample collection (mother report). For BMI, of the children in our study, 57% fell in the normal range (5th percentile to 85th percentile), 19% had over weight (85th to 95th percentile), 3% had underweight (< 5th percentile), and 21% had obesity (> 95th percentile). For analyses, BMI was computed using an age corrected z-score calculated based on the publicly available normalization procedures of the CDC [16]. Age was rounded to the nearest whole year, and sex was dichotomized assigning males as the reference group.

Analyses

Microbiome count data have distinct properties such as zero-inflation and over-dispersion [20]. Thus, mixture models with a negative binomial or Poisson error distribution were considered as possible analytic approaches for examining associations between MOD and FP and microbial abundance at various levels of taxonomy. Given that the Poisson distribution assumes the mean and variance are equal, we analyzed differences between the means and variances for each taxa and consistently found the variance was at least two-fold greater than the mean for each taxa (average variance to mean ratio = 515.12), making the negative binomial distribution more appropriate in order to handle over-dispersion in the data and ensure proper parameter estimation [20]. Moreover, negative binomial mixture models are appropriate for microbiome data given that the microbial data in this sample are nested within the host and individuals are nested within related pairs and within households [20]. Our model included a nested random effect allowing the intercept to vary among home and family and within home [21] to account for differences in bacterial abundance due to genetic relatedness and rearing environment. Additionally, we tested whether host gut microbiome alpha diversity (mean species diversity) was associated with MOD and FP. Shannon, Pielou, and Faith are continuous indices of alpha diversity which account for relative abundance and sequencing depth in different ways [22]. We examined whether MOD and FP were associated with these three metrics using a general linear model controlling for age, sex, BMI, and genetic relatedness and rearing environment. To assess whether the total count of ASVs present in each sample was related to our variables of interest, we performed a Poisson regression. We used permutational multivariate analysis of variance (PERMANOVA) to estimate the relative contributions of MOD and FP to beta diversity (pairwise differences in microbiome diversity estimated using Bray Curtis dissimilarity). All analyses were completed in R v3.4.3. The package glmmADMB [23,24] was used for all mixture models. The package vegan was used for PERMANOVA analysis [25].

Results

In our study’s subsample of adoptees and their siblings reared in the adoptive home and siblings reared in the biological home, 69% were delivered vaginally (n = 50), and 21% were breastfed (n = 15; see Table 2). We identified 11 phyla and 96 genera within the sample of 73 participants. Relative abundance of the most common phyla and genera across individuals within each MOD and FP group is depicted in Figs 1,2, 3 and 4. Results from negative binomial mixture models suggest that mode of delivery was unrelated to the presence of taxa at the phylum and genus levels after accounting for false discovery rate [26]. However, FP was significantly associated with abundance of the genus Bacteroides, as shown in Table 3 and Fig 3. Specifically, when children were breastfed as infants, the expected counts of the Bacteroides in the child’s gut microbiome were 0.46 fold those of children who were never breastfed (p < .0001). A box-plot with mean differences between breastfed and formula fed children on Bacteroides abundance is provided in Fig 5. There were no associations between any measures of alpha diversity, beta diversity, or total ASV count and MOD and FP.

Table 2

Mode of delivery and feeding practices by rearing environment.

	Adoptive home	Biological home	Total
Delivery type
Cesarean section	19% (14)	13% (9)	32% (23)
Vaginal delivery	46% (34)	22% (16)	68% (50)
Feeding practice
Formula fed	55% (40)	24% (18)	79% (58)
Breastfed	11% (8)	10% (7)	21% (15)

Number of children noted in parentheses.

Fig 1

Relative abundance of the most common phyla by feeding practice (FP).

Each vertical bar represents an individual. FP group is designated across the top of the figure. Samples are ordered within each group by descending Bacteroidetes abundance.

Fig 2

Relative abundance of the most common genera by feeding practice (FP).

Each vertical bar represents an individual. FP group is designated across the top of the figure. Samples are ordered within each group by descending Bacteroides abundance.

Fig 3

Relative abundance of the most common phyla by mode of delivery (MOD).

Each vertical bar represents an individual. MOD group is designated across the top of the figure. Samples are ordered within each group by descending Bacteroidetes.

Fig 4

Relative abundance of the most common genera by mode of delivery (MOD).

Each vertical bar represents an individual. MOD group is designated across the top of the figure. Samples are ordered within each group by descending Bacteroides abundance.

Table 3

Negative binomial mixture model of mode of delivery and breastfeeding for Bacteroides.

Fixed effects	Estimate	SE	z	p
Age	1.01	.02	.83	.41
Sex	0.87	.09	-1.53	.13
BMI	0.98	.05	-0.36	.72
Vaginal birth	0.87	.11	-1.19	.24
Breastfed	0.46	.20	-3.91	< .0001
Random effects	Variance	SD
Environment	.22	.47
Genetic relatedness	.22	.47

Fig 5

Boxplot of mean differences between formula feeding and breastfeeding for the genus Bacteroides.

Discussion

Using negative binomial mixture models to account for over-dispersion, genetic relatedness, and the shared rearing environment, we did not find any differences in child microbiomes associated with breastfeeding and vaginal delivery at the genus and phylum levels, except for the genus Bacteroides. Specifically, we identified a greater abundance of Bacteroides in the gut microbiomes of children who were not breastfed as infants compared to infants who were breastfed. This finding suggests that the direct effects of FP and MOD on the gut microbiome may become obsolete in childhood except for the influence of FP on the relative abundance of Bacteroides. This finding is meaningful given that Bacteroides is a predominate genus in the human gut microbiome (in fact, Bacteroides live and grow exclusively in the mammalian digestive tract) and are a known driver of gut maturation and diversity [27,28]. Moreover, Bacteroides have been shown to improve their host’s ability to fight infections by enteric pathogens and more generally improve immune tolerance [29,30]. However, Bacteroides have also been associated with problematic outcomes in the host.

Implications

Most research on the association between MOD and microbiome composition has been completed within the first three years of life, likely because it has been proposed that the gut microbiome converges to an adult-like state between the ages of 3 and 5 and remains stable in later life [31]. Previous research found that the microbiome maintained individual uniqueness but converged towards a relatively stable, adult-like trajectory after the age of 3 [28,32]. However, recent studies in older children suggest that the microbiome changes throughout childhood to support shifting developmental needs [28]. Our study suggests that changes in the microbiome into adolescence, may erase many of the effects of early life experiences on microbiome composition.

Feeding practices

Prior studies of infants and toddlers have found that breastfeeding is associated with greater abundances of the genera Bifidobacteria, Streptococcus, Bacteroides, Firmicutes, Lactobacilli-EnterocoiI [10,33-35] and the phyla Firmicutes, Bacteroidetes, and Actinobaceria [36]. Formula feeding has been associated with a greater abundances of Clostridium, Streptococcus, Enterococcus, and Veillonella [10,33,34] and the phylum Proteobaceria [36]. Thus, it was surprising that FP were related only to the abundance of Bacteroides in our sample (i.e., children who were formula fed harbored a greater abundance of Bacteroides in childhood compared to their breastfed counterparts). Bacteroides a predominate genus in the human microbiome, is a known driver of gut maturation and diversity [27,28], and has been shown to improve host resistance to pathogen colonization and improves human immune tolerance [29,30]. It should be noted that Bacteroides may, in some cases, cause harm to the host. For example, in the presence of inflammation, an abundance of specific Bacteroides species can enhance the pathogenicity of enterohemorrahagic E. Coli during inflammation [37,38] greater abundance of Bacteroides has also been linked to the prevalence of type 1 diabetes [39], and Bacteroides may cause infection if they escapes from the gut, potentially leading to septicemia [27].

Since some research suggests that Bacteroides are more abundant in breastfed infants [34], we might have expected that children with a history of breastfeeding rather than formula feeding would have harbored a greater abundance of Bacteroides. Instead we found that Bacteroides were more abundant for children with a history of formula feeding. This may be a result of early exposure to foods other than breast milk for infants who were formula fed. For example, Bacteroides are found in higher abundance in the gut microbiome of as children begin consuming solid foods [40], and higher abundance of Bacteroides is associated with a more mature (i.e., more adult-like) gut microbiome [28]. More research is needed to understand the associations between a higher abundance of Bacteroides and child outcomes, especially as is related to diabetes [39].

Mode of delivery

It is surprising that there was no evidence of differences in diversity or relative abundance of taxa in the gut microbiome associated with MOD in the current study, especially in light of differences observed in prior studies in the gut microbiota of infants and young children who were delivered by cesarean section and infants birthed vaginally [28]. Studies comparing the composition of the microbiome in infancy relative to MOD have revealed higher abundance of several genera in infants born vaginally compared to infants born by cesarean section [4]. For example, the microbes belonging to the Bacteroides, Bifidobacterium, Lactobacillus, Prevotella, and Snethia genera have all been found to be more abundant in infants delivered vaginally [5-7,41-44], whereas microbes belonging to the Blautia, Prevotella, Staphylococcus, Corynebacterium, Propionibacterium, and Clostridium genera are more abundant in the gut of infants delivered by cesarean section [7,41,42,44,45]. Of particular importance is the microbial presence and abundance of Clostridium difficile in cesarean section born infants, which is associated with health challenges including diarrhea and food poisoning [46]. Higher abundance of Clostridium has also been found in 7-year old children with a history of cesarean birth [47]. Interestingly, this same study did not find differences in microbiome diversity, the presence of Bacteroides, Bifidobacterium, or Lactobacillus. The absence of an association between MOD and the child gut microbiome in the current sample highlights the importance of observing the microbiome into later childhood to identify the persistence of microbial alterations as a result of differences in MOD. Our findings corroborate recent evidence suggesting that altering the microbiome of cesarean delivered infants to resemble vaginally delivered infants may not be a useful mechanism for improving individual host fitness [48]. For instance, research finding that cesarean section is associated with a higher incidence of problematic outcomes, such autoimmune diseases [6], has given rise to the practice of vaginal seeding for infants born by cesarean section [9]. Recent opinion has challenged this practice based on the dearth of well-designed studies on the association between cesarean section, microbiome composition and disease outcomes [48].

Additional considerations

Previous research reported effects of MOD and FP on the development of the infant microbiome; our research suggests that most of these effects may not be associated with the gut microbiome in childhood and adolescence. Additionally, potential environmental confounds may exist which are more salient predictors of microbial composition than early life experiences, such as specific aspects of the rearing environment which we have not accounted for in our study. It may also be that long term impacts of FP and MOD on gut microbial composition vary by geographical location, and samples in the current study were collected across a broad geographic range across the United States, such that comparisons of specific geographical regions were not feasible. Microbiota vary across geographic locations as a function of diet, cultural practices, and living situations [14,49]. Longitudinal studies of microbial composition in response to FP and MOD are needed in order to assess both the short- and long-term effects of early life experiences on the child and adolescent gut microbiome.

Strengths and limitations

Our study aimed to address the dearth of information on the effects of FP and MOD on gut microbiome composition in later childhood. There are several strengths to this approach, including the use of negative binomial mixture models to account for over-dispersion and genetic relatedness among siblings and home rearing environment. This is the first study, to our knowledge, to apply these techniques using a sibling-adoption design to account for rearing environment and genetic relatedness. Our study is also one of few that looks beyond the first four years of life to assess associations between FP and MOD and gut microbial composition [47,50,51] Some limitations to consider are that the current study had diminishing power to detect statistically significant associations in a sample size of n = 73. Larger studies must be completed in order to confirm our results, although our sample size was sufficient to detect an effect size of 0.4 or larger. Second, because of the use of an adoption sample, a lower proportion of children were ever breastfed compared to the general population of the United States (79%) [52], which increased variability in FP but limited our ability to explore differences in duration and exclusivity of breastfeeding and may limit generalizability and limited our ability to examine duration of breastfeeding and the use of breastmilk and formula simultaneously. Moreover, our data did not capture whether breastmilk came from other sources, such as friends, family, or community support breastmilk networks. Retrospective reports of feeding practices may also be inaccurate. Additionally, this study was unable to control for known influences of the gut microbiome, such as diet and antibiotic use [53,54]. Thus, the exclusion of these variables from our analytic models could have affected the results.

Conclusion

This work highlights that the effects of two early life experiences (MOD and FP), while important, do not necessarily impact the long-term development of the child gut microbiome. However, early feeding was related to the abundance of the genus Bacteroides in later childhood and adolescence, a known marker of gut maturity and diversity that provides benefits to the human immune system [27] but may also cause problems in the host [37,38,39]. This finding implies that early feeding may impart lasting effects on the gut microbiome well into childhood.

Child characteristics and microbiome data.

(CSV)

Click here for additional data file.

6 Mar 2020

PONE-D-20-02497

History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood

PLOS ONE

Dear Dr Leve,

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 Apr 20 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,

Corrie Whisner

Academic Editor

PLOS ONE

Journal Requirements:

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

1. 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.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified, if your study included minors under age 18, whether you obtained consent from parents or guardians.

3. To comply with PLOS ONE submission guidelines please deposit your sequencing data in a publicly available repository (you can find a list of repositories in the link here : https://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories).

4. Please include in your Methods section the date ranges over which you recruited participants to this study.

5. Please ensure that you refer to Figure 4 and 5 in your text as, if accepted, production will need this reference to link the reader to the figure.

6. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[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: Partly

Reviewer #2: Yes

**********

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

Reviewer #1: I Don't Know

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: Yes

Reviewer #2: Yes

**********

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: This paper adds to the literature on the effects of delivery mode and infant feeding on the gut microbiota by examining whether differences in diversity and bacterial composition can be seen in children aged 3-18 years by delivery and breast v formula feeding. While this is an important question, the relatively small sample size, with a wide range of child ages 3-18, and the lack of other control variables limit the generalizability of the study. Further, the authors make several strong claims that do not appear to be supported by the data and/or results.

First, the authors claim that this is the first paper to explore whether the effects of these early childhood exposures persist past age 4 (for example, pg 3 lines 44-45, pg 13, line 271). At least three other papers have examined these associations after age 4: Salminen S, Gibson GR, McCartney AL, et a influence of mode of delivery on gut microbiota composition in seven year old children. Gut 2004;53:1388-1389; Thompson, A. L., Houck, K. M., & Jahnke, J. R. (2019). Pathways linking caesarean delivery to early health in a dual burden context: Immune development and the gut microbiome in infants and children from Galápagos, Ecuador. American Journal of Human Biology, 31(2), e23219; and Nagpal R, Yamashiro Y: Gut Microbiota Composition in Healthy Japanese Infants and Young Adults Born by C-Section. Ann Nutr Metab 2018;73(suppl 3):4-11. doi: 10.1159/000490841

Second, the age range of the participants is quite wide and, while the authors control for age in their multilevel models of the factors associated with Bacteroides abundance it would be helpful to see whether diversity or any other measures differ by age group. Such an analysis would provide more support for their conclusion on page 10 that: “our study suggests that changes in the microbiome occur well into adolescence…” As currently described, this cross-sectional analysis doesn’t provide evidence that the microbiome changes in childhood and adolescence, just that there are differences in the colonization that may be associated with infant feeding practices. A similar wording issue is on pg 12, line 251 where the authors describe the effects as “diminishing over time.”

Similarly, the authors should be more explicit about the limitations/potential bias in their sample data. How likely are maternal reports of child size, breastfeeding and birth type to be accurate given the mixed history of the sample (some biological children, some adoptive children with likely varying ages at adoption)? Further, how likely are other variables to have influenced their findings (i.e. timing of complementary feeding, antibiotic use, etc) given the lack of other infant or childhood exposures?

The authors don’t appear to make use of the adoption-sibling design of the study other than controlling for clustering by household and differing genetic relatedness. This seems like a lost opportunity to make this study unique in the literature by providing an examination of how household environment may interact with early birth/feeding exposures to shape child microbiota.

A few more minor concerns:

Pg 9, line 185: “latent” seems to be the wrong word

Pg 13, lines 276: It is not clear that the authors are referring to breastfeeding initiation in the US (79%) compared to other measures (i.e. duration at 3 months, ebf at 6 months). The fact that the authors only collected “ever breastfed” appears to be the more limiting factor for looking at ebf or duration rather than the lower breastfeeding initiation prevalence than the US population.

Reviewer #2: There are several areas that can be improved with additional clarification.

1. Do not assume that the children that were not breastfed were formula fed, as that gives the impression of commercial formula use. Parents may have used other types of liquids to feed their children, or to have created their own types of 'baby milk'. I recommend terms such as breastfed and non-breastfed.

2. The discussion of the participants is confusing. A table displaying the children who were adopted or biological and reared with birth parents or adoptive parents would be easier to understand than the text. The term 'relatedness' is also not clarified as it could mean 'related through shared genetics' or 'related through being members of the same family'.

3. BMI needs to be stated as BMI percentile. A BMI of 20 does not mean anything to the reader since the children ranged in age from 3 to 18. It would be more helpful to know if the BMI% is at 50th% or 90th%.

4. Knowing the child's typical diet at the age when the fecal sample was collected would be useful. A food frequency questionnaire would have been a beneficial instrument for knowing if the child had a predominantly meat heavy or vegetable heavy diet, which can affect the gut microbiome.

5. It would be helpful to know the age of adoption. Some adoptees may have been initially breastfed. Cannot assume that if adopted, must have not been breastfed.

6. Acknowledge in additional considerations that knowing the current diet (important co-variate) could also contribute to analysis of these results.

**********

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.

9 Apr 2020

Dear Dr. Whisner,

Thank you for the opportunity to revise our manuscript entitled, “History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood”. We have reviewed the comments raised by you and reviewers and have provided our responses along with a tracked-changes version of the manuscript. Page numbers correspond with the tracked changes version of the manuscript.

Academic Editor

1. 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.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

We have reviewed our manuscript and file names to ensure consistency with PLOS ONE’s style requirements.

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified, if your study included minors under age 18, whether you obtained consent from parents or guardians.

We have provided additional details about participant consent on page 5.

3. To comply with PLOS ONE submission guidelines please deposit your sequencing data in a publicly available repository (you can find a list of repositories in the link here : https://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories).

In line with PLOS ONE submission guidelines, data are provided in the supporting information files.

4. Please include in your Methods section the date ranges over which you recruited participants to this study.

We have included the date ranges for data collection on page 5 (from July 2016 to September 2017).

5. Please ensure that you refer to Figure 4 and 5 in your text as, if accepted, production will need this reference to link the reader to the figure.

Figures 4 and 5 are now referenced in text on pages 7 and 8.

6. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

We have provided a caption for our supporting information after the acknowledgements on page 14. There is one supporting information file which is the dataset for this study.

Reviewer #1

1. First, the authors claim that this is the first paper to explore whether the effects of these early childhood exposures persist past age 4 (for example, pg 3 lines 44-45, pg 13, line 271). At least three other papers have examined these associations after age 4: Salminen S, Gibson GR, McCartney AL, et a influence of mode of delivery on gut microbiota composition in seven year old children. Gut 2004;53:1388-1389; Thompson, A. L., Houck, K. M., & Jahnke, J. R. (2019). Pathways linking caesarean delivery to early health in a dual burden context: Immune development and the gut microbiome in infants and children from Galápagos, Ecuador. American Journal of Human Biology, 31(2), e23219; and Nagpal R, Yamashiro Y: Gut Microbiota Composition in Healthy Japanese Infants and Young Adults Born by C-Section. Ann Nutr Metab 2018;73(suppl 3):4-11. doi: 10.1159/000490841

We have clarified that we do not believe we are the first to examine associations between early life experiences and child and adolescent microbiome composition and have included the appropriate citations on pages 3 and 13. We did not change text on page 13 because we believe this is the first published paper to utilize negative binomial mixture models and an adoption design to examine associations between early life experiences and the child/adolescent gut microbiome. We have added the citations the reviewer provided on page 13.

2. The age range of the participants is quite wide and, while the authors control for age in their multilevel models of the factors associated with Bacteroides abundance it would be helpful to see whether diversity or any other measures differ by age group. Such an analysis would provide more support for their conclusion on page 10 that: "our study suggests that changes in the microbiome occur well into adolescence..." As currently described, this cross-sectional analysis doesn't provide evidence that the microbiome changes in childhood and adolescence, just that there are differences in the colonization that may be associated with infant feeding practices. A similar wording issue is on pg 12, line 251 where the authors describe the effects as "diminishing over time."

We agree with the reviewer’s comment that our study could not address changes over time. Since associating age with the gut microbiome composition was not the focus of this study, we have removed language on pages 10 and 12 that infer change over time.

3. Similarly, the authors should be more explicit about the limitations/potential bias in their sample data. How likely are maternal reports of child size, breastfeeding and birth type to be accurate given the mixed history of the sample (some biological children, some adoptive children with likely varying ages at adoption)? Further, how likely are other variables to have influenced their findings (i.e. timing of complementary feeding, antibiotic use, etc) given the lack of other infant or childhood exposures?

We have provided additional information about the current study limitations on page 14. Text appears as follows:

“Retrospective reports of feeding practices may also be inaccurate. Additionally, this study was unable to control for known influences of the gut microbiome, such as diet and antibiotic use (53,54). Thus, the exclusion of these variables from our analytic models could have affected the results.”

4. The authors don't appear to make use of the adoption-sibling design of the study other than controlling for clustering by household and differing genetic relatedness. This seems like a lost opportunity to make this study unique in the literature by providing an examination of how household environment may interact with early birth/feeding exposures to shape child microbiota.

Since very few, if any, adoptees were breastfed, we do not have the equal representation needed in a crossed factorial design and thus are unable to run statistical tests for moderation by household environment. However, we would like to point the reviewer to Tavalire et al. (citation 13) which is currently under review at Science Advances and utilizes the adoption design to directly measure the effects of genetics versus the environment on microbiome composition.

A few more minor concerns:

Pg 9, line 185: "latent" seems to be the wrong word

Pg 13, lines 276: It is not clear that the authors are referring to breastfeeding initiation in the US (79%) compared to other measures (i.e. duration at 3 months, ebf at 6 months). The fact that the authors only collected "ever breastfed" appears to be the more limiting factor for looking at ebf or duration rather than the lower breastfeeding initiation prevalence than the US population.

We have addressed the minor concern highlighted by reviewer 1 on page 9. As noted on page 6, parents indicated the duration of time their child was breastfed but the responses were collapsed due to the low breastfeeding prevalence in our sample. This is also a limitation of our study that has been addressed on page 14 with the following text:

“Second, because of the use of an adoption sample, a lower proportion of children were breastfed compared to the general population of the United States (79%) (52), which increased variability in FP but limited our ability to explore differences in duration and exclusivity of breastfeeding and may limit generalizability and limited our ability to examine duration of breastfeeding and the use of breastmilk and formula simultaneously.”

Reviewer #2

1. Do not assume that the children that were not breastfed were formula fed, as that gives the impression of commercial formula use. Parents may have used other types of liquids to feed their children, or to have created their own types of 'baby milk'. I recommend terms such as breastfed and non-breastfed.

We acknowledge that infants who were not breastfed may not have been formula fed on page 5. We retained the term formula fed for consistency throughout the manuscript because using non-breastfed made some text more difficult to interpret. However, we clarify the definition of formula to include formula that may have been created or formulated by the infant’s rearing parent, rather than purchased as a marketed formula, to address the reviewer’s concerns. We used the following text:

“Parents were asked to report on whether their child was breastfed or formula fed. If parents indicated that their child was breastfed for any duration of time, they were classified as breastfed, whereas infants who were never breastfed were classified as formula fed. However, we acknowledge that infants who were not breastfed may not have consistently been formula fed. We use the term formula to include the wide variety of formula types, some of which may be created by the infant’s rearing parent, rather than purchased as marketed formula.”

2. The discussion of the participants is confusing. A table displaying the children who were adopted or biological and reared with birth parents or adoptive parents would be easier to understand than the text. The term 'relatedness' is also not clarified as it could mean 'related through shared genetics' or 'related through being members of the same family'.

We have provided a table on page 5 to clarify each group of children. We have clarified the term “relatedness” to say, “genetic relatedness”, throughout the manuscript.

3. BMI needs to be stated as BMI percentile. A BMI of 20 does not mean anything to the reader since the children ranged in age from 3 to 18. It would be more helpful to know if the BMI% is at 50th% or 90th%.

We have clarified that BMI was corrected for age using the CDC growth charts on page 4. The use of the CDC age-corrected growth charts makes the BMI directly comparable between ages.

4. Knowing the child's typical diet at the age when the fecal sample was collected would be useful. A food frequency questionnaire would have been a beneficial instrument for knowing if the child had a predominantly meat heavy or vegetable heavy diet, which can affect the gut microbiome.

We agree that child’s diet is an important predictor of microbiome composition and have noted this as a limitation of our study on page 14.

5. It would be helpful to know the age of adoption. Some adoptees may have been initially breastfed. Cannot assume that if adopted, must have not been breastfed.

Adoptions occurred, on average 3 days after birth. We have clarified this information on page 4. We agree that we cannot assume that children who were adopted were not breastfed and have clarified this on page 13.

6. Acknowledge in additional considerations that knowing the current diet (important co-variate) could also contribute to analysis of these results.

We have acknowledged this limitation on page 14.

Thank you again for the opportunity to revise our manuscript.

Sincerely,

Camille C. Cioffi

Submitted filename: Response to Reviewers[1].docx

Click here for additional data file.

27 May 2020

PONE-D-20-02497R1

History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood

PLOS ONE

Dear Dr. Leve,

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.

In revising your manuscript, please pay special attention to the comments provided by Reviewer 2, especially the need to present body mass index as a percentile instead of in kg/m^2. Please see the bulleted list of important changes I noted when reviewing the paper and the comments from reviewers:

For children, the World Health Organization suggests that this approach to reporting child size and body weight status are more accurate given that they are in a period of more intense growth. This is a best practice in both clinical and research settings, and should therefore be carried out in this paper before it will be deemed acceptable for publication.

The paper suggests that formula feeding is fine and may be beneficial given the promotion of gut maturation ia Bacteroides. This message needs to be softened throughout the manuscript as many health agencies support breastfeeding as the gold standard for gut and systemic child health. Please revise the manuscript to ensure that formula feeding is not viewed as equal or superior to breastfeeding.

Please submit your revised manuscript by Jul 11 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're 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.

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). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

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

We look forward to receiving your revised manuscript.

Kind regards,

Corrie Whisner

Academic Editor

PLOS ONE

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

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: (No Response)

**********

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: Partly

**********

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: (No Response)

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: (No Response)

Reviewer #2: I read your rationale why you used BMI only. But it lacks clarity on the predominant body size of the children. Please use BMI% for the BMI of 20 to make sense to readers. I know it's age corrected but it still doesn't tell me of the child's weight status; normal (5% to 85%) over/under weight, (85%-<5th%) obese >95%.

The child's rearing environment is confusing, especially Table 1. Does the box 'sibling genetically related and adoptive home mean that of the 25 adopted children, there were 17 genetically sibs who were reared together in an adoptive home? The the box 'child genetically unrelated to adopted child and biological home mean that in the biological home, 14 of children were there who were not related to the adoptees? For example, step-siblings and not half-siblings?

For feeding practices, did you also measure if the adoptive parents used breast milk from a milk bank or from friends/family/community support breast milk networks? That practice is becoming more common and could have occurred in the younger children in your sample.

Was the delivery mode measured by questions from adoptive or birth mother when records were not available?

In the results section, was the 21% breastfed combined from both genetically related and non-genetically related sibs?

On line 195, add 'gut' between human and microbiome.

Line 222 Mention also that bacteriodes can also cause serious infections if escape outside the gut, potentially leading to septicemia too.

Line 282 The percentage of breastfed varies by age of infant. Do you mean ever breastfed, exclusively breastfed, or breastfed to 6 months? more specificity needed.

Line 292 remove formula; it may be interpreted as 'better than breastfeeding' and that's not the approach supported by WHO or multiple health agencies.

**********

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

[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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

2 Jun 2020

Dear Dr. Whisner,

Thank you for the opportunity to revise our manuscript entitled, “History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood”. We have reviewed the comments raised by you and reviewers and have provided our responses along with a tracked-changes version of the manuscript.

Editor

1. For children, the World Health Organization suggests that this approach to reporting child size and body weight status are more accurate given that they are in a period of more intense growth. This is a best practice in both clinical and research settings, and should therefore be carried out in this paper before it will be deemed acceptable for publication.

We now report body size as body mass index as a percentile on line 122.

2. The paper suggests that formula feeding is fine and may be beneficial given the promotion of gut maturation in Bacteroides. This message needs to be softened throughout the manuscript as many health agencies support breastfeeding as the gold standard for gut and systemic child health. Please revise the manuscript to ensure that formula feeding is not viewed as equal or superior to breastfeeding.

We have softened the language throughout the manuscript to avoid suggesting that formula feeding may be equal or superior to breastfeeding, see lines 203, 240, and 305 in particular.

Reviewer #1:

No Comments

Reviewer #2:

1. I read your rationale why you used BMI only. But it lacks clarity on the predominant body size of the children. Please use BMI% for the BMI of 20 to make sense to readers. I know it's age corrected but it still doesn't tell me of the child's weight status; normal (5% to 85%) over/under weight, (85%-<5th%) obese >95%.

As noted above in the response to the Editor, we now report body size as body mass index as a percentile on line 122.

2. The child's rearing environment is confusing, especially Table 1. Does the box 'sibling genetically related and adoptive home mean that of the 25 adopted children, there were 17 genetically sibs who were reared together in an adoptive home? The box 'child genetically unrelated to adopted child and biological home mean that in the biological home, 14 of children were there who were not related to the adoptees? For example, step-siblings and not half-siblings?

The reviewer’s interpretation of Table 1 is accurate. We have added a note in the Table to clarify this for readers.

3. For feeding practices, did you also measure if the adoptive parents used breast milk from a milk bank or from friends/family/community support breast milk networks? That practice is becoming more common and could have occurred in the younger children in your sample.

Unfortunately, we did not measure if the adoptive parents used breast milk or from friends/family/community support breast milk networks. We have noted this limitation on line 286.

4. Was the delivery mode measured by questions from adoptive or birth mother when records were not available?

Delivery mode was measured by questions from adoptive or birth mother when records were not available. This information is provided on line 112.

5. In the results section, was the 21% breastfed combined from both genetically related and non-genetically related sibs?

The 21% was across all children in our study. We now provide clarification on line 147 by adding, “In our study’s subsample…”, and this percentage also appears in Table 2.

6. Minor changes

On line 195, add 'gut' between human and microbiome.

Line 222 Mention also that bacteriodes can also cause serious infections if escape outside the gut, potentially leading to septicemia too.

Line 282 The percentage of breastfed varies by age of infant. Do you mean ever breastfed, exclusively breastfed, or breastfed to 6 months? More specificity needed.

Line 292 remove formula; it may be interpreted as 'better than breastfeeding' and that's not the approach supported by WHO or multiple health agencies.

All minor changes noted by the reviewer have been made.

Thank you again for the opportunity to revise our manuscript.

Sincerely,

Submitted filename: Response to ReviewersR2.docx

Click here for additional data file.

11 Jun 2020

History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood

PONE-D-20-02497R2

Dear Dr. Leve,

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

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. 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 help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- 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.

Kind regards,

Corrie Whisner

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

23 Jun 2020

PONE-D-20-02497R2

History of breastfeeding but not mode of delivery shapes the gut microbiome in childhood

Dear Dr. Leve:

I'm 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 let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, 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.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Corrie Whisner

Academic Editor

PLOS ONE