The Escherichia coli strain that is known to produce the genotoxic secondary metabolite colibactin is linked to colorectal oncogenesis. Therefore, understanding the properties of such colibactin-positive E. coli and the molecular mechanism of oncogenesis by colibactin may provide us with opportunities for early diagnosis or prevention of colorectal oncogenesis. While there have been major advances in the characterization of colibactin-positive E. coli and the toxin it produces, the infection route of the clb + strain remains poorly characterized.
We examined infants and their treatments during and post-birth periods to examine potential transmission of colibactin-positive E. coli to infants. Here, analysis of fecal samples of infants over the first month of birth for the presence of a colibactin biosynthetic gene revealed that the bacterium may be transmitted from mother to infant through intimate contacts, such as natural childbirth and breastfeeding, but not through food intake.
Our finding suggests that transmission of colibactin-positive E. coli appears to be occurring at the very early stage of life of the newborn and hints at the possibility of developing early preventive measures against colorectal cancer.
Colorectal cancer (CRC) is the third most common form of cancer and the second most common cause of cancer mortality in the world . It is predicted that by 2030 approximately 2.2 and 1.1 million people will develop CRC and succumb to it . To reduce the number of CRC incidences and mortalities, it is vital to identify and mitigate the source of risk factors that contribute to the onset of CRC. Certain strains of Escherichia coli that harbor the gene cluster clb (also referred to as pks) responsible for the biosynthesis of the genotoxin colibactin have been linked to colorectal oncogenesis [3,4,5,6,7,8,9,10]. Recently, we created fluorescent probes  that are turned on specifically by ClbP, a peptidase required to activate the prodrug-like colibactin precursor [12, 13]. The probe allowed high-throughput screening of E. coli isolates from clinical samples, which led to the isolation of the high-colibactin producer E. coli-50 that will be useful in studying the properties of colibactin-positive (clb+) E. coli and colibactin . While we continue to elucidate the molecular mechanism of oncogenesis by clb + E. coli and colibactin, there is still a limited understanding on the infection route of the E. coli to the affected individuals. Since identification of the routes of infection would help develop measures to mitigate or prevent the infection, we initiated a screening effort to examine the prevalence of clb + E. coli among healthy individuals by analyzing their fecal samples  and are currently investigating the continued clb + E. coli infections that occur among healthy individuals. Also, it is known that the newborn gut microbiota starts to form upon exposure to the vaginal and maternal skin microbiomes after birth . A study based on a rat model showed that commensal E. coli strains, including clb + E. coli, are transmitted from mothers to neonates, where early colonization of neonate gut with genotoxic E. coli could influence the intestinal homeostasis at adulthood in a way that may put the individual at risk of colorectal cancer and other immune-mediated diseases . Therefore, we extended our screening to include newborns to study our main clinical objective, which is to determine if clb + E. coli could be transmitted to infants from mother or closely interacting caretaking adults, and if so, what is the source and the medium through which the strain is being passed on to the infants. Here we report that the colibactin-producing E. coli can indeed be rapidly transmitted from mother to child after birth, suggesting that a respectable number of healthy individuals may become predisposed to high risk of CRC at the very early stage of life.
In a previous study we conducted to identify the frequency of healthy adults who carry clb + E. coli, we surveyed 223 healthy adults from Tokyo metropolitan area in Japan for the presence of clb + E. coli in their fecal samples. We found that 60 participants (26.9 %) were positive . We extended the study further to investigate the timing at which individuals become infected with clb + E. coli. For the current study, we examined 51 infants (25 male, 26 female). From the subjects, one set of feces was collected at birth or within two to three days after birth, and another set was collected one month after birth. The samples were examined by PCR to detect the presence of clbB, the gene for one of the PKS–NRPS hybrid megasynthetases encoded in the colibactin biosynthetic gene cluster. We found that 8 out of 51 newborns or 15.7 % of the test samples harbored clb + E. coli immediately after birth (Fig. 1a, lanes 7, 10, 17, 22, 25, 34, 45 and 47 and Fig. 2). On the other hand, 16 of the 51 (31.4 %) tested positive for clb + E. coli one month after birth (Fig. 1b, lanes 7, 10, 17, 18, 22, 25, 26, 27, 28, 29, 31, 34, 35, 45, 47 and 49 and Fig. 2), indicating that 8 newborns or 15.7 % acquired the clb + E. coli strain during their first month. The clb-positive rate increased from 15.7 to 31.4 % by the end of the first month, reaching to the equivalent level of 26.9 % observed among healthy adults examined recently .
Perinatal transmission can happen in utero, in the birth canal or through breastfeeding. Because intrauterine transmission of E. coli in healthy pregnancies is considered to be infrequent , we next investigated the correlation between the method of delivery and the clb-positive ratio among the newborns we studied. Because we expected that the delivery method would only affect the clb-positive ratio among newborns to a few-days-old babies, we did not employ the data collected one month after birth. Regarding the birth canal transmission, clb + E. coli was detected in seven out of eight or 87.5 % of the infants that were born through natural delivery (Table 1). In contrast, only one of eight or 12.5 % of the infants that were delivered by Cesarean section was clb-positive (Table 1).
As to the breastfeeding-mediated transmission, we examined the correlation between the infant feeding mode and the clb-positive ratio. We screened the 43 infants who were determined to be clb-negative at birth to a few days after birth. After one month, eight of the 43 clb-negative infants presented clb-positive, while the remaining 35 (81.4 %) were not affected as determined by the PCR analysis of their fecal samples (Table 2). In total, there were 26 and 17 infants who were fed breastmilk alone and a mixture of formula and breastmilk, respectively. Among the eight clb-positive infants, seven were breastfed strictly over the one-month period, whereas only one newborn was fed a mixture of formula and breastmilk over the month. While 26.9 % of those given breastmilk alone became clb-positive, only 5.9 % of the infants given a mixed feed became clb-positive (Table 2). To check if intaking or handling of food items by caretakers could play a role in transmitting clb + E. coli to the infants, 58 different samples collected from food items including tap water that are commonly consumed by the demographics examined in the current study were screened for the presence of clb + E. coli (Table 3). The search failed to identify clb + E. coli except in a sample of cattle stomach. However, the strain found in the sample of cattle stomach belonged to the phylogroup B1, whereas all of the clb + E. coli strains isolated from human subjects thus far belonged to B2 (manuscript in preparation, Y.Y., Y.T., M.S., Y.I., N.M., M.Mutoh., H.I., H.S., K.Wakabayashi and K.Watanabe). Lastly, we did not observe any difference in the clb + E. coli infection ratio between the sexes of the infants either at birth or one month after birth in this study.
Our survey of infants identified that only 15.5 % of the screened infants harbored clb + E. coli immediately after birth (Fig. 1a, lanes 7, 10, 17, 22, 25, 34, 45 and 47 and Fig. 2). However, after one month the percentage doubled to 31.4 %, similar to the frequency found among healthy adults  (Fig. 1b, lanes 7, 10, 17, 18, 22, 25, 26, 27, 28, 29, 31, 34, 35, 45, 47 and 49 and Fig. 2). Once infected, it is expected that clb + E. coli remains within the system of the infected individual persistently. Thus, the adult-like clb-positive rate we found among the one-month-old newborns in this study suggests that healthy individuals become infected with clb + E. coli very early in their life stage in Japan. These results also indicate that the infants are getting exposed to the source of clb + E. coli under the ordinary living condition during its first month of life.
Next, we examined the source of perinatal transmission, where we focused on the correlation between the method of delivery and the clb-positive ratio among the newborns. The study found that while 87.5 % of the infants delivered by natural birth were clb-positive, only 12.5 % of those delivered by Cesarean section were clb-positive (Table 1). Those results indicated that a higher clb-positive ratio was observed among infants that were born through natural childbirth, similar to how the chance of infants acquiring the vaginal flora bacteria, including E. coli, increases by passing through the birth canal.
We also examined the possible role breastfeeding plays in the transmission of clb-positive E. coli to infants. Of the clb-negative infants that were strictly breastfed, 26.9 % became clb-positive one month after birth, while only 5.9 % of those given a mixture of formula and breastmilk turned clb-positive (Table 2). A simple survey of food items that are considered to be consumed typically by the demographics to which infant caretakers belong showed that none of the food samples examined was contaminated by the clb + E. coli strains that are isolated from human subjects. Therefore, intaking or handling of food items by caretakers being a potential source of clb + E. coli being transmitted to infants was considered less likely. The results implicate that the transmission of clb-positive E. coli to infants occurs mainly through mothers having close contacts with the infants.
In summary, the infection ratio of clb + E. coli is 31.4 % among the one-month-old infants studied, which is similar to the frequency found among healthy adults . It was also reported that 130 Swedish infants followed from birth to 18 months of age were determined to be 33 % clb-positive, very similar to our findings of 31.4 %, when their feces were analyzed using a similar PCR method . Our analysis identified that infants born by natural delivery had a higher incidence of being clb-positive than those born by Cesarean section. Similarly, those who were breastfed strictly showed a higher clb-positive frequency than those given a mixed feed. The fact that the mixed feed also contained breastmilk suggests that breastmilk itself was not the source of clb + E. coli. A simple survey of food items commonly consumed by the Japanese also indicated that clb + E. coli was probably not transmitted through contaminated food consumed or handled by caretakers of the infants. Rather, the most probable route of infection of the potentially oncogenic clb + E. coli strain appears to be through direct skin-to-skin contact, or skin-to-mouth contact involved in breastfeeding to be more specific, between mother and her infant (Table 2). The same can be said about the method of delivery, where infants born by natural birth would have a substantially higher frequency of direct skin-to-skin contact with its mother than those born by Cesarean section (Table 1). Taken together, similar to how the gut microbiota is transmitted from mother to infant , our results strongly imply that clb + E. coli might be transmitted from mother to newborn from the very early stage of life of the newborn through intimate contacts with the mother. Therefore, by implementing measures that can reduce the transmission of clb + E. coli from adults to infants, we may be able to lower the incidence of CRC in our population. For instance, we would be able to develop early preventive measures against colorectal cancer. Those measures could include procedures to prevent infection during childbirth, such as providing counselling to expecting couples to help them become aware of the condition, testing the vaginal flora and altering the flora by external interferences  or providing couples assistance in choosing Cesarean section for delivering their infants. Of course, if future research establishes safe treatments to eradicate clb + E. coli, such procedures could be performed before pregnancy or at a stable time during pregnancy to reduce the mother-to-infant transmission of clb + E. coli. To this end, we are currently analyzing the fecal samples from the mothers of the infants to fully understand the rates and modes of clb + E. coli transmission from mothers and her infants.
Participants and sample collection
The subjects were 51 healthy infants from Tokyo metropolitan area in Japan. The size of the subject pool was deemed adequate for the study based on the known frequencies of clb + individuals found among healthy individuals (approximately 26.9 % ) and child deliveries being carried out by Cesarean Sec. (24.4 % for 2005–2015 , and the rate is steadily increasing ) in Japan. If the frequency of clb + infants is similar to that of healthy adults, approximately three to four clb + infants are expected among 51 subjects (51 ⋅ 0.269 ⋅ 0.244 = 3.3). Fecal samples were collected at birth to two to three days after birth (at the time of discharge within a few days after birth) and one month after birth (at the time of the one-month checkup). The collected feces were immediately placed in a sealed container and stored in a − 20 °C freezer until DNA extraction was performed.
DNA extraction from fecal samples
DNA was extracted from the frozen fecal samples with the bead beating method using a GNOME DNA Isolation Kit (MP Biomedicals). DNA quality was assessed with an Agilent 4200 TapeStation (Agilent Technologies). After the final DNA precipitation step, the DNA samples were resuspended in TE buffer and stored at − 80 °C before the PCR analysis.
DNA extraction from food materials
Each food sample (10–50 g) obtained from grocery stores in Shizuoka, Japan, was added to 20 mL of EC medium (20 g peptone, 5 g lactose, 1.5 g bile salt, 4 g K2HPO4, 1.5 g KH2PO4 in 1 L H2O) in a sterilized bag. The mixture was incubated at 44.5 °C for 24 h. The mixture was filtered, and the filtrate was centrifugated at 10,000 g for 6 min at 4 °C. The collected precipitate was suspended in 1 mL of sterile Milli-Q water. The suspension was plated on a MacConkey agar medium and incubated at 37 °C for 16 h. The grown bacteria were cultured, and its genomic DNA was extracted by using a DNA isolation kit (QIAGEN). The isolated DNA was resuspended in TE buffer and stored at − 80 °C before the PCR analysis.
Confirmation of the presence of the clb gene cluster by PCR
The extracted fecal DNA was subjected to PCR (SapphireAmp Fast PCR Master Mix, Takara) and qualitatively analyzed for clbB (a 9.6-kilobase gene encoding one of the colibactin biosynthetic enzymes) present in the DNA extract by amplifying the gene fragment using a primer set of clbB-F: 5’-TGTTCCGTTTTGTGTGGTTTCAGCG-3’ and clbB-R: 5’-GTGCGCTGACCATTGAAGATTTCCG-3’ as described previously . The correlation was analyzed by comparing the presence or absence of the clbB gene with the subject’s birth method, diet content and sex.
Each experiment was performed at least three independent times. Representation of data as dot-plots and bar-and-whisker graphs is described in figure legends. The t test for determining the statistically difference between the expected and observed frequencies was calculated using JMP (SAS Institute Inc.) and the NORM.DIST function in Microsoft Excel version 16.16.25.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Ferlay J, Colombet M, Soerjomataram I, Mathers C, Parkin DM, Piñeros M, et al. Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer. 2019;144(8):1941–53.
Putze J, Hennequin C, Nougayrede JP, Zhang W, Homburg S, Karch H, et al. Genetic structure and distribution of the colibactin genomic island among members of the family Enterobacteriaceae. Infect Immun. 2009;77(11):4696–703.
Kawanishi M, Hisatomi Y, Oda Y, Shimohara C, Tsunematsu Y, Sato M, et al. In vitro genotoxicity analyses of colibactin-producing E. coli isolated from a Japanese colorectal cancer patient. J Toxicol Sci. 2019;44(12):871–6.
Yoshikawa Y, Tsunematsu Y, Matsuzaki N, Hirayama Y, Higashiguchi F, Sato M, et al. Characterization of colibactin-producing Escherichia coli isolated from Japanese patients with colorectal cancer. Jpn J Infect Dis. 2020;73(6):437–42.
Kawanishi M, Shimohara C, Oda Y, Hisatomi Y, Tsunematsu Y, Sato M, et al. Genotyping of a gene cluster for production of colibactin and in vitro genotoxicity analysis of Escherichia coli strains obtained from the Japan Collection of Microorganisms. Genes Environ. 2020;42:12.
Watanabe D, Murakami H, Ohno H, Tanisawa K, Konishi K, Tsunematsu Y, et al. Association between dietary intake and the prevalence of tumourigenic bacteria in the gut microbiota of middle-aged Japanese adults. Sci Rep. 2020;10:15221.
Watt S, Lanotte P, Mereghetti L, Moulin-Schouleur M, Picard B, Quentin R. Escherichia coli strains from pregnant women and neonates: intraspecies genetic distribution and prevalence of virulence factors. J Clin Microbiol. 2003;41(5):1929–35.
Yassour M, Jason E, Hogstrom LJ, Arthur TD, Tripathi S, Siljander H, et al. Strain-level analysis of mother-to-child bacterial transmission during the first few months of life. Cell Host Microbe. 2018;24(1):146–54.e4.
The authors are grateful toward all the supports and would like to thank everyone who contributed to this study, including the parents and infants who provided the samples. We also thank all of the study site staff for cooperation on this project.
This study was funded by the Development of Innovative Research on Cancer Therapeutics from Japan Agency for Medical Research and Development (AMED) (K.Watanabe, 16ck0106243h0001; 19ck0106475h0001), Innovative Areas from MEXT, Japan (K.Watanabe, 16H06449), the Takeda Science Foundation (K.Watanabe), the Institution of Fermentation at Osaka (K.Watanabe), the Princess Takamatsu Cancer Research Fund (K.Watanabe, 16-24825), Kobayashi Foundation for Cancer Research (K.Watanabe), the Yakult Bio-Science Foundation (K.Watanabe) and SECOM Science and Technology Foundation (K.Watanabe). The funding bodies had no role in the design of the study and collection, analysis, interpretation of data, or manuscript preparation.
Authors and Affiliations
Department of Pharmaceutical Sciences, University of Shizuoka, 422-8526, Shizuoka, Japan
Yuta Tsunematsu, Michio Sato & Kenji Watanabe
Laboratory of Vaccine Materials, Center for Vaccine and Adjuvant Research, Laboratory of Gut Environmental System, Health and Nutrition (NIBIOHN), National Institutes of Biomedical Innovation, 567-0085, Ibaraki-city, Japan
Koji Hosomi & Jun Kunisawa
Department of Pediatrics, Maternal and Child Health Center, Aiiku Clinic, 106-8580, Tokyo, Japan
Department of Human Nutrition, Tokyo Kasei Gakuin University, 194-0292, Tokyo, Japan
Department of Physical Activity Research, Health and Nutrition (NIBIOHN), National Institutes of Biomedical Innovation, 162-8636, Tokyo, Japan
Haruka Murakami & Motohiko Miyachi
School of Veterinary Medicine, Faculty of Veterinary Science, Nippon Veterinary and Life Science University, 180-8602, Tokyo, Japan
Department of Tumor Pathology, Hamamatsu University School of Medicine, 431- 3192, Shizuoka, Japan
Yuji Iwashita & Haruhiko Sugimura
Graduate School of Nutritional and Environmental Sciences, University of Shizuoka, 422-8526, Shizuoka, Japan
Noriyuki Miyoshi & Keiji Wakabayashi
Department of Molecular-Targeting Cancer Prevention, Kyoto Prefectural University of Medicine, 602-8566, Kyoto, Japan
YT, MS, HM, YY, NM, KW1 and KW2 conceived and designed the study. JK, KH, NS and ES collected the fecal samples. YT and KW2 designed and performed PCR analysis. YI, MM1, HI, HS and MM2 performed the statistical analysis. All authors analyzed and discussed the results. KW1 and KW2 prepared the manuscript. All authors have read and approved the manuscript.
A written letter of consent was prepared after obtaining informed consent from each of the parent to participate in this study. The study protocol was approved by the Ethical Committee of the National Institutes of Biomedical Innovation, Health and Nutrition, Japan (Approval No. 199-03).
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.