THE LANCET, Volume 335, Pages 569-571,
March 10, 1990.

Preliminary study of breastfeeding and bacterial
adhesion to uroepithelial cells


The oligosaccharide content of breast-milk and urine from ten nursing mothers and their babies, collected 30 days after delivery, was analysed by thin-layer and high performance liquid chromatography. Each mother's milk and urinary oligosaccaride profiles were very similar, and the pattern of oligosaccharides excreted by her infant was strongly correlated with that of her milk. The babies excreted 300-500 mg/day oligosaccharides and the mothers 500-800 mg/day. The effect of oligosaccharide fractions from a 500 ml pool of colostrum on bacterial adhesion to uroepithelial cells was tested on a strain of Escherichia coli isolated from an infant with urinary tract infection. The high-molecular-weight sialytated oligosaccharides had no effect but neutral oligosaccharides caused inhibition of bacterial adhesion which rose as the size of the oligosaccharides decreased. These findings suggest that breastfeeding may have a preventative effect on urinary tract infection in both mother and infant.

Lancet 1990; 335:569-771.


Since protective factors were demonstrated in human milk in 18921 there have been many attempts to prove that breastfeeding has a defensive role against various infections.2-4 Human milk gives protection against intestinal tract infections, helps the host defence of the respiratory apparatus, and reduces the incidence of otitis media.5-6 Moreover, breastfeeding gives effective protection to the newborn at high risk of infection7 and prevents neonatal sepsis.8

Here we present preliminary evidence suggesting that human milk may help to protect against infection of the urinary tract.


30 days after delivery, 24 h urine collection samples were made from ten nursing mothers and their babies; a sample of milk was also taken from each mother. All of the mothers were term, were healthy and were aged 18-27 years. The babies were all healthy and normal in their development during the first month. A pool of 500 ml human milk was also obtained by mixing several samples of colostrums. All samples were stored at -80°C. The oligosaccharides in the milk and urine samples were characterised by means of thin-layer chromatography (TLC)9 and high-performance liquid chromatography (HPLC). Milk samples were prepared for HPLC by addition of an equal volume of acetonitrile and centrifugation at 3500 g for 5 min. The supernatant was filtered through a 0.22 µm membrane (Millipore, Bedford, Massachusetts, USA. 20µl of the filtrate was injected into the HPLC system. Urine samples were prepared by filtration only.

HPLC used a Kontron 420 dual-piston pump (Basle, Switzerland) and a refractive index detector. The column was a 250 x 5 mm ‘Sperisorb S 5NH’ (Kontron). The elution solvent was water/acetonitrile (40/60 by volume) and the flow rate 1ml/min. An electronic integrator was used to measure the peak areas, which were expressed as concentration. A mixture of purified milk oligosaccharides (BioCarb, Lund, Sweden) was used as standard; the total amount of oligosaccharides was calculated from the peak areas.

Oligosaccharides were isolated from the 500 ml pool of colostrums10 and fractionated by gel filtration.11 Each fraction was analysed for total sugar12 and sialic acid content13 and characterized by TLC.9 Fractions with similar biochemical properties were combined in four pools and lyophilized.

For the study of bacterial adhesion we used as Escherichia coli strain isolated from the urine of a female infant with urinary tract infection; it was characterizsed as bearing mannose-resistant adhesions agglutinating human erythrocytes of group A, Rh+, P1.14 From the date of isolation until use, the bacterial cells were kept on deep agar slab cultures. Adhesion testing was done essentially according to Svanborg Edèn et al.15 Uroepithelial cells, obtained from the morning urinary sediment of healthy women, were washed and suspended in phosphate-buffered saline. To 105 such cells we added 108 bacteria and phosphate bacteria and phosphate-buffered saline to a final volume of 1 ml. After incubation at 37°C for 1 h with rotation at 20 rpm, unattached bacteria were removed by repeated washing in phosphate-buffered saline. Adherence was measured as the mean number of bacteria attached to 40 epithelial cells by means of light microscopy. The ability of pools of milk oligosaccharides to inhibit attachment was tested by preincubation of the bacteria and oligosaccharide suspensions (6mg/ml) for 30 min at 37°C, before the addition of the uroepithelial cells; the washing of the bacteria was omitted. Control cells not incubated with bacteria showed no adherent bacteria.


The urine sample from a 24-year-old healthy non-lactating woman contained only faint bands indicating carbohydrates, whereas substantial amount of sugars were seen in the urine from a lactating woman (fig 1). The pattern of oligosaccharides in the urine of the lactating mother and that of her child was very similar to that see in her milk in all ten subjects. In contrast, cow’s milk contained almost exclusively lactose, with only traces of other oligosacharides. The composition of oligosaccharides excreted in urine by the infant correlated strongly with that of the mother’s milk (fig 2).

By means of HPLC we calculated the amount of oligosaccharides excreted daily in the urine by breastfed babies and their mothers. Daily excretion by the babies was 300-500 mg and by the mothers 500-800 mg.

The extent of inhibition of bacterial adherence by the pooled milk oligosaccharides seemed to be related to the size of the oligosaccharides-pool 4 (containing trisaccharides and tetrasaccharides) had the greatest effect and pool 1 (containing high-molecular-weight sialylated oligosaccharides) had no effect (see table).


Human Milk contains several antiviral and antibacterial factors16 and in-vitro antimicrobial activities varies with the infant’s age according to the different pathogens most common in infants of that age.17

A fundamental role in the defence against infections has been attributed to the IgAs, but human milk contains many other non-immunoglobulin protective factors (eg, lysozyme, complement fractions, lactoferrin) whose biological role is still only partly understood.

Bacterial adhesion to epithelial surfaces is an important prerequisite for infection.18 Adhesion is generally established by binding of the bacteria to the specific receptors of the host cell surface. Most receptors are oligosaccharide residues of glycoproteins and glycolipids on the cell membrane.19

The protection against E. coli and Vibrio Cholarae affored by human milk is due the presence of glucidic compounds,20 and oligosaccharides from human milk can inhibit the adhesion of Streptococcus pneumoniae to human pharyngeal and buccal epithelial cells21 and bacterial adherence to pig intestineal epithelial cells in vitro.22

Human milk contains substantial amount of oligosaccharides (several g/l), especially in the colostral phase.9 23 24 Little is known about the metabolic fate of these substances in the newborn infant. Only traces of oligosaccharides are present in the urine of normal subjects with the exception of women in the last trimester of pregnancy, nursing mothers,25 26 and patients with some metabolic diseases.27

We have found substantial amounts of oligosaccharides in the urine of breastfed infants during the first month of life. These compounds have structural similarities to the receptors of host-cell surfaces, and each host-cell type has a different spectrum of oligosaccharide structure.28 Rosenstein et al29 showed that some oligosaccharides of human milk bound to E. coli isolated from a patient with urinary tract infection. This evidence, together with the inhibition of bacterial adhesion caused by a particular fraction of human milk oligosaccharides suggests that the protective effect against infections that human milk affords the baby could be extended to the urinary tract of not only the infant but also the mother.

Of course, caution is necessary in extrapolating these in-vitro findings to human beings.

We thank Prof C. Svanborg-Eden (University of Göteburg) for helpful comments and criticism and Dr P. Pierani and Dr L. Zampini for technical assistance.


 1. Ehrlich P. Ueber Immunität dirk Verebung und Saugung. Z Hygiene Infektionskr 1892;12:193-203.

 2. Cunningham AS. Morbidity in breast-fed and artificially fed infants. J Pediatr 1979;95:685-89.

 3. Larsen SA, Homer DR. Relation of breast versus bottle feeding to hospitalization for gastroenteritis in a middle-class US population J Pediat 1978;92:417-18.

 4. Wright JA, Walker WA. Breastmilk and host defence of the infant. In Grand RJ, Sutphen JL, Dietz WH, eds. Pediatric Nutrition. Boston: Butterworths, 1987:293-303.

 5. Jason JM, Nieburg P, Marks JS. Mortality and infectious disease associated with infant-feeding practices in developing countries. Pediatrics 1984;74(suppl):702.

 6. Kovar MG, Serdula MK, Marks JS, Fraser DW. Review of the epidemiologic evidence for an association between infant feeding and infant health. Pediatrics 1984;74(suppl):615-38.

 7. Narayanan I, Prakash K, Gujral VV. The value of human milk in the prevention of infections in the high-risk low-birth-weight infant. J Pediatr 1981;99:496-98.

 8. Winberg J, Wessner G. Does breast milk protect against septicemia in the newborn? Lancet 1971;i:1091-94.

 9. Coppa GV, Pierani P, Gabrielli O, Felici L, Catassi C, Giorgi PL. Chromatographic characterization of colostrums carbohydrates. Ital J Pediatr 1988:14:182-86.

10. Kabata A. Isolation of oligosaccharides from human milk. Meth Enzymol 1972;28:262-71.

11. Yamashita K, Mizuochi T, Kobata A. Analysis of oligosaccharides by gel filtration. Meth Enzymol 1982;83:105-26.

12 Dubois M, Gilles AK, Hamilton JK, Rebers PA, Smith F. Colorimetric method for determination of sugars-and related substances. Anal Chem 1956;28:350-56.

13. Warren L. The thiobarbituric acid assay of sialic acids. J Biol Chem 1959;234:1971-75.

14. Kallenius G, Bollby R. Adhesion of Escherichia coli to human periurethral cells correlated to mannose-resistant agglutination of human erythrocytes. FEBS Lett 1979; 5: 295-99.

15. Svanberg Eden C, Eriksson B, Hanson LA. Adhesion of Escherichia coli to human uroepithelial cells in vitro. Infect Immun 1977; 18: 767-74.

16. Hanson LA, Karlsson B, Jalil F, et al. Antiviral and antibacterial factors in human milk. In: Hanson LA, ed. Biology of human milk. New York: Raven Press Ltd, 1988: 141-57.

17. Dolar Sa, Boesman-Frinkelstein M, Frinkelstein RA. Antimicrobial activity of human milk against pediatric pathogens. J Infect Dis 1986; 154: 722-25.

18. Editorial. Microbial adhesion, colonisation and virulence. Lancet 1981; ii:508-10.

19. Karlsson KA. Animal glycolipids as attachment sites for microbes. Chem Phys Lipids 1986; 42: 153-72.

20. Holmgren J, Svennerholm AM, Lindband, N. Receptor-like glycocompounds in human milk that inhibit classical and El Tor Vibrio cholerae cell adherence (hemagglutination). Infect Immun 1983; 39: 147-54.

21. Andersson B, Porras O, Hanson LA, Lagergard T, Svanborg Eden C. Inhibition of attachment of Streptococci pneumoniae and Haemophilus influenzae by human milk and receptor oligosaccharides. J Infect Dis 1986; 153: 232-37. [PubMed]

22. Ashkenazi S, Mirelman D. Non immunoglobulin fraction of human milk inhibits the adherence of certain enterotoxigenic Escherichia coli strains to guinea pig intestinal tract. Pediatric Res 1987;22:130-34.

23. Montreuil J, Mullet S. Étude des variations des constituents glucodiques du lait de femme au cours de lactation [Study of the variations in the glucosidal constituents of human breast milk during the course of lactation]. Bull Soc Chim Biol 1960; 42: 365:77.

24. Viverge D, Grimmonprez L, Cassanas G, Bardet L, Bonnet M, Solere M. Variations of lactose and oligosaccharides in milk from women of blood types secretor A or H, secretor Lewis and secretor h/non secretor Lewis during the course of lactation. Am Nutr Metabol 1985; 29:1-11.

25. Date JW. The isolation of some oligosaccharides from the urine of pregnant and lactating women. Scand J Clin Lab Invest 1964; 16: 597-603.

26. Hallgren P, Lindburg BS, Lundblad A. Quantification of some urinary oligosaccharides during pregnancy and lactation. J Biol Chem 1977; 252: 1034-40.

27. Strecker G, Peers MC, Michalski JC, et al. Structure of nine sialyl-oligosaccharides accumulated in urine of eleven patients with three different types of sialidosis. Eur J Biochem 1977; 75:391-99.

28. Feiz T. Demonstration by monoclonal antibodies that carbohydrate structures of glycoprotein and glycolipids are onco-developmental antigens. Nature 1985;314:53-57.

29. Rosenstern IJ, Mirvochi T, Hounsel EF, Stoll MS, Childs RA, Feiz T. New type of adhesive specificity revealed by oligosaccharide probes in Escherichia coli from patients with urinary tract infection. Lancet 1988; ii: 1327-30.

ADDRESSES: Department of Paediatrics (Prof G. V. Coppa, MD, O. Gabrielli, MD, Prof P. Giorgi, MD, C. Catassi, MD) and Institute of Microbiology (M. P. Montanari, MD, Prof P. E. Varaldo, MD), University of Ancona Medical School, Ancona, Italy and USDA/ARS Children's Nutrition Research Center, Baylor College of Medicine, Houston, Texas, USA (Prof B. L. Nichols, MD). Correspondence to Prof G. Coppa, Department of Paediatrics, University of Ancona, Via Corridoni 11, 60123, Ancona, Italy.

(File revised 15 September 2006)

Return to CIRP library