Diarrhea is a major cause of illness and death among children in developing countries, especially in infants younger than 1 year of age.1 Among the causative agents there is a group of bacteria that have developed a special mechanism for infection, the type III secretion system (T3SS). The T3SS forms a needle-complex/injectisome through which secretes a set of effector proteins that are translocated into the host cell cytoplasm allowing the colonization or invasion of the small intestine.2,3 T3SS effector proteins required for enterocytes invasion in Salmonella and Shigella are called Sip (Salmonella invasion proteins) and Ipa (invasion plasmid antigens), respectively.4 In enteropathogenic Escherichia coli (EPEC), the T3SS effector proteins, denominated Esp (E. coli secreted proteins), are responsible for bacterial adherence/colonization to enterocytes resulting in the characteristics attachment and effacement (A/E) lesion.5 Cellular functions for the effector proteins of these 3 enteropathogenic bacteria have been previously described and are summarized on Table 1.
Breast-feeding reduces the risk of diarrhea and other infections in children living in developing countries. Ingestion of colostrum loads the infant gut with antibodies that protect against potentially lethal infant diarrhea.17 Secretory immunoglobulin A (sIgA) is the main immunoglobulin isotype in colostrum; it represents over 90% of the immunoglobulin present in milk.18 The protective effect of sIgA against various enteropathogens relates to its ability to inhibit cell adhesion,19 as well as its action against lipopolysaccharide.20,21 Antibodies against T3SS-secreted proteins appear to be key in this inhibition of bacteria-enterocyte adhesion.22,23
The aim of this study was to evaluate colostrum from puerperal women living in Lima for the presence of sIgA against the major proteins secreted by the T3SS of Salmonella, Shigella and EPEC. Such antibodies reflect previous exposure and immunological memory on the mother and provide insight into the range of antibodies consumed by infants in a developing country setting.
MATERIAL AND METHODS
Patients and Human Colostrums Samples Collection
A total of 76 human colostrum samples were collected during the first 5 days after birth from mothers with nonpremature newborns at Cayetano Heredia National Hospital in Lima, Peru. This is a referral hospital for urban and peri-urban communities of low and middle socioeconomic class serving the northern district of Lima. The mean age of women was 25 ± 6.1 years; the mean birth weight was 3434.1 ± 477.6 g; and the median newborn age at the time of colostrum collection was 72 hours (range: 48–120). The colostrum samples (5 ± 2 mL) were obtained by hand pumping using gentle pressure to move secretions toward the nipple into sterile flasks. Colostrum samples were centrifuged at 12,000 g for 15 minutes, and the top layer of fat was removed in order to obtain the liquid fraction.20 The samples were aliquoted and stored at –20°C until analysis. Participating mothers signed an informed consent before data and sample collection. The study and procedures were approved by the Ethics Research Institutional Committees at Cayetano Heredia National Hospital and Universidad Peruana Cayetano Heredia.
Bacterial Strains and Induction of Virulence Protein Expression
Strains used in this study included: Shigella flexneri M90T, Salmonella enterica ser. Typhimurium SL1344 and EPEC E2348/69. Additionally, Salmonella ser. Typhimurium SL1344 ΔhilA mutant, lacking the hilA gene, which is necessary for the expression of the genes encoding the T3SS and the effector proteins required for invasion. Samonella and Shigella strains were grown in Luria broth at 37°C, whereas EPEC was grown in Dulbecco’s minimal Eagle medium at 37°C. Each strain was incubated overnight in Luria broth at 37°C and at 350 rpm. The next day, these cultures were then diluted 1:5 in Luria broth and 1:100 in Dulbecco’s minimal Eagle medium, respectively, and grown for an additional 4.5–6 hours at 37°C to late-log/early-stationary phase (optical density [OD]600nm = 1.1–1.3).
Collection of Protein Components Secreted in Media
Supernatant protein extraction was carried out according to Komoriya et al24 with small modifications. Briefly, after desired OD was reached for each strain, 6 mL of bacterial culture were centrifuged at 12,500 g for 15 minutes, then 4 mL of supernatant were removed and 706 μL of trichloroacetic acid (25%, Merck) was added (final concentration 15%). This suspension was incubated at 4°C between 6 hours and overnight and then centrifuged at 12,5000 g for 15 minutes. The pellet was resuspended in Laemmli sample buffer, aliquoted and stored at −20°C until use.
Analysis of Protein Secretion Profiles of Bacterial Strains
Through a vertical denaturing electrophoresis (sodium dodecyl sulfate polyacrylamide gel electrophoresis at 12.5%), secretion profiles of all 3 bacteria were determinated with Coomassie blue. The presence of each specific protein was preliminary determined based on the molecular weight previously reported for each protein (Table 1) and confirmed by Western Blot using monoclonal and polyclonal antibodies against EspB and EspD, respectively,25 monoclonal antibodies against IpaB and IpaC26 and monoclonal antibodies against SipB and SipC (provided by Dr. Jorge E. Galan, Yale University). We did not had antibodies for the other proteins, therefore the diagnosis was presumptive. We were not able to adequately separate EspB and EspD because their molecular weights are too close; therefore, the band detected by antibodies to EspB or D was reported as EspB-D.
Western Blot and Quantification
The protein pellets resuspended in Laemmli sample buffer were separated and analyzed in a 12.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis and then electrophoretically transferred to 0.45 μm pore size nitrocellulose membranes. These membranes were incubated with blocking buffer (3% skim milk in tris buffer saline combined with 0.05% Tween 20) at room temperature for 2 hours and placed on a support: Mini-PROTEAN II Multiscreen Apparatus (Bio-Rad, Hercules, CA). The dilution used for the primary antibody (IgA present in human colostrum) was 1/10 for membranes with Salmonella and Shigella secreted proteins and 1/40 for membranes with EPEC secreted proteins. The membranes were incubated overnight with 300 μL of primary antibody at room temperature and then 3 times washed with milk buffer. The membranes were then incubated with 300 μL of secondary antibody (antihuman IgA peroxidase conjugated) in a dilution of 1/200 for Salmonella and Shigella strains and in a dilution of 1/800 for EPEC. After 2-hour incubation at room temperature, membranes were washed 3 times with tris buffer saline 0.05% Tween 20 and then another 3 times only with tris buffer saline. For developing, Horseradish Peroxidase Conjugate Substrate Kit (Bio-Rad, Hercules, CA) was employed; the reaction was stopped when the bands became visible.
In order to estimate the amount of anti-T3SS IgA present in each colostrum sample, we have quantified the intensity of each Western Blot band using the Quantity One v4.6.3 software (Bio-Rad, Hercules, CA), which measures the band intensity by OD (peaks density). Because the amount of the T3SS proteins loaded in each sodium dodecyl sulfate polyacrylamide gel electrophoresis was the same (1 large comb was use for each gel), and all colostrum samples had the same dilution, we have used the band intensity as a proxy for the amount of antibody present against each T3SS protein.
RESULTS AND DISCUSSION
Antibodies were detected against Salmonella antigens SipA 75/76(99%), SipC 62/76(82%) and SipB 31/76(41%); against Shigella antigens IpaC 70/76(92%), IpaB 68/76 (89%), IpaA 66/76(87%) and IpaD 41/76(54%); and against EPEC EspC 70/76(92%), EspB-D 65/76(86%) and EspA 41/76(54%) (Table 2). Thirty different patterns were found among 76 profiles analyzed. The most common pattern (n = 32, 42%) was the presence of antibodies to all except one of the major secreted proteins of Shigella, Salmonella and EPEC; 13 samples lacked antibodies to SipB, 6 lacked antibodies to EspA, 3 lacked antibodies to IpaD and 2 lacked antibodies to SipC. When combinations of proteins were evaluated, presence of sIgA against all Ipas was the most common combination, present in 54% (41 samples), followed by reactivity against all Esps in 53% (40 samples) and reactivity against SipA–SipC in 47% (36 samples). There were only 6 colostrum samples that lacked antibodies to all Shigella T3SS proteins, 4 that lacked antibodies to EPEC T3SS proteins and 1 that lacked antibody to the Salmonella T3SS proteins.
Antibodies to Salmonella T3SS effector proteins SipA and SipC were the most prevalent. These proteins are among the first to be released by the T3SS targeting host cells.2 On the other hand, responses to other components of the same translocon varied markedly in the same samples (eg, SipC was recognized in 82% but SipB in 41%). The difference in the expression of components of translocon SipB and SipC may be due to the fact that compared with SipC, SipB is secreted for a short period of time at the initial stage of the invasion.27 Therefore, SipB has a lower probability of interacting with the host immune system. Antibodies to Shigella T3SS effector proteins (IpaC and IpaB) were also very prevalent. Previous studies have shown that antibodies against these antigens are present in milk from women living in both high and low-endemic areas of Shigellosis,22,28 suggesting that milk mucosal immunity reflects the mother’s lifetime experience with these organisms.
An EPEC effector protein, EspC, was the most prevalent followed by EspB-D. Although the EspC is a autotransporter protein secreted by a type 5 secretion system, this can also be secreted through the T3SS, which improve its efficiency of translocation into the epithelial cell.15 The presence of sIgA against EspA has been described with large variability. Parissi-Crivelli et al29 found EspA antibodies in 16 of 21 (76%) of colostrum samples from Mexican women. On the other hand, Noguera-Obenza et al23 found a higher frequency in both Mexican women 68/73 (93%) and American women 45/50 (90%).
When we quantified the Western Blot band intensity for each T3SS protein and each colostrum sample, we found large variability on these measurements, but with a normal distribution in general (Table 3). This suggests that the amount of sIgA against each protein varies in each milk sample, probably reflecting the frequency, degree and time of mothers’ previous exposure to these pathogens.
Previous studies in developing countries have evaluated the presence of sIgA in human colostrum against the most frequent lipopolysaccharide serogroups of Salmonella, Shigella and EPEC, showing evidence of immunological memory and prior exposure to these pathogens, in addition to its possible protective role against infection.20,21 However, antibodies to specific lipopolysaccharide by definition do not protect all serotypes. Hence such antibodies represent a poor candidate for investigating milk protective antibodies against all Salmonella, all Shigella and all EPEC. Thus, the proteins of the shared virulence mechanism represent a more plausible set of antigens to study milk-related infant protection. The study of these T3SS antibodies in milk may aid in the understanding of the host immune response and the development of enteric vaccines.
Previously, Sip and Ipa proteins were reported as homologous proteins;6,7 however, analysis of amino acid sequences has determined them to be orthologous proteins with a low number of sequences conserved between them.30,31 Even though the structure of the proteins is different, their homologous functions could thus confer cross-reactivity. Its similarity is important to point out that antibodies against the T3SS effector proteins of EPEC are likely to also be relevant for protection from enterohemorrhagic E. coli or Shiga-toxin producing E. coli infection because the effector proteins are highly conserved between these bacteria. For example, based on BLAST search data through National Center for Biotechnology Information, the degree of aminoacid identity between E2348/69 and E. coli O157:H7 EDL933 for EspA is 81%, for EspB 61%, for EspC 53% and for EspD 74%. Thus, despite the fact that these proteins are not identical, it is likely that if antibodies protect against EPEC infection, they are also likely to protect against enterohemorrhagic E. coli or Shiga-toxin producing E. coli infection.
This study has several limitations. First, we used a qualitative method (presence or absence of an antibody) instead of quantitative method. Because amount may matter, we have tried to indirectly estimate the amount of antibody in each sample. Although this method provides some information, future studies are needed using specific quantitative methods (ie, enzyme-linked immunosorbent assay) or performing serial dilutions of all milk samples for the Western Blot analysis. Second, this study did not use purified antigens. Although this is an approach used in many studies, we were not able to discriminate between antibodies detecting EspB and EspD. Third, colostrum samples were collected from puerperal women at 1 hospital, which may not represent the immunological memory of women from different socioeconomic levels in Lima. Finally, we have not determined the functional protective effects of these antibodies, which is beyond the scope of this study. However, some previous studies have demonstrated the protective effect of specific colostrum and milk antibodies in vitro.19,32,33 Nevertheless, this is one of the few studies that have searched simultaneously for the T3SS effector protein antibodies in colostrum against these important enteric pathogens. Indeed the extraordinarily high frequency of antibodies detected in this study against these multiple pathogens also suggest that milk antibodies reflect the mothers’ long-term immunologic experience.
The authors thank Dr. Jorge E. Galan, Yale University, for kindly providing monoclonal antibodies against SipB and SipC proteins.
1. Fischer Walker CL, Perin J, Aryee MJ, et al. Diarrhea incidence in low- and middle-income countries in 1990 and 2010: a systematic review. BMC Public Health. 2012;12:220
2. Haraga A, Ohlson MB, Miller SI. Salmonellae interplay with host cells. Nat Rev Microbiol. 2008;6:53–66
3. Schroeder GN, Hilbi H. Molecular pathogenesis of Shigella spp.: controlling host cell signaling, invasion, and death by type III secretion. Clin Microbiol Rev. 2008;21:134–156
4. Cossart P, Sansonetti PJ. Bacterial invasion: the paradigms of enteroinvasive pathogens. Science. 2004;304:242–248
5. Ochoa TJ, Contreras CA. Enteropathogenic escherichia coli
infection in children. Curr Opin Infect Dis. 2011;24:478–483
6. Kaniga K, Trollinger D, Galán JE. Identification of two targets of the type III protein secretion system encoded by the inv and spa loci of Salmonella typhimurium that have homology to the Shigella IpaD and IpaA proteins. J Bacteriol. 1995;177:7078–7085
7. Kaniga K, Tucker S, Trollinger D, et al. Homologs of the Shigella IpaB and IpaC invasins are required for Salmonella typhimurium entry into cultured epithelial cells. J Bacteriol. 1995;177:3965–3971
8. Lara-Tejero M, Galán JE. Salmonella enterica serovar typhimurium pathogenicity island 1-encoded type III secretion system translocases mediate intimate attachment to nonphagocytic cells. Infect Immun. 2009;77:2635–2642
9. Demali KA, Jue AL, Burridge K. IpaA targets beta1 integrins and rho to promote actin cytoskeleton rearrangements necessary for Shigella entry. J Biol Chem. 2006;281:39534–39541
10. Hale TL. Genetic basis of virulence in Shigella species. Microbiol Rev. 1991;55:206–224
11. Moreira CG, Palmer K, Whiteley M, et al. Bundle-forming pili and EspA are involved in biofilm formation by enteropathogenic Escherichia coli
. J Bacteriol. 2006;188:3952–3961
12. Abe A, Heczko U, Hegele RG, et al. Two enteropathogenic Escherichia coli
type III secreted proteins, EspA and EspB, are virulence factors. J Exp Med. 1998;188:1907–1915
13. Iizumi Y, Sagara H, Kabe Y, et al. The enteropathogenic E. coli effector EspB facilitates microvillus effacing and antiphagocytosis by inhibiting myosin function. Cell Host Microbe. 2007;2:383–392
14. Daniell SJ, Delahay RM, Shaw RK, et al. Coiled-coil domain of enteropathogenic Escherichia coli
type III secreted protein EspD is involved in EspA filament-mediated cell attachment and hemolysis. Infect Immun. 2001;69:4055–4064
15. Vidal JE, Navarro-García F. EspC translocation into epithelial cells by enteropathogenic Escherichia coli
requires a concerted participation of type V and III secretion systems. Cell Microbiol. 2008;10:1975–1986
16. Mellies JL, Navarro-Garcia F, Okeke I, et al. espC pathogenicity island of enteropathogenic Escherichia coli
encodes an enterotoxin. Infect Immun. 2001;69:315–3124
17. Clemens J, Elyazeed RA, Rao M, et al. Early initiation of breastfeeding and the risk of infant diarrhea in rural Egypt. Pediatrics. 1999;104:e3
18. Morrow AL, Rangel JM. Human milk protection against infectious diarrhea: implications for prevention and clinical care. Semin Pediatr Infect Dis. 2004;15:221–228
19. Carbone CB, Carbone SB, Carneiro-Sampaio MMS. Secretory immunoglobulin A obtained from pooled human colostrum and milk for oral passive immunization. Pediatr Allergy Immunol. 2005;16:574–581
20. Hayani KC, Guerrero ML, Ruiz-Palacios GM, et al. Evidence for long-term memory of the mucosal immune system: milk secretory immunoglobulin A against Shigella lipopolysaccharides. J Clin Microbiol. 1991;29:2599–2603
21. Nathavitharana KA, Catty D, McNeish AS. IgA antibodies in human milk: epidemiological markers of previous infections? Arch Dis Child Fetal Neonatal Ed. 1994;71:F192–F197
22. Hayani KC, Guerrero ML, Morrow AL, et al. Concentration of milk secretory immunoglobulin A against Shigella virulence plasmid-associated antigens as a predictor of symptom status in Shigella-infected breast-fed infants. J Pediatr. 1992;121:852–856
23. Noguera-Obenza M, Ochoa TJ, Gomez HF, et al. Human milk secretory antibodies against attaching and effacing Escherichia coli antigens. Emerg Infect Dis. 2003;9:545–551
24. Komoriya K, Shibano N, Higano T, et al. Flagelar proteins and type III-exported virulence factors are predominant proteins secreted into the culture media of Salmonella typhimurium. Molecu Microbiol. 1999;34:767–779
25. Ochoa TJ, Noguera-Obenza M, Ebel F, et al. Lactoferrin Impairs Type III Secretory System Function in Enteropathogenic Escherichia coli
. Infect Immun. 2003;71:5149–5155
26. Gomez HF, Ochoa TJ, Carlin LG, et al. Human lactoferrin impairs virulence of Shigella flexneri. J Infect Dis. 2003;187:87–95
27. Gong H, Su J, Bai Y, et al. Characterization of the expression of Salmonella Type III secretion system factor PrgI, SipA, SipB, SopE2, SpaO, and SptP in cultures and in mice. BMC Microbiol. 2009;9:73
28. Cleary TG, West MS, Ruiz-Palacios G, et al. Human milk secretory immunoglobulin A to Shigella virulence plasmid-coded antigens. J Pediatr. 1991;118:34–38
29. Parissi-Crivelli A, Parissi-Crivelli JM, Girón JA. Recognition of enteropathogenic Escherichia coli
virulence determinants by human colostrum and serum antibodies. J Clin Microbiol. 2000;38:2696–2700
30. Matteï PJ, Faudry E, Job V, et al. Membrane targeting and pore formation by the type III secretion system translocon. FEBS J. 2011;278:414–426
31. Worrall LJ, Lameignere E, Strynadka NC. Structural overview of the bacterial injectisome. Curr Opin Microbiol. 2011;14:3–8
32. Fernandez RM, Carbonare SB, Carneiro-Sampaio MMS, et al. Inhibition of enteroaggregative Escherichia coli adhesion to HEp-2 cells by secretory immunoglobulin A from human colostrum. Pediatr Infect Dis J. 2001;20:672–678
33. Palmeira P, Carbonare SB, Amaral JA, et al. Colostrum from healthy Brazilian women inhibits adhesion and contains IgA antibodies reactive with Shiga toxin-producing Escherichia coli. Eur J Pediatr. 2005;164:37–43