Journal of Microbiology, Immunology and Infection (2011) 44, 1e7

ava i lab le at www.sc iencedi rec t .com

journa l homepage : www.e- jmi i . com

ORIGINAL ARTICLE

Human colostral phagocytes eliminateenterotoxigenic Escherichia coli opsonized bycolostrum supernatant

Eduardo Luzıa Franca a, Renata Vieira Bitencourt b, Mahmi Fujimori a,Tassiane Cristina de Morais a, Iracema de Mattos Paranhos Calderon c,Adenilda Cristina Honorio-Franca a,c,*

a Institute of Biological Sciences and Health, Federal University of Mato Grosso, Pontal do Araguaia, Mato Grosso, Brazilb Institute of Health Sciences, University Center of Planalto of Araxa, Araxa, Minas Gerais, Brazilc Postgraduate Program in Gynecology, Obstetrics and Mastology, Botucatu Medical School, Sao Paulo State University,Sao Paulo, Brazil

Received 2 March 2009; received in revised form 14 October 2009; accepted 17 December 2009

KEYWORDSColostrum;EnterotoxigenicEscherichia coli;Microbicidal activity;Phagocytes

* Corresponding author. Institute ofPontal do Araguaia, MT 78698-000, Br

E-mail address: adenilda@ufmt.br

1684-1182/$36 Copyright ª 2011, Taiwdoi:10.1016/j.jmii.2011.01.002

Background: Several elements in colostrum and humanmilk, including antibodies and nonspecificfactors with bactericidal and antiviral activity, may play an important anti-infectious and protec-tive role. In developing countries, enterotoxigenic Escherichia coli (ETEC) is the main etiologicalagent of diarrhea in low-socioeconomic level children. In the present work, we studied the func-tional activity ofmononuclear (MN) and polymorphonuclear (PMN) phagocytes of human colostrumagainst ETEC, as well as the interactions between these cells and colostral or serum opsonins.Methods: Colostrum samples were collected from 33 clinically healthy women between 48 and 72hours postpartum. We verified superoxide release in colostral MN and PMN using cytochrome Creduction methods, phagocytosis, and bactericidal activity using acridine orange methods andsuperoxide dismutase (SOD) in the colostrum supernatants.Results: Colostral MN and PMN phagocytes exposed to ETEC opsonized with colostrum superna-tants caused a significant increase (p< 0.05) in superoxide release. Phagocytosis by colostralPMN cells increased significantly (p< 0.5) when the phagocytes were incubated with both sourcesof opsonins (sera and colostrum). Increases in superoxide release in the presence of opsonizedbacteria triggered the bactericidal activity of the phagocytes. Phagocyte treatment with SODdecreased their ability to eliminate ETEC. Colostrum supernatant had higher SOD concentrations(p< 0.05) compared with normal human sera.

Biological Sciences and Health, Federal of University of Mato Grosso, Rodovia MT 100, Km 3,5 s/no,azil.(A.C. Honorio-Franca).

an Society of Microbiology. Published by Elsevier Taiwan LLC. All rights reserved.

2 E.L. Franca et al.

Conclusions: These results suggest that the ability of phagocytes to eliminate ETEC depends onthe activation of cellular oxidative metabolism; moreover, activation of colostral phagocytes islikely an additional breast-feeding protection mechanism against intestinal infections in infants.Copyright ª 2011, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. All rightsreserved.

Introduction

Colostrum is a rich source of nutrients containing severalimmunological components, antioxidants, and bioactivesubstances that play an important role in infant protectionagainst gastrointestinal and respiratory infections.1 Acutediarrhea is the second main cause of death in low-socio-economic level infants in developing countries.2 Enter-otoxigenic Escherichia coli (ETEC) is a major causative agentof death from diarrhea in children under five years of age.3

Many studies have shown that colostrum and human milkcontain important protective factors that combat pathogensin children.4e8 Human milk is particularly rich in secretoryimmunoglobulin A (sIgA),5 which blocks bacterial adherenceto human epithelial cells;7 neutralizes toxins;9 prevents viralinfections;10 and acts as an opsonin by increasing free radi-cals, phagocytosis, and the microbicidal activity of colostralcells.4,11 The protective role of colostral IgA and human milkhas been shown for various microorganisms.4,12

In addition to antibodies, soluble bioactive components,and anti-infectious factors, human colostrum contains largeamounts of viable leukocytes (1� 109 cells/mL in the firstdays of lactation), especially macrophages and neutro-phils.13 In addition to circulating leukocytes, these cellsproduce free radicals and have phagocytic and bactericidalactivity.14 In bacterial infections, phagocytes are known tobe the main cell lineage in host defense.15

Colostral macrophages have phagocytic activity, expressIgA receptors (FcaR; CD89),11,16 and produce oxygen-freeradicals.17 The bactericidal activity of colostral mono-nuclear (MN) phagocytes after opsonization with sIgA isequivalent to that of MN and polymorphonuclear (PMN)phagocytes from peripheral blood.4 Colostral neutrophils,in turn, have lower phagocytic and bactericidal activitycompared with the peripheral neutrophils.18 The IgAreceptors that the neutrophils express are free of gammachain associations (g-less FcaR) and may mediate thenoninflammatory effects of sIgA.11 Therefore, the main roleof sIgA is likely related to maternal protection.19

Human milk contains antibodies against a variety ofbacteria, and colostrum inhibits the adhesion of differentETEC serotypes.8 Components other than immunoglobulins(i.e. oligosaccharides, glycoproteins, and glycolipids) alsoseem to prevent the adhesion of ETEC strains to host cells.8

Phagocytes in human colostrum possibly act in conjunctionwith soluble components, thereby providing an additionalprotection mechanism against enterobacterial infection.However, studies on the functional activity of colostral MNand PMN phagocytes and their interactions with thebioactive components of colostrum are only partiallyunderstood. Considering that ETEC accounts for childdiarrhea worldwide, this study aimed at evaluating the

potential of colostral MN and PMN phagocytes for elimi-nating ETEC as well as the interaction between this path-ogen and soluble elements of colostrum.

Materials and methods

Subjects

After an informed consent was obtained, around 15 mL ofcolostrum was collected from clinically healthy women,18e35 years of age, between 48 and 72 hours postpartum4

at the Human Milk Bank of the Center for Female Care,Araxa, Minas Gerais (nZ 23) and at the Health SystemProgram of Barra do Garcas, Mato Grosso (nZ 10). All theprocedures were analyzed and approved by the ResearchEthics Committee of the “Centro Universitario do Planaltode Araxa”, Araxa, Minas Gerais, Brazil.

Separation of colostral cells

Colostrum samples were collectedmanually in sterile plastictubes and centrifuged (160�g, 4�C) for 10 minutes. Centri-fugation separated colostrum into three different phases:cell pellet, an intermediate aqueous phase, and a lipid-containing supernatant, as described by Honorio-Francaet al.4 Cells were separated by a Ficoll-Paque gradient(Pharmacia, Upsala, Sweden), producing preparations with95% of pure PMN cells and 98% of pure MN cells, analyzed bylight microscopy. Purified neutrophils and macrophageswere resuspended independently in serum-free medium 199at a final concentration of 2� 106 cells/mL.

Escherichia coli strain

The ETEC used was the ETEC 14e057:H7 serotype, providedby the Laboratory of Immunology of Intestine Mucosa of theBiomedical Sciences Institute of the University of Sao Paulo,Brazil. The stock culture was cultivated in trypic soy broth(Difco Laboratories, Detroit, MI, USA) for 18 hours at 37�C.Bacteria were washed twice in phosphate-buffered salineand adjusted to an approximate concentration of 1� 108

bacteria/mL as measured by turbidimetry at 540 nm, usinga spectrophotometer (Femto, Sao Paulo, Brazil). Thisbacterial concentration was previously determined bycolony unit counting on trypic soy agar (Difco Laboratories,Detroit, MI, USA).

Opsonin sources

Colostrum supernatantda pool of 10 colostral samples[immunoglobulin concentration (g/L): IgAZ 7.4, IgGZ 0.15,

Colostral phagocytes eliminate ETEC 3

and IgMZ 0.37] was defatted by repeated centrifugation at160�g for 10minutes, at 4�C and used for ETEC opsonization.Serada pool of normal human serum samples from 10volunteer donors [immunoglobulin concentration (g/L):IgAZ 2.64, IgGZ 14.0, and IgMZ 2.0] was prepared andused for ETEC opsonization.

Bacterial opsonization

The opsonization of ETEC was achieved according to thetechnique described by Bellinati-Pires et al.20 Colostrumsupernatants and sera from 10 individuals were collected,pooled, and frozen at �70�C. Immediately before use,colostral and serum aliquots were thawed and mixed withappropriate volumes of bacterial suspension to a finalconcentration of 2� 107 bacteria/mL in 10% of the opsoninsources. Another bacterial suspension prepared at the sameconcentration in medium 199 without opsonin was used asan untreated bacterial control. Both bacterial suspensionswere incubated for 30 minutes at 37�C and used in thebactericidal assays.

Release of superoxide anion

Superoxide release was measured by determining cyto-chrome C (Sigma, St Louis, USA) reduction as previouslydescribed.4,21 Briefly, MN and PMN phagocytes andbacteria, opsonized or not, were mixed and incubated for30 minutes for phagocytosis. Cells were then resuspendedin phosphate-buffered saline containing 2.6 mM CaCl2,2 mM MgCl2, and cytochrome C (2 mg/mL). Phorbol myr-istate acetate stimulation was performed as control at0.5 mg/mL. The suspensions (100 mL) were incubated for 60minutes at 37�C on culture plates. The reaction rates weremeasured by absorbance at 550 nm and the results wereexpressed as nmol/O2

�. All the experiments were performedin duplicate or triplicate.

Bactericidal assay

Microbicidal activity and phagocytosis were evaluated usingthe acridine orange method described by Bellinati-Pireset al.20 Equal volumes of bacteria and cell suspensions weremixed and incubated for 30minutes at 37�C under continuousshaking and in the presence or absence of superoxide dis-mutase (SOD; 140 units).22,23 Phagocytosis was stopped byincubation in ice. To eliminate extracellular bacteria, thesuspensions were centrifuged twice (160g, 10 minutes, 4�C),and the cellswere resuspended in serum-freemedium199andcentrifuged. The supernatantwasdiscardedand the sedimentdyed with 200 mL of acridine orange (14.4 g/L) for 1 minute.The sediment was resuspended in cold culture 199, washedtwice, and observed under immunofluorescence microscopeat 400� and 1,000� magnification. The phagocytosis indexwas calculated by counting the number of cells ingesting atleast three bacteria in a pool of 100 cells. To determine thebactericidal index, we stained the slides with acridine orangeand counted 100 cells with phagocytized bacteria. Thebactericidal index is calculated as the ratio between orangestained (dead) and green stained (alive) bacteria� 100.20 Allthe experiments were performed in duplicate or triplicate.

CuZn-SOD concentration (CuZn-SODdE.C.1.15.1.1)

The CuZn-SOD enzyme concentration was determined insera (nZ 10) and colostrum (nZ 10) samples using thenitroblue tetrazolium test (NBT; Sigma, St Louis, USA.) andspectrophotometrically read at 560 nm.24,25

A 0.5 mL volume of sera or colostrum samples wasplaced in test tubes, and 0.5 mL of standard hydroalcoholicsolutions (1:1 v/v) were prepared in other tubes. Bothsamples of sera/colostrum and the standard solutions wereadded to 0.5 mL of chloroform-ethanol solution (1:1 ratio)followed by 0.5 mL of a reactive mixture of NBT and dia-minoethanetetraacetic acid at a 1:1.5 ratio (v/v). After2.0 mL of carbonate buffer plus hydroxylamine wereadded, pH increased to 10.2.24 The tubes remained at roomtemperature for 15 minutes and underwent spectrophoto-metrical reading. The reactive mixture reached zero values(for 3.5 mL). The SOD was calculated by the followingrelationship: SODZ (Ab standard�Ab sample/Ab stand-ard)� 100Z % reduction of NBT/CuZn-SOD. The resultswere expressed in international units of CuZn-SOD.

Statistical analysis

We used analysis of variance to evaluate superoxide anionrelease, phagocytosis, and the bactericidal index accordingto the presence or absence of opsonized ETEC, in phago-cytes with or without SOD treatment and in MN andPMN phagocytes. Student’s t test was used to analyze SODenzyme concentration for two independent samples. Weconsidered p values< 0.05 to be statistically significant.

Results

Superoxide release by colostral phagocytes in thepresence of ETEC opsonized with colostrumsupernatant

Colostrum MN and PMN phagocytes exposed to ETEC opson-ized with colostrum supernatant increased superoxiderelease compared with phagocytes exposed to bacteriaopsonized with normal human sera and phorbol myristateacetate-stimulated cells (Table 1). When phagocytes wereincubated with nonopsonized ETEC, the superoxide releasedby MN phagocytes was higher than that of PMN and equal tothat released as phagocytes when exposed to bacteriaopsonized by human sera. PMN phagocytes exposed to non-opsonized ETEC had levels of superoxide release similar tothose of spontaneous release (Table 1).

Phagocytosis activity of colostrum MN and PMN cellsagainst ETEC opsonized by colostrum supernatant

Colostrum MN and PMN phagocytes have phagocytic activityfor ETEC. In the presence of opsonins (colostrum superna-tant or normal sera), phagocytosis increased significantly.Both sources of opsonins induced equivalent phagocytosisrates by colostral MN and PMN (Fig. 1A and B). To verify theeffects of superoxide release on the phagocytosis of ETEC,the phagocytic activity of colostral MN and PMN was

Table 1 Superoxide release by colostrum MN and PMNphagocytes in the presence or absence of ETEC

Phagocytes Superoxide release (nmol/O2�)

MN PMN

Spontaneous(without bacteria)

11.5� 0.7 11.6� 0.6

PMA (without bacteria) 15.8� 0.9** 14.7� 0.6**Bacteria incubated withmedium 199

15.0� 1.1** 11.4� 1.1*

Bacteria opsonized withcolostrum pool

21.7� 0.6** 16.3� 0.7*,**

Bacteria opsonized witha serum pool

15.7� 0.8** 14.0� 0.8**

Bacteria were opsonized with a colostrum supernatant pool ora normal serum pool. In control assays, MN and PMN cells werepreincubated with medium 199. The results represent themean � standard deviation of 8 experiments with cells fromdifferent individuals.*p< 0.05 comparing cell type, considering the same opsoninsource.**p< 0.05 comparing the treated groups with spontaneoussuperoxide released, considering the same cell type.ETECZ enterotoxigenic Escherichia coli; MNZmononuclear;PMAZ phorbol myristate acetate; PMNZ polymorphonuclear.

0

20

40

60

80

100

Medium 199 Serum Colostrum

Bacteria

Pha

gocy

tosi

s In

dex

(%)

SOD-SOD-

SOD+

0

20

40

60

80

100B

A

Medium 199 Serum Colostrum

Bacteria

Pha

gocy

tosi

s In

dex

(%)

SOD-SOD-

SOD+ **

*

*

Figure 1. Bacterial phagocytosis by colostral (A) MN and (B)PMN cells. Effects of SOD on ETEC phagocytosis by colostralphagocytes. Bacterial phagocytosis by MN or PMN cells fromcolostrum was determined using the acridine orange method inthepresenceor absenceof SOD (140units). Results represent themean � standard deviation of five experiments with cells fromdifferent individuals. *p< 0.05 comparing the opsonized groupswith the control groups, using a same SOD treatment or not.ETECZ enterotoxigenic Escherichia coli; MNZmononuclear;PMNZ polymorphonuclear; SODZ superoxide dismutase.

4 E.L. Franca et al.

analyzed in the presence of SOD, an enzyme that metabo-lizes anion superoxide. As shown in Fig. 1A and B, phago-cytosis index did not change when MN or PMN phagocyteswere incubated in the presence of SOD.

Elimination of ETEC opsonized with colostrumsupernatant by colostral MN and PMN phagocytes

Colostral MN and PMN phagocytes exerted bactericidalactivity against ETEC. When the bacteria were opsonizedwith colostrum supernatant or normal sera, cell activityenhanced significantly (p< 0.05). Both sources of opsonins(colostrum supernatant or normal sera) induced equivalentETEC elimination rates (Fig. 2). When the cells were pre-incubated in the presence of SOD, decreased bactericidalactivity (nonopsonized bacteria) was observed in both typesof cell. ETEC opsonized with colostrum supernatant in thepresence of MN or PMN phagocytes and preincubated withSOD had lower bactericidal activity. No effects of SOD wereobserved in ETEC elimination by PMN phagocytes usingserum-opsonized bacteria (Fig. 2).

CuZn-SOD concentration in human colostrumsupernatant

CuZn-SOD concentrations in colostrum supernatantincreased significantly (p< 0.05) compared with CuZn-SODlevels in sera (Fig. 3).

Discussion

In the present study, we show that colostral phagocytessignificantly eliminate ETEC and that this activity is

dependent on prooxidative cell metabolism. ETEC is themain cause of diarrhea in developing countries, accountingfor 800 million deaths a year, mainly in children under theage of 5 years.26

Breastfeeding has been described as an effective inter-vention for infant protection, especially against respiratoryand gastrointestinal infections.27 Several studies haveshown that the phagocytic and microbicidal activity ofcolostrum phagocytes12 is comparable with that of bloodphagocytes,13 with similar rates of phagocytosis andbactericidal function.4

Earlier studies have shown that colostrum supernatantcontains components that activate the prooxidativemechanisms of cells.4 The microbicidal activity of

0

20

40

60

80

100A

B

Medium 199 Serum Colostrum

Bacteria

Bac

teric

idal

Inde

x (%

)SOD-SOD+

0

20

40

60

80

100

Medium 199 Serum Colostrum

Bacteria

Bac

teric

idal

Inde

x (%

)

SOD-SOD+

** **

**

* *

** **

*

*

Figure 2. Bacterial elimination by colostral (A) MN or (B)PMN cells. Action of SOD on ETEC elimination by colostralphagocytes. Bacterial elimination by MN or PMN cells fromcolostrum was determined using acridine orange methods inthe presence or absence of SOD (140 units). Results representthe mean � standard deviation of five experiments with cellsfrom different individuals. *p< 0.05 comparing the opsonizedgroups to the control groups, using a same SOD treatment ornot; **p< 0.05 comparing the SOD-treated group to the sameopsonin source. ETECZ enterotoxigenic Escherichia coli;MNZmononuclear; PMNZ polymorphonuclear; SODZ super-oxide dismutase.

0

10

20

30

40

50

60

70

Serum Colostrum

Cu

Zn

-S

OD

(U

I)

*

Figure 3. Concentration (mean� standard deviation) ofCuZn-SOD in human sera and colostrum obtained in eightexperiments. SODZ superoxide dismutase.

Colostral phagocytes eliminate ETEC 5

colostrum phagocytes has been linked to the activation ofcellular oxidative metabolism and release of large amountsof free radicals, an extremely important phenomenonduring immune responses and inflammatory reactions.28

The results of this study confirm the importance ofsuperoxide anion for bacterial death. The increase insuperoxide release affects phagocytic and bactericidalactivity because SOD-treated phagocytes decrease ETECelimination. Other studies found that heat-inactivated SODdoes not affect the microbicidal activity of phagocytes.Similarly, heat-inactivated SOD fails to block peroxidaseinhibitors.29

Superoxide anion production, phagocytosis, and themicrobicidal activity of MN and PMN phagocytes was higher

in the presence of bacteria previously opsonized withcolostrum supernatant or normal sera. The immunoglobu-lins and their complements, in particular IgG and C3b,which are the main opsonins in the blood,30 are also foundin smaller amounts in human milk and colostrum.31 Otherstudies have already reported the elimination of opsonizedenteropathogenic Escherichia coli by MN phagocytes butnot by PMN phagocytes.4 In the case of MN phagocytes,microbicidal activity was stimulated only in the presence ofopsonins or other immunomodulatory agents.4

The functional activity of colostral PMN has been shownto be lower than that of both human blood PMN andcolostral MN.32 But in the present study, PMN and MNphagocytes had similar responses against ETEC previouslyopsonized with colostrum supernatant or sera. Although IgGis the predominant antibody type in sera, it is found inlower amounts in colostrum, as are proteins from thecomplement system.17

Both colostral MN and PMN phagocytes had higher super-oxide release and microbicidal activity in the presence ofETEC previously opsonized with colostrum supernatant.Because IgA is the predominant antibody class in colostrum,6

this finding suggests that IgA is the opsonin that stimulatesmicrobicidal activity by both types of phagocytes in humancolostrum. The significance of the biological activity of IgA,such as opsonin in colostrum secretion, is likely associated tothe formation of a complete microenvironment in whichsoluble and cellular components act together.4

An interesting point is that colostrum phagocytesexposed to nonopsonized ETEC had low bactericidalactivity. Some studies report that these cells havedecreased bactericidal activity because they lack nonspe-cific receptors on their surface. Hence, bacteria that areincubated with phagocytes may not be phagocytized andthus remain outside these cells.4 The fact that colostral MNphagocyte activity is opsonin dependent strengthens thehypothesis of an interaction between soluble and cellularcomponents from colostrum, with phagocytes playinga fundamental role in this interaction.

6 E.L. Franca et al.

According to our data, the release of superoxide anionwas higher in MN phagocytes than in PMN phagocytes, andthis difference was more marked when ETEC was previouslyopsonized with colostrum supernatant. The bactericidalactivity of MN phagocytes was also higher in the presence ofETEC previously opsonized with colostrum supernatant, asshown in Fig. 2. Similar results were found for the MN cellsof Giardia lamblia12 and enteropathogenic Escherichiacoli,4 whose phagocytic ability increased after exposure toa large amount of opsonins, suggesting that this cell type isfunctionally more active, especially after opsonization.6

The activation mechanisms of human colostrum phago-cytes may depend on stimulatory signals generated bycomplement proteins and antibodies in milk, especiallyIgA, complement receptor, and Fc receptor.33e35 Thecombined action of these factors mediates signals thatlead to degranulation, oxygen radical production, andphagocytosis.35

Human colostrum and milk are rich in biologicallyactive molecules that are essential for antioxidant func-tions. A number of milk enzymes are important as immuneprotectors.6 In the present study, cellular activationproduced a significant amount of superoxide, which isessential for microbicidal activity. SOD concentration incolostrum was higher than in human sera. Their solubleand cellular components act in a child’s gut without,however, provoking an inflammatory response.36 Theactivity of SOD in human milk appears to be 10e25 timeshigher than in sera36 and may change significantly duringlactation to meet the different needs of newborn devel-opment.37 Other studies show that the highest activity ofthis enzyme in milk occurs at three weeks of lactation andthat it decreases after 4 months of lactation.37 The anti-oxidant capacity of colostrum, which is vital for thenewborn in the first days of life, is higher than that oftransition and mature milk.1

The activation of human colostral phagocytes by opsoninsand consequent release of superoxide anion is an importantmechanism for child protection. This secretion exhibits highSOD enzyme content, which likely underlies phagocyteaction and avoids inflammatory reactions in a child’s intes-tine. The lack of antioxidant defenses may trigger an inten-sification of oxidative actions that culminate in a series ofdiseases, especially in premature infants.38

Owing to the immaturity of the immune system ofnewborns, many immunological components are notcompletely developed.5 Human milk not only providespassive immunity but can also directly modulate theimmune development of the child.39 Because of thenoninvasive characteristics of ETEC, a protective immuneresponse to the infection caused by this pathogen is likelymediated by the mucosal immune system.40 There aremany significant immune components in human milk, andthese factors interact with each other, acting synergisti-cally and providing protection to the newborn.5,14 Thenumber of viable cells and bioactive substances in thecolostrum suggests that this secretion is important toprotect the child. Because of the immature gastrointestinalsystem in newborns, cells received through the colostrumare not eliminated by digestive enzymes and remain intactin the intestinal mucosa for 4 hours, participating activelyin bacteria elimination.41,42

Colostrum phagocytes exert bactericidal activity againstETEC through interactions with soluble factors in colos-trum. The functional activity of phagocytes described inthis work reinforces the findings of other studies thatdemonstrate the importance of breastfeeding in combatingthese enterobacteriaceae.8,43 ETEC diarrhea affects mainlychildren under the age of 2 years, and its incidencedecreases over time.44

The ability of phagocytes to eliminate ETEC depends onthe activation of cellular oxidative metabolism; in addition,activation of colostral phagocytes is likely an additionalbreast-feeding protection mechanism for infants againstintestinal infections and may represent an important sourceof protection against infections during this phase of life andmay persist until maturity of the infant’s immune system iswell developed.

Acknowledgments

The authors are very grateful to the Postdoctoral Programof Sao Paulo State University. This research was supportedby FAPESP- The State of Sao Paulo Research Foundation(grant no. 2008/09187-8) and FAPEMAT- The State of MatoGrosso Research Foundation (grant no. 735593/2008 andNo453387/2009).

References

1. Zarban A, Taheri F, Chahkandi T, Sharifzadeh G. Pattern oftotal antioxidant capacity in human milk during the course oflactation. Iran J Pediatr 2007;17:34e40.

2. Kosek M, Bern C, Guerrant RL. The global burden of diarrhoealdisease, as estimated from studies published between 1992and 2000. Bull World Health Organ 2003;81:197e204.

3. Black RE. Epidemiology of diarrhoeal disease: implications forcontrol by vaccines. Vaccine 1993;11:100e6.

4. Honorio-Franca AC, Carvalho MP, Isaac L, Trabulsi LR, Carneiro-Sampaio MMS. Colostral mononuclear phagocytes are able tokill enteropathogenic Escherichia coli opsonized with colostralIgA. Scand J Immunol 1997;46:59e66.

5. Newburg DS. Innate immunity and human milk. J Nutr 2005;135:1308e12.

6. Hanson LA. Feeding and infant development breast-feedingand immune function. Proc Nutr Soc 2007;66:384e96.

7. Bessler HC, de Oliveira IR, Giugliano LG. Human milk glyco-proteins inhibit the adherence of Salmonella typhimurium toHeLa cells. Microbiol Immunol 2006;50:877e82.

8. Oliveira IR, Bessler HC, Bao SN, Lima RL, Giugliano RL. Inhi-bition of enterotoxicogenic Escherichia coli (ETEC) adhesionto CACO-2 cells by human milk and its immunoglobulin andnon-immunoglobulin fractions. Braz J Microbiol 2007;38:86e92.

9. Cravioto A, Tello A, Villafan H, Ruiz J, del Vedovo S, Neeser JR.Inhibition of localized adhesion of enteropathogenic Escher-ichia coli to HEp-2 cells by immunoglobulin and oligosaccharidefractions of human colostrum and breast milk. J Infect Dis1991;163:1247e55.

10. Wines BD, Hogarth PM. IgA receptors in health and disease.Tissue Antigens 2006;68:103e14.

11. Honorio-Franca AC, Launay P, Carneiro-Sampaio MMS,Monteiro RC. Colostral neutrophils express Fc alpha receptors(CD89) lacking gamma chain association and mediate nonin-flammatory properties of secretory IgA. J Leukoc Biol 2001;69:289e96.

Colostral phagocytes eliminate ETEC 7

12. Franca-Botelho AC, Honorio-Franca AC, Franca EL, Gomes MA,Costa-Cruz JM. Phagocytosis of Giardia lamblia trophozoitesby human colostral leukocytes. Acta Paediatr 2006;95:438e43.

13. Islam N, Ahmed L, Khan NI, Huque S, Begun A, Yunus ABM.Immune components (IgA, IgM, IgG immune cells) of colostrumof Bangladeshi mothers. Pediatr Int 2006;48:543e8.

14. Bhaskaram P, Reddy V. Bactericidal activity of human milkleukocytes. Acta Paediatr Scand 1981;70:87e90.

15. Tonetti MS, Imboden MA, Gerber L, Lang NP, Laissue J,Mueller C. Localized expression of mRNA for phagocyte-specific chemotactic cytokines in human periodontal infec-tions. Infect Immun 1994;62:4005e14.

16. Robinson G, Volovitz B, Passwell JH. Identification of a secre-tory IgA receptor on breast-milk macrophages: evidence forspecific activation via receptors. Pediatr Res 1991;29:429e34.

17. Rudiger A, Kuczera F, Kohler H, Schroten H. Superoxide aniongeneration in human milk macrophages: opsonin-dependentversus opsonin-independent stimulation compared with bloodmonocytes. Pediatr Res 2001;49:435e9.

18. Ozkaragoz F, Rudloff HB, Rajaraman S, Mushtaha AA,Schmalstieq FC, Godman AS. The motility of human milkmacrophages in collagen gels. Pediatr Res 1988;23:449e52.

19. Field CJ. The immunological components of human milk andtheir effect on immune development in infants. J Nutr 2005;135:1e4.

20. Bellinati-Pires R, Salgado MM, Hypolito IP, Grumach AS, Car-neiro-Sampaio MMS. Application of a fluorochrome-lysostaphinassay to the detection of phagocytic and bactericidal distur-bances in human neutrophils and monocytes. J Investig Aller-gol Clin Immunol 1995;5:337e42.

21. Pick E, Mizel D. Rapid microassays for the measurement ofsuperoxide and hydrogen peroxide production by macrophagein culture using and automatic enzyme immunoassay reader.J Immunol Methods 1981;46:211e26.

22. Babu U, Failla ML. Copper status and function of neutrophilsare reversibly depressed in marginally and severely copper-deficient rats. J Nutr 1990;120:1700e9.

23. McCord JM, Edeas MA. SOD, oxidative stress and humanpathologies: brief history and future vision. Biomed Pharmac-other 2005;59:139e42.

24. Novelli EL, Rodrıguez NL, Franca EL, Gebra LMN, Ribas BO. Highdietary carbohydrate and pancreatic lesion. Braz J Med BiolRes 1993;26:31e6.

25. Endo H, Nito C, Kamada H, Yu F, Chan PH. Reduction inoxidative stress by superoxide dismutase over expressionattenuates acute brain injury after subarachnoid hemorrhagevia activation of Akt/glycogen synthase kinase-3beta survivalsignaling. J Cereb Blood Flow Metab 2007;27:975e82.

26. Turner SM, Scott-Tucker A, Cooper LM, Henderson IR. Weaponsof mass destruction: virulence factors of the global killerenterotoxicogenic Escherichia coli. FEMS Microbiol Lett 2006;263:10e20.

27. Jones G, Steketee W, Black RE, Bhutta ZA, Morris SS. Howmany child deaths can we prevent this year? Lancet 2003;362:65e71.

28. Dizdaroglu M, Jaruga P, Birincioglu M, Rodriguez H. Freeradical-induced damage to DNA: mechanisms and measure-ment. Free Radic Biol Med 2002;32:1102e15.

29. Shen ZZ, Wu WJ, Hazen SL. Activated leukocytes oxidativelydamage DNA, RNA, and the nucleotide pool through halide-dependent formation of hydroxyl radical. Biochemistry 2000;39:5474e82.

30. Roskamp L, Pegoraro M, Luz PR, Crestani S, Vaz RS. Opsonicand nonopsonic receptors: a review. RUBS 2005;1:12e6.

31. Abdulla EM, Zaidi FE, Zaidi A. Immune factors in breast milk:a study and review. Pak J Med Sci 2005;21:178e86.

32. Buescher SE, McMlheran SM. Polymorphonuclear leukocytesand human colostrums: effects of in vitro exposure. J PaediatrGastroenterol Nutr 1993;17:424e33.

33. Vidarsson G, van de Winkel JGJ. Fc receptor and complementreceptor-mediated phagocytosis in host defence. Curr OpinInfect Dis 1998;11:271e2.

34. Keeney SE, Schmalstieg FC, Palkowetz KH, Rudloff HE, Le BM,Goldman AS. Activated neutrophils and netrophil activators inhuman milk. J Leukoc Biol 1993;54:97e104.

35. Vidarsson G, van Der Pol WL, van Den Elsen JM, Vile H,Jansen M, Dujs J, et al. Activity of human IgG and IgAsubclasses in immune defense against Neisseria meningitidisserogroup B. J Immunol 2001;166:6250e6.

36. L’Abbe MR, Friel JK. Superoxide dismutase and glutathioneperoxidase content of human milk from mothers of prematureand full-term infants during the first 3 months of lactation.J Pediatr Gastroenterol Nutr 2000;31:270e4.

37. Kasapovic J, Pejic S, Mladenovic M. Superoxide dismutaseactivity in colostrum, transitional and mature human milk.Turk J Pediatr 2005;47:343e7.

38. McCord JM. Human disease, free radicals, and the oxi-dant/antioxidant balance. Clin Biochem 1993;26:351e7.

39. Garofalo RP, Goldman AS. Cytokines, chemokines, and colony-stimulating factors in human milk: the 1997 update. BiolNeonate 1998;74:134e42.

40. Estrada-Garcıa MT, Jiang ZD, Adachi J, Mathew JJ, Dupont HL.Intestinal immunoglobulin a response to naturally acquiredenterotoxigenic Escherichia coli in US travelers to an endemicarea of Mexico. J Travel Med 2002;9:247e50.

41. Caspari GR. The influence of colostral leukocytes on the courseof an experimental Escherichia coli infection and serum anti-bodies in neonatal calves. Vet Immunol Immunopathol 1993;35:275e88.

42. Hugaes A, Brock JH, Parrott DMV, Cockburn F. The interactionof infant formula with macrophages: effect on phagocyticactivity, relationship to expression of class II MHC antigen andsurvival of orally administered macrophages in the neonatalgut. Immunology 1998;64:215e21.

43. Morrow AL, Rangel JM. Human milk protection against infec-tious diarrhea: implications for prevention and clinical care.Semin Pediatr Infect Dis 2004;15:221e8.

44. WHO. Collaborative Study Team on the Role of Breastfeeding onthe Prevention of Infant Mortality. Effect of breastfeeding oninfant and child mortality due to infectious diseases in lessdevelopedcountries: a pooledanalysis. Lancet 2000;355:451e5.