Identification of an active disaccharide unit of a glycoconjugate receptor for pneumococci attaching to human pharyngeal epithelial cells.

Glycoconjugates containing the disaccharide unit GlcNAc beta 1 leads to 3Gal beta were suggested as receptors for pneumococci adhering to human pharyngeal epithelial cells. The receptor activity was detected both by inhibition of adhesion by an excess of free oligosaccharide and by induction or increase of adhesion after coating of target cells with glycolipid. Studies with free natural and synthetic oligosaccharides identified the disaccharide GlcNAc beta 1 leads to 3Gal beta as one critical binding site. The specificity of recognition was shown inter alia by the lack of inhibitory activity of GlcNAc beta 1 leads to 4Gal beta, which differs only in the linkage of the two sugars. Specific interference with pneumococcal adhesion by administration of soluble receptor sugar may improve our understanding of the role of adhesion in vivo.

Globo-A--a new receptor specificity for attaching Escherichia coli.

Uropathogenic Escherichia coli strains designated as ONAP, based on their O negative A positive agglutination of human P1 erythrocytes, were shown to prefer the globo-A glycolipid as a receptor structure. The dependence on both the A terminal and the globoseries chain was confirmed by agglutination of human AP1, but not Ap or OP1 erythrocytes and by binding to the globo-A glycolipid on TLC plates. Neither Gal alpha 1----4Gal beta nor the A trisaccharide GalNAc alpha 1----3(Fuc alpha 1----2)Gal beta alone functioned as receptors. The bacteria thus appeared to recognize an epitope resulting from the combination of the terminal and internal structures.

Protective factors in milk and the development of the immune system.

The neonate is immature in certain immunologic functions. The slow development of secretory immunoglobin A (IgA) seems to be compensated by selective transfer of secretory IgM into exocrine secretions on mucous membranes during the first few months of life. Secretory IgA and secretory IgM antibodies against Escherichia coli and poliovirus are already found in the neonate, possibly in response to the maternal anti-idiotypic IgG antibodies transplacentally exposing the fetus. Via such a mechanism, food antibodies could occur before direct food exposure in the infant. Human milk provides large amounts of antibodies (as a crude comparison, about 50 times the amount of antibodies given to a patient with hypogammaglobulinemia). The milk antibodies, dominated by secretory IgA, protect especially against intestinal infections. The milk also contains oligosaccharide analogues to epithelial receptors for bacteria. They, as well as a number of milk components such as lactoferrin and lysozyme, may contribute to host defense. The food antibodies in human milk may influence the infant's immune response to foreign food proteins introduced during weaning.

Fever, bacteriuria and concomitant disease in children with urinary tract infection.

A total of 124 children aged 0.2 to 6 years were enrolled in a study of first time febrile urinary tract infection. The patient population was stratified in groups according to the stringency of criteria for fever and bacteriuria and the presence of concomitant disease. The major group of 88 patients consisted of children with fever greater than or equal to 38.5 degrees C measured at the hospital within 24 hours of diagnosis, bacteriuria verified by suprapubic bladder aspiration or repeated cultures of voided urine, but without concomitant disease. These children were mainly infected with attaching Escherichia coli specific for galactose alpha (1----4) beta galactose containing receptors and had laboratory evidence of inflammation. Another group of 11 children were distinguished with strictly defined bacteriuria and concomitant disease. These children were infected with nonattaching bacteria and had lower concentrations of C-reactive protein in serum and lower microsedimentation rates than the major group. Five of these children had a reduction in renal concentrating capacity. The study emphasizes the heterogeneity among patients with fever and bacteriuria but does not rule out the possibility of renal involvement in any subgroup.

Gamma-globulin treatment of recurrent acute otitis media in children.

This study examined the hypothesis that children prone to acute otitis media have a reduced concentration of circulating antibodies of the IgG2 subclass and that this defect can be compensated for by gamma-globulin treatment. Infants and children below 18 months of age with at least three episodes of acute otitis media were randomized to intramuscular gamma-globulin or no treatment and were followed for 6 months. We could demonstrate neither reduced IgG2 nor specific anti-polysaccharide antibody activity in the otitis-prone children. In contrast they had higher concentrations of IgG2 and antibodies to phosphorylcholine than did age-matched controls. There was neither a relationship between the IgG2 concentration and the number of otitis episodes prior to enrollment nor a reduction in otitis frequency in the gamma-globulin-treated group.

Role of adherence of Streptococcus pneumoniae in acute otitis media.

The adherence of Streptococcus pneumoniae to human nasopharyngeal epithelial cells was studied as a possible determinant in the development of acute otitis media (AOM). Pneumococcal isolates were obtained from the nasopharynx (NP) and middle ear fluid of infants followed from birth in a prospective study of pneumococcal carriage and infection. The adherence of 33 middle ear fluid isolates from 19 infants with AOM was compared with 143 strains recovered from NP cultures taken from each child both at the time of their acute infections and on other occasions. We studied 171 NP isolates from 29 "carrier" infants, who had no pneumococcal infections, for comparison. Adherence properties were not associated with any particular pneumococcal capsular types, nor were adherent strains more frequent among infants with AOM. There was no evidence to support the hypothesis that pneumococci associated with AOM have a special propensity for adherence. Adherence was a frequent characteristic of pneumococci recovered from the NP, especially in connection with upper respiratory tract infection, and may be required for the establishment of colonization but was not a property that discriminated between carriage strains and those causing AOM.

Mucosal immunity.

Mucosal defense is provided by a number of host factors countering the specific virulence factors of the many microorganisms infecting the mucous membranes. Secretory IgA antibodies presumably play an important role. Increase of the sIgA antibodies may most advantageously be attained by parenteral immunization, following mucosal priming. This was demonstrated in a rat model, where it was also noted that antigen injection into PP induced high milk IgA antibody levels. In man, parenteral vaccination against polio increased the sIgA antibody levels in the milk of mothers previously exposed naturally to the poliovirus. The response was relatively short-lived. In the previously unexposed, there was little or no response. By contrast peroral immunization with live poliovirus vaccine did not increase, or even decrease, the milk sIgA poliovirus antibody levels. Although salivary sIgA antibodies against antigens of colonizing E. coli appear during the first days of life, they are slow to increase. This deficiency is richly compensated for by all the sIgA antibodies that are provided the baby through the milk. No transfer of dimeric IgA into the milk could be shown in lactating rats, in contrast to what has been reported in mice. There is no evidence for a contribution to milk sIgA from serum in man. Close to parturition, human milk often contains some 7S IgA and various sizes of free SC, in addition to the dominating 11S sIgA. A few days later there is almost exclusively monomeric SC and 11S sIgA. IgG antibodies also play a role at the mucosal level. IgG2 antibodies against the bacterial polysaccharide capsule are as slow to appear as sIgA in ontogeny, possibly explaining the prevalence of infections with encapsulated bacteria and the poor response to polysaccharide vaccines in early childhood. Other defense factors preventing infections by way of mucous membranes may be important. Thus, oligosaccharides present in human milk seem to specifically prevent pneumococcal attachment to retropharyngeal cells. This anti-attachment capacity, in addition to that provided by milk and salivary IgA antibodies, may explain why breast-fed babies have less otitis media than formula-fed ones.

Secretory IgA antibodies to enterobacterial virulence antigens: their induction and possible relevance.

1) Milk and salivary s-IgA antibodies are via the homing of IgA producing cells from the Peyer's patches closely connected with antigenic stimuli in the intestine. This explains the presence in human milk of s-IgA antibodies against E. coli O and K antigens, V. cholerae and Shigella O antigens, E. coli and V. cholerae enterotoxins. These secretory antibodies can be induced by intestinal exposure and boosted by parenteral vaccination. 2) Preliminary data suggest that the IgA response in the urinary tract and possibly in the lung may be involved in the homing mechanism as well. 3) The protective role of the milk s-IgA antibodies to enterobacterial virulence antigens is strongly suggested, as is the protection mediated by urinary antibodies against urinary tract infections.

Influence of P blood group phenotype on susceptibility to urinary tract infection.

Bacterial attachment is an important event in the pathogenesis of urinary tract infection (UTI). Increased receptivity on the host cells has been suggested influence proneness to infection. The dual function of the globoseries of glycolipids both as receptors for attaching E. coli and as P blood group antigens lead us to examine the P blood group phenotype distribution in UTI prone patient populations. A correlation between the P1 blood group phenotype and susceptibility to UTI was found. Patients with recurrent pyelonephritis had 74/79 (94%), P1 compared to 75% in healthy controls. In contrast patients with asymptomatic bacteriuria (ABU) had a reduced frequency of P1, 43/74 (58%). P1 and P2 individuals differ in amount and composition of the globoseries of glycolipids on their erythrocytes. A similar difference in other tissues, e.g. uroepithelial cells might explain the association of P1 with UTI. There was, however, no significant difference in bacterial adherence to uroepithelial cells from P1 and P2 individuals. Other mechanisms explaining the increase in P1 individuals in recurrent pyelonephritis are discussed.

Escherichia coli pili as possible mediators of attachment to human urinary tract epithelial cells.

Presence of pili of fimbriae on Escherichia coli bacteria isolated from the urine of patients with urinary tract infection was related to the ability of the bacteria to attach to human uroepithelial cells. Piliated E. coli strains agglutinated guinea pig erythrocytes. D-Mannose and alpha-methyl-D-mannopyranoside inhibited this agglutination with all but one of the 12 strains tested. D-Mannose, D-galactose, alpha-methyl-D-mannopyranoside, and L-fucose did not afect attachment of piliated strains to uroepithelial cells. Heating as well as washing of piliated strains caused a parallel decrease of piliation and adhesive ability. Growth in glucose-enriched medium increased capsule formation but decreased piliation and adhesion. Capsulated strains retained their adhesive ability provided that pili extended outside the capsule.

Attachment of Proteus mirabilis to human urinary sediment epithelial cells in vitro is different from that of Escherichia coli.

The in vitro attachment of 335 Proteus mirabilis strains from various human sources to human urinary tract epithelial cells was measured. No significant difference in adhesive capacity was found between P. mirabilis strains isolated from the blood of 89 patients with bacteremia, the stools of 36 healthy subjects and 56 patients with diarrhea, and the urine of 62 adults and 92 children with bacteriuria. High mean adhesion values were observed in all groups. The P. mirabilis strains attached only to squamous cells and not to transitional epithelial cells, whereas most of the Escherichia coli strains tested attached to both cell types; strains from patients with acute pyelonephritis attached more often than those from patients with acute cystitis or asymptomatic bacteriuria. The attachment of P. mirabilis to squamous epithelial cells was high about day 15 of the menstrual cycle of the epithelial cell donor, but low at the beginning and the end of the cycle. In contrast, the attachment of E. coli to squamous and transitional epithelial cells did not vary significantly with the menstrual cycle of the cell donor. Differences in adhesion characteristics of E. coli and P. mirabilis may relate to the differences in clinical appearance of urinary tract infections produced by the two organisms.

Host resistance to mucosal gram-negative infection. Susceptibility of lipopolysaccharide nonresponder mice.

The effect of Lps on the resistance of mice to gram-negative infection was compared in two genetically different backgrounds; C3H and C57BL. To mimic the natural sequence of pathogenetic events, infection was induced via a mucosal surface (intravesically), with Escherichia coli which remained at the mucosal site and Salmonella typhimurium which invaded to e.g., livers and spleens. Susceptibility was assessed as the bacterial persistence in kidneys, bladders, livers, and spleens at various times after infection. The initial clearance of both bacterial species from the mucosal site was significantly impaired in Lpsd mice both in the C3H and C57BL backgrounds. In the C57BL mice, additional unknown determinants conferred increased resistance to mucosal infection compared to the C3H mouse. For S. typhimurium, these resistance factors and alleles at the Lps locus dominated over Ity as determinants of resistance to mucosal infection. The Itys genotype conferred a significant increase in the susceptibility only to systemic infection, especially in the Lpsd, Itys mice. These results demonstrate an important difference between the genetic determinants of host resistance at mucosal and systemic sites, and emphasize the role of LPS induced host defense mechanisms for bacterial clearance from mucosal surfaces.

Role of attachment for the virulence of Streptococcus pneumoniae.

Adherence of microorganisms to mucosal surfaces is a general phenomenon among microorganisms infecting the human host. Its role for persistence and colonization as well as production of local inflammation is well established. This paper describes the adhesion of Streptococcus pneumoniae to human epithelial cells. Strains from various anatomical sites or diseases are compared for attaching capacity. Isolates from the same host but at different times are also compared. The molecular mechanisms, the so-called adhesin-receptor interactions, are partially described. The pneumococcus recognizes a sugar sequence; GlcNAc beta 1-3Gal; on the surface of the host epithelial cell. Glycoconjugates containing this disaccharide act as receptors for adhering pneumococci. The adhesin in pneumococcal attachment is less well characterized. It is a heat and trypsin sensitive component, most likely a peptide, which forms a bridge between the receptor and an anchoring site in the pneumococcal cell wall. Receptor active saccharides are part of the adhesion-inhibitory activity found in human milk.

Evidence for separate genetic defects in C3H/HeJ and C3HeB/FeJ mice, that affect susceptibility to gram-negative infections.

Past studies have suggested a linkage between susceptibility to Salmonella typhimurium infection and the Lpsd genotype in C3H mice. Recently, this linkage was questioned by the finding that C3HeB/FeJ mice (Lpsn,Lpsn) were highly susceptible to systemic S. typhimurium infection. The present study shows a marked difference between C3H/HeJ and C3HeB/FeJ in their susceptibility to Gram-negative urinary tract infection. The number of E. coli and S. typhimurium recovered from the kidneys 24 hr after infection was 70 to 100 times higher in C3H/HeJ than in C3HeB/FeJ or C3H/HeN mice. Subsequently, in C3HeB/FeJ mice S. typhimurium multiplied to the level of C3H/HeJ mice, resulting in a shorter mean survival time of C3H/HeJ and C3HeB/FeJ compared with C3H/HeN mice. In contrast, E. coli remained localized to the urinary tract of C3H/HeJ mice but were eliminated from C3HeB/FeJ and C3H/HeN mice. Thus, experimental E. coli urinary tract infection appears to provide a method to differentiate the genetic defects of C3H/HeJ and C3HeB/FeJ mice. The results support an influence of the Lpsd genotype on clearance of Gram-negative bacteria from the kidneys of C3H mice.

Host response in women with symptomatic urinary tract infection.

The agreement between clinical signs and host response was analysed in 174 women with symptomatic urinary tract infection. C-reactive protein (CRP) confirmed the clinical diagnosis in that 94% of non-pregnant and 91% of pregnant women with acute pyelonephritis had serum levels greater than or equal to 30 mg/l, compared with only 5% of cystitis patients. There was a significant increase in the erythrocyte sedimentation rate (ESR) and reduction of the renal concentrating capacity in patients with acute pyelonephritis, although the overlap with the cystitis group was greater than for CRP. The transient decrease in urine osmolality was unrelated to age, as were CRP, ESR and the total white blood cell count. Pregnant women had higher ESR but lower CRP levels than non-pregnant women with acute pyelonephritis. The renal concentrating capacity was more reduced in those infected with Escherichia coli expressing adhesins specifically recognizing Gal alpha 1----4Gal beta-containing receptors on uroepithelial cells.

Adhesion of Escherichia coli to human uroepithelial cells in vitro.

Optimal conditions for in vitro adherence of Escherichia coli to uroepithelial cells, previously shown to more efficient for strains causing acute symptomatic than that for strains causing "asymptomatic" urinary tract infections, were investigated. Uroepithelial cells from fresh morning urine of healthy individuals and E. coli bacteria from patients with various forms of urinary tract infeciton were used. Adhesion was found to vary, between individuals and epithelial cell types, with epithelial cell viability, bacterial cultivation medium and growth phase, number of bacteria added to the epithelial cells, and incubation time and temperature. Adhesion was also influenced by variations in pH and osmolarity. Optimal test conditions were obtained with post-log-phase bacterial cultures grown on nutrient broth when 10(8) bacteria were added to 10(5) epithelial cells and incubated for 60 min. Considerable variation was found between experiments done on different days, whereas the variation between duplicates was small. The method described may provide a useful tool in the study of the host-parasite relationship in urinary tract infections.

Adhesiveness to urinary tract epithelial cells of fecal and urinary Escherichia coli isolates from patients with symptomatic urinary tract infections or asymptomatic bacteriuria of varying duration.

Adhesiveness to human urinary tract epithelial cells was high for Escherichia coli strains isolated from patients with acute pyelonephritis and acute cystitis, and low for asymptomatic bacteriuria strains detected at screening. Escherichia coli bacteria causing asymptomatic reinfections, detected near the onset of bacteriuria, adhered more than those detected at screening. No difference in the adhesive ability was found between fecal isolates of the strain causing urinary tract infection, isolated at or before onset of bacteriuria, and the urinary strain in symptomatic or asymptomatic patients. Normal fecal Escherichia coli from non-bacteriuric patients adhered less than all other strains tested.

Urinary tract infections caused by Proteus mirabilis in children. The antibody response to O and H antigens and Tamm-Horsfall protein and bacterial adherence to uro-epithelium.

Sera from seven girls with acute symptomatic pyelonephritis and nine children with acute symptomatic cystitis caused by Proteus mirabilis were analysed for antibodies against the bacterial O and H1 antigens and the Tamm-Horsfall protein. An increase in antibody levels against O antigen and Tamm-Horsfall protein was noted only in patients with acute pyelonephritis indicating that antibody determinations can be useful in differentiating between upper and lower urinary tract infection caused by Proteus in similarity to those caused by E. coli. In contrast no difference in adhesive ability was noted comparing Proteus strains causing acute pyelonephritis or cystitis.

New Method for isolation of immunologically pure pili from Escherichia coli.

A new technique for purification of bacterial pili was developed and applied to Escherichia coli strains isolated from the urine of patients with symptomatic urinary tract infections. After mechanical detachment from the bacterial cells, the pili were concentrated by precipitation with ammonium sulfate, dialyzed, and solubilized in buffer containing deoxycholate. The fraction containing the pili was purufied further by ultracentrifugation in a sucrose gradient and by elution through a Sepharose 4B column in 6 M urea buffer. The pilus filaments were not dissociated by concentrated urea and were eluted in the void volume of the column. The purified pili had a molecular weight of 17,000. The isoelectric point of the pili from one of the strains was 4.9, and about 43% of the amino acids were hydrophobic. Hyperimmunesera raised in rabbits against the purified pili did not contain detectable antibodies to the lipopolysaccharide O antigen or to the capsular polysaccharide K antigen of the homologous strain. The pili obtained by this purification procedure are free from other detectable bacterial surface antigens, and the purified pilus filaments are of relatively homogeneous size. This procedure enables purification of the pili also from flagellated strains.

Attachment of Escherichia coli via mannose- or Gal alpha 1----4Gal beta-containing receptors to human colonic epithelial cells.

The role of bacterial adhesion for the maintenance of the large-intestinal microflora has not been established. In this study, colonic cells from the adenocarcinoma cell line HT-29 or from surgical specimens were tested for the ability to bind Escherichia coli. The E. coli strains were manipulated by transformation or by mutagenesis to express either mannose-specific type 1 fimbriae (strains 506 MS and HU742) or Gal alpha 1----4Gal beta-specific P fimbriae (506 MR and HU824). Binding to HT-29 cells was seen with strains of either receptor specificity and was inhibited by alpha-methyl mannoside or globotetraosylceramide (GalNAc beta 1----3Gal alpha 1----4Gal beta 1----4Glc-ceramide), respectively. The Gal alpha 1----4Gal beta-specific strains interacted with a loosely surface-associated substance, which was sensitive to mechanical treatment and incubation at 37 degrees C, while the mannose-specific strains bound both directly to the cell and to the loosely associated substance. Isolated colonic epithelial cells bound the mannose-specific bacteria in high numbers, while the attachment of the Gal alpha 1----4Gal beta-specific strains depended on the elution method. Cells eluted sequentially with magnetic stirring were unable to bind the Gal alpha 1----4Gal beta-specific bacteria, while elution by a more gentle method resulted in binding of these strains to material loosely associated with the epithelial cells. Thus, the binding pattern of isolated colonic epithelial cells paralleled that of the HT-29 cell line. Conceivably, binding to mannose- and Gal alpha 1----4Gal beta-containing receptors could contribute to the maintenance of E. coli in the human large intestine.