Detection of campylobacter by immunofluorescence in stools and rectal biopsies of patients with diarrhoea.

Rabbit antiserum, elicited by the intravenous injection of a strain of Campylobacter jejuni heated to 100 degrees C, cross reacted strongly with all other thermophilic campylobacters tested as well as with "C pyloridis" and could be detected by indirect fluorescence with labelled anti-rabbit serum. Antisera to formalin killed cells did not do so. The correlation of positive stool culture with positive immunofluorescence of stools and rectal biopsies from patients with diarrhoea was 70-80%. Some inconsistent, weak reactions showing differently shaped organisms have been seen with some strains of Bacteroides fragilis. Wolinella spp reacted weakly, but one strain of Vibrio cholerae tested did not. Other intestinal organisms, commensals, and pathogens tested were negative.

Microbial and metabolic profile of achlorhydric stomach: comparison of pernicious anaemia and hypogammaglobulinaemia.

The microbial flora and some of its metabolites and enzymes in the stomach were compared in patients with achlorhydria, pernicious anaemia, and primary hypogammaglobulinaemia and in patients with dyspepsia with normal gastric acidity. Detailed analysis of the flora of the gastric juice and of the mucosa from the antrum, body, and fundus in six patients with hypogammaglobulinaemia (mean pH 8.2), seven patients with pernicious anaemia (mean pH 7.3), and five patients with dyspepsia (mean pH 1.9) yielded 22 different genera of bacteria, mainly from the patients with achlorhydria, the most common being streptococci, micrococci, staphylococci, veillonella, and lactobacilli. A similar flora was found associated with the mucosa at all three sites. Various metabolites were also looked for. beta Glucoronidase and C14 lipase were found in patients with hypogammaglobulinaemia but not in those with pernicious anaemia or dyspepsia. Volatile fatty acids were not found. Relatively high concentrations of ethanol were found in the patients with hypogammaglobulinaemia compared with those with pernicious anaemia (p = 0.02). Similar concentrations of dimethylamine were found in all three groups, but the concentrations of trimethylamine were much higher in patients with pernicious anaemia and hypogammaglobulinaemia. The high concentrations of some microbial enzymes and ethanol differentiated the group with hypogammaglobulinaemia from the rest, and these may bear some relation to the high incidence of gastric cancer in patients with hypogammaglobulinaemia.

The mechanism of protection of infant mice from intestinal colonisation with Campylobacter jejuni.

BALB/c mice, vaccinated intraperitoneally with a heat-killed (62 degrees C) suspension of Campylobacter jejuni before mating, completely protect c. 90% of their own infants from intestinal colonisation. This protection has now been investigated further in fostering experiments. Fostering by vaccinated dams within the first 24 h of life prevented intestinal colonisation in 50% of infants from non-vaccinated dams, and reduced colonisation in a further 25%. Infants from vaccinated dams, even if allowed to receive their own mothers' colostrum and milk, became susceptible to challenge when subsequently fostered by non-vaccinated dams. Immunity in experimentally infected infant mice depended upon the consumption of immune milk at and after the time of challenge. High concentrations of IgG antibodies specific for C. jejuni were found in the serum and mammary secretion of vaccinated dams, but there was very little specific IgA antibody.

Cross-protection of infant mice against intestinal colonisation by Campylobacter jejuni: importance of heat-labile serotyping (Lior) antigens.

An association of the heat-labile antigens detected by the Lior serotyping scheme with ability to protect infant mice against gastrointestinal colonisation with Campylobacter jejuni has been established. Overall, 39 (57%) of 68 infant mice challenged with a heterologous strain of the same Lior serotype as the vaccine strain were protected, compared with 40 (85%) of 47 infants protected against a homologous challenge. In contrast, none of the infant mice challenged with a strain carrying the same heat-stable antigens (i.e., of the same Penner serotype as the vaccine strain) were protected.

Nitrate- and nitrite-reducing bacteria in the achlorhydric stomach.

The microbial composition of samples of gastric juice from eight achlorhydric patients was determined by aerobic and rigorously anaerobic culture techniques. Bacteria from 16 genera were commonly isolated, but representatives of only three genera, (streptococci, neisseriae and haemophili) were isolated from every patient. Nitrate and nitrite were both reduced by veillonellae, haemophili, staphylococci, corynebacteria, lactobacilli, flavobacteria and fusobacteria, but the potential rate of nitrate reduction by suspensions of veillonellae, Haemophilus parainfluenzae and members of the Enterobacteriaceae were up to ten times more rapid than the rate of nitrite reduction. Conversely, although all Neisseria spp. reduced nitrite only some strains reduced nitrate. Streptococci did not reduce nitrate. Streptococcus sanguis reduced nitrite when grown with haematin; other streptococci did not reduce nitrite. Bacterial nitrate and nitrite reduction were active over the pH range 6-8, similar to the pH range of the achlorhydric stomach. From a knowledge of the composition of the bacterial flora and their potential rates of nitrate and nitrite reduction under prevailing conditions, predictions were made about the tendency of nitrite to accumulate during nitrate reduction. Studies of the transient accumulation of nitrite by mixed cultures of H. parainfluenzae and N. subflava were consistent with these predictions. Haemophili and veillonellae could be responsible for the accumulation of nitrite in the gastric juice of some patients, whereas streptococci and neisseriae would tend to remove nitrite from the stomach as rapidly as it formed.

Passive protection of mice against intracerebral infections with Bordetella pertussis.

The passive protection of mice against an intracerebral infection with Bordetella pertussis, by antiserum introduced directly into the brain with the infecting organisms, was compared with the protection afforded by intraperitoneal antiserum. The antibody effective by the intracerebral route is that which is adsorbed onto the infecting organisms, although it does not affect the viability of the organisms in vitro in the absence of complement.PASSIVE PROTECTION AGAINST ORGANISMS INTRODUCED INTRACEREBRALLY TAKES PLACE IN ONE OF TWO WAYS, DEPENDING ON THE SIZE OF THE CHALLENGE: (1) after 3-4 days' growth, the growth rate declines so that the number of organisms does not reach the figure lethal for the mouse; (2) the organisms do not appear to multiply, as their numbers decline from the moment of injection, so that the brain is sterile after 2 days.Many of the mice protected against a challenge of 50,000 organisms (ca. 100 LD 50) by intraperitoneal or intracerebral antiserum recover by the first mechanism. The second mechanism operates after a smaller challenge of 5000 organisms (ca. 10 LD 50), irrespective of whether the antiserum is given intracerebrally with the challenge, or intraperitoneally within several hours of challenge. Too much antiserum given intracerebrally with a 50,000 challenge, but not with a 5000 challenge, inhibits protection.In some mice, virulent antibody-treated organisms which have not been killed, grow very slowly over a long period, but are eventually eliminated from the brain.

Pertussis antibodies in the sera of children exposed to Bordetella pertussis by vaccination or infection.

Low agglutinin titres to pertussis suspensions were found in 99% of sera from a group comprising healthy adults and non-vaccinated, non-infected infants of 1-6 months of age. These are attributable to agglutinins to heat-stable antigens and/or heat labile agglutinogen 1, and cross-absorption tests must be done on the sera in order to distinguish between the two. Agglutinins to agglutinogens 2 and 3 were found in only about 20% of adult sera. Bactericidal antibody was low in titre or absent in all sera from non-exposed individuals.Raised bactericidal antibody titres and the presence of agglutinins 2 and 3 were attributed to exposure to Bordetella pertussis antigens, either as vaccine or as infection. The variation, amongst both vaccinated and infected children, was very great. A vaccinated child who became ill responded to the infection in much the same way as a non-vaccinated child. We were unable to relate the immunity of the child to the titres either of agglutinins or of the bactericidal antibody.The protective ability of sera from vaccinated or infected children measured in mice against small, lethal brain infections was also unrelated to the state of immunity in the children, but this protective ability was correlated with the complement-mediated bactericidal antibody titres of the sera.The distribution of agglutinins, bactericidal antibody, and anti-haemagglutinin in serum IgG and IgM was different in vaccinated and infected children.

The protection of infant mice from colonization with Campylobacter jejuni by vaccination of the dams.

Intraperitoneal vaccination of female mice, before mating, with a whole cell, heat-killed (62 degrees C) vaccine of Campylobacter jejuni allowed the mother to confer immunity to her young, challenged orally 4-6 days after birth with the homologous strain. There was no protection against a strain of another serotype. Heating the vaccine to 100 degrees C destroyed its protective properties. A vaccine prepared from an aflagellate variant of the original strain was as protective as the original vaccine against challenge with the flagellated strain. Anti-flagellar serum antibody titres of the dams did not correlate with protection of their young.

The influence of the route of immunization on the protection of mice infected intracerebrally with Bordetella pertussis.

The development of immunity in mice to Bordetella pertussis induced by intracerebral, intravenous or intraperitoneal vaccination was analysed in terms of the viable bacteria in the brain after intracerebral challenge, the serum antibodies, and protection against the sublethal infection of the lung that follows intranasal inoculation.A vaccine introduced intracerebrally was five to ten times more effective than that given intraperitoneally or intravenously, as measured for each route by the amount of vaccine required to protect half the mice against an intracerebral challenge 14 days later (ImD 50). Intracerebral vaccination induced higher antibody titres than vaccination by the other two routes. The survival of infected mice given 1-3 ImD 50 doses of vaccine intracerebrally 14 days before, followed a pattern similar to that after intraperitoneal or intravenous vaccination with up to 10 ImD 50 of vaccine: the numbers of organisms increased for 3 days and then declined. Injection of about four ImD 50 of vaccine intracerebrally produced a local immunity, resulting in an immediate kill of challenge organisms given 14 days later. Such an effect following intraperitoneal vaccination was achieved only against challenges with an avirulent strain. It is suggested that better stimulation of circulating antibody and local immunity in the brain together account for the better protection induced by intracerebral vaccine.Immunity to an intracerebral infection appears therefore to have at least three components, each specific for pertussis. The first, like that induced by intraperitoneal and intravenous vaccination, reaches a maximum in 2 or 3 weeks and is probably an expression of a general response by the animal operating not earlier than 3 days after infection. The second is a local immunity, appearing after the same interval. The third is a short-lived local immunity which has been described by previous workers; it immediately follows the injection intracerebrally of ten times less vaccine than that needed to protect against a challenge 14 days later and lasts only 2-3 days. The second and third types result in immediate sterilization of the infection.Mice recovering from sublethal brain infection with avirulent organisms were immune to a second infection with a virulent organism, but this was achieved not by the ability to kill the re-infecting organisms immediately on injection into the brain, but only after the 3-4 days lag such as follows intraperitoneal vaccination.

Stimulation of the development of bacteriostatic activity of guinea-pig milk against Escherichia coli.

The bacteriostatic activity of guinea-pig milk against various strains of Escherichia coli has been examined. Milk collected from sows suckling normal young was usually inactive, but the activity of milk from sows suckling young which had been orally infected with Esch. coli was significantly increased. The increase occurred in 2 phases: the first was found as early as 24 h after infection, suggesting stimulation via the teat canal (diathelic), but lasted only 2-3 days; the second occurred from 10 days post-infection onwards, and lasted until the end of lactation. The occasional bacteriostatic activity of milk from sows with normal young was not correlated with the presence in the faeces of naturally occurring Enterobacteriaceae, including Esch. coli.

Taxonomic distribution of the antigen eliciting bactericidal antibody for Bordetella pertussis.

Strains of Bordetella pertussis varied in their ability to elicit (in mice) an antibody bactericidal for an antiserum-sensitive strain of B. pertussis, although antibody was usually detectable after only one injection. High titres were produced by a course of seven injections with all strains of B. pertussis tested (six of phase I and three of phase IV) but not with three strains of other Bordetella species nor with two unrelated organisms, a finding of possible taxonomic value. Preliminary investigations have not revealed whether strain vaiations are due to quantitative or qualitative differences in either the bacterial lipopolysaccharide or the carrier protein necessary for antibody production, or whether they may be due to differences in heat lability of 'bactericidal antigen'.

Antibodies to Escherichia coli O antigens and the in-vitro bacteriostatic properties of human milk and its IgA.

Only a very small part of the iron-reversed bacteriostatic activity of milk against Escherichia coli, demonstrable in vitro, is due to its anti-O antibody. Most of its growth-inhibitory activity is due to another lactoferrin-dependent, non-specific system. IgA prepared from milk is bacteriostatic for E. coli in the presence of lactoferrin, if it contains O-antibody for the indicator strain and if the strain is susceptible. Susceptibility depends to some extent on virulence, since those inhibited by IgA antibody to their own O-antigens were enteropathogenic or enterotoxigenic, whereas the growth of commensal strains was inhibited only slightly or not at all.

Bacteriostasis of Escherichia coli by milk. III. The activity and stability of early, transitional and mature human milk collected locally.

Milk from 150 local mothers has been assayed for bacteriostatic activity for milk-sensitive and milk-resistant indicator strains of Escherichia coli. Activity is greatest in colostrum which is active directly against all strains of E. coli. One week after delivery of the baby, milk is active against the milk-sensitive strain and becomes active against the milk-resistant strain in the presence of physiological amounts of bicarbonate and iron-binding protein. This activity decreases within 2--4 days on keeping milk unheated at 4 degrees C but is preserved for at least 4 months and often up to 2 years in milk heated to 56 degrees C then stored at 4 degrees C or in milk frozen, unheated, at -28 degrees C provided it is not repeatedly thawed and frozen. Later lactation milks are usually indistinguishable in activity from 1-week post-partum milk but may be less stable on storage particularly if frozen. Lyophyilization in vacuo preserves activity of early-lactation milk for at least 6 months. Heating milk to above 65 degrees C causes a progressive loss of activity which can be partially restored by adding bicarbonate and iron-binding protein. Iron abolishes the activity of milk and reduces that of colostrum.

Bacteriostasis of Escherichia coli by milk. IV. The bacteriostatic antibody of human milk.

Bacteriostatic activity for milk-sensitive and milk-resistant strains of Escherichia coli is reduced when IgA is removed from milk by precipitation. Lysozyme is not involved in bacteriostasis and can be removed without loss of activity; heavy bentonite absorption however removes some lactoferrin causing partial loss of activity. The heat-labile antigen eliciting bacteriostatic antibody for E. coli is present in milk-sensitive and milk-resistant strains and in some other Enterobacteriaceae, e.g. salmonella; it cross reacts with the antigen in others, e.g. proteus and enterobacter. The antibody is therefore likely to be present in all human milk as a result of the normal commensal gut flora and with widespread activity.

The effect of the antigen which elicits the bactericidal antibody and of the mouse-protective antigen on the growth of Bordetella pertussis in the mouse brain.

The effect of antigens of Bordetella pertussis and their antibodies on brain infections by B. pertussis in mice are suppression of an infection immediately, so that the initial 90% loss due to leakage from the brain is maintained or the numbers of bacteria are reduced even further, sometimes with complete sterilization particularly after a small lethal challenge of 10 LD 50 (mechanism 1), and a delayed antibacterial activity in vivo which does not begin until 3 days after challenge (mechanism 2). The first, immediate reaction is over in 2-3 days; the second is maintained from 3-4 days onwards, and results in elimination of the bacteria and protection of mice.The parts played in vivo in overcoming infection in these two ways by two antigens and their respective antibodies have been investigated. These antigens are a lipopolysaccharide capable of eliciting an antibody which is bactericidal in vitro in the presence of complement called the ;bactericidal antigen', and the mouse protective antigen.Considering first passive immunity, bactericidal antibody elicited by isolated antigen, and of high titre in vitro, is only very weakly active by mechanism (1) in vivo. Brains are seldom sterilized and mice not therefore protected. Antisera to whole cell vaccines whether they contain the ;bactericidal antigen' or not, or the protective antigen or not can more easily reduce infections by mechanism (1), eliminating small lethal challenges in some mice which are protected. A passive, intracerebrally protective antibody (PIPA) different from other known antibodies, has been postulated to account for this. Antisera to whole cell vaccine which is protective as defined in the potency assay, can, in addition to this, protect mice by mechanism (2) not only against 10 LD 50 but also 100 LD 50 challenge, and is the only antibody which can do this.These antibodies have been investigated by injecting them with the challenging organisms. The antibody effects described above are given by antisera stimulated by several injections and also by the concentrated serum immunoglobulins of once vaccinated mice. The antibody, which is bactericidal in vitro only, is in the 7 S globulin fraction of the serum of once vaccinated mice. The protective antibody capable of overcoming small and large challenges is in the 19 S and 11 S globulins. The antibody, PIPA, protecting against small lethal challenges only is in the fraction A(2) containing mainly 11 S globulin.In active immunization experiments the suppression of infection which immediately follows intracerebral vaccination, but which only lasts 2-3 days (mechanism 1), is not dependent on either ;bactericidal' or protective antigens but on a component present in all our whole cell vaccines. Vaccines which also had protective antigen eliminated the remaining infection at 4-6 days after challenge by mechanism (2).As in passive immunity, only the protective antigen can completely overcome 100 LD 50. Suppression of a small, lethal, intracerebral infection given 14 days after intraperitoneal vaccination by mechanism (1) may however be correlated with protective antigen.

The effect of freezing and pasteurizing bovine milk on its ability to protect neonatal guinea-pigs against colonization of the small intestine by Escherichia coli.

The ability of frequent feeding of bovine milk diets to prevent the colonization of the small intestines of newborn guinea-pigs with orally inoculated Escherichia coli was tested. At 3--4 days small intestinal samples from suckled controls were frequently sterile or were colonized with only very low numbers of Esch. coli. No bovine milk diet exhibited a significant "protective" effect but the diets could, however, be ranged in order of effectiveness in decreasing colonization by Esch. coli. Raw, fresh bovine milk was best, followed by milk pasteurized at 56 degrees or 63 degrees, then boiled milk; frozen milk was the worst. Because of this last finding, neither the bacteriostatic lactoferrin-dependent activity nor the lactoperoxidase could be correlated with the ability to decrease the colonization of the small intestines by Esch. coli.

Differences in inhibition of the growth of commensal and enteropathogenic strains of Escherichia coli by lactotransferrin and secretory immunoglobulin A isolated from human milk.

Immunoglobulins from bovine and human colostrum and milk and lactotransferrin (LTF) from human milk were investigated for bacteriostatic activity against Escherichia coli growing in a tissue culture medium. When tested separately, LTF or secretory immunoglobulin A (sIgA) from pooled human milk showed only slight bacteriostatic activity against human commensal or enteropathogenic strains of E. coli. Together, they had a considerable bacteriostatic effect, but only against strains of enteropathogenic serotype. This activity of the sIgA from pooled human milk was consistent for all enteropathogenic serotypes tested, but sIgA isolated from individual milk samples was inactive against some serotypes, and this specificity was associated with antibody to the O antigens. The activity of the sIgA was stable to heat at 56 degrees for 2 h but was lost progressively on heating at 65 degrees for 10 min or longer. Bovine colostral IgGl was without bacteriostatic effect alone. Together with LTF, it was active against a strain pathogenic to calves but not against human enteropathogenic strains. Tests on rabbit antisera raised against commensal enteropathogenic strains of E. coli showed that for the enteropathogens the bacteriostatic activity (in association with LTF) was high and was specific for the serotype of the eliciting strain, but bacteriostatic activity was low or absent in the antisera to commensal strains in spite of the presence of high titres of agglutinating antibodies to these strains.

Bacteriostasis of Escherichia coli by milk. V. The bacteriostatic properties of milk of West African mothers in the Gambia: in-vitro studies.

Bacteriostatic activity was measured in 244 specimens of milk collected during 1977 throughout lactation of up to one year from 78 mothers; the activity varied from very good to fair and only seven were inactive. There was a wider range of activity than was found previously in milk from English mothers. Activity usually fell slowly during lactation but some of the Gambian mothers produced milk of very high activity, like that of colostrum into the second week of lactation, and two mothers did so at six and nine months; other mothers produced good-activity milk throughout lactation. The bacteriostatic activity varied little with the season but slight decreases from that expected were found after the high incidence of infant diarrhoea towards the end of the rainy season. The bacteriostatic activity of most of the milk tested could be prevented by iron salts but that of colostrum and some of the milks with high activity could not. Only these highly active colostra and milks were inhibitory in vitro when the inoculum was increased from 10(4) to 10(6) organisms per ml. These and less active milks were able to inhibit the smaller, standard inoculum for longer than 3 h with the addition of bicarbonate and extra iron-binding protein at the concentrations likely to be present in vivo. Both commensal and pathogenic E. coli were inhibited to a similar degree by these milks and there was no evidence of serotype specificity.

PIP: Bacteriostatic activity was measured in 244 specimens of milk from 78 mothers collected during 1977 throughout lactation of up to 1 year; the activity varied from very good to fair and only 7 were inactive. There was a wider range of activity than was found previous in milk from English mothers. Activity usually fell slowly during lactation but some of the Gambian mothers produced milk of very high activity, like that of colostrum, into the 2nd week of lactation, and 2 mothers did so at 6 and 9 months; others motherss produced good-activity milk throughout lactation. The bacteriostatic activity varied little with the season but slight decreases from that expected were found after the high incidence of infant diarrhea towards the end of the rainy season. The bacteriostatic activity of most of the milk tested could be prevented by iron salts but that of colostrum and some of the milks with high activity could not. Only these highly active colostra and milks were inhibitory in vitro when the inoculum was increased from 104 to 106 organisms/ml. These and less active milks were able to inhibit the smaller, standard inoculum for longer than 3 hours with the addition of bicarbonate and extra iron-binding protein at the concentrations likely to be present in vivo. Both commensal and pathogenic E. coli were inhibited to a similar degree by these milks and there was no evidence of serotype specificity.

Investigations on the role of flagella in the colonization of infant mice with Campylobacter jejuni and attachment of Campylobacter jejuni to human epithelial cell lines.

The biochemical and biological properties of the flagella of Campylobacter jejuni have been investigated using two variants selected from a flagellate, motile clinical isolate (strain 81116): a flagellate, non-motile variant (SF-1) and an aflagellate variant (SF-2). Phenotypic and biochemical analysis of the strains and amino acid analysis of the isolated flagella suggest that the variants differed from the wild-type strain only in the absence of flagella and/or motility. The aflagellate variant poorly colonized the gastrointestinal tract of infant mice but the flagellate, non-motile variant colonized the mice as successfully as the wild-type strain. 35S-labelled organisms were used to investigate the attachment of the variants to human epithelial cell monolayers in vitro. The flagellate, non-motile strain attached more efficiently to the cells than the wild-type strain or the aflagellate strain. Differences in attachment suggest that an adhesin is intimately associated with flagella of C. jejuni and that active flagella mediate only a tenuous association with host cells. This adhesin attached most efficiently to cells of intestinal epithelial origin and was not specifically inhibited by various sugars.

Comparison of different vaccines and induced immune response against Campylobacter jejuni colonization in the infant mouse.

The degree of protection conferred by vaccinated dams on infant mice against colonization by Campylobacter jejuni depended on the bacterial strain, preparation, and route of administration of the vaccine. In some instances of homologous protection, serum bactericidal titres correlated well with protection. However, boiled C. jejuni vaccine, which was non-protective, also elicited a strong bactericidal antibody response. Conversely, bactericidal activity could not be demonstrated against strains capable of cross-protection. There was a good correlation between high campylobacter-specific IgG response and bactericidal activity.