- Volume 7, Issue 2, 1974

One hundred specimens of nasopharyngeal secretion were collected from infants with acute respiratory disease and stored at -60°C before inoculation into tube cultures of bovine testis (BT) and HeLa (HL) cells. Of a total of 46 specimens which yielded respiratory syncytial virus, 45 did so in BT cells but only 32 in HL cells. The mean incubation time before a cytopathic effect was first detected was 6.5 days for BT cells and 10.6 days for HL cells.

Proteose-peptone agar was defective for the culture of H. influenzae from small inocula when sterilising temperatures exceeded 116°C. All 23 strains tested grew poorly or not at all on the medium sterilised at temperatures of 121°C or more for 30 min. The inhibition was more marked when 5% CO2 was present during incubation. The inhibitory effect disappeared after the medium had been stored at 4°C for 7 days before inoculation, or could be abolished by the addition of a reducing agent such as sodium dithionite 100 μg per ml.

The growth of H. influenzae may also be adversely affected by agar. Of three commercial agars tested, one batch of a particular brand was found to be inhibitory when used from a freshly opened bottle. Inhibition could not be demonstrated after the agar was washed or in the presence of sodium oleate (4 μg per ml).

Various authors have described a group of organisms resembling Citrobacter freundii, but differing in several significant respects. It has been suggested that these organisms should be regarded as a new species, and the names Citrobacter koseri, Levinea sp. and Citrobacter diversus have been proposed for them. Representative strains described as C. koseri, L. malonatica, and C. diversus were shown to form a biochemically homogeneous group, and an antigenic-classification scheme is proposed for them. Strains of L. amalo-natica differed biochemically and serologically from the other strains studied and are excluded from the antigenic scheme.

The changes that occurred in the faecal flora during the separate administration of lactulose and magnesium sulphate were studied in five patients with hepatic encephalopathy. Lactulose therapy produced significant increases in lactobacillus counts in three patients, and two of these patients showed clinical improvement. No systematic change occurred in the counts of other organisms during either treatment; in particular it was not possible to correlate clinical improvement with a reduction in the count of the two groups of organisms-enterobacteria and Bacteroides spp.-that are considered to be most active in the production of ammonia.

Among 95 strains of trimethoprim-resistant Gram-negative bacteria isolated from clinical material in various centres in the UK, one was a thymine-requiring variant of Escherichia coli and 10 carried transmissible trimethoprim R factors. The trimethoprim R factors belonged to three different compatibility groups. All gave rise to high levels of trimethoprim resistance. Eight belonged to group W, conferred resistance to trimethoprim and sulphonamides and appeared to be identical with one another and with those found in some London hospitals a year earlier. One belonged to a new group, Iδ, conferred resistance to trimethoprim and streptomycin and determined I pili. One belonged to group P and conferred resistance to trimethoprim alone. The W plasmid, carried by a particular klebsiella strain of capsular type 9, appeared to be established in several hospitals in London, but it also occurred sporadically and in other bacterial hosts. The other trimethoprim R factors were identified only in single bacterial strains.

Large amounts of high-titre E. coli α-haemolysin were produced in a nutrient broth medium. Production was enhanced by growth in a medium containing large molecular-weight components. The haemolysin was purified by “salting out” with ammonium sulphate precipitation at its iso-electric point (pI = 4.6) and by chromatography on Sephadex G-200. Evidence is presented that the haemolysin is a large protein (approximate molecular weight = 580,000) that was not easily dissociated into smaller subunits. Lethality in mice and rabbits was not observed after intravenous injection of α-haemolysin, but hard swellings appeared at injection sites after the intradermal administration of active haemolysin preparations. The haemolysin did not cause fluid accumulation in ligated intestinal segments in rabbits.

The interaction of Escherichia coli α-haemolysin with sheep erythrocytes was characterised in all experiments by a lag phase, and by a period of lysis that was linear between 20 and 80% haemolysis. The lag phase was inversely proportional to the logarithm of haemolysin concentration. Maximum rates of haemolysis were observed at 43°C and pH 8.8 to 9.0.

The optimal concentration of CaCl2 for haemolysis was between 1 mm and 10 mm. Evidence is presented that the first step in the haemolytic reaction is activation of α-haemolysin by calcium ions. This causes an alteration in the structural appearance of purified haemolysin molecules. It was also shown that EDTA inhibited haemolysis only when added to α-haemolysin before its activation by calcium ions. This finding suggests that, during activation, calcium ions are firmly bound to α-haemolysin.

Protection of mice against challenge with virulent Salmonella typhi could be induced by intraperitoneal or subcutaneous vaccination with heat-phenol-or acetone-inactivated suspensions of S. typhi strain Ty2 or of a Vi-producing strain of Escherichia coli, or with Vi antigen prepared from either of these.

In all cases, protection rose to a maximum at about 4 days after vaccination. By day 18 this had fallen to the level at day 2.

The rise and fall of Vi agglutinins in mice vaccinated either intraperitoneally or subcutaneously with heat-phenol- or with acetone-killed suspensions of S. typhi Ty2 or E. coli paralleled closely the rise and fall in protection, whereas the rise of O agglutinins corresponded with the decline in protection.

It was concluded that the potency tests for typhoid vaccines according to procedures SCM and IPM (WHO type-N procedures) in mice are chiefly a measure of the response to Vi antigen and consequently not a measure of factors responsible for eliciting immunity in man.

Cell-free preparations of “toxin” were made from two strains of Staphy-lococcus aureus, one known to be of high virulence and the other of low virulence in the mammary gland of cattle. These preparations were similar in their in-vitro properties and in their clinical and histopathological effect when inoculated into the mammary gland of mice. However, when live staphylococci were inoculated into the mammary gland of mice a difference in virulence corresponding to that seen in cattle was detected by the clinical response, and by the histopathological appearance, the number of organisms, and the amount of α-lysin present in the infected glands. The reasons for the failure of the low-virulence strain to express its potential virulence are discussed.

Non-sporing strictly anaerobic bacteria were isolated from human faecal samples from England, Scotland, USA, India, Uganda and Japan. Strains were assigned to genera on the basis of the Gram reaction and on the type of fatty acid produced from glucose. Analysis of these acids was by gas-liquid chromatography. The isolates were further classified on the basis of various biochemical reactions.

Bacteroides fragilis was the non-sporing anaerobe most frequently isolated from human faeces. Bifidobacterium adolescentis and Eubacterium aerofaciens were also frequently isolated. Stool samples from India, Uganda and Japan contained a larger proportion of eubacteria and fewer Bacteroides spp. than did the samples from England, Scotland and the USA.

The therapeutic effect of methicillin on skin lesions in guinea-pigs produced by two strains of Staphylococcus aureus showing heterogeneous resistance to methicillin, and by two strains of normal sensitivity, was compared. Methicillin was given intravenously immediately after the intradermal inoculation of staphylococci and the skin lesions were measured 24 hours later. The serum concentrations of methicillin were comparable to those obtained in man with standard dosage and were below or similar to the MIC of the “resistant” staphylococci when this was determined at 37°C. No difference was found after methicillin administration in the size of lesions produced by “resistant” and “sensitive” staphylococci.

Resistotyping is based on differences in the resistance of strains of a given species to a selection of chemicals used at critical concentrations. The method was applied to Staphylococcus aureus and the patterns of response were used for epidemiological enquiry. Amongst the chemicals used were metals, and these markers showed the plasmid-associated instability already known from genetic studies. Generally the resistotypes obtained and the phage-typing patterns gave parallel information. There were examples of staphylococcal outbreaks, however, in which resistotyping amplified or clarified the findings indicated by phage-typing patterns. Conversely, in other instances phage-typing patterns subdivided a common resistotype. This investigation confirms the validity of resistotyping as a method of epidemiological tracing applicable to widely differing bacterial species.

A previous study described the isolation of multiply-resistant strains of Escherichia coli from the gastro-intestinal tract after the ingestion of a strain carrying an R factor (Rl) that confers resistance to ampicillin (A), chloram-phenicol (C), kanamycin (K), streptomycin (S) and sulphadimidine (Su). One such strain carried the R factor R174 (ACKSSuT)–which was in addition resistant to tetracycline (T)–and another carried the factor R157 (AST).

The minimum inhibitory concentrations of these antibiotics for E. coli strains, isogenic but for the R factors, were consistent with the formation of factor R174 by recombination of the AST and ACKSSu resistance determinants of the other two R factors. The factor Rl 74 determined a streptomycin-phosphorylase enzyme also present in factor R157, together with the strepto-mycin-spectinomycin resistance determinant of factor Rl, and all three R-factors shared the same type of ampicillin resistance. Factor R174 carried all the other antibiotic-resistance determinants (TCKSu) of the parent plasmids and is thus probably a recombinent of factor Rl and R157.

DNA-DNA hybridisation showed that factor R174 contained almost all of the DNA-base sequences of factors R157 and Rl. There was also approximately 50% homology between the latter two R factors which may account for the occurrence of recombination.

Strains of Candida albicans isolated from the duodenum of infants with acute gastro-enteritis were inoculated into ligated segments of small intestine of infant rabbits. Twenty hours later, the lactase activity in the mucosa of these segments was compared with the lactase activity in similar segments into which sterile broth had been inoculated. Analysis of the results from 25 matched pairs of loops showed a statistically significant decrease in lactase activity in the loops that had received C. albicans. We suggest that the depression of lactase activity observed in infants with gastro-enteritis may be caused or perpetuated by the growth of C. albicans in the duodenum.

Two hundred and seven strains of bacteria from human faeces were screened for their ability to convert oleic acid to hydroxystearic acid in vitro. Ninety-four strains were able to perform this reaction. All but one of the enterococci tested were active, and some active strains were also found among Bacteroides strains (18%), bifidobacteria (32%), clostridia (50%), and enterobacteria (21%). Oleic acid-hydrating bacteria were also able to hydrate palmitoleic but not linoleic acid. Relationship could not be demonstrated between faecal hydroxystearic acid level and the number of bacteria in the faeces able to produce this acid.

Broth cultures of 24 out of 28 urease-positive strains of Clostridium sordellii tested lysed readily when mitomycin C (MC) was added early in the logarithmic phase of growth. Toxigenic strains of the species were more susceptible than the non-toxigenic strains to the effect of the drug. Phage-like particles with different morphologies could be obtained from cultures of all of five C. sordellii strains studied. The lytic agent or agents active on freeze-thawed cells of C. sordellii strains were also found in the MC lysates.

A multi-resistant strain of Staphylococcus aureus (no. 649MR) containing eight plasmids was constructed in vitro. Chloramphenicol resistance and pigment production were less stable in strain 649MR than when present singly in strain 649. Chloramphenicol resistance could not be transduced from strain 649MR to other strains. Recombination occurred between only two of the plasmids in strain 649MR; these determined (1) metal-ion resistance only (mir-r) and (2) resistance to erythromycin and production of penicillinase in addition to metal-ion resistance (pen +).

The plasmid-DNA content of the multi-resistant strain was only 60% of that expected from analysis of the individual plasmids in the host strain, and this could be only partially explained by interaction between the mir-r and pen + plasmids.

These findings suggest that under natural conditions the number of plasmids that a staphylococcal cell can maintain will be limited and that recombination will be confined to the “penicillinase plasmids”. The mechanism or mechanisms that limit the number of plasmids within the cell are not known, but factors additional to plasmid incompatibility must operate.

The existence of plasmids as multiple copies cannot be accounted for entirely as a means for ensuring their distribution at cell division, nor as provision of maximum levels of antibiotic resistance, but may be related to transfer between cells by transduction.