Capsules, flagella and fimbriae

Bacteriologists describe motile bacteria by where the flagella sit, because the arrangement is stable for a species and helps identification. A single flagellum at one end is monotrichous, as in Vibrio cholerae. A tuft of flagella at one end is lophotrichous, as in Pseudomonas fluorescens. Flagella at both ends are amphitrichous. Flagella spread over the whole surface are peritrichous, as in Escherichia coli, Salmonella and Proteus, and it is this arrangement that produces the spreading swarming edge seen on an agar plate. The filament itself is a hollow helix of flagellin subunits that grows at its tip and is powered by a basal body embedded in the membrane, which turns at high speed using the proton gradient rather than ATP. That motor reverses direction, and the reversal changes the swimming pattern: runs of smooth swimming alternate with brief tumbles, and by lengthening the runs when the concentration of an attractant rises the cell climbs a chemical gradient. This directed movement is chemotaxis, and it is what carries a pathogen towards a wound or up a ureter. The most important clinical lesson about the flagellum is not the number but the variation. Flagellar antigens can be switched on and off, a phenomenon called phase variation, so the antibody a host made against one flagellar type may not recognise the next generation. Salmonella alternates between two flagellin genes, and the identical trick appears in the HA and NA antigens of influenza virus and in the pilin genes of Neisseria gonorrhoeae. Antibodies against flagella can also be protective in a different way: they cross-link the filaments and immobilise the organism, which is why killed whole-cell vaccines such as the injected cholera vaccine were built partly around flagellar antigen. In the laboratory, the flagellar antigen pattern is read alongside the somatic antigen pattern in a serotyping scheme, and it was this combination, in the Kauffmann–White scheme, that first allowed public health teams to prove that scattered cases of typhoid belonged to one outbreak rather than to many unrelated infections. A nurse uses the same logic when reporting that three patients on one ward share a serotype: that is not three coincidences but one source, and it triggers a search for the common vehicle.

A pathogen that reaches a mucosal surface has not yet infected anything. To cause disease it must first attach, then multiply, then either invade or release a toxin, and each of those steps is a separate opportunity to stop it. Attachment deserves particular attention because it is the step that comes first and because it is often the only step for which the organism carries a dedicated tool. Uropathogenic Escherichia coli carries P fimbriae that bind a galactose-containing receptor on the lining of the urinary tract; the bacterium cannot grip a surface without them, and strains that lose the genes become far less able to cause pyelonephritis. Vibrio cholerae uses a toxin-coregulated pilus that not only anchors the organism to the small intestinal mucosa but also brings the bacteria close enough together to act as a receptor for the bacteriophage that carries the cholera toxin genes, so attachment and toxin production are physically linked. Neisseria gonorrhoeae varies its pilin proteins, which is why a person can be reinfected repeatedly. Attachment is also the step that allows a community to form. Once bacteria are held on a surface, they secrete an extracellular matrix, become a biofilm, and a biofilm is measurably more resistant to antibiotics and to the immune system than the same organisms growing free in liquid. This is why a device left in place is such a hazard and why removing it is a treatment. The vaccine idea follows directly. Because the adhesin is exposed on the outside of the cell, because it is essential rather than optional, and because antibody against it can physically block binding, vaccines have been designed to raise antibody against fimbrial proteins. The classic example is the oral cholera vaccine, which includes the B subunit of the cholera toxin and killed whole cells, and veterinary vaccines against enterotoxigenic Escherichia coli that raise antibody against the K88 and K99 fimbriae protect piglets by preventing attachment in the gut. Human trials of fimbrial vaccines have been less successful, and the reasons are instructive: there are many fimbrial types, so a vaccine must be multivalent; the strongest antibody response must be secreted onto the mucosa rather than merely present in serum; and some pathogens, such as pneumococcus, hide behind a capsule instead, so the capsule rather than the adhesin becomes the vaccine target. For nursing, the lesson is preventive and practical. Bacteria that cannot attach cannot colonise, so hand hygiene, catheter care, early removal of devices and the maintenance of skin and mucosal integrity are not general hygiene advice but precise attacks on the first step of infection.

Conjugation deserves a second look because it explains an observation that puzzles students: unrelated species in the same hospital share the same resistance pattern even when no antibiotic has ever touched one of them. The conjugative pilus, best studied as the F pilus of Escherichia coli, is longer than an ordinary fimbria and is encoded by the very plasmid it transfers. It contacts a recipient cell, retracts to bring the two envelopes together, and a conjugation pore is assembled through which a single strand of plasmid DNA passes while the complementary strand is synthesised. The recipient then builds the second strand, the plasmid circularises, and the cell is now a donor in its own right. The genes needed for this machinery are grouped in an operon, the tra operon, and they are carried on the plasmid, not on the chromosome. That arrangement matters because plasmids are accessory DNA: they can be gained, lost and exchanged without killing the cell, and one plasmid can carry several resistance genes at once, which is why a single conjugation event can make a bacterium resistant to three or four drug families simultaneously. Conjugation is the commonest route by which extended-spectrum beta-lactamase and carbapenemase genes move between Klebsiella pneumoniae and Escherichia coli in hospitals, and the same mechanism spreads plasmid-borne resistance in Staphylococcus aureus and Enterococcus. Other routes of horizontal gene transfer exist and should be distinguished from it. Transformation is the uptake of free DNA from the environment, which is how pneumococci acquire and exchange capsule genes. Transduction is transfer by a bacteriophage, which is how toxin genes reach Staphylococcus aureus and Escherichia coli O157. Conjugation is the only one of the three that requires direct cell-to-cell contact. Two practical consequences follow. First, the spread of resistance is not always caused by the antibiotic that is being used: a plasmid selected in one patient can travel through a ward on hands and equipment and appear in a different species, so infection control and antibiotic stewardship are two halves of one strategy. Second, free DNA and plasmids survive in the environment, so cleaning, hand hygiene, isolation of colonised patients and correct waste disposal act directly on the conditions that allow conjugation to happen. Nurses rarely see a pilus, but they decide whether the next contact occurs.