human body are usually protective, certain shifts in the composition of these populations can have pathological consequences. Diseases of this type are referred to as microbiota shift diseases. In chapter 5, “The Normal Human Microbiota,” examples of microbiota shift diseases are described in more detail, but for present purposes, one example should suffice: bacterial vaginosis.
Bacterial vaginosis is the term used to describe a shift in the vaginal microbiota from a predominantly Gram-positive population, dominated by Lactobacillus species, to a population of Gram-negative anaerobes. For a long time, this condition was not taken seriously by physicians because the only symptoms, if there were symptoms at all, were a sparse discharge, some discomfort, and in some women a fishy odor. Two papers in the New England Journal of Medicine in 1995 changed the status of bacterial vaginosis. One of these papers linked bacterial vaginosis with preterm births. This was an epidemiological association, not proof of a cause-and-effect relationship. The second paper described the result of an intervention study, in which antibiotics known to target Gram-negative anaerobes were administered to pregnant women with bacterial vaginosis, and the effect on the birth weight of the infant was determined. Antibiotic intervention that returned the vaginal microbiota to “normal” was associated with normal full-term births, whereas untreated women were significantly more likely to have preterm infants.
Soon after these first connections were made, bacterial vaginosis was linked to a higher risk of contracting HIV infections and other sexually transmitted diseases, just as chlamydial disease and gonorrhea had been shown previously. A major challenge for scientists trained in the analysis of diseases caused by a single species of microorganism is to learn how to deal technically and conceptually with polymicrobial diseases caused by shifts in bacterial populations consisting of hundreds of species. Undoubtedly, all of the species present are not equal contributors to the disease state, but the situation is far more complex than single-microbe infections.
A Brave New World of Pathogenesis Research
The H. pylori revolution captured the public imagination, but an even more important revolution has been the realization by research scientists that new molecular technologies are opening up a plethora of new opportunities to understand at the molecular level how infectious diseases develop. For several decades after the discovery of antibiotics, during a period in which a number of new vaccines were developed, it seemed sufficient to simply treat or prevent bacterial diseases. As long as antibiotics worked and vaccines were widely available, controlling bacterial infections at the practical level did not require in-depth information about the bacterium-host interaction. As physicians and scientists became concerned about increasing antibiotic resistance, there was a growing realization that a better understanding of the detailed interactions between the human body and the bacterial pathogen might lead to new treatment strategies.
Additionally, there was recognition that there are some diseases whose symptoms are caused by bacterially produced toxins that are not effectively treated by antibiotics. A good example is anthrax, a disease caused by Bacillus anthracis. The symptoms of this disease are caused by a protein toxin, produced and secreted as the bacteria multiply in the body. If the disease is diagnosed immediately and the right antibiotic is administered, the disease can be controlled. However, antibiotics do not inactivate the toxin, and if antibiotic therapy is delayed for even a few days, enough anthrax toxin will have been produced to cause death.
In the case of a disease called septic shock that develops when bacteria enter and proliferate in the bloodstream (sepsis), a nonprotein component of the Gram-negative bacterial cell surface, lipopolysaccharide (LPS) (see Box 1-4), acts as a toxin that leads to organ damage and death. Here too, antibiotics are only effective if they are administered very early in the infection before the bacteria lyse and release too much of this toxic material. Although anthrax is not a significant threat (other than as a potential bioweapon), septic shock continues to kill tens of thousands of people each year in the United States alone. As discussed in later chapters, new understanding has recently emerged about how the human body responds to LPS, as well as to nonprotein surface components of Gram-positive bacteria, such as lipoteichoic acid (LTA). There is growing hope that this knowledge will make possible new and more effective therapies.
Box 1-4.
A Brief Review of the Surfaces of Gram-Negative and Gram-Positive Bacteria—One Membrane versus Two
For convenience, bacteria are often divided into two main groups, Gram positive and Gram negative, based on their ability to retain a purple dye after being washed with organic solvent such as methanol or acetone (this procedure of coloring bacteria is called Gram staining). The LPS is located at the surface of Gram-negative cells and is composed of lipid A (composed of fatty acids and a disaccharide), the core oligosaccharide, and the repeating-unit O-antigen polysaccharide (panel A). The LPS forms the surface of the outer membrane, whose inner layer is composed of phospholipids. The thin peptidoglycan (also called murein) of Gram-negative species is located in the periplasm between the inner and outer membrane.
In contrast, the cell wall of Gram-positive species lacks LPS and an outer membrane (panel B). The thick peptidoglycan is composed of multiple, cross–linked layers. A substantial amount of the Gram-positive cell wall is made up of another anionic polymer called wall teichoic acid (WTA, or TA) and lipoteichoic acid (LTA), both of which contain repeat units of ribitol or glycerol phosphate linked to amino sugars, sometimes to amino acids, such as D-alanine, and sometimes to ternary amines, such as phosphorylcholine. TA is covalently bonded to the peptidoglycan, whereas LTA is linked to a lipid anchor in the cellular membrane.
Other bacteria have cell walls with compositions that are a variation on one of these two types plus a few different components. For instance, the cell wall of mycobacteria, including Mycobacterium tuberculosis (the causative agent of tuberculosis), is similar to Gram-negative bacteria, but mycobacteria have a thick outer membrane comprised of two leaflets: an inner leaflet composed of arabinogalactan and mycolic acid (hence the name “mycobacteria”) and an outer leaflet composed of phosphoglycolipids (panel C). Further details can be found in The Physiology and Biochemistry of Prokaryotes, 4th Edition by David White, James Drummond, and Clay Fuqua (Oxford University Press, 2011).
It should be noted that the differences in bacterial surfaces and their components between Gram-positive and Gram-negative bacteria often dictate how they interact with their environments and, in particular, how they interact with their hosts. Importantly, the host immune system responds quite differently to these components. The properties conferred by these components necessitate that hosts have alternative strategies to recognize and combat Gram-negative versus Gram-positive pathogens. The presence of the outer membrane in Gram negatives also creates an extra challenge for the bacteria to export molecules on its surface and to secrete its toxins and other virulence factors into the external medium.
A different type of question is why different people respond differently to the same bacterium. In some cases, the human response can range all the way from a virtual lack of symptoms in some individuals to severe illness and death in others. Taking this issue one step further, prior exposure to one microbe can impact the nature and degree of response to subsequent exposure to another microbe. Indeed, it is now well-established that maternal exposure to pathogens or their components (e.g., LPS) modulates the immune competence and immune response of offspring in species ranging from insects to plants to birds and mammals. Clearly, it would be helpful to understand this range of reactions so that people who are most susceptible to an infectious agent could be quickly identified and given priority in treatment.
These and similar practical problems with controlling bacterial infections have driven a new interest in the interaction between bacteria and the human body at the molecular level. Fortunately, a cornucopia of new molecular tools and paradigms has become available that have made it possible to explore the host-pathogen interaction in a detailed way. It has even become more feasible to investigate infections that involve more than one species of bacteria or infections in which the bacterial pathogen acts in an area of the body, such as the mouth or small