But there may be an important exception. In certain circumstances, contaminated water may prevent, rather than cause, the severe intestinal disease known as dysentery. That is the implication of a paper published recently in the Lancet by Dr David Sack and colleagues at John Hopkins University and the University of Maryland, Baltimore.
The link between impure water and intestinal infections is so pervasive that it has rarely been out of the news since August 1854 when the removal of the handle from the Broad Street pump in Soho, London, at the request of Dr John Snow, brought an end to a raging cholera epidemic.
Only last year, doctors at St Thomas's Hospital, London, discovered that a young girl who became severely ill after falling into the Thames had been infected with not only Salmonella but also three other disease-causing bacteria. Last month, microbiologists in London and Leicester, after finding food-
poisoning viruses in British bathing waters, called for an end to the discharge of raw sewage into the sea. And waterborne transmission of the bacterium Vibrio cholerae has played a major role in the cholera epidemic now affecting several South American countries.
Like cholera, bacterial dysentery is a potentially fatal disease, especially in the young, the old and the malnourished of all ages. The bacteria penetrate the lining of the large intestine, causing victims to pass, frequently and painfully, stools containing blood and mucus. High fever, chills, and abdominal cramps are among other symptoms. Contamination of water supplies from patients' faeces is a highly efficient mechanism through which this foul disease is spread.
Dr Sack and his collaborators have been especially interested in the contrast between two of the species of the bacterium Shigella responsible for the malady. S. sonnei is more common in industrialised societies, whereas S. flexneri occurs more often in developing countries. The disparity is surprising, given that they are transmitted in the same way. However, it also squares with evidence that as countries advance economically there is also an increase in the proportion of dysentery cases caused by S. sonnei.
There are various possible reasons for this pattern, none totally consistent with the evidence. The explanation which begins to seem most plausible is that in developing countries exposure to another bacterium, one that partially resembles S. sonnei, triggers the formation of antibodies that make a large proportion of the population immune to S. sonnei as well. In other words, such a microbe would comprise a naturally occurring vaccine. It would have to be relatively common and ingested by large numbers of people in the community - perhaps through polluted surface water. If this did occur, it would account for the contrast between the prevalence of S. sonnei and S. flexneri in developed and developing regions. It would also explain the comparative rise in the occurrence of S. sonnei as increasing prosperity facilitates improvements in water supplies.
When Dr Sack and his co-workers studied possible candidates as this 'natural vaccine' they settled on Plesiomonas shigelloides. A waterborne bacterium, P. shigelloides has been isolated from surface water samples in many developing countries. A typical resident in such a country, drinking about a litre of surface water a day, would consume about 105 cells of P. shigelloides. Only because special methods are required to grow it in the laboratory has the significance of the organism been overlooked in the past.
The candidature of P. shigelloides was greatly strengthened by the discovery that part of the cell wall of one particular strain, a molecule known as lipopolysaccharide (LPS), was identical with the LPS found in S. sonnei. Infection with this bacterium could, therefore, induce the formation of antibodies that would also deal with S. sonnei. The part of the dysentery bacterium that is most important in eliciting immunity is indeed its LPS. When the Baltimore researchers injected rabbits with P. shigelloides they found that it protected them against S. sonnei infection. Studies in volunteers will be necessary to determine whether the same thing occurs in humans.
Oddly, there is uncertainty as to whether P. shigelloides itself causes diarrhoea. Some studies have suggested that it does. Others have shown no such association. Inoculation of P. shigelloides, even in large quantities, into human volunteers has not led to any illness. Like artificial vaccines against conditions such as poliomyelitis and diphtheria, P. shigelloides would, of course, not need to produce symptoms in order to induce immunity. It may well be that the bacterium does cause mild inflammation of the intestinal wall, at the same time inducing antibodies protective against S. sonnei.
Dr Sack and his colleagues emphasise that their hypothesis, despite the strong and varied evidence in its favour, does not amount to an argument against the desirability of improving water supplies. The benefits of sewage treatment and water purification in excluding a whole host of microbial killers remain paramount. Nevertheless, P. shigelloides appears to be the exception that proves the rule.