Singing in the shower may never feel the same again, if one dwells on the results of a study by Norman Pace and colleagues (L. M. Feazel et al. Proc. Natl Acad. Sci. USA doi:10.1073/pnas.0908446106; 2009). While you trill to an aria from your favourite operetta, an audience of “microbes of potential public health concern” may well have prime seats in your shower head.

Shower heads are known to harbour microorganisms, mainly because they provide a warm, moist environment prone to the formation of biofilms — aggregates of microbes that adhere to surfaces. Previous studies have sought to identify the organisms that reside in shower heads by culturing them. But as most bacteria can't easily be grown in the laboratory, shower-head biofilm assemblages have remained largely uncharacterized.

In their study, Pace and colleagues use molecular techniques to analyse the microbes found in 45 shower heads obtained from nine cities in the United States. The authors sequenced bacterial genes isolated from each shower-head biofilm and identified the resident microbes by comparing their results with the genes of known bacteria. Such non-culture — or metagenomic — techniques have revolutionized our ability to characterize the microbial communities that live among us.

Pace and co-workers found that the shower heads contained many types of bacteria commonly found in water and soil. The striking result, however, was the abundance of non-tuberculous mycobacteria, in particular Mycobacterium avium — the amounts in the biofilm were 100-fold higher than those found in the associated water supply. About 20% of shower-head swabs harboured M. avium DNA sequences, and this figure rose to a staggering 78% when the authors used molecular techniques that specifically detect this organism.

Why are non-tuberculous mycobacteria so common in shower-head environments? These organisms are known to be chlorine-resistant, and they may actually be enriched by the treatments used to disinfect municipal water supplies. In fact, when the researchers attempted to clean one shower head harbouring a Mycobacterium gordonae species with bleach, they managed only to increase its relative abundance.

Non-tuberculous mycobacteria are opportunistic pathogens that can cause severe disease in immunocompromised people, and they can also infect healthy people. The rates of infection with these mycobacteria are increasing in resource-rich countries. Pace and colleagues' work lends support to the possibility that our predilection to soap in the shower, rather than soak in the bath, may be a contributory factor.