Internet Archaeol. 5. Gernaey et el. Detecting Ancient Tuberculosis

Introduction

Tuberculosis (TB) is currently the leading cause of death from a single infectious agent. This has been globally linked to an increasing drug resistance and a malevolent synergy with HIV infection (Snider et al. 1994). There is also an increase among those in homeless shelters and in inner cities (Farmer 1997) and those with apparently competent immune systems (Bhatti et al. 1995). TB is believed to be of great antiquity (Keers 1978) as many ancient civilisations have described and depicted the disease (Evans 1994). Perhaps the disease is cyclical in nature (Davies et al. 1997), reflecting a change in human herd immunity (Stead 1992).

Socio-economic factors

If there is a disease that signals poverty, it is TB. So much so that, in the United States, plans to combat the disease include national socio-economic policies to permit adequate resources to:

combat poverty and malnutrition,
rebuild the infrastructure of the public health system,
improve access to health care among rural and urban dwellers (Young et al. 1997).

In modern populations, whilst overcrowding has the strongest effect, other socio-economic factors such as income, poverty, unemployment and education level affect the prevalence of the disease (Cantwell et al. 1998). Why would these factors be important?

Tuberculosis in humans is usually caused by Mycobacterium tuberculosis (MTB) or Mycobacterium bovis (together known as the MTB complex) and is, principally, a disease of the respiratory system. As such, it is usually spread by coughing. Although the bacteria do not produce spores (heat, light and desiccation resistant bodies), they have a thick, lipid-rich, cell wall which may enable them to survive desiccation for a while. The healthy individual may be infected by contact with viable bacteria; the chances of infection are increased in crowded living conditions. Infection does not always lead to disease. What, then, induces disease formation?

Among generally healthy individuals, infection with MTB complex is likely to be asymptomatic, but prolonged contact with a diseased individual would result in exposure to increased bacterial numbers (Ridley 1988, 211). Low socio-economic status usually includes crowded living conditions, reduced income and a concomitant poor diet, all factors important in the transmission of the disease (Cantwell et al. 1998), but prolonged close contact is also required for disease transmission. Prolonged contact results in exposure to high bacterial numbers, and a large infective dose can overwhelm even a competent immune system (Elender et al. 1998).

In the early 20th century, prior to effective chemotherapy, TB frequencies were strongly associated with populations that were malnourished, overcrowded (or domiciled in poorly ventilated areas) and/or stressed, such as inmates in gaols and barracks or miners (Wilson and Miles 1946). We therefore suggest that TB prevalence may be an indication of the levels of poverty for the archaeological record. The concept of "poverty" is a relatively modern one, but did the condition exist, even if it was not recognised as such, in the past?. Modern "poverty" results in overcrowding and malnourishment, conditions that probably existed in the past, before the Victorian concept was introduced. Currently, there is no clear way of ascertaining the relative degrees of poverty between entire archaeological populations, only sometimes between individuals, as usually indicated by the presence or absence of grave goods. A chemical "screening test" which may indicate overall levels of poverty in a population would, therefore, be a useful tool in archaeology.

Recent studies to detect M. tuberculosis DNA in ancient bones have met with some success (e.g.Salo et al. 1994; Baron et al. 1996; Taylor et al. 1996; Faerman et al. 1997). In an unpublished pilot study by some of us (A. M. Gernaey, D. E. Minnikin, R. A. Dixon and C. A. Roberts) it was demonstrated that the presence of M. tuberculosis DNA could be confirmed in 1000 year-old archaeological human bone, by the detection of specific long-chain lipids, the so-called mycolic acids. These, and other lipids, have previously been shown to be effective "biomarkers" for modern clinical tuberculosis (Minnikin et al. 1993), using sensitive methods of analysis. To confirm whether mycolic acids are effective biomarkers for ancient TB, it was necessary to investigate a particularly well-documented site.