Finding hormone biomarkers in ancient skeletons
The widening horizon of molecular biomarkers from ancient bones and sediments may open new doors to the biology of ancient people.
A new paper by Aimée Barlow and collaborators shows the potential of recovering hormone biomarkers from skeletal remains that are hundreds of years old. The study examined bone, dentine, enamel, tooth root, and dental calculus from ten individuals from ancient Britain, dating from the first century CE to the nineteenth century. They found measurable levels of steroid hormones, including estrogen, progesterone, and testosterone, in at least some samples of all the types of hard tissues they examined.
The most interesting aspect of the study was the application of hormone data to uncover pregnancy. The study included two skeletons from women who died during late pregnancy, both evident from fetal remains in their uterus at the time of death. One was from the nineteenth century, and one from an early medieval cemetery. Barlow and coworkers found that one of these skeletons had very high levels of progesterone across samples, raising the possibility of a test that might reveal pregnancy in more fragmentary remains.
However the study has some important limitations. Different tissues had different preservation of steroids, and it’s not entirely clear why. Neither is it yet clear whether the uptake of steroids into bone or tooth tissues can be translated to measures of serum levels of these hormones that have physiological interpretations in living people. Indeed the hormones may have different effects in bone tissue that themselves may reveal some biology from ancient skeletons.
Such limitations point to some potent challenges. Even so, the recovery of biomarkers beyond DNA and protein may open up new ways of seeing the details of ancient lives.
Cortisol
Another set of recent studies have used similar methods to uncover evidence of cortisol in human teeth from archaeological contexts. Known as the “stress hormone”, cortisol is produced by the adrenal glands and functions not only in stress response but in regular daily cycles.
In a 2020 research study, Leslie Quade and collaborators sampled both enamel and dentine from archaeological teeth from Roman period burials of 65 individuals. They found detectable cortisol levels in 32 of their samples. A second study of deciduous teeth by Quade, working with a different team of collaborators, found a similar fraction of preservation in enamel and cortisol.
As in the case of estrogen, progesterone, and testosterone, it’s not clear whether these measures of cortisol have any interpretation in terms of the physiology or life history of the ancient individuals. Basic work on tooth levels of living people will need to be done before levels in archaeological bone can be interpreted.
A molecular approach to ancient pregnancy?
People who follow forensic anthropology and bioarchaeology may have an impression that skeletons provide a lot of evidence of pregnancy and childbirth. But actually such information is scarce.
For instance, anthropologists sometimes talk about “parturition scars” on the bony pelvis. The idea is that the ligaments that hold the sacrum and hip bones together are stressed by the birthing process, leaving marks that may be identifiable long after a mother’s death.
In reality, “parturition scars” simply don’t work the way that forensic anthropologists once assumed. Studies of skeletons with known histories of childbirth have shown that there is little or no predictive accuracy in such marks even for female individuals who have given birth many times. Even male individuals often have such marks on their pelvic bones.
As it stands, some researchers think that studies of the pelvis may provide a way of examining the average birth rate across large samples of individuals. But there is little chance of any accurate means of looking at birth histories from the bones of a single individual.
Newer methods suggest ways of looking at significant life events like birth. A couple of years ago I reviewed how researchers are using dental cementum to look at biological stress across the lifetime of Neanderthals. The idea here is that cementum growth is altered during times of increased metabolic or physiological stress, such as menarche, pregnancy, and chronic illness.
But different life events may alter cementum growth in similar ways, making it hard to disentangle one kind of event from another. That’s a problem with most studies of developmental and physiological stress.
This is one reason I think the new work on hormone survival in bones and teeth is worth watching. If biomarkers persist in hard tissues like cementum, it raises the possibility that high-resolution sampling might provide a way to match cementum growth with levels of progesterone or other steroids. Physiological processes involving hormonal changes, like menarche, pregnancy and lactation, would look quite different from long-term illness and other physiological stresses.
Challenges ahead
The studies reveal a number of challenges that will need to be overcome if researchers want to regularize these kinds of analyses in archaeology. Both the research groups relied on commercially available immunoassay kits, in which antibodies bind very small amounts of the target molecules. These are very sensitive to low levels of the target hormones, but neither the kits nor the bulk sampling strategies used in this study are really tuned to application on archaeological samples.
As a result, null results are really not informative. Understanding this is important when considering one interpretation of the new study by Barlow and coworkers that I’ve seen reported in the press. As I noted above, the study included two skeletons of individuals that were pregnant at the time of death. Neither of them showed any detectable testosterone in any samples. The study includes a statement recognizing that this lack of testosterone might conceivably be a signal: Maybe the bone of pregnant mothers lacks testosterone. But considering the limitations of the immunoassay kits and low preservation generally in archaeological bone, it’s much more likely that this finding is just a coincidence.
More broadly, most of the results from bone and teeth don’t have clear biological interpretation. None of the hormone measures distinguish female from male skeletons, for example.
What they do provide is a proof of principle. Now the interesting questions are those that enable connections with biology. Can the sampling of bone or tooth tissues be made more precise, enabling study of changes in hormone levels, whether across years or smaller time increments? Can the sensitivity of methods be extended?
And could there be information in coprolites or sediments that might allow these small molecules to yield biological information about past peoples?
I’m enthusiastic about these approaches. Small molecules have a high survival in archaeological contexts, and in particular steroids may survive for a long time in association with the mineral content of hard tissues. I don’t think it’s impossible that Neanderthals or other ancient humans of similar age might preserve signatures of these steroids.
If so, then I think it’s possible we’ll see the emergence of detailed chronologies of the life stresses and experiences of many ancient people. Experts on evolutionary health often hold that today’s environments are a significant mismatch for our deep heritage. I wonder if that idea will hold up once we see more of the health histories of ancient people.
References
Barlow, A., Craig-Atkins, E., Barker, E., Crawford, A., Cacciabue, D., & Hemer, K. A. (2025). First successful detection of oestrogen, progesterone and testosterone in multiple human hard tissues, and their use as potential biomarkers of pregnancy. Journal of Archaeological Science, 183, 106392. https://doi.org/10.1016/j.jas.2025.106392
Quade, L., Chazot, P. L., & Gowland, R. (2021). Desperately seeking stress: A pilot study of cortisol in archaeological tooth structures. American Journal of Physical Anthropology, 174(3), 532–541. https://doi.org/10.1002/ajpa.24157
Quade, L., Králík, M., Bencúrová, P., & Dunn, E. C. (2023). Cortisol in deciduous tooth tissues: A potential metric for assessing stress exposure in archaeological and living populations. International Journal of Paleopathology, 43, 1–6. https://doi.org/10.1016/j.ijpp.2023.08.001
Ubelaker, D. H., & De La Paz, J. S. (2012). Skeletal Indicators of Pregnancy and Parturition: A Historical Review. Journal of Forensic Sciences, 57(4), 866–872. https://doi.org/10.1111/j.1556-4029.2012.02102.x
Waltenberger, L., Rebay-Salisbury, K., & Mitteroecker, P. (2022). Are parturition scars truly signs of birth? The estimation of parity in a well-documented modern sample. International Journal of Osteoarchaeology, 32(3), 619–629. https://doi.org/10.1002/oa.3090