DOI: https://doi.org/10.21203/rs.3.rs-567955/v1
The La Chapelle-aux-Saints 1 skeleton of an old male Neanderthal is renowned for the advanced osteoarthritis of its spinal column and hip joint, and their implications for posture and lifestyle in these Mid- to Late Pleistocene humans. Reassessment of the pathologic lesions reveals erosions at multiple non-contiguous vertebrae and reactive bone formation extending far beyond the left hip joint, which suggests the additional diagnosis of brucellosis. This implies the earliest secure evidence of this zoonotic disease in hominin evolution. Brucellosis might have been transmitted via butchering or eating raw meat and is well compatible with the range of prey animals documented for Neanderthals. The associated infertility perhaps contributed to the demise of these late archaic humans.
Review of the pathologic findings initially reported in the La Chapelle-aux-Saints 1 Neanderthal 1–5 revealed alterations that appear attributable to the brucellosis. Today, brucellosis is globally the most prevalent zoonosis 6. It was first described by David Bruce on the island of Malta in 1885 7 and originally referred to as Malta fever. Evans 8 provided the appellation brucellosis and its discovery in Malta is reflected in one of its species names, Brucella melitensis. Hughes 9 suggests that Hippocrates described it in 1450 BCE. The first clinical description is attributed to Marston 10. Other names derived from its symptoms [undulant fever and Lazybones disease (in China)], the veterinarian who identified its cause (Bang’s disease), the employment of those affected by it (Corps disease), its epidemiologic geography (undulant Mediterranean fever or gastric remittent fever, Neapolitan fever, Gibraltar or Rock fever, Cyprus fever, Maltese fever, Crimean fever) 11–14.
Brucellosis is caused by an acid-fast non-motile, gram negative facultative intracellular coccobacillus related to animal (Anaplasma, Bartonella, Rickettsia and Wolbachia) and plant pathogens (Agrobacterium, Ochrobactrum and Sinorhizobium) 15,16. Although it occurs worldwide, it is predominantly recognized in the Mediterranean region, Asia and Latin America 17. There are at least 500,000 and perhaps 5–12 million new human cases each year 18,19.
Brucellosis has a long history. The earliest occurrence was claimed in a 2.1-2.5-million-year-old Australopithecus africanus skeleton, StW 431 20. A re-examination of the affected lumbar vertebrae L4 and L5 of StW 431 suggested, however, that the vertebral marginal lesions are more likely attributable to a limbus vertebra, i.e., an anterior disc herniation 5. Other occurrences of brucellosis have been described in 3200 − 3000 BCE Bab edh-Dhra in Jordan 21, 350 BCE – 1200 CE Nubia 22, 2nd -4th century BCE Butrint Albania 23, and 1st century Herculaneum 24. A study of mitochondrial genomes in early Bronze Age humans from the Novosvobodnaya North Caucasus site revealed Brucella abortus 25. We now examine evidence for its presence in a Neanderthal.
All preserved vertebrae of La Chapelle-aux-Saints 1 showed mild to advanced osteoarthritic changes. Except for the cervical vertebrae, the ventral margin of the vertebral bodies was damaged. In the upper thoracic and the lumbar vertebrae, the ventral portions of the vertebral bodies were reconstructed by Boule 1 with plasticine. Nevertheless, the first two thoracic vertebrae were relatively well preserved and, as the lower cervical vertebrae, sustained only minor damage see also 3,4,26.
Confluent vertebral endplate erosions with new bone formation were observed at the inferior vertebral surface of C5 (Fig. 1). Anterior endplate grooves with smooth bases were observed in C6. Note the marginal endplate erosions (arrows).
Erosions were observed at the right facet joint C4/C5 (Fig. 2), the right facet joint C7/T1 (Fig. 3), and T9/T10 (Fig. 4).
The left hip bone is fragmentary (Fig. 5). The articular surface in the weight-bearing posterosuperior portion of the acetabulum is eburnated. The acetabular margin bears large osteophytes and shows reactive new bone formation with erosions extending to the extraarticular portion of the lower ilium and upper ischium. The left proximal femur is not preserved. The right hip joint showed moderate osteoarthritic changes. Eburnation of the subchondral bone was also observed in both proximal humeri.
The possibility of osteoarthritis has been considered as an explanation for the vertebral and hip joint lesions observed in the La Chapelle-aux-Saints 1 Neanderthal 2–4. More precisely, La Chapelle-aux-Saints 1 was thought to have suffered from both osteoarthritis and spondylosis deformans, as the term osteoarthritis is reserved for the disease of diarthrodial joints, while that appellation has been replaced by the term spondylosis deformans for the intervertebral disk articulations 27–29. Both spondylosis deformans and osteoarthritis are recognized by production of osteophytes 30–34.
Spondylosis and facet joint osteoarthritis as well as osteoarthritis of the peripheral joints affecting both glenohumeral joints and both hip joints can undoubtedly be observed in La Chapelle-aux-Saints 1. Yet, the presence of erosions and reactive bone formation extending far beyond the left hip joint and the presence of actual subchondral and marginal erosions at the vertebrae precludes attribution of all damage to osteoarthritis 27,29. Although the term erosion has also been applied to the cartilage damage that osteoarthritis produces, this damage is characterized by mechanical abrasion and fissures rather than by the inflammatory process that produce erosions in spondyloarthropathy or infection 27,29,34,35.
The pattern of the pathological changes observed in La Chapelle-aux-Saints 1 is characteristic of brucellosis infection. Thus, vertebrae are involved in 2–53% of brucellosis infections 17,36,37. Typically, the disease initially affects a single vertebral endplate, mostly in the lumbar spine, but it can also be observed at multiple levels, affecting non-contiguous endplates and facet joints also in the thoracic and cervical spine 17,37. In addition, monoarthritis or oligoarthritis may occur, with the knee and hip joints being most commonly affected 12. The lesions have a specific erosive character and location seemingly pathognomonic for the disease, initially manifesting as an osteolysis/groove on the anterior aspect of superior vertebral endplates 17,38−41. Specifically referred to as Pedro i Pons’ sign 42,43, such grooves are not found with pyogenic (e.g., Staphylococcus) vertebral infections 44 or tuberculosis 27. Unfortunately, this characteristic destruction of the anterosuperior margin of the lumbar vertebrae cannot be observed in La Chapelle-aux-Saints 1 due to taphonomic damage of the anterior portion of all thoraco-lumbar vertebrae.
Examination of vertebrae C5 and C6 of La Chapelle-aux-Saints 1 revealed, however, erosive changes to the inferior vertebral surface (Fig. 1) that do not occur in spondyloarthropathy 27,29,45. The smooth character of the base of the groove in C6 evidences reactive new bone formation and not simply taphonomic bone loss (which would have exposed trabecular bone). Confluent erosion with new bone formation is quite different from the edge (Romanus) lesions noted with spondyloarthropathy 27,46. That appellation refers to apophyseal ring erosion, producing loss of the edge of the endplate in spondyloarthropathy, in contrast to production of a groove on its subchondral (disk-facing) surface. Such vertebral endplate grooves, however, are quite characteristic of the damage caused by brucellosis 17,38−43. In later stages of the infection, the erosive changes gradually expand over the entire vertebral endplate, producing the condition seen in vertebra C5.
Notation of erosions of the right facet joint of C4-C5 and C7-T1 and both facet joints of T9-T10 (Fig. 2–4) suggested differential diagnostic consideration of spondyloarthropathy or granulomatous infection. Facet joint involvement is found in up to one-third of individuals with brucellosis 6,39,47,48. However, such erosions of the facet joints and vertebral endplate are not anticipated with spondyloarthropathy 27,29,45. In addition, the transverse process of T1 and the laminae of T10 show large erosions characteristic of a space-occupying granulomatous process (see Figs. 3 and 4) 27,49−51. The latter are caused by infectious processes attributed to tuberculosis, fungal disease and brucellosis 27,29. Fungal disease produces erosions that do not respect variation in bone density, preserving a spheroid shape 49. Tuberculosis is characterized by zones of bone resorption 29,51. Neither pattern is observed in the La Chapelle-aux-Saints 1 Neanderthal.
The severe arthritic changes at the left hip joint have previously been attributed to trauma 3, but there is otherwise no evidence for trauma. The presence of erosions and new bone formation suggest that on top of the age-related primary osteoarthritis that might initially have been present (and which can also be observed in the right hip joint and both humeri), a secondary phenomenon, i.e., hip joint infection by brucellosis occurred. In fact, osteoarthritis-related changes would be confined to the joint region itself and do not extend to the extra-articular region as it is the case of La Chapelle-aux-Saints 1. Inflammatory processes that extend beyond the joint region are, however, not unusual in brucellosis, analogous to the paraspinal granulomas probably present in vertebrae T1 and T10 see also 12.
Neanderthals likely were infected by brucellosis during butchering of prey animals, not unlike the abattoir experience today 13, or by eating raw meat. The main causative organisms of brucellosis have been found in a wide range of wildlife. Particularly, Brucella melitensis has been reported in ibex and chamois, while B. abortus and B. suis have a preference for most other bovines and suids, including wild sheep, goats, horses, wild cattle, European bison, reindeer, roe and red deer, wild boar, but also hares and marmots 52–56. All these animals were important components of the Neanderthal diet 57–59. Only the two largest Neanderthal prey animals, mammoths and woolly rhinoceros, might not have been reservoirs for Brucella as inferred from their extant relatives 60–62. The host preference of the different Brucella species might suggest that La Chapelle-aux-Saints 1 was infected by B. abortus rather than B. melitensis. While the latter species often leads to acute pain and immobilization, B. abortus is known to cause a milder course of the disease. This might therefore explain the advanced stage of the arthritis in La Chapelle-aux-Saints 1 and his survival to the advanced age of perhaps well over 60 years 26.
Symptoms of brucellosis not only include fever, arthritis, endocarditis, neurologic defects, spinal deformity, reduced milk production, but also epididymitis/orchitis, infertility, still births and abortions 63. The latter could perhaps have been part of what led to the demise of Neanderthals, although the prevalence of brucellosis in Upper Palaeolithic modern humans is still unknown.
The La Chapelle-aux-Saints 1 burial of an old male Neanderthal was discovered in 1908 in a pit in the Bouffia Bonneval cave near the village of La Chapelle-aux-Saints in south-western France 1, together with dental remains of four additional individuals. Electron spin resonance (ESR) dating suggested a geological age of 47 ka ± 3 ka BP (early U-uptake model) or 56 ka ± 4 ka BP (linear U-uptake model) 64. The auricular surface of the ilium suggests an individual age of 62 ± 13 years 26. The partial skeleton preserves an almost complete skull with extensive post-canine tooth loss, a scapular fragment, portions of all long bones and both hipbones, as well as 19 variously complete vertebrae and a partial sacrum. The vertebrae include C1 and C2, C4 to T2, two fragmentary upper thoracic vertebrae (possibly T3 or T4 and T6 or T7), T8 to T12, L1, and L3 to S2 26,65.
The vertebrae of the La Chapelle-aux-Saints 1 Neanderthal were examined macroscopically and at 10× electronic magnification to characterize the presence and nature of vertebral endplate and facet alterations. In addition, CT scans of the vertebrae performed by the Musée de l’Homme were examined. All fossil bones were also surface scanned using a high-resolution surface scanner (PT-M4c, Polymetric GmbH, Darmstadt, Germany) 66.
The right hipbone is currently missing, and a cast as well as old photographs (curtesy of Eric Trinkaus) were examined.
Acknowledgements
We thank D. Grimaud-Hervé, L. Huet, V. Laborde, and A. Balzeau (Musée de l’Homme, Muséum Nationale d’Histoire Naturelle, Paris) for access to the La Chapelle-aux-Saints fossil remains. Financial support was provided by the Swiss National Science Foundation Grant Nos. 31003A_156299 and 31003A_176319 as well as the Mäxi Foundation, Switzerland.
Author contribution
BR and MH conceived the study, analysed the data and wrote the manuscript.
Competing interests
The authors declare no competing interests.
Data and materials availability
All data analysed during this study are included in this published article. The La Chapelle-aux-Saints 1 skeleton is housed at the Musée de l’Homme, Paris.