Australia: The Land Where Time Began

A biography of the Australian continent 

Denisovan Mandible from the Tibetan Plateau from the Late Middle Pleistocene

Denisovans are members of a hominin group who are known directly at present only from fragmentary fossils from a single site, Denisova Cave (Krause et al., 2010; Sawyer et al., 2015; Slon et al., 2017) in Siberia, the genomes of which have been studied. Also, they are known indirectly from their genetic legacy through gene flow into several East Asian populations (Browning et al., 2018; Sankararaman et al., 2016) from low altitudes and modern Tibetans living at high altitudes (Huerta-Sánchez et al., 2014). The ability to connect dispersed fossil hominids from Asia geographically and in time and to understand in a coherent manner their relation to recent Asian populations is hindered by the lack of morphologically informative Denisovan fossils. Included in this is understanding the genetic adaptation of humans to the Tibetan Plateau which is at high altitude (Chen et al., 2015; Meyer et al., 2017), was inherited from the Denisovans. In this paper Chen et al. report a Denisovan mandible, that was identified by analysis ancient protein (Walker, 2018; Walker, 2018), that was found in Baishiya Karst Cave, Xiahe, Gansu, China. Chen et al. determined that the mandible was at least 160,000 old by the use of U-series dating of an adhering carbonate matrix. Direct evidence of Denisovans outside the Altai Mountains is provided by the Xiahe specimen and the analysis of it provides insights into the mandibular and dental morphology of the Denisovans. It is indicated by the results of the analysis of Chen et al. that the Tibetan Plateau was occupied by archaic hominins in the Middle Pleistocene and that they adapted successfully to the hypoxic environments long before the arrival of modern Homo sapiens.

Denisovans are an extinct sister group of Neanderthals and are known only from fragmentary fossils that have been identified at Denisova Cave in Siberia. It has been found that their genomic legacy is present in several Asian, Australian and Melanesian populations (Browning et al., 2018; Reich et al., 2010; Qin & Stoneking, 2015; Skoglund & Jacobssen, 2011; Reich et al., 2011), which suggests that at some time in the past they might have been widespread. Denisovan introgression into Tibetans, Sherpas and neighbouring populations of the present includes positive selection for the Denisovan allele of the endothelial PAS domain-containing protein 1 (EPAS1), which provides adaptation to hypoxia at high altitudes in extant humans inhabiting the Tibetan Plateau (Huerta-Sánchez et al., 2014; Jeong et al., 2014; Hackinger et al., 2016). This adaptation that is derived from Denisovans is at present difficult to reconcile with the Denisova Cave at low latitude (700 m) and the earliest evidence of the presence of humans at high altitude on the Tibetan Plateau about 30-40 thousand BP (ky BP; taken as AD 1950) (Zhang et al., 2016; Yuan, Huang & Zhang, D, 2007; 2007; Zhao et al., 2009; Zhang, X. et al., 2018). Also, the relationships of various hominin fossils from the Middle Pleistocene and Late Pleistocene in East Asia with Denisovans are difficult to resolve because of morphological information on Denisovans being limited and the lack of palaeogenetic data on hominin fossils from the Middle Pleistocene in East Asia and tropical Oceania.

In 1980 the right half of a hominin mandible (the Xiahe mandible) was recovered from the Baishiya Karst Cave at an altitude of 3,280 m. Abundant stone artefacts from the Palaeolithic were recovered from a recent excavation, as well as animal bones with cut marks in the Baishiya Karst Cave. On the bottom of the specimen there is an in situ carbonate matrix which allowed the determination of a minimum age for the Xiahe mandible. 3 carbonate subsamples were collected for U-Th dating. The bulk ²³⁰Th age of 164.5 ± 6.2 ka does not differ statistically from the age of 155 ± 15 ka obtained from the uppermost and 163 ± 10 ka for the lowermost part of the crust (Student’s t-test, P<0.05). It is indicated by the consistency between the ages of subsamples from different places of the carbonate that this crust formed over a short period about 160 ka. Therefore, hominins had occupied the Tibetan Plateau at least as early as marine isotope stage 6 (MIS6) during the penultimate glacial phase.

No evidence has been found for the presence of ancient DNA in the Xiahe mandible. It has been revealed by palaeoproteomic analysis, however, the survival of an endogenous ancient proteome in a dentine sample, though not in a bone sample (Welker, 2018; Welker et al., 2016). The endogenous proteins are highly degraded and easily distinguished from contaminating modern protein. A phylogenetic tree that accurately reflects the phylogeny of the great apes, including the relationships between H. sapiens, Neanderthals and Denisovans was recovered from the proteome by phylogenetic analysis of the recovered proteome. The proteome of Xiahe is placed within this framework together with the Denisovan high-coverage genome (Meyer et al., 2012) (Denisovan Cave individual D3; Bayesian probability = 0.99). Also, the close relationship between the Xiahe proteome and Denisovans is further supported by the observed state of particular single amino acid polymorphisms within the Xiahe proteome. At present, only 1 high-coverage Denisovan genome is available (D3), which limits understanding of the proteome sequence diversity within Denisovans. Therefore Chen et al. assigned the Xiahe mandible to a population of hominins that are related closely to the Denisovans from Denisova Cave.

A carbonate matrix heavily encrusts the Xiahe mandible. This carbonate matrix is the only in situ material that can be associated directly with the mandible. The specimen was therefore scanned using micro-computed tomography and the carbonate matrix removed virtually by mannual segmentation. An archaic morphology is displayed by the Xiahe mandible that is rather common among hominins from the Middle Pleistocene. The fossil is metrically within the variation of this group. The body is very robust, relatively low and thick. Its height decreases slightly towards the back. There is a pronounced lateral prominence and well-developed anterior marginal tubercle. Internally, a protruding alveolar prominence overhangs a marked sub-alveolar fossa. The mylohyoid line runs in a low position from the mandibular foramen. The mental foramen is located under the P₄ and low on the body. There is no developed chin and the symphysis is receding strongly with an inclination angle of 69^o, though there is a triangular mental protuberance that is expressed weakly. Lingually, there is a primitive pattern of a small genioglossal fossa separating the upper and lower transverse tori. A geometric morphometric analysis, in which it falls within the sample of specimens from the Middle Pleistocene and at the limit of the Homo erectus distribution, close to the Chinese Lantian and Zhoukoudian HI.II and H1.12 specimens, confirms the primitive morphology of the Xiahe mandible. In contrast, it is well separated from H. erectus in geometric morphometric analysis of the shape of the dental arcade; it is less elongated than H. erectus and plots close to the primitive specimens of hominins from the Middle Pleistocene, Neanderthals and H. sapiens.

Also, the dental morphology fits within the variability of hominins from the Middle Pleistocene.  There are 5 well-developed primary cusps and a Y-fissure pattern on the M₁. In addition, it has 2 accessory cusps: tuberculum sextum (C6) and tuberculum intermedium (C7). There is no middle-trigonid crest on the outer enamel surface, though it is present at the enamel-dentine junction (EDJ) is a low continuous crest that dips at the sagittal sulcus (grade 2 in study that was published previously Baily et al., 2011). There are 2 wide roots - mesial and distal – each with bifurcated apices. The M₂ which was erupting but already in functional occlusion preserves the 5 principal cusps arranged in a Y pattern and a large C7. The between-group principal component analysis of the EDJ ridge and cervix shape groups the Xiahe M₂ with other specimens from the Middle Pleistocene, such as Mauer and Balanica, but away from the distribution of specimens of H. erectus.

The assignment of the Xiahe mandible to the Denisovans or a closely related sister population, orients the morphological comparisons towards the specimens of Denisovan from Denisova Cave, their sister group the Neanderthals and penecontemporaneous specimens from East Asia. The high mandibular body is lacking from Xiahe that is observed in specimens from Neanderthals and its symphyseal profile is more receding; however, the shape of the dental arcade that is broad anteriorly and flattened that is reminiscent of Neanderthals. A marked retromolar space is also present that is observed commonly in Neanderthals, though this feature might be linked to the agenesis of the M₃.

Its large dentition is the feature of the Xiahe specimen that best links it to the fragmentary fossils from Denisova cave. In the geometric morphometric form space, the size of the dental arcade of Xiahe is close to that of Tighénif 3 and Irhoud 11 and surpasses all other comparative specimens along between-group principal component 1, except for the KNM-WT15000. Analysis of the teeth of Xiahe confirms that it had a combination of a moderately large mandible with a dental arcade that is exceptionally large. The bucco-lingual diameter of the M₁ is larger than the mean of all the samples from Europe dating to the Middle Pleistocene. The mesio-distal diameter is larger than, as well as being outside the range of, all the comparative specimens that were available for this study. The bucco-lingual diameter of the M₂ of Xiahe is at the high end of the range of specimens of H. erectus. Its mesio-distal diameter is within the range of the sample from Europe that dates to the Middle Pleistocene but outside the range of the remainder of the comparative samples.

Xiahe is comparable to the mandible of Penghu 1 (Chang et al., 2015) in several aspects among the penecontemporaneous specimens from East Asia. A 3-D dimensional surface model of Penghu 1 is not available for geometric morphometric analysis though the bone metrics and dental morphology are very similar. Penghu 1 is similar to Xiahe in that it displays agenesis of M₃. Though the M₂ of Penghu 1 is smaller than that of Xiahe, the M₂ crown is wider mesio-distally than bucco-Lingually, in both individuals. The M₂ roots are notable in Xiahe and Penghu 1 specimens. As well as mesial and distal plate-like roots, there is a prominent accessory lingual root that branches from the mesial root below the cervix. This feature is of particular interest as mandibular molars with 3 roots are very rare in H. sapiens outside Asia, though they appear to occur much more frequently in recent Asian populations. Finally, the P₃ in both specimens display Tomes’ root, a feature that is occasionally observed in other fossil hominins. Among other non-H. sapiens specimens from China, the EDJ of the molars of both Xiahe and Xujiayao exhibit occlusal basins that are moderately complex, and the M₂ of Xiahe and the M3 of Xujiayao possess a similar protostylid crest (Xing et al., 2015). It is notable that as with the Xiahe mandible, morphological similarities to Neanderthals have also been described for cranial remains from Xujiayao and Xuchang. The idea that Denisovans are already represented in the fossil record of China, particularly such fossils as Penghu 1 and Xujiayao is reinforced by these observations. This hypothesis can be tested directly by ancient protein analysis of these specimens in the future.

According to Chen et al. as far as they know the Xiahe mandible is the first Denisovan to be identified outside Denisova Cave (Krause et al., 2010; Reich et al., 2010). The minimum age of the Xiahe mandible, about 160 ka, makes this mandible of comparable age to Denisova 2, which is chronologically the oldest Denisovan that is known at present from Denisova Cave (Sawyer et al., 2015; Slon et al., 2017; Douka et al., 2019). It has been estimated that Neanderthals and Denisovans diverged about 445-473 ka (Prüfer et al., 2014). It therefore remains to be documented when the early part of Denisovan lineage began. The Xiahe mandible demonstrates, however, that Denisovans or Denisovan-like populations have deep roots in central East Asia.

As far as is known, the Xiahe mandible represents the earliest hominin fossil on the Tibetan Plateau. This mandible is at least 120,000 older than the oldest known Palaeolithic sites, which are about 30-40 ka, in the region (Zhang, D. et al., 2016; Zhang, X. et al., 2018). It has generally been considered that successful colonisation of, and adaptation to, high altitude plateaus such as the Himalayas was restricted to recent H. sapiens (Zhang, X. et al., 2018), especially because of adverse conditions, such as a scarcity of resources, low temperatures and hypoxia. The Xiahe mandible demonstrates instead that the Tibetan Plateau was originally occupied by archaic hominins where they successfully adapted to such environments. It is suggested by the evidence that Denisovans or Denisovan-related populations accumulated adaptive alleles that are beneficial to the occupation of environments at high altitude over their presence on the Tibetan Plateau. In modern Himalayan populations archaic hominin alleles, such as the Denisovan-derived EPAS1 allele (Huerta-Sánchez, et al., 2014; Hackinger et al., 2016), may therefore derive from a local archaic hominin on or around the Tibetan Plateau, such as the Xiahe hominin population.

Evidence of deep evolutionary history of these archaic hominins within the challenging environment of the Tibetan Plateau is provided by the Xiahe Denisovan. The Xiahe mandibular and dental anatomy related other Chinese fossil hominins to the Denisovans and resolves several outstanding questions concerning this important hominin group. The analyses of Chen et al. paves the way towards a better understanding of the evolutionary history of hominins in East Asia from the Middle Pleistocene.

Sources & Further reading

Chen, F., et al. (2019). "A late Middle Pleistocene Denisovan mandible from the Tibetan Plateau." Nature 569(7756): 409-412.