生物組 計劃簡介

Origin of Modern Humans in China: a Molecular Approach

以分子生物學技術探討現代中國人的起源

1. Background

It has been widely agreed among anthropologists that living populations of humans (Homo sapiens) differ substantially in certain anatomical features (e.g., cranial size and shape) from the more archaic Upper Pleistocene representatives of the species. However, how the transition from archaic to modern humans has occurred is a highly controversial issue. There have been two major competing hypotheses (models) for the origin of modern humans. One is the so-called “Out of Africa” model, which postulates that modern humans originated in Africa and a group of individuals of this new species moved out of Africa and completely replaced all archaic populations in Europe and Asia (see Stringer and Andrews 1988). This hypothesis was based mainly on the fact that the earliest fossils of anatomically modern humans were found in Africa. The other hypothesis is the “multiregional evolution” model, which postulates that modern humans evolved in different areas from already differentiated ancestral groups of archaic Homo sapiens or Homo erectus (see Wolpoff et al. 1984; Frayer et al. 1993). This model was based mainly on the evidence of “regional continuity” in morphological evolution in fossil remains in China and Australia.

There are important differences between the “Out of Africa” and the “multiregional” models. The former stresses a complete replacement of all indigenous populations in the rest of the Old World by an African stock; the replacement was suggested to have started about 100,000 years ago (Stringer and Andrews 1988). Under this model, any anatomical features that might have evolved in different regions of the world in earlier Homo erectus or archaic Homo sapiens populations would have been lost, so that there would be no specific regional continuity of anatomical characteristics from, say, one million years ago through to the modern world. In contrast, the “multiregional” model predicts continuity of regionally developed characteristics. In terms of genetic changes, the differences between the two models are even more dramatic. Since the “Out of Africa” model assumes that the original African stock represented a new species and could not hybridize with any indigenous regional populations, no archaic humans or indigenous regional populations outside of Africa had contributed to the gene pool of modern humans. On the other hand, the multiregional model assumes no complete replacement of local gene pools and predicts regional genetic continuity.

The question of the origin of modern humans in China is particularly relevant to this controversy because a large part of the fossil record used to argue for the multiregional model came from China. More recently, Wu and Brauer (1993) reported several cranial features that existed both in the broad spectrum of African archaic Homo sapiens and the narrow spectrum of archaic Chinese variation. For instance, the most prominent point on the frontal squama was lower in Chinese specimens, while some were in a similarly lower position and others were higher in African crania. Recently, Wu (1998) found that in later Chinese crania, the condition of the above-mentioned features was the same as in the earlier archaic crania, indicating that Chinese populations continuously present a narrow spectrum of variation that occupies the same part of a broad spectrum. This continuity in morphological features is contradictory to the Out of Africa model. The similarity of the narrow spectrum of variation in archaic and more recent Chinese is best explained by descent, i.e., by the multiregional model.

Wu (1998) believes that dental evidence also supports the same explanation. Irish (1997) showed that many dental features strongly contrast Africans and East Asians. For example, the frequency of shovel-shaped incisors is much higher in East Asians than in Africans. Under the Out of Africa model, this observation implies a rapid increase in shoveling frequency, but the fact that all maxillary incisors dated to the Pleistocene found thus far in East Asia are shoveled implies a continuation of high frequencies from earlier archaic East Asian samples to modern ones. He further argues that the Out of Africa model is not supported by Paleolithic data. Under this model early modern invaders from western Asia should have brought Mousterian culture, because the earliest modern humans in Near East were associated with this distinctive culture. However, there has been only one site with Mousterian artifacts in the Chinese Paleolithic – it is Shuidonggou, in the northwestern part of China. The absence of a sharp contrast between earlier and later Paleolithic cultures in China is contradictory to the Out of Africa model. However, evidence of gene flow between China and other areas was suggested by some morphological features on some skulls in China such as the chignon-like structures in Liujiang and Ziyang, and more laterally facing of the antero-lateral surface of the spheno-frontal process of the zygomatic bone of the Upper Cave skull no. 102. These features have never been found in earlier skulls of China but are typical of Neandertals. For this reason, Wu (1998) proposed the model of “continuity with hybridization” for the origin of modern humans in China. In this model, continuity represents the main process while hybridization (with western invaders) is subsidiary.

On the other hand, Chu et al. (1998) argued that their study of the genetic profiles in 28 populations sampled from China suggested a single origin for these populations and claimed that modern humans originating in Africa constitute the majority of the current gene pool in China. However, this study seems to be the only extensive molecular study to date on this issue and is not sufficient for drawing a general conclusion. Moreover, the study was based on microsatellites, which are not very suitable for estimating the time depth of the genetic history of a population. As explained above, the Out of Africa model implies a shallow genetic history in Asia, whereas the multiregional model implies the existence of a deep history at certain genetic loci. Therefore, there is an urgent need for molecular data that are suitable for estimating the time depth of genetic history in Chinese populations and other East Asians.

2. Objectives

This subproject has two objectives. One objective is to train a young scientist who will have (1) excellent molecular biology knowledge and skills, (2) an adequate background in theoretical population genetics and evolution, and (3) a basic background in physical and cultural anthropology, and linguistics. As is clear from the above description, it requires a multidisciplinary approach to the question of the origin of modern humans in China. In addition to the urgent need for molecular data, there is a requirement for having the skills to conduct an adequate statistical analysis of molecular population data. Moreover, a good knowledge of the fossil, morphological, Paleolithic, and linguistic data that are involved in the controversy is useful for understanding the issue and for a better interpretation of the molecular data. Thus, we aim to train a young scientist who will be capable of using such a multidisciplinary approach to the issue.

The other objective is to obtain a substantial amount of molecular data that will be useful for resolving the controversy on the origin of modern humans in China. Specifically, we propose the following work:

(1) Select one noncoding X-linked region of ~2,000 base pairs (~2 kb) and four ~2 kb noncoding autosomal regions; each region selected should contain at least 2 Alu repeats.

(2) Sequence these five regions in 70 individuals: 40 East Asians (15 northern Hans, 15 southern Hans, 5 Mongolians, 3 Cambodians, and 2 Vietnamese), and 30 Africans (10 Nigerians, 10 black South Africans, and 10 pygamies). In addition, we will use one chimpanzee, one gorilla, and one orangutan as outgroup references.

(3) Use the molecular data to be obtained to estimate the levels of nucleotide diversity in East Asians and Africans, the genealogical relationships among haplotypes, and the age of the most recent common ancestor of the DNA sequences from Asia (and from Africa). As will be explained later, these analyses will provide much insight into the relationships among the modern humans in general and the origin of modern humans in China (east Asia) in particular.

We have not chosen any Europeans because there is now strong evidence that Europe was not the origin of modern humans and because inclusion of Europeans will increase considerably the amount of molecular work but will add little information to the issue. We have chosen 30 Hans but no Chinese minorities because minorities occupy less than 7% of the Chinese population. Instead, we include 5 Mongolians, 3 Cambodians, and 2 Vietnamese because these additional samples represent better East Asians than the Chinese minorities. Our strategy is to minimize the molecular work but maximize the genetic information.

3. Significance

Human evolution is a truly multidisciplinary subject, involving not only genetics and evolutionary biology but also cultural and physical anthropology, linguistics, paleontology, archeology, and statistics. We are not aware of any single scientist in Taiwan who possesses such broad knowledge and skills. We therefore propose to train such a young scientist. Although this scientist’s major expertise will be in molecular biology or molecular evolutionary biology, he/she will receive a basic training in each of the following areas: cultural and physical anthropology, linguistics, population genetics theory, statistics, and computer skills. With such a broad training, this scientist should be able to conduct statistical analyses of molecular data and should be able to connect the molecular data with data from other fields such physical and cultural anthropology, and linguistics.

The issue of the origin of modern humans in China (or East Asia) lies at the heart of the controversy over the origin of modern humans worldwide. Fossil remains from China have repeatedly challenged the Out of Africa model, suggesting continuity in morphological evolution with a minor contribution from the African stock (see above). If this hypothesis of “continuity with hybridization” based on fossil and Paleolithic data is also supported by molecular genetic data, then the evolution of modern humans is closer to the multiregional model than to the Out of Africa model. To date, Chu et al. (1998) seems to be the only genetic study of this issue. Although these authors claimed that their result suggested a major African contribution to the modern gene pool in China, their data is limited and did not provide an estimate of the time depth of the genetic history of Chinese or East Asians. At any rate, molecular evolution is a random process, so that data from many different genetic loci are needed to draw a general conclusion on the genetic history of a population.

The molecular genetic data to be obtained in this project is aimed to rectify the above deficiency. We have chosen noncoding regions because they are not directly subject to natural selection and can reflect the history of a population better than conding sequences. To increase the genetic information, each region will include at least two Alu repeats because Alu sequences evolve more rapidly than single copy sequences, so that a considerable amount of genetic information can be obtained from sequencing just a limited number of repeats. However, the information carried by a single repeat is limited, so we propose to sequence at least two adjacent repeats at each time. In total, more than 10 Alu repeats and nearby sequences will be obtained. Although haplotype determination is easier for X-linked sequences than for autosomal sequences, X-linked sequences have a smaller effective size than autosomal sequences and would have, on average, a shallower history than autosomal sequences. Therefore, autosomal regions are more suitable for our purpose than X-linked ones and we will study four autosomal regions and only one X-linked region.

The new sequence data will be useful for at least three purposes. First, they can be used to compute the levels of nucleotide diversity in Asian and African populations. Nucleotide diversity (p) is a common measure of genetic diversity and is defined as the number of nucleotide differences per site between two randomly chosen sequences from a population. A comparison of the p values in Africans and East Asians is informative. For example, if the p values are comparable, then East Asians would have not gone through a severe bottleneck and so the assumption of complete replacement of archaic Asians by an African stock as postulated in the Out of Africa model is not plausible. Second, the haplotypes can be used to infer genealogical relationships among Asians and Africans and can provide information about the level of African contribution to the gene pool in East Asia. Third, the data can be used to estimate the age (T) of the most recent common ancestor (MRCA) of East Asian samples (see Fu and Li 1997). A large T value (say, > 200,000 years) for a region would imply a deep genetic history of East Asians and so would be incompatible with the Out of Africa model. In short, the data to be obtained will shed much light on the issue under investigation.