From old markers to next generation: reconstructing the history of the peopling of Sardinia.

CONTEXT: For many years the Sardinian population has been the object of numerous studies because of its unique genetic structure. Despite the extreme abundance of papers, various aspects of the peopling and genetic structure of Sardinia still remain uncertain and sometimes controversial. OBJECTIVE: We reviewed what has emerged from different studies, focussing on some still open questions, such as the origin of Sardinians, their relationship with the Corsican population, and the intra-regional genetic heterogeneity. METHODS: The various issues have been addressed through the analysis of classical markers, molecular markers and, finally, genomic data through next generation sequencing. RESULTS AND CONCLUSIONS: Although the most ancient human remains date back to the end of the Palaeolithic, Mesolithic populations brought founding lineages that left evident traces in the modern population. Then, with the Neolithic, the island underwent an important demographic expansion. Subsequently, isolation and genetic drift contributed to maintain a significant genetic heterogeneity, but preserving the overall homogeneity on a regional scale. At the same time, isolation and genetic drift contributed to differentiate Sardinia from Corsica, which saw an important gene flow from the mainland. However, the isolation did not prevent gene flow from the neighbouring populations whose contribution are still recognisable in the genome of Sardinians.

Analysis of uniparental markers reveals a complex pattern of migration within Sardinia.

BACKGROUND: From the genetic viewpoint, Sardinia is well differentiated from other surrounding populations. In spite of a common ancestral origin, substantial genetic heterogeneity is observed within the island. Matrimonial pattern, as well as past migration movements, may account for the complex genetic structure of Sardinia. AIM: To compare data from uniparental markers in order to highlight the migration pattern of male and female lineages and check their congruence with the demographic data. SUBJECTS AND METHODS: Genomic DNA was obtained from 279 unrelated males selected from three isolated villages and from three open populations representative of North, Central and South Sardinia. The hypervariable region 1 of mtDNA was sequenced and 17 Y-chromosome loci were genotyped. Parameters of within and among populations diversity were calculated and analysis of migration was performed. RESULTS: When analysed as a whole population, demographic data show a balanced movement of males and females in Sardinia, unlike other Italian and European populations. Remarkably, when the island is divided into geographic areas, different migration patterns are clearly recognisable. Whereas North and Central Sardinia populations show a stronger male migration rate, the South Sardinia population shows a stronger female migration rate. CONCLUSION: Distinct migration patterns of male and female lineages affect the areas investigated differently. These past migration movements are major contributors to the complex genetic structure currently observed in the Sardinian population.

Analysis of 16 STRs of NOS gene regions and around in six Sardinian populations (Italy).

OBJECTIVES: The aims of this work are to provide first data on novel STRs at the NOS gene regions in human populations and to test for possible correlations with mortality rate by malaria in different areas of Sardinia (Italy). METHODS: In the present study, 16 STRs (13 analyzed for the first time in human populations) localized on three genes NOS were typed in 213 healthy individuals, unrelated for at least three generations, from six historical-geographical Sardinian areas. STRs alleles were determined through sequencing. Statistical analyses were performed by Genepop (v.4.0), Arlequin (v.3.5.1.2), R (v.2.15.1), Statistica (v.5.1), and PHASE (v.2.1) software packages. RESULTS: The number of alleles found for each locus ranged from 2 to 12 and their distribution is most often unimodal. All populations met Hardy Weinberg equilibrium after Bonferroni correction, with few exceptions. Analysis of genetic distances did not show strong genetic structuring of the investigated populations. Instead, the population genetic variability shows a positive and highly significant (P-value < 0.01) correlation between mortality determined by malaria infection and alleles (TGGA)7 of NOS2, (AAAAG)2 and (ATTT)10 of adNOS1, and (AAACA)11 of adNOS3 genes. CONCLUSIONS: The peculiar allele distribution found for several NOS alleles could be due to malaria infection that may have contributed to their frequencies, but we cannot exclude that the peculiar allele distribution of NOS might also be due to genetic drift, emphasized by isolation and founder effect.

The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269.

Recently, the debate on the origins of the major European Y chromosome haplogroup R1b1b2-M269 has reignited, and opinion has moved away from Palaeolithic origins to the notion of a younger Neolithic spread of these chromosomes from the Near East. Here, we address this debate by investigating frequency patterns and diversity in the largest collection of R1b1b2-M269 chromosomes yet assembled. Our analysis reveals no geographical trends in diversity, in contradiction to expectation under the Neolithic hypothesis, and suggests an alternative explanation for the apparent cline in diversity recently described. We further investigate the young, STR-based time to the most recent common ancestor estimates proposed so far for R-M269-related lineages and find evidence for an appreciable effect of microsatellite choice on age estimates. As a consequence, the existing data and tools are insufficient to make credible estimates for the age of this haplogroup, and conclusions about the timing of its origin and dispersal should be viewed with a large degree of caution.

Sampling strategies in a linguistic isolate: results from mtDNA analysis.

OBJECTIVES: Sampling strategies are crucial issues in population genetics and anthropological studies. The sampling choice is related to the research question and the type of markers used. In this research, we compared two different sampling strategies in the Sardinian linguistic isolate of Carloforte (Italy). METHODS: A first sampling (N = 49) was carried out through grandparents criterion: individuals selected for the study were born and resident in Carloforte, and unrelated for at least three generations. A second sampling (N = 50) was based on founders surnames (FS): selected participants were proved to be descendants of the village founders, and to have no ancestors in common, at least up to the grandparental generation. RESULTS: The group selected through FS showed a greater gene diversity, which was confirmed by both network and haplogroup analysis. Among the shared haplogroups, we find clear differences in their frequencies. Sampling through grandparents criterion showed essentially the same haplogroups found in Sardinia, and with similar frequencies. Interesting results came from genetic tree. The FS sampling clustered with Northern African populations and it is located very far from Italian and Sardinian populations, whereas the grandparents criterion sampling clustered with Italian populations and it is located close to the other Sardinian populations. CONCLUSIONS: Results showed that different sampling strategies can lead to contrasting results. As sampling through grandparents criterion is influenced by recent gene flow, we hypothesize that the difference observed with the two sampling strategies is due to the merging of Carloforte with Sardinian populations.

Selective neutrality analysis of 17 STRs in Mediterranean populations.

Detection of genes that have been targeted by natural selection is a powerful tool for predicting regions of the genome potentially linked with diseases and of interest in the field of genetic epidemiology. In recent years, several methods to detect patterns of natural selection have been developed. In general, these tests are based on different assumptions and parameters; hence, the detection of outlier loci with more than one statistical approach simultaneously will support the candidate status of a particular locus. In this study, we evaluated the presence of patterns of positive selection in 17 short tandem repeat loci genotyped in six different human populations from the Mediterranean area, for a total of 429 individuals. To identify patterns of selective pressure, we applied three different neutrality tests on the basis of different models, performing pairwise comparisons between populations. Results show the presence of one marker, a (CA)n repeat located in exon 29 of the NOS1 gene, which seems significant in the three different tests in two pairwise comparisons: Sicily vs Morocco and Balearic Islands vs Morocco. This suggests that this locus and its genome localization are candidates for further studies to investigate selective pressure, as well as for association studies.

Population data for Y-chromosome haplotypes defined by AmpFlSTR YFiler PCR amplification kit in North Sardinia (Italy).

The 17 Y-chromosomal short tandem repeats (STRs) included in the AmpFlSTR YFiler Amplification Kit (AB Applied Biosystems) (DYS19, DYS3891, DYS389II, DYS390, DYS391, DYS392, DYS393, DYS385, DYS437, DYS438, DYS439, DYS448, DYS456, DYS458, DYS635 and GATA H4.1) were typed in 100 samples from North Sardinia (Italy). A total of 91 different haplotypes were found, where 9 haplotypes were shared by two individuals. The overall haplotype diversity (HD) was 0.9982. DYS458 non-consensus alleles were found in one samples, and one in the DYS438. We found a double peak in one sample for the DYS19 with alleles 15/16. Population comparisons with available 10 YSTR loci data in Mediterranean Basin samples were undertaken, significant differences were observed between our sample and all the compared populations, except for a entire sample from Sardinia. Prediction of haplogroups showed I2al was found to be the most frequent haplogroup (33%) in our sample. Testing high-resolution Y-chromosome data sets it is useful in autochthonous population and micro-population studies to highlight the most informative loci for evolutionary aims.

Inter-individual genomic heterogeneity within European population isolates.

A number of studies carried out since the early '70s has investigated the effects of isolation on genetic variation within and among human populations in diverse geographical contexts. However, no extensive analysis has been carried out on the heterogeneity among genomes within isolated populations. This issue is worth exploring since events of recent admixture and/or subdivision could potentially disrupt the genetic homogeneity which is to be expected when isolation is prolonged and constant over time. Here, we analyze literature data relative to 87,815 autosomal single-nucleotide polymorphisms, which were obtained from a total of 28 European populations. Our results challenge the traditional paradigm of population isolates as structured as genetically (and genomically) uniform entities. In fact, focusing on the distribution of variance of intra-population diversity measures across individuals, we show that the inter-individual heterogeneity of isolated populations is at least comparable to the open ones. More in particular, three small and highly inbred isolates (Sappada, Sauris and Timau in Northeastern Italy) were found to be characterized by levels of inter-individual heterogeneity largely exceeding that of all other populations, possibly due to relatively recent events of genetic introgression. Finally, we propose a way to monitor the effects of inter-individual heterogeneity in disease-gene association studies.

The value of some Corsican sub-populations for genetic association studies.

BACKGROUND: Genetic isolates with a history of a small founder population, long-lasting isolation and population bottlenecks represent exceptional resources in the identification of disease genes. In these populations the disease allele reveals Linkage Disequilibrium (LD) with markers over significant genetic intervals, therefore facilitating disease locus identification. In a previous study we examined the LD extension on the Xq13 region in three Corsican sub-populations from the inner mountainous region of the island. On the basis of those previous results we have proposed a multistep procedure to carry out studies aimed at the identification of genes involved in complex diseases in Corsica. A prerequisite to carry out the proposed multi-step procedure was the presence of different degrees of LD on the island and a common genetic derivation of the different Corsican sub-populations. In order to evaluate the existence of these conditions in the present paper we extended the analysis to the Corsican coastal populations. METHODS: Samples were analyzed using seven dinucleotide microsatellite markers on chromosome Xq13-21: DXS983, DXS986, DXS8092, DXS8082, DXS1225, DXS8037 and DXS995 spanning approximately 4.0 cM (13.3 Mb). We have also investigated the distribution of the DXS1225-DXS8082 haplotype which has been recently proposed as a good marker of population genetic history due to its low recombination rate. RESULTS: the results obtained indicate a decrease of LD on the island from the central mountainous toward the coastal sub-populations. In addition the analysis of the DXS1225-DXS8082 haplotype revealed: 1) the presence of a particular haplotype with high frequency; 2) the derivation from a common genetic pool of the sub-populations examined in the present study. CONCLUSION: These results indicate the Corsican sub-populations useful for the fine mapping of genes contributing to complex diseases.

Frequencies of promoter pentanucleotide (TTTTA)n of CYP11A gene in European and North African populations.

The present work attempts to determine the distribution of CYP11A (TTTTA)n genotype and allele frequencies in 10 European and North African populations. This polymorphism has been associated with hyperandrogenism by several association studies. To our knowledge, this is the first study investigating the ethnic variation of this polymorphism. DNA was extracted from 868 whole-blood samples with the standard phenol-chloroform technique, and PCR reactions were carried out using fluorescent primers as described previously. PCR products were analyzed by an ABI 3,730 DNA Analyzer. A total of six alleles were identified, ranging from 220 bp (4 repeats [4R]) to 250 bp (10R). The most frequent allelic fragment size in all populations was 4R, with frequencies ranging from 47.9% (Sicily) to 62.8% (Tuscany and Germany). Allelic frequencies showed high heterogeneity between analyzed populations. We detected a significant gradient for alleles 4R and 8R. In this study, we report the allele frequency distribution of CYP11A (TTTTA)n showing a north-south geographic gradient. This result could be useful for epidemiological studies about hyperandrogenism.

Population genetic data on four STR loci, PAI (CA)n, GpIIIa (CT)n, PLAT (TG)14 (CA)12, and NOS2A (CCTTT)n, in Mediterranean populations.

In the present study, we have analyzed four highly polymorphic regions (STRs) chosen from four candidate genes involved in: (1) Platelet aggregation: alpha subunit of the platelet GpIIb/GpIIIa integrin complex (GpIIIa (CT)n; 17q21.31), (2) Coagulation fibrinolysis: Plasminogen Activator Tissue (PLAT5 (TG)14(TA)12; 8p12-q11.2) and Plasminogen Activator Inhibitor-1 (PAI-1 (CA)n; 7q21.3-q22), (3) Oxidative metabolism: the inducible nitric oxide (NO) synthase (iNOS) gene (NOS2A (CCTTT)n; 17cen-q11.2). Allele frequencies for these four STR loci were investigated in several Mediterranean populations. The population data deviate from the Hardy-Weinberg equilibrium in all populations for GpIIIa (CT)n polymorphism.

Overcoming the dichotomy between open and isolated populations using genomic data from a large European dataset.

Human populations are often dichotomized into "isolated" and "open" categories using cultural and/or geographical barriers to gene flow as differential criteria. Although widespread, the use of these alternative categories could obscure further heterogeneity due to inter-population differences in effective size, growth rate, and timing or amount of gene flow. We compared intra and inter-population variation measures combining novel and literature data relative to 87,818 autosomal SNPs in 14 open populations and 10 geographic and/or linguistic European isolates. Patterns of intra-population diversity were found to vary considerably more among isolates, probably due to differential levels of drift and inbreeding. The relatively large effective size estimated for some population isolates challenges the generalized view that they originate from small founding groups. Principal component scores based on measures of intra-population variation of isolated and open populations were found to be distributed along a continuum, with an area of intersection between the two groups. Patterns of inter-population diversity were even closer, as we were able to detect some differences between population groups only for a few multidimensional scaling dimensions. Therefore, different lines of evidence suggest that dichotomizing human populations into open and isolated groups fails to capture the actual relations among their genomic features.

Genetic isolates in Corsica (France): linkage disequilibrium extension analysis on the Xq13 region.

Genetic isolates with a history of a small founder population, long-lasting isolation and population bottlenecks represent exceptional resources in the identification of genes involved in the pathogenesis of multifactorial diseases. In these populations, the disease allele reveals linkage disequilibrium (LD) with markers over significant genetic intervals, therefore facilitating disease locus identification. This study has been designed to examine the background LD extension in some subpopulations of Corsica. Our interest in the island of Corsica is due to its geographical and genetic proximity to the other Mediterranean island of Sardinia. Sardinian isolates in which the extension of the background LD is particularly high have been recently identified and are now the object of studies aimed at the mapping of genes involved in complex diseases. Recent evidence has highlighted that the genetic proximity between the populations of Corsica and Sardinia is particularly true for the internal conservative populations. Given these considerations, Sardinia and Corsica may represent a unique system to carry out parallel association studies whose results could be validated by comparison. In the present study, we have analyzed the LD extension on the Xq13 genomic region in three subpopulations of Corsica: Corte, Niolo and Bozio, all located in the mountainous north-center of the island. Our results show a strong degree of LD over long distance for the population of Bozio and to a less extent for the population of Niolo. Their LD extent is comparable to or higher than that reported for other isolates.

Y-chromosome 10 locus short tandem repeat haplotypes in a population sample from Sicily Italy.

This study reports the first data on Y-chromosome-specific short tandem repeat (STR) haplotype frequencies, in the population of the island of Sicily (Italy), based on the combination of alleles at the following 10 Y-chromosome loci DYS19, DYS389I, DYS389II, DYS390, DYS391, DYS392, DYS393, DYS437, DYS438, and DYS439. In a total of 117 males, 108 unique haplotypes were observed, with 99 of them being singletons. The 10 locus haplotypes generated a diversity value of 0.9987 and discriminatory power (DP) of 92.30%. The data on the seven of the 10 polymorphisms (DYS19; DYS389I; DYS389II; DYS390; DYS391; DYS392 and DYS393) that have been most studied in worldwide populations were compared with similar data from neighboring Mediterranean populations in order to address the question of shared ancestry, gene flow and population affinities. Overall, results indicate Sicily is closest genetically to the mainland Italian population but also with evidence of a significant African component in the male gene pool. These findings are consistent with those obtained from other genetic markers (autosomal and mitochondrial DNA as well as the classical blood groups) and also with the recorded settlement history (either peaceful or due to invasion) of the island.

A pentanucleotide repeat polymorphism (TTTTA) in the apolipoprotein (a) gene--its distribution and its association with the risk of cardiovascular disease.

Apolipoprotein (a) is a component of lipoprotein (a). Several studies have shown the association between risk of coronary heart diseases and the size of apo(a) isoforms, although this issue is still controversial. Recent researches focused the attention on the pentanucleotide (TTTTA), highlighting a statistical correlation between low Lp(a) levels and high repeat numbers. In the present paper we studied the distribution of the apo(a) pentanucleotide polymorphism among populations from Corsica, and we then compared it with other populations from Europe, Africa and Asia. The results stressed out the usefulness of these markers in population genetics analysis. We later investigated the possible association of the apo(a) pentanucleotide polymorphism with serum lipid levels in two samples from Corsica (France): one comprises patients or individuals with high risk of future coronary heart disease and the other is a control sample. No significant differences between the two groups have been found, but the analysis of variance showed a significant association between different genotypes and cholesterol and LDL serum levels.

Molecular variation in endothelial nitric oxide synthase gene (eNOS) in western Mediterranean populations.

Endothelial nitric oxide synthase (eNOS or NOS3) is the main responsible for nitric oxide (NO) production in vascular system and different polymorphisms have been identified in epidemiological studies. Trying to test the eNOS genetic variation in general populations we studied the 27-bp VNTR in intron 4 and G894T substitution in exon 7 markers in 6 Western Mediterranean populations (3 from Iberian Peninsula, 1 from North Africa, and 2 from Sardinia) and a sample from Ivory Coast. The VNTR frequencies in Western Mediterranean and Ivory Coast fit well into the ranges previously described for Europeans and Sub-Saharans respectively, and a typical African allele has been detected in polymorphic frequencies in the Berber sample. The G894T substitution presents the highest frequencies described for the T allele in the North Mediterranean populations. Linkage disequilibrium is present between both markers in all populations except in the Ivory Coast sample. The variation found for these polymorphisms indicates that they may be a useful tool for population studies even at microgeographical level.

Molecular variation at functional genes and the history of human populations--data on candidate genes for cardiovascular risk in the Mediterranean.

A screening of 22 DNA polymorphisms has been performed in western Mediterranean populations (Iberian Peninsula, Morocco, and Central Mediterranean Islands). The analyzed markers correspond to polymorphic sites in several candidate genes for cardiovascular disease including apolipopoteins and their receptors (APOA1, APOB, APOE, APOC1, APOC2, LPA, and LDLR), genes implied in the hemostasis regulation (Factor VII, alpha and beta-fibrinogen, alpha and beta platelet-integrin, tissue plasminogen activator, and plasminogen activator inhibitor-1), and the angiotensin converting enzyme gene. The results are presented of a partial analysis carried out in following population samples: 6 from the Iberian Peninsula, 2 from Morocco, and 3 from Central Islands. The degree of inter-population diversity was significant and consistent with data from other kind of genetic polymorphisms. The apportionment of the allele frequency variance supported a geographic structure into three main regions: Central Mediterranean Islands, the Iberia Peninsula and North Africa. The genetic distance pattern is compatible with a south-to-north North African influence in the Iberian Peninsula and a remarkable gene flow from sub-Saharan Africa into Morocco. Epidemiologically, North Africa is characterized by high frequencies of LPA PNR alleles with high number of repeats (protective for cardiovascular risk) and high frequencies of the APOE*E4 allele (risk factor) as compared with European populations.

Linguistic, geographic and genetic isolation: a collaborative study of Italian populations.

The animal and plant biodiversity of the Italian territory is known to be one of the richest in the Mediterranean basin and Europe as a whole, but does the genetic diversity of extant human populations show a comparable pattern? According to a number of studies, the genetic structure of Italian populations retains the signatures of complex peopling processes which took place from the Paleolithic to modern era. Although the observed patterns highlight a remarkable degree of genetic heterogeneity, they do not, however, take into account an important source of variation. In fact, Italy is home to numerous ethnolinguistic minorities which have yet to be studied systematically. Due to their difference in geographical origin and demographic history, such groups not only signal the cultural and social diversity of our country, but they are also potential contributors to its bio-anthropological heterogeneity. To fill this gap, research groups from four Italian Universities (Bologna, Cagliari, Pisa and Roma Sapienza) started a collaborative study in 2007, which was funded by the Italian Ministry of Education, University and Research and received partial support by the Istituto Italiano di Antropologia. In this paper, we present an account of the results obtained in the course of this initiative. Four case-studies relative to linguistic minorities from the Eastern Alps, Sardinia, Apennines and Southern Italy are first described and discussed, focusing on their micro-evolutionary and anthropological implications. Thereafter, we present the results of a systematic analysis of the relations between linguistic, geographic and genetic isolation. Integrating the data obtained in the course of the long-term study with literature and unpublished results on Italian populations, we show that a combination of linguistic and geographic factors is probably responsible for the presence of the most robust signatures of genetic isolation. Finally, we evaluate the magnitude of the diversity of Italian populations in the European context. The human genetic diversity of our country was found to be greater than observed throughout the continent at short (0-200 km) and intermediate (700-800km) distances, and accounted for most of the highest values of genetic distances observed at all geographic ranges. Interestingly, an important contribution to this pattern comes from the "linguistic islands"( e.g. German speaking groups of Sappada and Luserna from the Eastern Italian Alps), further proof of the importance of considering social and cultural factors when studying human genetic variation.

Genome-wide scan with nearly 700,000 SNPs in two Sardinian sub-populations suggests some regions as candidate targets for positive selection.

This paper explores the genetic structure and signatures of natural selection in different sub-populations from the Island of Sardinia, exploiting information from nearly 700,000 autosomal SNPs genotyped with the Affymetrix Genome-Wide Human SNP 6.0 Array. The genetic structure of the Sardinian population and its position within the context of other Mediterranean and European human groups were investigated in depth by comparing our data with publicly available data sets. Principal components and admixture analyses suggest a clustering of the examined samples in two significantly differentiated sub-populations (Ogliastra and Southern Sardinia), as confirmed by AMOVA (F(ST)=0.011; P<0.001). Differentiation of these sub-populations was still evident when they were pooled together with supplementary Sardinian samples from HGDP and compared with several other European, North-African and Near Eastern populations, confirming the uniqueness of the Sardinian genetic background. Moreover, by applying several statistical approaches aimed at assessing differences at the SNP level, the highest differentiated genomic regions between Ogliastra and Southern Sardinia were thus investigated via an extended haplotype homozygosity (EHH)-based test to point out potential selective sweeps. Using this approach, 40 genomic regions were detected, with significant differences between Ogliastra and Southern Sardinia. These regions were subsequently investigated using a long-range haplotype test, which found significant REHH values for SNPs rs11070188 and rs11070192 in the Ogliastra sub-population. In the light of these results and the overlap of the different computed statistics, the region encompassing these loci can be considered a strong candidate to have undergone selective pressure in Ogliastra.

Strong intra- and inter-continental differentiation revealed by Y chromosome SNPs M269, U106 and U152.

More than 2700 unrelated individuals from Europe, northern Africa and western Asia were analyzed for the marker M269, which defines the Y chromosome haplogroup R1b1b2. A total of 593 subjects belonging to this haplogroup were identified and further analyzed for two SNPs, U106 and U152, which define haplogroups R1b1b2g and R1b1b2h, respectively. These haplogroups showed quite different frequency distribution patterns within Europe, with frequency peaks in northern Europe (R1b1b2g) and northern Italy/France (R1b1b2h).