You are on page 1of 14

tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 119:000 – 000 (2002)

Mitochondrial DNA Affinities at the Atlantic Fringe of Europe
Ana M. Gonza ´ lez,1* Antonio Brehm,2 Jose ´ A. Pe ´ rez,1 Nicole Maca-Meyer,1 Carlos Flores,1 1 and Vicente M. Cabrera
1 2

Departamento de Gene ´ tica, Universidad de La Laguna, 38271 La Laguna, Tenerife, Spain Human Genetics Laboratory, Center of Macaronesian Studies, University of Madeira, 9000 Funchal, Portugal KEY WORDS Western Europe; Portugal; Galicia; haplotypes; clines southern Portugal, and could be attributed to historic Black slave trade in the area and to a probable Saharan Neolithic influence. In fact, U6 haplotypes of specific North African origin have only been detected in the Iberian peninsula northwards from central Portugal. Based on this peculiar distribution and the high diversity ␲ value (0.014 Ϯ 0.001) in this area compared to North Africa (0.006 Ϯ 0.001), we reject the proposal that only historic events such as the Moslem occupation are the main cause of this gene flow, and instead propose a preNeolithic origin for it. Am J Phys Anthropol 119:000 – 000, 2002. © 2002 Wiley-Liss, Inc.

ABSTRACT Mitochondrial DNA analysis of Atlantic European samples has detected significant latitudinal clines for several clusters with Paleolithic (H) and Neolithic (J, U4, U5a1, and U5a1a) coalescence ages in Europe. These gradients may be explained as the result of Neolithic influence on a rather homogeneous Paleolithic background. There is also evidence that some Neolithic clusters reached this border by a continental route ( J, J1, J1a, U5a1, and U5a1a), whereas others (J2) did so through the Mediterranean coast. An important gene flow from Africa was detected in the Atlantic Iberia. Specific sub-Saharan lineages appeared mainly restricted to

Anthropological remains (Kozlowski, 1982) indicate that modern humans arrived in Eastern Europe around 40,000 years ago. They most probably came from the Near East, although a more or less simultaneous arrival from North Africa to the Iberian Peninsula has also been suggested (Ferembach, 1986). From the size and localization of their settlements, it has been deduced that European populations were sparse during the Paleolithic, suffering important constrictions and expansions due to climatic changes (Mellars, 1998). Notable densities were reached in the Mesolithic, mainly in southern areas, although population sizes did not increase greatly until the introduction of farming into Europe from the Near East in the Neolithic (Whittle, 1998). It has been proposed that these Neolithic farmers gradually displaced the indigenous Mesolithic hunters and gatherers (Ammerman and Cavalli-Sforza, 1973). Recently, the molecular variation in mitochondrial DNA (mtDNA) has been used to give a maternal genetic perspective of this European demographic history. Two main approaches have been used to accomplish this goal. Global quantitative analyses, such as those based on population genetic distances (Comas et al., 1997), and spatial autocorrelation (Simoni et al., 2000), have been congruent in showing a certain degree of population homogeneity in Central Europe and a detectable east-west clinal variation for the Mediterranean area. All these data have been interpreted as in agreement
©

with the Neolithic demic displacement hypothesis, although it has been pointed out that mtDNA does not offer the clear picture obtained by the use of autosomal and Y-chromosome markers. On the other hand, the establishment of phylogeographic relationships among mtDNA haplotypes has been a valuable tool to distinguish several major mtDNA lineages in west Eurasian populations (Macaulay et al., 1999). When the divergence times for the European founders of these clusters were calculated, it was found that the bulk of them originated in pre-Neolithic times, whereas those dated in the Neolithic were only around 15%, considerably reducing the Near East demic impact into Europe (Richards et al., 1996; Torroni et al., 1998). Furthermore, preliminary studies on mtDNA diversity in the Iberian Peninsula compared with other European samples
Grant sponsor: DGES; Grant number: PB93-0558; Grant sponsor: Gobierno de Canarias; Grant number: COF1999/019. *Correspondence to: Ana M. Gonza ´ lez, Departamento de Gene ´ tica, Facultad de Biologı ´a, Universidad de La Laguna, 38271 Tenerife, Spain. E-mail: amglez@ull.es Received 27 September 2000; accepted 17 July 2002. DOI 10.1002/ajpa.10168 Published online in Wiley InterScience (www.interscience.wiley. com).

2002 WILEY-LISS, INC.

also included were 50 non-Saami Finns (FI) from Sajantila et al. 1999).. When ambiguity persisted. 1998).tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 2 GONZA ´ LEZ ET AL. Scotland (SC) was represented by 874 sequences from the mainland (Helgason et al. This analysis involved amplifying a fragment embraced by primers L29 and H408 as in Vigilant et al. 78. 299 individuals from Portugal and 43 from Galicia (Northwestern Spain) were analyzed. it should be at this Atlantic fringe where the Paleolithic contribution to modern European populations is expected to be most conspicuous. PCR was amplified as in Pinto et al. 1999). and restricted with Alw44I. Sequencing Total DNA was isolated as described in Rudbeck and Dissing (1998). . In addition. We also screened samples for the 12308G mutation by using a mismatched primer that generates a HinfI site when the 12308G mutation is present (Torroni et al. 52°C. Anderson et al. All the molecular data are congruent with the idea that the cultural and demic impact from the Near East in the Neolithic was very attenuated on the Atlantic edge of Europe. 1996). To determine the relative affinities of the IP population with other Atlantic areas.. 137. several amplified segments were digested and restricted as detailed in the respective references. for the analysis of shared sequence types between regions. from position 15997–16399. These numbers were respectively enlarged with 100. MseI site at np 14766 (Lamminen et al. 1996. 8994 HaeIII.. 1996). MATERIALS AND METHODS Samples In this study. 1996). For this reason.. according to the origin of their most ancient known maternal ancestor.. In this paper. and NlaIII site at np 4577 (Torroni et al. 1993). (1997). were determined and aligned. sorted by sequence motif as H or U* (Torroni et al. 50 from Rousselet and Mangin (1998). 1996). and 100 UK Caucasians (Piercy et al. and South Portugal (SP). and 30 Mau- ritanians as detailed in Rando et al. using the following PCR conditions: 94°C... (1998). 10397 AluI (Chen et al. 1996). 15 sec. and 59 individuals from another study (Pereira et al. 15 sec. To discriminate for ambiguous haplogroup classification. Whenever Portugal was used as a total. 69 from Cornwall (Richards et al. Rocha et al. Blood samples from the Portuguese were taken from voluntary Army donors. and 10871 MnlI (QuintanaMurci et al. comprising samples from the following areas: 85 Berbers from northern Algeria analyzed for mtDNA sequences (Co ˆ rte-Real et al. 1996).. and 110 from Cali et al. (1996). 404 Germans were subdivided as: South Germany (SG). Using both population and molecular phylogenetic approaches. according to the Cambridge Reference Sequence (CRS. were also analyzed for polymorphic site 73 (Wilkinson-Herbots et al. it refers only to the pool of the above-mentioned Atlantic regions. Center Portugal (CP). as follows: North Portugal (NP). 92 unrelated individuals from the same region studied in Salas et al. all individuals.... the following additional sites were tested: AluI site at np 7025 (Torroni et al. and 23 Tuareg from Watson et al. 20 from the mtDNA Concordance database (Miller et al. 32 nonBerber Moroccans. 1996). (1997) and 199 described in Lutz et al. 1996). 25 west-Saharans. and one British unpublished sequence (J. we further searched for molecular clues of the past demographic relationships in this Atlantic area. and directly sequenced for both complementary strands as detailed in Rando et al.. 10394 DdeI. 1998. (Co ˆ rte-Real et al. (2001). 1997).. (2001). Continental Europe representatives included: 180 French (FR). moderate levels of mtDNA gene flow from North Africa to the Iberian Peninsula have also been detected (Co ˆ rteReal et al. personal communication) were included. 84. To enlarge this sample. When the Iberian Peninsula (IP) as a total is analyzed. Finally.. 54 individuals from throughout Portugal were added (Co ˆ rte-Real et al. 33 French sequences. Torroni et al. England (EN) was represented by 92 individuals from Wales. we studied the mtDNA variation in samples from the whole Western border of the Iberian Peninsula to determine the relative affinities and differences among them and other European populations. (1996). diversity levels and the geographic distribution of some autochthonous mtDNA European lineages revealed that a Paleolithic population expansion from Southwestern to Central and Northern Europe probably occurred after the last glaciation (Torroni et al. 1999). and North Germany (NG). Sites tested were: 2349 MboI. 3592 HpaI. Galician samples were collected from voluntary donors resident in the Canary Islands but with all their known maternal ancestors from Galicia. 90 additional North African sequences. The sequences of 403 bp of the first hypervariable segment (HVSI) of the control region of the mtDNA... Each one was assigned to a different geographic area. and 72°C. 4216 NlaIII (Macaulay et al. (1995) and 323 Norwegians (NO) from Helgason et al. 84. 2001).. the following samples were included in the analysis: a pool of 255 already published northwest African (NA) mtDNA sequences. 1996) and for cluster-diagnostic RFLP polymorphisms (Macaulay et al. (1998) were pooled (GA). in particular the Iberian Peninsula. 1996). (1989).. (1998).. comprising 105 German and 33 Danish as detailed in Richards et al... 2000). 1981).. 60 Moroccan Berbers. (1998). Rando et al.. Finally. 1999).. 15 sec. repeated for 35 cycles. 1995. 1996) showed greater diversity levels in Southern than in Northern European populations. To obtain more information for other haplogroup classification. Larruga. including 67 individuals from Hofmann et al.

Within haplotypes classified in superhaplogroup H* by their HVSI sequences. 1992). with coalescence ages in the early Upper Paleolithic..0% belonged to the U* group. the phylogeographic approach that joins phylogenetic clusterings of molecules with their geographic distribution. P Ͻ 0. as defined in Richards et al. and a probable European origin (Richards et al.001) also has its greatest frequency in Britain. is important in northwest Africa (26. Cluster U5. and matches within clusters. represented by haplotypes classified in L and M1 clusters.74. only the variation between positions 069 –365 was used. distances based only on the relative frequency of different molecular lineages between populations miss the sequence heterogeneity existing within lineages. and coalescence times (Forster et al.05). On the other hand. sub-Saharan African influence. and northern Germany.. P Ͻ 0.7%). Averaged distances between sequences do not take into account the fact that nucleotide differences between sequences belonging to different molecular lineages could be the same. geographical patterns of haplogroup variation were tested by spatial correlation. For example. On the other hand. and the linearized FST distance based on allele frequencies was used as implemented in the ARLEQUIN program. Germany.. However. In order to include other published samples in the comparisons. The neighbor-joining (NJ) tree (Saitou and Nei.7% in north Portugal. 1997). In this case. 4. We chose this distance for our population comparisons. different J derivatives have different geographic distributions.16. Due to the relatively low frequency of matches within haplogroups. P Ͻ 0. of North African origin (Rando et al. Pairwise population comparisons were estimated as distance values. and north. with representatives in all southern populations. in comparisons with samples where this sorting was not possible. we also grouped them as H*. P Ͻ 0. The significance tests of the variance components were obtained by comparison with a distribution of 1. On the other hand. diversity. Central (north Germany.03. 1. and south Portugal).83. 1981) is the most abundant haplotype in all samples. and 0. haplogroup J and the subhaplogroups T1. but keeping their geographic origins.01). and U6 in northern areas (␹2 ϭ 5. As expected.01) and J1b1 (␹2 ϭ 13.7% to the pre-HV cluster. Iberian (Galicia. 2.81.. Although not statistically testable due to their low frequencies. is only present in the northern population of Scotland. mtDNA was considered as one locus with as many alleles as different haplotypes detected. P Ͻ 0. version 1. subhaplogroup U6. For nonrecombinant systems such as mtDNA variation. and the Iberian Peninsula. and northwest Africa.10. The haplotype with the reference sequence (CRS.01). and France). although values range from 11. but has no significant level of quantification for the present relatedness between close populations. U5a1a. but give no clue about past interactions. but with inevitable shortcomings. Phylogeographic analysis All sequences used in this study were sorted into molecular clusters (haplogroups). 2000) and distributions in Europe to refine the global affinities obtained at population level. south Germany. reaches its highest frequencies for its ancestral motives in Britain (␹2 ϭ 11.. the populations studied were grouped into the following geographic areas: Northern (Norway and Finland).000 values calculated by randomization.7% in northwest Africa to 21. P Ͻ 0. being detected only in northwest Iberian Peninsula. center.9% to superhaplogroup R*. 1987) was calculated using the DnaSP program. whereas J2. among and between areas.001) when compared to other continental areas such as Scandinavia.1 (Schneider et al. using haplogroup frequency and latitude (represented by that of the main town in each sampled area) as variables. Several mean distances were used to deal with this issue.. The haplogroup frequencies and sample sizes for the populations analyzed are given in Table 1. 1999). France. The differential geographic distributions of these sub-Saharan African and northwest African haplogroups in the Iberian Peninsula are statistically significant: L and M1 clusters are more abundant in south Portugal (␹2 ϭ 9.. 2000). has a local presence in Europe. version 3. In a similar vein.22.. Western (Scotland and England). and perhaps U5a1 have been considered of Neolithic influence (Richards et al.0 (Rozas and Rozas. Distances based on the frequency of identical sequences between populations. The basal subhaplogroup J (␹2 ϭ 11. the highest frequencies for U5a1 are in the north (␹2 ϭ 8. although not appropriate to detect possible ancient connections. In addition. using the computer program ARLEQUIN. 1996) aims to cover this gap at a global scale.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 MITOCHONDRIAL DNA IN ATLANTIC EUROPE 3 Population comparisons AQ: 1 For the AMOVA analysis (Excoffier et al.001) are more abundant in northern areas including Scandinavia.1%) but negligible in Europe. P Ͻ 0. with the exception of south Portugal (11. 2000). Nucleotide diversity measured as ␲ values (Nei. 1998).6% to the HV* cluster according to the RFLP analysis. Anderson et al. U3. (2000). It is known that genetic distances between populations give us a measure of their present affinities. 1987) was used to depict graphical relationships among populations. were used in conjunction with their previously calculated ages (Richards et al. RESULTS Haplotype classification and distribution The Appendix lists our 342 sampled sequences according to their haplogroup status. T2. Relative frequencies of haplogroups. similar trends are also detectable for some subclusters be- T1 . seem the most adequate to fairly well detect relatively recent gene flow between populations. Britain. J1a (␹2 ϭ 8.

023 0.005 0.036 0.157 0. SG.005 0.003 0.042 0.003 0.038 0. North Africa.020 0.009 0.056 0.007 0.011 0. Thus.004 0.014 0.422 0.005 0.056 SP 0.033 CP 0.020 0.020 0.056 0.006 0.006 0.025 0.009 0.019 0.007 0.001 0.015 0.023 0.012 0. The bulk of the total variance (99.056 0. SP.0 0.0 266.008 0. NG.240 0.005 0.007 0.036 0.013 0.011 0.0 0.035 0.041 0.033 0.015 0.268 0.005 0.007 0.012 0. Norway.007 0.020 0.079 0. Paleolithic and Neolithic linages are heterogeneously distributed throughout the studied area.005 0.001 0.315 0.034 0.010 0.006 0.008 0.340 0.006 0.118 0.008 0.025 0.022 0.005 NA 0. Northwest Africa is significantly different from all other populations. on the contrary.014 0. but congruently.005 0.015 0.057 0.364 0.215 0.005 0.061 0.009 0. CP.022 184.057 0.019 0.003 0.239 0.020 NO 0. France.005 0.004 0.007 0.015 0.014 0.020 0. 0.019 0.046 0.004 0.007 0.100 0.023 0.031 0. SC.021 0.027 0. EN.011 0.020 0.022 0.006 0.023 0. which accounts for 87% of all matches used in the calculation of FST distances.019 0. Figure 1a shows the relationships among populations.005 0.019 0.005 0. England.015 0.183 0.022 NP 0.065 0.014 EN 0.265 0.019 0.012 0.019 0.007 135. masking the effect of other less frequent haplotypes.004 0. Galicia.006 0. However.015 0.049 0. TABLE 1. NO. south Portugal.097 349.0 0.004 0.01 level. In spite of this.1% attributable to differences among populations within areas.012 0.006 0.040 0.012 0.010 0.011 0.007 0.023 0.011 0.117 0. longing to haplogroup T. As expected.019 0.004 0.010 0.014 0.019 0.0 323.154 0.007 0.003 0.011 0. Distances calculated without the CRS (Table 2) significantly increased the heterogeneity between populations. whereas T5 is more prevalent in the south.009 0.005 0.014 0. curiously. and 0.0 1 FI.005 0.021 0.052 0.020 0. Finland appears as its most related pair.006 0. and showed somewhat different relationships among them in the generated NJ tree (Fig.023 0.080 0.043 0.012 0.034 FR 0.027 0.019 0.041 0.201 0.005 0.015 0.015 0.005 0.020 0.019 0.0 0. FR.032 0.052 0.005 0. Finland.045 0.164 0.007 0.005 0.009 GA 0. NA.014 SG 0.011 0.014 0. the geographic partition among areas is statistically signif- icant at the 0.022 0.016 0.023 0.0 0.002 0.019 0.040 0.191 0.004 0.030 0.037 0.015 0.016 0.001 874. north Germany.006 0.003 0.7%) is due to differences among individuals within populations.201 0.009 0.209 0.007 0.361 0.007 0.023 0.043 0.014 0. T4 shows greater frequencies in the north.026 0.015 0.010 0.023 0.008 0.004 50.0 262.015 0.003 0.007 0.073 0.011 0.034 0.043 0. Frequency of haplogroups and the cambridge reference sequence (crs) in the different populations analyzed1 Population Haplogroup CRS H*-CRS V U1 U1a U2 U3 U4 U5 U5a U5a1 U5a1a U5b U5b1 U6 U7 K J J1 J1a J1b J1b1 J2 T T1 T2 T3 T4 T5 B N1a N1b I W X C D M1 N/M/L3 L3b L3d L3e L1a L1b L1c L2 Sample size FI 0.027 0.019 0.043 0.006 0. now their closest relatives are the Iberian Peninsula F1 T2 .005 0.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 4 GONZA ´ LEZ ET AL.003 0.149 0.015 0.038 0. the cause for this peculiarity is that both have the lowest frequencies for the CRS sequence.011 0.050 0.005 0.385 0. 1b).010 0.003 0.022 0.006 0.0 140.006 0.005 0.2% to heterogeneity among areas.027 0.009 0.049 0.076 0.019 0.011 0.006 0.023 0.100 0.043 0.008 0.004 0.015 0.043 0.065 0.043 0.263 0. GA.028 0. northwest Africa continues to be the most divergent population.120 0.004 0.005 0.006 0. Scotland.019 0.011 0.019 0.015 0.009 0.019 162.016 0.011 0.005 0.011 0.003 0.002 0. based on FST values listed in Table 2.006 0.198 0.031 0.056 0.003 0.006 0.008 0.009 0.066 0.060 0.020 0.087 0.025 SC 0.120 0.005 0.023 0.006 0.004 0. NP.040 0.025 0. Population affinities AQ: 2 The AMOVA results showed that there is a very low level of mtDNA genetic structure in the geographic area analyzed.023 0.012 0.007 0.007 0.015 0.028 0.053 0.019 0.009 0.023 0.0 0.041 196.080 0.052 0.007 0.005 0. center Portugal.014 213.031 NG 0.012 0.038 0.0 0.003 0. north Portugal. south Germany.056 0.006 0.040 0.009 0. These results show that the western fringe of Europe is not a homogeneous mitochondrial landscape.217 0.009 0.009 0.

0029 0. *** P Ͻ 0.0017** 0.0016 0. NG.0024 0. Norway.0083*** 0. Galicia.0000 0.0000 0. North Portugal. globally.0018** 0.0012 0. an inspection of the matches between them shows that this affinity is due to the high frequencies in both for common haplotypes.0000 0.0018 0.0003 0.0077*** 0. Pairwise linearized FST for all haplotypes (above diagonal) and excluding Cambridge Reference Sequence (below diagonal)1 0.0029* 0. North Germany.0017* 0.0015* 0. Scotland.0000 0.0014* 0. U5b.0043 0. South Germany. GA.0024 0. more related to the Central European samples than to England and Scotland.0007 0.0024 0.0057*** GA 0. SC.0016 0. we analyzed matches only for the large geographic areas considered in the AMOVA analysis.0014* 0.01. FI.0009 0.0020* 0.0010 0.0028* 0.0063*** 0.0057* 0.0048*** 0.0016 0.0025 0.0032*** 0. In fact.0000 0. the Iberian Peninsula participates in 0.0012* 0. NP.0033** 0.0029 0.0053*** SG 0.0021* 0.0049* 0.0026* 0.0056** 0.3 Ϯ 3.0029** 0.0071* 0.0020* 0. CP.0056 0. NA. EN.7 Ϯ 1. Only 2. FR.0002 0. center Portugal.0003 0. In order to keep adequate sample sizes within haplogroups.0061* 0.0064* 0. T3 Within-haplogroup population affinities Sequence matches distribution within haplogroups. England.0022** 0.0054*** NP 0.0007 0.0019 0. SG.0024 0.0013 0.0022 0.5%.0012 0.0088*** 0.001.0051*** 0.0012* 0.0018* 0.0018 0.0%). (a) using all haplotypes and (b) excluding CRS.0032 0. T.0061* 0.0037*** 0.0 Ϯ 2. that together summed up an additional 10.0044*** 0.0010 0. Neighbor-joining trees constructed from FST distance matrix.0023 0.0020* 0.0011 0. However.0055*** CP 0.0009 0.0037*** 0.0020 0. T1.0086*** 0. center Portugal and north Germany cluster together in spite of their large geographic distance.0003 0. NO SC FI FI NO SC EN NG SG FR GA NP CP SP NA 0.0008 Fig.0000 0.0048*** FR 0. * P Ͻ 0.0003 0.0006 0.05. NO.0000 0.0014* 0. Table 3 indicates that matches due to the CRS are the overwhelming majority of matches between areas (87.0014 0.0042 0.0032** 0.0016* 0.0067** 0. South Portugal.0046* 0.0007 0. Finland. North Africa.0031** 0.0012* 0. Again.0015* 0.0007* 0.0083*** 0.0005 0.0021** 0. 1 samples. France.0032*** 0. SP.0000 0.0019 0.0008 0.0003 0.0071*** Abbreviations as in Table 1. the latter are.0019 0. K.0042* 0.0040* 0. all areas share matches due to the basic and more frequent haplotypes for the following clusters: H*. 1.0040* 0.0000 0.0030** 0.0044*** 0.0023 0.0030* 0.0022** 0.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 MITOCHONDRIAL DNA IN ATLANTIC EUROPE 0.0002 0.0010 0.0054** 0.0084*** EN 0.6% of the matches are specific between areas. and not to specific matches not shared with other areas.0073*** NG 0. ** P Ͻ 0. J.0000 0.0008* 0.0011 0.0005 0.0000 0.0055*** SP . and V. mainly in H* and K.0003 0. Furthermore.0080*** NA 5 TABLE 2. With rare exceptions such as the central motives of U5 and J1a.0010 0.0094*** 0.0005 0.

3 7.0 8.0 159.9 2.7 14.5 1.9 0. North Africa.0 6.1 11.0 175.0 8.8 1.1 4. Although gradients for clusters belonging to Js and U5 run T4-F2 .0 90.0 6.7 19.0 71.8 2. U6 haplotypes were not detected in southern Portugal (Table 1).9 1.0 19.7 72.7 15.2 7. I.4 0. However.1 4.3 NE ϫ IP 4.4 2.4 109.5 81.4 2. Latitudinal correlations.0 1. U5.2 3.1 37.0 4.2 2.2 0.0 61.1 9.8 3.4 2.8 BR ϫ NA 1.0 5.3 1.6 2.0 298.0 23.0 2.9 110.0 89. the Iberian Peninsula shares more matches with Britain than with the Continent.2 3.3 4. common haplotypes (present in all populations).996.3 CE ϫ IP 3.0 38. T.4 20.0 11. IP. On the one hand.9 3.4 0.507.4 46.2 0.0 0.8 1.0 3.1 74.0 25.2 NE.7 0. BR.4 3. U3.0 1.0 73. J1a.4 4.0 4.1 9.3 0. Iberian Peninsula.0 13.0 226.0 12.0 2.8 3.0 5.6 CRS Total % Common H*-CRS V U5b K J T T1 Total % Rare H*-CRS V U1a U2 U3 U4 U5 U5a U5a1 U5a1a U5b U6 K J J1 J1a J1b1 J2 T T1 T2 T3 T4 T5 N1b I W X M1 N/M/L3 L3b L3e L1b L2 Total % 1 3. Furthermore. J2.4 0.0 69.0 12. The highest frequencies for H* are found in Galicia. J1b1.0 1.236. J1b1.0 2.1 3.1 7.0 81.7 1.0 28.3 101.0 0.8 1. On the other hand.0 275. and V.4 2.0 3. and to the North for I.5 0.9 7.0 52.0 35.9 0.9 3.4 2.0 5.3 0.3 IP ϫ NA 2.0 2.7 BR ϫ CE 2.0 7.8 0.6 2.5 1.0 12.0 3.4 CE ϫ NA 1.5 1.0 100.0 90.0 10. whereas the J2 gradient is from south Portugal to the north.8 0.0 13.0 60.1 9.4 32.2 2. and the recent subclusters U5a1 and U5a1a.0 42.0 11.0 290.1 47. in particular for H*.0 0.4 3.2 37.0 9.4 109.0 70. TABLE 3. J. Significant latitudinal correlations were found for several haplogroups (Table 4 and Fig.5 1.0 307.3 1.0 0.0 81. the extensions of the clines are different.3 2.0 303. all the Js excluding J1b and J2.0 384.3 0.7 0. whereas matches in clusters J.0 11.4 0.1 9.0 16.0 0.8 2.4 5.9 36. U5a. there are matches within the specific northwest African sub-haplogrup U6 (Rando et al.0 2.1 2.5 NE ϫ NA 2.8 27. T2.0 6.0 55.9 1. U5a1a.3 3.0 61.6 1. NA.8 1.0 49.0 97. T1.0 29.4 0. 2).3 0.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 6 5 GONZA ´ LEZ ET AL.0 87.5 0.2 0.3 0.0 45.4 0. Quantitatively.2 0.0 78.8 18.7 0.0 0.8 16. Britain seems more related to Central Europe for H*. In fact.7 1.1 5.0 88.5 1.0 7.3 101. decreasing to the north but also to the south. At least two different African influences can be detected on the basis of the geographic areas and haplogroups implied.2 42.0 3.0 2.4 0.9 0.4 56. Britain.1 4.7 0.0 269.5 1.7 2.1 0.4 13.1 2.234.4 9.5 21.6 150.7 0.4 1.0 46. and rare haplotypes (shared by two or more populations).0 371.0 90.2 0.0 279.800.2 1.3 4. On the other.9 0.5 8. 1998) that occur only in the northern areas. and V approximate the Iberian Peninsula to the Continent.1 0.0 7.9 74.0 1.062.1 3.3 0. The match connection between northwest Africa and the Atlantic Iberian Peninsula deserves special attention.4 7.6 0.9 76.3 3.137. U4.4 0.0 85.0 84.4 4.6 1.0 10.0 84.0 12.9 2.0 34.626. Central Europe.0 11.3 121.0 1.9 2. affecting mainly southern Portugal.0 20.0 5. and X.936.1 0. CE.0 86.0 42..7 2.3 0.9 2.8 3.8 0.0 5.4 33.0 21.4 BR ϫ IP 3.1 0.0 27.2 68.1 3. there are matches due to haplotype sharing for sub-Saharan Africa lineages L and M1.0 1.9 1. The H* and J2 haplogroups show strong negative correlations.3 0.0 18. Frequencies of matches (ϫ 10 ) between areas for Cambridge Reference Sequence (CRS).7 122. significant positive correlations were found for subcluster U4.3 1.0 524.9 6.0 4.0 85.0 0. North Europe.3 3.4 0. and percentage (%) of matches between areas for each class1 Populations NE ϫ BR NE ϫ CE 3.0 1.9. all the matches between northwest Africa and Europe.3 1.9 4.8 34.3 198.9 0.8 98.3 0.0 2. and X.517.0 164.3 2.9 0.7 2.

.. have only been detected in Northern Europe. differing significantly only from northwest Africa. On the contrary. or U7 and J1b in the south. which is responsible for the majority of matches between populations. This shows TABLE 4.64 P Ͻ0.e. Correlation coefficients (R) and two-tailed significance values (P) obtained for widespread mtDNA clusters Cluster H* (without CRS) J2 Js (without J2 and J1b) U5a1 plus U5a1a U4 1 2 2 N1 8 11 11 11 11 R Ϫ0.001 0. Since the earliest human settlements in Norway and Finland were dated around 13. Finally. 1998). the most ancient subhaplogroup in Europe (Richards et al. 12308G and 16270T. Gray represents frequency of rest of haplogroups.97 Ϫ0.036 Number of pairs. Pie charts represent populations from Table 1. Black indicates frequency of haplogroup H* without CRS.. The significant latitudinal gradients found here for several mtDNA clusters with very different coalescence and expansion ages in Europe (Richards et al. and northwest Africa. but that more recent movements had a different impact on different areas. their presence in this area is most probably the result of an upward migration from more southern places. White includes frequencies of Js (excluding J2 and J1b) and U5a1 plus U5a1a.. 1999). DISCUSSION Global analysis shows European populations rather homogeneous at the mtDNA level. Although of Neolithic introduction in Europe.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 MITOCHONDRIAL DNA IN ATLANTIC EUROPE 7 parallel. there is an evident difference in their behavior. that all European populations shared an ancestral common background. the northern pattern for basal J can be explained in the same way. 1998). to which coalescence has been dated at recent Neolithic times (Macaulay et al. 1996.000 years ago (Mellars.. 2. some less frequent Caucasian subhaplogroups were only detected in localized areas such as U1a and U5b1 in the north.005 0. Cambridge Reference Sequence.78 0. The basic haplotypes. Frequency distribution for some mtDNA haplogroups that showed significant latitudinal clines.001 0. 2000) were also detected with synthetic maps Fig. Britain. significantly increases the spatial differentiation.000 –9.85 0.001 0. Reanalysis without the CRS. i. The haplogroup distribution of matches shows that the only young subcluster with a widespread representation is T1.81 0. there was no overall sharing for U5 haplotypes.

the main peculiarity of south Portugal is that this area harbors the highest frequency of X in Europe (␹2 ϭ 6.95%. 2000). a significant influence (at least 10%) of North Africans in the Iberian gene pool has also been admitted (Rocha et al. considered a European Paleolithic isolate (Bertranpetit et al. It seems. At the other end. as haplotypes belonging to the U6 lineage have not been found in other European samples (Richards et al.. sub-Saharan African L and northwest African U6 haplotypes should be uniformly distributed in the Iberian Peninsula. 1998. and 72 Equatorial Guineans (Pinto et al. other undetected haplogroups also participated in this pre-Neolithic northwest Africa gene flow... using the relative frequency of northwest African Y-chromosome-specific markers in Iberian samples (Flores et al. and for this reason. 2000). Rosser et al.. the negative gradients for J2 and the exclusive southern presence of J1b may be explained as signatures of the Mediterranean Neolithic wave into the Atlantic border. Flores et al..... 1999). obtained from classical markers (Cavalli-Sforza et al. The presence of U6 only in the northern Iberian Peninsula could represent the remnants of a pre-Neolithic expansion from northwest Africa to the Iberian Peninsula. P Ͻ 0. 1994).tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 8 GONZA ´ LEZ ET AL. for the first time. and possibly it was characteristic of all Paleolithic European settlers. however. This distribution cannot be totally explained by a historic genetic influence from the Moslem occupation (Pereira et al.. P Ͻ 0. With respect to northwest Africa. with a well-documented import in southern Portugal (Godinho. although both prehistoric and historical influences likely contributed to the sub-Saharan African haplotype pool present in the Iberian Peninsula. the haplotype composition of northwest Africa had to be similar to that of the present. 2000b). 2000a. These results suggest that. 1999. 1997). although matches between sub-Saharan and North African samples are only 0. The percentage obtained (0. Finally. we compared the proportion of sub-Saharan Africa haplotype matches between the Iberian Peninsula and northwest Africa (0. extend. 1995). 1994) and for Y chromosome haplogroups (Lucotte and Loirat. Rando et al. A high frequency of H* is the main characteristic of the Basques. The geographic gradient detected for this main haplogroup may be the result of a differential impact of posterior migrations on a common Paleolithic substrate in Europe.. 1999). 32 Yoruba (Watson et al. Rocha et al.0135 Ϯ 0. could have comparatively more influence in northern regions.. the former seems to be more important.35%) is roughly half of the former. and the younger clusters of U5a1 and U5a1a. it seems plausible that they could also have reached the Iberian Peninsula across the Gibraltar Strait. As southwestern areas were comparatively more populated than the north (Mellars.0055 Ϯ 0. 1997). 1983). .. 2000a). meaning that the L haplotypes de- tected in southern Portugal could be the result of African Neolithic influence in this region. in contrast with northern Iberia. are better explained as the result of the continental penetration of the Neolithic. J. 1998). To test this possibility.5%) from northwest Africa to the Iberian Peninsula estimated in a recent study of variation in the autosomic CD4 locus (Flores et al. personal communication).000 years ago. 2000).. is well-documented (Mart´ ıOliver. If we admit that this expansion brought. it has been proposed that the Neolithic in Europe had both a continental and a Mediterranean coastal introduction (Renfrew. 2000). Our results are in agreement with the gene flow (19. 1996. It has been pointed out that one of the most important demic influences on northwest African populations from the Sahara occurred around 7.. In comparisons within Europe. The sample includes 45 Bubis from Bioko and 49 individuals from Sa ˆ o Tome ´ (Mateu et al.0007). and in addition. and with the evidence of northwest African male input in Iberia calculated at around 20%. the significant positive correlations found for the rest of the Js. During that time. However. the majority of them (97%) are also shared with northwest Africa. For the same reason. 1987).01). and clarify the preliminary clues of an important mtDNA contribution from northwest Africa into the Iberian Peninsula (Co ˆ rte-Real et al.. north Portugal has the highest frequency for T1 (␹2 ϭ 16. where H* has lower frequencies. Most probably. since a strong Neolithic influence on Portugal. On the other hand. This might explain why Portugal and northwest Africa do not follow the H* gradient. Neolithic or more recent waves from the East. could also be a plausible alternative to explain the presence of these African haplotypes in this region (Pereira et al.90. Larruga. Furthermore. the recent history of the Black slave trade carried out by the Portuguese (mainly in the 15th and 16th centuries). Some molecular data are congruent with this picture. which have been interpreted as a result of the Neolithic expansion in Europe. The diversity value for U6 is significantly higher in the Iberian Peninsula (0. our results clearly reinforce.0014) than in northwest Africa (0. with respect to the sub-Saharan Africa lineages. Portugal deserves some particular comments. On the basis of archaeological evidence. and is the only region in the Iberian Peninsula where J1 has been detected. a finding that cannot be explained by a recent and limited gene flow.001)..75%) with those of the Iberian Peninsula and a sample of sub-Saharan Africans from the Gulf of Guinea.40. On the basis of the L1b frequencies detected in Spanish and Portuguese samples (2–3%) and those found in western Africa (10 –30%). 1996.. 1998). with the expansion of the Neolithic culture that flourished at that time in the Sahara (Dutour et al. sub-Saharan African haplotypes into the Maghreb. that these African waves did not reach farther than Galicia. the geographically localized distribution of matches and haplotypes of sub-Saharan African and northwest African origin in the Iberian Peninsula is noteworthy.

In spite of the simplicity of the method. different estimates give congruent results.3%) as the result of the northwest African gene flow at that time. as the pre-Neolithic frequency in that area. 13% would be left for the putative Saharan Neolithic gene flow. and the present frequency in the whole Iberian Peninsula (2. The value obtained (14%) could be as high as 35% using the data of Co ˆ rte-Real et al. However. and discarding possible genetic drift effects. our own data allow us to make minimal estimates of the maternal African pre-Neolithic. Tuareg.M. the Saharan Neolithic gene flow can be estimated as 13%. This value could rise to 23% when only south Portugal is taken into account. and Mauritanians) as the frequency of the African Neolithic. 1983).C. For the former. as historically documented (Godinho. and that of the Iberian Peninsula (6. Hypervariable segment I (HVSI) haplotypes and restriction fragment length polymorphisms (RFLPs) in Portuguese and Galician samples1 RFLPs 2 3 4 9 j 3 5 9 2 h 4 2 1 6 q 4 5 7 7 q 7 0 2 5 a 8 9 9 4 e 1 0 3 9 4 c 1 0 3 9 7 a 1 0 8 7 3 y 11 2 3 0 8 g 4 7 6 6 u GA 7 3 HVSI type H* CRS CRS CRS CRS CRS 093 114 126 129 141 147 148 153 172 179 189 189 189 221 222 240 261 262 264 265C 274 278 278 289 291 292 293 304 311 311 320A w Samples NP CP SP 1 24 2 1 1 1 1 1 1 1 1 1 1 1 1 1 Ϫ ϩ ϩ ϩ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ ϩ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ ϩ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ ϩ ϩ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ ϩ 4 14 4 19 ϩ Ϫ Ϫ Ϫ 2 1 2 1 2 1 1 1 1 1 1 1 1 Ϫ 1 1 1 1 1 1 2 1 1 2 1 4 2 1 1 1 2 ϩ Ϫ .tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 MITOCHONDRIAL DNA IN ATLANTIC EUROPE 9 In a similar way. This study was supported by grants of the DGES (PB93-0558) and Gobierno de Canarias (COF1999/ 019) to V. we consider only the mean value of the U6 frequency in northern African populations. if we admit a recent 10% of slave trade input into this region. proposed on the basis of archaeological and anthropological affinities (Barandiara ´ n. Neolithic. that in turn give molecular support to the earlier hypothesis of prehistoric northwest African influences in the Iberian Peninsula.8%) as the result of the putative Neolithic maternal gene flow. and Mauritanians (16%). or 27% with our north Portugal sample. Bandelt for his help and comments. ACKNOWLEDGMENTS We thank H. Saharans. (1996). The authors are grateful to the Por- APPENDIX. In the same vein. and/or recent slave trade input into Iberia. 1998). taking the actual frequencies for the sub-Saharan African haplogroups (51%) in southern northwest African samples (Tuareg. excluding Saharans.

RFLPs 2 3 4 9 j 3 5 9 2 h 4 2 1 6 q 4 5 7 7 q 7 0 2 5 a 8 9 9 4 e 1 0 3 9 4 c 1 0 3 9 7 a 1 0 8 7 3 y 11 2 3 0 8 g Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ ϩ Ϫ Ϫ ϩ Ϫ 4 7 6 6 u GA 7 3 HVSI type 320 335 344 355 362 093 189 093 263 093 270 093 319 111 311 124 220 124 362 129 242 129 362 188A 189 189 262 189 356 192 262 213 304 220T 311 221 291 235 291 269 270 278 311 278 311 292 318 293 311 311 319 093 189 293 093 189 311 114 311 362 126 264 362 126 361 362 129 344 362 172 192 311 172 233 362 180 219 221 189 356 362 213 254 304 217 218 311 235 291 293 126 291A 355 362 V (298) 298 153 298 189 298 239 298 240 298 256 298 258 298 298 311 298 319 240 298 311 U2 (051 129C) 051 129C 189 362 U3 (343) 343 390 343 356 390 w Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Samples NP 1 1 CP 2 SP 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Ϫ 1 2 2 Ϫ Ϫ Ϫ 2 1 1 1 1 1 1 1 1 1 1 1 Ϫ ϩ ϩ 1 1 1 1 1 1 1 1 22 1 1 1 1 1 1 1 1 1 1 4 2 2 4 1 1 2 2 1 1 2 0 1 1 1 0 1 4 1 1 1 1 2 1 3 39 47 1 66 2 U4 (356) 356 179 356 0 0 .tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 10 APPENDIX. (Continued) GONZA ´ LEZ ET AL.

tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 MITOCHONDRIAL DNA IN ATLANTIC EUROPE APPENDIX. (Continued) RFLPs 2 3 4 9 j 3 5 9 2 h 4 2 1 6 q 4 5 7 7 q 7 0 2 5 a 8 9 9 4 e 1 0 3 9 4 c 1 0 3 9 7 a 1 0 8 7 3 y 11 2 3 0 8 g 4 7 6 6 u GA 11 7 3 HVSI type U5a (192 270) 192 270 304 311 w Samples NP 1 1 CP SP 0 U5a1 (192 256 270) 192 256 270 192 256 270 362 0 U5a1a (256 270 399) 256 270 188 256 270 399 0 U5b (189 270) 189 270 189 270 189 192 270 189 270 390 093 189 192 270 093 189 192 270 311 U6 (172 219) 051 172 219 311 172 219 235 278 355 172 189 219 239 278 362 K (224 311) 224 311 093 224 311 192 224 311 224 234 311 224 235 311 224 304 311 224 311 320 093 110 224 218 224 311 222 224 311 093 156 224 ϩ 0 0 1 1 2 1 0 0 1 1 0 1 1 1 2 1 1 0 1 1 1 2 1 2 1 1 3 1 1 1 1 2 3 4 1 0 2 3 2 1 2 311 320 360 240C 311 1 1 1 1 4 4 1 1 1 2 1 1 0 1 1 1 5 0 0 1 1 1 0 1 1 1 3 1 2 1 1 2 1 1 1 1 6 1 1 1 1 3 3 1 1 7 3 8 1 1 1 1 J (069 126) 069 126 069 126 150 069 126 241 069 126 256 069 126 311 069 126 324 366 390 069 126 163 266 311 J1b (069 126 145 222 261) 069 126 145 222 256 261 278 0 J1b1 (069 126 145 172 222 261) 069 126 145 172 222 261 0 J2 (069 126 193) 069 126 193 278 069 126 193 319 360 T (126 294) 126 294 126 256 294 296 093 126 189 294 296 0 T1 (126 163 186 189 294) 037 126 163 186 189 126 163 186 189 294 126 163 171 186 189 294 126 163 186 189 261 294 126 163 186 189 249 294 311 1 .

(Continued) GONZA ´ LEZ ET AL. RFLPs 2 3 4 9 j 3 5 9 2 h 4 2 1 6 q 4 5 7 7 q 7 0 2 5 a 8 9 9 4 e 1 0 3 9 4 c 1 0 3 9 7 a 1 0 8 7 3 y 11 2 3 0 8 g 4 7 6 6 u GA 1 T2 (126 294 304) 126 294 304 126 172 294 304 126 294 296 304 051 126 294 296 304 093 126 294 296 304 126 147 294 297 304 126 294 296 304 344 126 294 296 304 365 093 126 271 294 296 304 T3 (126 292 294) 126 292 294 296 324 126 171 292 294 296 324 0 T5 (126 153 294) 126 153 294 126 153 274 293 294 296 N1a (147A/G 172 223 248 355) 086 147A 172 223 248 320 355 N1b (145 176G 223) 145 176G 223 390 0 I (129 223 391) 129 223 391 129 172 223 311 391 129 172 223 293 311 391 0 W (223 292) 223 292 223 292 311 223 292 311 223 292 318 223 292 320 223 292 362 192 223 292 325 129 209 223 292 295 311 343T X (189 223 278) 189 223 278 189 223 278 189 209 223 278 M1 (129 189 223 249 311) 129 189 223 249 311 129 186 189 218 223 249 311 0 L3 (223) 223 185 223 111 169 193 195 223 243 261 311 L3b (124 223 278 362) 223 278 362 124 189 223 234 278 0 L3e (172 223 320) 172 189 223 320 172 189 223 320 172 189 223 248 320 L1b (126 187 189 223 264 270 278 311) 187 189 223 264 270 278 311 ϩ 0 0 0 ϩ Ϫ Ϫ ϩ 1 1 0 0 1 1 0 0 Ϫ ϩ Ϫ ϩ 1 1 0 0 1 1 0 1 1 1 2 1 4 0 0 0 0 1 1 0 7 3 HVSI type 126 163 186 189 294 304 311 w Samples NP 1 4 1 1 2 1 1 1 1 1 1 7 2 1 1 1 1 3 CP 1 SP 4 2 1 1 0 1 0 1 Ϫ Ϫ ϩ 1 1 1 1 1 1 1 1 0 0 1 5 1 1 2 1 1 2 1 1 2 1 1 1 3 1 4 1 2 1 4 4 Ϫ ϩ ϩ Ϫ Ϫ Ϫ ϩ ϩ Ϫ ϩ Ϫ Ϫ ϩ Ϫ .tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 12 APPENDIX.

Hikey E. Prehistoria de la Pen´ ınsula Ibe ´ rica. Calafell F. Bertranpetit J. Staden R. Sala J. 1986. editor. p 245–256. In: Aumassip G. Mateu E. Torroni A. Gonza ´ lez AM. Am J Hum Genet 68:723–737. Heddouche A. Romano V. Bedeaux R. Coulson AR.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 MITOCHONDRIAL DNA IN ATLANTIC EUROPE APPENDIX. In: Arnaiz-Villena A. Douin. Issad MS. p 197–315. Bankier AT. Dupuy C. Herna ´ ndez M. del Rinco ´ n MA. Barandiara ´ n I. Schreier PH. Sequence and organization of the human mitochondrial genome. 2000a. p 33–50. Os descobrimentos e a economia mundial. (Continued) RFLPs 2 3 4 9 j 3 5 9 2 h 4 2 1 6 q 4 5 7 7 q 7 0 2 5 a 8 9 9 4 e 1 0 3 9 4 c 1 0 3 9 7 a 1 0 8 7 3 y 11 2 3 0 8 g 4 7 6 6 u GA 13 7 3 HVSI type 104 104 126 126 126 187 187 187 187 187 189 189 189 189 189 223 215 223 223 223 270 223 264 264 264 278 270 270 270 270 289 278 278 278 278 293 289 293 293 293 311 293 311 311 311 311 357 362 362 w Samples NP CP SP 1 1 1 1 2 0 0 1 1 0 3 1 1 1 2 1 1 ϩ ϩ ϩ Ϫ Ϫ Ϫ Ϫ L1c (129 187 189 223 278 294 311 360) 129 187 189 223 265C 278 292 294 311 360 0 L2 (223 278 390) 189 223 278 390 223 278 294 309 390 148 223 278 294 355 093 223 260 278 294 093 223 278 294 309 189 192 223 278 294 189 223 234 249 278 086 129 148 189 223 Total samples 1 390 309 311 309 294 278 390 320 316 295 300 390 390 390 354 390 0 43 1 83 1 1 1 1 7 137 1 78 Position numbers as in Anderson et al. Torroni A. Mertens S. tuguese Army Chief of Staff for allowing the collection of blood samples covering the whole territory LITERATURE CITED Ammerman AJ. Aholt S. 1997. London: Duckworth. Jaksch M. Sykes BC. Lisbon: Presenc ¸ a Editors. 1973. Hofmann S. northwest Africa. Cali F. 2nd ed. El Calcol´ ıtico y la Edad del Bronce. The history and geography of human genes. Barcelona: Ariel. A population model for the diffusion of early farming in Europe. Young IG. Chamla MC. Drouin J. 1997. 1994. and linguistics. Homo sapiens sapiens en Afrique: des origines au ne ´ olithique. Susanne C. Grebenart D. Barcelona: Ariel. Gonza ´ lez AM. Quattro JM. editors. Sykes B. Ferhat N. Moral P. Ould Khattar M. Le Roux MG. Origin and evolution of Native American mtDNA variation: a reappraisal. Excoffier L. Amblard S. Cavalli-Sforza LL. Bosch E. Sanger F. p 1–120. D’Anna R. Population genetics and disease susceptibility: characterization of Central European haplogroups by mtDNA gene mutations. In: Barandiara ´ n I. 1996. In: Renfrew C. Stefa ´ nsson K. Piazza A. 1981. Macaulay VA. Aumassip G. p 348 –358. del Rinco ´ n MA. editors. Cabrera VM. Maya JL. Princeton: Princeton University Press. Milieux. Analysis of mtDNA variation in African populations reveals the most ancient of all human continentspecific haplogroups. Comas D. 1995. 2000b. Pe ´ rez-Lezaun A. Co ˆ rte-Real HB. editors. son e ´ volution. Genetics 131:479 – 491. Bathily M. Gerbitz KD. Paris: CNRS. Ann Hum Genet 59:63– 81.A. Am J Hum Genet 59:935–945. Hum Genet 99:443– 449. 1998. Prehistoria de la Pen´ ınsula Ibe ´ rica. Mart´ ı B. Genetic affinities among human populations inhabiting the Subsaharan area. Wallace DC. correlations with D loop variants and association with disease. S. Tauveron M. Excoffier L. Cavalli-Sforza LL. Striedter KH. Forster P. Eperon IC. 1983. Northwest African distribution of the CD4/Alu microsatellite haplotypes. Quechon G. Genetic diversity in the Iberian Peninsula determined from mitochondrial sequence analysis. sa diversite ´ . Comas D. Bandelt HJ. Mitochondrial DNA variation and the origin of the Europeans. 1996. Bertranpetit J. Le peuplement prehistorique du Sahara. editor. Harding R. de Bruijn MHL. Dutour O. Chiavetta V. volume 4. Human mitochondrial DNA variation and the origin of the Basques. Those of HVSI less 16. El Paleol´ ıtico y el Mesol´ ıtico. Maya JL. Hariti G. Papiha S. AQ: 4 . In: Ferembach D. Hum Mol Genet 6:1835–1846. MtDNA and the islands of the North Atlantic: estimating the proportions of Norse and Gaelic ancestry. editors. Smith AJH. 1992. Vernet R.000. Ayala GF. Bosnes V. Int J Legal Med 114:229 –231.A. Bezold R. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA data. Calafell F. Paris: L’Harmattan. Menozzi P. Goodacre S. Godinho VM. hommes et techniques du Sahara prehistorique. New York: Kluwer Academic/Plenum Publishers. Anderson S. Thinon M. Roe BA. Chen Y-S. 1995. Flores C. Maca-Meyer N. Onrubia-Pintado J. L’homme. Paprotta A. p 39 –52. Prehistoric Iberia: genetics. Smouse PE. (1981). Santachiara-Benerecetti AS. Flugy A. anthropology. Dutour O. DeLeo G. Vernet R. Cambon-Thomsen A. Helgason A. Nature 290:457– 465. Mori F. and the Iberian Peninsula. Mart´ ı B. 2001. Nierlich DP. Ann Hum Genet 64:321–327. MtDNA control region and RFLP data for Sicily and France. Gaussen J. Ann Hum Genet 60:331–350. 2001. Cabrera VM. Underhill P. Richards MB. Proble ` mes actuels. Am J Hum Genet 57:133–149. S. 1994. In: Barandiara ´ n I. Bertranpetit J. Barrell BG. Ferembach D. Ward R. 1998. del Rinco ´ n MA. The explanation of culture change: models in prehistory. Flores C.

Pa ¨a ¨ bo S. Diversity of mtDNA lineages in Portugal: not a genetic edge of European variation. Maya JL. Bonne ´ -Tamir B. Rapid. Livshits LA. Gulf of Guinea. Pinto F. Saillard J. Petrozzi M. 1993. 1996. editor. Golge M. 1999. Whittle A. Sykes B. Tracing European founder lineages in the Near Eastern mtDNA pool. Richards M. 1998. Richards M. Bandelt H-J. Renfrew C. Prata MJ. Beckman L. Larruga JM. Savontaus M-L. Decorte R. Eur J Hum Genet 6:365–375. Switzerland. Rosser ZH. Hickey E. Larruga JM. Aula P. Ann Hum Genet 60:499 –508. Am J Hum Genet 62:1137–1152. Co ˆ rte-Real HB. Dimitrov D. MtDNA analysis of the Galician population: a genetic edge of European variation. simple alkaline extraction of human genomic DNA from whole blood. Wilkinson-Herbots HM. Anttinen T. Tyler-Smith C. Morelli L. Pettener D. 2000. MtDNA analysis reveals a major late Palaeolithic population expansion from southwestern to northeastern Europe. D’Urbano L.tapraid5/c0-phan/c0-phan/c01202/c00619d02a heckt Sϭ17 9/9/02 11:04 Art: 00-1074 14 GONZA ´ LEZ ET AL. 1997. Am J Hum Genet 61:691–704. Mart´ ı-Oliver B. Hickey E. Passarino G. Amorim A. Ann Hum Genet 60:321–330. Int J Legal Med 109:107–113. buccal epithelial cells. Genome Res 5:42–52. Cabrera VM. Hagelberg E. Los primeros agricultores-ganaderos. Sistonen P. 1996. Hill EW. Dissing J. 1999. McElreavey K. Cali F. Rozas R. Rando JC. Scozzari R. rather than by language. Pinto F. Benson N. Paleolithic and Neolithic lineages in the European mitochondrial gene pool. Bertranpetit J. 1989. A tale of two islands: population history and mitochondrial DNA sequence variation of Bioko and Sao Tome. Bertranpetit J. Scozzari R. Int J Legal Med 111:299 –304. Villems R. Geneva: Genetics and Biometry Laboratory. Grijalbo Mondadori. Zerjal T. Calafell F. Classification of European mtDNAs from an analysis of three European populations. Juvonen V. 1987. Nafa K. In: Cunliffe B. Int J Legal Med 106:85–90. Watson E. Rengo C. The application of mitochondrial DNA typing to the study of white Caucasian genetic identification. Barbujani G. Excavation in the Bacho Kiro Cave (Bulgaria). Harpending H. editor. Golge M. Sykes B. Bioinformatics 15:174 –175. Forster P. Issel-Tarver L. Torroni A. 2000. Vilarinho L. Salas A. Patsalis PC. Macaulay V. In: Cunliffe B. Parik J. DnaSP version 3: an integrated program for molecular population genetics and molecular evolution analysis. Rozas J. 1998. Beckman G. Dubrova YE. 2000. La revolucio ´ n del Paleol´ ıtico Superior.1: a software for population genetic data analysis. Nat Genet 23:437– 441. 1987. Jobling MA. Evgrafov O. Hum Biol 71:431– 437. Kivisild T. Loirat F.A. Guida V. Am J Hum Genet 67:1526 –1543. Pe ´ rez-Lezaun A. Beckman L. Mitchell RJ. Forster P. 1997. 1998. Papiha S. Schneider S. 1998. Bertranpetit J. Lavinha J. Mol Biol Evol 4:406 – 425. Alavantic D. Mitochondrial DNA polymorphisms: a study of 50 French Caucasian individuals and application to forensic casework. Kayser M. Santorelli FM. Demaine A. Rootsi S. Roessli D. El Neol´ ıtico. Rudbeck L. Prehistoria de Europa Oxford. Ann Hum Genet 61:507–518. New York: Columbia University Press. Torroni A. Santos FR. Gedde-Dahl T. Jeziorowska A. Heizmann JP. Pennington R. Archaeology and language. Location and frequency of polymorphic positions in the mtDNA control region of individuals from Germany. Amos W. Am J Hum Genet 65:1457–1459. Kocher TD. semen and forensic stains for PCR. Grijalbo Mondadori. Rychkov Y. Stefanescu G. Torroni A. Lukka M. Kozlowski JK. Mitochondrial footprints of human expansions in Africa. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Final report. Thomas M. Wilson AC. Armenteros M. I. Francalacci P. Sykes BC. Richards M. del Rinco ´ n MA. Guida V. Miller KWP. Savontaus ML. Herna ´ ndez M. Demaine A. p 138 –168. Am J Hum Genet 67:1251–1276. Genetic evidence of an early exit of Homo sapiens sapiens from Africa through eastern Africa. Prata MJ. Comas D. Hill E. 1987. Cooper G. Biotechniques 25:588 –592. Rocha H. Papiha S. WilkinsonHerbots H. Bonne ´ -Tamir B. and sub-Saharan populations. Gene and lenguages in Europe: an analysis of mitochondrial lineages. Mateu E. Huoponen K. Carracedo A. Geographic patterns of mtDNA diversity in Europe. Sullivan KM. Ann Hum Genet 62:241–260. Warsaw: Panstwowe Wydawnietwo Naukowe. 1998. 1996. 1999. 2000. Bosch E. Mitochondrial DNA analysis of northwest African populations reveals genetic exchanges with European. Macaulay V. Cruciani F. Forster P. Pandya A. 1995. Richards M. 1999. 1996. Aula P. 1998. Arenas J. Phylogeography of mitochondrial DNA in Western Europe. Foster P. Bradley D. Rousselet F. London: Penguin. Hurles ME. Tranebjaerg L. Comas D. Scozzari R. Bandelt H-J. Lucotte G. Bandelt H-J. 1999. Kueffer J-M. University of Geneva. About the “pathological” role of the mtDNA T3308C mutation. S. The emerging tree of West Eurasian mtDNAs: a synthesis of control-region sequences and RFLPs. 1998. de Knijff P. Nicholson J. Pereira L. Ann Hum Genet 64:491–506. Vona G. Norby S. Int J Legal Med 111:67–77. Dawson JL. Sajantila A. 1982. Glisic S. Triantaphyllidis C. Cruciani F. Romano V. Krumina A. Y-chromosome DNA haplotype 15 in Europe. Barcelona: Cr´ ıtica. Nei M. Peterlin B. Prehistoria de la Pen´ ınsula Ibe ´ rica. Barcelona: Cr´ ıtica. Semino O. Hedges R. Richards M. Mitochondrial DNA sequences in single hairs from a southern African population. Rychkov S. Moral P. Rubinsztein DC. Roewer L. Bandelt H-J. Lutz S. Kivisild T. 1997. Excoffier L. Vega E. Arlequin version 1. Bandelt HJ. 1998. 1998.A. Sellitto D. Pielberg G. Pereira L. Piercy R. Savontaus M-L. Rychkov O. p 47– 81. DiRienzo A. Near-Eastern. Stefanescu G. Meitinger TA. Sellitto D. Gill P. Kravchenko SA. Tolun A. Bandelt H-J. Macaulay V. In: Barandiara ´ n. Kucinskas V. Arroyo E. Previdere C. Gilissen A. Mangin P. Lamminen T. Herna ´ ndez M. editors. Santachiara-Benerecetti AS. Savontaus ML. Nei M. Calafell F. Barbujani G. Kalaydjieva L. Am J Hum Genet 59:185–203. 1998. Vigilant L. 1996. Cabrera VM. Lahermo P. Malaspina P. Flores C. Mart´ ı B. Brede G. Sykes BC. Huoponen K. p 121–195. A concordance of nucleotide substitutions in the first and second hypervariable segments of the human mtDNA control region. Sistonen P. Villems R. Am J Hum Genet 64:232– 249. Obinu D. Macaulay V. Campos Y. Sykes B. Am J Hum Genet 66:262–278. Lahermo P. Amorim A.A. Molecular evolutionary genetics. Gonza ´ lez AM. Bertranpetit J. editor. Bradley DG. Wallace D. Gonza ´ lez AM. Genetic relationship between the Canary Islanders and their African and Spanish ancestors inferred from mitochondrial DNA sequences. Rengo C. Mikelsaar AV. McElreavey K. S. Adojaan M. Weisser HJ. Quintana-Murci L. Lareu MV. Maria S. Torroni A. Saitou N. Co ˆ rte-Real H. Prehistoria de Europa Oxford. Mellars P. Simoni L. Nikoskelainen E. Vega E. Proc Natl Acad Sci USA 86:9350 – 9354. Site 73 in hypervariable region II of the human mitochondrial genome and the origin of European populations. Y-chromosomal diversity in Europe is clinal and influenced primarily by geography. 1997. Sykes B. Ann Hum Genet 62:531–550. MtDNA haplotype analysis in Finnish families with Leber hereditary optic neuropathy (LHON). Barcelona: Ariel S. Genetics 144:1835–1850. Eur J Hum Genet 5:271–279. Richards MB. . Lahermo P.