High frequency of lactose intolerance in a prehistoric hunter-gatherer population in northern Europe
- Helena Malmström†1,
- Anna Linderholm†2, 5Email author,
- Kerstin Lidén2,
- Jan Storå3,
- Petra Molnar3,
- Gunilla Holmlund4,
- Mattias Jakobsson1 and
- Anders Götherström1
© Malmström et al; licensee BioMed Central Ltd. 2010
Received: 25 September 2009
Accepted: 30 March 2010
Published: 30 March 2010
Genes and culture are believed to interact, but it has been difficult to find direct evidence for the process. One candidate example that has been put forward is lactase persistence in adulthood, i.e. the ability to continue digesting the milk sugar lactose after childhood, facilitating the consumption of raw milk. This genetic trait is believed to have evolved within a short time period and to be related with the emergence of sedentary agriculture.
Here we investigate the frequency of an allele (-13910*T) associated with lactase persistence in a Neolithic Scandinavian population. From the 14 individuals originally examined, 10 yielded reliable results. We find that the T allele frequency was very low (5%) in this Middle Neolithic hunter-gatherer population, and that the frequency is dramatically different from the extant Swedish population (74%).
We conclude that this difference in frequency could not have arisen by genetic drift and is either due to selection or, more likely, replacement of hunter-gatherer populations by sedentary agriculturalists.
The ability to drink milk as an adult occurs at a high frequency in present-day populations that practice dairying and cattle rearing [1–4]. It has been suggested that this correlation represents a case of gene-culture co-evolution, i.e. an adaptive genetic trait exposed to positive selection induced by cultural practices. The -13910*T allele associated with this trait in Europeans appears to have been the target of strong selection over a relatively short period of time [5, 6]. Such selection pressure could be one of several explanations for the high frequency of the derived allele (associated with allowing milk consumption in adulthood) in northern Europeans, the region where the allele is most common (74% in Sweden ). A single nucleotide polymorphism (SNP-13910 T/C), strongly associated with the ability to digest lactose in adulthood , has been used as a marker for the genetic trait. Notably, the allele frequency at the SNP and the extent of haplotype homozygosity around the particular SNP-allele indicate a history of a strong positive selection [5, 6, 9, 10], and it has been suggested that this selective pressure was attributable to the introduction of agriculture and animal domestication [10, 11].
The Pitted Ware Culture (PWC) was a major Neolithic hunter-gatherer population in Northern Europe and was partly contemporaneous with the farming TRB population (TRB after the German word Trichterbecherkultur, i.e Funnel Beaker Culture). The PWC are thought to have been present in Scandinavia between 5,400-4,300 years before present (BP) , which is later than the suggested initiation of selection for the T allele . In this study, we find that the frequency of the derived allele is low in the PWC (5%) compared to the frequency in the extant Swedish population, and that the change in frequency is incompatible with genetic drift as the sole explanation under a model of population continuity. Thus, a genetic component interacting with culture, such as the ability to digest milk as an adult, could have been the result of the replacement of the hunter-gatherer population by an agricultural population. Alternatively, positive selection could have dramatically increased the frequency of the derived allele in the PWC, allowing for population continuity from the PWC to the extant Swedish population.
Alleles at the polymorphic site -13910 in the lactase gene in 10 Middle Neolithic Scandinavian samples representing the Pitted Ware Culture (PWC).
U4 or H1b
U4 or H1b
U4 or H1b
U4 or H1b
DNA was extracted from approximately 100 mg of bone powder pre-treated with bleach . The samples were extracted in a laboratory specially dedicated to aDNA work at the Archaeological Research Laboratory in Stockholm, as well as in a newly built ancient DNA facility at the National Board of Forensic Medicine in Linköping that is separated from the department's PCR areas. The samples were extracted at least twice, and both water blanks and extracts from contemporary prehistoric seals were included as negative controls. The DNA extractions were performed using the method of Yang et al. .
To screen for DNA content indicating authenticity, real-time PCR was performed on the selected samples as well as on 98 negative controls (where 31 were DNA extracts from ancient non human mammals representing the same context as the selected samples and 67 were water blanks). An 80 bp fragment in the mitochondria was targeted (L4567F 3'CACTGATTTTTTACCTGAGTAGGCCT5', H4595R 3'CGAGGGTTTATTTTTTTGGTTAGAACT5'), and amplifications were carried out according to a previously described protocol .
The C/T polymorphism at position -13910 upstream from the lactase gene was amplified by two different fragments, a shorter 53 bp fragment and a longer 168 bp fragment. The two sets of primers shared the biotinylated forward primer (5'→3' GCTGGCAATACAGATAAGATAATG), but had different reverse primers for the 53 bp fragment (5'→3'GAGGAGAGTTCCTTTGAGGC), and the 168 bp fragment (5'→3' ATGCCCTTTCGTACTACTCCC). The system was originally set up to detect differences in preservation and authenticity of the material analyzed, relying on the fragmentation level . The PCR amplification was carried out using 5 μl of extract, 300 nM of each primer and the Illustra Hot Start Ready-To-Go mix (GE Healthcare Life Sciences). The PCR profile was 15 min at 95°C, followed by 43 cycles of 30 s at 94°C, 30 s at 54.6°C and 30 s at 72°C, with a final extension step of 15 min at 72°C. Pyrosequencing was used for allele identification. The sequencing primers were designed to anneal next to the SNP (5'→3'CCTTTGAGGCCAGGG). The pyrosequencing was performed according to supplier provided protocols, and as previously described in .
Real-time PCR results
The 14 selected samples yielded over a 1000-fold higher concentration of human DNA (average 23,307 molecules) than the negative controls (average 18 molecules) . When divided in non-human negative controls and water negative controls, the concentration of human DNA was still considerably lower in both cases compared to the 14 selected samples (non-humans 18 molecules, water blanks 18 molecules). When only considering the 10 samples yielding reproducible results, the concentration of human DNA also exceeded the negative controls by over a 1000 times (Figure 2).
The fourteen specimens all yielded replicable results and ten of these were successfully SNP-typed a minimum of four times (using both the 53 bp and the 168 bp fragment). Thus, even with thorough mitochondrial pre-selection, the success rate for the nuclear DNA sequences was down to 71%. As only one individual was heterozygote, it is not possible to calculate an allelic dropout rate, but from the four pyrosequencing results from the heterozygote individual, two replicate typing events expressed allelic dropout. A low number of positive results were detected in the negative controls, eight of the 59 Neolithic seals, seven of the 53 negative extraction controls and eight of the 47 PCR negative controls. In total, 14% of the negative controls were contaminated, and out of these, the frequency of the T allele was 52%. None of the contaminants could however be reproduced. The allele frequency in the negative controls was significantly different from the allele frequency in the ten archaeological samples used for further analysis (Fisher's Exact test, p < 0.001). The contamination frequency is in direct contrast to that of the short mitochondrial fragment, where 86% of the negative controls (98 negative controls, where 31 were nonhuman mammals and 67 water controls or PCR controls) expressed contamination, although to a more than 1000-fold lower quantity than the ancient human samples (Figure 2).
Genetic drift can cause differences in allele frequencies between two samples taken from the same population separated in time. Since the PWC and extant samples are separated by a substantial amount of time (around 4000 years), a simple Fisher's exact test would not account for genetic drift, but would only be relevant for the null-hypothesis of no difference between the two samples. To explore the possibility that genetic drift alone caused the difference in T allele frequency, we can model a prehistoric population (for example the PWC population) as being ancestral to a population existing today. To evaluate different demographic scenarios (e.g. constant population sizes or population expansion from the time of the prehistoric sample), we simulated an extant sample and a prehistoric sample using the program COMPASS . COMPASS simulates different temporal samples under a coalescent model, and allows a range of demographic assumptions, for example, population expansion. For a particular assumption about the demographic model, we ran a large number of replicate simulations, and kept only samples that contained exactly 144 derived alleles (the T allele) in the extant sample. In other words, we conditioned on observing exactly 144 derived alleles in the extant sample as in the empirical modern data set. Out of 1,000 replicate simulations (which all had 144 derived alleles in the extant sample), we counted the number of times zero or one derived allele was observed in the prehistoric sample. This fraction can be used to approximate the probability of observing a prehistoric sample with zero or one derived allele, conditional on observing 144 derived alleles in the extant sample.
To conclude, for a neutral site (a SNP) in an unstructured population, we note that, 1) assuming a constant population size (Figure 4C), or an exponentially growing population (Figure 4A), and for reasonable assumptions about the population sizes, the probability is very small of observing the configuration of one or no derived allele (out of 20) in the prehistoric sample and 144 derived alleles (out of 194) in the extant sample. 2). Even for extreme population histories, such as a constant population of 100 individuals until 225 generations ago, and where the population grew exponentially to 10,000 individuals (Figure 4B), or a constant size of 500 individuals (Figure 4D), the probability of observing one or zero derived alleles in the prehistoric sample is still small. Thus, it is unlikely that genetic drift caused the observed frequency in the PWC given the frequency in the extant Swedish population. This leaves either selection or differences in population structure between the prehistoric sample and the extant sample (i.e., the PWC is not an ancestral population to the extant Swedish population) as possible explanations.
Authentication of DNA sequences in prehistoric human remains has proved complicated, mainly because modern human contamination is always present to some degree [19–21]. Here we rely on a combination of several arguments for our acceptance of the results as authentic [21–23]. The most important argument is that the success rate, like the allele frequencies, differs among the prehistoric human remains, the present-day human samples and the negative controls. Further, our material had already been pre-screened for contamination, including a major contamination investigation in which human and nonhuman DNA was typed in large parallel series of human and animal assemblages from the same sites (Figure 3) .
The frequency of the derived T allele, strongly associated with the ability to consume unprocessed milk at adulthood, was significantly different between the prehistoric and the extant samples. For reasonable assumptions about population sizes our simulations show that it is highly unlikely that the frequency could have shifted over time due to drift in a population with constant size, or in an expanding population. Thus, only strong positive selection or differences in population structure between the PWC population and the population ancestral to the extant Swedish population would explain the discrepancy between the extant and the prehistoric data. We note that with the data on a single SNP presented here, it is not possible to discriminate between these two scenarios. Also, it should be noted that the prehistoric samples used in this study came from a non-agricultural population, existing in parallel with an agricultural population (TRB) that is potentially ancestral to the extant Northern European population . In Burger et al.  one Mesolithic, nine Neolithic and one Medieval samples were analyzed. In this study 9 of the 10 individuals were homozygous C at position -13.910 only the medieval individual was heterozygous for the -13.910-C/T polymorphism. These data show some similarity with our data as the Mesolithic individual does not show any lactase persistence.
Here we describe a possible scenario where cultural practices could have had a tremendous impact on the genetic composition of human populations. Before dairy farming was practiced in Scandinavia, the allele frequency at -13910 was not greatly affected. The neolithisation processes in Scandinavia provided a selection pressure on this specific locus on a scale that may actually have led to a population replacement in the area (as indicated by mitochondrial DNA ), or at least a major shift in the allele frequency at this particular locus. This would require that cultural changes were dependent upon demographic changes in this case, and that cattle farming presented a clear advantage to hunting/gathering. In such case the 'secondary product revolution' (i.e. the introduction of milk) in northern Europe would have a large effect on the derived allele at -13910. Thus, the development that led to the present-day North European lifestyle, which is heavily based on dairy and other farm products, may have been a process where cultural practices and genes interacted.
It is unlikely that the PWC and their ancestors used dairy products before the middle Neolithic, nor is there any archaeological evidence for such practices . The difference in allele frequency from the extant Scandinavian population may therefore be explained by the fact that PWC is not ancestral to the extant Scandinavian population. An alternative, but less likely, explanation posits an ancestral relationship, but with selection dramatically increasing the frequency of the derived allele.
The authors would like to thank the anonymous reviewers for their helpful comments and Rolf Quam for revising the language. AG was supported by the Royal Swedish Academy of Science. MJ is supported by the Swedish Research council FORMAS and Carl Trygger's foundation. The study was partly financed by the Swedish Research Council.
- McCracken RD: Lactase Deficiency: An Example of Dietary Evolution. Current Anthropology. 1971, 12: 479-10.1086/201234.View ArticleGoogle Scholar
- Holden C, Mace R: Phylogenetic analysis of the evolution of lactose digestion in adults. Human Biology. 1997, 69: 605-628.PubMedGoogle Scholar
- Bellwood P: First Farmers. The Origins of Agricultural Societies. 2005, Malden: Blackwell PublishingGoogle Scholar
- Beja-Pereira A, Caramelli D, Lalueza-Fox C, Vernesi C, Ferrand N, Casoli A, Goyache F, Royo LJ, Conti S, Lari M, Martini A, Ouragh L, Magid A, Atash A, Zsolnai A, Boscato P, Triantaphylidis C, Ploumi K, Sineo L, Mallegni F, Taberlet P, Erhardt G, Sampietro L, Bertranpetit J, Barbujani G, Luikart G, Bertorelle G: The origin of European cattle: Evidence from modern and ancient DNA. Proc Natl Acad Sci USA. 2006, 103: 8113-8. 10.1073/pnas.0509210103.PubMed CentralView ArticlePubMedGoogle Scholar
- Bersaglieri T, Sabeti PC, Patterson N, Vanderploeg T, Schaffner SF, et al: Genetic signatures of strong recent positive selection at the lactase gene. Am J Hum Genet. 2004, 74: 1111-1120. 10.1086/421051.PubMed CentralView ArticlePubMedGoogle Scholar
- Tishkoff SA, Reed FA, Ranciaro A, et al: Convergent adaptation of human lactase persistence in Africa and Europe. Nat Genet. 2007, 39: 31-40. 10.1038/ng1946.PubMed CentralView ArticlePubMedGoogle Scholar
- Kuokkanen M, Butzow R, Rasinperä H, Medrek K, Nilbert M, Malander S, Lubinski J, Järvelä I: Lactase persistence and ovarian carcinoma risk in Finland, Poland and Sweden. International Journal of Cancer. 2005, 117: 90-94. 10.1002/ijc.21130.View ArticleGoogle Scholar
- Enattah NS, Sahi T, Savilahti E, Terwilliger JD, Peltonen L, Jarvela I: Identification of a variant associated with adult-type hypolactasia. Nat Genet. 2002, 30: 233-237. 10.1038/ng826.View ArticlePubMedGoogle Scholar
- Coelho M, Luiselli D, et al: Microsatellite variation and evolution of human lactase persistence. Human Genetics. 2005, 117: 329-339. 10.1007/s00439-005-1322-z.View ArticlePubMedGoogle Scholar
- Itan Y, Powell A, Beaumont MA, Burger J, Thomas MG: The Origins of Lactase Persistence in Europe. PLoS Comput Biol. 2009, 5: e1000491-10.1371/journal.pcbi.1000491.PubMed CentralView ArticlePubMedGoogle Scholar
- Burger J, Kirchner M, Bramanti B, Haak W, Thomas MG: Absence of the lactase-persistence-associated allele in early Neolithic Europeans. Proc Natl Acad Sci USA. 2007, 104: 3736-3741. 10.1073/pnas.0607187104.PubMed CentralView ArticlePubMedGoogle Scholar
- Malmer M: The Neolithic of south Sweden: TRB, GRK, and STR. 2002, Royal Swedish Stockholm: Academy of Letters, History and Antiquities, Almqvist & Wiksell InternationalGoogle Scholar
- Malmström H, Gilbert MT, Thomas MG, Brandström M, Storå J, Molnar P, Andersen PK, Bendixen C, Holmlund G, Götherström A, Willerslev E: Ancient DNA reveals lack of continuity between neolithic hunter-gatherers and contemporary Scandinavians. Curr Biol. 2009, 19: 1758-62. 10.1016/j.cub.2009.09.017.View ArticlePubMedGoogle Scholar
- Malmström H, Svensson EM, Gilbert MTP, Willerslev E, Götherström A, Holmlund G: More on Contamination: The Use of Asymmetric Molecular Behavior to Identify Authentic Ancient Human DNA. Mol Biol Evol. 2007, 24: 998-1004. 10.1093/molbev/msm015.View ArticlePubMedGoogle Scholar
- Yang DY, Eng B, Waye JS, Dudar JC, Saunders SR: Improved DNA extraction from ancient bones using silica-based spin columns. American Journal of Physical Anthropology. 1998, 105: 539-543. 10.1002/(SICI)1096-8644(199804)105:4<539::AID-AJPA10>3.0.CO;2-1.View ArticlePubMedGoogle Scholar
- Anderung C, Bouwman A, Persson P, Carretero JM, Ortega AI, Elburg R, Smith C, Arsuaga JL, Ellegren H, Götherström A: Prehistoric contacts over the Straits of Gibraltar indicated by genetic analysis of Iberian Bronze Age cattle. PNAS. 2005, 102: 8431-8435. 10.1073/pnas.0503396102.PubMed CentralView ArticlePubMedGoogle Scholar
- Jakobsson M: COMPASS: A program for generating serial samples under an infinite sites model. Bioinformatics. 2009, 25: 2845-10.1093/bioinformatics/btp534.View ArticlePubMedGoogle Scholar
- Tenesa A, Navarro P, Hayes BJ, Duffy DL, Clarke GM, Goddard ME, Visscher PM: Recent human effective population size estimated from linkage disequilibrium. Genome Research. 2007, 17: 520-526. 10.1101/gr.6023607.PubMed CentralView ArticlePubMedGoogle Scholar
- Haak W, Forster P, Bramanti B, Matsumura S, Brandt G, Tanzer M, Villems R, Renfrew C, Gronenborn D, Alt KW, Burger J: Ancient DNA from the First European Farmers in 7500-Year-Old Neolithic Sites. Science. 2005, 310: 1016-1018.PubMedGoogle Scholar
- Sampietro ML, Gilbert MT, Lao O, Caramelli D, Lari M, Bertranpetit J, Lalueza-Fox C: Tracking down human contamination in ancient human teeth. Mol Biol Evol. 2006, 23: 1801-7. 10.1093/molbev/msl047.View ArticlePubMedGoogle Scholar
- Malmström H, Stora J, Dalen L, Holmlund G, Gotherstrom A: Extensive Human DNA Contamination in Extracts from Ancient Dog Bones and Teeth. Mol Biol Evo. 2005, 22: 2040-2047. 10.1093/molbev/msi195.View ArticleGoogle Scholar
- Richards MB, Sykes BC, Hedges REM: Authenticating DNA Extracted From Ancient Skeletal Remains. Journal of Archaeological Science. 1995, 22: 291-299. 10.1006/jasc.1995.0031.View ArticleGoogle Scholar
- Kolman CJ, Tuross N: Ancient DNA analysis of human populations. American Journal of Physical Anthropology. 2000, 111: 5-23. 10.1002/(SICI)1096-8644(200001)111:1<5::AID-AJPA2>3.0.CO;2-3.View ArticlePubMedGoogle Scholar
- Linderholm A, Malmström H, Lidén K, Holmlund G, Götherström A: Cryptic Contamination and Phylogenetic Nonsense. PLoS ONE. 2008, 3: e2316-10.1371/journal.pone.0002316.PubMed CentralView ArticlePubMedGoogle Scholar
- Eriksson G, Linderholm A, Fornander E, Kanstrup M, Schoultz P, Olofsson H, Lidén K: Same island, different diet: Cultural evolution of food practice on Öland, Sweden, from the Mesolithic to the Roman Period. Journal of Anthropological Archaeology. 2008, 27: 520-543. 10.1016/j.jaa.2008.08.004.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.