Neanderthals and humans interbred in Europe for 5,400 years

Schizophrenia affects approximately 1% of the world’s population and has accompanied humans through much of our recorded history (1, 2, 3, 4, 5, 6). This seemingly human-specific disorder is characterized by hallucinations and delusions (often involving language), thought disorders, and higher order cognitive dysfunctions. The mechanisms of schizophrenia are not well understood, but its heritability is high, between 60% and 80% (7), and the fecundity of affected people is reduced (8). Nevertheless, the prevalence of the disease seems to remain stable across generations, giving rise to the yet unresolved “evolutionary enigma” of schizophrenia (3, 4, 9, 10). Large variations in incidence across populations argue for environmental causes. However, by using standard, precisely drawn diagnostic criteria, the variation in incidence can be reduced (11). Classic explanations include a single, partially dominant gene with low penetrance giving slight physiologic advantages (12); balanced selection, where the gene variants conferring risk of the disease provide an advantage in particular environments; and hitchhiking, where disease variants are passed along with advantageous neighboring gene variants. Newer studies have focused on the polygenic nature of schizophrenia and have attributed the prevalence of the disease to the sporadic nature of complex disorders (13).


Figure 1

We first assessed the influence exerted on schizophrenia association propensity by the Neanderthal “character” of the DNA region of the SNP, as measured by the NSS score selection index (Supplemental Figure S1) (21). Using data from the recently published schizophrenia GWAS (20), we conditioned schizophrenia association p values on the NSS score. The conditional Q-Q (Figure 1A) and fold enrichment (Figure 1B) plots show that SNPs with negative NSS scores are enriched for associations with schizophrenia compared with SNPs with positive NSS scores. These results are nominally confirmed by the BPT (p = 2.40 × 10−2), and more robustly by the squared z-score regression against the NSS score (β = −.067, p = 7.30 × 10−9) (Table 1).

We carried out the same analyses on other phenotypes to assess the specificity of the evolutionary enrichment. As shown in Q-Q plots and fold enrichment plots (Figure 2), other phenotypes show mostly modest or scarce enrichment as a function of NSS compared with schizophrenia. The only other significant excesses of low p values were detected by the BPTs and the regression analyses for height and to some extent for body mass index (Table 1). Height in particular has effect size comparable to that of schizophrenia and possibly larger still, but its SE is larger. Targeted analyses of other psychiatric (attention-deficit/hyperactivity disorder, bipolar disorder, major depressive disorder) and neurologic (Alzheimer’s disease, migraine, multiple sclerosis) disorders revealed no measurable enrichment effect (Supplemental Figures S4 and S5). Schizophrenia has by far the largest NSS effect size among the psychiatric and neurologic GWASs, all of which have similar SEs (Supplemental Figure S8).To test the extent to which the effect on schizophrenia depends on the power of the GWAS from 2014 (20), we performed the same analyses on the smaller schizophrenia GWAS from 2013 (40), which is comparable in size to several of the other GWASs. The enrichment was diminished (Supplemental Figure S6) but remained nominally significant according to the regression analysis (β = −.038, p = 7.93 × 10−3). We also tested the censored (Supplemental Methods and Materials) schizophrenia GWAS summary statistics and still found a significant (regression coefficient p = 2.87 × 10−6) residual enrichment.


Figure 2

The effect of brain genes affiliation on enrichment was investigated further by testing whether brain genes with negative NSS scores are more enriched of associations with schizophrenia than any brain genes. The enrichment plots (Figure 3) for brain genes with negative NSS scores show a wider deflection from baseline, and the BPT shows a significant difference in the proportion of association p values in the lowest percentile (p = 5.5 × 10−3).


Figure 3

Previous studies of evolutionary factors of schizophrenia focused on small sets of genes (41, 42). The analysis of Bigdeli et al. (43) was more systematic but applied human accelerated regions as evolutionary proxy. Xu et al. (44), using special human accelerated regions, showed that genes next to human accelerated regions in primates were under greater selection pressure compared with other genes and are more likely to be associated with schizophrenia susceptibility loci. Green et al. (21), who reported the first Neanderthal draft sequence, introduced the selective sweep score and investigated its relationship with the disease association but only for the most significant genes singled out by their analysis. In the present study, we used the information from the original publication of Green et al. (21) and the more recent report by Prüfer et al. (22) on the complete Neanderthal genome sequence to identify evolutionary enrichment patterns with a polygenic approach. Another asset for our study was the availability of a large schizophrenia GWAS of >80,000 participants (20), which makes it feasible to investigate evolutionary factors in schizophrenia with adequate power.

The results presented here are in line with the idea of polygenic adaptation, which is believed to play a role in the development of many complex human diseases, as it likely happened in adaptation of humans to pathogens and in the variation of morphologic traits such as height (45, 46, 47). Classic selective sweeps, originating from strong selective pressure, are relatively rare in modern humans (48), and natural selection is not the only factor shaping human variation. Instead, polygenic selection involving subtle shifts of allele frequencies at many loci simultaneously has been suggested to be common for complex traits in humans (48). Selection acting simultaneously on many standing variants could be an efficient mechanism for phenotypic adaptation (49, 50). Given these premises, it is desirable to use analytical tools designed to capture polygenicity. The methods applied in our analyses were useful in studying polygenic factors in schizophrenia previously (28, 29, 32, 36). Our results indicate that many schizophrenia susceptibility factors in modern humans may have emerged after their divergence from Neanderthals.

Several of the genes found that were likely to have undergone positive selection in modern humans (21) are involved in cognitive functions. The enrichment of SNP associations observed for schizophrenia may be due to an overlap between swept genomic regions and brain and other central nervous system genes and the regulatory regions thereof. This question is addressed by the regression analysis in which protein coding annotations are accounted for (Table 1). Even the inclusion of brain genes annotation scores in the regression did not reduce the enrichment for schizophrenia. However, among the brain genes themselves, the ones with a negative NSS score were more enriched of associations with schizophrenia compared with other brain genes, let alone just any genes (Figure 3).

Biological Psychiatry August 15, 2016; 80:284–292

The gene microcephalin or MCPH1 is a critical regulator of brain size. In humans, homozygosis for loss-of-function mutations in this gene causes a condition known as primary microcephaly, characterized by a severe reduction in brain volume (3- to 4-fold) but also by a retention of the overall neuroarchitecture, without evident defects outside of the brain [1]. The exact biochemical function of the microcephalin protein has yet to be elucidated, but it likely plays an essential role in promoting the proliferation of neural progenitor cells during neurogenesis [1].

Microcephalin has been proposed as the target of positive selection in the evolutionary lineage leading from ancestral primates to humans [2]. This observation, coupled with the fact that this gene is a critical regulator of brain size, suggests that the molecular evolution of microcephalin may have contributed to the phenotypic evolution of the human brain [2]. In a recent study, Evans et al. [3] found that a class of haplotypes at the locus, dubbed haplogroup D (all sharing a G37995C transversion in the coding region, that results in the substitution of an aspartate with a histidine), shows a recent coalescence age (point estimate  = 37,000 years) despite a very high worldwide frequency (0.70). There is also a marked difference between Africa (with low D frequencies) and Eurasia (with high D frequencies). Taken at face value, these findings would imply a rapid rise in the frequency of haplogroup D, so rapid indeed to suggest that positive selection, and not only drift or a demographic expansion, has affected the frequency of haplogroup D after the spread from Africa of early modern humans [3]. Evans and colleagues proposed that haplogroup D originated from a lineage separated from modern humans for 1.1 million years and introgressed into the human gene pool by 37,000 years ago, probably from a Neanderthal stock [4].


Table 1. Sequencing results for the amplifications of mtDNA fragments on MLS Neanderthal sample.

However, simulation approaches have shown that two key hypotheses of this model, namely positive selection and admixture, can be relaxed as long as Eurasia was settled from an African population that was both subdivided and under expansion [5]. Interestingly, variation in neurocranial geometry have recently suggested significant levels of geographic structure among early modern humans from Africa [6]. In addition, no direct empirical evidence supports a third key component of the model, ie. that Neanderthals carried alleles of the D haplogroup.

It is now possible to empirically test for a possible origin of haplogroup D in Neanderthals, by sequencing a fragment of the microcephalin locus overlapping the polymorphic position G37995C. The presence of microcephalin haplogroup D in Neanderthals would be the first molecular evidence of introgression between the two human forms, in sharp contrast with the results of mitochondrial studies suggesting that they never admixed (reviewed in Ref. 7). To test that hypothesis, we took advantage of a well preserved sample that already delivered authentic sequences from mitochondrial (hypervariable region I) and nuclear (melanocortin 1 receptor, MC1R) loci in order to reveal the Neanderthal genotype at the microcephalin locus [8], [9].


Table 2. Sequencing results for the amplifications of nuclear loci on MLS Neanderthal sample.

Our data show that the vast majority of clones amplified from the MLS Neanderthal extract contain the ancestral allele of MCPH1. This result (together with the high proportion of ancestral allele at LCT promoter -13910 control position) appears in agreement with a previous analysis of the same extract [8]. However, in another previous report [9] a moderate level (7–25%) of clones exhibited the presumed Neanderthal allele at the MC1R locus, suggesting that contaminant alleles could have outcompeted endogenous alleles (at the nuclear but not at the mitochondrial level). If true, the high frequencies of ancestral alleles found at LCT and MCPH1 in the MLS extract would partially result from Neandertal and modern human templates. Importantly, the ancestral allele at MCPH1 is found at significant frequencies among current European populations (22%, International HapMap Project, hapmap.ncbi.nlm.nih.gov/index.html.en, refSNP rs930557) and a significant fraction of the Southern European population is actually lactase deficient (eg. 64,3% in Northern Italy [18]). In addition, the ancestral allele at LCT has already been recovered from moderate (<17%) to maximal levels in non-human extracts, extraction and/or PCR controls, even though this allele is currently rare among human populations from the same geographic area [23]. We however do not believe that the MCPH1 and LCT genotypes reported here result from contamination since several evidences support the authenticity of the results. First of all, high proportion of identical mtDNA sequences unequivocally Neanderthal-type were repeatedly obtained from the sample; even if there is not direct evidence that mtDNA and nuclear DNA contamination levels are equivalent (or not too different), low proportion of modern human mtDNA contaminants is a basic pre-requisite for nuclear loci investigation in Neanderthal samples [16], [17], [24]. Second, we produced for each positive amplification a large number of clone sequences (between 20 and 40), and we performed on selected amplicon ultra-deep sequencing by 454 technology to achieve still higher resolution in detecting both sporadic contamination and misincorporations due to damage. Third, the nucleotide misincorporation pattern in amplicon carrying the ancestral allele is effectively suggestive of ancient templates, opposite to amplicons with derived allele that do not exhibit this trend. Finally, none of the people who handled or worked on the MLS sample carried the ancestral allele at both -13,910 LCT promoter and 37,995 MCPH1 gene positions (Table S3). Recently, new methods based on Primer Extension Capture have performed in limiting sequence retrieval from contaminant origins in ancient human extracts [25], [26]. Unfortunately, such methods cannot be used in the present case due to extract depletion, but, however, the most plausible hypothesis supported by all our findings is that the MLS individual was homozygous for the ancestral, non-D, microcephalin allele.

This result does not prove that there was no interbreeding between anatomically archaic and modern humans in Europe, but certainly shows that speculations on a possible Neanderthal origin of the most common MCPH1 alleles [4] are currently unsupported by empirical evidence from ancient DNA. Of course, the possibility exists that MCPH1 was polymorphic in Neanderthals, and that different individuals could carry a different allele. Support to this hypothesis can only be sought by studying other Neanderthal samples, and it is likely that it would receive shortly great attention due to the release of the complete Neanderthal genome. A statistically robust estimation of allele frequencies in Neanderthals may never be possible, but to maintain that the D haplogroup of MCPH1 came from Neanderthals to modern humans it is crucial to find evidence of at least one haplogroup-D in Neanderthals. Therefore, at present we see no cogent reason to abandon simpler models in which the evolution of microcephalin diversity occurred entirely within populations of Homo sapiens sapiens.

The deep split in the gene genealogy described for MCPH1 [4] has been observed at other loci as well, including pyruvate dehydrogenase alpha 1 [27], Dystrophin [28], coagulation factor FVII [29], and two regions of chromosome X [30], [31]. These high levels of haplotype divergence, caused by several nucleotide substitutions between two major branches of the gene tree, have been interpreted as evidence for introgression followed by positive selection [32], but in fact that is not the only plausible explanation.

Lari M, Rizzi E, Milani L, Corti G, Balsamo C, Vai S, et al. (2010) The Microcephalin Ancestral Allele in a Neanderthal Individual. PLoS ONE 5(5): e10648. doi:10.1371/journal.pone.0010648