doi: 10.1038/s41559-020-1106-9. Epub 2020 Feb 24.
Emergence of human-adapted Salmonella enterica is linked to the Neolithization process
Felix M Key 1 2 3, Cosimo Posth 4, Luis R Esquivel-Gomez 5, Ron Hübler 4, Maria A Spyrou 4, Gunnar U Neumann 4, Anja Furtwängler 6, Susanna Sabin 4, Marta Burri 4, Antje Wissgott 4, Aditya Kumar Lankapalli 4, Åshild J Vågene 4, Matthias Meyer 7, Sarah Nagel 7, Rezeda Tukhbatova 4 8, Aleksandr Khokhlov 9, Andrey Chizhevsky 10, Svend Hansen 11, Andrey B Belinsky 12, Alexey Kalmykov 12, Anatoly R Kantorovich 13, Vladimir E Maslov 14, Philipp W Stockhammer 4 15, Stefania Vai 16, Monica Zavattaro 17, Alessandro Riga 16, David Caramelli 16, Robin Skeates 18, Jessica Beckett 18, Maria Giuseppina Gradoli 19, Noah Steuri 20, Albert Hafner 20, Marianne Ramstein 21, Inga Siebke 22, Sandra Lösch 22, Yilmaz Selim Erdal 23, Nabil-Fareed Alikhan 24, Zhemin Zhou 24, Mark Achtman 24, Kirsten Bos 4, Sabine Reinhold 11, Wolfgang Haak 4, Denise Kühnert 5, Alexander Herbig 25, Johannes Krause 26
Affiliations
- PMID:32094538
- PMCID: PMC7186082
- DOI: 10.1038/s41559-020-1106-9
Item in Clipboard
Emergence of human-adapted Salmonella enterica is linked to the Neolithization process
Felix M Key et al. Nat Ecol Evol.2020 Mar.
Display options
Format
Display options
Format
Abstract
It has been hypothesized that the Neolithic transition towards an agricultural and pastoralist economy facilitated the emergence of human-adapted pathogens. Here, we recovered eight Salmonella enterica subsp. enterica genomes from human skeletons of transitional foragers, pastoralists and agropastoralists in western Eurasia that were up to 6,500 yr old. Despite the high genetic diversity of S. enterica, all ancient bacterial genomes clustered in a single previously uncharacterized branch that contains S. enterica adapted to multiple mammalian species. All ancient bacterial genomes from prehistoric (agro-)pastoralists fall within a part of this branch that also includes the human-specific S. enterica Paratyphi C, illustrating the evolution of a human pathogen over a period of 5,000 yr. Bacterial genomic comparisons suggest that the earlier ancient strains were not host specific, differed in pathogenic potential and experienced convergent pseudogenization that accompanied their downstream host adaptation. These observations support the concept that the emergence of human-adapted S. enterica is linked to human cultural transformations.
Conflict of interest statement
The authors declare no competing financial interests.
Figures
(a) PCA of newly reported ancient individuals withsufficient data (in red) and selected published ancient and modernindividuals are projected onto principal components built with present-dayWest Eurasian populations (grey dots). (b) ADMIXTURE analysis(K=10) of newly reported ancient individuals and relevant published ancientand modern individuals sorted by genetic clusters. Overview ancient humangenetic data SupplementaryTable 1 and further analysis Extended Data Fig. 4. EHG, Eastern hunter gatherer; E, Early; M,Middle; HG, hunter–gatherer; N, Neolithic; C, Caucasus; S,Scandinavian; W, Western; BA, Bronze Age.
(a) ADMIXTURE analysis (K = 3 - 16) of newly reportedancient individuals (bold horizontal text) and published ancient and modernindividuals sorted by genetic clusters and geographic origin (Europe, NearEast and Caucasus, Asia, America, Africa). Each K was run five times and thereplicate with the highest likelihood is reported. Ancient MK3001 showsAsian genetic ancestry components represented by Nganasan, Kankanaey,Atayal, and Ami. (b) Box plot of five cross-validations (CV)values for every K calculated in ADMIXTURE. EHG, Eastern hunter gatherer; E,Early; M, Middle; HG, hunter–gatherer; N, Neolithic; C, Caucasus; S,Scandinavian; W, Western; BA, Bronze Age.
Maximum likelihood tree of the AESB including the high coverageancient genomes and 463S. enterica genomes, consideringall SNPs covered in at least 95% of strains (130,036 SNPs). New ancientgenomes are shown in red, and previously reported ancient genomes (Tepos) inpink.
Estimated recombination rate is shown as recombination event permutation event (r/m) and indicated on top of branch and by branch color.Recombination events have been inferred using all positions shared by 95% ofstrains from the AESB and are here reported for the SNPs shared by allstrains on the AESB (correspond to maximum likelihood phylogeny shown inFigure 2B). Maximum likelihood treeincluding all SNPs shared by at least 95% of strains from the AESB is shownin Extended Data Fig. 3.
Results of the date randomization test for two subsets of theHC2600_1272 cluster (agro-pastoralist branch). Circles represent meansubstitution rate estimations with error bars representing 95% highestposterior density (HPD) intervals. For each subset 10 date randomizationswere done. Significant temporal signal is indicated by non-overlapping HPDintervals between real data (red) and the randomizations (black), which isthe case for both subsets.
Proportion of pseudo Temporal signal analysis.gene-sharing (0-100%) between strains on theAESB is shown in tones of red. Strains are ordered by phylogenetic branchand coloured accordingly.
C to T changes indicated in red and G to A changes in blue, allother substitutions in grey. IV3002 and MK3001 are UDG-half treated, whichleads to observable damage only in the terminal positions. Plots generatedwith mapDamage2 (Jónsson H.et al, Bioinformatics2013).
(a) MUR009; (b) OBP001; (c)MUR019; (d) IKI003; (e) IV3002; (f)ETR001; (g) SUA004; (h) MK3001.
Previously published ancient genomes from 13th century Norway (Ragna)and 16th century Mexico (Tepos) are also shown.
(a) Maximum likelihood tree of the ancient genomes (>5Xcoverage) and 2,961 modernS. enterica genomes, including182,645 SNP positions in the core genome. Selected branches are identified basedon predicted serotype provided by EnteroBase and coloured according to hostspecificity (blue/orange), if they include ancient genomes (red), or notspecified (black). (b) Maximum likelihood tree of the AESBincluding the ancient genomes (>5X coverage) and 463S.enterica genomes, considering 37,040 SNP positions in the coregenome. New ancient genomes are shown in red, and previously reported ancientgenomes (Ragna, Tepos) in pink. Ancient human economy is indicated for all newlypresented genomes based on archaeological and ancient human genetic information.Low coverage ancient genomes (coverage <5X) are phylogenetically placed(red dashed line) based on all SNP positions covered once: MK3001: 17,324;OBP001: 26,657; and Ragna: 35,465. Modern genomes are collapsed based on theirpredicted serovar, eBurst group (closely related sequence types), or availablemetadata in EnteroBase. Host adapted serovars are coloured orange (incl. apictogram of the host species). Bootstrap values are shown in black at each node(1,000 bootstraps). Black dashed rectangles show extends of Para C Lineage andhierarchical cluster HC2600_1272. Enteritidis P125109 is used as theoutgroup.
(a) Topology of the AESB highlighting the hierarchical clusterHC2600_1272 (yellow), with symbols indicating ancient human economy ofS. enterica positive samples. Selected divergence timeestimates for the HC2600_1272 cluster are shown in years BP (95% highestposterior density intervals, see also Supplementary Table 2). (b) Gain-loss resultsfor all SPI’s and selected genes. For SPI’s, colour gradientrelates to the average of %-genes covered over 95% across all strains perbranch. For genes, colour gradient according to mean percentage covered acrossall strains per branch. Tepos: 16th century Mexican.
(a) Relative frequency of pseudogenes for each strain of the AESB.Ancient genomes are identified and host generalists are shown in blue and hostadapted in orange. (b) Illustration of model used to inferevolution towards host adaptation. (c) Simulated expectation ofcandidate genes (phoN in green,ydcK in beige)to harbour randomly distributed pseudogenization events using 10,000simulations. (d) Host adaptedS. enterica serovarsthat harbour aphoN orydcK pseudogene. BA:Bronze Age, Tepos: 16th century Mexican.
Comment in
- Getting sick in the Neolithic.Stone AC.Stone AC.Nat Ecol Evol. 2020 Mar;4(3):286-287. doi: 10.1038/s41559-020-1115-8.Nat Ecol Evol. 2020.PMID:32094537No abstract available.
References
- Fowler C, Harding J, Hofmann D. The Oxford Handbook of Neolithic Europe. OUP Oxford; 2015.
- Cockburn TA. Infectious diseases in ancient populations. Current Anthropology. 1971;12:45–62. - PubMed
- Armelagos GJ, Cohen MN. Paleopathology at the Origins of Agriculture. Academic Press Orlando, FL; 1984.
- Barrett R, Kuzawa CW, McDade T, Armelagos GJ. EMERGING AND RE-EMERGING INFECTIOUS DISEASES: The Third Epidemiologic Transition. Annual Review of Anthropology. 1998;27:247–271.
Publication types
MeSH terms
Grants and funding
LinkOut - more resources
Full Text Sources