HLA DNA sequences from archaeological bone

HLA DNA sequences from archaeological bone. In Charron, D. (ed.): HLA: Genetic Diversity of HLA Functional and Medical Implications, Vol 2. Paris: EDK, p. 280-2. ISBN 2-84254-003-4

Genetic

diversity

of HLA . Functional and Medical Implication Dominique Charron EDITOR

.....................

MEDICAL AND SCIENTIFIC

INTERNATIONAL PUBLISHER

Genetic diversity of HLA

HLA DNA sequences from archaeological bone. In Charron, D. (ed.): HLA: Genetic Diversity of HLA Functional and Medical Implications, Vol 2. Paris: EDK, p. 280-2. ISBN 2-84254-003-4

DNA sequences from archaeological bone M.P. Evison l ,2, D.M Smillie 2 , A. T. Chamberlain I 1 Department of Archoeology ond Prehistory, University of Sheffield, Sheffield S 10 2TN, UK 2 Trent Centre, National Blood Service, Langley Lone, Sheffield S7 5Jt'1 , UK

-

Analy s~ s of the HLA complex, when combined with DNA sf:xing methods , constitutes a unique tool fo r stu­ dying kinship and immunity in the archaeological record. Sequences recovered from mummified material have been assigned to class I and II HLA loci [1-4]. We analysed 92 teeth and bone specimens from archaeolo­ gical skeletons recovered from excavations ranging in date from the Palaeolithic to the Meu ieval Period, using standard PCR based HLA-DPB I typing and forensic DNA-sexing techniques.

Methods Precautions were taken against contamination with intrusive DNA. Equipment was cleaned by soaking in 0.5 % sodium hypochlorite solution for 1 hr. Pre- and post­ PCR activities ';"ere conducted in separate rooms and a laminar flow cabinet was used during the extraction step. Surfaces of bone or teeth were cleaned by washing with 0.05 % sodium hypochlorite solution or by abrasion with a grinding tip or drill bit, which was then discarded. Bone was ground to a fine powder in a coffee mill (Philips HR2811). Intact teeth were fractured longitudinally to expose the pulp cavity. We used a refinement of the silica method for DNA extraction [5]. Quanti ties of -1.0 g of bone or -0.1-0.7 g of tooth fragments were combined with 2.0 ml 0.5 M Na 2EDTA pH 8.0 and 25 )ll proteinase K (20 mg / ml) in 4 ml Khan tubes and mixed on a rota­ ry mixer for 48 hr at room temperature. Substrate residues were pelleted by centrifugation at 4,000 g for 5-15 min. Aliquots of 0.5 ml extract supernatant were bound to 20 )ll silica suspension using 1.0 ml 4 M guanidine isothio­ cyanate by mixing for 2 hr in 1.5 ml Eppendorf tubes on a rotary mixer at room temperat1 "·:> . DNNsilica matrix was pelleted by microcentrifugation for 20 seconds at 13,000 g and washed twice in 1.5 ml 70 % ethanol and

once in 1.5 ml acetone. The pellet was dried at 56 °C for 5 min. DNA was eluted from the silica into 115 )ll sterile filtered distilled water by heating at 56 °C for IS min, and vortexing every 5 min, to aid solution of the DNA. Silica was pelleted by centrifugation at 13,000 g for 2 min. Volumes of 105 jJl DNA solution were taken off to avoid inadvertent removal of unwanted silicate. Samples were stored in 0.5 ml Eppendorf tubes at - 20°e. All PCRs are of the sequence specific primer (SSP) type and were carried out using a Perkin-Elmer GeneAmp 9600 thermal cycler. PCR reaction mixtures and pro­ grams were optimi sed using modem DNA samples by D.S. Primer sequences are given in Table I. Table I. Ol igonucleotide primer sequences S'CCC TGG GCT CTG TAA AGA ATA GTG3' (Amel-A )

I

5' ATC AGA GCT TAA ACT GGG AAG CTG 3' (Amel-B ) I



S'GAG AGT GGC GCC TCC GCT CAT3' (DPB-AmpA)

S'GCC GGC CCA AAG CCC TCA CTC3' (DPB-AmpB)

S' ATG CTA AGT TAG CTT TAC AG3' (A)

S'ACA GTT TCA TGC CCA TCG TC3 ' (B)

The HLA-DPB I primers (DPB-AmpA and -AmpB) ari.plify a 327 -bp sequence of the polymorphic second exon [6]. The mtDNA analysis was included to provide a comparison of genomic and mtDNA survival. Primers A and B target a 121 bp segment of the non-coding region V [7]. The Arnel primers (Amel-A and -B ) target the X- Y homologous amelogenin gene [8]. The X homologue contains a 6 bp deletion. Products of 106 bp and 112 b. p are generated from the X and Y chromo­ somes, respectively. Blank controls were includpd in each extraction, purification and amplification ph , ·' ~ to allow detection of contamination with intrusive DNA. Positive controls were included in extraction (SO flJ

HLA DNA sequences from archaeological bone. In Charron, D. (ed.): HLA: Genetic Diversity of HLA Functional and Medical Implications, Vol 2. Paris: EDK, p. 280-2. ISBN 2-84254-003-4

Evolution of MHC. Ancient DNA

fable II. Table of results Amel

HLA-DPB I n

Forensic Archaeological

89 92

+

%

41 34

46 37

n

+

119

97 9

92

mtDNA %

82 10

Total

+

%

89

73

92

43

82 47

n

n

+

297 276

211 86

%

71 . 31

Results of PCR anal ys is for HLA-DPB I, amelogenin (Amel) and mtDNA region V (mtDNA) in arcryaeological tooth and bone SJmples. Results from a separate stUdy of forensic specimens are st. \wn for c,jmparison. Total number of samples (n), p."sitive sJmples (+) , generating a PCR produc( of the appropriate size, and % of positive samples are given.

modem blood in 450 /--II sterile filtered distilled water), purification (0.5 ml DNA solution at 10 ng / /--II) and amplification (5 /--II DNA solution at 10 ng / /--II) steps. All specimens were extracted and amplified in duplicate .

Results A degree of random contamination was experienced in extraction, purification and amplification blanks. Batches containing contaminated blanks were excluded from the study. A typical result is shown in Figure 1. Results are summarised in Table ll. Results of a study of forensic specimens are shown for comparison. As expected, mtDNA PCR products predominated. HLA­ DPB 1 sequences were successfully amplified less than half as commonly. Only limited success was achieved with amplification of amelogenin sequences, despite the shorter length product. Reproducibility of results in uplicate samples was poor, but tended to be better for mtDNA. Consistency was also poor, with samples often mplifying for only one primer system, usually mtDNA egion V. R:sults from ·UK archaeological specimens were generally poor, reflecting the poorer osteological tructure (assessed macroscopically). Teeth tended to be better substrate than bone. There was no correlation etween the age of the material and the presence of mplifiable DNA.

Ancient DNA results require scrupulous verification

9] .. It should be possible to confirm the amelogenin

exmg results by comparing them with the skeletal sex.

except two of the 39 amelogenin results were fema­ however, and sexing errors were exclusively "false . We have observed false female results in pre­ experiments with forensic specimens (data not

HLA-DPBI

mDNA region V

1 2 3 E B P \V

1 2 3 E B P \V

~~~...,

I ....... -

.

Figure 1. Photograph of 1.5 % agarose gel showing typical results for HLA-DPB I and mtDNA region V analysis. Aliquots of 4 fll DNA solution were added to I fll X5 gel loa­ ding buffer (40 % sucrose, 0.1 % bromophenol blue, 50 mM Na)EDTA, 50 mM Tris-HCI pH 7.6 and 5 % SDS) and run on a I ~5 % agarose gel (100 v., 20 min) containing 10 fll ethidium bromide (10 flg / ml). Single bands are generated by the ampli­ fication of 372 bp HLA-DPB I sequences and 121 bp sequences of the mitochondrial non-coding region V. Samples 1-3: DNA extracted from archaeological bone, E: extraction blank, B: silicate binding (purification) blank. P: PCR blank, W: molecular weight marker (MWVITT).

shown) and attributed these to preferential amplification of the shorter (X -chromosome) product in degraded samples. Stone et al. [10] reported the recovery of sequences of the amelogenin gene from a sample of 20 archaeological skeletons. They avoided the problem of allelic drop-out by amplifying a 112 bp region of ame­ logenin occurring without deletion on both the X and Y :-:hromosomes. They suggest that chance variation will influence whether the X or Y sequence tends to be amplified in males (in their comparison it again tends to be the X). Their attempts to apply the amelogenin sexing technique [8] we have used were not successfuL Lack of robusticity in the amelogenin PCR system may

HLA DNA sequences from archaeological bone. In Charron, D. (ed.): HLA: Genetic Diversity of HLA Functional and Medical Implications, Vol 2. Paris: EDK, p. 280-2. ISBN 2-84254-003-4

G enetic diversity of HLA

explain why its shorter length product is amplified less frequently than the longer HLA-DPB I product. We doubt that the results can be explained by a fortui­ tous pattern of inhibition or contamination. Contamination mi.ght be expected to generate false males at least as fre­ quently as false females and to effect each primer system similarly, especially if due to PCR carry over. False male results did not occur and there are distinct diffe­ rences in the pattern of results obtained from the diffe­ rent primer systems. If the pattern of results is due to inhibition of amplification of a substantial background of contaminating DNA in a subset of the samples, more overall consistency in the reslJlts might be expected. Ollr archaeological DNA samples rarely inhibit amplifi­ cation of modem DNA, further suggesting that this explanation is unlikely. It is possible that the apparent positive results are artefacts generated by in vitro recombination and "jumping PCR" occurring in degra­ ded and fragmented samples, containing a minimal quantity of potentially contaminated DNA substrate. We are currently resolving the HLA-DPB 1 alleles present in our sample by dot-blot hybridisatlon and direct . sequencing. Putative HLA-DPB 1 sequences were reco­ vered at a rate which would make analysis of polymor­ phisms viable in phylogenetic studies of archaeological skeletal material. Single copy nuclear DNA sequences are the most inter­ esting and difficult target for ancient DNA analysis. The proportion of positive results obtained from archaeolo­ gical material will be improved if teeth, which appear to be a better source than bone, are used, and if analysis is restricted to skeletons with good histological preserva­ tion. The silicate method of DN A extraction will remo­ ve inhibitors from most archaeological bone and tooth samples. Dedicated clean conditions, such as a clean room, are esselAtial to reduce the likelihood of contami­ nation. Cloning and sequencing of multiple PCR pro­ ducts will be necessary to establish the validity of HLA results from ordinary archaeological material. If pos­ sible, results should be verified via DNA independent controls .. Skeletal sex and the ankylosing spondylitis pathology serve this purpose. We suggest that the extent of contamination and negative results should be repor­ ted, and that positive results should be reproduced in an independent HLA laboratory. ~

Acknowledgements This study was supporled by NERC and Trent Celltre, Nari ollQ/ BlOod Service, where the laboratory work was carried out. We rhank Manthos Bessios, Birger Her:og, Nina Kvparissi-Aposro/ika, Maria Pappa and Beth Rega for access to archaeological marerial.

References 1. Lawl~r 6A, Dickel CD, Hauswirth WW, Parha.rn P. 2.

3. 4.

5.

6.

. AncientHLA genes from7 ,500-year-old archaeo;;)gi­ cal remains . Nature 1991; 349: 785-8. Hauswirth WW, Dickel CD, Rowold DJ, Hausv, ' ;~'1h MA. Inter- and intrapopulation studies of ancient human remains. Experientia J994; 50: 585-91. Del Pozzo G, Guardiola A. Mummy DNA fragment identified. Nature 1989; 339: 431-2. Woodward SR, King MJ, Chiu NM Kuchar MJ, C-iggs CWo Amplification of ancient nuclear DNA from teeth and soft tissues. peR Methods and Application.~ 1994; 3: 244-7. H6ss M, Paabo S. DNA extraction from Pleistocene bones by a silica-based purification method. Nucleic Acids Res 1993; 21 : 39 I3-4. Kimura A, Sasazuki T. Eleventh Internati onal Histocompatibility Workshop reference protocol for the HLA DNA-typing technique. In: Tsuji K, Aizawa M, Sasazuki T, eds. Proceedings of the Ele1'('l1th Intemational Histocompatibility Workshop ond Conference. Oxford: Oxford University Press, 1992.

7. Wrischnik LA, Higuchi RG, Stoneking M, Erlich HA., Amheim N, Wilson AC. Length mutations in human mitochondrial DNA - direct sequencing of enzymatically amplified DNA. Nucleic Acids Res 1987; J 5: 529-42. 8. Sullivan KM, Mannucci A," Kimpton CP, Gill PA. Rapid and quantitative DNA sex test: flourescence­ based PCR analysis of X- Y homologous gene amelo­ genin. BioTechniques 1993; 15: 636-4 I. 9. Richards MB, Sykes BC, Hedges REM. Authenticating DNA extracted from ancient skeletal remains. J Archaeol Sci 1995; 22: 291-9. 10. Stone AC, Milner GR, Paabo S, Stoneking M. Sex . determination of ancient human skeletons using DNA. ,l',m J Phys Anthropol 1996; 99: 231-8.