|
Chapter
2 Introduction This report details the ancient DNA analyses for two bone samples [97-1-16(MCa) and 97-1-12b(13)frag] from the human remains commonly referred to as "Kennewick Man" or "Ancient One", which were received by Dr. Frederika Kaestle (FAK) on May 19, 2000. In laboratory records these are referred to as "Kennewick Metacarpal" or "Kenn M-C" and "Kennewick Rib" or "Kenn Rib" respectively. In brief summary, four DNA extractions were attempted on each sample. Results of amplification attempts are most consistent with the inability to extract ancient DNA from these samples using available technology. Due to the sensitivity of the study of these remains, digital images were taken during the extraction procedures to document the protocol, and copies on a zip disk were submitted to the Department of the Interior (DOI), as were hard copies of the electrophoretic gel images. Detailed extraction and amplification protocols may be found in appendix A. All extractions and amplifications were performed by FAK, primary investigator (8 years experience with ancient DNA). Ancient DNA: Background DNA, the genetic material contained in our cells, is normally protected from the majority of damaging forces such as hydrolysis, oxidation, chemical damage, and ultra-violet radiation by our living bodies, and the majority of damage that does occur is corrected by the activities of proofreading enzymes and other mechanisms in the cell. However, upon death these protections and repair mechanisms break down, and the DNA quickly degrades into small fragments with frequent sites of molecular damage. Thus, although we can extract DNA from the remains of deceased individuals, the analysis of this genetic material is subject to a multitude of problems that are not issues when analyzing modern DNA. First, because the DNA has degraded, very few intact copies of particular regions of the genome remain for analyses. Only the extremely sensitive methods of the Polymerase Chain Reaction (PCR), a process that allows us to make millions of copies of a target fragment of DNA, makes ancient DNA analyses possible. In addition, although this process allows the amplification of long target fragments from high-quality modern DNA (thousands of nucleotide bases long, in some cases), the fragmentary nature of ancient DNA results in the inability to amplify fragments of ancient DNA longer than approximately 200-300 nucleotide bases in most cases. This limits the type of genetic tests that can be applied to ancient DNA to those that do not require long intact fragments of DNA (for example, the traditional 'DNA fingerprint' technique, which involves digesting long fragments of DNA with various restriction enzymes and exposing them to a DNA probe, would not be feasible). The low number of copies and degraded state of those copies creates an additional problem for ancient DNA research, that of contamination. Because modern DNA is of superior quality, in the sense that the enzymes utilized in genetic analyses have evolved or been designed by humans to interact with long fragments of undamaged DNA, and because modern DNA simply exists in higher concentration, the presence of even small amounts of modern DNA (shed skin cells, cells from the mouth in aerosolized saliva, follicles from shed hair, etc.) in or on laboratory equipment and reagents can easily overwhelm the small amount of ancient DNA extracted from deceased individuals. Thus, every measure to eliminate the possibility of such contamination, and to detect it when it (almost inevitably) occurs, is required for ancient DNA research. The replication of ancient DNA work in multiple laboratories is also helpful in detecting contamination, as it is extremely unlikely that identical DNA contamination will occur in two separate laboratories. A final problem that complicates ancient DNA analyses has little to do with the extracted DNA, but actually involves other factors, which are coextracted with the DNA from the sample. In general ancient DNA is extracted from tissue, mostly bone and teeth, which has been exposed to various environments for long periods of time. Often the bones absorb factors from the surrounding soil matrix such as humic acids, or byproducts of fungus or bacteria living in the soil. In addition, byproducts of the breakdown of the soft tissue, such as Maillard products, can be absorbed. Many of these factors will coextract with DNA (that is, they are not eliminated during the extraction procedure), and interfere with the functions of enzymes used to amplify the DNA during PCR (e.g., Taq polymerase). The inhibition of the Taq polymerase during PCR results in little to no amplification of the target ancient DNA fragment, and the appearance that the extract contains no DNA. Therefore, it is important to eliminate inhibition as a possible cause of a false negative result. This can be done in several manners. During many PCR reactions some of the primers (short fragments of DNA added to the reaction to target a specific region of the genome) may bind to each other and be amplified by the Taq polymerase, producing a short fragment (usually 40 to 80 nucleotide bases long) usually called primer-dimer. Therefore, the presence of primer-dimer indicates the proper activity of the Taq polymerase, even in the absence of amplification of the target fragment of ancient DNA. Laboratory Facilities The molecular anthropology laboratory facilities at Yale are located in two physically separate laboratory spaces, with separate air systems. Each room is regularly decontaminated with bleach, and ultra-violet (UV) lights mounted in the ceilings are used each evening as an additional measure to destroy DNA. The laboratory equipment [including fume hood, bio-safety cabinet, refrigerators, freezers (both -20 and -80 degree), centrifuges (including vacuum and refrigerated), Ultra-Violet cross-linker, water baths, heating blocks, incubators, vortexers, rockers, pH meters, balances, vacuum pump, dremel tool, thermal cyclers, vertical and horizontal electrophoresis rigs, microwave, digital imaging system, and electronic and manual pipetors] are dedicated equipment that are never transferred between laboratories and are regularly bleach and/or UV decontaminated. All disposable products such as pipette tips and microcentrifuge tubes are either purchased DNA-free or of similar quality (e.g., Dnase, Rnase and protease free), or are decontaminated prior to use, as are laboratory reagents. Water is purchased in DNA and HPLC grade and often additionally filtered through a 0.2 um vacuum filter. Laboratory reagents are purchased and prepared in small volume to minimize opportunities for contamination during use. Laboratory personnel wear protective gear including laboratory coats, latex gloves, shoe covers, hair covers and face masks to prevent contaminating the ancient DNA laboratory with their own DNA, or with amplified DNA from our own or other laboratories. Ancient DNA extraction and the preparation of amplification reactions are performed in one of the laboratories, while the actual amplification procedure (PCR) and subsequent analyses are performed in the second laboratory. Automated sequencing of amplified and sequenced DNA is performed in a third laboratory located in a separate building on a shared ABI 377 Automated Sequencer. Data analyses are performed using a Mac G4 and a Dell PC with a Pentium III processor, located in FAK's office. Extraction Detailed ancient DNA protocols may be found in Appendix A. The following summarizes the application of these protocols to the Kennewick samples in particular. The accession number of the rib sample was removed with a razor blade before analysis. The metacarpal fragment did not have an accession number on it. Surfaces of both bone samples were cleaned with 10% bleach and UV irradiated to eliminate modern DNA contaminants. Bone powder was removed from internal surfaces of the bone fragments using a Dremel tool. In both cases the bone fragments were quite brittle and easily broken, generally indicative of poor preservation of organics. The presence of soil matrix within each sample was noted, and was particularly prevalent in the rib sample. Between 0.124 and 0.358 grams of bone powder was utilized in each extraction. In three of the extraction procedures the bone powder was decalcified over the course of three days using 0.5 Molar EDTA, pH 8.0, with EDTA negative controls added each day. In the fourth extraction, to address the possibility that ancient DNA was being lost during the EDTA washes, no decalcification was performed. In all cases the powdered bone and negative controls were digested using proteinase K overnight at 55 degrees Celsius. It was noted that after digestion a significant portion of the powdered bone remained, which is typical of samples that are heavily mineralized, as was the case with these samples (Powell and Rose 1999). Under optimal conditions (i.e., little to no mineralization) decalcified bone powder will be completely digested by proteinase K. In one extraction, the digested bone powder and negative controls were filtered and concentrated using Centricon-100 filtration units (MilliporeTM), and silica extracted using the QiaQuick PCR Purification Kit (QiagenTM) (Yang et al.1998). In a second extraction the digested bone powder and negative controls were phenol-chloroform extracted and then filtered and concentrated using Centricon-100 filtration units. In a third extraction a small portion of the Centricon-100 filtered digested bone powder and negative controls were extracted according to the manufacturer's protocols using GeneReleaser (BioVenturesTM), a proprietary product designed to combine DNA extraction and amplification. In the fourth extraction, the digested but non-decalcified samples and negative control were filtered and concentrated using Centricon-100 filtration units and silica extracted using the QiaQuick PCR Purification Kit. Attempts were made to amplify short fragments of ancient DNA from each extraction using primers designed for ancient DNA analysis (see table 2 for primers and annealing temperatures). In cases of amplification inhibition by co-extracted PCR inhibitors (see above) it is sometimes possible to overcome inhibition through dilution. Therefore, all sample extracts except the GeneReleaser extracts were diluted to 1:10 and 1:50 concentrations and amplified. On occasion additional digestion of contaminating proteins after extraction can also overcome PCR inhibition. For this reason a small portion of the first (silica) extracts were re-digested with proteinase K in the presence of Tween-20 and amplified. Finally, amplifications were attempted using three of the extractions (excluding the GeneReleaser extract) in very high concentrations (approximately four times the usual concentration) in an attempt to compensate for possible low to absent concentration of ancient DNA. All amplifications were performed with high concentrations of Bovine Serum Albumin, which has been shown to reduce the affect of PCR inhibitors often co-extracted with ancient DNA. Negative amplification controls were utilized in all cases, and positive controls (consisting of modern DNA from the investigator, extracted in the post-PCR laboratory) were utilized in amplifications involving high concentrations of extract. In some cases, amplifications were exposed to DNAse I before primers and extracted DNA were added, a procedure that has been shown to eliminate contaminating DNA from the PCR reagents (Eshleman and Smith 2000). PCR products were separated on agarose gels through electrophoresis, stained with ethidium bromide, and visualized under UV light. Digital images of the electrophoretic gels were taken. Amplification Initial efforts focussed on amplifying mitochondrial DNA (mtDNA) rather than nuclear DNA (nDNA). As discussed above, because DNA degrades over time due to the effects of temperature, hydrolysis, radiation, oxidation, and acidity, ancient samples often possess only highly damaged DNA (Herrman and Hummel 1994). MtDNA is located not in the nucleus but rather in the mitochondria, the cellular organelles responsible for energy production. Most cells possess hundreds of mitochondria, each with several copies of the mitochondrial genome. MtDNA is therefore present in many more copies than nDNA, and the likelihood that a particular fragment of mtDNA will survive over time is therefore much greater than the likelihood that a particular fragment of nDNA will. In addition, mtDNA is inherited only through the female line (i.e., mothers pass their mtDNA on to all of their children but fathers pass no mtDNA on). This allows us to identify maternal lineages of mtDNA united by their possession of a set of common mutations. All individuals possessing a particular mtDNA lineage can theoretically trace their maternal ancestry back to a single woman. Many mtDNA lineages, commonly called haplogroups, are limited in distribution among modern humans (but note that some are present in a wide range of modern peoples). Studies of modern individuals who are maternally Native American have shown that the majority possess mtDNA belonging to one of five maternal lineages, or haplogroups, called A, B, C, D, and X (e.g., see Smith et al. 1999; Lorenz and Smith 1996; Merriwether et al 1995; Schurr et al. 1990), defined by the possession of the diagnostic mutations seen in Table 1. Although most modern Native American populations studied to date possess at least four of these haplogroups, haplogroup frequencies vary significantly between many groups, and some groups completely lack several of these haplogroups. In some cases, it appears that populations living within a restricted geographic region (such as the American Southwest) share similar frequencies of these haplogroups, while in other cases it appears that populations speaking related languages (such as the Algonquians) share similar haplogroup frequencies (Lorenz and Smith 1996; Malhi et al. 2000). In addition, within haplogroups particular single nucleotide mutations often occur in a relatively restricted area or population (Kaestle 1998; Schurr et al. 1999). Studies of ancient mtDNA from pre-Columbian Native Americans have shown that the majority of these individuals also possessed haplogroup A, B, C, D, or X (e.g., see Kaestle et al. 1999; O'Rourke et al. 1999; Stone and Stoneking 1998). Haplogroups A through D are also found among modern Asians, albeit in relatively low frequency, and haplogroup B is also found among modern Pacific Islanders (Hagelberg and Clegg 1993; Merriwether et al. 1996, 2000; Torroni et al 1993). The modern distribution of haplogroup X is less well characterized. Thus far it has been found in low frequencies in Europe and Western Asia, but a suite of additional mutations distinguish the American haplogroup X lineages from those in the Old World (Brown et al 1998; Smith et al. 1999). It is important to note that the possession of one of these haplogroups simply indicates that an individual has maternal Native American or Asian ancestors. What proportion of these ancestors derived from Native American or Asian populations, and the identity of the individual's paternal ancestors, is unknown. In addition, the terms Native American, Asian, European, etc. are not meant to imply racial differences or groups, but are instead fluid geographic categories describing fuzzy boundaries that may have shifted throughout prehistory. Table 1. Markers identifying the five common mitochondrial haplogroups found among modern Native Americans. A '+' indicates the presence of a restriction site, a "-" indicates the absence of a restriction site. 'np' = nucleotide position.
Amplifications were performed using primers that flank
regions of the mitochondrial genome known to contain mutations defining
the five maternal lineages present among modern Native Americans and
Asians (see table 2). The results of amplification attempts and restriction
digests of positive amplifications are presented in table
3. Table 2: Primers used for PCR amplification
In all cases, except the amplification attempts using
high concentrations of extracts, no amplification product was detected,
indicative of either the absence of DNA in the extracts or the presence
of a PCR inhibitor that co-extracted with the ancient DNA. Again, except
in the case of high concentration of extracts, no sign of contamination
was detected in either the extraction negative controls or the amplification
negative controls. In some cases, primer-dimer, an amplified product
of 40 to 80 nucleotide bases consisting of two to four repeats of the
approximately 20 base primer used in the reaction, was detected (see
figure 1). The ability of the DNA polymerase to amplify the primer-dimer
in the presence of the DNA extracts is typical of reactions lacking
template DNA, and is inconsistent with the presence of PCR inhibitors.
If PCR inhibitors were present, the Taq enzyme would be of highly reduced
activity, and would not catalyze the amplification of primer-dimer.
The presence of primer-dimer therefore attests to the normal functional
state of the Taq enzyme.
Thus, the results from the positive amplifications from both the Kennewick samples and negative controls are not consistent with a source that is a member of haplogroups A, B, C, D or X, but are consistent with a source that is a member of haplogroup I (such as FAK). Because ancient DNA is highly damaged, amplification success is generally correlated with the length of the amplified product (shorter fragments being easier to amplify). Note that there is no correlation between amplification success and fragment length in this case, suggesting that the amplified DNA is not ancient. These results are consistent with a very low level of contamination of the extracts with modern DNA, most likely from FAK, but possibly originating at the reagent or lab disposables manufacturer. Although disposable pipette tips and tubes utilized during the extraction are certified DNA-free by the manufacturer, in reality the manufacturer's quality control methods will only detect DNA presence above 5 x 10-11 mg (EppendorfT). In addition, although all reagents utilized are produced in extremely sterile conditions and additionally filtered and/or Ultra-Violet irradiated in the laboratory, low levels of DNA contamination might still be present. However, haplogroup I, defined by the loss of the DdeI restriction site at nucleotide position 1715 and distinguished from haplogroup X by the absence of the AccI restriction site at nucleotide position 14465, is very rare among Europeans (generally 5% or less in frequency), and absent in all other populations except for extremely rare examples in the middle east (Brown et al, 1998; Torroni et al., 1994,1996). Therefore the most likely source of this DNA is FAK. Because any contamination is extremely minor it is only detectable when large concentrations of DNA extract are utilized in amplifications, and even then it is only detectable in a minority of cases. The amplification process will only amplify the target fragment if at least one copy of that fragment is present in the DNA extract that is added to the reaction, and in fact is very inefficient at such low concentrations of DNA. Thus, if the DNA in the extract is at such low concentrations that the probability of obtaining even one copy in the amount added to the PCR reaction is quite low, the result will be only intermittent amplification. For example, if there are only five copies of the target DNA in the total of 50 microliters of extract, there is only a 10% chance that one of these copies will be contained in any one microliter portion of the extract. However, if much larger amounts of the extract are used in the amplification, as in the high concentration amplifications above, the probability that at least one copy of your target DNA will be present is much higher. Thus very low levels of contamination are likely to be detected only when high concentrations of DNA extract is used in the amplification process. Given that it is almost impossible to prevent contamination at these extremely low levels, most ancient DNA amplifications using high concentrations of extract will eventually detect contamination. In conclusion, the inability to amplify DNA except at extremely high extract concentrations (and then with corresponding sporadic amplification of negative controls), the lack of evidence for PCR inhibition, the inconsistencies in which extracts amplified at high concentration, the identical restriction digest results for Kennewick samples and negative extract controls to those of FAK, and the lack of correlation between amplification success and fragment length are most consistent with the absence or very low concentration of ancient DNA in the two bone fragments under analysis and the presence of very low levels of modern contamination in these extracts. Thus available technology and protocols do not allow the analysis of ancient DNA from these remains. However, it is important to note that the success in extraction and amplification of ancient DNA does not seem highly dependent on the age of the sample, and several humans of similar or significantly older age have successfully been analyzed, both from the Americas (e.g., Kaestle 1997, 1998; Stone and Stoneking 1996) and from elsewhere (e.g., Krings et al. 1997; Naumova et al. 1998). It must be emphasized that the lack of success in amplifying ancient DNA from one sample has little bearing on the probability of success in the analysis of another. REFERENCES CITED Anderson, S.. A. T. Bankier, B. G. Barrell, M. H. L.
De Bruijn, A. R. Coulson, J. Drouin, I. C. Eperon, D. P. Neirlich, B.
A. Roe, F. Sanger, P. H. Schreier, A. J. H. Smith, R. Staden and I.
G. Young Brown, M. D., S. H Hosseini, A. Torroni, H-J. Bandelt,
. C. Allen, T. G. Schurr, R. Scozzari, F. Cruziani, D. C. Wallace Eshleman, J. E. and D. G. Smith Hagelberg, E. and J. B. Clegg Herrman B. and S. Hummel Kaestle, F. A. Kaestle, F. A. Kaestle, F. A., J. G. Lorenz and D. G. Smith Krings M., A. Stone. R. W. Schmitz, H. Krainitzki,
M. Stoneking, S. Paabo Lorenz, J. G. and D. G. Smith Malhi, R. S., B. A. Schultz and D. G. Smith Merriwether, D. A., F. Rothhammer and R. E. Ferrell
Merriwether, D. A, F. Rothhammer, R. E. Ferrell Merriwether, D. A., W. W. Hall, A. Vahlne, R. E. Ferrell
Merriwether, D. A., F. A. Kaestle, B. Zemel, G. Koki,
C. Mogne, M. Alpers and J. Friedlaender Naumova O Yu., S. Yu. Rychkov O'Rourke, D. H., R. L. Parr and S W. Carlyle Powell, J. F. and J. C. Rose Schurr, T. G., S. W. Ballinger, Y. Gan, J. A. Hodge,
D. A. Merriwether, D. N. Lawrence, W. C. Knowler, K. M. Weiss, and D.
C. Wallace Schurr, T. G., R. I. Sukernik, Y. B. Starikovskaya
and D. C. Wallace Smith, D. G., R. S. Malhi, J. Eshleman, J. G. Lorenz,
and F. A. Kaestle Stone, A. C. and M. Stoneking Stone, A. C. and M. Stoneking Torroni, A., T. G. Schurr, M. F. Cabell, M. D. Brown,
J. V. Neel, M. Larsen, D. G. Smith, C. M. Vullo and D. C. Wallace Torroni, A., M. T. Lott, M. F. Cabell, Y-S. Chen, L.
Labergen and D. C. Wallace Torroni A., K. Huoponen, P. Francalacci, M. Petrozzi,
L. Morelli, R. Scozzari, D. Obinu, M-L. Savontaus and D. C. Wallace
Yang, D. Y., B. Eng, J. S. Waye, J. C. Dudar and S.
R. Saunders APPENDIX A Ancient DNA Extraction Protocols:
Decontamination of bone surfaces and powdering of sample:
Decalcification of bone:
NOTE: steps 11-15 are unnecessary if proteinase K digestion will be carried out using an EDTA proteinase K buffer. Proteinase K Digestion:
0.5 M EDTA pH 8.0 50mM (.05M) Tris pH 8.0 DNA Extraction: Many options exist, including Phenol Chloroform or Silica Phenol Chloroform extractions have the advantage of providing a larger quantity of DNA while Silica extractions have the advantage of eliminating more of the amplification inhibitors that commonly co-extract with DNA. Recently, another option, the BioVentures GeneReleaser kit, has become available, which involves a proprietary reagent that "sequesters cell lysis products which might inhibit polymerases" and "avoids lengthy protocols and excessive sample manipulations which may introduce contamination", (BioVentures, Inc., Murfreesboro, TN, product literature, 1996). This protocol prepares only enough extract for one amplification. Phenol Chloroform Extractions:
Silica Extractions (modification of Yang et al. 1998 AJPA 105(4):539-544):
Note: On occasion, DNA extracts from the Phenol Chloroform protocol are later run through the Silica extraction protocol to remove co-extracted PCR inhibitors. GeneReleaser Extractions
PCR Amplification: Amplifications are set up in the pre-PCR ancient DNA laboratory. A PCR cocktail is prepared that contains all reagents [0.6 mM primers, 1X manufacturer's buffer with 1.5 mM MgCl2, ddH2O, 1 mg/ml BSA, 200 nM dNTPs] except the Taq polymerase and the ancient DNA template was UV cross-linked for 10 minutes to eliminate any exogenous DNA contamination. Standard ancient DNA primers for the regions encompassing the haplogroup-defining mutations and the hypervariable segments are employed (for example, see Stone and Stoneking (1993) American Journal of Physical Anthropology 92:463-471, Parr, Carlyle and O'Rourke (1996) American Journal of Physical Anthropology 99:507-518). The equivalent of "hot start" amplifications are performed in 25 - 50 ml volumes using 3-10 ml ancient DNA extract and 1.5 - 3 units of Amplitaq Gold DNA polymerase (Perkin Elmer) or Platinum Taq DNA polymerase (Gibco BRL). Tubes are sealed and transported to the physically separate PCR laboratory. The amplifications are subjected to an initial 2-11 minute denaturation step at 95ºC, then 35 to 45 cycles in which reactions are denatured at 95ºC for 30-15 seconds, annealed at 49-59ºC for 30-15 seconds, and extended at 72ºC for 25-60 seconds, followed by a final extension step of 3-5 minutes at 72ºC. At least one negative control, consisting of PCR cocktail, Taq DNA polymerase, and 3-5 ul sterile water, is included in each amplification set. In some cases the DNAse predigestion protocol (Eshleman and Smith 2000) was utilized. In this protocol, the PCR cocktail is prepared without the primers but including the Taq. Amplification grade DNAse is added to the cocktail to digest any contaminating DNA, and after incubation at room temperature for 10 minutes the DNAse is heat inactivated at 70ºC. The primers are added and the cocktail is aliquoted into the PCR tubes. PCR then proceeds as usual. Haplogroup Assignment Sequence Reactions: Cloning: |
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
MJB/EJL