Content uploaded by Zeki Kaya
Author content
All content in this area was uploaded by Zeki Kaya on May 06, 2019
Content may be subject to copyright.
Content uploaded by Zeki Kaya
Author content
All content in this area was uploaded by Zeki Kaya on Apr 03, 2019
Content may be subject to copyright.
Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
54
Silvae Genetica 55, 2 (2006)
Closer relationships of P. uliginosa with dwarf mountain
pine mirrors also reported crossability. Peat-bog pine
share some traits with Scots pine due to introgression
but could lost crossability due to isolation.
Based on these data and earlier studies, close phyloge-
netic relationship between P. uliginosa and P. mugo
complex is suggested. This experiment is based on limit-
ed material and includes only one way of pollination.
Therefore, in context of hypothesized P. uliginosa hybrid
derivation, the obtained results unable to infer definite-
ly about putative parental species. Reliable experiments
should be founded on a more detailed material, repre-
senting more comprehensively examined populations of
different species and conducted for several years to
exclude influence of biotic and abiotic factors.
References
BORATYN
´SKI, A., K. BORATYN
´SKA, A. LEWANDOWSKI, Z.
GOL
⁄A¸Band P. KOSIN
´SKI (2003): Evidence of the possibili-
ty of natural reciprocal crosses between Pinus sylvestris
and Pinus uliginosa based on the phenology of repro-
ductive organs. Flora 198: 377–388.
BUSINSKY
´, R. (1999): Taxonomic essay in the Pinus mugo
complex and its hybrid population (in Czech). Acta
Pruhoniciana 68: 123–143.
CHRISTENSEN, K. I. (1987): Taxonomic revision of the
Pinus mugo complex and P. x rhaetica (P. mugo x
sylvestris) (Pinaceae). Nord. J. Bot. 7: 383–408.
DANIELEWICZ, W. and J. ZIELIN
´SKI (2000): Ochrona sosny
bl
⁄otnej Pinus uliginosa A. Neumann na terenie Borów
Dolnos
´la¸skich. Przegl. Przyr. 9: 113–124.
KORMUTAK, A. and M. LANAKOWA (1988): Biochemistry of
reproductive organs and hybridological relationships of
selected pine species (Pinus sp.). Acta Dendrob. 7–103.
KRZAKOWA, M., B. NAGANOWSKA and M. A. BOBOWICZ
(1984): Investigations on taxonomic status of Pinus
uliginosa Neumann. Bull. Soc. Pol. Acad. Sci. 24: 87–95.
LEWANDOWSKI, A., A. BORATYN
´SKI and L. MEJNARTOWICZ
(2000): Allozyme investigations on the genetic differen-
tiation between closely related pines – Pinus sylvestris,
P. mugo, P. uncinata and P. uliginosa (Pinaceae). Pl.
Syst. Evol. 221: 15–24.
LEWANDOWSKI, A., J. SAMOC
´KO, K. BORATYN
´SKA and A.
BORATYN
´SKI (2002): Genetic differences between two
Polish populations of Pinus uliginosa, compared to
P. sylvestris and P. mugo. Dendrobiology 48: 51–57.
MARCET, E. (1967): Über den Nachweis spontaner Hybri-
den von Pinus mugo Turra und Pinus sylvestris auf
Grund von Nedelmerkmalen. Ber. Schweiz. Bot. Ges.
77: 314–361.
MOULALIS, D., C. BASSIOTIS and D. MITSOPOULOS (1976):
Controlled pollinations among Pine species in Grece.
Silvae Genet. 25: 95–107.
MUIR, G., C. C. FLEMING and C. SCHLATTERER (2000):
Species status of hybridizing oaks. Nature 405: 1016.
PRUS-GL
⁄OWACKI, W. and J. SZWEYKOWSKI (1980): Serologi-
cal characteristics of some putative hybrid individuals
from a P. sylvestris x P. mugo swarm population. Acta
Soc. Bot. Pol. 49: 127–142.
PRUS-GL
⁄OWACKI, W. and J. SZWEYKOWSKI, 1983: Studies on
isoenzyme variability in populations of Pinus sylvestris
L., Pinus mugo Turra, Pinus uliginosa Neumann and
individuals from a hybrid swarm population. Bull. Sci.
Amis. Poznac
´. 22: 107–122.
SIEDLEWSKA, A. and W. PRUS-GL
⁄OWACKI (1995): Genetic
structure and taxonomic position of Pinus uliginosa
Neumann population from Wielkie Torfowisko
Batorowskie in Stol
⁄owe Mts. (Locus classicus). Acta Soc.
Bot. Pol. 64: 1–8.
STASZKIEWICZ, J. and M. TYSZKIEWICZ (1969): Naturalne
mieszan
´ce Pinus mugo Turra x Pinus sylvestris L. w
Kotlinie Nowotarskiej. Fragm. Flor. Geobot. 15:
187–212.
STASZKIEWICZ, J. (1993): Variability of Pinus mugo x P.
sylvestris (Pinaceae) hybrid swarm in the Tisovnica
nature reserve (Slovakia). Polish Bot. Stud. 5: 33–41.
WACHOWIAK, W., A. BA¸CZKIEWICZ, K. CELIN
´SKI and W. PRUS-
GL
⁄OWACKI (2004): Species-specific chloroplast DNA poly-
morphism in the trnV-rbcLregion in Pinus sylvestris
and Pinus mugo. Dendrobiology 51: 67–72.
WACHOWIAK, W., K. CELIN
´SKI and W. PRUS-GL
⁄OWACKI
(2005): Evidence of natural reciprocal hybridization
between Pinus uliginosa and P. sylvestris in sympatric
population of the species. Flora 200: 563–568.
Abstract
Ancient Taurus cedar (Cedrus libani A. Rich) wood
samples from the Tumulus of King Midas at the Gordion
archaeological site (about 2700 years old), near Ankara,
Turkey, and from the Al-Aksa Mosque (about 1500 to
1900 years old), Jerusalem, Israel, were characterized
by studying the sequences of ribosomal DNA (rDNA)
internal transcribed spacers (ITS1 and ITS2). After
DNA From Ancient Cedar Wood From King Midas’ Tomb, Turkey,
and Al-Aksa Mosque, Israel
By S. O. ROGERS1),2),*)and Z. KAYA1),3)
(Received 22th July 2005)
1) State University of New York, College of Environmental
Science and Forestry, Syracuse, NY, USA.
2) Current Address: Department of Biological Sciences, Bowling
Green State University, Bowling Green, OH, USA.
3) Current Address: Department of Biological Sciences, Middle
East Technical University, 06531 Ankara, Turkey.
*) Corresponding author: SOR; phone: 419-372-2333; fax: 419-
372-2024; E-mail address: srogers@bgnet.bgsu.edu
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
55
extraction of the DNA, the ITS regions were amplified
utilizing the polymerase chain reaction, followed by
sequencing, BLAST searches for similar sequences, and
phylogenetic analyses. Fifty-six sequences were
obtained. In BLAST searches of existing sequence data-
bases, most were closest to those from humans and
fungi. However, two sequences exhibited similarities
with conifer ITS sequences. One was an ITS1 region
from the Gordion wood specimen, and the other one was
an ITS2 region from the Al-Aksa wood specimen. Phylo-
genetic analyses indicated that both were closest to Tau-
rus cedar (C. libani, also known as Lebanon cedar) ITS
sequences from three recent samples of Taurus cedar
from two sites in Turkey. However, they exhibited many
differences from the recent C. libani rDNA ITS
sequences from Turkey, probably due to degradation of
the DNA in the ancient samples. The implications of the
results on future studies are discussed.
Key words: Ancient DNA, Midas, Gordion, Cedrus libani, wood.
Introduction
The major vegetation type in central Anatolia below
700 m of elevation is typical steppe. For the areas above
700 m of elevation, and usually at the edges of steppe
vegetation, scattered trees are common in a steppe-for-
est formation that can be fairly dense in higher eleva-
tions. Millennia of human influence on the landscape
make it hard to imagine how the natural vegetation of
central Anatolia once appeared. But pollen and charcoal
studies in archaeological sites in central Turkey such as
the ancient Gordion city site suggest that the major veg-
etation of the past consisted of juniper, pine and oak
species. The ancient Gordion city was the capital of the
Phyrigian Civilization and it is one of the few excavated
urban sites in central Turkey that date to the second
millennium B.C., and was occupied continuously for sev-
eral thousand years. Environmental deterioration,
caused by deforestation and grazing at this site and
other similar archaeological sites in the vicinity, has
been documented, and seems to be a relatively recent
phenomenon (MILLER, 1998; MILLER, 1999).
The discovery of rare wood in the Midas (the Phrygian
King) Tumulus indicates that there might have been
continuous forest from south to north including the Gor-
dion site. For instance, juniper (Juniperus spp.), pine
(Pinus nigra pallasiana), and Taurus cedar (Cedrus
libani) timbers are present in the Midas Tumulus. The
presence of Taurus cedar wood is especially interesting
since the closest natural cedar forest, that is in Afyon
province today, is about 100 km south of Gordion. The
closest natural occurrence of Taurus cedar forest to the
northeast (Niksar and Erbaa towns of Tokat Province) is
about 500 km from Gordion (BOYDAK, 2003; Figure 1).
Also, present in the Tumulus is furniture made from
boxwood (Buxus sempervirens). Again, the closest box-
wood stands are located in the Black Sea forests that
are about 125 km north of the Gordion site. These find-
ings suggest that Taurus cedar timbers either were car-
ried from long distances for wealthy and powerful per-
sons, such as in the case of the King Midas Tumulus
(MILLER, 1999), or the cedar forests once were continu-
ous from Afyon in the south, to Niksar (Tokat Province)
in the north, where the natural cedar stands can be still
seen today (AYTUG, 1988; BOYDAK, 2003). Similarly,
C. libani in Israel and surrounding countries consists of
very limited refugia, presumably being remnants of
more extensive forests. In either case, considering the
current distribution of Taurus cedar, as well as other
tree species that coexist with Taurus cedar in the Mid-
dle East, large portions of the forest genetic resources
may have been lost permanently. It is important to
establish the ancient distribution of C. libani in the
Middle East. It will be of interest to identify the alleles
that potentially have been responsible for adaptation of
Taurus cedar to changing environments and wood pro-
duction in the past several thousand years.
Figure 1. – Map of the natural range of C. libani in Turkey (shaded areas) and the location of the
Gordion archeological site. Afyon-SD and Isparta-KD on the map indicate the locations of the recent
C. libani samples of Afyon-Sultandag (ZK AS 31704 4) and Isparta-Kapidag (ZK IK 31704 4 and ZK
IK 31704 12), respectively.
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
56
Previous DNA studies (KAYIHAN, 2000) of Taurus cedar
indicated that there remains a large amount of genetic
diversity in the studied populations although the magni-
tude of diversity is lower than some other conifers occur-
ring in the same geographic regions (KANDEDMIR et al.,
2004; IÇGEN, 2003). It is possible that Taurus cedar
genetic resources in the Mediterranean regions have
diminished over the centuries due to human activities.
Ancient preserved specimens represent a resource of
potentially useful alleles that have become extinct. The
genetic composition of animal and plant species can be
deduced from their fossils (bones, wood, seeds, etc.) and
materials as old as tens of thousands of years old, as
well as living tissues, such as leaves, buds, embryos, and
other tissues using molecular methods developed for
ancient specimens (CANO and BORUCKI, 1995; GUGERLI et
al., 2005; HOFREITER et al., 2001; MAet al., 2000; PÄÄBO
1985, 1993; PARDUCCI and PETIT, 2004; ROGERS, 1994;
ROGERS and BENDICH, 1985, 1988, 1994; ROGERS et al.,
1989; SAVOLINEN et al., 1995). The great majority of
ancient DNA (aDNA) studies have been with animal tis-
sues, while molecular studies of wood are limited to a
few reports (DEGUILLOUX et al., 2002; HOFREITER et al.,
2001; KIM et al., 2004; PARDUCCI and PETIT, 2004; TANI et
al., 2003). A study of intraspecific aDNA from fossils and
archaeological materials (e.g., Taurus cedar wood from
the Midas Tumulus and other sites) could provide link-
ing information between palaeoecology, population
genetics, and phylogeography of species that should
improve our knowledge of past evolutionary processes
(GUGERLI et al., 2005; TANI et al., 2003).
The wood of Taurus cedar in the Midas Tumulus was
preserved to an extent that it could be identified by
AYTUG (1988). Thus, it is possible to investigate the
genetic structure of the buried wood materials and
assess genetic differentiation between the buried mate-
rials representing the past population and nearest pre-
sent population of the species. With this study, the
groundwork is established to sequence DNA from
ancient C. libani wood specimens. The specific objectives
of the current study were as follows: (1) to optimize the
DNA extraction procedure from ancient Taurus cedar
wood specimens from the King Midas Tumulus in Gor-
dion that are about 2700 years old and from specimens,
and from the Al-Aksa Mosque (removed and stored dur-
ing renovations from 1969-70 repair work) that are 1500
to 1900 years old; and (2) to characterize the DNA from
ancient Taurus cedar wood specimens using molecular
methods.
Materials and Methods
Specimens
Two ancient C. libani wood specimens were examined
in this study. One was from the Anatolian Civilization
Museum, Ankara, Turkey (provided by Mr. HI
.KMET
DENI
.ZLI
.and Mr. MEHMET AKALI
.N, Anatolian Civilization
Museum, Ankara, Turkey). Originally, it was part of the
King Midas Tumulus 100 km west of Ankara (Figures 1
and 2) and was estimated to be approximately 2700
years old (FILLEY et al., 2001). The other wood specimen
was from Al-Aksa Mosque, Jerusalem, Israel. It was
sampled from a piece of wood removed from the mosque
during renovations in 1969–70. It was approximately
1500 to 1900 years old (LIPHSCHITZ et al., 1997). Three
recent samples of megagametophyte tissue also were
examined in this study. They were collected for compari-
son to the sequences from the ancient Taurus cedar
specimens. They were from living C. libani trees sam-
pled at two sites, Afyon-Sultandag (elevation: 1400 m;
latitude: 38° 32’ 02’’ N; longitude: 31° 09’ 07’’ E; one spec-
imen), and Isparta-Kapıdag (elevation: 1600 m; latitude:
38° 05’ 23’’ N; longitude: 30° 42’ 20’’ E; two specimens),
both located southwest of Gordion in Turkey, respective-
ly). The specimens were designated as ZK AS 31704 4,
ZK IK 31704 4, and ZK IK 31704 12.
DNA extraction protocol
Prior to this study, the laboratory had never been used
to examine Taurus cedar tissues or nucleic acids. All
work surfaces were cleaned with Clorox (undiluted), fol-
lowed by 95% ethanol. All tubes, containers, and solu-
tions were autoclaved prior to use. Because wood is
porous, it is impossible to sterilize the surfaces of the
Figure 2. – Views of the Midas Tumulus. (a) Exterior of the Tumulus showing its large size, and rela-
tively arid surface. (b) Interior of the central chamber of the Tumulus, with logs and timbers
(juniper and pine) visible. The logs and timbers are in states of relatively good preservation. Cedar
wood (the source of wood for this study) is found further inside this Tumulus.
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
57
wood. Therefore, contaminants from fungi, humans, and
other organisms were expected. To minimize the concen-
trations of contaminants, the exposed surfaces of the
wood samples were removed with a sterile scalpel. Then,
a new sterile scalpel blade was used to remove wood
from the interior portion of the specimens. These interi-
or samples were stored in sterilized microfuge tubes at
–80°C prior to extraction of the DNA. Extraction was
accomplished using the CTAB (cetyltrimethylammoni-
um bromide, or hexadecyltrimethylammonium bromide)
method of ROGERS and BENDICH (1985, 1988, 1994). The
frozen pieces (ancient and modern, separately) were
ground in a cold (stored at –20 °C) sterile mortar and
pestle. Then, the ground samples each were transferred
into sterile microfuge tubes. For each mg of powdered
wood, 2 µl of 65°C extraction buffer (2% CTAB, 100 mM
Tris (pH 8.0), 20 mM EDTA, 1.4 M NaCl and 2%
polyvinylpyrrolidone (PVP)) was added and mixed thor-
oughly. First, the mixture was incubated at 65°C for 30
minutes. Then, the mixture was emulsified with an
equal volume of chloroform/isoamyl alcohol (24:1) and
centrifuged for 1 minute in a microfuge. The aqueous
phase was transferred into a new microfuge tube, fol-
lowed by addition of one-fifth volume of a 5%
CTAB/0.7M NaCl solution. A second chloroform/isoamyl
alcohol extraction was performed. Following 1 minute of
centrifugation in a microfuge, the aqueous phase was
transferred into a new microfuge tube and one volume
isopropanol was added, followed by gentle mixing. Fol-
lowing at least 2 hours at –4 °C, the precipitate was pel-
leted by centrifugation in a microfuge for 5 minutes. The
pellet was washed once with cold isopropanol, followed
by centrifugation (5 minutes). Then, the pellet was rehy-
drated in 50 µl of high salt TE (10 mM Tris (pH 8.0), 1
mM EDTA, 1 M NaCl) by heating at 65°C for 15 min-
utes and occasional vortexing. Two volumes of cold 95%
ethanol were added, followed by gentle mixing. The
DNA was allowed to precipitate at –20°C for at least 2
hours. Following centrifugation in a microfuge for at
least 15 minutes, the pellet was washed with 80%
ethanol. After centrifugation for 5 minutes, the ethanol
was decanted off and the pellet was dried using a
vacuum drier and was rehydrated in 20 µl 0.1X TE
buffer (1 mM Tris (pH 8.0), 0.1 mM EDTA). The extract-
ed DNAs were subjected to electrophoresis at 5 V/cm for
2 hours on 0.6% agarose gels (BioRad Laboratories,
Hercules, CA), with TBE (89 mM Tris, 89 mM boric acid,
2 mM EDTA) and 0.5 µg/ml ethidium bromide. Molecu-
lar weight markers (1kb ladder, Invitrogen, Inc., Freder-
ick, MD) were used as size markers on each gel.
Polymerase Chain Reaction (PCR) amplification/
reamplification
To characterize the extracted ancient DNA, primers
for PCR were used that amplify all or parts of the ribo-
somal DNA (rDNA) internal transcribed spacers (ITS1
and ITS2) of conifers. To amplify the entire region from
the 3’ end of the rDNA small subunit gene through the 5’
end of the rDNA large subunit gene (including ITS1, the
5.8S gene, and ITS2), primers prITS5 (GGAAG-
TAAAAGTCGTAACAAGG, forward primer) and prITS4
(TCCTCCGCTTATTGATATGC, reverse primer) (WHITE
et al., 1990) were used. Based on the sequences from
fresh megagametyphyte tissue of C. libani (KAYA,
unpublished data), the total length of the amplified frag-
ment generated using this primer pair is approximately
710–720 bp in length. For the ITS1 region, primers ITS-
Plant1 (TCCGTATGTGAACCTGCGG, forward primer)
and ITS-Gym2 (GGGGAATCCTGGTTAGTTC, reverse
primer) were used. For C. libani, these generate an
amplicon of approximately 350 bp in length. For the
ITS2 region, the primers, ITS-Gym3 (GCACCGATGAA-
GAATGTAGC, forward primer) and ITS-Plant4B
(GGGGAATCCTGGTTAGTTTC, reverse primer) were
utilized. This would generate an amplicon of approxi-
mately 430 bp in length.
The initial PCR reaction mixtures consisted of 25
pmol of each primer (primer pairs prITS4/prITS5, ITS-
Plant1/ITS-Gym2, and ITSGym3/ITS-Plant4B, all were
used in separate reactions), 10 mM dNTP (equimolar
amounts of each dNTP), 1.5 mM MgCl2, 5 units/µl Taq
DNA polymerase (GeneAmp PCR Amplification Kit,
Applied Biosystems, Foster, CA) and 1 µl of each DNA
solution. Several different temperature programs were
used in the PCR steps for this study using a program-
mable thermal cycler (Mastercycler, Eppendorf, New
York, NY). For the initial amplifications, the following
temperature regime was used: 5 minutes at 95°C, then
50 cycles of 1 minute at 95°C, 2 minutes at 45°C, a
ramp of 1°C per 8 seconds to 72 °C and 2 minutes at
72 °C. This was followed by 10 minutes at 72 °C. The
PCR amplified products then were stored at –20 °C until
needed. A 5 µl aliquot of each PCR reaction was subject-
ed to electrophoresis on an agarose gel (described
above), except the gels were 1%, and the running time
was 2.5 hours.
After examination of the gel photograph, the amplified
PCR products were used as templates for a second
round of PCR amplification (termed the first round of
reamplification), utilizing 2 µl of the initial PCR reac-
tions, and placed into new PCR reaction mixtures (as
above). The first reamplification temperature regime
was: 5 minutes at 95°C, then 50 cycles of 1 minute at
95°C, 4 minutes at 55°C, a ramp of 1°C per 8 seconds to
72°C and 4 minutes at 72°C. This was followed by 10
minutes at 72°C. All three primer sets were used. After
examination of the gel photograph, the amplified PCR
products were used as templates for another set of PCR
reamplification reactions, utilizing 2 µl of the PCR reac-
tions from the first set of reamplifications. The second
round used four temperature regimes that varied in
their annealing temperatures. The first was: 5 minutes
at 95°C, then 50 cycles of 1 minute at 95°C, 4 minutes
at 45°C, a ramp of 1°C per 8 seconds to 72 °C and 4 min-
utes at 72°C; followed by 10 minutes at 72°C. The sec-
ond, third, and fourth programs used 50, 55, and 57°C
(respectively) as the annealing temperature, rather than
45°C, but otherwise were the same as the first program.
Only two of the primer pairs were used (ITS-Plant1/ITS-
Gym2, and ITSGym3/ITS-Plant4B).
DNA sequencing
Single amplification bands from each successful
amplification and reamplification reaction were purified
and isolated on 1.5% low melting point agarose gels
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
58
(NuSive GTG, FMC Bio Products, Rockland, ME, USA)
using a freeze and centrifugation method (TAUTZ and
RENZ, 1983) to separate the DNA from the agarose. The
sequences from isolated amplified bands were deter-
mined employing the above primers using a Prism
Ready Dye Deoxy Terminator Cycle Sequencing Kit
(Applied Biosystems, Inc., Foster, CA) and an Automat-
ed DNA Sequencer (Applied Biosystems, Inc., Foster,
CA; Model 310).
BLAST searches and phylogenetic analyses
Sequences were compared with nineteen conifer rDNA
ITS sequences from NCBI (National Center for
Biolotechnology Information) (Larix gmelinii – GenBank
accession number AY523449, L. potaninii – AF538062,
Metasequoia glyptstroboides – AF387530, Picea glauca –
AF136621, Pinus echinata – AF367378, Pinus krempfii –
AF305061, Pinus manticola – AY619694, Pinus massoni-
ana – AF305063, Pinus palustris – AF305362, Pinus
taeda – AF367379, Pinus techunumanii – AF200524,
Pseudotsuga menziesii – AF041353, Sequoia semper-
virens – AF387521, Ta xus brevifolia = AF259295, Thuja
occidentalis = AY846284 and Tsuga mertensiana –
AY570231). Additionally, we determined the rDNA ITS
sequences from three Taurus cedar trees, using megaga-
metophyte tissue. BLAST searches of international
sequence databases (using Baylor College of Medicine
internet site, gapped BLASTN search, http://search-
launcher.bcm.tmc.edu/seq-search/nucleic_acid-
search.html; and the NCBI site, BLASTN search,
http://www.ncbi.nlm.nih.gov/) were used to determine
the most closely related species to each sequence.
Sequences from ancient Taurus cedar wood, Taurus
cedar megagametophytes (fresh tissues), and selected
the sequences (above) were aligned using ClustalW 1.8,
using the Baylor College of Medicine internet site
(http://searchlauncher.bcm.tmc.edu/multi-align/multi-
align.html). Phylogenetic analyses were conducted using
PAUP (Phylogenetic Analysis using Parsimony, Version
4.0b; SWOFFORD, 2000). The phylogram was generated
using maximum parsimony, heuristic search option,
with gaps scored as a fifth base. Maximum parsimony
analysis was also performed on the same data set scor-
ing the gaps as missing data.
Table 1. – Results from PCR amplification attempts, amplification totals, and sequencing results. In all cases the
results are shown with two numbers separated by a slash. The number on the right indicates the total number of
attempts (PCR, sequencing, or BLAST searches), while the number on the left indicates the number of successes for
each. N/A indicates no attempts.
aFor the initial amplifications, 1 µl of each DNA solution was used in 25 µl PCR reaction mixtures. For each of the
reamplification attempts, 2 µl of the previous amplification reactions were used in subsequent 25 µl PCR reaction mix-
tures. Annealing temperatures are noted in parentheses.
bPrimer pair 4/5 is prITS4 and prITS5, pair 1/2 is ITS-Plant1 and ITS-Gym2, and pair 3/4 is ITSGym3 and
ITS-Plant4B. Results indicate the number of reactions that generated bands visualized on agarose gels
(successes /attempts).
cThe annealing temperature for the 3/4 pair was 55°C, and that for the 1/2 pair was 57°C.
dPCR amplification bands were eluted from low melting point agarose gels (as described in the Materials and Methods).
In several cases, more than one band was evident. In these cases, each band was eluted separately. Each of the eluted
DNAs was subjected to PCR cycle sequencing and analysis on an ABI 310 automated DNA sequencer. The number of
readable sequences over the total number of sequencing reactions analyzed in the automated DNA sequencer is indi-
cated.
eThe number of sequences that were most similar to ITS rDNA sequences from gymnosperms is indicated compared to
the total number of sequences in BLASTN searches. The other sequences were closest to the following organism
sequences at NCBI: Gordion samples – ten were closest to human sequences, one was closest to fungal sequences, and
ten had no significant similarity with any other sequence; Al-Aksa samples – twelve were closest to human sequences,
nine were closest to fungal sequences, one was closest to an algal sequence, and eleven had no significant similarity
with any other sequence.
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
59
Results
The wood from the Gordion site was noticeably softer
than the sample from the Al-Aksa Mosque, and thus
grinding was much easier with the Gordion wood. This
probably is due to extensive degradation of the wood by
soft-rot fungi, described elsewhere (FILLEY et al., 2001).
Although signs of wood degradation were evident, of the
one-hundred sixty-nine attempts to amplify the rDNA
regions (Table 1), twenty-three (or 13.6%) yielded visible
bands on agarose gels (some yielded multiple bands).
From these, fifty-six readable DNA sequences were
obtained, twenty-two from the Gordion wood sample,
and thirty-four from the Al-Aksa wood sample. For the
sequences from the Gordion sample, in BLAST searches,
only one was closest to gymnosperm rDNA ITS
sequences. The others were closest to human sequences
(ten total), fungi (one), or were not similar to any other
sequences (ten). For the sequences from the Al-Aksa
sample, only one was closest to gymnosperm rDNA ITS
sequences. The others were closest to human (twelve),
fungi (nine), algae (one), or were not similar to any other
sequences (eleven).
The total length of accurately readable sequence
determined for the one Gordion wood sample was 181
nucleotides, and included the extreme 5’ end of the 5.8S
gene, and about half of the 3’ end of ITS1. The total
length of accurately readable sequence from the one Al-
Aksa wood sample was 265 nucleotides, and was com-
posed of much of the 5.8S gene and part of the 5’ end of
ITS2. These sequences were aligned with analogous
sequences from several species in the Coniferales (as
indicated in the Materials and Methods). Phylogenetic
analyses were performed using 479 nucleotide positions,
which included the 3’ half of ITS1, the entire 5.8S gene,
and the 5’ half of ITS2. Phylogenetic analyses with or
without gaps considered as a fifth base yielded similar
results (Figure 3). The branches with the two sequences
from ancient wood were always in the same position,
Figure 3. – Maximum parsimony tree based on sequences of rDNA ITS regions from selected
conifers, including DNA from modern C. libani (ZK AS 31704 4, ZK IK 31704 4, and ZK IK 31704 12)
and from the Gordion and Al-Aksa cedar wood specimens. The tree presented is one of nine most
parsimonious trees (1158 steps, CI = 0.6796, HI = 0.3204, RI = 0.6577, RC = 0.4470). All nine trees
are invariant with respect to the placement of the two sequences from the ancient wood. The
analyses were based on comparison 479 nucleotide positions for each of the 21 taxa in the data
matrix. The Gordion ancient wood sequence consisted of 181 nucleotides (plus 26 gap positions). The
Al-Aksa ancient wood sequence consisted of 265 nucleotides (plus 75 gap positions). Gaps were
included as a fifth base in the analysis that produced this tree. Analysis that considered gaps as
missing data resulted in seven most parsimonious trees (791 steps, CI = 0.6991, HI = 0.3009,
RI = 0.6801, RC = 0.4755). The placement of the Gordion and Al-Aksa sequences was exactly the
same for all sixteen trees.
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
60
nearest to the clade with the recent C. libani rDNA ITS
sequences. Maximum parsimony analyses with gaps
counted as a fifth base resulted in nine most parsimo-
nious trees, with length of 1158, consistency index (CI) =
0.6796, homoplasy index (HI) = 0.3204, retention index
(RI) = 0.6577, and rescaled retention index (RC) =
0.4470. Bootstrap analyses indicated that the C. libani
branch had 100% support and the branch with the two
ancient specimens had 97% support. When gaps were
considered as missing data, seven most parsimonious
trees resulted, with length of 781, CI = 0.6991, HI =
0.3009, RI = 0.6801, and RC = 0.4755. Support for the
two branches was 100% and 69%, respectively.
While the two sequences appear to be authentic Tau-
rus cedar sequences, the phylogenetic analysis (Figure
3) placed them distant from the Taurus cedar sequences
derived from currently growing plants in Afyon-Sul-
tandag and Isparta-Kapıdag (Figure 1). The sequence
data from both samples indicated that the DNA tem-
plates in the ancient Taurus cedar wood were damaged,
as evidenced by the low yield of sequences, and within
the readable sequences there were regions where
nucleotide determinations were difficult because the sig-
nals were weak.
Discussion
The results of the present study indicate that the
characterization of DNA from cedar wood as old as 2700
years old is possible. Further studies on this and other
ancient wood specimens may help to answer some of the
questions such as how the past distribution of the
species compares with the present. Such studies could
be useful in elucidating the origins of ancient wood spec-
imens in these and other archaeological sites. Further-
more, the alleles that may have been lost in the past
might be recoverable, although degradation will limit
such recovery efforts. The potential of wood for molecu-
lar genetic investigations was explored by DEGUILLOUX
et al. (2002) and it was concluded that younger tissues
from the outer part of the trunk were more likely to
yield DNA for PCR amplification than were samples
from older wood from the heartwood. Furthermore,
there was a better chance of obtaining longer amplified
DNA fragments compared with the older samples. For
most wood samples, the extracted DNA was degraded.
Genomic regions that gave the best results were those of
small size and present in high copy number, such as
chloroplast, mitochondrial, or repeated nuclear
sequences (GUGERLI et al., 2005). Nevertheless, TANI et
al. (2003) studied nucleotide diversity of several nuclear
genes which was estimated from DNA amplified from
fossil heartwood of six trees of Japanese cedar (Cryp-
tomeria japonica), buried for about 3600 years by com-
paring DNA sequences of nearby living trees. The previ-
ous aDNA studies from fossils and arcahealogical
remains of wood agree that authenticity of aDNA and
its degradation are the major concerns (http://
193.51.111.211/fossilva/index.htm; GUGERLI et al., 2005;
TANI et al., 2003; DEGUILLOUX et al., 2002).
The extent of contamination and degradation were
expected in this study. We and others have noted and
documented molecular damage and contamination in
ancient specimens (GUGERLI et al., 2005; PÄÄBO et al.,
2004; ROGERS et al., 1985, 1989, 2004, 2005; WAYNE et
al., 1999). The Gordion site and the Al-Aksa Mosque are
frequented by human visitors. Additionally, the samples
were all handled by an unknown number of people.
Thus, it is not surprising that most sequences obtained
were of human origin. In the future, researchers exca-
vating sites of antiquity might perform some of the sam-
pling using sterile methods to collect and store the speci-
mens. This probably would not reduce all forms of conta-
mination, but should reduce the amount of human DNA
contamination. Contamination from fungi also is of
major concern. Fungi are ubiquitous and readily travel
in the air, soil, and water. Thus, fungal contamination at
these sites is expected.
In this study, we saw evidence of degradation in two
ways. First, the Gordion wood sample was soft, indicat-
ing the action of fungi (or analogous processes). Second,
the DNA from each of the wood specimens was evident
from the low frequency of PCR amplification and the
weak DNA sequences. Degradation of the DNA in the
wood samples probably originated from at least three
sources. First, as wood matures, the amount of DNA and
RNA in the cells is reduced as these nucleic acids are
degraded in the natural maturation processes. Cambial
cells would be expected to retain more of a full comple-
ment of DNA and RNA sequences, which has been indi-
cated in other wood studies (DEGUILLOUX et al., 2002).
Since our samples consisted of mature wood without
cambium, part of the degradation might have originated
from the wood maturation process.
Second, the more rapidly the tissues dry, the less DNA
degradation will occur (ROGERS et al., 1989). Primarily,
this is because enzymes and other degradative processes
require hydration of the nucleic acids (LINDAHL,1993,
1997; ROGERS et al., 1989; 2004, 2005). Therefore, if the
wood dried rapidly and remained dehydrated, the nucle-
ic acids would have a higher probability of remaining
undamaged. Wood in the Midas Tumulus is buried
under tens of meters of soil. Although the site is semi-
arid, some rain water probably has seeped through the
soil from above. Additionally, the interior of the Tumulus
can be much cooler in the summer than outside. Thus, it
is possible that moisture from the air condenses inside
the Tumulus during warmer months. At Al-Aksa, it
appears that the preservation is somewhat better, since
this wood remained relatively hard. The constant supply
of dry air through the building, and protection from pre-
cipitation may have helped to preserve this wood.
Third, the cedar wood in the Midas Tumulus previous-
ly was examined microscopically (FILLEY et al., 2001)
and was infected with soft-rot fungi. The fungi had
degraded the wood to a significant degree. The degrada-
tion of the wood structure indicates that the probability
of nucleic acid degradation is higher than it would be
otherwise. Also, the presence of a large population of
fungi indicates that the wood was likely hydrated, at
least to an extent. The presence of water would increase
degradation via biotic as well as abiotic mechanisms. As
stated above, the Gordion wood samples were soft com-
pared to the Al-Aksa samples, which is consistent with
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
61
degradation by fungi. The hardness of the Al-Aksa sam-
ple is encouraging, since it indicates that this wood may
have less degradation both from fungi and the results of
hydration.
The large distances between the three Taurus cedar
sequences from fresh tissues and the ancient Gordion
and Al-Aksa Taurus cedar wood sequences seen in the
phylogram may be caused by one of several possibilities.
First, the DNA within the samples could be significantly
degraded. This is likely, since the amplification and
sequencing were difficult and inconsistent. Second, the
sequences could be from contaminants, such as from
species close to Taurus cedar. Other varieties of C. libani
exist in the Middle East, and it is possible (although
unlikely) that pollen or other contaminating cells
became embedded in the wood samples. Third, the wood
might be from an extinct different species of Cedrus.
Since no preserved wood or fossil evidence exists for this
possibility, it is unlikely that this is the case. Fourth,
the genetic diversity of Taurus cedar has decreased dur-
ing the past 2700 years. Long ago, the natural distribu-
tion of Taurus cedar may have been continuous from
Afyon-Sultandag to the Gordion site, but these natural
Taurus cedar forests were lost due to the many years of
human activities. Similarly, the forests were continuous
through other parts of the Middle East, including Israel.
Again, anthropogenic deforestation has severely limited
the extent of the C. libani forests. It is likely that the
genetic diversity of the species concurrently has been
reduced. Since the sequence distances are large in the
phylogenetic analyses between the ancient wood sam-
ples and the recent C. libani specimens, it is likely that
more than one of the above possibilities is responsible
for the sequence variation.
These results indicate that the wood can be utilized in
aDNA studies, but that degradation is likely, and the
specimens may be heavily contaminated through human
contact and microbial colonization (GUGERLI et al.,
2005). For future aDNA studies of archaeological wood
specimens, sample collection and preparation should be
carried out in aseptic conditions in order to avoid fur-
ther contamination. Dozens of DNA extractions and 169
PCR amplification reactions were attempted, from
which 56 sequences were obtained. Only two of the
resulting sequences were consistent with being from
cedar. However, since the samples are destroyed in the
process, we were restricted from sampling the wood
from many of the more suitable portions of the cedar
logs at the Gordion site. For future aDNA studies using
wood, samples taken from the areas close to bark, knots
or wounded areas would be more useful since these
areas are the most likely places that some intact
parenchyma cells (containing cytoplasm and nuclei)
could be found included among the xylem tissues during
wood formation. Additionally, the potential for contami-
nation would be reduced if specimens that are distant
from areas of high visitor traffic could be assayed. If we
had flexibility in sampling from the ancient cedar logs in
the Tumulus of King Midas in the Gordion site (Figure
2), the DNA amplification would be more likely to yield
DNA sequences from the cedar wood rather than from
contaminating organisms.
Acknowledgements
We express our gratitude to the Fulbright Scholar Pro-
gram for providing Prof. Z. KAYA a research grant at
SUNY ESF (Fulbright Research Grant # 24612) that
made possible our collaboration on genetic characteriza-
tion of ancient Taurus cedar wood samples from Gordion.
We are also grateful to the Anatolian Civilization Muse-
um Authorities (Assistant Director HI
.KMET DENI
.ZLI
.and
Archaeologist MEHMET AKALI
.N) for providing us the Gor-
dion cedar wood samples, and to Professor S. LEV-YADUN
of Haifa University, Israel for providing the Al-Aksa
mosque cedar wood specimens. We thank GANG ZHANG,
SEUNG-GEUK SHIN, and VINCENT THERAISNATHAN for their
help in preparation of parts of the manuscript.
References
AYTUG, B. (1988): Le mobilier funeraire du roi Midas I.
Pact 22: 357–368.
BOYDAK,M. (2003): Regeneration of Lebanon cedar
(Cedrus libani A. Rich.) on karstic lands in Turkey. For-
est Ecol. Manag. 178: 231–243.
CANO,R. J. and M. K. BORUCKI (1995): Revival and identi-
fication of bacterial spores in 25- and 40-million-year-
old Dominican amber. Science 268: 1060–1064.
DEGUILLOUX, M.-F., M.-H. PEMONG and R. J. PETIT (2002):
Novel perspectives in wood certification and forensics:
dry wood as a source of DNA. Proc. Royal Soc. Lond.
Ser. B 269: 1039–1046.
DEGUILLOUX, M.-F., M.-H. PEMONGE and R. J. PETIT (2004):
DNA-based control of oak wood geographic origin in the
context of the cooperage industry. Annals Forest Sci. 61:
97–104.
DEGUILLOUX, M.-F., M.-H. PEMONGE, L. BERTEL, A. KREMER
and R. J. PETIT (2003): Checking the geographical origin
of oak wood: molecular and statistical tools. Molec. Ecol.
12: 1629-1636.
DUMOLIN-LAPÈGUE, S., M.-H. PEMONGE, L. GIELLY,
P. TABERLET and R. J. PETIT (1999): Amplification of oak
DNA from ancient and modern wood. Molec. Ecol. 8:
2137–2140.
FILLEY, T. R., R. A. BLANCHETTE, E. SIMPSON and M. L.
FOGEL (2001): Nitrogen cycling by wood decomposing
soft-rot fungi in the “King Midas tomb,” Gordion,
Turkey. Proc. Natl. Acad. Sci USA 98: 13346–13350.
GUGERLI, F., F. PARDUCCI, L. and R. J. PETIT (2005):
Ancient plant DNA: review and prospects. New Phytol.
166: 409–418.
HOFREITER, M., D. SERRE, H. N. POINAR, M. KUCH and
S. PÄÄBO (2001): Ancient DNA. Nature Rev. Genet. 2:
353–359.
IÇGEN, Y. (2002): Genetic composition of Pinus brutia Ten.
forests established with seeds from seed stands and
orchards determined by using DNA markers. Ph.D. Dis-
sertation, Middle East Technical University, Depart-
ment of Biological Sciences, Ankara, Turkey.
KANDEDMIR, G., I. KANDEMIR and Z. KAYA (2004): The Pat-
tern of Genetic Variation in Turkish red pine (Pinus
brutia Ten.) seed stands as determined by RAPD mark-
ers. Silvae Genetica 53: 169–175.
KAYA, Z. and D. B. NEALE (1995): Linkage map based on
randomly amplified polymorphic DNA (RAPD) markers
in Pinus brutia Ten. Silvae Genetica 44: 110–116.
KAYIHAN, G. (2000): The genetic structure of Cedrus libani
populations determined by DNA markers. MS Thesis,
Middle East Technical University, Department of Biolo-
gy, Ankara, Turkey.
KIM, S., D. E. SOLTIS, P. S. SOLTIS and Y. SUH (2004): DNA
sequences from Miocene fossils: an ndhF sequence of
Rogers et. al.·Silvae Genetica (2006) 55-2, 54-62
DOI:10.1515/sg-2006-0009
edited by Thünen Institute of Forest Genetics
62
Magnolia latahensis (Magnoliaceae) and an rbcL
sequence of Persea pseudocarolinensis (Lauraceae).
Amer. J. Bot. 91: 615–620.
LINDAHL, T. (1993): Instability and decay of the primary
structure of DNA. Nature 362: 709–715.
LINDAHL, T. (1997): Facts and artifacts of ancient DNA.
Cell 90: 1–3.
LIPSCHITZ, N., G. BIGER, G. BONANI and W. WOOLPHI
(1997): Comparative dating methods: Botanical identifi-
cation and 14C dating of carved panels and beams from
the Al-Aksa Mosque in Jerusalem. J. Archaeolog. Sci.
24: 1045–1050.
MA, L. J., S. O. ROGERS, C. CATRANIS and W. T. STARMER
(2000): Detection and characterization of ancient fungi
entrapped in glacial ice. Mycologia 92: 286–295.
MILLER, N. F. (1999): Interpreting ancient environment
and patterns of land use: Seeds, charcoal and archaeo-
logical context. TUBA-AR II: 15–29.
MILLER, N. F. (1998): The macrobotanical evidence for
vegetation in the Near East, c 18000/16 000 BC to 4000
BC. Plaeorient. 23/2: 197–207.
PÄÄBO, S., H. POINAR, D. SERRE, V. JAENICKE-DESPRES,
J. HEBLER, N. ROHLAND, M. KUCH, J. KRAUSE, L. VIGI-
LANT and M. HOFREITER (2004): Genetic analyses from
ancient DNA. Ann. Rev. Genet. 38: 645–679.
PÄÄBO, S. (1993): Ancient DNA. Sci. Amer. 267: 60–66.
PÄÄBO, S. (1988): Molecular clonning of Ancient Egyptian
Mummy DNA. Nature 314: 644–645.
PARDUCCI, L. and R. J. PETIT (2004): Ancient DNA –
unlocking plants’ fossil secrets. New Phytol. 161:
335–339.
ROGERS, S. O. (1994): Phylogenetic and taxonomic infor-
mation from herbarium and mummified DNA,
pp. 47–67 in DNA Utilization, Intellectual Property
and Fossil DNA, edited by R. P. ADAMS, J. MILLER,
E. GOLENBERG, and J. E. ADAMS, Missouri Botanical
Gardens Press, St. Louis/MO.
ROGERS, S. O. and A. J. BENDICH (1985): Extraction of
DNA from milligram amounts of fresh, herbarium and
mummified plant tissues. Plant Mol. Biol. 5: 69–76.
ROGERS, S. O. and A. J. BENDICH (1988): Extraction
of DNA from plant tissues, pp. A6:1–10. In: Plant Molec-
ular Biology Manual, edited by S. B. GELVIN and
R. A. SCHILPEROORT, Kluwer Academic Publishers,
Boston/MA.
ROGERS, S. O. and A. J. BENDICH (1994): Extraction of
total cellular DNA from plants, algae, and fungi, pp.
D1:1–8. In: Plant Molecular Biology Manual, 2nd ed.,
edited by S. B. GELVIN and R. A. SCHILPEROORT, Kluwer
Academic Publishers, Dordrecht, The Netherlands.
ROGERS, S. O., L. MA, Y. ZHAO, C. M. CATRANIS,
W. T. STARMER and J. D. CASTELLO (2005): Recommenda-
tions for elimination of contaminants and authentica-
tion of isolates from ancient ice cores, pp. 5–21. In: Life
in Ancient Ice, edited by J. D. CASTELLO and S. O.
ROGERS, Princeton University Press, Princeton/NJ.
ROGERS, S. O., S. REHNER, C. BLEDSOE, G. J. MUELLER and
J. F. AMMIRATI (1989): Extraction of DNA from Basid-
iomycetes for ribosomal DNA hybridizations. Can. J.
Bot. 67: 1235–1243.
ROGERS, S. O., V. THERAISNATHAN, L. J. MA, Y. ZHAO,
G. ZHANG, S. G. SHIN, J. D. CASTELLO and W. T. STARMER
(2004): Comparison of protocols to decontaminate envi-
ronmental ice samples for biological and molecular
examinations. Appl. Environ. Microbiol. 70: 2540–2544.
SAVOLAINEN, V., P. CUÉNOUD, R. SPICHIGER, M. D. P. MAR-
TINEZ, M. CRÈVECOEUR and J.-F. MANEN (1995): The use
of herbarium specimens in DNA phylogenetics – evalua-
tion and improvement. Plant Syst. Evol. 197: 87–98.
SUYAMA, T., K. KAWAMURO, I. KINOSHITA, K. YOSHIMURA,
Y. TSUMURA and H. TAKAHARA (1996): DNA sequence
from a fossil pollen of Abies spp. from Pleistocene peat,
in http://www.madsci.org/ (University of Washington,
Medical School): The MadScientist Network.
SWOFFORD, D. L. (2000): PAUP: Phylogenetic Analysis
Using Parsimony, Version 4.03, Sinauer Associates,
Sunderland/MA.
TANI, N., Y. TSUMURA and H. SATO (2003): Nuclear gene
sequences and DNA variation of Cryptomeria japonica
samples from the postglacial period. Molec. Ecol. 12:
859–868.
TAUTZ, D. and M. RENZ (1983): An optimized freeze-
squeeze method for recovery of DNA from agarose gels.
Anal. Biochem. 132: 14–19.
WAYNE, R. K., J. A. LEONARD and A. COOPER (1999): Full of
sound and fury: The recent history of ancient DNA.
Ann. Rev. Ecol. Syst. 30: 457–477.
Abstract
By adding a penalty to a candidate’s breeding value
for its relationship with the selected individuals, two
indexes were constructed as criteria for stepwise selec-
tion of superior individuals from populations with a
hierarchical structure. The relationship was expressed
in terms of either family contribution or group coances-
try. One of the indexes was derived from an optimal
Stepwise Penalty Index Selection from Populations
with a Hierarchical Structure
By R.-P. WEI1), *)and D. LINDGREN2)
(Received 5th August 2005)
1) South China Agricultural University, Wushan, Guangzhou
510642, China, and Sino-Forest Corporation, 3815-29, 38/F.,
Sun Hung Kai Centre, 30 Harbour Road, Wanchai, Hong Kong.
2) Department of Forest Genetics and Plant Physiology, Swedish
University of Agricultural Sciences, S-901 83 Umeå, Sweden.
*) Correspondence author and address: RUN-PENG WEI, Sino-
Forest Corporation, 3815-29, 38/F., Sun Hung Kai Centre,
30 Harbour Road, Wanchai, Hong Kong. Tel.: +852-2514 2124.
Fax +852-2877 0062. Email: runpeng-wei@sinoforest.com
Wei et. al.·Silvae Genetica (2006) 55-2, 62-70
DOI:10.1515/sg-2006-0010
edited by Thünen Institute of Forest Genetics