"TV-"'6 m

" "~d\\

20002

 

LIBRARY
Michigan State
University

 

This is to certify that the
thesis entitled

GENETIC ANALYSIS OF THE GREAT CIRCLE FROM THE
TUMULUS AT KAMENICA, ALBANIA

presented by

LINDSEY NICOLE MURRAY

has been accepted towards fulfillment
of the requirements for the

MS. degree in Forensic Science

 

 

 

@figfl

 

Major Professor’s Signature
Sf/c)*/0 6

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

—.g.-.----.-.----------—-—

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

 

 

 

DATE DUE DATE DUE DATE DUE
AUG 0 5 2007
‘ sgp 2 0' 2009
, .27.: ‘3 I] .
? 052809
use 5) ‘ tin-in
H31}? g ‘W—r. JSU
2’) 4 '0 410

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/CIRC/DateDue.indd-p.1

 

GENETIC ANALYSIS OF THE GREAT CIRCLE FROM THE TUMULUS AT
KAMENICA, ALBANIA
By

Lindsey Nicole Murray

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTERS OF SCIENCE
School of Criminal Justice

2006

ABSTRACT
GENETIC ANALYSIS OF THE GREAT CIRCLE FROM THE TUMULUS AT
KAMENICA, ALBANIA
By
Lindsey Nicole Murray
Skeletal remains of 21 individuals dating to the 13m — 9th century B.C. and located

in the Great Circle section of a tumulus (ancient burial mound) in Kamenica, Albania,
were analyzed. Matemally inherited mitochondrial DNA (mtDNA) was examined to
determine maternal relatedness of these individuals for the purpose of understanding
ancient burial patterns and genetic continuity. MtDNA sequences were obtained for 18
of the 21 individuals analyzed. Of these, 14 displayed sequences that could be
distinguished from each other, indicating that they were not maternally related. MtDNA
sequences from 20 individuals currently living in Kamenica who self-proclaimed their
ancestry to be from this village were also obtained. All of these individuals could be
distinguished from each other, indicating that the residents currently in Kamenica are
genetically diverse, and that they were not direct (maternal) descendants of the ancient
peoples. European haplogroup studies were also performed on the ancient individuals to
determine whether those in the Great Circle originated from a common region of the
world. Hypotheses for possible burial patterns and social structure were made based on
these genetic analyses. Additionally, the results obtained were compared to previous
studies of the tumulus at Kamenica to examine how burial pattern and populations

throughout the centuries have changed.

ACKNOWLEDGEMENTS

I would first like to thank Dr. David Foran, who provided advice and support
throughout the research and writing aspects of the project. Thank you to Dr. Todd
Fenton for all of his input on anthropological issues and also for making the project
possible by allowing bone samples to be collected for DNA analysis. Thank you to Dr.
Steve Dow for his comments and review of this manuscript. Finally, thank you to the
individuals who helped with research in the laboratory, provided data, and assisted in the

collection of samples: Sara Jubelier, Stephanie Rennick, and Virginia Clemmer.

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. vi
LIST OF FIGURES ................................................................................ vii
INTRODUCTION ................................................................................. 1
Albanian Demographics ....................................................... -. ........... 2
Ancient Illyria ............................................................................... 2
Burial Customs ............................................................................. 4
The Tumulus at Kamenica ................................................................ 5
Genetic Analysis of Ancient Remains ................................................... 9
Previous Studies on Albanian Diversity ............................................... 11
Haplogroup Studies ...................................................................... 11
Obstacles Associated with Ancient DNA Analysis .................................. 13
The Goal of this Project .................................................................. 15
MATERIALS AND METHODS ................................................................ 16
Burials in the Tumulus at Kamenica ................................................... 16
Bone Selection .................................................................... 26
Bone Preparation ................................................................. 27
DNA Extraction .................................................................. 27
DNA Amplification ............................................................. 28
Buccal Swabs from Current Inhabitants of Kamenica ............................... 30
DNA Extraction .................................................................. 30
DNA Amplification .............................................................. 31
DNA Sequencing and Analysis ......................................................... 31
Single Nucleotide Polymorphism Analysis ........................................... 33
RESULTS .......................................................................................... 36
Bone Sampling and Preparation ......................................................... 36
PCR Amplification ........................................................................ 36
Analysis of mtDNA Sequences from Bone ........................................... 38
HVI Sequences ............................................................................ 39
HVII Sequences ............................................................................ 42
Ambiguous Results in mtDNA sequence Analysis ................................... 43
Analysis of mtDNA sequences from Buccal Swabs ................................. 46
Haplogroup Assignment ................................................................. 47
DISCUSSION ..................................................................................... 5]
Bone Preparation and DNA Amplification Success ................................. 51
Degradation and Mutations that Cause Ambiguity in Sequences .................. 53
Sequencing Success and Failure ........................................................ 55
Maternal Relatedness within the Great Circle ........................................ 56
Haplogroup Designation ................................................................. 58

iv

Genetic Comparison of Modern Day and Ancient Inhabitants of Kamenica . .....59

Comparison of Results to Previous Studies at Kamenica ............................ 60
Possible Burial Patterns in the Tumulus ............................................... 61
Summary ................................................................................... 62
APPENDICES ...................................................................................... 63
BIBLIOGRAPHY .................................................................................. 72

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Table 8:

Table 9:

LIST OF TABLES

Bones Processed for DNA ............................................................. 26
HVI and HVII Primer Pairs for Bone Samples 29
HVI and HVII Primer Pair for Buccal Swabs ....................................... 31
SNP Primer Pairs for Haplogroup H ................................................... 33
Bones that Successfully Sequenced ................................................... 38
HVI consensus sequences .............................................................. 41
HVII Consensus Sequences ............................................................ 43
Haplogroup H polymorphisms ........................................................ 47
Results for SNP Primer Extension .................................................... 49

vi

LIST OF FIGURES

Figure 1: Map Depicting the Great Circle and the Monumental Structures

of the Tumulus .............................................................................. 6
Figure 2: Map Depicting Graves Above the Great Circle and Monumental
Structures of the Tumulus ................................................................. 7
Figure 3: Photograph of the Great Circle .......................................................... 8
Figure 4: Photograph of the Monumental Structure 3 .......................................... 9
Figure 5: Photograph of Graves ................................................................... 16
a) Individual 347 .......................................................................... 16
b) Individual 349 ........................................................................... 17
c) Individual 351 ........................................................................... 17
d) Individual 356 ........................................................................... 18
e) Individual 357 ........................................................................... 18
0 Individual 358 ........................................................................... 19
g) Individual 367 .......................................................................... 19
h) Individual 370 .......................................................................... 20
i) Individual 373 ........................................................................... 20
j) Individuals 374 and 377 ............................................................... 21
k) Individuals 376 ......................................................................... 21
1) Individual 381 ........................................................................... 22
m) Individual 382 .......................................................................... 22
n) Individual 383 ........................................................................... 23
0) Individual 389 .......................................................................... 23

vii

p) Individual 390 ........................................................................ 24

q) Individual 391 ......................................................................... 24

r) Individual 392 and 393 ............................................................... 25

5) Individual 395 .......................................................................... 25
Figure 6: Semi-nested PCR ....................................................................... 37
Figure 7: Ambiguous bases from two different bones ........................................ 44
Figure 8: Ambiguous bases in one sequence .................................................... 44
Figure 9: Examples of Pull-Up .................................................................. 45
Figure 10: Examples of SNP Peaks .............................................................. 49

viii

Introduction

DNA analysis has become a vital tool for human identification and can be used to
link people to the scene of a crime, to link relatives to fallen soldiers, or to solve
questions of relatedness. The latter topic is the focus of this thesis, in which forensic
anthropologist and forensic molecular biologists work together in an effort to learn more
about the prehistoric people of Albania. Studying skeletal remains to determine age, sex,
population ancestry, and stature are the main objectives for physical anthropologists.
Forensic anthropologists further study skeletal remains to determine the trauma or
diseases to which they succumbed, and thus the cause of death. Molecular biologists can
contribute a great deal to these investigations by using DNA analysis to determine
ancestry and genetic relatedness of skeletal remains—information that cannot be
uncovered by anthropology alone.

There are several areas of the world that have not been thoroughly studied. Until
recently this was true of the small Mediterranean country of Albania. When the
communist government took over in 1944 relations among other nations, notably US and
Britain, deteriorated and the Albanian government prohibited entrance into the country
(Zickel and Iwaskiw 1994). As a result, research opportunities in Albania were limited;
therefore knowledge about the history and culture of Albania was also limited. In the
1950s and 60s several archaeological endeavors began; however these were restricted
only to Albanian researchers (Hammond 1967).

In 1990 the communist era ended, removing the entrance restrictions along with
it. With scientists from other countries allowed into Albania, new research ideas and

opportunities came about—including extensive excavation projects throughout the

country. In 2002 forensic anthropologists from Michigan State University became
involved with one of the large excavation projects, performing skeletal analyses on the
remains of the tumulus (burial mound) at Kamenica. The following year forensic
biologists fiom MSU joined the team to help in solving questions about Kamenica’s
history including genetic relatedness of the individuals buried there, their migration, and

the ancient burial patterns.

Albanian Demographics

Albania, approximately the size of Maryland, is the third smallest country in
Europe. Surrounding it is Montenegro and Serbia to the north, Macedonia to the east,
Greece to the south, and the Adriatic Sea to the west. The population size of Albania is
increasing, standing at about 3.54 million is 2005 (The World Almanac 2005). The
increase is due to a higher birth rate and a lower mortality rate rather than an increase in
emigration fiom other countries (Frucht 2005). The ethnic distribution is 95 % Albanian,
3% Greek, and 2% all others including Macedonian, Montenegrin, Serbian, and Roman.
This distribution according to the CIA World Fact Book (2006) has not dramatically
changed since a poll was taken in 1989, before the fall of communism in Albania.
Therefore, it is of interest to learn if the ethnic Albanian population has been consistent
for centuries and whether the current population descended from the Illyrians, a people

believed to have existed as far back as the 13th Century B.C.

Ancient Illyria
The Illyrians were an Indo-European population who appeared in the western part

of the Balkan Peninsula—including Greece, Macedonia and Italy—at the end of the

Bronze Era and the beginning of the Iron Age, around 1000 B.C. (Zickel and Iwaskiw
1994). Historians believe that one of the primary reasons the Illyrians settled in Albania,
specifically the area around the Korce basin in southeastern Albania, and continued to
thrive throughout the Iron Age, was due to the presence of fertile land, two river valleys,
and a large supply of copper and iron ores. These attributes would provide economic
means as well as water resources and adequate land for agriculture (Bejko in press).
During the Dark Ages most of the Illyrians disappeared when invaded by migrating
“barbarians”, except those living in Albania. This population was probably protected
from invasion by the large mountains surrounding the country, as well as a strong tribal
union, possibly allowing them to survive into modern days, maintaining their identity and
[ado-European language (Stipcivic 1977).

Recent archeological findings that date back to the time of Illyria indicate that the
Illyrians in Albania had a more advanced society than Illyrian populations in the Balkans,
Greece and Italy, especially concerning urbanization, social structure, ceramics, and
coinage (Stipcivic 1977). Based on the excavations and other historical finds it is
believed that the Illyrians of Albania were largely influenced by the neighboring Greeks
who founded trading colonies in Illyria (Zickel and Iwaskiw 1994). Excavations fi'om
burial mounds in Albania have provided further evidence of the prehistoric presence of
Illyrians. By combining the information discovered by anthropologists, molecular
biologists, and archaeologists a more defined history of the culture of these ancient
Illyrians and their connection with Greece, Italy, and other surrounding areas can be

established.

Burial Customs

Tumuli, or ancient graves that form large mounds, appeared to be the predominant
form of burial throughout Albania (Stipcivic 1977). There are apparent differences in the
burial patterns used throughout the history of Albania, possibly due to tribal diversity; but
for the most part tumuli are found in small groupings and can be up to 5 meters high and
40 meters in diameter (Bejko in press; Hammond 1967). The number of tumuli graves
generally range from as few as 20 to more than 180 individuals, although some larger
ones appear (Bejko in press).

During the Late Bronze Age and Early Iron Age bodies in the tumuli were placed
in a flexed position on their lefi or right side. Later during the Iron Age, an extended
position on their backs was commonly seen with the arms folded in various ways.
Typically individuals were buried alone; however there have been a few cases in which
two bodies were buried together (Bejko in press; Clemmer 2005). Although not
common, cremations were also detected throughout the tumuli, with individuals either
burned at the site or with ashes spread across the burial ground.

The most common tumulus grave type is that of a simple pit; however some were
more elaborate with wooden structures either on the sides or on the top and bottom of the
individual. Also common in many tumuli was a central grave that was covered or circled
with stones that usually signified the first individual interred in the burial ground. In
addition, there was often a ring of stones surrounding the entire tumulus and stones
placed methodically around the tumulus to indicate the end of use of the burial ground

(Bejko in press; Hammond 1967).

The Tumulus at Kamenica

There are several tumuli in Albania that date to the Late Bronze Age, including
those at Kuc i Zi, Pazhok, and Vajeze (Bejko 1994; Wilkes 1992). Just outside the city of
Korce lies the largest known tumulus in Albania—the tumulus at Karnenica—-thought to
be the largest excavation project in the Balkans (ICAA 2002/3). Skeletal remains from
395 graves have been uncovered, but there is an estimated one thousand or more graves
within the tumulus. Three of the archaeological features in the tumulus include the Great
Circle, Monumental Structure 1.1, and Monumental Structure 3 (Figure 1, Figure 2).
Anthropologists have estimated the remains to be from the late 13th century B.C. to the
6th century B.C.—the earliest being from the Great Circle. The tumulus, with dimensions
of 75m by 45m, includes a great number of graves with different methods of burial.

The most prominent feature of the tumulus is the large circle of stones denoting
the Great Circle, measuring 12.5 — 13 m in diameter (Figure 3). This circle was used as a
burial ground during the 13th to at least the 9th century B.C. The grave of most interest is
burial 395, which appears to be the first and therefore the earliest of the Great Circle,
hence the entire tumulus. It is found close to the center of the Circle and has large
wooden structures enclosing it. In addition, at least 40 other graves are positioned around
it, indicating that this individual may have held some position of importance, possibly as
a family leader, a leader of a tribe, or the first individual in the tumulus. Figure 3 shows
the location of the individual within the Great Circle; a closer photograph of this

individual can be found on page 25 of the Materials and Methods.

Figure 1: Map Depicting the Great Circle and the Monumental Structures of the
Tumulus.

The locations of graves within the tumulus are displayed. The large circle of stones in the
top left corner is the Great Circle, while the other two clusters of stones are the
Monumental Structures. Several burials fall in between these structures.

E
53$

‘1.

   

9
I
a

Figure 2: Map Depicting Graves Above the Great Circle and Monumental

Structures of the Tumulus.
The graves located above these sections of the tumulus are displayed. The structures from

the bottom are still apparent, but lost their fixed pattern as the mound grew.

 

I ‘5 ' if Iv
l“\_;.f‘¥ 'l|7 \
._ _ a. . j. '- . I.
v?" r ,_ if} 45": £73. . 1
rou- ‘ ' a"! 9")“ refit)” “¢. -Ja
. 0 I. V. I"1 ‘QE.’§,‘:‘J‘ . L'y'xg‘l 'Im .
fie . 1’34 3

Figure 3: Photograph of the Great Circle

The Great Circle section of the tumulus With stones outlining it. Note the individual
between the wooden structures in the center, believed to be the earliest member of the
tumulus—individual 395.

 

During the end of the 7“I century B.C. the form of burial in the tumulus changed,
when rock walls were used to define the graves, which were then filled in with stone
(Bejko in press; Figure 4). These often had a full circle of stones or an arch formation, in
which a single individual was placed. Archaeologists hypothesized that these
monumental structures were family graves (ICAA 2002/3). Genetic research by
Stephanie Rennick in 2005 supported this hypothesis by finding that many of these
individuals contained the same DNA sequence in a short region of the mitochondrial

genome.

Figure 4: Photograph of the Monumental Structure 3
Monumental structure 3 displaying the rock walls that were built up around each
individual.

 

Genetic Analysis of Ancient Remains

DNA analysis has a multitude of applications, one of the most important of which
is connecting individuals to their relatives. Within a cell there are two locations that
possess DNA—the mitochondria and the nucleus. Nuclear DNA has two copies, paternal
and maternal, while mitochondrial DNA (mtDNA) has one copy, which is inherited
maternally. Both types can be used to connect individuals to their relatives, but
mitochondrial DNA is the DNA of choice when studying ancient samples.

Mitochondria are responsible for producing ATP, the primary energy source for
the cell. The mitochondrial genome is circular and contains 16,569 bases that consist of

protein coding regions and a non-coding ‘control region’ that regulates replication of the

-9-

molecule. Within the control region are two hypervariable (HV) regions—HVI located
from base 16024 to 16365 and HVII located from base 73 to base 340—that differ widely
among unrelated individuals. In 1981, the entire human mitochondrial genome was
sequenced (Anderson et al.), and this sequence became a reference to which all other
mitochondrial sequences are compared. The control region contains approximately 1100
base pairs and averages about 3% difference among individuals within the two
hypervariable regions (Hummel 2003). These differences, or polymorphisms, have
accumulated over centuries, making the DNA sequences—termed haplotypes—identical
in maternal ancestors, but generally distinct from unrelated individuals. This variability
makes mtDNA a useful tool for comparing relatedness among individuals.

There are at least three other reasons that mtDNA is more useful for studying
ancient DNA than is nuclear DNA. First, mtDNA is haploid and is passed down from the
mother to the child as one haplotype, remaining the same throughout generations; thus
allowing for maternal lineages to be traced. In contrast, nuclear DNA is diploid,
recombining every generation to create DNA that is unique, and therefore is generally not
useful for determining genetic relatedness over time. Second, there is only one nucleus
and hence one DNA copy from each parent in a cell, while there are hundreds, possibly
thousands, of mitochondria in a cell. Because there is a potential for a high copy number
of mtDNA, there is a higher likelihood that it can be recovered from samples that are
very old and degraded. Third, the mitochondrion itself appears to be more robust than
the nucleus, and is able to protect DNA when the nucleus does not (F oran in press).

Using information about maternal relatedness obtained from mtDNA, hypotheses

about an ancient culture can be made, such as the cause of burial patterns within a

-10-

tumulus. For instance, if a familial burial exists then identical mtDNA haplotypes would
be expected, representing siblings and other maternal relatives. If burial pattern were
based on other themes (for example social status, such that the center individual was a
tribal leader, a religious leader, or any other important figure) then different haplotypes

would be expected throughout the tumulus.

Previous Studies on Albanian Diversity

Previous studies of mtDNA have not shown a great deal of genetic difference
between Albanians and other Europeans. In contrast, Albanians speak an Indo-European
language—a branch derived from the Illyrians—that is very different from other Indo-
European language types including Italic, Greek, Germanic, and Balto-Slavic. Belledi et
al. (2000) took a detailed look at this language separation to determine if the difference in
the Albanian language was tied to a difference in genetic structure of European
populations. They found that the language separation was not reflected by genetic
differences. In fact, the modern Albanian population was found to be very similar to the
other 17 European populations studied with regards to the amount of diversity
represented. Out of the 42 HV I sequences obtained from modern Albanians, a total of 36
shared polymorphisms seen in other European haplotypes, indicating that Albanians

today are not significantly different genetically from other Indo-European populations.

Haplogroup Studies

MtDNA can also be used to investigate if a population has changed over time.

Mitochondrial haplogroups—mtDNA types that contain polymorphisms known to be

-11-

specific to different regions of the world—can link individuals to a certain ancestry,
including European, African, Native American, etc. Virtually all Caucasians (European
and US) can be defined by 10 haplogroups (denoted H, I, J, K, M, T, U, V, W, X); the
most common being haplogroup H. People outside of these haplogroups would be non-
Caucasian.

There are two methods by which haplogroups can be determined. Specific
polymorphisms or sets of polymorphisms within the mitochondrial control region can be
examined through DNA sequencing (e.g., Allard et al. 2002). The other looks at a single
site in the mitochondrial genome (single nucleotide polymorphisms, or SNPs (snips)),
that define people from different regions of the world. SNPs can be assayed in a number
of ways—by sequencing a particular region where the polymorphism occurs, by using
restriction fragments that cut or fail to cut the DNA at a specific location based on the
DNA sequence present, or by primer extension—the method of choice in this study.
Primer extension utilizes a small DNA primer that anneals one nucleotide away from the
SNP site of interest. When the primer is extended, the first base incorporated will be
indicative of the haplogroup of that individual (V allone et al. 2004).

By examining SNPs the origin of the ancient inhabitants of Kamenica can
possibly be determined. In addition, by comparing the haplogroups of these ancient
people to modern day residents of this village, population continuity can be tested. If it
can be established that individuals have different haplogroups, it suggests that there has
potentially been influence from other populations throughout the centuries. However, if

individuals share a haplogroup, it indicates that there may have been little influence fi'om

-12-

other populations since the time of this ancient society, and that the current residents of

Kamenica may have descended from this ancient population.

Obstacles Associated with Ancient DNA Analysis

The likelihood of obtaining DNA from ancient samples is dependent on the
amount of DNA existing in the skeletal remains, and the level of DNA degradation that
the skeletal remains have experienced. The rate of degradation increases with the age of
the sample, but more importantly with the environment that it is exposed to, including
high moisture, high temperature, and acidic conditions (Kaestle and Horsburgh 2002).
DNA can bind to hydroxyapatite—a mineral in bone that gives it rigidity—slowing
degradation (O’Rourke 2000). However, microorganisms in soil colonize the skeletal
material and attack this mineral, inducing chemical changes in it; thus contributing to the
decay of DNA (Hummel 2003; Medicine Net, Inc. 2006). Likewise, microorganisms can
degrade DNA directly.

One of the most prevalent forms of post mortem DNA damage is deamination of
cytosine, where there is a loss of an amine group, causing the base to resemble a uracil.
When the DNA is replicated in the laboratory, this uracil will base pair with adenine,
which then base pairs with thymine resulting in an overall cytosine to thymine transition
(Gilbert 2003).

Heteroplasmy is another condition that occurs in mtDNA, leading to ambiguous
sequencing results. Heteroplasmy is the existence of two sequences within an individual.
This can be observed in various ways: as two bases present at one site in a mtDNA

sequence, as two different bases at the same site in two separate tissues, or as a

-13-

combination of these in one individual (Stewart 2001). There are also instances in which
length heteroplasmy can be seen. This generally occurs when an additional base is added
to a stretch of homopolymeric bases (repeated base in a sequence). The number of
insertions/deletions at these stretches can vary within an individual. Heteroplasmy is a
rare event, and typically occurs only at one site in an individual. However, post-mortem
mutations may mimic heteroplasmy at several sites in ancient samples.

When DNA is highly degraded it can be difficult to analyze. The Polymerase
Chain Reaction (PCR) is a method used by molecular biologists to amplify small
amounts of DNA, and is essential for DNA analysis of ancient material. However,
amplifying very small amounts of DNA may not be possible with just one round of PCR.
Increasing the number of cycles in a PCR reaction may help, however at a high cycle
number, spurious DNA amplification can occur. To prevent this, a process called nested
PCR can be used, in which two sequential DNA amplification steps are performed. In
the first round, DNA is amplified using one set of primers. If amplification does not
occur a second round can be performed with different primers, located internal to the first
set. This makes it less likely that either set will generate spurious DNA. Semi-nested
PCR is a similar process where only one primer is located internally. The two sequential
reactions with fewer cycles in each reduce the problems associated with overcycling.

Finally, as ancient DNA quantities are extremely low, it is important to take all
precautions to prevent contamination fiom modern sources. This includes the use of
facemasks, gloves, protective sleeves or lab coats, and performing DNA extractions and
amplification reactions in a PCR hood. Contamination can result from several sources

including other samples, the analysts, or fi'om PCR product carry over. For this reason,

-14-

several controls are conducted with each step of the analysis. A reagent blank contains
all the reagents used during the DNA extraction, and is processed like the samples
themselves. If there is evidence of DNA in this control, then at least one or more of the
reagents is contaminated and new reagents should be used. Positive and negative
controls are also conducted with each PCR reaction to verify successful amplification. If
a product appears in the positive control, all necessary reagents were present. The
absence of product in the negative control signifies that the reaction mix was not
contaminated. To prevent the possibility of PCR product carry over, all pre-amplification

reactions should be prepared in a separate room from post-amplification products.

The Goal of this Project

The goal of this study was to answer questions regarding the organization of the
tumulus at Kamenica and the relatedness of the individuals within the Great Circle by
comparing mtDNA haplotypes obtained from skeletal remains of those who were interred
nearly 3000 years ago. In addition, these findings were compared to previous genetic
studies of Monumental Structure 3, as well as to modern day inhabitants of Kamenica to
examine genetic continuity over time. This project can provide a more defined history of

the past Albanian population who lived during the Late Bronze Age.

-15-

Materials and Methods

Burials in the Tumulus at Kamenica
Samples from twenty-one individuals buried in the Great Circle were genetically
analyzed. These were chosen primarily based on the location of the grave and century of

interment. Figure 5 displays each of the burials that were included in the research:

Figure 5: Photograph of Graves

Each of the 21 individuals used in the research are displayed in the following
photographs (a-s). Location within or above the Great Circle is noted, as well as the sex
and estimated date in which the individuals were interred. An (’) indicates a possible
male/female.

(a) Individual 347:

Located in the southwest
portion above the Great
Circle.

Adult male estimated to
be from the 11 — 9th
century B.C.

 

-15-

 

(c) Individual 351:

Located in the southwest
portion of the Great
Circle.

Adult male estimated to be
from the 12 — 11th century
B.C.

(b) Individual 349:

Located in the northwest portion
above the Great Circle.

Adult of undetermined sex
estimated to be from the 11 — 9"
century B.C.

 

-17-

(d) Individual 356:

Located in the center/east
portion of the Great Circle.

Adult of undetermined sex
estimated to be from the 11 — 9th
century B.C.

 

(e) Individual 357:

Located in the center/west
portion above the Great Circle.

Adult male estimated to be from
the 11 — 9‘h century B.C.

 

-13-

 

(g) Individual 367:

Located in the center/west
portion above the Great
Circle.

Adult of undetermined sex
estimated to be from 13 —12th
century B.C.

(1) Individual 358:

Located near the center of the
Great Circle.

Adult male" estimated to be from
the 11 — 9" Century.

 

-19-

(h) Individual 370:

Located in the northeast portion
of the Great Circle.

Adult male‘ estimated to be from
the 12'h century B.C.

 

(i) Individual 373:

Located near the center of the
Great Circle.

Adult of undetermined sex
estimated to be item the 12'“
Century B.C.

 

.20-

(j) Individuals 374 and 377:

Located near the center of the
Great Circle.

374: Adult Female
(data could not be obtained)

377: Child (8 — lOyrs old)

Estimated to be from the 12'”
century B.C.

 

(k) Individual 376:

Located near the center of the
Great Circle.

Adult male estimated to be from
the 12“I century B.C.

 

-21-

0) Individual 381:

Located in the center/west
portion above the Great Circle.

Possibly 2 adult individuals of
undetermined sex estimated to
be from the 12‘” century B.C.

I

 

(In) Individual 382:

Located near the center of the
Great Circle.

Adult male estimated to be hour
the 12'11 century B.C.

 

-22-

 

(0) Individual 389:

Located near the
center/southwest portion of the
Great Circle.

Adult female estimated to be
from the 12'” century B.C.

(11) Individual 383:

Located in the southeast
portion of the Great Circle.

Adult male" estimated to be
from the 13th century B.C.

 

(p) Individual 390:

Located near the center/west
portion of the Great Circle.

Adult male estimated to be
ficm the 12th century B.C.

 

(q) Individual 391:

Located in the north portion
of the Great Circle.

Child (6 —— 8 years old)
estimated to be from the 13 —
12‘“ century.

 

-24-

 

(3) Individual 395:

Located near the center of the

Great Circle (earliest individual).

Adult male estimated to be from
the 13 —12th century.

(r) Individual 392 and 393:

Located in the north portion of the
Great Circle.

392: Adult Female
393: Child (3 — 4 years old)

Both estimated to be from the 13 —
12‘“ century B.C.

 

-25-

Bone Selection

Fifty-two bones representing twenty-one individuals were examined. Samples
were collected by a forensic biology student from Michigan State University in the
summer of 2003. Samples fi'orn each individual were chosen based on external quality of
the bone and bone types thought to harbor the most DNA. In particular, long bones such

as the femur and humerus, and the petrous portion of the temporal bone were selected

(Table 1).

Table 1: Bones Processed for DNA

The table lists the bones that were used for each individual. An ‘X’ indicates that
DNA extractions were performed on those bones. Bones other than femur,

humerus, and petrous are listed in the “Other” column.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Individual Bone
Femur Humerus Petrous Other

347 X X
349 X Tooth
35 l X X
356 X X Tibia
357 X X
358 X X
367 X Skull
370 X X X
373 X X X
374 X Skull
376 X X
377 X X
381 X Skull
382 X X
383 X X
389 X Tibia/Pelvis
390 X X X Tooth
391 X X Tooth
392 X X X
393 X Tooth
395 X X X

 

-26-

 

Bone Preparation

Bones were cleaned with sterile digestion buffer (20 mM Tris pH 8, 50 mM
EDTA, 0.5% SDS) and sterile H20 using a sterile cotton swab. Following the initial
cleaning, the outside of each bone was sanded with a Dremel Sanding Band in order to
remove any remaining dirt, ensuring that only DNA from inside the bone was extracted,
eliminating the possibility of obtaining DNA profiles from scientists who had handled the
material. The drill and drill pieces were scrubbed with 1% Liqui-Nox solution, soaked in
10% bleach for approximately 1 minute, and rinsed with 18.2 MST/cm Milli-Q water.
They were also exposed to short wave UV light at 2500 Joules/cm2 for 300 - 600
seconds. Bones were drilled at low speed in a PCR hood (Labconco) with the Dremel
tool and 1/16"I inch twist drill bit. Between 0.1 g and 0.2 g of bone powder was collected
in sterile 1.5 11L microcentrifuge tubes that had been UV irradiated for 300 seconds.
After each sample of bone powder was collected, the drill and drill bits were cleaned

using the procedure described above.

DNA Extraction

DNA from bone samples was extracted organically. Four hundred microliters of
digestion buffer and 4 11L of 20 mg/mL proteinase K were added to each tube containing
the drilled bone, which were incubated at 55°C overnight. In addition, 200 14L of
digestion buffer and 2 11L of proteinase K were added to an empty tube to act as a reagent
blank. Following incubation, an equal volume of phenol was added to each tube (400 11L
for bone samples, 200 14L for reagent blank), vortexed for 15 — 20 seconds, and

centrifuged at 14,000 rpm for 5 minutes. The aqueous layer was transferred to a new

-27-

microcentrifilge tube. A second phenol extraction was performed on each of the samples
if a colored (pink to dark purple) aqueous layer remained. A chloroform extraction of
equal volume followed. Samples were vortexed and centrifuged at 14,000 rpm for 5
minutes. The aqueous layer was placed into a YM-30 Microcon column (Millipore) and
centrifuged at 14,000 x g for 8 - 12 minutes depending on amount of retenate that had
flowed through. Three hundred microliters of TE buffer (10 mM Tris, 1 mM EDTA) was
added to the retenate, which was centrifuged at 14,000 x g for 6 — 9 minutes. This was
repeated two times for a total of three TE washes. Appropriate amounts of TE were
added to each sample to bring the final DNA volume to 20 uL. Samples were stored at

-20°C.

DNA Amplification

The primers used for PCR amplification of hypervariable regions I and II can be
found in Table 2. DNA was amplified in two successive PCR steps. The first round
consisted of 20 uL reactions using 19 uL of master mix, containing 2 pL of Hot Master
Taq Buffer (Eppendorf), 2 11L of 30 pig/11L bovine serum albumin (BSA), 2 11L of 2
ug/ 11L deoxynucleoside 5'-triphosphates (dNTPs), 1U Hot Master Taq Polymerase
(Eppendorf), 0.4 11L of 20 M forward and reverse primers for the hypervariable region
being amplified, and sterile H20. One microliter of template DNA from each bone
extraction was then added. From this reaction 1 uL was transferred to a second tube for a
1:20 DNA dilution. Positive and negative controls using the same master mix were also

included, using a 10 1.1L reaction volume.

-23-

Initially, the PCR primers used for HVI amplification were F16190 and R16410.
However, Belledi et al. (2000) in their study of Albanians found that much of the
sequence variability lies in a region before 16190. Therefore a new reverse primer,
which was designed fi'om the reverse complement of F 161 90 (R16207), was coupled
with F15989. PCR parameters were as follows: an initial denaturation step at 94°C for 2
minutes, followed by 30 — 38 cycles of 94°C for 30 seconds, annealing at 56°C for 1
minute, and extension at 72°C for 1 minute. Samples were held at 4°C until placed in a

-20°C ficezer.

Table 2: HVI and HVII Primer Pairs for Bone Samples
Primer pairs used for HVI and I-IVII for non- and semi-nested PCR. The primer name and
sequence are given in each block. (F = forward primer, R = reverse primer)

 

 

 

 

 

 

 

HVI HVII
First Round Primer Pair First Round Primer Pair
F16190 R16410 F15 R285
5’-ccatgcttacaagcaagt 5’-gaggatggtggtcaagggac 5’-caccctattaaccactcacg 5’-gttatgatgtctgtgtggaa
F 15989 R16207
5’-cccaaagctaagattctaat 5’-acttgcttgtaagcatgggg
Semi-nested Primer Pair Semi-nested Primer Pair

 

 

 

 

 

F 16057 R16207 F82 R285
Wm 5’-actt ctt cat 5’-ata catt c ac ct Shmtgatgtctgtgggaa

PCR products were electrophoresed on a 2% agarose gel, followed by ethidium

 

bromide staining and UV visualization. If a product was seen in the negative control, a
second PCR was performed with new reagents. If a product was seen in the reagent
blank, a second PCR attempt was performed with fewer cycles. If a product still

appeared, then DNA was re-extracted from the bone.

-29-

If no hands or very faint bands were present in sample lanes, semi-nested PCR
was performed. A master mix was prepared in the same manner as described above; with
the exception that BSA was replaced with filter sterilized Milli-Q water and a new
internal primer was used. One microliter of the amplified template DNA from the first
round of PCR was added to each tube containing the master mix. PCR parameters were
the same as in the first round—with fewer cycles (between 10 and 20) depending on the

amount of DNA product estimated from the agarose gel.

Buccal Swabs fi'om Current Inhabitants of Kamenica
DNA Extraction

DNA was extracted from nine buccal swabs obtained fiom inhabitants of
Kamenica who self-proclaimed their ancestry as being from the village. A small portion,
approximately 5 m2, was cut from each buccal swab with flame-sterilized scissors or
individually wrapped sterile scalpels. Samples were placed into labeled 1.5 uL
microcentrifuge tubes, along with 400 uL of digestion buffer and 4 11L of proteinase K.
Samples were incubated at 55°C overnight. In addition, 200 uL of digestion buffer and 2
1.1L of proteinase K were added to a tube for a reagent blank control.

An organic extraction was performed in the same manner as described for the
bone samples, with the exception that only one phenol extraction was necessary. The
aqueous layer was then removed to a sterile microcentrifuge tube. Two volumes (800
11L) of 95% ethanol and 1/10 volume (40 uL) of 3M Sodium Acetate were added to each
tube, vortexed, and placed in -20°C freezer for ~ 1 hour. Tubes were centrifuged for 15

minutes at 14,000 rpm and liquid was removed, leaving the DNA pellet intact. DNAs

-30-

were dried under vacuum for ~20 minutes. Twenty microliters of TE was added to each

sample and the DNA was stored at -20°C.

DNA Amplification

PCR was conducted in the same manner as the bone samples; however the entire
control region was amplified in one reaction using the primers in Table 3. Positive and
negative controls were included for each PCR reaction. Four microliters of PCR product
was electrophoresed on a 1.5% agarose gel to determine if amplification was successful

and to estimate the amount of DNA present.

Table 3: HVI and HVI] Primer Pairs for Buccal Swabs
Primer pair used for amplifying the whole control region for buccal swabs. The primer
name and sequence are given. (F = forward primer, R = reverse primer)

 

 

 

 

 

 

Primers
HVI HVI]
F 15989 R484
5’-cccaaagctaagattctaat 5’-tgagattagtagtatgggag

 

DNA Sequencing and Analysis

The remaining 16 uL of successfully amplified products were purified using
Montage PCR Centrifugal Filter Devices (Millipore). The final DNA volume was
brought to 400 11L with the addition of TE, and the columns were centrifuged at 1000 X g
for 15 minutes. Samples were resuspended to the initial volume with TE.

Sequencing reactions were prepared using 4 11L the CEO DTCS - Quick Start Kit
(a master mix containing the Reaction Buffer, dNTP’s, labeled ddNTP’s, and DNA

polymerase; Beckman), as well as 1 uL of 20 11M forward or reverse primer (primer used

-31-

in the non- or semi-nested PCR reaction for either HVI or HVII; refer to Tables 2 and 3)
and DNA template. The final volume was brought to 10 uL with filter sterilized Milli-Q
water.

Sequencing parameters varied based on the primers being used. Typical
sequencing parameters were: DNA denaturation at 96°C for 20 seconds, primer annealing
at 50°C for 20 seconds, and primer extension at 60°C for 4 minutes. For reverse primer
484 it was necessary to reduce the annealing temperature to 45°C. Reverse primer 16207
sequenced poorly, therefore a new annealing temperature was tested using a sequencing
gradient in increasing increments of 55°C to 65°C. The primer’s optimal annealing
temperature was found to be 61 .5°C. Primer elongation was raised to 65°C.

A stop solution containing 1 uL of 3M NaOAc pH 5.2, 0.5 uL glycogen, 0.2 11L
500 mM EDTA and 0.8 11L of water was added to each sample and mixed thoroughly by
vortexing ~20 seconds. Thirty microliters of cold 95% ethanol was added to each
reaction. Samples were vortexed for 20 seconds and centrifuged at 14,000 rpm for 15
minutes. The supernatant was removed, being careful not to disrupt the DNA pellet.
Two hundred microliters of 70% ethanol was added to each sample and centrifuged at
14,000 rpm for 3 minutes. The supernatant was removed and the 70% ethanol wash was
repeated, followed by ~25 minutes of vacuum drying. The pellet was resuspended in 40
11L of Sample Loading Solution (Beckman) and allowed to resuspend for l — 5 minutes
before it was vortexed and pipetted into the CEQ sample plates. One drop of mineral oil
was added on top of each sample. The samples were analyzed on a Beckman CEQ 8000
Genetic Analysis System, using the LFR 1 - 45 method: capillary temperature at 50°C,

denature at 90°C for 120 seconds, inject at 2.0kV for 15 seconds and separate at 4.2kV

-32-

for 45 minutes. For the buccal swabs a separation time of 60 minutes was performed
using the same parameters.

The sequences were aligned using BioEdit software (Hall, 1999) and compared to
the control region sequence determined by Anderson et al. (1981). Consensus sequences
were generated using the most prevalent base. When one sequence showed ambiguity
and another sequence did not, the non-ambiguous base was called. When no other
sequence was available, then the ambiguous bases were included in the consensus
sequence. These sequences were also used to help establish haplogroup (Allard et al.

2000).

Single Nucleotide Polymorphism Analysis

All 21 samples were tested for the H haplogroup using SNP primer extension. In
the first round of PCR, a product size of 80 bp was amplified (Table 4). SNP analysis
also employed the use of semi-nested PCR, using the H-SNP reverse primer, amplifying

a 61bp product.

Table 4: SNP Primer Pairs for Haplogroup H
Primer pairs used for Haplogroup H m the SNP primer extension reaction for mm and
semi-nested PCR. The primer name and sequence are given in each block.

 

SNP Primer Pair
H-Forward H-Reverse
S’jacatcgtactacacgacacg 5’-tgatggcaaatacagctcct

 

 

 

 

Semi-nested SNP Primer Pair

H—Forward H-SNP (Reverse)
5 ’-agacatcgtactacacgacacg 5’-tattgataggacatagtggaagtg

 

 

 

 

 

-33-

Ten microliter reactions were prepared with 1X Hot Master Taq Buffer, 1 11L of
30 ug/uL BSA, 1 uL of 2 ug/uL dNTPs, 0.4 11L of 20 11M forward and reverse primers,
luL template DNA, and 5.6 11L filter sterilized Milli-Q H20. For semi-nested PCR
reactions, BSA was not added, and 6.6 11L of H20 was used. The PCR parameters were:
an initial denaturation at 94°C for 2 minutes, followed by 40 cycles of 94°C denaturation
for 30 seconds, 57°C annealing for 30 seconds, 72°C primer extension for 30 seconds,
followed by a final elongation step at 72°C for 5 minutes. Four microliters of product
was electrophoresed on a 4% agarose gel to determine if amplification was successful
and to estimate the amount of DNA present.

Two microliters of Shrimp Alkaline Phosphatase (SAP) and luL Exonuclease I
were added to the PCR products and incubated at 37°C for 60 minutes, followed by
enzyme denaturation at 75°C for 15minutes. Primer extension was performed using a
CEQTM SNP-Primer Extension Kit (Beckman). A premix containing the provided DNA
polymerase, reaction buffer, and CEQ dye terminators (ddATP and ddGTP for
haplogroup H) was prepared. Each reaction contained 5.5 uL of premix, 1 uL of 0.2 pM
or 2 uM SNP primer, 0.5 11L template DNA, and 3uL of H20 to make a 10 uL reaction.
The reactions were subjected to 25 cycles of 96°C denaturation for 10 seconds, 55°C
annealing for 5 seconds, and 72°C elongation for 30 seconds. Again, 1 uL of SAP was
added to each reaction and incubated at 37°C for 60 minutes to remove unincorporated
dye terminators, followed by 15 minutes at 75°C to deactivate the enzyme.

Samples were analyzed on a CEO 8000 using the SNP-21 parameters (capillary

temperature 50°C, denature at 90°C for 60 seconds, inject at 2.0 kV for 30 seconds, and

-34-

separate at 6.0 kV for 21 minutes). Each sample contained 0.5 —- 1 uL of SNP DNA, 0.5
11L of size standard 80 (Beckman) and 38.5 — 39 11L of Sample Loading Solution. Results

were examined using the Fragment Analysis Program (CEQ) under the Default SNP

Parameters or with an adjusted slope threshold between 1 and 10.

-35-

Results

Bone Sampling and Preparation

The bones most often used for analysis in this research were femur (3 7% of the
samples) followed by skull petrous portion (27%) and humerus (21%). These bones are
harder and may protect the DNA from degradation more so than spongy bones, such as
that found in the pelvis. The remaining 15% of bones used were teeth, tibia, pelvis, and
skull. The long bones could easily be cleaned and drilled because they were cut at the
shaft, while the skull petrous portions had several crevices in which dirt and mud was
present creating difficulty when cleaning and drilling. Some of the bones—including the
long bones and petrous—were brittle and sufficient bone powder could easily be obtained

(as stated, 0.1 — 0.2g from each bone).

PCR amplification

A neat reaction (1 uL of DNA into a 20 11L PCR reaction) and a 1:20 dilution
were run for each PCR. The neat reaction amplified 22% of the time in the first round of
amplification for HVI and HVII. A 1:20 dilution of the DNA improved PCR success,
with an approximate 10% increase in the number of reactions that amplified. However,
in three cases (burials 391 F, 392 SP and 393 SP) a 1:200 dilution was necessary. The
three bone types most often tested showed a similar pattern in amplification success in
the first round of PCR, where none amplified more or less than the others; but after semi-
nested PCR the petrous was shown to have higher amplification success. HVI was more
likely to successfully amplify in the first round (approximately 23%), with HVII (12% in

the first round) requiring nesting more often. Overall, more than 70% of the samples

-36-

required semi-nested PCR. The increased PCR cycle number raised the chance of
contaminated results in control samples, particularly for HVII. Figure 6 displays a
typical semi-nested PCR product; two bands can be seen, with the lighter top band
resulting from continued amplification of the initial amplicon and the lower band
representing the semi-nested product.

After repeated attempts at amplification, DNA fi'om some bones still could not be
amplified. These included 357 F/SP, 374 SK, 389 F/P/T i, 392 F/H, and 393T
(abbreviation key for bones can be found in Appendix 1).

Figure 6: Semi-nested PCR. (a) 2% agarose gel fiom the first round of PCR. No bands
were produced. (b) 2% agarose gel after semi-nested PCR, showing positive results. The
lanes are labeled with the burial number and bone type. A (‘) indicates a 1:20 dilution.

Note that only the diluted samples were amplified. Negative (-) and positive (+) controls
are in the last two lanes of each gel.

(3)

   

-37-

Analysis ofmtDNA Sequences fi'om Bone

The bones in which DNA was successfully amplified can be found in Table 5.
Upon sequencing, ten bones (347 F/H, 349 T, 358 SP, 367 F, 370 F, 373 F, 374 F, 376
SP, 382 F), showed evidence of contamination, although this decreased throughout the
months of analysis presumably due to extra precautions, such as wearing a surgical mask

during all procedures, changing gloves often, and performing extractions and

amplification reactions in a PCR hood.

Table 5: Bones that Successfully Sequenced

Results for bones that were used and that produced analyzable sequences. Bones that did
not provide analyzable sequence are denoted by (-), while those that were successful are
denoted by (+). NA (Not Analyzed) represents the bones that were not used. This does

not include bones that amplified.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Burial Bone
No.

Femur Humerus Petrous Other
347 - - +
349 + NA NA Tooth -
351 NA - +
356 + + NA Tibia +
357 - NA -
358 + NA -
367 - (2) NA NA Skull +
370 - + +
373 - + +
374 - NA NA Skull -
376 + NA +
377 NA + +
381 NA + NA Skull +
382 - + NA
383 - NA +
389 - (2) NA NA Tibia - /Pelvis -
390 + + + Tooth +
391 + NA - Tooth -
392 - - +
393 NA NA + Tooth -
395 + + +

 

-33-

 

Eighty percent of the sequences that were interpretable in HVI were forward
sequences. Many of the reverse sequences contained a high baseline and therefore could
not be analyzed. One explanation for the lack of success may be that the primer-
annealing site for the reverse sequence is near a long stretch of the same base starting at
position 16193. A gradient was used to determine the optimal annealing temperature of
the primer by varying the temperature between 55°C and 65°C. The annealing
temperature was optimized at 61 .5°C and allowed four of the reverse sequences,
including 358 F, 377 SP, 381 SK, and 393 SP, to provide analyzable sequence.

Of the bones that provided usable sequence, 36% were the skull petrous portions
of the temporal bone. Data were produced in 11 out of the 14 instances that the petrous
was sequenced, making it successful 79% of the time. The humerus followed, making up
27% of the sequences and providing successful sequence 72% of the time. The femur
provided 23% of the sequences and was successful 37% of the time. A complete
sequence with no ambiguities was rare. The most common ambiguities present in these
individuals were CH“ and NO transitions (Y and R; refer to Appendix 1 for abbreviations
of ambiguous bases). K and M were also common but several of these ambiguities are

attributed to sequencing artifacts (discussed below).

HVI Sequences

Six of 16 individuals generated a consensus sequence that matched the Anderson
et al. (1981) reference sequence (Table 6) in HVI. There were several ambiguous sites
that appeared in more than one individual, including 16093 in which burials 373 and 395

have a y (a C was in one sequence and a T or a Y in the other). Position 16126 displayed

-39-

a T to C transition for individuals 367, 370, 377, and 382, while individual 390 had a Y at
this position. Other base ambiguities seen in more than one individual included: 16094
K, 16102K, 16122R, 16183M, and 16189 Y. Some individuals showed an ambiguous
base at the same position in different bones. For example, individual 390 had a K at
position 16102 and a Y at position 16126 in both the humerus and femur. A more
complete list of the ambiguities seen in each individual can be found in Appendix 2.
Five individuals showed unique polymorphisms. Individual 358 had a T to C
transition at site 16140 and a G to A transition at site 16196 and two others
polymorphisms at position 16058 and 16092, although the latter two were at sites for
which sequences were not attained for most individuals. Others were 16129A for
individual 383, 16174T for individual 377, 16179T for individual 367, and 16189C for

individual 382.

-40-

Table 6: HVI consensus sequences

Sequences for HVI are arranged in order of burial number. The shaded area represents
the region in which sequence could not be obtained. Lowercase letters represent sequence
differences from two different bones. Key for the ambiguous bases can be found in
Appendix 1.

 

Polymorphic
Site

 

> cameo“—
> oooocn~
—-] mecca-—
r-l wxooow—
v—J «IAOOO‘H
.—a too—cw—
u—i #Nt—Ox—
u-J CNN—bos—
Q cub—ow—
-—] Gav—cud
n Aq~o-
u-a ooq~o-
O \Ofl—O‘sfl
u-i Noe—ax.—
> woo—ow—
.—] \OOO—Oxvd
O cue—Ow—
Q axe—ow—
v—l xoxou—cx-a

Reference

 

351

 

356 A

 

358 T C C R

 

367 C T

 

370 G C

 

373 y m

 

376 M Y

 

377 C T

 

381 W

 

382 C CCC

 

383 A

 

390 KKbY

 

391

 

392

 

393

 

395 y

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-41-

HVII Sequences

HVII produced usable data less often than HVI with only eight of the 21
individuals having analyzable sequences, including burials 347, 349, 351, 370, 376, 377,
390, and 395 (Appendix 3). Table 7 shows the consensus sequences for these
individuals. Three sites within HVII had the same sequence polymorphisms in more than
one individual. For example, individuals 370 and 390 both have C to T transition at
position 150. One site with several polymorphisms was 195, with 4 individuals showing
differences from the reference: individuals 347 and 390 both have T to C transitions; the
reverse sequence for the petrous from individual 370 was called as M, while the forward
sequence for the humerus displayed a T. Since three different bases were seen in this
individual, the consensus sequence was called as h (either an A, C, or T). Individual 395
also showed polymorphisms at position 195, with a T to A transversion in the forward
sequence for the petrous and a T in the other sequences; therefore the consensus sequence
displays a w. The most common polymorphism in HVII was 263, in which a G was
present rather than an A. This polymorphism, however, is found in most Europeans and
therefore is not informative.

Other sites with polymorphisms, including ambiguous bases in more than one

individual, were 73 G/R, 152 CW, 153 AR, 158 C/Y, and 159 C/Y.

-42-

Table 7: HVII Consensus Sequences

Sequences for HVII are arranged in order of burial number. The shaded area represents
the region in which sequence could not be obtained. Lowercase letters represent sequence
differences from two different bones. Key for the ambiguous bases can be found in
Appendix 1.

 

 

 

 

 

 

 

 

 

 

Polyngomhic79111111111111222
5'“ 35555556788999036
023895023459453
Reference AACTATTACCACTTTAA
347 CA G C
349 C
351 R G
370 WT h G
376 Y CC G
377 Y Y Y
390 G T C GG
395 R YY w G

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ambiguous Results in mtDNA Sequence Analysis

Fifiy-nine percent of the individuals showed ambiguous bases, either in the same
sequence or at the same site in more than one sequence. For example, individual 395 had
Y’s at positions 158 and 159 in forward sequences for both the petrous and humerus.
Individual 373 had two different bases in different bones at position 16124, with an A in
one and a C in the other (Figure 7). The reverse sequence for the skull of individual 381

showed ambiguous bases, A and T, at position 16080 in the same sequence (Figure 8).

-43-

Figure 7: Ambiguous bases from two different bones.

MtDNA sequence from individual 373 showing ambiguous bases with (a) A at position
16124 in the petrous and (b) C at position 16124 in humerus. An arrow points to the
position of the ambiguous base. Images shown in color.

 

(a)60 70
ATATAGTACGGTA

/

 

 

 

 

(b)

80
ATAT C GTAC GGTA

/

 

 

 

Figure 8: Ambiguous bases in one sequence

MtDNA sequence from individual 381 showing base ambiguity with A and T in one
sequence at position 16080. An arrow points to the position of ambiguity. Image shown
in color.

 

30
AC TA AC C G C

./

 

 

 

Another complication with the analysis was base insertions. Individuals 351, 356,
381, and 395 contained small peaks that appear consistently throughout the sequence.
These were considered sequencing artifacts rather than insertions, because they generally
occurred near the same base each time, and could potentially be due to a high baseline
during analysis or pull-up. The latter can occur when there is an overlap of dyes that are
used to label a base, resulting in the dye nearest in the color spectrum to the one being
detected also being displayed. For example, in 381 pull-up from T results in a small G
peak beneath each (Figure 9a). Because it occurred near the same base each time, it was
not considered to be a valid insertion. The petrous of 395 for HVII had similar artifacts

with C’s appearing underneath some A’s (Figure 9b).

Figure 9: Examples of Pull-Up
(a) MtDNA sequence from individual 381 displaying pull-up from T. Arrows point to the
base being “pulled-up”. Images shown in color.

 

100 110 120
GTACATGAAAAACCCAAT CCACATGCAAAA

 

 

 

~10 50 60
3A TGAC TGCCAGCCACCATGAATTATG GTACG

/

 

 

 

-45-

(b) MtDNA sequence from individual 395 displaying pull-up from A. Arrows point to the
base being “pulled-up”. lrnage shown in color.

 

To..\TGTCGCAGT\TCTGTCTTTGATTCCTGCCTC

MAMMMW

Analysis of mtDNA sequences from Buccal Swabs

 

 

 

Twenty-one buccal swabs from modern residents of Kamenica were analyzed in
the current study and by Rennick (2005). Sequences for HVI and HVII were obtained for
twenty of these; one sample could not be amplified after two attempts to extract DNA.
Larger regions covering approximately 900 base pairs could be amplified from the buccal
swabs because the samples were fresh and contained ample DNA. Twenty different
haplotypes were observed; these sequences can be found in Appendix 4.

The buccal swabs had three sites in HVI in which the same polymorphisms were
seen in some bone samples. Three current residents of Kamenica and individuals 367,
370, 377, and 382 had a C at position 16126. Individual 383 had an A at position 16129
while one buccal swab had an R at this position. Five current residents and individual
382 had a C at position 16189; individual 376 had a Y. Three current residents displayed
the reference sequence, and therefore could not be differentiated from four individuals
from the tumulus: 381, 391, 392, and 393.

HVII has six polymorphic sites in which the buccal swabs and bone samples were
the same or could not be distinguished from each other. The most frequent was 263 G,
shared by all bone samples in which sequence was available (5 individuals) and 20

current residents; this is, as stated above, extremely common in Caucasians. Six current

-46-

residents had a G at position 73; individual 390 also had a G and individual 395 had an R.
Two current residents had a G at position 95 and individual 370 had a W at this position.
Four current residents and individual 347 had C’s at position 152. Two current residents
had a G at position 153 and individual 351 had an R. Three current residents and
individuals 347 and 390 had C’s at position 195. Individuals 370 and 395 also had

polymorphisms at site 195, but results were ambiguous.

Haplogroup Assignment

Two methods were used for haplogroup determination. The first involved
grouping individuals based on specific sequence differences found in the control region.
Haplogroup H is defined by 73A and should contain no polymorphisms characteristic of
any other haplogroups. Only two individuals had sequence in this region: 390 and 395,
which displayed G and R, respectively, indicating a non-H haplogroup. Twenty-four
additional polymorphisms have been associated with Haplogroup H, thirteen of which
fall into the region that was amplified in the current study. Five of the thirteen
polymorphisms were seen in the ancient samples (Table 8; Allard et al. 2002).
Table 8: Haplogroup H polymorphisms
Control region polymorphisms associated with haplogroup H. The Anderson et al.
(1981) reference base is given in the second column with the base associated with

Haplogroup H in the third column and individuals displaying those polymorphisms in the
last column. A (*) indicates ambiguous base at that position (Y).

 

 

 

 

 

 

 

 

 

Site Reference Polymorphic Individual(s)
Base Base

16129 G A 383

16189 T C 376*, 382

150 C T 370, 390

152 T C 347

195 T C 347, 390

 

 

 

-47-

The second method involved using SNP primer extension, which was attempted
on all 21 individuals, with results being obtained for 12 (Table 9). Two bones from an
individual were amplified if results from the first bone were not successful and between
one and three amplifications in total were attempted. Three individuals out of the 12
required analysis of two bones (374, 382, and 395). All 21 individuals amplified at least
once, however results for nine individuals could not be obtained due to absence of a SNP
peak or several peaks located at the wrong size location.

Six of the successfirl SNP results came fiom the petrous. The remaining six came
from either a femur (2), humerus (1), tooth (1), or skull (2). Nine individuals displayed a
non-H haplogroup (Figure 10a), with three individuals displaying ambiguous results
(374, 389, and 395). Individual 395 was amplified three times and was consistently
ambiguous (Figure 10b); further results from 374 and 389 were attempted but could not
be obtained. A peak was present for individual 351 at 27.56, which is higher than the

others possibly because the size standard was slightly off.

-43-

Table 9: Results for SNP Primer Extension

Location of peak and haplogroup assignment determined by SNP primer extension.
Ambiguous results mean that one peak indicated H haplogroup while another peak
indicated a non-H haplogroup.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Individual Size (nt) Haplogroup
Assignment
347 22.03 Non-H
349 22.44 Non-H
351 27.56 Non-H
367 20.50 Non-H
370 20.48 Non-H
373 21.35 Non-H
374 21.82; 21.85 Ambiguous
376 22.61 ‘ Non-H
382 21.75 Non-H
389 20.50; 21.82 Ambiguous
391 21.36 Non-H
395 21.94; 22.81 Ambiguous

 

The results for 347, 370, 376, and 382 are not consistent with the control region

SNPs; however these polymorphisms are also common in Haplogroups U, T, and J.

Figure 10: Examples of SNP Peaks
(a) An arrow points to the red SNP peak at location 21.75 indicating non-H haplogroup
for individual 382. Peaks at size 13 and 88 are the size standard. Image shown in color.

 

am an
‘E

"'1/

n 9'

 

 

 

u
i
‘L 1 l 1 1 r -— r'e-‘f’s-Nl-J r "n-‘l—afmr-‘J‘ “(Pr—Iv’f'_].'?H‘T—-fl‘"tr-‘fi-‘t‘qvt—T'TJ'l‘-"l_I"-T‘1"f"—f—-"t-":l_'x—~1-'~I-""1, a "T—-<r~'-r'~-r‘

-49-

(b): Ambiguous result for individual 395 with a red peak at 21.94, indicating non-H
haplogroup and a green peak at 22.81, indicating H haplogroup. Image shown in color.

 

 

 

\ 22.31 K
E ,,,,, 1}- . - , 1L

v‘ V

 

-50-

Discussion

The tumulus at Kamenica has been well studied for the past decade and now
includes genetic research on two of the large structures found there: Monumental
Structure 3 in 2005 and the Great Circle presented here. Using mtDNA to compare
relatedness and genetic continuity, a better understanding of burial patterns, migration,
and genetic diversity can be obtained. MtDNA haplotypes were used to determine if the
individuals in the Great Circle were maternally related to one another and whether the
change in burial structures that occurred in the 7th Century B.C. may be due to a new
wave of people moving into the region. Comparing haplogroups of the individuals buried
in the tumulus to modern day inhabitants of Kamenica can indicate whether the Illyrian

people who lived 3000 years ago have modern descendents in this region of Albania.

Bone Preparation and DNA Amplification Success

Because the bones were covered in soil, a rigorous cleaning process was
necessary before DNA could be extracted. The dirt was scraped off as much as possible
with cotton swabs and a spatula, however, the remaining dirt turned to mud when the
bone was cleaned with digestion buffer and water. The outer surface of the bone was
sanded off, but in bones with many crevices, such as the skull petrous portion, small
specks of mud still remained, which could potentially cause PCR inhibition. Many of the
bones were broken into relatively small pieces, making them difficult to grip while
drilling. In addition, because these bones were nearly 3000 years old, they were very

brittle and often times crumbled into pieces when handled. While the brittleness of the

-51-

bones allowed ample bone powder to be obtained, it also may have made it less likely to
yield amplifiable DNA (Kaestle and Horsburgh 2002).

The femur and skull petrous portions were used most often because they were
harder and thus easier to process, and may have been also more likely to possess intact
DNA (Misner 2004)—-although at times these bones were brittle as well. DNA fiom the
humerus and femur amplified at a similar rate (59% and 50%, respectively), but the
humerus had 48% fewer samples. DNA from the skull petrous portion amplified 67% of
the time, but usually only after semi-nested PCR. Because of the hardness of the petrous,
it is possible that the cells inside, or the DNA itself, was protected from the elements,
providing more robust DNA results. Rennick (2005), studying Monumental Structure 3,
found that femur DNA successfully amplified 92% of the time, followed by petrous
DNA, which successfully amplified 90% of the time and required fewer nested-PCRs.
The difference in amplification success rates in these two studies could be due to the age
of the samples (approximately 2 — 7 centuries apart), the environment that the bones were
exposed to, or the brittle nature of many of the femurs used in the current study.

The region of HV I amplified was less than 200 base pairs and in HVII was just
over 200. These short spans are not highly informative unless sequence differences are
unique. However, when working with ancient, often degraded DNA, smaller regions
must be amplified. Anything above 200 — 300 base pairs usually cannot be amplified
unless it is very well preserved (Hummel 2003). It is possible to amplify multiple
mtDNA regions with different primers, but this project focused on specific regions——as
the most pertinent information for the Albanian population has been reported to lie within

the first 200 bases of HVI (Belledi 2000).

-52-

HVII provided fewer results than HVI, partly due to less amplification success.
One reason for this may be that HVII in general is more susceptible to mutation. If this
were the case, then mutations may predominate at the primer binding site, thus the primer
will not be able to hybridize at the correct location and PCR would fail. Ideally, the
primer is located in a region that is known to have little variability, however, in a
hypervariable region—especially in ancient DNA that accumulates post-mortem

mutations—this is difficult to achieve.

Degradation and Mutations that Cause Ambiguity in Sequences

Post-mortem mutations are known to accumulate over time, thus the longer an
individual has been buried, the more mutations would be expected. Bones from more
distant centuries in the tumulus would predictably have more mutations or ambiguities
than those of more recent centuries, and especially more than samples from current
Kamenica inhabitants. An example from HVII provides some evidence of this trend,
where burial 395, fiom the 13 - 12th B.C., has a total of2l ambiguous bases from a
single bone (Petrous) and burial 347, from the 11 — 9‘h B.C., contains no ambiguous bases
in that same bone type. In addition, looking at the HV I and HVII sequences, only seven
ambiguous sites were detected in the modern inhabitants (6 of which were located in
HVII), whereas a total of 19 ambiguous bases were present in the consensus sequences ‘
for the bone samples. Of course this may also have resulted from the relative ease with
which the modern samples amplified and were sequenced, but overall it is a trend seen by

multiple researchers in the laboratory.

-53-

One of the more prevalent mutations in ancient samples that may result in
ambiguity are deaminations of cytosine to uracil, resulting in C to T transitions in
sequences following PCR. Deamination can also cause G to A transitions but less
frequently than C to T transitions (Hofreiter et al. 2001). When a site in one DNA
molecule is mutated and the same site in a different molecule is not, two results were
generated at that location. There were seven cases from the sequences obtained that
contain a C to T (or Y) transition and six instances of G to A (or R) transitions. When
sequences from other bones or other amplifications were included, two C to T transition
and five G to A transitions were eliminated. This observation reflects on the importance
of having multiple sequences to confirm results.

Another possible explanation for the presence of ambiguous bases is
heteroplasmy, which may have occurred in a few instances. For example, individual 395
had Y’s at position 158 and 159 in the forward sequences for both the petrous and
humerus bone. Individual 373‘ showed possible heteroplasmy in two different bones at
position 16124, with an A in the petrous and a C in the humerus. Since heteroplasmy is a
rare event and typically only occurs at one site in a sequence (Melton et al. 2005), these
ambiguities are still most likely due to post-mortem mutations, rather than potential
heteroplasmy.

Working with ancient DNA provides a higher risk for contamination, given the
very small amount of starting DNA that is present. Even when all reasonable precautions
are taken to prevent contamination, it can still occur from “background DNA” originating
from sources such as manufacturer reagents, pipettes and PCR tubes, or any of the other

items that were used during the procedures (Yang 2003). Since these contaminants are

-54-

typically from an individual much more recently alive, a higher copy number and less
degraded DNA will be present, causing it to be preferentially amplified. To avoid this,
supplies were UV irradiated whenever possible, and positive, negative and reagent
controls were always run. As DNA sequences were produced they were compared to
those of the researcher performing the procedures. Any sequences that could not be
differentiated from the researcher’s were not included in the results. If contamination
fi'om a specific outside source or a few sources was prevalent, the same sequence would
be expected in several individuals. Since all but four of the bone sequences were
different, and all could be differentiated from the researchers’, the results are considered

to have originated from the mtDNA in the ancient bones.

Sequencing Success and Failure

Each time a bone PCR product was present on an agarose gel, sequence analysis
was canied out. This did not always result in DNA sequences being generated however,
which could stem from a few factors. In HVI, the reverse primer at position 16207
anneals next to a long stretch of guanine bases. If the sequence had an A to G transition
at base 16189, a relatively common occurrence, the enzyme responsible for DNA
replication would have a difficult time accurately moving along the sequence. Some
sequences in this region could be obtained after the annealing temperature was raised to
an optimized 61 .5°C, but still less fiequently than forward sequences, with only six
analyzable reverse sequences included in the results.

Other possible explanations for sequence failure are either too much or too little

DNA in the sequencing reaction. Too much DNA can cause the current on the CEQ to

-55-

fail as a result of excess template DNA, which could interfere with the injection of the
sequenced fragments (Beckman 2004). Too little DNA would leave no sequenced
fragments to analyze. When only a faint band was seen on an agarose gel, the maximum
amount of template DNA was used in the sequencing reactions; at times this was still not
a sufficient amount to produce sequence data. Also, the sequenced DNA can be lost
during the ethanol precipitation stage, leaving no DNA to be analyzed. In most cases, at
least two sequences were obtained fi'om each individual, but 20% have only multiple
forward sequences, 24% have forward and reverse sequences but from only one bone,

and 32% have only one sequence.

Maternal Relatedness within the Great Circle

MtDNA haplotypes from each individual tested from the Great Circle were
compared to each other, as well as by approximate century of interment and the location
of the burial within the tumulus. Overall, 14 of 18 individuals in the Great Circle were
found not to be maternally related. Sixteen burials produced HV I sequences (T able 6) of
which six had consensus sequences that could not be differentiated from each other and
therefore could potentially be maternally related. Of these, two could be eliminated as
being maternally related having sequence polymorphisms unlike each other in HVII; the
other four individuals did not have available sequence from HVII to compare. The
remaining 10 individuals have polymorphisms throughout their sequences that
distinguish them from each other, indicating non-relatedness. The eight individuals for

which HVII data were obtained did not display the same sequence.

-56-

Three individuals of note are 391, 392, and 393, all of which have been dated
from approximately the 13 - 12th century B.C. Individual 392 (adult female) and 393 (3
- 4 year old child) were buried in the same grave, with the grave of 391 (8 — 10 year old
child) very near. DNA from these individuals was extracted on separate days and was
amplified separately. All sequences from HVI matched each other and the reference
sequence (no data from HVII were obtained), indicating that these individuals could be
related. It is interesting to consider if these represent a mother and her children buried
together; certainly the genetic data support this.

An important aspect of this research involved the adult male in the center and
earliest grave of the tumulus, individual 395. One of the goals was to consider the
possible reason for his placement and what he represented to this society—whether it was
as a family leader, a tribal or religious leader, or some other entity. If he were a family
leader and the Great Circle was a familial burial, then haplotypes from maternal relatives,
including a mother, her children, maternal cousins and the like, would be predicted in
individuals buried near to him. If this individual was a tribal leader and the burials
represented a social but not familial grouping, different haplotypes would more likely be
present throughout the burial ground. Individual 395 displayed one site of ambiguity in
HV I and four sites in HVH as well as the common polymorphism at 263 G. Sequences
from individual 395 did not match any of the other individuals’ tested when all
ambiguous sites were included. However, individual 376, estimated to be from the 12th
century B.C., was placed very near to burial 395, and in HVII, sites 158 and 159 had C/T
transitions for individual 376 and Y’s (C or T) for individual 395; both individuals also

have 263G. Position 195 was ambiguous in 395, with an A in one sequence and T’s in

-57-

the other sequences; 376 had T’s at this location and one additional site of ambiguity at
position 149. This individual appears most likely to be maternally related to burial 395,

although the ambiguous bases make such a conclusion tenuous.

Haplogroup Designation

Haplogroup H is the most common in Europeans and was tested in the individuals
fiom the Great Circle. The first method to define haplogroups compared control region
polymorphisms known to be common in Caucasians. Haplogroup H is defined in part by
73A (Allard et al. 2002); only two individuals, 390 and 395, had sequence in this region
and displayed G (not haplogroup H) and R (ambiguous), respectively. Owing to the latter
finding a conclusion on 395’s haplogroup could not be drawn. There are several other
polymorphisms in the control region that are associated with haplogroup H, e.g., sites
16129, 16189,]50, 152, and 195. Individuals 347, 370, 376, 382, 383, and 390 contained
some of these polymorphisms, indicating they could be haplogroup H. However, these
sites are poorly defined and several are shared with other haplogroups, including U, T, J,
I, W and K. Through SNP analysis, it was determined that none of the individuals listed
above were haplogroup H (except 383 and 390, fiom which no SNP data could be
obtained), and in fact, nine individuals displayed solid evidence of a non-H haplogroup.
Just as in mtDNA sequences, the SNP site can mutate, including deamination causing a C
to T transition. Since a C at position 7027 defines the H haplogroup using SNPs, a
deamination event would cause a non-H haplogroup to be assigned incorrectly, which
could explain some of the inconsistencies between the first and second method of

haplogroup determination.

-53-

Genetic Comparison of Modern Day and Ancient Inhabitants of Kamenica

Studying samples from modern day inhabitants of Kamenica and comparing them
to the ancient material can provide a bigger picture of genetic diversity and continuity
over time. Rennick (2005) found that the Albanian polymorphic sites determined by
Belledi et al (2000) were also present in 70% of the individuals from current residents of
Kamenica, suggesting they are genetically similar to other Albanians. When the entire
mtDNA region fiom current day Kamenicans was compared, none of their haplotypes
matched each other, indicating both a lack of maternal relatedness in the village, and a
broad level of genetic diversity. When the modern resident mtDNAs were compared
again, but using only the regions of HVI and HVII that were studied in the ancient
samples, one set of two and another set of three individuals were the same. The set of
three displayed the reference sequence in this region, and therefore could not be
differentiated from the four individuals that displayed the reference sequence in HVI
(381, 391, 392, and 393; HVII sequences could not be obtained for these individuals).

There were two common polymorphic sites present in both the ancient and
modern sample. Position 195 in HVII contained a T to C transition for modern samples
K7, K18 and K20, and individuals 347 and 390 fi'om the ancient samples had the same.
Position 16126 contained a T to C transition in K9, K15, and K20, while four ancient
individuals, 367, 370,377, and 382 displayed this transition as well. However, these
samples did not match when other sites were considered. The sites could be common
polymorphisms in Albanians that have endured through time, or they could represent
mutational ‘hotspots’ (Holland and Parsons 1999), with these changes occurring

independently in these individuals

-59-

Comparison of Results to Previous Studies at Kamenica

An earlier study of Monumental Structure 3 conducted by Rennick (2005)
indicated that the individuals buried there were maternally related, whereas haplotypes
from the individuals in the Great Circle indicate that the majority are not maternally
related, and show much higher mtDNA diversity overall. The monumental structure was
used two centuries later than the Great Circle; therefore the genetic change could be due
to completely different peoples inhabiting the area during that time. This lowering of
genetic variation could also result fi-om a combination of lack of gene flow (no migration
in or out of the region) and genetic drift, a population bottleneck created by a sharp and
sustained reduction in population size, or limited/selective female breeding. It could
equally have resulted in the way the tumulus was utilized (see below).

Clemmer (2005) looked at double burials throughout the tumulus and found that
87.5% of those buried together might be maternally related. Three double burials were
located directly within the Great Circle: 46, 56, and 375. The region of sequence
analyzed in HVI was outside the area used in the current study so no direct comparison of
the data can be made. There were two polymorphic sites in HVII that were common to
both studies, however the haplotypes were not the same. Therefore, while the individuals
in the double burials might be maternally related to each other, they do not appear to be
maternally related to any of the individuals analyzed for HVII in the current study of the

Great Circle, thus adding to the diversity of this region of the tumulus.

-60-

Possible Burial Patterns in the T umulus

Based on the findings that the individuals in the Great Circle are largely not
maternally related, it can be assmned that, in general, it does not represent a familial
burial plot. With four of the sequences possibly being maternally related, especially the
hypothesized mother and children, it could be that some related individuals were buried
together, but the Great Circle shows no genetic indication of being restricted to one
family. The lack of maternal relationship, however, raises the question of how this
society might have existed. For example, the findings of this study are not inconsistent
with what one would expect of a patrilocal family burial plot. In this type of society, the
male and his kin remain in a village, new females emigrate, and daughters move out of
the village, resulting in a wide diversity of mtDNA haplotypes. If this were the case, then
the earliest male in the Great Circle could have been the “grandfather” of the tumulus.
However, only a small number of burials were located within the Great Circle, indicating
that the tumulus was not used by all inhabitants. Rather, this area of the tumulus might
have been used to bury select individuals, where burial is based on achieved status.

It is possible that the first individual (burial 395) held a position of social
importance, for example as a tribal leader. Use of the tumulus for socially important
individuals is a viable possibility, considering the long span of time that this section of
the tumulus was engaged (13th - 9'“ century) and the very few individuals that were
buried there (~80). If the Great Circle was not based on a familial relatedness or
achieved status, and all individuals from the village were buried here, this suggests that
over a 400 year period only one person was buried every five years on average, and there

were only about 80 people in the village during this time. Of course, other individuals

-61-

were buried outside of the Great Circle over much of that period, however the entire
tumulus contains an estimated 1000 individuals (Fenton personal communication), fi'om a
span of approximately eight centuries. It seems unlikely that a functional village would
have this few dead, and instead indicates that only select individuals, perhaps of high
status, were buried at this location In addition, the rocks used for the monumental
structures and the outer circle from the Great Circle came from another location, showing

a great deal of effort put forth for the burials.

Summary
The findings from this study indicate that the individuals that were buried in the

Great Circle between the 13th and 9th century B.C. were, for the most part, not maternally
related. Because the haplotypes are very diverse, burial pattern may be based on
achieved social status or possibly a familial burial if the village was patrilocal; for which
several haplotypes would be expected in both cases. By the 7‘h — 6‘“ century B.C. burial
pattern changed and so did the mtDNA diversity, whereby the individuals in the
monumental structures may be maternally related. It is also possible that the earlier
population simply experienced a loss of genetic diversity, resulting in the genetic
similarity seen in the Monumental Structure. Comparing the haplotypes from the
individuals buried in the tumulus to modern day inhabitants shows a large increase in
mtDNA diversity since this ancient society. With the use of DNA, the organization of
the tumulus at Kamenica and the relatedness of the individuals within it were theorized;

thus a more defined history of the ancient Albanian people was put forth.

-62-

Appendices

-63-

Appendix 1: Key for Abbreviated terms.

Abbreviation used for each type of bone

 

 

 

 

 

 

 

 

 

Bone Type Abbreviation
Femur F

Skull Petrous Portion SP

Humerus H

Tibia Ti

Tooth T

Skull SK

Pelvis P

 

 

Abbreviation used for each sequence base

 

 

 

 

 

 

Base Abbreviation
Adenine A
Cytosine C
Guanine G
Thymine T

 

 

 

Abbreviation used for ambiguous bases

 

 

 

 

 

 

 

 

 

 

 

Code Bases

Y C or T

R A or G

M A or C

W A or T

S C or G

K G or T

H A,C or T
N A,C,G,or T

 

 

-64-

 

 

Appendix 2: HVI Sequences

Successful Sequences for each individual in HVI

F= forward sequence, R= reverse sequence. Key for bases and bone type can be found 1n Appendix 1 The shaded region indicates no available sequence. A (*) indicates only
one sequence available.

Polymorphic
Site

Reference Base
351 SP F (1
351 SP F (2)
351 SP R

1
6
0
5
8
A

351 Consensus

356 Ti F

356 F F ' ,.
356 H F '
356 Consensus

358 F F

358 F R

358 Consensus

367 SK F *

367 Consensus

370 SP F *

370 Consensus

 

65

 

 

Polymorphic

5

5

4
4
4
4

3

99002222

GT

AT

AATTTTTATTGATTAACGCTAACTCA

373 SP F

373 Consensus

376 Consensus
377 SP F
377 SP R

377 Consensus

381 SK F
381 SK R
381 Consensus

382 Consensus

383 SPF *

383 Consensus

 

390 Consensus

 

6
6
6
6
6
6
6
6
6
6
6
6
6
6
r0
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6

Polymorphic

589990022223444455566777788899

l

67
AATTTTTATTGATTACCGCTAACTCAATGT

1

8023425246980368

Reference Base

391 Consensus

392 SP F*

392 Consensus

393 SP F
393 sp R

393 Consensus

395 SP F

 

395 Consensus

 

 

 

Appendix 3: HVII Sequences
Successful sequences for each individual for HVII
F= forward sequence, R= reverse sequence. Key for bases and bone type can be found in Appendix 1. The shaded region indicates no available sequence

Polymorphic 7 9
Site 3 5

2
1
4
A

Anderson Base A

347 SP F

347 SP R

347 Consensus
F'R*

349 Consensus

351 SP F

351 SP R

351 Consensus

370 H F

370 SP R

370 Consensus

376 F F

376 F R

376 Consensus

377 H R*

377 Consensus

 

68

 

 

2222222222

1 1

I

7

1233336

1

I

0
9
9
9
9
oo
oo
oo
00
7
7
7
6
6
6
6
5
5
5
5
5
4
3
5
3

Polymorphic

404853457
TAAAGTAAA

6902389135903603494569
AndersonBaseAAGTCTATTTGAACTATAGACTTT

390 SP F
390 SP R

390 Consensus

395 SP F
395 SP R

 

395 Consensus

 

 

Appendix 4: Consensus sequences for each buccal sample for HVI. Only differences from the reference sequence are shown.

The shaded region indicates areas with no available sequence.

 

*Table provided by Stephanie Rennick (2005)

 

 

 

 

 

Appendix 5: Consensus sequences for each buccal sample for HVH. Only differences from the reference sequence are shown.
The shaded region indicates areas with no available sequence.

1 53 1
CTCTTAACTTACTTATC
K1 G C G
K2
K3
K4
K5
K6
K7
K8
K9
K10
K11
K12
K13
K14
K15
K16
K17
K18
K19
K20
K21

12>
('3

DOCUQQQQ>

“"‘x

QQQQQQQQQQ

C30

 

*Table provided by Stephanie Rennick (2005)

 

Bibliography

Allard, M.W., K. Miller, M. Wilson, K. Monson, and B. Budowle. 2002. Characterization
of the Caucasian Haplogroups Present in the SWGDAM Forensic mtDNA Dataset
for 1771 Human Control Region Sequences. Journal of Forensic Sciences. 47(6):
1 - 9.

Anderson, 8., AT. Bankier, B.G. Barrell, M.H.L. deBruijn, A.R. Coulson, J. Drouin, I.C.
Eperon, D.P. Nierlich, B.A. Rose, F. Sanger, R.H. Schreier, A.J.H. Smith,
R.Staden, and LG. Young. 1981 . Sequence and organization of the human
mitochondrial genome. Nature. 290:457 -— 465.

Beckman Coutler. 2004. CEQ 8000 Series: Genetic Analysis System Manual.

Bejko, L. 1999-2000. Mortuary customs in the late bronze age of south-eastern Albania.
Iliria.29(1— 2): 155 — 181.

Bejko, L. 1994. Some problems of the Middle and Late Bronze Age in southern Albania.
Bulletin of the Institute of Archaelogy. 31: 105 — 126.

Bejko, L. (In Press). The Late Bronze/Iron Age Tumulus at Kamenica-Albania

Belledi, M., E.S. Poloni, R. Casalotti, F. Conterio, I. Mikerezi, J. Tagliavini, and L.
Excoffier. 2000. Maternal and paternal lineages in Albania and the genetic
structure of Indo-European populations. European Journal of Human Genetics. 8:
480 — 486.

CIA World Fact Book. 2006. http://www.cia.gov/cia/publications/factbook/geos/al.htrnl.

Clemmer, V. 2005. Maternal Relatedness within Double Burials of an Ancient Albanian
Tumulus. Thesis for Degree of MS. School of Criminal Justice.

Frucht, R., Editor. 2005. Eastern Europe: An Introduction to the People, Lands, and
Culture, Volume 111. ABC Clio: Santa Barbar, CA.

Gilbert, M.T.P, E. Willerslev, A.J. Hansen, 1. Barnes, L. Rudbeck, N. Lynnerup, and A.
Cooper. 2003. Distribution Patterns of Postmortem Damage in Human
Mitochondrial DNA. American Journal of Human Genetics. 72: 32 — 47.

Hall, T.A. BioEdit: a user-friendly biological sequence alignment editor and analysis
program for Windows 95/98/NT. Nucl. Acids. Symp. 41: 95 — 98.

Hammond, N.G.C. 1967. Tumulus-Burial in Albania, the Grave Circles of Mycenae, and
the Indo-Europeans. Annual of the British School of Athens. 62: 77-105.

-72-

Hofreiter, M., V. Jaenicke, D. Serre, Av.A. Haeseler, and S. Paabo. 2001. DNA
sequences from multiple amplifications reveal artifacts induced by cytosine
deamination in ancient DNA. Nucleic Acids Research. 29(23): 4793 — 4799.

Holland, M. M. and Parsons, T. J. 1999. Mitochondrial DNA sequence analysis:
Validation and use for forensic casework, Forensic Science Review. 11:22-50.

Hummel, S. 2003. Ancient DNA Typing: Methods, Strategies, and Applications.
Germany: Springer-Verlag Berlin Heidelberg.

ICAA. 2002/3. The International Center for Albanian Archaeology Review.
http://www.icaa.org.al/images/ICAA%20Report%202002-3.pdf

Kaestle, F.A., and K.A. Horsburgh. 2002. Ancient DNA in Anthropology: Methods,
Applications and Ethics. Yearbook of Physical Anthropology. 45: 92 — 130.

Medicine NET (2006). http://www.medicinenet.com/hydroxyapatite/article.htm

Melton, T., G. Dimick, D. Higgins, L. Lindstrom, K. Nelson. 2005. Forensic
Mitochondrial DNA Analysis of 691 Casework Hairs. Journal of Forensic
Science. 50(1): 1-8.

Misner, L. 2004. Predicting mtDNA Quality based on Bone Weathering and Type.
Theseis for Degree of MS. School of Criminal Justice.

O’Rourke, D.H., M.G. Hayes, and SW. Carlyle. 2000. Ancient DNA Studies in Physical
Anthropology. Annual Review of Anthropology. 29:217 — 242.

Rennick, S. 2005. Genetic Analysis of a Monumental Structure within the Kamenica,
Albania Tumulus. Thesis for Degree of MS. School of Criminal Justice.

Stewart, J .E.B., C.L. Fisher, P.J. Aagaard, M.R. Wilson, A.R. Isenberg, D. Polanskey, E.
Pokorak, J .A. DiZinno, and B. Budowle. 2001 . Length variation in HV2 of the
human mitochondrial DNA control region. Journal of Forensic Science. 46(4):
862 — 870.

Stipcivic, A. (1977) The Illyrians: History and Culture. New Jersey: Noyes Press

Vallone, P.M., R.S. Just, M.D. Coble, J .M. Butler, T.J. Parsons. 2004 A Multiplex allele-

specific primer extension assay for forensically informative SNPs distributed throughout

the mitochondrial genome. International Journal of Legal Medicine. 118: 147-157

Wilkes, J. (1992) The Illyrians. Oxford: Blackwell Publishers

The World Almanac (2005). New York: Press Publishing Co.

-73-

Yang, D.Y., B. Eng, and SR. Saunders. 2003. Hypersensitive PCR, Ancient Human
mtDNA, and Contamination. Human Biology. 75(3): 355 — 364.

Zickel, R. and W.R. Iwaskiw. (1994) Albania: A country study, 2“(1 Ed. (1994) Library of
Congress: Federal Research Division.

-74-

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

IIHINHIWIllllllllll)llllllllllHlllllllllllllllhl