Avian Species Identification Using the ND2 Gene
and DNA Analysis from the External Surface of an Eggshell
Final reportAuthor: Natalie Timmers Version: 2
Avian Species Identification
Data
:Version:
2
Author:
Natalie Timmers
timm0027@flinders.edu.au
n.timmers@student.avans.nl
Place:
Flinders University
Sturt Road, Bedford Park
5001 Adelaide, SA
Department of Biological Sciences
Period:
20/02/2012 – 09/07/2012
Supervisors:
Adrian Linacre
adrian.linacre@flinders.edu.au
Kirsten Baken
Avian Species Identification
Foreword
This report is developed after a research project of 4.5 months at the Flinders University in Adelaide, South Australia. This research was done by a senior Forensic DNA Laboratory Science student from Avans University, Breda in the Netherlands.
This document mainly contains the results and corresponding discussion and conclusion. Furthermore, this report includes information about wildlife crime and legislation, species and species identifications, phylogenetics, mitochondrion and mitochondrial DNA, feathers & eggs and information about the methods employed. Chapter three contains the methods employed. This chapter is followed by the results and discussion in chapter four and the conclusion in chapter five. Chapter six is about further work that can be done. Finally, this report ends with a comprehensive bibliography.
A lot of gratitude goes out to Adrian Linacre for all the help that has been given. In addition, thanks go to all students working at the same lab for their help and to Kirsten Baken for her supervision.
Avian Species Identification
Index
1. Abstract...1 2. Introduction...3 2.1 Wildlife crime...3 2.1.1 Customs...4 2.2 Species...4 2.2.1 Phylogenetics...7 2.2.2 Species Identification...7 2.3 Mitochondrion...8 2.3.1 Mitochondrial DNA...8 2.3.2 Feathers...10 2.3.3 Eggs...112.4 Information about the methods...11
2.4.1 Primers...11 2.4.2 PCR...12 2.3.2.1 Touchdown PCR……….12 2.4.3 ExoSAP- it...13 2.4.4 Sequencing...13 2.4.4.1 Geneious……….13 2.4.5 GenBank...14 2.4.6 MEGA...14 3. Method...16
3.1 The analysed avian species with scientific name and accession number from GenBank....16
3.2 Feathers...18
3.3 Mitochondrial DNA extraction using the QIAamp DNA mini kit...19
3.4 PCR avian species identification using barbs as a DNA source...20
3.5 Sample visualization (2% gel) and preparation for purification...20
3.6 Purify PCR products before sequencing (Bioline Isolate PCR and Gel Kit)...21
3.7 Quantification...22
Avian Species Identification
3.9 Sequence comparison and tree construction...22
3.9.1 Tree construction with cyt b and COI sequences from GenBank...22
3.9.2 Comparison unknown samples to ND2 tree...22
3.10 Maternal trace DNA from the external surface of a chicken eggshell...23
3.10.1 Positive control...24
3.10.2 Forward primer design...24
3.10.3 Second round PCR...25
3.10.4 ExoSAP-it...25
4. Results...26
4.1 Quality of the sequence...26
4.2 Quantity of barbs and quality of DNA for successful sequencing...28
4.3 Different methods for constructing a phylogenetic tree...32
4.4 Phylogenetic tree with loci ND2...37
4.5 Comparison of a phylogenetic tree with loci cyt b and COI...38
4.6 Unknown samples...42
4.7 PCR optimization DNA extraction eggshell...44
4.8 Amplification several DNA extractions from an eggshell...45
5. Conclusion...49
6. Further work...50
Avian Species Identification
1. Abstract
Since the high financial value, low risk of being caught and the minor penalties for avian species trading international and national wide has become an increasing issue and more and more species are endangered because of that, it is important to identify the traded species . For this species identification, genes on the mitochondrial DNA are commonly used. Cytochrome b (cyt b) and Cytochrome Oxidase I (COI) are mostly used in mammalian species identification . However, a previous study proved that the ND2 gene is more distinctive between avian species.
The aim of this research project was to determine if the ND2 gene is indeed a suitable marker for avian species identification. Moreover determine whether maternal DNA could reliably be extracted from the external surface of a chicken eggshell. This can later be used for egg identification from other avian species.
As a DNA source, barbs of a feather were used and a swab was taken from the external surface of an eggshell. Mitochondrial DNA from these sources were extracted and a section of the mitochondrial DNA amplified by PCR. This amplification process from DNA extracted from feather barbs resulted in an amplicon of 561 bp. For the DNA trace identification of the eggshell a smaller amplicon was produced; being approximately 195 bp. The PCR samples were purified using the Bioline Isolate PCR and Gel Kit before sequencing the samples. The samples were prepared and send off to the
Australian Genome Research Facility (AGRF) from which the sequence data were used to build phylogenetic trees. Accession numbers from GenBank from the same species as the used samples were used to build a phylogenetic tree of the cyt b and COI loci. These trees were compared to the tree build from the ND2 samples. Blind Trial Testing was conducted using three unknown (to me) samples that were analysed and an attempt was made to determine what species it is, using the build phylogenetic tree from the ND2 gene. Two out of three unknown samples were identified
successfully.
Disregarding the differences in methods to build the phylogenetic tree, number of samples and the samples itself, COI seems to be a better locus for avian special identification in comparison to the cyt
b locus. The reason is that cyt b shows more anomalies. When comparing COI and ND2 as avian
species identification loci, both are equally accurate, as both loci show little intraspecies variation and cluster the families together.
Several different PCR settings were conducted after DNA extraction using a swab from the outer surface of a chicken eggshell. In the end a Touchdown PCR was used, giving an amplicon of 195 bp. After running a second round PCR, amplification was done successfully. However, after purifying using
Avian Species Identification
the Bioline Isolate kit no DNA was left or the sequence was a mixed profile. ExoSAP-it was used instead, but this as well resulted in mixed profiles. In conclusion, it seems possible to successfully amplify DNA extracted from the external surface of a chicken eggshell, but no certainty could be given if the extracted DNA is actually Gallus gallus (chicken).
This internship was carried out at the Flinders University in Adelaide, Australia, supervised by Adrian Linacre. Adrian Linacre is a Professor at the School of Biological Science department. One area of this department is Forensic Biology. Linacre’s main interests are developing more sensitive methods of human DNA typing and learning more about the transfer of biological material at crime scenes. Another area is the use of non-human DNA to assist in especially wildlife crimes. Many PhD and master students are contributing to these areas. The results of these projects are the reason that this department receives funding to do more research. This research contributed to the non-human DNA research area and was a continuation of a PhD student who looked at the whole mitochondrial genome for avian species identification.
This document starts with describing information about wildlife crime, species including
phylogenetics and species identification, mitochondrion, including its DNA and the DNA sources (feathers and eggshell) and information about the methods including the primers, PCR, sequencing, GenBank and MEGA. Chapter three includes the methods employed, followed by the results with the discussion, in chapter five the conclusion and chapter six further work. Finally this document shows the bibliography.
Avian Species Identification
2. Introduction
2.1
Wildlife crime
Wildlife crime includes trafficking in living specimen (or their eggs), hunting out of season, habitat destruction, cruelty to animals, poaching for meat or trophies and the use of animal parts in medicines . This occurs at an international and national level. Since humans are more aware of the increasing impact on the environment, the interest in the prevention of wildlife crime has increased . However, it is difficult to estimate the exact amount of illegal trade since there are not as many international surveillance teams as there are, for instance, for drug enforcement . It is contrary that the interest in these crimes by media and the public has increased, but the prosecution of the crimes still has a low priority, compared to other crimes such as the drugs trade. Furthermore, wildlife crime offers a high financial reward, with only a fraction of the risk of being caught. Even then, the penalties are light .
Currently, on national and international level, there is greater awareness of the need for the
conservation of the wildlife. Many endangered species are protected by the laws in many countries. For example in Australia, the import and export of wildlife and wildlife products is regulated under ‘Environment Protection and Biodiversity Conservation Act 1999’ (EPBC Act). This legislation is administered by the Australian Government Department of Sustainability, Environment, Water, Population and Communities and is enforced by Customs at the border. This legislation aims to ensure that Australia meets its obligations under CITES and the Convention of Biodiversity. Moreover, to protect wildlife that could be negatively affected by trade, and the promotion of the conservation of biodiversity in Australia and other countries. There are three main aspects of wildlife trade: import of live plants and animals, export of native species and the import and export of internationally endangered species . The legislation of the Netherlands is covered by the European Community (EC) Wildlife Trade Regulations and also covers the obligation for CITES. CITES is a well-known
international agreement between governments . CITES is an acronym for the Washington Convention on International Trade in Endangered Species of wild Fauna and Flora and it monitors since 1973 the trade in endangered wildlife species . Monitoring the trade aims to ensure that international trade in the flora and fauna does not threaten their survival. All international trades are controlled; this means all import, export and re-export goes through a licensed system, which is adopted by each participating country, currently there are 175.
Avian Species Identification
Around 5,000 animals species and 29,000 species of plants are protected by CITES . All avian species of the order Psittaciformes (Table one) are subjected to CITES legislation . These are listed in three different appendices. Appendix I includes species threatened with extinction. These species can only be traded in exceptional circumstances. Appendix II includes species that are not necessarily
threatened with extinction, but controlling the trade is there to prevent this from happening. Finally, Appendix III includes species that are protected in at least one country. This county requested the members of CITES for assistance in controlling the trade .
2.1.1
Customs
To protect Australia’s border by preventing the import and export of (endangered) flora and fauna, Customs (and/or its partner agencies) inspects and x-rays all arriving air and sea passenger’s baggage. Not only is the illegal trade in wildlife a serious threat to the viability of the native flora and fauna, it could also cause infectious diseases. Wildlife trafficking has already been associated with the spread of diseases which not only threatens the health of native avian species but also of humans. An example is the outbreak of SARS (Severe Acute Respiratory Syndrome) and the influenza virus.
Customs not only protects the border by detecting illegal movement, but also investigates and prosecutes those who smuggle wildlife. Under the EPBC Act, the maximum penalty for illegal importing or exporting wildlife is AUS$ 110,000 and/or up to 10 years imprisonment . However, as stated in 2.1 there is a high financial return with little chance of being caught and even then the penalties are light. Rarely does the maximum penalty for the alleged event meet the potential financial gains .
2.2
Species
In order to investigate if a trade is illegal it is important to identify the unknown species. The main question in wildlife crime is ‘what species is this’? . Before that question can be answered it is important to know what a species is. Unfortunately there is not one correct definition of a species. Many definitions of species are used, of which the ‘biological definition’ and the ‘phylogenetic definition’ are the most common ones . In short, the biological definition is based on gross
morphological features (visual traits) , whereas the phylogenetic definition is based on the outcome of evolution . The biological definition says that a species is defined by having a barrier to reproduce with other species and produce fertile offspring. Genes could flow within the species when mating, but not with others. This barrier isolates the species so the population evolves new genes which could make interbreeding difficult or impossible. Many generations later more barriers evolve and
Avian Species Identification
turn the isolated population into a distinct species. However this concept was not satisfactory enough for everybody since some species looked relatively distinct but were able to interbreed. On the other hand, some species appeared not to interbreed at all. So, the biological definition does not account for non-sexually reproducing organisms, neither for different species that do reproduce with one another. The phylogenetic definition takes this out of the equation by accounting for evolutionary history meaning species share a common ancestor with other species .
The biological definition is based on the development of the classification scheme in the 1700s, which is still used today. It was the Swedish botanist Carolus Linnaeus who developed the classification scheme. Linnaeus categorised all biological things into hierarchical groupings; Kingdom, Phylum, Class, Order, Family, Genus and Species. For these groups Linnaeus used a two part name system called the binomial system, consisting of a genus name (to which the species belong) and a specific epithet (refers to one species within the genus). The first letter is capitalized and the entire binomial is italicized. Latin scientific names are used to avoid ambiguity when communicating about the organisms. Moreover, it withstands any challenge in Court. This classification and naming of
organisms is called taxonomy. The named taxonomic unit at any level of the hierarch is called a taxon. This classification system was at the time based upon God’s creative scheme and changed
dramatically after ‘The Origin of Species’ in 1859. Now, the evolutionary framework is more common .
Table 2, presented on the next page, shows the scientific classification of several avian species in comparison to the human classification. All avian species and the human fall under the Kingdom ‘Animalia’, which is indicated with the dark red colour. Moreover, all avian species and the human have the same Phylum (‘Chordata’). Each grouping which shows a difference to one another has another colour. The human is separated from the avian species on Class level. All avian species fall under Class ‘Aves’ (Orange) and the human under ‘Mammalia’ (Dark yellow). The Ostrich has another Order and the difference between the other avian species start at Family level. All cockatoos have the same family name (Cacatuidae) but differ on genus level. All macaws, parrots and the lorikeet have the same family name (Psittacidae).
Avian Species Identification
Table 1: Comparison of the classification from several analysed avian species to the human. The colours indicate the differences in classification.
Yellow-tailed Black Cockatoo Gang-gang Cockatoo Yellow-crowned Parrot Moustache Parrot
Regent Parrot Little Lorikeet Blue & gold Macaw Green-winged Macaw Ostrich Human Kingdom Animalia Phylum Chordata
Class Aves Mammalia
Order Psittaciformes Psittaciformes Psittaciformes Psittaciformes Psittaciformes Psittaciformes Psittaciformes Psittaciformes Struthioni-formes
Primate Family Cacatuidae Cacatuidae Psittacidae Psittacidae Psittacidae Psittacidae Psittacidae Psittacidae Struthionidae Hominid Genus Calyptorhynch
us
Callocephalon Amazona Psittacula Polytelis Glossopsitta Ara Ara Struthio Homo
Avian Species Identification
2.2.1 Phylogenetics
Phylogenetic analysis is the study of relationships among species. For this a phylogenetic tree is used. This illustrates how the different species branch off from a common ancestor and in what order . For the tree, branches and nodes are used. There are two different nodes, internal nodes which
represent taxonomic units and external nodes (at the end of a branch) that represent the species name. The length of a branch usually corresponds with the distance from one species to another. The smaller the branch, the closer the species are related. A tree can be rooted or unrooted . A rooted tree contains an out-group; this is one taxa that is known or believed to be most distantly related from the other taxa. It is not advisable to choose an out-group that is too far related from the rest of the taxa, since this could result in topological errors. Topology is the order of the branching pattern. Moreover, an out-group that is not distant enough may give errors as well . An unrooted tree only represents the relationship between the organisms, but no evolutionary path. This means that an internal node does not need to represent a common ancestor of two external nodes . An example of a rooted and unrooted tree is shown in Figure 1.
Figure 1:Difference between a rooted and unrooted tree .
2.2.2
Species Identification
As stated above, it is important to know what species it is, when species or products derived from those species are being traded which go against the national and international legislation . When morphology is compromised, genetic species identification is attempted .
The loci for forensic species identification are primarily found within the mitochondrial DNA . The identification of species relies on the isolation and analysis of DNA markers that show sufficient variation between species, called interspecific variation, but are generally the same within species (intraspecific) .
Avian Species Identification
The main locus used for species identification, especially for mammalian species, was until recently cytochrome b (cyt b). Cyt b is 1140 bp long and encodes a protein of 380 amino in length. Many DNA sequences of this gene have already been registered on a DNA database like GenBank . More
recently, the use of cytochrome C oxidase I (COI) has increased. This gene is 1551 bp long. It was first used in the identification of invertebrate species and later became the aim of being the locus of choice for identification of all animal species. This was done by the Barcode for Life Consortium . The Barcode for Life is an international effort to standardize DNA loci for the use in species identification. The Barcode for Life Data System (BOLD) is an organisation that requires that the laboratory providing the DNA sequence works to certain standards .
Greater intraspecies variation occurs in the HVI and HVII region (this part does not encode a protein) compared to the coding regions of the mtDNA, which precludes the use in species identification.
A recent study proved that the Nicetinamide Adenine Dinucleotide dehydrogenase subunit 2 (ND2) is more useful for distinguishing avian species. The ND2 gene is 1041 bp .
2.3
Mitochondrion
Mitochondria are organelles that exist in the cytoplasm of eukaryotic cells. Mitochondria have two membranes, each with a phospholipid bilayer with a collection of proteins. The outer layer is smooth, but the inner layer is convoluted. Since this is a highly folded surface it gives the mitochondrial membrane a large surface area which enhances the productivity of cellular respiration. The inner membrane divides the mitochondrion into two internal compartments. One compartment, the intermembrane space, is the narrow region between the inner and outer membrane. The second one is called the mitochondrial matrix and is enclosed by the inner membrane. This matrix contains many different enzymes, but also the mitochondrial DNA and ribosomes. The main function of the
mitochondria is cellular respiration, which is the metabolic process that produces energy, adenosine triphosphate (ATP) .
2.3.1
Mitochondrial DNA
Mitochondrial DNA (mtDNA) is found in the mitochondrial matrix. The mitochondrial genome is a double-stranded circular DNA molecule and has two general regions: the coding region and the control region. The control region is often referred to as the ‘non-coding’ region or the D-loop. Here, the initial separation, or displacement, of the two strands of DNA during replication occur . The mitochondrial genome encodes for 37 genes. Thirteen of which encode proteins involved primarily in the oxidative phosphorylation (ND1, ND2, ND3, ND4, ND4L, ND5, ND6, ATP6, ATP8, COI, COII, COIII
Avian Species Identification
and Cyt b) , two rRNA and 22 tRNA genes. The 22 tRNA genes are relatively small and have little variation. The other 15 genes contain more variation .
Even though the mtDNA in all vertebrates is mainly invariant, the avian genome has a different order. The ND5 gene is followed by cyt b instead of the ND6 gene , as shown in Figure 2.
Figure 2: Left the mitochondrial genome in most avian species and right from a mammalian species. The order of the cytochrome b and ND6 gene are reversed .
Using mtDNA for species identification is because mtDNA has a number of advantages. The mtDNA is better protected than nuclear DNA. The double outer protein membrane protects the internal mtDNA from degradation, which is useful for old samples. Another advantage is that mtDNA is maternally inherited, so there is no recombination of the mtDNA . Moreover, a major advantage of mtDNA is the presence of multiple mitochondria within a single cell, containing multiple copies in comparison to two copies of nuclear DNA . This increases the chance of success when analyzing evidence samples that typically contain little cellular material. Finally, the mtDNA has a higher evolution rate compared to nuclear DNA, primarily as the mitochondria lack an error reading enzyme to repair the DNA bases added incorrectly during DNA replication. This means that the errors made in the circular loop of the DNA are not corrected, resulting in a five times higher evolutionary rate than nuclear DNA .
This error, a mutation, can be either a single base change or a whole alteration. This single base difference in the sequence of DNA can be defined as a Single Nucleotide Polymorphism (SNP). It is called a SNP when the alteration occurs in at least 1% of the population . The effects of these changes within a coding region depend on which base is altered. A mutation at the third base of a codon will not affect the resulting amino acid and is called a synonymous mutation. The second base of a codon cannot mutate without alteration of the encoded amino acid. Still, even when an amino acid is
Avian Species Identification
changed but falls in the same category (acid, basic, polar and non-polar amino acid) it does not necessarily have effect on the protein’s activity .
The variation in terms of a SNP is what shows the difference between different species and is therefore an important tool in species identification.
2.3.2 Feathers
Feathers were source of mtDNA for the first part of this research project. A feather is made up of a shaft, with a proximal part called the calamus, which is implanted in a socket in the skin of a bird. The calamus extends into the rachis, which bears numerous barbs. The barbs constitute the vane. Figure 3 shows the structure of a feather. Feather can be divided in three types; downy feathers (fine feathers for insulation), contour feathers and flight feathers .
Like hair, feathers derive from a follicle and are further comprised of dead cells. When a feather falls out, the follicle replaces it immediately. Moreover, the composition of a feather is like hair; it consists mainly of keratin. Keratin has large numbers of the sulphur containing amino acid cysteine, which is required for the disulphide bridges. These bridges give strength and rigidity to the feather and protect the DNA .
When identification based upon morphology is no longer possible because the feather is incomplete, dyed or damaged, DNA based species identification can be used. The calamus can be used for DNA extraction. However, the sample can then only be run once and in case of a living bird it will hurt. To do minimal damage to the specimen, barbs can be used since it contains enough mtDNA .
Figure 3: Structure of a feather (Blue and Gold Macaw), showing the calamus which extends to the rachis. From the rachis numerous barbs extend. The DNA is extracted from the barbs.
Avian Species Identification
2.3.3 Eggs
Smugglers often transport eggs prior to hatching, since it is much easier than transporting living birds. Unfortunately this creates a problem for the authorities in accurate identifying the species since the egg needs to be incubated until a morphological identification can be made. Correctly identifying the species is critical in the success of wildlife forensic prosecutions. Recently DNA has been successfully extracted from the eggshell or egg contents, even in highly degraded samples. However, the eggs were autoclaved beforehand and a small piece of eggshell was cut out. This means the chick does not survive and the eggshell is damaged which is in ancient museum pieces not ideal . The use of non-destructive sampling methods to collect genetic material from wildlife has recently increased. An example is swabbing the external surface of eggs, which is a simple and efficient method for the extraction of mtDNA. Several studies have determined that swabbing the outside of an egg is a reliable method for amplification of mtDNA .
2.4
Information about the methods
2.4.1
Primers
The following primerset has been used during this research where feathers were a DNA source. It resulted in an amplicon of 561 bp long.
Forward primer: 5’CATACCCCGAAAATGATGGT 3’ Reverse primer: 5’TGTGTYTGGTTKAGKCCTAT 3’
561 bp is a large amplicon for trace DNA from the eggshell. For that part of the research a new forward primer was constructed. Combined with the reverse primer shown above, the amplicon is 195 bp. The sequence of the forward primer is:
Forward primer II: 5’ CCCATTCCACTTCTGATTCCCAGAAGT 3’
These primers are located on the heavy-strand of the mitochondrial DNA. Figure 4 shows where the expected amplicon of the primer sets can be found within the mitochondrial DNA. The size of the gene is not on scale. In red the size of the smaller primerset is shown.
Avian Species Identification
Figure 4: Overlapping sites of the expected PCR product of the ND2 gene (larger amplicon) and in red the site of the smaller amplicon for eggshell amplification.
The primers are universal primers, specific for avian species. Some parts in the genome are conserved, which means that these parts are similar in a large number of species. In these regions there is sufficient variation between closely related species, but members of the same species have the same (or nearly the same) DNA sequence. These regions are therefore ideal sites for primer binding because the primer will bind even to DNA or, in this case, an unknown avian species, but does not amplify human DNA which makes contamination less of a problem .
As shown above, in the revers primer a ´K´ and ´Y´ are used. This primer is therefore called a degenerate primer. This is a mixture where one or more of the positions can have several possible bases. Every alternative base is represented by a specific letter. This specific letter is based on the International Union of Pure and Applied Chemistry (IUPAC) code. ‘Y’ is the amino acid code for Tyrosine and can be one of the pyrimidines (C and T). ‘K’ stands for Lysine and designates a G and a T. .
2.4.2
PCR
For the PCR the 2x MangoMix from Bioline was used. This reaction mix contains MangoTaqTM DNA
polymerase, dNTPs, red and orange reference dyes and a final concentration of 2.5mM MgCl2. This
makes it a complete ready to use mix and reduces the time required to set up the reaction
dramatically and so reduces the risk of contamination. The dyes make the mix already colour-coded for the gel electrophoresis, so no additional Orange G or SYBR Green is necessary .
The following PCR program for avian species identification using barbs as a DNA source was used. First 60 seconds at 95oC, followed by a denaturation step of 30 seconds at 95oC, an annealing step at
55oC for 30 seconds and an extension of 90 seconds at 72oC. These last three steps were repeated 39
times. Hereafter a five minute step at 72oC, ending with a 12oC infinitive hold.
For amplification of the DNA extracted from the eggshell a Touchdown PCR program was used; First 60 seconds at 95oC, followed by a denaturation step of 30 seconds at 95oC, an annealing step at
Avian Species Identification
one cycle. The second cycle was run with an annealing temperature of 59oC, followed by one cycle
with an annealing temperature of 57oC, 55oC and 53oC. The cycle with an annealing temperature of
53oC was repeated 35 times. These 40 cycles in total were followed by a five minute step at 72oC and
an infinitive hold at 12oC.
The primerset and PCR settings for the avian species identification using barbs as a DNA source were adopted from a previous PhD project , where the PhD student optimized the settings .
2.4.2.1 Touchdown PCR
A large number of variables can dramatically influence the outcome of PCR amplification. These variables include the concentration of Mg2+, dNTPs, primers, template and cycling parameters.
Touchdown PCR (TD-PCR) focuses on one variable, the annealing temperature. TD-PCR employs an initial annealing temperature above the estimated melting temperature (Tm) of the primers being used, then progressively transitions to a lower, more permissive annealing temperature over a couple of cycles. By the time the annealing temperature drops to a level where nonspecific priming would normally occur, the specific product produced with TD-PCR has a head start and should outcompete any nonspecific product .
2.4.3 ExoSAP- it
ExoSAP-IT™ is designed for fast and efficient purification of PCR products for downstream applications such as sequencing. ExoSAP-IT contains two hydrolytic enzymes, Exonuclease I and Shrimp Alkaline Phosphatase (SAP), which do not interfere with downstream applications. Exonuclease I removes residual single-stranded primers and any extraneous single stranded DNA and SAP removes the remaining dNTPs from the PCR mixture .
2.4.4
Sequencing
Before sequencing the samples were purified using the Bioline Isolate PCR and Gel Kit or ExoSAP-it. Both methods remove single stranded DNA, such as the primers, and dNTPs.
The samples were sent to the Australian Genome Research Facility Ltd (AGRF). This is Australia’s largest provider of genomics services and solutions and is a non-profit company . The 3730xl Genetic Analyser from Applied Biosystems was used.
Avian Species Identification
2.4.4.1 Geneious
Geneious is software from Biomatters Limited that enables to import and export file formats. Moreover the quality of the sequences can be determined, the sequence can be edited and can be translated into amino acids . It was used to determine the quality of the sequences and edited the sequence when necessary. When it is unclear what base it at a certain position, the base was altered into a degenerate base. Table 2 shows the degenerate bases in the first row and row two shows the multiple alternative bases.
Table 2: Degenerate bases with the corresponding alternative bases in the second row .
2.4.5
GenBank
GenBank is an example of a genetic sequence database, containing publicly available nucleotide sequences. It is a part of the International Nucleotide Sequence Database Collaboration (INSD), which also includes the DNA Databank of Japan (DBJ) and the European Molecular Biology Laboratory (EMBL). GenBank is available from the National Center for Biotechnology Information (NCBI) website (http://www.ncbi.nlm.nih.gov/nucleotide/). ‘Entrez Nucleotide’ can be used for searching the genome or a specific gene of a species. In Basic Local Alignment Search Tool (BLAST) a specific
(primer)sequence can be entered and it will search in which species and where the sequence occurs . A disadvantage of GenBank is that any person can publish sequences online and there is no certainty that the species is properly identified and the quality of the sequence is high.
2.4.6
MEGA
Molecular Evolutionary Genetics Analysis (MEGA) can be used to conduct: automatic and manual sequence alignments, building trees using several methods and estimating rates of molecular evolution . For aligning sequences, Clustal W and Muscle can be used. During this research project Clustal W was used for aligning sequences. This is a multiple sequence alignment program for DNA and proteins .
MEGA uses different methods for constructing a phylogenetic tree; Maximum Likelihood (ML), Neighbor-Joining (NJ), Minimum Evolution (ME), UPGMA (Unweighted Pair Group Method with Arithmetic Mean) and Maximum Parsimony (MP). These different methods construct different phylogenetic trees, even with the same samples. The reason for this is that the methods use different kind of data and a different algorithmic approach.
Avian Species Identification
There are two main approaches to construct a phylogenetic tree; Distance based and character-based methods. NJ and UPGMA are distant approaches. This approach has two advantages; it is fast and it yields only a single tree. UPGMA is the oldest method and is hardly used anymore. NJ has almost completely replaced UPGMA. The main disadvantage of this method (UPGMA) is that it weights every base change the same and it states that all taxa are equally distant from a root, both not likely to be correct. Character-based method constructs multiple trees, and then uses some criterion to decide the best tree. Secondly every base change has a difference in weight. For example it is not so
common to have a transversion (substitution of a pyrimidine for a purine or vice versa) compared to a transition. Character-based methods take this into account. This also applies to the base position. A change in the third base of a codon is more likely to happen, so this is of less value . Because of these differences it can result in different trees and no guarantee can be given which tree is the ‘true’ tree .
The aims of this project were to determine if the ND2 gene is a suitable marker for avian species identification and comparing the built phylogenetic tree with a phylogenetic tree based upon sequences from loci cyt b and COI using accession numbers from the online database GenBank. Moreover, determining whether maternal DNA could reliably be extracted from the external surface of eggs using a swab, the QIAamp DNA mini kit for extraction and the ND2 gene for amplification. This method can then be used to identify the species in the egg.
Avian Species Identification
3. Method
3.1
The analysed avian species with scientific name and accession number
from GenBank
Table 3 displays the common name of the samples used in the first part of this research project. This is followed by the scientific name (genus and species). The third, fourth and fifth column respectively show the accession number for the ND2, cyt b and COI genes. These numbers are adopted from GenBank. The accession numbers for the ND2 gene are there to indicate what has already been done before. For some species the entire mitochondrial genome is already known, others have the whole ND2 gene in GenBank and some only the partial ND2 sequence. The accession numbers for cyt b and COI are used to build a phylogenetic tree (displayed in Figure 22 and 23) to compare the interspecies and intraspecies variation between the three loci.
It is important to note that no certainty can be given about the origin of the given samples. The common names were written on an envelope and are based on morphology and experience by the person who gave the samples for research.
Table 3: Common name of the avian species used with the scientific name and the accession numbers from GenBank for the loci ND2, Cyt b and COI.
Common name Scientific name Accession number GenBank
ND2 gene
Cyt b COI
Black-capped Parrot
Pionites melanocephalus EU327648.1
(Partially ND2)
JQ175846.1 JQ175845.1 Blue & gold
Macaw Ara ararauna HQ629720.1 (Complete ND2) DQ150994.1 AY286204.1 JQ174061.1 AB570297.1 Bourke’s Parrot Neopsephotus bourkii EU327639.1
(Partially ND2)
Derbyan Parrot Psittacula derbiana AF346388.1
Eclectus Parrot Eclectus roratus EU327619.1 (Partially ND2) AB177965.1 AY220101.1 Gang-gang Cockatoo Callocephalon fimbriatum JF414338.1 (Partially ND2) JF414311.1 JF414312.1 JF414283.1 JF414284.1 Glossy black Cockatoo Calyptorhynchus lathami JF414241.1 (Complete mt genome) JF414310.1 JF414282.1 Green-winged Macaw
Ara chloropterus AF346367.1 AB570307.1
AB570310.1 Hahn’s Macaw Diopsittaca nobilis EU327618.1
(Partially ND2)
JQ174705.1 JQ174704.1
Avian Species Identification
AF370775.1 Indian-ringnecked Parakeet Psittacula krameri HQ629739.1 (Complete ND2) GQ996517.1 AY220110.1 JF498892.1 JF498891.1Little Lorikeet Glossopsitta pusilla
Moustache Parrot
Psittacula alexandri AY220105.1
AY220106.1 AB177970.1
Mulga Parrot Psephotus varius EU327657.1
EU407718.1 (Partially ND2)
Ostrich Struthio camelus NC_002785.1
(Complete mt genome)
L78809.1 U76062.1
Pink Cockatoo Cacatua leadbeateri HQ629723.1 (Complete ND2) JF414320.1 JF414319.1 JF414292.1 JF414293.1 Plum-headed Parakeet Psittacula cyanocephala GQ996508.1 AY220109.1 Quaker Parrot Myiopsitta monachus EU327635.1
(Partially ND2) AF370778.1 AF370779.1 FJ027896.1 FJ027892.1 FJ027895.1 Red-rumped Parrot Psephotus haematonotus HQ316882.1 (Complete ND2) Regent Parrot Polytelis anthopeplus HQ316880.1
(Complete ND2)
AF346386.1
Scarlet Macaw Ara macao EU327601.1
(Partially ND2) AF346366.1 AF346363.1 JQ174064.1 AB570305.1 JN801493.1 Scarlet-chested P arrot Neophema splendida JQ175546.1 JQ175545.1 Short-billed white-tailed BC Calyptorhynchus latirostris JF414243.1 (Complete mt genome) JF414303.1 JF414302.1 JF414275.1 JF414274.1 Sulphur-crested Cockatoo Cacatua galerita JF414344.1 (Partially ND2) AF346394.1 AB177977.1 JF414289.1 JF414290.1 Yellow-crowned Parrot
Amazona ochrocephala AY194464.1
(Complete ND2) AY194412.1 AY194410.1 AY194408.1 JQ174005.1 JQ174004.1 Yellow-tailed BC (Eyre Peninsula)
Calyptorhynchus funereus EU327606.1
JF414333.1 JF414332.1 (Partially ND2) JF414307.1 JF414278.1 JF414279.1
3.2
Feathers
During this project three types of feathers were used. The feathers are categorized as small, medium and large. Figure 5 is an example of a small feather. This downy feather is from the Red-rumped Parrot.
Avian Species Identification
Figure 5: Downy feather of the Red-rumped Parrot.
The next Figure is an example of a middle sized feather. This feather is from a Hahn’s Macaw. Five barbs were used for this extraction and the section is highlighted in red.
Figure 6: A feather of a Hahn’s Macaw, described above as a middle size feather. In red is highlighted where the barbs were cut off.
Figure 7 shows a feather from a Glossy Black Cockatoo. This is one of the large feathers. For the extraction five barbs were used, where the barbs were removed is highlighted in yellow.
Avian Species Identification
Figure 7: Feather of a Glossy Black Cockatoo, one of the large feathers. In yellow the place where the five barbs where cut of is circled.
3.3
Mitochondrial DNA extraction using the QIAamp DNA mini kit
Several barbs were cut off from each sample. Those were cut up in pieces and placed in a 1.5 mL tube. 360 µL of ATL buffer with 30 µL proteinase K (QIAGEN, 10mg/mL) and 10 µL of dithiothreitol (Sigma, DTT, 1M stock) was added to each sample. Before incubating overnight at 56oC (Thermo
Mixer Compact Eppendorf), the samples were mixed by vortex (Vortex Mixer Ratek). After incubating, the samples were mixed again and centrifuged (Eppendorf minispin) briefly. To all samples 400 µL of AL buffer was added, followed by mixing by vortex and centrifuging the samples briefly. 400 µL of 99% ethanol was added to each sample. Before mixing and centrifuging for one minute at 6000 x g, the samples were incubated one minute at room temperature. Per sample, the entire lysate was carefully transferred to a mini spin column and centrifuged for one minute at 6000 x g. The column was placed in a clean collection tube afterwards. 500 µL of AW1 buffer was added to the centre of the column. The samples were centrifuged for one minute at 6000 x g. The flow-through was discarded and the collection tube reused before the addition of 500 µL of AW2 buffer. Hereafter, the columns were centrifuged for three minutes at full speed and placed in a clean 1.5 mL tube. To the membrane 30 µL of AE buffer was added. Finally, before centrifuging for one minute at full speed, the samples were incubated at room temperature for one minute.
Avian Species Identification
3.4
PCR avian species identification using barbs as a DNA source
The sequences of the ND2 primerset that has been used for amplification is mentioned in 2.4.1 For each of the PCR reactions 2 µL of DNA was used. The reaction mix contained 10 µL of 2x ‘MangoTaqTm DNA polymerase (Bioline), 1 µL of forward primer and 1 µL reverse primer:
concentration of 10pmol (Stock is 100pmol, Geneworks) and 7 µL H2O (1 extra µL per sample).
Some samples were amplified in duplicate and combined for purification.
The following program for the thermal cycler (Biorad C1000) was used: 95°C for 60 sec
95°C for 30 sec;
55°C for 30 sec; 40 cycles 72°C for 90 sec;
72°C for 5min
12°C hold indefinitely.
3.5
Sample visualization (2% gel) and preparation for purification
5 µL ethidiumbromide (Biorad 161-0433, 10 mg/mL) was added to a 2% gel (Agarose AppliChem 9012-36-6 and TBE 10x, diluted to 1x from Biorad 161-0770). The complete sample was added to a well. Some samples were amplified in duplicate; these were added together in each well (2x 20 µL of sample). No additional loading buffer was necessary. For the DNA ladder, 2.5 µL of a 1 kb DNA ladder (Quick-Load ladder, 50 µg/mL, Biolabs) or 5 µL of Easy Ladder I (100-2000 bp, Bioline) was loaded. The gel was run for ±30 min with approximately 130V (Thermo Electron Corporation EC105) and visualized under UV light (Gel Doc EZ Imager, Biorad).
Running the smaller samples, a 100 bp DNA ladder (Promega) was used; 2 µL of the ladder together with 3 µL of the Loading Dye.
Avian Species Identification
Figure 8: Quick-Load 1 kb DNA ladder on the left, showing the kilobases and the mass in ng. Easy Ladder I, in the middle showing the size in bp and the amount in ng per band and on the right a 100 bp ladder. .
3.6
Purify PCR products before sequencing (Bioline Isolate PCR and Gel Kit)
A 50oC heating block (Talboys block heater) was preheated. Using a clean scalpel, the DNA fragment
was excised (maximum of 300mg per spin column). The slice was transferred to a 1.5 mL tube. To each sample 650 µL of Gel Solubilizer was added before incubating for 10 minutes at 50oC, or until
dissolved. This was followed by the addition of 50 µL of Binding Optimizer and mixing by vortex. 750 µL of the sample was transferred to a Spin Column A, which was placed in a collection tube
beforehand. The columns were centrifuged for one minute at 10,000 x g. The filtrate was hereafter discarded and the collection tube reused. This step was repeated if necessary. Furthermore, 700 µL of wash buffer A (ethanol was added beforehand) was added to each sample and these were
centrifuged for one minute at 10,000 x g. Again, the filtrate was discarded and the collection tube reused. This wash step was repeated, followed by a centrifuging step of two minutes at full speed. The collection tube was discarded and the Spin Column A was placed in a clean 1.5 mL elution tube. 30 µL of Elution Buffer was added directly to the Spin Column A membrane. Finally, the samples were incubated at room temperature for one minute and centrifuged for one minute at 6000 x g.
Some samples were purified with the QIAquick Gel Extraction Kit. This protocol is not displayed in this report since the Bioline Isolate PCR and Gel Kit appeared to work better, was cheaper and easier in use. The protocol for the QIAquick Gel Extraction Kit has the same construction as the Bioline Isolate protocol. The gel was cut out, a solution buffer was added, followed by a wash buffer and finally the samples were eluted.
Avian Species Identification
3.7
Quantification
For quantification of the purified DNA the Nanodrop (Thermo Scientific Nanodrop 1000
Spectrophotometer) and a 2% gel were used. 5 µL of purified DNA was loaded with 2.5 µL of 6x Blue and Orange Loading Dye (Promega) or 2.5 µL of 5x Red Loading Dye (Bioline). The DNA and loading dye were mixed together on a piece of parafilm (Pechiney Plastic Packaging). As a DNA ladder the 1 kb ladder or the Easy Ladder I was run with it.
3.8
Sequencing
The purified samples were send off to the Australian Genome Research Facility ltd (AGRF) for sequencing.
Beforehand, the samples were prepared by using 30-80 ng of DNA, 10 pmol of the forward primer, with a finale volume of 12 µL, using sterile H2O.
3.9
Sequence comparison and tree construction
The quality of the sequences was determined using Geneious. With Geneious, approximately 30 bases at the beginning and 15 bases at the end of the sequences were cut off. Degenerate bases (section 2.4.4) were edited into the sequence if necessary. The sequences were saved as a text file and imported in MEGA. Using Clustal W, the samples were aligned. None of the parameters were changed. After aligning the sequences a phylogenetic tree was build using five different methods; Maximum Parsimony (MP), Maximum Likelihood (ML), Neighbor-joining (NJ), Minimum Evolution (ME) and UPGMA (Unweighted Pair Group Method with Arithmetic Mean). No preferences for building the tree were altered.
3.9.1
Tree construction with cyt b and COI sequences from GenBank
The accession numbers in Table 3 were used for building a phylogenetic tree for the loci cyt b and COI. The numbers were copied into MEGA and aligned using Clustal W. No parameters were changed. The Minimum Evolution method was used to build the trees.
3.9.2
Comparison unknown samples to ND2 tree
Three unknown samples were analysed following the protocols above. The samples were added to the other ND2 data. An attempt was made to identify the unknown samples to validate the use of the ND2 locus for species identification.
Avian Species Identification
3.10 Maternal trace DNA from the external surface of a chicken eggshell
Several spots on a chicken egg were marked with a circle and a number was given for each swab (4 NG FLOQ swabs crime scene and cotton tips applicators from Livingstone) as shown in Figure 9. Marking the swabbed spots was important so a spot would not be swabbed twice. A wet swab was done first, followed by a dry swab on the same spot. Both ends of the swab were placed in the same tube.
Figure 9: Parts of the chicken egg are circled where a swab was taken for DNA extraction
The same protocol for DNA extraction was used as in 3.3, only with a few adjustments:
380 µL ATL was added to each sample, together with 10 µL proteinase K and 10 µL DTT. The samples were incubated for 15-20 minutes with regularly mixing by vortex in between.
A primerset, giving an amplicon of 195 bp was used as mentioned in section 2.4.1. 8 µL of DNA was added and another thermo cycle was used:
95°C for 60 sec
95°C for 30 sec;
61°C for 30 sec; 1 cycle 72°C for 90 sec;
95°C for 30 sec;
59°C for 30 sec; 1 cycle 72°C for 90 sec;
95°C for 30 sec;
57°C for 30 sec; 1 cycle 72°C for 90 sec;
95°C for 30 sec;
55°C for 30 sec; 1 cycle 72°C for 90 sec;
Avian Species Identification
95°C for 30 sec;
53°C for 30 sec; 36 cycles 72°C for 90 sec;
72°C for 5min
12°C hold indefinitely.
3.10.1 Positive control
A small piece of chicken breast was used as a positive control. This piece was placed in a 1.5 mL tube. The QIAamp DNA mini kit was used to extract DNA. 180 µL of ATL buffer with 20 µL proteinase K (QIAGEN, 10 mg/mL) was added. Before incubating for one hour at 56oC (Thermo Mixer Compact
Eppendorf), the samples were mixed by vortex (Vortex Mixer Ratek). The sample was frequently mixed by vortex until the tissue was completely lysed. 200 µL of AL buffer was added, followed by mixing by vortex and centrifuging the samples briefly. 200 µL of 99% ethanol was added to the sample. The sample was mixed thoroughly by pulse vortexing for 15 seconds and incubated at room temperature for 5 minutes. The sample was centrifuged briefly before transferring the entire lysate to a mini spin column. This column was centrifuged for one minute at 6000 x g. The flow-through was discarded and the column was placed in a clean collection tube. 500 µL of AW1 buffer was added to the centre of the column. The sample was centrifuged for one minute at 6000 x g, followed by the addition of 500 µL of AW2 buffer. Hereafter, the column was centrifuged again for one minute at 6000 x g. The flow-through was discarded and the column was centrifuged at full speed for three minutes to dry the membrane completely. The column was placed in a clean 1.5 mL tube. To the membrane 50 µL of AE buffer was added. Finally, before centrifuging for one minute at full speed, the samples were incubated at room temperature for one minute.
3.10.2 Forward primer design
Mega was used to align the avian species from the previous part of the project with the ND2 sequence from a chicken to determine a conserved region close to the reverse primer. The sequences of the avian species were compared to the sequence from the chicken and as shown in Figure 10; this part shows the least differences for all species close to the reverse primer (sequence reverse primer not shown in Figure).
Oligo calculators (http://www.basic.northwestern.edu/biotools/OligoCalc.ht mL and
Avian Species Identification
parameters, such as the formation of hairpins, primer dimer, percentage GC and the melting temperature.
Figure 10: Data sequence in MEGA showing the most conserved part of the ND2 close to the reverse primer, used to create a new forward primer resulting in a smaller amplicon.
3.10.3 Second round PCR
A second round PCR was conducted. The PCR products of all DNA extractions for the outer surface of the eggshell were combined (in results shown as sample A – E). 2 µL of this mixture was used for another PCR (or 4 µL when the final volume of the PCR was 20 µL). The same Touchdown PCR was run, only with a total volume of 10 µL; 5 µL Mango Mix, 2 µL of the PCR mixture, 1 µL each primer and 1 µL of H2O. A negative (5 µL Mango Mix, 1 µL each primer and 3 µL of H2O) and positive control
(5 µL Mango Mix, 1 µL of chicken DNA, 1 µL each primer and 2 µL of H2O) were run as well. This
resulted in a total cycle number of 80.
3.10.4 ExoSAP-it
After purifying the samples using the Bioline Isolate Kit, there was no DNA left. Even though there was enough DNA present beforehand. Therefore, some egg samples were purified using ExoSAP-it from USB. 5 µL of post PCR reaction product was mixed with 2 µL of ExoSAP-it, for a combined 7 µL reaction volume. The samples were incubated at 37oC for 15 minutes to degrade the remaining
primers and nucleotides. Afterwards the samples were incubated at 80oC for 15 minutes to inactivate
ExoSAP-it. The samples were then ready for sequencing.
4. Results
Sections 4.1 up to 4.6 show the results using barbs as a DNA source. Section 4.7 and 4.8areresults from DNA extraction from the eggshell.
Avian Species Identification
4.1
Quality of the sequence
The quality of the DNA sequence data was found to vary greatly. This did not seem to be affected by the quantity of DNA submitted to AGRF and neither by the quality. Figure 11 and 12 show differences in the quality of two sequences after analysis using the software Geneious.
Figure 11 is a partial sequence from the Blue and Gold Macaw I, visualised in Geneious. This
sequence has been used for building a phylogenetic tree as quality of the data is very good and there are no bases that might be miscalled. Geneious colour codes the quality of the sequences and the lighter blue the bases are coloured, the better the quality is. The better the quality means that more certainty can be given about what base is on what position. This part of the sequence is of good quality since the peaks are smooth and separated well.
Figure 11: The sequence of the Blue and Gold Macaw I, from Geneious. This sequence is used for building a phylogenetic tree, based upon the quality.
Figure 12 shows a part of the sequence from the Glossy Black Cockatoo I. This sequence is not useful for aligning and building a tree. The quality is low, indicated by dark blue colours. It is hard to tell if the two C’s in the red square in the Figure below are actually two C’s.
Avian Species Identification
Figure 12: Part of the sequence from a Glossy Black Cockatoo, visualized in Geneious.
As said in section 3.9 approximately 30 bases were cut off from the beginning of the sequence and 15 at the end. Figure 13 and 14 show examples of the beginning and end that are cut off.
Figure 13: First 30 bases cut off from sample 4-1 (top) and 4-2 (bottom) (samples shown in Table 4), sequence quality is low.
Figure 14: Sample 18 (top) and 42 (bottom) (Table 4) where the last 15 bases were cut off due to low quality.
Some bases in the middle of the sequence did not have a good quality, it was not clear what base it was. Figure 15 and 16 show an example of bases that were edited using Geneious, replaced by a degenerate base. Figure 15 shows a larger part of sample 42 (2). The degenerate primers are underlined with a yellow block. Figure 16 is zoomed in on a part of the sequence to clarify the degenerate bases. Table 2 shows which degenerate primer there are and what these stand for. For example, two peaks are visible in the black square; a green and blue one. A ‘T’ is indicated with a
Avian Species Identification
green peak and a ‘C’ is indicated with a blue peak. Assuming the base is either a ‘T’ or a ‘C’, the degenerate base ‘Y’ is used.
Figure 15: Partially sequence of sample 42 (2), the yellow blocks indicate a modification in a base. When it was unclear what the base was at that position a degenerate base replaced the base. Geneious was used to edit the sequence.
Figure 16: Same sequence as the previous Figure, only another part and zoomed in to clarify the adjustment made from a given ‘T’ to a degenerate ‘Y’, because the peak for a ‘C’ is almost as high.
4.2
Quantity of barbs and quality of DNA for successful sequencing
It seems that a high quality of DNA does not necessarily result in a good sequence, neither that a high concentration of DNA results in a good sequence. Table 4 shows each sample number and its
corresponding Latin name with the presence of a band on a gel after PCR, the quantification using the Nanodrop and/or Gel electrophoresis and the quality of the sequence given by the software
Geneious. Between 30 and 80 ng of DNA is necessary for sequencing (section 3.8) and approximately 50-70 ng was used for each sample. No attention was paid to the concentration of the extract for amplification. The quantity and quality of the DNA extracted from the feather barbs is shown in Table 4.
Avian Species Identification
extraction was done using the whole downy feather. This again resulted in no DNA amplification after visualisation on an agarose gel. A conclusion is that there is not enough DNA present in the downy feather to produce a PCR product using this primer set and using the QIAamp DNA mini kit to extract any DNA.
The extracted DNA product was run on a gel for sample 6, 12, 20-22 (2x), 25 (2x) and 40, since no amplification was noted on a gel after PCR. None of these samples showed that there was DNA extracted. As written above, for sample 20, 25 and 40 the downy feather might not have contained enough DNA. However, for the other samples a larger feather was used. It might be that these samples also did not contain enough DNA either.
When sample 21 and 22 were analysed again, there was a product after PCR. After purification almost no product was visible on a gel, which might indicate that for these samples there was a major DNA loss during gel purification. No third attempt was made, since the sequence data was already known for these samples.
Samples 18, 42 (2), 43-46 were amplified in duplicate and added together before purification. However, this does not seem to have a major advantage compared to the other samples in quantity and quality of the DNA.
Sample 46 and the ostrich feather both gave good sequences, even though the concentration differed (8 ng/ µL for the ostrich sample and 20 ng/ µL for sample 46). Moreover, sample 41 and 42 had the same quantification but there is a difference in the sequence quality. Sample 41 has a HQ% of >95, whereas sample 42 has a HQ% of 45%. In addition, the second time sample 42 was analysed the sequence quality was much better (85%) compared to the first analysis. Samples 16 and 17 both had a concentration of approximately 8 ng/ µL, but the sequence quality of sample 16 is not as good as the quality of sample 17. As stated above, this shows that the quality and quantity of a sample says little about the outcome of the sequence.
No conclusion can be drawn about the necessary concentration of purified DNA for reliable sequencing, especially knowing that the Nanodrop and Gelelectrophoresis are not accurate in determining the concentration.
Avian Species Identification
Table 4: Per sample the presence of a band after PCR, quantification using the Nanodrop and gel after purification for every extracted sample and the quality of the sequence based upon the software Geneious is indicated.
Sample nr Sample name Presence
band after extraction1 Presence band after PCR Quantification Nanodrop after purification (ng/ µL) Quantification gel after purification (ng/ µL)2 HQ %3 2-34 Calyptorhynchus funereus No information ±36 >95% 2-44 Calyptorhynchus funereus No information ±37 >95% 2-54 Calyptorhynchus funereus No information ±53 >95% 4-14 Psittacula alexandri No information ±38 >95% 4-24 Psittacula alexandri No information ±39 >95% 4-34 Psittacula alexandri No information ±30 >95% 6 Psittacula alexandri - - - X5 6 (2) - + ±15 0% 7 Diopsittaca nobilis + ±15 90-95% 8 Diopsittaca nobilis + ±15 95% 9 Diopsittaca nobilis + ±15 90-95% 10 Diopsittaca nobilis + ±15 >95% 11 Diopsittaca nobilis + ±15 >95% 12 Diopsittaca nobilis - - - - X 13 Ara chloropterus + ±12 90% 14 Ara chloropterus + ±12 85% 15 Calyptorhynchus funereus (Eyre Peninsula) + ±10 >95% 16 Calyptorhynchus funereus (East Coast) + ±8 70% 17 Calyptorhynchus funereus (Eyre Peninsula) + ±8 >95% 18 Primolius maracana + ±6 50% 19 Ara macao + ±11 90-95% 20 Psephotus varius - - - - X 20 (2) - - - - X 21 Cacatua leadbeateri - - - - X 21 (2) - + - - X 22 Cacatua - - - - X
1 Since there was no amplification, the extracted DNA samples were run on a 2% gel to determine whether there was DNA
present before amplification.
2 Nanodrop was either not available or gave results which seemed to make no sense (concentrations lower than 0 or a
minus ratio 260/280) A 2% gel was used to estimate the amount in ng/ µL. The estimation was done by supervisor Adrian Linacre.
3 Percentage of the overall quality of the sequence, results from Geneious 4
Avian Species Identification
leadbeateri 22 (2) - + - - X 23 Polytelis anthopeplus + ±9 >95% 24 Polytelis anthopeplus + ±7 80% 25 Psephotus haematonotus - - - - X 25 (2) - - - - X 26 Ara ararauna + ±12 >95% 27 Ara ararauna + ±11 >95% 28 Calyptorhynchus lathami + ±5 0% 29 Calyptorhynchus lathami + ±10 80% 30 Calyptorhunchus latirostris + ±4 X 31 Calyptorhunchus latirostris + ±6 90-95% 32 Neophema splendida + ±8 80% 33 Psittacula krameri + ±8 15% 34 Psittacula cyanocephala + ±20 >95% 35 Psittacula derbiana + ±<5 X 36 Amazona ochrocephala + ±15 >95% 37 Cacatua galerita + ±8 90-95% 38 Callocephalon fimbriatum + ±20 >95% 39 Pionites melanocephalus + ±20 >95% 40 Neopsephotus bourkii - - - - X 41 Glossopsitta pusilla + ±15 >95% 42 Primolius maracana + ±15 45% 42 (2) ±15 85% 43 Myiopsitta monachus + ±18 >95% 44 Myiopsitta monachus + ±18 >95% 45 Eclectus roratus + ±20 >95% 46 Eclectus roratus + ±20 >95%Ostrich Struthio camelus + ±8 >95%
Unknown sample I - + ±10 30% Unknown sample II - + ±20 75% Unknown sample III - + ±10 40%
Avian Species Identification
4.3
Different methods for constructing a phylogenetic tree
Since there are several computer software programmes to construct a phylogenetic tree based on DNA data, it is not possible to guarantee which one is the ‘true’ tree. Five different methods (Maximum Parsimony (MP), Maximum Likelihood (ML), Neighbor-joining (NJ), Minimum Evolution (ME) and UPGMA (Unweighted Pair Group Method with Arithmetic Mean)) were used in MEGA to construct a phylogenetic tree; these are displayed in Figures 17 to 20. These five methods are among the most commonly used programmes to compare two or more DNA sequences to determine the relationships. For constructing these phylogenetic trees DNA sequence data from the ND2 locus were used and these samples are shown in Table 4. The method MP was used to construct the
phylogenetic tree displayed in Figure 17. The green square indicates the scale for this tree. A 100% match between two samples would results in a scale of 0,00. A smaller scale, like the one for NJ (Figure 19; 0,02) shows a dissimilarity of 2% for every length of the scale. The meaning of the scale of MP is unknown. A major anomaly for both the MP and ML method is that they cluster the
Struhio camelus to the Neophema splendida, instead as an out-group, this is highlighted by a light
blue square. This indicates that these two trees are not reliable as the ‘true’ trees.
NJ and ME constructed the same tree and hence only one tree is shown (Figure 19). Next to it is a phylogenetic tree which has been constructed using the UPGMA method. The scale is different in both trees (indicated in green) and not all species are grouped in the same manner. The yellow square indicates the species that are grouped together in the same way for both of these methods. This is also the case for the MP and ML method. All macaws and some parrots are grouped together, which all belong to the same family (Psittacidae). It makes sense that these samples cluster together since these species are considered to share a common ancestor. Also the cockatoos are grouped together in all methods (NJ / UPGMA / MP / ML); only the clustering is different. All cockatoos are from the same family, Cacatuidae, so it is assumed that these samples would cluster together. The difference in the clusters generated between some samples in using the ML and MP method are highlighted with a lighter red square.
Moreover, the Psittacula cyanocephala and Eclectus roratus (indicated in brown), are both clustered together using all methods. Based upon this, the two species are genetically similar and it appears that these two species have a recent shared ancestor in common. Both are parrots and belong to the same family (Psittacidae) but have a different genus name. Since the parrots belong to the same family, it is plausible that these species cluster together. However, it would also be expected that these species do not cluster as closely as they are not members of the same genus.
Avian Species Identification
For all phylogenetic methods the data from the Amazona ochrocephala and the two Psittacula
alexandri cluster together. These data indicate that these samples have the same sequence even
though they are classified as being in different genera. There are three possible explanations for the clustering of these samples, which actually applies to every sample analysed. It is possible that the taxonomic classification is incorrect. A lot has already been changed since the development of the taxonomic classification system by Carolus Linnaeus in the 1700s . Maybe these samples should also be grouped in the same genus. Secondly, the name of the sample provided has been identified incorrectly. This original identification was done based upon morphology and experience of the zookeeper. Thirdly it might be that the ND2 gene is not the best loci for interspecies variation and if a different genetic locus has been used that these species may have been separated.
The main difference between the NJ/ME and UPGMA methods is the grouping of the Glossopsitta
pusilla and the Neophema splendida, which seem to cause a different clustering. UPGMA clusters
these two species together, whereas both NJ/ ME do not. Based upon the given classification names, it would make more sense for these samples not to cluster closely together since, while they are in the same taxonomic family, they are in different genera.
None of these trees can be identified as the ‘true’ tree, only differences between the methods can be recorded. One major anomaly using the ML and MP method is not having an out-group. This makes these trees not the most suitable trees. Both the UPGMA and the NJ/ME method show little intraspecies variation. Both methods group the Calyptorhynchus funereus EP closer to the
Calyptorhynchus latirostris than to the other samples from the same species. Besides that, both
methods indicate virtually no intraspecific variation.
Even though the UPGMA method is hardly used anymore in phylogenetics, the reason being this is the most simple method for construction a tree , it appears to be the best method for constructing a phylogenetic tree when using the data generated from ND2 samples. The UPGMA method exhibited the fewest anomalies when compared to the NJ/ME, MP and ML methods. Highlighted with a dark blue square, the families (Psittacidae and Cacatuidae) are separated as expected and the samples are clustered on family level and mainly on genus level as well. The NJ/ME method does not cluster the two families as clear as in the tree built using the UPGMA method.
Avian Species Identification
Figure 17: Phylogenetic tree generated using DNA sequence data from ND2 locus.
The samples used are listed in Table 4. MP was used to construct this tree. In green the scale (10) is highlighted and in blue the clustering of a parrot (Neophema splendida) with the out-group (Struthio camelus)
Figure 18: Phylogenetic tree generated using DNA sequence data from ND2 locus.
The samples used are listed in Table 4. ML was used to construct this tree. In green the scale (0.05) is highlighted and in blue the clustering of a parrot (Neophema splendida) with the out-group (Struthio
Avian Species Identification
Figure 19: Phylogenetic tree generated using DNA sequence data from ND2 locus.
The samples used are listed in Table 4. NJ/ME was used to construct this tree. In green the scale (0.02) is highlighted. Yellow, orange and red highlight grouping that are the same for this method compared to the UPGMA method. Only the Glossopsitta pusilla and the Neophema
splendida group differently (pink squares) resulting in a different clustering.
Figure 20: Phylogenetic tree generated using DNA sequence data from ND2 locus.
The samples used are listed in Table 4. UPGMA was used to construct this tree. In green the scale (0.00 to 0.10) is highlighted. Yellow, orange and red highlight grouping that are the same for this method compared to the NJ/ME one. Only the Glossopsitta pusilla and the Neophema splendida group differently (pink square) resulting in a different clustering.
