1. Evaluation of the phylogeny of derived Eupolypods (Monilophytes)
with special emphasis on Woodsiaceae, Dryopteridaceae and
Tectariaceae.
Vishnu Mohanan
B.A. (Mod) Thesis
2011
School of Botany
University of Dublin
Trinity College Dublin
2. Declaration
I, Vishnu Mohanan, declare that this thesis is my own work, except where otherwise stated,
and that it is words in length 9326, excluding the abstract and references.
3. Acknowledgements
I would like to sincerely thank Dr. Trevor Hodkinson and Dr. Steve Waldren for all the
guidance they have given me during the course of this project, without whom this project
would ever have been completed.
4. Table of Contents
Declaration......................................................................................................................................... 2
Acknowledgements......................................................................................................................... 3
Abstract ............................................................................................................................................... 6
Chapter 1: Introduction ................................................................................................................. 7
Leptosporangiates ......................................................................................................................................7
Polypod diversity ........................................................................................................................................9
Woodsiaceae ..............................................................................................................................................11
Dryopteridaceae.......................................................................................................................................11
Tectariaceae...............................................................................................................................................13
Aims...............................................................................................................................................................14
Molecular Systematics Approaches...................................................................................................15
Electrophoresis and PCR...................................................................................................................................15
Gene regions...........................................................................................................................................................15
Maximum Parsimony..........................................................................................................................................16
Bayesian Analysis.................................................................................................................................................17
Chapter 2:Materials and Methods............................................................................................18
Plant Material ............................................................................................................................................18
DNA Extraction..........................................................................................................................................18
Gel Electrophoresis..................................................................................................................................19
DNA Purification.......................................................................................................................................19
PCR.................................................................................................................................................................20
Taxon sampling and Alignment ..........................................................................................................23
Phylogenetic Analysis: Bayesian.........................................................................................................23
Phylogenetic Analysis: Maximum Parsimony................................................................................24
Multi-gene analysis..................................................................................................................................24
Visualising the Trees...............................................................................................................................24
Chapter 3:Results...........................................................................................................................26
Alignment....................................................................................................................................................27
Phylogenetic trees....................................................................................................................................27
atpB gene set (Figure 5).........................................................................................................................28
rbcL gene set (Figure 6) .........................................................................................................................29
atpB-rbcL gene set (Figure 7) ..............................................................................................................29
5. Chapter 4:Discussion....................................................................................................................49
Eupolypods II: Woodsiaceae (Fig)......................................................................................................49
Eupolypod I: Tectariaceae (Fig)..........................................................................................................49
Eupolypod I: Dryopteridaceae (Fig) ..................................................................................................50
Critique on the Methods Employed....................................................................................................52
Conclusion...................................................................................................................................................53
References...................................................................................... Error! Bookmark not defined.
Appendix......................................................................................... Error! Bookmark not defined.
6. Abstract
The principal aim of the study the overall phylogeny of the derived polypods within the
leptosporangiates and alleviate conflicts within authors on their position within the lineage
focusing specially on the Woodsiaceae, Tectariaceae and Dryopteridaceae families. The basal
leptosporangiates have been studied quite extensively, but the there has always been
controversy on circumscribing families within the polypods (Kramer et al., 1990, Hasebe et
al., 1995, Pryer et al., 2004, Smith et al., 2006).
This was done through extensive sampling where phyloegenetic analyses in the form of
Bayesian inference and Maximum Parsimony, was carried out on three data sets: atpB, rbcL
and a combined data set of atpB and rbcL.
Phylogenetic analyses showed the monophyly of the three families with the eupolypod I and
the eupolypod II clade being formed. Support was found to segregate Didymochlaena,
Leucostegia and Hypodematium from the Dryopteridaceae family as it renders the family
paraphyletic. Traditionally tectarioid fern Pleocnemia was found to have close affinity to
Lastreopsis and suggested to by circumscribed under the Dryopteridaceae family.
7. Introduction
Leptosporangiates
Earths vegetation changed dramatically from one dominated by gymnosperms and seed-free
vascular plants to one dominated by angiosperms over the course of about 80 million years.
As flowering plants rose to dominance, we saw other plant lineages relegated to the
sidelines and in some case driven to extinction (Lupia et al., 1999). Today, angiosperms
account for 96% of the vascular plant diversity with the other plant lineages contributing just
a handful of species or maybe a 100 species. The only exception to this is the
leptosporangiate clade, which consists of ≈ 9,000 species and more than 250 genera all
around the world. This is almost four times the number of extant species in all other non-
flowering lineages combined (Table 1)
Table 1: Estimated numbers of extant vascular plant lineages. Total species count represents a conservative
consensus drawn from several sources. Percentages in table indicate the contribution of lineages to total
species diversity. Adapted from (Schuettpelz and Pryer, 2009).
Vascular Plant lineage Total Species
Lycophytes
Quillworts 150 (< 1 %)
Clubmosses 380 (< 1 %)
Spikemosses 700 (< 1 %)
Ferns (Monilophytes)
Whisk ferns 12 (< 1 %)
Horsetails 15 (< 1 %)
Ophioglossid ferns 80 (< 1 %)
Marattioid ferns 150 (< 1 %)
Leptosporangiante ferns 9,000 (< 3 %)
Seed Plants (Spermophytes)
Ginkgo 1 (< 1 %)
Gnetales 80 (< 1 %)
Cycads 130 (< 1 %)
Conifers 600 (< 1 %)
Angiosperms 260,000 (< 96 %)
To understand the leptosporangiate clade, we have to hark our eyes back to 400 millions
years ago (mya) to the early-mid Devonian where a deep a phylogenetic dichotomy occurred
separating the lycophytes from the group that consists of all other vascular plants, namely
the euphyllophytes (Kenrick and Crane, 1997) (Figure 1). These euphyllophytes,
8. characterized by euphylls, (Stein, 1993) diverged into Spermophytes (seed bearing plants)
and Monilophytes (=Infradivision Moniliformopses, sensu Kenrick and Crane, 1997). The
monilophytes are united by the presence of a distinctive vasculature, different from the
spermophytes, where the protoxylem is confined to the lobes of the xylem strand (Stein,
1993). It is because of this they are known as monilophytes due to the Latin appellation of
moniliformis or ‘necklace like’. The monophyly of this group has been known for a while
through studies on its morphology including fossil taxa, studies of sperm ultrastructure and
analyses of DNA sequence data (Pryer et al., 2004). The extant members of the monilophytes
are found in five major lineages that we know today (Pryer et al., 2001): whisk ferns
(Psilotales), ophioglossid ferns (Ophioglossales), Mattaroid ferns (Marattiales), horsetails
(Equisetopsida) and leptosporangiate ferns.
Figure 1: Consensus tree showing the relationships among the major lineages of vascular plants sourced from
Pryer et al. (2004). Clades indicated on the tree: T = tracheophytes, L = lycophytes, E = euphyllophytes, S =
spermophytes, M = monilophytes. Black band refers to the membership of the respective lineages to the
corresponding taxon listed across the top (non-monophyletic taxa are in quotes)
The leptosporangiates, which are one of the most familiar of the monilophytes have a
monophyletic assemblage and consist of roughly 9000 species. These ferns are characterised
by several apomorphic characters that have been used extensively in taxonomy and
classification. These include, but not limited to, the presence of leptosporangia that
9. developed from a single cell and the occurrence of mature sporangial walls that are one cell
thick. Most of these sporangia possess a distinctive annulus that is used to eject spores for
dispersion and the size and shape of these annuli, as well as the shape of the sorus have
been used for taxonomic purposes as well.(Kramer et al., 1990)
The earliest known occurrence of leptosporangiate ferns is about 300 mya in the early
Carboniferous, before the evolution of angiosperms, which would have been at the end of
the Permian. Based on fossil records it has been hypothesised that leptosporangiates
underwent three successive radiations. The first occurred in the early Carboniferous and the
second in the Permian and early Triassic (Rothwell, 1987). The third radiation began in the
early Cretaceous and is ongoing today in the form of the polypod clade (Schneider et al.,
2004).
Polypod diversity
The polypod group of ferns is the heart of leptosporangiate diversity and constitutes more
than 80% of the living ferns. They have been shown to have diversified in the Cretaceous
after the angiosperms and this diversification occurred in the ‘shadow’ of the angiosperms
suggesting an ecological opportunistic response by the ferns (Schneider et al., 2004). With
the dominance of the angiosperms at the time, new niches were formed into which the ferns
diversified and this is evident by the fact that most of these ferns are epiphytic in nature.
Schneider et al. (2004) showed that in the early Cretaceous, the derived polypods split into
two major clades: the pteridoids (Pteridaceae and Dennstaedticeae) and the eupolypods
(Figure 2). The eupolypods diversified faster than the pteridoids in the Early Cretaceous,
forming the two clades we know today, the eupolypods I and the eupolypods II.
The Eupolypods are a hyperdiverse group found all across the world, constituting 67% of
fern diversity. Monophylly of this group has been well established through several single and
multi-gene phylogenetic analyses (Hasebe et al., 1994, Hasebe et al., 1995, Pryer et al., 2004,
Smith et al., 2006, Schuettpelz and Pryer, 2007), but this is not the case for the two lineages
that lie within.
Several authors have have suggested more work needs to be done on the families within
these two clades as the overall phylogenetic picture is far from equivocal (Murakami et al.,
1999, Pryer et al., 2004). This study will focus on the two eupolypod clades of the
leptosporangiates, specifically the three families of Woodsiaceae, Tectariaceae and
Dryopteridaceae. These three families resolved with very poor bootstrap support in Smith’s
10. (2006) consensus tree (Figure 2) and are plagued with conflicting conscriptions and
classification.
Figure 2: Consensus tree sourced from Smith (2006) outlining the major clades within the monilophytes.
Resolved nodes have a bootstrap support ≥ 70 except those drawn with dotted lines. Family, order, and class
names that correspond to our classification are indicated to the right. Common names for some larger clades
are indicated to the left.
11. Woodsiaceae
Woodsiaceae as defined by Smith (2006) is rendered paraphyletic according to Schuettpelz
(2007). Schuettpelz (2007) resolves Cystopteris and Gymnocarpium together (which Smith
(2006) puts in Woodsiaceae) as sister to the rest of the eupolypods with 100% bootstrap
support. Hemidictyum which Smith (2006) put in Woodsiaceae is resolved as sister to the
asplenioid ferns, whilst Woodsia is resolved as sister to a large clade of onocleoid, blechnoid
and athyrioid ferns.
The athyrioid ferns (85% of which are contained in the two genera Athyrium and Diplazium)
are resolved as a monophyletic clade with 96% bootstrap support. Schuettpelz (2007)
showed Deparia as being sister to the clade consisting of Athyrium and Diplazium using
multi-gene analyses. This study will try to comment on this conscription using two-gene
analyses.
Athyrium is resolved as being paraphyletic, however this might be the case of too few
sample sizes. Ma-Li used a higher sample size and resolved Athyrium as a monophyletic
clade (Ma-Li et al., 2003). The reverse is true for Diplazium, as it was resolved as a
monophyletic clade in Schuettpelz (2007), but a paraphyletic clade in Ma-Li (2003). Hopefully
by the end of this study, it will be possible to safely comment on the structure of the two
genera within the athyrioid ferns.
Woodsiaceae is described as having a terrestrial habit with a sub-cosmopolitan distribution,
characterised with creeping rhizomes that could be either ascending or erect. Petioles have
two elongate or crescent shaped vascular bundles facing one another that unite distall in to
form a gutter-shape. The leaf blades are monomorphic or sometimes dimorphic with pinnate
or forking veins. The sori are abaxial, round, J-shaped or linear, with the indusia either
absent or if present, reniform to linear in shape. The spores can be reniform, monolete,
perine-winged, ridged or spiny (Smith et al., 2006).
Dryopteridaceae
The Dryopteridaceae family is one of the most studied families within the eupolypods. It is
resolved by as a monophyletic clade (Smith et al., 2006, Schuettpelz and Pryer, 2007)
although both the authors call for further sampling within the clade before a clearer picture
is derived.
12. Didymochlaena, Hypodematium and Leucostegia form a clade outside the traditional
dryopterids and before Schuettpelz (2007), had never been analysed together. This is a
recent derivation as the three genera have traditionally been associated with other families,
namely dryopterid, athyrioid and davallid ferns respectively. If the above three genera are
included with the dryopterids, they would render the clade paraphyletic. This project will try
to find congruence with Schuettpelz (2007) and comment on whether they should be
conscripted within Dryopteridaceae.
Dryopteris, Polystichum and Elaphoglossum are extensively studied within Dryopteridaceae
with the three forming monophyletic clades within the family, with varying degrees of
support. Dryopteris and Polystichum compose a well-supported clade according to
Schuettpelz (2007) receiving 97% bootstrap support. They also found Polystichopsis (which is
usually synonymous with Arachinoides) sister to the clade consisting of dimorphic epiphytes
namely Cyclodium, Polybotyra, Maxonia and Olfersia. Smith (2006) did not comment on this
relationship, so it will be looked at in this study.
Lastreopsis, Rumohra and Megalastrum form a clade that is sister to the ‘former
lomariopsids’ clade that consists of Elaphoglossum, which resolved with 100% bootstrap
support in Schuettpelz (2007). Authors have asked for the resurrection of the family
Elaphoglossaceae, which narrowly defined is monophyletic (Skog et al., 2004a). However,
resurrecting this would render Dryopteridaceae paraphyletic. In this study we will comment
on whether there is a need to adopt Elaphoglossaceae as a family unto itself.
The Dryopteridaceae is described as having a terrestrial habit generally found growing on
rocks or epiphytic with a pan-tropical or temperate distribution. The rhizomes can be
creeping, ascending or erect and in some cases scandent with the petioles having numerous
round vascular bundles arranged in a ring-like fashion. Leaf blades are usually monomorphic
and in some cases dimorphic with scales, glands or hairs (rarely) with pinnate or forking
veins. The sori are usually round with the indusia round, reniform or peltate. In some
lineages the sori are fused and found to be lacking an indusium. The sporangium is 3-rowed
with short to long stalks and the spores are reniform, monolete or perine-winged (Smith et
al., 2006).
13. Tectariaceae
Schuettpelz (2007) commented very little on the phylogenetic structure of Tectariaceae.
They found that Arthropteris and Psammiosorus to be sister to the rest of the tectarioid
ferns with 75% bootstrap support. However Tectariaceae with the inclusion of the two
genera would be hard to describe morphologically (Smith et al., 2006).
There are still questions on the generic delimitations of Tectaria. Smith (2006) defines it as
being monophyletic if we include Heterogonium in Tectaria s.l., which is what Schuettpelz
finds as well. In this study, we aim to evaluate Tectariaceae with a higher sample size to see
the delimitation within the family.
It is described as having a terrestrial habit with a pan-tropical distribution, characterised with
short creeping to ascending rhizomes that are disctyostelic. The petioles are non-abscissing
with a ring of vascular bundles in cross-section. The leaf blades are pinnate, bipinnate or in
some cases de-compound with free or highly anastomosing veins. The spores are brownish,
reniform, monolete or ornamented (Smith et al., 2006).
14. Aims
The project aims to produce single gene phylogenetic trees (Huelsenbeck et al., 2001) for
Dryopteridaceae, Tectariaceae and Woodsiaceae utilising two chloroplast gene regions: rbcL
and atpB. This will be used to create a supermatrix (combined dataset) and will be used for
phylogenetic inference.
Major radiations amongst the families will be examined using the above-created trees. This
project will aim to evaluate classification of certain families and propose new ones if
necessary. The result from these studies can be then used for further studies into the
systematics of ferns, and help answer broader questions regarding leptosporangiates, like
points in time when key diversification events regarding morphology and habitat took place.
To answer these types of questions, one must have a strongly supported phylogeny, which I
hope this project will help in establishing.
The secondary aim is to obtain sequence data for several ferns that have yet to be
sequenced including Lastreopsis pacifica and to position these taxa on the phylogenetic
trees of the leptosporangiates.
15. Molecular Systematics Approaches
Electrophoresis and PCR
Gel electrophoresis is a method of detecting and visualising the extracted DNA material. The
theory behind it is that the DNA material, when put in the agarose gel and connected to a
power source, moves from the negatively charged electrode to the positively charged one.
This is because the DNA backbone contains phosphates that are negatvel charged and are
attracted to the positive electrode. The gel itself creates inertia by allowing the smallest
fragment move the fastest by being entangled the least in the agarose matrix. Ethidium
bromide is added to the gel binds with the DNA material, causing it to fluoresce under
ultraviolet light, allowing one to see the DNA material.
Polymerase chain reaction (PCR) is the process where one can amplify a single copy of a DNA
segment into potentially billions of identical copies. In addition to the several other
chemicals needed for the reaction to work, one of the most important is the primer, which is
generally 20 nucleotides long and are complementary in sequence to the ends of the target
DNA. The reaction can last several hours and typically consists of 20-35 cycles where DNA
material is repeatedly heated and cooled.
When it is heated, the DNA denatures and this breaks the hydrogen bonds that hold the
strand together. When it is cooled, the primers form hydrogen bonds and anneal with their
complementary sequences in the target DNA. The primers and DNA follow base-pairing rules
where Adenine (A) pairs with Thymine (T) and Cytosine (C) pairs with Guamine (G). When
the temperature is increased again, the Taq polymerase works optimally, polymerising the
strand by adding nucleotides to the 3’ end of each primer to the DNA strand. After every
cycle, the number of strands increases exponentially (Berg et al., 2008)
Gene regions
Chloroplast DNA (cpDNA) has been used extensively to infer plant phylogenies at various
taxonomic levels from interspecies relationships to intraspecies. Two of the most widely
used gene regions in the cpDNA for phylogenetic analyses are atpB and rbcL.
The atpB gene is located in the large single copy region of the chloroplast genome next to
the atpE gene downstream from the rbcL gene and is separated by approximately 650 base
pairs of intergenic spacers in ferns. It encodes the subunit of the ATP synthase, which is
involved in the synthesis of ATP. The rbcL gene which is roughly 1400 base pairs long, codes
16. for the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCo). Both of
these gene regions are valuable for comparative sequence studies as they are relatively
short enough for sequencing, but long enough to have important phylogenetic information
(Berg et al., 2008).
A test on the rate of synonymous nucleotide substitution in the gene (using rice and tobacco
as reference) yielded atpB with a Ka value of 0.62 and rbcL with a Ka value of 0.63, suggesting
similar rates of molecular evolution. Also, there are no reported insertions or deletions in
both the genes, allowing for easy alignment (Hoot et al., 1995)
Maximum Parsimony
Maximum parsimony (MP) is a discrete statistical method of inferring phylogeny. The data
for MP analyses comprises of individual nucleotide sites and for each site the goal is to
construct its evolution with the constraint of using the least amount of changes. This total
number of changes or mutations on a tree is referred to as the ‘tree length’ and is basically
the sum of the number of changes at a site (Page and Holmes, 1998).
There are several variables that are used to describe the trees generated through MP
analyses, and two of the most commonly used ones are consistency index (CI) and retention
index (RI). CI describes the extent of homoplasy within each tree, which defined as the the
backwards and parallel substitutions at each nucleotide site. This number varies between 0
and 1, where a number closer to one indicates that the nucleotide configuration at this site
is supportive of the tree under MP principle. RI is the measure of the amount of
synapomorphy of the tree and varies from 0 to 1 as well. Syanomorphy is defined as a
character that is shard by two or more taxa and their most common ancestor. In this case
however, the closer the number is to 0 means the tree is explaining the best amount of
synapomorphic characters.
Once the tree is generated through MP analyses, one can use Bootstrap (BS) to estimate the
reliability of an MP tree. It measures the occurrences of a split in a clade happening and
calculates a percentage based on it. A higher BS value shows a strongly supported node. For
example, a 70% BS value at a node shows that that split occurred at 70% of the trees
sampled (Nei and Kumar, 2000). This allows conclusions to be derived on the likelihood of
clades that have high or low bootstrap values.
17. Bayesian Analysis
Bayesian analyses for phylogenetics produces both a tree and measures of uncertainties
associated with that tree. This is done by first defining a model of molecular evolution for
the dataset and then running a unique algorithm on it for a certain number of iterations. This
algorithm is the Markov chain Monte Carlo (MCMC) and works by creating a conceptual
chain formed by following a series of self-generated steps.
It starts by a new ‘location in the parameter space’ being proposed using a stochastic
method. This ‘location in parameter space’ in phylogenetics is the description of the tree as
explained by the data and the model of molecular evolution specified. This is proposed as
the next link in the chain. The relative posterior probabilities of this new location are
calculated and if it’s higher than the previous chain, it is accepted. This becomes the next link
in the chain and the chain is updated with this new probability. By repeating this procedure
millions of times, a long chain is formed with a high posterior probability (Huelsenbeck et al.,
2001, Holder and Lewis, 2003).
18. Materials and Methods
Plant Material
Plant material for this project was provided by the Trinity College Botanic Gardens, Dartry
(Table 1). Freshest leaf material was chosen from individual plants, discarding any stem bits.
Table 2: Summary of the fresh plant material used. Family is based on the Smith (2006) conscription of the
respective taxa which has also been adapted by National Centre of Biotechnology Information (NCBI). Mass is
the amount of fresh material used for DNA extraction.
Plant Family Accession
No.
Source Mass (g)
Dryopteris tomentosum Dryopteridaceae 19990049 Vietnam 0.154
Cyrtomium falcatum Dryopteridaceae unknown 0.29
Dryopteris affinis Dryopteridaceae Ireland 0.205
Lastreopsis pacifica Dryopteridaceae 19980067 Pitcairn
Islands
0.15
Tectaria gemmifera Tectariaceae unknown 0.16
Athyrium felis-femina Woodsiaceae Ireland 0.17
DNA Extraction
Total DNA was extracted from each sample using a modification of the hot 2xCTAB method.
The protocol used for this method of extraction is outlined below, as well as elaborated in
Hodkinson et al (Hodkinson et al., 2007).
5ml of 2xCTAB buffer, along with pestle and mortar were preheated to 65°C in a water bath.
Plant material was cut into tiny pieces to facilitate grinding in the mortar. Green leafy parts
of the plant were selected and the stem parts were discarded. This was ground up with the
CTAB buffer to form slurry. The slurry was poured into chloroform-resistant tubes and
incubated in a water bath at 65°C for 10 minutes. 5 ml of CI was added to each tube after
this and gently mixed. The tubes were then placed at a horizontal position on a shaker for 30
minutes. After this time, the tubes were centrifuged at 4,000 rpm (c.3500 g) for 10 minutes.
The tubes were then carefully removed from the centrifuge to prevent the mixing of the
separated aqueous and organic material. The upper aqueous layer (containing DNA material)
of each sample was carefully extracted using a transfer pipette and poured into a 50 ml
conical-based tube. The rest of the contents of the centrifuge tube were discarded into the
hazardous waste unit. An equal volume of isopropanol was added to each of the samples
and the tubes were inverted gently to allow the DNA to precipitate out of the solution. The
19. samples were labelled and then placed in a freezer at -20°C for 5 days to allow the DNA to
precipitate out further.
The samples were removed from the freezer once after one week and defrosted. They were
centrifuged at 2000 rpm (c. 1800 g) for 10 minutes to cause DNA to form a pellet at the base
of the tube. The supernatant was carefully poured away, leaving the pellet in the base. 1.5ml
of ethanol (70%) was poured into the tube and gently shaken to dislodge and dissolve the
DNA pellet. The tubes were centrifuged once again at 2000 rpm for 5 minutes to cause the
DNA to form as a pellet at the base of the tubes again. The supernatant was poured off once
again and the tube was left inverted on absorbent paper to remove any of the excess
ethanol. The tubes were then left upright in a fume cupboard for 20 minutes. 0.5ml TE buffer
(10mM Tris-HCl; 1mM EDTA) was added to each tube to dissolve and resuspend the DNA
pellet. This was then mixed and transferred into a 1.5 ml centrifuge tube using a transfer
pipette and stored in a freezer at -70C.
Gel Electrophoresis
To test whether the extraction was successful and to measure the amount of crude total
DNA fragments, the samples were run through gel electrophoresis.
The gel consisted of 1.2% agarose in 1x TBE (0.445 M Tris-HCL; 0.445 Boric acid; 0.055 M
disodium EDTA). 80 ml of the gel was heated in a microwave oven to ensure no solid bits
remained in the jar. The cap was loosely tightened to prevent build up of gas. 2l of
ethidium bromide was added to the gel to facilitate binding of the DNA and allow
visualisation under UV light. After the solution had cooled, it was poured into an 11 x 14
casting boat which was sealed at the edges and allowed to cool for at least 30 minutes.
0.6l of loading dye was mixed with 0.5l of DNA sample on a parafilm base using a p20
micorpipette. This mixture was then transfered to one of the wells in the gel, using a
separate well for each sample. The gel was allowed to run for 30 minute in an
electrophoresis machine containing 1x TBE at 123 V. The resulting gel was placed on a UV
light box to visualise the DNA and a picture was taken using the inbuilt camera.
DNA Purification
On the successful extraction of DNA, the samples were then purified with a spin column
technique using a PCR purification kit called JETQUICK (GENOMED Inc) which contains all the
reagents necessary for the procedure.
20. 100l of each of the crude tDNA samples was pipetted into separate JETQUICK spin columns
which is sitting inside a 2ml reciever tube. 400l of the H1 binding solution (containing
Guanidine hydrochloride and Isopropanol) was pipetted into each of the spin columns and
the tubes were centrifuged for 1 minute at 13,000 rpm (c 12,000g). The solution collected in
the reciever tubes were discarded and the spin columns reinseted into it. 500l of H2 wash
buffer (containing Ethanol, NaCl, EDTA and Tris-HCl) was then pipetted into each of the spin
columns and centrifuged at 13,000 rpm for 1 minute. The solution collected in the reciever
tube and the reciever tube itself discarded and the spin columns placed in new reciever
tubes. 50l of sterile water was added to the spin columns and centrifuged at 13,000 rpm
for 2 minutes. The spin column was then discarded and the reciever tubes containing the
purified DNA were sealed, labelled and stored in a freezer at -20C.
PCR
Once the tDNA was sufficiently purified and was verified to contain DNA by running another
electrophoresis gel, polymerase chain reaction (PCR) was used to amplify the required DNA
regions. The gene regions of the plastid genome we were interested in were atpB and rbcL,
and respective primers [] were used to amplify these gene regions.
2l of purified tDNA was added to 1.5ml centrifuge tubes. A ‘master mix’ was made for each
of the PCR reactions to reduce the amount of pipetting containing all the required reagents
(Table 3). To each 1.5ml centrifuge tube, 48l of the ‘master mix’ was added. A negative
control was also introduced, by having one centrifuge tube containing 2l of sterile water
instead of DNA material. This will help determine if any contamination occurs during the
pipetting and PCR process.
21. Table 3: Reagents, volumes, concentrations and actual amounts that are added to the 1.5ml centrifuge tube to
make up the ‘master mix’.
Reagent Volume (l) Actual amount
DNA (purified, total) 2 c. 400 ng
Sterile water 31.75
5 x Buffer (pH 9.0, 100mM Tris,
HCl, 500 mM KCl, 1% Triton x-
100)
10 1 x buffer
dNTPs 1 10mM
Forward primer (100 ngl-1) 0.5 50 ng
Reverse primer (100 ngl-1) 0.5 50 ng
MgCL2 (25 mM) 4 2 mM
Taq polymerase (2.5 unit l-1) 0.25 0.625 units
This ‘master mix’ process was done for both of the gene regions to be amplified. The tubes
were pulsed in a centrifuge prior to loading into the thermal cycler to facilitate mixing. The
Taq polymerase was the last reagent to be added so that the least amount of reaction occurs
prior to being loaded into the PCR machine. The two DNA regions were amplified using a
GeneAmp PCR System 9700 thermal cycler (ABI, Applied Biosystems, Warrington,
Cheshire, UK). Parameters used for the PCR reaction are outlined in Table 4.
Table 4: PCR parameters for atpB and rbcL gene region (Schuettpelz and Pryer, 2007) executed on a GeneAmp
PCR system 9700 (Applied Biosystems). Soak time is synonymous with hold time where the PCR products are
held for an indefinite period of time.
Steps Temperature (o
C) Time (mins) Cycles
Premelt 94 5
Denaturation 94 1
x35Annealing 45 1
Extension 72 2
Final extension 72 10
Soak 4
The products of the PCR reaction were run on an agarose gel stained with ethidium bromide
to determine whether the PCR reaction was successful. The reaction however wasn’t initially
successful, so the PCR was run again, but changing some of the parameters or
concentrations of some of the chemicals. The cycles were reduced or the annealing
temperature increased to 52°C. The concentrations of the DNA and MgCl2 was also altered
22. accordingly. To optimise the PCR reaction, the amount of DNA was increased to 3µl and the
25mM MgCl2 was was increased to 5µl. In this case, the amount of water was reduced so the
total reaction mixture was 50 µl. However, this still gave negative results in the
electrophoresis gel.
23. Taxon sampling and Alignment
Exhaustive sampling was conducted for this study, where all available atpB and rbcL data for
the entire Dryopteridaceae, Tectariaceae and Woodsiaceae families were downloaded to
create matrices for the family. This involved extracting all the available nucleotide sequences
available for the respective families and sorting out the entries that had the required gene
regions. A list of species along with their respective accession numbers is available in
Appendix I.
Two DNA matrices were initially assembled for the purpose of analysis: [i] atpB (175 taxon; 1
outgroup) and [ii] rbcL (411 taxon; 1 outgroup) and the sequences were downloaded from
GenBank (National Centre of Biotechnology Information [NCBI]). The matrices were created
and aligned manually using the Mesquite Software package (Maddison and Maddison,
2010). No insertion or deletions were required and all characters were treated as equally
weighted and unordered. Osmunda cinnamomea from the Osmundaceae family was
selected as an outgroup due to its close relationship with the leptosporangiate clade (Smith
et al., 2006) and the respective sequences were added to the two matrices.
Phylogenetic Analysis: Bayesian
Bayesian Inference was performed on each of the three matrices using MrBayes version
3.1.2 (Huelsenbeck and Ronquist, 2001). Uniform prior probabilities were used and a
random starting tree. Prior to executing the matrices which were in nexus format [.nex], the
two matrices were edited to be compatible with MrBayes. This involved removing all the
characters that weren’t allowed in the software including numbers in the taxa block such as
accession numbers, parentheses etc. Once the matrices were compatible, they were
executed in the software using the execute command.
F81 model of molecular evolution (Felsenstein, 1981) was used initially to do a test run on
the two datasets and to see how long the analyses would run for. This was done by inputting
command lset nst=1 into MrBayes. Markov chains were run for 10,000,000 generations and
sampled every 1000 generations for the atpB and rbcL matrix. This was done by inputting
command mcmc ngen=10000000 samplefreq=1000. The first 3000 trees were the ‘burn-in’
of the chain.
The analyses was completed at a reasonable amount of time so, ModelTest was used to
calculate the most optimum model for molecular evolution, which was GTR + I + R.
24. A second set of Bayesian analyses was run using the above model and was executed in
MrBayes using the following command: lset nst=6 rates=invgamma. The Markov chains for
each of the dataset were run for 1,000,000 generations with sample frequency of 1000. The
‘burnin’ was set at 500. Excluding the ‘burn-in’ trees, all the trees sampled from the two
independent analyses were combined to produce two different consensus trees, which
formed the basis for calculation of Bayesian posterior probabilities (PP) for respective clades.
Phylogenetic Analysis: Maximum Parsimony
Maximum Parsimony (MP) tree building approach was performed on the two matrices using
PAUP 4.0b10 (Swofford, 2002), treating gaps as missing data.
For MP, unweighted analyses was performed on the sequences using heuristic searches for
the most parsimonious trees. The starting tree was obtained via stepwise addition utilising
tree-bisection-reconnection (TBR) as the swapping algorithm, with one tree held at each
step. 20 trees of length greater than and equal to 35 were held per replicate and Multrees
was in effect. Non-parametric bootstrap analyses (Felsenstein, 1985) was conducted to
evaluate the relative level of support for individual clades on the cladograms of each search.
1000 bootstrap replicates were used.
Multi-gene analysis
Once congruence was was found among the consensus trees generated through Bayesian
inference and maximum parsimony, the two matrices were combined to form one multi-
gene matrix. This matrix consisted of 138 taxa with one outgroup and was constructed using
the Mesquite Software package with the rbcL matrix fused to the 3’ of the atpB matrix.
The combined dataset was was analysed using a Bayesian Markov chain Monte Carlo
(B/MCMC) approach, as implemented in MrBayes version 3.0b. The optimum model for this
dataset was the same as the individual gene datasets: GTR + I + R. Markov chain for this
dataset was run for 1,000,000 generations, sampling every 1000 generations. Once the
analyses was complete, the ‘burnin’ was set to 500. The combined dataset was also analysed
using maximum parsimony and the same settings used in section 2.7 was used for this as
well.
Visualising the Trees
The trees generated through each of the analyses were viewed in both Mesquite (Maddison
and Maddison, 2010) and FigTree (Rambaut, 2009) along with their respective posterior
25. probability and bootstrap values. The consensus tree generated through MrBayes was
outputted as a [.con] file and was opened in FigTree for visualising and editing.
Once the maximum parsimony analyses was complete, the trees were saved with the
bootstrap values as a [.tre] file type. This was then opened in Mesquite initially and in Figtree
for editing.
26. Results
The total DNA (tDNA) was successfully extracted using the modified CTAB method and
fluoresced positively under ultra-violet light (Figure 3). Double banding was seen in sample 2,
3, 5 and 6 in the well.
Despite multiple attempts, amplification was unsuccessful using the rbcL primer listed in
(Bouchenak-Khelladi et al., 2008). Double banding, which was seen in the tDNA, was visible
again in the polymerase chain reaction products (Fig. 4). Figure 4 shows the last gel run with
an attempt to amplify the sequences using the rbcL sequence.
Figure 3: Total DNA flourescing under UV light
Figure 4: PCR products flourescing under UV light.
27. Alignment
The atpB and rbcL sequence alignments were straightforward with no insertions or
deletions. The atpB single gene set consisted of 175 taxa and 1369 characters, whilst the
rbcL single gene set consisted of 411 taxa with 1402 characters. The combined data of
consisted of taxa that had both atpB and rbcL gene regions available on GenBank. There
were 138 taxa in this combined set and 2680 alignment characters (Table 1). All the gene
sets included one species (in the outgroup) for rooting in phylogenetic analyses in the form
of Osmunda cinnamoneum.
Phylogenetic trees
Six trees (5 shown) were generated for analyses from the three datasets; a consensus 50%
majority rule tree by Bayesian inference and a strict consensus maximum parsimony tree
with bootstrap values being generated for each of the dataset. The number of
parsimoniously informative characters, tree length, contention and retention index for the
maximum parsimony trees are summarised in Table 5 as well as the model of nucleotide
evolution used for the Bayesian analyses.
Table 5: Summary of the three datasets used in the analyses. Included Taxa includes the one group of Osmunda
Maximum Parsimony Bayesian
Data
Set
Included
Taxa
Alignment
Length
Number of
informative
characters
(%)
Tree
Length
CI - RI Model for
nucleotide
evolution
atpB 175 1369 419 (30%) 2457 0.2869 –
0.7681
GTR + I + G
rbcL 411 1402 493 (35%) 4260 0.2190 –
0.8253
GTR + I + G
atpB-
rbcL
138 2680 802 (30%) 4510 0.3020 –
0.7508
GTR + I + G
The atpB single gene set (Figure 5) and the atpB-rbcL (Figure 7) combined multi gene set
resulted in Bayesian trees with nearly identical topology, with several minor differences at
genus level. The MP trees supported this topology with bootstrap support of 50% for the
major delineations.
28. The rbcL tree (Figure 6) resolved with several polytomies compared to the atpB and the
atpB-rbcL gene set. The overall topology of the polypods wasn’t resolved in this data set,
however generic level clades resolved well in the Bayesian tree with the MP tree supporting
the major delineations with 50% bootstrap.
atpB gene set (Figure 5)
The eupolypods were resolved relatively well in the atpB single gene tree.
The eupolypods II clade consisting of the Woodsiaceae family resolved with a posterior
probability (PP) value of 0.76 and a bootstrap (BS) value of 69% (Figure 1). Within
Woodsiaceae, Woodsia was resolved outside the rest of the family consisting of the athyrioid
ferns (at), which resolved into a paraphyletic group (PP 0.97, BS 55). Within the athyrioid
ferns, Athyrium was found to be monophyletic (PP 1.0, BS 88) and sister to the rest of the
athyroids consisting of a monophyletic Deparia clade (PP 1.0, BS 100) and a paraphyletic
Diplazium (PP 1.0, BS 60).
Leucostegia pallida and Didymochlaena truncatula resolved outside the euploypod clade
with no support found in the Bayesian or MP trees. However, Hemidyctum mamatum was
found to be sister to the Woodsiaceae family, with support found in the Bayesian tree (PP
0.76)
The eupolypod I clade was resolved in the Bayesian tree (PP 1.0) with generic level lineation
supported by the MP tree. Tectariaceae family resolved into a monophyletic group (PP 1.0,
BS 87), with Gymnopteris resolving outside the tectarioids (te). Tectaria was resolved
polyphyletic within the athyriods, with a clade consisting of Heterogonium and Ctenitopsis
being resolved within the tectarioids.
Sister to the Tectariaceae family is the Dryopteridaceae family. However, this relationship
was only supported by the Bayesian tree (PP 0.99). The family splits into two distinct clades,
the overall topology of which is only supported by the bayesian tree (PP 0.99). One clade
consists of the dimorphic climbers (PP 1.0, BS 93), Stigmatopteris plus Ctenitis (PP 0.55) and
Lastreopsis plus Megalastrum (PP 0.68) with former lomariopsids (fl). The former
lomariopsids resolve strongly into one group (PP 1.0 BS 98), and embedded in it is the large
Elaphoglossum clade (PP 1.0, 83), which is shown to be monophyletic here.
29. The second major clade within the family is the dryopteroid clade (dr) (PP 1.0, BS 57.7). The
dryopteroids resolve into two groups: one consisting of the Arachiniodes clade (PP 1.0 BS 95)
and the Dryopteris-Polystichum clade. This Dryopteris-Polystichum clade splits into two
groups: one consisting of the paraphyletic Dryopteris (PP 1.0 BS 100) and the second
consisting of the polyphyletic Polystichum (PP 0.89 BS 59.6).
rbcL gene set (Figure 6)
The single gene rbcL tree resolved a large tree with several polytomies, especially with
Polystichum. The overall topology of the eupolypods was not resolved, hence the trees
generated through this single gene set was used to analyse generic level delineations.
Arachinoides grouped with Lithostegia, Leptorumohra and Phanerophlebiopsis (PP 1.0, BS
80) and was found to be sister to paraphyletic Dryopteris (PP 0.5, BS 64). The dimorphic
climbers in the form of Polybotrya, Maxonia, Olfersia and Cyclodium, resolved into a strongly
supported group (PP 1.0, BS 100). Ctenitis (with Stigmatopteris) formed a paraphyletic group
(PP 0.7, BS 88).
Lastreopsis resolved into a monophyletic group (PP 0.7, BS 75) and was found to be sister to
the clade consisting of Elaphoglossum (PP 1.0, BS 75). The atyrioid ferns, consisting of
Deparia, Diplazium and Athyrium form one major group with Deparia resolving into a
monophyletic group (PP 1.0, BS 99), Diplazium paraphyletic (PP 1.0 BS 97) and Athyrium
monophyletic (PP 0.98 BS 53).
atpB-rbcL gene set (Figure 7)
The multi-gene trees resolved the overall topology of the eupolypods well. The eupolypods II
is sister to eupolypods I. The athyriod ferns (at) form a monolphyletic group (PP 1.0, BS 92.6)
with the monophyletic Deparia (PP 1.0, BS 100) sister to the clade consisting of a
monophyletic Athyrium (PP 1.0 BS 93.95) and a paraphyletic Diplazium. Woodsia is found to
be sister to the athyrioid ferns (PP 0.5). Gymnocarpium and Cystopteris form a basal clade
that is sister to the clade consisting of Woodsia and the athyrioid ferns (PP 1.0 BS 59.3).
Leucostegia and Didymochlaena resolve outside the eupolypod clade. However,
Hemidictyum was found inside the Woodsiaceae family in the Bayesian analysis, but no
support from the MP tree.
30.
31.
32. Figure 3a: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006). Circles are families discussed in the
study; e2: euploypods II, at athyrioid ferns.
33. Figure 3b: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006).
Circles are families discussed in the study.
34. Figure 3c: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006).
Circles are families discussed in the study.
35. Figure 3d: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006).
Circles are families discussed in the study; e2: euploypods II, at athyrioid ferns.
36. Figure 3e: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006).
Circles are families discussed in the study; e2: euploypods II, at athyrioid ferns.
37. Figure 3f: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006).
Circles are families discussed in the study; e2: euploypods II, at athyrioid ferns.
38. Figure 3g: Single gene atpB tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006).
Circles are families discussed in the study.
39.
40. Figure 4: Single gene rbcL tree. Major genera is described in a grey box and discussed within the stud
41. Figure 5a: Multi gene tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006). Circles are families discussed in the study
42. Figure 5b: Multi gene tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006). Circles are families discussed in the study.
43.
44. Figure 5c: Multi gene tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006). Circles are families discussed in the study
45.
46. Figure 5d: Multi gene tree generated by Bayesian inference and Maximum Parsimony. Boxes in grey are family circumscriptions by Smith (2006). Circles are families discussed in the study
48. 48
Eupolypods I resolves into two families (PP 1.0 BS 61.9), namely the Tectariaceae and the
Dryopteridaceae. In Tectariaceae, clade consisting of Arthropteris and Psammiosorus is found
to be sister to the tectarioid ferns (PP 1.0 BS 100). Within the tectarioid ferns, Pteridys is found
to be sister to the rest of the group (PP 1.0 BS 75.2), while the rest of the tectarioids resolve
into one group with strong support (PP 1.0 BS 100). Ctenitopsis and Heterogonium is found to
resolve within Tectaria, however with only Bayesian support (PP 0.5).
The Dryopteridaceae family resolves into one major group within the eupolypods II (PP 1.0 BS
70). The Bayesian tree shows a dichotomous split within the Dryopteridaceae family (PP 1.0),
which however is not supported by the MP tree. The group consisting of the dimorphic climbers
(dc) forms a strongly supported group within the family (PP 1.0 BS 100) as does Ctenitis which is
found to be monophyletic (PP 1.0 BS 100). A clade consisting of Rumohra, Megalstrum,
Lastreopsis, Ploecnemia and the former lomariopsids (fl) is resolved well within the family (PP
1.0 BS 98). The group consisting of Lastreopsis, Megalstrum and Rumohra is sister to the group
consisting of the Ploecnemia and the former lomariopsids (fl). The clade consisting of the
former lomariopsids is resolved well in both trees (PP 1.0 BS 98), with the Elaphoglossum found
in a sub-clade within it (PP 1.0 BS 96.1).
The dryopteroids form a sub-group within the Dryopteridaceae and is resolved in both trees (PP
1.0 BS 56.1). Within the dryopteroids, two delineations occur: one group consisting of
Polystichum and its sister group Cyrtomium (PP 1.0 BS 100) and another group consisting of
Dryopteris and its sister group Arachinoides (PP 1.0 BS 91.95). Arachinoides forms a strongly
supported group with Leptomohura and Lithostegia and is paraphyletic (PP 1.0 BS 100).
Arachinoides is sister to the clade consisting of Dryopteris, Acromohra, Nothoperanema and
Acrophorus (PP 1.0 BS 99), but Dryopteris itself is paraphyletic.
49. 49
Discussion
Eupolypods II: Woodsiaceae (Fig)
In the trees generated by both the atpB single gene set and the atpB-rbcL multi gene set,
Woodsiaceae resolved with strong Bayesian and maximum parsimony support. Cystoteris and
Gymnocarpium is found to be sister to the rest of the eupolypod II clade (Fig), which is
concordance with the current circumscription of Woodsiaceae (Smith et al., 2006, Schuettpelz
and Pryer, 2007). Smith (2006) tentatively placed Woodsia in the Woodsiaceae family and our
analyses is in agreement with this as it was resolved as being sister to the athyrioid ferns is all
three of the trees generated by the three gene sets.
The athyrioid ferns, which account for most of the diversity in Woodsiaceae are monophyletic
with 55% and 92.6 % BS support in the atpB and rbcL data set respectively and in accord with
published phylogenies on the athyrioid (Sano et al., 2000). Smith (2006) did not comment on
the structure within the athyrioid clade, but Schuettpelz (2007) found strong support through
their multi gene analyses on a split within the athyrioid ferns. This split was also found in our
multi-gene analysis, with Deparia being sister to Athyrium and Diplazium with 92.6% BS support
(Fig). This was also found in our single gene analyses of rbcL. (Fig). Athyrium is found to be not
monophyletic in our analyses (BSatpB
: 87.95; BSrbcL
: 98; BSatpB-rbcL
: 93.95), which is in agreement
with Schuettpelz (2007). Diplazium however is resolved as a monophyletic group (BSatpB
: 99;
BSrbcL
: 97; BSatpB-rbcL
: 100), which is in accordance with Sano et al. (2000) as well.
Eupolypod I: Tectariaceae (Fig)
In the trees generated by all the three datasets, Tectariaceae resolved with strong support as a
monophyletic sister group to the Dryopteridaceae (BSatpB
: 87; BSrbcL
: 95; BSatpB-rbcL
: 100) if we
use Smith’s (2006) definition of the family. Former circumscription of Tectariaceae within the
Dryoptridaceae family has been replaced by this relationship (Smith et al., 2006, Schuettpelz
and Pryer, 2007) as including it within an expanded Dryopteridaceae family renders the latter
polyphyletic.
Studies have shown that Arthropteris is closely related to Psammiosorus on the basis of spore
morphology (Tryon and Lugardon, 1991) and is also sister to the tectarioid ferns (Tsutsumi and
Kato, 2006). This is why Smith (2006) placed them in the Tectariaceae family. Our analyses has
found further evidence for this circumscription, with the two genera grouping together in all of
50. 50
our trees, and found as sister to the tectarioid ferns (Fig) in the multi-gene analyses with strong
bootstrap support (BSatpB-rbcL
: 100).
Generic delimitations in Tectaria s.p is still in doubt, as there has been very little published
study to understand this group. Our analyses found Tectaria to be paraphyletic with
Ctenitopsis, Heterogonium and Gymnopteris resloving within the Tectaria s.l group (BSatpB-rbcL
:
100).
Eupolypod I: Dryopteridaceae (Fig)
As per Smith’s (2006) circumscription of Dryopteridaceae, Didymochlaena, Leucostegia and
Hypodematium, not usually associated with the family, were included in these analyses
because they were categorised as such in GenBank. The reasons for its unorthodox positioning
in the Dryopteridaceae family are discussed in the Introduction and in this study, we find strong
support to re-evaluate their position and segregate them from the Dryopteridaceae family.
Didymochlaena, Leucostegia and Hypodematium resolve outside the eupolypods and including
them with the Dryopteridaceae family renders the latter paraphyletic. They were never
analysed together except for Schuettpelz (2007), but had been analysed individually before
(Hasebe et al., 1995, Schneider et al., 2004, Tsutsumi and Kato, 2006).
Within the family, we see a distinct clade consisting of ‘dimorphic climbers’ (BSatpB
: 100; BSatpB-
rbcL
: 100), that was resolved well in Schuettpelz (2007). Sister to this is Polystichopsis, which is
often synonimysed under Arachinoides due to morphological similarities. The reason for this is
unknown as Arachinoides is grouped well within the ‘former lomariopsids’ (fl). A possible
explanation would be a syanomorphy that is shared between Arachnoides and the clade
consisting of Polystichiopsis and the dimorphic climbers might have evolved twice in the
lineage.
Ctenitis and Stigmatopteris are found to be isolated within the family. Ctenitis however, forms
a monophyletic group with strong BS support in all three trees (BSatpB
: 100; BSrbcL
: 88; BSatpB-rbcL
:
100), which was supported in Liu et al.’s multi-gene analyses (Liu et al., 2007). This isolated
position is consistent with distinct morphological characters present in the genera such as tiny
teeth called ctenii on the mid rib of the blade or axis of inflorescence (Kramer et al., 1990).
Clade consisting of Megalastrum and Rumohra is found to be sister to the paraphyletic
Lastreopsis that is seen in all the trees and is in accordance with Schuettpelz (2007). This clade
in turn is sister to the clade consisting of Pleocnemia and the former lomariopsids. Pleocnemia
has been circumscribed in Tectariaceae according to Smith (2006). Liu et al. (2007) put forth the
51. 51
argument of including it within Dryopteridaceae due to its close affinity to Lastreopsis, which
was only recently moved into Dryopteridaceae (Smith et al., 2006), but failed to find strong
support for this relationship in their phylogenetic analyses. This study we found Pleocnemia
resolving as sister to the ‘former lomariopsids’ in the rbcL tree and the multi-gene tree (BSrbcL
:
76 BSatpB-rbcL
: 56). This is first analysis since Liu et al. (2007) that has compared Pleocnemia in
the context of Dryopteridaceae and Tectariaceae, and has found it nesting within the former.
This author would agree with Liu et al.’s (2007) suggestion to circumscribe the genera under
Dryopteridaceae, with the condition that more phylogenetic and morphological work is done
on it.
Sister to Pleocnemia is the clade consisting of the ‘former lomariopsids’ (fl). This clade consists
of polyphyletic Bolbitis, polyphyletic Lommagramma and a monophyletic Elaphoglossum.
Kramer added these three genera into a distinct family called Lomariopsidaceae, whilst Smith
(2006) included just Lomagramma in this family. Our study shows that this circumscription is
not necessary, and that the three genera form a strongly supported monophyletic group in all
three trees in the form of the ‘former lomariopsids’ group (BSatpB
: 98; BSrbcL
: 77; BSatpB-rbcL
: 98).
This monophyly is also supported in recent literature (Schuettpelz and Pryer, 2007, Liu et al.,
2007), however requires a morphological syanopomorphy to support its new circumscription.
The Elaphoglossum clade forms a monophyletic group within the ‘former lomariopsids’. This is
in concordance with published literature (Skog et al., 2004b) with the Amygdalifolia clade being
formed basal to the rest of the Elaphoglossum (BSatpB
: 60.95; BSrbcL
: 75; BSatpB-rbcL
: 95.1). This
strong support adds to the weight against re-circumscribing this group into a family of its own.
To do so, would render the Dryopteridaceae family parphyletic.
The two well-studied genera of Polystichum (Little and Barrington, 2003) and Dryopteris
(Geiger and Ranker, 2005) form a large clade along with Arachinoides and Cyrtomium known as
the dryopteroids (dr). Polystichum forms a large clade with Cyrtomium, Cyrtogonellum and
Cyrtomidictyum. Three sub-groups are resolved within this clade namely the Polystichum s.s.
clade, the Cyrtomium s.s. clade and the BCPC clade (Lu et al., 2007), with the BCPC clade sister
to the Cyrtomium s.s clade and the Polystichum s.s. clade. Our analyses failed to find this
topology, with the Cyrtomium s.s clade as sister to the BCPC and the Polystichum s.s. clade.
However, this topology is only seen in the multi-gene tree, with no bootstrap support.
The rest of the dryopteroid ferns consist of Dryopteris, Arachinoides and related taxa. Two
different topologies arose from our study. The atpB single gene tree gave us a configuration
where we have Arachnoides sister to Dryopteris, which in turn is sister to Cyrtomium and
52. 52
Polystichum (BSatpB
: 57.7). The multi-gene tree gave us a different topology where a clade
consisting of Arachinoides and Dryopteris is sister to Polystichum and Cyrtomium (BSatpB-rbcL
:
56.1). Published studies confirm the latter (Liu et al., 2007).
Leptorumohra and Phanerophlebiopsis are found within Arachinoides s.s, rendering the latter
paraphyletic (Figure). This is what Liu et al. (2007) found as well, and have asked for the
inclusion of the genera into the Dryopteridaceae family, which our study strongly supports.
Dryopteris s.s. is found to be polyphyletic, which is supported by all the trees in this study as
well as published literature (Liu et al., 2007, Schuettpelz and Pryer, 2007). The genus has been
studied extensively with faster evolving gene sequences than atpB and rbcL and authors have
come to the same conclusion of its polyphyly (Geiger and Ranker, 2005).
Critique on the Methods Employed
We were unable to obtain any DNA sequences at the end of this project. This is was partly due
to to use of incorrect primers. The rbcL utilised for this experiment had been optimised for use
in grasses and wasn’t suitable for the amplification of leptosporangiates. This was only realised
in hindsight and would’ve added immensely to this study. The Lastreopsis species made
available for this study had yet to be sequenced and its addition to the data set would’ve
helped the phylogenetic analyses substantially. The primers required for fern amplification are
outlined in Table 3.
Table 3: Primers required for amplification of DNA sequences in ferns adapted from Wolf’s research website (Wolf,
2010a, Wolf, 2010b).
Sequence 5’-3’
Gene region Forward Reverse
atpB TTG ATA CGG GAG CYC CTC TWA
GTG T
GAA TTC CAA ACT ATT CGA
TTA GG
rbcL ATG TCA CCA CAA ACA GAA ACT
AAA GCA AGT
TCA CAA GCA GCA GCT AGT
TCA GGA CTC
The taxon sampling, albeit exhaustive, was weighted too much on the species that had several
entries in GenBank. The rbcL dataset was clearly the largest because a lot more sequences were
available for it. If a representative taxon sampling were employed similar to Schuettpelz (2007),
53. 53
the polytomy seen in the rbcL tree would not have occurred. Sampling the lineages
proportional to the number of species they contain would’ve been a sounder strategy.
With regards to the methods employed in the phylogenetic analyses, due to time constraints,
and limitation on computational resources, quicker, but not necessarily thorough algorithms
were used. In the maximum parsimony analyses, the starting tree was obtained by simple step-
wise addition, whilst the optimum method would’ve been through random step-wise addition
(Nei and Kumar, 2000). Also, only 20 trees were held per replicate. This could’ve been increased
to 100, but would’ve greatly increased computational time.
In the Bayesian analyses, the Markov chain could’ve run a longer number of generations. With
such a high data set used in this study, setting the generations to 10,000,000 would’ve
produced a more stable result (Huelsenbeck and Ronquist, 2001).
Conclusion
Several points can be concluded about the result of this study. Firstly, all of the families
currently circumscribed under the derived polypod can be described by the multi-gene tree
consisting of atpB and rbcL sequences. The eupolypods form two distinct clades namely the
eupolypods I and the eupolypods II.
The eupolypod I consists of the two families of Tectariaceae and Dryopteridaceae, and we
found support for the monophyly of Tectariaceae and including it in Dryopteridaceae will
render the latter paraphyletic. Strong support was found for Didymochlaena, Leucostegia and
Hypodematium not to be included within Dryopteridaceae, as doing so will render the family
paraphyletic.
Strong support was also found to move the traditional tectariod fern Pleocnemia into
Dryopteridaceae and that it shows a close relationship with Lastreopsis.
Within the eupolypods II clade, Woodsiaceae resolved as a monophyletic group with strong
support being found for a split within the athyrioid ferns separating Deparia from the clade
consisting of a non-monophyletic Athyrium and the monophyletic Diplazium.
54. 54
Finally, this study illustrated that there are still holes in our understanding of leptosporangiates,
specifically the polypod. Uncertain relationships require further analyses and in tandem with
the phylogenetic analyses, morphometric studies should be employed so we have a clearer
understanding of the major delineations. With more the 25% of the fern genera unaccounted, it
is clear that much work needs to be done in the future as more species come to light (Schneider
et al., 2004).
55. 55
References
BERG, J., TYMOCZKO, J. & STRYER, L. 2008. Biochemistry (Looseleaf), W. H. Freeman.
BOUCHENAK-KHELLADI, Y., SALAMIN, N., SAVOLAINEN, V., FOREST, F., BANK, M. V. D.,
CHASE, M. W. & HODKINSON, T. R. 2008. Large multi-gene phylogenetic trees of the grasses
(Poaceae): Progress towards complete tribal and generic level sampling. Molecular
Phylogenetics and Evolution, 47, 488-505.
FELSENSTEIN, J. 1981. Evolutionary trees from DNA sequences: A maximum likelihood
approach. Journal of Molecular Evolution, 17, 368-376.
FELSENSTEIN, J. 1985. Confidence Limits on Phylogenies: An Approach Using the
Bootstrap. Evolution, 39, 783-791.
GEIGER, J. M. O. & RANKER, T. A. 2005. Molecular phylogenetics and historical
biogeography of Hawaiian Dryopteris (Dryopteridaceae). Molecular Phylogenetics and
Evolution, 34, 392-407.
HASEBE, H., OMORI, T., NAKAZAWA, M., SANO, T., KATO, M. & IWATSUKI, K. 1994. rbcL
gene sequences provide evidence for the evolutionary lineages of leptosporangiate ferns.
PNAS, 91, 5730-5734.
HASEBE, M., WOLF, P. G., PRYER, K. M., UEDA, K., ITO, M., SANO, R., GASTONY, G. J.,
YOKOYAMA, J., MANHART, J. R., MURAKAMI, N., CRANE, E. H., HAUFLER, C. H. & HAUK, W.
D. 1995. Fern Phylogeny Based on rbcL Nucleotide Sequences. American Fern Journal, 85,
134-181.
HODKINSON, T., WALDREN, S., PARNELL, J., KELLEHER, C., SALAMIN, K. & SALAMIN, N.
2007. DNA banking for plant breeding, biotechnology and biodiversity evaluation. Journal
of Plant Research, 120, 17-29.
HOLDER, M. & LEWIS, P. O. 2003. Phylogeny estimation: traditional and Bayesian
approaches. Nat Rev Genet, 4, 275-284.
HOOT, S. B., CULHAM, A. & CRANE, P. R. 1995. The Utility of atpB Gene Sequences in
Resolving Phylogenetic Relationships: Comparison with rbcL and 18S Ribosomal DNA
Sequences in the Lardizabalaceae. Annals of the Missouri Botanical Garden, 82, 194-207.
HUELSENBECK, J. P. & RONQUIST, F. 2001. MRBAYES: Bayesian inference of phylogenetic
trees. Bioinformatics, 17, 754-755.
HUELSENBECK, J. P., RONQUIST, F., NIELSEN, R. & BOLLBACK, J. P. 2001. Bayesian
Inference of Phylogeny and Its Impact on Evolutionary Biology. Science, 294, 2310-2314.
KENRICK, P. & CRANE, P. R. 1997. The origin and early diversification of land plants: a
cladistic study, Washington D.C., USA, Smithsonian Press.
56. 56
KRAMER, K., GREEN, P. & GöTZ, E. 1990. Pteridophytes and gymnosperms, Springer-
Verlag.
LITTLE, D. P. & BARRINGTON, D. S. 2003. Major evolutionary events in the origin and
diversification of the fern genus Polystichum (Dryopteridaceae). Am. J. Bot., 90, 508-514.
LIU, H. M., ZHANG, X. C., WANG, W., QIU, Y. L. & CHEN, Z. D. 2007. Molecular phylogeny of
the fern family dryopteridaceae inferred from chloroplast rbcL and atpB genes.
International Journal of Plant Sciences, 168, 1311-1323.
LU, J.-M., BARRINGTON, D. S. & LI, D.-Z. 2007. Molecular Phylogeny of the Polystichoid
Ferns in Asia Based on Rbcl Sequences. Systematic Botany, 32, 26-33.
LUPIA, R., LIDGARD, S. & CRANE, P. R. 1999. Comparing Palynological Abundance and
Diversity: Implications for Biotic Replacement during the Cretaceous Angiosperm
Radiation. Paleobiology, 25, 305-340.
MA-LI, W., ZHI-DUAN, C. & XIAN-CHUN, Z. 2003. Phylogeny of the Athyriaceae: evidence
from chloroplast trnL-F region sequences. Journal of Systematics and Evolution, 5, 416-426.
MADDISON, W. P. & MADDISON, D. R. 2010. Mesquite: a modular system for evolutionary
analysis. Version 2.73 ed.
MURAKAMI, N., NOGAMI, S., WATANABE, M. & IWATSUKI, K. 1999. Phylogeny of
Aspleniaceae Inferred from rbcL Nucleotide Sequences. American Fern Journal, 89, 232-
243.
NEI, M. & KUMAR, S. 2000. Molecular evolution and phylogenetics, Oxford University Press.
PAGE, R. & HOLMES, E. 1998. Molecular evolution: a phylogenetic approach, Blackwell
Science.
PRYER, K. M., SCHNEIDER, H., SMITH, A. R., CRANFILL, R., WOLF, P. G., HUNT, J. S. & SIPES,
S. D. 2001. Horsetails and ferns are a monophyletic group and the closest living relatives to
seed plants. Nature, 409, 618-622.
PRYER, K. M., SCHUETTPELZ, E., WOLF, P. G., SCHNEIDER, H., SMITH, A. R. & CRANFILL, R.
2004. Phylogeny and evolution of ferns (monilophytes) with a focus on the early
leptosporangiate divergences. Am. J. Bot., 91, 1582-1598.
RAMBAUT, A. 2009. FigTree: graphical viewer for phylogenetic trees. Version 1.3.1 ed.
Reading, U.K.
ROTHWELL, G. W. 1987. Complex Paleozoic Filicales in the Evolutionary Radiation of Ferns.
American Journal of Botany, 74, 458-461.
57. 57
SANO, R., TAKAMIYA, M., ITO, M., KURITA, S. & HASEBE, M. 2000. Phylogeny of the Lady
Fern Group, Tribe Physematieae (Dryopteridaceae), Based on Chloroplast rbcL Gene
Sequences. Molecular Phylogenetics and Evolution, 15, 403-413.
SCHNEIDER, H., SCHUETTPELZ, E., PRYER, K. M., CRANFILL, R., MAGALLON, S. & LUPIA, R.
2004. Ferns diversified in the shadow of angiosperms. Nature, 428, 553-557.
SCHUETTPELZ, E. & PRYER, K. M. 2007. Fern phylogeny inferred from 400
leptosporangiate species and three plastid genes. Taxon, 56, 1037-7E.
SCHUETTPELZ, E. & PRYER, K. M. 2009. Evidence for a Cenozoic radiation of ferns in an
angiosperm-dominated canopy. Proceedings of the National Academy of Sciences, 106,
11200-11205.
SKOG, J. E., MICKEL, J. T., MORAN, R. C., VOLOVSEK, M. & ZIMMER, E. A. 2004a. Molecular
Studies of Representative Species in the Fern Genus Elaphoglossum (Dryopteridaceae)
Based on cpDNA Sequences rbcL, trnL-F, and rps4-trnS. International Journal of Plant
Sciences, 165, 1063-1075.
SKOG, J. E., MICKEL, J. T., MORAN, R. C., VOLOVSEK, M. & ZIMMER, E. A. 2004b. Molecular
Studies of Representative Species in the Fern Genus Elaphoglossum (Dryopteridaceae)
Based on cpDNA Sequences rbcL, trnL‚ÄêF, and rps4‚ÄêtrnS. International Journal of Plant
Sciences, 165, 1063-1075.
SMITH, A. R., PRYER, K. M., SCHUETTPELZ, E., KORALL, P., SCHNEIDER, H. & WOLF, P. G.
2006. A Classification for Extant Ferns. Taxon, 55, 705-731.
STEIN, W. 1993. Modeling the Evolution of Stelar Architecture in Vascular Plants.
International Journal of Plant Sciences, 154, 229-263.
SWOFFORD, D. L. 2002. PAUP*: phylogenetic analysis using maximum parsimony (*and
other methods). 4.0 b10 ed. Sunderland, Massachusetts, USA.
TRYON, A. & LUGARDON, B. 1991. Spores of the Pteridophyta: surface, wall structure, and
diversity based on electron microscope studies, Springer-Verlag.
TSUTSUMI, C. & KATO, M. 2006. Evolution of epiphytes in Davalliaceae and related ferns.
Botanical Journal of the Linnean Society, 151, 495-510.
WOLF, P. G. 2010a. PCR amplification and DNA sequencing primers – fern atpB [Online].
Available: http://bioweb.usu.edu/wolf/atpB%20primer%20map.htm 2011].
WOLF, P. G. 2010b. PCR amplification and DNA sequencing primers – fern rbcL [Online].
Available: http://bioweb.usu.edu/wolf/rbcL%20primer%20map.htm 2011].
