Are the Parasites of Marsupials Ancient Heirlooms
or Recent Souvenirs?
Aime H. Rankin
Undergraduate, The University of Glasgow, Glasgow, Scotland.
Abstract
An organism’s parasites can be split into ancient ‘heirlooms’ and recently acquired
‘souvenirs’. Marsupials present as an ideal taxon for identifying these two categories, due to
experiencing extreme isolation in Australia and relatively little in South America. A list of
marsupial parasites was formed using the database BioNames and searching the literature.
Phylogenies stored on BioNames were used to identify the parasites’ closest relatives and
their hosts, as well as their country of residence. Sequences for marsupial parasites were
found on GenBank and used to create phylogenies using an automated BLAST program. The
parasites’ hosts were then superimposed upon the tree, which was interpreted for evidence
of evolutionary associations. It was found that few close relatives of marsupial parasites
parasitized exclusively marsupials and most had a worldwide presence, except for New
Guinean parasites. A positive correlation was found between the taxonomic distance
between parasites and their closest relatives and the taxonomic distance between their
hosts. Similarly, a positive correlation was found between the taxonomic distance between
parasites and their closest relatives and the geographical distance between their hosts. Of
the phylogenies generated for the marsupial parasites, only a few showed evidence of
evolutionary associations between parasites and hosts. These included the parasites
Sarcocystis, Heterodoxous octoseriatus, Brachylaima dasyuri, Macropodinium ennuensis and
M. yalabense. The majority of parasite phylogenies showed the hosts of the related
parasites in distantly related animal groups. Many phylogenies also included distantly
related taxa within the tree itself. It was thought that these unusual patterns in the
phylogenies could have been the result of sampling error (the absence of many marsupial
parasites from GenBank), the use of short sequences, or highly conserved genes. It was
concluded that to determine the true provenance of marsupial parasites, many more must
be sequenced and the data stored in the sequence databases.
Introduction
The parasites of an organism can be said to compose of a mixture of ‘heirlooms’ and
‘souvenirs’. Heirlooms reflect an ancient association between the parasite and the host,
with the parasite being passed on from the ancestral species through the generations. Thus
they share similar patterns of evolutionary history. Souvenirs are parasites that have been
acquired by the host more recently, perhaps through evolving new behaviours or by
inhabiting a new environment.
An increasing amount of parasite data is being added to sequence databases, with many
entries including the host animal of the parasite. By searching GenBank for the known
parasites of a group of hosts, evolutionary trees can be built from these sequences and used
to identify the close relatives of those parasites. It can then be asked what species plays
host to these related parasites. If they are found in association with close relatives of the
host then there might be evidence of cospeciation; however, if they are found on distantly
related taxa this may imply a recent association rather than an ancient one. The search for
heirlooms and souvenirs becomes an interesting one when the hosts have experienced a
significant ecological shift, such as moving between environmental extremes, or if their
environment has been invaded by foreign organisms and their parasites.
Marsupial animals make an interesting study group when exploring this particular theme of
heirlooms and souvenirs. Marsupials are a distinctive group of mammals that separated
from placental mammals around 160 million years ago in the Jurassic era (Yuan et al., 2011).
Although fossil marsupials have been found across the world, their current distribution is
restricted to South America, Australia, Indonesia and Papua New Guinea (Archer et al.,
2012). It is thought that marsupials migrated from South America to Australia by travelling
across Antarctica between 70 and 55 million years ago (Archer et al., 2012). Although South
American marsupials have had constant contact with placental animals throughout their
evolution, Australian marsupials have been relatively isolated since Australia broke away
from Antarctica some 35 million years ago (Archer et al., 2012). They have only recently
been introduced to placental animals with the arrival of man on the continent 50,000 years
ago (Bowler et al., 2003). This combination of recent contact following a period of isolation
make these animals ideally suited for separating which of their parasites are heirlooms and
of which are souvenirs.
The null hypothesis of the current study was that the parasites of any marsupial mammal
were likely to be more closely related to parasites found on other marsupials than to those
on placental mammals. That is, that marsupial parasites would be heirlooms.
If found that parasites were consistent with this hypothesis a second hypothesis would be
tested, namely that the marsupials and their parasites may have cospeciated. This could be
achieved by comparing evolutionary trees for both the host and the parasite. If cospeciation
occurred, the trees would be more similar than would be expected due to chance.
If there were parasites that were shared between marsupial and placentals, the hypothesis
that sharing of parasites would be related to degree of historical geographic isolation, would
be tested.
Materials and methods
Lists of marsupial host-parasite associations were obtained by searching through the
BioNames database (http:bionames.org) or by searching through the literature. Using
phylogenies on BioNames and from the literature, close relatives of these parasites were
found as well as their hosts and country of residence. Appropriate graphs and diagrams
were then created using programs such as Mintab to summarise this data. Then the
sequence database GenBank was searched for DNA sequences from those parasites. For
each parasite sequence studied from those in GenBank, similar sequences were retrieved
using BLAST and used to build a phylogeny (this process is automated using
http://iphylo.org/~rpage/phyloinformatics/blast). For each resultant phylogeny the hosts
for the parasites were identified, from GenBank or the literature. Through interpreting the
tree, it was determined whether marsupial parasites tended to cluster together on the tree,
suggesting evolutionary association with their hosts, or scattered across the tree, suggesting
episodes of host switching.
Results
First of all, it was found that the close relatives of the marsupial parasites, parasitized on a
broad range of chordate animal groups. The majority of these parasites were found to be on
placental animals or mammalian animals in general, a total of 28.2% and 27.2% of the
parasites respectively (Figure 1). Only 5.5% of the parasites’ relatives lived exclusively on
other marsupials and these came only from Australian Macropods (Figure 1). Interestingly, a
very small proportion of the parasites was found on Cephalopods, Insects or were free-living
(Figure 1).
The country in which the close relatives of the marsupial parasites resided was equally as
broad. The majority of the parasites’ close relatives (34%) had a worldwide presence (Figure
2). This was followed by 16.1% of the parasites residing in Australia (Figure 2). Only 3.8% of
the parasites were found in South America and even fewer still in Indonesia and New
Guinea (Figure 2). However, for the marsupial parasites of New Guinea, the majority of their
close relatives were also present in that region and in neighbouring Australia (Figure 2). As
much as half of Australian parasites had relatives in Australia (Figure 2). Very few of the
parasites from South American marsupials had relatives exclusive to that part of the world
(Figure 2).
When comparing the taxonomic distance between host and parasite pairs there was a slight
correlation, where an increase in taxonomic distance in one led to an increase in the other
(Figure 3). However, the taxonomic distance between the host pairs was greater than that
of the parasite pairs in general (Figure 3). Several outliers were also found. One of these
comprised Ophidascaris robertsi and O. moreliae, members of the same genus, yet their
hosts, marsupials and snakes respectively, are a whole class apart. Another interesting case
was that of apicomplexans Hepatozoon vivernus and Aggregata; while themselves only a
superfamily apart, their hosts, marsupials and Cephalopods respectively, are a whole
phylum apart.
Similarly, a trend of increasing taxonomic distance between parasite pairs with increasing
geographic distance between host pairs was observed (Figure 4). Again however, interesting
outliers were identified. Protozoans Toxoplasma and Hammondia hammondi, separated
only by an order, were found on hosts at distances of at least 16000km away from each
other (Figure 4). At the other end of the spectrum, lice Heterodoxus octoseriatus and
Laemobothrion atrum were separated by a whole class yet were found on hosts separated
by no geographic distance (Figure 4).
Of the trees generated by BLAST, a varied collection of parasite trees were created. The
majority did not seem to show any obvious pattern of marsupial hosts clustering around
certain groups of parasites. As already seen in Figure 1, it was found that the close relatives
of the marsupial parasites were often seen to be on distantly related groups such as birds or
reptiles. A very surprising find was the occurrence of very distantly related taxa placed
within parasite groups in some trees. A good example of this comes from the tree generated
for Parastrongyloides (Figure 5). Within this tree, frogs, mice, bears and even maize is
included in the parasite’s phylogeny (Figure 5). However, five trees out of the thirty-four
created did show clusters of marsupial parasites. These included Sarcocystis whose closest
relatives were found on marsupials, followed by birds and placentals (Figure 6).
Heterodoxous octoseriatus had relatives on marsupial hosts as well, followed by bird hosts
(Figure 7). Macropodinium ennuensis and M. yalabense had many close relatives parasitizing
marsupials, with more distant relatives on fish and placentals (Figure 9). Brachylaima
dasyuri and its close relatives are found on marsupials but its more distant relatives were
found on rodents, birds, amphibians and fish (Figure 8).
Figure 1: Diagram showing marsupial species and the hosts
of the close relatives of their parasites. The colour of the
line indicates the group to which each parasite belongs.
Figure 2: Diagram depicting the nationality of a marsupial and the
country in which the hosts of their parasites’ closest relatives live.
The colour of the line indicates the group to which each parasite
belongs.
Figure 3: Graph depicting the taxonomic distance between marsupial parasites and their closest relatives,
against the taxonomic distance between the marsupial host and the hosts of the parasites’ closest relatives.
Figure 4: Graph depicting the taxonomic distance between marsupial parasites and their closest relatives,
against the geographic distance between the marsupial host and the hosts of the parasites’ closest relatives.
Figure 5: Phylogeny of Parastrongyloides generated by
BLAST.
Figure 6: Phylogeny of Sarcocystis generated by BLAST.
Figure 7: Phylogeny of H. octoseriatus generated by BLAST.
Figure 8: Phylogeny of B. dasyuri generated by BLAST.
Figure 9: Phylogeny of Macropodinium ennuensis and M. yalabense generated by
BLAST.
Discussion
Using the BioNames database proved an effective way to compile parasite lists for marsupial
species. For any one marsupial species, its taxonomic and phylogenetic information was
easily at hand as well as any literature surrounding it, including what parasites infested it.
BioNames was also very helpful in providing trees in order to work out the close relatives of
the parasites identified and the literature to work out which host they parasitized and in
which country.
The resulting diagrams summarizing this process revealed that few relatives of marsupial
parasites were exclusive to marsupial hosts and that a majority were found to parasitize a
great number of vertebrate groups (Figure 1). This is interesting as it might be expected that
the close relatives would parasitize a closely related marsupial rather than a different class
of animal. A simple explanation for this could simply be down to sampling error. Using
phylogenies to work out a parasite’s closest relative has the downside that the tree may not
represent the true diversity of parasites actually present in the phylogeny. As not every
parasite discovered or studied will be sampled for sequencing, not every potential relative
can be included in the tree. This could result in two parasites appearing more closely related
than they actually are and imply relationships between sometimes equally distant hosts. For
example, Aggregata appears as the closest relative of H. vivernus yet this would suggest a
relationship between marsupials and its host Cephalopods, which is not likely since there
are no marine marsupials or for that matter, terrestrial Cephalopods. Also observed was
that there were a large number of parasites found on mammalian animals in general. This
would imply that they are not host specific parasites and that they may be able to survive on
a wider spectrum of hosts.
Furthermore, the geographic distribution of the parasites was also found to be very varied
(Figure 2). Only parasites of New Guinea had the majority of their relatives in the same area.
Reasons for this may be that the island and to a certain extent, its parasites, have been
relatively isolated since the end of last ice age when it broke away from Australia
(Gascoigne, 1998). This idea could also be applied to the half of Australian parasites having
relatives in the same area. Regards to the number of parasites found worldwide, it may be
that many of the parasites are not host specific such as Strongyloides and can infect a range
of vertebrates, hence their prevalence around the globe. Again, the sampling error will have
is likely to have affected these results. Geographically distant and widespread parasites may
not be as closely related to the marsupial parasites as they seem. This may explain why so
few parasites of South American marsupials appear to have close relatives in the same
region.
As might be expected, it was observed that the more taxonomically distant a parasite was
from its closest relative, the more taxonomically distant the hosts of these parasites would
be (Figure 3). Why the hosts of O. robertsi and O. moreliae were so taxonomically different
could be down to the fact that O. moreliae is a fairly recently discovered species. Although
marsupials have been found to be the intermediate host to the snake parasite O. robertsi, as
of yet, O. moreliae has only been experimentally introduced into marsupials and no wild
cases have yet been recorded (Sprent, 1970). Similarly, increasing geographic distance
between host pairs correlated with the taxonomic distance between parasite pairs (Figure
4). Similar results of this nature have been found in other studies looking at helminth
communities (Poulin, 2003).
Regards generating phylogenies using BLAST, the automated BLAST program was a very
efficient and easy method to use when creating trees. The abundance of phylogenies not
showing obvious clustering of marsupial hosts could be down to the sampling error
discussed earlier. As for the occurrence of distantly related taxa included in the tree itself,
there could be several reasons for this. Firstly the sequences may have come from very
conserved genes which would tend to be similar even in distant taxa. Secondly, it could be
that very short sequences were used and therefore more likely to be found repeated in
other genomes in general. Finally, it could also be the case that the sequence used may have
at some point integrated itself into the host’s genome making it appear as if the parasite
and host were closely related. In this respect, an improvement to the BLAST program would
be the addition of confidence values associated with the branches to give an idea of the
trustworthiness of the tree.
For the few trees that did exhibit marsupial hosts clustered around groups of parasites, this
could be evidence of cospeciation and further research should be carried out to test this.
Indeed, for the protozoan Sarcocystis (Figure 6) studies have already produced results to
suggest that Sarcocystids from Australia and South America are monophyletic, sharing a
common ancestor and have co-evolved with their marsupial hosts (Merino et al., 2010).
Similarly, for the digenean Brachylaima (Figure 8), a common ancestor for South American
and Australian parasites is postulated (Beveridge and Spratt, 1996). However, this theory
has been complicated by the discovery of a species from introduced rodents in Australia,
implying that the parasites may have come from a different line altogether (Beveridge and
Spratt, 1996). Another controversial theory has been the suggestion that the Boopiidae, a
family of lice including Australian H. octoseriatus (Figure 7), shares a common ancestor with
South American lice of marsupials (Barker, 1994). However, characters have been
discovered to suggest a sister-group of bird lice and Boopiids and that the transfer of bird
lice to marsupials creating the Boopiidae family is a relatively recent development (Barker,
1994). Although the phylogeny in this study suggests possible coevolution between H.
octoseriatus and its wallaby hosts, there is in actual fact very little evidence to suggest this
after comparing their phylogenies, due to the magnitude of host-switching events (Barker,
1994). Many parasite-host associations have been suggested for the Macropodinium genus
of Australian ciliates (Figure 9). Coevolution has been observed as well as host switching and
several speciation events are thought to have occurred due to vicariation in the Pleistocene,
when south and west marsupial populations split (Cameron and O’Donoghue, 2004).
Although, M. yalabense itself ‘missed the boat’ during this event as it failed to speciate
between the hosts Macropus giganteus and Ma. fuliginosus (Cameron and O’Donoghue,
2004). Another interesting theory is that association of these ciliates with their hosts is not
by co-descent but by the resource tracking of these ruminant animals (Cameron and
O’Donoghue, 2004). For each of these examples however, definite associations cannot be
assumed until robust parasite and host phylogenies are created.
To conclude, only a small number of the marsupial parasites investigated showed evidence
of cospeciation but of these parasites, most were present in literature that reinforced their
status as heirlooms. For the remainder of the parasites investigated, it was hard to establish
their provenance as there was much variety in the taxonomic and geographic distances
between themselves and the assumed close relatives. Perhaps most likely due to sampling
error, further study would be benefitted by closing the gaps in our sequence databases.
Acknowledgements
I would like to thank Professor R. Page for supervising this study. Financial support was
provided by the Head of College Scholars List Scheme.
References
Archer, M. and Kirsch, J. 2012. The Evolution and Classification of Marsupials. In: Armati, P.
J., Dickman, C. R. and Hume, I. D. eds. 2012. Marsupials. Cambridge: Cambridge University
Press. Ch.1.
Barker, S. C., 1994. Phylogeny and Classification, Origins, and Evolution of Host Associations
of Lice. International Journal for Parasitology, 24 (8), pp.1285-1291.
Beveridge, I. and Spratt, D. M., 1996. The helminth Fauna of Australasian Marsupials: Origins
and Evolutionary Biology. Advances in Parasitology, 37, pp.135-254.
Bowler, J. M., Johnston, H., Olley, J.M., Prescott, J. R., Roberts, R. G., Shawcross, W. And
Spooner, N. A., 2003. New Ages for Human Occupation and Climatic Change at Lake Mungo,
Australia. Nature, 421, pp. 837-840.
Cameron, S. L. and O’Donoghue, P. J., 2004. Morphometric and Cladistic Analyses of the
Phylogeny of Macropodinium (Ciliophora: Litostomatea: Macropodiniidae). Acta
Protozoologica, 43 (1), pp.43-53.
Gascoigne, I., 1998. Papua New Guinea (Cultures of the World). 2nd ed. New York:
Benchmark Books.
Luo, Z., Yuan, C., Meng, Q. and Ji, Q., 2011. A Jurassic Eutherian Mammal and Divergence of
Marsupials and Placentals. Nature, 426, pp. 442-445.
Merino, S., Martinez, J., Vasquez, R. A. and Slapeta, J., 2010. Monophyly of Marsupial
Intraerythrocytic Apicomplexan Parasites From South America and Australia. Parasitology,
137, pp.37-43.
Poulin, R., 2003. The decay of similarity with geographical distance in parasite communities
of vertebrate hosts. Journal of Biogeography, 30 (10), pp.1609-1615.
Sprent, J. F. A., 1970. Studies on Ascaridoid Nematodes in Pythons: The Life-history and
Development of Ophidascaris moreliae in Australian Pythons. Parasitology, 60 (1), pp.97122.