Content uploaded by Charles D Bell
Author content
All content in this area was uploaded by Charles D Bell on Feb 13, 2014
Content may be subject to copyright.
Cophylogeny and biogeography of the fungal parasite
Cyttaria
and its host
Nothofagus
, southern beech
Kristin R. Peterson
1
Donald H. Pfister
Department of Organismic and Evolutionary Biology,
Harvard University, 22 Divinity Avenue, Cambridge,
Massachusetts 02138
Charles D. Bell
Department of Biological Sciences, University of New
Orleans, 2000 Lakeshore Drive, New Orleans,
Louisiana 70148
Abstract
:The obligate, biotrophic association
among species of the fungal genus
Cyttaria
and their
hosts in the plant genus
Nothofagus
often is cited as a
classic example of cophylogeny and is one of the few
cases in which the biogeography of a fungus is
commonly mentioned or included in biogeographic
analyses. In this study molecular and morphological
data are used to examine hypotheses regarding the
cophylogeny and biogeography of the 12 species of
Cyttaria
and their hosts, the 11 species of
Nothofagus
subgenera
Lophozonia
and
Nothofagus
. Our results
indicate highly significant overall cophylogenetic
structure, despite the fact that the associations
between species of
Cyttaria
and
Nothofagus
usually
do not correspond in a simple one to one relation-
ship. Two major lineages of
Cyttaria
are confined to a
single
Nothofagus
subgenus, a specificity that might
account for a minimum of two codivergences. We
hypothesize other major codivergences. Numerous
extinction also are assumed, as are an independent
parasite divergence followed by host switching to
account for
C. berteroi
. Considering the historical
association of
Cyttaria
and
Nothofagus
, our hypothesis
may support the vicariance hypothesis for the trans-
Antarctic distribution between Australasian and
South American species of
Cyttaria
species hosted
by subgenus
Lophozonia
. It also supports the hypoth-
esis of transoceanic long distance dispersal to account
for the relatively recent relationship between Austra-
lian and New Zealand
Cyttaria
species, which we
estimate to have occurred 44.6–28.5 mya. Thus the
history of these organisms is not only a reflection of
the breakup of Gondwana but also of other events
that have contributed to the distributions of many
other southern hemisphere plants and fungi.
Key words:
Australasia, Leotiomycetes, long dis-
tance dispersal, South America, southern hemi-
sphere, vicariance
INTRODUCTION
The obligate, biotrophic relationship between species
belonging to the fungal genus
Cyttaria
(Ascomycota,
Leotiomycetes) and their hosts in the plant genus
Nothofagus
(Hamamelididae, Nothofagaceae) has
captured the attention of evolutionary biologists since
Charles Darwin, whose South American collections of
these fungi during the
Beagle
voyage became the first
two
Cyttaria
species to be described (Berkeley 1842,
Darwin 1839). After hearing from Joseph Dalton
Hooker that a third
Cyttaria
species had been found
in Tasmania, Darwin (1846) commented on the
‘‘singular relationship’’ between
Cyttaria
and
Notho-
fagus
in ‘‘distant parts of the world!’’
The history of
Nothofagus
is considered to be
important, even key, in understanding southern
hemisphere biogeography (Cracraft 1975, Darlington
1965, Steenis 1971).
Nothofagus
pollen is distinctive,
produced in copious amounts and is easily fossilized.
First appearing by the early Campanian of the Late
Cretaceous (,83.5 mya) (Dettmann et al. 1990), its
widespread, persistent, and abundant microfossil
record, and to a lesser extent its macrofossil record,
indicate that
Nothofagus
was widespread throughout
much of southern Gondwana before continental
breakup. In addition
Nothofagus
includes prominent
forest trees that presently exhibit a widespread,
disjunct southern hemisphere distribution, with
modern representatives in South America (southwest-
ern Argentina and Chile) and Australasia (southeast-
ern Australia, New Zealand, New Guinea and New
Caledonia) and with extinct taxa also known from
Antarctica (Dettmann et al. 1990, Hill 1991). This
disjunct, trans-Antarctic distribution, characterized by
areas of endemism, traditionally is hypothesized to be
entirely the result of vicariance and extinction
because the dispersability of its seeds has been
assumed to be insufficient to account for the
distribution of
Nothofagus
species. The phylogeny of
Nothofagus
should be a reflection of the geological
breakup sequence of southern Gondwana, according
to this hypothesis. It is generally accepted that New
Zealand was the first to separate ,80 mya and that
South America and Australia were connected by
Antarctica until ,35 mya (McLoughlin 2001). Thus
Submitted 2 Mar 2010; accepted for publication 22 Apr 2010.
1
Corresponding author. E-mail: krpeterson@post.harvard.edu
Mycologia,
102(6), 2010, pp. 1417–1425. DOI: 10.3852/10-048
#2010 by The Mycological Society of America, Lawrence, KS 66044-8897
1417
the phylogeny of
Nothofagus
should show that species
from Australia and South America are more closely
related to each other than to species from New
Zealand (but see Giribet and Edgecombe 2006).
However taxonomic arrangements and phylogenies
of
Nothofagus
have inferred a closer relationship
between Australian and New Zealand species (Dett-
mann et al. 1990, Hill and Jordan 1993, Hill and Read
1991, Linder and Crisp 1995, Manos 1997, Martin and
Dowd 1993).
Nothofagus systematics.—Nothofagus
comprises 35
extant species divided into four subgenera that
correspond to pollen types (Dettmann et al. 1990,
Hill and Jordan 1993, Hill and Read 1991): subgenus
Brassospora
(
brassii
type pollen) with 19 species in
New Caledonia and New Guinea, which do not host
Cyttaria
species; subgenus
Fuscospora
(
fusca
type [a]
pollen) with five species in southern South America
and Australasia, which do not host
Cyttaria
species;
subgenus
Lophozonia
(
menziesii
type pollen) with six
species in southern South America and Australasia,
all which host
Cyttaria
species; and subgenus
Nothofagus
(
fusca
type [b] pollen) with five species
in South America, all which host
Cyttaria
species.
Pollen types corresponding to the four modern
subgenera first appear contemporaneously in the
Antarctic fossil record in the late Campanian-early
Maastrichtian of the Late Cretaceous (,73 mya)
(Dettmann et al. 1990). The monophyly of each of
the four extant subgenera is supported by molecular
sequence data (Martin and Dowd 1993, Setoguchi et
al. 1997) and combined molecular and morpholog-
ical data (Manos 1997).
Nothofagus
subgenera that
host
Cyttaria
species,
Lophozonia
and
Nothofagus
do
not together form a single monophyletic group
(FIG. 1; Cook and Crisp 2005, Manos 1997, Martin
and Dowd 1993, Setoguchi et al. 1997); of the extant
subgenera
Nothofagus
is most closely related to
Brassospora
, representing the most recent divergence
among subgenera, and subgenus
Lophozonia
is sister
of the remaining three subgenera. Subgenus
Lopho-
zonia
exhibits a disjunct trans-Antarctic distribution,
with species occurring in Australia (including Tas-
mania), New Zealand and South America, while
subgenus
Nothofagus
occurs only in South America.
Thus modern day
Nothofagus
subgenera exhibit
three trans-Antarctic relationships: within subgenus
Fuscospora
(with species from Tasmania and New
Zealand more closely related to each other than to
South American species), within subgenus
Lophozo-
nia
(with species from Australia and New Zealand
more closely related to each other than to South
American species, and between subgenera
Nothofa-
gus
(with species from South America) and
Brasso-
spora
(with species from New Guinea and New
Caledonia species).
Cyttaria systematics.—
According to phylogenetic
analyses of combined morphological and nuclear
ribosomal RNA, mitochondrial ribosomal RNA and
TEF1
sequence data, genus
Cyttaria
comprises 12 taxa
representing three major clades (FIG. 1; Peterson and
Pfister 2010): one clade (A) occurs on subgenus
Nothofagus
, which occurs only in South America; the
second (B) occurs only in South America on both
subgenus
Nothofagus
and subgenus
Lophozonia
; and
the third (C) exhibits a trans-Antarctic distribution,
occurring in both South America and Australasia only
on subgenus
Lophozonia
. In total seven
Cyttaria
taxa
are endemic to southern South America (Argentina
and Chile), on subgenera
Lophozonia
and
Nothofagus
,
and the other five are endemic to southeastern
Australasia and New Zealand on subgenus
Lophozo-
nia
. Of the five Australasian taxa, two are found only
in Australia, including Tasmania, and three are found
only in New Zealand.
Associations between Cyttaria and Nothofagus.—
As-
sociations between species of
Cyttaria
and
Nothofagus
usually do not correspond in a simple one to one
relationship; several
Cyttaria
species may infect the
same
Nothofagus
species and a single
Cyttaria
species
may infect several
Nothofagus
species (FIG. 1). Gener-
ally an individual
Cyttaria
species is associated with
more than one (up to five)
Nothofagus
species, which
in turn are associated with more than one (up to
four)
Cyttaria
species. However hosts of
Cyttaria
species that are associated with multiple
Nothofagus
species invariably belong to the same subgenus (but
see FIG. 1). Furthermore with the exception of one
Cyttaria
species each major
Cyttaria
clade is associated
with a single
Nothofagus
subgenus. In no case is there
a
Cyttaria
or
Nothofagus
species common to both
South America and Australasia or between Australia
and New Zealand. Despite extensive searching,
Cyttaria
has not been found on any of the
Nothofagus
species from New Caledonia or New Guinea (Korf
1983). It is unlikely that undiscovered species exist
outside the current known range of
Cyttaria
. This is
because of the conspicuous nature of the typical
Cyttaria
species, with its spherical, honeycombed fruit
bodies (typically up to 4[–8] cm, depending on the
species) (FIG. 2), usually obvious perennial cankers
(depending on the species and type of canker
produced, globose cankers up to 1 m diam on large
branches and up to 1.5 m diam on large trunks, as
well as longitudinal cankers of up to 1 m long) and
often spectacular fruiting (up to 25 densely clustered
fruit bodies in a single group and a carpet of fallen
1418 MYCOLOGIA
fruit bodies up to 15 cm deep on the forest floor in
one species).
Cyttaria
species are presumed to be weak parasites
(Gamundı´ and Lederkremer 1989) that produce
trunk and branch cankers on
Nothofagus
trees. These
cankers arise due to localized, stimulated cambial
activity attributed to the presence of hyphae belong-
ing to
Cyttaria
species, found in the secondary
phloem and xylem, cambium and cortex of the hosts
(Gutie´rrez de Sanguinetti 1988, Wilson 1937).
Many have discussed the idea that parasites,
including
Cyttaria
(Humphries et al. 1986, Korf
1983), can serve as taxonomists to elucidate relation-
ships of their hosts. This idea implies that parasite
and host phylogenies should be congruent or based
on codivergence events. Confounding factors in this
association relating to the parasite include host
switching, extinction and speciation in the parasite
lineage but not the host lineage. Especially problem-
atic with respect to
Cyttaria
and
Nothofagus
is the fact
that most
Cyttaria
parasites and
Nothofagus
hosts do
not display a simple one to one correspondence. Also
of interest is the occurrence of
Cyttaria
species on two
relatively unrelated
Nothofagus
subgenera and their
absence from the other two.
The relationship between
Cyttaria
and
Nothofagus
often is cited as a classic example of cophylogeny (e.g.
Korf 1983) and is one of the few cases in which the
biogeography of a fungus is commonly mentioned.
Crisci et al. (1991), Seberg (1991), Linder and Crisp
(1995) and Sanmartı´n and Ronquist (2004) included
Cyttaria
in their biogeographical analyses. Humphries
et al. (1986) reconstructed five codivergence events,
including the concurrent origins of
Cyttaria
and
Nothofagus
, which they inferred was significant
cophylogeny in the associations between
Cyttaria
species and their
Nothofagus
hosts.
In this study we used molecular and morphological
datasets derived from our work for
Cyttaria
(Peterson
and Pfister 2010) and from Martin and Dowd (1993),
Manos (1997) and Setoguchi (1997) for
Nothofagus
to
test assertions of cophylogeny between the fungal
FIG. 1. Relationships between
Nothofagus
and
Cyttaria
. Branches in boldface are well supported (.0.95 posterior
probabilities or .70%bootstrap proportion). Lines connecting hosts and parasites represent associations; numbers in
parenthesis indicate number of associations per taxon. Shaded boxes indicate
Nothofagus
subgenera that host
Cyttaria
and the
Cyttaria
taxa associated with them. (
Cyttaria
clades A, B and C are discussed in the text and depicted in FIG. 2.) See Gamundı´
and Minter (2004a, b) and Peterson and Pfister (2010, TABLE I) for reports of possible additional hosts
N. dombeyi
and
N.
obliqua
for
C. espinosae
and
C. hookeri
respectively. Illustrations of
C. gunnii
and
N. cunninghamii
from Berkeley (1848) are
reproduced courtesy of the library of the Gray Herbarium, Harvard University, Cambridge, Massachusetts. AUS 5Australia,
NCA 5New Caledonia, NGU 5New Guinea, NZL 5New Zealand, SSA 5southern South America.
PETERSON ET AL.: COPHYLOGENY AND
C
YTTARIA
1419
parasite
Cyttaria
and its host plant
Nothofagus
. These
results were used to examine the biogeographic
history of
Cyttaria
.
MATERIALS AND METHODS
Parasite and host phylogenies.—
Character alignments and
GenBank sequences were obtained from studies on
Cyttaria
and
Nothofagus
. For
Cyttaria
an alignment of nuclear small
subunit (nucSSU), nuclear large subunit (nucLSU), mito-
chondrial small subunit (mitSSU) ribosomal RNA (rRNA)
and
TEF1
sequence data as well as morphological data were
obtained from our work (Peterson and Pfister 2010;
GenBank EU107178–203, -205–232, -234–249). For
Notho-
fagus
alignments of chloroplast DNA (cpDNA)
rbc
L
sequences (Martin and Dowd 1993; GenBank L13341–345,
-348, -350–358, -360, -362), nuclear internal transcribed
spacer (nucITS) rRNA sequences (Manos 1997; GenBank
U96849–857, -859, -863, -865–870), cpDNA
atp
B-
rbc
L
intergenic spacer sequences (Setoguchi et al. 1997; Gen-
Bank AF015687–690, -692, -696, -698–708) and morpholog-
ical data (Manos 1997) were obtained from the original
publications; the dataset of Knapp et al. (2005) was not used
because it did not include all
Nothofagus
species that host
Cyttaria
. Seventeen
Nothofagus
representatives for which
data were available for all four partitions were included.
This included all
Nothofagus
species that host
Cyttaria
taxa,
as well as select representatives from the other two
Nothofagus
subgenera. Methods for reconstructing phylog-
enies follow Peterson and Pfister (2010).
Cyttaria
phyloge-
nies were taken from Peterson and Pfister (2010), where
phylogenies resulting from the exclusion of morphological
data also were considered.
Divergence time estimation.—
Because the molecular data
could be rejected as evolving in clock-like (
P
,0.001),
based on a likelihood ratio test, we used two relaxed clock
dating methods: (i) penalized likelihood and (ii) an
uncorrelated method implemented in BEAST 1.4.8 (Drum-
mond and Rambaut 2007). An uncorrelated lognormal
(UCLN) model implemented in BEAST was used to infer
divergence times. Convergence of each chain to the target
distribution was assessed with Tracer 1.4 (Rambaut and
Drummond 2007) and by plotting time series of the log
FIG. 2. Hypothesized cophylogenetic reconstructions of
Cyttaria
taxa and
Nothofagus
hosts depicting major events. a.
Represents the concurrent origin of
Cyttaria
and
Nothofagus
. b. Represents the colonization by
Cyttaria
after the origin of
Nothofagus
to explain the absence of
Cyttaria
on
Nothofagus
subgenera
Brassospora
and
Fuscospora
, which do not host
Cyttaria
.
5s represent extinction or failure to track the host by
Cyttaria
. Events, labeled 1–10, are: circles, codivergence events; squares,
duplication; and triangles, host switching.
Cyttaria
clades A, B and C represent the three major monophyletic lineages. AUS 5
Australia, NCA 5New Caledonia, NGU 5New Guinea, NZL 5New Zealand, SSA 5southern South America.
1420 MYCOLOGIA
posterior probability of sampled parameter values. After
convergence each chain was sampled every 1000 steps until
50 000 samples were obtained. Model fit of the UCLN
relaxed clock models was assessed with Bayes factors as
implemented in Tracer 1.4. In addition divergence times
were estimated with penalized likelihood (PL, Sanderson
2002) as implemented in the program r8s (Sanderson
2003). The optimal smoothing parameter (l) was deter-
mined by cross validation. We constructed confidence
intervals for our PL based estimates with a bootstrap
resampling technique (Baldwin and Sanderson 1998). First,
the original dataset was resampled 1000 times with
SEQBOOT 3.6 (Felsenstein 2005). Each replicate was used
to re-estimate edge lengths on the optimal topology with
ML in PAUP* 4.0b10 (Swofford 2002). Replicate trees were
transformed in r8s and divergence time estimates summa-
rized across trees with the PROFILE command.
Peterson and Pfister (2010, FIG. 1) was used to estimate
divergence times relating to events pertaining to the
Cyttaria
clade.
Paleopyrenomycites
, widely regarded as the
most important fungal fossil (e.g. Lu¨cking et al. 2009,
Taylor and Berbee 2006), was used to calibrate divergence
among groups. It is uncertain to which fungal group this
400 myo fossil from the Rhynie Chert belongs. Taylor and
Berbee (2006) regard it as an ancestral member of the
Ascomycota, whereas Lu¨cking et al. (2009) proposed it as an
ancestral member of the Pezizomycotina. We therefore
tested both hypotheses.
Host-parasite associations.—
Data were taken from Calvelo
and Gamundı´ (1999), Gamundı´ (1971) and Rawlings
(1956).
Tree-based analyses of cophylogeny.—
These methods take
into account four basic types of cophylogenetic events:
codivergence (cospeciation), duplication (independent
speciation of the parasite), host switching and extinction;
the latter is a type of loss, which also includes two other
phenomena, missing the boat (failure to track all host
lineages following a speciation) and sampling failure
(failure of the researcher to observe parasites on their
hosts); also confounding is the scenario in which parasites
fail to diverge with their hosts (see review by Charleston and
Perkins 2006). Based on the one host-one parasite
assumption, these tree-based methods are unable to
accommodate widespread parasites (parasites associated
with more than one host); thus none were used in this
study to calculate significance values. TreeMap (Charleston
and Page 2002) however was used to provide a graphical
depiction of hypothesized cophylogenetic events.
Distance-based analysis of cophylogeny.—
COPYCAT (Meier-
Kolthoff et al. 2007), a wrapper or interface for ParaFit
(Legendre et al. 2002), was used to assess the null
hypothesis of random association between parasites and
hosts as performed in ParaFit. Host and parasite trees
including branch lengths plus an association file represent-
ing all 25 parasite-host combinations (because most
Cyttaria
taxa are associated with more than one host) were input
into COPYCAT, and tests of random association were
performed with 9999 permutations globally across both
phylogenies for each host-parasite association. Unlike
earlier methods, ParaFit is able to accommodate any type
of host-parasite association (Legendre et al. 2002), includ-
ing widespread parasites (as well as trees with polytomies).
Data-based analysis of cophylogeny.—
The incongruence
length difference (ILD) test (Farris et al. 1995) was used
to seek evidence that the
Cyttaria
and
Nothofagus
datasets
were not samples of the same phylogenetic history.
Complete parasite and host datasets were treated as
partitions, such that each parasite-host pair consisted of a
concatenation of all aligned data from a
Cyttaria
species
and an associated
Nothofagus
host species. All 25 parasite-
host taxa combinations were tested with 10 000 replicates of
the partition-homogeneity test in PAUP* 4.0b10 (Swofford
2002) with parsimony uninformative characters excluded.
RESULTS
Parasite and host phylogenies.—Cyttaria
phylogeny
(FIG. 1) recovered these notable clades are one
composed of the South American species
C. hookeri
and
C. johowii
(A), which forms a monophyletic sister
group with the remaining species, one composed of
the South American species
C. berteroi
,
C. darwinii
,
C.
exigua
and
C. hariotii
(B), which forms a monophy-
letic sister group with the remaining species, one
composed of the South American species
C. espinosae
plus the Australasian species (C), a monophyletic
Australian lineage and a monophyletic New Zealand
lineage.
Nothofagus
phylogeny (FIG. 1) recovered four
monophyletic groups corresponding to subgenera in
agreement with Manos (1997), Martin and Dowd
(1993) and Setoguchi et al. (1997) and for the
reduced set of taxa included here identical in
topology to the combined consensus tree presented
by Manos (1997). As indicated by earlier studies, the
two subgenera that host
Cyttaria
,
Lophozonia
and
Nothofagus
, did not form a clade. Instead subgenus
Lophozonia
(Australia including Tasmania, New Zea-
land and South America), the designated outgroup in
this study, was sister of the remaining subgenera. In
this and previous studies the next to diverge,
subgenus
Fuscospora
(Australia and Tasmania, New
Zealand, and South America), was sister of the final
two, subgenera
Brassospora
(New Caledonia and New
Guinea) and
Nothofagus
(South America). Thus two
subgenera,
Fuscospora
and
Lophozonia
, exhibit trans-
Antarctic distributions. Furthermore subgenera
Bras-
sospora
and
Nothofagus
together exhibit a third trans-
Antarctic distribution.
Divergence time estimation.—
Notable dates estimated
for the
Cyttaria
lineage (TABLE I) include the origin
of modern
Cyttaria
, estimated at 148.4–112.4 mya, the
divergence between Australian and New Zealand
PETERSON ET AL.: COPHYLOGENY AND
C
YTTARIA
1421
species estimated at 44.6–28.5 mya, the divergence of
the clade that includes
C. espinosae
at 81.7–61.5 mya
and the divergence of the clade that includes
C.
berteroi
at 60.6–31.7 mya. With the exception of the
divergence time estimate between the Australian and
New Zealand species of
Cyttaria
, the two placements
of the fossil
Paleopyrenomycites
produced times that
were considerably older than geological and other
events that would explain the divergences between
these groups of
Cyttaria
.Oneexampleofthis
phenomenon may be that our estimates for the
origin of
Cyttaria
predate the origin of
Nothofagus
,
which has been estimated at 93–83.5 mya (Cook and
Crisp 2005), and the origin of extant
Nothofagus
at
55–40 mya (Cook and Crisp 2005). These discrepan-
cies at least in part likely are due to the problem of
using a fossil far removed from our group of interest.
Distance-based analysis of cophylogeny.—
The distance-
based method ParaFit found highly significant overall
cophylogenetic structure between the
Nothofagus
and
Cyttaria
datasets (
P
,0.0001), regardless of correc-
tion method used in the DistPCoA setting (Lingoes,
Cailliez, or no correction). Tests of individual links
between host-parasite pairs found 23 (out of 25) links
to be significant (significant
P
values 50.0001–0.03).
The two nonsignificant links were those between
C.
berteroi
and its two hosts.
Data-based analysis of cophylogeny.—
The ILD test
detected no significant difference between
Cyttaria
and
Nothofagus
datasets (
P
50.97). That is there was
no evidence to indicate that they were not samples of
the same phylogenetic history.
DISCUSSION
Cophylogeny between Cyttaria and Nothofagus.—
Analyses of cophylogeny (distance-based ParaFit and
data-based ILD) indicate that
Cyttaria
taxa exhibit
highly significant cophylogeny with
Nothofagus
hosts,
even though associations between species of
Cyttaria
and
Nothofagus
usually do not correspond in a simple
one to one relationship.
Parafit and ILD do not provide graphical depic-
tions of probable cophylogenetic events. Given that
both tests detected highly significant cophylogeny
between
Cyttaria
and
Nothofagus
, we attempted to
reconstruct their cophylogenetic history with Tree-
Map and present two possible, general scenarios
(FIG. 2).
Cophylogeny scenario a.—
Two
Cyttaria
lineages each
are confined to a single
Nothofagus
subgenus (but see
FIG. 1), a specificity that accounts for a minimum of
two incidents of codivergence. Major incidents in the
cophylogeny (events 1–10, FIG. 2a) are summarized:
(i) an early codivergence at the concurrent origin of
Cyttaria
and extant
Nothofagus
leading to the South
American subgenus
Nothofagus
hosting
C. hookeri
and
C. johowii
(
Cyttaria
clade A) and subgenus
Lophozonia
hosting the lineage that gave rise to all other
Cyttaria
taxa (
Cyttaria
clades B and C); (ii) a codivergence
between the early diverging South American species
C. hookeri
and
C. johowii
(clade A) and their hosts in
subgenus
Nothofagus
; (iii) a duplication (indepen-
dent divergence) in the
Cyttaria
lineage on subgenus
Lophozonia
, which gave rise to (1) the South
American
C. espinosae
plus all of the Australasian
species (clade C) and (2) to the remaining
Cyttaria
species (clade B); (iv) host switching producing a
dichotomy between the South American
C. berteroi
(clade B), which parasitizes subgenus
Lophozonia
, and
the South American species
C. darwinii
,
C. exigua
and
C. hariotii
(clade B), which parasitize subgenus
Nothofagus
; (v) codivergence between
C. darwinii
,
C. exigua
, and
C. hariotii
(clade B) and their hosts,
subgenus
Nothofagus
; (vi) codivergence between
C.
berteroi
(clade B) and the South American lineage of
subgenus
Lophozonia
; (vii) codivergence leading to
the South American subgenus
Lophozonia
hosting
C.
espinosae
(clade C) and the Australasian subgenus
Lophozonia
hosting all Australasian members of
Cyttaria
(clade C); (viii) codivergence between
C.
espinosae
(clade C) and the South American subgenus
Lophozonia
; (ix) codivergence within the Australasian
members of
Cyttaria
(clade C) and subgenus
Lophozonia
, giving rise to the Australian species and
TABLE I. Divergence estimates (mya) for key splits. BEAST ages represent the mean of the 95%highest posterior density
(HPD). AUS 5Australia, NZL 5New Zealand
PL
a
BEAST
a
PL
b
BEAST
b
Origin of
Cyttaria
112.2 (101–123) 112.4 (64–157) 148.4 (133–162) 146.3 (84–178)
AUS/NZL split 28.5 (24–33) 33.9 (14–58) 37.8 (33–42) 44.6 (18–72)
C. espinosae
split 61.5 (55–67) 62.1 (29–96) 81.4 (73–88) 81.7 (38–102)
C. berteroi
split 31.7 (26–38) 47.5 (13–87) 41.9 (38–47) 60.6 (17–98)
a
Fossil placed at the divergence of the Ascomycota.
b
Fossil placed at the divergence of the Pezizomycotina.
1422 MYCOLOGIA
the New Zealand species; and (x) codivergence within
Australian members of
Cyttaria
(clade C) and
subgenus
Lophozonia
giving rise to
C. gunnii
sensu
stricto and
C. septentrionalis
and respective hosts,
N.
cunninghamii
and
N. moorei
. The absence of
Cyttaria
species on subgenera
Brassospora
and
Fuscospora
in
this scenario is explained by extinction or missing the
boat. If
C. espinosae
and
C. hookeri
also occur
respectively on
N. dombeyi
and
N. obliqua
(see
FIG. 1) two additional host-switching events also are
inferred.
Cophylogeny scenario b.—
This one (FIG. 2b) explains
the absence of
Cyttaria
species on subgenera
Brasso-
spora
and
Fuscospora
by inferring the infection of one
clade of
Nothofagus
with a subsequent host jump to
the other clade.
Biogeography of Cyttaria and Nothofagus.—
Given
fossil evidence that the four extant subgenera were
widespread, occurring in Antarctica, Australia and
South America before continental drift (Dettmann et
al. 1990), vicariance resulting from the breakup of
southern Gondwana comprise a plausible hypothesis
to explain the three trans-Antarctic distributions
exhibited by
Nothofagus
(FIGS. 1, 2). In support of
this hypothesis a biogeographic reconstruction for
Nothofagus
by Sanmartı´n et al. (2007) inferred that
these three Australasian-South American relation-
ships (within subgenus
Fuscospora
, within subgenus
Lophozonia
and between subgenera
Brassospora
and
Nothofagus
) arose from vicariance. Knapp et al.
(2005) estimated divergence at 65–36 mya between
the Australasian and South American species of
subgenus
Fuscospora
that were consistent with vicar-
iance, but their results for the divergence between
Australasian and South American species of subgenus
Lophozonia
were equivocal (38–21 mya, the lower
point being too recent to be indicative of vicariance).
In a similar study Cook and Crisp (2005) estimated
divergence consistent with vicariance between the
Australasian and South American species of subgen-
era
Fuscospora
at 45–30 mya and
Lophozonia
at 30 mya,
as well as between Australasian subgenus
Brassospora
and South American subgenus
Nothofagus
at 45–
30 mya. Considering the historical association of
Cyttaria
and
Nothofagus
, the cophylogenetic recon-
structions (FIG. 2) may support the vicariance hypoth-
esis for the trans-Antarctic distribution between
Australasian and South American
Cyttaria
species
hosted by subgenus
Lophozonia
. The codivergence of
one
Lophozonia
lineage with Australia and the other
Lophozonia
lineage with South America followed the
divergence of their associated
Cyttaria
parasites. Our
divergence estimates however are too early to support
the vicariance hypothesis to explain the disjunction of
Cyttaria
species in clade C present on subgenus
Lophozonia
, at 81.7–61.5 mya. The hypothesis of the
presence of
C. berteroi
from clade B on subgenus
Lophozonia
being the result of host switching also is
supported by our divergence estimates, at 60.6–
31.7 mya, later than the divergence among
C.
espinosae
and its Australasian relatives.
The riddle of Cyttaria.—
Subgenus
Nothofagus
, which
hosts
Cyttaria
, is sister of subgenus
Brassospora
, which
does not host
Cyttaria
. Following Hill (1996),
Swenson et al. (2001) discuss the riddle of the
presence of subgenus
Brassospora
in New Caledonia
and New Guinea. The riddle is its absence elsewhere,
according to Heads (2006).
Cyttaria
presents a third
riddle: Why is it present on subgenera
Lophozonia
and
Nothofagus
but absent on subgenera
Brassospora
and
Fuscospora
? The two subgenera that host
Cyttaria
are
not sister taxa and they are not devoid of
Cyttaria
(FIGS. 1, 2). Furthermore, although the current
distribution of subgenus
Brassospora
is restricted to
tropical latitudes in New Caledonia and New Guinea,
its species once were more widespread, occurring in
Antarctica, Australia and South America by the late
Maastrichtian (Dettmann et al. 1990). Subgenus
Fuscospora
, whose extant species co-occur with
Cyttaria
hosts in the other two subgenera, also was
widespread, occurring in Antarctica, Australia and
South America by at least the mid-Paleocene (Dett-
mann et al. 1990). In fact the four extant subgenera
were distributed widely from Australasia to Antarctica
and South America by the mid-Eocene (Dettmann et
al. 1990).
With respect to subgenera
Brassospora
and
Fusco-
spora
the absence of
Cyttaria
remains a riddle. It is
unknown whether the ancestors of subgenera
Bras-
sospora
and
Fuscospora
simply escaped colonization by
Cyttaria
or whether
Cyttaria
went extinct on the
ancestors of these subgenera. No fossils have been
attributed to
Cyttaria
(Korf 1983), but if the fossil
record of
Nothofagus
is any indication of what
happened to
Cyttaria
extensive extinction likely
contributed to the current distribution of
Cyttaria
.
The biogeographic reconstructions by Linder and
Crisp (1995) and Swenson et al. (2001), which predict
extinct lineages within at least the three major
lineages of
Nothofagus
(
Fuscospora
,
Lophozonia
and
Brassospora
-
Nothofagus
), demonstrate the effect of
extinction on the interpretation of biogeographic
patterns observed from extant species (see also Cook
and Crisp 2005, Manos 1997).
Long distance dispersal from Australia to New Zea-
land.—
In agreement with biogeographic, phyloge-
netic and taxonomic studies of
Nothofagus
(Cook and
Crisp 2005, Dettmann et al. 1990, Hill and Jordan
PETERSON ET AL.: COPHYLOGENY AND
C
YTTARIA
1423
1993, Hill and Read 1991, Knapp et al. 2005, Linder
and Crisp 1995, Manos 1997, Martin and Dowd 1993)
and
Cyttaria
(Sanmartı´n and Ronquist 2004), this
study found a closer relationship between Australian
and New Zealand
Cyttaria
taxa (FIGS. 1, 2) associated
with subgenus
Lophozonia
than would be predicted
based on the breakup sequence of southern Gond-
wana. Some suggest that this common pattern is a
result of propagules being carried by west wind drift,
which began in conjunction with Antarctic circumpo-
lar current (,35–28 mya), both of which could
account for dispersal across the great expanse of the
Tasman Sea (e.g. Sanmartı´n et al. 2007, Winkworth et
al. 2002). The divergence between Australian and
New Zealand species of extant subgenus
Lophozonia
was estimated by Cook and Crisp (2005) at 40–14 mya,
and Knapp et al. (2005) estimated divergence at 31–
16 mya. Thus, according to this hypothesis, earlier
Fuscospora
and
Lophozonia
fossil pollen from New
Zealand represent extinct lineages. Our divergence
estimates infer the separation of Australian and New
Zealand clades at 44.6–28.5 mya, consistent with the
long distance dispersal of
Cyttaria
between Australia
and New Zealand.
The fossil record suggests that several species of
Nothofagus
traveled from Australia to New Zealand via
transoceanic long distance dispersal (Hill 2001; Pole
1994, 2001). Although all pollen types were present at
some point in New Zealand, only one
Nothofagus
pollen type, now extinct, was present before New
Zealand drifted from Gondwana, and representatives
of subgenus
Lophozonia
, Australasian hosts of
Cyt-
taria
, did not appear in New Zealand until the Early
Eocene (Dettmann et al. 1990).
Conclusions.—
Because it was widespread in southern
Gondwana before the drift of Australia, New Zealand
and South America
Nothofagus
reflects the major
events that occurred during the formation of the
current biota of these regions, including vicariance,
extinction and transoceanic long distance dispersal.
This study demonstrates how, in turn, obligate
associates of
Nothofagus
such as
Cyttaria
can function
‘‘sufficiently accurate as taxonomists’’ (Korf 1983)
and perhaps hold keys to understanding more about
their hosts.
ACKNOWLEDGMENTS
We thank three anonymous reviewers for comments on this
manuscript. We thank M.J. Cafaro, G. Giribet and C.C. Davis
for advice and comments on this project and manuscript.
We also thank the library of the Gray Herbarium of Harvard
University for permission to reproduce the Berkeley
illustrations. We acknowledge financial support from the
Department of Organismic and Evolutionary Biology of
Harvard University and the Fernald Fund of Harvard
University Herbaria, as well as NSF PEET grant DEB-
9521944 to D.H. Pfister and M.J. Donoghue.
LITERATURE CITED
Baldwin B, Sanderson MJ. 1998. Age and rate of diversifi-
cation of the Hawaiian silversword alliance. Proc Natl
Acad Sci USA 95:9402–9406.
Berkeley MJ. 1842. On an edible fungus from Tierra del
Fuego and allied Chilean species. Trans Linn Soc
London 19:37–43.
———. 1848. Decades of fungi: decade XX. London J Bot 7:
572–579, 1 plate.
Calvelo S, Gamundı´ I. 1999. The record of
Cittaria exigua
(Cyttariales, Ascomycetes) at Tierra del Fuego (Argen-
tina). Hickenia 3:13–14.
Charleston MA, Page RDM. 2002. TreeMap (Jungle
Edition). Version 2.0.2 (Beta). http://taxonomy.
zoology.gla.ac.uk/%7emac/treemap/index.html.
———, Perkins SL. 2006. Traversing the tangle: algorithms
and applications for cophylogenetic studies. J Biomed
Inform 39:62–71.
Cook LG, Crisp MD. 2005. Not so ancient: The extant crown
group of
Nothofagus
represents a post-Gondwanan
radiation. Proc R Soc B 272:2535–2544.
Cracraft J. 1975. Historical biogeography and Earth
history—perspectives for a future synthesis. Ann Mo
Bot Garden 62:227–250.
Crisci JV, Cigliano MM, Morrone JJ, Roigjunent S. 1991.
Historical biogeography of southern South America.
Syst Zool 40:152–171.
Darlington PJ. 1965. Biogeography of the southern end of
the world; distribution and history of far-southern life
and land, with an assessment of continental drift.
Cambridge, Massachusetts: Harvard Univ. Press. 236 p.
Darwin C. 1839. Journal of researches into the geology and
natural history of the various countries visited by
H.M.S.
Beagle
, under the command of Capt. Fitz Roy,
R.N., from 1832 to 1836. London: Henry Colborn.
615 p.
———. 1846. Journal of researches into the natural history
and geology of the countries visited during the voyage
of H.M.S.
Beagle
round the world. New York: Harper &
Bros. 675 p.
Dettmann ME, Pocknall DT, Romero EJ, Zamaloa MC.
1990.
Nothofagidites
Erdtman and Potonie´, 1960: a
catalogue of species with notes on the paleogeographic
distribution of
Nothofagus
Bl. (southern beech). NZ
Geol Survey Paleontol Bull 60:1–79.
Drummond AJ, Rambaut A. 2007. BEAST: Bayesian evolu-
tionary analysis by sampling trees. BMC Evol Biol 7:
214–222.
Farris JS, Ka¨llersjo¨ M, Kluge AG, Bult C. 1995. Testing
significance of congruence. Cladistics 10:315–319.
Felsenstein J. 2005. SEQBOOT—bootstrap, jackknife or
permutation resampling of molecular sequence, re-
striction site, gene frequency or character data. http://
wwwsacsucsfedu/phylip2/seqboothtml.
1424 MYCOLOGIA
Gamundı´ IJ. 1971. Las Cyttariales sudamericanas (Fungi-
Ascomycetes). Darwiniana 16:461–510.
———, Lederkremer RM. 1989. Los hongos andino-
patagonicos del genero
Cyttaria
. Sus hidratos de
carbono. Cienc Investig 43:4–13.
———, Minter DW. 2004a.
Cyttaria espinosae
. IMI Descript
Fungi Bacteriol 160:1593.
———, ———. 2004b.
Cyttaria hookeri
. IMI Descript Fungi
Bacteriol 160:1597.
Giribet G, Edgecombe GD. 2006. The importance of
looking at small-scale patterns when inferring Gondwa-
nan biogeography: a case study of the centipede
Paralamyctes
(Chilopoda, Lithobiomorpha, Henicopi-
dae). Biol J Linn Soc 89:65–78.
Gutie´rrez de Sanguinetti MM. 1988. Exomorfologia y
anatomia de los tumores en
Nothofagus antarctica
(Fagaceae) atribuidos a
Cyttaria harioti
y
Cyttaria
hookeri
(Cyttariaceae). Symposio sobre
Nothofagus
(1987), An Acad Nac Cienc Exact Fis Nat 4:123–134.
Heads M. 2006. Panbiogeography of
Nothofagus
(Nothofa-
gaceae): analysis of the main species massings. J
Biogeogr 33:1066–1075.
Hill RS. 1991. Tertiary
Nothofagus
(Fagaceae) macrofossils
from Tasmania and Antarctica and their bearing on the
evolution of the genus. Bot J Linn Soc 105:73–112.
———. 1996. The riddle of unique southern hemisphere
Nothofagus
on southwest Pacific islands: its challenge to
biogeographers. In: Keast A, Miller SE, eds. The origin
and evolution of Pacific island biotas, New Guinea to
eastern Polynesia: patterns and processes. Amsterdam:
SPB Academic. p 247–260.
———. 2001. Biogeography, evolution and palaeoecology
of
Nothofagus
(Nothofagaceae): the contribution of the
fossil record. Aust J Bot 49:321–332.
———, Jordan GJ. 1993. The evolutionary history of
Nothofagus
(Nothofagaceae). Aust Syst Bot 6:111–126.
———, Read J. 1991. A revised infrageneric classification of
Nothofagus
(Fagaceae). Bot J Linn Soc 105:37–72.
Humphries CJ, Cox JM, Nielsen ES. 1986.
Nothofagus
and its
parasites: a cladistic approach to coevolution. In: Stone
AR, Hawksworth DL, eds. Coevolution and systematics.
Oxford, UK: Clarendon Press. p 55–76.
Knapp M, Stockler K, Havell D, Delsuc F, Sebastiani F,
Lockhart PJ. 2005. Relaxed molecular clock provides
evidence for long-distance dispersal of
Nothofagus
(southern beech). PLoS Biol 3:38–43.
Korf RP. 1983.
Cyttaria
(Cyttariales): coevolution with
Nothofagus
, and evolutionary relationship to the
Boedijnopezizeae (Pezizales, Sarcoscyphaceae). Aust J
Bot Suppl 10:77–87.
Legendre P, Desdevises Y, Bazin E. 2002. A statistical test for
host-parasite coevolution. Syst Biol 51:217–234.
LinderHP,CrispMD.1995.
Nothofagus
and Pacific
biogeography. Cladistics 11:5–32.
Lu¨cking R, Huhndorf S, Pfister DH, Rivas Plata E, Lumbsch
HT. 2009. Fungi evolved right on track. Mycologia 101:
810–822.
Manos PS. 1997. Systematics of
Nothofagus
(Nothofagaceae)
based on rDNA spacer sequences (ITS): taxonomic
congruence with morphology and plastid sequences.
Am J Bot 84:1137–1155.
Martin PG, Dowd JM. 1993. Using sequences of
rbc
L to study
phylogeny and biogeography of
Nothofagus
species.
Aust Syst Bot 6:441–447.
McLoughlin S. 2001. The breakup history of Gondwana and
its impact on pre-Cenozoic floristic provincialism. Aust
J Bot 49:271–300.
Meier-Kolthoff JP, Auch AF, Huson DH, Go¨ker M. 2007.
COPYCAT: co-phylogenetic analysis tool. Bioinfor-
matics 23:898–900.
Peterson KR, Pfister DH. 2010. Phylogeny of
Cyttaria
inferred from nuclear and mitochondrial sequence
and morphological data. Mycologia 102:1398–1416.
Pole M. 1994. The New Zealand flora—entirely long-
distance dispersal? J Biogeogr 21:625–635.
———. 2001. Can long-distance dispersal be inferred from
the New Zealand plant fossil record? Aust J Bot 49:357–
366.
Rambaut A, Drummond AJ. 2007. Tracer v1.4, available at
http://beast.bio.ed.ac.uk/Tracer.
Rawlings GB. 1956. Australasian Cyttariaceae. Trans R Soc
NZ 84:19–28.
Sanderson MJ. 2002. Estimating absolute rates of molecular
evolution and divergence times: a penalized likelihood
approach. Mol Biol Evol 19:101–109.
———. 2003. r8s: inferring absolute rates of evolution and
divergence times in the absence of a molecular clock.
Bioinformatics 19:301–302.
Sanmartı´n I, Ronquist F. 2004. Southern hemisphere
biogeography inferred by event-based models: plant
versus animal patterns. Syst Biol 53:216–243.
———, Wanntorp L, Winkworth RC. 2007. West wind drift
revisited: testing for directional dispersal in the
southern hemisphere using event-based tree fitting. J
Biogeogr 34:398–416.
Seberg O. 1991. Biogeographic congruence in the South
Pacific. Aust Syst Bot 4:127–136.
Setoguchi H, Ono M, Doi Y, Koyama H, Tsuda M. 1997.
Molecular phylogeny of
Nothofagus
(Nothofagaceae)
based on the
atp
B-
rbc
Lintergenicspacerofthe
chloroplast DNA. J Plant Res 110:469–484.
Steenis CGGJ. 1971.
Nothofagus
, key genus of plant
geography, in time and space, living and fossil, ecology
and phylogeny. Blumea 19:65–98.
Swenson U, Backlund A, McLoughlin S, Hill RS. 2001.
Nothofagus
biogeography revisited with special empha-
sis on the enigmatic distribution of subgenus
Brasso-
spora
in New Caledonia. Cladistics 17:28–47.
Swofford DL. 2002. PAUP*: phylogenetic analysis using
parsimony (*and other methods). Version 4.0b10.
Sunderland, Massachusetts: Sinauer Associates.
Taylor J, Berbee M. 2006. Dating divergences in the fungal tree
of life: review and new analyses. Mycologia 98:838–849.
Wilson JM. 1937. The structure of galls formed by
Cyttaria
septentrionalis
on
Fagus moorei
. Proc Linn Soc NSW 62:
1–8.
Winkworth RC, Wagstaff SJ, Glenny D, Lockhart PJ. 2002.
Plant dispersal NEWS from New Zealand. Trends Ecol
Evol 17:514–520.
PETERSON ET AL.: COPHYLOGENY AND
C
YTTARIA
1425
