Universidade de São Paulo
Biblioteca Digital da Produção Intelectual - BDPI
Departamento de Zoologia - IB/BIZ
Artigos e Materiais de Revistas Científicas - IB/BIZ
2014-03
Cryptic Diversity within Morphospecies
of Testate Amoebae (Amoebozoa:
Arcellinida) in New England Bogs
and Fens
Protist, Jena, v.165, n.2, p.196-207, 2014
http://www.producao.usp.br/handle/BDPI/45011
Downloaded from: Biblioteca Digital da Produção Intelectual - BDPI, Universidade de São Paulo
Protist, Vol. 164, 323–339, May 2013
http://www.elsevier.de/protis
Published online date 13 March 2013
ORIGINAL PAPER
Multigene Phylogenetic Reconstruction of the
Tubulinea (Amoebozoa) Corroborates Four of the
Six Major Lineages, while Additionally Revealing
that Shell Composition Does not Predict
Phylogeny in the Arcellinida
Daniel J.G. Lahra,b,1 , Jessica R. Grantb , and Laura A. Katza,b,2
aGraduate
Program in Organismic and Evolutionary Biology, University of Massachusetts, Amherst,
Massachusetts, 01003, USA
bDepartment of Biological Sciences, Smith College, Northampton, Massachusetts, 01063, USA
Submitted March 2, 2012; Accepted February 11, 2013
Monitoring Editor: Sandra L. Baldauf
Tubulinea is a phylogenetically stable higher-level taxon within Amoebozoa, morphologically
characterized by monoaxially streaming and cylindrical pseudopods. Contemporary phylogenetic
reconstructions have largely relied on SSU rDNA, and to a lesser extent, on actin genes to reveal
the relationships among these organisms. Additionally, the test (shell) forming Arcellinida, one of the
most species-rich amoebozoan groups, is nested within Tubulinea and suffers from substantial undersampling of taxa. Here, we increase taxonomic and gene sampling within the Tubulinea, characterizing
molecular data for 22 taxa and six genes (SSU rDNA, actin, ␣- and ␤-tubulin, elongation factor 2 and the
14-3-3 regulatory protein). We perform concatenated phylogenetic analyses using these genes as well
as approximately unbiased tests to assess evolutionary relationships within the Tubulinea. We confirm
the monophyly of Tubulinea and four of the six included lineages (Echinamoeboidea, Leptomyxida,
Amoebida and Poseidonida). Arcellinida and Hartmanellidae, the remaining lineages, are not monophyletic in our reconstructions, although statistical testing does not allow rejection of either group.
We further investigate more fine-grained morphological evolution of previously defined groups, concluding that relationships within Arcellinida are more consistent with general test and aperture shape
than with test composition. We also discuss the implications of this phylogeny for interpretations of
the Precambrian fossil record of testate amoebae.
© 2013 Elsevier GmbH. All rights reserved.
Key words: Testate amoebae; Arcellinida; multigene phylogeny; Tubulinea; Amoebozoa.
Introduction
1
Current Address: Department of Zoology, University of Sao
Paulo, Sao Paulo, 05508-090, Brazil
2
Corresponding author; fax +1-413 585 3786
e-mail [email protected] (D.J.G. Lahr), [email protected]
(L.A. Katz).
© 2013 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.protis.2013.02.003
The Tubulinea are a monophyletic lineage within
the Amoebozoa (Smirnov et al. 2005). Unlike
many other eukaryotic groups proposed in recent
years, these amoebae are marked by a putative synapomorphy (defining character): monoaxial
324 D.J.G. Lahr et al.
streaming of cytoplasm within roughly cylindrical
pseudopods (Smirnov et al. 2005). Some organisms in this group can produce several cylindrical
pseudopods, as in the genus Amoeba, while others have a single semi-cylindrical pseudopodial
protrusion that comprises the entire body, giving
them a slug-like (limax) shape, as in the genus
Saccamoeba. The Tubulinea are currently divided
into six major groups with defining morphological characteristics and varying levels of support in
molecular reconstructions. These are: Echinamoeboidea, Leptomyxida, Amoebidae, Hartmannellidae, Poseidonida and Arcellinida. The Tubulinida (comprised of Amoebidae + Hartmannellidae),
Echinamoeboidea, Leptomyxida and the recently
described Poseidonida are recovered with high levels of support in trees based on SSU rDNA and
actin gene datasets (see Lahr et al. 2011). The
Arcellinida are a species-rich assemblage characterized by the ability to produce a test (shell).
In molecular reconstructions, the Arcellinida are
either not monophyletic or exhibit low to moderate levels of support, with more comprehensive
taxon sampling tending to decrease support (Lahr
et al. 2011; Smirnov et al. 2011). Finally, a core
group of organisms within the Hartmannellidae are
often recovered with high levels of support, but
with the lineage Saccamoeba limax ATCC 30942
branching separately from this main group, rendering the Hartmannellidae paraphyletic (Bolivar
et al. 2001; Cavalier-Smith et al. 2004; Fahrni
et al. 2003; Lahr et al. 2011; Pawlowski and
Burki 2009; Smirnov et al. 2005; Tekle et al.
2008).
Taxonomic instability also impacts genera within
the Tubulinea. The genus Hartmannella (and Family Hartmannellidae by consequence) is probably
one of the most affected by recent molecular
reconstructions. Many small (10-30 ␮m) amoebae
that present a limax-like locomotive form were originally described as different species in the genus
Hartmannella (Page 1987). Based on morphological evidence, several species were later removed
from the genus (e.g. Nolandella (Page 1983) and
Echinamoeba (Page 1975)). Molecular studies
showed that Vermamoeba vermiformis, a common
freshwater and soil amoeba recently transferred out
of Hartmannella is in fact more closely related to
Echinamoeba than to other limax-shaped amoebae
that are now considered the “core-hartmannellids”
(Glaeseria, Saccamoeba) (Amaral Zettler et al.
2000; Fahrni et al. 2003). Further, marine species
of Nolandella and H. abertawensis were shown to
form the highly-supported Poseidonida (senior synonym of Nolandida (Smirnov et al. 2011)), distinct
from other hartmannellids (Lahr et al. 2011). Surprisingly, the sorocarpic (fruiting body producing)
slime mold Copromyxa protea was shown to be
very closely related to H. cantabrigiensis, prompting transfer of the latter to the genus Copromyxa
(Brown et al. 2011).
Arcellinida are conspicuous and abundant amoebae that occupy distinct tests (shells) that have
been argued to be valuable structures for both
species delimitation and phylogenetic inference
(Meisterfeld 2002). In this group, molecular
evidence does not corroborate morphological predictions in three significant and distinct instances:
1) some genera appear not to be monophyletic,
including Heleopera and Nebela (Lara et al. 2008;
Nikolaev et al. 2005); 2) relationships proposed
based on shell form and composition are not recovered – genera within the Suborder Arcellina, which
are defined based on the possession of an organic
membranous shell, are not monophyletic (e.g. Pyxidicula, Arcella and Spumochlamys (Lahr et al.
2011)), and finally; 3) at the most inclusive level,
increased taxonomic sampling results in reduced
support for the entire group, opening up the possibility that Arcellinida is not monophyletic (Lahr
et al. 2011). However, taxonomic sampling is still
far from comprehensive in this species-rich group,
making it difficult to evaluate these taxonomic instabilities.
A further limitation of previous work is that phylogenetic inference in the Tubulinea has relied
mostly on SSU rDNA and, to a lesser extent, on
actin genes (Fahrni et al. 2003; Lahr et al. 2011;
Smirnov et al. 2011). The problems associated with
single gene reconstructions are well known and
have been extensively dealt with elsewhere (e.g.
Baldauf et al. 2000; Philippe and Douady 2003).
The actin gene family, the second most sampled
marker, poses challenges for phylogenetic reconstruction due to high levels of paralogy present in
many members of the group (Lahr et al. 2010).
Here, we present a phylogenetic reconstruction of
the Tubulinea that capitalizes on sampling of SSU
rDNA, actin and additional genes coding for four
proteins: ␣- and ␤-tubulin, eukaryotic elongation
factor 2 (eEF2), and the regulatory protein 14-33. We provide molecular data for 22 taxa from all
six currently defined groups in the Tubulinea, with
greatest emphasis on the diverse Arcellinida (15
taxa), adding a total of 111 new gene sequences.
We perform phylogenetic reconstructions including a representative sample of eukaryotes to test
monophyly at higher taxonomic levels, as well
as specific hypotheses of evolution within the
Tubulinea.
Multigene Analysis of Tubulinea 325
Results
monophyletic, we chose the shortest branching paralog.
Genes Characterized
We characterized a total of 111 gene sequences
for 22 taxa (Table 1): nine SSU rDNAs for seven
taxa (752-1958 bp), 45 actin genes from 18 taxa
(∼796 bp), a total of 18 ␣-tubulin genes for 14
taxa (∼900 bp), 12 ␤-tubulin genes for 10 taxa
(∼450 bp), 9 eEF2 genes for 9 taxa and (∼840 bp),
and 15 14-3-3 genes for 10 taxa (∼500 bp). For
SSU rDNA, both Lesquereusia spiralis and Heleopera sphagni yielded multiple sequences: the 2
SSU rDNAs for H. sphagni are identical except that
one contains a group I intron; two of the three L.
spiralis SSU rDNAs are very similar (0.6% divergence) while a third averages 2.4% divergence from
the other two. As the DNA extraction for both taxa
was performed from a pool of individuals, the yield
of multiple SSU rDNAs may reflect intra-population
variation.
We found varying levels of paralogy in proteincoding genes. There is extensive paralogy of actin
genes as expected based on previous work on the
genus Arcella (Lahr et al. 2010), with 11 of the 18
taxa sampled here containing gene duplications.
For ␣-tubulin the taxa Difflugia cf. lacustris and
Quadrulella symmetrica contained paralogs; for ␤tubulin the taxa Difflugia cf. lacustris. and Chaos
carolinensis contained paralogs, for 14-3-3 the taxa
Hyalosphenia papilio, Nebela penardiana and Netzelia wailesi had paralogs (Table 1). We found no
indication of paralogy for the gene for eEF2.
Single Gene Tree Results
We performed individual phylogenetic reconstructions on each of the genes sampled, to look for
ancient gene duplication events in order to choose
appropriate genes for concatenation. Single-gene
trees made from protein coding genes have so far
shown limited utility for reconstructing deep relationships in Amoebozoa. Additionally, the variable
taxon sampling for each gene makes comparisons
difficult (Figure 2 shows phylogenies for protein
coding genes and Supplementary Figure S1 shows
the phylogeny for SSU rDNA). In most cases for
protein coding genes, paralogy seems to occur
independently at shallow levels, without evidence
for ancient duplications. For both actin and 14-3-3,
there is evidence of duplication events that predate the divergence of genera within the Nebelidae
and so we used the topologies generated here to
choose likely orthologs for concatenation. In other
cases where multiple paralogs for an isolate were
Concatenated Trees Results
General topology: The topology obtained from
the concatenated analyses (Fig. 3A) is largely
congruent with comprehensive eukaryotic analyses (e.g. Hampl et al. 2009; Parfrey et al. 2010;
Yoon et al. 2008) and Amoebozoa-specific reconstructions (Lahr et al. 2011; Shadwick et al. 2009;
Smirnov et al. 2005; Tekle et al. 2008). The
Tubulinea appear monophyletic without bootstrap
support (<50% BS, Fig. 3A), and four of the six
major included lineages receive moderate to high
support (Fig. 3B): Echinamoeboidea (73% BS),
Leptomyxida (99% BS), Amoebida (Hartmannellidae + Amoebidae 93% BS), Amoebidae (100%
BS) and Poseidonida (100% BS). Two remaining
lineages are non-monophyletic as in both cases
a single taxon branches separately from the main
group of taxa (Fig. 3B). For the Arcellinida, both
isolates of Heleopera sphagni branch separately
from the core group. The SSU rDNA and actin
sequences reported from our isolate of H. sphagni
from Massachusetts closely match previously published sequences from an H. sphagni isolate in
Switzerland, indicating that the position of H.
sphagni in our tree is unlikely due to misidentification or contamination. However, the approximately
unbiased (AU) test cannot reject the possibility that
Arcellinida sensu strictu (including H. sphagni) is
monophyletic (Table 3). The Hartmannellida are
also not monophyletic, with a highly supported core
group containing Saccamoeba lacustris, Glaeseria
mira and Copromyxa spp. (100% BS, Fig. 3B) more
closely related to the Amoebidae than to the lineage
Saccamoeba limax ATCC 30942. Again, an AU test
cannot reject the possibility that the Hartmannellida
sensu strictu (i.e. including S. limax ATCC 30942)
is monophyletic (Table 3).
Phylogeny of the Major Tubulinea
Lineages
The internal topologies of major Tubulinea lineages
are generally concordant with morphological observations as well as previous phylogenetic reconstructions (Lahr et al. 2011; Smirnov et al. 2005;
Fig. 3b). The few exceptions are detailed below.
Within the Echinamoeboidea, the genus Echinamoeba is monophyletic (100% BS) and our newly
isolated Vermamoeba vermiformis groups with the
other available V. vermiformis strain with characterized ESTs (100% BS). The Echinamoeboidea is
326 D.J.G. Lahr et al.
Table 1. Distribution of the 111 genes characterized from 22 taxa.
Taxon
Material Source Genetic SSU rDNA
Material
Actin
␣tub
␤tub
eEF2
14-3-3
Arcella gibbosa
Arcella hemisphaerica
C. operculata
Difflugia bryophila
Difflugia lanceolata
Difflugia cf. lacustris
Heleopera sphagni
Hyalosphenia papilio
Lesquereusia modesta
Lesquereusia spiralis
Nebela carinata
Nebela penardiana
Netzelia wailesi
Netzelia tuberculata
Q. symmetrica
V. vermiformis
Saccamoeba lacustris
Rhizamoeba saxonica
Nolandella hibernica
Chaos carolinense
Amoeba proteus
Nolandella sp.
Bear Swamp
CB 131310
CB 131934
Hawley Bog
Hawley Bog
Hawley Bog
Hawley Bog
Hawley Bog
Bear Swamp
CB 131334
Hawley Bog
Hawley Bog
Hawley Bog
Hawley Bog
Hawley Bog
Smith Coll.
CCAP 1572/4
CCAP 1570/2
CCAP 1534/10
CB 131324
CB 131306
ATCC 50913
JQ519417-18
JQ519421
JQ519499-501
JQ519422-27
JQ519428-30
JQ519431
JQ519437-43
JQ519444-45
JQ519449-51
JQ519447-48
JQ519452-54
JQ519455
JQ519432-36
JQ519457-58
JQ519456
JQ519420
JQ519419
-
JQ519459
JQ519460
JQ519463
JQ519464
JQ519465-67
JQ519469
JQ519468
JQ519471
JQ519472
JQ519473-75
JQ519470
JQ519476
JQ519462
JQ519461
-
JQ519477
JQ519480
JQ519481-82
JQ519483
JQ519485
JQ519486
JQ519487
JQ519484
JQ519488
JQ519478-79
-
JQ519489
JQ519490
JQ519492
JQ519495
JQ519497
JQ519494
JQ519498
JQ519491
JQ519496
JQ519397
JQ519399
JQ519400-02
JQ519403
JQ519404-05
JQ519407
JQ519410-13
JQ519408-09
JQ519414
JQ519406
JQ519416
JQ519398
-
c
c, e
e
c
c
c
c
c, a
c
c
c, a
c
c
c
c, a
c, a, e
c
c
c
c
c
c
JQ519503-04
JQ519506-08
JQ519509
JQ519511
JQ519505
JQ519510
JQ519502
-
Note: c – cDNA, a – genome amplification, e – genomic extraction.
Multigene Analysis of Tubulinea 327
Table 2. Genes selected for concatenation.
Taxon
actin
␣-tubulin
␤-tubulin
eEF2
14-3-3
A. proteus
A. gibbosa
A. hemisphaerica
C. carolinesis
C. operculata
Difflugia (comb)
V. vermiformis
H. sphagni
H. papilio
Lesquereusia (comb)
N. carinata
N. penardiana
Netzelia (comb)
Q. symmetrica
R. saxonica
S. lacustris
Nolandella sp. 50913
JQ519419
JQ519417
HM853688
JQ519420
JF694279
JQ519422
JQ519432
JQ519428
JQ519431
JQ519437
JQ519446
JQ519449
JQ519447
JQ519455
JQ519456
JQ519457
EU273446
JQ519461
JQ519459
JQ519460
JQ519462
JQ519463
JQ519464
JQ519470
JQ519469
JQ519468
JQ519471
JQ519472
JQ519473
JQ519476
EU273448
JQ519477
JQ519478
JQ519480
JQ519482
JQ519484
JQ519483
JQ519485
JQ519486
JQ519487
JQ519488
EU273450
JQ519489
JQ519490
JQ519491
JQ519492
JQ519494
JQ519493
JQ519495
JQ519497
JQ519498
JQ519496
JQ519397
JQ519398
JQ519399
JQ519400
JQ519406
JQ519403
JQ519404
JQ519407
JQ519413
JQ519408
JQ519414
JQ519415
JQ519416
-
Note: Accession numbers in italics were previously available in GenBank. (comb) indicates that genes from
multiple morphospecies within a genus were combined for concatenation.
not only monophyletic (Fig. 3B), but the AU test also
rejects the possibility of its grouping with any other
major tubulinid lineage (Table 3).
The topology of Leptomyxida is generally concordant with previous phylogenetic reconstructions,
with the exception of the position of the isolate
Rhizamoeba saxonica CCAP 1570/2. We recover
two highly supported groups within the Leptomyxida (Fig. 3B). The isolate Rhizamoeba saxonica
CCAP 1570/2, considered as the most morphologically accurate representative of the Rhizamoeba
genus (Smirnov et al. 2008), falls within a highly
supported group (93% BS) along with representatives of Paraflabellula, Flabellula and the isolate
‘Rhizamoeba’ sp. ATCC 50933 (Fig. 3B). This result
contrasts with previous reconstructions using only
SSU rDNA, where R. saxonica falls sister to all
other Leptomyxida (Dykova et al. 2008a; Smirnov
et al. 2008), or sister to the group comprised of Leptomyxa reticulata, Rhizamoeba neglecta and two
strains identified as Ripidomyxa sp. (Smirnov et al.
2009). We have sequenced two genes for this isolate (SSU rDNA and actin). The SSU rDNA was
identical to the previously published sequence for
this isolate (GenBank #EU719197). Because the
actin groups with Ancyromonas in the gene phylogeny (Fig. 2A), we ran an additional independent
maximum likelihood analysis using the concatenated dataset but only SSU rDNA to represent this
lineage, as a control in case the actin was a contaminant. The resulting topology was identical (not
shown). The second group within the Leptomyxida
(100% BS) contains the isolate Leptomyxa reticulata ATCC 50242, a Ripidomyxa sp. isolate, as
well as Rhizamoeba neglecta, consistent with the
reconstruction in Smirnov et al. (2009) and the SSU
rDNA single gene reconstruction (Supplementary
Fig. S1).
The Hartmannellidae topology recovered here
is congruent with previous reconstructions as
strain Saccamoeba limax ATCC 30942 branches
separately from a well-supported group of “core
hartmannelids” (Copromyxa cantabrigiensis, Saccamoeba lacustris CCAP 1572/4 and Glaeseria
mira plus Amoebidae, Fig. 3b). This result is consistent with the majority of previous reconstructions
(Amaral Zettler et al. 2000; Bolivar et al. 2001;
Brown et al. 2011; Cavalier-Smith et al. 2004;
Corsaro et al. 2010; Fahrni et al. 2003; Shadwick
et al. 2009; Tekle et al. 2008). However, the AU test
does not reject the possibility that the Hartmannellidae sensu strictu (i.e. including S. limax) could be
monophyletic (Table 3).
The Poseidonida appear monophyletic and
strongly supported (100% BS, Fig. 3B), with the
addition of a partial SSU rDNA sequence for
the isolate Nolandella hibernica CCAP 1534/10.
This is the strain used in the original description of the taxon, though the original designation
was Hartmannella hibernica in Page (1980) and
then transferred to Nolandella hibernica in Page
(1983). This result taxonomically validates the
328 D.J.G. Lahr et al.
Multigene Analysis of Tubulinea 329
Nolandellidae Lahr and Katz, 2011 and Poseidonida Lahr and Katz, 2011, since the type strain
is now shown to nest within the previously characterized lineages.
The Amoebidae are monophyletic (100% BS,
Fig. 3B). The SSU rDNA sequence for the isolate
of Chaos carolinense presented here is identical to
the previously-characterized SSU rDNA (accession
#AJ314607). Two groups emerge within the Amoebidae, corresponding to the genera Chaos and
Amoeba (both with <50% BS). This result is contradictory to previous reconstructions based solely on
SSU rDNA where lineages of Amoeba and Chaos
interdigitate (Smirnov et al. 2005).
The non-monophyly of the Arcellinida, with two
H. sphagni isolates branching separately from the
main group of Arcellinida (<50% BS, Fig. 3B), is
inconsistent with previous reconstructions of the
Arcellinida (Kudryavtsev et al. 2009; Lahr et al.
2010, 2011; Lara et al. 2008; Nikolaev et al. 2005).
However, highly variable taxon inclusion among
studies makes comparisons difficult. The AU test
does not reject the possibility that the Arcellinida
are monophyletic (Table 3). Two groups within the
Arcellinida have high support: a group uniting Netzelia and Arcella (86% BS, Fig. 3B) and a group
uniting the Hyalosphenidae and Nebelidae (84%
BS, Fig. 3B), also recovered previously (Lahr et al.
2011; Lara et al. 2008). These two highly supported
groups are in disagreement with the morphologically based classification of Meisterfeld (2002),
where the Hyalosphenidae and Nebelidae are independent lineages, and Arcella and Netzelia are in
separate clades due to differences in shell composition.
Three previously proposed groups within the
Arcellinida are not recovered: the Suborders
Arcellina and Difflugina, and the Family Lesquereusiidae. The Arcellina comprise amoebae
capable of producing organic membranous or
chitinous shells, and are represented in the current sampling by the genera Arcella, Pyxidicula
and Spumochlamys. The Arcellina appears polyphyletic (Fig. 3B): Arcella is in a well-supported
clade with Netzelia (86% BS), Spumochlamys is in
a non-supported clade with Difflugia and Pyxidicula
appears at the base of the Arcellinida clade, along
with Cryptodifflugia operculata. However, monophyly of the Arcellina or any combination of two taxa
cannot be rejected by AU tests (Table 3). The group
Difflugina comprises the majority of Arcellinida,
uniting amoebae that construct the shell by agglutination and are represented in the present study with
members of 9 out of 11 putative families (Heleoperidae, Hyalospheniidae, Difflugiidae, Nebelidae,
Lesquereusiidae, Paraquadrulidae, Centropyxidae,
Plagiopyxidae, Trigonopyxidae). The Suborder is
not monophyletic in reconstructions here (Fig. 3B),
and monophyly of the group is rejected by the
AU test (Table 3). The Lesquereusiidae, defined
as the Arcellinida capable of biomineralizing silica (Ogden 1979), originally included the genera
Lesquereusia, Quadrullela and Netzelia, with the
later additions of Microquadrulla and the marine
Pomoriella (Meisterfeld 2002). This group is not
monophyletic in our multigene reconstructions:
Quadrulella appears within the Nebelidae, Lesquereusia is sister to a Difflugia+Spumochlamys
clade, and Netzelia is in a well-supported position
sister to the genus Arcella. Additionally, AU tests
reject the possibility that Lesquereusiidae – the
genera Lesquereusia, Quadrulella and Netzelia –
is monophyletic (Table 3). However, the monophyly
of Netzelia and Lesquereusia cannot be rejected
(Table 3). The position of Quadrulella is confirmed
by phylogenetic analyses of cytochrome oxidase 1
(Kosakyan et al. 2012).
Monophyly of genera within the Arcellinida is variable: while the genera Arcella and Spumochlamys
are monophyletic (86% and 97% BS respectively,
➛
Figure 1. Images of organisms used in this study. a) Chaos carolinensis, stack of eight images under DIC; b)
two Amoeba proteus individuals, differential interference contrast (DIC); c, d) individuals of Difflugia bryophila
that were genome amplified, Hoffman Modulation Contrast (HMC); e) individuals of Heleopera sphagni that were
genome amplified, HMC image; f, g) details of Heleopera sphagni shell under scanning electron microscopy;
h, i) individuals of Lesquereusia modesta that were genome amplified, HMC images; j, k, l, m) individuals of
Quadrulella symmetrica that were genome amplified (j, k) and had their cDNA libraries constructed (l, m); n)
Lesquereusia spiralis individual that was genome amplified (HMC); o) Hyalosphenia papilio that was genome
amplified (HMC); p) Nebela carinata; q) representative individual from culture of Saccamoeba lacustris CCAP
1572/4 (DIC); r, s) representative individuals from Rhizamoeba saxonica CCAP 1570/2 (DIC); t, u) Netzelia
wailesi individual that was genome amplified; v, x) Netzelia tuberculata individual that was genome amplified,
although images don’t quite show the characteristic protuberances of the shell, these were prominent while
observing the living individual. Scale bars are 100 ␮m for a, b, c, e, i, n, o, p, t (bar for d is same as c); 50 ␮m
for h, j, k, l, m, u, v, x; 30 ␮m for f; 25 ␮m for q, r, s; and 10 ␮m for g.
330 D.J.G. Lahr et al.
Multigene Analysis of Tubulinea 331
Fig. 3B), the other three genera represented by
more than one species are not monophyletic
(Heleopera, Hyalosphenia and Nebela, Fig. 3B).
However the AU test cannot reject the monophyly
of any of these three genera (Table 3).
Discussion
The addition of 22 taxa combined with larger gene
sampling reveals a phylogeny that is generally
consistent with hypotheses on the six principal
lineages of Tubulinea (Fig. 3B), albeit still with
low resolution at deep nodes. The monophyly
of Echinamoebidae, Leptomyxida, Poseidonida,
“Hartmannellidae” (excluding Saccamoeba limax)
and Amoebidae that were previously recovered
in numerous SSU rDNA and actin gene reconstructions (Amaral Zettler et al. 2000; Brown et al.
2011; Cavalier-Smith et al. 2004; Corsaro et al.
2010; Dykova et al. 2008a, b; Fahrni et al. 2003;
Kudryavtsev et al. 2009; Lahr et al. 2010, 2011;
Lara et al. 2008; Nikolaev et al. 2005; Smirnov et al.
2005, 2009; Tekle et al. 2008) are confirmed here
with the addition of sequences for four protein coding genes (encoding ␣- and ␤-tubulins, eEF2 and
the 14-3-3).
The Arcellinida appear in our maximum likelihood tree as polyphyletic (Fig. 3B). However, this
split of Arcellinida does not receive substantial
bootstrap support and an AU test does not conclusively reject topologies where Arcellinida are
monophyletic (Table 3). Given still incomplete taxonomic sampling within the group coupled with lack
of statistical support, it would be premature to conclude that Arcellinida is not a natural group. With
the exception of a strongly supported relationship
between the Amoebidae and the “Hartmannelidae”,
a group termed Tubulinida (Smirnov et al. 2005), the
relationships among the six main lineages remain
uncertain. This is seen even with the addition of four
protein coding genes, as evidenced by low support
for the backbone of the tree (Fig. 3B).
The more comprehensive sampling presented
here, which emphasizes Arcellinida, enables more
fine-grained analysis within that group. Higher-level
relationships within the Arcellinida are currently
defined according to shell composition, though this
classification was proposed as explicitly provisional
(Meisterfeld 2002). The three more inclusive groups
are: 1) Difflugina, characterized by an agglutinated
shell composed of either collected particles (xenosomes, eg. Difflugia) or biomineralized particles
(idiosomes, eg. Lesquereusia); 2) Arcellina, characterized by a secreted organic membranous (eg.
Microchlamys) or chitinoid shell (eg. Arcella); and
3) Phryganelina, which are classified separately
by their distinguished pseudopodial morphology
rather than by features of the shell (Cryptodifflugia
and Phryganella) (Meisterfeld 2002). The Arcellina
(represented here by the genera Arcella, Pyxidicula
and Spumochlamys) do not appear monophyletic,
although monophyly is not rejected by the AU test
(Fig. 3B, Table 3). The monophyly of Difflugina is
rejected by the AU test and relatively high BS values (Fig. 3B, Table 3), indicating that agglutination
is either an ancestral character state in the group
(a symplesiomorphy) or evolved several times (a
convergence). The current topology indicates that
symplesiomorphy is a more likely hypothesis as
groups that secrete organic tests appear to derive
from taxa with presumably an ancestral state of
agglutinated tests (genus Arcella and the Nebelidae/Hyalosphenidae group; Fig. 4). The monophyly
of the Lesquereusiidae can also be rejected by AU
tests (Table 3), indicating at least two origins of
silica biomineralization within the Arcellinida (the
monophyly of Lesquereusia + Netzelia cannot be
rejected, but Quadrulella falls within the Nebelidae,
separate from the other two; Fig. 3B).
The current reconstruction, as well as other
recent phylogenies based on SSU rDNA, actin and
cox1 (Gomaa et al. 2012; Kosakyan et al. 2012;
Lahr et al. 2011) suggest that shell shape might be
more indicative of relationships within Arcellinida
than shell composition, as suggested by Gomaa
et al. (2012) (Fig. 4). Organisms with similar shell
shape group together. Arcella and Netzelia both
have shells that are round in cross-section with a
round aperture, while Quadrulella and other Nebelidae (Hyalosphenia, Nebela, Apodera, Porosia)
have vase-shaped shells that are flattened in crosssection and have ellipsoid apertures (Fig. 4). Of
further evolutionary interest, there are a number of
“intermediary” taxa, taxa displaying the shell shape
of one group and shell composition of another,
➛
Figure 2. Phylogenies for each of the protein coding genes surveyed in the present study, including all characterized paralogs. The taxa in each tree are the ones that were concatenated in the final analyses. Single gene
trees are shown for: A) actin; B) ␣-tubulin; C) ␤-tubulin; D) eEF2; E) 14-3-3 regulatory protein. Scale bar for
each phylogeny is indicated underneath the respective tree. Only bootstrap supports above 70% are shown.
An SSU rDNA tree is available as Supplementary Figure S1.
332 D.J.G. Lahr et al.
Figure 3. Maximum likelihood phylogeny of Amoebozoa based on five concatenated gene sequences: A) Tree
emphasizing general relationships between the Amoebozoa. B) Same tree emphasizing relationships within
the Tubulinea. Taxa in bold are novel data. Branches are drawn to scale, except in cases indicated by a break,
where branches where cut in half. Dashed lines indicate non-monophyletic groupings. Only bootstraps above
50% are shown.
which have yet to be sampled for molecular data.
Lesquereusia mimetica has the typical Lesquereusia shell, with the neck bent over the body of
the test. However L. mimetica’s shell is built with
roughly agglutinated material, in a manner more
similar to Difflugia (figs 21-28 in Lahr and Lopes
2007). Similarly, Difflugia gramen and D. achlora
have shells similar in shape to Netzelia (round shell
with lobed aperture) but agglutination is more like
Difflugia, (i.e., with the absence of idiosomes; figs
11-15 in Lahr and Lopes 2006). Netzelia themselves are able to both autogenously produce silica
and incorporate carbohydrate material after adding
a layer of silica (Anderson 1992). Pseudonebela
africana and Padaungiella nebeloides are both
shaped like pyriform Difflugia (vase-shaped shell,
with round cross-section and round apertures,
respectively; figs 1b-m in (Lahr and Souza 2011)
and figs 6-11 in (Todorov et al. 2010)) but the shell
composition is more akin to that of Nebela, with
agglutinated biomineralized plates.
Certain assumptions about test construction in
the Arcellinida may need to be revised in light
of combined results from the present work and
Multigene Analysis of Tubulinea 333
Profile
Composition
Org
Lesquereusia sp
Biom
Difflugia bryophila
Aggl
Spumochlamys spp
Org
Hyalosphenia elegans
Org
Hyalosphenia papilio
Org
Nebela carinata
Aggl
Porosia bigibbosa
???
Nebela tincta
Aggl
Quadrulella symmetrica
Biom
Apodera vas
Aggl
Nebela lageniformis
Aggl
Trigonopyxis arcula
Aggl
Bullinularia indica
Aggl
Centropyxis laevigata
Aggl
Heleopera rosea
Aggl
Argynia dentistoma
Aggl
Cryptodifflugia operculata
Org
Pyxidicula operculata
Org
Heleopera sphagni
Aggl
Outgroups
Tubulinida
Hartmannellidae
Poseidonida
Leptomyxida
Echinamoeba spp
Vermamoeba vermiformis
Incertae Sedis
Biom
Nebelidae
Netzelia wailesi
Novel Clade
Arcella spp
Aperture
Figure 4. Relationships among the Tubulinea, illustrating morphological traits. Shell composition for Arcellinida
is indicated in parenthesis. Branches were collapsed to polytomies where support is less than 70% in Figure 3.
Branches are not drawn to scale.
others (Gomaa et al. 2012; Kosakyan et al. 2012).
The siliceous plates in Nebela are assumed to
be collected either from the environment or from
prey organisms, rather than autogenously produced (Meisterfeld 2002). However results placing
Quadrulella amidst the Nebelidae prompt a reevaluation of this assumption as it is possible that at
least some members of the Nebelidae synthesize
silica. If so, a case of parallel evolution can be drawn
by comparing events in the two well supported
clades shown here (Fig. 4): Netzelia biomineralizes silica and its sister-group Arcella secretes an
organic shell; in the Nebelidae/ Hyalosphenidae,
genera Nebela, Quadrulella, Porosia and Apodera
biomineralize silica while the nested Hyalosphenia produce an organic shell. In other words, the
334 D.J.G. Lahr et al.
Table 3. Results from the approximately unbiased test (AU).
Constraint tested
wkh
au
wsh
Poseidonida+Hartmannellida s.s.
Poseidonida+Hartmannellida core
Poseidonida+Amoebidae
Poseidonida+Arcellinida s.s.
Poseidonida+Arcellinida core
Poseidonida+Leptomyxida
Poseidonida+Echinamoeboidea
Amoebidae+Hartmannellidae s.s.
Amoebidae+core Hartmannelidae
Amoebidae+Arcellinida s.s.
Amoebidae+core Arcellinida
Amoebidae+Echinamoeboidea
Amoebidae+Leptomyxidae
Hartmannellidae s.s.
Hartmannellidae s.s.+Arcellinida s.s.
Hartmannellidae core+Arcellinida core
Hartmannellidae s.s.+Echinamoeboidea
Hartmannellidae core+Echinamoeboidea
Hartmannellidae s.s.+Leptomyxida
Hartmannellidae core+Leptomyxida
Leptomyxida+Echinamoeboidea
Arcellinida s.s.
Arcellinida s.s.+Echinamoeboidea
Arcellinida core+Echinamoeboidea
Arcellinida s.s.+Leptomyxida
Arcellinida core+Leptomyxida
Difflugina
Lesquereusiidae
Lesquereusia+Netzelia
Arcellina
Arcella+Pyxidicula
Arcella+Spumochlamys
Spumochlamys+Pyxidicula
Nebela
Heleopera
Hyalosphenia
0.00
0.00
0.00
0.17
0.33
0.00
0.00
0.46
0.45
0.00
0.00
0.00
0.00
0.33
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.24
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.02
0.03
0.15
0.25
0.03
0.15
0.47
0.00
0.00
0.00
0.20
0.36
0.00
0.00
0.58
0.58
0.02
0.00
0.00
0.00
0.40
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.31
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.02
0.05
0.20
0.33
0.02
0.20
0.59
0.01
0.03
0.06
0.70
0.92
0.00
0.00
0.99
0.99
0.00
0.00
0.00
0.00
0.96
0.00
0.00
0.00
0.00
0.00
0.00
0.07
0.92
0.00
0.01
0.00
0.00
0.00
0.00
0.17
0.24
0.34
0.80
0.88
0.29
0.77
0.99
Notes: The constraints tested column lists the taxa that were tested for monophyly. Wkh – weighted KishinoHasegawa test; au – Approximatelly unbiased test; wsh – weighted Shimodaira-Hasegawa test. The tests are
listed in increasing order of conservativeness, that is, the wkh test is the least conservative, most prone to type
I error. The wsh is the most conservative, most prone to type II error. The au test is the most balanced test. In
bold are the relationships that can be rejected as monophyletic.
evolution of the ability to biomineralize silica in both
of these clades may have then been followed by
independent loss of this character in genera Arcella
and Hyalosphenia.
Within the Arcellinida there is extensive lack
of monophyly for morphologically well-established
genera: Nebela, Heleopera and Hyalosphenia all
do not appear monophyletic in the current and
previous reconstructions (Lahr et al. 2011; Lara
et al. 2008), although the AU test does not allow
rejection of monophyly for any of these taxa
(Table 3). All three genera are well defined morphologically, and it comes as a surprise that multiple
isolates end up in disparate portions of the tree
(Fig. 3B). One possibility is that some of the isolates are contaminants or misidentifications. This
possibility is unlikely, for in all cases two laboratories
have independently isolated cells and generated
sequences: the Mitchell laboratory (Gomaa et al.
2012; Lara et al. 2008) and our group (current
Multigene Analysis of Tubulinea 335
study). The Heleopera case is particularly interesting because members of this genus have a known
green-algal endosymbiont, and may have others.
Thus, there is still a possibility that multiple labs
have sequenced an internal organism, a resident
symbiont, or undigested food, and the true Heleopera sphagni SSU rDNA remains unknown. It is
also possible that adding representatives of the 44
unsampled genera in Arcellinida (Meisterfeld 2002)
will help resolve or even change the topology.
A lack of monophyly among less inclusive lineages (e.g., genera and species) runs rampant
in the Tubulinea beyond the Arcellinida. The
genus Hartmannella is probably the most striking
example, with representatives scattered in three
of the five major tubulinid lineages (Fig. 3B).
It is noteworthy that Copromyxa cantabrigiensis
and Vermamoeba vermiformis were only recently
removed from the genus (Brown et al. 2011;
Smirnov et al. 2011). Taking into account that the
type strain Hartmannella hyalina is lost (Brown et al.
2011; Page 1967), it will be extremely difficult to
determine which of the many lineages should retain
the taxon name and Hartmannella may qualify as
nomen nudum. Perhaps the best solution will be to
invalidate the genus, by transferring or proposing
novel genera for each of the three major lineages.
Hartmannella abertawensis stands out as an immediate candidate to be transferred to Nolandella,
given its stable position in the current reconstruction as well as morphological characteristics and
ecology. However this should only be done after an
appropriate morphological assessment. Another
case of a non-monophyletic genus is Rhizamoeba,
at least given the current taxon and gene sampling.
Current results corroborate non-monophyly found
in other SSU rDNA reconstructions (Dykova et al.
2008a; Smirnov et al. 2009; Smith et al. 2008).
Observations of diverse testate amoebae in
the Precambrian combined with the phylogeny
of Arcellinida presented here generate several
hypotheses on the early evolution of this group.
Arcellinida fossils in marine sediments from
750mya represent some of the oldest, unambiguous records of extant eukaryotic lineages (Porter
and Knoll 2000; Porter et al. 2003). There is
considerable taxonomic diversity in these marine
sediments, including specimens that morphologically similar to Arcella, Lesquereusia, Difflugia and
Heleopera (Bosak et al. 2011; Porter et al. 2003).
These fossil marine morphospecies interdigitate
with the freshwater species characterized for this
study, yet only a few extant marine representatives
have been described (e.g. Pomoriella valkanovi
(Golemansky 1970)). Further, the monophyly of
Arcellinida (with or without Heleopera sphagni)
and the marine Poseidonida cannot be rejected
(Table 3). Synthesis of these observations lead to
two related possibilities: 1) Arcellinida evolved in
a marine environment, perhaps from a common
ancestor with the Poseidonida, and subsequently
invaded the freshwater environments in which they
are now common after diversification; 2) there is
considerable diversity of extant marine Arcellinida
yet to be discovered.
Conclusions
The phylogeny of Tubulinea presented here corroborates previously defined higher-level groups,
but does not resolve the relationships among them.
Similarly, previously well-supported groups remain
so, while others, such as Arcellinida, still require
further study. Better resolution may come from
increased sampling of both genes and taxa: for
example, over 40 genera in the Arcellinida have
not been sampled for molecular data. Alternatively, heterogenous rates of evolution or other
complexities of genome evolution in these lineages
may confound attempts to reconstruct phylogenetic relationships with existing methodologies. The
main advances provided by the approach taken
here lie in the resolution of lower-level relationships: 1) Rhizamoeba saxonica CCAP 1570/2 has
been placed at a distinct position than previous
reconstructions that relied on SSU rDNA alone;
2) genera within Amoebidae were recovered as
monophyletic for the first time; and 3) extensive
non-monophyly has been demonstrated for additional genera, which will require additional studies
to resolve.
Methods
Taxon sampling: Amoebae were obtained by two methods: 1)
culturing of newly isolated or deposited strains and 2) isolation,
photo-documentation and genome amplification or construction of cDNA libraries of individuals or small groups of freshly
isolated organisms (Table 1, Fig. 1). Arcella hemisphaerica,
Cryptodifflugia operculata and Vermamoeba vermiformis were
isolated and cultured as previously described (Lahr et al. 2010;
Lahr and Katz 2009). Chaos carolinensis (Cat. no 131324),
Amoeba proteus (Cat. No 131306) and Lesquereusia spiralis
(Cat. no 131334, listed as Difflugia) cultures were obtained
from Carolina Biological Supply. These amoebae were cleaned
by multiple transfers of sterilized pond water and allowed
to sit overnight to finish digestion of prey organisms before
being subjected to cDNA construction. Arcella gibbosa, Difflugia bryophila, Difflugia lanceolata and Difflugia cf. lacustris,
Heleopera sphagni, Hyalosphenia papilio, Lesquereusia modesta, Nebela carinata, Nebela penardiana, Netzelia wailesi,
336 D.J.G. Lahr et al.
Netzelia tuberculata and Quadrulella symmetrica? were isolated from natural sources (details in Table 1), cleaned through
successive transfers, photodocumented and then subjected to
genome amplification and/or construction of cDNA libraries.
Saccamoeba lacustris CCAP 1572/4, Rhizamoeba saxonica
CCAP 1570/2 and Nolandella hibernica CCAP 1534/10 were
obtained from the Culture Collection of Algae and Protozoa.
These cultures were grown according to instructions from the
repository, and large numbers of amoebae were harvested for
cDNA construction.
DNA and cDNA isolation: Genetic material was obtained
by three methods: 1) genomic extraction; 2) genomic amplification and 3) construction of cDNA. For genomic extraction
(gEXT), cultures were grown either in liquid media or agar
plates as previously described (Lahr et al. 2011), amoebae were
harvested and cleaned either through several washes or by filtering, and subjected to a standard phenol/chloroform protocol
(Lahr et al. 2011). For genomic amplification (gAMP), one or a
small group of organisms was isolated, cleaned through washes
in sterile water, left overnight to finish digestion of prey organisms, and subjected to amplification using a Repli-g Genomic
amplification kit (Qiagen, Cat. No. 150023) following manufacturer’s directions. The same strategy for isolation and cleaning
of organisms was adopted for construction of cDNA libraries,
but in the final step organisms were subjected to lysis and
cDNA contruction protocol through a SuperScript III Cells Direct
kit (Invitrogen, Cat. No. 18080-200), following manufacturer’s
instructions.
Amplification of target genes, cloning and sequencing:
We performed amplification of genes of interest using Phusion
Hot Start DNA polymerase (New England BioLabs, Cat. no.
F540), following manufacturer’s instructions. Small subunit ribosomal DNA (SSU rDNA) and actin genes were amplified with
previously described primers (Medlin et al. 1988; SnoeyenbosWest et al. 2002; Tekle et al. 2007). The remaining genes were
amplified with newly designed primers (␤-tubulin: TGG GCT
AAG GGT CAY TAY ACN GAR GG, CTC CGT TTC CCN GGN
CAR YTN AA, GAA GAA GTG NAG NCK NGG RAA NGG,
GGT GTA CCA GTG NAR RAA RGC YTT; ␣-tubulin: GGC AAG
GAG GAC GCN GCN AAY AAY TWY GC, TTG AAG CCT GTC
GGR CAC CAR TCN ACRAAY TG, ACC TTC GCC GAC RTA
CCA RTG NAC RAA NGC; 14-3-3: CTG AGC AAG CTG ARM
GNT AYG ANG ARA TGG, GTT GCC TAC AAR AAY GTY RTY
GGN GC, AGT GCA AGA CCN ARN CGG ATN GGG TG, GCG
ATG GCA TCA TCG AAN GCN TGR TTN GC; ef2␣: GAA GTC
ACT GCT GCN CTN CGN GTN ACN GA, GGT GTT TGC GTC
CAA ACN GAR ACN GTN CT, CGC CCG AAG GCA TAG AAN
CGN CCY TTR TC, AAA TCT CCA GGT GNA GYT CNC CNG
CNC C). In general, reactions were performed on serial dilutions
of starting material (1x, 1:10x, 1:100x, 1:500x) to determine
the lowest amount of starting DNA necessary for amplification,
in an attempt to minimize the formation of chimeras. Detailed
conditions are explained in (Lahr and Katz 2009). Successfully
amplified products were then gel isolated using the Millipore
Ultra Free DA spin column, and cloned using the Zero Blunt
TOPO cloning kits (Invitrogen, Cat. No. K280020) according
to manufacturer’s instructions. Colonies were then screened
by PCR and generally 8 positive colonies were sequenced in
an ABI3100 sequencer (Applied Biosystems, Foster City, CA,
USA) at the Smith College Center for Molecular Biology.
Many of the gene sequences generated in this study are pioneering for this territory of the tree. We carefully analyzed each
sequence generated using several BLAST strategies and treesearch algorithms. For each sequence, we specifically looked
for evidence of contamination from other eukaryotic organisms.
Specifically, the sequences originating from single cell protocols
are more prone to contamination, while sequences generated from cell cultures are unlikely to be contaminants. We
have discarded roughly 40% of sequences generated. However, many sequences have shown significant divergence from
other available amoebozoan sequences (principally from the
genomes of E. histolytica and D. discoideum) but did not
show significant similarity to anything else considered contaminating. Some sequences showed equidistance between
Amoebozoa and other eukaryotes. These sequences were
maintained in our analyses, and will need further sequencing of
closely related groups before contamination can be determined
for certain. Genbank accession numbers for sequences used
are: SSU rDNA JQ519502-511, actin JQ519417-458, ␣-tubulin
JQ51459-476, ␤-tubulin JQ51477-488, 14-3-3 JQ519397-416,
eEF2 JQ519489-498.
Analytical methods: With the resulting set of 111 new
sequences (Table 1), we reconstructed the phylogeny of each
gene independently, both to determine possible ancient paralogy as well as a point of comparison for concatenated
reconstructions. Taxon sampling for Amoebozoa is identical to
the dataset used in Lahr et al. (2011), with the addition of the
taxa sampled in the current study. Broader eukaryotic sampling
comprises a dataset of representative organisms named 10-16
proposed by Parfrey et al. (2010) and this dataset is available
for download at Treebase (www.treebase.org). For SSU rDNA
and each protein coding gene, alignments were constructed
in SeaView (Galtier et al. 1996; Gouy et al. 2010) with alignment algorithm MAFFT (Katoh et al. 2009) using the L-INS-I
setting. Alignments were then subjected to automated removal
of ambiguously aligned sites using the software GUIDANCE
(Penn et al. 2010). We performed maximum likelihood phylogenetic reconstruction for each gene using RAXML HPC 7.2.7
(Stamatakis 2006; Stamatakis et al. 2008) as implemented in
the online server CIPRES (Miller et al. 2009). We ran 100 rapid
bootstrap searches using the GTRCAT approximation followed
by a slow maximum likelihood search using the GTRGAMMA
model for the SSU rDNA partition and the LG model with
gamma distribution of site heterogeneity for the protein partition. The most appropriate model for amino-acid evolution was
determined using model testing implemented both in the software ProtTest 3.0 (Darriba et al. 2011) and the online server
Datamonkey (Delport et al. 2010), which gave similar results.
Bootstrap values of the GTRCAT search were then plotted on
the best tree found by maximum likelihood search for comparative analysis.
Each gene phylogeny was analyzed to determine which paralogs should be used for concatenation. In most cases, there
was no indication of ancient paralogy so we chose the shortest
branching paralog for concatenation (Table 2). In the few cases
where duplication predated species divergence, we took care
to choose orthologous genes (Table 2). We concatenated all six
genes into one file, consisting of 224 taxa and 3075 sites. The
dataset is missing 42% of genes (out of 224 taxa, the matrix
has 222 taxa with information for SSU rDNA, 176 for actin, 111
for ␣ tubulin, 99 for 14-3-3, 97 for ␤ tubulin and 82 for eEF2).
We performed the analyses on this concatenated dataset using
RAXML HPC 7.2.7 as described above, but with two partitions:
one for the SSU rDNA gene and one large partition with all 5
protein coding genes and LG model of substitution with gamma
distribution of site heterogeneity, as determined by the software ProtTest 3.0 (Abascal et al. 2005; Darriba et al. 2011).
We have also performed a slower, more accurate search in
RAXML HPC 7.2.7 consisting of 650 multiparametric bootstraps
using GTRGAMMA (as opposed to rapid bootstraps based on
the GTRCAT approximation), followed by 100 maximum likelihood searches, each starting from an independent maximum
Multigene Analysis of Tubulinea 337
parsimony starting tree. The resulting topology from this slower,
more accurate approach was identical to the topology obtained
from the faster approach, and the bootstrap values had minimal
variation. The faster approach is at least one order of magnitude less time consuming. We present results from the slower
approach in Figure 3.
With the results from the unconstrained reconstruction at
hand, we designed constraints to several proposed groups as
well as non-monophyletic groups to be tested by the approximately unbiased test (AU (Shimodaira 2002, 2004)). The AU
test provides a statistical measure whether the current dataset
is able to reject the monophyly of constrained groups. We generated maximum likelihood reconstructions with constraints for
each of 36 hypotheses that were not monophyletic in the most
likely tree. Parameters for tree searching in RAxML HPC 7.2.7
were identical to the standard reconstruction (here the advantages of a less computationally intensive approach become
critical, hence for this analyses we used the resulting concatenated tree from the fast approach). All trees were then
compared to the best tree found on the standard analysis
using RaxML to calculate per-site log likelihoods. The per-site
likelihoods were then analyzed in CONSEL (Shimodaira and
Hasegawa 2001) with standard parameters to obtain p-values.
Acknowledgements
The authors would like to thank O. Roger Anderson,
Benjamin Normark, Michael Hood, David Patterson
and Laura W. Parfrey for extensive discussion of the
results presented in this manuscript. We appreciate the careful revisions made by two anonymous
referees and specially Associate Editor Sandra Baldauf, which significantly improved earlier versions
of this manuscript. DJGL has been funded by CNPq
GDE Fellowship 200853/2007-4 and a CNPq PostDoctoral Fellowship 501089/2011-0. Funding for
the work comes from an NSF RUI Systematics
grant (DEB RUI: 0919152) and NSF Assembling
the Tree of Life grant (DEB 043115) to Laura A.
Katz.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.protis.2013.02.003.
References
Abascal F, Zardoya R, Posada D (2005) ProtTest: selection of best-fit models of protein evolution. Bioinformatics
21:2104–2105
Amaral Zettler LA, Nerad TA, O’Kelly CJ, Peglar MT, Peglar
MT, Gillevet PM, Silberman JD, Sogin ML (2000) A molecular
reassessment of the leptomyxid amoebae. Protist 151:275–282
Anderson OR (1992) The effects of silicate depletion and subsequent replenishment on the cytoplasmic fine structure of the
silica-secreting testate amoeba Netzelia tuberculata in laboratory culture. J Morphol 211:285–293
Baldauf SL, Roger AJ, Wenk-Siefert I, Doolittle WF (2000)
A kingdom-level phylogeny of eukaryotes based on combined
protein data. Science 290:972–977
Bolivar I, Fahrni JF, Smirnov A, Pawlowski J (2001) SSU
rRNA-based phylogenetic position of the genera Amoeba and
Chaos (Lobosea, Gymnamoebia): The origin of Gymnamoebae
revisited. Mol Biol Evol 18:2306–2314
Bosak T, Lahr DJG, Pruss SB, Macdonald FA, Dalton
L, Matys E (2011) Agglutinated tests in post-Sturtian cap
carbonates of Namibia and Mongolia. Earth Planet Sci Lett
308:29–40
Brown MW, Silberman JD, Spiegel FW (2011) Slime molds”
among the Tubulinea: Molecular systematics and taxonomy of
Copromyxa. Protist 162:277–287
Cavalier-Smith T, Chao EEY, Oates B (2004) Molecular phylogeny of Amoebozoa and the evolutionary significance of the
unikont Phalansterium. Eur J Protistol 40:21–48
Corsaro D, Michel R, Walochnik J, Muller KD, Greub G
(2010) Saccamoeba lacustris, sp nov (Amoebozoa: Lobosea:
Hartmannellidae), a new lobose amoeba, parasitized by
the novel chlamydia ‘Candidatus Metachlamydia lacustris’
(Chlamydiae: Parachlamydiaceae). Eur J Protistol 46:86–95
Darriba D, Taboada GL, Doallo R, Posada D (2011) ProtTest
3: fast selection of best-fit models of protein evolution. Bioinformatics 27:1164–1165
Delport W, Poon AFY, Frost SDW, Pond SLK (2010) Datamonkey 2010: a suite of phylogenetic analysis tools for
evolutionary biology. Bioinformatics 26:2455–2457
Dykova I, Kostka M, Peckova H (2008b) Morphology and
SSU rDNA-based phylogeny of a new strain of Saccamoeba
sp (Saccamoeba Frenzel, 1892, Amoebozoa). Acta Protozool
47:397–405
Dykova I, Fiala I, Peckova H, Dvorakova H (2008a) Phylogeny
of Flabellulidae (Amoebozoa: Leptomyxida) inferred from SSU
rDNA sequences of the type strain of Flabellula citata Schaeffer, 1926 and newly isolated strains of marine amoebae. Folia
Parasitol 55:256–264
Fahrni JF, Bolivar I, Berney U, Nassonova E, Smirnov A,
Pawlowski J (2003) Phylogeny of lobose amoebae based on
actin and small-subunit ribosomal RNA genes. Mol Biol Evol
20:1881–1886
Galtier N, Gouy M, Gautier C (1996) SEAVIEW and
PHYLO_WIN: Two graphic tools for sequence alignment and
molecular phylogeny. Comput Appl Biosci 12:543–548
Golemansky V (1970) Rhizopodes nouveaus du psammon
littoral de la Mer Noire (Note préliminaire). Protistologica
6:365–371
Gomaa F, Todorov M, Heger TJ, Mitchell EAD, Lara E (2012)
SSU rDNA phylogeny questions the monophyly of Arcellinida
(Amoebozoa) and reveals that Difflugia, the largest Arcellinida
genus, is paraphyletic. Protist 163:389–399
Gouy M, Guindon S, Gascuel O (2010) SeaView Version 4: A multiplatform graphical user interface for sequence
alignment and phylogenetic tree building. Mol Biol Evol 27:
221–224
338 D.J.G. Lahr et al.
Hampl V, Hug L, Leigh JW, Dacks JB, Lang BF,
Simpson AGB, Roger AJ (2009) Phylogenomic analyses
support the monophyly of Excavata and resolve relationships
among eukaryotic “supergroups”. Proc Natl Acad Sci USA
106:3859–3864
Katoh K, Asimenos G, Toh H (2009) Multiple alignment of
DNA sequences with MAFFT. Methods Mol Biol 537:39–64
Kosakyan A, Heger TJ, Leander BS, Todorov M, Mitchell
EAD, Lara E (2012) COI barcoding of nebelid testate
amoebae (Amoebozoa:Arcellinida): extensive pseudocryptic
diversity and redefinition of the Hyalospheniidae Schultze. Protist 163:415–434
Ogden CG (1979) Siliceous structures secreted by members of
the subclass Lobosia (Rhizopodea:Protozoa). Bull Br Mus Nat
Hist (Zool) 36:203–207
Page FC (1967) Taxonomic criteria for limax amoebae, with
descriptions of 3 new species of Hartmannella and 3 of
Vahlkampfia. J Eukaryot Microbiol 14:499–521
Page FC (1975) New family of amebas with fine pseudopodia.
Zool J Linn Soc 56:73–89
Page FC (1980) A light-microscopical and electronmicroscopical comparison of limax and flabellate marine
amebas belonging to 4 genera. Protistologica 16:57–78
Kudryavtsev A, Pawlowski J, Hausmann K (2009) Description and phylogenetic relationships of Spumochlamys perforata
n. sp and Spumochlamys bryora n. sp (Amoebozoa,
Arcellinida). J Eukaryot Microbiol 56:495–503
Page FC (1983) Marine Gymnamoebae. Institute of Terrestrial
Ecology, Cambridge, UK
Lahr DJG, Katz LA (2009) Reducing the impact of
PCR-mediated recombination in molecular evolution and environmental studies using a new-generation high-fidelity DNA
polymerase. Biotechniques 47:857–866
Parfrey LW, Grant J, Tekle YI, Lasek-Nesselquist E, Morrison HG, Sogin ML, Patterson DJ, Katz LA (2010) Broadly
sampled multigene analyses yield a well-resolved eukaryotic
tree of life. Syst Biol 59:518–533
Lahr DJG, Lopes SGBC (2006) Morphology, biometry, ecology and biogeography of five species of Difflugia Leclerc, 1815
(Arcellinida: Difflugiidae), from Tiete River, Brazil. Acta Protozool 45:77–90
Pawlowski J, Burki F (2009) Untangling the phylogeny of
amoeboid protists. J Eukaryot Microbiol 56:16–25
Page FC (1987) The Classification of Naked Amebas (PhylumRhizopoda). Arch Protistenkd 133:199–217
Penn O, Privman E, Ashkenazy H, Landan G, Graur D,
Pupko T (2010) GUIDANCE: a web server for assessing alignment confidence scores. Nucleic Acids Res 38:W23–W28
Lahr DJG, Lopes SGBC (2007) Ultrastructure and biometry
of three lobose testate amoebae of the family Lesquereusiidae
(Tubulinea: Arcellinida) based on specimens from Sao Paulo,
Brazil. Acta Protozool 46:339–348
Philippe H, Douady CJ (2003) Horizontal gene transfer and
phylogenetics. Curr Opin Microbiol 6:498–505
Lahr DJG, Souza MBGe (2011) Occurrence of the lobose testate amoeba Pseudonebela africana (Amoebozoa, Arcellinida)
in the Brazilian “cerrado”. Eur J Protistol 47:231–234
Porter SM, Knoll AH (2000) Testate amoebae in the Neoproterozoic Era: Evidence from vase-shaped microfossils in the
Chuar Group, Grand Canyon. Paleobiology 26:360–385
Lahr DJ, Nguyen TB, Barbero E, Katz LA (2010) Evolution of
the actin gene family in testate lobose amoebae (Arcellinida)
is characterized by 2 distinct clades of paralogs and recent
independent expansions. Mol Biol Evol 28:223–236
Porter SM, Meisterfeld R, Knoll AH (2003) Vase-shaped
microfossils from the Neoproterozoic Chuar Group, Grand
Canyon: A classification guided by modern testate amoebae.
J Paleontol 77:409–429
Lahr DJG, Grant J, Nguyen TB, Lin JH, Katz LA (2011) Comprehensive phylogenetic reconstruction of Amoebozoa based
on concatenated analyses of SSU rDNA and actin genes. PloS
ONE 6(7):e22780
Shadwick LL, Spiegel FW, Shadwick JDL, Brown MW, Silberman JD (2009) Eumycetozoa = Amoebozoa?: SSUrDNA
phylogeny of protosteloid slime molds and its significance for the amoebozoan supergroup. PloS ONE 4(8):
e6754
Lara E, Heger TJ, Ekelund F, Lamentowicz M, Mitchell
EAD (2008) Ribosomal RNA genes challenge the monophyly of the Hyalospheniidae (Amoebozoa: Arcellinida). Protist
159:165–176
Medlin L, Elwood HJ, Stickel S, Sogin ML (1988) The
characterization of enzymatically amplified eukaryotic 16S-like
rRNA-coding regions. Gene 71:491–499
Meisterfeld R (2002) Order Arcellinida Kent, 1880. In Lee JJ,
Leedale GF, Bradbury P (eds) The Illustrated Guide to the Protozoa. 2nd ed Allen Press, Lawrence, Kansas, USA, pp 827–860
Miller M, Holder M, Vos R, Midford P, Leiebowitz
T, Chan L, Hoover P, Warnow T. (2009). “The
CIPRES Portals.”. Retrieved 2009-08-04, 2009, from
http://www.phylo.org/sub sections/portal
Nikolaev SI, D. Mitchell EA, Petrov NB, Berney C, Fahrni
J, Pawlowski J. (2005) The testate lobose amoebae (Order
Arcellinida Kent, 1880) finally find their home within Amoebozoa. Protist 156:191–202.
Shimodaira H (2002) An approximately unbiased test of phylogenetic tree selection. Syst Biol 51:492–508
Shimodaira H (2004) Approximately unbiased tests of regions
using multistep-multiscale bootstrap resampling. Ann Statist
32:2616–2641
Shimodaira H, Hasegawa M (2001) CONSEL: for assessing
the confidence of phylogenetic tree selection. Bioinformatics
17:1246–1247
Smirnov AV, Nassonova ES, Cavalier-Smith T (2008) Correct identification of species makes the amoebozoan rRNA tree
congruent with morphology for the order Leptomyxida Page,
1987; with description of Acramoeba dendroodea n. g., n. sp.,
originally misidentified as ‘Gephyramoeba sp.’. Eur J Protistol
44:35–44
Smirnov AV, Chao E, Nassonova ES, Cavalier-Smith T
(2011) A revised classification of naked lobose amoebae
(Amoebozoa: Lobosa). Protist 162:545–570
Multigene Analysis of Tubulinea 339
Smirnov A, Nassonova E, Fahrni J, Pawlowski J (2009)
Rhizamoeba neglecta n. sp (Amoebozoa, Tubulinea) from the
bottom sediments of freshwater Lake Leshevoe (Valamo Island,
North-Western Russia), with notes on the phylogeny of the
order Leptomyxida. Eur J Protistol 45:251–259
Smirnov A, Nassonova E, Berney C, Fahrni J, Bolivar I,
Pawlowski J (2005) Molecular phylogeny and classification of
the lobose amoebae. Protist 156:129–142
Smith HG, Bobrov A, Lara E (2008) Diversity and biogeography of testate amoebae. Biodivers Conserv 17:329–343
Snoeyenbos-West OLO, Salcedo T, McManus GB, Katz
LA (2002) Insights into the diversity of choreotrich and
oligotrich ciliates (Class: Spirotrichea) based on genealogical analyses of multiple loci. Int J Syst Evol Microbiol 52:
1901–1913
Stamatakis A (2006) RaxML-VI-HPC: Maximum likelihoodbased phylogenetic analyses with thousands of taxa and mixed
models. Bioinformatics 22:2688–2690
Stamatakis A, Hoover P, Rougemont J (2008) A rapid
bootstrap algorithm for the RaxML web servers. Syst Biol
57:758–771
Tekle YI, Grant J, Katz LA (2007) Placement of diverse amoeboid lineages in the eukaryotic tree of life and the evolution of
‘Amoebozoa’. J Phycol 43:56
Tekle YI, Grant J, Anderson OR, Nerad TA, Cole JC, Patterson DJ, Katz LA (2008) Phylogenetic placement of diverse
amoebae inferred from multigene analyses and assessment of
clade stability within ‘Amoebozoa’ upon removal of varying rate
classes of SSU rDNA. Mol Phylogenet Evol 47:339–352
Todorov M, Golemansky V, Meisterfeld R (2010) Is Difflugia
nebeloides (Amoebozoa: Arcellinida) really a Difflugia? Redescription and new combination. Acta Zool Bulg 62:13–20
Yoon HS, Grant J, Tekle Y, Wu M, Chaon B, Cole J, Logsdon J, Patterson D, Bhattacharya D, Katz L (2008) Broadly
sampled multigene trees of eukaryotes. BMC Evol Biol 8:
14
Available online at www.sciencedirect.com