Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
RESEARCH
Open Access
Evolution of vacuolar proton pyrophosphatase
domains and volutin granules: clues into the
early evolutionary origin of the acidocalcisome
Manfredo J Seufferheld1*, Kyung Mo Kim2,3, James Whitfield4, Alejandro Valerio5 and Gustavo Caetano-Anollés2
Abstract
Background: Volutin granules appear to be universally distributed and are morphologically and chemically
identical to acidocalcisomes, which are electron-dense granular organelles rich in calcium and phosphate, whose
functions include storage of phosphorus and various metal ions, metabolism of polyphosphate, maintenance of
intracellular pH, osmoregulation and calcium homeostasis. Prokaryotes are thought to differ from eukaryotes in that
they lack membrane-bounded organelles. However, it has been demonstrated that as in acidocalcisomes, the
calcium and polyphosphate-rich intracellular “volutin granules (polyphosphate bodies)” in two bacterial species,
Agrobacterium tumefaciens, and Rhodospirillum rubrum, are membrane bound and that the vacuolar protontranslocating pyrophosphatases (V-H+PPases) are present in their surrounding membranes. Volutin granules and
acidocalcisomes have been found in organisms as diverse as bacteria and humans.
Results: Here, we show volutin granules also occur in Archaea and are, therefore, present in the three
superkingdoms of life (Archaea, Bacteria and Eukarya). Molecular analyses of V-H+PPase pumps, which acidify the
acidocalcisome lumen and are diagnostic proteins of the organelle, also reveal the presence of this enzyme in all
three superkingdoms suggesting it is ancient and universal. Since V-H+PPase sequences contained limited
phylogenetic signal to fully resolve the ancestral nodes of the tree, we investigated the divergence of protein
domains in the V-H+PPase molecules. Using Protein family (Pfam) database, we found a domain in the protein,
PF03030. The domain is shared by 31 species in Eukarya, 231 in Bacteria, and 17 in Archaea. The universal
distribution of the V-H+PPase PF03030 domain, which is associated with the V-H+PPase function, suggests the
domain and the enzyme were already present in the Last Universal Common Ancestor (LUCA).
Conclusion: The importance of the V-H+PPase function and the evolutionary dynamics of these domains support
the early origin of the acidocalcisome organelle. In particular, the universality of volutin granules and presence of a
functional V-H+PPase domain in the three superkingdoms of life reveals that the acidocalcisomes may have
appeared earlier than the divergence of the superkingdoms. This result is remarkable and highlights the possibility
that a high degree of cellular compartmentalization could already have been present in the LUCA.
Reviewers: This article was reviewed by Anthony Poole, Lakshminarayan Iyer and Daniel Kahn
Background
According to the theory of serial endosymbiosis [1], the
symbiotic history of mitochondria and plastids started
more than 1.5 billion years ago when a primitive eukaryotic cell engulfed a bacterium [2]. These endosymbionts gave rise to contemporary organelles through
* Correspondence: [email protected]
1
Department of Crop Sciences, University of Illinois at Urbana-Champaign,
Urbana IL USA
Full list of author information is available at the end of the article
complex, and poorly understood molecular evolutionary
events. It is now widely accepted that an a-proteobacterium was the ancestor of mitochondria [3]. Similarly, a
single endosymbiotic association between a cyanobacterium and a mitochondriate eukaryote has been proposed
as the origin of the chloroplast [2]. Although the endosymbiotic origin of plastids is well established [4,5], the
events that drove the evolution of other sophisticated
membrane-bound cellular compartments remain
© 2011 Seufferheld et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
unclear. In fact, other mechanisms may be responsible
for the generation of organelles.
Prokaryotic cells have been simplistically portrayed in
the general scientific literature as a plasma membrane
“sack” containing DNA, filled with cytoplasm and surrounded by a cell wall. However, provocative findings
have challenged the notion that the prokaryotic cell
lacks a sophisticated cytoplasmic organization. For years,
it was believed that the shape of bacterial cell was maintained by the organization of cellulose fibers within the
cell wall. However, a protein similar to actin, which was
once thought to be exclusive to the eukaryotic cytoskeleton, was discovered in bacteria [6-8]. This protein
called MreB self-assembles into filaments forming a
cytoskeleton-like structure that maintains the shape of
bacterial cells.
Even more recently, the idea that organelles similar to
those present in eukaryotes are absent in bacteria was
challenged by the discovery of an organelle similar to
the acidocalcisome of unicellular eukaryotes within the
bacterium Agrobacterium tumefaciens [9]. The existence
of this bacterial organelle has also been confirmed in
the photosynthetic bacterium Rhodospirillum rubrum
[10]. The acidocalcisome is an acidic calcium-storage
organelle that was first described in trypanosomes, but
has now been found in the cells of diverse organisms
ranging from bacteria to higher eukaryotes [11]. This
membrane-enclosed organelle is characterized by its
acidic nature, high electron density, and high content of
polyphosphates (polyP) including pyrophosphate (PPi),
calcium, magnesium, and other elements. In addition,
the organelle contains a variety of cation pumps including Ca2+/H+ and H+ pumps. In particular, the vacuolar
proton translocating pyrophosphatase (V-H+PPase) proteins have been localized in the acidocalcisomes of bacteria, parasitic protozoans, algae, plants, and recently in
cockroaches [12]. The widespread distribution of these
proteins, which also contain ancient, highly conserved
protein motifs, suggests that V-H+PPase arose early in
the evolution of life on Earth. As mentioned before, the
main components of the acidocalcisome are polyphosphates, which includes PPi. V-H+PPases use PPi as an
energy source to pump protons into the acidocalcisome
lumen, thereby generating an electrochemical gradient.
This strongly suggests that both the V-H+PPase and the
acidocalcisome share similar evolutionary histories.
Acidocalcisomes are morphologically and chemically
similar to the structures historically described as volutin
or polyphosphate bodies, which have been identified in
a variety of microorganisms, including bacteria, archaea,
algae, and protozoans [13]. Meyer [14] described volutin
granules in bacteria more than 100 years ago and much
later they were identified as acidocalcisomes [15]. In
protists, this organelle has essential roles in the
Page 2 of 15
regulation of intracellular Ca2+, pH, and osmotic homeostasis, in which polyP is a key player. However, the
study of polyP has been neglected for years. An important feature of polyP is its negative charge, which
enables it to interact with nucleic acids and act as a regulator of RNA polymerases [16] and proteinase activity
[17]. In addition, these negatively charged polymers can
serve as a binding template for proteins, amino acids,
polysaccharides and many cations [18-20].
Other membranous structures have also been reported
in bacteria as well [21]. However, these organelles are not
found in the eukaryotes (e.g., magnetosomes of magnetotactic bacteria [22]). Thus, the acidocalcisome is the first
and the only organelle demonstrated to be present in
both bacteria and eukaryotes, and its morphological,
structural and ultrastructural, chemical and biochemical,
and molecular links to volutin granules may suggest it is
indeed universal. The conservation of this organelle in
prokaryotes and eukaryotes suggests that it has important
functions that still await discovery [11].
Debates about organellar evolution have been revitalized and streamlined during the post-genomic revolution. The availability of multiple complete genome
sequences from the whole spectrum of life, along with
corresponding protein sequence data sets and more
sophisticated bioinformatics tools opens a window into
the intricate mechanisms of cellular evolution. In the
present study, we use sequences and domain information from a diagnostic protein of the acidocalcisome, the
proton pump V-H+PPases, to infer the phylogenetic history of this protein and an ancient origin of the acidocalcisome. Within protein molecules, certain parts can
interact more strongly with each other than with other
parts, and are usually highly conserved, making up modules, called protein domains [23]. These segments fold
compactly, appear repeatedly in different proteins, and
often combine with other domains. Because domains
can harbor functional centers or can modulate biological
activities, the combination of domains in proteins adds
inherent diversity to the biological make up of an organism [24]. The combination of domains in proteins can
help clarify the origin and the evolutionary histories of
proteins and proteomes [25]. Using Bayesian reconstruction methods, we build phylogenetic trees base on protein sequences that shed light on the evolution of V-H
+
PPase proteins. We then examine the distribution of
the main functional V-H + PPase domain on the tree
using Protein family (Pfam) database definitions.
Here, we delineate the connections between the evolution of V-H+PPases and volutin granules as opportunity
to infer the likely origin of the acidocalcisome. In addition, we discuss the implications that these findings may
have on our understanding of the cellular complexity of
the ancestor of cellular life.
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
Results and Discussion
Taxonomic distribution of V-H+PPase pumps
V-H+ PPase is commonly present in the membrane of
acidocalcisomes which couple PPi hydrolysis to active
proton transport across the organellar membrane. The
enzyme is found in organisms from all superkingdoms
of life and should be considered ancient. Its ubiquitous
nature already suggests it appeared before the last universal common ancestor (LUCA) of diversified life [26].
The V-H+PPase is a highly conserved enzyme that acidifies the lumen of the acidocalcisome [11]. V-H+PPase
has been detected by immunogold labeling localization
in acidocalcisomes of bacteria, and several eukaryotic
microorganisms [9,10,27-33]. Recently, an organelle that
shares similar characteristics of the acidocalcisome,
including the presence of V-H+PPases in its surrounding
membrane, was found inside the protein storage vacuole
of seeds [34]. Additionally, an acidocalcisome-like organelle with membrane-bound V-H+PPases has been identified in the egg yolk of cockroaches [12]. These studies
show that a bona fide organelle linked to V-H + PPase
function exists in bacteria, microbial eukaryotes, and
higher eukaryotes.
Volutin granules: A universal phenomenon
Volutin-polyP bodies occur in organisms spanning an
enormous range of phylogenetic complexity from Bacteria and Archaea to unicellular eukaryotes to algae to
plants to insects to humans. Volutin/polyphosphate
electron dense granules exhibit varied internal patterns
suggestive of sponge-like electron-dense spheres, and
when fixed they look as empty or partially empty
vacuoles. The volutin granules shown in a number of
microorganisms appear to be identical to acidocalcisomes of Agrobacterium and Rhodospirillum and eukaryotes. While it may come as a surprise that volutin
electron-dense granule from different organisms bear
close similarity to acidocalcisomes, now we know that
acidocalcisomes are virtually identical in size, composition and morphology to volutin-polyP bodies found in a
vast array of organisms, including Archaea. Although
the volutin granules present in Methanosarcina have
not been yet confirmed to be acidocalcisomes, they are
morphologically and chemically similar to the acidocalcisomes of Agrobacterium and Rhodospirillum [9,10].
The volutin granules of Methanosarcina have the same
chemical profile, morphological characteristics (sponge
like-structure) and high levels of phosphorous compounds and calcium [35-37] than acidocalcisomes. In
addition, some of the Agrobacterium acidocalcisomes
(Figure 1, panels A and B) appear like partially or empty
vacuoles due to the fixation/staining protocol that may
promote the diffusion of the electron dense material out
Page 3 of 15
of the acidocalcisome. Remarkably, the same phenomenon is observed in the volutin granules of Methanosarcina (Figure 1, panels C and D).
Therefore, volutin-polyphosphate bodies electron
dense granules are distributed throughout phylogeny
and are universal. Consequently, our results support the
hypotheses that the volutin granules of Archaea are
likely acidocalcisome structures and that acidocalcisomes are universally distributed.
Phylogenetic analysis of V-H+PPase
The tree of V-H+PPase amino acid sequences is star-like
and without much deep internal topological structure,
suggesting the existence of limited phylogenetic signal
in the sequence needed to dissect deeper phylogenetic
relationships (Figure 2). It also suggests a relatively
recent history of sequence diversification in the molecule. The eukaryal, bacterial and archaeal V-H + PPase
sequences largely formed monophyletic groups and
eukaryotic sequences mostly grouped according to
established organismal classification, with unicellular
eukaryotes being placed at the base of the clade and
unicellular algae and plants diversifying later. However,
there were several instances of sequences that group in
a way contradictory to accepted classification. These atypical patterns could be explained by lateral gene transfer
(LGT) events among Bacteria, Archaea and Eukarya,
some of which may have occurred early in evolution.
Alternatively, these unusual placements might simply be
the result of phylogenetic error associated with using
only a single gene for analysis [38].
The fact that V-H+PPase sequences group many familiar clades monophyletically, and are also ubiquitous
among superkingdoms although with poor phylogenetic
signal, suggests these proteins are truly ancient. To
further explore their ancestral nature we focused on
structure, which is highly conserved in evolution [39].
We therefore, identified protein domains in the V-H
+
PPase sequences, mapped them onto the V-H +PPase
tree, and determined how many species in the three
superkingdoms harbored these domains.
Evolution of Pfam domains in V-H+PPase proteins
V-H + PPase proteins have not been crystallized, and
there is no structural entry associated with them in the
Protein Data Bank (PDB). This absence probably stems
from the limitation of NMR and X-ray crystallographic
techniques in their ability to acquire high-resolution
structures from proteins associated with membranes
[40]. In fact, only 1% of structural PDB entries are membrane proteins. We nevertheless considered the possibility that some other known structure could match the
structure of V-H+PPase. However, domain search using
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
Page 4 of 15
Figure 1 Electron micrographs thin sections of Agrobacterium tumefaciens(a &b) and Methanosarcina acetivorans (c & d). In panel (a),
the arrow shows the partially filled acidocalcisome of A. tumefaciens containing electron dense material. In panel (b), the arrow shows an empty
A. tumefaciens acidocalcisome. In panel (c), the arrow shows the electron dense volutin granule of M. acetivorans. In panel (d), the arrows show
empty, partially, and completely filled volutin granules of M. acetivorans.
Hidden Markov Models (HMMs) of fold superfamily
and fold family structural recognition revealed no hits.
It has already been established that V-H+PPase proteins
have domains with sequence motifs that are highly conserved in evolution [41], which suggests that catalytic
domains harboring the motifs in V-H + PPase are very
ancient. Because Pfam attempts to classify protein
domains and families using sequence alignments and
profile HMM analysis of sequence motifs [42], we
explored the distribution of Pfam domains in the
sequences of V-H + PPase proteins. A Pfam search
revealed that all V-H + PPase sequences have a single
domain that is annotated as inorganic H+ pyrophosphatase, PF03030. This domain in our analysis is distributed
in 279 species including Eukarya, Bacteria, and Archaea
(Figure 2). The domain PF03030 has a conserved motif
(57 amino acids) that contains a functional active site
for the V-H + PPase (Additional File 1). It has already
been established that a consensus sequence of the motif
is highly conserved across more than three hundred VH + PPase protein sequences obtained from completed
genomes and Sargasso Sea metagenomic sequences [41].
Our data indicate that the domain PF03030 is conserved
in all three superkingdoms, supporting our hypothesis
that the protein appeared before the LUCA. The taxonomic incongruences present in Figure 2 are clearly the
result of secondary evolutionary events of LGT that are
derived or simply arise from poor phylogenetic signal in
V-H+PPase sequence.
Origin of the acidocalcisome
Several hypotheses attempt to answer the question of
when and how the acidocalcisome arose. The canonical
explanation that the acidocalcisome was acquired either
through lateral transfer or via an endosymbiotic event
is, however, unlikely. Furthermore, sequence and
domain analysis of the V-H+PPase and the universal distribution of volutin granules support the appearance of
the volutin granules and possibly the acidocalcisome
before the initial branching of the tree of life (Figure 2).
Neither LGT nor direct filiation from the a-proteobacteria appears to be a likely explanation for the organelle’s acquisition in eukaryotes. Instead, the most
parsimonious explanation would be to posit that the
volutin granule/acidocalcisome appeared very early,
before the branching of the three superkingdoms of life,
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
Page 5 of 15
Bacteria (231 species)
Eukarya (31 species)
1.00
1.00
1.00
1.00
0.83
Ruthia magnifica A1AVD4
Gemmatimonas aurantiaca C1A776
Geobacter lovleyi B3E4Y7
Geobacter sulfurreduc
ens Q747H5
Geobacter meta
llireducens Q39
Pelobacter
QM2
propionicus
A1ATG4
Geobacte
r uraniiredu
Geobac
cens A5GC
ter bemi
38
djiensis
Desulfo
YP_002
coccus
137390
oleovor
Desulfa
ans A8
ZXZ6
Chtho tibacillum alk
niobac
enivora
ter fla
Methy
ns B8
vus B4
FLB6
la
Therm cidiphilu
CVZ9
m infe
Meth omicrobi
rnorum
um ro
yloco
B3DZS
Syn
ccus
se
9
tr
capsu um B9K
Des ophoba
YZ6
latus
u
ct
Q60
Meth lforudis er fum
9P4
a
Pha anoco audaxv roxidan
s A0
cco
s
ides iator B
L
Vig eolus
M
L7
1I0V
n
aure
burt
1
Me a rad
onii
us
ia
d
YP_
Nic icago ta O2 P2161
565
6
212
trun
684
Ha otian
4
c
a
lo
Ka stac rusti atula
li
B
c
h
d
a
6
y
ium
Ch
DX
sc
Q1
Ch enop folia aspic 97Z6 D7
He enop odium tum A a A4L
Cu vea odiu
rub 1E9 AP4
Py curb bras m gla rum B0
Ho rus ita m ilien ucu Q8L
Tr rde com os sis Q m Q 5B2
Ho iticu um mu chata 6R 1W
2P
4
n
Th rde m a bre is Q BA U3
4
M ellu um est visu 8G A33
a
14
Zy lu ng vu ivum bula T22
9
Pic go s do iella lgar
A tum
ea phy me sa e Q 9LR
Q8
4Q
sit llum stic lsu 9FS Z1
I7
ch
a g
1
en xan A9 inea 2
sis th X9 Q
C0 oxy A3 6T5
53
PR lum
N4 A
7X
Y7
8
Vesicomyosocius okutanii A5CXT3
Q3SIS2
Thiobacillus denitrificans
ha YP_747021
Nitrosomonas eutrop
B25
multiformis Q2Y
Nitrosospira
T52
tophilia B2F
omonas mal
PH20
Q8
Stenotroph
nopodis
658
onas axo
Xanthom
NP_638
48
mpestris
onas ca
ca Q47B L8
Xanthom
aromati
X6
omonas
nes C3
7
Dechlor
formige
A4G7X
bacter
xydans
Oxalo
HN8
co
2S
ni
A
as arse leiphilum
1XZ65
iniimon
nii B
8
petro
Herm
cholod s Q21YE 5
libium
n
Methy
othrix
CYM
Lept
duce
rrire
s C5
Y6
rax fe aradoxu A1VM 9
4
dofe
ns
xp
MA
Rho
ovora lenivora us B9 012
Vari
a
5
re
phth rax eb 00423 79
s na
ovo e YP_ A1WM T0
ona
Acid
M
rom
e
na
Pola
ave isenia e C5B 77
8
rax
ra
re
ovo
594 I8
cte
rne
Acid hroba ter tu s YP_ A4TZ 3
p
ri
ac
e
4S
ine
inib cellula ldens 2W 65
Q
2
Verm Tered
tra
wa
m
1
63
u
is
ia in
ph netic 0012 G9G 03
son
2 74
gry
_
g
Law illum
ma ii YP ns Q _45 67
spir irillum ittich ivora s YP 6170 S1
eto
_
sp
tic ali
IM
sw
gn
YP B6 905 3
eto
na oma litor
Ma
6
gn
is
mo
ar ter
um 42 Z9 2
ns
Ma
go
c
m
hin obiu oba laske nten YP_ 1U LU 0
Q 4F
r
Sp
9
g
a
ce m
hin Eryth yxis
m bru ique e Q 7HX 24
sp
b
u
p
illu
vo
go spir m r er u biqu ns A 4RB 176
No
t
8
o
lu
u
hin
a
B 0
od piril ibac ter
or m 42 675 6
Sp
s
c
1
u
_ 5
Rh
tiv
g
o
a
en ine NP _7 C3
ela gib
od
Rh us P ela vam zuc tus YP Q0
t
P
la
m en ris m
ida
iu
iu
c
a
lum er es m tun
nd
cu act cr lis
Ca
p
iba lob ter au ne
c
rv
y
ic
s
Pa hen loba ar ona
M
P au
m
C
ho
yp
H
Archaea (17 species)
1.00
0.95
0.99
1.00
1.00
1.00
0.52
1.00
Th
er
m
1.00
1.00
1.00
51
44
C4
CA
I8 dtii
8
r
W
a
P
0.54
0.53
1.00
9Z inh 4I6
3
5
Q
1
58
a s re m A 6E HJA
11
0.64
Q
4
u
llin a
10
0.65
ra on fant Q4 is A F0
00
r
s
co m
D
P_
1.00
a ydo ia in ajo ilien 9N
X
r
m z
n
1.00
6
ila
1.00
77
iQ
ha m
C hla ma nia bra ruz oph 380 4U0 9
0.65
C ish ma nia a c rm AK
Q0 221
0.99
Le ish ma om the ii A enii
00
50
Le ish nos ena ond ters YP_
59
71
1.00
Le ypa ym a g gpe ans X5
1.00
03
Tr trah lasm bor rrog 0S8 P_0 0
1.00
Te xop pira inte xa B us Y TSB
1.00
s
o
c
2
a
T pto pir bifle anti is A
1.00
0.87
Le ptos pira r atl ens
0.96
Le ptos acte ngha DJK7
8
1.00
1.00
Le oceib ia do a A9 LY3
Y9
W6
378
Cr kdon lgicid tii A0 A4BZ A5FJ
1.00
060
e
_01
Do rdia a fors ensii oniae
1.00
ZP
1.00
a
sis
Ko mell ter irg johns
den YJW4
c
ra
n
m
a
la
G
u
b
0.7
1.00
1.00
C0
ab
teri
lari
Po obac oekiell gleum ES25
v
1.00
3
64
Fla wenh terium cus B
1.00
4SE
u
c
ti
is A
1.00
Lee seoba s asia
oform 7
ilu
1.00
vibri
3L
Chry eboph
s
ri
Q3B
o
7
hlo
0.50
lum
Am
coc
QP0
o
e
3
te
B
th
lu
s
8L2
rvum
Pro ictyon
B4S
d
1.00
m pa
uarii
V1
Pelo baculu
1.00
0.96
aest
ro
B3QV
1.00
Chlo
chloris alassium
03
heco
th
A831
1.00
Prost erpeton
nea BA
rra
oh
ite
or
C8
ed
1.00
Chl
1.00
aria m abulum Q9ST
ul
ab
Acet
acet
Z2
0.83
ularia
1.00
ZE
ab
A1
et
Ac
rina
196855
1.00
0.93
1.00
illa ma
YP_003
0.80
Microsc
ormata
alea bif
Robiginit
a A6GBL9
tis pacific
41
Plesiocys
ZP_049696
nucleatum
ium
cter
Fusoba
lis Q5LIL5
Bacteroides fragi
1.00
solvens CBK67775
xylani
0.50
roides
Bacte
1.00
1.0
Bacteroides ovatus A7LVP2
1.00
Bacteroides caccae A5ZKN0
1.00
Bacteroides thetaiotaomicron Q8A294
Bacteroides stercoris B0NSQ0
1.00
1.00
1.00
Bacteroides eggerthii B7ALD5
1.00
1.00
1.00 Bacteroides intestinalis
B3C7L3
1.00
Bact
eroides unifo
1.00
rmis A7UYB2
1.00 Bacteroides
0.951.00
plebeius B5C
1.00
YN8
Bacteroid
es coproc
1.00 Bacte
ola
B3JI04
roides do
0.76
rei C3R6
Bactero
Q7
ides vu
1.00 Pa
lga
tus A6
rabacte
L2M4
roides
Paraba
merda
cteroi
0.85 A
e A7AD
des di
kkerm
83
stason
Verru ansia m
is A6LI
ucinip
com
1.00 1.00
E5
icrobi
hila B
Blast
2ULG
ae ba
o
Plan pirellula
2
cteriu
ct
m B5J
Can omyce marina
Q
A
T8
d
s ma
3ZR
Can idatus
C4
ris A
K
d
Bra idatus uenen 6BZZ1
1.00 1.00
ia s
c
Clo
tuttg
aca
Clo hyspir
artie
a
m
s
1.00
nsis
Clo tridium hyody onas
a
Q
s
s
c
ente
1PZ
idam
lep
Clo tridiu
1.00
tu
R6
riae
inov
Ba stridiu m me m A7
C0R
ora
VN
c
ns B
142
H8
Sh teroid m ph thylpe
0VH
u
y
n
S9
1.00 1.00 0.77 MOstre ttlewor es capiltoferme tosum C
0.50
0.56 1.00
0.50
th
n
0.50
o
lo
0
ta
c
e
EI3
ia
0.65
0.50
sa sus A ns A
4
Me than occu
tell
1.00
1.00
0.99
Me than ocull s luc
es 6NPF 9KSE
1.00
9
im
C
e
th
7
o
1.00
us
Me an sa
arin 4GB
A7
Ru tha oco rcina maris us
nig A4R
ac
D min nos rpu
0.98
e
Co orea oco pirillu sculu tivo ri YP RD6
1.00
1.00
Ru pro for ccu m h m la rans _001
m
s
c
N
bre
04
C m
oc ic
la un
P
C lost inoc cus ige cta gate anu _61 5961
Do lost ridiu occ co nera ris B i YP m Y 8750
0.66
5
P
_
re ridiu m us me ns
0.50
s C B0 CSC 503 _001
a
lo m hyle gna
83
02
0.50
ng sc
vu 0B GAF 0
5
m
0.50 0.76
94
ica ind on s A 9C
50
1.00
7 9
te en ae
7
1.00
na s
C B1
1.00
A6 B0N 0C Z7
4K
B
D
0.92
FT R
6
2
7
1.00
0.78
0.71
1.00
1.00
1.00
0.99
1.00
Ac
id
ob
ac
M
et
ha Me teriu
no
oa
m
th
na
ca
Ca
Clo sph an
e
p
ldic
Ca
ro
a o
ba strid eru regu sula
rb
ell
ulo
ium la
la
An oxy cte
tu
p
r
s
a
d
ir
Ca
e
ibr ten be alu boo m C
ldic upto roc
a
ij
g
c
er str ne 1F
ell
c
r
h
ell
ulo kron um ium ong inck is B i A7 569
e
8
I
t
s
o
Dic irup tsk herm pac nsis ii A6 GJ 486
tor yen
tyo
op ificu Q8 M3 H9
s
sa
g
Co
cc is Y hilum m B RC H6
D lom
pro
X1
the ictyo us th haro P_0
B 7R
glo
lyti 04 9M 706
rm
e
F
r
m
o
0
erv
m
c
Q
Clo
ido bacte us tu oph us A 245 79
str
idiu
ba
rp
rg ilum 4X 60
c
m
the teriu roteo idum B5 HY4
Ce
narc rmoc m no lytic B8E YF3
us
do
ellu
0W 4
hae
Nit
s
Ca
B
m
u
ros
ldiv
YP um A 5Y8 5
opu m sy
irga
0
_0
mb
mil
01 7HM 2
ma
io
us
quil
ma sum 0378 Q6
in
ritim
49
A
The gensis
us 0RW
rmo
YP
Pyro
A9
83
_
p
A
bac
ro
001
1
ulum teus
Pyro
540 E8
te
bac
0
isla
Pyro
ndic nax A9 35
bacu ulum a
um
JQ
rs
lum
aero enaticu A1RV I4
Kosm philum m A4W 93
NP_
o
IY1
5
Eubac toga ole
aria 59532
te
Bacte
C5C
roides rium elig
IC6
ens
pectin
Petro
ophilu C4Z6Y1
toga m
s B7A
obilis
SN2
YP_0
Therm
01
otoga
lettinga 568076
Thermo
e A8F6
sipho afr
U1
Thermos
icanus
ipho me
B7IEX6
lanesien
sis A6LP
Thermotog
21
a neapol
itana Q5
Thermotoga
CBQ3
naphthophil
a YP_003346
312
Thermotoga
maritima Q9S
5X0
Marinitoga piezophila
B7RDQ4
Thermotoga petrophila A5IKP7
Desulfitobacterium hafniense YP_521069
Heliobacterium modesticaldum B0TGJ8
0.97
1.00
1.00
Single copy
Two copies
1.00
1.00
6
6
PB
Z8
2K H8 C0C 2
C
V
5M
5
8
us 0E hica 5Z
35
51
sin ii C op m A 926 512 321
m all otr
1
u
01
M
riu m h en rios B0 _00 _0
t
P
g
te
P
e
ac eriu dro ven cca ii Y ns Y
d
e
a
rib ct y
H2
73
O uba tia h rium s c
lan dig G7
BH
E lau cte tipe rem llire C4 s C2 X6
5
B uba ros us o eta tiva ticu
7C
CH
E ae hil s m fec oly
C2 B6W
8
An alip hilu de lact dius alis
TK
ia
p
Alk ali oph ccus tetra ogen A3M
Alk iotr oco cus ydr ntis
Ab aer coc us h lidifo YP6
An aero cocc ca A0P Q9
An aero culum ovyi Q898 IDY6
C4
An oba m n tani
m
r
iu
0
Py strid m te tyricu 0S16
iu
6
510
820
Clo strid m bu na B SMK
676
B1C
g
46
003
Clo stridiu ma ra A8 minis
YP_ s B0KB
ia
Clo egold as mic oriho thranii
licu
rc
ano
Fin imon tis ste ter ma
9
deth
rv
c
Y
u
a
6
e
b
Pa erofus
A
r ps
ero
s B2
acte
ana
Ana
hilu
rmo naerob ermop 0GFB8
a
th
The
C
ius
rmo
ilus
The naerob alkaliph
YF6
L99
a
B8C
Q67
Natr bacter x orenii
hilum
io
ri
ermop
Deth ermoth
m th
th
Halo obacteriu
4BSF8
833
ilis A
bi
s C0N
Sym
s mob
occu
oxidan
Nitroc phaga thio
Q3J9Y1
lo
oceani
Methy
coccus
A4A7F7
litoralis
Nitroso
gibacter
Q3A315
Congre
nolicus
Q1JY39
ter carbi
toxidans
Pelobac
monas ace
Desulfuro
YP_629923
s xanthus
Myxococcu
MC1
eriovorus Q6M
Bdellovibrio bact
01213266
opropionicum YP_0
Pelotomaculum therm
s YP_001114313
Desulfotomaculum reducen
Carboxydothermus hydrogenoformans YP_359158
O
M
0.99
M
c
et
1.00
hy eth ean
1.00
lo
1.00
ba ylob icau
M
et
ct
ac lis
M
0.91
1.00
hy
lob ethy Met eriu ter ale
0.66
iu
1.00
lob hylo m
ac
ra m n xan
te
a
Xa
b
0.99
0.51
riu
cte
ac dio od drii
0.95
nth
m
A
t
r
t
u
e
o
iu
A
Rh
1.00
z
o
1.00
c
od bac orhiz hlo m e rium lera lans 3U
0.65
t
r
op
C
n
x
se er a obiu ome tor pop s B B8 C6
1.00
u
q
I
1.00
ud
Oli
om totro m c than uen uli B 1LZ PI6
a
go
s
ic
o
p
1 A
u
1.00
Br trop nas hicu lino um C5A ZIM 8
ad
1.00
d
ha
pa
s
B
y
0.99
lus YP ans 7K XH 5
0.57
Nit rhizo carb
U 1
_
tr
1.00
rob
biu oxid is Y 001 A8IB J9
0.99
Nit
4
ac
rob
ter m ja ovo P_5 190 H1
ac
0.98
p
win
on ran 325 75
ter
1.00
icu s B
3
o
h
g
5
a
1.00
La
1.00
bre mbu radsk m Q 6JG
1.00
8
rg
Aura
1.00
I6
nz
y
L
9
1.00
Sta abren ia ale ensis i Q3S K83
ntim
ppia
zia
ona
R
x
Q
agg andri 1QL T3
sm
agg
ang
i
C
re
0.97
ano regata gata B9QV 2
1.00
xyd
A0
ans ZP_0
N 43 1.00 1.00
B
154 MF0
ZP
Meth eijerin
1.00 1.00
5
_
c
4
0
kia
yloc
indic 12273 02
ella
25
a
s
Agro
Rhiz ilvestris B2ICH
bact
1.00 1.00
eriu
m tu obium lo B8EN 6
Och
B6
mefa
ti Q
roba
9
cien
Ochro
ctrum
s Q8 83A3
bact
an
1.00
U
rum
1.00
interm thropi Q G67
8VR
ed
1.00
Bruce ium C4W Z2
0.51
FQ4
lla su
is B0C
Brucel
L9
la
1.00
6
Brucell
0.80
canis
1.00 1.00 0.63
1.00 1.00
A9MAF
a meli
1.00
tensis
1.00
1.00
Q8YG 1
Brucell
H4
a ovis
1.00
A5VPV3
Brucella
1.00 0.65
ceti C0G5
0.66
1.00
Brucella
abortus B2 N5
1.00
S5
19
Agrobacter
1.00 1.00
ium vitis B9J
0.96
V82
Sinorhizobium
medicae A6U
7Q0
Rhizobium melilo
ti Q8VRZ3
Agrobacterium radiobacter
B9JCY4
1.00
Rhizobium leguminosarum Q1MIT1
1.00 1.00
1.00
Rhizobium etli Q2KA06
1.00
Frankia alni YP_716705
195
Thermobifida fusca YP_288
s B1VM24
Streptomyces griseu
PP4
clavuligerus B5G
Streptomyces
1.00
iralis B5HA91
esp
tina
pris
1.00 1.00
s
_825793
Streptomyce
rmitilis NP 18746
1.00
yces ave
_069
Streptom
1.00
ceus ZP
yces svi
Q6BCL0
tom
lor
ep
co
Str
coeli
D5
myces
s A0LW 03
Strepto
lolyticu
XC
1.00
us cellu tropica A4
3
therm
8M5X
pora
Acido
ola A
Salinis
49
arenic 011026
0.51
ra
po
9
_0
is
1.00
a YP
6023 2
Salin
3
ae
1
0
hr
yt
_0
Y0
ra er
s YP
C5B
1.00
olyspo iotoleran vernae KCY8
arop
2
d
ca
2
ch
B
ra
ia
ac
W
1.00
m
S
us
P
rg
inutu ii A7N 89
ococc eutenbe
1.00
mm
Kine
37
B
olz
robiu astenh 00246 W3
imic
_
8VN K8
us c ns YP
Elus
x
0.50
Q
1.00
le
eif
ga
cus C2KP 5
gre
Ros
G
ntia
0.50
s ag
ura ulieris Q6A7 J7
0.86
sa
3
exu
rofl
s m cnes Q02 TP5
exu
0.99
rofl iluncu
a
s
Chlo
Z
lo
tu
1
b
Ch
rium usita ae B QJ7
1.00
Mo
0.67
r
7
rr
6Y
acte
nib bacte s te ae B 8I4X 0
tu ph
pio
0.75
li
B
Y3
itu
So
Pro
Op nym ticum Q0A 93
1.00
ho loly lfei 909 E0
ic
tr
o
1.00
5
Z
llu
do
1.00
ce as w YP_ Q08 DT8
eu
2
s
ps
K
a
on
ium
1.00
0.50
es strid hom satili tiac Q8 GM 89
S
r
n
oid
0.50
ter
Clo ntrop r ve ura idum B4 0YT D3
1.00
0.50
a ep
e
ac
m s Q S4 A5
ob
Sy acte ella
t
r
z
2
o
A
1.00
orib mat bium ratif idan r Q 5ID G5
K
5 7
e B
1.00
x
1.00
h
o
ig
tus
St hlor clat rroo rub nei B1L C8 5
0.99
ida
C
4A 12
fe cter oo m
eo
1.00
1.00
nd
a
Q
b
0.50
u
Y
a
C
ph bium iba m ofil e Q4
1.00
n
lin du pt rola i
0.50
yo loro
1.00
1.00
Sa fun cry me aud
ict
h
1.00
d
o
C
lo
pr um on hab
1.00
1.00
Pe
e
z
0.50
uli
1.00
id cha hy m c
c
1.00
c
r
A a
s
iu
r
io od
0.50
0.99
o
id m
1.00
K
s
an
1.00
0.83
Cy Pla
State of PF03030 on
terminal branches
1.00
0.50
1.00
0.88
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.51
1.00
0.68
0.56
1.00
1.00
Figure 2 Unrooted phylogram from Bayesian analysis of 279 V-H+PPase sequences from the three superkingdoms of life. Terminal
branches of V-H+PPases are labeled with a solid square when the Pfam domain was found duplicated in the sequence. If the domain were
found single, the terminal branch has no square. All the branches are labeled with the name of the species and accession number. Eukaryote
species are represented with black letters, Bacteria with blue and Archaea with red.
i.e., that the volutin granule/acidocalcisome was likely
present in the LUCA. Because acidocalcisomes have
been confirmed in Eukarya and Bacteria, and volutin
granules that are remarkably similar to acidocalcisome
like-structures have been observed in Archaea [35-37],
these findings support the view that this organelle might
have been conserved in all superkingdoms. Even if
volutin granules of Archaea are not confirmed to be
genuine acidocalcisomes and represent ancient forms of
the acidocalcisomes organelle, the universal presence of
the organelle precursor (volutin granules) suggests that
the LUCA had a more complex cytoplasm organization
than previously thought. Moreover, the PF03030 domain
which carries the V-H+PPase proton pumping molecular
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
function crucial for the acidification of the acidocalcisome using PPi as substrate (see below) is distributed
universally. For these reasons, we hypothesize that this
organelle or its precursors were very early cellular innovations that occurred prior to major organismal lineage
diversification events. As the lineages of the tree of life
began to diverge, this organelle or its precursors were
inherited in the three domains by direct descent from
the LUCA. Under this scenario, the putative absence of
the organelle in an organism must be explained as a
loss.
It is outside of the scope of the present work to discuss the characteristics of the LUCA, but the above
hypothesis implies that the ancestral lineage may have
had a high degree of cellular sophistication, as others
have suggested [43-46]. For example, Collins [47] argued
that introns and spliceosomal machinery in the eukaryotic ancestor might have been similar in overall complexity to that seen today. Likewise, protein architecture
offers another clue because a substantial fraction of
domain structures in the protein world are common to
all organismal superkingdoms and are truly ancient [23].
While the 3D structure of the proton V-H + PPase
pumps is not yet available, a recent analysis revealed
that sequences from this protein family has a very
ancient and strongly conserved 57-residue block with
unique properties [41]. This finding also supports the
idea that V-H+PPases appeared very early in evolution
of life. Our analysis of the distribution of the main Pfam
domain along the V-H + PPase phylogenetic tree reinforces this hypothesis (Figure 2). In addition, one of the
main functions of the acidocalcisome, the modulation of
polyP homeostasis, could offer further insight into its
evolutionary link with V-H +PPases. The most striking
feature of acidocalcisome is the capacity to store polyP
and PPi. This polymer is present in every living cell, and
was long thought to be a molecular fossil conserved
from a prebiotic time with no remaining cellular function in the cytoplasm of cells. However, this molecule
has been rediscovered, and evidence is mounting for the
many functions of polyP. PolyP is an anionic, negatively
charged molecule. Its length varies from the dimeric
pyrophosphate to chains of thousands of orthophosphate residues linked by phosphoanhydride bonds,
similar to those found in ATP. PolyP is a molecule that
was apparently present on Earth before life appeared
[48,49]. When the planet was still hot, phosphate might
have occurred largely in the form of polyP, which was
formed by a wide variety of abiotic processes including
the dehydration of orthophosphate at high temperatures
[50]. However, it was recently proposed that the prebiotic chemistry of phosphorus was shaped by phosphorous
(P) from extraterrestrial material, called schreibersite,
(Fe,Ni)3P. This mineral reacts with water to form more
Page 6 of 15
soluble reduced P compounds; thus, the oxidation of
schreibersite in water forms several potentially prebiotic
phosphorous species including phosphite, pyrophosphate
and triphosphate, and phosphonates [51]. Alternatively,
it has been suggested that in the early Earth, cyanate
might have mediated the synthesis of PP i from CaHPO4·2H2O (brushite) [52]. Brushite is the only solid species formed between pH 6 and 7, which is thought to
have been in the pH range of the earlier marine environment [53].
PP i and polyP can be used as a source of energy.
However, the developments of proton gradients are a
more sustainable strategy to generate potential energy
available for work in a cell. Proton gradients are as universal as proteins and nucleic acids in living organisms.
Therefore, the acidocalcisome, acting like an energy
source, might have played a critical function for survival
of the LUCA cells. Interestingly, V-H + ATPases and
ATPsynthases, which are capable of creating proton gradients across the cell membranes, are among the most
ancient protein structures that are known [23]. In addition, it has been suggested that the catalytic and noncatalytic subunits found in these proteins evolved from the
same enzyme already present in the LUCA [54-56].
Remarkably, the V-H+ATPase was found in the acidocalcisome of several eukaryotic species together with the
V-H+PPase [11,57]. Both V-H+PPase and V-H+ATPase
couple the transport of protons into cellular compartments and enhance the acidification of the lumen of
this organelle [11]. Moreover, V-H+ATPases can “run in
reverse” to synthesize ATP when subjected to a sufficiently large H+ electrochemical gradient [55], while VH+PPsaes can have the same ability to regenerate pyrophosphate[58,59]. PPi can be used as an alternative to
ATP, because of its similarity to the ATP bond structure. In addition, PPI acts as a biological energy donor
in photosynthesis [60,61]. Therefore, it has been suggested that PPi might have been involved as an energy
carrier during the early evolution of life [62]. Taken
together, these observations suggest that the LUCA
might have used the polyP stored in the acidocalcisome
as a substrate for its V-H+PPases, which together with
the V-H +ATPase make the acidocalcisome capable of
sustained energy generation. As a result, the energy
demands of the LUCA in the harsh environments of
early Earth might have been met by a combination of
polyP hydrolysis, creation of a H+ gradient, and regeneration of ATP and PP i . Moreover, polyP may have
facilitated the assembly and orientation of the key molecules such as phospholipids, nucleic acids, and proteins
[63]. The multifunctionality of polyP makes this molecule a unique link between living organisms and the
inorganic world. PolyP structural and physico-chemical
characteristics are likely the reason this molecule was
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
selected as a core component of living cells during evolution [18]. Although the chemical identity of polyP has
remained unchanged, its function might have become
more specialized during the organism diversification
from the ancestral cells to apparently simple bacterial
cells and the platelets of mammals. Now a large body of
emerging evidence has begun to elucidate the role of
polyP in a wide variety of physiological processes including physical and chemical stresses. Indeed, one of the
main functions of the acidocalcisome is linked to cellular responses to environmental stresses [64]. Therefore,
it would be entirely plausible to hypothesize that the
presence of the acidocalcisome in the LUCA might have
been advantageous for adapting to the environment of
the early Earth.
Conclusions
The sequence and domain analysis of the V-H+PPases
together with the characteristics and universal distribution of the volutin granules and its main component,
polyP, offer pertinent clues about the evolutionary history of the acidocalcisome. Our findings suggest that the
V-H + PPases, which are universally distributed and
tightly associated with the acidocalcisome, originated
prior to the divergence of the three superkingdoms.
Interestingly, volutin granules-polyphosphate bodies that
have been detected in Methanosarcina acetivorans and
other Archaea [35-37] are remarkably similar to acidocalcisomes of Agrobacterium and Rhodospirillum [9,10].
Confirmation of the existence of the acidocalcisome in
the Archaea using biochemical, enzymological and cytological characterization will be significant because it will
further support the hypothesis that this organelle or its
precursors were indeed inherited directly from the
LUCA. Further characterization of the prokaryotic acidocalcisome may confirm the presence of other proteins
previously detected within the eukaryotic acidocalcisome. These findings could be used for a more comprehensive analysis of the evolutionary history of the
acidocalcisome. We trust these results will shed new
light in understanding the putative complexity of the
LUCA and the early evolution of the three superkingdoms of life.
Methods
Sequences and phylogenetic analysis
Sequences for vacuolar pyrophosphatases were downloaded in FASTA format obtained from the Pfam database (see below), and were aligned with Clustal W [65].
Ambiguously aligned sites were removed using BioEdit
ver. 7.0.9 [66].
We initially collected 98 V-H+PPase sequences from
organisms spanning the three superkingdoms of life
including the unicellular parasites Trypanosoma,
Page 7 of 15
Toxoplasma and Plasmodium [26,27] and the green
algae Chlamydomonas reinhardtii [28], where acidocalcisomes were first identified. The Pfam search indicated
that all of the 98 sequences have one or two PF03030
(V-H+PPase) domain. Consequently, we retrieved all the
protein sequences that harbor the domain from the
Pfam database. As a result, 776 protein sequences in
total were obtained and then aligned and edited, and
imported into phylogenetic reconstruction using Bayesian MCMC methods. By manually examining the topology of the phylogenetic tree, we removed duplicated
sequences with no genetic distance and sequences that
are not identified taxonomically. Subsequently,
sequences that have paralogous relationships in the tree
were totally removed in this analysis because it is difficult to know which sequences of paralogs are true
orthologs for the pyrophosphatase on the tree. We
finally selected 279 out of 776 pyrophosphatase
sequences that consist of 31 eukaryal, 231 bacterial and
17 archaeal species (Additional File 2). The multiple
sequence alignment for the sequences followed by deletion of ambiguously aligned sites resulted in 1,652 sites.
The conservation scores of individual aligned sites were
calculated using JALVIEW version 2 [67].
From the multiple sequence alignments manually edited, phylogenetic trees were reconstructed using
MrBayes version 3.1.2 [68], which runs two independent
Markov chain Monte Carlo (MCMC) chains for each
analysis and checks for convergence between them. The
JTT [69] model of amino acid replacement was used,
and 4 chains were run for each analysis, sampling trees
every 100 generations. 2,000,000 generations were run,
and 5,000 trees (25%) were discarded like burnin. The
resulting tree files were summarized into a consensus
tree showing posterior probabilities, visualized, and
color-coded using Dendroscope version 2 [70]. We
excluded the aligned sites that include gaps in the multiple sequence alignment and calculated the overall mean
distance among the 279 sequences of the alignment
using MEGA5 with the JTT model [71]. For each of the
non-aligned 279 sequences, the topology of transmembrane helices was predicted using SCAMP with default
parameters [72]. We also used several other algorithmic
implementations, including TMMOD, an HMM-based
transmembrane protein prediction implementation [73],
and TOPCONS, a consensus prediction tool of membrane protein topology [74].
Protein domain analysis
We used the Pfam database version 24.0 to assign protein domains to the vacuolar pyrophosphatase sequences
[42]. On the Pfam website, http://pfam.sanger.ac.uk/
family/PF03030#tabview=tab6, we performed a batch
search for the sequences against both the Pfam-A
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
HMMs using an E-value of 10-4 [75]. To assign protein
domains at the levels of fold superfamily and fold family
that are defined in the Structural Classification of Proteins [76], we downloaded a MYSQL database from the
SUPERFAMILY database, version 1.73 [77] and searched
the sequences against HMMs of fold superfamilies and
fold families using the iterative sequence alignment and
modeling system (SAM) method [78]. As with the Pfam
search, an E-value of 10-4 was used because it represents
an ideal cutoff to minimize false positive assignments in
the HMM searches [79].
Electron Microscopy
Agrobacterium tumefaciens cells were washed with
Dulbecco’s PBS and fixed for 1 h with 2.5% grade II
glutaraldehyde, 4% freshly prepared formaldehyde,
0.03% CaCl2, and 0.03% picric acid in 0.1 m cacodylate
buffer, pH 7.2. Bacteria were post-fixed with OsO4 for
45 min and then for 15 min with potassium ferricyanide, washed, and treated with 2% uranyl acetate for
30 min. Subsequently, samples were dehydrated by
successive incubations of 6 min with increasing concentrations of ethanol (10, 25, 36, 75, 95, and 100%) at
room temperature. Epoxy embedding was carried out
by resuspending the sample once in 1:1 ethanol/acetonitrile, twice in 100% acetonitrile, and then 30 min in
1:1 epoxy/acetonitrile, 1.5 h in 3:1 epoxy/acetonitrile,
and 4 h in 100% epoxy. Embedded samples were polymerized for 20 h at 85°C. Epoxy blocks were ultrathinsectioned, sections were deposited on 300-mesh copper
grids and grids were stained with uranyl acetate for 30
min and triple lead stain (lead citrate, lead nitrate, and
lead acetate) for 1 min. For Methanosarcina acetivorans, cells were processed for electron microscopy as
described in [80].
Competing interest statement
The authors declare that they have no competing
interests.
Reviewers’ comments
Reviewer’s report 1
Anthony Poole (University of Canterbury, New Zealand)
The work presented here makes an interesting case for
the existence of compartmentation early in evolution,
based on the presence of volutin granules/acidocalcisomes in all three domains of life. The authors present
data indicating that vacuolar pyrophosphatases are
found in all three domains of life, and interpret this as
consistent with their conjecture that acidocalcisomes
predate the divergence of the three domains. It is clear
that the phylogeny of vacuolar pyrophosphatases cannot
be used to strongly argue for presence of this gene in
the LUCA, partly because of poor resolution in the
Page 8 of 15
dataset, and partly because of unambiguous evidence for
horizontal gene transfer within the dataset. While there
are some clear cases of late transfers in their phylogeny,
it is harder to establish events deeper in the tree. Mapping the domain architecture of PF03030 onto this tree
does not obviously strengthen the evidence, though it is
striking to note the high level of conservation of the
domain itself.
While it certainly seems reasonable to expect that a
complex organelle may only have arisen once in evolution (meaning acidocalcisomes should be considered
ancient), such expectation is not evidence in itself. In
further testing this conjecture, it will therefore be interesting to identify a broader set of common signature
proteins associated with this organelle, and examine
their evolutionary histories. In addition, examining
whether patterns of gain and loss of both the organelles
and key associated proteins can be ascertained using
more detailed phylogenies and character analyses within
a phylogenetic framework (as was successfully demonstrated for hydrogenosomes - see Embley et al. Phil.
Trans. R. Soc. Lond. B (2003) 358, 191-203) will provide
an important means of testing the conjecture presented
here.
The following major claims are made on the basis of
the data presented:
1. That Archaea possess membrane-bound acidocalcisomes/volutin granules.
2. That volutin granules evolved prior to the divergence of the three domains.
These are bold claims. However, my reading of the
manuscript leads me to conclude that the authors do
not currently have sufficient evidence to fully support
either of these claims. I address each in turn.
1. That Archaea possess membrane-bound acidocalcisomes/volutin granules.
Figure 1 presents electron micrographs of Methanosarcina acetivorans cells (panels c & d) wherein intracellular structures are clearly visible as either black
(electron dense) or white (putatively empty). The resolution of panels c & d make it difficult to see any evidence
for a membrane; plate d in particular is pixellated. I
therefore find it difficult to evaluate the presented evidence as being either supportive or dismissive of a
membrane surrounding these structures. No other cell
biological evidence is given to support the interpretation
presented in the paper, in notable contrast to the careful
and extensive experimental evidence previously presented by the lead author for the characterization of
acidocalcisome structures in the bacterium Agrobacterium tumefaciens (J Biol Chem 278, 29971-8). The results
in figure 1 therefore are far too preliminary to be presented in support of acidocalcisomes or acidocalcisomelike structures in Methanosarcina.
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
I note that proton pyrophosphatase pumps have previously been identified in Archaea, having been experimentally characterized in Pyrobaculum aerophilum
(Drozdowicz et al. 1999 FEBS Lett 460:505-12) – it
seems somewhat surprising that no reference to this
paper is made. From what I can ascertain, no data indicate an association of these pumps with any internal
membrane structure in Pyrobaculum. I am therefore
curious as to whether acidocalcisome-like structures
have been characterized in this archaeon. Finally, given
that those authors were able to express the Pyrobaculum
pyrophosphatase in yeast, have antibodies have been
successfully raised against this enzyme? This is helpful
information, since my reading of the literature indicates
that these pumps are known to associate with acidocalcisomes, though not exclusively; with that in mind, it is
rather a leap to equate presence of H+-PPases in
Archaea with the presence of acidocalcisomes. Experiments of the sort presented in the above-noted JBC
paper are essential for any claim for the existence of
such structures in Archaea.
Authors’ Response: In this paper we neither directly
claim that the volutin granules of Archaea are acidocalcisomes nor do we explicitly extrapolate acidocalcisome
presence based on identification of V-H+-PPases in Genbank. What we are postulating is that the volutin granules observed with electron microscopy in
Methanosarcina are indeed very similar to the volutin
granules of Agrobacterium and Rhodospirillum, which
now are known to be acidocalcisomes. In addition, our
hypothesis is based on previous publications where
micrographs of the volutin granules of Methanosarcina
and other Archaea such as Acidithiobacillus ferrooxidans
and Sulfolobus metallicus are structurally very similar to
the acidocalcisomes found in bacteria. Our intention of
showing these structures is to provide a visual connection
between volutin granules (which we do have direct evidence of) and acidocalcisomes (an inference that requires
considerably more evidence to demonstrate).
We consider that the information we present in this
paper is significant because it provides concrete facts and
clues about the possible presence of this organelle in
Archaea. In addition, this work prompts further inquiry
and exploration of this organelle in other organisms
(including the full characterization of the volutin granules in Archaea as suggested by the reviewer). The full
characterization of the volutin granules of Methanosarcina is not a trivial matter. Based on similar previous
studies, it might well take 1-2 years of work to complete.
Therefore, the present scope of this paper is not to prove
that the volutin granules of Methanosarcina are acidocalcisomes, but instead illustrate our hypothesis that
they are promising candidates (among others) to be acidocalcisomes. We appreciate the note about the work of
Page 9 of 15
Drozdowicz et al. (1999). The reference has now been
included. Originally, we did not include this reference
because the discovery of V-H+PPase in Pyrobaculum was
made more than 10 years ago and we considered this
information to be common knowledge (we did include
this species in our analysis). Interestingly, Drozdowicz et
al. (1999) also made an early suggestion of the possibility
that the V-H+PPases could have originated in the last
universal common ancestor. However, and as the
reviewer points out, no data is presented in the paper
that indicates an association of these pumps with any
internal membrane structure in Pyrobaculum. This is
not surprising. At the time of the publication of this
paper, the association between volutin granules and
acidocalcisomes in bacteria had not been reported.
Moreover, the absence of organelles (similar to eukaryotic
organelles) in bacteria was at that time already considered a dogma in microbiology. Consequently, the search
for a membrane bound organelle in bacteria was not
undertaken. In the case of Agrobacterium for example,
the presence of volutin granules and vacuolar pyrophosphatases were described a long time ago, but it was not
until 2003 when the volutin granules of Agrobacterium
were rediscovered as acidocalcisomes and the V-H+PPase
was located in their surrounding membrane. Even more
eloquent is the fact that in Rhodospirillum rubrum, the
microbe where the V-H+PPase was first discovered, the
volutin granules were never investigated until 2004,
when they were confirmed as acidocalcisomes and the VH+PPase also localized in their surrounding membranes.
2. That volutin granules evolved prior to the divergence of the three domains.
The authors also present genomic evidence (H+-PPase
sequences) in support of acidocalcisomes in archaea.
The implicit argument in this part of the paper is that
one can extrapolate acidocalcisome presence based on
identification of H+-PPases in Genbank.
There are two interesting points here. First, it is nice
to see that the authors are honest regarding the limits
of phylogeny in this particular instance. The previously
inflammatory acronym MUTOG (Myth of the Universal
Tree from One Gene – coined by David Penny) does
need to be taken seriously following the recognition that
most genes do not carry sufficient phylogenetic signal to
trace the relationships between species across the three
domains. Moreover, with HGT an indisputable factor in
gene distribution, this caution is not only welcome, but
necessary. Against this backdrop, that a clearly resolved
three-domain tree cannot be generated for the V-H
+
PPases (for instance, the archaeal sequences do not
form a monophyly: Methanococcoides is on alone
branch, Pyrobaculum and Caldivirga form a clade, and
there is a third clade of five sequences, including Methanosarcina acetivorans, that connects the tree at the large
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
central polytomy) is not evidence against an ancient preLUCA origin. Having said that, it is not evidence in
favor of it either. The dual forces of mutation and gene
transfer are having a major impact on the reconstruction of the deep evolutionary past and it may be that
signal has in most cases been heavily degraded, to the
point that it cannot be informatively identified.
The second point lies at the heart of this paper. The
triumph of endosymbiotic theory has been the recognition that some key eukaryotic organelles did not evolve
de novo, but are instead the result of one cell becoming
resident within another. The emerging view of complex
membrane structures has therefore been one of continuity (TCS), though this may yet be challenged by the
recognition that both bacteria and archaea can harbor
complex cell structures. In the case of acidocalcisome/
volutin granule organelles, it seems plausible to suggest
a deep evolutionary origin, given the presence of both
the H+PPases and these organelles across representatives
from all three domains of life. This would suggest that
multiple losses have occurred throughout evolution,
from a more compartmentally complex common ancestor. What we still don’t know, and are challenged to
think about by this paper, is whether complex internal
membrane structures such as these could arise multiple
times independently. In this respect, comparative analyses do have an important role to play in establishing
what the minimum number of components required to
support an acidocalcisome or volutin granule may be.
The variability in membrane number and architecture
observed across Planctomycetes (Fuerst) does suggest a
greater capacity for evolution of novel membranes than
suggested by a stringently ‘genetic’ view (sensu CavalierSmith) of membrane evolution. More fine-scale examination of whether the distribution of key proteins localized to acidocalcisomes/volutin granules completely
overlaps with the existence of these organelles. In addition, looking at evolution within a group (e.g. relatives
of Agrobacterium tumefaciens) may help elucidate
whether or not horizontal gene transfer does play a role
in the emergence of such complex cell structure. It
might be time to ask just whether our assumptions
about how ‘hard’ it is to evolve a compartment are in
need of scrutiny.
Authors’ Response: We agree with the reviewer that the
poor phylogenetic resolution of the tree generated from
V-H+PPase gene sequences, a common finding in many
genes that exist in a genome, tells little about the ancestrality of the V-H+PPase gene. Lack of phylogenetic signal
in the V-H+PPase sequences is probably the result of (i)
high sequence conservation of segments that correspond
to the Pfam 03030 domain that are present in each and
every one of the sequences examined, including sites
necessary to maintain fold structure and sites needed for
Page 10 of 15
catalytic activity, and (ii) segments of sequence that are
highly variable and prone to sequence saturation that
contain highly degraded phylogenetic signal and probably arise from the complex and frustrated interplay
between stability and function that exists in proteins
[81]. In contrast, the universal distribution of the Pfam
03030 domain in the tree of V-H+PPase sequences and
in every other V-H + PPase sequence we examined provides strong support to the suggestion that the functional
and structural core of this domain was already present
in LUCA and was vertically distributed without loss in
all lineages of the tree of life. The size of the domain (an
average of 603 amino acids covering 94% of the sequence
with a model length of 682 amino acids) was calculated
based on start and end positions on the HMM (Pfam)
alignment of sequences obtained by searches in the Sanger database. The average identity of the complete alignment is 44%. Despite sequence variability, the domain is
therefore highly conserved and universally present. More
information about seed sequences and PF03030 domain
distributions can be found elsewhere. (http://pfam.sanger.
ac.uk/family/PF03030#tabview=tab6).
The second point raised by the reviewer is important
for the genesis of organelles and cellular structure. The
proteome of LUCA was recently reconstructed and was
shown to embody the functions of a complex organism,
with pathways for the biosynthesis of membrane sn1,2
glycerol ester and ether lipids [82]. An ancient volutin/
acidocalcisome structure present in LUCA implies gradual evolution of internal cellular structure in the universal ancestor of cellular life. By definition, LUCA
implies the absence of cellular lineages. Consequently, a
process of endosymbiosis between lineages similar to that
used to explain the origin of mitochondria or plastids
could not be responsible for the rise of volutin/acidocalcisome organelles without invoking endosymbiosis was an
ancient and pervasive property of LUCA and that these
polyphosphate harboring cellular structures were remnants of that cellular state. Alternatively, cellular structure other than an external membrane may be the
logical consequence of energetic and architectural constraints operating on membrane organization [83]. In
vitro experiments of membrane biogenesis show considerable diversity of cellular shape and structure, with cellular engulfment being a common outcome [84,85].
The argument presented here, that compartmentation
is much older than we may expect based on the presumed late origin of the eukaryote nucleus and endomembrane system, has been given an additional boost in
this manuscript from Seufferheld and colleagues. However, the jury is certainly still out – on the present data,
it is difficult to distinguish between either a deep evolutionary origin with numerous losses, or ‘convergent’ origins (of the organelles, not the H+PPases). We do not
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
yet know whether independent recruitment (via horizontal transfer) of key functionalities, such as the H
+
PPases, are sufficient for the independent evolution of
a new membrane-bound organelle. The authors make
the reasonable assumption that loss is more likely, and
bolster their argument by relating their provocative
model a possible early role for pyrophosphate/polyphosphate in the origin and early evolution of life. The
obvious problem with scenarios that invoke loss from a
more complex ancestral state is that one also needs to
explain the original gain. The present manuscript does
not give us an answer, but this and other results do
challenge us to keep an open mind.
Authors’ Response: A scenario of multiple parallel or
convergent origins of volutin/acidocalcisome cellular
structures in lineages is possible but is less parsimonious
than a single origin in LUCA. The reviewer however is
right that a more comprehensive analysis of molecular
machinery responsible for key functionalities of this compartment will help understand the role that horizontal
gene transfer and molecular recruitment may have in
shaping cellular structure. Phylogenomic analysis of the
acidocalcisome-associated protein repertoire could be
very informative in this regard.
Reviewer’s report 2
Lakshminarayan Iyer (NIH/NLM/NCBI)
Seufferheld and colleagues perform a phylogenetic
analysis of the V-H+ PPase domain, a component of the
acidocalcisome in the eukaryotes and bacteria. They
argue that the acidocalcisome was present in LUCA
along with the V-H+ PPase.
Critique:
The phylogenetic analysis of the V-H+ PPases suggests
a complicated evolutionary history with loss and possible lateral transfer playing a role in their distribution.
However, reconstruction to LUCA is not strongly supported. Firstly this domain family is not widespread in
the archaea. Further, the major archaeal branches (e.g.
eury- and cren- archaea) also do not cluster with each
other. Even within the bacteria, homologs from monophyletic groups do not group with each other. Given the
presence of this family in diverse bacteria, I would find
it more parsimonious to reconstruct the evolutionary
history of this family as having evolved in the bacteria
with subsequent lateral transfer to the archaea and
eukaryotes.
The paragraph on the Pfam domains does not fit into
the paper since these proteins show no domain complexity. Further, in this case the Pfam domain deposited
in the database only captures the key catalytic motif of
the protein. If the authors wish to discuss the domain in
terms of sequence, they can write about the various
Page 11 of 15
transmembrane helices and the context of the catalytic
residues.
Authors’ Response: Our phylogenetic analyses of the VH+PPase proteins include the amino acid sequence of the
entire membrane-associated protein and not only of the
domain regions. Furthermore, our phylogenetic tree
describes the evolution of the sequence in 279 organisms
spanning the three superkingdoms of life, Archaea (17
species), Bacteria (231), and Eukarya (31). These 279
sequences were selected from an analysis of a larger set
of 776 sequences, following removal of paralogs or
sequences with no genetic distance or that have no clear
taxonomical descriptions. We find that the V-H+PPases
are present in every one of the 776 organism analyzed
and that the Pfam domain PF03030 is always present
and captures the main sequence profile of the structure
of these proteins. The claim of the reviewer that the
domain is absent in Archaea and that its evolutionary
history follows a complex series of losses and possible
gains through LGT is therefore incorrect. The protein
and associated domain is simply present everywhere.
While variants could have been recruited by LGT and
replaced, the protein was never lost in any lineage examined. We agree that LGT events are possible. They could
have been responsible for the lack of phylogenetic signal
that limits the correct grouping of organisms in the V-H+
PPase tree. However, the topology of the tree has clearly
defined groups that match taxonomical classification,
including the Euryarchaeota, Proteobacteria, Thermotogales, Actinobacteria and Firmicutes. Some cases show
non-monophyletic groupings that could have resulted
from LGT. Nevertheless, it is important to notice that in
several of these non-monophyletic groupings the phylogenetic support is very poor 0.5 to 0.76. While the tree of
the vacuolar pyrophosphatase is one gene tree and has a
largely star-like internal topology, it is significant that
the phylogenetic reconstruction of this gene maintains
essentially the three superkingdoms of life.
An analysis of the mean distance between protein
sequences using the JTT model with gaps excluded in the
analysis showed that levels of sequence identity were
64.4%, much larger the 30% expected in the presence of
high conservation levels. Using JALVIEW, we extracted
conservation scores (in a scale from 0 to 11) for every
nucleotide site with a proportion of gaps that is less than
25%. Analysis of the 523 sites (out of the total 1,652 in
the alignment) that fulfills that criterion, shows the vast
majority of sites have conservation scores above 7 (251
sites, 48%) and many have scores above 10 (90 sites,
17%) (Figure 3). These results support the notion that VH+PPases have highly conserved protein sequences and
that the variable regions have little or highly degraded
phylogenetic signal. This explains the star-like
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
Page 12 of 15
Figure 3 Sequence conservation and transmembrane helical (TMH) make up of V-H+PPases. (A) Sequence conservation of the set of
aligned V-H+PPase sequences illustrated with sequence conservation scores obtained with JALVIEW. (B) Pfam domain definitions and TMH
regions in the V-H+PPase of R. rubrum. The plots show TMH definitions obtained with TMMOD and TOPCONS. TMMOD TMH definitions are given
together with posterior probabilities of the existence of a TMH region. The TOPCONS definitions are from a consensus
appearance and poorly supported nature of the V-H +
PPase tree and suggests the V-H+PPase domain is old,
and from a structural perspective, is highly canalized.
This conclusion, together with its wide if not universal
distribution in life, suggests that the most parsimonious
explanation of the origin of this membrane-associated
protein is in LUCA and not in any lineage of the diversified world of organisms, as the reviewer claims.
While a minority of sequences contain a duplicated
PF03030 arrangement or are associated with another
domain (PB008043), most contain a single Pfam domain.
Thus, and as the reviewer correctly points out, our
analyses show V-H + PPases do not exhibit a complex
domain organization. Complex domain arrangements
are the hallmark of recently evolved proteins in the protein world [25]. Consequently, the domain make up of
V-H+PPases provides additional clues in support of their
ancient origin. Remarkably, an analysis of trans-membrane helices (TMH) of the protein shows also high levels
of structural conservation. The number of TMHs ranges
9-11with an average of 10.2 ± 0.75 (SD) and TMHs hold
an average of 243.2 ± 20.30 amino acid residues. TMHs
therefore represent on average 42.1 ± 3.46% of the protein. Figure 3 illustrates the position of the two PF03030
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
domain and the 10 TMHs in the sequence of R. rubrum
YP426905, the model system for acidocalcisome studies.
The V-H+PPases are therefore evolutionarily conserved
at different levels, including sequence, domain structure,
and TMH make-up, a fact that speaks about their
ancient origins.
Page 13 of 15
3.
4.
5.
6.
Reviewer’s report 3
Daniel Kahn (Laboratoire de Biométrie et Biologie Évolutive, Université Lyon, France)
This reviewer provided no comments for publication
Additional material
Additional file 1: Multiple alignments of PF03030 domain
sequences distributed in the three superkingdoms (Archaea,
Bacteria, and Eukarya). Shown above is the strongly conserved 57residue region of the V-H+-PPase identified by Hedlund et al [53] and
viewed using JALVIEW provided by the Pfam database.
Additional file 2: Taxa and accession numbers used in the
phylogenetic analyses.
7.
8.
9.
10.
11.
12.
13.
Abbreviations
V-H+PPases: Vacuolar proton-translocating pyrophosphatases; Pfam: Protein
family; LUCA: Last universal common ancestor; LGT: Lateral gene transfer;
PDB: Protein Data Bank; HMMs: Hidden Markov Models; SAM: Sequence
alignment and modeling system
Acknowledgements
The authors thank Dr. Kevin Sowers for providing the Methanosarcina
pictures and Dr. Yoshie Hanzawa for comments. This material is based upon
work supported by the Cooperative State Research, Education and Extension
Service, U. S. Department of Agriculture, under projects ILLU-875-389 (to
MJS) and ILLU-802-314 (to GCA), and by the National Institute of Health
under project AI079478-02 (to MJS) and the National Science Foundation,
under project MCB-0749836 (to GCA). Opinions, findings, and conclusions or
recommendations expressed in this material are those of the authors and
do not necessarily reflect the views of the funding agencies.
Author details
1
Department of Crop Sciences, University of Illinois at Urbana-Champaign,
Urbana IL USA. 2Evolutionary Bioinformatics Laboratory, Department of Crop
Sciences, University of Illinois at Urbana-Champaign, Urbana IL USA. 3Korean
Bioinformation Center (KOBIC), Korea Research Institute of Bioscience and
Biotechnology (KRIBB), Daejeon 305-806, Korea. 4Department of Entomology,
University of Illinois at Urbana-Champaign, Urbana IL, USA. 5Department of
Evolution, Ecology and Organismal Biology, Museum of Biological Diversity,
1315 Kinnear Rd., Columbus, OH 43210, USA.
Authors’ contributions
MJS, JW and GCA conceived and designed the experiments. MJS, JW, GCA,
KMK and AV performed the experiments and/or participated in data analysis.
MJS, JW, GCA, and KMK wrote the paper. All authors read, edited and
approved the final manuscript.
Received: 21 September 2011 Accepted: 5 October 2011
Published: 5 October 2011
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
References
1. Margulis L: Symbiotic theory of the origin of eukaryotic organelles;
criteria for proof. Symp Soc Exp Biol 1975, 21-38.
2. Martin W, Russell MJ: On the origins of cells: a hypothesis for the
evolutionary transitions from abiotic geochemistry to chemoautotrophic
27.
prokaryotes, and from prokaryotes to nucleated cells. Philos Trans R Soc
Lond B Biol Sci 2003, 358:59-83, discussion 83-55..
Dyall SD, Brown MT, Johnson PJ: Ancient invasions: from endosymbionts
to organelles. Science 2004, 304:253-257.
Bhattacharya D, Yoon HS, Hackett JD: Photosynthetic eukaryotes unite:
endosymbiosis connects the dots. Bioessays 2004, 26:50-60.
Stiller JW: Plastid endosymbiosis, genome evolution and the origin of
green plants. Trends in Plant Science 2007, 12:391-396.
Jones LJ, Carballido-Lopez R, Errington J: Control of cell shape in bacteria:
helical, actin-like filaments in Bacillus subtilis. Cell 2001, 104:913-922.
van den Ent F, Amos LA, Lowe J: Prokaryotic origin of the actin
cytoskeleton. Nature 2001, 413:39-44.
Shih YL, Rothfield L: The bacterial cytoskeleton. Microbiol Mol Biol Rev
2006, 70:729-754.
Seufferheld MJ, Vieira MC, Ruiz FA, Rodrigues CO, Moreno SN, Docampo R:
Identification of organelles in bacteria similar to acidocalcisomes of
unicellular eukaryotes. J Biol Chem 2003, 278:29971-29978.
Seufferheld MJ, Lea CR, Vieira M, Oldfield E, Docampo R: The H
+
-pyrophosphatase of Rhodospirillum rubrum is predominantly located
in polyphosphate-rich acidocalcisomes. J Biol Chem 2004,
279:1193-51202.
Docampo R, de Souza W, Miranda K, Rohloff P, Moreno SN:
Acidocalcisomes - conserved from bacteria to man. Nat Rev Microbiol
2005, 3:251-261.
Motta LS, Ramos IB, Gomes FM, de Souza W, Champagne DE, Santiago MF,
Docampo R, Miranda K, Machado EA: Proton-pyrophosphatase and
polyphosphate in acidocalcisome-like vesicles from oocytes and eggs of
Periplaneta americana. Insect Mol Biol 2009, 39:198-206.
Kornberg A: Inorganic polyphosphate: toward making a forgotten
polymer unforgettable. J Bacteriol 1995, 177:491-496.
Meyer A: Orientierende untersuchungen ueber verbreitung.
morphologie, und chemie des volutins. Bot Zeit 1904, 62:113-152.
Vercesi AE, Moreno SNJ, Docampo R: Ca2+/H+ Exchange in acidic vacuoles
of Trypanosoma brucei. Biochem J 1994, 304:227-233.
Shiba T, Tsutsumi K, Yano H, Ihara Y, Kameda A, Tanaka K, Takahashi H,
Munekata M, Rao NN, Kornberg A: Inorganic polyphosphate and the
induction of rpoS expression. Proc Natl Acad Sci USA 1997,
94:11210-11215.
Nomura K, Kato J, Takiguchi N, Ohtake H, Kuroda A: Inorganic
polyphosphate stimulates Lon-mediated proteolysis of nucleoid proteins
in Escherichia coli. Cell Mol Biol 2006, 52:22-29.
Kulaev I, Kulakovskaya T: Polyphosphate and phosphate pump. Annu Rev
Microbiol 2000, 54:709-734.
Kulaev IS, Bagabov VM, Kulakovskaia TV, Andreeva NA, Lichko LP,
Trilisenko LV: [Peculiarities of metabolism and functions of highmolecular inorganic polyphosphates in yeasts as representatives of
lower eukaryotes]. Mol Biol (Mosk) 2005, 39:567-580.
McInerney P, Mizutani T, Shiba T: Inorganic polyphosphate interacts with
ribosomes and promotes translation fidelity in vitro and in vivo. Mol
Microbiol 2006, 60:438-447.
Cannon GC, Bradburne CE, Aldrich HC, Baker SH, Heinhorst S, Shively JM:
Microcompartments in prokaryotes: carboxysomes and related
polyhedra. Appl Environ Microbiol 2001, 67:5351-5361.
Komeili A, Vali H, Beveridge TJ, Newman DK: Magnetosome vesicles are
present before magnetite formation, and MamA is required for their
activation. Proc Natl Acad Sci USA 2004, 101:3839-3844.
Caetano-anollés G, Wang M, Caetano-anollés D, Mittenthal JE: The origin,
evolution and structure of the protein world. Biochem J 2009,
417:621-637.
Moore AD, Bjorklund AK, Ekman D, Bornberg-Bauer E, Elofsson A:
Arrangements in the modular evolution of proteins. Trends Biochem Sci
2008, 33:444-451.
Wang M, Caetano-Anollés G: The evolutionary mechanics of domain
Organization in Proteomes and the rise of modularity in the protein
world. Structure 2009, 17:66-78.
Drozdowicz YM, Lu YP, Patel V, Fitz-Gibbon S, Miller JH, Rea PA: A
thermostable vacuolar-type membrane pyrophosphatase from the
archaeon Pyrobaculum aerophilum: implications for the origins of
pyrophosphate-energized pumps. Febs Letters 1999, 460:505-512.
Moreno SN, Zhong L: Acidocalcisomes in Toxoplasma gondii tachyzoites.
Biochem J 1996, 313(Pt 2):655-659.
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
28. Marchesini N, Luo S, Rodrigues CO, Moreno SN, Docampo R:
Acidocalcisomes and a vacuolar H+-pyrophosphatase in malaria
parasites. Biochem J 2000, 347(Pt 1):243-253.
29. Ruiz FA, Marchesini N, Seufferheld MJ, Govindjee , Docampo R: The
polyphosphate bodies of Chlamydomonas reinhardtii possess a protonpumping pyrophosphatase and are similar to acidocalcisomes. J Biol
Chem 2001, 276:46196-46203.
30. Docampo R, Ulrich P, Moreno SNJ: Evolution of acidocalcisomes and their
role in polyphosphate storage and osmoregulation in eukaryotic
microbes. Philos Trans R Soc London [Biol] 2010, 365:775-784.
31. Rodrigues CO, Scott DA, Docampo R: Characterization of a vacuolar
pyrophosphatase in Trypanosoma brucei and its localization to
acidocalcisomes. Mol Cell Biol 1999, 19:7712-7723.
32. Scott DA, de Souza W, Benchimol M, Zhong L, Lu HG, Moreno SNJ,
Docampo R: Presence of a plant-like proton-pumping pyrophosphatase
in acidocalcisomes of Trypanosoma cruzi.. J Biol Chem 1998,
273:22151-22158.
33. Lemercier G, Dutoya S, Luo S, Ruiz FA, Rodrigues CO, Baltz T, Docampo R,
Bakalara N: A vacuolar-type H+-pyrophosphatase governs maintenance of
functional acidocalcisomes and growth of the insect and mammalian
forms of Trypanosoma brucei. J Biol Chem 2002, 277:37369-37376.
34. Jiang L, Phillips TE, Hamm CA, Drozdowicz YM, Rea PA, Maeshima M,
Rogers SW, Rogers JC: The protein storage vacuole: a unique compound
organelle. J Cell Biol 2001, 155:991-1002.
35. Scherer PA, Bochem HP: Energy-dispersive X-ray microanalysis of the
methanogen Methanosarcina barkeri ’Fusaro’ grown on methanol and in
the presence of heavy metals. Curr Microbiol 1983, 9:187-193.
36. Remonsellez F, Orell A, Jerez CA: Copper tolerance of the
thermoacidophilic archaeon Sulfolobus metallicus: possible role of
polyphosphate metabolism. Microbiology 2006, 152:59-66.
37. Alvarez S, Jerez CA: Copper ions stimulate polyphosphate degradation
and phosphate efflux in Acidithiobacillus ferrooxidans. Appl Environ
Microbiol 2004, 70:5177-5182.
38. Nichols R: Gene trees and species trees are not the same. Trends Ecol
Evolut 2001, 16:358-364.
39. Illergard K, Ardell DH, Elofison A: Structure is three to ten times more
conserved than sequence-A study of structural response in protein
cores. Proteins-Structure Function and Bioinformatics 2009, 77:499-508.
40. Elofsson A, von Heijne G: Membrane protein structure: prediction versus
reality. Annu Rev Biochem 2007, 76:125-140.
41. Hedlund J, Roberto C, Margareta B, Herrick B, Bengt P: Analysis of ancient
sequence motifs in the H+-PPase family. FEBS Journal 2006,
273:5183-5193.
42. Finn RD, Tate J, Mistry J, Coggill PC, Sammut SJ, Hotz HR, Ceric G,
Forslund K, Eddy SR, Sonnhammer EL, Bateman A: The Pfam protein
families database. Nucleic Acids Res 2008, 36:D281-288.
43. Kurland CG, Collins LJ, Penny D: Genomics and the irreducible nature of
eukaryote cells. Science 2006, 312:1011-1014.
44. Poole A, Jeffares D, Penny D: Early evolution: prokaryotes, the new kids
on the block. Bioessays 1999, 21:880-889.
45. Penny D, Poole A: The nature of the last universal common ancestor.
Curr Opin Genetics Dev 1999, 9:672-677.
46. Embley TM, Martin W: Eukaryotic evolution, changes and challenges.
Nature 2006, 440:623-630.
47. Collins L, Penny D: Complex spliceosomal organization ancestral to
extant eukaryotes. Mol Biol Evol 2005, 22:1053-1066.
48. Tiedemann H: “Killer” Impacts and life’s origins. Science 1997,
277:1687-1688.
49. Kornberg A, Rao NN, Ault-Riche D: Inorganic polyphosphate: a molecule
of many functions. Annu Rev Biochem 1999, 68:89-125.
50. Yamagata Y, Watanabe H, Saitoh M, Namba T: Volcanic production of
polyphosphates and its relevance to prebiotic evolution. Nature 1991,
352:516-519.
51. Pasek MA: Rethinking early earth phosphorus geochemistry. Proc Natl
Acad Sci USA 2008, 105:853-858.
52. Hagan W, Parker A, Steuerwald A, Hathaway M: Phosphate solubility and
the cyanate-mediated synthesis of pyrophosphate. Origins of Life and
Evolution of Biospheres 2007, 37:113-122.
53. Grotzinger JP, Kasting JF: New Constraints on precambrian ocean
composition. The Journal of Geology 1993, 101:235-243.
Page 14 of 15
54. Jekely G: Did the last common ancestor have a biological membrane?
Biology Direct 2006, 1:35.
55. Drory O, Nelson N: The emerging structure of vacuolar ATPases.
Physiology 2006, 21:317-325.
56. Forterre P, Benachenhou-Lahfa N, Confalonieri F, Duguet M, Elie C,
Labedan B: The nature of the last universal ancestor and the root of the
tree of life, still open questions. BioSystems 1992, 28:15.
57. Scott DA, Docampo R: Characterization of isolated acidocalcisomes of
Trypanosoma cruzi. J Biol Chem 2000, 275:24215-24221.
58. Rocha Facanha A, de Meis L: Reversibility of H+-ATPase and H
+
-pyrophosphatase in tonoplast vesicles from maize coleoptiles and
seeds. Plant Physiol 1998, 116:1487-1495.
59. Au KM, Barabote RD, Hu KY, Saier MH Jr: Evolutionary appearance of H
+
-translocating pyrophosphatases. Microbiology 2006, 152:1243-1247.
60. Siu PML, Wood HG: Phosphoenolpyruvic carboxytransphosphorylase, a
CO2 fixation enzyme from propionic acid bacteria. J Biol Chem 1962,
237:3044-3051.
61. Heinonen JK: Biological role of inorganic pyrophosphate Springer; 2001.
62. Baltscheffsky M, Baltscheffsky H: Alternative photophosphorylation,
inorganic pyrophosphate synthase and inorganic pyrophosphate.
Photosynth Res 1995, 46:87-91.
63. Brown MR, Kornberg A: Inorganic polyphosphate in the origin and
survival of species. Proc Natl Acad Sci USA 2004, 101:16085-16087.
64. Seufferheld MJ, Alvarez HM, Farias ME: Role of polyphosphates in
microbial adaptation to extreme environments. Appl Environ Microbiol
2008, 74:5867-5874.
65. Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through
sequence weighting, position-specific gap penalties and weight matrix
choice. Nucleic Acids Res 1994, 22:4673-4680.
66. Hall TA: BioEdit: a user-friendly biological sequence alignment editor and
analysis program for Windows 95/98/NT. Nucl Acids Symp Ser 1999,
41:95-98.
67. Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ: Jalview
version 2 - a multiple sequence alignment editor and analysis
workbench. Bioinformatics 2009, 25:1189-1191.
68. Ronquist F, Huelsenbeck JP: MrBayes 3: Bayesian phylogenetic inference
under mixed models. Bioinformatics 2003, 19:1572-1574.
69. Jones DT, Taylor WR, Thornton JM: The rapid generation of mutation data
matrices from protein sequences. Comput Appl Biosci 1992, 8:275-282.
70. Huson D, Richter D, Rausch C, Dezulian T, Franz M, Rupp R: Dendroscope:
An interactive viewer for large phylogenetic trees. BMC Bioinformatics
2007, 8:460.
71. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S: MEGA5:
Molecular evolutionary genetics analysis using maximum likelihood,
evolutionary distance, and maximum parsimony methods. Mol Biol Evol .
72. Burnsel A, Viklund H, Falk J, Lindahl E, von Heijne G, Elofsson A: Prediction
of membrane-protein topology from first principles. Proc Natl Acad Sci
USA 2008, 105:7177-7181.
73. Kahsay R, Liao L, Gao G: An improved hidden Markov model for
transmembrane protein topology prediction and its applications to
complete genomes. Bioinformatics 2005, 21:1853-1858.
74. Burnsel A, Viklund H, Hennerdal A, Elofsson A: TOPCONS: consensus
prediction of membrane protein topology. Nucleic Acids Res 2009, 37:
W465-W468.
75. Durbin R, Eddy SR, Krogh A, Mitchison G: Biological sequence analysis:
probabilistic models of proteins and nucleic acids. Cambridge:
Cambridge University Press; 1998.
76. Murzin AG, Brenner SE, Hubbard T, Chothia C: SCOP: A structural
classification of proteins database for the investigation of sequences
and structures. J Mol Biol 1995, 247:536.
77. Wilson D, Pethica R, Zhou Y, Talbot C, Vogel C, Madera M, Chothia C,
Gough J: SUPERFAMILY–sophisticated comparative genomics, data
mining, visualization and phylogeny. Nucl Acids Res 2009, 37:D380-386.
78. Gough J, Chothia C: SUPERFAMILY: HMMs representing all proteins of
known structure. SCOP sequence searches, alignments and genome
assignments. Nucl Acids Res 2002, 30:268-272.
79. Wilson D, Madera M, Vogel C, Chothia C, Gough J: The SUPERFAMILY
database in 2007: families and functions. Nucl Acids Res 2007, 35:
D308-313.
Seufferheld et al. Biology Direct 2011, 6:50
http://www.biology-direct.com/content/6/1/50
Page 15 of 15
80. Sowers KR, Ferry JG: Isolation and characterization of a methylotrophic
marine methanogen, Methanococcoides-Methylutens Gen-Nov, Sp-Nov.
Appl Environ Microbiol 1983, 45:684-690.
81. Caetano-Anollés G, Mittenthal JE: Exploring the interplay of stability and
function in protein evolution. Bioessays 2010, 32:655-658.
82. Kim KM, Caetano-Anollés G: The proteomic complexity and rise of the
primordial ancestor of diversified life. BMC Evol Biol 2011, 11:140.
83. Ingber DE: The origin of cellular life. Bioessays 2000, 22:1100-1170.
84. Morris CE: How did cells get their size? Anat Rec 2002, 268:239-251.
85. Hanczyc MM, Fujikawa SM, Szostak JW: Experimental models of primitive
cellular compartments: encapsulation, growth, and division. Science 2003,
302:618-622.
doi:10.1186/1745-6150-6-50
Cite this article as: Seufferheld et al.: Evolution of vacuolar proton
pyrophosphatase domains and volutin granules: clues into the early
evolutionary origin of the acidocalcisome. Biology Direct 2011 6:50.
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit