Ancien sal e n me agenomics: acking changes in
mic obes and hei i uses om he unde g ound o he
su ace
MªDolo es Ramos-Ba be o,
1
Tomeu Vi e ,
2
Ane Zabale a,
3
Ece Senel,
1,4
Ma ía Goma iz,
1
Iñaki An igüedad,
3
Fe nando San os ,
1
Manuel Ma ínez-Ga cía ,
1
Ramon Rossell
o-M
o a
2
and Jose a An
on
1
*
1
Depa men o Physiology, Gene ics and Mic obiology,
Uni e si y o Alican e, 03690 San Vicen del Raspeig,
Alican e, Spain.
2
Ma ine Mic obiology G oup, Depa men o Animal and
Mic obial Di e si y, Medi e anean Ins i u e o Ad anced
S udies (IMEDEA; CSIC-UIB), Espo les, Illes Balea s,
07190, Spain.
3
Hyd o-En i onmen al P ocesses G oup, Geology
Depa men , Science and Technology Facul y,
Uni e si y o he Basque Coun y UPV/EHU, Leioa,
48940, Spain.
4
Depa men o Biology, Ins i u e o G adua e P og ams,
Eskisehi Technical Uni e si y, Yunusem e Campus,
Eskisehi , 26470, Tu key.
Summa y
Mic obial communi ies in hype saline unde g ound
wa e s de i e om ancien o ganisms apped wi hin
he e apo i ic sal c ys als and a e pa o he poo ly
known sub e anean biosphe e. He e, we cha ac e -
ized he i al and p oka yo ic assemblages p esen in
he hype saline sp ings ha dissol e T iassic-Keupe
e apo i e ocks and eed he Añana Sal Valley
(A aba/Ala a, Basque Coun y, Spain). Fou unde -
g ound wa e samples (a ound 23% o al salini y)
wi h di e en le els o exposu e o he open ai we e
analysed by means o mic oscopy and me a-
genomics. Cells and i uses in he sp ing wa e had
lowe concen a ions han wha a e no mally ound in
hype saline en i onmen s and seemed o be mos ly
inac i e. Upon exposu e o he open ai , he e was an
inc ease in ac i i y o bo h cells and i uses as well
as a selec ion o phylo ypes. The unde g ound wa e
was inhabi ed by a ich communi y ha bou ing a
di e se se o genes coding o e inal binding p o-
eins. A o al o 35 i al con igs om 15 o 104 kb,
ep esen ing pa ial o o al i al genomes, we e
assembled and hei e olu iona y changes h ough
he sp ing sys em we e ollowed by SNP analysis
and me agenomic island acking. O e all, bo h he
i al and he p oka yo ic assemblages changed
quickly upon exposu e o he open ai condi ions.
In oduc ion
The s udy o mic obial communi ies p esen in hype -
saline sp ings has he po en ial o un eil he di e si y o
li e in sub e anean biomes. This unde g ound wo ld
accoun s o up o one-fi h o Ea h’s biomass bu cons i-
u es only a ound 8% o 16S RNA genes in public da a-
bases (Anan ha aman e al., 2016). Hype halophilic
unde g ound communi ies likely de i e om ancien
o ganisms apped wi hin he e apo i ic sal c ys als
(V eeland e al., 2000; Baa i e al., 2010). When he
hype saline sp ing wa e is used o eed sola sal e ns,
he sys em can be used o ack he changes expe i-
enced by he unde g ound communi y when fi s exposed
o open-ai and sunligh i adia ion condi ions. This moni-
o ing is especially in e es ing in he case o ex eme
en i onmen s, in which ecological and e olu iona y
empos o e lap (I anzo e al., 2017, 2019). In ac , mic o-
bial communi ies ha e been p oposed o e ol e as e in
ex eme han in ‘mild’en i onmen s (Li e al., 2014).
Finally, gi en he high concen a ions o i uses no mally
ound in hype saline en i onmen s and he ole o i uses
in con olling hei mic obial hos s (Guixa-Boixa eu
e al., 1996), s udies o mic obial di e si y and e olu ion
in ex eme en i onmen s should be ex ended o encom-
pass he i al ac ion.
Howe e , despi e he unique ecology o unde g ound
sys ems, only a ew hype saline unde g ound sp ings
ha e been s udied. Examples o such en i onmen s a e
he Andean hype saline wa e s eeding Ma as sal e ns
(Ma u ano e al., 2006) o An a c ic hype saline
Recei ed 7 Ap il, 2021; e ised 26 May, 2021; accep ed 6 June,
2021. *Fo co espondence. E-mail [email p o ec ed]; Tel.
+34-965903870; Fax. +34 965909569.
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s
use and dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modifica ions o
adap a ions a e made.
En i onmen al Mic obiology (2021) 23(7), 3477–3498 doi:10.1111/1462-2920.15630
oligo ophic sp ings (Colangelo-Lillis e al., 2016). He e,
we ha e cha ac e ized he mic obial and i al assem-
blages o he Añana Ancien Sal Valley hype saline
sp ings.
The Añana Sal Valley (A aba/Ala a, Basque Coun y,
Spain) consis s o se e al saline wa e sp ings. This
wa e has been used o p oduce sal con inuously o a
leas he las 7000 yea s (Pla a and E kiaga, 2018). The
sal alley lies on he no he n edge o an ac i e diapi ic
s uc u e (19 km
2
), whe e T iassic-Keupe e apo i e
ocks (gypsum and hali e) slowly ascend (F anko ic
e al., 2016) cu ing h ough Lowe C e aceous-Te ia y
ocks, and inco po a e ca bona e and ophi ic ocks wi hin
a clayey mass. Consequen ly, he hyd ogeological con-
ex o he diapi and, in pa icula , o he sal alley is
qui e complex. The sal alley is he main discha ge a ea
o he diapi ’s g oundwa e . The e a e wo ypes o wa e
in he sp ings o he alley, saline wa e and b ackish
wa e .
The saline wa e , which eeds he sal p oduc ion c ys-
allize s, is d ained by fi e sp ings, o which he h ee
mos impo an (Hon ana, El Cau i o and San a Eng acia)
a e he subjec o his a icle; in he la e wo, he ou le
o he wa e is condi ioned as a pool (Fig. 1), so ha he
wa e does no flow di ec ly o he ou side. Thei com-
bined discha ge is abou 2.5 L s
1
, emaining e y con-
s an h oughou he yea , indica i e o he aqui e ’s high
egula ing capaci y. All he saline sp ings ha e p ac ically
iden ical physico-chemical cha ac e is ics, indica ing he
same deep o igin (260 m, acco ding o I iba and
Abalos, 2011) and he same access ou e o he su ace.
The elec ical conduc i i y o hese wa e s is 180–
220 mS cm
1
(wi h e y li le fluc ua ion and no annual
cyclici y), and he empe a u e is 15–18 C (a e age ai
empe a u e, 11 C), all indica i e o deep and slow flows
(as also indica ed by T i ium alues). The wa e s a e high
in chlo ide-sodium (Cl: 140–160 g L
1
) bu wi h a signifi-
can p esence o sul a es (4–4.8 g L
1
) (unpublished da a
p o ided by he Hyd o-En i onmen al P ocesses G oup
o he Uni e si y o he Basque Coun y).
Ou wo k using samples om unde g ound saline
wa e s showed ha hese wa e s ha bou a di e se com-
muni y o p oka yo es and i uses, some o which we e
selec ed upon exposu e o open ai condi ions, which
likely igge ed he ac i i ies o bac e ial cells and hei
i uses.
Fig 1. Samples analysed in his wo k.
B inesampleswe e akendi ec ly om
he sp ing (HU), a pool ed om abo e
wi h HU wa e (HP) and wo pools (SE,
EC) ed om he bo om; all he poin s
wi h unde g ound wa e had he same
o igin. The olume o he wa e in he
pools is kep cons an by allowing he
wa e ou flow by ou le s (as e isks) con-
nec ed o c ys allize ponds h ough a
se ies o pipes. F: flux, V: olume, RT:
calcula ed esidence imes. Sub-
indexes e e o he names o he sam-
pling poin s. [Colo figu e can be
iewed a wileyonlinelib a y.com]
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3478 M. D. Ramos-Ba be o e al.
Resul s and discussion
The sys em and gene al mic obial ea u es
Fou wa e samples we e aken o his s udy: h ee om
he sp ings La Hon ana (HU, ‘Hon ana upwelling’), San a
Eng acia (SE) and El Cau i o (EC), and one om he
shallow puddle ed by he unde g ound wa e eme ging
a HU, which in u n eeds some c ys allize ponds used
o sal p oduc ion (Fig. 1). Thus, he same unde g ound
b ine eme ging a HU also eme ges a he bo om o SE
and EC pools whe e i is empo a ily s o ed and used o
eeding he c ys allize s. As shown in Fig. 1, hese pools’
walls a e co e ed wi h conc e e and isola ed om soil.
B ine inpu and ou pu flows a e kep equal, so he ol-
umes o hese pools emain cons an . Volumes, flow
a es and calcula ed esidence imes a e shown in Fig. 1.
No e ha , om a hyd ogeological poin o iew, he h ee
sp ings ha e he same ‘unde g ound’o igin and he
men ion o ‘su ace’o ‘open ai ’in his wo k e e s only
o he exposu e o hese wa e s o he condi ions o he
pools whe e hey we e empo a ily s o ed.
O e all, esidence imes in he wa e ese oi s a e
a he sho ( om less han 1 h in he La Hon ana puddle
o a ound 3 h in EC and SE pools). Howe e , hese a e
a e age alues and small amoun s o cells o i uses
may emain o longe imes in he ese oi s. No e ha :
(i) HP is di ec ly exposed o he en i onmen (bo h ai
and clay soil) since i is ed om he sou ce HU; (ii) HU
has ne e been exposed o he ex e io condi ions as i s
wa e s flow di ec ly om he unde g ound sp ing; and
(iii) SE and EC pools, which a e ed om he bo om, con-
ain a mix u e o esh and s o ed ( o an a e age o
ca. 3 h) g oundwa e . The e o e, wa e in SE and EC has
been less exposed o he open ai /su ace condi ions
han ha om HP. As shown in Table 1 he main compo-
nen o he ou analysed b ines is NaCl, in good ag ee-
men wi h p e ious esul s (see abo e).
To al cell and CARD-FISH coun s wi h gene al p obes
o Bac e ia and A chaea we e de e mined o he ou
samples (Supplemen a y Fig. S1; Table 1). O e all, he
analysed samples, excep HP, ha bou ed a ound
10
6
cells ml
1
, which was lowe han wha is no mally
ound in coas al hype saline wa e s (San os e al., 2012;
Boujelben e al., 2012b; Di Meglio e al., 2016; Roux
e al., 2016). La Hon ana sp ing wa e (HU), eshly
eme ged om he unde g ound, ha bou ed 2.68 10
6
cells ml
1
, which is g ea e han he densi y o cells in an
inland hype saline sp ing eeding he Pe u ian sal e ns o
Ma as, which had a ound 100 cells ml
1
(Ma u ano
e al., 2006), a ac ha may be ela ed o he ela i ely
high empe a u e o Añana unde g ound saline wa e ,
which is conside ed o be highe han ha eco ded a
he eme ging poin s (I iba and
Abalos, 2011). No unex-
pec edly, he ac ion o cells de ec ed wi h CARD-FISH
was e y low in he HU sample and inc eased when he
unde g ound b ine was exposed o he open ai condi-
ions. Howe e , he numbe o cells de ec ed by CARD-
FISH ne e su passed 50% o he o al cell coun s. The
Table 1. Cha ac e is ics o he samples om Añana Ancien Sal Valley used in his wo k.
Samples
Pa ame e s La Hon ana upwelling (HU) La Hon ana pond (HP) San a Eng acia (SE) El Cau i o (EC)
pH 7.34 6.68 6.68 6.69
%Sal
a
24 23 23.8 23
Cl
156.98 163.60 156.59 150.19
B
0.10 0.09 0.09 0.09
NO
3
0.04 0.04 0.04 0.03
SO
42
4.36 4.48 4.32 4.23
Na
+
100.09 95.92 96.04 92.18
K
+
0.47 0.45 0.44 0.43
Mg
2+
0.81 0.37 0.25 0.28
Ca
2+
1.19 1.29 1.55 1.35
VLP ml
1b
7.00 10
4
2.16 10
4
2.09 10
6
9.30 10
5
4.40 10
5
1.73 10
5
2.24 10
5
1.01 10
5
A chaea ml
1c
1.60 10
5
1.26 10
5
1.23 10
7
2.10 10
6
2.10 10
6
5.77 10
5
1.18 10
6
4.05 10
5
Bac e ia ml
1c
3.20 10
5
2.53 10
5
4.80 10
6
2.00 10
5
1.30 10
6
3.46 10
5
1.06 10
6
9.85 10
5
Cells ml
1d
2.68 10
6
8.27 10
5
2.57 10
7
1.96 10
6
6.75 10
6
3.00 10
5
3.43 10
6
1.72 10
6
% A chaea
c
6.0 47.9 31.1 34.4
% Bac e ia
c
11.9 18.7 19.3 30.9
%ND
e
82.1 33.5 49.6 34.7
VLP/cell 0.03 0.08 0.07 0.07
a
Measu ed in si u wi h a hand e ac ome e .
b
Epifluo escence coun s ollowing Syb -Gold s aining.
c
CARD-FISH.
d
DAPI.
e
Pe cen age o cells no de ec ed by FISH wi h A chaea/Bac e ia p obes. Concen a ions o ions a e gi en in g L
1
.
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3479
numbe o 16S RNA molecules (i.e. ibosomes) pe cell
equi ed o de ec a bac e ial cell by CARD-FISH is
a ound a ew ens, depending on he sample condi ions
(Hoshino e al., 2008), which is e y low e en o ac i e
bac e ia which equen ly ha bou mo e ibosomes
(Amann and Fuchs, 2008). The e o e, he low de ec abil-
i y le els in he samples could indica e ha a la ge p o-
po ion o he cells migh be inac i e, do man o dead,
especially o he eme ging b ine (HU).
The numbe o i uses (coun ed as i us-like pa icles,
VLPs) was ema kably low in he HU, wi h a i us o cell
a io o 0.03. This a io is ex emely low compa ed o
abo eg ound high sal sys ems, which no mally ha bou
om 10 o up o hund eds o i uses pe cell (Di Meglio
e al., 2016). In he case o A c ic hype saline sp ing
b ines, he a io is lowe han 10 and i uses always ou -
numbe mic obes (Colangelo-Lillis e al., 2016). Howe e ,
in o he unde g ound eshwa e s, his a io is a ound 0.1,
wi h a ound 10
4
VLPs ml
1
(Roudnew e al., 2013; Pa -
ikka e al., 2017). This could again be ela ed o he low
me abolic ac i i y o do mancy o he a ailable hos s,
since i uses ely on hei hos ene gy esou ces o hei
eplica ion (Mahmoudabadi e al., 2017). Acco dingly,
i al pa icle concen a ions in HP, SE and EC inc eased
compa ed o HU. As wi h he cells, HP p esen ed he
highes VLP concen a ion while SE and EC, filled wi h a
mix o eshly upwelled and s o ed wa e , p esen ed in e -
media e concen a ions.
Me agenomic analysis o cell and i us assemblages
O e all cha ac e is ics o i al and cellula me agenomes
The mic obial (cells and i uses) communi ies in he ou
Añana wa e samples we e analysed by me agenome
sequencing. Cha ac e is ics o he me agenomes a e
shown in Supplemen a y Table S1. All me agenomes
could be conside ed o good quali y acco ding o he
c i e ia o Rod iguez-R and Kons an inidis (2014), wi h
sequencing co e ages abo e 60% as calcula ed by Non-
pa eil (Rod iguez-R and Kons an inidis, 2014). The i al
me agenomes had a e y low 16S RNA gene signal
(below 0.02% o 16S RNA eads) and hus cellula con-
amina ion could be conside ed negligible (Roux
e al., 2013; Enaul e al., 2017).
P io o assembly and anno a ion, unassembled eads
om all me agenomes we e compa ed ‘all e sus all’,in
o de o p o ide an o e all pic u e o changes ha he
unde g ound i al and cell communi ies unde wen when
exposed o he ai . In addi ion, compa isons using
unassembled da a o e come biases in oduced by me a-
genomic assembly since in many cases assembly does
no cap u e he whole ex an di e si y o cell and i us
popula ions in a sample (Ma inez-He nandez
e al., 2017). Changes expe ienced by he cellula and
i al assemblages upon eme ging om he b ine we e
analysed by calcula ing he pe cen age o sha ed
sequences (Fig. 2A and B) as well as he pai wise o e all
ANIs among he co esponding me agenomes (Fig. 2C).
O e all, bo h conside ing he sequences sha ed and he
simila i y o he sha ed sequences, he cellula assem-
blage seemed o change mo e han he i al ac ion. As
shown in Fig. 2B, SE and EC we e he closes cell
me agenomes, as could be expec ed om hei simila
en i onmen al condi ions and in ag eemen wi h 16S
RNA genes and MAG analyses (see below), whe eas
HP and HU we e he closes i al me agenomes.
As also shown in Fig. 2D, be ween 1.4% and 3.5% o
he i al me agenome eads we e also p esen in he
co esponding cell me agenomes. Al hough his i al sig-
nal could also co espond o i uses a ached o he cells
o o p o i uses, i can be aken a p oxy o i uses epli-
ca ing inside cells (Hallam e al., 2004; Ghai e al., 2010;
L
opez-Pé ez e al., 2017) and hus as a p oxy o i al
ac i i y. The highe p esence o i al eads in HP com-
pa ed o HU samples could hus be indica ing ha
i uses a e mo e ac i e in he puddle han in he sp ing
b ine, as u he discussed below.
P oka yo ic di e si y: he unde g ound seed bank
To gain insigh in o he axa p esen in he mic obial com-
muni ies o he samples, eads co esponding o 16S
RNA gene agmen s we e e ie ed om he ou
me agenomes (Supplemen a y Da ase 1). As shown in
Fig. 3A, mo e sequences we e assigned o A chaea
(mainly o he class Halobac e ia) han o Bac e ia,
hough he bac e ial assemblage was mo e di e se; bo h
o hese esul s a e consis en wi h o he s udies o
hype saline en i onmen s (An
on e al., 2002; Bake
e al., 2010; Ven osa e al., 2015; Mo a-Ruiz e al., 2018).
BLAST analyses (da a no shown) o ope a ional phylo-
gene ic uni s (OPUs), accoun ing o >1% o he
sequences in each sample, indica ed ha he OPUs we e
closely ela ed o sequences e ie ed om o he hype -
saline en i onmen s wo ldwide, ei he e es ial o
aqua ic. Ve y ecen ly, a new species o Alpha-p o e-
obac e ia (Al e e y h obac e mu iae) has been isola ed
om he SE sp ing (Muniozgu en e al., 2021). In ou
da ase his bac e ium co esponded o OPU 114 (Supple-
men a y Da ase 1) ha was de ec ed, albei a e y low
abundance, in HU, as also indica ed by ec ui ing he
eads om he ou cellula me agenomes gene a ed in
his wo k (da a no shown) o A.mu iae genomic
sequence.
In good ag eemen wi h he analyses men ioned abo e
(Fig. 2C), HU was he mos di e gen and mos di e se
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3480 M. D. Ramos-Ba be o e al.
sample acco ding o he composi ion o i s mic obio a
(Fig. 3B). The communi y s uc u e in HU was also di e -
en in e ms o he p esence o dominan phylo ypes,
since HU had ewe phylo ypes ha domina ed. Fu he -
mo e, he eshly eme ged unde g ound wa e
(HU) ha bou ed he highes p opo ion (56%) o sample-
specific OPUs (i.e. OPUs no p esen in he es o he
samples, Fig. 3C). These sample-specific axa could co -
espond o he dead o do man cells ha did no con ib-
u e o he ac i e and i al in ec ion-suscep ible ac ion o
he mic obiome. The subsu ace wa e seemed hus o
be ac ing as a seed bank o OPUs (i.e. phylo ypes) o
he su ace sys em since mos OPUs p esen in he h ee
sampled wa e ese oi s we e also p esen in he sp ing
(Figs 3C and 4). Thus, he e we e ‘unde g ound’and
‘su ace’OPUs, as well as some pool-specific ones,
albei hese we e no mally no e y abundan in he
co esponding me agenomes. As highligh ed in Fig. 4,
‘unde g ound’phylo ypes, which d opped sha ply o
e en disappea ed in he samples exposed o he su ace
condi ions, included OPUs 342, 459, 397, 398,
425, co esponding espec i ely o he candida e phylum
Pa cubac e ia, he Nanohaloa chaeo a,Haloplanus sp.,
Halolamina sp. and Halomic oa cula sp. Supplemen a y
da ase S1 p o ides a de ailed desc ip ion o OPU abun-
dances and changes h ough he Añana pools. Espe-
cially ema kable was he abundance o membe s o he
candida e phylum Pa cubac e ia which seem o be
au och honous o g oundwa e s (He, e al. 2021). Many
o hese HU-specific OPUs could co espond o membe s
o he ancien g oundwa e dissol ing he sal diapi ,
gi en hei low abundances and hei simila i ies o meso-
philic bac e ia (see Supplemen a y da ase 1). In any
case, mos o hese specific OPUs we e p esen a e y
low abundances in HU.
Among he ‘su ace’OPUs he mos ou s anding case
was ha o Salinibac e sp. ( ep esen ed by OPU 218),
which expe ienced a old-change o o e 32 in he pools
compa ed o he upwelling. Finally, he p esence in HU o a
ela i ely high numbe (close o 9%) o OPUs ela ed o he
chlo oplas o he g een algae Dunaliella sp. was in iguing
(Fig. 3A). As expec ed, he p opo ion o his OPU was
highe in he o he pools, whe e Dunaliella could cons i u e
he mos impo an p ima y p oduce (O en, 2014; Kee hi
e al., 2018), eeding he he e o ophic bac e ia, and espon-
sible o he a leas a en- old inc ease o cells in HP in
compa ison wi h he HU. Howe e , he p esence o a
pho o ophic euka yo ic o ganism in he da k unde g ound
Fig 2. A Compa ison o Añana me agenomes.
A. Compa ison ‘all e sus all’among cellula (MG) and i al (MV) me agenomic eads. Numbe s indica e he pe cen age o sha ed sequences.
The di ec ion o he compa ison is om op o he bo om. Fo ins ance, he cellula me agenome o HU sha es 26% o i s sequences wi h he cel-
lula me agenome o HP (*), while he cell me agenome o HP sha es 34% o i s sequences wi h he cellula me agenome o HU (**).
B. ‘All e sus all’PCA (p incipal componen analyses) o Añana i al and cellula me agenomes based on % o eads sha ed be ween
me agenomes.
C. G aph ep esen a ion o ANI (a e age nucleo ide iden i y om sha ed eads) om compa ison all e sus all i omes and cell me agenomes o
Añana sal e n. Only hi s o e 70% o co e age we e conside ed.
D. Vi al signal in cell me agenomes in Añana samples. The pie cha s show he pe cen age o sequences sha ed be ween i al-cellula
me agenome pai s o each sampling poin .
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3481
wa e was a he di ficul o explain and could be ela ed o
ei he a do man s a e eco e ed om he dissolu ion o he
e apo i e o o a po en ial o a mixo ophic li es yle o
Dunaliella unde ce ain en i onmen al condi ions.
Wha e e he case, he p esence o pho o ophic mic obes
in unde g ound sys ems has been epo ed p e iously,
such as o iable cyanobac e ia in deep con inen al sub-
su ace (Reisse , 2007; Puen e-S
anchez e al., 2018).
Fig 3. Cha ac e iza ion o he mic obial communi ies based on 16S RNA eads e ie ed om he cellula me agenomes.
A. Taxonomic dis ibu ion a he genus le el.
B. Communi ies clus e ed using coo dina ed analysis o he weigh ed UniF ac dis ance ma ix.
C. Venn diag ams showing he gene a common o he di e en me agenomes. [Colo figu e can be iewed a wileyonlinelib a y.com]
Fig 4. OPU dis ibu ion in he ou me agenomes (*see Supplemen a y da ase 1), o de ed acco ding o hei abundances in he HU sample.
Some OPUs a e ma ked as examples. HU is shown in ed, HP in blue, SE in yellow and EC in g een. [Colo figu e can be iewed a
wileyonlinelib a y.com]
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3482 M. D. Ramos-Ba be o e al.
The anno a ion o he assembled me agenomes con-
fi med he dominance o A chaea (Supplemen a y
da ase 2) and allowed he e ie al o 18 (almos ) com-
ple e 16S RNA gene sequences ha could be used o a
mo e accu a e phylogene ic analysis (Supplemen a y
Fig. S2). I is wo h men ioning ha some o he 16S
RNA gene sequences eco e ed in his wo k we e o e
99% iden ical o sequences ob ained in a p elimina y
s udy by dena u ing g adien gel elec opho esis (DGGE)
om San a Eng acia sp ing in 2011 (in pu ple in Supple-
men a y Fig. S2 and Supplemen a y da ase 1) indica ing
he pe sis ence o he co esponding species in he
sys em.
Me agenome anno a ion indica ed ha COG dis ibu-
ion was simila among he h ee sp ings and
HP. Howe e , i also confi med ha he communi y o
HU was he mos di e gen o he ou analysed sam-
ples (Supplemen a y Fig. S3 A). These esul s indi-
ca ed ha no only was he phylogene ic di e si y
wide in HU compa ed o he es o samples, bu he e
we e genes ha we e specifically p esen in he eshly
upwelled unde g ound wa e while o he s we e
en iched in he wa e s exposed o he su ace. Among
he me abolic ca ego ies de ec ed in he samples, he
case o e inal binding p o eins (RBP, also known as
hodopsins) was especially ema kable. RBPs
e ie ed om Añana me agenomes (Supplemen a y
Fig. S4) included he h ee ypes no mally ound in
hype saline sys ems: (i) he ou wa d-di ec ed p o on
pumps, such as he a chaeal bac e io hodopsin (BR),
he bac e ial p o eo hodopsin (PR), and Salinibac e
xan ho hodopsin (XR), (ii) he inwa d-di ec ed chlo ide
pump halo hodopsin (HR), and (iii) he senso y ho-
dopsins (SRI and SRII) in ol ed in ligh sensing o
pho o axis (O en, 2002). The p esence o ligh - ela ed
genes in he unde g ound wa e was e y in iguing
(as was he p esence o he g een algae Dunaliella
men ioned abo e), bu e en mo e in iguing was he
high di e si y o RBPs. As shown in Supplemen a y
Fig. S4, RBPs in HU we e no only abundan bu we e
also mo e di e se han in he es o samples. Fu he -
mo e, many o he RBP coding genes we e specific o
HU and we e no ound in he es o samples, excep
o XR- ela ed genes. The lowe p esence o XR in HU
was no unexpec ed since i is p oduced by
Salinibac e spp. (and close ela i es) which we e p e-
sen only in e y low le els in HU. O e all, he unde -
g ound wa e ha bou ed di e se no el RBP coding
genes as indica ed by hei dis ances om p e iously
desc ibed hodopsins. The sp ing is hus also ac ing
as an RBP seed bank (Baa i e al., 2010). We can
specula e ha his seed bank o igina ed in he ancien
communi y apped in he unde g ound e apo i e
ocks.
Reco e y o mic obial genomes om me agenomes:
changes in he unde g ound communi y when exposed
o he su ace condi ions
A e 16S RNA gene ead eco e y and me agenome
anno a ion, he nex s ep was he assembly and analysis
o mic obial genomes o me agenome-assembled
genomes (MAGs). A o al o nine MAGs o di e en qual-
i y (acco ding o Kons an inidis e al., 2017) we e eco -
e ed by co-assembly o he ou me agenomes (wi h he
excep ion o MAG A2_MAG005, which was eco e ed
om EC). Taken oge he , he MAGs did no cons i u e a
dominan p opo ion o he o al me agenomic sequences
( om 1.96% o he eads in HU o 18.1%, 17.8% and
23.03% in HP, SE and EC espec i ely) bu we e s ill
use ul o acking genome dynamics ac oss he sys em.
The main cha ac e is ics o he MAGs a e shown in
Table 2 and Supplemen a y Fig. S5. All MAGs ep esen-
ed new species o known gene a o o unclassified
highe axonomic ca ego ies. A close look a he anno a-
ion o he MAGs (Supplemen a y da ase 3) indica ed
in e es ing ea u es o ecological ele ance such as he
p esence o many genes in ol ed in de oxifica ion o
a sena e o o he me als, capabili y o anae obic ni a e
espi a ion, use o phosphona es, p esence o di e en
ypes o RBPs (see abo e), among o he ai s.
In an a emp o ack he communi y changes associ-
a ed wi h he exposu e o sp ing wa e o su ace condi-
ions, we ec ui ed me agenome eads o he MAGs in
he di e en pools and calcula ed ANI a 95% iden i y as
a p oxy o he o e all in a-popula ion di e si y o he
MAG- ela ed genomes (S a nawski e al., 2017). Co e -
age, abundance and single nucleo ide polymo phism
(SNP) a es we e also calcula ed (Table 2).
O e all, excep o MAG021 ha was specific oSE, he
beha iou o he MAGs in he wo pools SE and EC was
simila , indica ing ha he hyd ogeologic con ex o he sys-
em exe ed some de e minis ic e ec s on he popula ions
ep esen ed by he MAGs. Mos MAGs seemed especially
success ul in he wa e bodies exposed o he su ace. As
expec ed, he MAG co esponding o a new genus wi hin
he Salinibac e aceae was especially abundan in he HP,
SE and EC samples, whe e hey eached densi ies abo e
3.8 10
4
cells ml
1
(calcula ed as in S a nawski e al.
2017), as shown in Table 2.
Quan i ying he numbe o SNPs pe Mb showed ha
some popula ions (MAG002 and MAG015) seemed o
ha e di e sified upon exposu e o su ace condi ions,
whe eas o he popula ions we e mo e homogeneous
despi e hei inc eased abundances (e.g. MAG005). The
high homogenei y o he popula ion ep esen ed by
MAG013 (Ec o hio hospi aceae), which was p esen in
he HP, SE and EC (see also Supplemen a y Fig. S6),
was demons a ed by his MAG ha ing ze o SNPs in i s
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3483
Table 2. S a is ics o MAGs eco e ed om cellula me agenomes.
MAG002 (A)
a
MAG003 (A) MAG004 (A) MAG005 (A) MAG013 (B) MAG015 (A) MAG021 (B)
a
MAG022 (A) A2_MAG005 (B)
No. con igs 233 379 249 516 575 605 776 632 974
No. bases 2 329 966 3 073 719 1 117 659 2 039 487 2 226 472 1 323 474 2 544 609 2 774 985 2 272 857
%GC 64.5 66.9 62.7 62 64.7 65 56.5 66.4 65.3
Comple eness 100 92.99 61.5 50 85.8 23.1 82.1 53.8 34.9
Con amina ion 3.8 5.51 0 3.8 4.7 0 1.9 0 3.8
Closes ela i e
genome by
MiGA
Na onomonas
moolapensis
Haloa cula sp. JP
L23 NZ
CP050014
Halo ub um sp.
PV6 NZ
CP030064
Halo ub um
lacusp o undi
ATCC 49239 NC
Spi ibac e sp.
2438 NZ
CP046046
Haloa cula
hispanica
ATCC 33960
Thiohalobac e
hiocyana icus NZ
AP018052
Haloa cula sp. JP
L23 NZ
CP050014
Salini enus lu ea
NZ
NCRC00000000
ANI 83.6 ND ND ND ND ND ND ND ND
AAI 81.0 74.8 71.7 63.0 52.4 54.6 48.6 74.9 79.9
Taxonomy Na onomonas sp.
new species
Haloa cula sp.
new species
Halobac e iaceae
new genus
Halo ub aceae new
genus
Ec o hio hodospi-
aceae new
genus
Halobac e iales
new amily
γ-P o eobac e ia new
o de
Haloa culaceae
new genus
Salinibac e aceae
new genus
Abundance
in HU
b
0.08%, 2.14 10
3
0.05%, 1.3410
3
0.13%, 3.48 10
3
0.23%, 6.16 10
3
0.01%,
2.68 10
2
0.45%, 1.21 10
4
0.01%, 2.68 10
2
0.07%, 1.88 10
3
0.004%, 1.07 10
2
Abundance
in HP
b
0.81%, 2.08 10
5
0.52%, 1.34 10
5
1.90%, 4.88 10
5
1%, 2.57 10
5
0.52%,
1.34 10
5
0.45%, 1.16 10
5
0.02%, 5.14 10
3
2.3%, 5.91 10
5
0.15%, 3.86 10
4
Abundance
in SE
b
2.20%, 1.49 10
5
0.98%, 6.62 10
4
1.00%, 6.75 10
4
1%, 6.75 10
4
0.41%,
2.77 10
4
0.26%, 1.76 10
4
0.95%, 6.41 10
4
0.02%, 1.35 10
3
0.8%, 5.40 10
4
Abundance
in EC
b
2.67%, 9.16 10
4
1.91%, 6.55 10
4
1.89%, 6.48 10
4
1.4%, 4.80 10
4
0.42%,
1.44 10
4
0.29%, 9.95 10
3
0.03%, 1.03 10
3
0.02% 7.55 10
2
1.33%, 4.56 10
4
HU
c
99.3, 0.7x, 9 97.6, 0.5x 39.6 99.0, 0.9x, 6.8 98.8, 1.7x, 271.6 99.4, 0.1x, 0 99.1, 3.3x, 5.2 99.1, 0.59x, 0 98.5, 0.59X,19 98.6, 0.03X, 0
HP
c
99.4, 3.0x, 15.4 99.1, 1.9x, 11 99.1, 6.9x, 9.7 99.1, 4.6x, 70.1 99.7, 2.0x, 0 99.1, 1.9x, 7.5 99.1, 0.09x, 0 98.5, 9.9X, 1 97.9, 0.6X, 0
SE
c
99.5, 17.9x, 207.2 99.5, 6.9x, 13 99.1, 6.7x, 20.9 99.2, 8.5x, 126.9 99.5, 3.1x, 0 99.0, 2.1x, 98.9 99.3, 7.3x, 13.36 96.4, 0.18X, 41 99.3, 6.4X, 2.1
EC
c
99.4, 11.8x, 133.9 99.5, 8.3x, 6.1 99.1, 8.0x, 27.2 99.2, 7.1x, 54.9 99.5, 2.1x, 0 98.9, 1.4x, 68.7 99.2, 0.15x, 0 96.1, 0.15X, 24.5 99.3, 6.5X, 7
All MAGs we e e ie ed om co-assembly o he ou me agenomes, excep o A2_MAG005, which was e ie ed om EC.
ND: no de e mined.
a
(A): A chaea; (B): Bac e ia.
b
Abundance in % nucleo ides ec ui ed/Mb ollowed by he calcula ed popula ion size in cells ml
1
(as in S a nawski e al. 2017).
c
Fo each o he ou ponds he ollowing pa ame e s a e p o ided in his o de : ANI , co e age o he bin, and SNPs Mb
1
o bin.
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3484 M. D. Ramos-Ba be o e al.
2.2 Mb genome, which may indica e ha his bac e ium
was a ecen colonize o he pools (Mezi i e al., 2019).
Vi al di e si y
Taxonomic composi ion o me agenomes om he ee
i us assemblages (Supplemen a y Figs S7–S11) indi-
ca ed ha hese communi ies ha bou ed i uses ela ed
o hose iden ified in o he hype saline en i onmen s,
al hough wi h a conside able le el o no el y since only
3% o i al con igs could be assigned o known i us
g oups. The compa ison o he unc ional ca ego ies in
he ou i al me agenomes (Supplemen a y Fig. S3B)
did no show clea di e ences among hem. Howe e ,
he low p opo ion o anno a ed genes ( om 10% o 19%,
see Supplemen a y Table S1) in he ou i al assem-
blages hampe s any comp ehensi e compa ison.
The assembled ac ion ep esen ed om 18% o 36%
o he i al me agenome o al eads (Table S1). This con-
s i u ed a conside able ac ion o he i al assemblage,
compa ed o o he i al me agenomes (Hayes
e al., 2017; Villamo e al., 2017). A e he assembly o
he ou i al me agenomes, he non- edundan con igs
la ge han 15 kb we e selec ed o an in-dep h analysis.
A o al o such 30 con igs ( om 15 056 o 104 118 kb,
a e age 30 329, see Table 3) we e conside ed bona fide
i al genomes based on he c i e ia desc ibed p e iously
(Roux e al., 2015, Cou inho e al., 2019a; Cou inho
e al., 2019; Sil ei a e al., 2020). The p esence o hese
i al genomes in he di e en pools was analysed by con-
ig ec ui men as shown in Fig. 5A (see Supplemen a y
Table S2 o abundances). Only ou bona fide i al
genomes we e ound in all samples (see Supplemen a y
Fig. S12 o hei anno a ion). In mos cases, i al
genomes, al hough p esen in di e en samples, we e
assembled only om one o hem. This ex ensi ely
epo ed ‘me agenomics anomaly’(Ma inez-He nandez
e al., 2017; Ramos-Ba be o e al., 2019) is likely ela ed
o he mic odi e si y o i al genomes, as discussed
below.
Is he i al ac ion in he unde g ound wa e in ec i e?
The concen a ion o i uses in he HU wa e was e y
low, likely oo low o allow o in ec ion o hos cells. A
ough es ima e o in ec ion a es in HU, calcula ed as in
Colangelo-Lillis e al. (2016), yields as li le as 0.0001
con ac s pe mic obe pe day. Fu he mo e, he ac i i y o
he cell ac ion, as shown by CARD-FISH, seemed also
e y low, which could also hampe i al p opaga ion.
Indeed, none o he i al genomes assembled om he
ou i al me agenomes we e de ec ed in he HU cellula
me agenome, as shown in Fig. 5B, which indica ed ha
likely hese i uses we e no eplica ing inside HU cells.
This in e ence assumes ha he p esence o i al
genomes in he cell me agenomes can be aken, wi h
some limi a ions (i.e. p esence o p ophages, i us
a ached o he cells, e c.), as a p oxy o ac i e i al epli-
ca ion inside he cells. All hese esul s ein o ce he idea
ha some o he cells and i uses eco e ed in HU could
ha e an ancien o igin and we e eleased by he dissolu-
ion o he sal c ys als o he e apo i ic ock p ima y
sou ce o he b ines.
The ques ion is hus whe he he VLPs obse ed in HU
co esponded o in ec i e pa icles o we e jus emnan s
o ancien in ec ion e en s which we e p ese ed in he
b ine o we e dissol ed om he sal c us . In o de o
in es iga e his poin , we can assume ha in ec i e
i uses, once gi en he oppo uni y o find hei hos s, will
eplica e wi hin he hos cy oplasm. Thus, he sea ch o
HU i uses in he cellula me agenomes o he Añana
sys em would un eil whe he o no hey we e eplica ing.
As shown in Fig. 5B, some o he i uses om HU could
be de ec ed by con ig ec ui men in cellula
me agenomes o HP, poin ing o he in ec i i y o he
i uses p esen in he unde g ound b ines.
To u he explo e his possibili y, he eads sha ed
be ween he HU i al me agenome and he cellula
me agenomes om he es o he pools we e ex ac ed
and assembled, yielding a o al o 10 i al con igs la ge
han 10 kb (Supplemen a y Table S3). We will e e o
hem as ‘ a ge ed eplica ing’HU i al genomes. The
i al na u e o hese con igs was confi med and hei
pu a i e hos s de e mined. As shown in Fig. 6, hese
‘ eplica ing’ i al genomes we e abundan in he ee i al
ac ion o HU (excep o i al genome h), despi e hei
low abundances in he cell ac ion. In con as , mos o
hese i uses we e p esen in he cellula me agenomes
o ‘su ace’samples, in abundances anging om 0.01%
o 0.12% o me agenome nucleo ides (Supplemen a y
Table S2). The p esence o HU i uses eplica ing inside
HP cells was especially ema kable, since i indica es
ha he ac i a ion o he co esponding i al pa icles was
qui e as (no ice ha HU wa e alls di ec ly on HP).
Taken oge he , hese esul s sugges ha i al pa icles
p esen in he unde g ound wa e e y likely co -
esponded o in ec i e i us.
Dynamics o he i al communi ies: e olu ion in ac ion
Gi en he hyd ogeology o he sys em, we can assume
ha i al genomes p esen bo h in HU and an addi ional
sample came om he unde g ound b ines. Thus, ack-
ing he changes in i al genomes om HU o a su ace
pool can shed ligh on he empo and mode o i al e olu-
ion. A possible app oach o moni o e olu ion in ac ion
would be o ack SNP changes in he co e genome o
hese i al popula ions. Fu he mo e, he s udy o he i al
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3485
The mo Scien ific) and 50 ng o samples we e Sange
sequenced using he 907R p ime .
Sequencing, ead imming and cleaning, ead joining,
communi y co e age and assembly
Sequencing o cell and i al DNAs was pe o med
using an Illumina Mi-seq Nex e a XT 300 2bp
pai ed-end un. Pai ed-end eads we e joined using
Fas q-join om he ea ools sui e (A ones y, 2011) o
cell me agenomes, and q2 a om IDBA 1.1.1 assem-
ble o he i al me agenomes (Peng e al., 2012).
Reads we e quali y assessed and immed using
PRINSEQ so wa e (Schmiede and Edwa ds, 2011).
Reads sho e han 50 bp and wi h a quali y lowe han
20 we e disca ded. Addi ionally, he i al me agenome
eads we e cleaned o con aminan sequences (Sal e
e al., 2014) (including sequences om human-
associa ed bac e ia, euka yo es and humans); con ami-
nan de ec ion was pe o med using Diamond BLASTx
(e- alue 0.0001 (Buchfink e al., 2014) and Kaiju (que y
co e age >70%) (Menzel e al., 2016) bo h agains he
NCBI n da abase ( eleased on 01/01/2020). The mos
abundan sequence con aminan co esponded o
P opionibac e ium acnes, one o he mos common
Illumina ki con aminan sequences (Sal e
e al., 2014). Due o he high pe cen age o con amina-
ion om P opionibac e ium spp.( om0.1% o5%o
he o al eads), i ome da a we e cleaned using wo
s eps. Fi s , all i ome eads we e ec ui ed sepa a ely
agains he P opionibac e ium acnes NC008065.1
genome and ma ching eads we e emo ed (que y co -
e age >70% ID >90%). Second, as p e iously men-
ioned, con aminan sequences we e also de ec ed by
DIAMOND (Buchfink e al., 2014) and Kaiju (Menzel
e al., 2016). Finally, all con aminan sequences we e
ex ac ed ( om 1% o 19% o o al eads) using Fas A.
fil e .pl om he en eomics package (Rod iguez-R and
Kons an inidis, 2016) o ge he clean i omes used in
he s udy. The Nonpa eil ool (Rod iguez-R and
Kons an inidis, 2014) was used o es ima e he co e -
age o he communi y in each me agenome da ase
wi h de aul pa ame e s. De no o assemblies o
immed eads we e gene a ed using he IDBA 1.1.1
assemble (Peng e al., 2012) wi h he ‘-p e_co ec ion’
op ion. In o de o eco e addi ional comple e i al
genomes om he assembly sequences, con igs we e
ex ended by he ‘map o e e ence’op ion (minimum
ma ch iden i y 99%) o he Genious 6.1.8 pla o m
(Kea se e al., 2012) acco ding o he ecommenda ion
o Villamo e al. (2017)..
Classifica ion o ‘bona fide’ i al sequences in i al
me agenomes
Vi al con igs la ge han 15 kb we e checked o gua an-
ee hei quali y ( hey we e conside ed i al when hey
ha bou ed i al genes and did no align comple ely wi h a
cellula genome). In his wo k, we ha e conside ed as a
bona fide i al genomes hose sequences which
achie ed he ollowing equi emen s: he sequences
should be included in one o he Vi So e ca ego ies
( om 1 o 6) (Roux e al., 2015); hey should ha e a i al
hi by DIAMOND BLASTx agains he NCBI n da abase
(e- alue <0.00001); a leas 8% o he p edic ed genes in
a gi en con ig should be anno a ed as i al genes by
DIAMOND BLASTp; las ly, BLASTn sea ches we e pe -
o med o disca d any con igs ha likely o igina ed om
cells. The edundan i al sequences we e de ec ed by
Cd-hi (cdhi -es -c 0.9 –n 8) (Li and Godzik, 2006) and
omi ed by Fas A.fil e .pl (Rod iguez-R
and Kons an inidis, 2016).
The in silico i us–hos p edic ion was pe o med using
mul iple ools based on BLAST iden ifica ion o i us–hos
homologous ORFs (Cou inho e al., 2017), CRISPR sys-
ems, RNA iden ifica ion by G RNAdb (Chan and
Lowe, 2009) and oligonucleo ide equencies (Edwa ds
e al., 2016). In he se o pla o ms used o he i us–
hos pai iden ifica ion, IMG/V3 (Roux e al., 2020) and
PHISDe ec o (Zhang e al., 2019) we e included. Addi-
ionally, i us–hos pai s we e manually assigned o
each bona fide i al genome by checking gene simila i-
ies and genome syn eny wi h hei pu a i e hos using
he Re Seq non- edundan da abase (Saye s e al., 2016;
Cou inho e al., 2017).
Cell and i us me agenome anno a ion and compa ison
Func ional anno a ion o p edic ed genes om cell and
i us assembled me agenomes was done by JGI (IMG/M
ER: h p://img.jgi.doe.go /me ) (Hun emann e al., 2014)
and DIAMOND BLASTp (Buchfink e al., 2014). Fo i us
me agenomes he NR NCBI da abase was used
(Decembe 2019 upda e); he Unip o da abase
(Ba eman e al., 2015) was used o cell me agenomes.
Raw cell and i al me agenomes ‘all e sus all’compa i-
sons we e pe o med by BLASTn s and-alone using he
same pa ame e s desc ibed below.
Reco e y o d a genomes om cell me agenomes
MaxBin so wa e (Wu e al., 2014) was used o bin d a
genomes om cell me agenomes. Cell con igs la ge
han 1000 bp we e g ouped in bins based on
e anucleo ide equencies and con ig sequencing dep h
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3492 M. D. Ramos-Ba be o e al.
le els. The quali y o he eco e ed bins (comple eness
and con amina ion) was analysed wi h CheckM (Pa ks
e al., 2015) and he HMM.essen ial. b sc ip om
en eomics package (Rod iguez-R and Kons an inidis,
2016), which iden ified lineage-specific ma ke genes in
each bin. Po en ial con amina ion o MAGs was emo ed
ollowing he u o ial a ailable by An io’s ools 5.5 (E en
e al., 2015). P edic ed genes om MAGs we e anno-
a ed agains bo h UniP o da abases (SwissP o and
T EMBL) using he DIAMOND BLASTp ool (Buchfink
e al., 2014). Vi al con igs la ge han 20 000 bp we e
ex ac ed and checked by BLASTn agains he NCBI Re -
e ence Sequence (Re Seq) da abase (h p://www.ncbi.
nlm.nih.go /Re Seq) o be confiden ha hey we e i al
con igs. MAG quali y and axonomic a filia ion we e addi-
ionally checked by he MiGA pla o m (Rod iguez-R
e al., 2018) and Genome axonomy da abase (Chaumeil
e al., 2019).
Abundance o eco e ed d a i al and cell genomes
The abundances o eco e ed d a genomes we e
assessed by agmen ec ui men using he immed
eads o each sample. The eads o cell and i al
me agenomes we e mapped agains he e e ence mic o-
bial ( i al o cell) genomes using s and-alone BLASTn
wi h a cu o o 70% que y co e age, e- alue ≥10
1
and
he ‘bes hi ’op ion. Nex , cell and i al abundances we e
calcula ed conside ing only hi s wi h iden i ies ≥98% (cell)
o 95% ( i us). F agmen ec ui men da a we e plo ed
by using he en eomics.R package in he R s a is ical
ool (Rod iguez-R and Kons an inidis, 2016). MAG di e -
si ies we e calcula ed by ANI (a e age nucleo ide iden-
i y o mapped eads) as p e iously desc ibed (Mezi i
e al., 2019) and SNPs alues we e calcula ed using map
o e e ence (≥95% iden i y, minimum co e age 20and
0.25 o polymo phism equency) in he Genious 6.1.8
pla o m (Kea se e al., 2012). The mu a ion a e was cal-
cula ed as he numbe o SNPs/genome size (Mbp).
T ee econs uc ions based on 16S, 23S and 5S
RNA genes
Gene agmen s om me agenome immed eads ha
con ained he 16S RNA gene we e ex ac ed using
Pa allel-META 2.4 so wa e. Sequences we e g ouped
in ope a ional axonomic uni s (OTUs) a 98.7% simila i y
using UCLUST (Edga and Ba eman, 2010) implemen ed
in QIIME sc ip pick_closed_ e e ence_o us.py (Capo aso
e al., 2010). Fo phylogene ic pu poses, he longes
sequence om each OTU was aligned using he SINA ool
(SILVA Inc emen al Aligne (P uesse e al., 2007) and
we e added o he e e ence da abase SILVA REF123
using he pa simony me hod implemen ed in he ARB
6.0.2 so wa e (Ludwig e al., 2004). The OTUs we e hen
clus e ed in o OPUs (Mo a-Ruiz e al., 2016).
The almos -comple e 16S RNA, 23S RNA and 5S
RNA genes om assembled con igs we e ex ac ed
using RNAmme 1.2 Se e (Lagesen e al., 2007). The
16S and 23S RNA genes we e added by pa simony o
he SILVA SSU and LSU REF 123 da abases espec-
i ely. Sequences we e used o econs uc de no o ees
using he neighbou -joining algo i hm and Jukes-Can o
co ec ion wi h he close ela i e sequences selec ed
om he SILVA da abase.
The ob ained 5S RNA genes we e compa ed using
he online NCBI BLASTn p og am and 5S RNAdb (h p://
combio.pl/ na/) o selec he closes ela i e sequences.
Sequences we e aligned using he MUSCLE 3.8.31
(Edga , 2004) p og am and phylogene ic econs uc ion
was pe o med using he algo i hm neighbou -joining in
ARB 6.0.2 (Ludwig e al., 2004).
Vi al genome di e si y analyses
To calcula e he pe cen age o polymo phisms o i al
genomes, aw eads om each Añana i al me agenome
we e mapped agains he selec ed i al genomes using
he sensi i e local mode o Bow ie2 (Langmead and
Salzbe g, 2012). Then, in o de o ob ain coun s o syn-
onymous and nonsynonymous mu a ions in each ORF,
he ou pu files we e analysed by Di e siTools (h p://
josephhughes.gi hub.io/Di e siTools/). As ecommended
by Cou inho e al. (2019a) and Cou inho e al. (2019), o
es ima e he pe cen age o polymo phic si es and pN/pS
a ios (Schloissnig e al., 2013), only he eads wi h co -
e age equal o highe han 5 we e conside ed, and codon
mu a ions mus ha e been de ec ed a leas ou imes in
a leas 1% o mapped eads. The ORFs included in he
analyses mus be p esen in he ou Añana i al
me agenomes. Only i al genomes wi h a leas eigh
ORFs ha passed he men ioned c i e ia we e included
in he di e si y analyses. The esul ing mu a ion da a
we e used o es ima e he pe cen age o polymo phic
si es and pN/pS a ios, which we e calcula ed as p e i-
ously desc ibed by Schloissnig e al. (2013). The pN and
pS alues equal o ze o we e included in he s udy, con-
sequen ly, we ob ained pN/pS =0 (maximum pu ifica-
ion) and pN/pS =∞(di e sifica ion), and o his eason
he da a we e ans o med by RS udio (h p://www.
s udio.com) o he subsequen analyses. The sequenc-
ing dep h o each i al genome in each i al me agenome
was calcula ed using he BLASTn ou pu fil e by co e -
age (equal o >70%) and iden i y (equal o >95%) using
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3493
Blas Tab.seqde ph.pl om en eomic box (Rod iguez-R
and Kons an inidis, 2016).
S a is ical analysis o i al mic odi e si y analyses
SNPs and pN/pS a ios we e compa ed be ween pai s o
samples using Wilcoxon ank-sum es s implemen ed in
he coin package (Ho ho n e al., 2006) o RS udio (h p://
www. s udio.com). Fi s , he significan di e ences (p-
alue equal o >0.05) be ween samples we e es ed
using he ‘ wo side’op ion, which was hen ollowed by
he ‘less and g ea e ’op ion o asce ain he di ec ion o
he change.
Sequence da a
Raw sequence eads om each me agenome a e pub-
licly a ailable a he eposi o y o Disco e y En i onmen
(Cy e se) h ough he ollowing links: h ps://de.cy e se.
o g/dl/d/25B33A44-98A0-4ABA-B706-B71C1FFEBEDB/
Ramos-Ba be o.zip o cell me agenomes and h ps://
da a.cy e se.o g/da -anon/iplan /home/dolo es/Ramos_
Ba be o_e _al_2021_pa _1/1_Ramos_e _al_2021_Clean_
me a i omes_sequences. a o i al me agenome.
Addi ionally, he bona fide i al genomes and MAGs
a e a ailable in he link h ps://da a.cy e se.o g/da -
anon/iplan /home/dolo es/Ramos_Ba be o_e _al_2021_
pa _2/7_Ramos_e _al_2021_bonafide_ i al_and_MAGs_
sequences. a .
Acknowledgemen s
We would like o hank Añana Sal Valley Founda ion, and
Andoni E kiaga Agi e, i s di ec o a he ime o sampling,
o hei kind help. Thanks o Lei e A ana, Edo a Loma and
Kika Colom o hei help wi h sampling and o Edua do Gon-
z
alez-Pas o o elling us abou he Añana Sal Valley. We
hank Hea he Maughan o he p o essional English edi ing
and he c i ical eading o he manusc ip and Es he Rubio-
Po illo o he help wi h s a is ical analyses. This wo k was
unded by he Spanish Minis y o Science, Inno a ion and
Uni e si ies g an MICROMATES (PGC2018-096956-B-C41
and C44, o J.A./F.S. and R.R.-M.), which was also
suppo ed wi h Eu opean Regional De elopmen Fund
(FEDER) unds, and by he Gene ali a Valenciana g an
PROMETEO/2017/129.
REFERENCES
Amann, R., and Fuchs, B.M. (2008) Single-cell iden ifica ion
in mic obial communi ies by imp o ed fluo escence in si u
hyb idiza ion echniques. Na Re Mic obiol 6: 339–348.
Anan ha aman, K., B own, C.T., Hug, L.A., Sha on, I.,
Cas elle, C.J., P obs , A.J., e al. (2016) Thousands o
mic obial genomes shed ligh on in e connec ed
biogeochemical p ocesses in an aqui e sys em. Na
Commun 7:1–11.
An
on, J., O en, A., Benlloch, S., Rod ı! Guez-Vale a, F.,
Amann, R., and Rossell
o-Mo a, R. (2002) Salinibac e
ube gen. no ., sp. no ., a no el, ex emely halophilic
membe o he Bac e ia om sal e n c ys allize ponds. In
J Sys E ol Mic obiol 52: 485–491.
A ones y, E. (2011) Command-line ools o p ocessing bio-
logical sequencing da a. ea-u ils. Exp ession Analysis,
Du ham, NC.
Baa i, H., Gue mazi, S., Gha sallah, N., Sghi , A., and
Amma , E. (2010) Mic obial communi y o sal c ys als
p ocessed om Medi e anean seawa e based on 16S
RNA analysis. Can J Mic obiol 56:44–51.
Bake , B.J., Comolli, L.R., Dick, G.J., Hause , L.J., Hya , D.,
Dill, B.D., e al. (2010) Enigma ic, ul asmall, uncul i a ed
A chaea. P oc Na l Acad Sci U S A 107: 8806–8811.
Ba eman, A., Ma in, M.J., O’Dono an, C., Mag ane, M.,
Apweile , R., Alpi, E., e al. (2015) UniP o : a hub o p o-
ein in o ma ion. Nucleic Acids Res 43: D204–D212.
Bouche , Y., Co de o, O.X., Takemu a, A., Hun , D.E.,
Schliep, K., Bap es e, E., e al. (2011) Local mobile gene
pools apidly c oss species bounda ies o c ea e endemic-
i y wi hin global Vib io chole ae popula ions. MBio 2:
e00335-10.
Boujelben, I., Goma iz, M., Ma ínez-Ga cía, M., San os, F.,
Peña, A., L
opez, C., e al. (2012a) Spa ial and seasonal
p oka yo ic communi y dynamics in ponds o inc easing
salini y o S ax sola sal e n in Tunisia. An onie Van Leeu-
wenhoek 101: 845–857.
Boujelben, I., Ya za, P., Almansa, C., Villamo , J.,
Maalej, S., An
on, J., and San os, F. (2012b) Vi ioplank on
communi y s uc u e in Tunisian sola Sal e ns. Appl En i-
on Mic obiol 78: 7429–7437.
Buchfink, B., Xie, C., and Huson, D.H. (2014) Fas and sen-
si i e p o ein alignmen using DIAMOND. Na Me hods
12:59–60.
Capo aso, J.G., Kuczynski, J., S ombaugh, J., Bi inge , K.,
Bushman, F.D., Cos ello, E.K., e al. (2010) QIIME allows
analysis o high- h oughpu communi y sequencing da a.
Na Me hods 7: 335–336.
Chan, P.P., and Lowe, T.M. (2009) G RNAdb: a da abase o
ans e RNA genes de ec ed in genomic sequence.
Nucleic Acids Res 37:93–97.
Chaumeil, P.-A., Mussig, A.J., Hugenhol z, P., and Pa ks, D.H.
(2019) GTDB-Tk: a oolki o classi y genomes wi h he
Genome Taxonomy Da abase. Bioin o ma ics 36:1925–1927.
Colangelo-Lillis, J., Wing, B.A., and Why e, L.G. (2016) Low
i al p eda ion p essu e in cold hype saline A c ic sedi-
men s and limi s on ly ic eplica ion. En i on Mic obiol
Rep 8: 250–260.
Coleman, M.L. (2006) Genomic Islands and he ecology and
e olu ion o P ochlo ococcus. Science 311: 1768–1770.
Co de o, O.X., and Polz, M.F. (2014) Explaining mic obial
genomic di e si y in ligh o e olu iona y ecology. Na Re
Mic obiol 12: 263–273.
Cou inho, F.H., Edwa ds, R.A., and Rod íguez-Vale a, F.
(2019a) Cha ing he di e si y o uncul u ed i uses o
A chaea and bac e ia. BMC Biol 17:1–16.
Cou inho, F.H., Rosselli, R., and Rod íguez-Vale a, F.
(2019) T ends o Mic odi e si y Re eal Dep h-Dependen
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3494 M. D. Ramos-Ba be o e al.
E olu iona y S a egies o Vi uses in he Medi e anean.
mSys em 4: e00554–19.
Cou inho, F.H., Sil ei a, C.B., G ego acci, G.B.,
Thompson, C.C., Edwa ds, R.A., B ussaa d, C.P.D., e al.
(2017) Ma ine i uses disco e ed ia me agenomics shed
ligh on i al s a egies h oughou he oceans. Na
Commun 8:1–12.
Di Meglio, L., San os, F., Goma iz, M., Almansa, C.,
L
opez, C., An
on, J., and Ne cessian, D. (2016) Seasonal
dynamics o ex emely halophilic mic obial communi ies in
h ee A gen inian sal e ns. FEMS Mic obiol Ecol 92:fiw184.
Edga , R.C. (2004) MUSCLE: mul iple sequence alignmen
wi h high accu acy and high h oughpu . Nucleic Acids
Res 32: 1792–1797.
Edga , R.C., and Ba eman, A. (2010) Sea ch and clus e ing
o de s o magni ude as e han BLAST. Bioin o ma ices
26: 2460–2461.
Edwa ds, R.A., McNai , K., Faus , K., Raes, J., and
Du ilh, B.E. (2016) Compu a ional app oaches o p edic
bac e iophage-hos ela ionships. FEMS Mic obiol Re
40: 258–272.
Enaul , F., B ie , A., Bou eille, L., Roux, S., Sulli an, M.B.,
and Pe i , M.A. (2017) Phages a ely encode an ibio ic
esis ance genes: a cau iona y ale o i ome analyses.
ISME J 11: 237–247.
E en, A.M., Esen, O.C., Quince, C., Vineis, J.H., Mo ison,
H.G., Sogin, M.L., e al. (2015) An i’o: An ad anced anal-
ysis and isualiza ion pla o m o ’omics da a. Pee J 2015:
e1319.
F anko ic, A., Eguiluz, L., and Ma ínez-To es, L.M. (2016)
Geodynamic e olu ion o he Salinas de Añana diapi in
he Basque-Can ab ian Basin, Wes e n Py enees. J S uc
Geol 83:13–27.
Ga cia-Do al, C., and an Raaij, M.J. (2012) S uc u e o he
ecep o -binding ca boxy- e minal domain o bac e io-
phage T7 ail fibe s. P oc Na l Acad Sci U S A 109:
9390–9395.
Ghai, R., Ma in-Cuad ado, A.-B., Mol o, A.G., He edia, I.G.,
Cab e a, R., Ma in, J., e al. (2010) Me agenome o he
Medi e anean deep chlo ophyll maximum s udied by
di ec and osmid lib a y 454 py osequencing. ISME J 4:
1154–1166.
Goma iz, M., Ma ínez-Ga cía, M., San os, F., Rod iguez, F.,
Capella-Gu ié ez, S., Gabald
on, T., e al. (2015) F om
communi y app oaches o single-cell genomics: he dis-
co e y o ubiqui ous hype halophilic Bac e oide es gene -
alis s. ISME J 9:16–31.
Guixa-Boixa eu, N., Calde
on-Paz, J., Heldal, M.,
B a bak, G., and Ped
os-Ali
o, C. (1996) Vi al lysis and
bac e i o y as p oka yo ic loss ac o s along a salini y g a-
dien . Aqua Mic ob Ecol 11: 215–227.
Hallam, S.J., Pu nam, N., P es on, C.M., De e , J.C.,
Rokhsa , D., Richa dson, P.M., e al. (2004) Re e se
me hanogenesis: es ing he hypo hesis wi h en i onmen-
al genomics. Science 305: 1457–1462.
Hayes, S., Mahony, J., Nau a, A., an Sinde en, D.,
Hayes, S., Mahony, J., e al. (2017) Me agenomic
app oaches o assess bac e iophages in a ious en i on-
men al niches. Vi uses 9: 127.
Hoshino, T., Yilmaz, L.S., Nogue a, D.R., Daims, H., and
Wagne , M. (2008) Quan ifica ion o a ge molecules
needed o de ec mic oo ganisms by fluo escence in si u
hyb idiza ion (FISH) and ca alyzed epo e deposi ion-
FISH. Appl En i on Mic obiol 74: 5068–5077.
Ho ho n, T., Ho nik, K., Van De Wiel, M.A., and Zeileis, A.
(2006) A lego sys em o condi ional in e ence. Am S a
60: 257–263.
Hun emann, M., Ma omma is, K., I ano a, N.,
Mikhailo a, N., O chinniko a, G., Schaumbe g, A., e al.
(2014) The JGI pipeline o anno a ion o mic obial
genomes and me agenomes he JGI pipeline o anno a-
ion o mic obial genomes and me agenomes. Cali Digi
Lib :2–4.
I anzo, J., K upo ic, M., and Koonin, E.V. (2017) A ne wo k
pe spec i e on he i us wo ld. Commun In eg Biol 10:
e1296614.
I anzo, J., Wol , Y.I., Koonin, E.V., and Sela, I. (2019) Gene
gain and loss push p oka yo es beyond he homologous
ecombina ion ba ie and accele a e genome sequence
di e gence. Na Commun 10:1–10.
I iba , V., and
Abalos, B. (2011) The geochemical and iso o-
pic eco d o e apo i e ecycling in spas and sal e ns o
he Basque Can ab ian basin, Spain. Appl Geochem 26:
1315–1329.
Kea se, M., Moi , R., Wilson, A., S ones-Ha as, S.,
Cheung, M., S u ock, S., e al. (2012) Geneious basic: an
in eg a ed and ex endable desk op so wa e pla o m o
he o ganiza ion and analysis o sequence da a. Bioin o -
ma ics 28: 1647–1649.
Kee hi, S., Kodu u, U.D., Ni ala, S.S., and Pa ine, N.R.
(2018) The he e o ophic eubac e ial and a chaeal co-
inhabi an s o he halophilic Dunaliella salina in sola sal-
e ns ed by bay o Bengal along sou h eas e n coas o
India. Saudi J Biol Sci 25: 1411–1419.
Kons an inidis, K.T., Rossell
o-M
o a, R., and Amann, R.
(2017) Uncul i a ed mic obes in need o hei own axon-
omy. ISME J 11: 2399–2406.
Lagesen, K., Hallin, P., Rødland, A., S ae eld , H.-H.,
Rognes, T., and Usse y, D.W. (2007) RNAmme : consis-
en and apid anno a ion o ibosomal RNA genes.
Nucleic Acids Res 35: 3100–3108.
Langmead, B., and Salzbe g, S.L. (2012) Fas gapped- ead
alignmen wi h bow ie 2. Na Me hods 9: 357–359.
Li, S.-J., Hua, Z.-S., Huang, L.-N., Li, J., Shi, S.-H., Chen, L.-
X., e al. (2014) Mic obial communi ies e ol e as e in
ex eme en i onmen s. Sci Rep 4: 6205.
Li, W., and Godzik, A. (2006) Cd-hi : a as p og am o clus-
e ing and compa ing la ge se s o p o ein o nucleo ide
sequences. Bioin o ma ics 22: 1658–1659.
L
opez-Pé ez, M., Ha o-Mo eno, J.M., Gonzalez-Se ano, R.,
Pa as-Mol
o, M., and Rod iguez-Vale a, F. (2017)
Genome di e si y o ma ine phages eco e ed om Medi-
e anean me agenomes: size ma e s. PLoS Gene 13:
e1007018.
Ludwig, W., S unk, O., Wes am, R., Rich e , L., Meie , H.,
Yadhukuma , A., e al. (2004) ARB: a so wa e en i onmen
o sequence da a. Nucleic Acids Res 32:1363–1371.
Mahmoudabadi, G., Milo, R., and Phillips, R. (2017) Ene -
ge ic cos o building a i us. P oc Na l Acad Sci U S A
114: E4324–E4333.
Ma inez-He nandez, F., Fo nas, O., Lluesma Gomez, M.,
Bolduc, B., de la C uz Peña, M.J., Ma ínez, J.M., e al.
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3495
(2017) Single- i us genomics e eals hidden cosmopoli an
and abundan i uses. Na Commun 8: 15892.
Ma u ano, L., San os, F., Rossell
o-Mo a, R., and An
on, J.
(2006) Mic obial di e si y in Ma as sal e ns, a hype saline
en i onmen in he Pe u ian Andes. Appl En i on
Mic obiol 72: 3887–3895.
Menzel, P., Ng, K.L., and K ogh, A. (2016) Fas and sensi-
i e axonomic classifica ion o me agenomics wi h Kaiju.
Na Commun 7:1–9.
Mezi i, A., Tsemen zi, D., Rod iguez-R, L.M., Ha , J.K.,
Ka ayanni, H., Ko mas, K.A., and Kons an inidis, K.T.
(2019) Quan i ying he changes in gene ic di e si y
wi hin sequence-disc e e bac e ial popula ions ac oss a
spa ial and empo al i e ine g adien . ISME J 13:
767–779.
Mizuno, C.M., Ghai, R., and Rod iguez-Vale a, F. (2014) E i-
dence o me a i omic islands in ma ine phages. F on
Mic obiol 5: 27.
Mizuno, C.M., Rod iguez-Vale a, F., Kimes, N.E., and
Ghai, R. (2013) Expanding he ma ine i osphe e using
me agenomics. PLoS Gene 9: e1003987.
Mo a-Ruiz, M.d.R., Ci uen es, A., Fon -Ve de a, F., Pé ez-
Fe n
andez, C., Fa ias, M.E., Gonz
alez, B., e al. (2018)
Biogeog aphical pa e ns o bac e ial and a chaeal com-
muni ies om dis an hype saline en i onmen s. Sys Appl
Mic obiol 41: 139–150.
Mo a-Ruiz, M.d.R., Fon -Ve de a, F., O fila, A., Ri a, J., and
Rossell
o-M
o a, R. (2016) Endophy ic mic obial di e si y o
he halophy e A h ocnemum mac os achyum ac oss
plan compa men s. FEMS Mic obiol Ecol 92:fiw145.
Muniozgu en, M.A., Balles e os, I.M., Gamboa, J., Seoane, S.,
Alonso, R., Lao den, L., e al. (2021) Al e e y h obac e
mu iae sp. no ., isola ed om hype saline Añana Sal Valley
sp ing wa e , a con inen al halassohaline ype sola sal e n.
In J Sys E ol Mic obiol 71:1–9.
Mu lu, M.B., Ma ínez-Ga cía, M., San os, F., Peña, A.,
Gu en, K., and An
on, J. (2008) P oka yo ic di e si y in
Tuz Lake, a hype saline en i onmen in inland Tu key.
FEMS Mic obiol Ecol 65: 474–483.
Muyze , G., De Waal, E.C., and Ui e linden, A.G. (1993)
P ofiling o complex mic obial popula ions by dena u ing
g adien gel elec opho esis analysis o polyme ase chain
eac ion-amplified genes coding o 16S RNA. Appl En i-
on Mic obiol 59: 695–700.
O en, A. (2002) Molecula ecology o ex emely halophilic
A chaea and bac e ia. FEMS Mic obiol Ecol 39:1–7.
O en, A. (2014) The ecology o Dunaliella in high-sal en i-
onmen s. J Biol Res 21:23–23.
Pa ikka, K.J., Le Romance , M., Wau e s, N., and
Jacque , S. (2017) Deciphe ing he i us- o-p oka yo e
a io (VPR): insigh s in o i us-hos ela ionships in a a i-
e y o ecosys ems. Biol Re 92: 1081–1100.
Pa ks, D.H., Imel o , M., Skenne on, C.T., Hugenhol z, P.,
and Tyson, G.W. (2015) CheckM: assessing he quali y o
mic obial genomes eco e ed om isola es, single cells,
and me agenomes. Genome Res 25: 1043–1055.
Pla a, A., and E kiaga, A. (2018) El sis ema de p oducci
on
de sal de Añana. Valle Salado (A aba, País Vasco).
Uni e sidad del País Vasco, Bilbao, España pp. 170.
Peng, Y., Leung, H.C.M., Yiu, S.M., and Chin, F.Y.L. (2012)
IDBA-UD: a de no o assemble o single-cell and
me agenomic sequencing da a wi h highly une en dep h.
Bioin o ma ics 28: 1420–1428.
P uesse, E., Quas , C., Kni el, K., Fuchs, B.M., Ludwig, W.,
Rg Peplies, J., and Glö Ckne , F.O. (2007) SILVA: a com-
p ehensi e online esou ce o quali y checked and
aligned ibosomal RNA sequence da a compa ible wi h
ARB. Nucleic Acids Res 35: 7188–7196.
Puen e-S
anchez, F., A ce-Rod íguez, A., Ogge in, M.,
Ga cía-Villadangos, M., Mo eno-Paz, M., Blanco, Y., e al.
(2018) Viable cyanobac e ia in he deep con inen al sub-
su ace. P oc Na l Acad Sci U S A 115: 10702–10707.
Ramos-Ba be o, M.D., Ma in-Cuad ado, A.-B., Vi e , T.,
San os, F., Ma inez-Ga cia, M., and An
on, J. (2019)
Reco e ing mic obial genomes om me agenomes in
hype saline en i onmen s: he good, he bad and he ugly.
Sys Appl Mic obiol 42:30–40.
Reisse , W. (2007) The Hidden Li e o Algae Unde g ound.
Do d ech : Sp inge , pp. 47–58.
Rod iguez-R, L.M., Gun u u, S., Ha ey, W.T., Rossell
o-
Mo a, R., Tiedje, J.M., Cole, J.R., and Kons an inidis, K.T.
(2018) The mic obial genomes a las (MiGA) webse e :
axonomic and gene di e si y analysis o A chaea and
bac e ia a he whole genome le el. Nucleic Acids Res
46: W282–W288.
Rod iguez-R, L.M., and Kons an inidis, K.T. (2014) Nonpa-
eil: a edundancy-based app oach o assess he le el o
co e age in me agenomic da ase s. Bioin o ma ics 30:
629–635.
Rod iguez-R, L.M., and Kons an inidis, K.T. (2016) The
En eomics Collec ion: A Toolbox o Specialized Analyses
o Mic obial Genomes and Me agenomes. Pee J P ep .
Rod iguez-Vale a, F., Mizuno, C.M., and Ghai, R. (2014)
Tales om a housand and one phages. Bac e iophage 4:
e28265.
Roudnew, B., La e y, T., Seymou , J., Smi h, R., and
Mi chell, J. (2013) Spa ially a ying complexi y o bac e ial
and i us-like pa icle communi ies wi hin an aqui e sys-
em. Aqua Mic ob Ecol 68: 259–266.
Roux, S., Enaul , F., Hu wi z, B.L., and Sulli an, M.B. (2015)
Vi So e : mining i al signal om mic obial genomic da a.
Pee J 3: e985.
Roux, S., Enaul , F., Ra e , V., Colombe , J., Be a el, Y.,
Augue , J.-C., e al. (2016) Analysis o me agenomic da a
e eals common ea u es o halophilic i al communi ies
ac oss con inen s. En i on Mic obiol 18: 889–903.
Roux, S., K upo ic, M., Deb oas, D., Fo e e, P., and
Enaul , F. (2013) Assessmen o i al communi y unc-
ional po en ial om i al me agenomes may be hampe ed
by con amina ion wi h cellula sequences. Open Biol 3:
130160.
Roux, S., P
aez-Espino, D., Chen, I.-M.A., Palaniappan, K.,
Ra ne , A., Chu, K., e al. (2020) IMG/VR 3: an in eg a ed
ecological and e olu iona y amewo k o in e oga ing
genomes o uncul i a ed i uses. Nucleic Acids Res 49:
D764–D775.
Rubino, F., Ca be y, C., Wa e s, M., Kenny, D., McCabe, M.
S., and C ee ey, C.J. (2017) Di e gen unc ional iso o ms
d i e niche specialisa ion o nu ien acquisi ion and use
in umen mic obiome. ISME J 11: 932–944.
Sal e , S.J., Cox, M.J., Tu ek, E.M., Calus, S.T.,
Cookson, W.O., Mo a , M.F., e al. (2014) Reagen and
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3496 M. D. Ramos-Ba be o e al.
labo a o y con amina ion can c i ically impac sequence-
based mic obiome analyses. BMC Biol 12:1–12.
San os, F., Ya za, P., Pa o, V., Mesegue , I., Rossell
o-
M
o a, R., and An
on, J. (2012) Cul u e-independen
app oaches o s udying i uses om hype saline en i on-
men s. Appl En i on Mic obiol 78: 1635–1643.
Saye s, E.W., Aga wala, R., Bol on, E.E., B is e , J.R.,
Canese, K., Cla k, K., e al. (2016) Da abase esou ces o
he Na ional Cen e o bio echnology in o ma ion. Nucleic
Acids Res 44:D7–D19.
Schä e , H., Be na d, L., Cou ies, C., Leba on, P.,
Se ais, P., Pukall, R., e al. (2001) Mic obial communi y
dynamics in Medi e anean nu ien -en iched seawa e
mesocosms: changes in he gene ic di e si y o bac e ial
popula ions. FEMS Mic obiol Ecol 34: 243–253.
Schloissnig, S., A umugam, M., Sunagawa, S., Mi e a, M.,
Tap, J., Zhu, A., e al. (2013) Genomic a ia ion landscape
o he human gu mic obiome. Na u e 493:45–50.
Schmiede , R., and Edwa ds, R. (2011) Quali y con ol and
p ep ocessing o me agenomic da ase s. Bioin o ma ics
27: 863–864.
Sil ei a, C.B., Cou inho, F.H., Ca alcan i, G.S., Benle , S.,
Doane, M.P., Dinsdale, E.A., e al. (2020) Genomic and
ecological a ibu es o ma ine bac e iophages encoding
bac e ial i ulence genes. BMC Genomics 21: 126.
Smillie, C.S., Smi h, M.B., F iedman, J., Co de o, O.X.,
Da id, L.A., and Alm, E.J. (2011) Ecology d i es a global
ne wo k o gene exchange connec ing he human mic o-
biome. Na u e 480: 241–244.
S a nawski, P., Ba aillon, T., E ema, T.J.G., Jochum, L.M.,
Sch eibe , L., Chen, X., e al. (2017) Mic obial communi y
assembly and e olu ion in subseafloo sedimen . P oc
Na l Acad Sci U S A 114: 2940–2945.
Ven osa, A., de la Haba, R.R., S
anchez-Po o, C., and
Papke, R.T. (2015) Mic obial di e si y o hype saline en i-
onmen s: a me agenomic app oach. Cu Opin Mic obiol
25:80–87.
Villamo , J., Ramos-Ba be o, M.D., Gonz
alez-To es, P.,
Gabald
on, T., Rossell
o-M
o a, R., Mesegue , I., e al.
(2017) Cha ac e iza ion o ecologically di e se i uses
in ec ing co-occu ing s ains o cosmopoli an hype -
halophilic Bac e oide es. ISME J 12: 424–437.
V eeland, R.H., Rosenzweig, W.D., and Powe s, D.W.
(2000) Isola ion o a 250 million-yea -old halo ole an bac-
e ium om a p ima y sal c ys al. Na u e 407: 897–900.
Wilkins, L.G.E., E inge , C.L., Jospin, G., and Eisen, J.A.
(2019) Me agenome-assembled genomes p o ide new
insigh in o he mic obial di e si y o wo he mal pools in
Kamcha ka, Russia. Sci Rep 9: 3059.
Wu, Y.-W., Tang, Y.-H., T inge, S.G., Simmons, B.A., and
Singe , S.W. (2014) MaxBin: an au oma ed binning
me hod o eco e indi idual genomes om me agenomes
using an expec a ion-maximiza ion algo i hm. Mic obiome
2: 26.
Zhang, F., Zhou, F., Gan, R., Ren, C., Jia, Y., Yu, L., and
Huang, Z. (2019) PHISDe ec o : a ool o de ec di e se in
silico phage-hos in e ac ion signals o i ome s udies.
bioRXi 661074. h ps://doi.o g/10.1101/661074.
Suppo ing In o ma ion
Addi ional Suppo ing In o ma ion may be ound in he online
e sion o his a icle a he publishe ’s web-si e:
Supplemen a y Da ase 1.
Supplemen a y Da ase 2.
Supplemen a y Da ase 3.
Supplemen a y Da ase 4.
Supplemen a y Da ase 5.
Supplemen a y Da ase 6.
Supplemen a y Fig. S1. (A) Abundances o cells and
i uses in Añana samples. (B) Examples o pho omic o-
g aphs used o coun ing ( om op o bo om): SE DAPI s a-
ined cells, SE CARD-FISH wi h bac e ial p obe, HU CARD-
FISH wi h a chaeal p obe, Syb -Gold s ained i uses om
EC. Ba : 5 μm.
Supplemen a y Fig. S2. Phylogene ic ee econs uc ion
using he neighbou -joining algo i hm on 16S RNA genes
eco e ed om assembled me agenomes.
Supplemen a y Fig. S3. A. Me agenome anno a ion. 1)
glycosyl ans e ase in ol ed in cell wall biosyn hesis, 2)
uncha ac e ized memb ane p o ein, 3) signal ansduc ion
his idine kinase, 4) pu a i e ansposase, 5) nucleo ide-
binding uni e sal s ess p o ein, UspA amily, 6) pu a i e
memb ane p o ein, 7) cell di ision con ol p o ein 6, 8) ans-
posase, 9) ansc ip ion ini ia ion ac o TFIIB, 10) p edic ed
ansc ip ional egula o , 11) me hyl-accep ing chemo axis
p o ein, 12) DNA polyme ase I, 13) PAS domain-con aining
p o ein, 14) DNA-binding ansc ip ional egula o , L p amily,
15) b anched-chain amino acid anspo sys em pe mease
p o ein, 16) ABC-2 ype anspo sys em ATP-binding p o-
ein, 17) small-conduc ance mechanosensi i e channel, 18)
ansi ional endoplasmic e iculum ATPase, 19) 23S RNA.
A chaeal LSU, 20) UDP-glucose 4-epime ase, 21)
uncha ac e ized conse ed p o ein, 22) ansposase and
inac i a ed de i a i es, 23) uncha ac e ized p o ein con-
se ed in a chaea, 24) uncha ac e ized p o ein conse ed in
bac e ia, 25) uni e sal s ess p o ein UspA and ela ed
nucleo ide-binding p o eins, 26) his idine kinase-, DNA gy -
ase B-, and HSP90-like ATPase, 27) p edic ed memb ane
p o ein, 28) PAS old, 29) Si e-specific ecombinase Xe D,
30) Response egula o con aining CheY-like ecei e , AAA-
ype ATPase, and DNA-binding domains, 31)
glycosyl ans e ase, 32) T ansc ip ional egula o s 33) K
+
anspo sys ems, NAD-binding componen , 34) py idine
nucleo ide-disulphide oxido educ ase, 35) p edic ed DNA
binding p o ein, 36) ca ion anspo ATPase. B. Vial
me agenome anno a ion. a) DNA polyme ase, b) phage ail,
c) HNH endonuclease, d) MULTISPECIES: hypo he ical, e)
e minase la ge, ) si e-specific DNA-me hyl ans e ase, g)
phage po al, h) capsid p o ein, i) PBSX amily, j) po al
Halo i us, k) si e-specific in eg ase, l) pu a i e DNA, m) AAA
amily, n) ABC anspo e , o) esponse egula o , p) DEAD/
DEAH box, q) phage majo , ) ATP-binding p o ein, s)
y osine- ype ecombinase/in eg ase, ) ape measu e, u)
fib onec in ype, ) pu a i e ATPase, w) adenine me hyl-
ans e ase, x) IS200/IS605 amily, y) phage e minase.
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
Vi al and mic obial me agenomics o unde g ound b ines 3497
Supplemen a y Fig. S4. Phylogene ic ee econs uc ion
using he neighbou -joining algo i hm wi h he e inal binding
p o eins (RBP, also known as hodopsins) eco e ed om
assembled con igs o each cellula me agenome. Ba plo s
indica e he coun s pe million (CPM) eads mapping o each
ype o RBP.
Supplemen a y Fig. S5. MAGs om ex emely halophilic
mic oo ganisms ob ained om Añana me agenomes. Fo
each MAG, bin me ics (numbe o con igs wi hin he bin,
N50 o he ob ained bins, bin comple eness and deg ee o
con amina ion, GC con en ) and My Taxa e alua es MAGs in
windows o en genes and de e mines he axonomic a filia-
ion o each one. Addi ionally, My Taxa p o ides in o ma ion
abou possible cases o ho izon al gene ans e o con ami-
na ion in s udied genome, hese egions a e ma ked as ed
ci cles in he figu e.
Supplemen a y Fig. S6. Añana MAGs ec ui men plo in all
cell me agenomes.
Supplemen a y Fig. S7. Taxonomic composi ion o i al
me agenomic eads.
Supplemen a y Fig. S8. Taxonomic composi ion o i al
con igs om he HU i al me agenome.
Supplemen a y Fig. S9. Taxonomic composi ion o i al
con igs om he HP i al
me agenome.
Supplemen a y Fig. S10. Taxonomic composi ion o i al
con igs om he SE i al me agenome.
Supplemen a y Fig. S11. Taxonomic composi ion o i al
con igs om he EC i al me agenome.
Supplemen a y Fig. S12. Anno a ion o cosmopoli an i al
genomes.
Supplemen a y Table S1. Cha ac e is ics o he cellula
and i al me agenomes analysed.
Supplemen a y Table S2. Vi al abundances and ANI o
i al con igs in cell and i us me agenomes.
Supplemen a y Table S3. Gene al cha ac e is ics and abun-
dance o HU i us subse in Añana i omes and cell me agenomes
Supplemen a y Table S4.Dis ibu ion o pN/pS calcula ed
o i al genomes om all Añana me a i omes
© 2021 The Au ho s. En i onmen al Mic obiology published by Socie y o Applied Mic obiology and John Wiley & Sons L d.,
En i onmen al Mic obiology,23, 3477–3498
3498 M. D. Ramos-Ba be o e al.
