ORIGINAL ARTICLE
Me ap o eogenomic insigh s beyond bac e ial
esponse o naph halene exposu e and
bio-s imula ion
Ma ı
´a-Eugenia Guazza oni
1,9,11
, Flo ian-Alexande He bs
2,9
,I a
´nLo es
3,9
, Ja ie Tamames
4,9
,
Ana Isabel Pela
´ez
3
, Nie es Lo
´pez-Co e
´s
1
, Ma ı
´a Alcaide
1
, Me cedes V Del Pozo
1
,
Jose
´Ma ı
´a Viei es
1
, Ma in on Be gen
2,5
, Jose
´Luis R Gallego
6
, Ra ael Ba giela
1
,
A an xa Lo
´pez-Lo
´pez
7
, Die ma H Piepe
8
, Ramo
´n Rossello
´-Mo
´ a
7
, Jesu
´sSa
´nchez
3,10
,
Jana Sei e
2,10
and Manuel Fe e
1,10
1
Depa men o Bioca alysis, Ins i u e o Ca alysis, CSIC, Mad id, Spain;
2
Depa men o P o eomics,
UFZ-Helmhol z-Zen um u¨ Umwel o schung GmbH, Leipzig, Ge many;
3
A
´ ea de Mic obiologı
´a, IUBA,
Facul ad de Medicina, Uni e sidad de O iedo, O iedo, Spain;
4
CSIC, Cen o Nacional de Bio ecnologı
´a,
Mad id, Spain;
5
Depa men o Me abolomics, UFZ-Helmhol z-Zen um u¨ Umwel o schung GmbH,
Leipzig, Ge many;
6
IUBA, A
´ ea de P ospeccio
´n e In es igacio
´n Mine a, Uni e sidad de O iedo, Mie es,
Spain;
7
IMEDEA (CSIC-UIB), Espo les, Spain and
8
Helmhol z Zen um u¨ In ek ions o schung–HZI,
B aunschweig, Ge many
Mic obial me abolism in a oma ic-con amina ed en i onmen s has impo an ecological implica ions, and
ob aining a comple e unde s anding o his p ocess emains a ele an goal. To unde s and he oles o
biodi e si y and a oma ic-media ed gene ic and me abolic ea angemen s, we conduc ed ‘OMIC’
in es iga ions in an an h opogenically in luenced and polya oma ic hyd oca bon (PAH)-con amina ed
soil wi h (Nbs) o wi hou (N) bio-s imula ion wi h calcium ammonia ni a e, NH
4
NO
3
and KH
2
PO
4
and he
comme cial su ac an I eysol, plus wo naph halene-en iched communi ies de i ed om bo h soils
(CN2 and CN1, espec i ely). Using a me agenomic app oach, a o al o 52, 53, 14 and 12 dis inc
species (acco ding o ope a ional phylogene ic uni s (OPU) in ou wo k equi alen o axonomic
species) we e iden i ied in he N, Nbs, CN1 and CN2 communi ies, espec i ely. App oxima ely 10
ou o 95 dis inc species and 238 ou o 3293 clus e s o o hologous g oups (COGs) p o ein
amilies iden i ied we e clea ly s imula ed unde he assayed condi ions, whe eas only wo species
and 1465 COGs con o med o he common se in all o he mesocosms. Resul s indica ed dis inc
biodeg ada ion capabili ies o he u ilisa ion o po en ial g ow h-suppo ing a oma ics, which esul s
in bio-s imula ed communi ies being ex emely i o naph halene u ilisa ion and non-s imula ed
communi ies exhibi ing a g ea e me abolic window han p e iously p edic ed. On he basis o
compa ing p o ein exp ession p o iles and me agenome da a se s, in e -alia in e ac ions among
membe s we e hypo hesised. The u ilisa ion o cu a ed da abases is discussed and used o i s
ime o econs uc ‘p esump i e’ deg ada ion ne wo ks o complex mic obial communi ies.
The ISME Jou nal (2013) 7, 122–136; doi:10.1038/ismej.2012.82; published online 26 July 2012
Subjec Ca ego y: in eg a ed genomics and pos –genomics app oaches in mic obial ecology
Keywo ds: biodi e si y; bio-s imula ion; label- ee p o ein quan i ica ion; me agenomics; me ap o eomics;
polya oma ic hyd oca bon
In oduc ion
Polya oma ic hyd oca bons (PAHs) a e widely dis-
ibu ed in he en i onmen owing o hei abundance
in c ude oil and hei widesp ead use in chemical
manu ac u ing (Ka
¨s ne , 2000). PAHs a e pollu an s o
g ea conce n owing o hei oxici y, mu agenici y
and ca cinogenici y. A numbe o mic oo ganisms, ia
he ac ion o Rieske non-haem i on oxygenases, ha e
he abili y o g ow on PAHs as a sole ca bon and
ene gy sou ce (Beil e al., 1998; Roling e al., 2003;
Wi zig e al.,2006;Penge al., 2008).
The majo i y o e o s aimed a unde s anding
mic obial esponses o a oma ic compounds ha e
Co espondence: J Sa
´nchez, A
´ ea de Mic obiologı
´a, IUBA,
Facul ad de Medicina, Uni e sidad de O iedo, O iedo, Spain.
E-mail: [email p o ec ed]
o J Sei e , Depa men o P o eomics, UFZ-Helmhol z-Zen um
u
¨ Umwel o schung GmbH, Leipzig, Ge many.
E-mail: [email p o ec ed]
o M Fe e , Depa men o Bioca alysis, Ins i u e o Ca alysis,
CSIC, Ma ie Cu ie 2, Mad id 28049, Spain.
E-mail: [email p o ec ed]s
9
These au ho s con ibu ed equally o his wo k.
10
These au ho s con ibu ed equally o his wo k.
11
Cu en add ess: Labo a o y o Molecula Ecology, Cen o de
As obiologı
´a (CSIC-INTA), Ca e e a de Ajal i km 4, To ejo
´nde
A doz 28850, Mad id, Spain.
Recei ed 9 Janua y 2012; e ised 11 June 2012; accep ed 11 June
2012; published online 26 July 2012
The ISME Jou nal (2013) 7, 122–136
&
2013 In e na ional Socie y o Mic obial Ecology All igh s ese ed 1751-7362/13
www.na u e.com/ismej
been ocused on genomic (Jime
´nez e al., 2002;
Nogales e al., 2008; Pucha"ka e al., 2008), an-
sc ip omic (Domı
´nguez-Cue as e al., 2006; Yus e
e al., 2006) and p o eomic (San os e al., 2004;
Segu a e al., 2005; Kim e al., 2006; Ku ba o e al.,
2006; Toma
´s-Galla do e al., 2006) analyses con-
duc ed in pu e cul u es. Among PAH-(including
naph halene, luo ene, phenan h ene, py ene and
dibenzo u an, o ci e some) mine alising s ains, a
numbe o bac e ia ha e ecei ed special a en ion.
Those include membe s o he common gene a
o Pseudomonas,Sphingomonas,Cycloclas icus,
Bu kholde ia,Rhodococcus,Pola omonas, some
no el gene a o Nep unomonas and Janibac e ,some
he mophilic bac e ia o Noca dia and Bacillus,
some anoxic bac e ia o Del ap o eobac e ia and
Alcaligenes, and high molecula weigh PAH-
deg ading bac e ia o Mycobac e ium,S eno opho-
monas and Pas eu ella; a lis wi h known pu e
bac e ial cul u es capable o deg ading PAHs can be
seen in a ecen e iew by Lu e al. (2011).
Howe e , a la ge numbe o cul u e-independen
echniques ha e shown ha pollu an -deg ading
o ganismsen ichedin helabo a o yo endono
ha e an impo an ole in he in si u biodeg ada ion o
pollu an s and ha he di e si y o pollu an -deg ad-
ing o ganisms in pollu ed en i onmen al samples is
much g ea e han he di e si y de e mined om
cul i a ion (Abulencia e al., 2006; Liu e al., 2009;
Bo onin and Koshele a, 2010; Yagi e al., 2010; Liang
e al., 2011). Unde his scena io, i is ele an o use
molecula mic obial ools o de ine key ca abolic
playe s a con amina ed si es o p edic pollu an
deg ada ion ne wo ks in he en i onmen and o
sugges me hods o a ional in e en ion associa ed
wi h he implemen a ion o bio emedia ion (Vilchez-
Va gas e al., 2010). Howe e , he numbe o in e-
g a i e ‘omic’ in es iga ions ha ha e been ca ied
ou in PAH-associa ed mic obial communi ies is
limi ed (Kweon e al., 2007; Powell e al., 2008;
Selesi e al., 2010), because o he incomple e
genomic in o ma ion and cu a ed da abases a ailable
(Pe
´ ez-Pan oja e al., 2009, 2012).
Taking all o he abo e in o ma ion in o consid-
e a ion, we pe o med a ho ough and holis ic (o
ecosys ems biology app oach) phylogene ic, unc-
ional and p o eomic analysis o he key playe s in a
sample o s ongly an h opogenically in luenced,
PAH-con amina ed soil (N) and a naph halene-
en iched communi y de i ed om his soil (CN1).
This s udy was ca ied ou using also samples o he
same soil bio-s imula ed wi h calcium ammonia
ni a e, NH
4
NO
3
and KH
2
PO
4
, and he comme cial
su ac an I eysol (I ey In e na ional Inc., Campbell
Ri e , BC, Canada) (he ein e e ed as Nbs), and he
naph halene-en iched communi y de i ed om i
(CN2). In all cases, he biodeg ada ion ne wo ks o
he espec i e whole communi ies we e econ-
s uc ed. This s udy cla i ied he genomic and
p o eomic basis o he pu pose o unde s anding
mic obial biodi e si y, ecology and unc ion in
esponse o bo h PAHs ( ep esen ed by naph halene)
and bio-s imula ion. I should be no ed ha ou
me ap o eomic in es iga ions we e es ic ed o
naph halene-en iched communi ies because o hei
lowe complexi y as compa ed wi h he soil samples
as well as he la ge assembled sequences ob ained
h ough di ec py o-sequencing o bo h samples.
Ma e ials and me hods
Gene al me hods and ‘OMIC’ da a analysis and
p ocessing
Full desc ip ions o he ma e ials and me hods used
o soil cha ac e isa ion; hyd oca bon analysis; DNA
ex ac ion; cons uc ion o 16 S RNA gene clone
lib a ies, sequencing and phylogene ic analysis;
dena u ing g adien gel elec opho esis analysis;
me agenomic se up and sequencing, assembly and
gene p edic ion; me ap o eomic se up and p o ein
ex ac ion, sepa a ion and iden i ica ion and da a
p ocessing a e a ailable in he Supplemen a y
Ma e ials and me hods.
Soil sample collec ion and p epa a ion o bio-
s imula ed soil and en ichmen cul u es
Soil samples (he ein e e ed o as N) we e ob ained
om a pa cel on he no he n Ibe ian Peninsula
con iguous o a chemical plan (Lugones, O iedo,
Spain; 4011803300N, 313903100W, a an al i ude o
300 m). The chemical plan (Supplemen a y Figu e
S1) was used o se e al decades o he p oduc ion
o naph halene, phenols and o he compounds om
coal a p ocessing as well as o manu ac u ing
esins. A conside able amoun o o he chemical
p oduc s (pes icides, sol en s, e c.) was s o ed,
al hough p obably no manu ac u ed, in he plan .
In 1989 he plan was closed and hen used o yea s
o s o e chemical was e, pa icula ly polychlo i-
na ed biphenyls, coolan s and o he unspeci ied
p oduc s. In 2006 mos o he buildings we e
demolished and he cha ac e isa ion o soil con am-
ina ion s a ed. Thus, a signi ican amoun o he
PAH de ec ed we e concei ably o med by he
en iched bac e ial popula ions p esen in he soil
h ough a na u al a enua ion e ec . The op soil
was sampled a a dep h o 0–30 cm on Feb ua y 2007
(soil empe a u e 18 1C). Th ee eplica es (1 kg each)
we e collec ed wi hin a 1 m dis ance, and he
samples we e kep in open plas ic bags in he da k
a 4 1C. Vege a ion and o he non-soil ma e ials,
including cobbles, we e emo ed p io o homo-
genising he samples. Immedia ely a e acqui ing
he soil samples, hey we e sie ed (2-mm mesh
size), ollowed by mixing o ep esen a i e subsam-
ples o he iplica es samples, and 100 g aliquo s
o sample N we e used o chemical analysis. In
addi ion, 10 g aliquo s we e s o ed a 20 1Cin
s e ile lasks o DNA- and p o eome-based analyses.
Bio emedia ion was pe o med o e app oxi-
ma ely 900 m
3
o he con amina ed soil (N), whe e
Mic obial esponse o polya oma ics
M-E Guazza oni e al
123
The ISME Jou nal
9 ons o dehyd a ed calcium ammonia ni a e and
3 ons o a mix u e o dehyd a ed NH
4
NO
3
and
KH
2
PO
4
combined wi h 4500 l o he comme cial
su ac an I eysol we e applied o he homogenised
soil. The C:N:P a io ha was applied was 100:10: 1,
as ecommended o bio emedia ion pu poses
(Gallego e al., 2007a). The ea men was pe o med
o e 231 days, and he biopile was ou inely
wa e ed and illed o main ain humidi y (15–20%)
and adequa e oxygena ion. Samples o he op
bio-s imula ed soil, he ein named Nbs, we e aken
and p ocessed using he same p o ocols as desc ibed
abo e o he o iginal soil.
En ichmen cul u es we e pe o med in Bushnell
Haas (Sigma Chemical Co., S Louis, MO, USA)
mine al medium ha con ained naph halene a a
concen a ion o 0.1% (w/ ) as he sole ca bon
sou ce as desc ibed p e iously (Gallego e al.,
2007b). The composi ion o he medium was as
ollows: MgSO
4
H
2
O (0.20 g l
1
), CaCl
2
2H
2
O
(0.02 g l
1
), KH
2
PO
4
(1.0 g l
1
), K
2
HPO
4
(1.0 g l
1
),
NH
4
NO
3
(1.0 g l
1
) and FeCl
3
(0.05 g l
1
). Two
di e en inocula we e used. The CN1 en ichmen
cul u e was ob ained by inocula ing 1 g o he
pollu ed soil (N) in o a lask con aining 100 ml o
he cul u e medium; he CN2 cul u e was inocula ed
wi h 1 g o he same soil ha had been subjec ed o
a massi e bio emedia ion (bio-s imula ion) p ocess
(Nbs). The en ichmen cul u es we e incuba ed a
30 1C and 250 .p.m., in which 0.1% ( / ) o he
cul u e was ans e ed o esh medium each week.
Resul s and discussion
Sample cha ac e is ics
An ag onomic analysis o he soil N e ealed a
loamy clay soil wi h a pH o 8.2 and a conduc i i y
o 0.13 dS m
1
clea ly con aining low amoun s o
he ypically p edominan ions (calcium, magne-
sium, po assium and sodium); he de ec ed na u al
o ganic ma e , ni ogen and phospho us le els
in he soil (Supplemen a y Table S1) a e cha ac-
e is ic o in e ile soils. Gas ch oma og aphy–mass
spec ome y (GC–MS) was used o quan i y he
le els o he 16 EPA (En i onmen al P o ec ion
Agency) p io i y PAHs p esen in he soil
(Supplemen a y Figu e S2). To ally, he soil exhib-
i ed an a e age concen a ion o 805 mg PAH pe kg
(Table 1), which is in he ange o sligh ly lowe
o he le el ound in p e iously epo ed PAH
con amina ed soils ( ha is, 1667 mg kg
1
in Nı
´
Chadhain e al. (2006); 589 mg kg
1
in Richa dson
e al. (2012); 335–8645 mg kg
1
in Tha amani e al.
(2012); 3000 mg kg
1
in Lo s e al. (2012)). The bio-
s imula ed soil exhibi ed a o al concen a ion o
221.6 mg kg
1
(a e age) o he 16 EPA p io i y PAHs
and he quan i ied compounds a e lis ed in Table 1.
The deg ada ion a e o naph halene in N was oo
low and was di icul o be es ablished owing o he
long ime anscu ed in he p esence o he
con aminan s; howe e , once submi ed o bio-
s imula ion, we calcula ed he deg ada ion a e o
naph halene in soil om ime 0 o 231 days:
1.818 mg kg
1
soil pe day.
Dena u ing g adien gel elec opho esis was used
o su ey he de elopmen o he mic obial commu-
ni ies in he en ichmen cul u es and o deduce
he ime poin a which a s able communi y was
ob ained. We obse ed ha he dena u ing g adien
gel elec opho esis p o iles o CN1 en ichmen
cul u es changed be ween he 45 and 60 ans e s
(Supplemen a y Figu e S3); changes in he p o iles
we e also obse ed o CN2 a e 35 and 60 ans e s.
These changes we e subsequen ly main ained and
samples subjec ed o a leas 60 ans e s we e
he ein used o u he in es iga ions. The majo
bands we e excised and sequenced, and hei
a ilia ion e ealed he p esence o mic oo ganisms
iden i ied as membe s o he gene a Ach omobac e ,
Fla obac e ium and Acido o ax in bo h en ich-
men s, and Pseudomonas,Mic obac e ium,Lysobac-
e and o an endosymbion o Acan homaeba in
CN2. Resul s indica ed ha conso ia CN1 and CN2
we e able o deg ade 97% naph halene a an ini ial
concen a ion o 1000 p.p.m. wi hin 72 and 80 h,
espec i ely.
Bac e ial di e si y and composi ion bluep in s
DNA isola ed om each in es iga ed mic obial
communi y was employed o a PCR-based 16S
ecombinan DNA ( DNA) gene di e si y su ey o
he communi y s uc u es in he samples. Fo his
pu pose, clone lib a ies we e cons uc ed as
Table 1 Le el o 16 EPA p io i y PAHs p esen in he o iginal (N)
and bio-s imula ed (Nbs) soils
PAH Concen a ion (mg kg
1
)
Soil N Soil Nbs
Naph halene 607 174
Acenaph hylene 1.60 0.40
Acenaph hene 16.60 4.40
Fluo ene 20.80 5.96
An h acene 21.20 4.99
Phenan h ene 46.40 24.23
Fluo an hene 30.40 4.73
Py ene 19.40 4.02
Benzo(a)an h acene 12.00 2.83
C ysene 12.50 2.64
Benzo(b) luo an hene 4.40 1.34
Benzo(k) luo an hene 3.10 0.90
Benzo(a)py ene 4.90 1.27
Indene(1,2,3-cd)py ene 3.40 1.58
Benzo(g,h,i)pe ylene 1.50 0.82
Dibenzo(a,h)an h acene 0.79 0.18
Abb e ia ions: N, non-bio-s imula ed polya oma ic hyd oca bon-
con amina ed soil; Nbs, bio-s imula ed polya oma ic hyd oca bon-
con amina ed soil; PAH, polya oma ic hyd oca bon.
Gas ch oma og aphy–mass spec ome y (GC–MS) was used o
quan i ica ion.
Mic obial esponse o polya oma ics
M-E Guazza oni e al
124
The ISME Jou nal
desc ibed in he Supplemen a y Ma e ials and
me hods, and he clones we e ully sequenced o
ob ain he almos comple e cloned 16S DNA gene
sequence. Addi ionally, and because o he low
yields in he clone lib a ies o Nbs, we used he 16S
DNA gene pa ial sequences ob ained in he
me agenome su ey. Fo his pu pose, we used only
hose sequences ha had a leng h 4600 nucleo ides.
All sho e sequences we e disca ded. A o al o 670
sequences ( ha is, N: 212; Nbs: 261 (86 clones þ175
454 pa ial sequences); CN1: 90; CN2: 107) we e
ob ained, and analysed. The o e all phylogene ic
composi ion in he lib a ies is shown in Figu es 1–3,
in which all ope a ional phylogene ic uni s (OPUs)
a ilia ed wi h wel e phyla o he domain Bac e ia,
namely, P o eobac e ia ollowed o much lowe
ex end by Bac e oide es,Ve ucomic obia,Fi mi-
cu es,Ac inobac e ia,Chlo o lexi,Cyanobac e ia,
Planc omyce es,Spi ochae es and he candida us
phyla OP8, TM7 and WCHB1 (Figu e 4).
The clones we e classi ied in o 124 bac e ial
ope a ional phylogenic uni s (OPUs) (Lo
´pez-Lo
´pez
e al., 2010) (Table 2 and Supplemen a y Figu e S2).
Highe alues o Shannon–Wiene ’s and Good’s
co e age indexes indica ed ha communi ies N and
Nbs we e much mo e complex (51 OPUs and 53
OPUs we e de ec ed, espec i ely) han he CN1 and
CN2 (13 OPUs and 12 OPUs, espec i ely) commu-
ni ies, which we e domina ed by ew mic obial
species o s ains and exhibi ed a a he simple
s uc u e (Figu es 1–3). I should be ecalled ha
each o he dis inc OPUs de ec ed was iden i ied as
a pu a i e single species owing o he high sequence
iden i y (Ya za e al., 2008). Howe e , i should also
be no ed ha he sequencing su ey co e ed o e
75% o he expec ed OPU di e si y in all cases as i
can be deduc ed om he high Good’s co e age in all
samples (Table 2), and he a e ac ion cu es
indica ing closeness o sa u a ion (Supplemen a y
Figu e S4). A la ge p opo ion o he ob ained
sequences a ilia ed wi h known amilies and
clus e ed wi h b anches ep esen ed by cul u ed
mic oo ganisms ha ha e p e iously been ound in
con amina ed en i onmen s (Figu es 1–3 and
Supplemen a y Figu es S5 and S6), and hus, hey
do no appea o be speci ic o he soil and he
en ichmen samples ha we e in es iga ed he e.
Wha e e he case, despi e soil cha ac e is ics
and speci ic pollu an s composi ion may di e ,
he biodi e si y ound in he o iginal soils (N and
Figu e 1 A neighbou -joining ee o he p o eobac e ial 16S
DNA gene sequences om he gene sequences o igina ed om N,
Nbs, CN1 and CN2 communi ies. Fo he econs uc ion, only he
almos comple e sequences o he clones om he ou samples
had been used. Besides, pa ial 454 sequences o 4600 nucleo-
ides we e inse ed in he ee using he Pa simony ool
implemen ed in ARB. The numbe o sequences comp ised in
each iden i y clus e (OPU) is speci ied o each sample. The
colou code o he 16S DNA sequences is as ollows: N (g een),
Nbs (pu ple), CN1 (blue), CN2 ( ed). Whe he an OPU con ains
sequences o mo e han one sample, he OPU denomina ion is
w i en in black ollowed by he code colou and sequence
numbe o each s udied sample.
Mic obial esponse o polya oma ics
M-E Guazza oni e al
125
The ISME Jou nal
Nbs) (Shannon indexes o 2.81 and 3.32, espec-
i ely) a e in he ange o ha ound in o he PAH-
con amina ed soils. Fo example, Shannon indexes
o 1.74–2.78 (Vi as e al., 2008), 2.38–2.96 (Tha amani
e al., 2012) and 4.87–5.01 (Ma in e al., 2012)
we e obse ed in soils ha had been his o ically
con amina ed wi h PAHs.
The dominan g oup among he Gammap o eo-
bac e ia consis ed o membe s o he genus Pseudo-
monas (Figu es 2 and 4). The N sequences we e
dis ibu ed a he e enly wi hin his genus, wi h
app oxima ely 40% o he o al gammap o eo-
bac e ial sequences b anching in he P. luo escens,
P. ede iksbe gensis and P. amygdale lineages,
ollowed by hose a ilia ed wi h he Pseudoxan ho-
monas spp. (20%) and P. pu ida (14%) lineages
(Supplemen a y Table S2). A simila e en dis ibu-
ion was obse ed in he sample Nbs, being as well
he membe s o Pseudomonadaceae he mos abun-
dan . Howe e , in his case, he Pseudomonas
species composi ion shi ed owa ds he majo
ep esen a i es P. uomou ensis,P. pe ucinogena,
P. s u ze i genomo a 1, and ye undesc ibed
Pseudomonas species. These esul s clea ly con as
wi h he dis ibu ion o he CN2 sequences, whe e
jus ep esen a i es o P. s u ze i (46%) lineages
(Supplemen a y Table S2) we e obse ed in acco -
dance wi h his being one o he majo ep esen ed
lineages in Nbs. Finally, in CN1, no ep esen a ion
o he Pseudomonas genus was ound, bu six
sequences ep esen ing Pseudoxan homonas we e
iden i ied (6% o he o al) (Figu es 1 and 4 and
Supplemen a y Table S2); his is in ag eemen wi h
p e ious obse a ions made in PAH mic ocosm
expe imen s ha sugges ha , al hough, Pseudomonas
spp. can be easily en iched and use a a ie y o
o ganic subs a es, hey may be ep essed by o he
P o eobac e ia (Rhee e al., 2004).
The alphap o eobac e ial sequences de ec ed in
N we e widely dis ibu ed among Rhizobium,
De osia,No osphingobium,Sphingomonas and
B e undimonas, which we e he mos ep esen ed
;;)1VG1UPO (Pseudomonas s u ze i 1N 12Nsb 48CN2
sbN3)(2UPO Pseudomonas alcaligenes
Pseudomonas o i idis, AY953147
;)3VG(3UPO Pseudomonas s u ze i 6N 1Nsb
,).ps(4UPO Pseudomonas s u ze i 8Nbs 1CN2
;)C41M(5UPO Pseudomonas s u ze i 5N 10Nbs
Nbs132 (OPU 96)
bsN12)(6UPO Pseudomonas uomue ensis
Pseudomonas mendocina, Z76674
N-179 (OPU 7)
Pseudomonas a gen inensis, AY691188
Pseudomonas la escens, U01916
Nbs117 (OPU 97)
sbN8)(8UPO Pseudomonas pe ucinogena
OPU 9 ( sp.) 20NbsPseudomonas
)(01UPO Pseudomonas guineae 1Nbs4N,
Pseudomonas anguillisep ica, X99540
Pseudomonas bo bo i, AM114527
N56)(11UPO Pseudomonas amygdali
N-193 (OPU 12)
Pseudomonas sp. VET-8, EU781734
Pseudomonas b assicacea um, AF100321
Pseudomonas luo escens, AJ308308
Pseudomonas o ien alis, AF064457
N22)(31UPO Pseudomonas pu ida
Pseudomonas japonica, AB126621
OPU 14 sp.) 3Nbs(Pseudomonas
Azo obac e inelandii, AB175657
N-05 (OPU 15)
Pseudomonas balea ica, U26418
uncul u ed bac e ium, GQ979955
OPU 16 ( sp.) 3NbsPseudomonas
10%
N clones
CN1(N) clones
CN2 (Nbs) clones
Nbs clones and 454 sequences
Figu e 2 A subse o he ee shown in Figu e 1, whe e he genealogical composi ion o he membe s o he amily Pseudomonadaceae is
shown. Layou , econs uc ion and colou codes used a e he same as in Figu e 1.
Mic obial esponse o polya oma ics
M-E Guazza oni e al
126
The ISME Jou nal

gene a (Figu e 1 and Supplemen a y Table S2). On
he o he hand, membe s o his lineage we e ba ely
ep esen ed in Nbs wi h jus se en sequences, being
B e undimonas he mos ep esen ed. None o hese
gene a we e de ec ed in any o he en ichmen
cul u es (Figu e 1). In con as , mos o he alpha-
p o eobac e ial clones in CN1 (73%) we e a ilia ed
wi h Azospi illum species, ha is, Azospi illum
o yzae, whe eas only h ee alphap o eobac e ial
sequences (ou o a o al o 108) we e ound in
CN2, and only one was a ilia ed wi h Azospi illum
(Figu e 5 and Supplemen a y Table S2).
Finally, he wo N clones (ou o 214) ep esen ing
Be ap o eobac e ia a ilia ed wi h Acido o ax and
Ach omobac e spp. (Supplemen a y Table S2). On
he o he hand, Nbs sequences we e impo an ly
ep esen ed in his lineage wi h 27 sequences (10%),
a ilia ed wi h Te a hiobac e and Acido o ax.In
his ega d, he pe cen age o Be ap o eobac e ia
ound in he en ichmen cul u es was much highe
(Figu es 1 and 4); we obse ed an o e ep esen-
a ion o sequences a ilia ed wi h he gene a
Comamonas and Ach omobac e in CN1 and
Ach omobac e in CN2 (Figu es 1 and 5). In addi-
ion o Del a-andEpsilon-p o eobac e ia, membe s
o Tene icu es, Ve ucomic obia, Fi micu es and
Cyanobac e ia we e only de ec ed in samples N
and Nbs (Figu es 3 and 4).
The abo e da a demons a ed ha he CN1 and
CN2 communi ies displayed conside ably di e en
phylogene ic s uc u es (Figu es 4 and 5), sha ing
only wo OPUs: Be a- and Gamma-p o eobac e ia
ep esen a i es o he deni i ying Ach omobac e
( ha is, Ach omobac e spanius) and Azospi illum
( ha is, Azospi illum doebe eine ae) gene a, espec-
i ely. Whe eas he i s one was mos abundan in
CN2 (29 s 7 sequences), he second was pa icu-
la ly en iched in CN1 (35 s 1 sequences)
(Supplemen a y Table S2). In addi ion o supplying
biosu ac an s, he addi ion o di e en ni a e
compounds may ha e s imula ed he g ow h o
deni i ying Ach omobac e spp. and P. s u ze i
(Song and Wa d, 2003) in he bio-s imula ed com-
muni y, whe eas deple ion o ni a e may ha e
a ou ed he p esence o he ni ogen- ixing Azos-
pi illum species in he non-s imula ed communi y.
In addi ion o deni i ica ion, P. s u ze i is well
known as naph halene deg ade , which has been
isola ed in such hyd oca bon en ichmen s om a
wide ange o en i onmen al samples (Rossello
´-
Mo a e al., 1994), and was al eady an impo an
componen o he o iginal Nbs soil. Howe e , he
possibili y ha he selec i e deso p ion and libe a-
ion o ecalci an compounds om soils du ing he
bio-s imula ion p ocess may ha e key unc ions in
shaping he genomes and communi y s uc u es
o he associa ed mic oo ganisms should no be
uled ou . The indings o p e ious s udies using
BTEX biodeg ada ion co-cul u es (Shim e al., 2005)
ag ee wi h his hypo hesis. In any case, al hough
bac e ia ela ed o Pseudomonas, Ach omobac e
Figu e 3 A neighbou -joining ee o non-p o eobac e ial 16S
DNA gene sequences om he clone lib a ies es ablished om N,
Nbs, CN1 and CN2 communi y DNA. Layou , econs uc ion and
colou codes used a e he same as in Figu e 1.
Mic obial esponse o polya oma ics
M-E Guazza oni e al
127
The ISME Jou nal
and Comamonas spp. ha e been widely associa ed
wi h PAH deg ada ion in a numbe o con amina ed
ecosys ems and mic obial conso ia (Goyal and
Zyls a, 1997), he p esence o he Azospi illum
species in con amina ed soils and PAH-deg ading
bac e ial conso ia has only p e iously been
epo ed in a hea ily c eoso e-con amina ed soil
(Vin
˜as e al., 2005). This is o special in e es
because i is known ha he ecen ly sequenced
Azospi illum sp. B510 ha bou s gene se s o
deg ada ion pa hways, hough hei unc ions ha e
no ye been expe imen ally elucida ed (Kaneko
e al., 2010).
Due o he clea shi s in communi y composi ion
de ec ed among he in es iga ed condi ions, i is
impo an o es ablish he oles and deg ada ion
capabili ies o he indi idual mic obial membe s o
he communi ies o unde s and he o e all unc ion
o each communi y and i s membe s.
Genomic signa u es associa ed wi h he ae obic
deg ada ion o a oma ics
DNA isola ed om each mic obial communi y was
sequenced using a Roche GS FLX DNA sequence , a
Li esequencing SL (Valencia, Spain) which p o-
duced 1 471 821 eads (N: 165 791; Nbs: 438 821;
CN1: 448 713; CN2: 418 293) wi h an a e age leng h
pe ead o 395 bp o CN1 and CN2, 336 bp o N
and 531 o Nbs. Acco dingly, a o al o 55.7, 233.3,
177.2 and 165.2 Mbp o aw DNA sequences o N,
Nbs, CN1 and CN2 we e ob ained, which we e
assembled in o 2.6 Mbp (4335 con igs), 17.9 Mbp
(16 032 con igs), 20.0 Mbp (20 809 con igs) and 13.0
Mbp (9915 con igs), espec i ely. CN1 (266) and
CN2 (256) con ained a highe numbe o con igs
wi h leng hs g ea e han 10 kbp compa ed wi h Nbs
(68) and o majo ex en N (only 1); his may be due
Figu e 4 Compa ison o bac e ial phylo ypes (cu -o o 497% sequence iden i y) based on he 16 S DNA sequences ex ac ed om N,
Nbs, CN1 and CN2 communi y DNA. The pe cen ages o bac e ial phylogene ic lineages de ec ed we e based on OPUs. (a) Pe cen age o
16S DNA sequences dis ibu ed by phyla. (b) Con ibu ion o dominan g oups among Gammap o eobac e ia.(c) Rela i e dis ibu ion o
majo bac e ial phylo ypes (based on 16S DNA gene sequences (OPUs)) o soil communi ies CN1 and CN2 a he genus le el.
Table 2 S a is ical indexes o he ou samples
N Nbs CN1 CN2
Taxa (OPUs) 51 53 13 11
Indi iduals (sequences) 212 261 90 107
Dominance_D 0.13 0.06 0.20 0.30
Shannon_H 2.81 3.32 2.02 1.50
Good’s_G 0.76 0.80 0.86 0.90
Abb e ia ions: CN1, naph halene-en iched communi y de i ed om
N soil; CN2, naph halene-en iched communi y de i ed om Nbs soil;
OPU, ope a ional phylogene ic uni ; N, non-bio-s imula ed
polya oma ic hyd oca bon-con amina ed soil; Nbs, bio-s imula ed
polya oma ic hyd oca bon-con amina ed soil.
PAST so wa e 1.82b (Hamme e al., 2001) was used o calcula e he
s a is ical indices o he bac e ial sequences. The ollowing o mulas
we e used. Shannon–Weine index: H ¼S(ni/n ) Ln (ni/n ), whe e
ni is he numbe o sequences o a pa icula OTU and n is he o al
numbe o sequences. Dominance_D: D ¼S(ni/n )2. Good’s co e age:
C¼1(ni/n ), whe e ni is he numbe o OTUs obse ed exac ly once
and n is he o al numbe o sequences.
Mic obial esponse o polya oma ics
M-E Guazza oni e al
128
The ISME Jou nal
o he high di e si y o he N sample, which esul ed
in a la ge pool o eads ha could no be assembled
(Supplemen a y Figu e S7). Fu he in o ma ion
ega ding he ob ained me agenome cha ac e is ics
is gi en in Supplemen a y Table S3. Wi h a h esh-
old o highe han 95% iden i y and an aligned
leng h o mo e han 150 bp, 91.7% ( o N), 95.2%
( o Nbs), 92.3% ( o CN1) and 89.9% ( o CN2) o
he p edic ed genes we e assigned o pa icula
gene a, he analysis o which showed esul s
compa able o hose ound in he 16S DNA assign-
men s (see Figu e 5 o CN1 and CN2 compa ison).
App oxima ely 89% o he p edic ed genes (63 974
in o al) we e assigned o COG p o ein amilies, and
65% we e assigned o KEGG pa hways
(Supplemen a y Table S3). Among he 1538 ( o
N), 2751 ( o Nbs), 2775 ( o CN1) and 2507 ( o CN2)
KEGG o hologs de ec ed, 891 we e sha ed be ween
he wo en ichmen cul u es, and 66% o hese we e
pu a i ely a ibu ed o Ach omobac e species.
We i s calcula ed he o e ep esen a ion o
unc ions be ween me agenomes; o ha , we
applied a z- es o independen p opo ions p o-
posed by Li (2009) o s a is ically analyse he
changes o he unc ional ca ego ies be ween sam-
ples. I should be no iced ha he compa ison was
es ic ed o samples Nbs, CN1 and CN2 because
he lowe numbe o py o sequences and COGs in
sample N may ha e impac on he compa a i e
esul . We ound a a he s able dis ibu ion o he
unc ional ca ego ies composi ion be ween he sam-
ples wi h signi ican di e en con ibu ion o only
238 ou o 3293 COGs (Supplemen a y Table S4).
Among hem, 56/23 did show a signi ican inc ease
in he bio-s imula ed soil as compa ed wi h he
CN1/CN2 en ichmen s, 58/26 inc eased in CN1 as
compa ed wi h Nbs/CN2 and 41/55 we e o e -
ep esen ed in CN2 as compa ed wi h Nbs/CN1.
O e all, we obse ed ha bo h bio-s imula ed com-
muni ies we e pa icula ly en iched in COGs ela ed
o ‘Replica ion and epai ’ and ‘T ansla ion’ (29 s 4
dis inc COGs in o al in CN1). O e ep esen a ion
o such unc ions is ypical o mic obial commu-
ni ies de eloped unde e y dynamic en i onmen al
condi ions, such as bio-s imula ion; he addi ion o
di e en ni a e compounds may ha e s imula ed a
compe i ion be ween as -g owing o ganisms and
o ganisms capable o me abolising poly-a oma ics.
By con as , i is no ewo hy ha COGs ela ed o
‘T ansc ip ion’ we e mos likely cha ac e is ics o
en ichmen cul u es (13 and 7 COG en iched in
CN1 and CN2, espec i ely), whe eas only wo
(COG0789 and COG1974) we e en iched in Nbs
(Supplemen a y Table S4). This sugges s a plausible
scena io in which he s ess caused by a oma ic
compounds du ing he naph halene cul i a ion led
o an en ichmen o genomes con aining ansc ip-
ional elemen s ha could be equi ed o s ess
endu ance (Domı
´nguez-Cue as e al., 2006) and/o
ac i a ion o unc ions equi ed o subs a e up ake.
Ten dis inc COGs wi hin he ‘Cell wall, memb ane,
en elop biogenesis’ ca ego y we e ound o e ep e-
sen ed in CN1, a numbe much highe ha hose
ound in Nbs ( ou COGs) and in CN2 ( wo COGs)
(Supplemen a y Table S4). In addi ion, CN1 was
also pa icula ly en iched in ‘ABC T anspo Sys-
ems’ by meaning o genes dis ibu ed in 11 dis inc
o e ep esen ed COGs, o which only h ee and wo
we e ound in Nbs and CN2, espec i ely. COG0318
ace yl-CoA syn he hase was also associa ed o CN1
communi y; enzymes o he COG0318 ca alyse he
o ma ion o ace yl-CoA om ace a e, sugges ing
ha ace a e may be a majo end p oduc used o
p oduce ace yl-CoA eeding he K ebs pa hway in
membe s o CN1 (in ag eemen wi h p o eomic da a
ha will be discussed below). By con as , CN2 was
pa icula ly en iched in genes coding p o eins
Figu e 5 Rela i e abundances and dis ibu ions o gene coding
sequences in he CN1 and CN2 communi ies based on axonomic
bins o bac e ial 16S DNA gene sequences (OPUs) (blue),
axonomic bins o me agenome-de i ed genes encoding p o eins
ha could be assigned a axonomic anno a ion ( ed) and
axonomic ca ego ies o p o eins ha we e iden i ied in he
me ap o eomes (g een). As shown, he me agenome and he 16S
DNA analyses yielded cong uen esul s. Fu he , only a mino
numbe o bac e ial membe s seemed o be me abolically ac i e
in bo h communi ies, by meaning o exp essed p o eins binned
o hem.
Mic obial esponse o polya oma ics
M-E Guazza oni e al
129
The ISME Jou nal
con aining Fe–S clus e s, namely COG3210 (la ge
exo-p o eins in ol ed in haem u ilisa ion o adhesion),
COG1049 (aconi ase B), COG0348 (poly e edoxin) as
well as COG4773 (ou e memb ane ecep o o e ic
cop ogen and e ic- hodo o ulic acid) (Supplemen a y
Table S4), indica ing i on being pa icula ly ele an
o CN2 communi y membe s. Taken oge he , he
mos no able obse a ions d awn om all hese da a
se s a e ha bio-s imula ed communi ies appea o
be ex ensi e ese oi s o genes in ol ed in eplica-
ion, epai and ansla ion and i on adhesion
and u ilisa ion whe eas bac e ial componen s o
non-s imula ed communi ies seems o be mos
ac i e in cell wall, memb ane and en elop biogen-
esis; in addi ion, en ichmen mos likely a ou ed
ansc ip ional e en s.
The ob ained co e age was su icien o p oduce a
u he pa ial heo e ical me abolic econs uc ion
(including bo h exclusi e and common capabili ies) o
he ae obic a oma ic ca abolic ou es in he in es iga ed
communi ies; al hough hese econs uc ions we e
incomple e and likely ep esen composi e cell
ne wo ks, he in o ma ion ob ained may be su i-
cien o achie e a be e unde s anding o how he
ou communi ies beha e ega ding hei biodeg a-
da ion capabili ies on a genomic scale. To do ha , a
p ope unc ional assignmen o he p edic ed genes
was pe o med using an in-house da abase con ain-
ing p o ein sequences wi h biochemical unc ions
shown o be in ol ed in biodeg ada ion ( o de ails
see Supplemen a y Ma e ials and me hods). Acco d-
ing o his p o ocol, 428 (N: 38; Nbs: 115; CN1: 132;
CN2: 143) open- eading- ame agmen s we e iden-
i ied as showing close sequence simila i y o genes
ha encode enzymes known o be in ol ed in he
ae obic me abolism o a oma ics ia di- and i-
hyd oxyla ed in e media es and, acco dingly, p e-
sump i e unc ions we e assigned. The o e all
gene ea u es and p esump i e ca abolic ou es
mos likely associa ed o hem, a e shown in
Figu e 6 Po en ial ae obic deg ada ion ne wo ks o a oma ics ia di- and i-hyd oxyla ed in e media es in he ou in es iga ed communi ies.
The colou code used o he espec i e pa hways is as ollows: black, all samples; g een, N and Nbs (speci ic o soil communi ies); blue, CN1
(N) (speci ic o non bio-s imula ed communi ies]; ed, CN2 (Nbs) (speci ic o bio-s imula ed communi ies). Illus a ions we e c ea ed by
ChemD aw g aphic p og amme Chem D aw Ul a 8.0 (Camb idgeSo ; h p://www.camb idgeso .com/) based on subs a e speci ici y o
enzymes lis ed in Supplemen a y Table S5, and he co esponding me abolic pa hways we e es ablished based on bibliog aphic eco ds. I
should be no iced ha he iden i ica ion o a pa icula ac i i y in CN1 and CN2 implies i s p esence in N and Nbs, espec i ely, al hough hei
signa u es we e no iden i ied in he la e mos likely owing o hei highe biodi e si y and low co e age o he me agenome. Acco dingly,
hose ac i i ies we e indica ed as CN1 (N) and CN2 (Nbs).
Mic obial esponse o polya oma ics
M-E Guazza oni e al
130
The ISME Jou nal