Ma a MARIALVA
Disse a ion p esen ed o ob ain he Ph.D deg ee in E olu iona y
Biology
Ins i u o de Tecnologia Química e Biológica An ónio Xa ie | Uni e sidade No a de Lisboa
Oei as, No embe , 2016
E ec s o en i onmen al and gene ic
ac o s on ansposable elemen ac i i y
Resea ch wo k coo dina ed by:
!
Aos meus pais…

!
3!
The esea ch s udy desc ibed in his hesis was inancially suppo ed by
Fundação pa a a Ciência e a Tecnologia (FCT) ellowship
SFRH/BD/51882/2012. This hesis was p in ed wi h suppo om Ins i u o
Gulbenkian de Ciência.
!
"!
The s udy desc ibed in his Ph.D. hesis was pe o med a Ins i u o
Gulbenkian de Ciência (IGC), Oei as, Po ugal and was inancially
suppo ed by Fundação pa a a Ciência e a Tecnologia (FCT), Po ugal
(SFRH/BD/51176/2010).
The wo k on his hesis esul ed in he ollowing publica ion and some
w i en con en o he hesis was in oduced in o he ollowing publica ion:
Yilmaz, B., Po ugal, S., T an, T. M., Gozzelino, R., Ramos, S., Gomes, J., e
al. (2014). Gu Mic obio a Elici s a P o ec i e Immune Response agains
Mala ia T ansmission. Cell, 159(6), 1277–1289. doi:10.1016/j.cell.2014.10.053
!
5!
ROAD TO EXTINCTION
“ca y he weigh o c ush unde i ”
Un il we a e no mo e (B ea he)
Moonspell

!
13!
Summa y
He i able pheno ypic a ia ion is he aw ma e ial o e olu ion o occu h ough
na u al selec ion. I is a pe asi e p ope y o li ing o ganisms and, in o de o
unde s and e olu iona y p ocesses, we mus s udy he ac o s ha media e he
p oduc ion o he i able pheno ypic, and he e o e geno ypic a ia ion.
T ansposable elemen s (TEs) a e mobile DNA sequences ha cons i u e a la ge
p opo ion o mos euka yo ic genomes and ha can d i e he p oduc ion o
adap i e gene ic a ia ion. The e a e many s udies on ansposon mobili y, bu
hese gene ally include only ew en i onmen al condi ions, TEs and gene ic
backg ounds. As a consequence, we s ill know e y li le abou he impac o
gene ic and en i onmen al ac o s ha can a ec ansposon dynamics in na u al
popula ions. He e, I p oposed o o e come hose limi a ions by analysing he
impac o a mul i ude o ac o s on ansposon mobili y using D osophila
melanogas e as a model o ganism.
To iden i y loci ca ying na u al allelic a ian s a ec ing TE copy-numbe , we
pe o med a genome-wide associa ion s udy, using a mapping panel o D.
melanogas e wild-de i ed geno ypes (Chap e 2). We hen selec ed ou D.
melanogas e s ocks in ec ed wi h Wolbachia – a ma e nally ansmi ed α-
P o eobac e ia, which ende s he hos esis an o in ec ion by RNA i uses – o
es he e ec o symbio ic bac e ia on he exp ession o ansposons ha
esemble e o i uses. We u he asked whe he TE ac i i y in he p esence o
he endosymbion was in luenced by empe a u e and hos gene ic backg ound,
hus in es iga ing gene ic-by-en i onmen -by-en i onmen (GxExE) in e ac ions
(Chap e 3). This analysis was ex ended o o he abio ic ac o s; no ably,
exposu e o oxida i e s ess and hea y me als in he die , and hei in e ac ion
wi h he p esence o Wolbachia (ExE in e ac ions) (Chap e 4).
E en hough he piRNA pa hway cons i u es he hallma k o o ganismsʼ
p o ec ion agains TE ac i i y in he ge mline, ac ing h ough he p oduc ion o
small RNAs ha in e ac wi h PIWI amily p o eins (piRNAs), we we e only able
o ind ew piRNA pa hway loci associa ed wi h ansposon copy-numbe
(Chap e 2). Addi ionally, his pa hway was also ound o be obus o
!
14!
en i onmen al pe u ba ions (Chap e 4). Based on he analysis o he
associa ed loci and on he TE esponses o pe u ba ion, we p opose ha
changes in ansc ip ion ac o ac i i y could be a he basis o geno ype- and
en i onmen al-speci ic exp ession o ansposons (Chap e 2, 3 and 4).
To u he assess he impac o Wolbachia in he con ex o TE in asion o
na u al popula ions, and gi en he known co ela ion be ween a ecen
eplacemen o Wolbachia s ains and he sp ead o he TE P-elemen , we
cha ac e ized a panel o 86 na u al geno ypes acco ding o he p esence o
Wolbachia and suscep ibili y o P-elemen in asion (Chap e 5). This analysis
sugges ed ha he p esence o he endosymbion migh ha e con ibu ed o
p omo e P-elemen sp ead in na u al popula ions h ough bene icial and
geno ype-speci ic e ec s on ecundi y and/o pe cen age o eclosion.
In summa y, his hesis p o ides no el insigh abou he complexi y o
in e ac ions es ablished be ween TEs, en i onmen al condi ions and hos
geno ypes. We also show ha inc eased TE exp ession in o a ies can
po en ially lead o no el inse ions in he nex gene a ion and ha egula ion a
addi ional s eps o he ansposon “li e cycle” can add an ex a laye o
complexi y in he abili y o p oduce gene ic a ia ion.
!
15!
Resumo
Toda a a iação eno ípica que é he dada encon a-se codi icada no genoma e
cons i ui uma p op iedade uni e sal de odos os sis emas biológicos. Essa
a iação pode se il ada e/ou induzida po de e minadas condições ambien ais.
Sendo c ucial pa a e olução, o seu es udo é de e minan e pa a comp eende os
p ocessos e olu i os.
T ansposões são sequências de DNA com p op iedades mó eis que compõem
uma impo an e p opo ção dos genomas de euca io as, e que in e êm na
p odução de a iação gené ica adap a i a. Embo a exis am es udos ele an es
que desc e em a mobilidade des es elemen os, os ac o es gené icos e
ambien ais que a i am ansposões são ainda amplamen e desconhecidos. Es a
ese em como p incipal objec i o comp eende a dinâmica de ansposição em
di e sos genó ipos de D osophila melanogas e , e como es a é a e ada pela
exposição a ac o es bió icos e abió icos.
Em p imei o luga , es ámos a associação en e o núme o de cópias de
ansposões e a a iação alélica que seg ega em populações na u ais de D.
melanogas e , que exis em num painel de linhas es abelecidas no labo a ó io
(Capí ulo 2). Des as linhas, o am selecionadas pa a pos e io análise algumas
que con inham in ecção com Wolbachia – uma α-P o eobac e ia ansmi ida
ma e nalmen e e que con e e ao hospedei o esis ência a in ecções i ais –
pe mi indo assim o es udo da in e ação en e o endossimbion e, a empe a u a e
o genó ipo (AxAxG) do hospedei o na modulação da exp essão de ansposões
que pa ilham semelhanças com e o í us (Capí ulo 3). Pa alelamen e, e de
o ma a en iquece o es udo com ou os ac o es abió icos, oi ambém incluída
na análise a in e ação en e a p esença de Wolbachia e a espos a a s esse
oxida i o ou a die as icas em me ais pesados (AxA) (Capí ulo 4).
Embo a o p ocesso de p odução de piRNAs (pequenos RNAs que econhecem
sequências de ansposões e que in e agem com p o eínas PIWI) cons i ua o
p incipal modo de ep essão de elemen os mó eis na linha ge minal de
euca io as, obse ámos que poucos alelos em genes elacionados com es e
mecanismo se encon am associados ao núme o de cópias de ansposões
!
16!
(Capí ulo 2). A análise de possí eis e ei os ambien ais ambém e elou que
es e mecanismo é obus o a pe u bações bió icas e abió icas (Capí ulo 4).
Tendo em con a a al a de ligação en e o p incipal mecanismo de ep essão de
ansposões e a espos a des es elemen os a ac o es gené icos e ambien ais,
p opomos que al e ações na a i idade de ac o es de ansc ição possam es a
na base da espos a dos ansposões obse ada, sendo es a especí ica pa a os
ac o es es ados (Capí ulo 2, 3 e 4).
Pa a es uda o impac o do endossimbion e Wolbachia du an e o p ocesso de
in asão de ansposões em populações na u ais, i emos em con a a co elação
es abelecida en e a subs i uição de genó ipos de Wolbachia e a in asão do
ansposão P em populações de D. melanogas e (Capí ulo 5). A nossa análise
de 86 genó ipos na u ais, a aliando a p esença de Wolbachia e a sua
suscep ibilidade à in asão de elemen os P, suge e que a p esença do
endossimbion e pode e p omo ido a in asão des e elemen o, ao con e i
e ei os bené icos na ecundidade e pe cen agem de eclosão, e ei os que são
especí icos pa a cada genó ipo.
Em suma, es a ese e elou in e ações complexas, es abelecidas en e
ansposões, condições ambien ais e genó ipos, onde cada elemen o mó el
esponde de o ma especí ica a cada ac o es ado. Os nossos esul ados
suge em ambém que um aumen o da exp essão de ansposões nas gónadas
pode á conduzi a no as inse ções, que se ão ansmi idas à ge ação seguin e,
e ealçam as al e ações ambien ais como um ac o impo an e na p odução de
a iação gené ica he edi á ia mediada po ansposição.
!
17!
Chap e I
Gene al In oduc ion

!
19!
En i onmen al condi ions shape e olu ion by ac ing bo h as an agen o na u al
selec ion, in ol ed in he so ing o gene ic a ia ion, and as a ac o in luencing
he p oduc ion o no el gene ic a ia ion, h ough e ec s on ecombina ion and
mu a ion a es. Du ing he pas ew yea s, much a en ion has been de o ed
owa ds he explo a ion o he gene ic basis o adap i e e olu ion1,2, and
impo an insigh s ha e eme ged om expe imen al e olu ion s udies ha
a emp o link h ee key componen s o e olu ion: geno ype, pheno ype and
i ness3–7. Howe e , mos o hose s udies ypically use a he simple
en i onmen s, wi h changes in ew a iables, ha all sho o p ope ly
ep esen ing many o he ecological in e ac ions ha media e adap a ion in wild
popula ions8. Also, ew s udies succeed a connec ing a ia ion in pheno ypic
ai s wi h a ia ion in allele equencies as en i onmen al condi ions change in
na u e1,9,10. New echnologies p o ide ools ha will allow be e explo a ion o
he gene ics o adap a ion o changing en i onmen s by allowing he as gene ic
mapping o pheno ypic ai s, and genome-wide analyses o popula ions e ol ing
in he wild8,11–15. Those app oaches can help o p ope ly es ablish he link
be ween geno ype, pheno ype and i ness, and o unde s and how and why
allele equencies change in na u e.
The e olu iona y success o a popula ion depends on a ailable gene ic
a iabili y ha is ini ially p oduced h ough mu agenesis and ea anged by
ecombina ion. In aspeci ic a ia ion in mu a ion and ecombina ion a es can
also be subjec ed o na u al selec ion as popula ions adap o a new
en i onmen 16,17. In addi ion, he e is pheno ypic plas ici y o hose ai s, as
in e ed by he en i onmen al abili y o in luence he equency o mu a ion
and/o ecombina ion in same geno ype18–26. En i onmen al ac o s such as
empe a u e and nu i ional in ake a e impo an playe s in he es ablishmen o
no el gene ic a ia ion18–20,24–26.
!
20!
1.1 Response o s ess ul en i onmen s
En i onmen al he e ogenei y is a pe asi e p ope y o all ecological niches, and
condi ions a e called s ess ul when hey esul in educed indi idual i ness27.
Subop imal condi ions caused by ei he bio ic o abio ic ac o s can be g ouped
in o ch onic, mild o acu e based on he se e i y and/o du a ion o he s ess28.
Mos s udies abou he e ec s o dynamic en i onmen s ocus on acu e s esses
ha equi e immedia e and p ecise cellula esponses.28–30.
In o de o ci cum en s ess and e-es ablish homeos asis, o ganisms need o
sense s ess and induce p ope signal ansduc ion pa hways ha ac i a e
e ec o p ocesses28. Signalling pa hways end o be mo e conse ed han
sensing and e ec o mechanisms, and gene ally espond o a a ie y o
en i onmen al ac o s28,31. Fo example, mi ogen-ac i a ed p o ein kinase
(MAPK) amily elemen s a e ac i a ed by se e al s imuli such as gamma ays,
hea shock, in ec ion, ul a iole ligh , osmo ic and oxida i e s ess28,31–37. How
speci ici y is ob ained when he same pa hway esponds o di e en s esses
may ely on 1) he di e en ial ac i a ion, in space and ime, o he MAPK
p o eins, and 2) he simul aneous induc ion o signal ansduc ion pa hways ha
a e exclusi e o each en i onmen and may c ea e a s imulus-speci ic ne wo k o
esponsi e elemen s28.
1.1.1 Mechanisms o esponse o empe a u e
In he wild, popula ions a e exposed o empo al and spa ial luc ua ions in
empe a u e ha may a ec o ganismal pe o mance and i ness38–41. Indi iduals
e ol ed di e en mechanisms o sense he mal condi ions and igge e ec o
esponses upon exposu e o sub-op imal empe a u es42–45. The ela i e
con ibu ion o di e en sensing mechanisms depends on he species, p e ious
exposu e and/o he ampli ude o he mal pe u ba ion42,46–48.
Hea shock p o eins (Hsps) a e he hallma k o empe a u e-media ed cellula
p ocesses and p omo e p ope p o ein olding in esponse ex eme insul s such
as hea , oxida i e s ess, in ec ion and p esence o hea y me als49. In esponse
!
21!
o he mal pe u ba ion, Hsps can unde go con o ma ional changes in a
he mosenso domain ha a e necessa y o ac i a e hei chape one ac i i y42. In
addi ion, Hsps can also indi ec ly de ec changes in empe a u e ia hei high
binding a ini y o hea -induced dena u ed p o eins45. The ec ui men o Hsps o
empe a u e-media ed p o ein agg ega es esul s in he ac i a ion o hea shock
ac o s (HSFs) ha con e sely elici he ansc ip ion o Hsp genes (Figu e
1.1)50–52.
Figu e 1.1. E ec o p ocesses media ing
esponse o empe a u e s ess. In he
absence o he mal pe u ba ions, hea
shock ac o (HSF) is s abilized in he
cy osol h ough he in e ac ion wi h hea
shock p o eins (Hsps)49 (A). Upon
empe a u e shock, se e al p o eins acqui e
non-na i e con o ma ions ha ul ima ely
esul in he o ma ion o agg ega es. As
Hsps a e ec ui ed o media e p ope p o ein
olding, HSF is eleased and ac i a ed by
o ming homo ime s53–57. HSF ime s a el
o he nucleus and induce ansc ip ion o
empe a u e esponsi e genes, including
Hsps, ha con ain hea shock cis- egula o y
elemen s (HSE)58. A he same ime, he
mi ogen-ac i a ed p o ein kinase (MAPK)
pa hway is induced59 (B). This media es
HSF phospho yla ion and alle ia es i s
in luence in gene exp ession. Addi ionally,
as cy osolic p o eins acqui e p ope
con igu a ion, Hsps a e eleased and
con ibu e o e- ep essing HSF (C).
!
28!
Figu e 1.4. Mechanisms o TE jumping. DNA elemen s (A) mo e ia a cu -and-pas e
mechanism mainly media ed by ansposase ac i i y. Each DNA elemen encodes own
ansposase ha speci ically ecognizes e minal in e ed epea s a each TE end (yellow
a ows), media ing DNA clea age and ansposon excision. T ansposase also ca alyzes
TE in eg a ion in a new genomic a ea. LTR (B) and non-LTR (C) elemen s a e
e o ansposons and mo e ia an RNA in e media e. LTR e o ansposons a e lanked
by long e minal epea s (yellow a ows) and code o : p o ease, i al pa icle coa (Gag),
e e se ansc ip ase and in eg ase p o eins. Hos ansc ip ional machine y is used o
exp ess LTR elemen s and a i al-like pa icle is o med in he cy oplasm wi h Gag
isola ing TE-exp essed p o eins and mRNA. Re e se ansc ip ion occu s inside hose
pa icles. A e , cDNA is anspo ed o he nucleus whe e in eg ase ca alyzes ansposon
inse ion (B). Non-LTR elemen s code o wo enzymes: endonuclease (ORF1 p o ein)
and e e se ansc ip ase (ORF2 p o ein). cDNA p oduc ion occu s a he nucleus while
e e se ansc ip ase uses mRNA as empla e and DNA o e hang (p oduced by
endonuclease ac i i y) as p ime (C). Aside de ailed in o ma ion o he e e se
ansc ip ion p ocess, li le is known abou o he s eps o non-LTR ansposi ion122,126.
This igu e was adap ed om Le in and Mo an (2011)122.
AB C

!
29!
DNA ansposons a e ancien TEs ound in p oka yo es and almos all
euka yo es ha anspose wi hou p oducing an RNA in e media e127. Those a e
mos ly lanked by e minal in e ed epea s (TIRs) (Figu e 1.4A) ha can a y in
leng h and sequence, and be used o classi y DNA elemen s128,129. In plan s, he
i s bases o TIRs a e composed by highly conse ed CACTA mo i s ha a e
eplaced by CCC sequences in animals and ungi130.
In sho , DNA elemen s ha code o ansposase mobilize by a cu -and-pas e
mechanism and can only inc ease in copy numbe when ansposing du ing
DNA eplica ion o by sis e ch oma id ecombina ion131. Al e na i ely, hey can
also exploi he gap epai machine y upon excision while he old inse ion is
es o ed using lanking homologous egions as p ime s o DNA syn hesis and
he sis e ch oma id con aining TE inse ion as empla e132.
1.2.2 TE ac i i y upon en i onmen al pe u ba ions
Ou knowledge abou wha igge s TE ansposi ion is s ill a he incomple e.
Ba ba a McClin ock p oposed ha , mobile elemen s would unc ion as genomic
a chi ec s by becoming ac i e in esponse o en i onmen al s esses133. She
belie ed ha jumping o ansposons could elici he exp ession o genes
necessa y o ci cum en s ess. Only ecen ly ha e disco e ies unco e ing a link
be ween en i onmen al ac o s and TE ac i i y134–139 con i med he iew. This
includes da a demons a ing ansc ip ional induc ion o plan Tn 1 and T o1 LTR
elemen s in esponse o se e al bio ic (inocula ion wi h i al, bac e ia and ungal
pa hogens) and abio ic (me hyl jasmona e, CuCl2 and salicylic acid)
condi ions140–143. Also, he p esence o cis-ac ing elemen s wi hin he 5ʼ
sequence o LTRs was p o en o media e TE ac i a ion du ing en i onmen al
change144. Con e sely, Tn 1 inse ions ound in se e al obacco plan s we e
shown o p omo e chlo ide- esis an pheno ypes123,145.
I is impo an o no e ha he li e cycle o e o ansposons includes o he s eps
han ansc ip ion such as ansla ion, e e se ansc ip ion and in eg a ion, and
ha en i onmen al ac o s can a ec any o hose le els o in luence
ansposi ion. Fo example, change in he phospho yla ion s a e o Ty5 in eg ase
!
30!
ha is media ed by nu ien dep i a ion a ec s TE inse ion p e e ence wi hou
in luencing TE exp ession146,147. When in eg ase is phospho yla ed, Ty5
p e e en ially inse s in o he e och oma in minimizing he chances o damage
coding sequences. A e exposu e o nu ien dep i a ion, in eg ase is
dephospho yla ed and Ty5 is inse ed in o a eas o ac i e gene exp ession. A
mechanism whe eby en i onmen al pe u ba ion can egula e TE ac i i y o
in eg a ion could ha e been selec ed o i TE deploymen is some imes, bu no
always, ad an ageous. The s udy o TE- esponse o en i onmen al luc ua ions
in o he model o ganisms is needed o unde s and he conse a ion and/o
gene ali y o such mechanisms.
The e ha e been se e al a emp s o demons a e induc ion o mobile elemen s
associa ed wi h s ess ul condi ions in D. melanogas e 148. These include
ci cums an ial e idence o TE esponse o hea 149–151, e hanol, adia ion, and
i al in ec ion152,153 hough esul s we e s ongly dependen on he ansposon
and gene ic backg ound analyzed. While he e is e idence o he impo ance o
en i onmen al ac o s in he egula ion o TE ac i i y, empi ical esea ch on o he
hos o ganisms and ansposon ypes is needed o elucida e he occu ence and
impo ance o such s a egies.
1.2.3 TEs as sou ce o gene ic a ia ion
The amous colou pheno ypes o co n ke nels i s desc ibed by Ba ba a
McClin ock esul om TE ins abili y and inse ion in essen ial pigmen a ion
pa hway genes154. In ha s udy, ansposi ion occu ed du ing ke nel
de elopmen and he e o e induced non-he i able mu a ions123,154. In ac , only
TE ac i i y in he ge mline will con ibu e o he i able gene ic a ia ion, which is
wha is ele an o e olu ion. The ansposon-media ed mu a ion a e can ange
om 10-3 and 10-5 pe elemen pe gene a ion, depending on he elemen ,
whe eas he classical nucleo ide subs i u ion a e luc ua es a ound 10-8 and 10-9
pe nucleo ide pe gene a ion155. Upon ansposi ion, TEs can in luence
exp ession o neighbou ing genes, bo h a he ansc ip ional and pos -
ansc ip ional le els (Figu e 1.5). In e es ingly, se e al mobile elemen s adding
!
31!
new egula o y egions a e sca e ed along he genome and can play an
impo an unc ion in he es ablishmen o gene ic ne wo ks156,157. The p e e ence
o DNA elemen s o Hsp p omo e s in D osophila158 and he conside able
impac o LTR e o ansposons in he ansc ip ional ne wo k o human p53159
p o ide good examples o his.
Figu e 1.5. T ansc ip ional and pos - ansc ip ional modi ica ions media ed by TE
inse ions. Gene exp ession and/o unc ion can be a ec ed by no el TE inse ions.
Those can be media ed by he induc ion o ansc ip ional (A-E) o pos - ansc ip ional
modi ica ions (F-I). Gi en ha ansposons con ain hei own p omo e s and cis- egula o y
egions, TE inse ions ups eam o a gene can in oduce an al e na i e ansc ip ion s a
si e (A), dis up exp ession when inse ed wi hin gene egula o y elemen s (B) o add new
egula o y elemen s (C)160–166. In addi ion, TE in eg a ion can also p omo e an i-sense
ansc ip ion (D)167 o ec ui ch oma in- emodeling p o eins ha e en ually silence gene
and TE ac i i y (E)168. Con e sely, TE in eg a ion may media e pos - ansc ip ional
A
B
C
D
E
F
G
H
I
!
32!
modi ica ions by in oducing al e na i e polyadenyla ion signals (F)169,170, no el miRNA
a ge si es (G)171 o al e na i e splicing e en s (H)172,173. TE sequences can also be
inco po a ed as an al e na i e exon (p ocess called exoniza ion) ha can esul in he
ansla ion o a new p o ein iso o m, o media e mRNA deg ada ion by he nonsense-
media ed decay (NMD) i new s op codons a e inse ed (as e isk) (I)163,174. This igu e
was modi ied om Fescho e (2008)156.
The con ibu ion o TE ac i i y o he p oduc ion o gene ic a ia ion is a gene al
p ope y o li ing o ganisms. In ac , ac i e ansposons ha e been ound in he
genome assemblies o ma ine plank on175 as well as in se e al plan 176–178 and
ish179,180 species. Addi ionally, mo e de ailed s udies on he selec i e alue o
TE inse ions seg ega ing in D. melanogas e na u al popula ions e ealed ha
many a e adap i e wi h pu a i e unc ion in p ocesses such as pes icide
esis ance181,182, oxida i e s ess183, amino acid me abolism, immune esponse
o i al in ec ion and eye de elopmen 184. Mo eo e , expe imen al e olu ion in
bac e ia has shown changes in TE ac i i y and selec ion o no el inse ions
du ing adap a ion o gu en i onmen 6. In summa y, he onse o e idence o he
impo ance o mobile elemen s in adap a ion dis inguishes hose sequences as
enginee s o ansc ip ional ne wo ks and pu a i e media o s o di e si ica ion.
1.2.4 TEs and no el y
The bulk o s udies in e olu iona y biology en ail esea ch abou adap a ion,
specia ion and he o igin o no el ies – lineage speci ic ai s wi h adap i e
alue185. This las esea ch p og am ad anced he eme gence o a new
discipline (e olu iona y de elopmen al biology) ha ocuses on he gene ic basis
o in a- and in e -speci ic mo phological a ia ion186. Du ing he pas yea s, i
has become clea ha TEs a e a he ounda ion o se e al no el ai s and
hence demand u he a en ion. Among hose, esea che s ga he ed e idence
o he e olu ion o conspicuous male in colou pa e ns ha a e in ol ed in
ma ing o cichlids187, he o igin o mammal placen a188 and he di e si ica ion o
ui colou and shape189; all pheno ypes de i ing om ansposon inse ions ha
esul ed in cis- egula o y changes o neighbou ing genes. Aside hese examples
!
33!
o how pa icula TE inse ions a ec ed gene exp ession o media ed new
genomic unc ions, he e is also documen a ion o a ious “domes ica ion
e en s” ha p oduced new p o ein-coding genes om ansposases156. In
pa icula , he V(D)J ecombina ion eac ion necessa y o adap i e immuni y
elies on ansposase-de i ed RAG1 ha eme ged in jawed e eb a es190. O he
examples also ound in me azoans deal wi h he e olu ion o highly conse ed
PAX p o eins ha unc ion in eye de elopmen and cephaliza ion191 as well as
cen ome e-associa ed p o ein B (CENP-B) ha is ela ed wi h cell-cycle192.
Addi ionally, a s iking scena io o con e gence be ween dis an lineages
desc ibes he pu a i e ecu en appea ance o ansposase-de i ed p o eins
ha a e in ol ed in a - ed ligh signalling in yeas , insec s and
angiospe ms156,193,194. All in all, i is clea ha TEs a e c i ical media o s o
di e si ica ion and inno a ion.
1.3 Mechanisms o ep ess TE ac i i y
Despi e he impo ance o mobile gene ic elemen s in c ea ing adap i e a ia ion
and con ibu ing o mo phological di e si y, hey can h ea en genomic in eg i y
and educe i ness when ac i e in he ge mline195. As a esul , me azoan species
ha e e ol ed mechanisms o p omo e speci ic TE ecogni ion and guide
silencing h ough he use o small RNAs196,197. Small RNAs can be g ouped in o
h ee main classes ha di e in hei biogenesis p ocess and
biological/biochemical unc ion. Small in e e ing RNAs (siRNAs) and mic o
RNAs (miRNAs) a e ubiqui ously exp essed and de i e om double s anded
RNAs p ocessed by he RNA-clea ing enzyme Dice , whe eas he gonadal
speci ic PIWI-in e ac ing RNAs (piRNAs) a e p oduced h ough Dice -
independen mechanisms and de i e om bo h single and double s anded
RNAs198,199. Mos in o ma ion a ailable abou piRNA ac i i y comes om s udies
in D. melanogas e ha is cu en ly used as e e ence o TE-silencing sys ems
in o he species197. The main unc ion o piRNAs is o media e homology-
dependen ansposon silencing while in e ac ing wi h p o eins ha belong o he
PIWI amily – Piwi, Aube gine (Aub) and A gonau e 3 (AGO3)200–202. piRNAs

!
34!
belong o a g oup o sequences ha a e pa icula ly di e se and ansc ibed om
in e genic inac i e epe i i e elemen s ha collec i ely map o genomic clus e s.
In D. melanogas e he e a e 569 desc ibed piRNA clus e s
(h p://pi nabank.ibab.ac.in/), and only some a e exclusi ely exp essed in he
ge mline o males and emales203–205. Con e sely, o he clus e s speci ically
ep ess TE ac i i y in he soma ic issues o o a ies, whe eas some ep ess bo h
ge mline and soma ic exp ession o mobile elemen s203.
Figu e 1.6. P og essi e oocy e de elopmen in D. melanogas e o a ioles.
Oogenesis ini ia es a he ge ma ium ha is localized a he an e io ip o o a ioles.
The e, ge mline s em cells (GSC) di ide o p oduce oocy e and nu se cells; soma ic s em
cells (SSC) gi e ise o ollicula cells. When he egg chambe exi s he ge ma ium, a
laye o ollicula cells is p o ec ing he oocy e and nu se cells ha main ain a common
cy oplasm.
1.3.1 piRNA pa hway in he o a y o D. melanogas e
D osophila o a ies a e composed o mul iple o a ioles each made up o a se ies
o egg chambe s wi h an e io -pos e io p og essi e de elopmen (Figu e 1.6).
The soma ic suppo cells es ablish an in ica e ela ionship wi h he ge mline by
eeding de elopmen al signals and nu ien s in o he oocy e and nu se cells
syncy ium206. This close in e ac ion be ween soma ic ( ollicula cells) and
ge mline (oocy e and nu se cells) issue exposes he oocy e genome o h ea s
ha can de i e om bo h cell ypes. In ac , se e al ansposons a e able o
exp ess in he ge mline whe e mo emen can di ec ly d i e he i able
ansmission195,207–209, bu e o ansposons exp essed in ollicula cells can also
o m e o i al pa icles ha e icien ly in ade he oocy e210–213. To minimize TE
S age 10 S age 14S age 8
Ge ma ium S age 3 S age 5
Egg chambe
Oocy e
Nu se cells Follicula cells
GSC SSC
Egg
An e io Pos e io
!
35!
access o he emale game ocy e, he piRNA pa hway unc ions as a genome
su eillance sys em o bo h cell ypes. In e es ingly, piRNA biogenesis di e s
be ween soma ic and ge mline issue leading o dis inc silencing p ocesses
used o ep ess TEs in di e en cell ypes203.
In ollicula cells o e 90% o he piRNA pool is an isense o ansposons ha
a e speci ically ac i e in soma ic gonadal cells and belong o he gypsy amily214.
Mos a e p oduced om wo X-ch omosomal clus e s – he lamenco locus and
clus e 20A – and de i e om long single-s anded ansc ip s ha a e
p ocessed in o smalle sequences214,215. In e es ingly, he p oduc ion o hose
genomic gua dians is independen o he exp ession o ac i e ansposons and
hence cons i u es he p ima y piRNA biogenesis214. The biogenesis s ep akes
place in he cy oplasm, whe e he Piwi p o ein (only elemen o he PIWI- amily
exp essed in ollicula cells) is loaded wi h piRNAs216. Piwi-piRNA complexes
hen mig a e o he nucleus o media e ansc ip ional ep ession o mobile
elemen s217. The genomic place o TE inse ions is ecognized by
complemen a i y wi h piRNA sequences, and Piwi di ec s he accumula ion o
he e och oma ic ma ks a hose genomic places h ough di ec in e ac ion wi h
he e och oma in p o ein 1a ha is a cen al playe in he e och oma ic gene
silencing (Figu e 1.7)168.
!
36!
Figu e 1.7. P ima y piRNA biogenesis in ollicula cells. piRNA p oduc ion in o a ian
soma ic issue consis s o p ima y biogenesis esul an om p ocessing o wo main
clus e s (clus e 20A and lamenco, ed box). Those a e unidi ec ionally ansc ibed and
p ocessed in o smalle sequences (piRNAs). Piwi loaded wi h piRNAs mig a es o he
nucleus o p omo e TE ep ession by ec ui ing ch oma in- emodeling p o eins. A pu a i e
TE inse ion in he X-ch omosome is highligh ed in he blue box.
In ge mline cells he a chi ec u e o piRNA biogenesis is complex and in ol es
he p oduc ion o bo h sense and an isense small RNAs. Those o igina e om
se e al piRNA clus e s ha a e sca e ed along he genome and comp ehend an
ex ensi e collec ion o agmen ed TE sequences215. Clus e s ha a e exp essed
in he soma ic and ge mline issue o o a ies (such as he p e iously desc ibed
clus e 20A) a e unidi ec ionally ansc ibed; ge mline-speci ic clus e s (such as
he well desc ibed clus e 42AB) a e bidi ec ionally ansc ibed and p oduce
double-s anded ansc ip s ha a e ully clea ed un il piRNAs203,214. Those a e
hen loaded in o all h ee PIWI amily p o eins: Piwi and Aub bind o an isense
!
37!
and AGO3 o sense piRNAs. In addi ion o he p ima y biogenesis, piRNAs
unde go a loop o ampli ica ion ha is mos ly d i en by Aub and AGO3. Those
wo p o eins a e en iched a ound he nucleus whe e Aub-piRNA complexes
a ge and di ec TE ansc ip s o clea age. As a consequence, sense piRNAs
a e p oduced and inco po a ed by AGO3, ha has he po en ial o speci ically
igge piRNA clus e p ocessing and inc ease he abundance o an isense
piRNAs loaded in o Aub215. In pa allel, Piwi-piRNA complexes main ain unc ion
o p omo e ch oma in- emodelling changes a he loci whe e ansponsons a e
inse ed. In e es ingly, ma e nal ansmission o an isense small RNAs
associa ed wi h bo h Piwi and Aub wo ks as a ca alys o p oduce piRNAs
c ucial o ep ess TEs in he nex gene a ion207. The na u e o Aub and AGO3
engagemen o p oduce piRNAs in he ge mline named his ampli ica ion loop as
ping-pong cycle215. This mechanism ep esen s he hallma k o he ge mline
piRNA pa hway as i is conse ed in a spec um o species ha ange om
sponges o mammals (Figu e 1.8)218–221.
Figu e 1.8. P oduc ion o piRNAs in he ge mline. Se e al piRNA clus e s, including
clus e 42AB, a e bidi ec ionally exp essed in he ge mline ( ed box). As a consequence,
bo h sense and an isense piRNAs a e p oduced and loaded in o all PIWI- amily p o eins.
!
44!
ch omosome o D osophila melanogas e . Gene ics 530–532 (1926).
21. Hende son, S. A. & Edwa ds, R. G. Chiasma equency and ma e nal
age in mammals. Na u e 218, 22–28 (1968).
22. P ies , N. K., Roach, D. A. & Galloway, L. F. Ma ing-induced
ecombina ion in ui lies. E olu ion (N. Y). 61, 160–167 (2007).
23. Hussin, J., Roy-Gagnon, M. H., Gend on, R., Andel inge , G. & Awadalla,
P. Age-dependen ecombina ion a es in human pedig ees. PLoS Gene .
7, (2011).
24. Neel, J. V. A Rela ion be ween La al Nu i ion and he F equency o
C ossing o e in he Thi d Ch omosome o D osophila Melanogas e .
Gene ics 26, 506–516 (1941).
25. Bjedo , I. e al. S ess-induced mu agenesis in bac e ia. Sicence 42,
1404–1409 (2003).
26. Mulle , H. J. The Measu emen o Gene Mu a ion Ra e IN D osophila, I s
High Va iabili y, and I s Dependence Upon Tempe a u e. Gene ics 279–
357 (1928).
27. A mb us e , P. & Reed, D. H. Inb eeding dep ession in benign and
s ess ul en i onmen s. He edi y (Edinb). 95, 235–242 (2005).
28. de Nadal, E., Amme e , G. & Posas, F. Con olling gene exp ession in
esponse o s ess. Na . Re . Gene . (2011). doi:10.1038/n g3055
29. Azad, P., Zhou, D., Russo, E. & Haddad, G. G. Dis inc mechanisms
unde lying ole ance o in e mi en and cons an hypoxia in D osophila
melanogas e . PLoS One 4, e5371 (2009).
30. De Angelis, M. & Gobbe i, M. En i onmen al s ess esponses in
Lac obacillus: A e iew. P o eomics 4, 106–122 (2004).
31. Wagne , E. F. & Neb eda, A. R. Signal in eg a ion by JNK and p38 MAPK
pa hways in cance de elopmen . Na Re Cance 9, 537–549 (2009).
32. Kamada, Y., Jung, U. S., Pio owski, R. & Le in, D. E. The P o ein-
Kinase C-Ac i a ed Map Kinase Pa hway o Saccha omyces-Ce e isiae
Media es a No el Aspec o he Hea -Shock Response. Genes De . 9,
1559–1571 (1995).
33. Lemai e, B. & Ho mann, J. The Hos De ense o D osophila
melanogas e . Annu. Re . Immunol. 25, 697–743 (2007).
34. Mi le , R. Oxida i e s ess, an ioxidan s and s ess ole ance. 7, 405–410
(2002).
35. Inoue, H. e al. The C . elegans p38 MAPK pa hway egula es nuclea
localiza ion o he ansc ip ion ac o SKN-1 in oxida i e s ess esponse.
2278–2283 (2005). doi:10.1101/gad.1324805.2278

!
45!
36. Ko un, Y., Chiu, W. L., Tena, G. & Sheen, J. Func ional analysis o
oxida i e s ess-ac i a ed mi ogen-ac i a ed p o ein kinase cascade in
plan s. P oc. Na l. Acad. Sci. U. S. A. 97, 2940–2945 (2000).
37. de Nadal, E., Alepuz, P. M. & Posas, F. Dealing wi h osmos ess h ough
MAP kinase ac i a ion. EMBO Rep. 3, 735–740 (2002).
38. Thomas, M. B. & Blan o d, S. The mal biology in insec -pa asi e
in e ac ions. T ends Ecol. E ol. 18, 344–350 (2003).
39. Ragland, G. J. & Kingsol e , J. G. The e ec o luc ua ing empe a u es
on ec o he m li e-his o y ai s: Compa isons among geog aphic
popula ions o Wyeomyia smi hii. E ol. Ecol. Res. 10, 29–44 (2008).
40. Lazza o, B. P., Flo es, H. a., Lo igan, J. G. & You h, C. P. Geno ype-by-
en i onmen in e ac ions and adap a ion o local empe a u e a ec
immuni y and ecundi y in D osophila melanogas e . PLoS Pa hog. 4, 1–9
(2008).
41. Coline , H., Sinclai , B. J., Ve non, P. & Renaul , D. Insec s in Fluc ua ing
The mal En i onmen s. Annu. Re . En omol. 60, 123–140 (2015).
42. F anzmann, T. M., Menho n, P., Wal e , S. & Buchne , J. Ac i a ion o he
Chape one Hsp26 Is Con olled by he Rea angemen o I s
The mosenso Domain. Mol. Cell 29, 207–216 (2008).
43. Hamada, F. N. e al. An in e nal he mal senso con olling empe a u e
p e e ence in D osophila. Na u e 454, 217–220 (2008).
44. McClung, C. R. & Da is, S. J. Ambien he mome e s in plan s: F om
physiological ou pu s owa ds mechanisms o he mal sensing. Cu . Biol.
20, 1086–1092 (2010).
45. Rich e , K., Haslbeck, M. & Buchne , J. The Hea Shock Response: Li e
on he Ve ge o Dea h. Mol. Cell 40, 253–266 (2010).
46. Dhaka, A., Viswana h, V. & Pa apou ian, A. T p Ion Channels and
Tempe a u e Sensa ion. Annu. Re . Neu osci. 29, 135–161 (2006).
47. Klinke , B. & Na be haus, F. Mic obial he mosenso s. Cell. Mol. Li e Sci.
66, 2661–2676 (2009).
48. Fede , M. E. & Ho mann, G. E. Hea -shock p o eins, molecula
chape ones, and he s ess esponse: E olu iona y and Ecological
Physiology. Annu. Re . Physiol 61, 243–282 (1999).
49. Ake el , M., Mo imo o, R. I. & Sis onen, L. Hea shock ac o s: in eg a o s
o cell s ess, de elopmen and li espan. Na . Re . Mol. Cell Biol. 11,
545–555 (2010).
50. Gasch, a P. e al. Genomic exp ession p og ams in he esponse o
yeas cells o en i onmen al changes. Mol. Biol. Cell 11, 4241–4257
!
46!
(2000).
51. Caus on, H. e al. Remodeling o yeas genome exp ession in esponse
o en i onmen al changes. Mol Biol Cell 12, 323–337 (2001).
52. Shamo sky, I. & Nudle , E. New insigh s in o he mechanism o hea
shock esponse ac i a ion. Cell. Mol. Li e Sci. 65, 855–861 (2008).
53. Shi, Y., Mosse , D. D. & Mo imo o, R. I. Molecula chape ones as HSF1-
speci ic ansc ip ional ep esso s. Genes De . 12, 654–666 (1998).
54. Bale , R., Welch, W. J. & Voellmy, R. Hea shock gene egula ion by
nascen polypep ides and dena u ed p o eins: hsp70 as a po en ial
au o egula o y ac o . J. Cell Biol. 117, 1151–1159 (1992).
55. Ab a aya, K. & Biology, C. The human hea shock p o ein hsp70
in e ac s wi h HSF , he ansc ip ion ac o ha egula es hea shock
gene exp ession iii ; ii ! i ! iiiii ii. 1153–1164 (1992).
doi:10.1101/gad.6.7.1153
56. Goodson, M. L. & Sa ge, K. D. Hea -inducible DNA binding o pu i ied
hea shock ansc ip ion ac o . J. Biol. Chem. 270, 2447–2450 (1995).
57. La son, J. S., Schue z, T. J. & Kings on, R. E. In i o ac i a ion o
pu i ied human hea shock ac o by hea . Biochemis y 34, 1902–1911
(1995).
58. Pa ke , C. S. & Topol, J. A D osophila RNA polyme ase II ansc ip ion
ac o con ains a p omo e - egion-speci ic DNA-binding ac i i y. Cell 36,
357–369 (1984).
59. Chu, B., Soncin, F., P ice, B. D., S e enson, M. A. & Calde wood, S. K.
Sequen ial Phospho yla ion by Mi ogen-ac i a ed P o ein Kinase and
Glycogen Syn hase Kinase 3Rep esses T ansc ip ional Ac i a ion by
Hea Shock Fac o -1. J. Biol. Chem. 271, 30847–30857 (1996).
60. Raison, J. K. The in luence o empe a u e-induced phase changes on
he kine ics o espi a o y and o he memb ane-associa ed enzyme
sys ems. J. Bioene g. 4, 285–309 (1973).
61. Phillips, R. S. Tempe a u e modula ion o he s e eochemis y o
enzyma ic ca alysis: P ospec s o exploi a ion. T ends Bio echnol. 14,
13–16 (1996).
62. B auchi, S., O io, P. & La o e, R. Clues o unde s anding cold sensa ion:
he modynamics and elec ophysiological analysis o he cold ecep o
TRPM8. P oc. Na l. Acad. Sci. U. S. A. 101, 15494–15499 (2004).
63. Voe s, T. e al. The p inciple o empe a u e-dependen ga ing in cold-
and hea -sensi i e TRP channels. Na u e 430, 748–754 (2004).
64. Rosenzweig, M. e al. The D osophila o holog o e eb a e TRPA1
!
47!
egula es he mo axis. Genes De . 19, 419–424 (2005).
65. Finkel, T. & Holb ook, N. J. Oxidan s, oxida i e s ess and he biology o
ageing. Na u e 408, 239–247 (2000).
66. Buchon, N., Sil e man, N. & Che y, S. Immuni y in D osophila
melanogas e - om mic obial ecogni ion o whole-o ganism physiology.
Na . Re . Immunol. 14, 796–810 (2014).
67. Be e idge, D. J. Wha Is Oxida i e S ess? Me abolism 49, 3–8 (2000).
68. Finkel, T. Oxidan signals and oxida i e s ess. Cu . Opin. Cell Biol. 15,
247–254 (2003).
69. Allen, R. G. & T esini, M. Oxida i e s ess and gene egula ion. F ee
Radic. Biol. Med. 28, 463–499 (2000).
70. Zhong, M., O osz, a & Wu, C. Di ec sensing o hea and oxida ion by
D osophila hea shock ansc ip ion ac o . Mol. Cell 2, 101–108 (1998).
71. Singh, V. & Aballay, A. Hea Shock and Gene ic Ac i a ion o HSF-1
Enhance Immuni y o Bac e ia. 5, 2443–2446 (2006).
72. Bu don, R. H. Supe oxid and hyd ogen pe oxide in ela ion o mammalian
cell p oli e a ion. 18, 775–794 (1995).
73. Beckman, K. B. & Ames, B. N. The ee adical heo y o aging ma u es.
Physiol. Re . 78, 547–581 (1998).
74. Sai oh, M. e al. Mammalian hio edoxin is a di ec inhibi o o apop osis
signal- egula ing kinase (ASK) 1. EMBO J. 17, 2596–2606 (1998).
75. Adle , V. e al. Regula ion o JNK signaling by GSTp. EMBO J. 18, 1321–
34 (1999).
76. Lee, S. R., Kwon, K. S., Kim, S. R. & Rhee, S. G. Re e sible inac i a ion
o p o ein y osine phospha ase 1B in A431 cells s imula ed wi h
epide mal g ow h ac o . J. Biol. Chem. 273, 15366–15372 (1998).
77. Wang, X., McCullough, K. D., F anke, T. F. & Holb ook, N. J. Epide mal
g ow h ac o ecep o -dependen Ak ac i a ion by oxida i e s ess
enhances cell su i al. J. Biol. Chem. 275, 14624–14631 (2000).
78. Polyak, K., Xia, Y., Zweie , J. L., Kinzle , K. W. & Vogels ein, B. A model
o p53-induced apop osis. Na u e 389, 300–305 (1997).
79. Johnson, T. M., Yu, Z. X., Fe ans, V. J., Lowens ein, R. a & Finkel, T.
Reac i e oxygen species a e downs eam media o s o p53-dependen
apop osis. P oc. Na l. Acad. Sci. U. S. A. 93, 11848–11852 (1996).
80. de Ba y, H. A. Die E scheinung de Symbiose. (1879).
81. Relman, D. A. ʻTil dea h do us pa ʼ: coming o e ms wi h symbio ic
ela ionships. Na . Re . Mic obiol. 6, 721–724 (2008).
!
48!
82. We en, J. H., Baldo, L. & Cla k, M. E. Wolbachia: mas e manipula o s o
in e eb a e biology. Na . Re . Mic obiol. 6, 741–751 (2008).
83. Baumann, P., Mo an, N. A. & Baumann, L. Gene ics o Aphid
Endosymbion s. Bioscience 47, 12–20 (1997).
84. Taylo , M. J. & Hoe au , A. Wolbachia bac e ia o ila ial nema odes.
Pa asi ol. Today 15, 437–442 (1999).
85. Fos e , J. e al. The Wolbachia genome o B ugia malayi: Endosymbion
e olu ion wi hin a human pa hogenic nema ode. PLoS Biol. 3, 0599–0614
(2005).
86. C aig, S. C. & S ephen, J. G. T anso a ial inhe i ance o endosymnio ic
Bac e ia in clams onhabi ing deep-sea hyd o he mal en s and colds-
seeps. P oc Na l Acad Sci U S A 90, 5695–5699 (1993).
87. S ewa , F. J. & Ca anaugh, C. M. Bac e ial endosymbioses in Solemya
(Mollusca: Bi al ia) - Model sys ems o s udies o symbion -hos
adap a ion. An onie an Leeuwenhoek, In . J. Gen. Mol. Mic obiol. 90,
343–360 (2006).
88. Ushe , K. M., Kuo, J., F omon , J. & Su on, D. C. Ve ical ansmission o
cyanobac e ial symbion s in he ma ine sponge Chond illa aus aliensis
(Demospongiae). Hyd obiologia 461, 15–23 (2001).
89. Sha p, K. H., Eam, B., John Faulkne , D. & Haygood, M. G. Ve ical
ansmission o di e se mic obes in he opical sponge Co icium sp.
Appl. En i on. Mic obiol. 73, 622–629 (2007).
90. T am, U. & Sulli an, W. Role o Delayed Nuclea En elope B eakdown
and Mi osis in Wolbachia-Induced Cy oplasmic Incompa ibili y. Science
(80-. ). 296, 1124–1126 (2002).
91. S ou hame , R. & Kazme , D. J. Cy ogene ics o mic obe-associa ed
pa henogenesis and i s consequences o gene low in T ichog amma
wasps. He edi y (Edinb). 73, 317–327 (1994).
92. Kose, H. & Ka , T. L. O ganiza ion o Wolbachia pipien is in he
D osophila e ilized egg and emb yo e ealed by an an i-Wolbachia
monoclonal an ibody. Mech. De . 51, 275–288 (1995).
93. Fe ee, P. M. e al. Wolbachia u ilizes hos mic o ubules and Dynein o
an e io localiza ion in he D osophila Oocy e. PLoS Pa hog. 1, 0111–
0124 (2005).
94. Se bus, L. R. & Sulli an, W. A cellula basis o Wolbachia ec ui men o
he hos ge mline. PLoS Pa hog. 3, 1930–1937 (2007).
95. Sheehan, K. B., Ma in, M., Lesse , C. F., Isbe g, R. R. & New on, L. G.
Iden i ica ion and Cha ac e iza ion o a Candida e Wolbachia pipien is
Type IV E ec o Tha In e ac s wi h he Ac in Cy oskele on. MBio 7, 1–12
!
49!
(2016).
96. O e, M., Ueyama, M. & Yamamo o, D. Wolbachia P o ein TomO Ta ge s
nanos mRNA and Res o es Ge m S em Cells in D osophila Sex-le hal
A icle Wolbachia P o ein TomO Ta ge s nanos mRNA and Res o es
Ge m S em Cells in D osophila Sex-le hal Mu an s. Cu . Biol. 26, 1–10
(2016).
97. Chin apalli, V. R., Wang, J. & Dow, J. a T. Using FlyA las o iden i y
be e D osophila melanogas e models o human disease. Na . Gene .
39, 715–20 (2007).
98. Go a , M. e al. The D osophila immune esponse agains G am-
nega i e bac e ia is media ed by a pep idoglycan ecogni ion p o ein.
Na u e 416, 640–644 (2002).
99. Choe, K. M., We ne , T., S o en, S., Hul ma k, D. & Ande son, K. V.
Requi emen o a pep idoglycan ecogni ion p o ein (PGRP) in Relish
ac i a ion and an ibac e ial immune esponses in D osophila. Science
(80-. ). 296, 359–362 (2002).
100. Räme , M., Man uelli, P., Pea son, A., Ma hey-P e o , B. & Ezekowi z,
R. A. B. Func ional genomic analysis o phagocy osis and iden i ica ion o
a D osophila ecep o o E. coli. Na u e 416, 644–648 (2002).
101. Kaneko, T. e al. Monome ic and polyme ic g am-nega i e pep idoglycan
bu no pu i ied LPS s imula e he D osophila IMD pa hway. Immuni y 20,
637–649 (2004).
102. Takehana, A. e al. Pep idoglycan ecogni ion p o ein (PGRP)-LE and
PGRP-LC ac syne gis ically in D osophila immuni y. Embo J 23, 4690–
4700 (2004).
103. Kaneko, T. e al. PGRP-LC and PGRP-LE ha e essen ial ye dis inc
unc ions in he d osophila immune esponse o monome ic DAP- ype
pep idoglycan. Na . Immunol. 7, 715–723 (2006).
104. B umell, J. H. & Scidmo e, M. a. Manipula ion o ab GTPase unc ion by
in acellula bac e ial pa hogens. Mic obiol. Mol. Biol. Re . 71, 636–652
(2007).
105. Roy, C. R. Exploi a ion o he endoplasmic e iculum by bac e ial
pa hogens. T ends Mic obiol. 10, 418–424 (2002).
106. Se bus, L. R., Caspe -Lindley, C., Landmann, F. & Sulli an, W. The
gene ics and cell biology o Wolbachia-hos in e ac ions. Annu. Re .
Gene . 42, 683–707 (2008).
107. Vo onin, D., Dudkina, N. V & Kisele a, E. V. A new o m o symbio ic
bac e ia Wolbachia ound in he endoplasmic e iculum o ea ly emb yos
o D osophila melanogas e . Dokl Biol Sci 396, 227–229 (2004).

!
50!
108. McClin ock, B. Induc ion o Ins abili y a Selec ed Loci in Maize. Gene ics
38, 579–599 (1953).
109. McClin ock, B. The o igin and beha iou o mu able loci in maize.
Gene ics 36, 344–355 (1950).
110. O gel, L. E. & C ick, F. H. C. Sel ish DNA: he ul ima e pa asi e. Na u e
284, 604–607 (1980).
111. Dooli le, W. F. & Sapienza, C. Sel ish genes, he pheno ype pa adigm
and genome e olu ion. Na u e 284, 601–603 (1980).
112. Kidwell, M. G. & Lisch, D. R. Pe spec i e: ansposable elemen s,
pa asi ic dna, and genome e olu ion. E olu ion (N. Y). 55, 1–24 (2001).
113. González, J., Macphe son, J. M. & Pe o , D. a. A ecen adap i e
ansposable elemen inse ion nea highly conse ed de elopmen al loci
in D osophila melanogas e . Mol. Biol. E ol. 26, 1949–1961 (2009).
114. González, J. & Pe o , D. A. The adap i e ole o ansposable elemen s
in he D osophila genome. Gene 448, 124–33 (2009).
115. Ko le , R., Be ancou , A. J. & Schlö e e , C. Sequencing o pooled DNA
samples (Pool-Seq) unco e s complex dynamics o ansposable
elemen inse ions in D osophila melanogas e . PLoS Gene . 8, (2012).
116. Fis on-La ie , A.-S., Ba ón, M. G., Pe o , D. A. & González, J. T-lex2:
geno yping, equency es ima ion and e-anno a ion o ansposable
elemen s using single o pooled nex -gene a ion sequencing da a.
Nucleic Acids Res. 43, e22 (2014).
117. Kaminke , J. e al. The ansposable elemen s o he D osophila
melanogas e euch oma in: a genomics pe spec i e. Genome Biol. 3, 1–
20 (2002).
118. In e na ional Rice Genome Sequencing P ojec . The map-based
sequence o he ice genome. Na u e 436, 793–800 (2005).
119. Daboussi, M.-J. & Capy, P. T ansposable elemen s in ilamen ous ungi.
Annu. Re . Mic obiol. 57, 275–299 (2003).
120. Human Genome Sequencing Conso ium. Ini ial sequencing and analysis
o he human genome. Na u e 409, 860–921 (2001).
121. Wicke , T. e al. A uni ied classi ica ion sys em o euka yo ic
ansposable elemen s. Na . Re . Gene . 8, 973–982 (2007).
122. Le in, H. L. & Mo an, J. V. Dynamic in e ac ions be ween ansposable
elemen s and hei hos s. Na . Re . Gene . 12, 615–627 (2011).
123. Fescho e, C., Jiang, N. & Wessle , S. R. Plan T ansposable Elemen s:
Whe e Gene ics Mee s Genomics. Na . Re . Gene . 3, 329–341 (2002).
!
51!
124. Ca , M., Bensasson, D. & Be gman, C. M. E olu iona y Genomics o
T ansposable Elemen s in Saccha omyces ce e isiae. 7, (2012).
125. Kim, J. M. e al. T ansposable Elemen s and Genome O ganiza ion : A
Comp ehensi e Su ey o Re o ansposons Re ealed by he Comple e
Saccha omyces ce e isiae Genome ? Sequence T ansposable Elemen s
and Genome O ganiza ion : A Comp ehensi e Su ey o
Re o ansposons Re ea. 464–478 (1998). doi:10.1101/g .8.5.464
126. Kazazian J ., H. H. Mobile Elemen s: D i e s o Genome E olu ion.
Science (80-. ). 303, 1626–1632 (2004).
127. Fescho e, C. & P i ham, E. J. DNA ansposons and he e olu ion o
euka yo ic genomes. Annu. Re . Gene . 41, 331–68 (2007).
128. Kempken, F. & Windho e , F. The hAT amily: a e sa ile ansposon
g oup common o plan s, ungi, animals, and man. Ch omosoma 110, 1–
9 (2001).
129. Fescho e, C. & P i ham, E. J. Non-mammalian c-in eg ases a e encoded
by gian ansposable. BMC E ol. Biol. 4, 22–32 (2004).
130. DeMa co, R., Venancio, T. M. & Ve jo ski-Almeida, S. SmTRC1, a no el
Schis osoma mansoni DNA ansposon, discloses new amilies o animal
and ungi ansposons belonging o he CACTA supe amily. BMC E ol.
Biol. 6, 89 (2006).
131. G eenbla , I. M. & B ink, A. Twin mu a ions in medium a iega ed
pe ica p maize. Gene ics 47, 489–501 (1961).
132. Nassi , N., Penney, J., Pal, S., Engels, W. R. & Gloo , G. B. E icien
copying o nonhomologous sequences om ec opic si es ia P-elemen -
induced gap epai . Mol. Cell. Biol. 14, 1613–25 (1994).
133. Nobel Lec u es: Physiology o Medicine 1981-1990. (Wo ld Scien i ic
Publishing Co. P e. L d., 1993).
134. Ikeda, K., Nakayashiki, H., Takagi, M., Tosa, Y. & Mayama, S. Hea
shock, coppe sul a e and oxida i e s ess ac i a e he e o ansposon
MAGGY esiden in he plan pa hogenic ungus Magnapo he g isea.
Mol. Gene . Genomics 266, 318–325 (2001).
135. Kimu a, Y. e al. OARE-1, a Ty1-copia e o ansposon in oa ac i a ed
by abio ic and bio ic s esses. Plan Cell Physiol. 42, 1345–1354 (2001).
136. Salaza , M., González, E., Casa e o, J. A., Casacube a, J. M. & Ruiz-
La a, S. The p omo e o he TLC1.1 e o ansposon om Solanum
chilense is ac i a ed by mul iple s ess- ela ed signaling molecules. Plan
Cell Rep. 26, 1861–1868 (2007).
137. Se an , G., Penne ie , C. & Lesage, P. Remodeling yeas gene
ansc ip ion by ac i a ing he Ty1 long e minal epea e o ansposon
!
52!
unde se e e adenine de iciency. Mol. Cell. Biol. 28, 5543–5554 (2008).
138. Wood ow, P., Pon eco o, G., Cia miello, L. F., Fuggi, A. & Ca illo, P.
T d1a p omo e is in ol ed in DNA-p o ein binding by sal and ligh
s esses. Mol. Biol. Rep. 38, 3787–3794 (2011).
139. Casacube a, E. & González, J. The impac o ansposable elemen s in
en i onmen al adap a ion. Mol. Ecol. 22, 1503–1517 (2013).
140. Pou eau, S., G andbas ien, M.-A. & Bocca a, M. Mic obial elici o s o
plan de ence esponses ac i a e ansc ip ion o a e o ansposon. Plan
J. 5, 535–542 (1994).
141. G andbas ien, M. A. Ac i a ion o plan e o ansposons unde s ess
condi ions. T ends Plan Sci. 3, 181–187 (1998).
142. Takeda, S., Sugimo o, K., O suki, H. & Hi ochika, H. A 13-bp cis-
egula o y elemen in he LTR p omo e o he obacco e o ansposon
T o1 is in ol ed in esponsi eness o issue cul u e, wounding, me hyl
jasmona e and ungal elici o s. Plan J. 18, 383–393 (1999).
143. Begui is ain, T., G andbas ien, M.-A., Puigdomènech, P. & Casacube a,
J. M. Th ee Tn 1 Sub amilies Show Di e en S ess-Associa ed Pa e ns
o Exp ession in Tobacco. Consequences o Re o ansposon Con ol
and E olu ion in Plan s. Plan Physiol. 127, 212–221 (2001).
144. Kuma , A. & Benne zen, J. L. Plan e o ansposons. Annu. Re . Gene .
33, 479–532 (1999).
145. G andbas ien, M. a, Spielmann, A. & Caboche, M. Tn 1, a mobile
e o i al-like ansposable elemen o obacco isola ed by plan cell
gene ics. Na u e 337, 376–380 (1989).
146. Dai, J., Xie, W., B ady, T. L., Gao, J. & Voy as, D. F. Phospho yla ion
Regula es In eg a ion o he Yeas Ty5 Re o ansposon in o
He e och oma in. Mol. Cell 27, 289–299 (2007).
147. Ebina, H. & Le in, H. L. S ess Managemen : How Cells Take Con ol o
Thei T ansposons. Mol. Cell 27, 180–181 (2007).
148. Ga cía Gue ei o, M. P. Wha makes ansposable elemen s mo e in he
D osophila genome? He edi y (Edinb). 108, 461–468 (2012).
149. Junako ic, N., F anco, C. Di, Ba san i, P. & Palumbo, G. T ansposi ion o
Copia-like Nomadic Elemen s Can Be Induced by Hea Shock. J. Mol.
E ol. 24, 89–93 (1986).
150. Ra ne , V. A., Zabano , S. A., Kolesniko a, O. V & Vasilye a, L. A.
Induc ion o he mobile gene ic elemen Dm-412 ansposi ions in he
D osophila genome by hea shock ea men . P oc. Na l. Acad. Sci. U. S.
A. 89, 5650–4 (1992).
!
53!
151. A naul , C., Loe enb uck, C. & Biémon , C. T ansposable elemen
mobiliza ion is no induced by hea shocks in D osophila melanogas e .
Na u wissenscha en 84, 410–4 (1997).
152. Jouan-Du ou nel, I., Cosse , F., Con amine, D., Ve die , G. & Biémon , C.
T ansposable Elemen s Beha io Following Vi al Genomie S ess in
D osophila melanogas e Inb ed Line. J. Mol. E ol. 43, 19–27 (1996).
153. Nabi ochkin, S. D. e al. Onco i al DNAs induce ansposi ion o
endogenous mobile elemen s in he genome o D osophila melanogas e .
Mu a . Res. 403, 127–36 (1998).
154. Wessle , S. R. Pheno ypic Di e si y Media ed by he Maize T ansposable
Elemen s Ac and Spm. Science (80-. ). 242, 399–405 (1988).
155. Biémon , C. & Viei a, C. Gene ics: junk DNA as an e olu iona y o ce.
Na u e 443, 521–524 (2006).
156. Fescho e, C. T ansposable elemen s and he e olu ion o egula o y
ne wo ks. Na . Re . Gene . 9, 397–405 (2008).
157. Cowley, M. & Oakey, R. J. T ansposable Elemen s Re-Wi e and Fine-
Tune he T ansc ip ome. PLoS Gene . 9, (2013).
158. Walse , J.-C., Chen, B. & Fede , M. E. Hea -shock p omo e s: a ge s o
e olu ion by P ansposable elemen s in D osophila. PLoS Gene . 2,
e165 (2006).
159. Wang e al. Species-speci ic endogenous e o i uses shape he
ansc ip ional ne wo k o he human umo supp esso p o ein p53. P oc.
Na l. Acad. Sci. U. S. A. 104, 18613–18618 (2007).
160. Ma iño-Ramí ez, L., Lewis, K. C., Landsman, D. & Jo dan, I. K.
T ansposable elemen s dona e lineage-speci ic egula o y sequences o
hos genomes. Cy ogene . Genome Res. 110, 333–341 (2005).
161. Van De Lagemaa , L. N., Land y, J. R., Mage , D. L. & Meds and, P.
T ansposable elemen s in mammals p omo e egula o y a ia ion and
di e si ica ion o genes wi h specialized unc ions. T ends Gene . 19,
530–536 (2003).
162. Mikkelsen, T. S. e al. Genome o he ma supial Monodelphis domes ica
e eals inno a ion in non-coding sequences. Na u e 447, 167–77 (2007).
163. Beje ano, G. e al. A dis al enhance and an ul aconse ed exon a e
de i ed om a no el e oposon. Na u e 441, 87–90 (2006).
164. San angelo, A. M. e al. Ancien exap a ion o a CORE-SINE e oposon
in o a highly conse ed mammalian neu onal enhance o he
p oopiomelanoco in gene. PLoS Gene . 3, 1813–1826 (2007).
165. B i en, R. J. Cases o ancien mobile elemen DNA inse ions ha now
! !

!
61!
2.1 Abs ac
T ansposable elemen s (TEs) a e mobile DNA sequences ha cons i u e a la ge
ac ion o euka yo ic genomes, and media e he p oduc ion o gene ic and
pheno ypic a ia ion1–5. Because ansposi ion h ea ens genomic in eg i y and
ypically has dele e ious e ec s, o ganisms e ol ed a p o ec i e mechanism o
speci ically es ain TE ac i i y in he ge mline6–12, he piRNA pa hway. Despi e
much wo k on his pa hway, he gene ic basis o na u al a ia ion in TE ac i i y
emain unknown5. To s udy he loci ha con ain a ia ion a ec ing TE ac i i y in
na u al popula ions we pe o med a genome-wide associa ion s udy using as
pheno ype he numbe o blood, gypsy5, mdg1, oo, I-elemen , jockey, H-
elemen and pogo inse ions ha a e mo e likely o esul om ecen TE
ac i i y13. We obse ed li le o e lap be ween alleles associa ed wi h he ac i i y
o di e en ansposons and e y ew a ec ed piRNA pa hway genes. Ou
analysis on blood ac i i y led us o p opose ha he ansc ip ion ac o Sin3A
migh ha e played an impo an ole in a ec ing he exp ession o his TE.
Con e sely, he bu s o gypsy5 ac i i y obse ed in a single geno ype was
ela ed o alleles a ec ing se e al genes unc ioning in ollicle cell de elopmen
and ha included he ansc ip ion ac o Poin ed. This s udy iden i ied an
unan icipa ed se o egula o s and highligh ed he impo ance o genome-wide
associa ion s udies o add ess ques ions ega ding he gene ic ac o s ha
in luence TE ac i i y in he wild.
2.2 In oduc ion
T ansposable elemen s (TEs) a e mobile DNA sequences ha we e i s
disco e ed by Ba ba a McClin ockʼs e olu iona y wo k on he e ec o genomic
ins abili y in he con ex o colo pigmen a ion in maize ke nels14. They a e
abundan , di e se and ac i e componen s o mos euka yo ic genomes, and
while ansposi ion e en s a e o en dele e ious, TEs can con ibu e o gene ic
a ia ion ha can be adap i e and ele an o e olu ion1,3,4,15–18. Recen s udies
ha e also implica ed TEs as media o s o pheno ypic di e si ica ion con ibu ing
!
62!
o he e olu ion o no el ai s such as placen a19, ish spo s20 and colo
pigmen a ion in B i ish peppe ed mo hs21.
TEs a e di ided in o wo main ca ego ies ha g oup mobile elemen s acco ding
o hei sequence p ope ies, which e lec how ansposi ion occu s.
Re o ansposons esemble e o i uses and mo e ia an RNA in e media e ha
adds new copies o he p o ile o al eady exis ing inse ions; DNA elemen s a e
excised and in eg a ed in o new genomic places hus he numbe o copies is
main ained a each cycle o mo emen 22,23. The copy-and-pas e and cu -and-
pas e mechanisms o ansposi ion desc ibed abo e include a scope o dis inc
s eps such as ansc ip ion, ansla ion and in eg a ion, upon which hos s can ac
o con ol TE-speci ic ac i i y24–26. Mo eo e , he disco e y o a conse ed
pa hway – he piRNA pa hway – ha p oduces small RNAs able o a ge and
educe exp ession o mobile elemen s ( e o ansposons and DNA elemen s),
con e ged e o s o he s udy o TE ansc ip ional egula ion6–9.
The piRNA pa hway unc ions as a genome su eillance sys em ha speci ically
minimizes TE access o emale and male game ocy es6,27. In D. melanogas e
o a ies, he biogenesis o PIWI-in e ac ing RNAs (piRNAs) di e s be ween
soma ic ( ollicula epi helium ha es ablish an in ica e ela ionship wi h
ge mline) and ge mline (oocy e and nu se cell syncy ium) issue whe e he
silencing p ocess is specialized o ep ess TEs exclusi e o each cell ype28.
Al hough ansposons ha a e ac i e in he ge mline issue ha e di ec access o
he oocy e, TEs exp essed in ollicula cells need o o m i al-like pa icles in
o de o be able o in ade he emale game ocy e29,30.
Despi e he ex ensi e in o ma ion ega ding ep essi e mechanisms, he gene ic
ac o s ha modula e TE ac i i y in na u al popula ions emain elusi e5. He e, we
aimed o iden i y loci in hos genomes ha ca y allelic a ia ion con ibu ing o
changes in TE ac i i y and ha leads o di e ences in he numbe o ecen TE
inse ions. We bene i ed om he D osophila melanogas e Gene ic Re e ence
Panel (DGRP) ha includes a ious ully sequenced geno ypes con aining
alleles ha seg ega ed in a na u al popula ion13. The in silico analysis o he
sequence o 146 o hose isogenic lines p o ided in o ma ion on TE inse ions,
!
63!
including he posi ion o inse ions, he elemen inse ed, and whe he o no
each inse ion was sha ed wi h he D. melanogas e e e ence genome ( om
ano he popula ion o lies). TE inse ions we e classi ied as no el when hey
we e no sha ed wi h e e ence genome; hese we e also ypically no sha ed
be ween di e en DGRP geno ypes and a e, hus, mo e likely o esul om
ecen TE ac i i y. We pe o med a genome-wide associa ion s udy (GWAS)
using he numbe o no el inse ions o di e en ansposons (blood, gypsy5,
mdg1, oo, I-elemen , jockey, H-elemen and pogo) as pheno ype. To do he
wo k, we i s alida ed in silico p edic ions o TE composi ion o na u ally-
de i ed sequenced genomes and hen used hose p edic ions as a pheno ype in
he GWAS s udy. Al hough we had hypo hesized o ind allelic a ian s o piRNA
genes, we obse ed li le o e lap be ween loci con ibu ing o di e ences in he
numbe o no el inse ions and hose encoding genes o he piRNA pa hway.
Analysis o he GWAS esul s sugges ha ac i i y o he TE blood, a
e o ansposon exp essed in ge mline and soma ic issue o o a ies, is
egula ed by he ansc ip ion ac o Sin3A, and ha a bu s in he ac i i y o TE
gypsy5, a e o ansposon speci ically exp essed in ollicula cells, may be
ela ed o allelic a ia ion o genes impo an o he de elopmen o ollicula
epi helia.
2.3 Ma e ials and Me hods
TE iden i y o no el inse ions in DGRP geno ypes
The DGRP is composed by nea ly 200 wild-de i ed geno ypes, 146 o which
ha e a ailable in o ma ion abou TE inse ions ha we e cha ac e ized as
sha ed o no el acco ding o whe he hey a e p esen o absen , espec i ely, in
he e e ence genome13. Fo he no el inse ions, b eakpoin es ima es (posi ion
o inse ion) and TE iden i y we e assessed based on sequence a ailable o le
and igh sides o he inse ion, as desc ibed in e 13. Fo each ansposon
iden i ied in a no el inse ion, he in silico anno a ion p o ided se e al TEs, each
wi h an associa ed p obabili y. He e, o each no el inse ion, we used as TE
iden i y ha wi h highes es ima ed p obabili y.
!
64!
Con i ma ion o TE iden i y o no el inse ions
To con i m in silico p edic ions o TE composi ion, we selec ed 129 p edic ed
no el inse ions in 12 DGRP geno ypes including 13 o TE blood, 4 o copia, 7
o Doc, 7 o gypsy5, 7 o H-elemen , 5 o hoppe , 3 o F-elemen , 25 o I-
elemen , 5 o Juan, 19 o mdg1, 2 o opus, 17 o pogo, 10 o oo and 5 o
T anspac elemen s. Lis o TE inse ions and espec i e DGRP lines es ed he e
can be ound in Supplemen a y Table 2.1. We ex ac ed gDNA om RAL-021,
RAL-321, RAL-357, RAL-358, RAL-381, RAL-443, RAL-761, RAL-790, RAL-810,
RAL-812, RAL-892 and RAL-908 lies, pool o en males pe sample. DNeasy
Blood & Tissue Ki (Qiagen) was used o he ex ac ion, ollowing
manu ac u e ʼs p o ocol. Sample concen a ion was measu ed in Nanod op ND-
1000 Spec opho ome e and samples we e s o ed a -20°C un il use.
Based on he genome sequence a ailable o he DGRPs, we designed p ime s
o he lanking egion o each no el inse ion and ob ained amplicons by
longPCR (Roche) using 4ng o gDNA in 15µl o eac ion wi h 0.5µM p ime s, 2%
DMSO, 1x bu e , 0.5mM dNTPs, 0.21µl o enzyme in 15µl o eac ion. We used
s anda d ampli ica ion p og am o longPCR wi h 10 min o elonga ion ime;
empe a u e o annealing and p ime sequences can be ound in
Supplemen a y Table 2.1. Size o inse ion was ob ained by elec opho esis in
1% aga ose gel and classi ied as “la ge ”, “expec ed” o “smalle ” aking in o
accoun he canonical TE sequences desc ibed o D osophila ansposons
( e sion 9.42). We assessed TE iden i y by sequencing he amplicons wi h
BigDye Te mina o p o ocol (Te moFishe ) o SUPREME un™ (NZYTech) using
he same o wa d p ime s used o ampli ica ion (Supplemen a y Table 2.1).
To de ec p esence o alse nega i es in Mackay e al. p edic ions o TE
inse ions we pe o med a PCR using as empla e gDNA om lines whe e
inse ions we e p edic ed o be absen (Supplemen a y Table 2.2). As posi i e
con ols, we pe o med same PCRs using ei he gDNA om a line wi h p edic ed
inse ions (RAL-321; check o p ime s) o p ime s o gene Rpl32 (check o
empla e). In nega i e con ol eac ions o each p ime pai gDNA was eplaced
by wa e . The gDNA ex ac ion o RAL-109, RAL-161, RAL-237, RAL-350, RAL-
!
65!
362, RAL-555, RAL-776 and RAL-908 was pe o med as desc ibed abo e. Each
PCR eac ion con ained 0.25U o GoTaq (P omega), 1x eac ion bu e , 1.5mM
MgCl2, 0.4µM p ime s and 0.4ng o gDNA. The he mal cycling p o ocol was he
same o all eac ions: 10 min a 95°C; 35 cycles o 95°C o 30 sec, 60°C o 1
min and 72°C o 30 sec; 5 min a 72ºC. P ime sequences a e desc ibed in
Supplemen a y Table 2.3.
Genome Wide Associa ion S udy (GWAS)
We pe o med a GWAS analysis using he numbe o no el inse ions o blood,
gypsy5, mdg1, oo, I-elemen , jockey, H-elemen , and pogo as pheno ypes. This
in o ma ion is a ailable o 146 o he DGRP geno ypes. The analyses we e
done on he DGRP2 websi e (h p://dg p2.gne s.ncsu.edu/)13,31. Fo he genes
con aining alleles ound o associa e o a ia ion in numbe o no el inse ions
(p<7e-05), we an gene on ology (GO) analyses using GO conso ium algo i hm,
and Bon e oni co ec ion o mul iple es ing32.
RNA ex ac ion and cDNA syn hesis
To assess blood, gypsy5, and poin ed exp ession we chose nine ly s ocks wi h
di e en numbe o no el inse ions: RAL-021, RAL-237, RAL-321, RAL-357,
RAL-358, RAL-375, RAL-391, RAL-790 and RAL-908. To elimina e Wolbachia
om he lines RAL-021, RAL-237, RAL-321 and RAL-790, lies we e ea ed wi h
e acycline (0.05mg/ml) o wo gene a ions, and gu lo a con en homogenized
as p e iously desc ibed33. We only used o his s udy indi iduals wi hou
endosymbion in ec ion. Flies we e kep in s anda d co nmeal ood34 a 25°C,
60% humidi y and 12h day-nigh cycles.
We collec ed o a ies om se en days old emales and pooled bo h o a ies om
eigh emales pe sample. O a ies we e ob ained om wo independen
expe imen s, one wi h eigh and ano he wi h ou eplica e samples pe
geno ype. Dissec ions we e done in esh cold PBS 1x and o a ies we e di ec ly
s o ed in 400μl o T izol (Ambicon) a 4ºC un il homogenized by pes els. Tissue
collec ion and homogeniza ion was done in he same day; samples we e hen

!
66!
s o ed a -80ºC un il RNA ex ac ion. Fo RNA ex ac ion, we used Di ec -zol
RNA Minip ep ki (Zymo Resea ch) ollowing manu ac u e ʼs ins uc ions, and
o al RNA was elu ed in 25μl o RNase- ee wa e (Sigma). RNA pu i y and
concen a ion was measu ed wi h Nanod op ND-1000 Spec opho ome e ;
A600/A800 abso bance was a ound wo o all samples and concen a ions
anged om 200 o 800ng/µl. All RNA samples o he same geno ype we e
p ocessed on he same day.
Fo cDNA syn hesis we emo ed con amina ions o gDNA om 1µg o RNA
using DNAse (P omega) ea men and ollowing manu ac u e ʼs ins uc ions. We
hen ollowed Re e se T ansc ip ion Sys em (P omega) p o ocol using 0.02µM
Oligo dT p ime s o p oduce cDNA. Reac ion was incuba ed a 42°C o 60min
and hea ed a 95°C o 5min. cDNA was dilu ed 1:10 in RNAse- ee wa e and
s o ed a -20°C un il quan i a i e eal- ime PCR (qPCR).
qPCR
We measu ed le els o gypsy5, poin ed, Rpl32 and Tbp exp ession by qPCR
using a CFX384 he mal cycle (BioRad). Fo each eac ion, we used 5µl iQ™
SYBR® G een supe mix (BioRad), 1µl o 4µM p ime s and 4µl o dilu ed cDNA
(1:10). P ime sequences a e desc ibed in Supplemen a y Table 2.3. We an
wo echnical eplica es pe sample, bo h in he same qPCR pla e. The he mal
cycling p o ocol was he same o all eac ions: 2 min a 50°C; 10 min a 95°C;
40 cycles o 95°C o 30 sec, 60°C o 1 min and 72°C o 30 sec. In each pla e,
we an s anda d cu es (one s anda d cu e o each gene o TE es ed) wi h
se ial dilu ions o known amplicon concen a ions ha we e ampli ied by PCR
using 0.25U o GoTaq (P omega), 1x eac ion bu e , 1.5mM MgCl2, 0.4µM
p ime s and 1µl o cDNA (1:10). PCR p oduc s we e cleaned using NucleoSpin®
Gel and PCR Clean-up (Mache ey-Nagel), and elu ed in 15µl o bu e NE. We
measu ed concen a ions in Nanod op ND-1000 Spec opho ome e . The
minimum amoun o nucleic acids ha can be de ec ed by qPCR o each p ime
pai was es ima ed by ex ending dilu ions o ou s anda ds un il he ela ionship
be ween he known amplicon concen a ions and he quan i ica ion cycles (Cqs)
!
67!
eached a pla eau. Samples ha had Cq alues ou side he linea de ec ion
(Supplemen a y Figu e 2.1) and wi h s anda d de ia ion abo e 0.5 o he
echnical eplica es we e elimina ed om he analysis. We used de aul
h eshold se ings o ob ain he Cq o each eac ion, and ans o med Cq alues
in o absolu e a ge concen a ions using he espec i e s anda d cu es. This
me hodology allowed us o con ol o pla e e ec s and o di e ences in p ime
e iciency. qPCR mel ing cu es we e analyzed o con i m speci ici y o ampli ied
p oduc s and nega i e con ols ne e showed de ec able ampli ica ion. Fo he
analysis o exp ession da a, we used No mFinde 35 and geNo m36 algo i hms o
check exp ession s abili y o wo pu a i e e e ence genes, Rpl32 and Tbp,
commonly used o no malize gene exp ession a di e en en i onmen al
condi ions and geno ypes37,38. S abili y alues ob ained o Rpl32 (No mFinde ,
0.58; geNo m, 1.32) we e o e all be e han hose ob ained o Tbp
(No mFinde , 0.92; geNo m, 1.32). We he e o e no malized TE exp ession o
Rpl32.
S a is ical analyses
All s a is ical analyses we e done in R ( e sion 3.0.3)39.
Fo he analysis o gypsy5 and poin ed exp ession in di e en DGRP lines, we
used linea models (lm) wi h logno mal dis ibu ion. Mul iple compa isons we e
pe o med wi h leas -squa es means (lsmeans), and compac le e display (cld)
was used o g oup geno ypes ha a e no s a is ically signi ican (p>0.05). Da a
was ob ained om wo independen expe imen s (one wi h eigh and ano he
wi h ou biological eplica es) ha we e plo ed and analyzed oge he .
Analysis o TE exp ession om a ailable nex -gene a ion RNA sequencing
da a
To s udy he impo ance o Sin3A o egula ing blood exp ession we analyzed
a ailable nex -gene a ion RNA sequencing lib a ies (GSE72172 and GSE68775)
deposi ed on NCBI40. We ex ac ed aw da a o S2 (GSM1857008,
GSM1857009 and GSM1857010), Sin3A-187HA (GSM1857011, GSM1857012
!
68!
and GSM1857013), GFP RNAi (GSM1681056, GSM1681057 and
GSM1681058) and Sin3A KD (GSM1681059, GSM1681060 and GSM1681061)
samples. Those samples we e ob ained om ou es ablished cell cul u e lines:
wi hou any cons uc (S2), wi h GFP RNAi cons uc , o e exp essing Sin3A-187
(Sin3A-187HA), and ano he down- egula ing all iso o ms (Sin3A KD). Reads
we e aligned o ansposon canonical sequences ( e sion 9.42) o D.
melanogas e . We ex ac ed he numbe o eads o each TE and analyzed he
da a using edgeR by i ing ead coun s o a nega i e binomial dis ibu ion. To
e alua e he e ec o o e exp essing Sin3A-187 (Sin3A-187HA), we used S2
samples as nega i e con ols; o e alua e he e ec o knocking down Sin3A
(Sin3A KD) iso o ms we used GFP RNAi samples as nega i e con ols. We
plo ed old-change o TE exp ession wi h alse disco e y a e in e io o 0.1.
2.4 Resul s
Con i ma ion o TE inse ions in DGRP lines
We s udied a o al o 129 no el inse ions ha included ansposons wi h
di e en sequence p ope ies: e o ansposons wi h and wi hou long e minal
epea s (LTR and non-LTR elemen s), and DNA elemen s. Fo all 129 p edic ed
inse ions (100%), we con i med p esence o ansposon sequences, and o
113 o hem (87.6%) we con i med he mos likely ansposon iden i y as pe he
in silico p edic ions13 (co ec TE iden i y, Figu e 2.1). Fo eigh o he 16 cases
whe e we did no con i m he mos likely p edic ed iden i y, ou sequence
ma ched ha o elemen s wi h same sequence p ope ies and se en ma ched
sequence iden i y o he second mos likely p edic ed TE (Supplemen a y
Table 2.4). Fo he emaining nine cases, he inse ed TE was no wi hin he lis
o p edic ed iden i ies o he speci ic posi ions.
Fo he cha ac e iza ion o TE inse ions acco ding o leng h ( ela i e o he
leng h o he TEs iden i ied by ou sequencing da a) we obse ed ha 58.9%
showed p ope size o inse ion, 33.3% we e smalle and 7.8% la ge (Figu e
2.1).
!
69!
We also assessed he occu ence o alse nega i es in Mackay e al. p edic ions
by es ing p esence o TE inse ions in DGRP geno ypes ha we e p edic ed o
lack hose elemen s (Supplemen a y Table 2.2). We de ec ed p esence o
es ed TEs in he genome o all lines assessed (Supplemen a y Figu e 2.2).
Figu e 2.1. Cha ac e iza ion o no el TE inse ions acco ding o sequence iden i y
and size. Fo each no el inse ion we used as TE iden i y ha wi h highes es ima ed
p obabili y o inse ion13. We hen cha ac e ized acco ding o sequence iden i y and size
inse ions assigned o LTR (blood, copia, gypsy5, mdg1, opus, oo and T anspac), non-
LTR (Doc, F-elemen , I-elemen and Juan) and DNA (H-elemen , hoppe and pogo)
elemen s. No el TE inse ions we e g ouped acco ding o ou abili y o con i m TE iden i y
es ima ed by Mackay e al. (expec ed and unexpec ed) and acco ding o size (smalle ,
expec ed and la ge ). Ba cha s show he numbe o blood, mdg1, opus, oo, T anspac,
Doc, F-elemen , I-elemen , Juan, H-elemen , hoppe and pogo inse ions wi h expec ed
and unexpec ed TE iden i y, and wi h smalle , expec ed and la ge sizes. When TE
iden i y was de ined as unexpec ed we included he name o he ansposon ob ained
ins ead. In hose cases, TE size was cha ac e ized in o smalle , expec ed o la ge
acco ding o he canonical sequence o he TE we iden i ied he e.
!
76!
Figu e 2.4. Model p oposed o he e ec o alleles associa ed wi h gyspy5 bu s in
ac i i y. Scheme illus a ing genes impo an o ollicula cell de elopmen in D.
melanogas e o a ies (A and B). A s age 6 and 8 o oocy e de elopmen , No ch is
exp essed in he ollicula epi helia ha p o ec s he egg chambe . A s age 6, he
in e ac ion o Sm e wi h Su(H) inhibi s No ch signaling and p omo es eplica ion o
ollicula cells (a1)60. La e in de elopmen , Sm e is dissocia ed om Su(H) ha ,
consequen ly, ec ui s co-ac i a o s such as Mas e mind (b1)60. This e en esul s in he
inhibi ion o mi o ic cycles and ollicula cells en e endo eplica ion o become polyploid62.
The pola iza ion o ollicle cells is also essen ial o main ain epi helial unc ions h oughou
oogenesis; cell adhesion (Sho gun and Fas3) and ac in-binding (Enabled and Rhea)
p o eins a e impo an o he es ablishmen and main enance o cellula apical-basal
pola i y (a2)47–49,59. The in e ac ion be ween oocy e and ollicula cells is c ucial o he
es ablishmen o an e iopos e io and do so en al pola i y in he oocy e, and is mainly
media ed by wa es o Gu ken-EGFR signaling (A and B)63. A s age 6, an ea ly signal o
Gu ken de ines he an e iopos e io axis (A), whe eas a la e signal o Gu ken a s age 8
esul s in do so en al pola i y (B). When he EGFR signals o he ollicula cells i
induces he exp ession o se e al genes including poin ed (a3)64. We p opose ha , some
alleles associa ed wi h gypsy5 ac i i y may ha e enhanced EGFR signaling pa hway
h ough e ec s in EGFR, Poin ed, H s and D k abundance and/o ac i i y (C). Also,
alleles a ec ing No ch signaling ( h ough e ec s in Sm e , Mas e mind, H s, kib a and
pigs) and cell pola i y ( h ough e ec s in Fas3, Sho gun, Rhea and Enabled) could ha e
esul ed in inc eased cell numbe (d1) o genomic con en (d2). Those e ec s could ha e
a di ec ( h ough ac i a ion o gypsy5 exp ession by he ansc ip ion ac o Poin ed, C) o
an indi ec ( h ough inc ease in he numbe o cells exp essing gypsy5 o inc ease in he

!
77!
numbe o gypsy5 copies in polyploid nuclei, D) impac in gypsy5 exp ession. In o de o
inc ease in copy-numbe , gypsy5 needs o be exp essed and o o m i al-like pa icles
ha a e able o mig a e om ollicle cells o he oocy e29,30. We u he sugges ha he
dis up ion o epi helial in eg i y media ed by e ec s in cell cycle and cell-cell adhesion (D)
could ha e bene i ed he mo emen o gypsy5 i al-like pa icles o he oocy e. Fo mo e
in o ma ion abou he alleles a ec ing genes ela ed o ollicle cell de elopmen see Table
2.2.
Figu e 2.5. Alleles a ec ing poin ed exp ession may explain bu s in gypsy5
ac i i y. Poin ed binding si es (GGAA) ound in 5ʼLTR sequences o gypsy5 and ZAM
(A). Gypsy5 exp ession ela i e o Rpl32 (loga i hmic scale) ob ained o RAL-237, RAL-
!
78!
321, RAL-358, RAL-375 and RAL-790 (B). We could no de ec gypsy5 exp ession in
RAL-021, RAL-357, RAL-391, and RAL-908. Exp ession o poin ed ela i e o Rpl32
(loga i hmic scale) in RAL-021, RAL-237, RAL-321, RAL-357, RAL-358, RAL-375, RAL-
391, RAL-790 and RAL-908 (C). We used linea model (lm) o e alua e he di e ence in
gypsy5 and poin ed exp ession be ween DGRP lines. Mul iple compa isons we e
pe o med using leas -squa es means (lsmeans). We used compac le e display (cld) o
g oup geno ypes ha a e no s a is ically signi ican (p>0.05), in he g aphs ep esen ed
by he same le e . Da a was collec ed in wo independen expe imen s; in he g aph we
show esul s o bo h expe imen s. Each ci cle in he boxplo s ep esen s a pool o eigh
pai s o o a ies; iangles co espond o he mean and black line o he median.
2.5 Discussion
DGRP geno ypes: a lib a y o na u al TE inse ions
TEs a e ac i e DNA sequences ha cons i u e a la ge p opo ion o mos
genomes, making up o 4% o he D. melanogas e genome, o example65. The
sequence di e si y and epe i i e na u e o ansposons makes hei analysis
challenging, especially when using nex -gene a ion sequencing da a ha
p o ides ai ly sho sequencing eads66,67. Fo his eason, he empi ical
con i ma ion o in silico p edic ions o TE inse ions is essen ial.
In his s udy, we con i med p esence o mobile elemen s a 129 posi ions
p edic ed o ha e no el TE inse ions13. The obse a ion o 0% o alse posi i es
and ou con i ma ion o 87.6% o he TE iden i ies wi h highes p obabili y13,
s ongly sugges s ha he in o ma ion a ailable o TE inse ions in DGRP
geno ypes is eliable, and his panel can he e o e be used as a lib a y o he
na u al a ian s o ansposon inse ions. The inabili y o con i m 12.4% o he
mo e p obable TE iden i ies could, howe e , esul om signi ican di e ences
be ween ou and p e ious13 app oaches o de e mine ansposon iden i y.
Al hough ou es ima ion o iden i y was based on he sco e o he alignmen
wi h canonical TE sequences alone (gi ing g ea e weigh o a be e aligning
TE), Mackay e al. p edic ions included he equency o ha elemen in he
e e ence genome as well (gi ing g ea e weigh o mo e equen elemen s). In
addi ion, ou iden i ies a e mos ly based in in o ma ion o one side o TE
!
79!
inse ion ough Mackay e al. p o ide es ima es o bo h sides. Fo he cases
whe e iden i ies wi h highes p obabili y a e only de ec ed in one side o he
inse ion, and wi h second mos likely TE being de ec ed in bo h sides, i would
be impo an o econside using in silico es ima es om bo h sides (le and
igh ) o assign TE iden i ies o each posi ion o inse ion. As o he case o ou
ou misiden i ied blood inse ions, his new app oach would p o ide be e
p edic ions o TE iden i ies (es ima ed imp o emen o 3.1%).
Fo he cha ac e iza ion o TE inse ions acco ding o size, we obse ed ha ,
al hough 58.9% had expec ed sequence leng h, 33.3% we e smalle and 7.8%
la ge han he p edic ed sizes o he canonical sequences o he espec i e
mobile elemen s. La ge inse ions could esul om he in eg a ion o mo e han
one ansposon in he same genomic posi ion. In con as , smalle inse ions
could de i e om 1) sequence loss a e TE mobiliza ion and/o 2) he ac i i y o
incomple e and non-au onomous elemen s ha sha e sequence p ope ies wi h
o he au onomous ansposons22.
Rega ding ou esul s o he p esence o alse nega i es in TE p edic ions by
Mackay e al., we sugges ha he lack o genomic in o ma ion o he
ch omosome 4 and possibly Y ch omosome in he DGRP could explain ou
de ec ion o speci ic TEs in geno ypes p edic ed o lack hem. A ailable da a o
ansposon inse ions in he ou h and Y ch omosomes is o pa icula in e es
gi en he high p e alence o epe i i e sequences in hose ch omosomes68,69.
Impo an ly, and al hough Mackay e al. p edic ions had ne e been
expe imen ally alida ed be o e, we only es ed a e y small p opo ion o
inse ions (129 o app oxima ely 50 housand no el inse ions o all
ansposons).
Few SNPs a ec ing piRNA pa hway genes a e associa ed wi h TE copy
numbe
A g ea deal o wha is known abou TE ep ession in he ge mline is g ounded
on s udies ha use D. melanogas e as a model o ganism6,11,12. Those desc ibe
he exis ence o piRNAs ha cons i u e he hallma k o ansposon inhibi ion in
!
80!
he gonads and collec i ely map o clus e s ha a e en iched o agmen s o TE
inse ions. No ably, he piRNA biogenesis is media ed by specialized p ocesses
ha di e be ween ge mline and soma ic issues o D osophila o a ies28.
In ollicula cells, he p oduc ion o hose genomic gua dians is independen o
he exp ession o ac i e ansposons and cons i u es he p ima y piRNA
biogenesis70. The biogenesis s ep akes place in he cy oplasm, whe e he Piwi
p o ein (only elemen o he PIWI- amily exp essed in ollicula cells) is loaded
wi h piRNAs71. Piwi-piRNA complexes hen mig a e o he nucleus o media e
ansc ip ional ep ession o mobile elemen s72. Only ecen ly, p ima y
biogenesis o piRNAs has been linked o TUDOR domain-con aining p o eins
(such as e eno) ha modula e Piwi unc ion h ough p o ein-p o ein
in e ac ion73.
In ge mline cells, he a chi ec u e o piRNA biogenesis is complex, as i in ol es
he p ima y biogenesis and a loop o ampli ica ion ha is mos ly d i en by Aub
and AGO3. Those wo PIWI- amily p o eins a e en iched a ound he nucleus
whe e Aub-piRNA complexes a ge and di ec TE ansc ip s o clea age. As a
consequence, sense piRNAs a e p oduced and inco po a ed by AGO3, ha has
he po en ial o speci ically igge piRNA clus e p ocessing and inc ease he
abundance o an isense piRNAs loaded in o Aub6.
In e es ingly, ou GWAS o TE copy-numbe e ealed e y ew alleles a ec ing
piRNA pa hway genes. In ac , we only ound one single nucleo ide
polymo phism (SNP) in e eno coding sequence, and ano he SNP in an in on
o AGO3, ha we e associa ed wi h he ac i i y o gypsy5 and pogo,
espec i ely. Those esul s a e o pa icula in e es gi en he close ela ionship
es ablished be ween he unc ion o hose p o eins and he place o gypsy5 and
pogo exp ession. In o he wo ds, e eno p o ein unc ions in he p ima y piRNA
biogenesis and in he cell ypes whe e gypsy5 is speci ically exp essed, whe eas
AGO3 ampli ies he abundance o piRNAs in he ge mline whe e pogo is
exclusi ely exp essed. Howe e , gi en ha pe u ba ions in he unc ion o hose
wo p o eins is desc ibed o esul in o e all uncon olled TE ac i i y73,74, we
would expec hose SNPs wi hin e eno and AGO3 sequences o be associa ed
!
81!
wi h a ia ion in he ac i i y o se e al ansposons. The TE-speci ic associa ion
o bo h loci o a ia ion in gypsy5 and pogo copy numbe was he e o e
unexpec ed and we u he sugges ha pleio opic e ec s be ween hose and
o he alleles could explain hose esul s.
Alleles in CG17839, unc-13-4A and Sin3A may explain di e ences in blood
ac i i y
Da a a ailable om p e ious s udies40,41 allowed us o con i m ha CG17839,
unc-13-4A and Sin3A a e able o egula e blood exp ession. Those esul s
sugges ha alleles a ec ing such genes could po en ially in luence blood
exp ession and lead o changes in blood copy numbe . Al hough he molecula
unc ion o CG17839 and unc-13-4A is s ill unknown, Sin3A is ecognized o i s
ansc ip ion ac o ac i i y. Sin3A lacks DNA-binding mo i s bu can ac as
sca old o he assembly o nume ous ansc ip ion ac o s75. We hen p opose
ha he non-synonymous mu a ion ound o be associa ed wi h a ia ion in blood
copy numbe migh a ec Sin3A abili y o in e ac wi h some o hose ac o s,
whe eas he SNP in in onic sequence may a ec i s exp ession. Bo h loci could
he e o e impac blood exp ession h ough e ec s in he ac i i y and/o
exp ession o his p o ein. Howe e , ou analysis o he o e exp ession o Sin3A
e ealed e ec s in he exp ession o many ansposons including blood and oo.
How alleles in Sin3A could media e speci ic e ec s in blood (as hose we e no
ound o be associa ed wi h a ia ion in oo ac i i y) is s ill la gely unknown. We
p opose ha he non-synonymous mu a ion may ha e a ec ed he in e ac ion
be ween Sin3A and co- ac o s able o in luence blood bu no oo exp ession.
This hypo hesis needs, howe e , o be es ed oge he wi h he unc ional
alida ion o hose loci.
Alleles associa ed wi h appa en bu s o gypsy5 ac i i y in RAL-790
The assessmen o TE popula ion dynamics in di e en species sugges s ha
ansposons mo e h ough wa es o inc eased ac i i y76,77. Howe e , e y ew
episodes o spon aneous bu s s in ansposi ion we e empi ically de ec ed.

!
82!
Those include a couple o examples om D. melanogas e isogenic lines
main ained in labo a o y condi ions78–80, and c osses be ween s ains wi h
incompa ible TE composi ion81,82. He e, o he i s ime, we epo ed a bu s in
TE ac i i y (gypsy5) in a geno ype (RAL-790) ha was collec ed om he wild
and con ained alleles seg ega ing in na u al popula ions. Some o he loci
associa ed wi h a ia ion in he numbe o gypsy5 inse ions we e ound o be
wi hin o close o genes ela ed o ollicle cell de elopmen . Gi en ha gypsy5 is
speci ically exp essed in hose cells, we p opose a model sugges ing ha alleles
a ec ing hose genes could esul in he ac i a ion o gypsy5 exp ession and/o
p omo e i s mobiliza ion o he oocy e (Figu e 2.4C and D). In his model, he
SNPs associa ed wi h gypsy5 ac i i y may ha e caused he in ensi ica ion o
EGFR signaling cascade ha can esul in mo e exp ession o he ansc ip ion
ac o poin ed (Figu e 2.4C). Also, e ec s in he No ch signaling pa hway,
oge he wi h p oblems in cell adhesion, could a ec he abundance o ollicula
cells and he in eg i y o he epi helium, ei he by in ensi ying cellula di ision
(Figu e 2.4d1) o endo eplica ion (Figu e 2.4d2). In summa y, we sugges ha
highe exp ession o poin ed, inc ease in ollicula cell numbe o highe genome
abundance could inc ease gypsy5 exp ession. Fu he mo e, we p opose ha he
dis up ion o he soma ic epi helium could acili a e he mo emen o gypsy5
i al-like pa icles om he ollicula cells o he oocy e.
Acco ding o ou model, RAL-790 should ha e highe exp ession o gypsy5 and
poin ed when compa ed o ha o o he DGRP lines. Al hough ha expec a ion
was con i med o TE exp ession, he le els o poin ed exp ession we e no
enough o explain gypsy5 bu s in ac i i y. We sugges ha , alleles a ec ing he
ac i i y o o he p o eins necessa y o he ac i a ion o Poin ed (such as hose
ha media e pos - ansla ional modi ica ions and ac i a ion h ough
phospho yla ion)83 could in luence he ac i i y o his ansc ip ion ac o wi hou
a ec ing i s exp ession. Howe e , u he analyses a e needed o cla i y his
hypo hesis.
I is also impo an o no e ha , al hough ZAM also con ained cis- egula o y
elemen s o Poin ed, we only obse ed highe copy-numbe o gypsy5 in RAL-
!
83!
790. We sugges ha he numbe o egula o y elemen s o Poin ed dic a es he
sensi i i y o changes in ha ansc ip ion ac o : a ansposon con aining mo e
binding si es would be mo e sensi i e o inc ease in Poin ed exp ession and/o
ac i i y.
In sum, ou GWAS analysis o he numbe o no el TE inse ions p o ed o be a
powe ul ool no only o ind na u al a ian s associa ed wi h, bu also o ob ain
unan icipa ed candida e egula o s o , ansposon ac i i y. Howe e , ou da a
p o ide less insigh a he mechanis ic le el and u u e s udies o he allelic
e ec s o gene exp ession and/o p o ein unc ion a e c i ical o ou
comp ehension o he gene ic ac o s a ec ing TE ac i i y in na u al popula ions.
Acknowledgemen s
We would like o acknowledge Rena o Al es o helping wi h bioin o ma ic ools
and Daniel Sob al o aligning nex gene a ion sequencing eads agains TE
canonical sequences. Wen Huang and Michael Magwi e o answe ing all ou
ques ions ega ding he D osophila Gene ic Re e ence Panel, and o Osca Ruiz
and El i a La uen e o eading his chap e . We a e also hank ul o Casey
Be gman o impo an eedback on his p ojec .
Au ho con ibu ions
The au ho o his hesis pe o med some expe imen s o con i m TE iden i y, did
s a is ical and bioin o ma ic analysis, and oge he wi h Pa ícia Beldade
designed all expe imen s. And é Al es checked p esence o alse nega i es,
con i med iden i y and size o some no el inse ions, and epea ed he
expe imen o con i m gypsy5 and poin ed exp ession. Ana Eugénio pe o med
all o he expe imen s.
!
85!
Supplemen a y In o ma ion
Supplemen a y Figu e 2.1. Range o quan i ica ion cycles (Cqs) o blood and
Rpl32 ha we e used in he analysis. Fo each sample wi h known amplicon
concen a ion we ob ained he amoun o cycles ha we e necessa y o de ec exp ession
by qPCR using de aul h eshold se ings. We an wo echnical eplica es and checked
ha he ela ionship be ween he known amplicon concen a ions (log) and he
quan i ica ion cycles (Cq) was linea . The ange o Cqs used o quan i y TE o gene
exp ession is wi hin he linea ampli ica ion. Same me hodology was used o de ec
exp ession o gypsy5, Tbp and poin ed.
!
92!
Supplemen a y Table 2.1. Con .
Posi ion o inse ion
DGRP
line
Ta
(ºC)
Fo wa d p ime
sequence
Re e se p ime
sequence
mdg1
3L:10095528..100955
32
RAL-357
60
AGAGGCCTGTGAGATC
CTTT
AAGTGTTGCCCCATTCGAA
3L:20245853..202458
56
RAL-357
60
TGCCCGTACATTTATGC
AGC
CGTGTGTGTGGAGCGATC
3R:5798180..5798183
RAL-357
60
AGGGCACACAGTTTTG
GTAC
CTCGAATGGCAACCTGTAC
T
3R:12117132..121171
36
RAL-357
62
CCAGAACGTCGACCAA
CC
GTGACCTAACGCAATACAC
AA
3R:23192134..231921
45
RAL-357
62
CCTCAACAAGTCGGCG
GG
CCCCATTCCAACGCACAA
X:16472629..1647264
3
RAL-357
60
TAAAGGCGAGAGGAAA
GAGC
TTTTGCCCCGTTCGAGTG
2L:19305699..193057
03
RAL-908
62
AACTCCACCGACTTGC
TTTT
TGGCTCTGACATGACCTAG
A
2R:11793310..117936
99
RAL-908
60
AAAGGAGCAGGAGAAG
GAGT
GAGGCAAAACAACGAGCAT
C
3L:676970..677090
RAL-908
60
ATATTCACGGTTTTGGC
CAG
GAGGGGCCGAGTCCTAAAT
3L:13514785..135147
89
RAL-908
60
CGCGCTTTGTTAACGT
GG
CATATTGGCTCGCGTTTTGT
3R:5737883..5737886
RAL-908
60
GAACACGCCGAGGATA
GC
CGAAAATCTACTCAGCCTG
CT
3R:8620803..8620806
RAL-908
60
CTGACTGAGAAATGGG
GCG
TCACCAAAGTTGCAGTCGA
A
3R:16918669..169186
73
RAL-908
62
ACCGGGTGCCAAGGAA
TG
CTTCCTCTTCGCTTCAGTG
G
3R:18045826..180458
29
RAL-908
57
ACTCTACGGCAAATGA
ATGGG
GGCGTGACCCTTACTGAA
X:5024909..5024914
RAL-908
60
ACAGAGAGCGACGAAA
GC
CGATGCGATGTCATGTCC
opus
2L:7336324..7336327
RAL-908
60
GCATGACGATTACGTG
GCTA
ACAACCAAACGCTTTTCAC
C
2R:17950637..179507
06
RAL-443
62
ATATGTCCTCGCCTGA
CCTG
GTTTCCACTGCACAGCCAT
A
oo
3R:17957511..179575
20
RAL-358
62
CCCACATAGTTGCATG
CCA
GTGTGTGGCAGTGGTAGT
3R:7884988..7884992
RAL-358
60
CAGTGCCCCATTGTTA
CGA
CGAACTCATTGCAACTCCC
T
3R:17515331..175153
34
RAL-358
60
TCGTTCTCTCAGCAACC
AAC
AGAGAACCGAAACGCTACT
G
X:21270475..2127047
7
RAL-908
62
TTGGCGGGGAACAAAT
ACAC
TGCCCCTTACCGCTCCAT
2L:9204672..9204675
RAL-908
60
TCTTGTTTCCCCGCATC
ATG
TCTACCTCCGGAGCTCACA
2L:9361897..9361901
RAL-908
60
CAACTGATGGGGTAAG
CAATG
GCGAGCAAAGACAGCACTA
2L:9705174..9705181
RAL-908
60
CGGTGCTTGAGGTGTC
CC
CACTTCAAGCCTGGTTACC
C
2L:12191481..121914
88
RAL-908
60
CACGACGTTAGAGCTG
GAAC
CACGCATCAAGTCAGGGTT
2L:12434248..124342
66
RAL-908
60
GAATTTCGGGCTGGTT
TTCA
CCATGATGTCCTGTGCTCC
2L:12902268..129022
74
RAL-908
60
GAATTTCGGGCTGGTT
TTCA
CCATGATGTCCTGTGCTCC

!
93!
Supplemen a y Table 2.1. Con .
Posi ion o inse ion
DGRP
line
Ta
(ºC)
Fo wa d p ime
sequence
Re e se p ime
sequence
T anspac
2R:12108723..121087
29
RAL-810
60
ATAAACCGCAGCAAAA
GTGG
CCAATGGATTTTCGAGAGG
A
3L:12856836..128568
42
RAL-908
60
CCCACTTCCTCTTCCAC
TCA
AGTCGACCAGGGACAATGA
C
X:345626..345630
RAL-357
RAL-892
58
TAACGATGGTGGCTGC
TACA
AAGGAAAGCGATTCAAGAC
C
X:15333499..1533350
4
RAL-443
60
CTGCAACTTTCCATGG
CTTT
ACAGCTTTCCCCTTCTGGA
T
Re o ansposons: non-LTR elemen s
Doc
2L:11138677..111387
34
RAL-357
RAL-892
60
AAAATCCATTCGGCAAA
CTG
TCGATCAGCGCCTAGTATC
A
3R:7873179..7873180
RAL-357
RAL-381
60
ATTGTCTGCGCAACTGT
CTG
ATGAATTCGTCTGCCTGTC
C
2R:7936918..7936946
RAL-810
60
CGAAGACATCAGTCCT
GCAA
CCGCTGACTGTGATTGCTA
A
3R:10301789..103018
22
RAL-908
60
GCACGAGACTCACACA
GGAA
TTATGGCCATTGTACGCTG
A
2L:1399467..1399478
RAL-908
60
TGCATCTGTGTGCGTAT
GTG
GCACTTTTTGCCTCTGTTCC
F-elemen
3R:6947766..6947992
RAL-810
60
TAGGCGCTGTTATTGAA
ACC
CAGTGAAAGTGGGTGCAAA
3L:7996067..7996071
RAL-908
60
GGGATTTGCTCTTGCT
CTTG
GCCATGGTCGAAACAAAAC
T
3R:9994118..9994126
RAL-908
60
GCTTGTCAAAGGGTCC
AAGA
TGTTATGTGCGCGAACTTG
T
I-elemen
2R:10573276..105732
86
RAL-357
60
TCCGTCGGCTCTTATTT
GTC
CGTCTTACACTCGCAGCAA
A
2R:13861586..138616
00
RAL-810
60
CCCAGATTCGCAATAC
CAAA
AACAAAAGCAACCACCAAG
G
3L:5068011..5068084
RAL-908
56
TCGAATTGATACAACCC
CAAT
CTACTACGGCGGTGTTGGT
T
3R:11405981..114059
87
RAL-357
60
GGCAGTGCAAACAAAA
ACAA
CTGAGGCCAAGGACTTATG
C
3R:14271683..142716
93
RAL-810
56
ACCTCATAGGGGGTGC
TTTT
TTGGAAGTGAAGGCTTTGA
A
2L:11388784..113887
93
RAL-790
60
CTTGAAGACCGGGTAC
TTCT
GCTGGCATATCTTCTCCGA
C
2R:7961110..7961122
RAL-790
60
AACCCAGGGAGCTAAG
TAGA
ATTGAATGTCCGGGGATCT
T
2R:12732718..127327
30
RAL-790
60
GCCGAACAGCATATAC
CCT
TCAAGCGTGTTTCCTCGAT
2R:14371197..143712
09
RAL-790
60
TACCCGCCACTCAATTA
TCC
AGTATGGCGTCGAGTGTG
3R:4277178..4277189
RAL-790
60
CTTTAAGCACCACGAG
ACGA
ACCCAAATGCAAAGCCGT
3R:14671496..146715
08
RAL-790
58
TAAATAAGTGCCTTCGC
CCC
TGTATCTCGGCTGTCTCCA
3R:21118597..211186
11
RAL-790
62
CCGGAGCCTGGTAGTT
CTT
TATCTTTCTCCCAGCCCGTC
3R:21756397..217564
00
RAL-790
60
CGCGCCTAGAACTATG
CAA
ACACACTAGCAAGCACTGG
2R:2648988..2649001
RAL-908
60
CCCCTCGTTTTCTACTG
CTAC
GTGGGTGTGGAACTCTGTG
2R:7224003..7224015
RAL-908
58
TAGCCTGCTTTTGTGTG
GAG
CTTGCAAGTTGGTTTTGGG
G
!
94!
Supplemen a y Table 2.1. Con .
Posi ion o inse ion
DGRP
line
Ta
(ºC)
Fo wa d p ime
sequence
Re e se p ime
sequence
I-elemen
3L:4501669..4501675
RAL-908
60
ACAGAAGTACAGTGAG
CGT
GGGGAGGTTCAATTGGTCA
3L:5068011..5068084
RAL-908
60
AGTCAATTCGCCTAGTA
CCAC
CAATTGGAGCTGCATCCTT
T
3L:10088380..100883
81
RAL-908
60
AGGGCACTTTCCTCGA
GA
ATATGCTTTGATACGGCGC
T
3L:22408430..224084
42
RAL-908
60
GTCTTGACGCCTTGCC
TAG
CACTGCATTTCAAACGCTC
G
3R:8071617..8071628
RAL-908
58
GAACACGCCGAGGATA
GC
CGAAAATCTACTCAGCCTG
CT
3R:11192794..111928
06
RAL-908
60
GCCTTCCACACGCATC
TG
GATGCCCGCACTGAGAGA
3R:22596610..225966
22
RAL-908
60
ACCCAAGTTCTTGTCG
CC
GCGGCGCATCAACTAATG
3R:22850421..228504
34
RAL-908
60
TCCGCTGGAGAAATTG
CAT
AATCTAAAAGGGGCTGCCA
X:302449..302464
RAL-908
60
TAAGTCACAACCCTACA
GCA
TGACAGCAGTTGGGATCAA
X:10027285..1002729
8
RAL-908
60
CGCTTACACTGTATTTG
CCC
GCCACCGTCTCTACTTGC
Juan
2R:5572279..5572291
RAL-810
60
CTAACACGTTTCCGCC
AAGT
TTCGAGGGTGTGGGTGTAT
T
3L:17698435..176984
91
RAL-357
RAL-761
60
TCAAGTCCCAGATGCA
CTCA
ATGTGGAACTTGGAGGATG
C
X:18924423..1892443
6
RAL-908
58
TCGAAGCCATTGCTATT
TTTG
TGACACCTATTCCTCAGACT
CG
3L:16130441..161304
56
RAL-443
60
CAATCGCCTAGATCGC
TTGT
AGTAGCAGGTCGCCTTGAA
A
DNA elemen s: TIR elemen s
hobo
3R:10703373..107033
79
RAL-357
60
CTCCCAAGGATTCTGT
CCAA
AATGTTTCCCAAAGCTGAC
G
3R:22133514..221335
21
RAL-357
60
GGGTCTGAAAGCAGCT
ATGG
CATTGTTCTTGGCTGACGA
A
2R:12871101..128711
13
RAL-810
60
TCAACGCTGAAAAGTAT
GCAA
GCAGATGATGTTGGCTTGA
A
3L:19555349..195553
54
RAL-761
RAL-810
RAL-812
60
AGCTTTAGCCACAGCC
ACAT
GAGAGGCACGCAGGTAAGA
C
3R:22134885..221348
92
RAL-908
60
CAAAGGCAGGGCTAAC
AAAA
CACAAGTGGGAGCATCAAC
A
hoppe
2L:2390528..2390533
RAL-357
60
ACCCATCAGACTTCCA
CGAC
GGAATCGCCTACAGAAGCT
G
3R:13938897..139389
01
RAL-357
RAL-812
60
TCGATTTGGCTGGAAA
CTCT
ATGCTGAACACGATGTGGA
A
3R:15010776..150107
80
RAL-908
60
GGGTACAATCAAATCG
AGCTTC
GCGAAAACTGCACTCAATC
A
X:4110151..4110442
RAL-810
60
CTTCGTTTCATTTGGCC
ATT
TGTGCCAAAAACACAGGCT
A
pogo
2R:11767856..117678
57
RAL-810
RAL-381
60
GGCTACGACATTTCCG
TTGT
AACCTATTCCTTGCGGACC
T
3L:21388576..213885
78
RAL-810
RAL-812
60
TTCAATACGGATTTGCC
ACA
GCAAAAATAAGGGCCATCC
T
3R:11292096..112920
97
RAL-357
60
GTTGAGCAAACAGACC
CACA
GGAGCCTCATAATCCGGTC
T
!
95!
Supplemen a y Table 2.1. Con .
Posi ion o inse ion
DGRP
line
Ta
(ºC)
Fo wa d p ime
sequence
Re e se p ime
sequence
pogo
3R:2926760..2926761
RAL-357
60
AACTCGAATCTGGCTC
GAAA
AGTGGCCTTATCGATTGGA
A
X:6015256..6015259
RAL-908
RAL-381
60
GATGTTTCGTGTGGCT
GTTG
GCAGTCGCTGCAGTTTGAT
A
3L:1571691..1571694
RAL-358
60
TGCACATGACTGGATT
CACA
CACACGAACATTGCTCCGA
2L:12470302..124703
09
RAL-908
60
TTAGAAAGCAAGTACC
GGCA
TTCTGCCATCGTGTTGCC
2L:13770337..137703
38
RAL-908
60
GTGGGGCCTCATAGAT
ACGT
TATGCGCTGAGGTACACTT
G
2L:13780514..137805
15
RAL-908
60
CCGGCCCATGTTAAGC
TTTA
GAGGCAGCGGATCAATTCA
2L:15151373..151513
77
RAL-908
58
TGTTCGGGTGGTAAAA
GCG
CAACGTTCCCAGGACACC
2L:18860089..188600
91
RAL-908
60
ACGCCTCGGATTTGAC
ATC
CCGTTGGCATTTTGTGGAT
A
2R:1749442..1749443
RAL-908
60
ATAGCACATTCAGCCA
CACG
TGCTGAATTCGGAAAGAGC
T
2R:5741587..5741594
RAL-908
60
GGTTTCGATTCGGTATT
GGTTG
ATTGCTTCGTGTTAGGACC
C
2R:6426442..6426443
RAL-908
60
CTGAGTCGAGCTGGTA
GGT
CGACATTTTCTGCGGCCG
Posi ion o inse ions is based on he assembly numbe i e o D. melanogas e genome.
Ta s ands o annealing empe a u e
Supplemen a y Table 2.2. DGRP lines wi h no inse ions de ec ed o blood, copia,
gypsy5, opus, jockey, H-elemen and pogo elemen s.
DGRP line
blood
copia
gypsy5
opus
jockey
H-elemen
pogo
RAL-109
X
X
RAL-161
X
RAL-237
X
X
X
X
X
X
X
RAL-350
X
RAL-362
X
X
RAL-555
X
RAL-776
X
RAL-808
X
!
96!
Supplemen a y Table 2.3. P ime s used o de ec alse nega i es and o qPCR
eac ions.
Ta ge
Fo wa d p ime sequence
Re e se p ime sequence
Re
blood
AACAATAGAAAGAAGCCACCGAAC
AGTCATGGACTATTGAGGGTGTTG
73
copia
TGCCAGAGAGCAAGTTCAGA
GCAAACCCAATTTGTCTCGT
gypsy5
GCCCAGAGACAACGACAGAA
CTGTCTTTGCTGTCCCGGAT
H-
elemen
CATTAAGTCGGAAGGCCAAA
CTTGCTCTTCCGCTATCCAC
jockey
GCGGATTAACAAGGGGCTCT
CCTGGGAGATAGATGCGCTG
opus
CGAGGAGTGGGGAGAGATTG
TGCGAAAATCTGCCTGAACC
85
pogo
CCAGCGATAACGAAGAAAGC
GCTGCAAACCCATCCTTAAA
poin ed
GGGGCGTGCTGTTGTTGATG
TCGCTGGGACTGGGCTACTTC
86
Rpl32
ATGCTAAGCTGTCGCACAAATG
GTTCGATCCGTAACCGATGT
Tbp
GGCAAAGAGTGAGGACGACT
GAGCCGACCATGTTTTGAAT
38
Re , e e ence
!
97!
Supplemen a y Table 2.4. P obabili ies es ima ed by Mackay e al. o he iden i y
o TE inse ions.
Posi ion o inse ion
DGRP
line
Mos likely
TE
P obabili ies o LEFT
eads
P obabili ies o RIGHT
eads
3L:
22887401..22887413
RAL-321
blood
R 1b, 0.33
C 1a, 0.28
blood, 0.93
R 1b, 0.07
3R:7387389..7387392
RAL-358
blood
blood, 0.47
S alke 2, 0.42
S alke 2, 0.42
1360, 0.21
2L:14056691..14056894
RAL-321
blood
blood, 0.6
S alke 2, 0.39
S alke 2, 0.52
INE-1, 0.25
X:10922194..10922200
RAL-790
blood
blood, 0.6
S alke 2, 0.39
S alke 2, 0.35
1360, 0.23
2L:1788665..1788852
RAL-357
mdg1
mdg1, 0.58
HMS-Beagle, 0.19
Emp y
3R:23192134..23192145
RAL-357
mdg1
Emp y
mdg1, 0.86
T anspac, 0.13
2R:19347971..19347974
RAL-357
mdg1
mdg1, 0.45
C 1a, 0.23
Emp y
X:5024909..5024914
RAL-908
mdg1
mdg1, 0.83
T anspac, 0.12
Ide ix, 0.79
INE-1, 0.21
X:21270475..21270477
RAL-908
oo
Emp y
oo, 0.57
pogo, 0.43
2L:12191481..12191488
RAL-908
oo
oo, 0.81
oo, 0.55
INE-1, 0.29
3R:17515331..17515334
RAL-358
oo
INE-1, 0.81
oo, 0.17
oo, 0.99
3R:7873179..7873180
RAL-357
Doc
Doc, 0.96
INE-1, 0.49
1360, 0.3
3R:7873179..7873180
RAL-381
Doc
Doc, 0.97
INE-1, 0.69
1360, 0.16
3R:6947766..6947992
RAL-810
F-elemen
INE-1, 0.8
F-elemen , 0.9
3L:4501669..4501675
RAL-908
I-elemen
INE-1, 0.74
I-elemen , 0.12
I-elemen , 1
2R:6426442..6426443
RAL-908
pogo
pogo, 0.58
oo, 0.19
pogo, 0.69
oo, 0.12
We only include in o ma ion o he inse ions analyzed in his s udy and ha con ained a
di e en TE iden i y han he one p edic ed by Mackay e al. Posi ion o inse ions is
based on he assembly numbe i e o D. melanogas e genome.

!
98!
Supplemen a y Table 2.5. Numbe o no el inse ions pe TE pe geno ype. TEs
ha had mo e han en no el inse ions a leas in one DGRP line (Maximum) a e
highligh ed in iole . F om hose, we chose eigh TEs o he GWAS analyses (highligh ed
in o ange). This able is p o ided in digi al o ma .
Supplemen a y Table 2.6. Alleles associa ed wi h blood, gypsy5, mdg1, oo, I-
elemen , jockey, H-elemen and pogo ac i i y. Table illus a es op alleles (single p-
alue<7e-05) associa ed wi h numbe o no el inse ions o each TE. We ob ained allelic
posi ion using he assembly numbe i e o D. melanogas e genome. This able is
p o ided in digi al o ma .
Supplemen a y Table 2.7. Gene on ology (GO) analysis o genes con aining alleles
associa ed wi h gypsy5 ac i i y. GO analysis was pe o med o assess en ichmen o
ce ain biological p ocesses in he g oup o genes con aining loci wi h allelic a ia ion
a ec ing blood ac i i y. Signi ican GO e ms we e ob ained a e Bon e oni co ec ion o
mul iple es ing. En ichmen o o a ian ollicle cell de elopmen (GO:0030707) genes is
highligh ed in o ange. This able is p o ided in digi al o ma .
!
99!
Re e ences
1. Ba ón, M. G., Fis on-La ie , A.-S., Pe o , D. A. & González, J.
Popula ion Genomics o T ansposable Elemen s in D osophila. Annu.
Re . Gene . 48, 561–581 (2014).
2. Vicien , C. M. T ansc ip ional ac i i y o ansposable elemen s in maize.
BMC Genomics 11, 601 (2010).
3. González, J., Lenko , K., Lipa o , M., Macphe son, J. M. & Pe o , D. a.
High a e o ecen ansposable elemen -induced adap a ion in
D osophila melanogas e . PLoS Biol. 6, 2109–2129 (2008).
4. Lisch, D. How impo an a e ansposons o plan e olu ion? Na Re
Gene 14, 49–61 (2012).
5. Ga cía Gue ei o, M. P. Wha makes ansposable elemen s mo e in he
D osophila genome? He edi y (Edinb). 108, 461–468 (2012).
6. B ennecke, J. e al. Disc e e Small RNA-Gene a ing Loci as Mas e
Regula o s o T ansposon Ac i i y in D osophila. Cell 128, 1089–1103
(2007).
7. A a in, A. A., Sachidanandam, R., Gi a d, A., Fejes-To h, K. & Hannon,
G. J. De elopmen ally egula ed piRNA clus e s implica e MILI in
ansposon con ol. Science (80-. ). 316, 744–747 (2007).
8. Kawaoka, S. e al. The Bombyx o a y-de i ed cell line endogenously
exp esses PIWI/PIWI-in e ac ing RNA complexes. RNA 15, 1258–64
(2009).
9. Lau, N. C., Ohsumi, T., Bo owsky, M., Kings on, R. E. & Blowe , M. D.
Sys ema ic and single cell analysis o Xenopus Piwi-in e ac ing RNAs
and Xiwi. EMBO J. 28, 2945–2958 (2009).
10. Khu ana, J. S. e al. Adap a ion o P Elemen T ansposon In asion in
D osophila melanogas e . Cell 147, 1551–1563 (2011).
11. A a in, A. e al. Double-s anded RNA-media ed silencing o genomic
andem epea s and ansposable elemen s in he D. melanogas e
ge mline. Cu . Biol. 11, 1017–1027 (2001).
12. Kalmyko a, A. I., Kleno , M. S. & G ozde , V. A. A gonau e p o ein PIWI
con ols mobiliza ion o e o ansposons in he D osophila male
ge mline. 33, 2052–2059 (2005).
13. Mackay, T. F. C. e al. The D osophila melanogas e Gene ic Re e ence
Panel. Na u e 482, 173–178 (2012).
14. McClin ock, B. The o igin and beha iou o mu able loci in maize.
Gene ics 36, 344–355 (1950).
15. Amine zach, Y. T., Macphe son, J. M. & Pe o , D. A. Pes icide
!
100!
Resis ance ia T ansposi ion-Media ed Adap i e Gene T unca ion in
D osophila. Science (80-. ). 309, 764–767 (2005).
16. Dabo n, P. J. e al. A Single P450 Allele Associa ed wi h Insec icide
Resis ance in D osophila. Science (80-. ). 2253, 2253–2256 (2013).
17. Ba oso-Ba is a, J. e al. The Fi s S eps o Adap a ion o Esche ichia coli
o he Gu A e Domina ed by So Sweeps. PLoS Gene . 10, (2014).
18. Guio, L., Ba ón, M. G. & González, J. The ansposable elemen Ba i-
Jheh media es oxida i e s ess esponse in D osophila. Mol. Ecol. 23,
2020–2030 (2014).
19. Eme a, D. & Wagne , G. P. T ansposable elemen ec ui men s in he
mammalian placen a: Impac s and mechanisms. B ie . Func . Genomics
11, 267–276 (2012).
20. San os, M. E. e al. The e olu ion o cichlid ish egg-spo s is linked wi h a
cis- egula o y change. Na . Commun. 5, 5149 (2014).
21. Ho , A. E. anʼ e al. The indus ial melanism mu a ion in B i ish
peppe ed mo hs is a ansposable elemen . Na u e 534, 102–105 (2016).
22. Wicke , T. e al. A uni ied classi ica ion sys em o euka yo ic
ansposable elemen s. Na . Re . Gene . 8, 973–982 (2007).
23. Le in, H. L. & Mo an, J. V. Dynamic in e ac ions be ween ansposable
elemen s and hei hos s. Na . Re . Gene . 12, 615–627 (2011).
24. Cu cio, M. J. & Ga inkel, D. J. New lines o hos de ense : inhibi ion o
Ty1 e o ansposi ion by Fus3p. T ends Gene . 15, 43–45 (1999).
25. Dai, J., Xie, W., B ady, T. L., Gao, J. & Voy as, D. F. Phospho yla ion
Regula es In eg a ion o he Yeas Ty5 Re o ansposon in o
He e och oma in. Mol. Cell 27, 289–299 (2007).
26. Siebel, C. W. & Rio, D. C. Regula ed splicing o he D osophila P
ansposable elemen hi d in on in i o: soma ic ep ession. Science
(80-. ). 248, 1200–1208 (1990).
27. Kibano , M. V e al. A no el o ganelle, he piNG-body, in he nuage o
D osophila male ge m cells is associa ed wi h piRNA-media ed gene
silencing. Mol. Biol. Cell 22, 3410–3419 (2011).
28. Malone, C. D. e al. Specialized piRNA Pa hways Ac in Ge mline and
Soma ic Tissues o he D osophila O a y. Cell 137, 522–535 (2009).
29. Meignin, C., Das ugue, B. & Vau y, C. In e cellula communica ion
be ween ge m line and soma ic line is u ilized o con ol he ansc ip ion
o ZAM, an endogenous e o i us om D osophila melanogas e .
Nucleic Acids Res. 32, 3799–3806 (2004).
30. Tou e , F., Guiguen, F. & Te zian, C. Wolbachia In luences he Ma e nal
!
101!
T ansmission o he gypsy Endogenous Re o i us in D osophila
melanogas e . MBio 5, 1–10 (2014).
31. Huang, W. e al. Na u al a ia ion in genome a chi ec u e among 205
D osophila melanogas e Gene ic Re e ence Panel lines Na u al
Va ia ion in Genome A chi ec u e Among 205 D osophila melanogas e
Gene ic Re e ence Panel Lines. Genome Res. 24, 1193–1208 (2014).
32. The Gene On ology Conso ium. Gene On ology: ool o he uni ica ion o
biology. Na . Gene . 25, 25–29 (2000).
33. Ch os ek, E. e al. Wolbachia Va ian s Induce Di e en ial P o ec ion o
Vi uses in D osophila melanogas e : A Pheno ypic and Phylogenomic
Analysis. PLoS Gene . 9, 1–22 (2013).
34. Mendes, C. C. & Mi h, C. K. S age-Speci ic Plas ici y in O a y Size Is
Regula ed by Insulin/Insulin-Like G ow h Fac o and Ecdysone Signalling
in D osophila. Gene ics (2015). doi:10.1534/gene ics.115.179960
35. Ande sen, C. L., Jensen, J. L. & Ø n o , T. F. No maliza ion o eal- ime
quan i a i e e e se ansc ip ion-PCR da a: a model-based a iance
es ima ion app oach o iden i y genes sui ed o no maliza ion, applied o
bladde and colon cance da ase s. Cance Res. 64, 5245–5250 (2004).
36. Vandesompele, J. e al. Accu a e no maliza ion o eal- ime quan i a i e
RT-PCR da a by geome ic a e aging o mul iple in e nal con ol genes.
Genome Biol. 3, 1–12 (2002).
37. Pon on, F., Chapuis, M.-P., Pe nice, M., Swo d, G. A. & Simpson, S. J.
E alua ion o po en ial e e ence genes o e e se ansc ip ion-qPCR
s udies o physiological esponses in D osophila melanogas e . J. Insec
Physiol. 57, 840–850 (2011).
38. Ma a, B. P., Bi ne -Ma hé, B. C. & Al es-Fe ei a, M. Ge ing eal wi h
eal- ime qPCR: A case s udy o e e ence gene selec ion o
mo phological a ia ion in D osophila melanogas e wings. De . Genes
E ol. 221, 49–57 (2011).
39. R-Co e-Team. R - A language and en i onmen o s a is ical compu ing.
h ps://www. -p ojec .o g/ (2012).
40. Saha, N., Liu, M., Gajan, A. & Pile, L. A. Genome-wide s udies e eal
no el and dis inc biological pa hways egula ed by SIN3 iso o ms. BMC
Genomics 17, 111 (2016).
41. Czech, B., P eall, J. B., McGinn, J. & Hannon, G. J. A ansc ip ome-wide
RNAi sc een in he D osophila o a y e eals ac o s o he ge mline
piRNA pa hway. Mol. Cell 50, 749–61 (2013).
42. Sha ma, V., Swamina han, A., Bao, R. & Pile, L. A. D osophila SIN3 is
equi ed a mul iple s ages o de elopmen . De . Dyn. 237, 3040–3050
(2008).

!
109!
3.1 Abs ac
T ansposable elemen s (TEs) a e impo an con ibu o s o pheno ypic
di e si ica ion1–3 and ep esen an impo an sou ce o adap i e gene ic
a ia ion4,5. Howe e , a la ge p opo ion o inse ions is elimina ed due o
dele e ious e ec s6, and hos s bene i om mechanisms ha speci ically a ge
TEs in he ge mline, and ha a e able o con ol he p oduc ion o he i able
gene ic a ia ion7–11. Despi e ecen ad ances in he unde s anding o hose
p ocesses, he en i onmen al and gene ic ac o s ha igge ansposi ion in
D osophila melanogas e emain poo ly desc ibed12. In addi ion, s udies abou
he impo ance o bio ic13–15 and abio ic16–18 pe u ba ions in TE ac i i y p o ided
con lic ing esul s as each assess e ec s in di e en ansposons and in di e en
gene ic backg ounds. In his wo k, we o e came hose p oblems by es ing
GxExE in e ac ions in he ac i i y o di e en ansposons. In pa icula , we
s udied he e ec o a bio ic (Wolbachia) and an abio ic ( empe a u e) ac o in
he exp ession o e o ansposons (blood, mdg1 and oo) in he o a ies o ou
di e en ully sequenced, wild-de i ed geno ypes (geno ypic ac o s)19. How
changes in empe a u e du ing adul li e in luenced TE esponses o he
p esence o Wolbachia (ExE in e ac ion) was also included in he analysis.
In e es ingly, we obse ed geno ype- and TE-dependen e ec s o Wolbachia
and empe a u e in blood, mdg1 and oo exp ession. Mo e speci ically, he
p esence o endosymbion and he mal pe u ba ions we e able o educe,
induce o con e no e ec in he exp ession o ansposons. We discuss he
in luence o hos ansc ip ion ac o s as media o s o hose esponses.
3.2 In oduc ion
T ansposable elemen s (TEs) a e conside ed impo an d i e s o genome
e olu ion. They cons i u e a la ge p opo ion o hos DNA and ep esen a majo
sou ce o gene ic a ia ion6,20. Al hough ecen s udies ecognized ha TEs a e
impo an con ibu o s o pheno ypic di e si ica ion1–3, and ha se e al inse ions
ha e adap i e alue4,5, a la ge p opo ion is s ill elimina ed due o pu a i e
dele e ious e ec s6.
!
110!
When o e -ac i e in he ge mline, TEs can damage genomic in eg i y and lead
o s e ili y21. As a consequence, hos s ha e e ol ed s a egies o speci ically
ecognize mobile elemen s and con ol ansposi ion in ge minal cells7–11.
Al hough ecen ad ances ha e unco e ed many aspec s o such mechanisms,
he en i onmen al and gene ic ac o s ha igge jumping emain la gely
unknown12. Also, and hough ansposon inse ions in luence hos adap a ion o
en i onmen al pe u ba ions22, he e is no consensus abou he impo ance o
speci ic s esses, bio ic13–15 and abio ic16–18, in D. melanogas e ansposi ion
a es. Inconsis encies ob ained o TE esponse o hea shock16–18 and sca ce
s udies on ansposon ac i i y in lies in ec ed wi h na u al pa hogens14 highligh
he demand o u he a en ion o his subjec .
D. melanogas e genome con ains se e al inse ions o di e en mobile
elemen s ha can be g ouped in o wo main classes6,23. Those classi y TEs
acco ding o sequence p ope ies, and di e en ia e e o ansposons om DNA
elemen s23. Re o ansposons mo e ia an RNA in e media e, and sha e
s uc u al commonali ies wi h e o i uses; DNA elemen s mo e ia a non-
eplica i e cu -and-pas e mechanism, and a e usually lanked by e minal
in e ed epea s24. They bo h exp ess p o eins ha a e c ucial o comple e each
cycle o ansposi ion, and o media e in eg a ion in hos genome25. DNA
elemen s code o a ansposase p o ein, while e o ansposons and
e o i uses usually exp ess a i al-pa icle coa , e e se ansc ip ase,
ibonuclease H and in eg ase25. T adi ionally, e o ansposons a e di ided in o
wo main g oups based on he p esence o absence o long e minal epea s
(LTRs). In D. melanogas e , mos copies belong o he LTR g oup6 ha can
con ain cis- egula o y elemen s wi hin 5ʼ LTR o TE open eading ames
22.
Those egula o y elemen s a e necessa y o igge exp ession o ansposons,
and a e ypically ecognized by hos ansc ip ion ac o s26–28. In e es ingly,
expe imen s done in plan species p o ed impo ance o such elemen s in
media ing TE esponse o speci ic en i onmen al s esses22,26–30. This idea was
ne e empi ically es ed in me azoans.
!
111!
He e, we p opose o s udy he e ec o he in e ac ion be ween wo
en i onmen al ac o s (one bio ic and ano he abio ic) in he ac i i y o
ansposons o D. melanogas e o a ies. Fo he bio ic ac o , we es ed he
impo ance o Wolbachia – an obliga o y in acellula bac e ia ha in ec s se e al
a h opod and nema ode species31–33. The s udy o he e ec o his ma e nally
ansmi ed endosymbion in he ac i i y o ansposons is o pa icula in e es
gi en 1) Wolbachiaʼs abili y o inc ease hos su i al o RNA i al in ec ion34,35
and 2) he s uc u al commonali ies be ween e o i uses and e o ansposons.
In acco dance wi h hese wo obse a ions, we hypo hesize ha he p esence o
Wolbachia in D. melanogas e o a ies could educe e o ansposon exp ession,
simila o ha obse ed o i al load34. Also, and gi en ha some Wolbachia-
media ed hos pheno ypes (such as male killing in D. bi ascia a36) a e sensi i e
o empe a u e, we decided o u he es o which ex en he mal pe u ba ions
du ing adul li e in luenced Wolbachia-induced TE esponses. The exp ession o
h ee e o ansposons (blood, mdg1 and oo) was assessed in o a ies collec ed
om ou ully sequenced, wild-de i ed geno ypes. We obse ed geno ype- and
TE-dependen e ec s o Wolbachia and empe a u e in blood, mdg1 and oo
exp ession. Mo e speci ically, he p esence o endosymbion and he mal
pe u ba ions we e able o educe, induce o con e no e ec in he exp ession o
ansposons. We u he sugges ha hos ansc ip ion ac o s may media e
hose esponses.
3.3 Ma e ials and Me hods
D osophila melanogas e s ocks and main enance
O egon-R s ock was ob ained om Rui Ma inho labo a o y, and was na u ally
clea ed o Wolbachia in ec ion. Flies om he D osophila Gene ic Re e ence
Panel (DGRP)19 (RAL-021, RAL-237, RAL-321 and RAL-790) ha we e na u ally
in ec ed wi h he same s ain o Wolbachia (wMel-like geno ype37), we e
pu chased om Blooming on s ock cen e . To ob ain gene ically iden ical
geno ypes in ec ed o no wi h Wolbachia, hose lies we e ea ed wi h
e acycline (0.05mg/ml) o wo gene a ions, and gu lo a con en homogenized
!
112!
as p e iously desc ibed38. Flies we e kep in s anda d co nmeal ood39 a 25°C,
60% humidi y and 12h day-da k cycles, unless o he wise men ioned.
Tempe a u e ea men s
To assess TE esponse o empe a u e change a di e en ime poin s, we
collec ed 0-8h old O egon-R lies and kep i e emales wi h wo males o h ee
days a 25ºC. Males we e hen emo ed and emales placed a 25ºC (con ol
condi ion) o 29ºC (expe imen al condi ion). O a ies we e dissec ed a day one,
wo, h ee and ou a e empe a u e change, o RNA ex ac ion. We pooled
eigh pai s o o a ies pe sample, ou eplica es pe condi ion.
To es he e ec o D. melanogas e gene ic backg ound in TE esponse o
empe a u e and Wolbachia, we collec ed 0-8h old RAL-021, RAL-237, RAL-321
and RAL-790 lies in ec ed o no wi h endosymbion . Fo each geno ype and
Wolbachia s a us, i e emales we e kep wi h wo males o h ee days, a 25ºC.
Males we e hen emo ed and emales placed a di e en empe a u es, 25ºC
o con ol, 21ºC and 29ºC o expe imen al condi ions. O a ies we e dissec ed
a day ou a e he mal change o RNA ex ac ion. We pooled eigh pai s o
o a ies pe sample, six o eigh eplica es pe condi ion.
RNA ex ac ion and cDNA syn hesis
O a ies we e dissec ed in esh cold PBS 1x, and di ec ly s o ed a 4ºC in 400μl
o T izol (Ambicon) un il homogenized by pes els. Tissue collec ion and
homogeniza ion was done in he same day; samples we e s o ed a -80ºC un il
RNA ex ac ion. We polled eigh pai s o o a ies pe sample. Fo RNA ex ac ion
we used Di ec -zol RNA Minip ep ki (Zymo Resea ch) and o al RNA was elu ed
in 25μl o RNase- ee wa e (Sigma). RNA pu i y and concen a ion o each
sample was measu ed wi h Nanod op ND-1000 Spec opho ome e ; A600/A800
abso bance was a ound wo o all samples, and concen a ions anged om
200 o 800ng/µl. All RNA samples o he same geno ype we e p ocessed on he
same day.
!
113!
Fo cDNA syn hesis we emo ed con amina ions o gDNA om 1µg o RNA
using DNAse (P omega) ea men . Fo cDNA syn hesis, we ollowed Re e se
T ansc ip ion Sys em (P omega) p o ocol, using 0.02µM Oligo dT p ime s.
Reac ion was incuba ed a 42°C o 60 min and hea ed a 95°C o 5 min. cDNA
was dilu ed in RNAse- ee wa e (1:10) and s o ed a -20°C un il quan i a i e
eal- ime PCR (qPCR).
qPCR
We measu ed le els o mRNA exp ession by qPCR in CFX384 he mal cycle
(BioRad). Fo each eac ion we used 5µl iQ™ SYBR® G een supe mix
(BioRad), 1µl o 4µM p ime s and 4µl o dilu ed cDNA (1:10). P ime sequences
a e desc ibed in Supplemen a y Table 3.1. We un wo echnical eplica es pe
sample, pe pla e. The mal cycling p o ocol was he same o all eac ions: 2 min
a 50°C; 10 min a 95°C; 40 cycles o 95°C o 30 sec, 60°C o 1 min and 72°C
o 30 sec. We used de aul h eshold se ings o ob ain he quan i ica ion cycle
(Cq) o each eac ion. qPCR mel ing cu es we e analyzed o con i m speci ici y
o ampli ied p oduc s and nega i e con ols ne e showed de ec able
ampli ica ion.
To es he dynamic o TE exp ession in O egon-R o a ies upon exposu e o
25ºC and 29ºC, we used P a l me hod40 ha co ec s exp ession o he
e iciency o each p ime pai . We used Rpl32 as e e ence gene, gi en ha he
alues ob ained o exp ession s abili y using No mFinde 41 and geNo m42
algo i hms we e be e o ha gene (Rpl32: No mFinde =0.04, geNo m=0.60;
Tbp: No mFinde =0.05, geNo m=0.61). Fold-change o exp ession was
de e mined o each TE, using as no malize he a e age ela i e exp ession
ob ained o samples collec ed in day one, and kep a 25ºC.
To assess le els o TE and gene exp ession o DGRP samples, we an in each
pla e s anda d cu es (one s anda d cu e o each gene o TE es ed) wi h
se ial dilu ions o known amplicon concen a ions. To ob ain amplicons o
s anda d cu es, we ampli ied a ge egions by PCR using 0.25U o GoTaq
(P omega), 1x eac ion bu e , 1.5mM MgCl2, 0.4µM p ime s (Supplemen a y

!
114!
Table 3.1) and 1µl o cDNA (1:10). PCR p oduc s we e cleaned using
NucleoSpin® Gel and PCR Clean-up (Mache ey-Nagel), elu ed in 15µl o bu e
NE, and concen a ions measu ed in Nanod op ND-1000 Spec opho ome e .
The minimum amoun o nucleic acids ha can be de ec ed by qPCR o each
p ime pai was es ima ed by ex ending dilu ions o ou s anda ds un il he
ela ionship be ween he known amplicon concen a ions and he quan i ica ion
cycles (Cqs) eached a pla eau. Samples ha had Cq alues ou side he linea
de ec ion and wi h s anda d de ia ion abo e 0.5 o he echnical eplica es we e
elimina ed om he analysis. To ob ain le els o exp ession we ans o med Cq
alues in o absolu e a ge concen a ions using espec i e s anda d cu es.
This me hodology allowed us o con ol o pla e e ec s, and o di e ences in
p ime e iciency. Fo he analysis o exp ession da a, we used No mFinde 41
and geNo m42 algo i hms o check exp ession s abili y o wo e e ence genes
commonly used o no malize gene exp ession a di e en en i onmen al
condi ions and geno ypes, Rpl32 and Tbp43,44. S abili y alues ob ained o
Rpl32 (RAL-021: No mFinde =0.26; geNo m=0.36; RAL-237: No mFinde =0.24;
geNo m=0.22; RAL-321: No mFinde =0.07; geNo m=0.19; RAL-790:
No mFinde =0.03; geNo m=0.13) and Tbp (RAL-021: No mFinde =0.33;
geNo m=0.33; RAL-237: No mFinde =0.15; geNo m=0.20; RAL-321:
No mFinde =0.09; geNo m=0.21; RAL-790: No mFinde =0.08; geNo m=0.13)
we e simila in bo h algo i hms. We hen decided o use Tbp as con ol gene o
RAL-021, and Rpl32 o RAL-237, RAL-321 and RAL-790 as hose seem o be
mo e s able o he espec i e DGRP lines (Supplemen a y Figu e 3.1).
gDNA ex ac ion
gDNA om RAL-021, RAL-321 and RAL-790 male adul s was ex ac ed using
DNeasy Blood & Tissue Ki (Qiagen). Tissues we e homogenized in 180µl o
bu e ATL, using pes els. We used 200µl o bu e AE o elu e gDNA and ea ed
all samples wi h 0.2mg/ml o RNAse (In i ogen) o 20 min a 55ºC. gDNA
concen a ion was measu ed in Nanod op ND-1000 Spec opho ome e , and
samples s o ed a -20°C un il nes ed PCR.
!
115!
Nes ed PCR
We designed p ime s o he lanking egion o blood o oo inse ions p edic ed
wi hin CG31751, CG10073, CG31974, CG8920, CG4374, CG5440, CG11319,
CG2022, CG5681, CG17883, CG33966, CG5210, CG9194, po , Sema-1a,
Ac 42A, Dg, Ta 12, Fas2, Topo s, ws and dnc genes (Supplemen a y Figu e
3.2A). P ime sequences a e desc ibed in Supplemen a y Table 3.2. We
ob ained amplicon by longPCR (Roche): 4ng o RAL-021, RAL-321 o RAL-790
gDNA, 0.5µM p ime s, 2% DMSO, 1x bu e , 0.5mM dNTPs, and 0.21µl o
enzyme pe 15µl o eac ion. We used s anda d ampli ica ion p og am o
longPCR wi h 10 min o elonga ion ime, and 60ºC as annealing empe a u e.
Size o inse ion was ob ained by elec opho esis, and 1µl o he p oduc o
longPCR (1:100) used as empla e o a second PCR eac ion: 0.25U o GoTaq
(P omega), 1x eac ion bu e , 1.5mM MgCl2 and 0.4µM o p ime s o blood o
oo (Supplemen a y Table 3.1). This me hodology allowed us o con i m
whe he blood o oo inse ions ha we e p edic ed o ha e landed inside genes
and ha we e pu a i ely exp essed oge he wi h p o ein-coding mRNAs
con ained sequences de ec ed by qPCR.
S a is ical analyses
All s a is ical analyses we e done in R ( e sion 3.0.3).
To assess he e ec o empe a u e and ime o exposu e (ca ego ical ixed
a iables) in TE ac i i y, we used linea models (lm) wi h logno mal dis ibu ion
(O egon-R da ase ). In e ac ion be ween empe a u e and ime was emo ed i
no con ibu ing o explain di e ences in exp ession. We es ed he signi icance
(p<0.05) o he e ec o empe a u e a each ime poin by using leas -squa es
means (lsmeans). When ime o exposu e in luenced TE exp ession, ega dless
o he e ec o empe a u e, we ob ained an es ima e and signi icance o ha
e ec , using same models desc ibed abo e and including ime o exposu e as
con inuous ixed a iable.
To es o he e ec o Wolbachia and empe a u e (ca ego ical ixed a iables)
on ela i e TE and gene exp ession (dependen a iables) in RAL-021, RAL-237,
!
116!
RAL-321 and RAL-790 o a ies, we used lm wi h logno mal dis ibu ion. When
he in e ac ion be ween hose wo ac o s (Wolbachia and empe a u e) was
s a is ically signi ican we used lsmeans o es o he e ec o Wolbachia a
each empe a u e. Each ansposable elemen , gene and geno ype was
analyzed sepa a ely.
Cha ac e ize place o TE inse ions
We ex ac ed a ailable in o ma ion o blood, mdg1 and oo inse ions o RAL-
021, RAL-237, RAL-321 and RAL-790 ha we e deposi ed online
(h p://www.hgsc.bcm. mc.edu/p ojec s/dg p/19) and used Va ian E ec
P edic o ool45 (Ensembl) o u he cha ac e ize inse ions acco ding o hei
genomic posi ion (e.g. coding s. non-coding). Fo hose ha landed inside o in
he neighbo egions o gene sequences, we un a gene on ology (GO) analysis
o assess en ichmen o inse ions a ec ing genes wi h pa icula molecula
unc ions. We used GO conso ium algo i hm, and Bon e oni co ec ion o
mul iple es ing46.
P edic ion o ansc ip ion ac o binding si es
We sea ched o ansc ip ion ac o binding si es in blood, mdg1 and oo
canonical sequences using Consi e47 and TFBIND48 algo i hms. We only
conside ed cis- egula o y elemen s ha we e p edic ed by bo h p og ams, as
p e iously desc ibed49. As Consi e and TFBIND use da abases o mammalian
binding si es, we only show ansc ip ion ac o s ha a e known o ecognize
conse ed cis- egula o y elemen s.
3.4 Resul s
Wolbachia s a us and empe a u e pe u ba ions in luence TE ac i i y in a
geno ype-dependen way
In o de o assess he dynamic o TE esponse o empe a u e, we exposed
adul emales o 29ºC (and compa ed hem wi h con ol emales ha we e
main ained a 25ºC) o di e en pe iods o ime. We obse ed ha TE
!
117!
exp ession was a ec ed by empe a u e, and ha hose e ec s we e dependen
on he ime o exposu e and on he TE analyzed ega dless o sequence
p ope ies (Supplemen a y Figu e 3.3). Gi en ha mo e TEs esponded o
empe a u e a e ou days o exposu e, we hen decided o use his ime poin
o analyze he e ec o Wolbachia on TE ac i i y o ou geno ypes upon
exposu e o di e en empe a u es (21ºC, 25ºC and 29ºC).
In e es ingly, he dynamic o TE esponse o en i onmen al ac o s (bio ic and
abio ic) was speci ic o each mobile elemen and geno ype analyzed
(Supplemen a y Figu e 3.4 and Figu e 3.1). Al hough he exp ession o blood,
mdg1 and oo in one geno ype (RAL-790) was obus o en i onmen al
pe u ba ions, i esponded o one o bo h ac o s o he o he gene ic
backg ounds analyzed (RAL-021, RAL-237 and RAL-321). Also, and when he e
was an e ec , mobile elemen s o he same geno ype esponded in he same
way o each en i onmen (Wolbachia o empe a u e) bu no be ween hem. In
pa icula , we obse ed h ee ypes o esponses ha we e geno ype-, TE- and
en i onmen -speci ic, and included dec ease, inc ease and no e ec in TE
exp ession (Figu e 3.1). Only o one o he geno ypes analyzed (RAL-321) we
ha e obse ed Wolbachia-media ed esponses in TE exp ession ha we e
dependen on empe a u e (Supplemen a y Figu e 3.4).
To unde s and o which ex en en i onmen al ac o s also a ec oocy e
p og ession, we assessed he exp ession o CG15279 ha is mos ly exp essed
in la e s ages o oocy e de elopmen 50 (Figu e 3.2A). We obse ed geno ype-
dependen e ec s o Wolbachia s a us and empe a u e pe u ba ions in he
exp ession o CG15279 (Figu e 3.2A and Supplemen a y Figu e 3.5). In
pa icula , o mos geno ypes (RAL-237, RAL-321 and RAL-790) we de ec ed
e ec s o bo h Wolbachia and empe a u e, and o only one geno ype (RAL-
021) we ha e obse ed an e ec o he in e ac ion be ween hese wo ac o s.
!
124!
E ec o Wolbachia in TE exp ession and i al load may be ela ed
Wolbachia is an endosymbion able o sp ead in na u al popula ions o D.
melanogas e 61, and o con e p o ec ion agains i uses34,35. We hypo hesized
ha he p esence o his endosymbion in D. melanogas e o a ies could also
educe e o ansposon exp ession, simila o ha obse ed o i al load34.
Al hough he mechanism unde lying he ipa i e in e ac ion be ween hos ,
Wolbachia and i us is s ill la gely unknown, ecen s udies ha e unco e ed a
possible associa ion be ween Wolbachia-induced oxida i e s ess and i s abili y
o p omo e hos su i al upon i al in ec ion62,63. I is he balance be ween he
p oduc ion and des uc ion o eac i e oxygen species (ROS) ha e lec s he
deg ee o oxida i e s ess64. As one possibili y, we hypo hesize ha Wolbachia
may be able o educe hos abili y o es ablish an ioxidan de ences h ough
nega i e e ec s in ansc ip ion ac o s ha p omo e cellula esponses o
oxida i e s ess esul ing in inc eased le els o ROS.
In e es ingly, blood and oo exp ession in RAL-021 o a ies ollowed ha o ou
p edic ion and could be associa ed wi h p esence o cis- egula o y elemen s o
a ansc ip ion ac o ha media es cellula esponses o oxida i e s ess (Cap-n-
colla )65. This obse a ion sugges s ha TE and i al esponses o Wolbachia
s a us may sha e impo an playe s possibly ela ed o he induc ion o oxida i e
s ess.
En i onmen al pe u ba ion as a d i e o he p oduc ion o gene ic
a ia ion
Inse ions o mobile elemen s in luence hos adap a ion and esponse o
en i onmen al pe u ba ions22. Acco dingly, i has p e iously been sugges ed
ha , when exposed o s ess ul condi ions, o ganisms could bene i om he
p oduc ion o no el adap i e gene ic a ia ion h ough he ac i a ion o
ansposons66. He e, we p oposed o explo e he e ec o bio ic and abio ic
ac o s in D. melanogas e TE ac i i y, using exp ession in he ge mline as a
p oxy o ansposi ion a es. Also, and gi en he eplica i e p ope ies o
e o ansposons, we sugges ha inc eased TE ac i i y in he ge mline can

!
125!
culmina e in highe mu a ion bu den ha is ansmi ed o he nex gene a ion.
In e es ingly, we obse ed ha he abili y o p oduce he i able gene ic a ia ion
is no only dependen on he en i onmen , bu also on he gene ic backg ound.
Al hough some geno ypes we e obus o en i onmen al changes and had no
po en ial o igge en i onmen ally media ed TE ac i i y (RAL-790) o educed
TE exp ession in esponse o pe u ba ion (RAL-021 esponse o Wolbachia
s a us), we obse ed ha one DGRP line (RAL-021) was able o p omo e blood,
mdg1 and oo ac i i y in esponse o he mal challenges (inc ease in TE
exp ession wi h dec ease in empe a u e). This s udy p o ides e idence ha
some en i onmen al pe u ba ions, bu no all, may p omo e p oduc ion o
he i able gene ic a ia ion ha can inc ease popula ionsʼ abili y o adap o new
en i onmen s (e ol abili y).
In summa y, he cu en s udy shows ha he egula ion o TE ac i i y can be
e y complex, as geno ypic and en i onmen al ac o s g ea ly in luence he
ac i i y o e o ansposons. Addi ionally, speci ic ime-dependen esponses o
ex e nal s imuli we e also obse ed. We consequen ly p opose ha di e ences
in he D. melanogas e gene ic backg ound, ime o exposu e o s ess, age o
adul s and TE analyzed, could be a he basis o he p e ious inconsis encies
ound in TE esponses o hea shock16–18.
Acknowledgemen s
DGRP lines we e ob ained om Blooming on ly s ock cen e , and O egon-R
lies om Rui Ma inhoʼs lab. The au ho s hank o Alisson Gon ijo o p o iding
p ime s agains CG15279, and membe s o Alisson Gon ijo, Chis en Mi h and
Élio Sucena labo a o ies o sugges ions and impo an discussions. We would
also like o acknowledge Osca Ruiz o eading his chap e .
!
126!
Au ho con ibu ions
The au ho o his disse a ion ob ained da a o O egon-R, did s a is ical and
sequence analysis, and w o e his chap e . Ana Eugénio ob ained da a o
DGRP lines; Ma a Ma ial a and Pa ícia Beldade designed all expe imen s.
!
127!
Supplemen a y In o ma ion
Supplemen a y Figu e 3.1. Exp ession o con ol genes (Rpl32 and Tbp). Rpl32 and
Tbp exp ession (loga i hmic scale, log) in ou DGRP lines (RAL-021, RAL-237, RAL-321
and RAL-790) in ec ed (Wolb+) o no (Wolb-) wi h Wolbachia and exposed o di e en
empe a u es (21ºC, 25ºC and 29ºC). Each do in box plo s ep esen s a biological
eplica e (pool o eigh pai s o o a ies); iangles ep esen he mean.
89:!$;>
89:!;<"
89:!<;>
89:!"=$
!"#$
!%#&
!%#$
!&#&
!&#$
'()*! '()*+'()*!'()*+ '()*! '()*+
!"#$%,-./0-112(3,4)(56
89:!$;>
89:!;<"
89:!<;>
89:!"=$
!7#&
!7#$
!"#&
!"#$
!%#&
&'",-./0-112(3,4)(56
25ºC 29ºC
21ºC
'()*! '()*+'()*!'()*+ '()*! '()*+ '()*! '()*+ '()*!'()*+ '()*! '()*+ '()*! '()*+'()*!'()*+ '()*! '()*+
'()*! '()*+'()*!'()*+ '()*! '()*+ '()*! '()*+'()*!'()*+'()*! '()*+ '()*! '()*+'()*!'()*+'()*! '()*+ '()*!'()*+ '()*!'()*+ '()*! '()*+
!
128!
Supplemen a y Figu e 3.2. Con i ma ion o TE inse ions included in ma u e
mRNAs. Scheme o nes ed PCR o ampli y TE inse ions ha a e wi hin gene sequences
(A). P ime s used o he i s PCR eac ion (PCR.1) we e designed o ecognize gene
sequences a ound he inse ion (Supplemen a y Table 3.2); p oduc o i s PCR was
used as empla e o he second PCR eac ion (PCR.2) using p ime s designed o
ecognize nucleo ide sequences inside he TE (Supplemen a y Table 3.1). Aga ose gel
(1%) wi h amplicons ob ained om he i s (yellow) and second ( ed) PCR eac ions (B).
Blood and oo nega i e con ols (C -) we e pe o med using a PCR eac ion wi hou
gDNA bu wi h p ime s ampli ying espec i e TE sequences (Supplemen a y Table 3.1).
We used GeneRule ™ 1kb DNA ladde (The mo Fishe Scien i ic).
Ta ge gene
PCR.1
PCR.2
T ansposable elemen
blood C -
oo C -
CG31751
CG5440
CG31974
Sema-1a
CG5681
CG11319
Ac 42A
CG8920
CG10073
Dg
CG17883
Topo s
CG5210
CG9194
CG33966
CG4374
Ta 12
CG2022
ws
po
dnc
Fas2
A B
!
129!
Supplemen a y Figu e 3.3. E ec o empe a u e on TE exp ession a di e en ime
poin s. Rela i e exp ession (P a l me hod) o blood, He -A, hobo, Juan, mdg1, opus, oo
and Ti an elemen s, a day 1, 2, 3 and 4 a e empe a u e change. Blood, mdg1, opus,
oo and Ti an a e e o ansposons wi h LTRs (LTR elemen s), HeT-A and Juan a e
e o ansposons wi hou LTR sequences (non-LTR elemen s), hobo is a DNA elemen .
We assessed he e ec o empe a u e and ime (ca ego ical ixed a iables) using linea
models wi h dependen a iable ( ela i e TE exp ession) in loga i hmic scale. The e ec
o empe a u e a each ime poin was es ed using leas -squa es means (p<0.05).
S a is ical signi ican compa isons ob ained om his analysis a e highligh ed in he
boxplo s. We also assessed he e ec o ime (con inuous ixed a iable) in TE
exp ession (loga i hmic scale) by using linea models (TEs wi h exp ession ha is
a ec ed by he age o emales: Juan, es ima e=-0.185, p- alue=0.002; Ti an , es ima e=-
0.430, p- alue=0.0003). Each ci cle in he boxplo s ep esen s a pool o eigh pai s o
o a ies; iangles co espond o he mean, and black line o he median.
B0..9
=(%!>
A.B.
?<1/
@97#
.*<,
+..
%-+1/2
!"
!#
$
#
!"
!#
$
#
%&'()*+(,,-./'+(012-3('2.'4*05"'60.78
91:'# 91:'" 91:'5 91:';
91:'# 91:'" 91:'5 91:'; 91:'# 91:'" 91:'5 91:'; 91:'# 91:'" 91:'5 91:';
25ºC 29ºC
p<0.05
p<0.05
p<0.05
p<0.05
LTR elemen
LTR elemen LTR elemen LTR elemen LTR elemen
non-LTR elemen non-LTR elemen DNA elemen

!
130!
Supplemen a y Figu e 3.4. E ec o Wolbachia and empe a u e on TE exp ession
o RAL-021, RAL-237, RAL-321 and RAL-790. Rela i e exp ession o blood, mdg1 and
oo e o ansposons, in RAL-021, RAL-237, RAL-321 and RAL-790 o a ies. We es ed
he e ec o Wolbachia, empe a u e and he in e ac ion be ween bo h (ca ego ical ixed
a iables) in ela i e TE exp ession using linea models wi h logno mal dis ibu ion. Da a
o each TE and gene ic backg ound was analyzed sepa a ely; s a is ically signi ican
ac o s a e shown in boxplo s wi h espec i e p- alues. When he in e ac ion be ween
Wolbachia and empe a u e was s a is ically signi ican ( o he case o blood and mdg1
exp ession in RAL-321) we used leas -squa es means (lsmeans) o es o he e ec o
Wolbachia a each empe a u e. In ha case, colo ed lines ep esen s a is ically
signi ican e ec s o Wolbachia. Each ci cle in he boxplo s ep esen s a pool o eigh
o a y pai s; iangles co espond o he mean and black line o he median. Blood and
mdg1 a e exp essed in he ge mline and soma ic issue o o a ies whe eas oo is only
exp essed in he ge mline (see Supplemen a y Figu e 2.3 o mo e de ails).
!"##$ %$&' (##
)*+),-.(,//0#1)(,"2304,)3#)!"#$%)5"#&6
7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9
25ºC 29ºC
21ºC
ABC8;@=
ABC8@;'
ABC8=D<
8:
8;
<
;
8=
8>
8?
8:
8@
8;
8=
8>
8?
8:
8@
8;
7#"!8 7#"!9 7#"!8 7#"!9 7#"!8 7#"!9
ABC8<;'
8@
8;
8'
<
'
;
8;
<
;
:
<
;
:
>
)*+),-.(,//0#1)(,"2304,)3#)&'")5"#&6
Wolb p=8.8e-04
Temp p=5.9e-06
Wolb*Temp p=0.03
Wolb p=3.6e-05
Temp p=0.14
Wolb*Temp p=0.04
Wolb p=0.52
Temp p=2.9e-04
Wolb*Temp p=0.14
Wolb p=5.1e-04
Temp p=7.8e-06
Wolb*Temp p=0.21
Wolb p=0.38
Temp p=0.002
Wolb*Temp p=0.90
Wolb p=0.04
Temp p=7.97e-06
Wolb*Temp p=0.61
!
131!
Supplemen a y Figu e 3.5. E ec o Wolbachia and empe a u e on he exp ession
o CG15279. Rela i e exp ession o CG15279 in RAL-021, RAL-237, RAL-321 and RAL-
790 o a ies. We es ed he e ec o Wolbachia, empe a u e and he in e ac ion be ween
bo h ac o s (ca ego ical ixed a iables) in ela i e exp ession using linea models wi h
logno mal dis ibu ion. Da a o each gene ic backg ound was analyzed sepa a ely;
s a is ically signi ican ac o s a e shown in boxplo s wi h espec i e p- alues. Each ci cle
in he boxplo s ep esen s a pool o eigh o a y pai s; iangles co espond o he mean
and black line o he median.
!"
!#
!$
!%
!&
!'
>?@!'&"
>?@!&'<
>?@!"A=
!"#$%&'
()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+, ()*+! ()*+,
25ºC 29ºC
21ºC
-./01223)4501*6738157)5()*+%59*):;
>?@!='<
!&
!'
!<
=
<
'
-./01223)4501*6738157)5,-)59*):;
Wolb p=0.55
Temp p=0.08
Wolb*Temp p=0.01
Wolb p=0.04
Temp p=6.89e-05
Wolb*Temp p=0.30
Wolb p=0.04
Temp p=1.75e-04
Wolb*Temp p=0.15
Wolb p=5.51e-04
Temp p=0.01
Wolb*Temp p=0.11
!
132!
Supplemen a y Figu e 3.6. T ansc ip ion ac o exp ession. Exp ession o CREB,
do sal, E2F1, o k head, Pebbled, cap-n-colla , PAX-6, slow bo de cells and se en up
ansc ip ion ac o s in esponse o se e al en i onmen al pe u ba ions (cold, hea ,
cadmium, coppe , zinc, e hanol, ca eine, pa aqua , o e one and Sindbis i us) (A).
Exp ession o ansc ip ion ac o s in o a ies collec ed om ma ed emales (B).
Exp ession in esponse o en i onmen al s imuli was ob ained om modENCODE
ea men exp ession da a67; exp ession in o a ies was acqui ed om modENCODE
issue exp ession da a67. No exp ession, 0-3 eads; low exp ession, 4-10 eads;
mode a e exp ession, 11-25 eads; mode a ely high exp ession, 26-50 eads; high
exp ession, 51-100 eads.
!"#$% &'()*+% #,-.% /'(0%12*&% 3244+2&% 5*67875'++*(% 39:7;% )+'<%4'(&2(%
52++)%
)2=28%>6%
!"#$%&'($)*+,-.!,/.)$01+,223"/)(-.-
!'+&
?2*@
!*&AB>A
!'662(
CB85
#@1*8'+
!*//2B82
3*(*D>*@
"'@2@'82
EB8&4B)%=B(>)
!"#$% &'()*+% #,-.% /'(0%12*&% 3244+2&% 5*67875'++*(% 39:7;% )+'<%4'(&2(%
52++)%
)2=28%>6%
!"#$%&'($)*3224,)$01+,223"/)(-.-
F=*(GH%A*@2&
I'&2(*@2%2J6(2))B'8
I'&2(*@2+G%1BK1%2J6(2))B'8
?BK1%2J6(2))B'8
L'<%2J6(2))B'8
!
133!
Supplemen a y Table 3.1. P ime s used o qPCR
Ta ge
Fo wa d p ime sequence
Re e se p ime sequence
Re
blood
AACAATAGAAAGAAGCCACCGAAC
AGTCATGGACTATTGAGGGTGTT
G
68
He -A
CGCGCGGAACCCATCTTCAGA
CGCCGCAGTCGTTTGGTGAGT
69
hobo
CATTAAGTCGGAAGGCCAAA
CTTGCTCTTCCGCTATCCAC
Juan
GGGGCAAAATTCTCAATGAA
GCGGAATATATGTGGGTTGC
mdg1
GTCAGAAGGAGGCCATTCAGGAATTT
GTTGCTGGCGGTTTCTGTTATTGT
CAA
70
opus
CGAGGAGTGGGGAGAGATTG
TGCGAAAATCTGCCTGAACC
7
oo
CGTCTGCAATGTACTGGCTCT
CGGCACTCCACTAACTTCTCC
7
Ti an
AAATGTTTGCCCCATCTCAG
GGGTCAGTGTGGCGTTATTT
CC15279
TCGTCTATCCAAATTGGTCTTACTC
AATGGCCACGATCATCCA
Rpl32
ATGCTAAGCTGTCGCACAAATG
GTTCGATCCGTAACCGATGT
Tbp
GGCAAAGAGTGAGGACGACT
GAGCCGACCATGTTTTGAAT
44
Re , e e ence
Supplemen a y Table 3.2. P ime s used o con i m TE inse ions
Ta ge
Fo wa d p ime sequence
Re e se p ime sequence
Ac 42A
ATATAACGCCGCTCGTTCTC
ATCTCGCCTGAGCTCTACCA
CG10073
TTACGTTTGGACGTGGTTGA
CCATTCACACCACGTTTGTC
CG11319
TGACAACGATGAGGAACCAA
ATCGTCCCTGGTGTAGATCG
CG17883
CCCTCCACACCAGTTTTCTT
CAGTCACATGGTTCGATTGG
CG2022
TGACACGTTTTTCAGGGATG
GCGGGTAGAGATTATGCCTTA
CG31751
CCGTTAGGCTTTTCCGTTTT
TTTTGTTTGCGATCATCTCG
CG31974
AACAAGGCGCTAATTCGATG
AAGAACACGGGTGCAAAAAC
CG33966
ATGTCGATGTGGACAAAGCA
ATGAGACCAGCAGCTCGATT
CG4374
CGCCTTGATGTCGATCTTCT
TGCTTCACCAGTTCCAATGA
CG5210
AGAGGACTCGTCGTGGCTAA
GTGCTGCATGTTGGCTTAAA
CG5440
ACGGCACGATAAAGAAAAGG
ATGATCGTCGAGGTCCACTC
CG5681
TGAACGGATGTCCAAGTCAA
CCCCTGCACCATGATTAGTT
CG8920
TATTGGGCCTTTCGAATCTG
GGATTGGTTCTGCCATGTCT
CG9194
CGACCTCCAAACATTTTCGT
AGCAGATGGCAGCTACACAG
Dg
GTTTCTTCGGCACATCACCT
CAACTGCGGAATGAGACTGA
dnc
AAATGCAACACTTGCCTTCC
CACTTCCGATTTCGTTGGTT
Fas2
AGACGGTCGGTGAAACTTTG
GTTCATTCTTCCGCTTCTGC

!
141!
Chap e IV
TE esponse o s ess ul condi ions is dependen
on he ype o s ess and Wolbachia s a us: a
s udy o he e ec o empe a u e, cadmium and
oxida i e s ess
!
143!
4.1 Abs ac
En i onmen al pe u ba ions educing i ness – s ess ul condi ions1 – a e
uni e sal o all na u al habi a s. Upon exposu e o h ea s, popula ions can
bene i om esponses ha inc ease gene ic a ia ion2,3, including hose d i en
by ansposi ion4,5. Despi e g owing e idence ha sugges ansposable
elemen s (TEs) as impo an media o s o adap i e gene ic a ia ion6–8, he
en i onmen al ac o s ha in luence mobiliza ion emain poo ly unde s ood9.
He e, we aimed a s udying he impac o di e en en i onmen al cues, including
bio ic and abio ic ac o s, on ansposi ion in he ge mline. We exposed
D osophila melanogas e adul emales o s ess ul condi ions ha included
se e al empe a u es, le els o oxida i e s ess and doses o hea y me al. We
hen assessed exp ession o i e TEs (blood, Juan, pogo, oo and Ti an ) in
o a ies. Gi en he abili y o Wolbachia o a ec many aspec s o hos biology,
and o sp ead in na u al popula ions by ma e nal ansmission10–12, we u he
assessed he impac o his endosymbion in a ec ing TE esponse o he
a o emen ioned abio ic ac o s. We obse ed ha TE ansc ip ion le els
depended on he en i onmen , Wolbachia s a us, and TE analysed. Al hough TE
esponses o oxida i e s ess we e no obus and had signi ican di e ences
be ween independen expe imen s, we obse ed obus esponses o blood and
Ti an o changes in empe a u e, and o Juan and Ti an o cadmium. We
p opose ha he main ep esso o TE ac i i y in he ge mline ( he piRNA
pa hway) is no a ec ed by en i onmen al pe u ba ions and con ols he ac i i y
o mos mobile elemen s, whe eas ew ansposons would speci ically espond
o pe u ba ions (namely by e ec s in he ac i i y o ansc ip ion ac o s). This
en i onmen -speci ic TE ac i i y in o a ies ha esul ed in new ansposon
inse ions ansmi ed o he nex gene a ion may p oduce gene ic a ia ion ha
is adap i e o he inducing en i onmen .
4.2 In oduc ion
Wild popula ions a e o en exposed o en i onmen al pe u ba ions ha can
shape e olu ion by ac ing as agen s o na u al selec ion, in ol ed in he so ing
!
144!
o gene ic a ia ion ha esul s in ans-gene a ional changes in allelic
equencies, and as ac o s in luencing he p oduc ion o no el gene ic a ia ion,
h ough e ec s on ecombina ion13,14 o mu a ion a es2,3. Despi e g owing
e idence ha ansposable elemen s (TEs) cons i u e an impo an sou ce o
adap i e gene ic a ia ion6–8, li le is known abou he en i onmen al ac o s ha
in luence TE mobiliza ion9.
Mos o he in o ma ion ega ding he egula ion o mobile elemen s in me azoan
species desc ibes a conse ed pa hway ha con ols TE ac i i y in he
ge mline15. Those o ganisms speci ically ep ess TEs by p oducing small RNAs
homologous o TE sequences and ha in e ac wi h PIWI amily p o eins
(piRNAs)15. In e es ingly, ecen s udies e ealed commonali ies be ween he
mechanisms ha con ol TE ac i i y and he pa hways igge ed in sub-op imal
condi ions16. Those included he induc ion o hea shock esponses and he
piRNA pa hway ha ails o con ain TE ac i i y in he absence o Hsp90 (a hea
shock p o ein in ol ed in p o ein olding)17–19. As many en i onmen al changes
esul in mis olded p o eins20, i is concei able ha , in s ess ul condi ions, he
ec ui men o chape ones such as Hsp90 o mis- olding co ec ion, would esul
in elaxed piRNA su eillance and in inc eased ansposon ac i i y. I his
happens, we should see de egula ion o di e en TEs, including
e o ansposons and DNA elemen s21. In his wo k, we aimed a s udying he
impac o se e al en i onmen al cues, including bio ic and abio ic ac o s, on he
ac i i y o e o ansposons and DNA elemen s in he ge mline o D osophila
melanogas e . We also assessed whe he ou challenges could a ec he piRNA
pa hway.
D. melanogas e is able o colonize di e se habi a s22 and ha e adap ed o
se e al bio ic and abio ic s esses known o igge hea shock esponses and o
p oduce excessi e amoun o eac i e oxygen species (ROS)23–25. These
en i onmen al condi ions include in ec ion10,26,27, empe a u e pe u ba ion 28–30
and exposu e o hea y me als31. We he e o e decided o e alua e TE
exp ession a e exposu e o di e en ypes o en i onmen al pe u ba ion ha
included empe a u e changes as well as exposu e o an oxidan agen and an
!
145!
hea y me al. Also, gi en ha Wolbachia – a bac e ial symbion common in
insec s – is able o in luence many aspec s o hos biology, and o sp ead in
na u al popula ions by ma e nal ansmission10–12, we assessed he impac o he
p esence o his endosymbion on TE esponse o he abio ic ac o s. Those
e ec s we e es ed in ui ly o a ies, because ansposi ion occu ing in he
ge mline can add o he i able gene ic a ia ion in a popula ion.
We obse ed ha ansposon ac i i y is en i onmen -speci ic, and sugges ha
empe a u e, and inges ion o cadmium, may in luence abili y o p oduce
he i able gene ic a ia ion by changes in TE exp ession. In addi ion, and
con a y o ou expec a ions, piRNA pa hway seems o be obus o
en i onmen al change, and keep con olling he bulk o mobile elemen s, while
only ew speci ic TEs espond o pe u ba ions. We discuss ha ansc ip ion
ac o s ela ed wi h pa icula pa hways igge ed by en i onmen al change migh
di ec ly a ec TE ac i i y, and ha Wolbachia may play an impo an ole in
a ec ing ansposon sensi i i y o ac i a ion.
4.3 Ma e ials and Me hods
D osophila melanogas e s ocks and main enance
All expe imen s we e done in a gene ic backg ound esul an om a c oss
be ween Ha wich emales (wi h and wi hou Wolbachia) and w[1118] males ( ee
o Wolbachia). w[1118] and Ha wich s ocks we e p o ided by Luís Teixei a.
Ha wich lies we e na u ally in ec ed wi h wMel Wolbachia s ain. To ob ain a
Ha wich line ee o in ec ion, we ea ed lies wi h e acycline (0.05mg/ml added
o hei ood), o wo gene a ions. Gu lo a con en was homogenized be ween
e acycline ea ed and non- ea ed lies, as p e iously desc ibed32. We c ossed
SpnE616/TM3 and SpnE100.37/TM3 lies (p o ided by Ví o Ba bosa) o ob ain a
posi i e con ol (SpnE616/ SpnE100.37) and nega i e con ols (SpnE616/TM3 o
SpnE100.37/TM3) o Bicaudal-D agg ega e o ma ion in o a ies. We kep lies in
s anda d co nmeal ood33 a 25°C, 60% humidi y and 12h-12h day-nigh cycles,
unless o he wise men ioned.

!
146!
Tempe a u e, cadmium and oxida i e s ess ea men
Fo empe a u e, cadmium and oxida i e s ess ea men s we collec ed eigh -
hou -old lies (wi h o wi hou Wolbachia) and kep i e emales wi h wo males
pe ial, a 25ºC. A e h ee days, males we e emo ed, emales ans e ed o
new ubes, and placed in all di e en expe imen al condi ions. Females we e
hen kep a each expe imen al condi ion o ou days (Supplemen a y Figu e
4.1). Fo empe a u e ea men s, emales we e placed ei he a 21ºC, 25ºC o
29ºC. Fo exposu e o hea y me als, lies we e ed wi h one o h ee di e en
doses o cadmium (0mM, 0.05mM, o 0.1mM), and o exposu e o oxida i e
s ess-inducing condi ions, emales we e ed wi h di e en doses o pa aqua
(0mM, 1.25mM, o 2.5mM), a commonly used d ug o igge he p oduc ion o
ROS34. Cadmium and pa aqua solu ions (Sigma) we e di ec ly mixed in he ly
ood. To con i m ha emales ed, we included 2% o blue ege able colo ing in
he ood and checked o he p esence o blue colo in he ly abdomen and gu s.
We assessed ecundi y (numbe o eggs laid pe emale) a day ou a e
en i onmen al pe u ba ion o es whe he changing empe a u e, exposing lies
o hea y me al doses (cadmium) o o an oxidan agen (pa aqua ) a e s ess ul
o D. melanogas e emales (wi h o wi hou Wolbachia) (Supplemen a y
Figu e 4.1A). Also, a e ou days o exposu e o each expe imen al condi ion,
o a ies we e dissec ed o RNA ex ac ion o assessmen o TE exp ession (see
below and Supplemen a y Figu e 4.1B). To con i m e ec o empe a u e on
TE copy numbe in he nex gene a ion, we collec ed emb yos laid on day ou
a e empe a u e change, o genomic DNA (gDNA) ex ac ion (see below and
Supplemen a y Figu e 4.1C).
RNA ex ac ion and cDNA syn hesis
O a ies we e dissec ed in esh cold PBS 1x, and di ec ly s o ed a 4ºC in 400µl
T izol (Ambicon) un il homogeniza ion using pes els ha was done in he same
day. Be o e homogeniza ion we polled 8-10 pai s o o a ies pe sample, and
used 6-8 eplica es pe condi ion, pe expe imen . Homogenized samples we e
s o ed a -80ºC un il RNA ex ac ion. Fo RNA ex ac ion we used Di ec -zol RNA
!
147!
Minip ep ki (Zymo Resea ch), and o al RNA was elu ed in 25µl o RNase- ee
wa e (Sigma). RNA pu i y and concen a ion o each sample, was measu ed
wi h Nanod op ND-1000 Spec opho ome e ; A260/A280 abso bance was
a ound wo o all samples, and concen a ions anged om 200 o 1000ng/µl.
All RNA samples o each expe imen we e p ocessed on he same day. We
cleaned 1µg o RNA om gDNA con amina ion, using DNAse (P omega)
ea men , ollowing manu ac u e ʼs ins uc ion. Fo cDNA syn hesis, we ollowed
Re e se T ansc ip ion Sys em (P omega) ins uc ions, using 0.02µM o Oligo dT
p ime s. Reac ion was incuba ed a 42°C o 60min, and hea ed a 95°C o
5min. cDNA was hen dilu ed in RNAse- ee wa e (1:10) and s o ed a -20°C
un il quan i a i e eal- ime PCR (qPCR).
gDNA ex ac ion
gDNA om emb yos was ex ac ed using DNeasy Blood & Tissue Ki (Qiagen).
Fo each sample, we pooled all eggs laid by a g oup o 20 emales
(Supplemen a y Figu e 4.1), six eplica es pe condi ion. Tissues we e
homogenized in 180µl o bu e ATL, using pes els. We used 200µl o bu e AE
o elu e gDNA and ea ed all samples wi h 0.2mg/ml o RNAse (In i ogen) o
20 min a 55ºC. gDNA concen a ion was measu ed in Nanod op ND-1000
Spec opho ome e , and samples s o ed a -20°C, un il qPCR.
qPCR
We measu ed le els o TE exp ession and TE copy numbe by qPCR: CFX384
(BioRad) o ABI Quan S udio-384 (Applied Biosys ems) he mal cycle s. Fo
each eac ion we used 5µl o iQ™ SYBR® G een supe mix (BioRad), 0.4µM
p ime s, 4µl o dilu ed cDNA (1:10) o gDNA (4ng/µl). We assessed he
exp ession o i e TEs (blood, Juan, pogo, oo and Ti an ) and h ee con ol
genes (Rpl32, Tbp and asa). P ime sequences a e desc ibed in
Supplemen a y Table 4.1. We an wo echnical eplica es pe sample, pe
pla e. The he mal cycling p o ocol was he same o all eac ions: 2 min a 50°C;
10 min a 95°C; 40 cycles o 95°C o 30s, 60°C o 1 min and 72°C o 30s.
!
148!
In each pla e we an eac ions on se ial dilu ions o known a ge amplicon
concen a ions o build s anda d cu es. To ob ain s anda ds, we ampli ied
a ge egions o each p ime pai (Supplemen a y Table 4.1) by PCR using
0.25U o GoTaq (P omega), 1x eac ion bu e , 1.5mM MgCl2, 0.4µM p ime s
and 1µl o cDNA (1:10). PCR p oduc s we e cleaned using NucleoSpin® Gel and
PCR Clean-up (Mache ey-Nagel), and elu ed in 15µl o bu e NE. We measu ed
concen a ions in Nanod op ND-1000 Spec opho ome e , and did 13 se ial
dilu ions (1:5) om 0.125ng/µl un il 6.4e-10ng/µl. Fo each p ime pai , we
es ima ed he minimum amoun o nucleic acids de ec able by qPCR by
ex ending dilu ions o ou s anda ds un il he ela ionship be ween he known
amplicon concen a ions and he quan i ica ion cycles (Cqs) eached a pla eau.
We used de aul h eshold se ings o he qPCR machines o ob ain he Cq o
each eac ion. Cq alues o all a ge s and expe imen al samples we e wi hin
he linea ampli ica ion. We ans o med Cq alues in o absolu e a ge amplicon
concen a ions using he espec i e s anda d cu es. This me hodology allowed
us o con ol o pla e e ec s and p ime e iciency. qPCR mel ing cu es we e
analyzed o con i m speci ici y o ampli ied p oduc s, and nega i e con ols ne e
showed de ec able ampli ica ion.
Fo he analysis o exp ession da a, we used No mFinde 35 and geNo m36
algo i hms o check exp ession s abili y o wo e e ence genes commonly used
o calcula e ela i e gene exp ession a di e en en i onmen al condi ions, Rpl32
and Tbp37,38. We also es ed asa exp ession s abili y, which is a ge mline-
speci ic gene39. Fo bo h algo i hms used, s abili y alues ob ained o asa
(No mFinde , 0.39; geNo m, 0.84) we e be e han hose o Rpl32
(No mFinde , 0.57; geNo m, 0.95) and Tbp (No mFinde , 0.69; geNo m, 1.09).
Howe e , asa exp ession in he second expe imen o Pa aqua ea men s is
mo e a iable han ha obse ed o Rpl32 and Tbp (Supplemen a y Figu e
4.2). We he e o e decided o use Tbp (No mFinde , 0.46; geNo m, 0.40) as
e e ence gene o ha da ase ins ead o asa (No mFinde , 0.81; geNo m,
0.62) o Rpl32 (No mFinde , 0.63; geNo m, 0.47). The median o ela i e
exp ession in con ol samples (25ºC, 0mM o cadmium, o 0mM o pa aqua ),
!
149!
ee o Wolbachia, was used o calib a e each ea men ( empe a u e, cadmium
and pa aqua ) pe expe imen and ansposon (P a l me hod)40.
To assess TE copy numbe in he genomic DNA o emb yos, we used Rpl32 (a
known single copy gene) as e e ence.
An ibody s aining o o a ies
O a ies we e dissec ed in esh cold PBS 1x, and ixed in 200µl o 3.2%
o maldehyde (Sigma), 0.5% NP40 (Sigma) dilu ed in PBS 1x, and 600µl o
hep ane (Sigma) o 20 min wi h gen le (400 pm) shaking. To a oid unspeci ic
an ibody (Ab) binding and pe meabilize issues, o a ies we e kep o 1 hou a
400 pm, in 1% BSA (Sigma), 0.01% T i onX (Sigma) dilu ed in PBS 1x. We hen
added a cock ail o an i-Bicaudal-D 4C2 (1:10) and an i-Bicaudal-D 1B11 (1:10)
(De elopmen al S udies Hyb idoma Bank) solu ions and incuba ed o e nigh a
4ºC. Alexa488 an i-mouse was used as seconda y Ab, and dilu ed (1:500) in
PBS 1x, con aining 0.2% Tween (Sigma) (PBT). O a ies we e kep in ha
solu ion o 2 hou s, in he da k a 400 pm. Nuclei we e hen s ained wi h To o-3
(1:500) o 7 min in he da k. We pe o med washes wi h PBT be ween all s eps
men ioned abo e; las washes o he p o ocol we e done wi h PBS. All p o ocol
was pe o med a oom empe a u e, unless s a ed o he wise. O a ies we e
moun ed in glyce ol, and images acqui ed in Leica SP5 in e ed con ocal.
S a is ical analyses
All s a is ical analyses we e done in R ( e sion 3.2.1).
To es he e ec o abio ic ac o s ( empe a u e, cadmium and pa aqua ) and
Wolbachia s a us (ca ego ical ixed a iables) on ecundi y, we used gene al
linea models wi h nega i e binomial dis ibu ion (glm.nb). The in e ac ion
be ween ea men and p esence o Wolbachia was included in he analysis;
each abio ic ac o was analyzed sepa a ely. Gi en ha Wolbachia s a us was
ne e a ac o con ibu ing o explain he da a, we hen assessed di e ences
be ween ea men s using non-pa ame ic Wilcoxon ank es . Fo empe a u e
ea men s (21ºC, 25ºC and 29ºC) we pe o med all pai wise compa isons