Ma ia Adelina Je ónimo
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
The o igin and de elopmen o
no el y: eyespo s and immuni y
Resea ch wo k coo dina ed by:
Decla ação/Decla a ion
Es a disse ação é o esul ado do meu p óp io abalho
desen ol ido en e Maio de 2011 e Maio de 2015 no labo a ó io da
D a. Pa ícia Beldade, Ins i u o Gulbenkian de Ciência em Oei as,
Po ugal, no âmbi o do P og ama da FCT de Dou o amen o (2010).
Com base no abalho desen ol ido, a minha o ien ado a e eu
espe amos subme e dois a igos cien í icos pa a publicação.
This disse a ion is he esul o my own esea ch, ca ied ou
be ween May 2011 and May 2015 in he labo a o y o D . Pa ícia
Beldade, Ins i u o Gulbenkian de Ciência in Oei as, Po ugal, unde
he FCT Doc o al P og amme (2010). F om his wo k, my supe iso
and I expec o submi wo pape s o publica ion.
Apoio Financei o/Financial Suppo
Apoio inancei o da FCT e do FSE no âmbi o do Quad o
Comuni á io de Apoio, bolsa de dou o amen o #
SFRH/BD/73658/2010, dos p ojec os a ibuidos a P. Beldade
(PTDC/BIA-BEC/ 099808/2008 e PTDC/BIA-EVF/2170/2012) e da
Fundação Calous e Gulbenkian.
Financial suppo o his hesis was p o ided by he FCT and
FSE hou gh he Quad o Comuni á io de Apoio, doc o al ellowship #
SFRH/BD/73658/2010, esea ch g an s o P. Beldade (PTDC/BIA-
BEC/ 099808/2008 and PTDC/BIA-EVF/2170/2012) and Fundação
Calous e Gulbenkian.

I
Acknowledgemen s
I would like o hank he IGC o gi ing me he oppo uni y o do
my Ph.D..
To all elemen s o my hesis commi ee: in he e y beginning
Vic o Ba bosa, An ónio Jacin o, and mo e ecen ly Élio Sucena and
Jo ge Ca nei o.
To all he p esen and pas membe s o Va ia ion: De elopmen &
Selec ion Lab, E olu ion and De elopmen Lab and De elopmen ,
E olu ion and he En i onmen lab. Thank you o all you help,
pa ience and ui ul discussions ha con ibu ed a lo o his wo k.
Fo all my iends and amily ou side IGC ha encou aged me
h oughou his jou ney. Thank you o you pa ience and you
cu iosi y abou my wo k.
Finally, I would like o hank o Pa ícia Beldade, he supe iso
ha ga e me he oppo uni y, he guidance and he eedom o di e
in o he wonde ul wo d o e olu ion and de elopmen .
II
Summa y
The in ini ude o o ms ac oss li ing beings always ascina ed
scien is s and laymen alike. Pheno ypic a ia ion is he aw ma e ial
o na u al selec ion and is ans e sal in biological sys ems.
E olu iona y no el ies a e lineage speci ic ai s wi h adap i e alue.
How such ai s a ise in o ganisms is s ill an open ques ion. To
unde s and his p ocess we used he bu e ly eyespo s as a model,
which a e e olu iona y no el y. Ou es ing hypo hesis was ha
bu e lies had ewi ed ancien gene ic mechanisms used o he
wound- esponse o de elop eyespo s, co-op ion. The majo ac
suppo ing his hypo hesis is ha wounding an ea ly pupal wing can
de elop an eyespo -like pa e n a ound wound si e. This pinpoin s he
exis ence o gene ic and cellula commonali ies be ween eyespo
de elopmen and wound esponse.
Fi s , we cha ac e ized he gene ic wound esponse on he wings
o ou model sys em, Bicyclus anynana, compa ing he gene ic
p o iles o wounded and non-wounded wing pai s (Chap e 2).
Wounded wings we e en iched o immuni y- ela ed genes. These
genes a e exp essed in ci cula pa e ns a ound wound si es.
Addi ionally, we obse ed ain exp ession pa e ns a he p esump i e
eyespo cen e , especially o he an imic obial pep ide gene Glo e in
2. An ennapedia, a ansc ip ion ac o in ol ed in eyespo
de elopmen was also obse ed a wound si es. A p esump i e
eyespo cen e s and wound si es he e a e highe cell densi y due o
mig a ion o hemocy e-like cells.
Secondly, we explo ed he ole o immuni y in o he de elopmen
o wound-induced pa e ns (Chap e 3). We manipula ed he immune
sys em ac i a ion du ing pupal de elopmen and quan i ied he do sal
Summa y
_____________________________________________________________________________________________________________________________________________________________________________________________________________
III
adul wing pa e ns. Wounds wi h high immuni y ac i a ion de eloped
la ge wound-induced eyespo s. This enla gemen is media ed
h ough local immuni y. Wi h his expe imen s we no iced ha no only
do sal wound-induced pa e ns had changed, bu also he en al wing
pa e ns we e di e en .
Thi dly, we cha ac e ized he immuni y-media ed de elopmen al
plas ici y, one geno ype p oducing di e en pheno ypes acco ding o
he en i onmen al condi ions, on en al wing pa e ns (Chap e 4). As
in low de elopmen al empe a u e indi iduals, high immuni y
ac i a ion induced long de elopmen al ime, smalle eyespo s, lowe
wing colo con as , and lowe ecdyson ho mone le els. This e ec
was independen o he o gan whe e he immuni y ea men was
applied. To ou knowledge, his is he i s ime ha immuni y is
media ing de elopmen al plas ici y in bu e ly wing pa e ns.
Las ly, we explo ed he immuni y e ec s on wing size and shape
(Chap e 5). Con a y o low de elopmen al empe a u e, wings o high
immuni y indi iduals we e smalle . Howe e , his e ec only occu ed
when immune ea men was applied in one o he de eloping wings,
which educed he p oli e a ion a e. High immuni y and low
de elopmen al empe a u e induce simila shape changes, pulling he
wing shape in he same di ec ion. This indica es ha di e en s esses
ac i a e a common de elopmen al p og am de e mining wing shape.
In summa y, his wo k ein o ces he hypo hesis o wound gene ic
ci cui y co-op ion o he e olu iona y o igin o bu e ly eyespo s.
Mo eo e , i unco e ed he c ucial ole o immuni y in he bu e ly
wing pa e ns de elopmen . I also open up new pe spec i es o
unde s and he o igin and di e si ica ion o no el ies, de elopmen al
plas ici y and he ecological meaning o pheno ypic a ia ion on
bu e ly wing pa e ns.

Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
1
Chap e 1 – Gene al In oduc ion
The endless di e si y o mo phological o ms ac oss li ing beings,
he ecological niches colonized and he good ma ch be ween he wo
has always ascina ed scien is s and laymen alike. Pheno ypic
a ia ion is a uni e sal p ope y o biological sys ems and he aw
ma e ial o na u al selec ion. The molecula and gene ic mechanisms
and he en i onmen al ac o s unde lying he p oduc ion o pheno ypic
a ia ion a e a key subjec in ecological e olu iona y de elopmen al
biology (eco-e o-de o). One o he mos s udied ai s in his ield is
body pigmen a ion. The e is g ea in a-speci ic a ia ion and in e -
species di e si y in body pigmen a ion, and his ai is ecologically
ele an and expe imen ally ac able. He e, we ocused on a model
o s udies o he o igin and modi ica ion o no el ai s (i.e. ai s o
which he e a e no clea homologs in o he lineages) and o
de elopmen ally plas ici y (i.e. he en i onmen al egula ion o
de elopmen ha can esul in he p oduc ion o dis inc pheno ypes
wi hou gene ic di e ences): wing pigmen a ion pa e ns in bu e ly
wings.
1.1. E olu iona y no el ies
Among he di e si y o mo phological ai s, no el ai s such as
lowe s, bi d ea he s, mammalian placen a, insec wings, bee le ho ns
and bu e ly eyespo s (e.g. Figu e 1.1) a e especially ascina ing and
hei o igin is especially challenging o explain. The e is no single
uni e sally accep ed de ini ion o wha cons i u es an e olu iona y
no el y and o how ha can be asce ained. Di e en au ho s conside
di e en le els o biological o ganiza ion and ocus on di e en c i e ia
(Ra osa 1991), including 1) lack o clea homology in he ances al
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
2
species (Mulle & Wagne 1991), 2) new ecological unc ion wi hin he
lineage (Massimo Pigliucci 2008), o 3) new ecological unc ion ha
pe mi new unc ions and open up new adap i e zones. He e, we
conside as no el ai s hose ha a e es ic ed o a lineage and ha e
a speci ic adap i e alue.
Figu e 1.1- Examples o e olu iona y no el ies. Flowe s in plan s, hai in mammals,
ea he s in bi ds and scale-based wing pa e ns in bu e lies.
1.1.1. No el ai s on bu e ly wing pa e ns
The beau y and as onishing di e si y o bu e ly wing pa e ns
ha e been inspi ing scien is s o cen u ies, and ha e been used o
add ess ques ions in disciplines as di e en as sys ema ics and
e olu ion, de elopmen and gene ics, biochemis y, and immuni y.
Du ing las decades se e al e o-de o (B ake ield e al. 1996; Saenko
e al. 2011; Hines e al. 2012) and eco-de o (P udic e al. 2011; P udic
e al. 2015) s udies ha e inc eased ou unde s anding abou
de elopmen , gene ics and ecological ele ance o bu e ly wing
pa e ns. Howe e , many ques ions emain. Fo example, he
e olu iona y o igin, he gene ic egula ion and he ole o he
en i onmen du ing de elopmen o he bu e ly wing pa e ns.
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
3
Bu e ly colo pa e ns a e o med by he wo-dimensional
a angemen o single colo scales. Those pa e ns a e e olu iona y
no el ies and ha e allowed wing su aces o pe o m new ecological
unc ions in he mo egula ion and isual communica ion (Beldade &
B ake ield 2002). The di e si y be ween species is d ama ic, bu he e
is also a ia ion e en wi hin a popula ion conside ing di e en wing
su aces, di e en sex and di e en seasonal o ms (Nijhou 1991).
Homology ela ionships be ween pa e n elemen s a e ep esen ed in
he "nymphalid g oundplan" (Schwanwi sch 1929; Nijhou 1991). In
his model, di e en wing pa e n elemen s, such as eyespo s,
che ons and bands, a e o ganized in pa allel se ies; indi idual
elemen s a e epea ed along he an e io -pos e io wing axis wi hin so-
called wing cells ha co espond o wing compa men s bo de ed by
eins. The ex en o which wing pa e n elemen s can e ol e in an
independen way will impac how easily hey can espond o di e en
selec ion p essu es and i will also impac he di e si ica ion o
bu e ly wing pa e ns (Nijhou 1991; Nijhou 2001).
Se e al s udies ocusing on he gene ic mechanisms unde lying
wing pa e ns de elopmen and e olu ion ha e p o ided impo an
insigh in o homology o di e en wing pa e n elemen s (Mon ei o e al.
2006; Shi ai e al. 2012; Saenko e al. 2011). Among he di e en
ypes o wing pa e n elemen s, eyespo s ha a e composed o ings
o con as ing colo s a e a guably hose mos s udied a se e al le els,
such as molecula , gene ic, de elopmen al, ecological and
e olu iona y.
1.1.2. De elopmen al plas ici y on bu e ly wing pa e ns
Pheno ypic plas ici y is he abili y o an o ganism o eac o
en i onmen al inpu wi h a change in pheno ype (Wes -Ebe ha d
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
4
2003). I is also de ined as he abili y o a single geno ype o p oduce
mo e han one al e na i e o m in di e en en i onmen al condi ions,
whe he he al e na i e pheno ypes a e con inuous o discon inuous
(S ea ns 1989). De elopmen al plas ici y is a pa icula case o
pheno ypic plas ici y whe e en i onmen al inpu du ing p e-adul
de elopmen induces al e na i e pheno ypes in he adul s age
(Beldade e al. 2011). Tempe a u e-dependen sex di e en ia ion in
u les (Pieau e al. 1999), p eda o -induced change in head
mo phology o Daphnia (Boe sma e al. 1998), and nu i ional
egula ion o ho ns size in sca ab bee le (Moczek & Emlen 2000) a e
examples o de elopmen al plas ici y.
De elopmen ally plas ic ai s ypically ha e a pa icula pe iod in
which o ganisms sensing a di e en en i onmen shi hei
de elopmen al p og ams, and en i onmen al changes ou side his
sensi i e pe iod do no induce pheno ypic changes (Beldade e al.
2011). The sensi i e pe iod p ecedes he de elopmen o he inal
adul pheno ype by a pe iod o ime ha may be as b ie as a ew days
o as long as se e al mon hs (Nijhou 1999). Fo example, he
sensi i e pe iod o he seasonal adul wing pa e n polyphenism in
bu e lies is du ing la e la al ins a s and ea ly pupal s age (Nijhou
1999; Kooi & B ake ield 1999).
This s udy ocuses on he eyespo s o Bicyclus anynana, which
a e no el and de elopmen ally plas ic ai s. B. anynana is an A ican
bu e ly li ing in opical en i onmen s wi h wo main seasons: he we
and he d y seasons which a e e y dis inc in e ms o empe a u e,
humidi y and ood a ailabili y (Windig e al. 1994). This bu e ly
exhibi s clea seasonal polyphenism in wing pa e n and o he ai s
(B ake ield & F ench 1999; B ake ield e al. 2009; Oos a e al. 2011).
We season la ae p oduce adul s wi h conspicuous en al wing
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
5
pa e ns wi h la ge ma ginal eyespo s, while d y season la ae
p oduce adul s wi h dull b own colo s and e y small eyespo s (Figu e
1.2A). These wo al e na i e wing pa e ns a e belie ed o co espond
o seasonally di e en s a egies o a oid p eda ion. While he
ma ginal la ge eyespo s o he we -season bu e lies a e hough o
a ac he p eda o ’s a en ion o he wing ma gin and away om he
ulne able body, he all-b own d y-season bu e lies a e c yp ic
agains a backg ound o d y lea es (Lyy inen e al. 2003; P udic e al.
2015; Beldade e al. 2011). Thus, his seasonal polyphenism p o ides
an adap i e esponse o he al e na ing seasonal en i onmen s
inc easing he indi idual’s i ness.
S udies ha e shown ha he empe a u e du ing la e la al
de elopmen , which p edic s he season o adul s, is an impo an
en i onmen al cue de e mining he de elopmen o he adul wing
pa e n pheno ypes h ough changes in ho monal dynamics (Kooi &
B ake ield 1999; Koch e al. 1996; Ma eus e al. 2014) (Figu e 2.2C
and D). The eyespo s on he en al su ace o B. anynana wings, he
one exposed o p eda o s when he bu e ly is es ing, a e he ones
mos he mally plas ic (B ake ield e al. 2009; Ma eus e al. 2014)
(Figu e 2.2A). The do sal eyespo s, which a e implica ed in ma e
choice, a e much less plas ic, a leas in ela ion o empe a u e
(B euke & B ake ield 2002; Wes e man e al. 2014; Ma eus e al.
2014) (Figu e 2.2B).
1.1.3. Eyespo de elopmen
Du ing he pas decades, a numbe o eco-e o-de o s udies
ocused on bu e ly eyespo s and p o ided insigh in o hei
e olu iona y o igin, ecology and de elopmen al plas ici y (e.g. Saenko
e al. 2011; P udic e al. 2015; Ma eus e al. 2014). The expe imen al

Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
6
ac abili y o de eloping eyespo s is an ad an age in e o-de o
s udies. Fo example, easy de ec ion and access o he posi ion o
eins and de eloping eyespo cen e s on he do sal o ewing o
de eloping pupae allows o e y p ecise su gical manipula ions unde
a simple s e eoscope.
Figu e 1.2- Seasonal polyphenism in B. anynana. In he lab, lowe de elopmen al
empe a u es (e.g. 19°C) lead o he p oduc ion o adul s wi h wing pa e ns
esembling hose o he d y season, while wa me de elopmen al empe a u es
(27°C) lead o he p oduc ion o adul wing pa e ns esembling hose o he we
season. Tempe a u e du ing de elopmen media es pheno ypic plas ici y o en al
wing pa e ns (A) and has a much smalle e ec on do sal pa e ns (B). Ven al
pa e ns a e mo e plas ic han do sal pa e ns in esponse o en i onmen al
empe a u es (Ma eus e al. 2014) D) Di e en 20E dynamics in he hemolymph o
pupae om he di e en seasonal o ms (c . Oos a e al. 2011).
Wing de elopmen s a s du ing he i s la al ins a when he
wings g ea ly inc ease in size and he ena ion sys em inc eases in
complexi y. Du ing his s age, he eyespo de elopmen s a s wi h
de e mina ion o eyespo cen e s o oci (Figu e 1.3) (B ake ield e al.
1996; Saenko e al. 2011). A se ies o genes ha e been associa ed
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
7
wi h his phase o eyespo de elopmen , including An ennapedia
(An p), Dis al-less (Dll), No ch (N), eng ailed (en), hedgehog (hh),
cubi us in e up us (ci), pa ched (pa ch) and spal (sal) (Saenko e al.
2011; Mon ei o e al. 2006; Da id N Keys e al. 1999). La e , in ea ly
pupal wings he e is he posi ioning o he eyespo ings a ound he
ocus (B ake ield e al. 1996; Wi kopp & Beldade 2009). Su gical
manipula ions o pupal wings ha e led o he sugges ion ha ocal
cells p oduce (o deg ade) a di usible mo phogen-like signal ha
de e mines he a e o su ounding cells (Nijhou 1980; Mon ei o e al.
1994; F ench & B ake ield 1995). Depending on he concen a ion o
he mo phogen-like signal hey a e exposed o, su ounding cells
exp ess di e en (combina ions o ) ansc ip ion ac o s (du ing ea ly
pupal s age), which co espond o he p oduc ion o di e en colo
pigmen s (in la e pupae). In B. anynana, Dll and Sal map he black
ing and en he golden ing (B une i e al. 2001; Beldade e al. 2002).
The iden i y o he mo phogen-like signal igge ing he exp ession o
hese ansc ip ion ac o s is s ill unknown. The syn hesis o pigmen s
and hei deposi ion is he las s age o eyespo de elopmen and
akes place only du ing he las days o he pupal li e (Figu e 1.3).
Figu e 1.3- B. anynana eyespo de elopmen du ing la al and pupal s ages. Focal
de e mina ion occu s du ing he las la al s age (B ake ield e al. 1996). In ea ly
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
8
pupae, he eyespo cen e p esumably p oduces (o deg ades) a mo phogen-like
signal inducing he ansc ip ion o “posi ioning” genes in ings a ound he ocus
(B une i e al. 2001). In he las days o he pupal s age, pigmen syn hesis akes
place (Iwa a & O aki 2016). The sensi i e pe iod o induce an ec opic eyespo in his
bu e ly is om 6 o 18 hou s pos -pupa ion o 27°C ea ing empe a u e indi iduals
(B ake ield & F ench 1995).
1.1.4. The e olu iona y o igin o bu e ly eyespo s
The o igin o e olu iona y no el ies is one a key ques ion in e o-
de o (Mulle & Wagne 1991; Mülle 2007; Moczek 2008). The ac o s
p omo ing he di e si ica ion o no el ai s ha e been s udied o a
long ime (S ebbins 1970), bu he gene ic and de elopmen al
mechanisms unde lying hei o igin ha e become he ocus o
esea ch only in he las decades (Wagne & Lynch 2010; Saenko e
al. 2008). “Co-op ion” has been p oposed as he main mechanisms
explaining he o igin o e olu iona y no el ies (T ue & Ca oll 2002).
This mechanism e e s o an e olu iona y p ocess by which e olu ion
ecycles ances al gene ic ci cui ies sha ed ac oss lineages o
p oduce no el ai s (T ue & Ca oll 2002; Gan o nina & Sánchez
1999). The e a e se e al s udies showing ha he o igin o no el ai s
la gely elies on exis ing genes and gene ne wo ks ha a e e-
deployed and subsequen ly modi ied o gi e ise o no el
mo phological s uc u es (Moczek & Nagy 2005; Saenko e al. 2011;
P um 2005). Gene ecycling highligh s he pleio opy o gene ic
ne wo ks, which a e in ol ed in a ious p ocesses du ing
de elopmen and in di e en body o gans. Recen whole genomic
s udies ha e aised he possibili y ha lineage-speci ic genes can also
play a ole (Khal u in e al. 2009; Donoghue e al. 2011; Zhou e al.
2015).
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
9
Scien is s ha e highligh ed he co-op ion o di e en gene ic
ci cui ies o he e olu iona y o igin o bu e ly eyespo s, such as leg-
de elopmen , wing ma gin de e mina ion, an e io -pos e io wing
compa men aliza ion, emb yonic de elopmen and wound esponse
(Held 2013; Mon ei o 2015; Saenko e al. 2011). In his hesis I ha e
ocused on he commonali ies be ween eyespo de elopmen and
esponse o epide mal wounds. The ac ha some bu e lies de elop
o ganized colo pa e ns a ound wound si es esembling he adul
eyespo pa e ns (B ake ield & F ench 1995; Mon ei o e al. 2006;
O aki 2011) e eals commonali ies be ween eyespo de elopmen and
he esponse o wounding. While i is common o ha e da k
pigmen a ion spo s a wounded si es (Tang 2009; Le esque e al.
2012; Binggeli e al. 2014), he p oduc ion o o ganized colo pa e ns
cen e ed on wound si es is less common, bu no es ic ed o
bu e lies (Ohno & O aki 2012). In B. anynana, ec opic eyespo s can
be induced by wound on pupal wings om six o 18 hou s pos
pupa ion, wi h 12 hou s pos -pupa ion co esponding o he highes
p obabili y o induc ion o bu e lies ea ed a 27°C(B ake ield &
F ench 1995).
These obse a ion o wound-induc ion o ec opic eyespo s led o
he hypo hesis o co-op ion o he wound esponse gene ic ci cui y o
he e olu iona y o igin o eyespo s (Mon ei o e al. 2006). The genes
in ol ed in he di e en ia ion o ec opic eyespo s ha e no been
ex ensi ely in es iga ed and, hus, i is unclea whe he na i e
eyespo s and ec opic eyespo s a e using he same gene ic ci cui y.
Mon ei o and co-wo ke s in es iga ed he gene exp ession o some
known eyespo - ela ed genes and saw ha en, Sal and Dll we e
exp essed in cells a ound wound si es se e al hou s a e wounding
(Mon ei o e al. 2006). In his case, eyespo s a e like “pain ed sca s”
on wings. This wo k p esen ed he e s a ed o o in es iga e he
Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
16
change he cell a e o su ounding cells, ac ing much like an eyespo
o ganizing cen e . P e ious s udies e ealed ha se e e wounds
could change he likelihood o p oduc ion o an ec opic eyespo
a ound a wound si e, bu no i s size (B ake ield & F ench 1995). Ou
esul s sugges a new ole o immune- ela ed pa hways in o he
eyespo size egula ion, which does no in e e e in he likelihood o
p oduc ion o an ec opic eyespo .
In Chap e 4, we desc ibe he e ec o an immune challenge in
pupae on B. anynana adul wing pa e ns. This e ec phenocopies he
well-known e ec o coole de elopmen al empe a u es on wing
pa e n, as well as on ecdysone dynamics. We explo ed he e ec o
immune ac i a ion wi h di e en bac e ial ypes and upon wounding o
di e en body pa s on de elopmen al ime, ecdysone dynamics, and
adul wing pa e ns. These expe imen s open new pe spec i es o a
ole o immune challenge as a media o o plas ici y in wing colo
pa e ns.
In Chap e 5 we explo ed he e ec s o immune challenge on
o gan size and shape and discuss i s impac on in e o gan
communica ion and eadjus men o g ow h du ing de elopmen .
Typically, he in es men in immune de ense is associa ed wi h
deple ion o ene ge ic esou ces and, consequen ly, wi h ade-o s
wi h a ious li e his o y ai s. We compa ed he e ec s o immune
challenge om wound on di e en body pa s on size and shape o
adul wings. We u he in es iga ed he mechanisms behind he
educ ion o wing size by analyzing he e ec o he immune challenge
on he numbe o mi o ic cells. We also compa ed di e ences in wing
size and shape esul ing om immune challenge wi h di e ences in
wing size and shape be ween indi iduals ea ed a di e en
empe a u es. These expe imen s allowed us o unde s and how he

Chap e 1
_____________________________________________________________________________________________________________________________________________________________________________________________________________
17
wo ac o s a ec ing wing de elopmen ( empe a u e and le el o
immune challenge) a ec di e en axes o wing g ow h.
Finally, in Chap e 6, we discuss he main con ibu ions o his
wo k and p esen possible u u e di ec ions o esea ch ha could
u he help us unde s anding he gene ic basis o e olu iona y
no el ies and how immune challenge can ac as a cue inducing
pheno ypic change in bu e ly wing pa e ns.
1.4. Acknowledgemen s
We would like o hank he ‘e o-de o’ communi y a Ins i u o
Gulbenkian de Ciência, including he labs headed by Pa ícia Beldade,
Ch is en Mi h, and Élio Sucena, as well as Magda A ilano o ui ul
discussions h oughou my PhD.
18
Chap e 2 – Molecula and cellula
commonali ies be ween wound
esponse and pigmen a ion
pa e ning
2.1. Summa y
The esponse o epide mal wounding is an e olu iona ily
conse ed p ocess and is ypically igh ly connec ed o pigmen a ion.
An in e es ing example occu s in some Lepidop e ans ha can
de elop o ganized pigmen a ion pa e ns a ound wound si es ha
esemble na i e pa e n elemen s called eyespo s. This sugges ed he
exis ence o molecula and cellula commonali ies du ing de elopmen
o bo h wound-induced and na i e eyespo s. To e eal hese
commonali ies, we s udied he wound esponseon pupal de eloping
wings o he bu e ly Bicyclus anynana. Mic oa ay analysis
compa ing gene exp ession in wounded s. non-wounded pai s o
wings e eled en ichmen o an imic obial pep ides (AMPs) and
melanin pa hway enzymes, bo h ela ed wi h inna e immuni y in
insec s. We alida ed and ex ended he mic oa ay analysis h ough
quan i a i e RT PCR (qPCR), and in si u hyb idiza ion (ISH). We also
ound ha he cell o ganiza ion o he epide mis was simila a wound
si es and a he cen e o p esump i e eyespo s: mo e cells be ween
do sal and en al epi helia, exp essing mo e Ac in, An ennapedia
(An p) and Glo e in 2 (Glo 2) (an AMP). Highe cell densi y is no a
consequence o highe cell p oli e a ion a ound wound si es
sugges ing cell mig a ion owa ds wound si es. The cell shape and
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
19
gene exp ession o hose cells sugges s ha hey a e hemocy es. Ou
expe imen s b ough new insigh on o a speci ic wound esponse
(pos -g ow h and p e-adul s age) and also ein o ce he hypo hesis o
he e olu iona y o igin o eyespo s h ough co-op ion o he ancien
wound esponse gene ic mechanisms.
2.2. In oduc ion
Mul icellula o ganisms can epai hei epi helia a e an inju y.
This p ocess is essen ial o main ain homeos asis (Da is & Engs öm
2012) and es o e he in eg i y o he ou e ba ie . Wound esponse
in eg a es e olu iona ily conse ed cellula and gene ic mechanisms
o issue epai and immuni y (Moussian & U 2005; Wood e al. 2002).
Du ing he las decades, insec s and in pa icula D osophila
melanogas e , ha e become a key model sys em o explo e he
mechanisms o wound healing, epi helial epai and egene a ion
(Wood e al. 2002; Bosch e al. 2005; Jiang e al. 2011; Moussian &
U 2005). The e a e ex ensi e simila i ies be ween insec species
(K au z e al. 2014; Ele he ianos & Re enis 2011; Kanos e al. 2004)
and e en be ween insec s and mammals wound esponse (Moussian
& U 2005).
A e wounding, insec s o m a melanin clo o seal wounds and
ap mic oo ganisms, blocking hei en y in o he insec open body
ca i y (La ine & S and 2002; Ele he ianos & Re enis 2011). The
mic oo ganisms a e subsequen ly agglu ina ed, immobilized, and
killed by a ious lec ins and an imic obial pep ides (AMPs)
(Ele he ianos & Re enis 2011). To es o e epi helium in eg i y,
epide mal cells sp ead along and h ough he clo un il hey mee and
ees ablish a con inuous epi helial shee (Galko & K asnow 2004).
Du ing his p ocess, he e is eo ganiza ion o epi helial s uc u e and
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
20
cell’s cy oskele on adjus men s (Moussian & U 2005; Jacin o e al.
2001). Insec immuni y consis s o a panel o humo al and cellula
de ense mechanisms, ac ing local and sys emically, which oge he
p o ide e icien p o ec ion agains pa hogens. The humo al
componen o he immune sys em include he ansc ip ional ac i a ion
o genes ha lead o he p oduc ion o e ec o molecules ha
ecognize (Pa e n Recogni ion Pa e ns, PRPs) and kill pa hogens
(AMPs, eac i e in e media es o oxygen and ni ogen, and
melaniza ion o hemolymph) (La ine & S and 2002). AMPs a e
p oduced in he insec a body and eleased in o he hemolymph, bu
also in epi helial cells and in he insec ´s blood cells, hemocy es
(Lemai e & Ho mann 2007). Hemocy es, a e in ol ed in p ocesses
such as phagocy osis, o ma ion o mul icellula hemocy e agg ega es,
nodula ion and encapsula ion o in ading pa hogens, and p oduc ion
o oxygen and ni ogen eac i e species (S and 2008; Ele he ianos &
Re enis 2011; Yi e al. 2014). Hemocy es a e ec ui ed o he wound
si e and engul damaged epi helial cells and pa hogens (Wood e al.
2006; La ine & S and 2002; Losick e al. 2013). These wound
esponse p ocesses a e highly conse ed, bu he e a e also
pa icula i ies o wound esponse in di e en species and in di e en
de elopmen al s ages and/o issues (Bosch e al. 2005; Gu ne e al.
2008).
An in e es ing pa icula i y o wound esponse occu s in some
Lepidop e ans, he insec o de o bu e lies and mo hs. Wounds in
de eloping pupal wings can lead o he o ma ion o o ganized
pigmen a ion pa e ns cen e ed a wound si es (Nijhou 1985;
B ake ield and F ench 1995; O aki 2011). The wound esponse
p ocess is ypically and b oadly igh ly connec ed o pigmen a ion, and
p esence o di e en pigmen a ion pa ches and spo s a wound si es is
e y common (Le esque e al. 2012; Binggeli e al. 2014). In his
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
21
pa icula wound esponse in a p e-adul and pos -g owing
de elopmen al s age, he e is he o ma ion o well-o ganized
pigmen a ion pa e ns ha esemble na i e wing eyespo s, pa e n
elemen s composed o concen ic ings o di e en colo s (Figu e
2.1A).
Bu e ly eyespo s ha e been p ime models o e o-de o s udies
(Beldade & B ake ield 2002) including s udies o he o igin and
di e si ica ion o e olu iona y no el ies, lineage speci ic ai s (Beldade
& B ake ield 2002; Saenko e al. 2008). Eyespo s de elop a ound
o ganizing cen e s ha , ea ly in pupal wings a e hough o p oduce
di usible signals ha induce su ounding epide mal cells o o m ings
o di e en colo s (F ench & B ake ield 1995). Wound si es also
beha e like o ganize s, eleasing signal o he su ounding and
p omo ing ec ui men o immune cells and eepi heliza ion
(Nie hamme e al. 2009; Bidla e al. 2009; Galko & K asnow 2004).
The ac ha wound si es can induce he p oduc ion o eyespo -like
pa e ns lead o he sugges ion ha he e olu iona y o igin o bu e ly
eyespo s migh be ela ed o he co-op ion o wound esponse gene ic
ci cui y conse ed ac oss insec s (Mon ei o e al. 2006). Co-op ion is
a p ocess by which e olu ion ecycles ances al gene ic ci cui ies
sha ed ac oss lineages o p oduce no el ai s (T ue & Ca oll 2002;
Gan o nina & Sánchez 1999). Melanogenesis is a good example o
his pleio opy; being in ol ed in body pigmen a ion, bu also in o he
impo an physiological unc ions such as immuni y (Kim e al. 2013;
Wi kopp & Beldade 2009). Since immuni y is ac i a ed a e wounding,
he melanin pa hway genes a e good candida es o b idge he wound
esponse and he de elopmen al mechanisms o o m eyespo -like
pa e ns. Howe e , un il now he exp ession o hese genes o o he
immune ela ed genes was no epo ed in ea ly s ages o eyespo
de elopmen . Ea ly s udies e ealed ha he abili y o induce

Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
22
eyespo s a ound wound si es is es ic ed in ime (6-18h pos -pupa ion,
o B. anynana indi iduals de eloping a 27°C) and space (on he
dis al hal o he do sal su ace o o ewings) ( B ake ield and F ench
1995).
In his s udy we aimed a cha ac e izing he molecula and cellula
commonali ies be ween wound esponse and eyespo de elopmen ,
which unde lie he hypo hesis o co-op ion o wound esponse gene ic
ci cui y in o he e olu iona y o igin o eyespo s. We wounded B.
anynana pupal wings o compa e gene exp ession p o iles o
wounded and non-wounded wings, and we in es iga ed he spa ial
pa e ns o gene exp ession o “wound genes” and “eyespo genes” a
eyespo cen e s and wound si es espec i ely. We also analysed he
cellula o ganiza ion a wound si es and p esump i e eyespo cen e s.
Ou esul s shown ha , genes (AMPs and melanogenesis enzymes)
and cells o inna e immune sys em (hemocy e-like cells) a e p esen a
wound si es and also a p esump i e eyespo cen e s. An p, which is
known o be in ol ed in eyespo de elopmen , is also exp essed a
wound si es. This sugges s ha inna e immuni y mechanisms migh
ha e been co-op ed du ing e olu ion o eyespo de elopmen .
2.3. Ma e ials and Me hods
2.3.1 Animals
B. anynana wild- ype labo a o y s ock was ea ed as desc ibed in
(B ake ield e al. 2009). B ie ly, he bu e lies we e main ained a
27°C (+/- 0.5ºC) wi h 65% (+/-1%) humidi y on a 12h ligh /da k cycle.
Eggs we e collec ed in young maize plan s om adul cages
(45x45x45cm; BugDo m-44545) wi h app oxima ely 400 indi iduals
ed wi h esh banana on op o we co on. La ae we e ed wi h esh
maize plan s and main ained a densi ies o 150 o 200 indi iduals pe
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
23
cage (same size as adul cages). Du ing he i h la al ins a we
sexed la ae keeping only emales o his s udy. P e-pupae we e
collec ed on o 25 well pla es and pupa ion ime was eco ded du ing
he nigh ia ime-lapse pho og aphy (one pho o e e y 10 min; Canon
1000D digi al came a, Hahnel Giga T P o 2.4GHz wi eless ime
emo e con ol).
2.3.2 Wing wounding and dissec ions
The expe imen s had wo g oups: non-wounded indi iduals (NW)
wi h no manipula ion, and wounded (W) indi iduals, which we e
wounded wice o gene exp ession mic oa ay analysis and qPCR.
Wounds we e in lic ed wi h a ungs en needle (ca . no. 501317; Wo ld
P ecision Ins umen s) on he do sal su ace o he igh o ewing; in
he wo ein-bound wing compa men s below he an e io eyespo ,
app oxima ely hal way be ween he wing ma gin and he no mal
loca ion o he eyespo s (Figu e 2.1B). In expe imen s o spa ial
pa e ns, in si u hyb idiza ion (ISH), luo escen in si u hyb idiza ion
(FISH)) and immunohis ochemis y (IHC) we wounded as be o e, bu
only did one wound on he i s wing compa men below he an e io
eyespo . All wounds we e in lic ed a 12 hou s pos -pupa ion,
acco ding o B ake ield and F ench his is he bes ime poin o induce
an ec opic eyespo (B ake ield & F ench 1995). A e wounding, all
pupae we e e u ned o 27°C un il u he manipula ions.
Fo gene exp ession mic oa ays analysis and qPCR we
dissec ed he wings emo ing he a ached ex e nal cu icle. We
dissec ed wings a wo di e en ime poin s o gene exp ession
mic oa ay analysis ( ou and eigh hou s pos -wounding (Figu e 2.1B)
and i e di e en ime poin s o qPCR analysis ( wo, ou , six, eigh
and 10h pos -wounding). Fo he gene exp ession spa ial pa e ns
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
24
(ISH) and cellula s uc u e analysis (IHC) we dissec ed he wings
keeping he a ached ex e nal cu icle, which helps o main ain he
wing in eg i y.
2.3.3 RNA isola ion and cDNA p epa a ion o mic oa ay
The dis al hal o each dissec ed o ewing (wi hou he cu icle)
was used o RNA isola ion. We had wo eplica e pai s o samples
(wounded and non-wounded wings) o each o wo ime poin s (16h
and 20h pos -pupa ion, co esponding o 4h and 8h pos -wounding,
espec i ely). To al RNA was ex ac ed wi h TRIzol (In i ogen)
ollowing manu ac u e ’s ins uc ions. RNA in RNAse- ee wa e was
checked o concen a ion and pu i y (A260/A280 a io o > 1.8) by
spec opho ome y (NanoD op), and o in eg i y by unning in a 1.1 %
aga ose gel. We used o al RNA o syn hesized and ampli y cDNA
using he O a ion Sys em (NuGEN) wi h oligo-dT p ime and ollowing
manu ac u e 's ins uc ions. We hen pu i ied he cDNA using
Quiaquick cleanup ki (Qiagen) wi h a inal elu ion olume o 30 µl in
1x TE bu e . Pu i y and concen a ion o cDNA samples was
assessed in NanoD op and aga ose gel. All cDNA samples me he
equi emen s o NimbleGen-Roche hyb idiza ion (A260/A280>1.7,
A260/A230>1.5, >1µg cDNA/sample) and we e s o ed a -20°C p io
o shipmen o NimbleGen-Roche (Madison, WI, USA) on d y ice. Cy3
labeling o cDNA was done a NimbleGen-Roche s a ing wi h 1 µg o
quali y-con olled cDNA (Agilen Bioanalyze ) and using Cy3 andom
p ime s and Klenow enzyme. Each sample o cDNA wi h added Cy3
andom p ime s was hea ed o 98°C o 10 min and hen cooled on ice
o 10 min, a e wha Klenow and dNTPs we e added be o e
incuba ion a 37°C o 2h . Reac ion was s opped by adding 10 µl
EDTA. Labeled cDNA was hen p ecipi a ed and used in hyb idiza ion.
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
25
2.3.4 Mic oa ays, hyb idiza ion and slide scanning
To measu e gene exp ession le els, we used Cus om Nimblegen-
Roche mic oa ays (Gene Exp ession 4x72K A ays) wi h ea u es
designed o ep esen 15830 B. anynana gene objec s, co esponding
o con igs o single ons esul ing om he assembly o >200,000 ESTs
(mos ly om de eloping wings) desc ibed elsewhe e (Beldade e al.
2006; Beldade e al. 2009), as well as a numbe o o he genes ha
had been p e iously cloned and implica ed in B. anynana wing
pa e ning. A gene can be ep esen ed by mul iple gene objec s,
which a e named wi h C[numbe ], s anding o con ig; S[numbe ],
s anding o single on, a con ig uniquely associa ed o a UniGene (c. .
(Beldade e al. 2006); and P[numbe ] o hose p e iously cloned by P.
Beldade. The cus om a ay con ained 76,697 60me p obe ea u es
including 69,921 co esponding o he B. anynana genes (each gene
objec being ep esen ed by 1-6 p obes) and a numbe o di e en
ypes o con ols (including 2,000 andom p obes). Mic oa ay
hyb idiza ion, scanning and image ex ac ion was pe o med a
Nimblegen-Roche ollowing s anda d p o ocols (NimbleGen 2011).
NimbleGen p o ided access o he image iles as well as o he
co esponding PAIR iles. We used he non-no malized luo escence
alues ( aw in ensi ies) o u he analysis.
2.3.5 Da a no maliza ion and le els o gene exp ession
The aw in ensi ies p o ided o each mic oa ay ea u e we e
no malized wi h he so wa e ANAIS, a web-based ool o he
p ocessing o NimbleGen exp ession da a (Simon & Bio 2010). B ie ly,
aw in ensi ies we e no malized o in a-a ay (RMA backg ound
co ec ion) and in e -a ay (quan ile) a ia ion. The same ool was also
used o do he “p obe o gene con e sion” (using median alue o all
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
32
650bp. Fo ISH, he p obes we e labeled wi h digoxigenin (DIG) by in
i o ansc ip ion wi h T7 o SP6 RNA polyme ase (P omega) in a
eac ion con aining DIG-UTP (DIG RNA labeling mix, Roche) and
ollowing manu ac u o ’s ins uc ions. Fo FISH, we labeled ple p obe
wi h DIG as be o e and Glo 2 p obe wi h bio in, using a BIO RNA
labeling mix (Roche) ollowing he same p o ocols desc ibed o ISH.
We de ec ed DIG and bio in wi h sheep an i-DIG (Roche) and mouse
an i-bio in (Roche) p ima y an ibodies, espec i ely. These p ima y
an ibodies we e hen de ec ed wi h seconda y an ibodies, an i-sheep
and mouse coupled wi h Alexa 488 and 594 (In i ogen), espec i ely,
ollowing he p o ocol desc ibed in (Kosman e al. 2004).
ISH images we e acqui ed in a s e eoscope (Leica MZ6
S e eoscope) coupled o a digi al came a (Leica DFC420 C Digi al
Came a SW Ki ) and analyzed in Fiji-ImageJ so wa e (Schindelin e al.
2012). FISH images we e acqui ed in he SP5 Leica SP5 con ocal,
using he 40x 1.3NA Oil imme sion objec i e and o mo e de ail o
cells we applied zoom. Z-s acks we e hen con e ed in o a single
image, and i s b igh ness and con as we e adjus ed in Fiji so wa e
(Schindelin e al. 2012).
2.3.11 Immunohis ochemis y (IHC)
We used wounded and con ol o ewings wi h o e laying cu icle
ha we e collec ed a 4 and 8 h pos -wounding. The wings we e ixed,
dehyd a ed (5min sequen ial washed wi h 25%, 50%, 75% and 100%
e hanol) and s o ed a -20°C un il use. Immedia ely be o e s a ing he
immunohis ochemis y p o ocol desc ibed in (Saenko e al. 2011), he
wings we e ehyd a ed (5min sequen ial washes wi h 100%, 75%,
50% and 25% e hanol). We used DAPI o To-P o and hodamine
phalloidin (In i ogen) and ollowed manu ac u e ’s ins uc ions, o

Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
33
s ain nuclei and ac in, espec i ely. We used an ibodies agains An p
4C3 (dilu ion 1:200) (Condie e al. 1991), ob ained om he
De elopmen al S udies Hyb idoma Bank, Eng ailed (En) 4F11
(dilu ion 1:50) (Pa el e al. 1989) and phospho-his one H3 (Se 10) 6G3
(Co ez e al. 2001) om Cell Signaling Technology (dilu ion 1:500).
We de ec ed hem wi h an i-mouse Alexa 488 (An p, and En) and an i-
mouse Alexa 647 (phospho-his one H3 (Se 10)) (dilu ion 1:200)
(In i ogen).
Fo IHC we used he Ligh shee mic oscope (Gualda e al. 2013)
a ailable a Ins i u o Gulbenkian de Ciência. Ligh shee mic oscopy is
a luo escence mic oscopy echnique, whe e he illumina ion is done
pe pendicula ly o he de ec ion. The echnique shapes he
illumina ion lase beam in o a ec angle and hen ocuses i down only
in one di ec ion, using a cylind ical lens (SPIN). This o ms a hin
"shee o ligh " igh in he ocal plane o he de ec ion objec i e,
illumina ing he whole sample plane a he same ime. A e s aining,
wings we e in oduced in a glass cylinde illed wi h o liquid aga ose
(0.1%), a e solidi ica ion he glass cylinde was emo ed and he
wings we e imaged wi h he 4x and/o he 10x objec i es wi h 0.13
and 0.30NA, espec i ely. Because o he wing cu a u e, i was
di icul o ob ain images wi h eyespo s and wound si es ocused on
he same plane.
Mi o ic cells we e de ec in he Zeiss’ S e eo Luma luo escen
s e eoscope (wings dissec ed 1h30 min a e wounding) in z-s ack
se ies ha we e con e ed in o a single image h ough z-p ojec ion
maximum in ensi y in Fiji so wa e (Schindelin e al. 2012). Mi o ic cells
we e quan i ied in wings dissec ed a 4h pos -wounding. Images we e
acqui ed in a SP5 Leica SP5 con ocal, using 10x 0.40NA objec i e. Z-
s acks we e con e ed in o a single image, and i s b igh ness and
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
34
con as we e adjus ed in Fiji so wa e (Schindelin e al. 2012). The
wings we e imaged in he an e io dis al a ea wi h an e io eyespo
and wound si e o co esponding wing egion. We quan i ied he a ea
o luo escence o DAPI and o phospho-his one H3 (Se 10) in he
o al image a ea (1550 µm2) o i e indi iduals. The luo escence
in ensi y was adjus ed equally in all images o each luo opho e using
he h eshold in ensi y ool o Fiji, which selec s all he pixels in he
image abo e ce ain luo escence in ensi y. We saw ha he e was
in e ac ion be ween he wing (wounded and con ol) and he o al a ea,
so we could no use he ANCOVA o he s a is ical analysis. Thus, we
used a glm model o es i despi e hese in e ac ion, he e we e
di e ences in he a ea o mi o ic cells be ween wounded and con ol
wing aking he a ea o o al cells in he image as co a ian
(glm(mi o ic cells a ea ~ wing * o al a ea)).
2.4. Resul s and Discussion
Co-op ion is a ecu en e olu iona y s a egy o de elop lineage
speci ic ai s (T ue & Ca oll 2002). He e, we explo ed he ec ui men
o genes ela ed wi h wound esponse, an ancien e olu iona ily
conse ed p ocess (Moussian & U 2005; Ne es e al. 2015; Vilmos &
Ku ucz 1998), o he de elopmen o he Lepidop e ans eyespo s,
lineage speci ic ai s (Beldade & B ake ield 2002; Saenko e al. 2010).
We also in es iga e he exp ession o genes ela ed wi h eyespo
de elopmen a he wound si e and he cellula s uc u e a he wound
si e and p esump i e eyespo cen e . We s a ed compa ing he gene
exp ession p o iles (mic oa ays, Figu e 2.1 and 2.2), dynamics
(qPCR Figu e 2.3) and spa ial pa e ns (ISH, Figu e 2.4 and FISH 2.5)
o pai s o wounded and non-wounded wings. A e we compa ed he
exp ession o a ge genes ela ed wi h eyespo de elopmen (Figu e
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
35
2.6) and he cellula s uc u e a wound si es and p esump i e eyespo
cen e s (IHC, Figu e 2.7).
2.4.1 Wound esponse induces he exp ession o AMPs and
melanogenesis enzymes in B. anynana ea ly pupal wings
The wound esponse in insec s is well-s udied in D osophila,
mainly du ing emb yonic and la al de elopmen al s ages (Lemai e &
Ho mann 2007; Mace e al. 2005). The gene ic esponse du ing pupal
s age, a pos -g ow h and p e-adul s age, is poo ly unde s ood in
insec s (Regan e al. 2013). He e, we used he bu e ly B. anynana,
which de elops o ganized pigmen a ion pa e ns a ound wound si es
(B ake ield & F ench 1995) (Figu e 2.1A), o unde s and he
“gene ali ies” and he “pa icula i ies” o pupal wound esponse. We
compa ed gene exp ession p o iles o wounded (manipula ed wings,
M) s. con ol (C) wings o he same indi idual (pai ed samples) o
access gene exp ession changes induced by wounding pupal wings in
a manne ha ypically leads o he p oduc ion o wound-induced
ec opic eyespo s (B ake ield & F ench 1995). M and C wings o wo
biological eplica es we e collec ed a 4 and 8h pos -wounding (Figu e
2.1A) and mRNA le els he ein used on a cus om-made mic oa ay o
B. anynana genes.
This expe imen was a pilo es pe o med in 2007 and designed
o es and alida e he Cus om Nimblegen-Roche mic oa ays made
wi h B. anynana gene objec s. Fo economic easons, i was decided
use only wo biological samples and compa e wounded and non-
wounded wings in wo di e en ime poin s (Figu e 2.1B). I was
expec ed ha pai ed wings clus e ed oge he as well as same ime
poin wounded indi iduals. Indeed, his was he case (Figu e 2.1C
bo om). I was also expec ed ha wounded wings exp essed some
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
36
well-known genes ela ed wi h insec s wound esponse, which also
happen (Figu e 2.2B). This expe imen was enough o alida e he B.
anynana Cus om Nimblegen-Roche mic oa ays as a me hodology o
s udy gene exp ession in his model sys em. Howe e , wi h only wo
biological eplica es i was no enough o make a p ope mic oa ay
s a is ical analysis (Pawi an e al. 2005). Ne e heless, once we we e
s udying he wound esponse in his model sys em we decided o look
a his da a, conside ing i s limi a ions and adop ing a conse a i e
app oach ha educes alse posi i es a he expense o inc easing
alse nega i es hi s. A e no maliza ion analysis (Figu e 2.1C op), we
conside ed as di e en ially exp essed gene objec s ha we e
exp essed a leas wice ( old change (FC)>2; up- egula ed), o hal
(FC<2; down- egula ed) on M s. C wings in bo h biological eplica es.
Wi h hese c i e ia, we ob ained a o al o 7821 di e en ly exp essed
gene objec s, co esponding o 3160 a 4h and 4661 a 8h.
The compa ison o gene exp ession in pai ed wings e ealed ha
he majo i y o di e en ly exp essed genes we e no common o he
eplica e indi iduals (93% a 4h and 92% a 8h)(Figu e 2.1D). The
a ia ion in wound esponse gene exp ession migh indica e o he
le els o a ia ion such as 1) de elopmen al ime: he ime in hou s
ha we quan i ied (4 and 8h pos -wounding) migh no co espond o
he same biologic de elopmen al ime in bo h biological eplica es; 2)
gene ic: biological eplica es migh ha e conside able gene ic
di e ences since we we e using an ou b ed s ock, and 3)
de elopmen al and gene ic: a conjuga ion o he wo p e ious
a ia ion sou ces.
We only p oceeded wi h he analysis o genes objec s ha we e
di e en ly exp essed in bo h biological eplica es (Figu e 2.1D). This
conse a i e app oach educes alse posi i es, which is c i ical
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
37
because he limi a ions in he expe imen al design. Conside ing only
hose genes, we ob ained a lis o 624 gene objec s, 20% (126 genes)
we e up- egula ed in M s. C wings a 4h, 40% (249) a 8h, and 18%
(112) and 22% (137) we e down- egula ed a 4 and 8h, espec i ely.
F om he 624 gene objec s di e en ly exp essed in M s. C wings
o bo h eplica es only 34% we e anno a ed (214 genes). 60% o
anno a ed genes we e up egula ed (52 a 4h and 74 a 8h) and 40%
down egula ed (33 a 4h and 55 a 8h) (Figu e 2.2A). All anno a ed
genes we e used in a gene en ichmen analysis o GO biological
p ocesses. This analysis e ealed ha in o al o bo h ime poin s
he e we e 48 genes in he en iched biological p ocesses
(hype geome ic es wi h Benjamini & Hochbe g False Disco e y Ra e
co ec ion (p- alue<0.005)), dis ibu ed o e 52 di e en GO e ms
(one gene can be included in one o mo e GO e ms). A 4h he e
we e 20 genes ep esen ing 36 en iched GO e ms and a 8h he e
we e 29 genes ep esen ing 40 en iched GO e ms. In o al, 24 GO
e ms we e en iched in bo h ime poin s (Figu e 2.3B) con aining
se en up- egula ed genes ( h ee an imic obial pep ides (AMPs) and
ou melanogenesis ela ed genes) ha we e common o bo h ime
poin s. The only down- egula ed en iched genes a e om GO e ms
ela ed wi h cu icle de elopmen a 8h.
The same gene is ypically associa ed wi h se e al GO e ms, so
we g ouped he GO e ms lis in h ee gene al ca ego ies acco ding
wi h hei biological p ocesses: pep ides ha kill cells o o he
o ganisms o modula e his esponse we e g ouped as AMPs,
enzymes ela ed o simila o melanogenesis pa hway enzymes as
melanogenesis, and none o hese as o he . A 4h he e we e se en
en iched genes in he AMP ca ego y and six a 8h, bu a 4h all genes
we e ela ed wi h ac i a ion o immune sys em and a 8h he e we e

Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
38
also nega i e egula ion o immune esponse (Figu e 2.2C). The
en ichmen o melanogenesis genes was highe o bo h ime poin s
ela i e o he o he wo ca ego ies wi h same numbe o genes (13),
including known melanogenesis pa hway enzymes such as ple and
Ddc.
Figu e 2.1- Wound esponse ansc ip omic analysis. A) Pheno ypic e ec s on adul
wing do sal side. Epide mal wounding on wings o 12h old pupae induces he
o ma ion o an o ganized pigmen a ion pa e n a ound wound si es (*) simila o
na i e eyespo s (a ows). B) Expe imen al design: pupae we e wounded wice in one
o he o ewings (as e isks in g een d awing ep esen ing a pupa) a 12h pos -
pupa ion and bo h hei wings (wounded/manipula ed, M, and con ala e al/non-
wounded, C) we e collec ed 4 o 8h pos -wounding. We had wo indi iduals pe ime
poin ( ep esen ed as wo shades o ligh (4h) o da k (8h) blue). C) Quali y con ol
g aphics o he mic oa ay da a. Boxplo o in ensi ies a e da a no maliza ion ( op)
and hie a chical clus e o no malized in ensi ies (bo om). D) Fo each collec ion ime
poin , he Venn diag am ep esen ed he numbe o di e en ially exp essed genes, a
leas wo- old di e ence in gene exp ession be ween C and M wings o each
indi idual (each ci cle). Upwa ds and downwa ds a ows ep esen s genes o e -
exp essed and unde -exp essed, espec i ely, in M s. C wings o each indi idual.
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
39
The anno a ed genes ha we e di e en ially exp essed in M s. C
wings we e mainly om wo main ca ego ies: AMPs and
melanogenesis pa hway enzymes (Figu e 2.2B and C). Bo h
ca ego ies ha e well-known unc ions in insec wound esponse and
in ec ion (Lemai e & Ho mann 2007; Tanaka & Yamakawa 2011).
The o he di e en ially exp essed genes wi hou anno a ion can be
conside ed candida es o “pa icula i ies” o his wound esponse,
which migh co e speci ic wound genes ac ing du ing pupal s age
ac oss insec s, and also speci ic genes ha p omo e he de elopmen
o pigmen a ion colo pa e ns in Lepidop e ans. In his lis also can be
ep esen ed new “gene al” conse ed wound esponse genes.
We alida ed and ex ended he mic oa ay esul s h ough qPCR.
We selec ed di e en ly exp essed genes and quan i ied he gene
exp ession a 4 and 8h pos -wounding as in mic oa ay analysis. We
added h ee mo e ime poin s (2, 6, and 10h pos -wounding) o be e
unde s and he gene exp ession dynamics o hese genes a e
wounding. The c i e ia o selec genes in his analysis we e exp ession
di e ence be ween M and C wings we conside ed only genes wi h a
leas one in on, which allows in on spanning p ime design o
di e en ia e he cDNA and gDNA ampli ica ion p oduc s. We selec ed
cec6 and cecD, wo AMPs, om AMP ca ego y and ple and p i1 om
melanogenesis ca ego y. We also included non-anno a ed genes (see
supplemen a y File S2.1 – a ailable only online) wi h high sequence
simila i y wi h genes o p o ein domains desc ibed o be in ol ed in
insec de ense mechanisms. Glo 2 is an AMP (Kaneko e al. 2007)
and lys-lp1 is a lysozyme like p o ein, bo h a e e ec o gene in
immune sys em playing a ole in killing bac e ia (C a a e al. 2015). c-
l10 is a (Ca(2+))-dependen lec in and is in ol ed in pa hogen
ecogni ion (Cambi e al. 2005). P o-in6 is a p o ein wi h a ypsin
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
40
inhibi o -like cys eine- ich domain; p o eins con aining his domain
play an
Figu e 2.2- Wounded wings a e en iched o genes encoding o AMPs and
melanogenesis pa hway enzymes. A) Numbe o anno a ed and non-anno a ed genes
di e en ially exp essed in M s. C wings. B) Hea map able ep esen ing he gene
en ichmen analysis o he lis o anno a ed genes. GO e ms o up- egula ed genes
a e in ed; hose o down- egula ed genes a e in blue. Whi e ep esen s he non-
en iched GO e ms. The numbe s ep esen he p opo ion o genes di e en ly
exp essed ela i e o genes o hese GO e m p esen in he mic oa ay. Nega i e
alues ep esen down- egula ed GO e ms. C) Pe cen age and numbe o genes ha
a e ela ed wi h AMPs, melanogenesis and o he han hese biological p ocesses.
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
41
impo an ole in immune esponse and an icoagula ion (Zeng e al.
2014; Jin e al. 2011) . Based on hese pu a i e unc ions we g ouped
he Glo 2, lys-lp1 and c- l10 oge he wi h cecD and cec6 in he AMP
ca ego y. p o-in6 was g ouped in melanogenesis ca ego y because in
insec s his pa hway is in ol ed in coagula ion p ocesses (Má kus e
al. 2005). Acco ding o mic oa ay esul s, h ee o he eigh genes
we e up- egula ed a 4 and 8h pos -wounding: Glo 2 (4h: 14.73, 5.51
and 8h: 8.04, 6.50; esul s o wo biological eplica es), p i1 (4h: 3.83,
3.20 and 8h: 3.45, 4.05) and ple (4h: 9.34, 10.59 and 8h: 16.43,
23.14); ou genes only a 4h: cec6 (4h: 2.12, 2.18), cecD (4h: 2.32,
2.63), c- l10 (4h: 6.67, 5.93) and p o-in6 (4h: 6.89, 2.65); and one
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
48
(B ake ield & F ench 1995), which migh indica e ha mo e hemocy es
a p esump i e eyespo cen e s p oduce la ge eyespo s. Fu he mo e,
we only ound hese pa ches o cells in he pos e io eyespo he
la ge one; a he an e io eyespo he p esence o hese cells was no
di e en om o he wing a eas. Fu u e s udies in na u al mu an s wi h
di e ences in eyespo size (P Beldade e al. 2008) migh elucida e i
la ge eyespo s a e co ela ed wi h he p esence o highe numbe o
hemocy e-like cells a p esump i e eyespo cen e s.
2.4.3 An p, an “eyespo gene”, is exp essed a wound si e
We es ed wo “eyespo genes”, En and An p, bo h known o be
exp essed du ing la al and pupal s age a p esump i e eyespo
cen e s (Mon ei o e al. 2006; Saenko e al. 2011; Tong e al. 2014)
and En also du ing pupal de elopmen a p esump i e golden ing
(B une i e al. 2001). Ou esul s demons a ed ha An p is also
exp essed a eyespo cen e s du ing ea ly pupa (16h pos -pupa ion)
(Figu e 2.6B). In Sa y inae bu e lies his Hox gene is he i s known
gene o be exp essed a la al p esump i e eyespo cen e s (Shi ai e
al. 2012; Saenko e al. 2011) and i s exp ession migh be essen ial
h ough de elopmen o p oduce an eyespo .

Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
49
Figu e 2.5- Cells o di e en shape and exp essing Glo 2 a e p esen be ween do sal
and en al epi helia a wound si e and p esump i e eyespo cen e . A) Glo 2 and ple
exp ession in epi helial cells is educed ela i e o i ’s exp ession a wound si es (B).
A he wound si e, he exp ession pa e n o Glo 2 is di use because he e a e cells
s aining o Glo 2 be ween do sal and en al epi helia (D); ple exp ession is e y
in ense and es ic ed o cells nea wound si es. C) Glo 2 exp ession is inc eased a
p esump i e eyespo cen e (a ow) bu ple is no . De ailed images o he squa e
sec ions ( om 1 o 9) a e shown in E. G ay images a e Glo 2 (cen e ) and ple
(bo om) exp ession. E1,E 2, E3, E7 and E8 a e images om cells in he en al
epi helium and E4, E5, E6 and E9 om space be ween bo h epi helia. Images om
A,B and C a e z-p ojec ions o 19 - 23 s acks. D is a lowe ampli ica ion o wing a ea
shown in B o Glo 2 whe e we excluded he s acks con aining en al cells. In E1 o
E9 he images a e z- p ojec ions o 2 - 4 s acks. D awings on E1 and E5 ep esen he
app oxima e cellula shape. A owheads a e poin ing cells exp essing Glo 2. The
d awings on he op indica e he app oxima e posi ions o he images and he c ossed
a ows he wing o ien a ion; an e io (A), pos e io (P), dis al (D), p oximal (P ). Scale
ba s a e 50µm and images om 1 o 9 a e squa es o 50x50µm. Images a e om an
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
50
indi idual dissec ed 8h pos -wounding, bu simila esul s we e ound a 4h pos -
wounding.
As in p e ious s udies, En was exp essed a eyespo cen e s and
golden ings, bu a his ime poin (4h pos -wounding) i was no
exp essed a wound si e (Figu e 2.6A). In con a y, An p was
exp essed a wound si es (Figu e 2.6B). As a ansc ip ion ac o his
p o ein is ac i e when i is exp essed in he nucleus as i occu s a
eyespo cen e (Figu e 2.6B). Un o una ely, in his expe imen ou
nuclea s aining did no wo ked and a he wound si es he An p
s aining is dispe sed. We i s hypo hesized ha An p exp ession a
wound si es was unspeci ic binding o he an ibody o he wound si e,
bu he nega i e esul o En (an an ibody also p oduced in a mouse,
un acco ding same IHC p o ocol) a wound si es indica es ha his is
no he case (Figu e 2.6). We also obse ed spo s o Ac in a eyespo
cen e s, wound si es and also sp ead in an appa en andom pa e n
o e he wing (Figu e 2.6B). These Ac in spo s a e loca ed be ween
epi helia and some o hem co-loca e wi h cells exp essing An p
(Figu e 2.6). The ac in s aining a wound si e is e y in ense as well as
An p (co-localiza ion in yellow, Figu e 2.6B).
An p is exp essed in insec hema opoie ic cells ha can
di e en ia e in plasma ocy es, c ys al cells and in cases o inju y o
in ec ion in lamellocy es (Mandal e al. 2007; G igo ian e al. 2011). In
o he expe imen s we saw ha hemocy es isola ed om young pupae
s ongly s ain o Ac in (Supplemen a y Figu e S2.2). Toge he , his
indica es ha he exp ession o An p a wound si e 4h pos -wounding
is e lec ing he p esence o hema opoie ic p ecu so cells and hose
migh be essen ial in hemocy e di e en ia ion and wound healing. In
his expe imen , we classi ied An p as an eyespo gene (Saenko e al.
2011), bu in ac his ansc ip ion ac o is a pleio opic p o ein wi h
mul iple unc ion du ing limb de elopmen (Shiga e al. 2002),
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
51
neu onal ne wo k (Jolio e al. 1991) and also in hema opoiesis
(G igo ian e al. 2011; Mandal e al. 2007). Ea ly a e wounding i
migh be necessa y o p omo e hemocy e di e en ia ion. The unc ion
o his ansc ip ion ac o a eyespo cen e s o p oduce an o ganized
pigmen a ion pa e n is no known. The new me hodologies o genome
edi ing can be an impo an ool o unco e he biological unc ion o
An p du ing eyespo de elopmen . Ne e heless, e en wi hou his
ool u he expe imen s can be pe o med including IHC o An p and
nuclei, and image acquisi ion wi h a highe magni ica ion objec i e o
mo e de ail o cells exp essing An p a wound si es.
Figu e 2.6- An p, bu no En, is exp essed a wound si e. A) En is exp essed a
p esump i e eyespo cen e and golden ing bu i was no de ec ed a wound si e. B)
An p is exp essed a p esump i e eyespo cen e s and wound si es. Ac in and An p
a e exp essed a wound si e (colocaliza ion in yellow). Top images a e a Z-p ojec ion
and bo om images a e o hogonal iews o wound si e. Blue a ows indica e he
o ewing an e io eyespo cen e , ed a ow he pos e io eyespo , a owheads he
ac in spo s, he as e isk (*) he wound si e. The d awings indica e he app oxima e
posi ions o he images and he c ossed a ows he wing o ien a ion; an e io (A),
pos e o (P), dis al (D), p oximal (P ). Scale ba s a e 100µm.
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
52
2.4.4 The wound inhibi s cell p oli e a ion
The epi helium inju y induces changes in cell and issue
o ganiza ion (Razzell e al. 2011; K au z e al. 2014). Ou p e ious
expe imen s (Figu e 2.5 and 2.6) ha e sugges ed ha p esump i e
eyespo cen e s and wound si es ha e highe cell densi y han o he
a eas o wing epi helia. Highe cell densi y a wound si es can be due
highe cell p oli e a ion a e o cell mig a ion owa ds wound si e. Cell
p oli e a ion a e can inc ease a wound si es o es o e and e-
epi helize he wounded epi helium, gene ally his is he las phase o
wound healing (Reinke & So g 2012; Razzell e al. 2011). To
in es iga e whe he mo e cells a wound si es and eyespo cen e s a e
a consequence o locally inc eased cell p oli e a ion s. cell mig a ion,
we used a mi o ic ma ke . A bo h si es he epi helium was hicke han
neighbo ing egions and had mo e cells (Figu e 2.7A and B). A
p esump i e eyespo cen e he e is an ou g ow h owa ds he en al
epi helium in compa ison wi h neighbo ing a eas (Figu e 2.7A). The
same occu s a wound si es (Figu e 2.7B).
To unde s and i we e he e highe cell p oli e a ion we s ained
mi o ic cells in wounded and non-wounded indi iduals om 30min o
4h pos -wounding. He e, we a e showing he s aining a 1h30min
(Figu e 2.7C op), bu we obse ed he same end in all es ed
indi iduals. The e a e ewe p oli e a ing cells a ound wounds in
compa ison wi h nea by wing. The wound was associa ed o inhibi ion
o cell p oli e a ion ela i e o he con a la e al non-wounded wing. To
quan i y cell p oli e a ion we used i e indi iduals dissec ed 4h pos -
wounding and compa ed he a ea o mi o ic cells in wounded s.
con ol wing o same indi idual. Wounded wings had signi ican ly
lowe a ea o mi o ic cell ela i e o espec i e con ala e al wings
(lines in Figu e 2.7C bo om). These esul s indica e ha he highe
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
53
cell densi y a ound wound si es is likely due o cell mig a ion, which is
common in he in lamma o y p ocess a e wounding (Wu e al. 2009;
Williams 2007).
All nuclea RNA syn hesis is ep essed du ing he mi o ic phase o
he cell cycle (Ha l 1993; Spence 2000). A e wounding, cells ha e
o eadjus hei ansc ip ion and ini ia e he syn hesis o wound
esponse genes such as AMPs and melanogenesis enzymes. I is
possible ha cells a ound wound si es a e mo e ansc ip ionally
ac i e in compa ison wi h o he cells in he wing, which educes he
p obabili y o mi osis. Double s aining wi h a mi o ic ma ke and a
ansc ip ion ma ke , such as RNA polyme ase II, can con i m his
hypo hesis.
Figu e 2.7- High cellula densi y a p esump i e eyespo cen e s and wound si es is
no consequence o highe cell p oli e a ion. A p esump i e eyespo cen e s (A) and
wound si es (B) he nuclea s aining is mo e in ense ela i e o o he wing a eas. The
o hogonal iews in he igh show he nea wing epi helium ( op; line 1) and he
in e es zone (bo om; line 2). C) Lowe cell p oli e a ion a wound si es. Top images
a e s aining mi o ic cells in he wounded (le ) and he con ala e al non-wounded wing
o he same indi idual ( igh ) (1h30min pos -wounding). Bo om g aphic is he
quan i ica ion o mi o ic cells in wings pai s (connec ed wi h a line) 4h pos -wounding.
The e a e ewe mi o ic cells a ound wound si e in compa ison wi h same a ea in
con ol wing (glm model wi h o al cells a ea imaged as co a ian ; F=7.34, p-
alue=0.035). Blue a ows indica e he o ewing an e io eyespo cen e , he as e isk

Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
54
(*) he wound si e and dashed lines he wing eins posi ion. D awings indica e he
app oxima e posi ions o he images and he c ossed a ows he wing o ien a ion;
an e io (A), pos e io (P), dis al (D), p oximal (P ), do sal (Do), en al (V). Scale ba
in A and B is 100µm and in C is 500µm.
2.4.5 The e a e cellula simila i ies be ween wound si es and
p esump i e eyespo cen e s
The p esence o mo e cells and lowe cell p oli e a ion a e
(Figu e 2.7) indica e ha inc eased cell densi y a wound si es likely
esul s om cell mig a ion, which is consis en wi h cellula immune
esponse (S and 2008). Indeed, analysis o gene exp ession spa ial
pa e ns e ealed he p esence o cells ha migh be hemocy es
(Figu e 2.5), wi h simila shape o plasmo ocy es and g anolocy es
(Ka anagh & Ree es 2004). The p esence o plasma ocy es is
especially in ense a e a pa asi izing wasp ha induces he mig a ion
o plasma ocy es owa ds pa asi ic egg (Mo ime ; Russo e al. 1996).
Ou wounding ea men migh be compa ed o he wasp o iposi o
wounding; his physical damage can be su icien o induce he same
ype o cellula immune esponse. In d osophila, s udies du ing pupal
s age ha e demons a ed ha a e lase epide mal damage he e
we e ac i e mig a ion o hemocy es unde he epi helium, o en om
he so-called ‘sessile pa ches’ (Regan e al. 2013). Howe e , he
cellula iden i y o such cells migh be con i med o con i m his
hypo hesis.
Hemocy e-like cells we e also p esen a he p esump i e eyespo
cen e s o ming clus e s (Figu e 2.5). A eyespo cen e s and wound
si es Ac in s aining was mo e in ense ela i e o o he wing a eas
(Figu e 2.6). Ac in is a molecule in ol ed in cell cy oskele on
a chi ec u e, when cells need o eadjus hei shape as in wound
healing hey change ac in loca ion and pola iza ion (Wood e al. 2002;
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
55
Razzell e al. 2011). Mo ile cells, such as hemocy es, a e ich in ac in
(Supplemen a y Figu e S2.2) and also ac i a e he same kind o
p ocesses o mo e owa ds wound si es (Sampson e al. 2013; Wood
e al. 2006). Eyespo cen e s ha e mo e Ac in s aining bu also ha e
mo e cells; hese Ac in- ich cells can be hemocy es. Ne e heless,
Ac in- ich pa ches o cells a e no exclusi e o wound si es and
p esump i e eyespo cen e s; hey a e sp ead h oughou he wing
(Figu e 2.6). S ill, hemocy e-like cells exp essing Glo 2 agg ega ed in
pa ches we e obse ed only a wound si es and p esump i e eyespo
cen e s. This can indica e he e ogenei y o hemocy e-like cells in
de eloping wings ha exp ess di e en combina ion o genes.
Al oge he , wounding induces an in lamma o y esponse wi h
mig a ion o hemocy e-like cells exp essing Glo 2 o he wound si e,
which inc eases he cell densi y. The same ype o cells is also
p esen a he p esump i e eyespo cen e s, which also ha e highe
cell densi y wi h a simila cell o ganiza ion as a he wound si es as
exempli ied in scheme o Figu e 2.8.
Figu e 2.8- Scheme o cellula s uc u e o wing epi helia, wound si e and
p esump i e eyespo cen e . A wound si es and eyespo cen e s he hickness o
wings is highe because hey ha e hemocy e-like cells be ween do sal and en al
epi helia.
Chap e 2
_____________________________________________________________________________________________________________________________________________________________________________________________________________
56
2.5. Conclusions
Bu e ly eyespo s a e e olu iona y no el ies and ep esen an
excellen oppo uni y o unde s and he e olu iona y o igin o no el
ai s. In his chap e we aimed o unde s and i he e we e cellula and
gene ic commonali ies be ween wound esponse and eyespo
o ma ion. The exis ence o such commonali ies suppo s he
hypo hesis o co-op ion o wound esponse gene ic ci cui y o de elop
bu e ly eyespo s. Ou esul s ein o ce his hypo hesis; we
demons a ed ha he wound esponse and he de elopmen o
eyespo s sha e cellula and gene ic mechanisms.
2.6. Acknowledgemen s
We would like o hank An hony Long and Nicolien Pul o help
wi h ob aining he mic oa ay da a. Daniel Sob al, Paulo Almeida, and
Rena o Al es (IGC’s Bioin o ma ics Uni ) o gene anno a ion and
en ichmen analysis. Jo g Becke (IGC) and Adeline Simon (ANAIS
p ojec ) o ad ising in he mic oa ay analysis. Ma a Ma ial a o he
guidance in qPCR, Da id Duneau and Nelson Ma ins o he
discussions and sugges ions abou s a is ical analysis. The Uni o
Imaging a IGC o all he assis ance du ing image acquisi ion and
analysis, in pa icula o Emilio Gualda in Ligh shee SPIN mic oscope
assis ance. Élio Sucena, Ma a Ma ial a, El i a La uen e, Alexand e
Lei ão and Magda A ilano o discussions h oughou his p ojec .
57
Chap e 3 – Immuni y egula es he
size and scale composi ion o
wound-induced pigmen a ion
pa e ns
3.1. Summa y
The ea ly de elopmen o o ganisms in ol es a se ies o igh ly
egula ed p ocesses ha de e mine he adul pheno ype. The di e si y
ac oss species and also he a ia ion o a species is e y s iking in
bu e ly wing colo pa e ns, which a e o med by single colo scales
o ganized in a mosaic. The scale de elopmen and he de e mina ion
o i s pigmen s occu du ing pupal de elopmen . Eyespo s a e bu e ly
colo pa e ns elemen s combining di e en colo ings o ganized
a ound a cen al pupil. Du ing pupal s age he wing cells a e a ed o
p oduce di e en scale mo phologies and pigmen s. In some
bu e lies an epi helial wound du ing ea ly pupal de elopmen can
change he a e o wing cells, which p oduce scales wi h di e en
colo s a ound wound si es, wound-induced eyespo s (WIE). He e, we
used he lab model Bicyclus anynana o in es iga e he e ec s on cell
a e a e inducing di e en le els o immune sys em ac i a ion. We
obse ed ha he ea men wi h hea -killed bac e ia a wounding is
e y e icien o ac i a e local and sys emic immuni y. In con as wi h
p e ious s udies inducing di e en le els o wounding se e i y, he
ac i a ion o highe immuni y le els did no induced mo e, bu induced
la ge WIP. This indica es ha immuni y a ec s he numbe o cells
changeing hei de elopmen al p og am a ound wound si es, bu i is
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
64
NanoD op, and o in eg i y by unning in a 1% aga ose gel. We
syn hesized he cDNA aking 400ng o RNA as empla e in a 10ul o al
olume eac ion using 40U o he e e se ansc ip ase (M-MLV
RNase H Minus, P omega) and 0.5µg oligo(dT) ollowing
manu ac u e ’s ins uc ions. The cDNA p oduc was dilu ed o a inal
olume o 50µL and used as empla e.
We used he 7500 Real-Time he mal cycle (Applied Biosys ems)
o un all qPCR. The eac ion mixes we e op imized o a o al olume
o 10uL wi h 5uL iQTM SYBR G een supe mix (BioRad), 3.6uM o
each p ime , and 1uL o cDNA empla e. The PCR p og am included
an ini ial s ep o 95ºC o 10min, ollowed by 40 cycles o 95ºC o 20
sec, 59ºC o 20 sec, and 72ºC o 30 sec. The speci ici y o he qPCR
eac ions was moni o ed wi h mel ing cu es ob ained om he
dissocia ion cu e (59-95ºC), using SDS so wa e ( e sion 1.4;
Applied Biosys ems). We quan i ied he exp ession o h ee a ge
genes: one an imic obial pep ide (AMP), Glo e in2 (Glo 2), a
melanogenesis enzyme, pale (ple) and a po en ial melanogenesis-
ela ed enzyme, p imo1 (p i1) (see Chap e 2). We designed p ime s
o ampli y amplicons spanning an in on o hese genes and also an
endogenous con ol gene, Ribosomal-like p o ein 10A
(RpL10A)(p ime sequences, amplicon size and homology desc ibed
in Chap e 2- Supplemen a y Figu e S2.1)
Gene exp ession le els we e de e mined ou and eigh hou s
pos -wounding in ou eplica e pupae (each wi h wounded and con ol
wings) and o each biological sample we e un h ee echnical
eplica es. We had ou wounding ea men s (S, NS, Ec6, Ec7) and a
e e ence g oup (NW). We quan i ied he exp ession le els in ou
biological eplica es pe ea men pe ime poin using he mean o
echnical eplica es ( ejec ing he biological samples in which he

Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
65
epea abili y o echnical eplica es we e low: di e ences be ween
echnical eplica es highe han 1.5 cycles). We used he ΔΔ C
me hod o quan i y he gene exp ession (Li ak & Schmi gen 2001;
P a l & P a l 2001). B ie ly, wi h C alues we i s no malize he gene
exp ession o a ge genes (Glo 2, ple and p i1) ela i e o he in e nal
con ol gene (Rpl10A), we ob ained he ΔC alues and calcula ed he
ela i e gene exp ession. A e we compa ed he ela i e gene
exp ession in each ea men (NW, S, NS, Ec6, Ec7) wi h he a e age
o ela i e exp ession o NW g oup o each gene (Glo 2, ple and
p i1) in each ime poin ( ou and eigh hou s) and issue (wounded
and con ol wing), so he esul s a e shown as old change di e ence
ela i e o NW g oup.
We impo ed he esul s o R so wa e (R De elopmen Co e
Team 2015) o g aphical ep esen a ion and s a is ical analysis. Fo
each gene, ime poin and issue we es ed o he e ec o ea men
and echnical eplica e on gene exp ession using he model: Gene
exp ession ~ T ea men * Technical eplica e, which was simpli ied
whene e he e ec o echnical eplica es was no s a iscaly
signi ican . We ans o m ou da a o mee he Shapi o-Wilk no mali y
es and B own-Fo sy h homogenei y o a iance es (p- alue<0.05)
and a e checking ha he e we e no in e ac ions be ween
explana o y a iables we used an ANOVA (ao (Fold Change ~
T ea men * echnical eplica e)). In cases whe e he a iance was no
cons an we used a gene alized linea model (GLM) wi h Gamma
dis ibu ion. When he e we e signi ican di e ences among
ea men s (p- alue<0.01) we used he lsmeans R package (Len h &
He é 2015) o pos -hoc pai wise compa isons and conside ed
di e en gene exp ession be ween ea men s when p- alue<0.01.
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
66
3.3.6 Pheno ypic measu emen s
A e s e ching wings, adul bu e lies we e sac i iced (-20ºC o
a leas 2h s). Wings we e de ached om he ho ax and he do sal
side was scanned (Epson Pe ec ion V600 Pho o). Images we e
impo ed o Fiji (Schindelin e al. 2012) and wings we e classi ied o
p esence/absence o WIE. A e ha , we measu ed he o al a ea o
WIE ( esponse a iable), as well as o al wing and do sal an e io
na i e eyespo o con ol wing (used as co a ian s). We also
measu ed he sca a ea (absence o colo scales) a he wound si e
and calcula ed he colo scales a ea sub ac ing he sca a ea o he
o al WIE a ea (Figu e 3.1). All measu emen da a we e impo ed o R
(R De elopmen Co e Team 2015) o do hei g aphical
ep esen a ions and s a is ical analyses. We es ed he e ec o
ea men on de elopmen p obabili y, o al a ea, sca a ea and o al
colo scales a ea o WIE, aking in conside a ion possible co a ian s
wi h wings a ea and do sal na i e eyespo a ea, and also conside ing
he ba ch o he expe imen ( wo ba ches un in wo di e en mon hs).
To know i he ea men has an e ec on WIE o ma ion
(p esence/absence) we used a GLM wi h a binomial e o dis ibu ion
and a logi link: glm(P esence WIE ~ T ea men + Wing A ea + Do sal
na i e eyespo A ea + Ba ch, binomial). In he WIE a ea ai s we
es ed he e ec o ea men on WIE aking in conside a ion possible
co a ian s wi h wing a ea o na i e eyespo a ea and he expe imen
ba ch. In his analysis we con i med ha he esiduals o he da a o
loga i hmic ans o med da a showed no signi ican depa u e om
no mali y (Shapi o-Wilk es ) o om homogenei y o a iances
(B own-Fo sy h es ). A e we used ANOVA o es o he e ec o
ea men ( ac o wi h ou le els: S, NS, Ec6 and Ec7) on hese a ea
ai s: lm(WIE a ea ~ T ea men + Wing A ea + Do sal na i e eyespo
A ea + Ba ch). Whene e he T ea men had a s a is ical signi ican
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
67
e ec on WIE we pe o med a pai wise compa ison be ween ea men
g oups h ough lsmeans R package (Len h & He é 2015).
3.3.7 Local and Sys emic e ec o immune sys em ac i a ion
Ins ead o wounding one o ewing and ha ing he con a-la e al
o ewing as he con ol, in his expe imen we wounded bo h wings.
We compa ed he WIE in wo ea men g oups: Con ol and Tes
g oup. The wounding ea men in he igh wing is di e en in Con ol,
NS-like wound and Tes g oup, Ec7-like wound, which acco ding o
gene exp ession esul s (Figu e 3.3) do ac i a e di e en le els o
local and sys emic immuni y, bu he wound on he le side is simila
in bo h Con ol and Tes g oups. Thus, i he componen s o sys emic
immuni y we e essen ial o inc ease he WIE size, he WIE on he igh
wing and also in he le wing o Tes g oup a e la ge han WIE in
Con ol g oup. A e wounding, pupae we e e u ned o 27ºC un il
eclosion. Adul s we e sac i iced, wings collec ed, images we e
acqui ed and WIE we e measu ed as be o e (Figu e 3.1). We es ed
he e ec o T ea men (Con ol and Tes g oup), Wing side (Righ and
Le ) and he in e ac ion o T ea men *Wing side on he WIE size,
aking he wing size as a possible co a ia e, using a GLM wi h a
no mal dis ibu ion ( esiduals wi h no mal dis ibu ion es ed wi h
Shapi o es ): glm(WIE ~ T ea men * Wing Side + Wing Size). We
used he lsmeans R package o pos -hoc pai wise compa isons
(Len h & He é 2015) conside ing di e en g oups when p- alue<0.01.
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
68
Figu e 3.1- Expe imen al design. A) E ec s o immune sys em ac i a ion on gene
exp ession, su i al and WIE o ma ion. In a ime line we i s wounded (g een a ow)
and applied a d op o PBS (S and NS) o hea -killed E. coli solu ion (Ec6 and Ec7) ( o
a be e desc ip ion o ea men s see he main ex )), a e we collec ed samples o
gene exp ession analysis ( ed a ows) and he emaining indi iduals we e used o
analyze he ea men e ec s on su i al and WIE (a e eclosion). Fo gene
exp ession quan i ica ion we collec ed om he same indi idual he mos an e io and
dis al wing po ion o wounded and con ol wing (da k g een a ea), which was used o
RNA isola ion. B) E ec s o local and/o sys emic immune ac i a ion on WIE. Con ol
and Tes g oups we e wounded on bo h wings, he con ol g oup wi h NS-like wounds
on le and igh wings and Tes g oup wi h NS-like wound in he le and Ec7-like
wound on he igh wing. The e ec o sys emic immune sys em ac i a ion was
measu ed as he size o WIE on le wings o adul bu e lies. T ea men g oups a e
o de ed by he inc easing p obabili y o immune sys em ac i a ion om he ligh e o
he da ke box colo s. NW – non-wounded, S – s e ilized, NS – non-s e ilized, Ec6 –
106 cells/µL E. coli, Ec7 –107 cells/µL E. coli.
3.4. Resul s and Discussion
P e ious s udies ha e shown ha wounding bu e ly pupal wings
can induce he o ma ion o eyespo -like pa e ns (Nijhou 1991;
B ake ield & F ench 1995; O aki 2011). These obse a ions indica e
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
69
ha wound esponse induces cell a e changes in pupal wing epi helial
cells ha o he wise would de elop b ownish backg ound co e scales.
In his case, p ocesses in ol ed in wound esponse, such as immuni y
and issue epai migh impac eyespo o ma ion a ound wound si es.
B ake ield and F ench had epo ed ha he p obabili y o induce an
eyespo -like pa e n was highe upon a se e e epi helial damage, bu
he size o ha eyespo s was no a ec ed (B ake ield & F ench 1995).
He e we manipula ed he immuni y ac i a ion a he wound si e and
analyzed he eyespo -like pa e s in adul wings o in es iga e he ole
o immuni y on cell a e de e mina ion upon wounding. Non-s e ilized
condi ions (NS) ha e been used o induce he o ma ion o eyespo -
like pa e ns by p icking pupal wings (Nijhou 1991; B ake ield &
F ench 1995; O aki 2011), so we ook his wound ype as e e ence
and compa ed o he ea men s hough o induce highe and lowe
immune esponse. We show ha a ea men wi h hea -killed bac e ia
ac i a es inna e immune sys em in B. anynana (Figu e 3.2 and 3.3)
and ha ea men induces he de elopmen o la ge eyespo -like
pa e ns (Figu e 3.4C), bu was no a ec ing he p obabili y o i s
de elopmen (Figu e 3.4A). We also demons a ed ha he
enla gemen o hese pa e ns is ela ed wi h local, bu no sys emic
immune esponse (Figu e 3.4D).
3.4.1 The numbe o melano ic spo s and he mo ali y
inc ease upon ea men wi h hea -killed bac e ia
The e a e mul iple ways o ac i a e he immune sys em: wi h li e,
killed o cell wall componen s o mic oo ganisms, as well as wi h
mechanical damage (Lemai e e al. 1997; Rao & Yu 2010; Cla k e al.
2011). Insec s ha a e immune-challenged ac i a e he inna e
immuni y, and i he immune challenge is high i is possible de ec

Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
70
changes in hemolymph colo and he p esence o melano ic spo s wi h
a naked eye (La ine & S and 2002) (Figu e 3.2A). To induce di e en
le els o he immune sys em ac i a ion pos -wounding, we i s es ed
i a ea men wi h hea -killed E. coli was e ec i e. Immedia ely a e
wounding, we applied on op o he wound 0.5 µL ei he o s e ile
PBS1X, in he NS con ol g oup o o a solu ion wi h h ee di e en
concen a ions o hea -killed E. coli, 105, 106, and 107 cells/µL o he
Ec5, Ec6, and Ec7 g oups, espec i ely. Wi h a naked eye, i was
isible ha wings o he Ec7 ea men had mo e melano ic spo s
(Figu e 3.2A). The quan i ica ion e ealed indeed ha melano ic spo s
numbe was highe in ea men g oups s imula ed wi h highe
concen a ion o bac e ia (Ec6 and Ec7) (F=230, d. .=3, p-
alue<0.0001)(Figu e 3.2B). Ec5 g oup was no s a is ically di e en o
NS. The e o e, o inc ease he immune sys em ac i a ion a e
wounding we selec he wo highe hea -killed bac e ia ea men s,
Ec6 and Ec7.
This simple me hodology can be an al e na i e app oach o
quan i y he immune esponse on bu e ly pupal wings wi hou he
adi ional gene exp ession quan i ica ion, which equi es mo e ime,
human and economic esou ces. Quan i ying he melano ic spo s
h ough cu icle is also possible because we can see da ke a eas
unde he cu icle (Figu e S3.1). Since his me hodology does no
equi e pupal sac i ice, we can co ela e he immune esponse upon
wounding on pupal s age wi h adul pheno ype o each indi idual in a
popula ion, which migh allow, o example, a i icial selec ion on
in ensi y o immune esponse upon challenging.
The balance be ween immune sys em ac i a ion and inhibi ion is
i al o main ain homeos asis. Upon an immune h ea men , he
absence o an immune esponse can allow pa hogens o p oli e a e
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
71
uncon ollably inducing sep ic shock and e en ually kill he hos , while
an excessi e immune esponse can lead o anaphylac ic shocks wi h
deadly consequences o he hos (Cauwels 2007). In insec s also
occu simila scena ios o sep ic and anaphylac ic shocks. Fo
example, se e al s udies had demons a ed ha he numbe o
pa hogens inc eases o e ime in immune challenged indi iduals
(A ilano e al. 2011; Qui oz-Cas añeda e al. 2015; Singh e al. 2014),
and in silkwo m was demons a ed ha a e challenge wi h hea -killed
bac e ia o cell wall componen s he indi iduals induced an excessi e
ac i a ion o he inna e immuni y, which inc eased he hos mo ali y
isk (Ishii e al. 2010). In his s udy he au ho s demons a ed ha
ea men wi h melanogenesis inhibi o s inc eased he hos su i al,
which shows ha melanin pa hway o e -ac i a ion is ha m ul o he
hos (Ishii e al. 2010).
He e, we saw ha he Ec7 ea men induced highe mo ali y
du ing pupal s age (
χ
2 = 31.8, d. .=3, p- alue<0.001) (Figu e 3.2C).
This esul sugges s ha hea -killed bac e ia induced an exagge a e
immune esponse. We did no use li e bac e ia in o de o a oid he
sep ic shock-like scena io, bu no he anaphylac ic shock-like
scena io. Typically, Ec7 indi iduals had da ke hemolymph (obse ed
du ing dissec ions o qPCR sampling), u ning he pupa a da k g een
colo . In gene al, da k g een pupae e e o he no mal colo in one o
wo days, hose ha do no e e he colo usually e e each he
adul s age. All oge he hese obse a ions sugges ha , as in
silkwo m expe imen s (Ishii e al. 2010), he hea -killed bac e ia
ea men induced he o e -ac i a ion o melanogenesis. Some
indi iduals dec eased he melanin p oduc ion, es o ing he
homeos asis, bu some indi iduals main ained he melanin p oduc ion
longe inc easing he eac i e oxygen species (ROS) le els. High
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
72
le els o ROS a e necessa y o kill in ade pa hogens, bu also induce
apop osis in hos cells ha , i gene alized in whole body, de egula e
he in e nal homeos asis and lead o he hos dea h (Ishii e al. 2010;
Nappi & Ch is ensen 2005).
3.4.2 Hea -killed bac e ia ea men s inc ease local and
sys emically he exp ession le els o immune- ela ed genes
The immune sys em ac i a ion can be quan i ied h ough di e en
me hods, such as enzyma ic ac i i y, numbe o immune cells, and
gene exp ession (Neyen e al. 2014). He e, we quan i ied he
exp ession o immune- ela ed genes a e di e en wounding
ea men s. To induce highe le els o immune sys em we selec ed he
wo highe hea -killed bac e ia dosages es ed in p e ious expe imen
(Ec6 and Ec7) and o induce a lowe ac i a ion we s e ilized he cu icle
o wounded pupae o minimize possible na u al in ec ions. A e
wounding we collec ed wings a wo di e en ime poin s (4 and 8h) o
quan i y he ex en o local (wounded wing) and sys emic (con ol
wing) immune sys em ac i a ion in all ea men g oups (S, NS, Ec6
and Ec7) and compa e i wi h non-manipula ed indi iduals (NW) (see
expe imen al design Figu e 3.1). We measu ed he exp ession o
h ee immune ela ed genes, which we e o e exp essed a e
wounding a ound wound si es (see Chap e 2 - Figu e 3.3): Glo e in 2
(Glo 2), pale (ple), and p imo-1 (p i1). In gene al, he exp ession
le els inc eased a e wounding ela i e o NW indi iduals: a highe
hea -killed bac e ia dosage added o he esh wound induced highe
gene exp ession o immune ela ed genes in bo h wounded (Figu e
3.3A) and con ol wings (Figu e 3.3B) a 4 and 8h pos -wounding. This
analysis e ealed ha S and NS ea men s ac i a ed immuni y o
simila le els (Figu e 3.3A and B). Thus, he cu icle s e iliza ion does
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
73
no dec ease signi ican ly he immuni y ac i a ion upon wounding,
which indica es ha wounding pe se is ac i a ing immuni y pa hways.
I had been shown be o e ha he mechanical damage o p icking an
epi helium induces a basal ac i a ion o he immune sys em (Má kus
e al. 2005). This basal ac i a ion o an immune esponse migh be
su icien o deal wi h po en ial oppo unis ic pa hogen in asions a
wound si es. In con as , he Ec7 ea men was s a is ically di e en
om NS ea men (excep o he exp ession o p i1 a 4h s and ple a
8h s) leading o highe up- egula ion o immuni y- ela ed genes, bo h
locally in he wounded wing, as well as in he con a-la e al con ol
wing (Figu e 3.3 A).
The gene exp ession dynamics o Glo 2 and p i1 was simila :
local, exp ession le els we e al eady inc eased a 4h and a e s ill high
a 8h pos -wounding. Sys emic exp ession le els a e simila , bu
highe a 4h han a 8h. The ple exp ession dynamics is sligh ly
di e en : a 8h local exp ession le els a e al eady dec easing and
sys emic exp ession le els a e al eady in basal le els (Figu e 3.3A
and B). This indica es ha in B. anynana du ing ea ly pupal s age he
local esponse is longe han sys emic esponse and he sys emic
esponse migh ha e wo phases: i s phase wi h ac i a ion o genes
ela ed wi h melanogenesis, and he second phase wi h ac i a ion o
genes ela ed wi h AMP and o he pep ides/enzymes p oduc ion.
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
80
known as co e scales because hey a e co e ing ano he scales
laye , g ound scales, which a e ligh e in colo and smalle han co e
scales (Nijhou 1991). F equen ly, a he wound si e he e is no
de elopmen o colo ed co e scales, only g ound scales do de elop,
he sca a ea (Figu e 3.4B, solid line). Thus, du ing he wound healing
p ocess wing cells a wound si e lose hei abili y o de elop co e
scales o swi ch hei de elopmen al p og am o de elop only g ound
scales ins ead. This pheno ype was also obse ed by B ake ield and
F ench, hey obse ed ha a e se e e wounding he sca a ea was
la ge han in mild wounding (B ake ield & F ench 1995). In ou
expe imen , in ec ed-like wounds also de eloped la ge sca s han NS
wounds (Figu e 3.4E). This implica es immune sys em and/o issue
epai ela ed pa hways impac he ype o scales ha de elop upon
wounding. In B. anynana, a ound 40 hou s pos -pupa ion ac in- ich
s uc u es s a o be sec e ed in scale building cells, which a e he
u u e wing scales (Supplemen a y Figu e S3.2). Scale p oduc ion
equi es high quan i ies o ac in (Dinwiddie e al. 2014). A e
wounding an ac in ich cable is o med, which also equi es high le els
o ac in and is c i ical o gene a e a pulling o ce ha con ac s he
wound edges o close he wound (Wood e al. 2002; Bosch e al.
2005; K au z e al. 2014). The ac in equi emen s o wound healing
(12-20h pos -pupa ion) migh dec ease he a ailable ac in le els o
scale de elopmen (36-48h pos -wounding) and igge ing he
de elopmen o small g ound ins ead o la ge co e scales a he
wound si e.
In se e e wounds his hypo hesis can be easily explained: he
wound is la ge and o close he wound i is necessa y mo e ime and
high ac in le els o o m he ac in cable, so he ac in a ailabili y o
build scales will be less and sca a ea la ge . In in ec ed-like wounds
we do no expec a la ge ac in cable since he hole has he same size

Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
81
as in NS wounds. Howe e , he in lamma o y esponse is longe
(Figu e 3.3) and can delay he healing p ocess (Diegelmann & E ans
2004), which migh ex end in ime he p esence o he ac in cable a
wound si e. Ano he hypo hesis o explain he la ge sca s in in ec ed-
like wounds is he wing cellula dead a wound-si e, which can be
induced a e melanogenesis o e -ac i a ion (Nappi & Ch is ensen
2005). Melanogenesis pa hway is necessa y du ing wound healing
and also o neu alize pa hogens (Binggeli e al. 2014; Nam e al.
2012). Du ing melanin syn hesis he e is p oduc ion o cy o oxic
byp oduc s, which kill in ade cells, bu hey a e also ha m ul o hos
cells (Nappi & Ch is ensen 2005; Ishii e al. 2010). We saw ha 1)
in ec ed-like wounds had la ge melaniza ion a eas a wound si e
(Figu e 3.2A), 2) ple exp ession, a melanogenesis pa hway gene, was
eigh imes highe han in NS (NS=3 and Ec7=26, old change ela i e
o NW a ou hou s pos -wounding) (Figu e 3.3A), and 3) high
exp ession le els o ple a e main ained in wounded wing a leas
du ing mo e ou hou s (un il eigh hou s pos -wounding) (Figu e 3.3A).
In in ec ed-like wounds melanogenesis o e -ac i a ion migh induce
hos cell dea h inc easing he opening size, and consequen ly highe
ac in le els a e equi ed and also he wound healing p ocess is
delayed. This and p e ious expe imen s by B ake ield & F ench
(B ake ield & F ench 1995) indica e ha inhibi ion o co e scale
de elopmen migh be ela ed wi h he ime necessa y o es o e he
wing epi helium, which migh be p opo ional o he numbe o
damaged cells, ac ing equi emen s and o he ex en o in lamma o y
esponse.
In ou expe imen s we did no explo e in de ail he scale
composi ion o he sca a ea, so u u e expe imen s a e necessa y o
unde s and whe he in sca a ea we do ha e less and smalle scales
o same numbe o scales bu smalle . To quan i y i i ’s necessa y o
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
82
peal he wing scales and coun he inse ions (socke cells) o he
scales in he wing.
Figu e 3.4- Local immune sys em ac i a ion induces la ge , bu no mo e WIE. A)
P obabili y o WIE de elopmen . The immune sys em manipula ion did no a ec ed
he likelihood o WIE de elopmen ; (χ2 = 1.02, d. .=3, p- alue >0.05. B) Sec ion o
an e io dis al do sal wing wi h an e io na i e eyespo and he WIE. We measu ed
h ee ai s o WIE: o al a ea (dashed line), sca a ea ( ull line) and co e scales a ea
(space be ween dashed and ull line). (*) wound si e; scale ba = 1mm. C)To al size o
WIE. Highe bac e ial dosage induces la ge WIE; F3,162=15.4, d. .=162, p- alue
<0.001. D) Local, bu no sys emic, ac i a ion o immune sys em leads o enla ged
WIE. The Tes g oup p oduced la ge WIE on igh wing, bu no on le wing;
F1,107(G oup*Wing Side)=9.57, p- alue <0.003, Con ol: N=25; Tes : N=31. E) G ound
scales a ea. High bac e ial dosages inhibi he de elopmen o co e scales;
F3,162=18.0, d. .=162, p- alue <0.001. F) Co e scales a ea. Highe bac e ial dosage
ans o med mo e backg ound b ownish scales in o black o golden scales; F3,162=
10.8, d. .=162, p- alue<0.001. Each do ep esen s he a ea o a biological eplica e
and iangles a e he mean o each ea men , he ho izon al line in he boxplo is he
median, boxplo limi s a e he uppe and lowe qua iles and e o ba s a e he
maximum (uppe ) and minimum (lowe ) alues excluding ou lie s. Di e en le e s
indica e ea men pai wise compa isons s a is ically di e en (lsmeans p- alue<0.05).
The sample size in C), E) and F) we e he indi iduals p oducing WIE on A).
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
83
3.4.6 Reepi helializa ion and immuni y ela ed-genes migh
ha e dis inc oles in WIE de elopmen
Wounding an ea ly pupal de eloping wing a ec s he cell a e o
wounded and su ounding wing cells in wo ways: 1) scale- ype
de elopmen (g ound o co e scales in ocus-like cells), and 2)
pigmen - ype p oduc ion (black o golden ins ead a ound wound si e).
He e we quan i ied he a ea o 1) non-backg ound scales a ound
wound si e, o al WIE a ea (Figu e 3.4C), 2) he sca a ea, only g ound
scales (Figu e 3.4E), and he black and golden co e scales a ea, he
di e ence be ween o al and sca a ea (Figu e 3.4F). All h ee ai s
we e la ge in g oups wi h s onge ac i a ion o immuni y (Ec6 and
Ec7 g oups) (Figu e 3.4C, E and F). These esul s indica e ha he
immuni y-associa ed gene ic pa hways impac bo h cell a e decisions:
scale- ype de elopmen and pigmen - ype p oduc ion. As e e ed
be o e, se e e wounds wi hou addi ional immune challenge a ec ed
he p obabili y o eyespo -like pa e n and scale- ype de elopmen , bu
no he o al numbe o cells wi h di e en pigmen a ion (B ake ield &
F ench 1995). The e o e, ac o s inc easing he sca a ea a e
equi alen ly ac i a ed in se e e and in ec -like wounds, bu ac o s
de e mining o ma ion o no o ma ion o an eyespo like pa e n and
he o al numbe o cells changing he cell a e a e s onge in se e e
wounds and in in ec ed-like wounds, espec i ely.
The o ma ion o an eyespo is igh ly co ela ed wi h he
exp ession o se e al ansc ip ional ac o s du ing la e la al and
ea ly pupal de elopmen (B ake ield e al. 1996; B une i e al. 2001;
Saenko e al. 2011). An ennapedia (An p) in one o hese ansc ip ion
ac o s and we saw in p e ious chap e ha some cells we e
exp essing i a he wound edge 4h pos -wounding. The exp ession o
An p a wound si e can be co ela ed wi h he se e i y o wounding
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
84
and wi h he numbe o cells damaged and also wi h he cell a e
decision o de elop o no an eyespo -like pa e n. To es his
hypo hesis i is necessa y quan i y he numbe o cells exp essing
An p a wound si e in mild and se e e wounds.
Se e al s udies in D osophila ha e shown ha se e e epi helial
damage ac i a e highe le els o melanin pa hway enzymes ela i e o
mild inju ies (Nam e al. 2012; Binggeli e al. 2014). O he s udy had
demons a ed ha a e inju y Dpp, a TG -β signal, ac s as an an i-
in lamma o y o dec eases he exp ession o AMPs, his down-
egula ion o immune sys em is pa icula ly s ong in s e ile inju ies
(Cla k e al. 2011). dawdle (daw), ano he TG -β signal, also inhibi he
immune esponse, ep essing he melanogenesis pa hway, specially
a e in ec ion (Cla k e al. 2011). Thus, mild wounds educe AMPs
exp ession and in ec ed-like wounds ep ess he o e -ac i a ion o he
melanogenesis pa hway. Se e e wounds p esumably s imula e highe
le els o melanogenessis han mild wounds, bu AMPs le els migh be
simila , o only sligh ly high, because in absence o in ec ion Dpp
ep esses he AMP p oduc ion. In ec ed-like wounds induced la ge
WIE and se e e wounds did no , bo h wound ypes can ac i a e high
le els o issue epai mechanisms (explained abo e) and
melanogenesis, bu only in ec ed-like wounds can induce high le els
o AMPs. In eyespo de elopmen models a mo phogen-like signal
egula es he eyespo size; highe le els p oduce la ge eyespo s
(F ench & B ake ield 1992; Nijhou 1991). The e o e, mo phogen-like
signals egula ing he p oduc ion o AMPs migh be good candida es
o be he mo phogen-like signal p oduced a he eyespo cen e ha
con ol he pigmen - ype de elopmen . Fu he mo e, he cy o oxic
e ec s o melanogenesis, which migh be co ela ed wi h he ime
equi ed o es o e o epi helium, egula e he scale- ype de elopmen .
Toge he , i sugges s ha di e en p ocesses o wound healing
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
85
esponse a e a ec ing di e en aspec s o eyespo like pa e ns
de elopmen . We p opose ha he numbe o damaged cells a
wounding con ols he es ablishmen o an eyespo -like o ganizing
cen e (Figu e 3.5B); he du a ion o issue eepi helializa ion a ec s
he scale- ype de elopmen (Figu e 3.5C) and he le el o immuni y
ac i a ion a ec s he pigmen - ype p oduc ion (Figu e 3.5D).
Figu e 3.5- Possible models o mo phogen-like molecule(s) p oduc ion upon di e en
wound ypes and cell a e de e mina ion on ea ly pupal wing cells. A) Model o
de elopmen o WIP a e mild, se e e and in ec ed wounds. Black cu e ep esen s
he concen a ion o mo phogen-like and o he lines ep esen he h eshold-like
esponse in wing epi helium. B-D) cell a e decisions upon wounding o es ablish an
eyespo -like cen e (B), he scale- ype (C) and he pigmen - ype (D) migh ely on
numbe o damaged cells, du a ion o epi helium eepi helializa ion and immuni y le el,
espec i ely.

Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
86
3.5. Conclusions
Fo simplici y, biologis s always y o compa men alize gene ic
ne wo ks, o example, as de elopmen al o immune ne wo ks.
Howe e , he cases o blu ing on ie s be ween bo h a e qui e
common. Fo example he Toll pa hway, which was i s desc ibed as
a de elopmen al pa hway essen ial o es ablish he do so en al axis
in in e eb a es, was la ely desc ibed o be also essen ial in immune
p ocesses (Valanne e al. 2011).
He e, we s udied ano he case o blu ing on ie s, be ween
eyespo de elopmen and wound esponse, whe e he gene ic
ci cui y desc ibed as wound- ela ed migh be essen ial du ing
bu e ly eyespo de elopmen . Ou and also p e ious esul s indica e
ha he numbe o wounded cells migh es ablish an eyespo -like
cen e ; he du a ion o eepi hilializa ion p ocess migh a ec he ype
o scales p oduced; and he le el o immuni y ac i a ion a ec s he
colo o co e scales. Du ing e olu ion Lepidop e ans migh ha e co-
op ed he ini ial wound signaling o es ablish eyespo -o ganizing
cen e s, he issue epai gene ic ci cui y o con ol he scale
mo phology; and immuni y gene ic ci cui y o egula e eyespo size.
Fine- une adjus men s o hese gene ne wo ks migh ha e been
c ucial o di e en ia ion o wing cells h ough e olu ion and con ibu e
o di e si ica ion o eyespo -like pa e ns in e ms o i s size, colo ,
numbe , posi ion, numbe o ings and also scale mo phology.
3.6. Acknowledgmen s
We hank An ónia Mon ei o o he B. anynana lab s ock and
Isabel Go do o he E. coli bac e ial s ain used in hese expe imen s,
EVO-DEVO IGC communi y o he use ul discussions, Luis Teixei a
o he sugges ions in expe imen al design o es he e ec s o local s.
Chap e 3
_____________________________________________________________________________________________________________________________________________________________________________________________________________
87
sys emic e ec o immune sys em and Da id Duneau and Jo ge
Ca nei o o he sugges ions o analyze he qPCR.
88
Chap e 4 – Immuni y up- egula ion
in pupae phenocopies e ec s o
low empe a u e on de elopmen
ime, wing pa e ns, and pupal
ecdysone le els
4.1. Summa y
In seasonal en i onmen s, luc ua ions a e p edic able and
mul icellula o ganisms wi h seasonal polyphenism can cope be e .
Seasonal polymo phic o ganisms sense abio ic and bio ic cues
associa ed wi h a speci ic en i onmen and change hei de elopmen
o p oduce he mos adjus ed pheno ype. Changes in body
pigmen a ion a e common esponses in his de elopmen al p ocess
and usually hey a e egula ed h ough changes in ho monal le els. In
insec s, pigmen a ion is igh ly associa ed wi h he en i onmen al
empe a u e and in e nal immune s a e sensed du ing de elopmen .
He e, we used a bu e ly model sys em (Bicyclus anynna) o explo e
he e ec s on pigmen a ion a e manipula ion o in e nal immune
s a e on a p e-adul s age. We saw ha immuni y phenocopied
pa ially he e ec s o low de elopmen al empe a u e and ha his
e ec was independen o he immune challenge issue/o gan and
bac e ial- ype. High sys emic immuni y induced 1) longe pupal
de elopmen al ime, and educed 2) eyespo size, 3) o e all wing
da kness, 4) wing colo con as and 5) pupal 20-hyd oxyecdysone
(20E) le els. These esul s demons a ed ha immuni y also induce
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
89
pheno ypic plas ici y in B. anynana. Fu u e expe imen s explo ing he
in e ac ion be ween pigmen a ion, empe a u e and immuni y, and i s
egula ion h ough ho mones migh shed ligh in o he di e si ica ion o
bu e ly wing pa e ns.
4.2. In oduc ion
Many mul icellula o ganisms ha e mul iple li e cycles du ing one-
yea leng h and di e en gene a ions ha e o su i e in comple ely
di e en en i onmen al condi ions. En i onmen al luc ua ions can
a ec d as ically he a ailabili y o di e en nu ien s, he p esence o
de imen al subs ances o o he ex e nal abio ic and bio ic ac o s like
empe a u e and pa hogens (Beldade e al. 2011). Howe e ,
mul icellula o ganisms ha e e ol ed mechanisms o cope wi h his
en i onmen al a ia ion in o de o su i e and assu e pheno ypically
obus de elopmen al ou come (B aendle & Félix 2009). Thus, he
ex e nal en i onmen can a ec de elopmen and lead o he
p oduc ion o dis inc pheno ypes om he same geno ype, his
mechanism is called pheno ypic plas ici y (Pigliucci 2001; Beldade e
al. 2011). Seasonal polyphenism, in which di e en o ms o a species
a e p oduced a di e en imes o he yea , is a common o m o
pheno ypic plas ici y among insec s (Hazel 2002). In seasonal habi a s,
pheno ypic plas ici y may e ol e as a esul o con as ing bu
p edic able seasonal selec ion p essu es, esul ing in di e en
pheno ypes being exp essed in each season (Roskam & B ake ield
1999; Oos a e al. 2014). The en i onmen al cues inducing seasonal
polyphenism can be e y di e se. Fo example, abio ic cues:
empe a u e, pho ope iod, and salini y; and bio ic cues: nu i ion,
c owding, p esence o p eda o s and pa hogens (Beldade e al. 2011;
Poina & Yano iak 2008).
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
96
To alida e hese wo me hodologies, we measu ed he o e all
da kness and colo con as in non-wounded indi iduals ea ed a 19C
and 27C. The low de elopmen al empe a u e pheno ype is
conside ed o be da ke and wi h less colo con as (Beldade &
B ake ield 2002). We obse ed he same pa e n o hese wo ai s
using hese new me hodologies (see Figu e 4.6A).
Da a we e impo ed o R and we used ANCOVA aking he o al
numbe o pixels and he ansec size as co a ian es o o e all
da kness and colo con as , espec i ely. To see he e ec s o
empe a u e on hese wo ai s we used he ollowing models:
lm(o e all da kness ~ empe a u e + o al numbe o pixels) and
lm( colo con as ~ empe a u e + ansec size). The e ec s o
bac e ia ype and dosage we e accessed h ough hese models:
lm(o e all da kness ~ bac e ia * dosage + o al numbe o pixels) and
lm(colo con as ~ bac e ia * dosage + ansec size). Residuals we e
no mal (Shapi o es ) and a iance was cons an (B own-Fo sy h es ),
so we used ANCOVA wi hou da a ans o ma ion. We conside ed
signi ican e ec s when p- alue<0.01, and we used he lsmeans R
package o pos -hoc pai wise compa isons (Len h & He é 2015)
conside ing di e en g oups when p- alue<0.05.
4.3.5 Quan i ica ion o 20E le els
We measu ed in e nal le els o 20E in immune challenged
(con ol and ea men s wi h 107 hea killed E. coli o S. au eus as
explained abo e) and non-manipula ed pupae ea ed a 27°C and
19°C a 8.5, 12 and 24% o o al pupal de elopmen al ime (Figu e
4.2C ed squa e). We ex ac ed hemolymph by smashing he pupae
wi h a pes le in an emp y 1.5mL eppendo ube p e-wa med o 60°C.
The ube was immedia ely placed a 60°C o 20 min o inac i a e he

Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
97
hemolymph enzymes and a oid melaniza ion p ocesses (Geo ge &
C is o alo 1972). Tubes we e a e placed on ice and hen cen i uged
(10 min a 4C) a ull speed o emo e cell deb is. The supe na an
was collec ed in o a new cold ube and 300µL o ice-cold 100%
me hanol we e added o p o ein p ecipi a ion. Samples we e again
cen i uged and he supe na an placed in a new cold ube. This
p ocess was epea ed un il no palle was deposi ed upon
cen i uga ion ( ypically, h ee imes) and he samples we e s o ed a -
80°C un il u he p ocessing. Be o e ho mone quan i ica ion wi h EIA
enzyme immunoassay (Cayman Chemical Co., Ann A bo , MI, USA),
he me hanol was elimina ed om he samples by using a o a y
e apo a o , a maximum speed (RT). Du ing his p ocess we checked
he samples e e y 30 min o emo e he d ied ones, ypically he
me hanol e apo a ion equi ed 1h ± 30 min. D ied samples we e hen
dissol ed in assay bu e . Acco ding o he manu ac u e 's ins uc ions,
he quan i ica ion o 20E le els equi es con ols (blanks) and s anda d
calib a o samples, o gene a e a calib a ion cu e. S anda d calib a o
samples we e gene a ed using se ial dilu ions o comme cially
a ailable 20E (SciTech chemicals, e : S5314-001; 0.5 µg/µL in 100%
e hanol) (six on Pla e 1 and eigh on Pla e 2). The e hanol om
s anda ds was elimina ed as me hanol in hemolymph samples, and
a e d ied s anda ds was dissol ed also in assay bu e . Abso bance
o con ols, s anda ds, and hemolymph samples was measu ed in
iplica ed a a wa eleng h o 405 nm (VICTOR Mul ile el Pla e
Reade spec opho ome e ). We calcula ed he a e age o
abso bance o con ols, s anda d and hem lymph samples. Acco ding
o manu ac u e ins uc ions, we emo ed he blanks abso bance o all
samples and used he s anda d samples o do a calib a ion cu e
(Figu e S4.2), which was used o calcula e he 20E concen a ion in
hem lymph samples.
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
98
Da a we e impo ed o R and we used an ANOVA s a is ical es
o see he e ec s o immune challenge on 20E le els, which we e
no malized o he minimum 20E le els in each pla e. We used he
ollowing model: a o(20E le els ~ T ea men * de elopmen al ime +
pla e) and he leans R package o pos -hoc pai wise compa isons
(Len h & He é 2015) conside ing di e en g oups when p- alue<0.05.
4.3.6 Manipula ion o 20E le els
To he immune-challenged pupae (c . me hods desc ibed abo e),
we also applied 0.5 µl o a solu ion o 20E in PBS. We i s es ed a
dosage ha had p e iously been used success ully o pa ly escue
e ec s o low de elopmen al empe a u e: 250ρg injec ed in o he
abdomen o pupa a 8% o hei pupal li e du a ion (Ma eus e al.
2014). In ou expe imen , applica ion o 250 pg o wing o ho ax
wounds lead o 100 % pupal mo ali y. We hen es ed a se ies o
lowe doses: 250, 125, 62.5 and 31.25ρg o 20E pe pupa. 62.5ρg
was he highes dose ha did no lead o 100% mo ali y. To es i his
dosage could es o e he pheno ype o high de elopmen al
empe a u e, we measu ed h ee wing pa e n ai s (size o hindwing
i h eyespo , o e all da kness and colo con as ) in ee di e en
ea men s 1) wound wi h PBS (PBS), 2) solu ion wi h 107 hea -killed E.
coli (Ec7), and 3) same bac e ial solu ion supplemen ed wi h ho mone
(HOR) (Figu e 4.2C g ay squa e). The ho mone s ock solu ion was
dissol ed in e hanol, so we added o PBS and bac e ial solu ion he
same olume o e hanol. T ea ed pupae we e le o comple e hei
de elopmen a 27°C, adul wings we e emo ed, and wing pa e n
ai s we e measu ed and analyzed o as explained abo e.
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
99
Figu e 4.2- Expe imen al design o unde s and he A) ea men posi ion e ec s; B)
bac e ia ype and dosage e ec s and C) o measu ed and manipula e 20E le els. Top
a ow ep esen s de elopmen al ime om pupa ion (P) o eclosion (E) wi h ea men
(T) and sampling (S) ime poin s. In hese expe imen s we ea ed he indi iduals
always a 12hou s pos -pupa ion. To analyze ea men posi ion e ec s (A) we ea ed
indi iduals on wings o on ho ax (posi ion is indica ed wi h (*) on pupae scheme) wi h
con ol (0.5µL o s e ile PBS – PBS) o immune challenge (0.5µL o 107 E. coli
solu ion – Ec7) solu ions and measu ed he pupal de elopmen al ime and six
eyespo s (wi h ou eyespo ai s each) on bo h sides (L-le and R-Righ ). We
e alua ed he e ec s o bac e ia ype and dosage (B) a e ea men on ho ax. We
used h ee di e en bac e ia ypes: E. coli (Ec – blue); S. au eus (Sa – yellow) and a
balanced mix o bo h bac e ia solu ions (ES – g een); in h ee di e en dosages:
applica ion o 0.5 µL o 105, 106 and 107 bac e ia cells solu ions (da ke colo s
ep esen highe dosages). Each ea men g oup is iden i ied wi h a combina ion o
he bac e ia ype and he dosage; e.g. ea men wi h 0.5µL o 105 E. coli solu ion –
Ec5. We analyzed he e ec s on pupal de elopmen al ime, on i e eyespo s (wi h
ou eyespo ai s each) on he igh side, on hindwing colo s (o e all da kness and
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
100
colo con as ) and on numbe o do sal hindwing whi e pupils. 20E le els we e
quan i ied in non-wounded indi iduals ea ed a 19°C and 27°C and in ho ax- ea ed
indi iduals wi h PBS (con ol g oup) and Ec7 and Sa7 ( eddish squa es). E ec s o
20E manipula ions we e analyzed on i h eyespo o he hindwing and also in colo
ai s (o e all da ckness and colo con as ). FwD-Fo ewing do sal su ace; FwV-
Fo ewing Ven al su ace; HwV-Hindwing Ven al su ace.
4.4 Resul s and Discussion
The he mal de elopmen al plas ici y in B. anynana is well-known
(B ake ield e al. 1998; Oos a e al. 2014; Ma eus e al. 2014).
Howe e , o ou knowledge, o he en i onmen al cues inducing
de elopmen al plas ici y in his species a e no desc ibed. Ou esul s
showed ha immune challenged ea ly pupae phenocopy he low
de elopmen al empe a u e e ec in ou di e en aspec s: 1) longe
pupal de elopmen al ime (Figu e 4.3), 2) educed eyespo size
(Figu e 4.1, 4.4, 4.5), 3) educed colo con as (Figu e 4.6B2) and 4)
educed le els o 20E (Figu e 4.7). These e ec s we e independen o
he immune-challenged o gan (he e only es ed wing and ho ax)
(Figu e 4.4), and despi e only G am- Bac e ia induced an ex ended
esponse, G am+ Bac e ia had simila e ec s on wing pa e ns (Figu e
4.5) and 20E le els (Figu e 4.7)
4.4.1 High bac e ial dosage ea men educes he numbe o
“ as -li ing” indi iduals
The p esence o pa asi es and pa hogens cause subs an ial
i ness cos s o hei hos s. Thus, hos s ha e e ol ed e ec i e
immune sys ems, which a e ene ge ically expensi e o main ain and
o ac i a e (Saas amoinen & Ran ala 2013). T ade-o s be ween
immune de ense and o he li e-his o y ai s a e common
(Saas amoinen & Ran ala 2013; Val onen e al. 2010). Fo example,
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
101
indi idual’s nu i ional esou ces and body condi ion can in luence
immune in es men di ec ly, in his case indi iduals wi h ewe
esou ces being able o alloca e less o immuni y (Val onen e al.
2010). He e, we in es iga ed he e ec s o in e nal immune s a e on
pupal su i al and de elopmen al ime. We conside ed he e ec s o
1) he immune challenge posi ion (wing and ho ax), 2) he bac e ia
ype (G am-:Ec, G am+:Sa and bo h G am+ and G am- oge he : ES),
and 3) he bac e ia dosage (105,106,107 bac e ial cells)(Figu e 4.2).
We quan i ied he numbe o pupae ha eached he adul s age in
each day and se e al imes du ing a day (each ~20min a e ligh
ime). Fo same age pupae, eclosion occu ed du ing wo days
(classi ied as as -de elopmen (FD) ( i s day) and slow-de elopmen
(SD)(second day). Pupae ha did no eclosed du ing his ime window
we e dead (classi ied as non-adul (!)). Among eclosed pupae he e
we e indi iduals ha s a ed he me amo phosis p ocess bu could no
comple ed i , and o he s ha could no s e ch co ec ly he wings.
Those indi iduals we e also classi ied as non-adul s. Rega dless he
wounding posi ion (Figu e 4.3A; Eclosion Day*T ea men : χ2 = 43.129,
d. .=2, p- alue<0.0001) and bac e ia ype (Figu e 4.3C; Eclosion
Day*T ea men : χ2 = 20.63, d. .=6, p- alue=0.002), high bac e ial
challenged pupae eclosed p e e en ially on he second day. Only e y
ew indi iduals om highe bac e ia dosage g oups we e FD (Figu e
4.3A and C). The p opo ion o SD was no di e en in compa ison
wi h con ol indi iduals (Figu e 4.3A and C). I mo ali y among FD and
SD was equi alen and bo h had delayed he de elopmen al ime
(acco ding o classic ade-o heo y (Ande son & May 1982)), we
should obse e pupae eclosing on a hi d day, which did no occu ed.
Thus, ou esul s a e consequence o highe mo ali y among FD and
same de elopmen al ime o SD o inc eased de elopmen al ime o
FD and highe mo ali y o SD..

Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
102
The pace o li e heo y p edic s ha “ as -li ing” indi iduals ha e
high ene gy u no e , sho de elopmen al ime and in es ela i ely
li le in immune de ense in a o o g ow h and ea ly ep oduc ion
(Niemela e al. 2013; Reznick e al. 2002). On he o he hand,,“slow-
li ing” indi iduals ha e lowe ene gy u no e , long de elopmen al
ime and in es mo e esou ces in o cos ly immune de ense (Niemela
e al. 2013; Reznick e al. 2002). Acco ding o his heo y, he “slow-
li ing” animals use mo e slowly-deployed while “ as -li ing” use mo e
apidly-deployed immune p ocess (P e i ali e al. 2012).
Melanogenesis is a as immune esponse ha occu s in a h opods
a e a wounding o immune challenge i is essen ial o wound healing
and con ol o in ade s (Ele he ianos & Re enis 2011; Binggeli e al.
2014). Ne e heless, i o e ac i a ed i can induce he hos dea h
(Ishii e al. 2010). Du ing he expe imen we obse ed ha a e high
bac e ial dosage ea men , he e was an in ense da kening a wound
si e and pupae we e da ke , especially du ing he ea men day. Two
days a e ea men , he e we e pupae ha had al eady eco e ed he
o iginal g een colo and o he s ha we e s ill da ke . Usually, hese
da ke pupae (s ill ali e by he hi d day pos -pupa ion) ne e eached
he adul s age. Thus, a e bac e ial ea men “ as -li ing” pupae
migh ha e o e ac i a ed melanogenesis p oducing highe le els o
cy o oxic by-p oduc s, which induced de elopmen al de ec s and
consequen ly he pupae dea h.
The e we e mo e pupae eclosing on he i s day in wing- han in
ho ax-wounded indi iduals in PBS ea men , o Ec7 indi iduals he
di e ences we e no signi ican , bu s ill he p opo ion o i s day
eclosions was lowe (Figu e 4.3B; Posi ion: χ2 =97.57, d. .=1, p-
alue<0.0001). In D osophila i was shown ha mo ali y is highe a e
immune challenge (in asep ic condi ions and wi h in ec ion) on ho ax
han in abdomen (Chambe s e al. 2014). This indica es ha he
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
103
wound healing p ocess is di e en in di e en o gans, and can esul
in di e ences in su i al (Chambe s e al. 2014) and pe haps in
de elopmen al ime.
We saw ha ho ax-wounded indi iduals de eloped one day
as e ( ou o i e days) han wing-wounded indi iduals ( i e o six
days)(Figu e 4.3B). This expe imen was done in wo di e en phases:
i s we un he expe imen wounding indi iduals on wings (du ing
win e ) and second on ho ax (summe ). We saw ha non-wounded
indi iduals had simila de elopmen al ime as he PBS ea ed
indi iduals (pe sonal obse a ions) in bo h expe imen al ba ches.
Thus, i is no he wounding posi ion ha is inducing such
de elopmen al di e ences. This migh be a consequence o bu e ly
eeding sys em in he lab. Bu e lies we e ed on young maize plan s
g owing in a g eenhouse, which a he ime had no e y s ic
empe a u e con ol. In summe maize plan s g ew as e han in
win e and he nu ien composi ion on plan lea es migh be sligh ly
di e en . This hypo hesis was no es ed in ou lab ye , bu i was
shown in o he Lepidop e a species ha di e en g own condi ions o
hos -plan s can g ea ly in luence de elopmen al ime (Baue eind &
Fische 2013).
T ea men s wi h di e en bac e ia- ype did no induced
di e ences on eclosion day (Figu e 4.3D; Eclosion Day*Bac e ia Type:
χ2 =0.44, d. .=2, p- alue= 0.8020). Howe e , highe bac e ial dosages
g oups eclosed p e e en ially on he second day (Figu e 4.3D;
Eclosion Day* Dosage: χ2 = 43.57, d. .=3, p- alue< 0.0001). The
lowes bac e ial dosage did no induced di e ences ela i e o PBS
ea men g oup and he in e media ed dosage only induced
di e ences wi h G am- bac e ia (E. coli, Ec). This migh indica e ha
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
104
G am- Bac e ia a e mo e e icien o o e ac i a e he immune sys em
in his bu e ly,
Figu e 4.3- E ec s on su i al and pupal de elopmen al ime a e di e en immune
challenges. The mo ali y and de elopmen al de ec s we e inc eased in immune-
challenged indi iduals independen o he ea men posi ion (A) and bac e ia ype (B),
bu hey we e signi ican only in highe dosages. Among he indi iduals ha had
eached adul s age, he o al de elopmen al ime was di e en in immune-challenged
indi iduals. This e ec was independen o he ea men posi ion (C) and bac e ia
ype (D), and i was s a ically signi ican in highe bac e ial dosages (107 bac e ia cells
in all bac e ia ypes and also in106 o Ec). p- alue <0.001***, p- alue <0.01**, p-
alue<0.05*; sample sizes (N) a e indica ed o each ea men g oup in each
expe imen ; ea men posi ion is ep esen ed by an as e isk (*) on pupae scheme; Ec:
E. coli; Sa: S. au eus; ES: balanced mix o bo h bac e ia (Ec and Sa) solu ions;
ea men wi h 0.5 µL o 105, 106 and 107 bac e ia cells solu ions a e ep esen ed as 5,
6 and 7 numbe s a e bac e ia ype, wi h da ke colo s ep esen ing highe dosages.
4.4.2 Sys emic immuni y is a de elopmen al plas ici y igge
A e Chap e ’s 3 expe imen s, we no iced ha indi iduals ea ed
wi h Bac e ia on do sal wings had changed hei en al wing pa e ns
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
105
(Figu e 4.1). Wing pa e ns we e simila o hose o low de elopmen al
empe a u e. (Figu e 4.1). To unde s and i hese pheno ypic e ec s
we e a wing-speci ic esponse o a gene al esponse o in e nal
immuni y s a e, we induced wounds on ho ax and compa e
pheno ypes be ween wing- and ho ax- ea ed g oups (Figu e 4.2A).
We hypo hesized ha e ec s a e wing- ea men we e wing-speci ic
i a e ho ax- ea men wi h same s imuli, he e we e no pheno ypic
di e ences be ween con ol (PBS) and ea men (Ec7) g oups. Fo
his compa ison we measu ed he a ea o colo ings in six eyespo s
pe side (le and igh ) ob aining ou eyespo ai s pe eyespo :
golden, black, and whi e a eas and o al eyespo a ea (sum o golden,
black and whi e a eas) (Figu e 4.2). Resul s a e show in Figu e 4.4
and a e ep esen ed as he pe cen age o di e ence ela i e o PBS.
The e a e signi ican ly di e en esponses ( ed ci cles in Figu e 4.4;
do plo s g aphics in Supplemen a y Figu e S4.3), in which he size o
eyespo ai s we e smalle , in wing- and in ho ax- ea ed g oups.
Thus, educ ion o na i e eyespo size is a esul o di e ences in
in e nal immune s a e and can be induced in di e en o gans o he
han wings.
P e ious s udies ha e shown ha he eyespo cen e p oduces
one (o mo e) mo phogens ha di use o he su ounding cells which,
depending on he mo phogen concen a ion hey expe ience, become
a ed o p oducing di e en colo s (B une i e al. 2001; F ench &
B ake ield 1992). The cen e s o la ge eyespo s p esumably p oduce
highe quan i ies o mo phogen ha sp ead h ough mo e cells
(Mon ei o e al. 1997; Pa ícia Beldade e al. 2008; Beldade &
B ake ield 2002). In na i e eyespo s, sys emic immuni y migh ha e
educed he p oduc ion and/o mig a ion o his ocal signal inducing
he o ma ion o smalle eyespo s. On he o he hand, on wound-
induced eyespo -like pa e ns, he local immuni y had he opposi e
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
112
esponse: no esponse, local o sys emic esponse. The sys emic esponse was
subdi ided in wo classes: independen (sys emic) o dependen o wounding posi ion
(sys emic-wing o - ho ax). See da a poin dis ibu ion in Supplemen a y Figu e S4.3.
Sample sizes a e be ween 23 and 60.
P e ious expe imen s es ing o he e ec o de elopmen al
empe a u e and ho mone manipula ions on eyespo colo ings had
shown di e ences in de elopmen al plas ici y among di e en ings
be ween do sal and en al eyespo s and be ween an e io and
pos e io wing compa men s (Ma eus e al. 2014). I has been shown
ha bu e lies wi h d y season pheno ype (smalle eyespo s) in a we
season en i onmen dye mo e a e a p eda o a ack (P udic e al.
2015). Thus, an in ec ion du ing ea ly pupal de elopmen migh ha e
wo i ness cos s associa ed: 1) he esou ces used du ing
de elopmen in immuni y canno be la e used on ep oduc ion o
locomo ion and 2) he isk o p eda ion du ing adul s age migh
inc ease.
4.4.6 Immune challenge le el, bu no ype, a ec ed eyespo
size
Di e en ypes o bac e ia ac i a e p e e en ially di e en immune
sys em pa hways. Mos G am- bac e ia a e ich in meso-
diaminopimelic acid con aining pep idoglycan (DAP-PGN), while
G am+ bac e ia in lysine-con aining pep idoglycan (Lys-PGN). DAP-
PGN and Lys-PGN p e e en ially ac i a e he IMD o Toll pa hways,
espec i ely (Tanaka & Yamakawa 2011). Howe e , he e is c oss alk
be ween bo h pa hways ha ac syne gis ically a e an immune
challenge (Tanji e al. 2007). We es ed i immune challenge wi h h ee
di e en hea -killed bac e ia solu ions (G am-: Ec, G am+: Sa, and a
mix o bo h bac e ia: ES) in di e en dosages (0.5 µL o a solu ion wi h

Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
113
105, 106 and 107 bac e ial cells) induces di e en pheno ypic
esponses in wing pigmen a ion (Figu e 4.2). We saw ha despi e
G am+ bac e ia had induced mo e signi ican di e ences ela i e o
con ol g oup (Figu e 4.5). This migh indica e ha he pa hways
con olling G am- pa hogens, e.g. IMD, a e mo e in ol ed in eyespo
de elopmen han pa hways con olling G am+ pa hogens, e.g. Toll.
Lowe expe imen al dosages (ca. 105 and 106 bac e ial cells) did
no signi ican ly dec ease he size o eyespo ai s (da k blue ci cles,
Figu e 4.5), bu highe dosage (ca. 107 bac e ial cells) esul ed in
smalle eyespo s ( ed ci cles, Figu e 4.5). Mo eo e , lowe dosages
had he opposi e e ec inc easing he size o some eyespo ai s (ligh
blue ci cles, Figu e 4.5). This indica es ha ac i a ion o immune
sys em does no always mean smalle eyespo s. Tes ing di e en
bac e ial dosages and also di e en ime poin s can cla i y his
hypo hesis.
The ole o Toll and IMD pa hways a e in ec ion is well
documen ed (De G ego io e al. 2002; Lemai e & Ho mann 2007;
Tanaka & Yamakawa 2011). Toll pa hway we e also desc ibed ha ing
a c ucial ole in do sal/ en al pa e ning du ing emb yonic
de elopmen (Bel in & Ande son 1996). IMD pa hway is mo e
desc ibed in egula ion o immune- ela ed p ocesses (Cos a e al.
2009; Myllymäki e al. 2014), bu elish a ansc ip ion ac o in ol ed
in IMD signaling has shown o be in ol ed in aging and
neu odegene a ion de elopmen al p ocesses (Se oude e al. 2002;
Chincho e e al. 2012). In B. anynana, Toll and IMD pa hways (o
ela ed pa hways) migh be c ucial in eyespo size egula ion. Ou
esul s indica e ha IMD migh dec ease eyespo size and bo h IMD
and Toll oge he migh inc ease i .
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
114
Figu e 4.5- E ec o bac e ia ype and dosage on di e en eyespo ai s. Fo each
ai , bac e ia ype and dosage combina ion, he ci cles ep esen he magni ude o
di e en ela i e o PBS (ci cle size; scale on he igh side), Di e en colo s ep esen
di e en di ec ions o eyespo size a ia ion ( ed and da k blue o smalle and ligh
blue o la ge eyespo s) and s a is ical signi icance ( ed signi ican and wo shades o
blue o non-signi ican di e ences; lsmeans p- alue≤0.01). We used an ANCOVA
s a is ical es o es he model: lm ( ai size ~ T ea men + wing size). Eyespo ai s
a e o ganized pe wing: do sal o ewing, en al o ewing and en al hindwing
eyespo ai s. See da a poin dis ibu ion in Supplemen a y Figu e S4.4. Sample sizes
a e be ween 23 and 31.
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
115
4.4.7 Immuni y up- egula ion only pa ially phenocopies he
low de elopmen al empe a u e synd ome
In insec s, pigmen a ion plays essen ial ole in he mo egula ion,
sexual selec ion, species ecogni ion and immuni y (Wi kopp &
Beldade 2009). Melanin pigmen s play a key ole in bo h colo a ion o
he insec in egumen and in de ense agains pa hogens and an
inc easing body pigmen a ion is cos ly and associa ed wi h highe
esis ance o in ec ion (Dubo skiy e al. 2013; Hassall 2014; Ba nes &
Si a-Jo hy 2000). He e, we obse ed ha he wing pa e ns o B.
anynana changed d ama ically a e an immune s imula ion (Figu e
4.1). We s a ed analyzing changes in eyespo size, because
di e ences we e ob ious and eyespo s a e easy o measu e. Howe e ,
his ea men also induced o he changes in wing pigmen a ion ha
we e no so easy o measu e. We ied o ind some me hodologies o
quan i y possible di e ences in gene al wing pigmen a ion, such as
he o e all da kness o he wing and he colo con as o e p oximal-
dis al axis ( o a be e explana ion see Ma e ials and Me hods). To
alida e his me hodologies we used non-manipula ed indi iduals
ea ed a 19°C and 27°C and compa ed hem, expec ing low
empe a u e indi iduals o ha e da ke and less colo con as
pheno ypes. Indeed, his was he case (Figu e 4.6A1 and A2).
The highes dosages o bac e ia dec eased he o e all wing
da kness, con a y o low empe a u e indi iduals (Figu e 4.6B1). On
he o he hand, he colo con as o same indi iduals dec eased as in
low empe a u e, bo h p esen ed less colo con as (Figu e 4.6B2).
Melanic s ains a e usually mo e esis an o pa asi es, bu in he
absence o pa hogens, he hea y de ense in es men s esul in a
lowe biomass, dec eased longe i y and lowe ecundi y (Dubo skiy e
al. 2013; Ba nes & Si a-Jo hy 2000). Melanin p oduc ion a e immune
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
116
challenge and hose used o wing pigmen a ion equi e
p ophenoloxidases (Ce enius & Söde häll 2004). Deploymen o
phenoloxidases o immune de ense could come a he cos o hei
a ailabili y o p oducing body colo s. Howe e , bo h p ocesses in ou
expe imen s a e occu ing in di e en imes: immuni y du ing he i s
day and pigmen a ion du ing he las wo days o pupal de elopmen .
We also analyses he numbe o do sal hindwing whi e pupils.
This ai is in ol ed in sexual selec ion: in bu e lies ea ed a low
empe a u es males p e e o ma e wi h emales mo e do sal hindwing
whi e pupils (Wes e man e al. 2014). The numbe o such pupils was
shown highe in emales ea ed a 17°C han a 27°C (Wes e man e
al. 2014). In ou s udy, we ound he same end be ween low (19°C)
and high (27°C) empe a u e ea ed emales (Figu e 4.6C). In immune
ea ed g oups he e we e no signi ican di e ences in his ai , and
con a y o low empe a u e e ec do sal whi e pupils dec eased in
highe bac e ial dosage independen ly o he bac e ia ype (Figu e
4.6D).
4.4.8 Highe immune challenge educes 20-hyd oxyecdysone
o le els close o hose cha ac e is ic o lowe de elopmen al
empe a u es
The mal plas ici y is co ela ed wi h, and media ed by changes in
20-hyd oxyecdysone le els (Koch e al. 1996; Oos a e al. 2011;
Ma eus e al. 2014). This ho mone is also known o egula e insec
immuni y (Regan e al. 2013; Rus e al. 2013; Tian e al. 2010),
p omo ing bo h cellula and humo al immune esponses (Regan e al.
2013; Rus e al. 2013; Ahmed e al. 1999). He e, we saw ha lowe
de elopmen al empe a u es induced de elopmen o smalle
eyespo s (Figu e 4.1) and da ke wings (Figu e 4.6A1) wi h lowe
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
117
con as (Figu e 4.6A2). The immune challenge applied o pupae
phenocopies he e ec s on eyespo size (Figu e 4.4 and 4.5) and
con as (Figu e 4.6B2), bu no wing o e all-da kness (Figu e 4.6B1).
We hypo hesized ha he mechanism media ing he wing
pigmen a ion changes induced by immune challenge was also al e ed
le els o 20-hyd oxyecdysone (20E). We induced immune challenge
a 8% o pupal de elopmen al ime (12h pos -pupa ion a 27°C) wi h
PBS (con ol o wounding), G am- (Ec) and G am+ (Sa) bac e ia.
A e , we collec ed hemolymph o quan i y ho mone le els a 8.5, 12
and 24% o de elopmen al ime in ea ed indi iduals, as well as in
non-wounded indi iduals ea ed a 27°C o 19°C.
Figu e 4.6 –Highe bac e ial dosage induce same e ec on wing colo con as bu
opposi e e ec on o e all da kness and do sal hindwing whi e pupils ela i e o low
de elopmen al empe a u e. A) Low de elopmen al empe a u e indi iduals ha e
o e all da ke hindwings (A1) wi h lowe colo con as o e p oximal-dis al axis (A2)
ela i e o high de elopmen al empe a u e indi iduals. B) High dosage immune-

Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
118
challenge indi iduals ha e o e all ligh e hindwings (B1) wi h lowe colo con as o e
p oximal-dis al axis (B2) ela i e o con ol g oup. C) Low de elopmen al empe a u e
inc eases he numbe o do sal hindwing whi e pupils, he numbe o indi iduals wi h
h ee whi e pupils is highe a 19°C han a 27°C. D) Immune challenge did no
a ec ed he numbe o do sal hindwing whi e pupils, bu i he e is a end on his ai
i is opposi e o low de elopmen al empe a u e indi iduals. p- alue<0.001***, p-
alue<0.05*, p- alue>n.s.; same le e indica es no signi ican di e ence (p- alue
>0.05), while di e en le e s indica e signi ican di e ences (p- alue <0.05); sample
sizes in A and C a e indica ed (N), and in B1 and in D we e he same as indica ed on
Figu e 4.3 (N), bu in B2 we emo ed ou lie s, so sample size in his analysis was
di e en : PBS=30, Ec5=28, Ec6=24, Ec7=23, Sa5=23, Sa6=24, Sa7=21, ES5=23,
ES6=28, ES7=19.
One hou pos immune challenge (8.5% o pupal de elopmen al
ime) he e we e no di e ences among pupae ea ed a 27°C. 4h la e
(co esponding o 12% o pupal de elopmen ), immune challenged
pupae ea ed a 27°C had 20E le els close o hose o non-
challenged pupae om lowe de elopmen al empe a u es (Figu e
4.7). One day a e bac e ial ea men (24% o pupal de elopmen al
ime), pupae had in e media e 20E le els: lowe han 27°C and highe
han 19°C (Figu e 4.7). The con ol ea men pe se did no a ec ed
20E le els in any o es ed ime poin s (Figu e 4.7). This esul
indica es ha 20E is media ing he immune-induced de elopmen al
plas ici y.
In se e al insec s, such as D. melanogas e (Dip e a) and
Helico e pa a mige a (Lepidop e a), inc eased le els o 20E induce
highe exp ession o AMP (Fla e al. 2008; Wang e al. 2014).
Howe e , i was epo ed ha in B. mo i 20E le els dec ease he AMP
gene exp ession (Tian e al. 2010; Tanaka & Yamakawa 2011). We
know om p e ious chap e ha AMPs le els a e high in G am-
ea ed pupae, a leas un il eigh hou s pos -immune ea men (see
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
119
Chap e 3 - Figu e 3.3). He e, we saw ha six hou s pos -immune
challenge he le els o 20E we e al eady educed, indica ing ha in B.
anynana he p oduc ion o AMPs do no equi e highe le els o his
ho mone. In his case, he immune sys em egula ion in B. anynana
h ough 20E migh be simila o B. mo i. Ne e heless, o con i m his
hypo hesis u he quan i ica ions a e equi ed: 1) 20E le els in ime
poin s be ween ze o and six hou s pos -immune challenge; and 2)
AMPs gene exp ession le els, in 20E manipula ed indi iduals.
Figu e 4.7- 20E le els dec eased in immune-challenged pupae owa ds le els ha
a e simila le els o low de elopmen al empe a u e pupae. F(T ea men :
De elopmen al ime)=4.3, d. .=8, p- alue= 0.0001, ollowed by mul iconpa ison
analysis, alpha=0.05: same le e indica es no signi ican di e ence (p- alue >0.05),
while di e en le e s indica e signi ican di e ences (p- alue <0.05)
4.4.9 Ho monal ea men o immune-challenged pupae
induced high mo ali y
The ho monal le els in B. anynana a e esponsible (a leas
pa ially) o wing pa e n di e ences be ween indi iduals ea ed a
19°C s. 27°C (Koch e al. 1996; Ma eus e al. 2014). S udies ha e
demons a ed ha low empe a u e indi iduals injec ed wi h 20E had
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
120
inc eased eyespo size ela i e o con ol injec ions (Ma eus e al.
2014; Koch e al. 1996). Because immune challenged pupae ha e
phenocopied he e ec o low de elopmen al empe a u es in wing
pigmen a ion (Figu e 4.1) and in in e nal 20E le els (Figu e 4.7), we
hypo hesized ha he immune challenge-induced changes in he la e
could be he cause o he o me . We designed an expe imen wi h
applica ion o he same dose o ho mone as es ed in (Ma eus e al.
2014) ha had been shown o cause wing pheno ype o low
empe a u e indi iduals o look like hose o high empe a u e.
Howe e , ou expe imen al design was di e en in whe e ( ho ax
ins ead o abdomen) and how ( opic applica ion on wound ins ead o
injec ion) and when (8% o de elopmen ins ead o 3%) he ho mone
was applied. Su p isingly, ega dless whe he 20E was applied wi h
PBS o solu ion o dead bac e ia, we had 100% pupal mo ali y. The
20E used he e was pu chased om a di e en company, we
wonde ed i di e ences in pu i y migh esul in di e ences in e ec i e
dose and ied di e en doses o ou assay. We ound ha applica ion
o 62.5 ρg pe indi idual was he highes o ou es ed dosages ha
did no esul in 100% mo ali y (Fig 8A). E en hough mo ali y was
s ill high o his dose (Fig 8B), we could ob ain adul s o sco e wing
pigmen a ion pheno ypes. Pheno ypic measu emen s showed ha his
ho monal ea men was no su icien o es o e he eyespo size o
immune-challenged indi iduals. To con i m his i s in e p e a ion he
20-hyd oxyecdysone le els need o be measu ed in con ols, and
ho mone ea ed indi iduals o know i in e nal le els o ea ed
indi iduals a 12% o de elopmen a e a leas as highe as PBS
con ols. Howe e , his in e p e a ion e en i 20E le els we e
con i med is qui e isky wi h such high mo ali y.
A p e ious s udy has demons a ed ha same immune challenge
induced on ho ax and on abdomen ha e di e en impac s on lies:
Chap e 4
_____________________________________________________________________________________________________________________________________________________________________________________________________________
121
ho ax inju y is mo e se e e han abdomen inju y (Chambe s e al.
2014). Du ing hese expe imen s we go he idea ha he melaniza ion
a e a ho ax immune challenge was as e and mo e ex en han an
abdomen o wing immune challenge. E en i his canno explain
e e y hing i migh be explaining pa o i . In o he insec s was shown
ha 20E egula es d as ically he immune sys em (Fla e al. 2008;
Wang e al. 2014; Tian e al. 2010). Fo example highe 20E le els in
wo di e en Lepidop e a ha e opposi e consequences: in Helico e pa
a mige a he e is an inc ease and in B. mo i a dec ease o AMPs
p oduc ion (Wang e al. 2014; Tian e al. 2010). Because we saw ha
six hou s pos -immune challenge 20E le els we e educed, i is e y
likely ha in B. anynana 20E is egula ing he immune esponse as in
B. mo i. In his scena io, i is no ob ious he eason o such a high
mo ali y a e an immune challenge wi h a non-in ec ious agen . We
obse ed ha h ee days a e ea men , pupae ha had no es o ed
he o iginal g eenish colo main aining he blackish colo did no
managed o each adul s age. The blackish pupae colo we hink was
ela ed wi h he hemolymph melaniza ion, and consequen ly hose
pupae had o e ac i a ed his pa hway o ha m ul le els o he hos . I
his hypo hesis is ue low le els o 20E a e no limi a i e o p oduce
high le els o AMPs (Chap e 3 - Figu e 3.3), bu hey migh be c ucial
o melanogenesis down egula ion.
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
128
insec s) causes a delay in me amo phosis (S iepe e al. 2008). This
allows epai o he damaged disc, wi h i s cells eadjus ing
p oli e a ion a es o egene a e wo adul wings simila in size and
shape (Smi h-Bol on e al. 2009; Shingle on 2010). In i o
expe imen s wi h wing imaginal discs cul u ed in a g ow h-pe missi e
en i onmen showed ha hey au onomously s op cell di ision a
app oxima ely he same size as ha in la al en i onmen (B yan &
Le inson 1985). This suppo s he idea ha each o gan does ha e a
a ge size ha should be achie ed be o e me amo phosis.
He e, we used a bu e ly, B. anynana, as a model sys em o
unde s and i immune s ess (wounding wi h bac e ial challenge)
du ing pupal s age a ec s wing size and shape. We wounded one o
he wings wi h di e en hea -killed bac e ia, a ea men ha ac i a es
immune sys em (Chap e 3 - Figu e 3.3), and analyzed adul wings.
Wi h his expe imen we we e able o know i he e was in e o gan
communica ion and de elopmen al coo dina ion o all ou wings a
his s age. Geome ic mo phome ics a e analy ical me hodologies
ha allow esea che s o analyze shape di e ences in i s bo h
componen s: allome ic and non-allome ic (Deba e al. 2003;
Klingenbe g 2011). T ai shape changes a e e y common and can be
e y d as ic ac oss species (Gidaszewski e al. 2009; Cama go e al.
2015; Ou omu o e al. 2012). In con as , wi hin a popula ion shape
a ia ions a e a e and disc ee (Deba e al. 2003; Blacks one e al.
2013). We used geome ic mo phome ics o analyze he con ibu ion
o allome ic and non-allome ic componen s o shape a ia ion. A
possible mechanisms con ibu ing o wing size and shape a ia ion is
p oli e a ion a e. In Lepidop e ans, wo o h ee cell di isions occu
du ing pupal s age (Nijhou e al. 2014). To unde s and i ea men
was a ec ing p oli e a ion du ing his s age we quan i ied he numbe
o mi o ic cells in immune challenged and con ol indi iduals. Finally,

Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
129
we quan i y shape a ia ion associa ed wi h di e en de elopmen al
empe a u e (a known en i onmen al abio ic cue inducing wing size
and shape di e ences (Deba e al. 2003)), and compa ed he
di ec ion o non-allome ic shape changes o bo h s esses
( empe a u e and immuni y).
5.3. Ma e ial and Me hods
5.3.1 Biological Ma e ial
B. anynana inb eed line (a sub popula ion om An ónia
Mon ei o's lab; used o whole genome sequencing) was ea ed a 19
o 27°C as desc ibed elsewhe e (B ake ield e al. 2009). B ie ly, he
bu e lies we e main ained a 27°C (+/- 0.5°C) wi h 65% (+/-1%)
ela i e humidi y wi h 12h s cycle ligh /da k. Eggs we e collec ed om
ne cages (45x45x45cm; BugDo m-44545) wi h app oxima ely 400
adul indi iduals in e y young maize plan s. To educe he p obabili y
o pa asi e ansmission, we bleached he eggs (5 min in a 20%
bleach solu ion) immedia ely a e collec ion. Adul bu e lies we e ed
wi h esh banana on op o we co on and la ae we e ed wi h young
maize plan s and main ained in cages iden ical o hose used o
adul s a a densi y o 150 o 200 indi iduals pe cage. La ae we e
sexed and only emales we e used in his s udy. P e-pupae we e
collec ed on o 25 well pla es and pupa ion ime was eco ded du ing
he nigh ia ime-lapse pho og aphy wi h a pho o aken each 10 min
(Canon 1000D digi al came a, Hahnel Giga T P o 2.4GHz wi eless
ime emo e con ol).
5.3.2 Wounds and Immune Challenge
We wounded only emales ea ed a 27°C. Wounds we e in lic ed
wi h a 0.25mm diame e ungs en needle (Wo ld P ecision
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
130
Ins umen s) in he igh wing (in he wing cell (wing compa men
be ween wo adjacen eins) below he an e io eyespo ,
app oxima ely hal way be ween he wing ma gin and he no mal
loca ion o he eyespo s) and in he ho ax (in he middle ( ela i e o
le and igh and an e io and pos e io ) o he hi d do sal ho acic
segmen ). Pupae we e e u ned o 27°C un il adul eclosion o un il
u he manipula ion o hemolymph ex ac ion. The wing-wounded
expe imen was un on Janua y and he ho ax-wounded expe imen
on July o 2014.
We induced h ee di e en immune challenges: one con ol
(PBS1X – PBS), one bac e ia ype (Esche ichia coli – Ec (MC4100-
CFP)) in wo di e en dosages: 10exp6 (Ec6) and 10exp7 (Ec7) hea -
killed bac e ia cells/µL dilu ed in PBS). 0.5µL o each solu ion we e
added o he esh wound. A e immune challenge ea men
indi iduals we e e u ned o 27°C un il adul eclosion o un il
dissec ion.
Figu e 5.1- Expe imen al design o unde s and he A) ea men and posi ion e ec s
on wing size; B) he a ec on allome ic and non-allome ic shape changes in wing-
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
131
ea ed indi iduals and he p oli e a ion a e bac e ial ea men ; and C) he a ec on
allome ic and non-allome ic shape changes in di e en ea ing empe a u es. Top
a ow ep esen s de elopmen al ime om pupa ion o eclosion wi h ea men (12
hou s pos -pupa ion) and sampling ime poin s ( ou hou s pos - ea men ). To
analyze ea men posi ion e ec s (A) we ea ed indi iduals on wings o on ho ax
(posi ion is indica ed wi h (*) on pupae scheme) wi h con ol (0.5µL o s e ile PBS –
PBS) o immune challenge (0.5µL o 106 – Ec6, and 0.5µL o 107 E. coli solu ion –
Ec7) solu ions and measu ed he o al size o adul wings. Di e ences in wing shape
we e analyzed using geome ic mo phome ics me hods. In wings- ea ed indi iduals
(B), we analyzed e ec s o ea men in ea ed wings, he e ec o
ea men *wounding, and he e ec s o bac e ial ea men on p oli e a ion du ing
pupal de elopmen (ligh ed squa e). To analyze he e ec o empe a u e on wing
shape, we analyzed adul wings o indi iduals ea ed a 19°C and 27°C.
5.3.3 Wing A ea and landma ks measu emen s
A e s e ching hei wings, adul bu e lies we e sac i iced (-20°C
o a leas 2h s). Wings we e de ached om he ho ax and do sal
sides we e scanned (Epson Pe ec ion V600 Pho o). Images we e
impo ed o Fiji (Schindelin e al. 2012) and he same colo h eshold
was applied o all images, we selec ed whole wing a ea as
measu emen . Da a we e impo ed o R (R De elopmen Co e Team
2015) o do hei g aphical ep esen a ions and s a is ical analysis.
We used an ANOVA o es i bac e ial ea men s had a
signi ican e ec on wing a ea (ao (wing a ea ~ T ea men *Side)).
When ANOVA showed signi ican e ec s o ea men on wing a ea,
we compa ed ac oss ea men s using a pai wise compa ison es
(lsmeans) (Len h & He é 2015).
The shape analysis, geome ic mo phome ics, is based on he
ela i e posi ion o p ede ined landma ks, which can ca y impo an
phylogene ic, de elopmen al, and unc ional in o ma ion (Ca dillo &
Reymen 2010). In bu e lies, and in insec s in gene al, he landma ks
mos o en measu ed a e he loca ions whe e wing eins mee he
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
132
wing ma gin o o he eins because hey a e easye o iden i y (Figu e
5.2A). In his analysis we ollowed he s anda d ecommenda ions o
selec ion o landma ks (13 in o al) (Zeldi ch e al. 2012): 1) homology,
landma ks numbe one o i e a e ein in e cep ions o bi u ca ions in
he p oximal side o he wing and om numbe six o 13 a e ein
in e cep ions wi h he dis al ma gin; 2) adequa e co e age o he o m,
landma ks co e ed p oximal and dis al pa s o he wing; 3)
epea abili y, landma ks we e easy o ind in all indi iduals; 4)
consis ency o ela i e posi ion; 5) coplana i y o landma ks (Zeldi ch
e al. 2012) and 6) app op ia ed sample size (numbe o landma ks=<
(N -4)/2), o maximize he wing shape in o ma ion and he s a is ical
powe o his analysis) (Zeldi ch e al. 2012). Images we e loaded o
Fiji-ImageJ (Schindelin e al. 2012) and 13 landma k we e placed o
each wing. The landma k coo dina es we e loaded in Mo phoJ
(Klingenbe g 2011) o geome ic mo phome ic analysis. Va ia ion in
shape was examined by using geome ic mo phome ics based on
gene alized leas squa es P oc us es supe imposi ion me hods
(Klingenbe g & Mcin y e 1998), which analyze shape by
supe imposing con igu a ions o landma ks o indi iduals o achie e
an o e all bes i .
To e i y epea abili y o he measu men s, we ook 10 bu e ly
wings ( ou PBS, h ee Ec6 and Ec7) and placed all landma ks h ee
imes independen ly. We quan i ied a iance among ea men s and
be ween h ee se s o measu men s h ough P oc us es ANOVA
analysis in Mo phoJ (Klingenbe g & Mcin y e 1998; Klingenbe g 2011).
Resul s showed ha he a iance be ween measu men s is negligeble
(p- alue>0.05) o cen oid size (di e ences in shape co ela ed wi h
size) and also p oscus es dis ance (di e ences in shape wi hou size
e ec ) (Tab S1). Thus, since biological di e ences be ween wings a e
much highe han echnical e o s we can use his me hodology in ou
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
133
shape analysis. F om he landma ck coo dina es we ob ained an
a e age posi ion o each ladma k ( ed do s on Figu e 5.2B), which
ep esen s he wing shape wi hou loca ion, scale and o a ion e ec s.
This wi e ame ep esen a ion was used as wing shape e e ence in
u he analysis.
We analyzed shape a ia ion o unde s and he e ec s o : 1)
di e en immune challenged o gans; igh o ewings o wing and
ho ax immune challenged indi iduals, 2) wounding and immune
challenge in wing- ea ed indi iduals; igh (wounded) and le (non-
wounded) o ewings, 3) de elopmen al empe a u e; igh o ewings o
non-wounded indi iduals ea ed a di e en empe a u es (19 and
27°C). Fo hese h ee compa isons we an he ollowing se o
analyses: 1) de ec ion o possible ou lie s o wing shape (e.g. w ong
landma ks coo dina es), 2) P oc us es bes i , 3) gene a ion o a
co a iance ma ix, 4) p incipal componen analysis, 5) canonical
a ia ion analysis and 6) disc iminan analysis (DA) (Klingenbe g
2011). The ou pu o DA is he compa isons be ween pai s o
ea men o he P oc us es dis ance (shape di e ence) and he
s a is ical signi icance o i is gi en a e 1000 andom pe mu a ion es .
Va ia ion in shape was de e mined as o al (bo h componen s; black
double a ow in Figu e 5.2C), allome ic ( a ia ion co ela ed wi h size;
ed double a ow in Figu e 5.2C) and non-allome ic ( a ia ion in
shape no co ela ed wi h size; blue double a ow in Figu e 5.2C). The
o al shape a ia ion was calcula ed om he o iginal landma k da a.
The co a iance ma ix o hese da a was used in a eg ession model
( a ia ion ~ wing size) o ob ain he da a o allome ic shape a ia ion
( eg ession p edic ed alues) and non-allome ic shape a ia ion
( eg ession esiduals alues). All mo phome ic and s a is ical
analyses we e pe o med using Mo phoJ so wa e, including ou ines

Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
134
o he P oc us es i as well as he pe mu a ion and boo s ap
p ocedu es (Klingenbe g 2011).
Figu e 5.2- Landma k posi ion in he o ewing. A) Image o a do sal o ewing om B.
anynana and espec i e scheme, wi h he 13 landma ks used in his s udy
supe imposed as solid ed do s. B) Wi e ame scheme o landma k posi ion a ia ion
a e p oc us es supe imposi ion o all 10 indi iduals used o e o measu men s ( ed
do s ep esen he a e age o each landma k). The wing shape wi e ame was d awn
in Mo phoJ and he ela i e wing posi ion o an e io and pos e io eyespo ings and
wound si e (*) we e added manually. C) G aphical ep esen a ion o o al, allome ic
and non-allome ic shape a ia ion be ween wo hypo he ical popula ions. The
di e ence be ween he a e age (di ched lines) o bo h popula ions is he o al shape
a ia ion (black double a ow). Solid blue and ed lines ep esen allome ic shape
a ia ion. Shape a ia ion in blue ela i e o he ed popula ion, i is explained by he
allome ic e ec ( ed double a ow) and o he emaining e ec s ( esiduals): non-
allome ic shape a ia ion (blue double a ow).
5.3.4 Immunohis ochemis y
We used wounded and con ol o ewings wi h o e laying cu icle
collec ed a 4 hou s pos - ea men o know i he e was a educ ion o
mi o ic cells in immune challenged indi idual (Ec7). The wings we e
ixed, dehyd a ed (5min sequen ial washed wi h 25%, 50%, 75% and
100% e hanol) and s o ed a -20°C un il u he p ocesses.
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
135
Immedia ely be o e s a ing he immunohis ochemis y p o ocol
desc ibed in (Saenko e al. 2011; B ake ield e al. 2009), wings we e
ehyd a ed (5min sequen ial washed wi h 100%, 75%, 50% and 25%
e hanol). We used DAPI (In i ogen; 1(5mg/mL): 1000(PBS1X) and
ollowed manu ac u e ’s ins uc ions o s ain nuclei. To s ain mi o ic
cells, we use he an i-phospho-his one H3 (Se 10) mouse an ibody
(1:500; Cell Signaling Technology) de ec ed wi h an an i-mouse
seconda y an ibody coupled wi h Alexa 647 (In i ogen; 1:200 dilu ion).
Con ocal Z-se ies s ack images we e acqui ed on a Leica SP5
con ocal, using 20x 0.7NA objec i e. Images we e acqui ed aking he
wound si e o equi alen a ea in he non-wounded wing a image
cen e . Z-s acks we e hen con e ed in o a single image, and i s
b igh ness and con as we e adjus ed in Fiji so wa e (Schindelin e al.
2012). We quan i ied he wing a ea wi h luo escence in ensi y abo e
a ce ain colo h eshold (equal h eshold o all images o each
luo escence s aining) in o al image a ea (2.4 mm2). These
luo escen a eas we e aken as an app oxima ion o o al numbe o
wing cells in he image (DAPI luo escence) and numbe o mi o ic
cells in he image. DAPI a ea was aken as co a ian and he an i-
phospho-his one H3 (Se 10) as esponse a iable in he s a is ic
analysis. We used a GLM model o es i he numbe o mi o ic cells
was signi ican di e en in ea men and wing ype aking o al numbe
o cells as co a ian . The minimal model was (lm(numbe o mi o ic
cells ~ ea men + wing + o al cell numbe )). A e we hen un
pai wise compa isons (package lsmeans) o ind di e ences ac oss
ea men s and wings (Len h & He é 2015).
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
136
5.4. Resul s and Discussion
Wing size and shape a e c ucial ai s o ligh (Sane & Dickinson
2001; Swaddle 1997), and a e known o be a ec ed by gene ic and
en i onmen al ac o s (Zeng & Ve heyen 2004; Deba e al. 2003). In
p e ious chap e s, we ha e shown ha ea men wi h hea -killed
bac e ia inc eased he exp ession o immuni y ela ed genes (Chap e
3) and also saw ha his ea men induced gene al wing pigmen a ion
changes in all wings (Chap e 4 - Figu e 4.1). He e, we in es iga e he
e ec s o same ea men s on wing size and shape. We saw ha
when bac e ial ea men is applied on one wing du ing ea ly pupal
de elopmen all adul wings we e smalle , bu no when same
ea men is applied on ho ax (Figu e 5.3). Thus, he esponse a e
wing ea men is a wing-speci ic esponse and implies in e o gan
communica ion/ egula ion. We hypo hesized ha immune s ess on
pupal wings migh a ec also adul wing shape. To add ess his
hypo hesis we analyzed wing shape in i s wo componen s: allome ic
(di e ences in shape in ela ion o size) and non-allome ic
(di e ences in shape no co ela ed wi h size) in wing- ea ed
indi iduals compa ing 1) wounded wings o di e en ea men pai s
(Figu e 5.4) and 2) wing pai s (wounded and non-wounded wings) o
same ea men (Figu e 5.5). These analyses e ealed ha he
bac e ial ea men a ec ed wing shape in i s bo h componen s and
ha he wing damage only a ec s he non-allome ic shape
componen . To unde s and he possible mechanism unde lying he
de elopmen o smalle wings we quan i ied cell p oli e a ion in
wounded and con ala e al wings in con ol and bac e ia ea ed
g oups (Figu e 5.6). We saw ha bac e ial ea ed g oup had lowe
mi o ic densi y in bo h wounded and con ala e al wing, ela i e o
con ol g oup. Finally, we analyzed he wing shape o wo di e en
Chap e 5
_____________________________________________________________________________________________________________________________________________________________________________________________________________
137
de elopmen al empe a u es (Figu e 5.7A) and saw ha low
de elopmen al empe a u e induce simila non-allome ic wing shape
changes as in bac e ial ea men g oups.
5.4.1 Bac e ial ea men , on de eloping wing, educes he
o al a ea o all adul wings
In p e ious chap e s, we ha e shown ha ea men wi h hea -
killed bac e ia inc eased he exp ession o immuni y ela ed genes
such as an imic obial pep ides and melanogenesis pa hway genes
(Chap e 3 - Figu e 3.3). We also saw ha his ea men induced local
(Chap e 3 - Figu e 3.4) and gene al wing pigmen a ion changes
(Chap ee 4; Figu e 4.1). He e, we in es iga e he e ec s on wing size
o wing- and ho ax- ea ed indi iduals wi h same bac e ial ea men .
Bac e ial ea men educed wing size ela i e o he con ol ea men
(PBS), bu only when bac e ia we e applied on de eloping wings
(Figu e 5.3). This sugges s an in e ac ion be ween ea men and
a ec ed o gan and also an in e o gan communica ion a e wing
ea men o de elop same size o le and igh wings. This also
sugges s ha du ing his de elopmen al s age he bu e ly is no
longe able o es o e he a ge size o damaged wing, bu i can s ill
eadjus he size o o he wings, all eaching a new inal size. The
eadjus men o a new adul size migh be impo an o p oduce
symme ic wings, which in u n, a e impo an o ligh pe o mance
(Hambly e al. 2004).
The eadjus men o size o all wings can also e lec o he
phenomena; such as he ealloca ion o ene ge ic esou ces. Immune
sys em ac i a ion induces a ba e y o esponses including he
p oduc ion o high concen a ions o an imic obial e ec o s (Wu e al.
2010; Fe andon e al. 1998; Lemai e e al. 1997). Moun ing an