Fluorescent sperm marking to improve the fight against the pest insect Ceratitis capitata (Wiedemann; Diptera: Tephritidae).

Abstract

The Sterile Insect Technique (SIT) involving area-wide release of mass-reared and sterilized pest insects has proven successful to reduce, control and eradicate economically important pest species, such as the Mediterranean fruit fly (medfly). For the efficient application, effective monitoring to assess the number and mating success of the released medflies is essential. Here, we report sperm-specific marking systems based on the spermatogenesis-specific Ceratitis capitata beta2-tubulin (Ccbeta2t) promoter. Fluorescent sperm can be isolated from testes or spermathecae. The marking does not cause general disadvantages in preliminary laboratory competitiveness assays. Therefore, transgenic sperm marking could serve as a major improvement for monitoring medfly SIT programs. The use of such harmless transgenic markers will serve as an ideal initial condition to transfer insect transgenesis technology from the laboratory to field applications. Moreover, effective and easily recognizable sperm marking will make novel studies possible on medfly reproductive biology which will help to further improve SIT programs.

Full text

Page 1

RESEARCH PAPER
New Biotechnology?Volume 25,Number 1?June 2008
Fluorescent sperm marking to improve
the fight against the pest insect Ceratitis
capitata (Wiedemann; Diptera:
Tephritidae)
Francesca Scolari1,3, Marc F. Schetelig2,3, Sabrina Bertin1, Anna R. Malacrida1,
Giuliano Gasperi1and Ernst A. Wimmer2
1Dipartimento di Biologia Animale, Universita ` di Pavia, Piazza Botta 9, 27100 Pavia, Italy
2Department of Developmental Biology, Go ¨ttingen Center for Molecular Biosciences, Johann-Friedrich-Blumenbach-Institute of Zoology and Anthropology,
Georg-August-University Go ¨ttingen, GZMB, Justus-von-Liebig-Weg 11, 37077 Go ¨ttingen, Germany
The Sterile Insect Technique (SIT) involving area-wide release of mass-reared and sterilized pest insects
has proven successful to reduce, control and eradicate economically important pest species, such as the
Mediterranean fruit fly (medfly). For the efficient application, effective monitoring to assess the number
and mating success of the released medflies is essential. Here, we report sperm-specific marking systems
based on the spermatogenesis-specific Ceratitis capitata b2-tubulin (Ccb2t) promoter. Fluorescent sperm
can be isolated from testes or spermathecae. The marking does not cause general disadvantages in
preliminary laboratory competitiveness assays. Therefore, transgenic sperm marking could serve as a
major improvement for monitoring medfly SIT programs. The use of such harmless transgenic markers
will serve as an ideal initial condition to transfer insect transgenesis technology from the laboratory to
field applications. Moreover, effective and easily recognizable sperm marking will make novel studies
possible on medfly reproductive biology which will help to further improve SIT programs.
Introduction
TheMediterraneanfruitfly(medfly),Ceratitiscapitata(Wiedemann;
Diptera:Tephritidae),isoneofthemostdevastatingandeconomic-
allyimportantinsects[1].Amosteffectiveandenvironmentallysafe
approach to the species-specific control of this pest is the Sterile
InsectTechnique(SIT)[2],whichreducesapestpopulationbymass
release of reproductively sterile male insects into a wild-type (WT)
populationofthesamespecies.Thisleadstothedecreaseofprogeny
by the competition of radiation-sterilized males with WT males for
WT females [3]. In SIT programs besides the mass rearing, steriliza-
tion and releasing of the pest species, monitoring is of major
importance.Todirectlymonitortherelativesizeofwildpopulations
and to calculate the ratio of released to wild insects, data from field
trapsareused.Currentlymass-rearedpupaearesterilizedanddusted
with fluorescent dyes for field monitoring in medfly SIT programs
[4]. Thus, sterilized flies can be distinguished from WT flies when
recaptured in traps within the release area. However, this marking
procedure for monitoring implies several disadvantages: the fluor-
escent dyes are expensive, dangerous for human health and error
prone [5]. Morphological markers could be alternatives to dusted
flies, but they are often associated with a loss of overall competi-
tiveness [6]. However, flies dusted with fluorescent dyes or carrying
morphological markers cannot be used to get data on the mating
statusofWTfemales,whichisimportanttoassesstheefficacyofSIT
programs.WhetheraWTfemalehasmatedtoaWTmale,areleased
sterile male or perhaps to both, can be checked by analyzing the
progenyoflive-trappedfemales[7]orbycomparingthespermhead
lengthsofsperminthespermathecaeofmatedfemales[8].Sterilized
sperm heads are shorter than WT ones [9]. This size differentiation
despite being widely applied (Beatriz Jorda ˜o Paranhos and Donald
O.McInnis,personalcommunication)iscumbersomeandmethods
suchasanalyzingtheprogenyoflive-trappedfemalesareevenmore
laborious.
For monitoring, transgenetically marked sperm can serve as an
excellent alternative. Such marking systems have been described
for the mosquitoes Anopheles stephensi [10] and Aedes aegypti [11].
In both systems a spermatogenesis-specific b2-tubulin (b2t)
Research Paper
Corresponding author: Wimmer, E.A. (ewimmer@gwdg.de)
1These authors contributed equally.
76
www.elsevier.com/locate/nbt 1871-6784/$ - see front matter ? 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.nbt.2008.02.001

Page 2

promoter from Anopheles gambiae or Aedes aegypti drives testes-
specific expression of a fluorescent marker. These systems thus
combine inheritable fluorescence with sex-specific marking of the
male testes. Here, we report the development of two testes-specific
markers for medfly using the newly isolated spermatogenesis-
specific medfly b2t promoter driving the expression of a red or
green fluorescent protein. These markers will make improved
monitoring possible for SIT programs and enable novel studies
on medfly reproductive biology related to the mating behavior of
this polyandrous species, such as sperm transfer, sperm storage,
sperm use, sperm precedence and sperm competition [12,13].
Results
Isolation of medfly b2-tubulin promoter and generation of
testes-specific marker constructs
To generate a sperm-specific marker system in the medfly, we
isolated the Ccb2t (Genbank accession number EU386342) by
degenerate primer PCR, RACE and inverse PCR. We fused the
Ccb2t promoter (Genbank accession number EU386341) to the
genes of the ‘fast-folding’ fluorescent proteins turboGFP (tGFP
[14]) and DsRedExpress (DsRedEx [15]) to generate sperm-specific
markers. These were then used to engineer four constructs based
on piggyBac vectors carrying polyubiquitin (PUb)-driven EGFP
or DsRed germline transformation markers [16,17]: pBac{fa-
f_attP-Ccb2t-tGFP_a_PUb-DsRed} (#1258), pBac{f_attP-Ccb2t-DsRe-
dEx_a_PUbnlsEGFP} (#1259), pBac{<af_attP-Ccb2t-tGFP_af<_PUb-
DsRed} (#1260) and pBac{>f_attP-Ccb2t-DsRedEx_a>_PUbnlsEGFP}
(#1261). All four constructs carry a minimal attachment P (attP) site
[18], which will enable site-specific integration to potentially
modify the transgenic situation. Moreover in the transformation
vectors #1260 and #1261 the sperm-specific marker was flanked
by gypsy insulator elements ( > = gypsy element in 50–30orienta-
tion; Fig. 1), which should reduce varieties in expression strength
of the sperm-specific markers owing to position effects mediated
New Biotechnology?Volume 25,Number 1?June 2008RESEARCH PAPER
FIGURE 1
Sex-specificmarkedaliveanddeadmales.Bothalive(A1–3,B1–3,C1–3;dorsalview)anddead(A4–6,B4–6,C4–6;ventralview,threemonthsafterdeath)maleswere
observedwithdifferentfiltersets:coldlightsource,DsRedwidefilterandEYFPfilter.(A)AliveanddeadWTmalesshownoorweakautofluorescencewiththeEYFPand
DsRedwidefilter.(B)Schematicrepresentationofthetransgene#1260:pBac{<af_attP-Ccb2t-tGFP_af<_PUb-DsRed}(nottoscale).Alive(B2)anddead(B5)malesshow
redfluorescenceinthehead,thoraxandabdomen.Testesshowstronggreenfluorescence(B3andB6;yellowarrowheads).Indeadmalesbothmarkersarestableforat
least three months (B5 and B6). Images were taken on males from line #1260_F-1_m-2. (C) Schematic representation of the transgene #1261: pBac{>f_attP-Ccb2t-
DsRedEx_a>_PUb-nls-EGFP} (not to scale). The transgene contains the Ccb2t promoter fused to DsRedEx and a PUb-driven EGFP [16]. Ventral head (not shown) and
dorsalthorax(C3)showgreenfluorescence.Testes(C2andC5;yellowarrowheads)arestrongredfluorescent.Theredtestesmarkerisstableforatleastthreemonthsin
thedeadmales(C5).EGFPfluorescenceisnotvisibleindeadmales(C6).Imagesweretakenonmalesfromline#1261_F-5_m-4.A52-bpattPsiteisindicatedasblack
bar; white arrowheads represent 0.4-kb gypsy fragments (50–30orientation). BamHI restriction sites are shown by vertical arrows.
www.elsevier.com/locate/nbt
77
Research Paper

Page 3

by different genomic integration sites. These four vectors were
used to germline transform medfly. The vector #1260 (Fig. 1B) was
injected into 445 embryos of which 120 larvae hatched and 23
survived to adulthood (11 G0 females and 12 G0 males). Four
female crossings (2–3 G0 females crossed to 5 WT males; F-1–F-4)
and 4 male crossings (3 G0 males crossed to 15 WT females; M-1–
M-4) were set up. G1 adults were screened for tGFP and DsRed
fluorescence. Three of the female crossings (F-1, F-2 and F-3) and 1
of the male crossings (M-3) had fluorescent progeny with different
red patterns and all transgenic males showed testes-specific tGFP
expression. Thirteen male or female flies with different DsRed
expression patterns (2 for F-1, 8 for F-2, 2 for F-3 and 1 for M-3)
were backcrossed twice to WT to recognize possible multi-inser-
tions by different PUb–DsRed patterns. The vector #1261 (Fig. 1C)
was injected into 376 embryos of which 85 larvae hatched and 14
survived to adulthood (10 G0 females and 4 G0 males). Five female
crossings (2 G0 females crossed to 5 WT males; F-1–F-5) and 4 male
crossings (1 G0 male crossed to 6–15 WT females; M-1–M-4) were
set up. G1 adults were screened for DsRedEx and EGFP fluores-
cence. Two of the female crossings (F-2 and F-5) had fluorescent
progeny showing different green fluorescent expression patterns
with all transgenic males having testes-specific DsRedEx expres-
sion. Eight male or female flies with different EGFP expression
patterns (6 for F-2 and 2 for F-5) were backcrossed twice to WT to
recognize potential multi-insertions by different PUb–EGFP pat-
terns. For injections with piggyBac vectors #1258 and #1259, lack-
ing the flanking gypsy insulators, no transgenic flies could be
identified, despite a similar number of injected embryos, larval
hatchings, adult survival and fecundity (data not shown).
Sex- and tissue-specific marker expression
Males transformed with the construct #1260 expressed Ccb2t
promoter-driven tGFP in the testes (Fig. 1B3). The PUb–DsRed
markervariedintheredfluorescencepatternbetweenthedifferent
lines as previously described for PUb–DsRed [19], whereas the
pattern itself was stable for each independent line. Both body
and testes markers are highly stable in alive flies (Fig. 1B2 and 3) as
well as in dead flies three months after death (Fig. 1B5 and 6). By
contrast, alive or dead WT flies show no or only weak autofluor-
escence (Fig. 1A1–6).
Males transformed with the construct #1261 expressed Ccb2t
promoter-driven DsRedEx in the testes (Fig. 1C2 and 5). The green
fluorescence pattern varied because of position effects of the PUb–
EGFP marker in different lines. The testes-specific DsRedEx expres-
sion was stably detectable in dead males three months after death
(Fig. 1C5), whereas the green EGFP fluorescence lost intensity two
to three days after eclosion in alive flies (data not shown) and was
not detectable in dead flies (Fig. 1C6).
Strategy for generating homozygous lines
To verify transgenic lines (preliminarily screened for homozygous
condition by fluorescence patterns and intensity) as genetically
homozygous, we followed a strategy including three molecular
techniques. First, Southern blots to BamHI-digested genomic DNA
of the lines #1260_F-1_m-2 and #1260_F-3_m-1 using a DsRed-
specific probe (Fig. 2A) and of the lines #1261_F-5_m-4 and
#1261_F-5_m-5 using a EGFP-specific probe (Fig. 2B) indicated
single copy integration of the respective constructs. Second, 50
and 30insertion site sequences were isolated by inverse PCR and
verified correct piggyBac-mediated integration at canonical TTAA
target sites (see Supplementary Methods). Third, characterization
offliesbymultiplexPCR[20]wascarriedouttoverifyandestablish
homozygous lines. For this, we designed two specific primers to
the piggyBac ends and one primer for each flanking sequence
(Fig. 2C diagram). From ten single crossings of putative homo-
zygous flies, egg collections were taken separately. Thereafter a
multiplex PCR was performed separately on genomic DNA from
each single parent. Therefore, we were able to identify each parent
molecularly as a homozygous or heterozygous individual for the
lines #1260_F-1_m-2, #1260_F-3_m-1 and #1261_F-5_m-5. The
progeny of all the single couples in which both parents were
molecularly proven homozygotes were grouped to reduce bottle-
neckeffectsinthegenepool.Inline#1261_F-5_m-4, thetransgene
integrated into a mariner-like transposon. Therefore, the multiplex
PCR could not be used to identify homozygous situations and this
line was checked for homozygous condition by more time-con-
suming test crosses to WT.
Mating ability of transgenic testes-marked medflies
To assess whether the testes-marked homozygous medfly lines
show any major biological fitness problem, we tested their ability
to mate and sire progeny in preliminary laboratory competitive-
ness tests. Transgenic males were tested at a 1:1:1 ratio competing
with WT males for copulation with WT females (Table 1). G-test
analyses showed that the ten replicates run for each line were
statistically homogenous (#1260_F-3_m-1: G = 4.7. #1260_F-1_m-
2: G = 1.6. #1261_F-5_m-5: G = 4.2. #1261_F-5_m-4: G = 3.0. df = 9
and P > 0.05 for all lines). #1260_F-3_m-1 males made 47% of the
totalmatings andaG-test onthe totalmating number resulted ina
nonsignificant difference between WT and transgenic perfor-
mance (G = 0.5; df = 1; P > 0.05), which indicates the absence of
a negative load for transgenic males in obtaining copulations.
However, transgenic males from the other three transgenic lines
had significant less matings compared to WT (Table 1; #1260_F-
1_m-2:
G = 18.4. #1261_F-5_m-5:
G = 102.5. df = 1 and P < 0.001 for all three lines). The results
obtained from control cages in which only WT males were present
showed thatthe total mating frequency was independent from the
composition of the experimental cage and a G-test (G = 5.91;
df = 4; P > 0.05) indicated that the presence of transgenic males
did not disturb the mating tendency of the females.
Moreover, we compared the copulation latency [21] of trans-
genic and WT males and observed that transgenic males were not
slower than WT in gaining copulation (#1260_F-3_m-1 versus WT:
t = 0.3, df = 175. #1260_F-1_m-2 versus WT: t = 0.8, df = 184.
#1261_F-5_m-5 versus WT: t = 1.2, df = 163. #1261_F-5_m-4 versus
WT: t = 0.1, df = 166. P > 0.05 for all lines).
G = 81.7.#1261_F-5_m-4:
Adult longevity and paternity success
As an additional assessment of the overall fitness of the testes-
marked homozygous medfly lines, we compared the longevity of
WT and transgenic flies. Adult longevity of both #1260_F-3_m-1
and #1260_F-1_m-2 males was nonsignificantly different from WT
as shown by t-tests (#1260_F-3_m-1 versus WT: t = 0.23. #1260_F-
1_m-2 versus WT: t = 0.23. df = 48 and P > 0.05 for both lines). On
the contrary, #1261_F-5_m-5 and #1261_F-5_m-4 males showed a
RESEARCH PAPER New Biotechnology?Volume 25,Number 1?June 2008
78
www.elsevier.com/locate/nbt
Research Paper

Page 4

significantly shorter survival (#1261_F-5_m-5 versus WT: t = 2.58.
#1261_F-5_m-4 versus WT: t = 2.15. df = 48 and P < 0.05 for both
lines) (Fig. 3A). Nonsignificant differences in female adult long-
evity were observed between WT and transgenic lines (#1260_F-
3_m-1 versus WT: t = 1.83. #1260_F-1_m-2 versus WT: t = 1.11.
#1261_F-5_m-5 versus WT: t = 0.48. #1261_F-5_m-4 versus WT:
t = 1.75. df = 48 and P > 0.05 for all lines) (Fig. 3B).
Furthermore, to analyze the effect of transgenic marking on the
reproductive capacity, we compared the proportion of progeny
sired by transgenic or WT males in a laboratory competition assay
(Fig. 3C–F). In two experimental replications, males from lines
#1261_F-5_m-5 and #1261_F-5_m-4 did show reduced competi-
tiveness in these tests. However, males from lines #1260_F-3_m-1
and #1260_F-1_m-2 demonstrated to have great success in siring
progeny in competition to WT males, which indicates that the
transgenic marking does not generally cause a reduction in com-
petitiveness.
Sperm-specific expression of fluorescent markers
To see whether the Ccb2t promoter-driven fluorescence could
serve to follow individual sperm, we analyzed dissected testes
and spermathecae in detail (Fig. 4). The testes-limited fluorescence
in medfly (Fig. 1) confirmed the spermatogenesis-specific expres-
sion of b2t [22] and showed that the isolated 868-bp 50region
New Biotechnology?Volume 25,Number 1?June 2008 RESEARCH PAPER
FIGURE 2
Strategy for generating homozygous transgenic lines. BamHI-digested genomic DNAs isolated from indicated medfly lines were hybridized by using a DsRed (A)
oranEGFP(B)probe.WTgenomicDNAwasusedasacontrolforboth.Asinglebandineachlaneindicatesuniqueintegrationofthetransgenes.(C)Astrategyfor
generating homozygous lines is shown for #1260_F-3_m-1 as an example. Eggs from ten single mating pairs were collected separately. Subsequently, genomic
DNAwas isolated from each single parent fly and usedas template forindependent multiplex PCR reactions with two primers specific to the piggyBac ends (black
arrows) and two primers depending on the integration site sequence as derived by inverse PCR (hatched arrows); specific piggyBac primers have been designed
for individual lines to minimize interference with the integration site-specific primers. In homozygous situation (homo) two bands of 474 bp (amplified by fs-14f/
fs-9r) and 254 bp (amplified by fs-8f/fs-11r) are expected, while in the heterozygous individuals (hetero) an additional band of 166 bp (amplified by fs-8f/fs-9r) is
expected,whichcorrespondstotheWTgenomicsituation.Theprogenyofthesinglematingpairsinwhichbothparentswerehomozygouswerepooledtogether
and maintained as homozygous line. M: 50 bp ladder.
www.elsevier.com/locate/nbt
79
Research Paper

Page 5

contains the regulatory elements responsible for sex- and tissue-
specific expression. Fluorescent spermatid bundles were isolated
from males of lines #1260_F-3_m-1, #1260_F-1_m-2 (Fig. 4B3),
#1261_F-5_m-5 and #1261_F-5_m-4 (Fig. 4C2). Dissected sper-
mathecae from WT females mated to transgenic males from
#1260_F-3_m-1 or #1260_F-1_m-2 (Fig. 4B6) and #1261_F-5_m-5
or #1261_F-5_m-4 (Fig. 4C5) carried respective green or red fluor-
escent sperm and additional fluorescent material transferred dur-
ing mating. By contrast, WT females mated to WT males did not
show such red or green fluorescent materials (Fig. 4A5 and 6) and
single sperm separated from such spermathecae showed neither
red nor green autofluorescence (Fig. 4A8 and 9). Single sperm
derived from transgenic males showed the respective green or
red fluorescence restricted to the sperm tail. Sperm of #1260_F-
3_m-1 showed in addition to green also red fluorescence, which is
most probably due to a position effect on the PUb–DsRed marker
(data not shown). The detectability of fluorescently marked sperm
in spermathecae and the results that marked sperm does not in
principle interfere with mating and paternity success indicate the
potential use of such a marker for monitoring purposes in medfly
SIT programs.
Studying reproductive biology
Furthermore, the ability to easily detect marked sperm after the
transfer from males to females enables effective studies of repro-
ductive biology, such as sperm transfer and sperm use. To demon-
strate this, we compared the number of transgenic or WT sperm
stored by WT females after successful copulation (Table 2). The
number of females with empty spermathecae [23] resulted to be
statistically heterogeneous across the lines (G = 25.75; df = 4;
P < 0.001). This was because of both asignificant difference within
the transgenic lines (G = 15.51; df = 3; P < 0.001) and between
transgenic (pooled together)
P < 0.001). Successful sperm transfer appeared particularly com-
promised for #1261_F-5_m-5 males.
Sperm count data were pooled from two replicates covering two
generations and statistically tested for homogeneity (WT: t = 0.44;
df = 48. #1260_F-3_m-1: t = 1.37; df = 59. #1260_F-1_m-2: t = 1.25;
df = 70. #1261_F-5_m-5: t = 1.34; df = 52. #1261_F-5_m-4: t = 1.69;
df = 40. P > 0.05 for all lines). ANOVA across the four transgenic
linesdidnotresultinanysignificantdifference(F = 1.46;df = 3,131;
P > 0.05). However, ANOVA showed that WT and transgenic lines
werestatisticallyheterogeneous(F = 3.93;df = 4,171;P < 0.01).The
comparison between transgenic lines (pooled together) and WT
showed a high significance (F = 11.41; df = 1, 171; P < 0.001),
demonstrating that the number of sperm released by transgenic
males was significantly lower than the number transferred by WT.
In an identical but independent experimental set-up, we exam-
ined the time line of WT or transgenic sperm use (Fig. 5). A G-test
indicatesthatlarvalhatch-ratesdeclinedovertimeforWTaswellas
all the transgenic lines (WT: G = 136.6. #1260_F-3_m-1: G = 542.7.
#1260_F-1_m-2: G = 712.2. #1261_F-5_m-5: G = 958.3. #1261_F-
5_m-4:G = 760.3.df = 2andP < 0.001forall).Ateacheggcollection
andWT(G = 10.23;df = 1;
RESEARCH PAPER New Biotechnology?Volume 25,Number 1?June 2008
TABLE 1
Mating ability of transgenic males competing with WT males
Transgenic lines
, mated to transgenic < (tot matings)
#1260_F-3_m-1
47% (177)NS
#1260_F-1_m-2
34% (186)***
#1261_F-5_m-5
16% (165)***
#1261_F-5_m-4
13% (168)***
Relative mating success of transgenic males in competition with WTmales forcopulations
with WT females. For each transgenic line,25 males were maintained in a cage with 25 WT
females and 25 WT males and the number of total matings was obtained in 10
experimental replications. The percentage of matings attributed to transgenic males is
reported. G-tests were used to reveal significant differences in mating number between
transgenic and WT males (df = 1 in all the G-tests; significance levels: ***=P < 0.001;
NS = nonsignificant).
FIGURE 3
Adult longevityandpaternity successof the transgenicflies. Meansurvival of WTandtransgenicmales (A)andfemales (B). Bars represent average number (?SE) of
survival days. Twenty-five flies were monitored for each line. *, transgenic lines with significantly shorter lifespan than WT. (C–F) For each of the four indicated
transgenic lines, 25 males were maintained in a cage with 25 WT females and 25 WTmales to determine the relative success in siring progeny of transgenic males
competingwithWTmales.Eggslaidwithin24hourswerecollectedeverytwodaysfortwoweeksstartingthreedaysaftereclosion.Theexperimentwasdonetwicefor
each transgenic line (black and gray bars). The percentage of progeny attributed to transgenic males is shown with total number of flies on top of the bars.
80
www.elsevier.com/locate/nbt
Research Paper

Page 6

New Biotechnology?Volume 25,Number 1?June 2008 RESEARCH PAPER
FIGURE 4
Fluorescence in testes, spermathecae and single sperm. Spermatid bundles dissected from male testes (1–3), mechanically opened spermathecae from WT
females mated to WT or transgenic males (4–6) and single sperm from these spermathecae (7–9) are depicted with phase contrast or epifluorescence with the
Zeiss filters sets 13 and 20. (A) The WTspermatid bundle (A1–3) and single sperm (A7–9) show no red or green autofluorescence. The intact WT spermatheca is
darkblack(notshown)andwhenbrokenopenshowsweaklyredandgreenautofluorescentingmaterial(A4–6).(B)Thespermatidbundleandthespermfromthe
spermatheca (female mated to a male of line #1260_F-1_m-2) show strong green fluorescence (B3, 6 and 9). Additional green fluorescent material is detected
alongthespermandinthespermatheca(B9and6).(C)Thespermatidbundleandthespermfromthespermatheca(femalematedtoamaleofline#1261_F-5_m-
4) show red fluorescence (C2, 5 and 8). Additional strongly red fluorescent material is visible along the sperm and in the spermatheca (C8 and 5).
www.elsevier.com/locate/nbt
81
Research Paper

Page 7

time point (3, 10 and 15 days after the mating), the proportion of
hatchedeggswasstatisticallyheterogeneousacrossthelines(df = 4;
P < 0.001 in all the G-tests). This was because of both a significant
difference within the transgenic lines(df = 3; P < 0.001 in all the G-
tests) and between transgenic (pooled together) and WT (df = 1;
P < 0.001 in all the G-tests). On the one hand, this suggests that
transgenicspermwasnotasefficientlyusedovertimeasWTsperm,
butontheotherhand,thisalsoindicatesthattransgenicspermwas
still used 15 days after the mating.
Discussion
With this study we provide a powerful sperm-marking system for
medfly. Our aim was to obtain direct sperm visualization, both for
applied and basic purposes: first, to provide an efficient tool for
improvingthemonitoringofongoingSITprogramsandsecond,to
improve the analysis of medfly mating physiology regarding
sperm transfer, storage, competition and use. We show that the
endogenous b2t promoter could be used in the medfly to generate
fluorescently labeled sperm as was previously described for mos-
quitoes [10,11]. The fluorescent marking makes it possible to
separate males from females and to clearly identify the testes as
well as single sperm at adult stages. In contrast to the mosquito
studies, however, the fluorescence in the testes is only sporadically
detectable at the last larval stage despite the use of fast-folding
fluorescentproteins, whichistherefore notsuitabletocreate larval
sexing and sex-separation techniques in medfly. For the clean and
rapid establishment of homozygous transgenic lines – which are
necessary for assays on sperm competition – we developed a
molecular method based on multiplex PCR that can be applied
to nonmodel organisms. When the homozygous transgenic lines
were compared to WT in different preliminary laboratory compe-
titiveness tests, several lines showed reduced fitness indicating
that the genetic bottleneck during the lines’ establishment, the
presence of the transgene, or its particular insertion into the
genome can cause fitness costs [24]. In spite of this, males of
one sperm-marked line, #1260_F-3_m-1, showed no significant
reduction in their overall fitness transmitting their genes to the
next generation in a competitive way. This shows the advantage of
creating many independent transgenic lines to choose competi-
tive ones. Actually, our constructs introduced attP sequences,
which will make it possible to use the site-specific integrase system
from phage phiC31 [25] to add additional functional transgenes
into this fit sperm-marked line #1260_F-3_m-1 at the same geno-
mic position [26]. Moreover, line #1260_F-3_m-1 shows both red
and green fluorescence in the sperm – probably on account of a
position effect on the PUb-driven DsRed body marker – that
indicates that the strong expression of fluorescent proteins is
not harmful to the sperm’s function. The laboratory competitive-
ness tests performed here can only be a first step into the evalua-
tion of the fitness and stability of such transgenic lines, which
need to be followed by mass rearing tests and field cage evaluation
by setting competition between transgenic and endemic, locally
sampled, WT flies [27]. A simple distinction between WT sperm
and sperm from released sterile males will greatly improve mon-
itoring SIT programs, which is a key component to run this
environment-friendly pest management successful and cost effec-
tive. The combination of the presented fluorescent markers could
serve excellently for this purpose. Owing to the stability of the
usedfluorescentproteins[28],boththeredfluorescence(DsRed)in
the body as well as the green fluorescence in the testes (tGFP) are
stable several months after death. Therefore, even dead flies cap-
tured in traps within the SIT release area could be scored to
calculate the ratio of released to wild insects. In addition, the
sperm marking would make it possible to more easily assess the
mating status of trapped females, which would allow for control-
ling the mating efficiency of the released sterile males.
The high quality control during the production of sterile insects
for SIT as well as the fact that sterilized insects are incapable of
passing their genes on to the next generations provide ideal initial
conditionsforfirstreleasesofgeneticallyengineeredinsectsintothe
field [29,30]. Before fertile transgenic insects can be released in
economic and medical field applications, ecological risk assess-
ments are necessary. Nevertheless, to assess such potential risks,
at some point transgenic insects will have to be released. For this
purpose,SITprogramsthatusesterilizedtransgenicinsectscarrying
harmless, fluorescent-protein-based markers – as described in this
study – will be most appropriate. Owing to the sterility, exposure of
the environment to the transgenes will be limited to the lifespan of
the released individuals. Transgene constructs containing only
fluorescent protein markers will be most suitable to transfer trans-
genic technology from the laboratory to the field, since they will
improveSITapplicationsbysimplifyingthemonitoring,shouldnot
provide advantages to the carrier organism and should allow the
RESEARCH PAPERNew Biotechnology?Volume 25,Number 1?June 2008
TABLE 2
Average number of mating-transferred sperm
Medfly lines% , with
sperm (n tot)
Average sperm
number ? SE
2364 ? 217
1854 ? 232
1625 ? 174
1164 ? 176
1532 ? 198
WT
82 (50)
#1260_F-3_m-1
64 (61)
#1260_F-1_m-2
71 (72)
#1261_F-5_m-5
27 (54)
#1261_F-5_m-4
60 (42)
Percentage of females that had a mating with actual sperm transfer is indicated and the
sample size (number of mating engaged and dissected females) is shown in parenthesis.
Average number (?SE) of sperm counted from dissected spermathecae of WT females
mated to WT or transgenic males. The average sperm number is calculated on the total
number of samples, excluding the females whose mechanically broken spermathecae
were found empty.
FIGURE 5
Time course of larval hatching. The hatching rate was determined on egg
collections of three time points: 3, 10 and 15 days after mating between WT
females and either WT or transgenic males had occurred. Each point
represents the mean proportion (?SE) of hatched embryos on total eggs laid
within 24 hours. The respective sample size (number of females scored) is
indicated in the inlet legend.
82
www.elsevier.com/locate/nbt
Research Paper

Page 8

identification of carriers at later stages. Such initial transgenic SIT
programs could then provide the first ecological insights into the
fitnessoftransgenicinsectsandtheirdistribution[31].Furthermore,
theavailabilityofhomozygousmedflylineswithdifferentlymarked
sperm (tGFP or DsRedEx) will enable novel studies on the repro-
ductive biology of this polyandrous species. Thus old and new
questions on mating physiology, cryptic female choice, sperm
competition, sperm quality and other postcopulatory sexual beha-
viors can now be addressed [32]. And in turn, an enhanced knowl-
edge of these basic biological processes will provide additional
means to further improve environmentally safe, biological pest
management approaches.
Materials and methods
Details on inverse PCR, cloning strategy, primers, germline trans-
formation, imaging, Southern blot analyses and fitness tests are
shown in Supplementary Methods.
Medfly samples
WT and transgenic medfly lines were maintained under standard
rearing conditions [33]. The WT strain Egypt-II was obtained from
the FAO/IAEA Agriculture and Biotechnology Laboratory (Seibers-
dorf, Vienna, Austria). The fitness tests were performed on homo-
zygous transgenic lines maintained in the lab for six to ten
generations.
Isolation of medfly b2-tubulin
On the basis of protein alignments of b2-tubulin genes from Droso-
phila melanogaster, Drosophila hydei, Anopheles gambiae, Danio rerio,
GallusgallusandHomosapiens,degenerateprimersweredesignedto
possibly exclude or at least minimize the isolation of b1- and b3-
tubulinsequences.ToisolatepolyA+mRNAfromtestes,wedissected
36 males, put the testes into RNAlater solution (Ambion, Austin)
and extracted polyA+RNA using Micro Poly(A) Pure (Ambion,
Austin). Three cDNA pools were generated using this testes-specific
mRNA: (i) a double-strand cDNA using BD SMART PCR cDNA
Synthesis Kit (BD Biosciences, Heidelberg); (ii) a 50single-strand
cDNA for RACE and (iii) a 30single-strand cDNA for RACE using the
BD SMART RACE cDNA Amplification Kit (BD Biosciences, Heidel-
berg). A 523-bp fragment of the b2-tubulin was isolated with degen-
erateprimersmfs-100andmfs-102onthedouble-strandcDNApool
(1.5 min at 948C; 31 cycles of 30 s at
948C,1 minat508C,1.5 minat688C;and5 minat688C).Toamplify
the 50- and 30-UTR of b2-tubulin we did RACE PCR on a 50and a 30
single-strand cDNA pool using the BD SMART RACE cDNA Ampli-
fication Kit (BD Biosciences, Heidelberg) and the gene-specific
primers mfs-150 for the 50-end and mfs-155 for the 30-end. The
genomicupstreamregionofCcb2twasthenisolatedbyinversePCR,
sequenced and finally an 868-bp 50region including the Ccb2t
promoterwasamplifiedbyPCRfromgenomicDNA(Supplementary
Fig. 1).
Multiplex single-fly PCR
To identify homozygous transgenic individuals, genomic DNA of
single flies was isolated as described [34]. Five microliters of the
single-fly homogenates were used for multiplex PCR reactions
(2 min at 948C; 25 cycles of 30 s at 948C, 30 s at 588C, 1 min at
728C;and5 minat728C)usingBDAdvantage2PCR(BDBiosciences,
Heidelberg). Each multiplex primer mix contained four primers: a
WT-specific forward primer (primer fs-15f for line #1260_F-1_m-2,
fs-8f for #1260_F-3_m-1 and fs-27f for #1261_F-5_m-5), a WT-spe-
cific reverse primer (primer fs-16r for line #1260_F-1_m-2, fs-9r for
#1260_F-3_m-1 and fs-28r for #1261_F-5_m-5), a piggyBac-specific
forward primer (primer fs-20f for line #1260_F-1_m-2, fs-14f for
#1260_F-3_m-1 and fs-30f for #1261_F-5_m-5) and a piggyBac-
specific reverse primer (primer fs-18r for line #1260_F-1_m-2, fs-
11r for #1260_F-3_m-1 and fs-29r for #1261_F-5_m-5). Each PCR
reaction was analyzed on a 2.0% agarose gel for the presence or
absence of WT and transgene insertion-specific bands.
Acknowledgements
We thank Alfred M. Handler for providing plasmids, Gerald Franz
for medfly strains and Hendrikje Hein, Ludvik Gomulski, Claudio
Quaroni, as well as Paolo Siciliano for technical assistance. We
thank Carlo Matessi for statistical advice. This work was initiated
during an IAEA funded meeting of the Co-ordinated Research
Project (CRP) ‘The Use of Molecular Tools to Improve the
Effectiveness of SIT’. The work was supported by the Robert Bosch
Foundation (to E.A.W.), the DAAD-VIGONI (to E.A.W. and G.G.)
and the PRIN 2006 (to G.G.).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.nbt.2008.02.001.
References
1 White, I.M. and Elson-Harris, M.M. (1992) Fruit flies of economic significance: their
identification and bionomics. CAB International
2 Enkerlin, W.R. in Sterile insect technique - principles and practice in area-wide integrated
pest management. (eds. V.A. Dyck, J. Hendrichs and A.S. Robinson) 651–676
(Springer, Dordrecht, NL; 2005)
3 Knipling, E.F. (1955) Possibilities of insect control or eradication through the use of
sexually sterile males. J. Econ. Entomol. 48, 459–462
4 Enkerlin, W. et al. (1996) Increased accuracy in discrimination between
captured wild unmarked and released dye-marked adults in fruit fly
(Diptera: Tephritidae) sterile released programs. J. Econ. Entomol. 89,
946–949
5 Hagler, J.R. and Jackson, C.G. (2001) Methods for marking insects: current
techniques and future prospects. Annu. Rev. Entomol. 46, 511–543
6 Robinson, A.S. and Hendrichs, J. in Sterile insect technique - principles and practice in
area-wide integrated pest management. (eds. V.A. Dyck, J. Hendrichs and A.S.
Robinson) 727–760 (Springer, Dordrecht, NL; 2005)
7 Katsoyannos, B.I. et al. (1999) Method of assessing the fertility of wild Ceratitis
capitata(Diptera:Tephritidae)femalesforuseinsterileinsecttechniqueprograms.J.
Econ. Entomol. 92, 590–597
8 McInnis, D.O. et al. (1994) Population suppression and sterility rates induced by
variable sex ratio, sterile insect releases of Ceratitis capitata (Diptera: Tephritidae) in
Hawaii. Ann. Entomol. Soc. Am. 87, 231–240
9 McInnis,D.O.(1993)Sizedifferencesbetweennormalandirradiatedspermheadsin
mated female Mediterranean fruit flies (Diptera: Tephritidae). Ann. Entomol. Soc.
Am. 86, 305–308
10 Catteruccia, F. et al. (2005) An Anopheles transgenic sexing strain for vector control.
Nat. Biotechnol. 23, 1414–1417
11 Smith,R.C.etal.(2007)Testis-specificexpressionoftheb2tubulinpromoterofAedes
aegypti and its application as a genetic sex-separation marker. Insect Mol. Biol. 16,
61–71
12 Bonizzoni,M.etal.(2006)Ispolyandryacommoneventamongwildpopulationsof
the pest Ceratitis capitata? J. Econ. Entomol. 99, 1420–1429
New Biotechnology?Volume 25,Number 1?June 2008RESEARCH PAPER
www.elsevier.com/locate/nbt
83
Research Paper

Page 9

13 Yuval, B. et al. (1995) Sperm transfer and storage in the Mediterranean fruit fly
Ceratitis capitata (Diptera: Tephritidae). Isr. J. Zool. 42, 88–89
14 Shagin, D.A. et al. (2004) GFP-like proteins as ubiquitous metazoan superfamily:
evolution of functional features and structural complexity. Mol. Biol. Evol. 21, 841–
850
15 Bevis, B.J. and Glick, B.S. (2002) Rapidly maturing variants of the Discosoma red
fluorescent protein (DsRed). Nat. Biotechnol. 20, 83–87
16 Handler, A.M. and Harrell, R.A. (1999) Germline transformation of Drosophila
melanogaster with the piggyBac transposon vector. Insect Mol. Biol. 8, 449–457
17 Handler, A.M. and Harrell, R.A. (2001) Transformation of the Caribbean fruit fly,
Anastrepha suspensa, with a piggyBac vector marked with polyubiquitin-regulated
GFP. Insect Biochem. Mol. Biol. 31, 199–205
18 Thorpe, H.M. et al. (2000) Control of directionality in the site-specific
recombinationsystemoftheStreptomycesphagephiC31.Mol.Microbiol.38,232–241
19 Handler, A.M. and Harrell, R.A. (2001) Polyubiquitin-regulated DsRed marker for
transgenic insects. Biotechniques 31, 820–828
20 Chamberlain, J.S. et al. (1988) Deletion screening of the Duchenne muscular
dystrophy locus via multiplex DNA amplification. Nucleic Acids Res. 16, 11141–
11156
21 Moehring, A.J. and Mackay, T.F. (2004) The quantitative genetic basis of male
mating behavior in Drosophila melanogaster. Genetics 167, 1249–1263
22 Michiels,F.etal.(1989)A14 bppromoterelementdirectsthetestisspecificityofthe
Drosophila beta-2 tubulin gene. EMBO J. 8, 1559–1565
23 Taylor, P.W. et al. (2000) Copula duration and sperm storage in Mediterranean fruit
flies from a wild population. Physiol. Entomol. 25, 94–99
24 Marrelli, M.T. et al. (2006) Mosquito transgenesis: what is the fitness cost? Trends
Parasitol. 22, 197–202
25 Groth, A.C. et al. (2004) Construction of transgenic Drosophila by using the site-
specific integrase from phage phiC31. Genetics 166, 1775–1782
26 Wimmer, E.A. (2005) Insect transgenesis by site-specific recombination. Nat. Meth.
2, 580–582
27 Scott, T.W., Rasgon, J.L., Black, W.C.I. and Gould, F. in Bridging laboratory and
field research for genetic control of disease vectors. (eds. B.G.J. Knols and C. Louis)
171–181 (Frontis, Wageningen, NL; 2005).
28 Handler, A.M. (2002) Prospects for using genetic transformation for improved SIT
and new biocontrol methods. Genetica 116, 137–149
29 Robinson, A.S. et al. (2004) Insect transgenesis and its potential role in agriculture
and human health. Insect Biochem. Mol. Biol. 34, 113–120
30 Wimmer, E.A. (2005) Eco-friendly insect management. Nat. Biotechnol. 23,
432–433
31 Wimmer, E.A. (2003) Innovations: applications of insect transgenesis. Nat. Rev.
Genet. 4, 225–232
32 Snook, R.R. (2005) Sperm in competition: not playing by the numbers. Trends Ecol.
Evol. 20, 46–53
33 Saul, S.H. (1982) Rosy-like mutant of the Mediterranean fruit fly, Ceratitis capitata
(Diptera: Tephritidae), and its potential for use in a genetic sexing program. Ann.
Entomol. Soc. Am. 75, 480–483
34 Baruffi, L. et al. (1995) Polymorphism within and between populations of Ceratitis
capitata: comparison between RAPD and multilocus enzyme electrophoresis data.
Heredity 74, 425–437
RESEARCH PAPERNew Biotechnology?Volume 25,Number 1?June 2008
84
www.elsevier.com/locate/nbt
Research Paper