![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
|
Vol. 9, Issue 5, 471-481, May 1999
LETTER
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ABSTRACT |
|---|
 Top
 Abstract
 Introduction
RESULTS
DISCUSSION
METHODS
References
|
|---|
Drosophila melanogaster larvae usually react against eggs of the parasitoid wasp Leptopilina boulardi by surrounding them with a multicellular melanotic capsule. The genetic determinism of this response has been studied previously using susceptible (non-capsule-forming) and resistant (capsule-forming) strains. The results suggest that differences in their encapsulation response involve a single gene, resistance to Leptopilina boulardi (Rlb), with two alleles, the resistant one being dominant. Rlb confers specific protection against Leptopilina boulardi and is thus probably involved in parasitoid recognition. Recent studies have localized this gene on the right arm of the second chromosome and our aim was to precisely determine its genetic and molecular location. Using strains bearing deletions, we demonstrated that resistance to Leptopilina boulardi is conferred by the 55C; 55F3 region and that the 55E2-E6; F3 region is particularly involved. A physical map of the 55C; 56A region was then constructed, based on a set of overlapping cosmid and P1 phage clones. Using single and double digests, cross hybridization of restriction fragments, and location of genetically mapped genes and STSs, a complete, five-enzyme restriction map of this 830-kb region was obtained.
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
RESULTS
DISCUSSION
METHODS
References
|
|---|
Insects possess a complex immune system, with both humoral and
cellular components to protect themselves from a foreign
entity. Bacterial invasion for example induces the
production of an array of peptides with antimicrobial or antifungal
activity as well as the activation of the prophenol oxydase cascade
(Hoffmann 1996
; Hoffmann and Reichart 1997). In the case of parasitism
by a larger parasite, an encapsulation response is provoked that
involves the deposition of eumelanin resulting from the prophenol
oxydase system activation as well as the adhesion of numerous hemocytes around the egg of the parasite (Carton and Nappi 1997). No experimental data have clearly identified the components of the melanotic
encapsulation response that kill the parasite, even if production of
cytotoxic radicals is probably involved (Nappi et al. 1995), but the
presence of the capsule always indicates the failure of parasitism.
Despite the recent advances in understanding these immunological
processes and the demonstration that the immune response is not
completely aspecific (Lemaitre et al. 1997), little is known about how
insects recognize a parasite and which factors underlie the specificity of the response.
Drosophila melanogaster is parasitized by several wasps
including Leptopilina boulardi (Cynipidae) and Asobara
tabida (Braconidae). In both of these systems, virulent
(immune-suppressive) and avirulent (non-immune-suppressive) strains of
parasites have been obtained. Virulent strains always succeed in
escaping or suppressing the host immune response, whereas parasitoids
from avirulent strains are encapsulated by some of the hosts. The use
of an avirulent strain has allowed selection of two isofemale lines of
D. melanogaster with different immune responses to L. boulardi (Carton and Boulétreau 1985). Flies from the
resistant (R) strain encapsulate eggs of the parasite within
24 hr after infestation, whereas flies from the susceptible
(S) strain do not exhibit the encapsulation response. Experiments using R and S strains have demonstrated
that induction of the phenoloxydase cascade by parasitism occurs only
in the R strain (Nappi et al. 1991). The resistance gene(s)
thus act at an early phase in the immune response. Parasitism does not induce the production of antibacterial peptides in the R
strain, which indicates that the antibacterial and the antiparasite
immune responses are controlled by different pathways (Cousteau et al. 1996). Very interestingly, comparative infestation experiments with
L. boulardi and A. tabida have shown that host
resistance is highly specific (Carton and Nappi 1997). The S
strain susceptible to L. boulardi is able to encapsulate the
eggs of A. tabida, demonstrating that it is not
immune-incompetent but simply unable to recognize L. boulardi
(Vass et al. 1993). The molecular mechanisms sustaining this
specificity are currently completely unknown. Characterization of the
resistance gene(s) and determination of its function would greatly help
in understanding how insects recognize the nonself.
In the systems studied, the resistance of D. melanogaster to
avirulent parasitoids has been demonstrated to be autosomal and monogenic. Two major resistance genes have been described that confer
resistance to L. boulardi
resitance to L. boulardi (Rlb) (Carton and Nappi 1997) and A. tabida-resistance to A. tabida (Rat) (Orr et
al. 1997; Benassi et al. 1998), respectively. These genes are dominant
and both localized on the second chromosome of D. melanogaster. The genetics of resistance to Leptopilina has been further studied using the S and R strains
and Rlb has been localized in the region 55-56
(Carton and Nappi 1997). It should be noted that the 55A-56A
region of the Drosophila genome contains several genes
involved in immunological processes. The prophenol-oxydase-encoding
gene, A1, has been located in 55A (Fujimoto et al.
1995), the immune deficiency (imd) gene (Lemaitre et
al. 1995) in 55CD, and the gene encoding the diptericin
antibacterial peptide in 56A (Wicker et al. 1990).
This paper reports genetic experiments using strains bearing deletions, which allowed us to obtain a more precise localization of Rlb. We demonstrated that deletion of the 55C; 55F3 region and especially of the E2-E6; F3 region affects the resistant phenotype considerably. Further characterization of the 55C-56A region was then undertaken by taking advantage of the existence of two D. melanogaster genomic libraries, based on cosmid and P1 phages. A complete, fine restriction map of the Rlb-containing region was obtained that constitutes an essential tool for the molecular cloning of the Rlb gene.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
RESULTS
DISCUSSION
METHODS
References
|
|---|
Genetic Localization of Rlb
As mentioned above, classical genetic studies have shown that Rlb is completely dominant. Nevertheless, using deficiency (Df) strains bearing deletions in the 55-56 region of the second chromosome [Df(2R) strains], we observed that, unexpectedly, when the R allele of the Rlb gene (including its regulatory regions) is facing a deletion, the resistant phenotype is disturbed. The important implication of this phenomenon for our studies is that deletions can be useful for localizing the dominant Rlb gene. To perform this localization, five D. melanogaster strains bearing deletions were used.
Individuals of each deleted strain [Df(2R)/CyO] were crossed
with individuals of the R strain, homozygous for Rlb.
F1 hybrids [Df(2R)/R and CyO/R] were then
infested by L. boulardi as were control individuals from the
deleted strains. The number of individuals used in each experiment is
reported Figure 1. The controls were used to confirm
the infestation ability of the parasitoid and the susceptible phenotype
of the deleted strains. The encapsulation ability was determined by
dissecting larvae and recording melanotic encapsulation, and the
F1 encapsulation rates (ER) were calculated as stated by
Carton et al. (1992). At the larval stage, when infestation occurs,
CyO/R individuals cannot be distinguished from
Df(2R)/R individuals. The ER were then corrected, considering
that half of the F1 progenies consist of
CyO/R-resistant (capsule-forming) larvae.
|
The localization of the right and left limits of the five deletions were obtained in the FlyBase databank. These data have sometimes be specified or changed in the course of our experiments: For example, the deleted region in Df(2R)Pc4 considered previously as 55A1-55F was further specified as 55A1-55F3. When in doubt, the different data were taken into account.
Encapsulation rates recorded for the Df(2R)Pc4,
Df(2R)Pc17B, Df(2R)Pc66, Df(2R)P34, and
Df(2R)P111B strains were, respectively, 4.8% (126 individuals), 0% (249 individuals), 0.02% (124), 3.9% (152 individuals), and 0% (32 individuals). As these results do not differ
significantly from the ER usually obtained in the S strain
(Carton et al. 1992), the deleted strains were considered as
susceptible subsequently.
The length of the region involved in the resistance phenomenon was
first defined using the Df(2R)Pc4 strain: The encapsulation rate of Df(2R)/R heterozygous individuals fell under 20% as
compared with 80%-95% found in the R homozygous strain or
in the F1 between the R and S strains (Fig.
1; Carton et al. 1992). This demonstrated that the 55A1;
55F3 region is involved in resistance to L. boulardi. This region was then restricted to 55C1; 55F3 using
the Df(2R)Pc17B and Df(2R)Pc111B strains. The
corrected ER obtained in their F1 progenies with the
R strain were of the same order as the ER of the R
strain itself (94.7% and 91%, respectively; Fig. 1). These results
showed that deletions of the 54E8-F1; 55B9-C1
[Df(2R)Pc17B] or the 55A1; 55C1-C3
[Df(2R)Pc111B] region do not impair the resistant phenotype.
On the contrary, the corrected ER was impaired greatly (52%) when
hybrids between the R strain and Df(2R)P34 were
infested. The region deleted in this strain had been first located in
55E2-4; 56B2-C1 but recent FlyBase data report a
55E6-F3; 56C1-C11 localization. In combination with
the previous results, this allowed us to conclude that deletion of the
E2-E6 F3 region has a great impact on the ability
of Drosophila to recognize and encapsulate
Leptopilina eggs. Finally, F1 hybrids between the
R strain and the Df(2R)Pc66 strain were infested. The
corrected encapsulation rate was 100%. This deletion had been first
localized to the 55D2-E1; 55E3-E4 region and is now
reported to be in the 55D2-E1; 56B2 position. Nevertheless, other experiments reported only a slight alteration of
the D region in the Pc66 strain (Georgel, pers.
comm.), and we considered subsequently a 55D2-E1;
55E1-E4 localization. As a consequence, our main conclusion
is that the E2-E6; F3 region is clearly involved in
resistance. Nevertheless, the ER is significantly higher in
R/P34 hybrids (52%) as compared with R/Pc4 hybrids
(19%), which indicates that another region, located on the left side of the Df(2R)Pc66 deletion (55C1-C3;
55D2-E1) or on its right side (55E1-E4) could also
have an impact on the resistance phenotype.
Mapping of the 55C-55F Region
Covering of the Region
The genome of D. melanogaster is studied by the European Drosophila Genome Program (EDGP) using cosmid clones (Siden-Kiamos et al. 1990; Kafatos et al. 1991) and by the Berkeley
Drosophila Genome Program (BDGP) using P1 phage clones
(Smoller et al. 1991; Hartl et al. 1994; Kimmerly et al. 1996). The
results on the genetic localization of Rlb interested us in
the mapping of the 55C-55F region. Two contigs of P1 clones
were available in this region, otefin (55BC) and
diptericin (55F-56AB) but the region between the two
was not covered. Furthermore, some of the clones had not been
localized, the position of others had to be confirmed, and the limits
of the complete region had to be determined (Rubin 1996).
Cosmid Map
We first attempted to recover the complete 55C-55F region using the cosmid library. Thirty-one cosmids predicted to cover this region or not localized yet were obtained from I. Siden Kiamos [Foundation for Research and TechnologyHellas (FORTH) Heraklion, Crete]. Their DNA was digested with BsrgI and NotI
enzymes and transferred. BsrgI and NotI cut,
respectively, at the positions 91 and 5144 in Lorist6 (5156 bp) and a
5-kb restriction fragment containing almost the entire vector is
obtained (Cross and Little 1986; Gibson et al. 1987). This fragment is
thus the only one common to all the clones. The blots were hybridized
with radioactively labeled DNA of each clone and cross-hybridizations
were analyzed. Cosmids containing repeated DNA were detected by
high-stringency hybridization of the Southern blots with radioactively
labeled Drosophila genomic DNA. Several bands were detected
for five cosmids (86B6, 152H3, 65H8, 195C7, 128B5) which indicated that
they contain repeated DNA. This result was taken into account during
the analysis. Finally, four contigs containing, respectively 2, 4, 11, and 5 cosmids were determined but the complete region could not be
obtained (Fig. 2). The use of the P1 library was then
necessary to confirm cosmid overlaps and solve remaining gaps.
|
In Situ Localization of Cosmids
Five cosmids, chosen to define the limits of the 55C-55F region, were hybridized on Drosophila polytene chromosomes. Their localization was compared with data from the EDGP. As expected, 119C2, 27B7, and 59G5 hybridized in 55C1-2/C3/C4-5, 55D1-2/D3, and 55 F5-F11/56A, respectively, which indicated the limits of the entire 55C-55F region (Figs. 2 and 3). On the contrary, the 55G2 clone, localized in 56A1-B7 by the EDGP program, was found to hybridize in 55F1-2/F3. Sequence data found on a 55G2T STS in the EMBL databank (accession no. Z5065) were analyzed with a BLAST program (Infobiogen Bisance, Dessen et al. 1990). The results demonstrated
clearly that 55G2 shares homologies with a P1 clone localized in
52E1-E8 (DS03910) and with sequences of the retrotransposon
Burdock (EMBL accession no. U89994). Therefore, the contradictory
results are most probably caused by differences between strains of
D. melanogaster concern ing the presence of this transposable
element. This confirms the high rate of D. melanogaster middle
repetitive DNA as a source of false positives in clone localization. The 55G2 clone was not further considered. A last clone, 75A9 had been located both in 90E and 55C
but only the first localization (90E) was confirmed.
|
P1 Map
Thirty-three P1 clones believed to cover the 55C-56A region were obtained from M. Ashburner and C. McKimmie (University of Cambridge, UK). Their DNA was digested with SfiI and NotI enzymes and transferred. SfiI and NotI cut at the 39 and 15978 positions, respectively in the Ad10SacBII vector and a 16-kb restriction fragment containing almost the entire vector was obtained (Pierce et al. 1992; Sternberg 1990, 1994). This
fragment is thus common to all the clones with the exception of the P1
8374, 2561, 1552, and 2599 which correspond to the pNS582 Ad10tet14
vector. Radioactively labeled DNA of each P1 or cosmid clone was
hybridized on Southern blots of all the P1 clones and Southern blots of
all the cosmid clones. Controls showing hybridization patterns of the
two vectors (Lorist6 and Ad10SacBII) onto these blots were also
realized. Cross hybridizations were finally analyzed to determine the
P1 contigs. As shown in Figure 2, the three gaps found in the cosmid map were solved with P1 clones, and the cosmid contigs were confirmed. Nevertheless, the 55C-55F region could not be covered
completely using only the P1 library as three contigs (4, 4, and 10 P1)
were detected. The two gaps were solved with cosmid clones. The
55BC-56A contig, made of P1 and cosmids is represented Figure 2.
Restriction Map of the Region
The restriction mapping of the region with five enzymes was realized to obtain a chromosomal walk, and with the objective of future subcloning and sequencing of the region. The DNA extracted from each cosmid and P1 clone located in the region was digested with NotI, BsrgI, XhoI, AscI, and SfiI and double-digested with NotI-BsrgI, NotI-XhoI, XhoI-BsrgI, AscI-XhoI, and XhoI-SfiI. Migrations were done on both 0.7% (or 0.5%) and 1.2% agarose gels. Southern blots were hybridized with radioactively labeled DNA of the vector and the clone itself and the restriction map of each clone was drawn. Genomic DNA fragments inserted into the Lorist cosmid vector ranged from 35 to 45 kb and fragments inserted into the Ad10SacBII P1 vector from 80 to 85 kb. Southern blots of each clone were then hybridized with probes of other clones located in the same part of the map. This allowed us to detect cross-hybridizing fragments between the clones. When necessary, fragments corresponding to the ends of inserted DNA were used as probes on Southern blots of other clones to detect small cross hybridizations. The complete restriction map of the 55BC-56A region finally drawn from covering of individual maps is ~830 kb long and contains 357 restriction sites (Fig. 4).
|
Genes and STS Localization
The 55C-56A region was analyzed further by localizing STSs and genes predicted to be in the region. FlyBase databank localization of the STSs Dm0406, Dm0810, Dm0556, Dm2546, Dm0881, Dm0843, Dm406, Dm0835, and Dm1275 was, respectively, 55C1-2, 55D1-D2, 55E6-E7, 55E9-F1, 55F1-F2, 55F1-F2, 55F6-F13, and 56A1-A3. Fragments corresponding to these STSs were obtained by PCR experiments, radioactively labeled, and hybridized on Southern blots of all the clones located in the same part of the map. As shown on Figure 4, the chromosomal walk was confirmed by the fine localization of these STSs. Nevertheless, a clear discrepancy was sometimes recorded between in situ localization data of STSs and clones. According to FlyBase data, a few genes localized in the 55C-56A region or near this region have been cloned and sequenced. Fragments of these genes were cloned and sequenced following amplification with specific primers. Two fragments (222 and 780 bp) were obtained for otefin (55C1-C13), a 395-bp fragment was obtained for diptericin (56A1-A3), a 195-bp fragment was obtained for enabled (56B), a 270-bp fragment was obtained for coracle (56C), a 170-bp fragment was obtained for 5HT1-A (56AB), a 250-bp fragment was obtained for 5HT1-B (56AB), and a 1021-bp fragment was obtained for three rows (55A1-A4). All these DNA fragments were radioactively labeled and used as probes on Southern blots of cosmid and P1 clones. The precise localization of otefin, diptericin, 5HT1-A, and enabled is reported Figure 4. The three rows fragment was found to be located on the 128B5 cosmid which is not part of the studied region but is most probably located in 55A. No hybridization was detected for 5HT1-B and coracle. The localization of the diptericin and otefin genes on the 1962 and 2561 P1 clones confirmed the limits of the 55C-56A contig. 5HT1-A and enabled are located outside this region, in 56B.| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
RESULTS
DISCUSSION
METHODS
References
|
|---|
Genetic Localization of Rlb
D. melanogaster resistance to parasitism by L. boulardi is determined by a major gene, Rlb, located on
the second chromosome, and whose expression is completely dominant.
Nevertheless, preliminary experiments designed to localize Rlb
demonstrated that some deletions in a heterozygous form are able to
disturb the otherwise Rlb-dominant phenotype, that is, the
encapsulation ability. No molecular hypothesis explaining this
phenomenon has been proposed, yet but it might be reminiscent of the
transvection effects recorded in Drosophila. Transvection as
"the complementation of heteroalleles allowed by their proximity"
was first described by Lewis (1954) for the Ultrabithorax
(Ubx) locus. Where alleles subject to transvection have been
characterized, it appeared that one allele carried a mutation in the
regulatory region of the gene, and the other in its coding region. It
is thus likely that transcription of the intact coding region on one
homolog is activated by the intact enhancer on its pairing partner
(Bender et al. 1983). When the heteroalleles are no longer able to
interact, as is the case following chromosomal rearrangements, the
transvection is disrupted. In the case of Ubx, it was
demonstrated recently that such rearrangements reduce the expression of
both alleles and not only the one encoding the active product
(Goldsborough and Kornberg 1996). This result suggests strongly that
somatic pairing is not only involved in the rescue of gene expression
from mutant alleles, but may also play a role in the normal expression
of this gene. This phenomenon might occur by a variety of molecular
mechanisms and Henikoff (1997) refers to the ability of a gene to sense
its paired state by the umbrella term of "trans-sensing." For
example, the occurrence under certain circumstances of physical
interactions between enhancer and promoter sequences located on
homologous chromosomes has been pointed out by Geyer et al. (1990).
Our working hypothesis is that such somatic pairing might have a role in the expression of Rlb. Chromosomal deletions including the Rlb region could prevent the usual short range pairing between the Rlb alleles (including regulatory regions) and result in a dramatic reduction in the level of expression of the Rlb allele. The lowered level of the Rlb+ allele product would be responsible for the reduced rate of encapsulation in the R/Df(2R) heterozygous strains.
Whatever its molecular underlying, this phenomenon allowed us to precisely localize the region involved in resistance. The five strains bearing deletions that we used were first demonstrated to be susceptible to parasitism by Leptopilina. Then, the encapsulation rate of the hybrids R/Df(2R) were obtained and compared, taking into account the fact that half of the F1 progenies must be of CyO/R genotype and therefore of resistant phenotype. An important number of individuals has been tested (Fig. 1) and the differences recorded between the ERs were high enough to show immediately if a deletion has an impact on the resistance phenotype or not.
When using deletions in genetic studies, the accuracy of in situ localization of their limits is of great importance. The limits of the deletions that we used were found in the FlyBase databank but some of them had been refined or changed in the course of our experiments. When determining the Rlb-containing region, we used all the known data to keep it as large as possible. The ERs were as high as in the R strain for three of the four deleted strains: thus, the 55A; 55C1-3 and 55D2-E1; 55E1-4 regions seem not to be involved in the resistance phenotype. On the contrary, the 55E2-6; 55F3 region is likely to be implied since the R/Df(2R)P34 individuals showed an ER of only 52%. The fact that resistance might not be determined completely by this region is suggested by the comparison of this ER (52%) with the ER (19%) of individuals heterozygous for the 55A-55F3 deletion (Fig. 1). The 55C1-C3; 55D2 region could be responsible for this effect. Nevertheless, the right limit of the Df(2R)Pc66 deletion and the left limit of the Df(2R)P34 are not defined precisely. Part of the 55E2-E6 region may not be included in the Pc66 or Pc34 deletions and play a role in determining the resistant phenotype.
The detailed interpretation of these data will remain difficult until the trans-sensing effects will be better understood. For example, the ER of R/Df(2R)Pc4 individuals (19%) is not in the same range as the ER of the S strain (0%-5%) but this could result from either the way deletions affect the expression of Rlb on the homologous chromosome or it could indicate that another region outside the 55A-55F3 part of the chromosome is also involved in the resistance phenomenon. Finally, resistance to Leptopilina appears to be determined in a great part by the 55E2; 55F3 region even if the involvement of the 55C1-C3; 55D2-E1 region can not be excluded completely.
Physical Mapping of the 55C-55F Region
To map this region we used a cosmid library (EDGP) and a P1 phage library (BDGP) as well as numerous data classified in FlyBase including in situ localization data. For instance, the diptericin P1 contig was known to cover 55F-56A and the otefin contig was located in 55B-C. Both cosmid and P1 clones, located in the region or not localized yet, had to be used to obtain the complete 55C-55F region. Having identified the clones containing repeated DNA, we used direct cross-hybridization and restriction mapping to determine the contigs. The main limitation of this bottom-up mapping technique is the difficulty to detect small overlaps between clones. Nevertheless, unique restriction sites were available in the vectors that allowed us to separate them from the inserted DNA and comparison of the restriction maps allowed us to determine clone overlaps with good accuracy. To solve remaining ambiguities, the ends of some clone inserts were prepared as probes and used in hybridization experiments.
Four contigs covering the 55C-55F region were obtained with
the cosmids and the gaps (75, 30, and 45 kb) were further solved using
P1 phages. Using the P1 library, three contigs were determined in the
55BC-56A region. Again, the two gaps (75 and 25 kb) were covered by cosmid clones. Therefore, the limit of the bottom-up technique was not responsible for the nonrecovering of the complete region from a given library. The fact that some gaps remain in the
cosmid map as well as in the P1 map of the 55C-56A region can
be explained in two ways. First, even if all the clones located in this
region have been a priori provided, other clones corresponding to the
missing regions may be contained in the libraries. Nevertheless, the
fact that different regions are missing in each library seems more
likely. The cosmid library represents four times the genome of D. melanogaster (Siden Kiamos et al. 1990) but instability has been
described for some inserts that are replicated by multicopy replicons
and this could account for the failure to recover certain genomic
segments in cosmid cloning systems (Smoller et al. 1991). One of the
advantages of the P1 vector is supposed to be the avoidance of this
instability with the use of a one-copy replicon and the framework map
obtained by Hartl et al. (1994) included 85% of the euchromatic
genome. Nevertheless, it can be noticed that in their chromosome 2R
map, no clones were obtained for part of the 55A band.
To confirm the limits of the recovered region, we used two complementary approaches: in situ mapping of some of the clones and localization of known STSs and genes on the physical map. The right limit of the region was confirmed by in situ localization of 59G5 in 55F5-F13/56A, and of the diptericin gene (56A) on P1 1962. The localization of 27B7 in 55D1-2/D3 and of the STSs Dm0810, Dm0556, Dm2546, Dm0881, Dm0406, Dm0843, Dm0835, and Dm1275 gave confirmation of different parts of the map. Finally, the left limit of the covered region was obtained with in situ localization of 119C2 in 55C1-2/C3-4/C5 and of the otefin gene (55C1-C13) on the 27G2 and 9D5 cosmids. FlyBase localization of the P1 8860 at the left end of our contig is 55B. Occurrence of false localization of clones because of middle repetitive DNA was demonstrated for the cosmid 55G2.
The 55C-56A region has been mapped with three rarely cutting
enzymes, AscI, NotI, SfiI, and two
frequently cutting enzymes, XhoI and BsrgI. A total
of 357 restriction sites was obtained. The size of P1 inserts varied
from 80 to 85 kb, whereas the cosmid inserts size ranged from 35 to 45 kb, which did not differ from the expected 85 and 35 kb (Siden Kiamos
et al. 1990; Smoller et al. 1991). Because of the number of
BsrgI fragments, the relative position of a few neighbored
BsrgI-BsrgI fragments remained uncertain. Nevertheless, the opportunity to test the restriction maps arose recently with the publication by the BDGP of sequence data regarding six P1 clones located in 55F-56A (1552, 9119, 8204, 7069, 2599, and 1962) and three P1 clones in 55BC (8374, 2561, and
8860). Only one inversion between two BsrgI-BsrgI
fragments was recorded.
The complete restriction map is reported Figure 4 with the precise
localization of the diptericin, otefin,
5HT1-A, and enabled gene fragments. The size of the
mapped 55BC-56A region is ~830 kb. According to Sorsa
(1988), the total DNA contents of the band 55 is 1316 kb (with
284 kb in 55C, 103 kb in 55D, 298 kb in 55E, and 220 kb in 55F). The 56A band would represent 93 kb and the complete 55C-56A region would then cover ~900
kb. The 70-kb difference could be explained partly by the fact that the
56A region is not recovered completely in our contig but on
the other part, we also included part of the 55B region in our
map. It is then likely that the DNA content in each band had been
slightly overestimated previously.
Based on our physical map, we estimated the maximal size of the
E2; F3 region as ~300 kb. The expected number of
genes in the Drosophila genome ranges from 12,000 (Miklos and
Rubin 1996) to 43,000 (Louis et al. 1997), this last number probably
being overestimated. As the size of the euchromatic genome is
1.2 × 108 bp, one to three genes should be found every
10 kb and the region could contain 30-90 genes. Nevertheless, the size
of this region is probably overestimated because of the limits of in
situ hybridization data and because of the possible discrepancy between
the localization of the clones and the localization of the limits of
the deletions. Our first aim will be to determine which of the clones
are included in which deletion.
To carry on the cloning project of Rlb, we have the choice
between a genetic and a physical approach. For instance, a genetic study was chosen by Collins et al. (1997) for the pen1 gene
responsible for the encapsulation of malarial parasites. In our case,
some sequences of the 55E-55F region are available. Our
future efforts will be based on the analysis of these sequences and on
the characterization of the regions of the contig expressed at the
larval stages when parasitism occurs. The potential function of the
genes determined will be considered, taking into account the recent
advances underlining the homologies between the immunological responses
of mammals, insects, and plants (Baker et al. 1997; Dushay and Eldon
1998; Vilmos and Kurucz 1998). Depending on the number of clones
finally retained and the estimated number of candidate genes, it will be possible to decide if a genetic strategy involving the use of
microsatellites markers will be or not necessary.
Whatever the final strategy, this physical mapping represents the first step toward the cloning of an insect resistance gene to its parasitoid, a type of gene that has never been cloned to date.
| |
METHODS |
|---|
|
Top
Abstract
Introduction
RESULTS
DISCUSSION
METHODS
References
|
|---|
Origin of Insect Strains
The origin of the avirulent strain of L. boulardi (Gif
stock no. 486) and the resistant D. melanogaster strain (strain 940), as well as their rearing conditions have been described elsewhere (Carton et
al. 1992). The Df(2R)Pc4, Df(2R)Pc17B, Df(2R)P34, Df(2R)Pc66, and
Df(2R)Pc111B fly strains were obtained from the Bloomington Stock Center.
Bioassay Procedures
The working hypothesis was that chromosomes bearing the resistant
allele could display lower resistance levels when combined with
Df chromosomes missing the Rlb gene (or its
regulatory regions). To test this hypothesis, crosses were performed
between each deleted strains [Df(2R)] and the resistant
R strain. Hybrid progenies as well as individuals from the
deleted strains were submitted to parasitism by exposing batches of 50 second-instar larvae hosts to L. boulardi females. Infestation
and rearing of infested larvae were conducted as described in Carton et
al. (1992). Dissection of infested larvae were carried out 3 days after
infestation to determine the encapsulated or nonencapsulated status of
the parasitoid egg. The encapsulation rate ER was calculated as the
ratio of the encapsulated egg number to the recovered egg number.
Superparasitized larvae were included in the counts, as encapsulation
rate do not vary if calculated with monoparasitized or superparasitized
larvae. The ER of transheterozygous individuals bearing the deletion
was calculated considering that they represent only 50% of the
F1 progeny. Any significant change from a 100% value could
be interpreted as a role of the deleted region in this character.
Origin of Clones and Experimental Procedures
Cosmids were provided by I. Siden Kiamos and originated from a
library made of OregonR adult DNA (Siden Kiamos et al. 1990). P1 phages
were provided by C. MacKimmie and M. Ashburner and originated from a
library made of y, cn bw sp strain adults DNA
(Smoller et al. 1991). In situ localization of P1 clones was found in
Hartl et al. (1994) or provided by the BDGP. In situ localization of cosmid clones was provided by the EDGP. Data on the contigs assembled by STS mapping come from Kimmerly et al. (1996). Data concerning the
STSs and gene sequences were found in the FlyBase databank. The P1
sequences were obtained from the BDGP databank.
Unless otherwise indicated, all molecular procedures were performed as
described by Ausubel et al. (1994).
Cosmid and P1 DNA Extraction
Cosmid clones were grown in 2YT-kanamycin broth as indicated in Siden-Kiamos et al. (1990). P1 clones were grown in LB-kanamycin broth
for 3 hr, then supplemented with 5 ml of 0.1 M IPTG and shaken for another 3 hr, as reported in FlyBase. Clone and vector DNA was isolated by an alkaline extraction procedure followed by
CsCl-gradient separation. Standard techniques including purification by
ultracentrifugation on ethidium bromide-cesium chloride gradients were
used for extraction of genomic DNA.
Obtaining Contigs and Restriction Mapping
Standard techniques were used for DNA digestion with restriction enzymes (1 µg of DNA in each lane), gel electrophoresis (0.5% or 0.7% and 1.2% Seakem agarose) and Southern blotting onto Nylon+ (ICN products) membranes. The probes were random-primed labeled with [
-32P]dATP (ICN products) using the Klenow fragment of
DNA polymerase I (Promega) and used at a concentration of 106
cpm/ml of hybridizing solution. Hybridizations were carried out at
65°C in 0.5 M at pH 7.2 Na2HPO4-NaH2PO4, 7% SDS, and 1 mM EDTA. The final washing was done in 0.2× SSC, 0.1% SDS
at 65°C. Hybridized filters were autoradiographed with Fuji RX films
at 
80°C. Restriction maps were drawn using Plasmid Artist TM
1.13 software.
Fragments corresponding to the end of clone inserts were obtained by
standard fragment extraction procedures (Ausubel et al. 1994),
quantified, radioactively labeled, and used as probes.
Localization of STS and Gene Fragments
Fragments corresponding to STS DM0881, 0843, 0556, 0835, 1275, 2546, 0810, and 0406 were obtained by PCR experiments using the primers and amplification conditions described in FlyBase. Fragments of genes located in the region were obtained by PCR using 300 ng of D. melanogaster genomic DNA. The DNA was dissolved in 10 mM Tris-HCl at pH 9, 3 mM MgCl2, 50 mM KCl, 0.1% TritonX-100, 150 mM each dATP, dCTP, dGTP, and dTTP, 0.1 mM of each oligonucleotide, in a 100-µl reaction volume with 5 units of Taq Polymerase (Promega). Each PCR was carried out in a programable thermal controller (Perkin-Elmer) for 30 cycles. The cycle was: denaturing at 94°C for 1 min, annealing for 1 min, and extending at 72°C for 30 sec. At the end of the thirtieth cycle, the heat denaturation step was omitted and extension was allowed to proceed at 72°C for 3 min. Primers containing restriction sites (EcoRI, BamHI) were used to obtain the following fragments: a 1065-bp three rows fragment (NTc1 5'-GGAATTCCGACGGCACTGCATATTGG-3' and NTc2 5'-CGGGATCCCGCAATCAATAGACGTTT-3' TA 61°C), a 395-bp diptericin fragment (NDipt1 5'-GGAATTCCCTGCAGCAAAGGTATCA-3' and NDipt2 5'-CGGGATCCCGAAGCTTAGAAATTCGGA-3' TA 57°C), a 195-bp enabled fragment (Nena1 5'-GGAATTCCCTGCAGCAGTTCAAGCTC-3' and Nena2 5'-CGGGATCCCGGATCCTTCTTTTCTGC-3' TA 60°C), a 270-bp coracle fragment (Ncora1 5'-GGAATTCCGAGACGCCCACATCCG-3' and Ncora2 5'-CGGGATCCCGTAGTCGCCCATCTCCG-3' TA 64°C), a 170-bp 5HTA-1 fragment (NHTA1 5'-GGAATTCCCTGCAGCGTATCGAGCA-3' and NHTA3 5'-CGGGATCCCGTCGACGATGGATGCGTT-3' TA 61°C), a 250-bp 5HTB-1 fragment (NHTB1 5'-GGAATTCCTGCAGAACAGTGATCGGAG-3' and NHTB2 5'-CGGGATCCCGGATCCGGGTTATGCAAAAT-3' TA 60°C). The primers Ote1 5'-GAGACGCCCACAGATCCG-3' and Ote1b 5'-CGTAGTCGCCCATCTCCG-3' were used to obtain a 220-bp otefin fragment and the Ote1 and Ote2 5'-AGTGCGACCCTTGTAGCG-3' primers were used to obtain a 627-bp otefin fragment. In both cases, the TA was 52°C. The DNA fragments were eluted using a Promega PCR Prep Kit, and the concentration of each was estimated on agarose gels. The otefin fragments were subcloned with the pGEM-T vector (Promega) and the other fragments with a M13mp18 phage vector digested with BamHI and EcoRI. Each DNA fragment was sequenced, for both strands, in a Li-Cor automatic sequencer, using a Sequitherm Excel II long-read sequencing kit (Epicentre Technology) and labeled universal and reverse IRD41 primers in the conditions described by the suppliers. Sequences were analyzed with Infobiogen Bisance programs (Dessen et al. 1990).
STS and gene DNA fragments were labeled with 32P by random
priming and used at a concentration of 106 cpm/ml of
hybridizing solution. Hybridizations and washings were carried out as
described previously.
In situ hybridization
In situ hybridization of biotinylated probes (Boehringer kit) to salivary gland polytene chromosomes was adapted from Engels et al. (1986). This technique permits identification of the sites of
homologous DNA.
| |
ACKNOWLEDGMENTS |
|---|
We thank I. Siden-Kiamos and C. MacKimmie for providing the cosmid and P1 clones, P. Georgel for personal communication on the Df(2R)Pc66 limits, and E. Huguet and J.M. Drezen for helpful comments on the manuscript. This work was supported by grants from the CNRS (UPRES-A 6035), the Ministère de Enseigment National de la Recherche et de la Technologie (MENRT) (Génome et Interactions durables). We are grateful for the assistance provided by European Community (EC) (grant no. AIR36CT9-1433) and the French CNRS-INRA program Biodiversité. M. Hita is supported by an EC grant.
The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
| |
FOOTNOTES |
|---|
3 Corresponding author.
E-MAIL poirie{at}univ-tours.fr; FAX 33247366966.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
RESULTS
DISCUSSION
METHODS
References
|
|---|
Received October 26, 1998; accepted in revised form March 24, 1999.
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. P. Sorrentino, J. P. Melk, and S. Govind Genetic Analysis of Contributions of Dorsal Group and JAK-Stat92E Pathway Genes to Larval Hemocyte Concentration and the Egg Encapsulation Response in Drosophila Genetics, March 1, 2004; 166(3): 1343 - 1356. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
