| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Tropical Medicine, and Biomedical Sciences Program, Tulane University Health Sciences Center, New Orleans, Louisiana, USA;
Division of Infectious Diseases, Queensland Institute of Medical Research, Brisbane, Queensland, Australia; and
Department of Microbiology, Immunology and Tropical Medicine, George Washington University Medical Center, Washington, DC, USA
1Correspondence: Microbiology, Immunology and Tropical Medicine, George Washington University Medical Center, Ross Hall, Rm. 448, 2300 Eye St., NW, Washington, DC 20037 USA. E-mail: mtmpjb{at}gwumc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: schistosome • transgenesis • VSVG • luciferase • provirus • functional genomics
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. Indeed, transgenesis techniques have been developed for a number of model and medically and economically important species, including the malaria parasite, mosquito vectors of infectious diseases, and human cell lines (3
4
5)
. By contrast, transgenic technologies and tools for schistosomes and other parasitic helminths are not as well advanced (6
7
8
9
10
11
12
13)
. Among the potential tools for gene manipulation, transgenesis mediated by retroviruses offers a potentially tractable method to introduce foreign genes, a means to determine the importance of schistosome genes, including those that could be targeted in novel interventions and the potential to undertake large-scale genetic analysis by insertional mutagenesis.
Schistosomiasis is considered the most important of the human helminthiases in terms of both morbidity and mortality (14
, 15)
. Advances in molecular genetics and immunology hold the promise to control the spread of schistosomiasis and to guide development of new tools to combat this neglected tropical disease. At present, control of schistosomiasis largely relies on chemotherapy with praziquantel but wide-scale use of this medication has led to concerns about development of drug resistance (16)
. No vaccine is yet available for control of schistosomiasis, although new antigens are being developed (17)
. It is anticipated that the genome sequences of two of the main species of schistosomes that parasitize humans will be published soon, Schistosoma japonicum and Schistosoma mansoni, following on reasonably comprehensive descriptions of their transcriptomes and proteomes (18
19
20
21)
.
We modified a murine leukemia virus (MLV) vector to incorporate the reporter gene firefly luciferase under control of endogenous schistosome gene promoters, including the actin promoter, which was then incubated with developmental stages of S. mansoni, including schistosomula. The vesicular stomatitis virus glycoprotein (VSVG) -pseudotyped replication incompetent retroviruses transduced the cultured schistosomes, leading to integration of proviral forms of the retrovirus into schistosome chromosomes. A PCR technique utilizing endogenous mobile elements as anchors was employed to recover proviral integration junctions, which revealed that MLV showed primary sequence specificity for a gGATcc-like target motif and that the proviral transgenes were located within or near protein encoding genes and other sites. Furthermore, the actin promoter drove transcription of the luciferase transgene in schistosomula and adult mixed-sex schistosomes. These findings definitively demonstrated chromosomal integration of transgenes and somatic transgenesis of schistosomes mediated by pseudotyped retrovirus, and provide a foundation for functional genomics investigations of schistosomes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Schistosomula were washed 3x in wash medium and cultured at 37°C under 5% CO2 in air in modified Basch’s medium (23)
supplemented with washed human erythrocytes, 2 µl erythrocytes/ml culture medium. Adult schistosomes perfused from mice were cultured at 37°C under 5% CO2 in air in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 100 U of penicillin and streptomycin, and washed human erythrocytes (as above). Schistosomula and adult schistosomes were maintained in culture for up to 3 wk.
Transduction of schistosomes with pseudotyped retrovirus
Plasmid constructs were assembled from the pLNHX-based plasmid pLNHX-SmSL-luciferase (24)
by insertion of the cassette encoding the S. mansoni actin gene promoter and the firefly luciferase gene (25)
to derive pLNHX-SmACT-Luc. The actin gene promoter was amplified using Platinum Taq DNA Polymerase High Fidelity (Invitrogen, Carlsbad, CA, USA) designed with restriction sites for XhoI and BglII in the primer clamp region to directionally clone the promoters into pLNHX. The S. mansoni SL RNA gene promoter was removed using XhoI and BglII and replaced with the actin gene promoter to derive pLNHX-SmACT-Luc. Production of VSVG-pseudotyped virions in packaging cells was undertaken as described previously (24)
. In brief, GP2–293 packaging cells (Clontech, Mountain View, CA, USA) (modified HEK-293 cells) were transfected with plasmids pLNHX-SmACT-Luc or pLNHX-SmSL-Luc along with a plasmid encoding VSVG, delivered in liposomes (Lipofectamine 2000, Invitrogen). Subsequently, pantropic virus in the culture supernatants was concentrated by high-speed centrifugation (Sorvall SS-34 rotor, Thermo Scientific, Waltham, MA, USA; 50,000 g, 90 min, 4°C). The pellet of concentrated virions (from 50 ml of packaging cell culture supernatants) was resuspended in TNE (50 mM Tris, 130 mM NaCl, 1 mM EDTA, pH 7.8) at 4°C overnight, after which the virions were stored at –80°C. Viral titers were determined using target NIH-3T3 fibroblast cells cultured in the presence of the antibiotic geneticin. Schistosomula (103–104) or mixed-sex adult S. mansoni (10–100) were cultured in 35-mm tissue culture wells in 2 to 3 ml of medium containing 
200 µl of virion stock with an infectivity of 1 x 104 to 1 x 107 colony forming units (CFU)/ml, i.e., in the range of one to several thousand virions per worm.
Detection of provirus in the schistosome genome
Total genomic DNAs (gDNAs) were isolated from transduced or control schistosomes using the AquaPure system (Bio-Rad, Hercules, CA, USA). Primers specific for genes encoding neomycin phosphotransferase II, 5'-TGTGCTCGACGTTGTCACTGAA and 5'-ATGAATCCAGAAAAGCGGCCA, and firefly luciferase, 5'-GTGCCAGAGTCCTTCGATAG and 5'-ACAACTTTACCGACCGCGCC, were employed to amplify proviral reporter transgenes. Primers for the S. mansoni cytochrome oxidase 1 (cox I) gene [GenBank AF101196; National Center for Biotechnology Information (NCBI), Bethesda, MD, USA], 5'-TGAGTGTCATTTTAGGGTGGTG and 5'-ACAAACCAATGAAAATATCCAAGA were used to confirm the integrity of schistosome gDNAs. At the time of harvest of the cultured schistosomes, an aliquot of culture media was also collected for analysis to investigate the presence of residual, contaminating pLNHX plasmids carried over in virion inocula. PCR products were separated by electrophoresis through 1% agarose, stained with ethidium bromide, visualized under UV illumination and digital images captured (Versa-Doc, Bio-Rad). PCRs were carried out using Master Mix (Promega, Madison, WI, USA) reagents, as well as 35 thermal cycles of 94°C for 1 min, 52°C for 1 min, and 72°C for 2 min. PCR products were Southern blotted onto Zeta-Probe (Bio-Rad) nylon membranes. A 4.5 kb, KpnI fragment of pLNHX-SmSL-Luc (24)
was labeled with 32P-dCTP by random oligomer priming (RadPrime DNA Labeling System, Invitrogen); this probe includes transgenes encoding neomycin phosphotransferase and luciferase. Southern blots were hybridized to the labeled probe for 18 h and washed at high stringency (26)
, and hybridization signals were detected by autoradiography and X-ray film (Biomax; Kodak, Rochester, NY, USA).
In addition, schistosome gDNAs and plasmid pLNHX-SmACT-Luc were cleaved with NcoI and separated by electrophoresis through 1% agarose, and the fragments were transferred by capillary action to Zeta-probe nylon membranes (Bio-Rad). A gene probe was prepared by isolation of a 5.3 kb KpnI fragment pLNHX-SmACT-Luc, which includes the luciferase coding sequence. The 5.3 kb KpnI fragment was isolated from the digestion products by agarose gel electrophoresis, eluted from the gel, and labeled with 32P-dCTP. Southern blots were hybridized to this probe for 18 h and washed at high stringency at 65°C, and hybridization signals were detected as above.
Reverse transcription PCR
Pellets of schistosomula were homogenized using a motorized pestle, after which total RNA was isolated from the homogenates using the Versagene RNA Purification kit (Gentra, Minneapolis, MN, USA). The RNA preparations were incubated with RNase-free DNase I (Gentra) to remove contaminating gDNA. cDNAs were synthesized from the DNase-treated RNA using iSCRIPT reverse transcriptase (Bio-Rad). Endpoint PCR was performed on dilutions of cDNA and control DNase I-treated RNA using the Promega Mastermix system, and 35 thermal cycles of 94°C, 1 min, 52°C, 1 min, and 72°C, 2 min. The control S. mansoni housekeeping gene cox I (GenBank AF101196), neomycin phosphotransferase II gene (neo), and firefly luciferase gene were amplified using the primers described above. Negative controls included total (virus exposed schistosome) RNA that had not been reverse transcribed as a template, and reactions in which water replaced the template nucleic acids. Amplification products were sized by electrophoresis through 1% agarose, visualized after staining with ethidium bromide and UV illumination, and the images were recorded (Versa-Doc, Bio-Rad). PCR products were transferred to nylon membranes by Southern blot analysis and probed with the 32P-labeled pLNHX-SmSL-Luc/KpnI probe (above).
Retrotransposon-anchored PCR (RAP)
We developed an anchored PCR-based approach, RAP, to investigate retrovirus integrations into the schistosome genome. In brief, RAP employs a primer specific for the luciferase (luc) transgene from the donor pLNHX construct in tandem with second primer specific for endogenous retrotransposons present at high copy number and interspersed throughout the genome of natural populations of S. mansoni (27)
. Specifically, the primers included sequences specific for the retrotransposons SR1 and fugitive (28
29
30
31
32
33)
; SR1 reverse 5'-CGTCGTGGTACCAACCTTTGCACTCATC, fugitive reverse 5'-CAACTGCGTTACTGTCTCCTCGAAACTG, and luciferase left 5'-CTGCGAAATGCCCATACTGTTGAGC. Primers targeting other endogenous retrotransposons were also tested, as described previously (11)
. The RAP primers were employed with 100 ng template gDNA from populations of MLV-transduced schistosomula and Platinum Taq DNA Polymerase High Fidelity (Invitrogen). The RAP cycling conditions were 94°C for 2 min followed by 38 cycles of 94°C for 30 s, 57.5°C for 30 s and 68°C for 10 min, with a final extension at 68°C for 10 min. RAP products were analyzed by ethidium-stained agarose gel (1%) electrophoresis, after which the products were transferred to nylon (Zeta-probe) and hybridized under stringent conditions to a luciferase gene probe. This probe was amplified from pGL3 Basic (Promega) with primers Luc-F 5'-ATGGAAGACGCCAAAAACAT and Luc-R 5'-TACACGGCGATCTTTCCGCC, and cycling conditions of 1 min at 94°C followed by 38 cycles of 30 s at 94°C, 30 s at 50°C, 2 min at 68°C, and a final extension at 72°C for 7 min. The 1.6 kb product was isolated by agarose gel electrophoresis, eluted from the gel and labeled with 32P-dCTP using the random oligomer priming method. From the RAP products that appeared to be positive in the Southern hybridizations, fragments ranging from 8 to 2 kb in size were isolated from a crystal violet gel and ligated into plasmid pCR-XL TOPO (Invitrogen) to establish libraries of integration junction fragments. About 1000 randomly selected colonies were screened by colony hybridization. Nylon membranes were probed with the labeled luciferase gene fragment; from the 1000 colonies, 10% were positive for luciferase-encoding sequences. Minipreps of plasmids from the 100 positive colonies were isolated (GenElute Plasmid Miniprep Kit, Sigma-Aldrich, St. Louis, MO, USA), digested with EcoR I to release the inserts, and sized by agarose gel electrophoresis; the fragments were then transferred to nylon, and the membranes were probed with the labeled luciferase probe, as above. Nucleotide sequences of the positive clones were determined using vector-specific and insert-specific primers.
Luciferase activity assay
Schistosomula and adult schistosomes were transduced in vitro with retrovirus by exposure to MLV-VSVG virions. One day later, media were replaced, and the worms were cultured for 2 to 7 days. At this point, transduced schistosomes were removed from culture, washed 3x, and stored at –80°C until needed. To prepare soluble lysates for investigation of reporter firefly luciferase activity, pellets of schistosomula were subjected to sonication (5x5 s bursts, output cycle 4, Heat Systems–Ultrasonics, Plainview, NY, USA) in 250 µl CCLR lysis buffer (Promega). Protein concentration of soluble sonicates were determined using the bicinchoninic acid assay (Pierce, Rockford, IL, USA). Aliquots of 100 µl of sonicate were injected into 100 µl luciferin substrate (Promega) at 23°C and mixed; the relative light units (RLUs) were determined in the luminometer 10 s later. Duplicate samples were measured; results are presented as the average of the duplicate readings per microgram of soluble protein. Recombinant luciferase (Promega) was included as a positive control (11)
.
Immunolocalization of luciferase
Schistosomula were harvested from tissue culture into 1.5 ml microcentrifuge tubes by centrifugation for 2 min at 10,000 g, washed in PBS (3x each, 1–2 min) by centrifugation to remove culture media, and fixed in 10% formalin in PBS for 20 min at 23°C or overnight at 4°C. Schistosomulum tissues were permeabilized by incubation in 0.2% Triton X-100 in PBS for 30 min, washed 2x in PBS to remove the detergent, and incubated in 1% nonfat milk in PBS for 30 min at 37°C to block unbound antibody sites. After two further PBS washes, schistosomula were incubated in primary antibody (goat anti-luciferase, diluted 1:200; Promega) for 18 h at 4°C or for 45 min at 37°C. Following removal of the primary antibody (PBS, 3x washes), schistosomula were incubated in secondary antibody [Alexa Fluor 594 donkey anti-goat (red), Molecular Probes, Eugene, OR, USA; 1:250] for 90 min at 37°C and washed 3x in PBS. Samples were mounted in 50% glycerol/PBS and viewed with an inverted microscope (Eclipse TS100; Nikon, Tokyo, Japan) fitted with a red filter (Texas Red) at 540–580 nm excitation, 595 nm emission. Images were captured with a CoolPix 5700 camera (Nikon). In addition, to determine the percentage of the populations of schistosomula positive for luciferase, 120 to 160 individual schistosomula were examined for fluorescence at each time point.
Bioinformatics
Sequences obtained from RAP clones were analyzed with assistance of the Accelrys Gene version 2.0 software (Accelrys, San Diego, CA, USA) and/or or other online bioinformatics tools, as appropriate. Searches of putative integration junction sequences were performed using BLAST search algorithms at the NCBI (http://www.ncbi.nlm.nih.gov/BLAST/) and Sanger Schistosoma mansoni Genome Project (http://www.sanger.ac.uk/Projects/S_mansoni/) databases.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A) were amplified from the genome of schistosomula transduced by pLNHX-SmSL-Luc virions (Fig. 1B
). Schistosomula were infected by soaking in 1 x 105 to 1 x 106 CFU/ml pLNHX-SmSL-Luc virions and were harvested 5 days postinfection. In addition to the detection of ethidium-stained products of the expected size, the identity of the PCR products was confirmed by Southern hybridization analysis to a labeled pLNHX-SmSL-Luc/KpnI gene probe (Fig. 1B
, lanes 1 and 2). The final wash before harvest was investigated by PCR for the presence of contaminating pLNHX-SmSL-Luc plasmids; neither neo- nor luc-specific targets were amplified, indicating absence of residual plasmid carried over in the virion inoculum (Fig. 1B
, lanes 4 and 5). Cox I-specific products were amplified (Fig. 1B
, lane 3), verifying the integrity of the gDNAs. A direct PCR approach had been employed to detect MLV integrations into chromosomes of Anopheles gambiae cell line MOS-55 exposed, as here, to VSVG pseudotyped MLV virions (34)
.
|
Southern blot analysis indicated integration of MLV transgenes in the schistosome genome
Because direct PCR analysis of the transduced schistosomes suggested that mobilization of VSVG-MLV was feasible in schistosome tissues (Fig. 1B
), a Southern hybridization experiment was undertaken to further investigate integration of MLV into schistosome chromosomes. gDNA isolated from 20,000 schistosomula, transduced with pLNHX-SmACT-Luc virions, was digested with NcoI, after which the fragments were resolved by electrophoresis, transferred to nylon, and hybridized to a labeled pLNHX-SmACT-Luc/KpnI gene probe. NcoI cleaves twice within pLNHX-SmACT-Luc, releasing a fragment of 1.9 kb that should hybridize to the probe. Hybridization to gDNA from transduced schistosomula was apparent; a robust signal was evident at 1.9 kb (Fig. 1C
, lane 2), indicating the presence of numerous copies of the MLV provirus within chromosomes of the schistosome population. Two other strong bands of hybridization, at 1.7 and 2.5 kb were evident, along with minor bands of hybridization from 3 to >10 kb in size (Fig. 1C
, lane 2). By contrast, no signal was seen in gDNA from untreated (nontransduced) schistosomes (Fig. 1C
, lane 1). Signals of variable length larger than 1.9 kb likely represented NcoI fragments of proviral transgene and schistosome chromosomes adjacent to sites of MLV integration. The other two major bands, at 1.7 and 2.5 kb, indicated the presence of NcoI sites in the schistosome chromosomes near the 5'- or 3'-LTRs of integrated copies of MLV. The pLNHX-SmACT-Luc/KpnI probe extends beyond the 1.9 kb NcoI fragment of pLNHX-SmACT-Luc (Fig. 1A
); it would hybridize to all three fragments of an integrated proviral form of the pLNHX-SmACT-Luc construct released by NcoI cleavage of schistosome gDNA: 1) 
1.6 kb, 2) 1.9 kb, and 3) 2.4 kb. This is clearly evident as the three major bands in Fig. 1C
, lane 2. Other minor bands of hybridization from 3 to >10 kb in size were apparent, indicating integrations of a number of other copies of the provirus transgene. By contrast, a band of hybridization of 6.5 kb, seen strongly in lane 3, was absent from the genomic DNA of the transduced schistosomula (lane 2), demonstrating that the signals apparent in lane 2 represented integration events and not spurious hybridization of any pLNHX-SmACT-Luc vector plasmid that may have contaminated the virion inoculum.
Retrotransposon-anchored PCR retrieved MLV transgene integrations
Similar preparations of gDNA isolated from schistosomula transduced with pLNHX-SmACT-Luc virions were employed as templates for an anchored PCR-based approach that we term RAP (11)
. In brief, RAP employs a primer specific for the luc transgene from the pLNHX-SmACT-Luc cassette in tandem with second primer specific for one of the endogenous retrotransposons known to be present at high copy number in the genome of natural populations of S. mansoni. Specifically, the retrotransposon-specific primers targeted the SR1, SR2, Boudicca, fugitive and SM
elements. Figure 2
A provides a schematic representation of the RAP protocol. We analyzed the patterns of the resulting PCR products in ethidium-stained gels (representative example shown in Fig. 2B
) and by Southern hybridization analysis of the RAP products to a 32P-labeled luciferase gene probe (Fig. 2C
). The patterns of hybridization indicated the presence of amplicons representing integration events of MLV into S. mansoni chromosomes. Subsequently, we cloned a number of these Southern hybridization-positive PCR bands into pCR-XL-TOPO, including fragments amplified with retrotransposon SR1 paired with luciferase transgene-specific primers (lane 5, Fig. 2C
) and fragments amplified with retrotransposon fugitive paired with luciferase transgene-specific primers (lane 3, Fig. 2C
). Only one round of RAP was required to amplify apparent integration junctions.
|
Widespread retrovirus integrations into schistosome genome
We investigated 100 positive clones (from libraries of RAP products in pCR-XL-TOPO screened with a luciferase gene probe) by restriction digest pattern analysis. Approximately 30 distinct patterns were identified, and clones representative of each of the patterns were sequenced. From these sequences, we confirmed 16 discrete integrations of MLV into schistosome chromosomes (Table 1
). The sequences of these transgene integration sites have been assigned the GenBank accession numbers ET202011–ET202026. The accession numbers include annotations of the genomic locations of the transgenes as determined by Blast analyses of the NCBI (www.ncbi.nlm.nih.gov/BLAST) and Sanger Institute (www.sanger.ac.uk/cgi-bin/blast/submitblast/s_mansoni/omni) databases. In brief, MLV proviral transgenes were found in the vicinity of endogenous schistosome retrotransposons SR2, fugitive, Perere-2, and Saci-4 (35)
(e.g., ET202011, ET202012, ET202018), and also near other repetitive sequences, including the Sm7 satellite sequences and the Sh122 repetitive sequence (ET202021, ET202016). MLV provirus transgenes were also located near to or within the introns of protein-encoding genes, including those encoding immunophilin, zinc finger protein Sma-Zic, glutathione-S-transferase, cathepsin D, and hypoxanthine-guanine phosphoribosyl transferase (e.g., ET202013, ET202024, ET202015) (Table 1)
. The transgene integration junction sequences definitively documented somatic transgenesis of the S. mansoni genome by VSVG-pseudotyped MLV retrovirus, and indicated a widespread distribution of MLV provirus integrations throughout the schistosome genome.
|
Intact and mutated provirus transgenes
As illustrated in Table 1
, 16 integration junctions were recovered using the RAP method and sequenced. The RAP targeted endogenous retrotransposons and the luciferase transgene. Eight integration junctions were recovered using the SR1 retrotransposon anchor and 8 using the fugitive retrotransposon anchor, in both cases paired to a transgene-specific primer directed in the antisense direction on the luciferase gene. Of the 16 integration junctions—and although we have only recovered and sequenced the 5' region (left hand side) of the provirus as a consequence of the RAP method—7 of the 16 included the complete (intact) MLV provirus. These included the intact 5'-LTR and downstream sequences (including the psi encapsidation signal, neomycin resistance gene, actin gene promoter and luciferase reporter gene). By contrast, 9 of the 16 (56%) were mutated MLV provirus forms, where the 5'-LTR and some adjacent downstream residues of the provirus had been deleted. In most of the mutated proviral transgenes, the entire 5'-LTR, the psi encapsidation signal and about one-third of the neo gene were deleted. The remainder of neo, the actin gene promoter, and luciferase reporter gene were integrated into the schistosome genome. More specifically, of the 8 integrations recovered with the SR1 anchor, 3 were intact, while 5 were mutated. For the 8 integrations recovered with the fugitive primer, 4 were intact and 4 were mutated. Schematic representations of these intact and mutated proviral transgenes are shown in Fig. 3
.
|
Integration target motif, gGATcc
At each of the 16 integration junctions, the motif gGATcc, or a close variant thereof, occurred at the site of integration. As shown in Fig. 3
, the gGATcc-like motif was located adjacent to the 5' terminus of the MLV. In the 7 intact of 16 integrations, this motif was flanked by the terminal residues of the intact 5'-LTR of MLV, TTTGAAAGA ... . Moreover, the gGATcc-like motif was present at the site of integration of the 9 of 16 insertions, which were 5'-LTR truncated. It appeared that gGATcc or perhaps the central trinucleotide core of this motif, GAT, or even the central dinucleotide, AT, was a preferred target for MLV integration. In addition, there was a T residue flanking the gGATcc motif in all of the 16 MLV integrations, i.e., T was the terminal 5'-residue of the intact or mutated MLV proviral transgene. Because we isolated only one side of the MLV provirus, we do not yet know whether target site duplications had occurred at both the 5' and 3' termini of the integration; target site duplications of 4–6 bp are characteristic of retrovirus integrations into mammalian chromosomes (36)
.
Transgenes transcriptionally active
We employed reverse transcription PCR to investigate transcription activity of the transgene reporter genes. We analyzed total RNA from groups of schistosomula exposed to 105 or 106 CFU/ml virions. More specifically, 1000 schistosomula were cultured with 20,000 or 200,000 pLNHX-SmSL-Luc virions, which correlated to 20 or 200 virions/schistosomulum.
Schistosomula were infected 1 or 6 days after cercarial transformation by soaking in pLNHX-SmSL-Luc virions, and harvested 2 to 5 days later. Transcripts encoding neo and luc were detected (Fig. 4
; lanes 1 and 4, luc; lanes 2 and 5, neo) in both inoculum size treatment groups, although there was no obvious difference in intensity of the signals. Cox I was included as a positive control for the integrity of the RNA (Fig. 4
, lanes 3 and 6). A negative control without cDNA gave no product (Fig. 4
, lane 7). Additional control reactions were carried out employing RNA rather than cDNA as the template to ensure the absence of gDNAs; the results were negative, indicating the total RNA preparations were free of contaminating gDNA (not shown). Identities of the neo and luc products were confirmed by Southern hybridization to a labeled 4.5 kb KpnI fragment of pLNHX-SmSL-Luc. The presence of these transcripts indicated that the MLV 5'-LTR and the SmSL promoter were active in the transduced schistosomes since the 5'-LTR drives transcription of neo and SmSL was expected to drive transcription of luc (Fig. 4)
.
|
Luciferase confirms productive transduction of schistosomes and developmental activity of actin gene promoter
In addition to transgene transcription, we investigated reporter luciferase enzyme activity in tissues of the transduced schistosomes. Initially, adult worms perfused from mice 2 to 3 wk previously were exposed to pLNHX-SmSL-Luc or pLNHX-SmACT-Luc virions (8x103 and 8x104 CFU/ml, respectively) for 1 day. Four to 6 days later, soluble extracts of transduced worms were examined for luciferase activity. Luciferase was detected in the adult worms transduced with pLNHX-SmACT-Luc virions, in worms that had been in culture either for 2 (Fig. 5
A, lane 2; 6.0 RLU/µg) or 3 (Fig. 5A
, lane 4; 8.8 RLU/µg) weeks after perfusion from mice. By contrast, far less luciferase was detected in the worms transduced with the pLNHX-SmSL-Luc virions (Fig. 5A
, lanes 1 and 3, 0.6 and 0.6 RLU/µg). These findings indicated that the S. mansoni actin gene promoter was more active than the SmSL promoter in driving luciferase from proviral MLV transgenes (Fig. 5A
).
|
Given the superiority of the actin promoter in driving luciferase activity in the adult worms, we investigated transduction of blood-stage schistosomula with pLNHX-SmACT-Luc virions. Furthermore, because Pearce and colleagues (25)
had demonstrated developmental regulation of the actin gene promoter, we investigated the influence of the age of the developing schistosomulum on transgene activity. Schistosomula that had been maintained in culture for 1, 5, 10, 15, and 20 days after mechanical transformation from cercariae were exposed to 40 pLNHX-SmACT-Luc virions per schistosomulum, after which transgene reporter activity was investigated using two procedures. First, 3 days after virus transduction, soluble extracts of schistosomulum tissue were prepared and analyzed for luciferase activity. Elevated activity was evident in the 10- and 15-day-old schistosomula, 64.9 and 50.9 RLU/µg, respectively (Fig. 5B
, lanes 4 and 5). By contrast, only minimal luciferase activity was recorded from the 1-, 5- and 20-day-old schistosomula, 1.7, 1.6, and 1.5 RLU/µg, respectively (Fig. 5B
, lanes 2, 3, and 6). Second, an immunostaining technique was deployed using an antiluciferase specific antibody and an Alexa Fluor 594 labeled secondary antibody to monitor transgene expression in similarly aged groups of schistosomula. We visualized the presence of luciferase in each age group of developing schistosomes, i.e., in 1-, 5-, 10-, 15- and 20-day-old schistosomes (Fig. 5C
); the percentages of the populations positive for luciferase were 8.7% (day 1), 18.3% (day 5), 23.3% (day 10), 20.7% (day 15), and 11% (day 20). Further, on days 10 and 15, schistosomula were generally brighter. Control 20-day-old schistosomula, which were not exposed to virions but were processed through the same immunostaining process, showed no or minimal (11.3%) Alexa Fluor 594 fluorescence (Fig. 5C
). Together, the markedly elevated luciferase enzyme activities, the stronger immunofluorescence, and the larger proportion of the population immunostaining for luciferase in 10- and 15-day-old schistosomula suggested that there was an authentic influence of schistosomulum developmental age on the activity of the S. mansoni actin gene promoter.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, which has been developed over the past two decades as a tool for forward genetics studies of model organisms, including the zebrafish (2
, 38)
, and as a vector for human gene therapy (39
, 40)
. In addition, earlier studies have indicated that MLV pseudotyped with the glycoprotein of VSV could facilitate transgenesis of a diverse array of other model and pathogenic eukaryotic species, including surf clams, oysters, mosquitoes, and amoebae (34
, 41
42
43)
. In a preliminary report, we suggested that VSVG-pseudotyped MLV could infect developmental stages of the human blood fluke S. mansoni (24)
. Here, we have now definitely confirmed that VSVG-MLV can transduce blood-stage forms of S. mansoni, that integration of the provirus efficiently occurs soon after exposure of the worms to even small numbers of virions, that integration may exhibit an unusual genome target site preference, and that reporter provirus transgene activities can be driven in schistosome cells by the retrovirus LTR or endogenous schistosome gene promoters.
Direct PCR analysis and Southern hybridization indicated that MLV proviral transgenes had integrated into the schistosome chromosomes. Subsequently, we employed an anchored PCR protocol to search for MLV provirus integrations, a technique termed RAP, which uses direct PCR targeting nondigested template DNA (11)
. RAP is similar to the Alu-PCR procedure that has been employed to locate adenovirus within human chromosomes (44)
. Luciferase gene probe hybridization to RAP products indicated the presence of MLV transgenes in the vicinity of endogenous schistosome retrotransposons. Nucleotide sequence analysis of integration junctions verified insertion of MLV provirus at disparate sites in schistosome chromosomes. At a more global level, this outcome confirmed the utility of VSVG-MLV for transgenesis of schistosomes. Characterization of MLV proviral integrations into schistosome chromosomes also indicates there is no evolutionary block to at least some of the molecular steps of the MLV life cycle in organisms as distant as flatworms and mammals. It has been suggested that blocks to key steps in the developmental cycle likely would limit the utility of MLV in lower vertebrates and invertebrates (45)
.
We isolated 16 discrete integration events. Because RAP was used to recover these integrations, all of the events characterized here were localized to the vicinity of the endogenous retrotransposons SR1 and fugitive employed as RAP anchors. Both of these retrotransposons occur at high copy number in the schistosome genome and are presumed to be interspersed (28
, 31)
. Nonetheless, we were able to employ the schistosome genomic sequence flanking the integration sites to attempt to identify the target sites of these integration events, using BLAST analysis of the draft S. mansoni genome. MLV proviral transgenes were found near endogenous schistosome retrotransposons and other repetitive sequences and nearby or within the introns of protein-encoding genes. In mammalian cells, MLV displays moderate target specificity, in particular for the transcription start sites of protein-encoding genes, CpG islands and the 5' ends of genes, including the first intron. Other retroviruses exhibit other preferences; HIV-1 targets transcriptionally active regions of RNA polymerase II transcribed genes, ASLV weakly targets active genes and HTLV-1 does not specifically target transcription units and transcription start sites (2
, 46
47
48)
.
About one-half of the MLV proviral transgenes recovered displayed an intact 5'-LTR, whereas others were mutated by deletions of the 5'-LTR and adjacent genes. Deletions of 5'-LTR regions of proviral retroviruses are known to occur, with the deletion generally occurring before integration of the provirus (36
, 49)
. Nonetheless, all 16 of the proviral transgenes were likely to display transcriptionally active reporter genes because the actin gene promoter and firefly luciferase coding regions were intact in all of them. Hallmarks of integration into mammalian chromosomes include duplication of 4–6 bp of host chromosome flanking the 5'- and 3'-LTRs of the retrovirus. We have been unable to establish whether this duplication also occurs in schistosome chromosome integrations because we have so far only examined one side of the transgene. In addition, usually there is deletion of 2 bp from the terminus of the 5'- and 3'-LTRs at integration of MLV and other retroviruses into mammalian chromosomes (50)
. This deletion did not occur in schistosome cells since sequence analysis of the 7 intact proviral transgenes demonstrated the presence of the entire 5'-LTR. In this regard, integration of MLV into schistosome chromosomes diverged from the mammalian paradigm. There is minimal information available on the integration of MLV into chromosomes of other invertebrates. However, MLV integration into mosquito chromosomes resulted in the characteristic 2 bp deletion, and moreover, no target site specificity was apparent (34)
.
Whereas preferences are exhibited by different retroviral species in relation to target site selection, a target site primary sequence motif is unapparent at integration of MLV, HIV-1, and other retroviruses into mammalian chromosomes. Indeed, primary sequence does not determine target selection. By contrast, remarkably, all 16 of 16 MLV integrations into schistosome chromosomes were located at a gGATcc-like motif. The core trinucleotide GAT or dinucleotide AT may be a recognition site for integration, in similar fashion to recognition sites exhibited by transposons, e.g., TTAA by piggyBac, TA(T/A)TA by Tn7, and TA by mariner elements (51
52
53)
. In any event, the apparent target sequence specificity suggested that the MLV integrase may operate with schistosome cellular cofactors, and if so, the integrase-cofactor complex may endow target site specificity on MLV integration. It is also possible that the viral integrase is not functional in schistosome cells and that schistosome DNA repair or other enzymes, rather than viral integrase, insert the proviral transgene, endowing the target site specificity on MLV proviral integrations into schistosome chromosomes.
Reporter gene silencing of MLV provirus transgenes is known in some contexts (37
, 54)
; however, the actin promoter drove reporter luciferase activity, and the 5'-LTR drove transcription of neo. These findings indicated, at least in somatic cells, that reporter gene silencing was not occurring and might not be an impediment to MLV-mediated functional genomics of schistosomes. Given that the virions employed here were replication incompetent, we anticipate that only those schistosome cells that were transduced with MLV virions would have expressed the proviral transgenes. The tissue- and organ-specific localization of these cells is not yet known, as we have yet to undertake detailed immunolocalization studies of transduced schistosomes. However, it is likely that surface- and/or gut-located cells were transduced. Exogenous double-stranded RNA is taken in through the gut of blood stage schistosomes, rather than traversing the surface (10)
. The schistosome tegument (surface) is covered by a unique lipid bilayer (55)
. Whether pseudotyped virions can cross this barrier awaits elucidation. Nonetheless, VSVG pseudotyped MLV virions clearly can efficiently transduce schistosomes and so the resolution of the transduction pathway, across the tegument or through the gut, will likely also provide insights into the nature of the schistosome surface and the barrier presented by the enigmatic surface double lipid bilayer (21)
.
In overview, the present findings demonstrated VSVG-pseudotyped MLV mediated transduction of blood stages of S. mansoni, chromosomal integration of retroviral transgenes, transgene activity, and somatic transgenesis. It has been suggested that evolutionary blocks would constrain the utility of MLV in nonmammalian taxa (45)
. By contrast, we conclude now that essential post-transduction steps of the MLV retrovirus developmental cycle that were postulated to be disabled or absent in nonmammalian species are clearly functional in schistosome cells, including presumably the activity of retroviral reverse transcriptase and integrase, and the assembly of the preintegration complex. Indeed, we consider that comparison of host cell accessory factors for retroviral activity in cells as phylogenetically distant as mammals and schistosomes could inform our understanding of the molecular pathogenesis of retroviral infection of human and other vertebrate cells. The atypical target site preference displayed by MLV in schistosome chromosomes provides a phenomenon where comparisons would likely be informative. Finally, given that there is an intact RNAi pathway in schistosomes (10)
and that RNAi induced visible and lethal phenotypes can be achieved (56
, 57)
, it may be feasible to establish lines of transgenic schistosomes carrying hairpin gene constructs in order to advance functional genomics approaches for these parasites. Also, given that MLV integrated nearby or within protein-encoding genes, insertional mutagenesis screens and gene trapping are additional avenues that can be addressed in the future if heritable lines of transgenic schistosomes can be established.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 17, 2008. Accepted for publication March 20, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
