Snails, species identification and susceptibility for schistosome infection - B. glabrata snails were collected from a small stream in an endemic site for transmission of S. mansoni, in the south east of Brazil, Barreiro, Minas Gerais, (19^oS 59 min/44^oW 02 min). Offspring of these snails are maintained as a laboratory strain designated as BB02 (Biomphalaria from Belo Horizonte, Minas Gerais, Brazil 2002).

The species identity of BB02 snails was determined by PCR-RFLP. The ITS1-5.8S-ITS2 sequence region was PCR amplified from DNA of individual snails using primers (all primers are shown 5' -3' ) ETTS2: TAA CAA GGT TTC CGT A GG TGA A and ETTS1: TGC TTA AGT TCA GCG GGT. The amplicons were digested with DdeI and restrictions patterns obtained from BB02 snails were compared to the characteristic banding pattern specific for B. glabrata (Vidigal et al. 1998). Also sequences from the 16SrDNA and ND1 genes of one BB02 snail were amplified by PCR, using primer pairs 16Sar: CGC CTG TTT ATC AAA AAC AT - 16Sbr: CCG GTC TGA ACT CAG ATC ACG T (Palumbi et al. 1996) and SNDF1F2: CGR AAA GGA CCT AAY AGT TGG - SND1R4: ART CRA ATG GYG CHC GAT TAG, respectively. (R=A/G Y=C/T H= A/C/T). The sequences from these amplicons were obtained by direct sequencing and analyzed relative to previously generated phylogenies of Biomphalaria isolates, all according to DeJong et al. (2003). The sequences of 16S rDNA and NADH dehydrogenase 1 were deposited in GenBank under accession numbers AY737280 and AY737281, respectively.

Members of the F1 generation derived from field collected snails were tested for susceptibility to two different S. mansoni strains (LE, SJ) at the Section of Molluscs Rearing at the Centro de Pesquisas René-Rachou in Belo Horizonte, Brazil. Groups of 50 juvenile snails (3-6 mm) were exposed individually to 10 miracidia. The parasite-susceptible BB01 strain of B. glabrata (maintained over 10 years in the laboratory in Brazil) was used as a control for miracidial infectivity. At 4 weeks post exposure, snails were exposed to artificial light for 30 min and the shedding of cercariae was recorded. Non-shedding snails were dissected to check for developing sporocysts. BB02 B. glabrata were similarly tested for susceptibility to the NMRI strain of S. mansoni at the Biomedical Research Institute (MD, US).

Preparation of HMW genomic DNA from BB02 B. glabrata - Initial comparisons disclosed that relative to whole body or the digestive gland, the ovotestis of B. glabrata was optimal for generation of monocellular suspensions as required to obtain high molecular weight DNA (Luo & Wing 2003). However, the DNA yield from a single snail was insufficient to generate a BAC library. B. glabrata is a simultaneous hermaphrodite and offspring were generated by selfing to minimize haplotype diversity. One newly hatched BB02 snail (< 3 mm shell diameter) was kept in isolation to generate F1 progeny by self-fertilization (sF1). A selfed F2 generation (sF2) derived from the sF1 was similarly obtained.

High molecular weight (HMW) genomic DNA was isolated from forty sF2 snails (10-12 mm shell diameter). Following cleaning and removal of shells, live snails were kept briefly in ½199 medium (physiological buffer for snail cells; medium 199 (Sigma) diluted 1:1 [v/v] with distilled water) until all snails were processed. From 4 snails at a time, the ovotestes were dissected and pooled in 800 µl of ½199 in 1.7 ml Eppendorf tubes on ice. All the following manipulations were performed gently to minimize damage to cells and mechanical shearing of DNA. The tissues were disrupted with 3 strokes of a polypropylene pellet pestle (Kontes). The resulting cell suspensions were pooled in a 50 ml Falcon tube on ice. No sediment was evident after 1 h. Cells were pelleted (400 g, 5 min at 4°C) and the cleared supernatant fluid was reduced to 600 µl. The cells were resuspended uniformly by tapping the side of the tube and incubated for 3 min at 42C. Then, 600 µl of 1% Seakem agarose (FMC) in ½199, (pre-warmed to 42°C) was mixed with the cells using minimal agitation. The monocellular cell suspension in agarose was transferred (using a cut-off, wide bore pipette tip) into disposable CHEF plug moulds (Bio-Rad) to obtain plugs with uniform cell numbers embedded in an agarose matrix, and placed on ice for 20 min. The 13 resulting plugs were transferred to 50 ml NDS (0.5 M EDTA, 10 mM Tris, 1% w/v N-lauroyl sarcosine, pH 9.5 (NaOH), supplemented with 1 mg/ml proteinase K (Invitrogen) and incubated overnight at 50°C in a rotary hybridization oven. This treatment lysed the cells while the agarose matrix protected high molecular weight genomic DNA from mechanical shearing. The medium was replaced by NDS and again incubated overnight at 50°C with rotation. DNA quality and susceptibility to HindIII digestion were evaluated by contour-clamped homogeneous electric field (CHEF) gel electrophoresis.

Generation of the BG_BBa BAC library - The methods of Luo and Wing (2003) were used to produce the BAC library. Briefly, following testing to determine optimal conditions, HMW DNA embedded in plugs was partially digested with HindIII. Following separation on CHEF gels twice, DNA fragments in the 150-300 kilobase (kb) range were eluted and ligated into pAGIBAC1. This BAC vector carries a resistance marker for chloramphenicol and incorporates a high signal for blue/white screening of non-insert transformants. The resulting constructs were introduced into DH10B-T1 phage resistant Escherichia coli cells by electroporation and plated on LB containing 12.5 µg/ml chloramphenicol and 80 µg/ml X-gal, 100 µg/ml IPTG for blue/white screening. Guided by video recognition of successful transformants, clones were picked and gridded into 384 well plates by a Q-bot (Genetix). Clones were stored as glycerol cultures at _80°C. Also, the clones from the BAC library were inoculated on four 22.5 ´ 22.5 cm Hybond N+ filters (Amersham) in high density, double spots and 4 ´ 4 patterns with a Q-bot (Genetix). The resulting filters a, b, c each contained 18432 clones in duplicate in six fields, the last filter (d) held 6528 clones in the same layout. The membranes were placed on LB agar plates containing 12.5 µg/ml chloramphenicol and incubated overnight to obtain colonies of 1 to 2 mm diameter. The membranes were placed (colony side up) on absorbent filter paper (Whatman Cat. No. 3030 700) soaked in the following solutions: (1) solution 1 (0.5 N NaOH, 1.5 M NaCl) for 7 min; (2) solution 2 (1.5 M NaCl, 0.5 M Tris-HCl, pH 8.0), 7 min; (3) air dry for more than 1 h; (4) solution 3 (0.4 N NaOH), 20 min; (4) solution 4 (5X SSPE), 7 min, and air dried overnight. The complete library (as frozen stocks), high density filters, and individual clones are available at cost from AGI. Protocols for screening of high density BAC library filters and address determination of positive signals are publicly available from AGI (www.genome. arizona.edu).

Isolation and sequencing of BAC DNA - At AGI, BAC DNA was isolated from 1.2 ml 2 ´ YT (Fisher) overnight cultures using alkaline lysis (96-well format) with a Quadra 96 Model 320 (Tomtec). Both ends of BAC inserts were sequenced using T7: TAA TAC GAC TCA CTA TAG GG as ``forward'' primer and BES_HR: CACT CAT TAG GCA CCC CA as the ``reverse'' primer. Cycle sequencing (BigDye Terminator v 3.1, Applied Biosystems) was performed using PTC-200 thermal cyclers (MJ Research) in 384-well format applying 150 cycles of 10 s at 95°C, 5 s at 55°C, and 2.5 min at 60°C. Extension products were purified by CleanSeq magnetic beads (Agencourt). Samples were eluted into 20 µl of ddH₂0 and separated on ABI 3730xl capillary sequencers with default conditions. Sequence data were collected by data collection software (Applied Biosystems), and transferred to a UNIX workstation. Sequences were base-called using the program Phred (Ewing & Green 1998, Ewing et al. 1998); vector and low-quality (Phred value <16) sequences were removed using the program Lucy (Chou & Holmes 2001). The methods applied at UNM included Montage BAC96 (Millipore) and Perfectprep BAC 96 (Eppendorf) for isolation of BACs. BAC ends were sequenced (Big Dye v. 3.1, ABI), also with T7 and BES_HR primers, using Biometra T-gradient thermal cyclers in 96 well format. The temperature profile was 1 min at 94°C, 100 cycles of 30 s at 94°C, 1 min at 55°C, 1 min at 72°C, and 7 min at 72°C. Following cleanup (Montage SEQ96; Millipore), extension products were read on an ABI 3700. Sequencher (GC codes) was used to remove vector sequences and edit chromatograms by eye.

Quality control of the BAC library - To estimate the average insert size of the BAC library, BACs were extracted from 361 randomly selected clones at AGI. The DNA was digested to completion with NotI (3 h/37°C) and separated on 1% CHEF gels to determine the size of the insert DNA. These data were applied to calculate the estimated coverage of the genome of B. glabrata by the BAC library. Absence of insert DNA was monitored to determine the proportion of empty vector in the BAC library.

The non-redundancy of BAC inserts was tested by sequencing (AGI) both termini of a random set of 192 clones. The clones were arbitrarily selected from wells A01, A02, A03 from plates 1-32, and well B23 from plates 1-96 in which the library is stored.

The representation of the genome of B. glabrata in the BAC library was investigated by screening the BAC library for sequences representing low- or single copy genes of B. glabrata (UNM). The probe sequences were selected from the literature, or chosen arbitrarily (see Table II). The probes were amplified by PCR from genomic or cDNA templates, labeled with ³²P a dCTP (Perkin Elmer) by random priming (Prime-it RT, Stratagene), and used as hybridization probes to screen filters that contained spotted BAC clones. The initial screening of high density filters representing the whole library (as available from AGI) was performed with two sets of five pooled probes (see http://www.genome.arizona.edu/information/protocols /index.html). The filters were prehybridized at 65°C for at least 4 h with hybridization buffer (0.5 M sodium phosphate pH 7.2, 7% SDS, 1 mM EDTA, 10 µg/ml sheared salmon DNA). After an exchange with fresh buffer, pre-hybridization was continued for 2 h. The probes were added and hybridized (>18 h, 65°C). The filters were washed sequentially with 2X SSC, 1X, and 0.1X SSC (all containing 0.1% SDS), 2 times each (20 min, 65°C), then autoradiography was performed. Positive clones were identified and obtained from AGI as bacterial stab cultures. These clones were used to manually prepare macroarrays (96 well format) applying similar methods as described above for the high density filters. The macroarrays were screened with individual probes to determine which clones contained specific target sequence. The BAC clones were also end-sequenced.

Contig alignment of BACs by fingerprinting - The BACs from clones that strongly hybridized the low- or single copy probes were subjected to the fingerprinting methods described by Luo et al. (2003). The resulting digestion patterns were compared for similarities to identify and contiguously align (partially) overlapping BACs using FPC software for the contig assembly (Soderlund et al. 2000). Also see http://www.genome.arizona.edu/BAC_special_projects/

Computational analysis and annotation of BAC end sequences - A contig analysis of the BAC end sequences was performed using Sequencher (GC codes). The clustering criteria were arbitrarily set at 98% identity over 100 nucleotides. The AT content was calculated for all non-redundant (sequence contigs were used instead of individual cluster mates) BAC end sequence data combined. BLAST searches were used to investigate the likelihood that BAC inserts were of snail origin, as well as to uncover similarities between BAC end sequences and the protein and nucleotide databases of GenBank, with special consideration of sequence entries from B. glabrata. E-values £ 10⁴ were considered significant. Discrepancies in sequence similarities between genomic and cDNA sequences were analyzed for the presence of non-coding sequences, including introns. Repetitive sequences were identified by direct inspection of sequence data and by analysis of results from BLAST searches. The BAC end sequence data were submitted as genome survey sequences (GSS) to GenBank under accession numbers CZ547921-CZ548269; DX360039-DX360203.