Construction of the P1-Based Framework

Dan Hartl and his group (who were members of the BDGP until August 1995) used the approach of in situ hybridization to Drosophila chromosomes in order to assign more than 2600 P1s to genomic localizations (Hartl, D.L., D.I. Nurminskky, R.W. Jones, and E.R. Lozovskaya. 1994. PNAS 91: 6824-6829). The polytene chromosomes of the salivary gland allow exquisite resolution, on the order of 50-100 kb. The data from these experiments suggest that about 2300 of the localized clones represent euchromatic genomic regions. The hybridization results of the other 300 clones suggest that these clones (about 15% of those assayed) might correspond to heterochromatic regions or euchromatic regions rich in repeats. These experiments resulted in a framework physical map that covers a bit more than 75% of the euchromatic genome. Furthermore, the mapped P1 clones provide a ready source of mapped STS markers.

The Bacteriophage P1 genomic library The Drosophila melanogaster genomic P1 libraries were constructed using genomic DNA isolated from a mixture of both male and female animals of the strain iso1 (y; cn bw sp) by David Smoller while a member of Dan Hartl's laboratory at Washington University, St. Louis, MO (Smoller, D.A., D. Petrov, and D.L. Hartl. 1991. Chromosoma 100: 487-494). High molecular weight DNA extracted from nuclei was partially digested with Sau3A then size fractionated using a 10%-40% sucrose gradient to select fragments in the range of 75 to 100 kb.

Recombinant P1 clones were generated in two separate ligation experiments using two similar P1 cloning vectors, pNS582tet14Ad10 (Sternberg, N. 1990. PNAS 87: 103-107) and pAd10sacBII (Pierce, J.C., B. Sauer, and N. Sternberg. 1992. PNAS 89: 2056-2060). These two P1 vectors differ by the insertion of a fragment containing the Bacillus amyloliquifaciens sacB gene into the BamHI site of the tetracycline resistance gene of pNS582tet14Ad10 to create pAd10sacBII. The presence of the sacB gene allows of a positive selection for clones with inserts by growth on sucrose (Gay, P., D. Le Coq, M. Steinmetz, T. Berkelman, and C.I. Kado. 1985. J. Bacteriol. 164: 918-921). Cloning into the BamHI site of pAd10sacBII disrupts expression of the sacB gene and allows growth on sucrose-containing media. If no insert is acquired, the expression of the sacB gene kills E. coli cells in the presence of sucrose. The sacB insert in the pAd10sacBII vector also provides on one side of the BamHI cloning site a T7 promoter and NotI restriction site, and on the other side an SP6 promoter and an SfiI restriction site. The ligation reactions were performed using phage arms generated by BamHI and ScaI digestion of the vector and an equimolar amount of digested genomic DNA. The ligation products of the pNS582tet14Ad10 reaction were packaged into phage heads in vitro and used to infect E. coli strain NS3145 then plated on LB plus kanamycin media to select clones. The resulting colonies were subsequently screened for inserts by growth on LB media containing tetracycline. Colonies that failed to grow on this media were presumed to have acquired genomic inserts. For the pAd10sacBII ligation reaction the strain NS3529 was used for infection after in vitro packaging into phage heads. These recombinant clones were selected by plating on LB media containing kanamycin and 5% sucrose.

The genomic P1 clones were picked by hand using sterile toothpicks into individual wells of microtiter plates filled with LB media with kanamycin and 10% glycerol. After growth at 37 degrees C overnight the plates were manually replicated using a 96-pin hand tool to provide additional copies of the library. The plates were then covered with plate sealers and stored at -70 degrees C. A total of 40 plates from the pNS583tet14Ad10 ligation and 99 plates from the pAd10sacBII ligation was picked for a total of 139 plates. The master P1 genomic library used for the in situ hybridization experiments and the STS mapping experiments described in this section was composed of 40 plates (3840 clones) from the pNS582tet14Ad10 ligation along with 56 plates (5376 clones) from the pAd10sacBII ligation. In total, these 96 microtiter plates contained 9216 clones, an estimated five-fold coverage of the genome. An additional 96 microtiter plates from other pAd10sacBII ligation reactions were later picked at LBNL using the robotic colony picker designed by Dr. Joe Jaklevic and members of the LBNL Automation Group.

The average insert size of clones generated from the two ligation reactions was discerned by individual growth and plasmid DNA preparation of a sampling of clones from each followed by restriction digestion and CHEF gel analysis. For the pNS582tet14Ad10 clones the average insert size was 83.0 ± 6.2 kb (N=25) and for the pAd10sacBII clones the average insert size was 82.5 ± 5.8 kb (N=20) (Hartl, D.L., D.I. Nurminskky, R.W. Jones, and E.R. Lozovskaya. 1994. PNAS 91: 6824-6829).

Naming of P1 clones Each P1 clone has a unique identifying number of the form DSnnnnn (where n is a number between 0 and 9). This is the name that should be used in all publications; this number will never change. In addition, each P1 clone has a position in the arrayed library that is expressed as a microtiter plate number and position number within that plate (see below). This number (Librarylocation) refers to the position of the P1 clone in a particular representation of the library. If the library were re- arrayed in the future-for example, in order of the position of the clones in the final map-then these numbers would change. The Librarylocation of each clone given here indicates its position in the current arrayed representation of the library (see Obtaining Materials). In each plate the clones in the wells 1 through 40 (A1 through D4) contain inserts in the P1 vector NS582 tet14ad10. In each plate the wells 41- 96 (D5 through H12) contain inserts in the P1 vector ad10sacBII.

Preparation of P1 probes and in situ hybridizations P1 DNA from individual clones was prepared by an alkaline lysis protocol. The DNA was labeled by a random hexamer method to incorporate biotinylated dNTPs. Polytene chromosomes were prepared from the Drosophila strain Oregon R. After denaturation and dehydration pre-treatments the biotinylated probe was hybridized to the polytene chromosomes overnight at 37 degrees C in 1.4 x SSC, 7% dextran sulfate, 35% N,N- dimethylformamide, and 0.6 mg/mL sonicated salmon DNA. After washing the chromosomes, the hybridization signal was developed using a horseradish peroxidase conjugate and 3,3'-diaminobenzidine. Finally the chromosomes were stained with Giesma stain and embedded in Permount. The slides were visualized by light microscopy.

Analysis of in situ hybridization results Analysis of in situ hybridization results. The in situ localization of 2653 clones from the master library has been acquired in the process of making the framework map. Of these clones, 2317 hybridized strongly to a single (or occasionally two or three) euchromatic sites. Occasionally a clone hybridized to a single location strongly and many more weakly. This result is expected if the clone carries one or more dispersed repetitive elements; such sequences are common in the Drosophila genome. In such cases the localization was ascribed to the strong site of hybridization. About 5% of the total clones analyzed hybridized to two or three distinct euchromatic sites. These clones could represent either chimeric clones constructed in the ligation reaction whereby two disparate pieces of genomic DNA became ligated into the same vector, or clones that contain duplicated genes. A third explanation of these results is cross- contamination of adjacent wells in a microtiter plate. The initial set of probes that were made by Hartl and colleagues were done so by growth of clones directly from the microtiter well, not from a single colony. To address this issue directly, Hartl later showed that some of these multiple site in situ hybridization results vanished when the probes were re-made from P1 DNA isolated from re-streaked single colonies rather than the primary library well. Together these euchromatic clones provide a 1.5-fold coverage of the euchromatic genome and should statistically cover nearly 80% of the euchromatic genome. The remaining set of 336 clones hybridized to the chromocenter (the pericentric region and the Y chromosome, typically under-replicated in salivary gland cells) and/or multiple euchromatic sites.