A time-saving alternative to conventional library plating and screening
Anuradha Mehta Donna Driscoll
Department of Cell Biology, Cleveland Clinic Foundation, Cleveland, OH
We chose Stratagenes Human Universal cDNA Library 1 (HUCL) to isolate a full-length cDNA clone encoding a novel low-abundance protein called Apobec-1 Complementation Factor (ACF) for several important reasons: The simple two-step hybridization protocol allowed us to quickly screen a large number of clones, normalization increased the chance of finding low-abundance cDNAs, the librarys varied tissue base aided in isolating a protein that was widely expressed but only in small amounts, and the average size of the library clones ranged within ACFs expected sequence size. We successfully isolated ACF cDNA, which was crucial in allowing us to define the minimal protein requirements for the C to U editing of apolipoprotein-B mRNA.
Purification of proteins and identification of their cognate cDNAs are powerful tools for studying the molecular mechanisms underlying biological systems. Highly sensitive mass spectrometric techniques provide valuable peptide sequence information from very limited amounts of protein, which can be used to screen the wealth of information available in the NCBI databasesincluding both nonredundant and Expressed Sequence Tag (EST) databases. In the case of a novel protein, identi fying the correct complete cDNA sequence is often hindered by the short sequences of the reported ESTs that are biased towards the 5 end. Since the EST cDNA inserts are generally not full-length and the clones are not always available, there is still a need to screen libraries to isolate a full-length cDNA. However, conventional library screening is tedious and time consuming, especially for a low-abundance protein.
To circumvent these problems, we used Stratagenes HUCL arrays to isolate a full-length clone that encodes Apobec-1 Complementation Factor (ACF), a novel RNA-binding protein that is involved in the post-transcriptional editing of apolipoprotein B (apo-B) mRNA in mammals. The HUCL array is a collection of more than 145,000 primary clones in 1536 pools of 96 clones each, which are arranged on four master membranes. Because of the librarys large representation of high-quality cDNA clones, human genes can be quickly identified and subsequently isolated.
Editing apo-B mRNA involves a site-specific deamination reaction that converts cytidine6666 to uridine and generates a premature translational stop codon.2 Although the catalytic subunit of the editing enzyme was discovered in 1993, the identity of the RNA-binding subunit has been elusive, despite intensive efforts by several groups. We previously characterized and purified a 65-kDa protein from baboon kidney that had all the properties expected for the RNA-binding subunit. 3,4 The protein was purified over 140,000-fold, and silver staining of the final purified fraction showed the presence of a single protein of 65 kDa.4 To identify the protein, the 65-kDa band was subjected to tryptic digestion and microsequence analysis by mass spectrometry.
We obtained partial sequences from 23 peptides that were derived from two independent experiments. We designed highly degenerate primers from these partial peptide sequences, but our initial attempts at PCR-cloning were unsuccessful. The peptide sequences were compared with EST Genbank databases translated in all six reading frames (tBLASTn). At least nine of the peptides were identified in partial sequences from the four EST cDNAs (Figure 1). However, only two of the human EST cDNAs were available from the ATCC (Accession numbers N77737 and AA678055). We completely sequenced the EST cDNAs and found that each contained inserts of about 1 kb. Unfortunately, the sequences were nonoverlapping so we were unable to establish whether the two ESTs corresponded to the same cDNA.
To isolate the ACF cDNA, we used the two EST clones as probes to screen a human cDNA library. Stratagenes HUCL was chosen because it offered several benefits over conventional protocols. First, a large number of clones can be screened conveniently without the intensive labor of multiple platings and hybridizations. Second, ACF is a low-abundance protein and likely, the message encoding ACF is also in low abundance. The primary clones represented in the HUCL arrays are selected after extensive normalization that reduces the representation of high-abundance messages; hence, the likelihood of identifying low-abundance cDNAs is increased. Third, the HUCL was pooled from 29 different human tissues. This was advantageous since we and others5,6 have reported that the ACF activity is widely but not ubiquitously expressed. Normalization of pooled cDNAs from multiple tissues therefore addresses the question of tissue-specific variations in expression levels of ACF. Finally, the average insert size of the clones is 1.7 kb, a range suitable for identifying a 1.75-kb sequence predicted to code for ACF, a protein of 65 kDa.
The cDNA insert from the 5' EST clone (AA678055) was labeled with 32P-CTP and used as probe to screen the four HUCL master membranes.7 Hybridization was performed for an hour at 68C, and the blot was subjected to autoradiography. After adequate exposure, the first probe was stripped, and the HUCL membranes were reprobed with the 3' end (N77737) EST cDNA. Both nonoverlapping probes hybridized to the single cDNA pool at position F6 on membrane D (Figure 2). The coordinate position (F6) derived from the first screening was used to obtain the corresponding secondary membrane containing 96 individual clones from Stratagene. The secondary membrane was also screened sequentially using the two ESTs as probes. Both ESTs hybridized to a single clone at coordinate N8 of the secondary membrane. The purified cDNA clone corresponding to this position was obtained from Stratagene as a bacterial stab culture. Therefore, in two simple hybridization steps, we were able to identify a single cDNA that contained both EST sequences.
The HUCL clone contained an in sert of 1.94 kb and encoded an open reading frame of 586 amino acids.7 The predicted amino acid sequence contained 20 of the 23 peptides that were obtained by mass spectrometry, which was strong evidence that we had isolated the correct sequence. Based on DNA sequencing, the ACF cDNA encoded a novel protein with a predicted molecular mass of 64,274 Da.7 The N-terminal region of ACF showed the presence of RNA recognition motif (RRM)-type RNA binding domain, which is consistent with its function as the RNA-binding subunit of the editing enzyme. The C-terminal region of ACF (residues 304-586) was unique.
Once we had the clone in hand, we were able to quickly establish that ACF is required for editing apo-B mRNA both in vitro and in vivo. Taken together, our results support a model of the editing enzyme in which ACF is the RNA binding subunit that docks the catalytic subunit to edit the cytidine6666 of apo-B mRNA.
We determined that screening of Stratagenes HUCL was a straightforward and highly effective alternative to conventional library plating and screening. Using the HUCLs normalized cDNAspooled from multiple human tissuesgreatly facilitated our cloning of ACF, a very low-abundance protein. The single positive ACF clone was readily detected against a very low background after hybridization. The final purified clone was available as a bacterial stab for further characterization after only two simple screenings.
1. Lelias, J-M., et al. (1998) Strategies 11: 29-32.
2. Chan, L., et al. (1998) Strategies 11: 29-32.
3. Mehta, A ., Banerjee, S., and Driscoll, D. M. (1996) J. Biol. Chem. 271: 28294-28299.
4. Mehta, A. and Driscoll, D. M. (1998) Mol. Cell Biol. 18: 4426-4432.
5. Driscoll, D. M. and Zhang, Q. (1994) J. Biol. Chem. 269: 19843-19847.
6. Yamanaka, S., et al. (1994) J. Biol. Chem. 269: 21725-21734.
7. Mehta, A., et al. (2000) Mol. Cell Biol. 20: 1846-1854.
Purchase of these products includes a limited license to use these products for research purposes. For information on commercial use licenses, please contact Stratagene's Product Manager for libraries at (800) 424-5444.