mkw6d@krebs.acc.Virginia.EDU (Martyn K. White) (11/13/90)
To: GRAZIAWD@snyplava.bitnet Subject: Re: Problem Query Newsgroups: bionet.molbio.methds-reagnts In-Reply-To: <9011121617.AA05492@genbank.bio.net> Organization: University of Virginia Cc: Bcc: In article <9011121617.AA05492@genbank.bio.net> you write: >Date sent: 12-NOV-1990 11:01:52 > >Recently we were having problems searching for an interferon regulatory >protein clone in a lambda gt10 library. So, we decided to access the >quality of our library by screening for bovine actin cDNA (SP64-actin). >The plaques were unusual, i.e., quite small. So we decided to extract >DNA from the library using the procedure in Current Protocols in Mol. Bio. >Vol 1 (1.13.7) and hybridize the DNA blot (DNA -/+ EcoR1) to [32]P SP64- >actin. Results showed the majority of signal at 2kb (insert is 2kb) with 3 >larger bands of much less intensity.However, one these signals was at ca. 50kb >for both -/+ EcoR1. Surprise and confused! Since the procedure requires the >use of DNase & RNase treatment to lysate prior to phenol extraction, could >this be the problem. Any suggestions or help would be appreciated. We are >in the process of extracting the DNA without DNase & RNase. Send responses >to GRAZIAWD@SNYPLAVA.BITNET. Thanks in advance. > >Regards, > Bill :-) > >+-------------------------------------------------------------------------+ >| Voice: (518) 846-7144 | William D. Graziadei, PhD | >| Fax: (518) 846-7144 Ext 40 | In Vitro Cell Biology & | >| (518) 564-7827 | Biotechnology | >| Bitnet: GRAZIAWD@SNYPLAVA | SUNY Plattsburgh/Miner Inst| >|Internet: GRAZIAWD@SPLAVA.CC.PLATTSBURGH.EDU| Chazy, New York 12921 | >|-------------------------------------------------------------------------| >| EMail saves ca. 5x10e8 tree cells which would cover a std printed page. | >+-------------------------------------------------------------------------+ I think the problem is that in libraries in lambda GT10, one or both of the EcoR1 sites flanking the insert can be lost by some unknown mechanism. This occurs at a surprisingly high rate and is very annoying. Thus I have isolated clones and then tried to subclone them using EcoR1 only to find that in a significant percentage of the clones EcoRI will not cut at either one or both sites. This is due to a mutation in the sequence and not to "dirty DNA". Thus your three extra bands would be: actin + short arm of GT10 (1 EcoR1 site lost) actin + long arm of GT10 (other site lost) actin + both arms (both EcoR1 sites lost) You could test this (if you are interested) by cutting with both HindIII and BglII which should give you a single band which is somewhat bigger than the expected size of the insert (if my theory is correct). HindIII and BglII flank the "EcoR1" sites. Clones with this problem can be subcloned using these two restriction enzymes into suitable polylinker-containing plasmid and sequenced at their termini using lambda GT10 primers (BRL). -- Martyn K. White, Box 441, U.Va, Cville VA 22908 (804) 924 8710 mkw6d@krebs.acc.Virginia.EDU mkw6d@Virginia.EDU This is my .signature and I will say what I like. Martyn K. White, Box 441, U.Va, Cville VA 22908 (804) 924 8710 mkw6d@krebs.acc.Virginia.EDU mkw6d@Virginia.EDU
LAFLAMME@MAINE.BITNET (11/15/90)
I too have found that the EcoRI sites one expects to find flanking a cDNA insert are often non-functional. This proved to be the norm and not the exception in my last experiments involving gt10. I was able to subclone HindIII/BglII fragments into HindIII/BamHI cut vectors. Good Luck!. ********************************************************************* * * * * Dan Laflamme * * * Dept. of Biochemistry * * * University of Maine * * * Orono, Maine 04469 * * * (207) 582-2818 * * * * * *********************************************************************