restriction enzymes

[Chris Boyd]

Ashok Aiyar (aiyar at aiyar.dyn.ml.org) wrote:
: On 2 Aug 1998 17:57:03 -0700, Markus Schneemann wrote:

: >I'm looking for a restriction enzyme that cuts human (eukaryotic)
: >genomic DNA into smaller pieces than it does with bacterial (gram +
: >actinomyces, GC-rich) chromosomal DNA.
: >The purpose is to generate a bacterial chromosomal library from
: >intracellularly grown bacteria (grown inside human macrophages) and to get
: >rid of the human genomic DNA that may contaminate the bacterial DNA.
: >If the human DNApieces are considerably smaller (let's say, below 10-15kb)
: >than the bacterial ones (>25kb) the packaging of cosmids (Stratagene
: >SuperCos) should exclude any DNA below 20kb (so they promise...).
: >Does anybody know of such a restriction enzyme ? 

: If the genomic DNA in your bacteria is dam-methylated, then you can 
: digest the total "genomic" DNA preparation with DpnII, a four-cutter
: that cuts the sequence GATC when it is unmethylated.  

: There is no equivalent to dam-methylase in mammalian cells, and therefore 
: human DNA is cut efficiently by DpnII.  In contrast, many eubacteria 
: dam-methylate their genome, and the resultant GAmTC is resistant to DpnII.

Where is your evidence for this assertion?  I'd be surprised if
dam-methylation were present in more than a minority of species outside
the Enterobacteriacae.

: If the bacterial genomic DNA is indeed methylated, you could first
: digest exhaustively with DpnII, to cut the contaminating human genomic DNA
: into fragments far smaller than 10 kb, and then cut the undigested
: bacterial genomic DNA with the enzyme of your choice to yield fragments
: that can be packaged into phage heads after ligation with lambda arms.

It's a nice idea, but in this case take a look at...

  TI: RESTRICTION ANALYSIS OF ACTINOMYCETES CHROMOSOMAL DNA
  AU: NOVELLA_IS, MARIN_I, SANCHEZ_J 
  JN: CANADIAN JOURNAL OF MICROBIOLOGY, 1996, Vol.42, No.2, pp.201-206
  AB: Actinomycetes DNAs were digested with restriction enzymes to study
      the presence of methylated bases. Analysis showed that the  
      enterobacterial Dam and Dcm systems are absent. Methylation at the  
      internal cytosine in CCGG sequences, typical of eukaryotes, was also
      absent. We also tested 18 restriction endonucleases recognizing six
      base pair sequences (all of which were inhibited by methylation).
      Results showed a higher number of restriction sites for enzymes  
      recognizing CG-rich sequences (CG endonucleases) than for enzymes
      patterns with CG endonucleases were quite uniform, with the
      remarkable exception of XhoI, which yielded a small number of DNA   
      bands. The study performed with AT endonucleases allowed 
      differentiation of three groups of enzymes based on different degrees
      of chromosomal sensitivity. One group (BglII and BglII) produced
      restriction patterns with more abundant restriction sites than  
      expected, a second group (ClaI, EcoRI, and EcoRV) yielded the
      predicted number of DNA fragments, and the third group (HpaI,     
      HindIII, XbaI, and DraI) produced an unexpectedly low number of
      fragments. Some individual cases of resistance to particular enzymes
      could be explained by the presence of restriction-modification  
      systems with the same specificity.    

Your best bet is either to take not of the results of the above or
repeat this kind of study with a panel of (four-cutter?) enzymes
against chromosomal DNA from your particular actinomycete and identify
one that didn't cut.  Use this to chop up the human DNA preferentially
as Ashok suggested.

Best wishes,
-- 
Chris Boyd                      | from, but not \ MRC Human Genetics Unit,
Christopher.Boyd at hgu.mrc.ac.uk  | on behalf of  /  Western General Hospital,
http://www.hgu.mrc.ac.uk/Users/Christopher.Boyd \   Edinburgh, EH4 2XU, SCOTLAND