hygromycin phosphotransferase

Brian Foley brianf at med.uvm.edu
Thu Jun 16 09:30:55 EST 1994

Bernard Murray (bernard at elsie.nci.nih.gov) wrote:

: Just curious....
: 	Why do you need the antibody?  Wouldn't it be sufficient to test the
: hygromycin sensitivity of the transfected lines, or are there complicating
: factors?  

	Yes, this works fine for me.  My untransfected cells (CHO-K1) 
give a typical dose-response curve when tested after 24 hours of exposure
to various concentrations of Hygromycin (0.1 unit/mL; 1 U/mL; 10 U/mL,
100 U/mL, 1000 U/mL) For the assay, I count 35-S methionine incorporated
into TCA-precipitatable material (protein) during a one hour pulse
labeling, followed by a 1 hour cold media chase.
	I plant the cells directly in sterile glass scintilation vials,
so the whole assay is quite simple.  The replicate vials are grown 12 hours,
then treated with hygromycin for 24 hours, then pulse-labeled and chased
for one hour each, then TCA-precipitated right in the vial.  The
5% TCA is aspirated off, and scintilation fluid is added.  I typically
plant duplicate or triplicate vials for each point on the curve for
publishable data.  If I am just checking a few transfections, I pick 
fewer hygromycin concentrations, but stll do duplicate or triplicate
vial samples.  You never know when one vial will get broken or something.
	Transfected cells show a measurable increase in hygromycin
resistance.  Transfected cells which have been selected in 300 Units/mL
hygromycin for two weeks, show a marked increase in resistance, even if
they are grown in non-selective media for a period of time before the

: Alternatively, how about cutting out the hyg gene from the plasmid
: and using it to probe Northerns?

	That would also be a good idea.  Other good ideas are to
pick some individual clones so you know what a high and low producer look
like, rather than assaying a whole population of transfected cells, which
just gives you an average value.  This applies to Northerns or to any
other assay.   Also make sure that production of your protein of interest 
correlates with production of the selectable marker.  This is not always 
the case, even when both transcripts use the same promoter.  Differences 
in mRNA processing rates, and translation rates can be very significant.  
Also keep in mind that the Km and Ka of hygromycing phosphotransferase 
for hygromycin may be significantly different than your protein of 
interest for its substrate.  So even if you have equal levels of protein 
production, the hygromycin resistance may be impressive, while activity 
of your protein is still non-detectable.

: 	While we are on the topic, has anyone any experience of cells
: becoming hygromycin resistant without transfection with this gene?  

	Yes.  It happens at a very low level (between 1 in 10 million cells
and one in 100 million cells becomes resistant) in CHO-K1.  The 
resistance is due to mutation(s) in one of the ribosomal RNA genes.  There
are several papers published on this.

: I believe
: that G418 is not a substrate for classic multidrug resistance (pgp/P170/mdr)
: but I do not know if hygromycin has been tested.
: 		Bernard

	I doubt it is a substrate for P170 tranport, but there a many
other ways in which cells can become resistant to agents.  I work with
resistance to bacterial toxins, and for diptheria toxin there are at
least 15 genes involved in toxin sensitivity.  There are gene products needed
for toxin binding to the cell, for entry of bound toxin into the cell, for
modifying the EF-2 substrate, and for other factors as yet unknown.  
Mutations in these genes all result in a recessive phenotype (a cell-cell
hybrid between the toxin-resistant and wild-type cells results in a 
toxin-sensitive hybrid).  Then there are other mutations which provide
dominant toxin resistance.  These include mutations in the elongation
factor 2 gene.

	This points out the need to have realatively good transfection
efficiency when selecting for transfected cells.  If you only get one
or a few colonies on your plate, they could be due to mutations in
an untransfected cell, rather than due to transfection.  Always do a couple
of control plates treated with another plasmid lacking the selectible marker.

	The mutation rate for different genes in one cell line, or the 
same gene in different cell lines is highly variable.  So just because 
mutation to hygromycin resistance is low in CHO-K1 do not assume that 
this is true in your cel line.  I have some cell lines that mutate to
diphtheria toxin resistance at such a high rate that I cannot work with 
them.  This even happens with recently-cloned sub-strains of that line, 
eliminating the possibility that it is just due to a mixed population of 
cells, some of which already carry a mutation.

: Bernard Murray, Ph.D.
: bernard at elsie.nci.nih.gov (National Cancer Institute, NIH, Bethesda MD, USA)

*  Brian Foley               *     If we knew what we were doing   *
*  Molecular Genetics Dept.  *     it wouldn't be called research  *
*  University of Vermont     *                                     *

More information about the Methods mailing list