Western blotting variability

[Bernard Murray]

In article <3347C7B7.6307 at sccwrp.org>, jeffb at sccwrp.org says...
>
>Hello
>I'm trying to get a semi-quantitative western blotting procedure to work 
using alkaline 
>phosphatase conjugated secondary antibody and image analysis.  A 4 point 
standard calibration is 
>included in all gels, and the unknown samples are within the linear range. 
 Three replicates of 
>each sample are run on three different gels.
>The problem is with between blot variability (CV=25%).  I have also tested 
within blot 
>variability, and this too can be high.
>I'm using a BioRad mini-protean II electrophoresis setup (BioRad 10% ready 
gels)and mini 
>transblot module (overnight tank transblotting) with supported 
nitrocellulose.
>Any suggestions to reduce the variability will be appreciated.
>Thanks you
>        Jeff

Below is a *personal* opinion.  Be warned.
I did a direct comparison of detection with HRP and AP linked second
antibodies.  I found that AP gave the most sensitive detection (BCIP/NBT for
colour or Attophos and a fluorimager for chemifluorescent.  I didn't get
around to trying any of the chemiluminescent substrates.  In contrast
detection with HRP linked second antibodies (chloronaphthol for colour
or ECL for chemiluminescent) was less sensitive but more linear and more
reproducible.
	I tried two different AP and HRP second antibodies but I fully
admit that the antibody moiety may be the major source of the difference.
At the time I attributed it to the nature of the enzymes - HRP kills itself
during catalysis and so is self limiting, whereas AP continues its activity
for some time so the amount of enzyme on the membrane may not be rate
limiting.  For my own purposes I now use AP for semiquantitative/detection
purposes and HRP for formal quantification.
	This opinion may be heretical as far as some people are concerned
but you did ask for suggestions....
		Bernard

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