S35 thermal cycler contamination

Michael Finney finney at Frodo.MGH.Harvard.EDU
Fri Jul 23 16:38:01 EST 1993

In my alternate identity as chief scientist of MJ Research, I've become 
concerned about reports of S35 contamination of thermal cyclers.  What 
follows is a summary of what we think is going on, some questions that 
we would like to ask anyone who has information, and some suggestions 
for controlling and cleaning up contamination.

Radioactive contamination has been reported by people using S35 (but not 
P32 or P33), principally for cycle sequencing.  What appears to be 
happening is that the thiophosphate nucleotides break down chemically, 
yielding a sulfur-containing product that is volatile at high 
temperatures.  One possibility is H2S, but there are a number of others.

Whatever it is, the radioactive compound can get out and contaminate 
things in two different ways.  First, it can get out around the rim of a 
tube.  This is likely to be a problem only if you are using an oil-less 
configuration (an oil overlay should prevent volatilization) and the 
amount of contamination should vary with the brand or lot of tubes.  
Second, it can get out through the wall of the vessel.  The 
polycarbonate plates sold by MJR and other vendors are not permeable to 
liquids, but they are slightly permeable to gasses (such as water 
vapor).  Thus we would not be surprised if some S35 could go through the 
walls of the plates.  Thin-wall polypropylene tubes are NOT permeable to 
water vapor, so we would be somewhat surprised if S35 compounds could 
get through them (but we can't absolutely rule it out either).  These 
tubes are usually used without oil, so we would rather suspect leakage 
around the rims.

We don't know anything about the chemical identity of the offending 
molecule (ask NEN or Amersham!) but there may well be different amounts 
in different lots or brands of label and the amount may increase with 
the age of the label or the number of times it's been freeze-thawed.  A 
few years ago, there was a similar problem with S35-methionine for 
metabolic labeling of proteins.  Does anyone know how this was resolved? 

I would like to ask anyone who has done S35 cycle sequencing to drop me 
a quick note (by email; I will post results) answering if possible the 
following (and anything else useful):
1  Did you experience contamination?
2  Where was the contamination (block, lid, gloves, etc.)?
3  How severe was the contamination?
4  What type and brand of reaction vessel did you use?
5  Did you use an oil overlay?
6  Did you use oil in the thermal cycler block?
7  What was the temp. and time of denaturation, and # of cycles?
8  What was the brand of the label?
9  How old was the label?

Finally, what to do about it.  This is preliminary, and includes some 
helpful suggestions that were previously posted.
1  Use P33.  Yes, I know it's expensive, but it will definitely solve 
the problem and it works great.
2  Put your cycler in a hood.  If the S35 can get out of the tube, 
you're breathing it, and that's not good.
3  Keep denaturing times and temperatures to a minimum.  This will limit 
both volatilization and the pressure inside vessels.
4  For plates, use a small amount of oil in the thermal cycler block.  
This will limit the amount of gas (including water vapor) that gets out.  
An easy way to do this is to dip the underside of a plate into a small 
dish (pipet tip box top?) of mineral oil before putting the plate into 
the cycler.
5  Use oil overlays.  I know it's inconvenient, but it's better than 
losing your radioactive license.
6  This is speculative.  Add something to the reactions that will sop up 
the guilty compound.  Possibilities include dithiothreitol (DTT), 
oxidized DTT, and Fe II or Fe III.  Be judicious with the iron, as it 
can attack DNA, especially in the presence of EDTA.

As for decontamination of thermal cyclers, we don't know of anything 
that works well.  Detergents apparently have little value, so the S35 
may well be covalently bonded.  Reasonable candidates are mild bases 
(NH4OH), reducing agents (DTT), or oxidizing agents (bleach, hydrogen 
peroxide).  Don't use bleach on bare metal (all MJR cyclers have 
anodized blocks, so bleach can be used _occasionally_).  We would love 
to hear from anyone who has something that works.

Mike Finney
finney at ochre.mgh.harvard.edu

More information about the Methods mailing list