On Tue, 23 Jan 2001, Carrie Kyle wrote:

> John Kiernan wrote:
>> "Note that the name PLP was poorly chosen by the inventors
>> of this mixture, because the solution does not contain
>> any paraformaldehyde. For a good paper on how it works,
>> see DC Hixson et 4 al 1981 J Histochem Cytochem 29:561-566."
> 
> I beg to differ with this statement.  I used PLP fixative for years when I was working in a research lab and it is made with paraformaldehyde.  A quick search produced the following reference if you're interested:
> MacLean IW, Nakane PK.  Periodate-lysine-paraformaldehyde fixative.  A new fixative for immunoelectron microscopy.  J Histochem Cytochem 1974;22:1077-84.

Yes indeed. The MacLean and Nakane paper gave the first account of this
fixative, which is always made using paraformaldehyde. The name PLP is
confusing because it conveys the impression that paraformaldehyde, a
substance that does not dissolve in any liquid, is present in the 
fixative solution. 

When paraformaldehyde is heated in water in the presence 
of sufficient added acid or base it appears to dissolve
when the temperature reaches about 60C. What actually happens is
that the large paraformaldehyde molecules decompose yielding as
product methylene hydrate, HOCH2OH, which is the gem-diol or
hydrated form of formaldehyde. This is the compound present in
all formaldehyde-containing fixatives, whether the starting material
is paraformaldehyde or formalin. The only difference is that a
4% formaldehyde fixative made from formalin contains about 1%
methanol. This is from the 10% included with commercial formalin
as a preservative to retard polymerization, which eventually leads
to precipitation of insoluble paraformaldehyde. This is often seen
in bottles of formalin that have stood in a cold place (such as a
British laboratory in winter). 

An old bottle of formalin contains more than 10% methanol and also 
some formic acid. These are the products of a Cannizzaro reaction
in which two formaldehyde molecules mutually oxidize and reduce
one another, the products being methanol and formic acid. A
buffered formaldehyde fixatve made with old formalin is therefore
likely to contain somewhat less formaldehyde (actually methylene
hydrate) than one made from new formalin or from paraformaldehyde,
and also more than 1% methanol and some formate ions. The 
concentration of formaldehyde in a fixative is not critical, and
there is uncertainty about whether a little methanol and formate
make any difference. The reason for using paraformaldehyde as a
starting material is that it gives a solution of known composition.

The L in PLP is also inappropriate because by the time the working
solution is used it contains little or no lysine, a fact apparently 
not appreciated by MacLean & Nakane (1974) but recognized by Hixson 
et al. (1981) [refs at top of this email]. The lysine in "PLP" reacts
rapidly with formaldehyde, forming polymers that can combine with 
proteins to make cross-links of varying length - an action comparable
to that of the polymers present in glutaraldehyde-containing fixatives. 
Aldehyde groups of glycoproteins, formed by periodate oxidation of
their sugar side-chains, cross-link directly to other protein molecules 
of the tissue, not to lysine derived from the fixative as was originally
postulated.

It would probably be possible to make up a chemically correct
name and abbreviation for MacLean & Nakane's fixative, but nobody
would use it, so why bother? 
 
----------------------------------------
John A. Kiernan
Department of Anatomy & Cell Biology
The University of Western Ontario
London,  Canada   N6A 5C1
   kiernan@uwo.ca
   http://publish.uwo.ca/~jkiernan