Confusion about the prefixes D- and L-
All of the pages in this series are in urgent need of updating.
The biochemical principles have not changed, of course, but textbooks have: some of
those that were current when I first prepared these pages in 2000 have appeared in new editions, and
others have ceased to be widely used. New books have appeared that are not discussed. Unfortunately
I do not have easy access to any of the commonly used textbooks, as I work in a research (not teaching)
environment in a country where English is not the everyday working language. I could buy them, of course,
but that would represent rather a large investment for the sake of a few web pages.
Accordingly
I should be grateful if someone would collaborate with me in the revision. If you have access to
all of the textbooks published in English in the past ten years (say 1996 or later)
that are commonly used for teaching biochemistry, and if you would like to
help, please contact me at acornish@ibsm.cnrs-mrs.fr.
An example of the problem
Campbell, pp. 69–70:
The two possible stereoisomers of another chiral compound, L- and
D-glyceraldehyde, are shown for comparison with the corresponding forms of alanine. These two forms of glyceraldehyde are the basis of the classification of amino acids into L and
D forms. The terminology comes
from the Latin laevus and dexter, meaning left
and right,
respectively. The two stereoisomers of each amino acid are designated as L and
D amino acids on the basis of their similarity to the glyceraldehyde standard. In the L form of
glyceraldehyde the hydroxyl group is on the left side of the molecule, and in the D form it is on the right side,
as shown in perspective in Figure 3.3 (a Fischer projection). In an amino acid, the position of the amino group on the left or right side of the
*gkalpha carbon determines the L or
D designation. The amino acids that occur in proteins are all
of the L form.
After a giving a correct account of the (RS) system often used in natural products
chemistry, Mathews, van Holde and Ahern (p. 281)
give the following explanation of why it is little used by biochemists:
Although the R–S convention is more general, it is little used by biochemists. It becomes
difficult to apply in the common situation in which a molecule contains more than one asymmetric
carbon atom.
The stereochemical principles involved
In the early days of stereochemistry the only relevant property that could easily be measured was optical activity,
i.e. the capacity to rotate the plane of polarized light clockwise or
anticlockwise. These were called dextrorotatory and laevorotatory respectively, and represented by the prefixes
d- and l- respectively (Note. Lower-case letters, not D and
L). Ordinary glucose, for example, is dextrorotatory and is thus d-glucose, whereas ordinary fructose is laevorotatory and
is thus l-fructose, and these two sugars are sometimes called dextrose and laevulose respectively to reflect their
optical properties.
It was later realized that opposite
optical activity meant opposite (mirror-image) arrangement of the atoms in space, or configuration.
However, only very much later did it become technically
possible to determine this arrangement—i.e. for a long time it was known that the two forms of lactic acid were
mirror images of one another but there was no way of knowing which was which.
During this long period, chemical methods were developed that allowed different molecules to
be related seterochemically to one another. So,
for example, although there was no way of knowing the actual arrangement of atoms around the asymmetric carbon atoms in
serine and alanine obtained by protein hydrolysis, it was
possible to deduce to deduce that both of these amino acids (as well as all the other asymmetric amino acids obtained
by protein hydrolysis) had the same configuration. Thus if the configuration of natural serine was defined
as L it followed that all the others were L as well. It is more complicated for the hexoses, including
glucose and fructose, because
each of these contains several asymmetric carbon atoms. However, similar experiments to relate the sugars to one another showed that
although there were variations at some of the asymmetric atoms, the configuration at C5 was always the same. d-Glucose (optical symbol) was
then defined as D-glucose (configuration symbol), which meant that l-fructose was also D-fructose.
Not surprisingly, this system of symbols led to confusion, and in consequence the old d- and l- prefixes have been abandoned. You
should never see them except in very old textbooks or historical accounts. In modern usage they are replaced by (+) for d, and (-) for l. The
modern symbols are not common in biochemical sources because biochemists are not often interested in the polarizing properties of
molecules. By contrast, they are quite often interested in the configurations of molecules, i.e. the arrangements of atoms around a chiral atom,
and so the configurational prefixes D and L are in common use. More important, biochemists frequently deal
with series of related molecules, the amino acids, for example, or the hexoses. They need, therefore, symbols that express in a clear way
whether two related molecules have the same or different configurations. When chemists work with the same sorts of molecules they have
the same requirements, and they also continue to use D and L.
The terms dextrorotatory and laevorotatory refer to an optical property
that can be observed and measured without any knowledge of the detailed molecular structure. The
modern configurational symbols D and L echo their historical
origins in measurements of optical rotations, but they have no clear relationship to the normal
meanings of right and left. Even though it is of course possible to represent a molecule on paper
in such a way that a particular group is on the left or the right, the actual molecule in three
dimensions has no left
or right
side except in relation to some arbitrary convention.
In addition, however, chemists often need to define a configuration unambiguously in the absence of any reference compound, and for this
purpose the alternative (R,S) system is ideal, as it uses priority rules to specify configurations. These rules sometimes lead to absurd results
when they are applied to biochemical molecules. For example, as we have seen, all of the common amino acids are L, because they
all have exactly the same structure, including the position of the R group if we just write the R group as R. However, they do not all
have the same configuration in the (R,S) system: L-cysteine is also (R)-cysteine, but all the other L-amino
acids are (S). However, this just reflects the human decision to give a sulphur atom higher priority than a carbon atom, and does not reflect
a real difference in configuration. Worse problems can sometimes arise in substitution reactions: sometimes inversion of configuration can result
in no change in the (R) or (S) prefix; and sometimes retention of configuration can result in a change of prefix.
It follows that it is not just conservatism or failure to understand the (R,S) system that causes biochemists to continue with D
and L: it is just that the DL system fulfils their needs much better. As mentioned, chemists also use D
and L when they are appropriate to their needs. The explanation
given above by Mathews, van Holde and Ahern
of why the (R,S) system is little used in biochemistry is thus almost the exact opposite
of reality. This system is actually the only practical way of unambiguously representing the
stereochemistry of complicated molecules with several asymmetric atoms, but it is
inconvenient with regular series of molecules like amino acids and simple sugars.
Acknowledgement. I thank Professor Ulrich Müller (Marburg, Germany) for pointing out that the term asymmetric centre
introduced by
Cahn, Ingold and Prelog is illogical and self-contradictory, as the concept of a centre of symmetry only makes sense when there is symmetry. This and similar
terms have therefore been eliminated from this discussion.
Textbook checklist
Other common errors in textbooks
List of books considered
