How a virus is neutralised depends upon several interacting factors. The most important of these, but by no means the only ones, are the interaction of a specific epitope, the properties of the cognate antibody, and the nature of the host cell. Variation in any one of these may change the mechanism by which neutralisation occurs and may obviate neutralisation altogether. The purpose of this article is to present an overview of neutralisation; for a more detailed discussion the reader should consult Dimmock' for references up to 1993 and more recent reviews2--'
DEFINITIONSNeutralisation is the loss of infectivity which ensues when antibody molecule(s) bind to a virus particle, and usually occurs without the involvement of any other agency. As such this is an unusual activity of antibody paralleled only by the inhibition of toxins and enzymes. However, neutralisation by some antibodies may require or be assisted by, for example, complement or cells bearing receptors for the Fc portion of the antibody molecule. The term neutralisation is best used only in relation to in uifro systems as there is little information about how antibodies act in uiuo. The latter is defined here as protection.
NEUTRALISATION EPITOPES
Exposed epitopesMost epitopes are exposed and, in contrast to cryptic epitopes (see below), are available on the surface of a virion to bind antibody at all times. Most but not all viral neutralisation antigens are proteins. Figure I shows the relationship between epitope, antigenic site and antigen. An antigenic site is a collection of overlapping epitopes, but its definition is limited by the number of mAbs which are available for this purpose. In fact it is debatable whether or not discrete antigenic sites exist or' whether an antigen is a continuum of epitopes. However, the concept of an antigenic site highlights that there are regions of immunodominance to which the immune system responds by preference. Not all epitopes mediate Abbreviations used: CDR, complementary determining region. neutralisation. This is an important fact which emphasises that binding of antibody to an epitope is not in itself sufficient for neutralisation, and that neutralisation epitopes have the unique functional property which leads to loss of virus infectivity. This fact is further and very well demonstrated by some neutralising antibody-escape mutants which are still able to bind the selecting mAb (Table I; The epitope itself is unchanged, and it is likely that a mutation in another part of the molecule prevents neutralisation; this in turn implies that neutralisation via this epitope depends on the mAb being able to trigger an event within the neutralisation protein which occurs subsequent to binding.
Cryptic epitopesAlthough postulated to occur some years ago, evidence for the existence of such epitopes has only recently been obtained. As their name indicates, cryptic epitopes are not normally available to bind antibody but are revealed by conformational changes in the virion which occur when it interacts with its environme...