Here's a reponse I got from Ian York; I'm posting it for him because
his posting software is not working at the moment...
References: <3v40i0$o2b at grape.epix.net> <3v5ual$egh at cisunix1.dfci.harvard.edu> <3vp30o$ifl at jive.cs.utexas.edu>
Organization: Dana-Farber Cancer Institute
Cc:
Status: OR
In article <3vp43l$1cn at jive.cs.utexas.edu> wilson at cs.utexas.edu (Paul Wilson) writes:
>>I now suspect I didn't grasp the "antibody mediated" thing. Are you saying
>that Ebola does *well* when there are antibodies to it? E.g., because
>it attacks its attackers successfully and uses them as trojan horses?
OK, let me amplify the story a little bit. First, as far as Ebola is
concerned, this is almost entirely speculative. This is built on the
observation (reported in ProMed mailing list) that the one monkey that
survived Reston infection, died more rapidly than naive animals when
exposed to Ebola. (Correct me if I've got this wrong, please.) So (1)
this is a single observation with (2) minimal controls and (3) I don't
know the circumstances of the reinfection: there may have been lingering
effects from the Reston infection, or whatever.
Given that, I still thought it was of interest to mention that
antibody-mediated enhancement of infection could be occuring here. The
classic example of this is dengue virus; feline infectious peritonitis
and HIV may also make use of it as well, and there are other examples. if
you're interested in a review of the phenomenon, take a look at:
Morens DM.
Antibody-dependent enhancement of infection and the pathogenesis of
viral disease.
Clinical Infectious Diseases. 19(3):500-12, 1994
To simplify a bit, what you need is two things: a virus that can
replicate in cells expressing Fc receptors, and non-neutralizing
antibody. The Fc-expressing cells include monocyte/macrophages and some
lymphocytes. Since Ebola replicates in monocytes, it meets the first
requirement.
The non-neutralizing antibody thing is more complicated. I don't have
room to go into the mechanisms by which antibody can neutralize
extracellular virus; there are many. Suffice it to say that it is
possible for antibody to bind to certain epitopes of a virus without
neutralizing it. (In fact, for many if not most viruses, the majority of
antibodies may be non-neutralizing; since the usual response is very
widley polyclonal, though, most antibody responses to most viruses are
neutralizing.)
Some viruses are less susceptible than others to extracellular
antibody-mediated neutralization: FIP is highly resistant. In other
cases, high or normal levels of antibodies neutralize but as the
concentration drops (or if the response is directed mainly against a
limited subset of epitopes) the neutralizing ability drops. The latter is
more typical of dengue: the classic situation is when a person is infected
with one serotype of dengue and then a second. The antibody response to
the first strain is neutralizing to that strain; but it does not
neutralize the second strain, even though the antibodies can bind. Thus
you have all the requirements: antibody binding without neutralizing to a
virus which can replicate in monocytes. The monocyte adsorbs the
antibody-coated virus more efficiently than just the plain virus, so
infection of monocytes is more efficient, and you're off to the races.
I think the Reston/Ebola situation may be analagous to dengue; whether
this is true for antibodies raised to ebola itself is undecidable now.
Again this is all speculative based on a single inadequate (but
intriguing) experiment.
Ian
--
Ian York (york at mbcrr.harvard.edu)
"As things are, most Usenet postings are of no interest to socially
well-adjusted people over the age of twelve." -Nature 376:200 (1995)
--
| Paul R. Wilson, Comp. Sci. Dept., U of Texas @ Austin (wilson at cs.utexas.edu)
| Papers on memory allocators, garbage collection, memory hierarchies,
| persistence and Scheme interpreters and compilers available via ftp from
| ftp.cs.utexas.edu, in pub/garbage or http://www.cs.utexas.edu/users/oops/)
