Index Home About Blog
From: B. Harris)
Subject: Re: Viral Isolation: Pasteur Institute Rules Need Revision (fwd)
Date: Sat, 11 Oct 1997

In <> Greg Nigh
<> writes:

>I would like to pose an open question:
>How were the proteins that are considered HIV proteins derived? IOW, how
>was it determined that the HIV proteins *are* HIV proteins? From what
>material were those proteins derived? How was it *confirmed* that they
>were in fact produced by the retrovirus?

   The retrovirus has been cloned.  You take RNA from the sucrose
gradient and turn it into DNA with reverse transcriptase.  You get a
roughly 10 kB piece of DNA.  That DNA has pretty much the same sequence
whenever you do the experiment, so it's not some random piece of RNA.
It's a special piece.  It has the genes of a retrovirus, with common
sequences found in other lentiviruses.  It codes for a number of
proteins.  You can clone the DNA into bacteria and they MAKE the
proteins.  SDS-PAGE and various immune blot methods tell you that these
are the same proteins that infected cells secrete, and which are in the
sucrose gradient along with the RNA.  Thus, they must be viral
proteins.  You can raise antibodies to them, label these with gold, and
see with EM what they stick to.  Yep, they stick to those particles
that look like retroviruses from the sucrose gradient.  Also they stick
to the particles which look like retrovirus which bud from cells WHEN
you infect with the cloned set of DNA genes.   What the heck more do
you need?

                                    Steve Harris, M.D.

From: B. Harris)
Subject: Re: How Know You Have a Virus in Culture (was Re: Chemo  
	(Ultrastructure of HIV))
Date: Sat, 23 Aug 1997

In <> Greg Nigh
<> writes:

>This is not accurate. It is accurate to say that particles have been
>pictured which bud from cells and even that may (or may not) have the
>characteristic features of retroviruses. However, particles budding
>does not necessarily mean that they are what we think they are.

   Comment:  Let me try again.

   There is this family of proteins which are known as "HIV
proteins," but which we can call "X-proteins" for he sake of
discussion, here.  There isn't any question of their identificat-
ion or their structure or their "isolation."  We know their
sequences exactly, and we produce them routinely by genetic
engineering.  For example, in the 3rd generation ELISA tests for
>HIV-antibodies, the proteins for use in the tests are actually
made by recombinant techniques in microbes, using genes for the
"X-proteins."  We know what these proteins, are, okay?
We have given these "X proteins" names like HIVp24, HIVp17, and
HIVgp160.  This is not misidentification, because these proteins
are called these names by definition.  Ultimately the name does not
matter, so long as we can agree as to the nature of these proteins,
and the fact that they aren't a part of normal cells, and must
come from elsewhere.

  Do you have any problem with this?

  Now, these X proteins are viral proteins, and they are not endogenous
viral proteins.  We know this because the genes that code for the
proteins are NOT present in normal cells as those of an endogenous
virus would of course be.  There isn't any where they can hide from a
cell DNA digest and Southern blot.  Or from PCR techniques targetted to
find them.   However, these genes can be transferred to normal cells in
a filtered, cell-free extract, and this transfer causes cultured cells
to suddenly start actively making all these X proteins, which can then
be collected, sequenced, and/or identified by various means (SDS-PAGE,
for instance).  So the ability of cells to make these proteins is
infectious, and can be transferred from culture to culture.  Moreover,
such cultures secrete the genes to make these proteins, contained in
lipid particles of a characteristic density, just like those secreted
particles that contain the proteins themselves.  Addition of a
cell-free extract containing these particles to cells in culture,
causes the cells to incorporate the genes, and to make more of the

   A virus, by definition, is a package of genes and proteins
secreted by cells, which causes other cells to make more of the same
such packages.  Our X-proteins and the genes that code for them fit
this definition, and X-proteins are thus viral proteins, and the genes
that code for them are viral genes.  Moreover, limiting dilution assay
shows that only one virus type is involved with this protein family,
since dilution of cell-free extracts of X-proteins and their genes
reaches the point where the extract either transforms a culture into
making the whole family of X-proteins and genes, or else does nothing.
If more than one virus was involved, dilution would eventually
reach the point that cultures might be induced to only make some
of the viral X proteins in question, and not others.  But all
these X-proteins we're talking about go together, always.  They
are the proteins of one virus species, clearly.

    That's really all that is necessary.  We are talking about a
new virus species, defined by the family of viral proteins it
causes cells to make in culture.  Science has given this virus
the name "HIV."  That's the end of that story.   That's what HIV

    You'll notice that I haven't talked about antibodies or
electron microscopes.  They are not needed to tell if you have
captured a particular virus in culture.  However, we have used
such things to see what "HIV" looks like.  The HIV proteins can
be used to make antibodies that react with HIV proteins, but not
other proteins in cells.  These antibodies can be stuck to heavy
metal atoms so that they can be seen on electron micrographs.
If they are added to cell in cultures where HIV proteins are not
made, the antibodies don't stick to anything, and we see nothing
special about the cells.  However, in cultures where virus has
been added, the cells are budding small particles into the culture
media fluid, and HIV-protein antibodies are seen to stick heavily to
these structures.  It is thus eminently reasonable (and to decide
otherwise would be perverse) that these structures are where the
HIV-proteins are, and are thus the HIV virus.  It helps that they
look like virus particles.

    Now, remember, we have shown that have a virus
(self-propagating collection of proteins and genes) in HIV
cultures, which we call HIV.  It's always possible that what we
see on electron microscopy isn't the virus which we know we're
growing, and is some other virus--- but it's rather unlikely.
If it's some other virus, how come it has such an affinity for
antibodies which we carefully made to HIV-proteins, which are the
viral proteins we're interested in?  You tell me.

                                  Steve Harris, M.D.

Index Home About Blog