Immunotoxin Treatment for AIDS?

A San Francisco immunologist has proposed using immunotoxins, a technology already tested in humans as an
experimental cancer treatment, in the treatment of AIDS and ARC. Alan S. Levin M.D. of Positive Action Healthcare in San Francisco, who is presenting the idea this week at a meeting of the AIDS Medical Resource Center Physicians Association in Chicago, based the proposal on a model (theory) of the development of AIDS, a model he derived from recently published work of a number of researchers. (For background on Dr. Levin and Positive Action Healthcare, see AIDS Treatment News #47, January 1, 1988).

Even though the proposed AIDS treatment has not yet been synthesized--let alone have any clinical data supporting it--we are interested in the proposal for several reasons:

* The technology of immunotoxins has developed far enough that creating one for AIDS would be almost routine; according to Dr. Levin it could probably be ready for human testing within six months, if unforeseen problems do not intervene. We do not know of any other researchers developing an AIDS treatment based on immunotoxins; a computer search of the literature turned up only one reference, a published letter which suggested this approach in 1986.

* Dr. Levin is in an excellent position to develop the treatment. His wife, Vera S. Byers Ph.D., M.D., is an expert
in making monoclonal antibodies and immunotoxins. Dr. Levin has already started to assemble a scientific team for the
project.

* Aside from a treatment possibility, the theory itself summarizes what a number of leading researchers are now
learning about AIDS, and addresses several troubling questions such as how HIV can cause AIDS when it only infects a tiny fraction of T-helper cells (an issue raised by retrovirologist Dr. Peter Duesberg, who has argued that HIV could not cause AIDS). Levin's model also explains why no animals get AIDS, even if they can be infected with HIV; this question could be important for developing strategies for future research, for example by suggesting more emphasis on human immunology rather than animal models.

The Theory

We interviewed Dr. Levin for this article but explained the theory in our own words; any errors are our own
responsibility. This explanation is complex because the theory depends on several concepts from immunology; readers who find it confusing can skip to the "Proposed Treatment" section, below.

The central element of Dr. Levin's theory is that part of the protein of HIV imitates part of a human protein which has a
key role in the immune system. (Proteins consist of sequences of simpler chemicals known as amino acids; a virus may happen to have the same amino acid sequence within one of its proteins as is found in a human protein.) This unfortunate coincidence not only enables HIV to infect human T-helper and certain other cells; it also explains --according to the theory presented here--how later immune dysfunctions develop, through mechanisms
well known to immunologists. And fortunately this explanation suggests several approaches to treatment.

The theory arose from a question. AIDS depletes the T- helper cells, but HIV infects very few of them at any one time,
less than one in ten thousand. The body could easily replace these cells. Therefore if HIV is causing the damage (and there is much evidence that HIV is the sine qua non of AIDS, whether or not it is the sole cause or may also require cofactors) it must be depleting the cells by some means other than directly infecting and killing them.

Levin's model is not new research but rather a coherent summary of a number of findings of others. His theory proposes that HIV does its damage primarily by causing cells to produce an AIDS-virus protein, called gp 110 (or gp 120, another name used for the same substance), which circulates in the blood. Gp 110 (gp 120) includes the sequence which allows HIV to attach to the CD4 molecule--which is the receptor site on the T-helper (T-4) group of white blood cells.

This sequence causes the gp 110 (gp 120) protein to attach to T-helper cells, just as HIV does when it infects the cells. This protein cannot reproduce, as it is not the whole HIV virus, only part of it. But it causes damage in other ways.

The gp 110 (gp 120) molecule can attach itself to more than one cell, causing clumps of T-helper cells to fuse
together into giant clusters called syncytia. These cells stop working and soon die. In this way, a single cell which is
infected by HIV and secreting the gp 110 (gp 120) protein can cause the destruction of many healthy T-helper cells elsewhere in the body.

In addition, the immune system recognizes gp 110 (gp 120) as a foreign protein, and forms antibodies against it.
Presumably these antibodies do help to reduce its concentration.

But since gp 110 (gp 120) contains the sequence by which HIV matches the CD4 molecule (the receptor on the T-helper cell), an antibody against this AIDS-virus protein can also act against the human protein sequence which normally uses the CD4 receptor site. As a result, communication between T-helper cells and certain other cells, such as macrophages (see below), becomes disrupted.

(It might be helpful in understanding this process to visualize an antibody and its corresponding antigen as a key
which fits the lock. Like the key and the lock, the antibody and antigen fit because the shapes match exactly. When the
body is invaded by a foreign organism (the lock), it produces an antibody (the key) which precisely fits the protein from
that organism--and usually no other.

But unfortunately the AIDS-virus protein gp 110 (gp 120) - - the lock--happens to fit a key which is already in normal use
in the body (the CD4 molecule, which is the T-helper receptor site to which the AIDS virus attaches itself in order to enter
and infect the cell). Therefore when the immune system makes new antibodies to gp 110 (gp 120), these new keys not only fit this AIDS-virus protein, as intended. They also fit the "lock" (found on other cells such as macrophages) which was intended to use the "key" on the T-helper cell. Therefore these antibodies intended to attack the AIDS-virus protein also attack the normal cells such as macrophages, interfering with communication between the macrophages and T-helper cells and perhaps doing other damage as well.)

Normally macrophages, which are large immune-system cells which can circulate in the blood or remain in various organs of the body, engulf foreign organisms such as bacteria or fungi. Later they present the foreign proteins to T-helper cells, which instruct other cells to make antibodies specifically targeted against the invading organisms. The antibodies which the immune system produces against gp 110 (gp 120) interfere with this process.

After the macrophages present antigen to the T-helper cells, these cells in turn secrete chemicals called growth
factors (such as colony-stimulating factors, abbreviated CSF), which cause the maturation of new T-cells and other cells.
When the antibodies to gp 110 (gp 120) interfere with the macrophages, these chemicals are not secreted in the normal
amounts, resulting in further depletion as the new cells which should have matured do not do so. (One new treatment, called GM-CSF, is being tested for a deficiency of white blood cells often found in AIDS, especially after treatments with certain drugs such as AZT. GM-CSF is a colony stimulating factor given to correct for a shortage of this substance, a shortage which could be caused by the mechanism outlined above.)

Besides the damage described above, the AIDS-virus protein gp 110 (gp 120) causes harm indirectly in still another way. Normally after the immune system makes antibodies, it makes antibodies against the antibodies themselves. These, called antiidiotype antibodies, help to turn the immune response off.

But this AIDS-virus protein contains the sequence which matches the CD4 molecule on the T-helper cell. Therefore its
antiidiotype antibody can also act against the T-helper cell--further interfering with the normal functioning of the immune
system.

All of these forms of damage come about because the HIV virus--and its gp 110 (gp 120) protein--has a sequence which mimics a key human protein in the immune system.

And according to this theory, the reason animals do not develop AIDS from HIV is that they have different proteins,
which HIV does not mimic.

Proposed Treatment: An Immunotoxin

An immunotoxin consists of an antibody chemically attached to a toxin--a preparation intended to kill specific cells. The antibody targets the right kind of cells and binds to them, but the antibody itself does not kill the cell. The toxin kills
the cell; the antibody is custom made to bring the toxin only to a specific group of cells, a group which must be
immunologically distinct in some way.

For AIDS, the proposed treatment will use an antibody which seeks out the gp 110 (gp 120) protein but not any normal
human protein. This antibody will carry a toxin to the HIV- infected cells which are producing gp 110 (gp 120); these cells
can be recognized because they have this protein on their surface. The immunotoxin will kill these infected cells,
greatly reducing the amount of gp 110 (gp 120) being produced. Dr. Levin believes that it should be possible to reduce the level of this AIDS-virus protein by 100 times through the use of an immunotoxin.

This treatment will not be a complete cure; it would not eliminate the virus entirely, as some cells infected with HIV
show no outside evidence of infection. And of course there may be unforseen difficulties in creating a practical drug. But
the approach, already in use for cancer, certainly deserves serious attention for AIDS.

We do not know of anyone developing this approach until now. If you have heard of any work on immunotoxins for AIDS, please call John S. James, (415) 255-0588, or Dr. Alan Levin at Positive Action Healthcare, (415) 788-7545.

Note: for a copy of Dr. Levin's presentation on immunotoxins send a #10 self-addressed stamped envelope (2
ounces postage) to: Alan S. Levin M.D., Positive Action Healthcare, 450 Sutter St., Suite 1138, San Francisco, CA
94108.

**************************************************