快猫短视频

FOR almost four years, research into HIV has been dominated by a single
theory about how the virus causes the catastrophic collapse of the immune
defences that leads to AIDS. But the consensus on this theory is now crumbling,
thanks in part to the work of a Dutch team led by immunologist Frank Miedema. If
the Dutch team is right, the consequences will be profound. People with HIV may
hope for new types of treatment. And some of the most cherished dogmas of a
multibillion-dollar research industry may be overturned.

The prevailing view about how HIV causes AIDS is that every day the virus
makes billions of copies of itself and, in doing so, kills billions of the key
defence cells that it infects, a class of T cell known as CD4 cells. These vital
cells orchestrate the body鈥檚 immune response. Every day the infected person鈥檚
immune system attempts to replace these cells. After years of waging this
immunological war, the body eventually fails to keep pace with the virus and the
numbers of CD4 cells become dangerously low, leaving the body unable to defend
itself against microorganisms and cancerous cells.

But Miedema and his colleagues at the University of Amsterdam see things
differently. They agree that the number of CD4 cells ultimately dwindles, but
not because the virus is killing them off. In their view, the virus impairs the
body鈥檚 ability to produce new CD4 cells, and鈥攃ritically鈥攊t traps
existing cells in lymph nodes and other tissues, preventing their movement in
and out of the bloodstream. As large numbers of CD4 cells become trapped in this
way, and the body fails to produce a sufficient number of new ones, the
dwindling population of circulating cells becomes increasingly restricted in its
range and ability to respond to different invading microbes.

Naturally, the champions of the prevailing theory dispute the Dutch ideas.
David Ho, chief architect and proponent of the accepted view, at the Aaron
Diamond AIDS Research Center in New York, has told colleagues that 鈥渢he whole
field would have to be turned upside down if they were right鈥�. But elsewhere,
the controversial Dutch theory is gaining ground. Indeed, it builds on ideas
that have been circulating since about 1990, among researchers such as Yvonne
Rosenberg at TherImmune, a company in Maryland, and John Sprent, at the
University of California, San Diego. Earlier this month, Miedema鈥檚 latest
findings were aired at a major international meeting in Glasgow on new therapies
for HIV鈥攁 sign that the ideas are attracting interest from those at the
sharp end of AIDS treatment.

The widely accepted view, that HIV is a mass murderer of cells, first took
hold in 1995, when Ho and his colleagues in New York, and another group led by
George Shaw of the University of Alabama at Birmingham, published two seminal
papers in the journal Nature (vol 373, p 117 and p 123). These papers
reported that there was a large and rapid turnover of CD4 cells in people with
HIV infection, and that therapy with a powerful cocktail of antiviral drugs
brought about huge and immediate increases in the numbers of these cells. The
fact that the cells bounced back so quickly was due, Ho and Shaw reasoned, to
the effects of the antiviral drugs. By stopping HIV from building new copies of
itself, the drugs stopped the virus from killing (or 鈥渓ysing鈥�) cells, while new
cells continued to be produced at a rapid rate. This compelling idea offered a
simple explanation for how HIV could wreak such havoc. Overnight, the theory
became dogma.

Then, in November 1996, Miedema proposed an alternative view. His work at
that time centred on telomeres. These are the small sections of DNA at each end
of a chromosome that are shortened with each cycle of cell division. Miedema and
his colleagues reasoned that if CD4 cells were being constantly destroyed,
then the unremitting cell division needed to supply the new cells would wear
away their telomeres.

Yet the length of the telomeres turned out to be stable. 鈥淭his means that
cells are not being turned over in massive numbers,鈥� Miedema said at the time
(快猫短视频, Science, 7 December 1996, p 20). 鈥淥ur data cannot be
interpreted any other way.鈥� He suggested that if the cells are disappearing but
not being destroyed, then HIV must be hitting their production instead.

Ho disagreed. He said that an enzyme called telomerase, which rebuilds
telomeres in cells that need to carry on dividing indefinitely, such as
reproductive cells, is overactive in people with HIV. The enzyme is active in
their immune cells, where normally it is absent. He argued that this
overactivity could explain why the telomeres do not shorten. But Miedema鈥檚 group
has tested T cells from people with HIV and has found no evidence of increased
telomerase activity. Ho retorts that their tests are not sufficiently sensitive,
and that special assays are needed.

Ho鈥檚 views find support from Tomas Lindahl, a telomere specialist at
Britain鈥檚 Imperial Cancer Research Fund. 鈥淚 don鈥檛 think the telomere
argument鈥�is very strong,鈥� he says. 鈥淭elomerase activity is notoriously
difficult to measure.鈥�

Vanishing cells

Indeed, other researchers now suggest that Miedema may have misinterpreted
his original results. They believe that he found the average length of telomeres
to be stable because he missed those cells that were disappearing most
rapidly鈥攖he very cells that would have the shortest telomeres if they were
turning over at the rate Ho suggests.

Whether the telomere research is significant or not, a growing number of
researchers now believe that HIV does prevent the production of new T cells.
Mike McCune at the University of California, San Francisco, suggests that the
site of this inhibition could be the thymus, the organ where CD4 cells develop
(see Diagram). But the Dutch group and others were increasingly convinced that
there was another possibility. If T cells were disappearing from the blood,
perhaps it was not just because new cells were failing to appear. It could also
be that existing cells were being hidden away in other tissues. Miedema and his
colleagues were puzzled by the flood of CD4 cells rushing into the blood that Ho
and others had observed when infected people start to take antiviral drugs. They
knew that the rise was rapid and then reached a plateau, and so they argued that
it could not be due to the production of new cells because this would lead to a
slow, more sustained increase. Instead, it must be due to the release of
existing CD4 cells trapped in lymph nodes and elsewhere.

Their own experiments supported their hunch. When they analysed T cells in
the blood of people with HIV as they started antiviral treatment, they found the
same steep rise of CD4 cells, reaching a plateau within three weeks. The
findings also appear to explain a phenomenon that has puzzled doctors, namely
that the more advanced a person鈥檚 HIV infection, the greater the initial rise in
their CD4 cell count when they start antiviral therapy. This, says Miedema, is
because more and more cells become trapped as infection persists. If Ho and Shaw
were right, the increase in CD4 cells should be modest in such people, because
the virus would have killed so many of their cells.

But the nature of the newly appeared cells gave the Dutch team further
support. They were virtually all so-called CD4 memory cells鈥攖hat is, cells
that had already come into contact with antigens from specific invading
microbes. What is more, so-called naive CD4 cells鈥攖hose that have not yet
met an antigen鈥攄id not immediately appear. These findings strengthen the
argument that antiviral drugs were not preventing HIV from killing cells, but
simply releasing into the blood mature CD4 cells that had been trapped
elsewhere.

An obvious response to the suggestion that CD4 cells are disappearing from
blood into lymph tissue might be: 鈥淲hy not count them?鈥� Unfortunately, this is
easier said than done. Removing lymph tissue is awkward and unpleasant鈥攁nd
may be unhelpful to patients whose immune systems are already disrupted. Equally
important, researchers would not know exactly how many CD4 cells a patient had
in the first place, and therefore would have no baseline figure with which to
compare their estimate. Finally, even in healthy individuals, the number of CD4
cells in the bloodstream is a tiny proportion鈥攂etween 1 and 2 per
cent鈥攐f the total. So even if the researchers measured their decline in
the bloodstream and estimated their numbers in the lymph nodes over a period of
time, the margin of error would probably be too wide for the counts to be
meaningful.

The Dutch group now has the backing of a growing number of immunologists.
Brigitte Autran at the Piti茅-Salpetri猫re Hospital in Paris has
found that, in people with HIV who take powerful drug cocktails, the immune
system appears to be able to take a break from the damaging effects of the virus
and boost its numbers of naive CD4 cells. This implies that the unchecked virus
does indeed prevent the production of new CD4 cells. And, in the latest move,
also reported at the Glasgow meeting, Miedema found that the immature
鈥減rogenitor鈥� cells that eventually mature into T cells are also disrupted. When
his team took progenitor cells from people with HIV and from uninfected people,
and put them into mice to mature in the thymus, they found that cells from
HIV-infected people matured more quickly鈥攕uggesting that the virus is
cranking up the immune system into excessive activity from the earliest
stage.

Taken together, says Mario Roederer of Stanford University, who studies
T-cell dynamics, these findings are 鈥渢he final nails in the coffin鈥� for the
theory put forward by Ho and Shaw. Roederer believes the virus completely
鈥渞earranges鈥� the immune system, rendering it ineffective and drastically
reducing the repertoire of CD4 cells available to fight off infections.

Beyond killing

Roederer points to a string of observations showing that HIV鈥檚 effects go
well beyond simply killing cells. First, it causes a decline in naive CD4 cells,
even though鈥攗nlike the activated CD4 cells that HIV uses for replication,
these cells are relatively resistant to being killed by the virus. So whereas
healthy people have a large range of subtly different subclasses of CD4 cells in
place to fight the vast array of infections they face, HIV-positive people have
a sharply reduced repertoire. Second, another class of T cell, the CD8 cell,
which cannot be infected, also declines in numbers. Third, says Roederer, it is
not only CD4 cells that bounce back after treatment starts鈥擟D8 cell counts
rise too.

Roederer is critical of Ho and his colleagues for focusing only on lysis and
failing to look at the other evidence at their fingertips. If they had analysed
what happens to CD8 cells as well as CD4 cells, says Roederer, their conclusions
would have been different. 鈥淚t would have blown their lysis theory out of the
飞补迟别谤.鈥�

Because it has long been known that HIV kills CD4 cells when it replicates
inside them, many researchers have been too ready to assume that this is the
reason they all die. But, in Roederer鈥檚 view, this is just 鈥渁 very ugly
coincidence that has led us astray from understanding鈥�this disease鈥�.

Ashley Haase, an immunologist at the University of Minnesota in Minneapolis,
agrees that Ho and his colleagues may have oversimplified the story. But he
defends their papers in Nature as pivotal. Their research alerted
scientists to the fact that HIV starts to cause disease immediately after
infection, rather than sitting latent until the onset of AIDS, as was once
thought.

Indeed, few researchers will go as far as Roederer in pronouncing Ho and
Shaw鈥檚 theory dead. But many鈥攊ncluding some who work closely with
Ho鈥攏ow agree that it cannot explain everything. Alan Perelson, a
theoretical biologist at Los Alamos National Laboratory in New Mexico, who
worked with Ho on his 1995 paper in Nature, is one of them. 鈥淭here is
clearly evidence that you have increased trapping in the tissue,鈥� he says.

Nevertheless, says Perelson, there is still evidence that CD4 cells are
turning over at an unusually high rate. He and others have found that CD4 cells
are three times as likel to divide in infected people as in uninfected people.
But these results don鈥檛 show why these cells are disappearing in the first
place, and therefore cannot substantiate Ho and Shaw鈥檚 assumption that HIV is
directly killing them.

However, Ho, Perelson and others have now produced more evidence. They
studied macaques infected with the simian immunodeficiency virus, a relative of
HIV. Using a marker to identify cells that have just divided, they found that
CD4 cells divided more rapidly in infected macaques than in uninfected animals.
And by measuring the rate at which the marker disappeared, they found that the
cells died off more quickly. Miedema and Ho are now racing to repeat the
experiments in people.

If the monkey results are borne out in humans, there will be convincing
evidence that HIV does indeed kill some CD4 cells, says Haase. But the question
is, how many? If CD4 cells decline simply because HIV is killing them, he says,
there should be many more infected cells than there are. Haase should know
because he has made a sustained effort to count the number of CD4 cells in the
body and how many are typically infected.

Tissue trap

By measuring the amount of HIV鈥檚 genetic material in various representative
tissue samples from infected people, and extrapolating from these samples to the
entire body, Haase estimates that at most 1 in 2500 cells, maybe fewer, is
infected with HIV. This is nowhere near enough for direct cell killing to
account for the depletion in their numbers that leads to AIDS.

Haase says that his own work now shows that large numbers of CD4 cells are
becoming trapped in lymph tissue, and he believes that HIV also disrupts the
production of new cells. In common with a number of other researchers, he also
believes that HIV may cause the loss of uninfected CD4 cells by triggering
abnormally high levels of cell suicide, or apoptosis鈥攁 separate process
that has been a subject of research throughout the 1990s (鈥淗ow does HIV cause
AIDS?鈥� 快猫短视频, 18 July 1992, p 31).

Taken together, these findings clearly suggest that HIV keeps the immune
system in a state of constant activation, and unbalances it in four ways: by
trapping mature cells, by stopping the production of new cells, by triggering
abnormally high rates of apoptosis and by killing a small but significant number
of cells directly. Their combined impact leaves the immune system depleted and
unable to cope with opportunistic infections.

How will any of this affect treatments? Would better knowledge of the ways in
which the virus disturbs the immune system enable researchers to rebuild it,
broadening the depleted repertoire of CD4 cells? Roederer, at least, thinks that
drugs that directly affect the immune system will be needed.

Others go further. Jay Levy at the University of California, San Francisco,
worries that prolonged treatment with cocktails of antiviral drugs might even
fool the immune system and 鈥減ut it to rest鈥�, by keeping levels of HIV so low in
the body that they fail to trigger any immune responses at all. This might make
individuals who stop taking the drugs even more vulnerable. He argues that
immune-restoring treatments should be given alongside antiviral drugs. Already,
some researchers are working on novel approaches such as developing genetically
engineered T cells to replace lost CD4 cells.

But in the short term, most researchers agree on one thing. The key problem
with HIV is that it triggers a disastrous loss of CD4 cells in the blood. How
the cells are lost may still be a subject for debate, but no one doubts that the
priority is to stop their loss in the first place. This will require better,
less toxic antiviral drugs. But given the new evidence that HIV also imbalances
the immune system and disrupts CD4 cell production, drugs that manipulate the
immune system are also needed. And, with resistance to antivirals crushing hopes
of a cure, the need for those drugs is more pressing than ever.

When a healthy person gets infected with a microorganism, say through a cut
in the foot, circulating T cells are a key part of the immune response to it. In
HIV infection, trapped cells are unable to respond in the same way