Archive for the ‘A: Pathogenesis of HIV disease’ Category

AIDS pathogenesis: HIV disease and Positive feedback: An additional comment.

August 31, 2010 Leave a comment


This blog more or less duplicates that at the site, explained in the “about” page above.

HIV Disease and Positive Feedback.  An additional comment.

AUGUST 31ST 2010

A previous post focussed on the positive feedback interaction between HIV replication and immune activation.   HIV replication and immune activation reciprocally enhance each other.

While HIV infection is an essential cause of the immune activation that’s characteristic of HIV disease, there are other factors that also contribute to it.   In that post as well as in the blog I write on the POZ magazine website, I described some of these additional factors that can add to immune activation.   As noted, viruses of the herpesvirus family, cytomegalovirus (CMV) in particular are the most important of these worldwide, while in parts of Africa certain endemic infections may be of great significance in contributing to immune activation.

Since sustained immune activation, involving both innate and adaptive immunity is at the heart of the pathogenesis of HIV disease an understanding of how it is perpetuated is critical.

Evidence for activation of innate immunity was noted in 1981, the year that AIDS was first reported, in the detection of large amounts of alpha interferon in the circulation of patients.  We even knew then that interferon alpha and gamma could  induce an enzyme, indole 2,3-dioxygenase  (IDO),  (IDO was known to be responsible for the inhibition of toxoplasma gondii by depletion of  tryptophan  in cells treated with gamma interferon) but we did not know then that this enzyme could contribute to the loss of T lymphocytes.   Another observation of historical interest is that even before AIDS was first reported in 1981, interferon was known to preferentially inhibit CD4 lymphocyte proliferation in mixed lymphocyte culture.

Since immune activation and its effects, including  inflammation, are harmful if sustained,   there are mechanisms that can  dampen it.

But in HIV disease, immune activation persists with continued deleterious consequences.

The reason I’m revisiting this now is that there is a question that continues to be bothersome.

HIV disease is not the only infection associated with long standing immune activation.

Several endemic infections in Africa are also associated with sustained immune activation, certainly not all – some even have a dampening effect on immune responses. TB is another example of an infection associated with chronic immune activation.   In none of these conditions is there such a profound loss of CD4 lymphocytes as in HIV disease.  While individuals with active pulmonary  TB have been reported to have lower CD4 counts than healthy individuals, the numbers were well above 500.

Is the difference between sustained immune activation associated with HIV and that associated with other chronic infections in HIV negative individuals a matter of degree – is it a quantitative difference?

Could the  mechanisms that dampen and check  immune activation be impaired in HIV disease?   These mechanisms include the secretion of cytokines that have anti-inflammatory properties, such as IL-10, IL-13, and  TGF-beta, among others.  Specialized immune system cells can also dampen immune activation.  Tregs, a subset of T lymphocytes, have such a dampening effect.   Although there are conflicting reports on the relationship of Tregs to HIV disease, it is known that HIV targets some of  these particular T lymphocytes.

This graphic comes from my earlier post on positive feedback characteristics of HIV disease.

In this diagram HIV pathogenesis is represented by a circular process moving in a clockwise direction.  It is started by HIV infection and can be propelled by a positive feedback association  between HIV replication with immune activation.      Immune activation is reinforced by CMV, and in certain settings, by some endemic infections.  This is represented by the + sign in the diagram.      Immune activation is retarded by those influences that dampen the immune response, including anti-inflammatory cytokines and Tregs, represented by the – sign in the diagram.

Here is a revised version of this diagram:

HIV disease progression is represented as moving clockwise in a circle, reinforced by sources of immune activation other than HIV and retarded by Tregs and other mechanisms that dampen immune responses.  Tregs  act as brakes, but HIV can directly make the brakes less effective.

Could critical differences between HIV disease and other infectious causes of long standing immune activation where CD4 numbers are relatively preserved, be  the preferential targeting of Tregs by HIV and a different pattern of cytokine secretion?

I wonder if this revised representation of HIV disease lends itself to a more formal modelling process.

In this particular model a disease process is represented by a circular motion in a clockwise direction, with forces that both propel and retard it.  Some predictions can be made.

The degree of immune activation at the time of HIV seroconversion would favour more rapid HIV disease progression.  The set point – the level from which CD4 lymphocytes decline following an acute HIV infection, would be lower, and the subsequent  rate of CD4 decline higher when HIV infection occurs in a person where there already is a higher degree of immune activation, compared to an individual where this is not the case.  There already is  some evidence in support of this possibility.

It’s well established that HIV disease progresses more rapidly with increasing age.  Could an explanation for this be that immune activation increases with age – indeed, it’s been suggested that immune activation  contributes to the aging process.

HIV disease progresses more rapidly in individuals with active TB.  CMV viremia was noted to carry an adverse prognostic significance in HIV disease very early in the epidemic.  There are but two  examples, but there are many more of  of a more rapid course of HIV disease in the setting of other  infections caused by bacteria, protozoa, viruses and helminthes.  Some are referred to in a previous post.

Are Treg numbers at seroconversion and for a period immediately afterwards  related to subsequent disease progression?

Could treatment with anti CMV agents during acute HIV infection retard subsequent disease progression?

There already  is some evidence that treatment of HIV during acute infection might slow the subsequent course of HIV disease.

The utility of any model of a disease process lies in its ability to provide a common explanation for disparate observations as well as to make predictions that can be tested by an analysis of available data or by experimentation.

Viewing HIV disease as a process with a positive feedback interaction between HIV replication and immune activation with forces that both enhance and retard this interconnection,  provides a useful descriptive framework as well as testable predictions.

AIDS Pathogenesis: HIV disease has characteristics of positive feedback systems

April 2, 2010 Leave a comment

2nd April,  2010

There is a similar and slightly extended version of this post on the blog I have on the POZ website. It’s in two parts:

Part 1

Part 2

HIV infection and many other infections caused by a wide variety of microorganisms have a mutually enhancing relationship that is characteristic of positive feedback systems.

Although the reciprocal enhancing effects of HIV and other infections have been frequently described since the late 1980s, it is useful to explicitly recognize these as positive feedback systems as this highlights the implications they have for treatment of individuals and for control of the epidemic.  Explicitly recognizing the positive feedback characteristic of HIV disease also provides a way of looking at pathogenesis that can suggest further studies, both clinical and laboratory, that might advance our understanding of mechanisms of disease acquisition.

This is an illustration of positive feedback.    A stimulates B which in turn stimulates A. In this way the effects of A and B are increased.

The infections associated with the immunological disorders of HIV disease are generally, but not solely, caused by microorganisms that replicate within cells.     Many of the organisms that cause these infections survive in healthy people without causing disease, prevented from doing so by a competent immune system.   When the immune system fails these infectious agents start to divide.   They may then cause disease.  An additional effect of some of these active infections is to accelerate the replication of HIV.  Several mechanisms are responsible for this effect, which can then result in further immunological deterioration.

In addition, co-infection with many of the pathogens that also affect individuals with intact immune systems can also promote HIV replication.

Not all co- infections result in a more rapid progression of HIV disease.  Many have no effect and a few have even been reported to cause a temporary improvement of HIV disease.    This may be the case with measles, scrub typhus and a form of transfusion associated hepatitis.   But more often, when an effect of a co-infection has been noted, it has been to promote HIV disease progression.

Different co-infections can therefore affect the course of HIV disease in different ways.  Some may have no impact on the course of HIV disease; a few may possibly cause a temporary amelioration.   Those that are able to accelerate it are highly prevalent in HIV infected individuals.

Worldwide, viruses of the herpes family are probably the most important of the co-infections that interact with HIV in a mutually enhancing fashion. .    Virtually all adults are infected with some of these viruses that usually exist in a latent or dormant state.  They are readily activated in the setting of HIV infection and then promote further HIV replication by a number of different mechanisms.

In developing nations a range of different endemic infections, depending on geography, may be just as important; many can also accelerate HIV disease progression.  Conversely, HIV infection can promote progression of some  endemic infections.

Several different mechanisms have been uncovered that can explain the effects of co-infections on promoting HIV replication.    With such a wide range of infections, the precise ways in which each do this will vary in detail.

However there is one characteristic possessed by all HIV potentiating infections.  This is their ability to add to the immune activation that is a feature of progressive HIV disease.

By now I think it is generally accepted that chronic immune activation not only results from HIV infection but is a major contributor to the pathogenesis of HIV disease.   A state of sustained high level immune activation is the basis of the chronic inflammation and immunologic deterioration characteristic of progressive   HIV disease.

But what exactly is immune activation?

Immune activation refers to those changes that take place in the immune system when exposed to an infectious agent that allow it to eliminate or control the infection.  Essentially, the immune system is activated from a resting state to fight an infection.   Generally this process will last for days until the infection is overcome, and usually but not always, is followed by a lifelong immunity to the infectious agent.

However in progressive HIV disease the immune system continues to be activated at a high level and it is this sustained immune activation that eventually results in disease.   An activated state of the immune system is characterized by differentiation of precursor immune system cells.  Differentiation is the process by which these cells develop specialized functions.   Examples of cells that have acquired specialized functions are those that produce specific antibodies, or those with the ability to kill other cells infected with specific microorganisms.   Proliferation of immune system cells is an important characteristic of an activated state.  This is usually a short-term response subsiding with control of the infection that stimulated it.  But in progressive HIV disease, proliferation is sustained, probably with episodic cycles of further accelerations, and this continued proliferation contributes to the loss of immune system cells.

These cellular changes, differentiation and proliferation, are associated with the secretion of a variety of cytokines.  Cytokines are molecules that can change the behaviour of cells by binding to specific receptors on their surfaces, for example, causing them to divide.  Once released, cytokines not only attach to receptors on other cells but can also come back and attach to the receptors on the cell that produced it.

The cytokines that are released   have widespread effects.  Importantly, they include those that are associated with inflammatory changes, – the pro-inflammatory cytokines.    With respect to positive feedback, pro-inflammatory cytokines including IL-6 and TNF alpha are able to accelerate HIV replication.

A part of the immune system, the innate immune system, responds immediately to infection by recognizing molecular patterns common to different organisms.  The more familiar adaptive immune system responds to specific characteristics unique to each organism.

The innate immune system is also activated in untreated HIV infection.   Interestingly effects of activation of innate immunity were recognized very early in the epidemic, even before HIV was discovered, and so are among the earliest recognized AIDS related immunological abnormalities.  Activated innate immunity is responsible for the large amounts of alpha interferon in the circulation of people with untreated HIV/AIDS, first noted in 1981, the year this disease first came to our attention[i].   At that time the origin of this endogenous interferon was not known.   For a period, elevated levels of beta 2- microglobulin were regarded as an adverse prognostic marker.  This molecule can be regarded as a surrogate marker for interferon.   The association of interferon with abnormalities characteristic of this disease – including low CD4 numbers was also reported in the first 2-3 years of the epidemic[ii].   Over twenty years later mechanisms have been discovered that can explain the participation of interferon in the disease process[iii].

Interferon appearing in the circulation in untreated HIV disease may even be the first marker of immune activation noted, although not recognized as such when first observed

The changes that occur on activation of the immune system are associated with many other markers that can be measured.    Different molecules appear on the surface of activated cells.  These can be detected and measured, as can the cytokines associated with immune activation.

These measurements can tell us the extent of immune activation.   Importantly, the degree of immune activation parallels the rate of HIV disease progression.

Although it is now accepted that the consequences of continued activation and proliferation of immune system cells contribute to the loss of CD4 cells and the development of disease, the precise way it does so is not yet known, although there  are a number of different mechanisms  that could account for it.  The   associated inflammation also has adverse effects beyond the immune system.   For more detailed information on these mechanisms there are references to two reviews at the end of this article[iv].

Sustained immune activation is therefore at the heart of HIV/AIDS pathogenesis.   It is the sustained nature of the activated state that is critical.  Short lived states of immune activation are of course beneficial allowing us to recover from infections.  But in progressive HIV disease the process continues at variable rates.   Understanding what causes continued immune activation is central to an understanding of the pathogenesis of HIV disease.

What causes Immune activation?

While infection with HIV may start the process, other causes of immune activation are almost certainly also necessary to keep it going.

The following all contribute:

1:            The immune response to HIV itself.   This includes both innate and adaptive immune responses.  As noted above, adaptive responses are the familiar specific antibody and cell mediated responses that provide generally lifelong immunity to specific infectious agents.  Innate responses depend on recognition of molecular patterns common to several organisms.

Some suggest that HIV contributes directly to immune activation through binding of some of its proteins to immune system cells.

2:            Microbial products that can penetrate into the intestinal wall as a result of damage caused by HIV.  These microbial products then activate immune system cells.

3:            Other infections.

Some like active herpesvirus infections or the more traditional opportunistic infections can be seen as indirect effects of HIV infection.

Others are infections that can cause disease in people with intact immune systems like the endemic infections in developing nations. Some of these can be more severe in the setting of HIV infection.

Infections that can accelerate HIV replication include those caused by bacteria, viruses, protozoa and helminths.

Those that promote HIV disease progression can  usefully be  described in three categories.

A:            Herpes virus infections.   These are probably the most important worldwide.  Virtually 100% of adults are infected with some of them.     They represent infections that are more often latent, but are  readily activated  in HIV infected individuals.

B:            Endemic infections caused by a variety of different microorganisms than promote HIV disease progression and HIV replication.   These are important in developing nations.

C:            Other infections.  These include the opportunistic infections, as well as those that can affect people with intact immune systems.  TB may be the most important.  HIV infected individuals are much more susceptible to active TB infections than those who are HIV uninfected.  HIV transcriptional activity and viral loads have been noted to be higher in people with active TB.

Here is a little more detail about these three classes of infection:

A:  Herpesviruses.

There are eight members of the herpesvirus family that can infect humans.   Herpes simplex virus types 1 and 2 (HSV-1, HSV-2) are perhaps the most familiar.  Cytomegalovirus (CMV) and the Epstein-Barr virus (EBV) infect close to 100% of adults.   Varicella-Zoster virus (VZV) causes chicken pox on initial infection and shingles when reactivated. Of the three remaining human herpes viruses HHV-6, HHV-7, and HHV-8, the last is associated with Kaposi’s sarcoma.

With all of the herpes viruses, once infected, individuals carry them for the rest of their lives, usually in a dormant or inactive state.  All can be periodically reactivated with or without symptoms.

Humans and herpes viruses have co-existed for evolutionary periods and are well adapted to each other.   The immune system generally maintains these viruses in a latent sate so that they cause no harm.  Reactivation does occur periodically but is generally limited.  Virtually 100% of adults will carry some viruses of the herpesvirus family, usually in a dormant or latent state.

The impaired immunity characteristic of HIV disease however results in reactivation of herpes virus infections. In progressive HIV disease these viruses become active and through a variety of mechanisms, including their contribution to immune activation, promote the replication of HIV.    Cytomegalovirus (CMV) may be the most important of the herpesviruses that promote HIV disease progression.  It can be part of a positive feedback system in its interactions with HIV.

HIV → latent herpes infections  →active herpes infections → HIV

It is not only through their contributions to immune activation that herpes viruses promote HIV replication.   In addition to the pro-inflammatory cytokines that have this effect, herpes virus gene products can directly activate HIV if a cell is infected with both viruses.  This process, called transactivation works both ways; HIV can also activate herpes viruses.

In addition herpes infections cause a receptor (Fc) to appear on cell surfaces that allows HIV to enter it.  In this way cells that do not possess CD4 molecules can become infected with HIV.   Active CMV infections can also exert a mildly immunosuppressive effect.

Herpesviruses, particularly CMV are singled out because they probably play a significant role in the pathogenesis of HIV disease.  CMV infections are so common that it is hard to find HIV infected individuals who are free from it so that they can be compared to those who are not.   But as early as 1991 this was done with HIV infected haemophiliac patients, when it was noted that those also infected with CMV had a much more rapid progression of their HIV disease[v].

That CMV may play a role was suggested by many very early in the epidemic.  A multifactorial model for the development of this disease published in 1983 before HIV was discovered suggested a major role for CMV and EBV[vi].    The considerable evidence for a role for herpesviruses, particularly for CMV, did not disappear with the discovery of HIV.   The interactions of CMV and other herpes viruses with HIV that have been discovered may now explain their role.

Large studies on the effects of acyclovir on the course of HIV infection have provided compelling evidence that active infection with these viruses can be regarded as part of the disease process for most HIV infected individuals.    Investigators focussed on HSV-2 undoubtedly because it is the most common cause of genital ulcers.   The dose of acyclovir used would also have suppressed  HSV-1, which is even more prevalent than HSV-2 and may be more sensitive to acyclovir.  HIV viral loads and the rate of HIV disease progression were reduced in individuals receiving acyclovir compared to those receiving placebo.  Although genital ulcer recurrences were suppressed by acyclovir, the drug had no effect on the transmission of HIV.

The effects of acyclovir on HIV probably resulted from suppression of active herpes infection.  This is entirely consistent with a model that places HIV and herpesviruses in a positive feedback relationship.

EBV and CMV are much more resistant to acyclovir than HSV-1 and 2.   But it cannot be excluded that this drug did not have some effect in also diminishing reactivations of CMV and EBV.   If samples from the trial have been stored appropriately, this can be looked at.  EBV reactivation patterns are easily recognized, CMV virus isolation is possible and even detection and quantification of activated T lymphocytes would tell us something.

B:  Endemic infections:

These are singled out because of their high prevalence in some parts of the developing world.

These infections affect significant proportions of the population, they tend to be chronic and persist in the absence of treatment.    The specific infections will depend on geography and many are transmitted by insects.   Many of these can also accelerate   HIV disease progression, and some also progress more rapidly in the setting of HIV infection[vii].

C:   Other infections:

On an individual level, some episodic infections can promote HIV replication.   An acute febrile illness may increase HIV viral loads, but this is a transient effect lasting for the duration of the infection.

Most of the serious opportunistic infections occur late in the course of HIV disease, and may promote even further disease progression.

TB deserves special consideration because of its high prevalence in HIV infection.  Susceptibility to TB is increased even at higher CD4 levels. Active TB can then promote further HIV replication thus becoming a partner with HIV in a positive feedback interaction[viii].

A role for immune activation in a positive feedback system:

One way to look at the process of disease acquisition in HIV infection assigns a central role to immune activation.

Immune activation not only results from HIV infection, it can also promote further replication of HIV.

HIV replicates more efficiently in activated immune system cells.  Secondly, the pro-inflammatory cytokines that are associated with an activated immune system   can directly stimulate HIV replication.   Progressive HIV disease and immune activation are therefore components of a positive feedback system in this way.

HIV disease → Immune activation → HIV disease → Immune activation

The process starts with HIV infection, and is promoted by other infections , some of which are activated by HIV infection.

Whatever is driving immune activation is driving HIV disease.

The following diagram illustrates this.

Looking at the course of HIV infection in this way has a number of implications.


In the above diagram the course of HIV disease is represented by a self perpetuating cycle proceeding in a clockwise direction.   In addition to the elements that have positive effects in driving the process, there will also be those that retard the cycle.  This is illustrated in the next diagram which focuses for simplicity on the immunological control of HIV infection and of those infections that add to immune activation.    Of course there are other mitigating factors, for example, genetic factors conferring varying degrees of resistance resulting from receptor polymorphism.

In the diagram, the connection of HIV with CMV and other herpes viruses is probably constant and indicated by a red arrow.   The connection of HIV with endemic and associated infections is indicated by a blue dotted line, because HIV infection does not increase susceptibility to all of them, nor does it accelerate the progression of all.

The positive feedback cycle starts with HIV infection.  At least some of the determinants of the rate of disease progression may be found in the conditions that exist at the time of initial infection that promote or retard the cycle.

There is evidence that the degree of immune activation at the time of seroconversion predicts future disease progression.[ix] [x] It may also be an important determinant of what is called the set point.  This is the point following initial infection with HIV, from which CD4 numbers decline.

The degree of immune activation at seroconversion thus influences the starting CD4 level; the rate of subsequent decline is influenced by the degree of immune activation  in a system where once started, conditions can exist where  immune activation increases with falling CD4 numbers, in a self perpetuating and accelerating fashion.   Whatever the outcome, it will be the balance of positive and negative influences.

In the earliest years there were reports of EBV reactivation preceding HIV seroconversion[xi].

I have not seen any follow up of this interesting report.   It at least suggests that there might even be   situations in which active herpes infections could sometimes promote seroconversion.  They certainly produce signals that can activate HIV transcription from proviral DNA.

Treatment and prevention.

The role of immune activation in driving HIV disease is generally accepted now.   There are sources of immune activation other than HIV and some of these can be controlled.

Attempts to identify and control additional sources of immune activation may be critical in the fight against HIV/AIDS.

Perhaps the most significant benefit in this respect concerns the developing world, where there are so many additional sources of immune activation. Even ascariasis, infestation with the common intestinal round worm is associated with significant immune activation.   Worldwide prevalence is estimated to be about one billion, with 173 million in sub-Saharan Africa.

Many highly prevalent endemic infections can promote HIV replication.  Controlling these are perfectly appropriate targets in the fight against HIV/AIDS, and of course this would independently improve the lives of millions of individuals.

Measures to control endemic infections include traditional public health interventions, such as the provision of sanitation and clean water and the control of insect vectors. Effective drugs are sometimes inexpensive.  Peter Hotez has written an article entitled “Africa’s 32 cent solution to AIDS”.[xii] This refers to the price of Praziquantel , effective in treating  schistosomiasis as a single dose.

The lives of impoverished populations are ravaged and shortened by these infections. Many of these infections also interact with HIV to compound the devastation they cause.  Poverty, multiple endemic infections and HIV are intimately intertwined and in many instances reciprocally affect each other.

Recent and ongoing studies will probably lead to the routine use of drugs that are effective against herpes virus infections.       Trials of valacyclovir to reduce HIV viral loads are in progress. Given the ubiquitous nature of herpes infections, the use of acyclovir as adjunctive therapy might be warranted even in the absence of recurrent herpetic ulcers.  Valacyclovir unfortunately is not yet available as a generic medication.

Unfortunately EBV and CMV are much more resistant to these drugs.   The development of agents less toxic than valgancyclovir is important.   Valgancyclovir has already been shown to reduce immune activation in HIV infected individuals as measured by a reduction in activated CD8 lymphocytes.

In summary it is useful to explicitly recognize the positive feedback interactions between HIV and other infections that can promote its replication, some of which are in turn promoted by HIV.    Control of the AIDS epidemic in Africa must include measures to prevent and treat multiple endemic infections that affect hundreds of millions of individuals.

[i] This is of particular interest to me as I was involved in the discovery of large amounts of interferon in the circulation of people with HIV/AIDS in 1981, the year the disease was a first described.

[ii] Fig 1 shows CD4 counts in relation to serum interferon  . Presented 1986 at the 2nd international aids conference in Paris.


[iv] Immune activation and inflammation in HIV-1 infection:  causes and consequences.

V.Appay and D. Sauce

J.Pathol. 2008; 214: 231-241

(This is an important  review)

HIV immunopathogenesis and strategies for intervention.

M. Cadogan and A Dalgleish

Lancet Infectious diseases. 2008: 8: 675-84



[vii] Endemic infections in Africa have everything to do with HIV/AIDS:






Endemic Infections in Africa have everything to do with HIV/AIDS and are a long neglected therapeutic target.

June 6, 2009 1 comment

An article with the striking title “Africa’s 32 Cents Solution for HIV/AIDS” was just published in PLoS Neglected Tropical Diseases.  It can be seen here:

This dramatic title refers to the cost of treatment of schistosomiasis with praziquantal.

Schistosomiasis is an infection caused by parasitic worms, or helminths., of the genus  Schistosoma.    Most of the 200 million cases of schistosomiasis in the world occur in Africa.

The species, Schistosoma haematobium is estimated to infect about 112 million people in sub Saharan Africa.  So its high prevalence puts it in the same class as that of TB, malaria and HIV.  It is responsible for a huge burden of morbidity particularly in children and young adults.

S. haematobium  has a complicated life cycle, some of which takes place in snails.  People are infected by organisms released by snails living in fresh water. These organisms can penetrate the skin of any body part that is immersed in snail infested water.  S. haematobium affects the urinary tract.  The disease it causes is commonly called bilharzia.

I was very conscious of its danger as a child growing up in Zimbabwe, with signs at several small lakes around Bulawayo warning one not to swim in them because of the danger of bilharzia.

Peter Hotez and colleagues article is a welcome addition to the already substantial literature that strongly suggests that many endemic infections, not only with helminths, but also with bacteria, protozoa and viruses can increase the transmission of HIV and most probably  have a detrimental effect on the course of HIV infection.

This paper concentrates on the local effects of S.haematobium on the female genital tract , where lesions caused by  schistosome egg deposition result in mucosal patches, that can bleed during sexual intercourse. The authors state “Presumably, the schistosome egg granulomas produce genital lesions and mucosal barrier breakdown to facilitate HIV viral entry” and go on to compare this to the process by which herpes simplex ulcers increase susceptibility to HIV.

This does seem obvious – there is a mucosal break, so HIV has a way in.

In fact in the case of herpes simplex, this seemingly obvious connection is probably not correct.   The large Partners in Prevention study, recently completed, found that acyclovir, a drug effective in treating herpes does not reduce the risk of HIV transmission.  The drug however was associated with a reduction in the number of recurrences of herpetic ulcerations, and significantly slowed HIV disease progression.  I have written about this in another post.

As with herpes simplex, it is possible that systemic effects of schistosomiasis, may be much more significant, or at least as significant, as local effects in enhancing the transmission of HIV.    Of course, both local and systemic effects may play a role in enhancing HIV transmission.  The systemic effects include an impairment of virus specific immune responses; immune activation may also increase susceptibility to HIV and promote its replication.

The influence of associated infections on the infectivity of HIV extends far beyond that of schistosomiasis.  Peter Hotez  (the lead author of the above article) has done a great service by bringing attention to a number of devastating neglected tropical diseases.  This important article can be seen in the Lancet of May 2nd, 2009, (Lancet 2009 373;1570-1575).

The title of the article is:

“Rescuing the bottom billion through control of neglected tropical diseases”

By Peter J Hotez, Alan Fenwick, Lorenzo Savioli and David Molyneux

I have copied this table from the above article:


These are incredibly huge numbers.

Many of these infections occur in children and young adults and not only  have an impact on life expectancy, but significantly are the cause of chronic debility particularly in young people.

Some also have an activating effect on HIV replication by several mechanisms, some of which  have been understood for well over ten years.  The resulting acceleration of HIV infection,  by  increasing  HIV viral loads,  as well as by other mechanisms increases the transmission of this virus.

The health of hundreds of millions of individuals could be improved by efforts to prevent and treat these infections.  These infections are also appropriate therapeutic targets in the fight against HIV/AIDS.

Despite a great deal of evidence for the interaction of multiple bacterial, viral, protozoal and helminthic infections and HIV,  this association has been inexplicably neglected in providing  additional approaches to controlling the epidemic..

I had what might be described as a  misfortune to have been a member of President Mbeki’s panel on AIDS, an almost surreal experience I should write about.  The following is an excerpt from something I wrote for this panel almost 10 years ago:

“The crucial difference in Africa, as opposed to the US, is the high prevalence of associated infections. These include STDs, TB, malaria and other protozoal infections, helminthic and bacterial  infections. Such infections would supply sustained signals, such as IL-1  IL-6 and TNF, known to activate HIV.  Some can also upregulate the expression of chemokine co receptors required for HIV entry.  Some of these infections are  somewhat immunosuppressive themselves, an effect contributed to by the secretion of IL-10.37 Sexual transmission of HIV is also known to be facilitated by a high viral burden.38 This would also be the consequence of the HIV activating effect of frequent associated infections in Africa.”

This was almost 10 years ago, and since then literature has continued to accumulate documenting the detrimental interactions between HIV and multiple infectious agents.

About two years ago I made a presentation at the Prevention Research Center at Berkeley, trying to understand why endemic diseases had been so neglected in our attempts to control AIDS, particularly in Africa.  I thought that part of the problem was poor interdisciplinary communication and understanding.   Specifically, there might be difficulties in   communications between public health experts and microbiologists.   Possible public health implications of the findings of microbiologists might not be perceived without additional explanation.  I illustrated this with a specific article.

I used an excellent article to illustrate this problem.

The article is called “Contribution of Immune Activation to the Pathogenesis and transmission of HIV type 1 infection” and the authors are Stephen Lawn, Salvatore Butera and Thomas Folks.   (Clinical Microbiology Reviews. Oct 2001 14; 753-777)

This is part of what I said in California  in trying to illustrate the difficulty in communication:

“Of great interest – because of its implications for disease control was the discovery that other infections, viral, bacterial, protozoal and helminthic, could influence the course of HIV disease.  Generally the effect was to enhance HIV replication, but a few seemed to ameliorate – at least temporarily, the course of infection.  Scrub typhus, measles and perhaps a form of viral hepatitis, may have a  transient beneficial effect on HIV disease, but these are exceptional cases. Most co-infections have the opposite effect.

We now come to an example of observations made by microbiologists and work done at a molecular level with enormous implications for the control of AIDS in Africa.   This example is a review (cited above)  explaining in great technical detail how the replication of HIV can be enormously enhanced by concurrent endemic infections, and how this not only accelerates the progression of HIV disease, but also facilitates its transmission. The authors show in molecular detail how many viral, bacterial, protozoan and helminthic infections can affect HIV replication.  Included among these are common intestinal worms and water borne bacterial infections, causing severe diarrhea particularly in infants.  The discussion is largely concerned with the possible beneficial effect of drugs that might counteract this enhancement of HIV replication. There is one short sentence on public health interventions that might eliminate this problem altogether. It is of particular interest because of its brevity in a rather long article.   There is also a curious statement that where antiretroviral drugs are unavailable, measures to control endemic infections may be a useful approach.  This comment is reproduced below, and somehow ignores the significance of the implication that control of these endemic infections requires no other justification than as a measure to control AIDS.

This paper, because of its immunological and molecular detail is not too likely to find its way to an epidemiologist or public health expert,  but for one trained in these technicalities, I would suppose the public health implications would be immediately evident.

This particular paper also is a great illustration of the compartmentalization of information, and the difficulties of interdisciplinary communication.

Below is an illustration from the body of the article: there is much more just like this.  A person with no training in molecular biology or virology would not be likely to spend any time with this illustration.


However if one turned a few pages the following diagram may just be of some interest. But again this is unlikely.

The part that would be of interest to a public health professional , if noted,  is contained in the large arrow at the bottom right of the illustration.  In this rather complex diagram it would be quite easy for the public health expert to be sufficiently distracted so that the bottom right hand corner would be easily missed.


There is a long discussion, quite technical in nature, but at least the authors find space for the following brief comment.

“Prevention and Treatment of Coinfections

The widespread use of HAART in the treatment of HIV-

infected persons in westernized countries has resulted in a

phenomenal decrease in the incidence of opportunistic infec-

tions and has greatly increased survival. For these individuals,

the antiretroviral drugs are the major determinant of prognosis

and the potential cofactor effect of opportunistic infections is

now a more minor consideration. However, the vast majority

(>95%) of the world’s HIV-infected people do not currently

have access to antiretroviral drugs. Most of these people live in

developing countries, where the quality and access to health

care is often limited and where there is a high incidence of

endemic infectious diseases such as malaria, TB, and infections

by helminths and waterborne pathogens which may adversely

affect HIV-1 disease progression. Prevention or early treat-

ment of these diseases may therefore represent an important

strategy in addressing the HIV-1 epidemic in developing coun-

tries”. –

In the above quotation, the authors are overoptimistic in their assertion that the cofactor effect of opportunistic infections is now a more minor consideration in developed countries.  Valacyclovir, a drug that inhibits the replication of  many members of the herpes virus group, but has no direct effect on HIV was reported to reduce HIV viral loads in the absence of antiretroviral therapy. In the developed world, active herpes virus infections are common in the setting of HIV infection, although most will be asymptomatic. For example, Cytomegalovirus, Epstein Barr Virus and Human herpes virus type 6 are not infrequently found to be active in HIV infected individuals. Valacyclovir will have an effect on these viruses, and may well find a place in the treatment of HIV infection in developed countries.  Indeed it may not be uncommon for experienced physicians here (in the US) to prescribe related anti herpes medications to their HIV infected patients. I certainly do.

There is another aspect, a little more difficult to establish and perhaps altogether conjectural.  This is that we are presented with the question of why we need AIDS to justify interventions that have long been established to themselves improve the health of populations.  These include the provision of sanitation and clean water, the control of malaria and TB, and something as simple as getting rid of worms.  In the public’s assessment of the health needs of developing countries the information that is used is largely to be found in popular media, newspapers, magazines and TV.  Those who report in turn receive information from professional sources, and maybe it is here that the interdisciplinary barriers to communication I have been talking about have their effect. Thus the AIDS epidemic is perceived to be the greatest threat to the future of Africa, even though malaria kills more people, and common endemic infections contribute to an abysmal life expectancy.   (This was written 2-3 years ago and was probably incorrect even at that time;  estimates are that today there are  1.5-2 million deaths from AIDS in Africa, with close to 1 million deaths from malaria.  Malaria though  is responsible for a greater  number of deaths in children under 5 years of age).

It continues to be remarkable that although evidence has existed for years that many of these infections can interact with HIV infection to increase its infectivity and accelerate disease progression, those who advocate for, and allocate funds to fight HIV/AIDS seem oblivious to the relevance and implications of these interactions.

This effort of course needs absolutely no justification, but its funding is small compared to the resources that have been made available to combat HIV/AIDS –  but from all that has been described funding for these endemic infections is in fact also funding to fight HIV/AIDS “.

Those were comments made 2-3 years ago.

While malaria and tuberculosis are now receiving attention and are included with AIDS in some programs,   many other endemic infections  continue to be neglected.

Going back much further in time,  interest in the activating effects of associated infections on HIV replication began within the first 10 years of the epidemic.  This started with the demonstration that proinflammatory cytokines, TNF alpha or IL 6, for example could greatly accelerate HIV replication.

Of course these cytokines appear in the course of many different infections.  When viral load tests became available this effect was well understood by patients and physicians in N America and Europe. It became common wisdom that an HIV infected person who had a febrile illness, or had even received a flu vaccine  should delay viral load testing because the infection or vaccination was frequently associated with temporary rises in HIV viral loads.

The implications for geographic areas where the infections were far from temporary seemed to escape notice.

Thus endemic infections in Africa do have everything to do with HIV/AIDS.  There are numerous preventative and therapeutic measures available to control many of these infections,  and some are inexpensive.  Even something as simple as deworming may be useful.  Ascaris lumbricoides, the common intestinal round worm also is associated with immune activation and is easily got rid of.  There is a report that doing this with a drug called albendazole actually raised CD4 counts. (Walson JL et al. Albendazole treatment of HIV-1 and helminth co-infection: a randomized, double-blind, placebo-controlled trial. AIDS 22:1601-1609, 2008).

The person who has been studying immune activation and the association of parasitic infestations and AIDS for the longest time is  Zvi Bentwich.   I can’t remember when his first  publication on this issue appeared but by the mid 1990s he was publishing on this association in Ethiopian immigrants to Israel.   Zvi Bentwich deserves the greatest credit for his early recognition of the importance of this association, its significance regarding immune activation and for his continuing contributions.   He pointed out the relevance of schistosomiasis to AIDS  (and TB) at least 10 years ago.

The connection of so many endemic infections with AIDS  in Africa is also a connection of poverty with AIDS.  I saw an absurd and instantly forgettable paper entitled something like “Poverty does not cause AIDS” a few years ago.    Of course poverty is not the direct  cause of ascariasis,  schistosomiasis, tuberculosis, or any number of devastating infections.  Poverty is a very significant factor in  the acquisition of these infections, and as such can certainly be regarded as having a causative role.

The lives of impoverished populations are ravaged and shortened by these infections. Many of these infections also interact with HIV to compound the devastation they cause.  Poverty, multiple endemic infections and HIV are intimately intertwined and in many instances reciprocally affect each other.  For example the debility associated with schistosomiasis has an impact on an individual’s productivity, with economic consequences not only for the individual but for the larger community.

Controlling the AIDS epidemic in Africa must also include measures to prevent and treat the multiple endemic infections that affect hundreds of millions of individuals.

To conclude this post I want to recommend a book published about four years ago by Eileen Stillwaggon, a professor of economics.  It is called “AIDS and the ecology of poverty” and is published by the Oxford University Press.

Herpes Viruses and HIV: Some early History and a Bit about Safe Sex

May 17, 2009 1 comment

[The relationship between herpes viruses and HIV disease is also discussed in a subsequent post: ]

The relationship between herpes simplex virus type 2 and HIV is in the news again.   This time the press reports are that while acyclovir failed to suppress transmission of HIV it did cause a 17% reduction in HIV disease progression.

This reduction in disease progression was assessed by noting differences between the treated and placebo group in the numbers whose CD4 count dropped below 200, and who died.  A reduction in HIV viral load was also observed in those treated with acyclovir.

The concept on which this study was based is absolutely solid.

Herpes simplex virus type 2 is the most frequent cause of genital ulcers, and the presence of genital ulcers is associated with enhanced transmission of HIV.

The failure of acyclovir to suppress HIV transmission is a disappointment, but the study should not be seen as a failure.

There is no doubt that anti herpes drugs can suppress the recurrent herpes ulceration that some individuals experience. This was observed in the study.

Herpes viruses – and not just herpes simplex virus,  have an impact on the course of HIV infection.  This study provides yet another demonstration that treating herpes virus infections has a beneficial effect on the course of HIV disease.

Valtrex, a drug related to acyclovir was reported to reduce HIV viral loads in infected women in 2007.

“Reduction of HIV-1 RNA Levels with Therapy to Suppress Herpes Simplex Virus” and it appeared in the New England Journal of medicine .

(NEJM 2007  356:790)

It is possible that the association of herpetic genital ulcers with HIV transmission is not as direct as generally assumed.  The reasonable suppositions included the possibility that the ulcers provided a portal of entry for HIV in the uninfected partner, that there was an accumulation of CD4 cells in the ulcer that provided a good target for HIV, or even that in the infecting partner HIV was present in greater concentrations in the ulcer.

These assumptions about the reasons for increased HIV transmission may all be mistaken.

We do know with some confidence that transmission of HIV is related to viral load in the infecting partner.  It may be that the assumptions outlined above derive from observing an increased frequency and duration of genital ulcers in individuals with higher viral loads who are therefore more infectious not by virtue of the ulcers.

An individual with higher HIV viral loads  will more easily transmit the infection,  and also experience  more frequent recurrences  herpetic ulcers.  This of course only applies to HIV infected individuals.

As far as individuals who are not HIV infected are concerned, a direct causative association between herpetic ulcers and HIV infection may also be spurious.

Herpes simplex infections are ubiquitous but immunological mechanisms generally control the infection so that it remains latent and not manifested.

Sometimes individuals know what provokes a recurrence.  Recurrences can be associated with febrile illnesses.  It is completely reasonable to suggest that the effects of some  intercurrent  infections may cause  both  herpetic recurrences and increase susceptibility to HIV.

Whatever infection  causes the fever may also increase susceptibility to HIV, possibly by an association of the  infection with perturbed immunological function.    Transient immunological perturbations  can accompany many viral and tropical infections and so may not only disturb herpes simplex latency but also increase susceptibility to HIV.

For some reason, interest in the relation of HIV to herpes viruses seems to have been almost completely confined to herpes simplex virus type 2.  At least regarding what is reported to the public.

However the herpes virus family includes other members which have long been thought by some – including myself, to play an important role in HIV disease.

Cytomegalovirus  (CMV) and the Epstein Barr virus (EBV) are perhaps the two that are most important.  These viruses are also sensitive to the anti herpes drugs used in these two trials.

Since infections with CMV and EBV  are so widespread how can effects of acyclovir and Valtrex   on reducing  HIV viral loads be attributed to an effect of these drugs  on herpes simplex type2?

I cannot recall that these two other members of the herpes virus family – or even a third, HHV6  were even mentioned in the papers demonstrating effects of acyclovir and Valtrex on HIV viral loads.

It is entirely possible that suppression of  two viruses,  CMV and EBV, contributed, perhaps to the greatest extent,  to the anti HIV effects seen.

One can only hope that sera from these studies were frozen and stored.  Such samples could provide information on an effect of these drugs t on EBV reactivation and on active CMV infections.

As an historical comment, acyclovir was tried as a treatment for AIDS in 1987  around the time AZT was introduced.

There were several studies of differing design over for some years from about 1987, some based on the hypothesis that CMV contributed to disease progression.

AZT was tried with or without acyclovir, but the results were contradictory. Interestingly AZT also inhibits EBV replication.

One study, ACTG 204, which compared two doses of acyclovir with Valtrex was stopped because 25% of those taking Valtrex died compared to 20% taking acyclovir.

Some observational studies (including the MACS study) found that there was some survival benefit among those taking acyclovir.  Another retrospective observational study found no benefit.

Nothing much can be made of these contradictory early results.

But now, with newer techniques for measuring HIV activity by viral load assays, we   have very clear evidence that treating herpes virus infections has a beneficial effect on HIV infection.

With the advent of the newer potent antiviral drugs, interest in anti- herpes drugs did wane, until there was a renewed interest in the past few years in connection with herpes simplex virus 2 and genital ulcer disease,  Unfortunately most of the  emphasis is on herpes simplex virus, when suppression of CMV and EBV may be as – or I believe,  of even greater importance.

Actually there had been  interest in CMV and EBV in relation to AIDS from the time the disease was first reported in 1981.

I have been involved in AIDS research and  treating patients with this disease from the time it started and so can  provide some historical perspective on the interest in herpes viruses,  that dates to the late 1970s, even before AIDS was described and long before HIV was discovered.   At this early time epidemiological studies on the prevalence of infection by CMV among sexually active gay men were undertaken in the US.

As another historical interlude,  interest in herpes viruses also provided the basis for safer sex, as it is understood today.  As remarkable as this may seem, the first published and disseminated proposal to use condoms to prevent the transmission of AIDS had nothing to with HIV.   Condom use was proposed a few years before this virus was discovered, and had everything to do with herpes viruses, specifically CMV.

From about 1978 I had the opportunity to observe and treat a very large number of men who were to be the first to succumb to this new disease.

I knew that over 90 % of gay men attending a clinic for sexually transmitted diseases around that time had antibodies to CMV compared to 54% of heterosexual men.   By 1983  over 40% of a cohort of gay men in New York City carried CMV in their semen.   Amongst my patients, studies on EBV carried out by David Purtilo at the University of Nebraska showed an extraordinary high prevalence of reactivated EBV infections.  (Epstein Barr Virus and chronic lymphadenopathy in make homosexuals with Acquired Immunodeficiency Syndrome. H Lipscomb et al.  AIDS Research 1983 1: 59)

At that time – 1981-1982, many of the patients I was taking care of experienced reactivated EBV infections as determined by serological methods,  and were excreting CMV in semen. Of course they were also infected with HIV , but this could not be known at that time.

But from what was known about CMV and EBV it was reasonable to postulate that these viruses were somehow implicated in the disease.  It was thus possible to propose a way to at least prevent the sexual transmission of CMV.

This formed the basis for the first published recommendations for condom use.

With two of my patients, Michael Callen and Richard Berkowitz a booklet was written called “How to have sex in an epidemic: One approach”.

The appropriate  title  was  coined by Richard.

The twenty fifth anniversary of the publication of this booklet, that was essentially produced and widely distributed by four individuals, and funded by a single person, went almost completely unnoticed in 2007.    Although it is  in fact a landmark event in the history of the epidemic.

Richard is only now receiving some acknowledgement for this life saving proposal  because a documentary film called Sex Positive has brought attention to  his achievement.

An account of our collaboration in producing the safer sex guidelines can be seen by following this link.

Safer sex recommendations.

Michael Callen is remembered by many for his activism.   There is even a clinic in New York City named for him and Audre Lorde .

I actually worked there as a physician for a short period, and with very few exceptions, the health care providers and others working there had no idea of who he was, let alone his contribution to safer sex.

I just visited the Callen Lorde website, and indeed there is a photograph of Michael and of Audre Lorde with a few words about each, but no mention of Michaels contribution to safer sex.

Thus herpes viruses, at least CMV had a role in the development of safer sex recommendations.

As it turns out herpes viruses – CMV and EBV included, have a great deal to do with AIDS.    This is quite apart from their multiple clinical manifestations as opportunistic pathogens.  Both of these viruses almost definitely contribute to pathogenesis.

Evidence that some herpes viruses can play a critical role in HIV disease progression has accumulated  for many years.

In fact some evidence for this  was already apparent when AIDS was first described.

This considerable body of evidence did not disappear with the discovery of HIV, but was relatively neglected.

As work on HIV proceeded we gained some understanding of the ways in which herpes viruses can interact with HIV to accelerate disease progression, increase HIV infectivity and thus enhance its transmission.

I should now describe some of the interactions that exist between herpes viruses, particularly CMV and EBV, and HIV.

Many, perhaps most of these interactions also involve herpes simplex viruses types 1 and 2.

The role of CMV in immune system activation, a major force in driving HIV infection.

The systemic effects of CMV and EBV infections are most probably of great importance in this respect.

Systemic effects resulting in immune system activation and activation of HIV replication may also  accompany reactivated herpes simplex virus infecteions.

Among the systemic effects of active herpes virus infections are the secretion of pro inflammatory cytokines.  These circulate and attach to specific receptors on the cell surface. A consequence of this is that certain sequences on DNA will be activated resulting in the transcription of HIV DNA and ultimately the production of new HIV particles.  So, this is but one way in which an active herpes virus infections can promote the replication of HIV.  The general mechanisms are described in a previous post..

An important and interesting  paper that also deals with   EBV and CMV in relation to HIV replication was published by V Appay and colleagues.  It can be seen  by clicking the following link.


I am  reproducing some excerpts from Dr Appay’s paper here as the descriptions are very clear and there are references.  The references can be seen in the complete text seen by following the above link.

“HIV-1 also causes immune activation and inflammation through indirect means. Antigenic stimulation during HIV-1 infection may be induced by other viruses, such as CMV and EBV”

“In addition, inflammatory conditions occurring during HIV infection (eg release of proinflammatory cytokines) may also participate in

the reactivation of latent forms of CMV and EBV. Recent studies have shown significant activation of EBV- and CMV-specific CD8+ T cells during HIV-1 acute infection [40,41] . Hence, sustained

antigen mediated immune activation occurs in HIV-1-infected

patients, which is due to HIV-1, but also to other viruses (and may be restricted to CMV and EBV)”.

“CMV has been associated with strong and persistent expansions of T cell subsets that show characteristics of late differentiation and replicative exhaustion [94-96]. The anti-CMV response appears

to monopolize a significant fraction of the whole T cell repertoire [97], so that it might compromise the response to other antigens by shrinking the remaining T cell repertoire and reducing T cell diversity. CMV infection is actually extremely common in HIV-1- infected individuals and its recurrent reactivation may put further stress on their immune resources. Interestingly, CMV-seropositive subjects generally experience more rapid HIV disease progression than CMV seronegative subjects [98]”.

Herpes virus (including herpes simplex) infected cells express Fc receptors on their surface.  These receptors can bind certain sequences on antibody molecules. If these antibodies are attached to HIV, a portal for entry of HIV is provided on herpes infected cells that do not possess CD4 molecules on their surface. This process has in fact been demonstrated.

Transactivation  of HIV by herpes viruses.

In cells infected with both viruses herpes virus gene products can activate HIV and promote its replication. The transactivation is reciprocal as HIV can promote herpes virus replication.

Acyclovir and Valtrex have no direct effect on HIV except under one unusual circumstance,  yet both have been demonstrated to reduce HIV viral loads.

In the early 1980s when we had no effective measures against  this disease I treated my patients with high dose acyclovir.

There then  was evidence, albeit theoretical and indirect for a role for these viruses in this new disease.

In the absence of clear evidence from clinical studies, and given the gravity of the disease, it seemed completely appropriate to be guided by these theoretical considerations, particularly involving a drug that is so free of toxicity.

But interestingly,  at that time,  none of these theoretical considerations placed much importance on HSV 2.

The practice of medicine in those years, dealing with such a mysterious and deadly disorder of unknown causation , demanded responses that could only be based on one’s best judgment.

Fortunately I also had had some experience in the transplant field and was also able to provide bactrim to my patients years before recommendations for its use were issued.

But it was not until potent antiviral drugs became available that we were able to make significant and life saving, rather than life extending  interventions.

What I have written of this experience with bactrim in the early years can be seen by following this LINK

In the light of later evidence, I believe it is possible I was able to provide some small benefit in prescribing high dose acyclovir in those very early years.

[i]   Acyclovir, when phosphate is added to it, acts like the nucleoside analogues active against HIV, drugs like AZT, D4T, 3TC etc.   But this drug has a truly remarkable quality.  The cellular enzyme that  adds phosphate to make drugs of this type active,  does not work on acyclovir as it does on AZT, 3TC and other anti HIV nucleoside analogues.   But an enzyme, thymidine kinase that is encoded by herpes viruses, and therefore only appears  in herpes virus infected cells  has the ability to add the phosphate group and turn acyclovir into an active drug.  This is the reason why acyclovir is such a safe drug.  It only disrupts DNA synthesis in herpes virus infected cells, where of course this effect is desirable; it has no effect on uninfected cells.

However, if  the same  cell happens to be infected with HIV and a herpes virus, the herpes thymidine kinase will phosphorylate acyclovir, which now can work to  terminate  HIV DNA synthesis just as 3TC , AZT and similar drugs do when phosphorylated by the cellular enzyme.

This effect , only observed in doubly infected cells in the laboratory is unlikely to be of much significance in the body.

Interferon: Another Historical Digression

This is about something I wrote in 1964, which was recently reproduced and is now available on line.

It can be seen by clicking on this link:

1964 interferon article.

Seeing this 45 year old document prompted me to write this post.

It is about interferon and has nothing to do with AIDS, at least not in any immediately obvious fashion.

It is an interesting story, about at least one of the ways in which science progresses.  It is a story of how an apparently insignificant change in an experiment can sometimes lead to very significant advances.  In this instance, about how cytokines exert their effects.

Cytokines are protein or peptide molecules released by cells which then attach to the surfaces of other cells.  As a consequence, the behaviour of the cells to which they attach is altered.  In this respect cytokines are similar to hormones.

Generally, each cytokine will only attach to a specific receptor on the surface of the cell.

When a cytokine attaches to its matching receptor, a cascade of events is set in motion resulting in the activation of specific sequences in nuclear DNA.

Messenger RNA molecules are then transcribed from specific DNA sequences and these direct the synthesis of specific proteins that ultimately are the molecules that cause the particular effects produced by the cytokine.

Therefore, as the picture below demonstrates, cytokines are not themselves the molecules that directly mediate the effects they cause.  Through a complex series of signalling events in the cell, set in motion by the binding of the cytokine to its receptor, specific proteins are made by the cell.  These proteins are the actual mediators of the cytokine’s effects. [1]

In the illustration, the right angled arrow in the nucleus represents the messenger RNAs which will direct the synthesis of these proteins.

HIV DNA is integrated into host DNA.   Should certain cytokines,  IL-6 or TNF alpha for example,  attach to their receptors on the cell membrane,  a series of events follow, ultimately resulting in sequences in nuclear DNA being activated which in turn causes HIV DNA to make RNA which directs the synthesis of HIV proteins and ultimately of new HIV particles.


Since many of those cytokines that can activate HIV in this way are produced during the course of many different infections, this then is but one of the several ways in which HIV replication can be enhanced by many different concurrent infections.  TB and malaria are among them, as are the bacterial diarrheal infections associated with a lack of sanitation and clean water.   Controlling these many  HIV enhancing infections,  is  with the exception of TB,  a neglected target in the fight against the epidemic.

Interestingly,   discoveries about the ways in which cytokines exert their actions have largely been made since AIDS was first recognized in 1981.

Thus HIV research has progressed in tandem with research on molecular cell biology.  There have been reciprocal benefits.  HIV research has both contributed to our understanding of molecular cell biology, as well as itself being advanced by discoveries in this field.

Interferon was the very first cytokine to be discovered.  It was discovered by Alick Isaacs and Jean Lindenmann .   Actually it was not really discovered as a specific molecule; the term interferon was coined by Alick  Isaacs in 1957, to describe an activity – an antiviral activity released by virus infected cells. It was perhaps a bit premature to assume that this activity resided in a single molecule. But that was what we all thought at the time; it was nonetheless a concept that facilitated research as probably did the coining of the word “interferon” to describe this antiviral substance.

We now know that there are many types of interferon, and we therefore should properly speak of the interferons.  Also, as is the case with cytokines generally, the interferons have multiple effects, but the antiviral effect is how it was first recognized and also measured.

Alick Isaacs  was my mentor in the laboratory study of viruses; I shared a lab with him and  worked on the mechanism of the antiviral action of interferon.

In 1963, we had no idea about how interferon exerted its antiviral effect. We at least knew that it did not directly inactivate viruses.  Molecular biology – at least as far as eukaryotic cells were concerned, had hardly developed.

The 1964 article that can be seen by following the link at the beginning of this post resulted from the work of Joyce Taylor.1964 interferon article.

Joyce Taylor is a biochemist.  She also worked in Alick’s lab in 1963.  It was rather unusual,  in those days for a biochemist to be working in a lab concerned with animal viruses. Animal virology was just beginning to employ biochemical methods.

Joyce was attempting to show that interferon blocked the synthesis of viral RNA.   This of course required the use of biochemical techniques to identify and measure viral RNA.

She was able to demonstrate that viral RNA was not made in cells treated with interferon. This was accomplished by using a compound that blocked DNA directed cell RNA synthesis, actinomycin D.  It was necessary to use actinimycin D because there is so much background  cellular  DNA directed  RNA synthesis that unless this can be stopped it would be impossible to observer viral RNA synthesis. She used an RNA virus (SFV) that was unaffected by actinomycin D.

Joyce very clearly showed that the synthesis of SFV RNA was blocked in cells treated with interferon.  as with the availability of actinomycin D,  she was able to detect and measure viral RNA.

We are now coming to the happy, but at the time seemingly  insignificant change in the sequence of steps in an experiment,  that had such far reaching consequences.

This is how Joyce did her experiments.  Cells were exposed to interferon for some hours, and then the SFV virus added with actinomycin D, to allow the measurement of viral RNA synthesis by removing the background of cellular RNA synthesis.  As mentioned,  in this way, Joyce was able to show very clearly that pre-treatment of the cells with interferon blocked the synthesis of SFV RNA.

One day, because Joyce had to leave early and she did not want her technician to handle actinomycin D, she added this drug with the interferon, at the beginning of the period of interferon treatment  .  Nobody at that time would have thought that this would make the slightest difference.   It is this change in the time when the actinomycin was added that was critical, but it was not at all expected to have any effect.

But it did have an extraordinary effect.  When cells were treated with interferon in the presence of actinomycin D it had no antiviral effect.  At first it was thought that an inactive preparation of interferon was used, but the same result was obtained when the experiment was repeated.

The significance of the change in the order in which actinomycin was added was that now, while the cells were exposed to interferon, DNA directed RNA synthesis was also blocked.

The implications of this were quite extraordinary.  At that time, 1963 and 1964, the foundations of our understanding of basic molecular cell biology were being worked out mostly in bacterial systems.  The structure of DNA had been worked out, messenger RNA discovered (although there is some dispute as to who discovered it) and there was some understanding of derepression – that is the ability of certain molecules to cause the synthesis of specific proteins by bacteria.

The result of the changed order of Joyce’s experiment suggested that something similar might be happening in animal cells- that interferon was inducing the synthesis of a specific messenger RNA which in turn directed the synthesis of a  protein  responsible for its antiviral effect.  This is what prompted me to write the short article that can be seen by clicking the link at the beginning of this post.

What was described in 1964 was  in fact the first demonstration that cytokines exert their effect by attaching to a receptor on the cell surface,  and  as a result of this,    specific regions on cellular  DNA are activated,and  RNA synthesized.  Work showing that this RNA is responsible for the synthesis of proteins followed immediately.

Robert Friedman, was visiting the laboratory from the National Cancer Institute, and we worked together to show that not only RNA synthesis, but also protein synthesis was required for interferon action – and as was to be found, for the action of all cytokines.

Joyce Taylor remembers this story somewhat differently, but I trust that my version is correct.  I have repeated it so frequently since the events in question, as an illustration of how science sometimes progresses.

Joyce changed the order of adding reagents. As a result we knew that interferon action needed cell the participation of cell DNA and the synthesis of RNA.  Bob Friedman and I then showed that interferon action also required cell protein synthesis. Ian Kerr who was also in the lab around that time, and others then showed a part of what changes interferon induced in cells.

Interferon was the tool by which a signalling pathway was demonstrated that could account for the effects exerted on nuclear DNA by a molecule interacting with its receptor at the cell surface.   Ian Kerr was a key contributor to this work.

This post was not directly connected with HIV/AIDS.  But cytokines are most certainly  connected with HIV/AIDS.  This will be the subject of future posts.

[1]  The genetic code is defined by the sequence of the four bases that make up genomic DNA. A particular sequence of three nucleotides can be regarded as a code component  which ultimately defines a particular  amino acid; amino acids are the building blocks of proteins.  The DNA code is conveyed from the nucleus to the protein synthesizing apparatus in the cell cytoplasm in the form of messenger RNA. This RNA molecule is made from a DNA template and exactly reflects the nucleotide sequence of the section of DNA from which it is transcribed.


HIV Infection in HIV Antibody Negative Individuals

April 1, 2009 2 comments
  • HIV infection in HIV antibody negative individuals

The possibility that there are individuals who are infected with HIV but who are negative on the test for HIV antibodies has always been theoretically possible. Considerable evidence has accumulated for many years that there are indeed such individuals. Despite the importance of this phenomenon, it receives relatively little comment.

It sometimes seemed to me ever since I first tried to discuss this possibility in the mid 1980s that there was a wilful discouragement of any discussion of this topic.

In 1989 David Imagawa reported that that 31 of 133 HIV antibody negative men showed the presence of HIV.

In 27 of them this persisted for 36 months despite remaining seronegative  [1]. This resulted in a vigorous response culminating in what almost looked like a retraction by the authors. At that time many unsuccessful attempts to replicate these results were reported, and the findings of David Imagawa were generally presented as due to technical errors, such as incorrect specimen labelling. In view of many subsequent findings, the likelihood is that David Imagawa and his colleagues were correct. The furious response to Imagawa’s paper is an indication of how non rational considerations can influence the progress of science. This is of course nothing new.

Curiously in a recent book, Imagawa’s findings are included in a list of what are stated to be errors and controversies in the HIV/AIDS epidemic that impeded scientific progress [2]

What in fact impeded progress was a rigid adherence to what was only a hypothetical, not an empirical model of the course of HIV infection.

David Imagawa died shortly after this controversy, and sadly did not live to see that his initial conclusions were absolutely consistent with what has been learned of the complexity and diversity of individual responses to HIV infection.

I certainly experienced considerable resistance and disbelief when I raised the possibility of silent HIV infections. In the late 1980s I took part in a NPR program, and was quite abruptly dismissed by another scientist (I have forgotten who) when I raised the absolutely reasonable theoretical possibility of persistent latent infections in antibody negative individuals.

Apart from very few exceptions there was an almost complete lack of interest in HIV seronegative, but infected individuals, by science writers; there was no shortage of community commentators who also seemed to be oblivious or uncaring of this phenomenon.

To be sure there were occasional reports of seronegative but infected individuals. Gus Cairns, a UK journalist wrote about this in the UK magazine, Positive nation. I wrote something about this as a result of an interview with him in 2000, which he published. I have scanned the article. I was unable to make a perfect copy, but a legible version can be seen by clicking HERE.

In the US reports confirming the existence of seronegative infected people continued to receive very little comment; what little there was was generally quite hostile..

Today this issue was again brought to my attention by an article I saw reporting the presence of HIV proteins and HIV RNA in cervical biopsies from women who were persistently HIV seronegative , at least for the duration of the study which was one year [3]. They did not have antibodies to HIV despite being infected; of course it is possible that they are in an unusually long “window period” and will eventually seroconvert.  If we use “window period” in this sense then we  can speak of a distribution of window periods of different lengths, including an indefinite one.

I expect that, as is usual this report will provoke little or absolutely no interest.

But it is enormously interesting; (just one of many questions: can these women infect their male partners?)

Seeing this article is the reason why I decided to make this issue the subject of this post.

It was no great surprise when evidence appeared that there were some individuals who were HIV infected but remained negative on the HIV antibody test. It must be said that there were probably more papers in the early years in which silent HIV infections in HIV antibody negative individuals was not observed.

In another approach, reports started to appear that HIV antibody negative individuals had T lymphocyte responses to HIV which means that they were exposed to the virus, not necessarily that they were infected – although that is quite a real possibility. Some early papers, before 2000, including those showing T cell responses can be seen by clicking HERE . There was quite an extensive literature at that time, but most, as mentioned reported that there was no such thing as a silent antibody negative infection, apart from the short window period following infection.

Why has the possibility of prolonged latency always been theoretically possible?

As part of its life cycle HIV is turned into DNA and is then incorporated into the host genome. In infected cells it effectively becomes part of our genetic material. Once inserted into human DNA, it must be activated to start the process of making new virus particles. Cellular signals that start the process of activating HIV DNA include cytokines, which are messenger molecules produced and released by cells, which can then act on other cells to evoke a variety of responses. Amongst these HIV activating cytokines are those that are called proinflammatory cytokines.  These appear during the course of many different infections.  Once HIV DNA is activated, and at least some of its proteins made, these then mediate further activation.

There are some other factors that can activate HIV DNA.

Alloantigens are antigens expressed on foreign cells. When these antigens are in contact with a cell containing integrated HIV DNA, activation occurs; HIV DNA is transcribed and new viral particles made. In earlier days HIV was isolated from infected lymphocytes in this way. Latently infected lymphocytes were induced to produce HIV by culturing them together with lymphocytes from an uninfected donor.

It is the nature of HIV infection that it is frequently acquired in situations which involve exposure to foreign cells (to alloantigens). This may be exposure to semen in sexual transmission, or blood cells in the case of infection by shared needles, or by blood transfusion.

Herpes viruses have the ability to activate HIV if a cell is infected with both viruses. I suppose this must happen but I imagine doubly infected cells may not be found  too frequently. Of course active herpetic infections in non HIV infected cells may be associated with the production of pro inflammatory cytokines, which circulate and can activate HIV DNA in a cell at a distance.

There is absolutely no reason not to expect that in some circumstances incorporation of HIV DNA into human DNA will result in a state of stable integration. This means that HIV DNA remains in the genome, it is not activated, and no virus is produced. Since antibodies are made as a response to viral proteins, and as none are made, the HIV antibody test will be negative.

So it was no surprise when such individuals were again reported in 1999 [4]. These individuals remained in good health and were reported to be antibody negative as long as they were observed [5].

We cannot know if these individuals may seroconvert (or maybe already have), but what is established is that stable integration of HIV DNA without seroconversion can occur. In such individuals limited expression of HIV can occur, at least sufficient to induce, if not antibodies, a cellular immune response.

The presence of such cellular immune responses in HIV antibody negative individuals is further evidence consistent with HIV DNA persistence, but in itself does not indicate this.

Demonstration of cell mediated immunity to HIV:

Apart from the identification of antibodies, specific immunity to HIV can also be detected by a much more elaborate test that measures cellular immunity rather than immunity determined by detecting specific anti HIV antibodies. In this case what is measured is the ability of lymphocytes to recognize HIV. They will do so only if they have been exposed to the virus, which would obviously be the case if they were taken from an infected individual.

The detection of such lymphocyte responses in the mid 1990s was one of the first indications that there may be infected people who don’t make antibodies. Other interpretations are that the infection was overcome, or that that the individual was infected with defective virus.

Gene Shearer was I believe the first to report this phenomenon. HIV antibody negative sexual partners of HIV positive people, as well as individuals who had occupational contact with HIV were among those showing these responses.

It is unknown how widespread this phenomenon of silent HIV infection is. It may be exceedingly rare. It is also unknown if this condition of stable integration is really just a prolonged “window” period that always follows all HIV infections.

But it is entirely possible that there are individuals in whom the ability to control HIV is such that they will remain healthy and HIV negative.

A number of different  outcomes of HIV infection are possible:

Some of the factors that influence this:

Host genetic factors.

Size of the inoculum – the amount of infecting virus.

Route of infection

The particular virus strain.

The presence of associated systemic infections.

these provide signals activating HIV proviral DNA. In the case of some tropical infections there may be cytokines (IL 10) that blunt immune responses.

Sexually transmitted infections with genital ulcers.

Double infection of a cell with HIV and herpes viruses – probably an unusual occurrence.

Exposure to alloantigens; a theoretical possibility.

These are some of the known influences.

Maybe the most common outcome is a productive infection where viral DNA is activated within a few weeks.

But this scenario is also possible:

Infection is followed by insertion of HIV DNA into cellular nuclear DNA. Possibly with small inoculums, and in the absence of strong or sustained activation signals, the proviral DNA remains silent. This has been observed.

Or this one:

There is a limited burst of viral production, not sufficient to elicit an antibody response but enough to induce a cell mediated response with the generation of lymphocytes that recognize HIV antigens and can kill HIV infected cells. HIV seronegative individuals with such specific lymphocyte responses have certainly been observed. In this case if there is an incipient burst of HIV production, the producing cells are promptly killed. Each time this happens the cellular immune response is primed and strengthened. Such a mechanism has been well studied in EBV infections. This common virus is totally unlike HIV, but it does similar things. It remains present in B lymphocytes rather than T lymphocytes for life. The mechanism of persistence is quite different – EBV is not a retrovirus. But the majority of individuals carry this virus – which in rare situations can have lethal effects, in their B lymphocytes for life. We have evolved many mechanisms to keep this virus in check. The ability of some types of lymphocytes to kill EBV infected cells which start to make virus is well understood. Similar mechanisms must exist for HIV – but obviously for most, are insufficiently effective. But in those with very limited HIV production these killer lymphocytes may actually be what allows such rare fortunate individuals to remain HIV seronegative.

With this outcome, one can view the infection as actually having an immunizing effect.

If there were not yet enough reason to study the phenomenon of persistently seronegative HIV infection, this is an important one. What are the circumstances that produce this outcome?

So, for many reasons individuals who are seronegative but have lymphocyte responses to HIV are of great interest.

Yet another scenario is one of stable integration, but where some HIV proteins, but not complete virus, are produced. Maybe the women referred to whose cervical biopsies contained HIV antigens might be in this category. This is a strange situation as antigens were detected but these women apparently did not develop antibodies.

Another very early observation that can be explained by the prior presence of integrated HIV  DNA that is only activated by a subsequent non HIV  infection is the finding that  episodes  of EBV reactivation may precede HIV seroconversion. [6].  This raises the possibility that at least some illnesses associated with primary HIV infection are nothing of the sort. They instead may represent rather non specific viral infections that activate already present integrated HIV DNA, and thus  followed by HIV seroconversion. This is a completely plausible scenario. Of course self reported sexual histories may sometimes  not be too reliable, but nontheless, I well recall an older gay male patient of mine who told me that he had had no sexual contact for years, he had several negative HIV tests over a period of a few years, and then tested positive.  I wondered  then if he may possibly have been infected years before, that he carried latent HIV DNA and this was subsequently activated by some febrile illness. I know this is only an anecdote, and that individuals can be guarded about their sexual histories.  I wonder if others have had similar experiences?

I think around 1996  a description of the course of infection was produced. Everyone interested in this disease will have seen this picture: Here it is again:


This may represent a typical course of infection.  But HIV disease is probably so variable in the course it can take that there may well not be such a thing as a typical infection.

This depiction does however give the impression that there is,  and discourages an appreciation of the probably  immense variations in the course of  HIV disease.  The notion of a “standard” course of HIV disease has  had implications for treatment.  Recommendations are made that take no account of  individual  rates of disease progression;  a one size fits all approach has been adopted.

The  rapid acceptance that there is a typical – or an  average  course of HIV infection is particularly odd as not only is the disease new – we have no precedents of human retroviral diseases (apart from HTLV-1 associated disease);  the techniques used to study the disease are themselves new. The ability to identify T lymphocyte subsets with monoclonal antibodies is about as old as the HIV epidemic. So we had no idea then of the variation in T subset numbers in health and disease. Other immunological and virological techniques were, and continue to be introduced as the epidemic is proceeding.

A model was constructed before sufficient evidence was available to justify it.  It really had no empirical basis; moreover it seemed to utterly ignore what we knew of other chronic viral diseases.  For example, hepatitis B and Hepatitis C can both have very variable courses.  These can range from clearing the infection, running a fulminant course ending fatally  to the establishment of a chronic active state which may progress at varying rates.  If we were to construct a model of the course of HIV disease only about  12 to 15 years after the disease was first seen, why on earth did we not consider the precedents of other chronic viral diseases?   Thus we might have  included the real possibility that some exposures may result in infections that may be cleared , as well as the now demonstrated situation where silent antibody negative infections occur.    The picture shown above – and presented in every text on HIV disease may indeed represent the most common course of HIV infection. But even this is not  known.

HIV infection, like other chronic viral infections  can progress in different ways. If we were more open to this there may have been greater interest and funding into research that investigates the various factors that influence how the disease progresses. This has obvious therapeutic implications  –  for example as proinflammatory cytokines promote HIV replication, the control of endemic infections in some areas where they are highly prevalent is absolutely relevant to the control of HIV infection.  Steps as simple as the provision of sanitation and clean water may well have an impact on the control of HIV infection in some geographical areas.  Had we not been so tied to the notion of  a fixed course of HIV infection, we might have placed importance on the individualization of therapy, not only considering a fixed CD4 count as a signal to start therapy, but also considering each individuals rate of disease progression.

HIV disease is in this sense like  every other infectious disease, the course of which  to a greater or lesser extent can be influenced by many different factors , including host factors, factors related to the pathogen, the particular variant , the size of the infecting dose, the route of infection amongst many others.

I have often wondered why there has been such resistance to not only the reasonable idea, but also to actual evidence that HIV disease  does not necessarily  take the course  shown above.

In conclusion, the study of prolonged HIV seronegativity in infected people is important. Some reasons are:

1. There are obvious implications for vaccine development.

2. Seroprevalence may significantly underestimate the prevalence of HIV infection.

3. Understanding the phenomenon will advance our understanding of the pathogenesis of this disease, which in turn will open new therapeutic approaches.

4. There are instances of infected people remaining seronegative and in good health.


Imagawa, D.T., M.H. Lee. S.M Wolinsky. et al..

Human immunod­eficiency virus type 1 infection in homosexual men who remain seronegative for prolonged periods.

New England Journal of Medicine 1989 320:1458-1462.


Scientific Errors and Controversies in the U.S. HIV/AIDS Epidemic: How They Slowed Advances and Were Resolved

By Scott D. Holmberg

Published by Greenwood Publishing Group, 2008


Human Immunodeficiency Virus (HIV) Antigens and RNA in HIV-Seronegative Women with Cervical Intraepithelial Neoplasia
Jayasri Basu, Seymour L. Romney, Ruth H. Angeletti, Sten H. Vermund, Edward Nieves, Anna S. Kadish, Magdy S. Mikhail, and George A. Orr

The publisher of this journal kindly sends me the contents of each issue as I started this journal around 1983 and was its first editor, seeing it through its first two volumes. It was then simply called AIDS Research.


Zhu T, Corey L, Akridge R, Change Y, Feng F, Kim J, Alef C, Mcelroy J, Mullins J, Mcelrath J.

Evidence for HIV-1 latent infection in exposed seronegative individuals.

Abstract No.8, 6th Conference on Retroviruses and Opportunistic Infections. Chicago. 1999.


Persistence of extraordinarily low levels of genetically homogeneous human immunodeficiency virus type 1 in exposed seronegative individuals.

Journal of virology, {J-Virol}, Jun 2003, vol. 77, no. 11, p. 6108-16,

Zhu-Tuofu, Corey-Lawrence, Hwangbo-Yon, Lee-Jean-M, Learn-Gerald-H, Mullins-James-I, McElrath-M-Juliana.


Some individuals remain inexplicably seronegative and lack evidence for human immunodeficiency virus type 1 (HIV-1) infection by conventional serologic or virologic testing despite repeated high-risk virus exposures. Here, we examined 10 exposed seronegative (ES) individuals exhibiting HIV-1-specific cytotoxicity for the presence of HIV-1. We discovered HIV-1 DNA in resting CD4(+) T cells (mean, 0.05 + /- 0.01 copies per million cells) at multiple visits spanning 69 to 130 weeks in two ES individuals at levels that were on average 10(4)-to 10(6)-fold lower than those of other HIV-1-infected populations reported. Sequences of HIV-1 envelope and gag genes remained markedly homogeneous, indicating little to undetectable virus replication. These results provide the evidence for HIV-1 infection in ES individuals below the detection limit of standard assays, suggesting that extraordinary control of infection can occur. The two HIV- infected ES individuals remained healthy and were not superinfected with other HIV-1 strains despite continued high-risk sexual exposures to multiple HIV-infected partners. Understanding the mechanisms that confer diminished replicative capacity of HIV-1 in these hosts is paramount to developing strategies for protection against and control of HIV-1 infection.

Schattner, A, Hanuka N, Sarov B, Sarov I, Handzel Z, Bentwich Z.

Sequential serological studies of homosexual men with and without HIV infection. Epstein-Barr virus activation preceding and following HIV seroconversion.

Clin Exp Immunol 1991; 85: 209-13.

HIV disease and alpha interferon

March 21, 2009 Leave a comment

I had intended to continue writing about individualization of treatment for HIV infection with an emphasis on the variability of the natural history of HIV disease. Instead, I will make an historical digression. I’ll do this from time to time. An account of the diverse AIDS related issues with which I have been involved since 1981 (and even before the epidemic was first recognized) is on my web site I’m slowly adding content. This should speed up; a lucky circumstance has provided me with access to professional web advice. I have until now had to rely on “how to” articles to get the site going. Surprised that at 76 I have got this far. At the outset I had said that one purpose of this blog was to bring attention to the web site, and that is one reason for this short introduction.

This post is concerned with the connection between alpha interferon and AIDS. I should say connections; there are many.

Today, most people will probably only be aware of interferon in connection with the treatment of hepatitis C in HIV infected and uninfected individuals. The benefit conferred by interferon treatment to many people with Hepatitis C, even those co infected with HIV is tremendous. The FDA first approved interferon alpha for the treatment of hepatitis C in 1991.

There are many connections between interferon and AIDS; the first of these became evident in 1981, the year the epidemic was recognized.

People with AIDS produce large amounts of interferon themselves. The sustained production of large amounts of interferon by untreated HIV infected people with more advanced disease is not only a part of the disease, but the most compelling evidence suggests that its various actions contribute to producing some of the abnormalities associated with it. But, paradoxically, in the early years of the epidemic more was injected into patients in attempts to treat the underlying disease. (There is an important difference between this and treating Kaposi’s sarcoma or Hepatitis C in coinfected people with interferon).

That interferon can be seen as both contributing to the disease and also as a means of treating it makes for a confusing but interesting story [1]

It will be helpful to start with a very brief description of the interferon system.

The interferons – there are several types, are a family of proteins produced by vertebrates. They are cytokines, the name given to polypeptide or protein molecules produced by cells which act as signals that can influence their behavior, and that of other cells distant from the producing cell. In acting at sites distant from where they are produced, cytokines are akin to hormones.

All the interferons share some common properties, and it is easier to write about interferon in the singular. Interferon is best known for its broad antiviral effect. It is produced by cells in response to viral infections, and circulates to render other cells resistant to infection, thereby playing a central role in recovery.

Most viruses are sensitive to interferon and it was once hoped that interferon might prove to be a broad spectrum antiviral agent, similar to broad spectrum antibiotics that act against bacteria. There was great difficulty in purifying interferon for human use but in 1980 recombinant DNA technology permitted the manufacture of large amounts of pure interferon by inserting the gene for interferon into bacteria or yeast. Apart from its antiviral effect, interferon has numerous other effects, particularly on the function of the immune system. It can inhibit the growth of certain cancers, and has an inhibitory effect on new blood vessel formation. It therefore has been effective in the treatment of AIDS related Kaposi’s sarcoma. Unfortunately its clinical utility is more limited than originally hoped. Its greatest success is in the treatment of hepatitis C. Here are some brief reviews:

To return to interferon and its connections with HIV:

This paradoxical situation – in which interferon is seen at the same time to be harmful and helpful, has given rise to some peculiar interpretations. Research directions have been influenced; on balance the desire to treat HIV disease itself with interferon (as opposed to treating Kaposi’s sarcoma and Hepatitis C in coinfected individuals) has probably inhibited research into its role in pathogenesis. It is notable that the overproduction of alpha interferon, a striking abnormality in people with AIDS, known since 1981, was barely studied, let alone discussed, in the first years of the epidemic.

This strange story of a substance seen by some to be harmful, and by others to be beneficial in the same circumstances, is best told in the light of my own experiences in both fields; in AIDS and in interferon.

The two areas that have occupied my professional life have been the laboratory study of the mechanism of interferon’s antiviral action, and clinical work in HIV disease, providing direct medical care to a very large number of HIV infected people as well as conducting clinical and laboratory research on this disease.

In the strangest of circumstances these two fields came together as early as 1981. In that year, AIDS was first recognized, although I and others had already noted early manifestations of what was to be called AIDS among our patients.

The first information I received concerning the occurrence of Kaposi’s sarcoma in several gay men in New York City came from Dr Joyce Wallace. She had received a biopsy report of a diagnosis of Kaposi’s sarcoma in one of our patients. Joyce had called the National Cancer Institute to ask if there was a physician in New York who was familiar with what was then a very rare condition. She had been told that there were – I think at that time, about 20 men with this condition in New York who were under the care of Dr Alvin Friedman-Kien.

This was quite astounding. Unsurprisingly, I did not immediately connect this with what I had been seeing in my own practice, – enlarged lymph nodes, enlarged spleens, low white blood cell counts, low blood platelets among other abnormalities.

Jan Vilcek, who was head of the virology lab at NYU is an old friend and interferon colleague, and I knew Alvin, because he also worked in Jan’s laboratory. So I immediately called to obtain more information about this remarkable news.

Given my training and experience as a microbiologist and the nature of my practice, there was no question that I needed to contribute to the response.

So, in 1981, I also started to work in the virology laboratory at NYU.

I divided my time. Mornings were spent in the lab, and patients seen in the afternoons.

My work in the lab was initially focussed on cytomegalovirus (CMV) as there was evidence that many gay men at risk for this new disease were actively infected with this ubiquitous virus and excreting it at rates higher than noted in others. There also was literature at that time suggesting that CMV was involved in the development of non AIDS related Kaposi’s sarcoma (an idea that was not to hold up).

Once in the lab a strange circumstance brought interferon back into my life in connection with this new disease.

I read a preprint of a paper of Jan Vilcek’s where he described the ability of an antibody to lymphocytes – specifically anti CD3, to induce the synthesis of gamma interferon. Because of other observations that were made on our first patients, I had the idea that we would find gamma interferon in the circulation of patients with AIDS. This incident has been recorded and published by Jan Vilcek and an extract of the article can be seen by clicking here.

This is just the relevant part from a longer article, appearing in the annual “Interferon” series published by Academic Press and edited by Ion Gresser.

It explains how we came to look for interferon in the blood of people with AIDS. Alvin Friedman-Kien provided sera from his patients, and Jan Vilcek provided just about everything else. Gene De Stefano, who is the lead author on the paper we finally published,[2] was a student working in Jan’s lab. The author’s names are in alphabetical order, as this seemed the best way to deal with the matter of precedence, as so many collaborators had become involved. I had sent sera to Robert Friedman in Bethesda, another old friend and interferon colleague, and we joined forces in pursuing this work.

As Jan Vilcek’s account describes, my idea proved to be wrong, the interferon we found was not gamma interferon, but alpha interferon.

Many years later gamma interferon was detected in the circulation of people with AIDS.

This was the first of many connections between interferon and AIDS, a connection made in the first year of the epidemic.

It immediately suggested that the sustained presence of large amounts of interferon in the circulation might be contributing to pathogenesis, and that there was an autoimmune component to AIDS. Apart from AIDS, at that time the only other situation in which there was the sustained presence of large amounts of interferon was in auto immune diseases such as lupus. Also, as individuals with various diseases, including Hepatitis C were treated with interferon, auto immune complications were noted among them.

Since I will be critical of some aspects of AIDS research in relationship to interferon it is very important that, before I get into this, I make the following point very clearly.

Interferon has been of inestimable value to people infected with Hepatitis C, including those coinfected with HIV. Interferon in combination with ribavirin has been able to cure many individuals of Hepatitis C infection. It has thus been life saving, as the consequences of Hepatitis C infections can include liver cirrhosis and liver cancer. It is probably the case that interferon’s greatest clinical triumph has been in the treatment of hepatitis C. At one time it was also the only available treatment for Hepatitis B.

So, to emphasize the point, interferon for the treatment of hepatitis C in HIV infected individuals can be life saving. It may be useful in some instances of Kaposi’s sarcoma unresponsive to antiretroviral drugs.

But I believe it has absolutely no place in the treatment of HIV disease itself. There are early reports of benefits conferred by interferon treatment [3] but there is also a great deal of persuasive evidence that long term treatment is hazardous[4]. (This article contains numerous references supporting a role in pathogenesis for interferon).

So this is an illustration of the Jekyll and Hyde view of interferon. Does it mediate some of the pathological features of HIV disease, or should we use it to ameliorate these features?

On balance, I believe the evidence supports the view that overproduction of alpha interferon contributes to the manifestations of HIV disease. In specific instances, particularly in Hepatitis C in coinfected individuals and in some cases of AIDS related Kaposi’s sarcoma, the benefits of interferon most definitely outweigh the risks. This is particularly true in people with higher CD4 counts.

Nonetheless overproduction of interferon is a feature of AIDS. But It took many years for work to be done to identify the interferon producing cell. This was achieved by Frederick Siegal in 1999.

Quite early in the epidemic, AIDS was described as a disease characterized by a dysregulation of cytokine production. Interferon is a cytokine, in fact the very first to have been described, but it rarely appeared in the list of cytokine abnormalities associated with AIDS.

Here are some of the biological effects of interferon that resemble features characteristic of HIV disease:

Interferon inhibits the development of white blood cells and platelet and red blood cell precursors. It causes fevers. It stimulates the production of a molecule called beta2microglobulin, which was used as an adverse prognostic marker in AIDS. It affects lipid metabolism and can cause an increase in serum triglycerides, observed in AIDS patients before the era of HAART. It modulates the activity of B cells, which make antibodies, and B cells are overactive in AIDS.

But perhaps of greatest interest is the ability of Interferon α to selectively inhibit the proliferation of the CD4 lymphocyte subset, a finding that was published as early as 1983.[5]It also has a slight stimulatory effect on CD8 lymphocytes.

This is the “dark and sinister” side to interferon.

Given these effects of interferon it is hard to understand what the researchers hoped to achieve by injecting yet more of into people who were already full of it.

Two reasons were given for administering interferon. Firstly interferon has antiviral properties. This rationale was resistant to the obvious problem that despite large amounts of interferon in the circulation, HIV continued to replicate. Indeed, as the disease progressed and viral production increased, so did the levels of interferon.

The second reason given was that cells taken from people with AIDS could not be stimulated to produce interferon in the test tube. This was an early finding of Dr Siegal.

The inability of cells from people with advanced HIV disease to make interferon in the test tube is actually exactly what is to be expected. It has been known for many years that when cells are exposed to large amounts of interferon for long periods, they cannot be stimulated to make interferon. They are in what is called a refractory state. The authors describing the inability of patient’s cells to make interferon seemed to not consider this, and so the strange idea that the inability to make interferon was an intrinsic abnormality in AIDS was advanced as a reason to administer interferon.

The inability to induce interferon production in cells derived from people with AIDS is indeed strange as the circulation from which they are removed is full of it. The interferon must come from somewhere. It is possible that it comes from cells in solid tissue. The reason for suggesting this is that membrane fragments from HIV infected cells are excellent inducers of interferon. This suggests that in the body, interferon may be made by cells that are in apposition to HIV infected cells in solid tissue.

This may be more difficult to study now. AZT promptly removes interferon from the circulation, and this is probably true for all effective antiretroviral drugs.

The prompt removal of interferon by antiretroviral treatment must make one wonder if this is at least part of the reason for the benefits of treatment. Inhibition of HIV replication, associated with the loss of circulating interferon definitely suggests that HIV is responsible for the high levels of interferon.

In this connection here are some results that we observed:


The solid line represents HIV levels, and the dotted line interferon levels. These two individuals, A and B were treated with AZT for one week at weekly intervals. Both interferon and HIV levels promptly decline when on AZT and just as quickly go up when AZT is removed.


These three people started AZT at time 0. Both HIV (p24) and interferon rapidly decline.


These are individuals on continuous AZT therapy. Interferon rapidly declines in all, but returns at varying times despite continued treatment with AZT. Is it possible that the transient and variable duration of benefit experienced, coincides with the period when interferon is absent?


This is one person on continuous AZT treatment. Interferon starts to return and rise before 18 weeks. P24 only returns after 33 weeks. However this does not necessarily mean that interferon returns to the circulation before HIV. P24 measurements are not that sensitive and if PCR had been used HIV may have been detectable much earlier.

This is turning out to be a long post, and I will just make a few more points and end it.

When cells are exposed to interferon for prolonged periods several changes are noted in addition to their diminished capacity to make interferon when stimulated to do so. The antiviral action of interferon depends on the attachment of interferon to a specific cell receptor. The number of interferon receptors is reduced in cells taken from patients with AIDS, most probably as a result of exposure to endogenous interferon, and this may partly explain the diminished antiviral effect of interferon in advanced disease. This finding also has implications about possibly diminished effects of added interferon

From the point of view of interferon’s antiviral action only, it might seem advantageous for interferon to always be present. But there are active mechanisms to turn off its production, usually after a matter of days, which supports the view that prolonged exposure to interferon can be detrimental. Its many actions – other than its antiviral action are in fact deleterious. Apart from untreated HIV disease, lupus, an autoimmune disease is also associated with the sustained production of interferon α. There are studies in this disease on the mechanisms that sustain interferon production that may also have relevance to HIV disease. In HIV disease, it may of course be the persistence of HIV, but the opportunity to do study this has probably been lost as antiretroviral treatments remove interferon from the circulation.

When the question was asked, why add more interferon to people who already had lots of it, the answer was that the interferon already in the patient, (the endogenous interferon), was different to that to be injected. The basis for this claim of difference was that endogenous alpha interferon was unstable in acid, unlike conventional interferon.

But endogenous alpha interferon did everything that conventional interferon did – most importantly it had the same antiviral properties. Further, there was evidence that the acid instability was not an intrinsic property of the interferon molecule[6].

The neglect in pursuing a possible role in pathogenesis of high levels of circulating interferon was connected with a desire to use it to treat people with AIDS. This was a strange initiative. Apart from Kaposi’s sarcoma and hepatitis C it helped nobody in the long term and subjected people to extremely unpleasant side effects. Considering what interferon can do, one must wonder what effect it might have had on disease progression in the longer term.

Here is an extract from a transcript of a meeting in New York City that Dr Fauci attended to answer questions. This is the response to a question about administering interferon to people who already had more than enough of it in their circulation:


No. I think that acid-labile alpha interferon is an abnormal form of alpha interferon that really doesn’t have the same effects as the kind of interferon we’ve been infusing. It’s almost as though it’s two different drugs. It’s very confusing, because that’s been in the literature and in the paper a lot. It really is different. It’s different. It isn’t the same. There are some similarities, obviously, because it’s the same type of species of an agent, but there are some differences. Whether or not it’s doing harm or good, we don’t know, because there’s so many other things going on ………………………


You said there are differences, and then you went on. But I didn’t hear what the differences were between the two.


Yeah. In vitro effects. Joe, you look like you had a question about that.


I’m not aware that there are any biological differences between acid-labile interferon and conventional interferon. Acid-labile interferon is neutralized by antibodies to conventional interferon. There’s been a recent report that, as you know, in lupus, a similar interferon appears, and there’s quite some conjecture that indeed it may play a role in pathogenesis. More recently, from Jan Vilcek’s lab, there’s been a demonstration that the acid lability may be due to another protein that sticks to it. If that’s so in lupus, my guess is that there’s no reason to think it’s different in AIDS. As far as I know the biological properties of acid- labile interferon are identical to those of –


Yeah, well –

The other argument for using treatment with interferon was that cells from AIDS patients could not make interferon. As noted, the problem with this justification is that the people from whom these cells were taken were full of interferon, which had to come from somewhere. So if the cells taken from the patients were unable to be stimulated to make interferon, other cells are actually overproducing it.

There are still some attempts to treat HIV infection (as opposed to hepatitis C and Kaposi’s sarcoma) with interferon. It is possible that a place may eventually be found for its use, but this would almost surely be on temporary basis and in those who do not already have interferon in their circulation.

The presentation made in 1991 from which the figures in this post were taken can be seen by clicking HERE; there are a few contemporary annotations.

[1] “For, like the character of Dr Jekyll and Mr Hyde, interferon , while possessing great virtues, has a dark and sinister side” Susan Krown, in “Interferon 7” 1986 Academic press p 185-211

[2] DeStefano-E, Friedman-R-M, Friedman-Kien-A-E, Goedert-J-J, Henriksen-D, Preble-O-T, Sonnabend-J-A, Vilcek-J

Acid-labile human leukocyte interferon in homosexual men with Kaposi’s sarcoma and lymphadenopathy. The Journal of infectious diseases, {J-Infect-Dis}, Oct 1982, vol. 146, no. 4, p. 451-9, ISSN: 0022-1899.


Some immunologic parameters in homosexual patients with Kaposi’s sarcoma (KS) or unexplained lymphadenopathy resemble findings in patients with autoimmune diseases such as systemic lupus erythematosus (SLE). Many patients with SLE have an unusual acid-labile form of human leukocyte interferon (HuIFN-alpha) in their serum. Sera from 91 homosexual men were tested for the presence of HuIFN. Of 27 patients with KS, 17 had significant titers of HuIFN in their serum. Ten of 35 patients with lymphadenopathy and three of four patients with other clinical symptoms also had circulating HuIFN. In contrast, only two of 25 apparently healthy subjects had serum HuIFN. All 32 samples of Hu IFN had antiviral activity on bovine cells, a characteristic of HuIFN-alpha, and all of 14 representative samples tested were neutralized by antibody to HuIFN-alpha. In addition, the HuIFN-alpha in six of eight representative patients was inactivated at pH 2 and therefore appears to be similar to the HuIFN-alpha found in patients with SLE. These findings suggest that an autoimmune disorder may underly lymphadenopathy and KS in homosexual men.

[3] Marroni M., Gresele P., Landonio A. et al. Interferon-alpha is effective in the treatment of HIV-1-related, severe, zidovudine resistant thrombocytopenia. A prospective, placebo-controlled, double-blind trial. Ann Intern Med 1994; 121(6): 423–429.

Skillman D. R., Malone J. L., Decker C. F. et al. Phase I trial of interferon alfa-n3 in early-stage human immunodeficiency virus type 1 disease: evidence for drug safety, tolerance, and antiviral activity. J Infect Dis 1996; 173(5): 1107–1114.

Rivero J., Fraga M., Cancio I., Cuervo J., Lopez-Saura P. Long term treatment with recombinant interferon alpha-2b prolongs survival of asymptomatic HIV-infected individuals. Biotherapy 1997; 10(2): 107–113.

Mauss S., Klinker H., Ulmer A., et al. Response to treatment of chronic hepatitis C with interferon alpha in patients with HIV-1 is associated with higher CD4+ cell count. Infection 1998; 26(1):16–19.

Yabrov,A. It is hazardous to treat HIV patients with interferon-a

Medical Hypotheses (2000) 54(1), 131–136

[5] Selective effects of alpha interferon on human T-lymphocyte subsets during mixed lymphocyte cultures.

Scandinavian journal of immunology, {Scand-J-Immunol}, Jun 1983, vol. 17, no. 6, p. 559-67, ISSN: 0300-9475.

Hokland-M, Hokland-P, Heron-I, Schlossman-S-F.


Mixed lymphocyte reaction (MLR) cultures of human lymphocyte subsets with or without the addition of physiological doses of human alpha interferon (IFN-alpha) were compared with respect to surface marker phenotypes and proliferative capacities of the responder cells. A selective depression on the T4 (inducer) T-cell subset could be demonstrated as a sequence of events: decreased fluorescence intensity of the T4 inducer cells (day 2 of culture), decreased percentages of T4 cells as demonstrated by cell cytofluorometry (days 3-6 of culture) , and decreased 3H-thymidine incorporation of purified T4 cells and decreased numbers of T4 cells harvested from IFN MLRs (days 5-6 of culture). In contrast, it was shown that the T8 (cytotoxic/suppressor) subset in MLRs was either not affected or slightly stimulated by the addition of IFN. The depression of the T4 cells by IFN was accompanied by a decrease in the number of activated T cells expressing Ia antigens. On the other hand, IFN MLRs contained greater numbers of cells expressing the T10 differentiation antigen. In experiments with purified T-cell subsets the IFN effect was exerted directly on the T4 cells and not mediated by either T8 suppressor cells or monocytes. These findings are discussed in relation to other immunoregulatory effects of IFN-alpha.

[6] Endogenous “acid labile” interferon is neutralized by monoclonal antibodies against conventional “acid stable” interferon. The amount of interferon in a preparation was measured by observing how much the interferon containing sample could be diluted before it lost antiviral activity. Samples from patients contained antiviral molecules other than alpha interferon, and those that synergized with alpha interferon to increase its effects. Gamma interferon and TNF would be examples. These are destroyed by acid – not the interferon.