now.jpg (1640 bytes)

Virus Activity in the Lymph Tissue; Reservoirs of virus: provirus DNA and Memory CD4 lymphocytes

written by Jules Levin

There is an abundance of data supporting that potent antiretoviral therapy can lower plasma viral load to <50 copies/ml. But, there are reservoirs for the virus in lymph tissue, the brain, cerebral spinal fluid, and other so-called "compartments" of the body. It has been estimated that 99% of the total pool of virus in an individual's body is in the lymph tissue, at least subsequent to seroconversion or primary infection. Research conducted to evaluate viral activity in lymph tissue has been ongoing. Reports were presented at the 3rd International Congress on Drug Therapy in HIV Infection at Birmingham UK in November 1996, followed by reports at the 4th Retroviral Conference in January 1997.

In the current issue of Antiviral Therapy, Dr. Giuseppe Pantaleo of the University of Lausanne-Switzerland, discusses HIV infection in the lymph tissue during the different phases of disease progression. He said it is crucial that therapeutic regimens inhibit viral replication in both the blood and the lymph tissue. Winston Cavert, MD and co-investigators said in the May 8 1997 issue of Science (1), "ultimately, the impact of treatment can only be assessed completely in the lymphoid tissue where most of the virus is produced by CD4 cells, macrophages and other lymphoid mononuclear cells (MNCs) and is stored in immune complexes on the surface of follicular dendritic cells (FDCs)."

Pantaleo went on to say, as is widely recognized, that there appears to be three general phases of HIV progression for infected individuals: primary infection, a long period of clinical latency or chronic infection, and AIDS, which may be characterized by immuno-suppression with a susceptibility to opportunistic infections and other illnesses.

During the primary infection phase, HIV is widely disseminated in the lymph tissue, where the lymph tissue serves as a host for initial viral replication, dissemination to other parts of the body, and in due course, the establishment of chronic infection (2, 3). Over the course of the three phases of progression the location of virus in the lymphoid tissue changes. During primary HIV infection there are a greater number of virus expressing cells (MNCs) consisting mostly of CD4s, and the immune complexes stored in the FDCs are hardly detected.

During chronic infection, in the so-called asymptomatic stage of infection, there is a large amount of virus in these immune complexes trapped in the FDCs, but the individual virus expressing cells are also present. However, minimal changes are detected in the amount of MNCs containing HIV proviral DNA during the transition from primary infection to chronic infection. This indicates a pool of latently-infected cells is generated during early infection.

Proviral DNA. Recently published research by Dr. Robert Siliciano (9), of the NIH, suggests there is a latent pool of virus with a slow decay rate (an estimated half-life of 5-7 months). In his study only half of the 14 study participants were receiving antiretroviral therapy. Ongoing research will address the effect of treatment on these cells. An implication of such an additional latent reservoir for virus, which Siliciano calls integrated proviral DNA in resting CD4 t-cells, is that a possible time frame for the potential for eradication might be prolonged by a number of years. This information, if confirmed, casts doubt on the potential for eradication. But, a few researchers think the proviral DNA that Siliciano is talking about may be non-infectious. Siliciano's research finds these cells have a much longer decay rate than both productively infected cells and other latently infected cells identified by Ho and others. An additional object of follow-up research should be to try and understand how long these cells may last.

Memory CD4 lymphocutes. Recently at the US AIDS Conference in Miami, Florida, Dr. David Ho discussed this reservoir of virus. His laboratory at the Aaron Diamond AIDS Research Center in New York City initiated several studies over the past 2 years to explore how far down they could knock the virus and the potential for eradication of HIV, by treating early in disease stage with a potent antiretroviral regimen. Studies were initiated in individuals during both primary infection and chronic infection. In Miami, he said that after 18-26 months of treatment, just about 99% of the lymph tissue sections they've examined were negative for lymphocytes with HIV RNA. The virus has largely been eliminated but there remains a tiny bit of virus. He said, we are beginning to understand what type of cells they are and believes they are CD4 lymphocytes that are quiescent. They are sitting around carrying HIV. A fraction of them, for reasons we don't understand, will turn on to release HIV every day. But, when a patient is on potent therapy they cannot spread. The cell then dies. But, we need to worry about cells that remain quiet and might turn on - a day, month or year later. His lab is trying to understand the lifespan and decay characteristics of such cells so they can understand how long treatment might have to be administered before the cells might die out. This won't be easy because memory lymphocytes generally live for a long time.

Recognizing that this cell population is another obstacle, they are trying to devise an approach with stimulants or cytokines to activate these memory t-cells, that are in the resting state, while drugs are being administered. Since the amount of these cells appears to be small, Ho suggested it is possible that the body's immune system could cope with them but that theory has not yet been tested. The preferable strategy would be to induce better immune responses in individuals. They are now in the process of planning to immunize their study subjects with viral proteins or viral constructs that will essentially act as vaccines to see if they could help protect these individuals.

Back to Pantaleo. Pantaleo continued, as HIV progresses into the more advanced or AIDS defining phase, the amount of virus containing immune complexes trapped in the FDCs is dramatically reduced, and there is a progressive increase in the number of individual virus expressing cells. Cavert believes this occurs only in very late AIDS. During this phase, the architecture of the lymphoid tissue is progressively disrupted and destroyed, and the mechanism of virus trapping becomes less efficient (3, 8). The ability of the immune system to contain HIV replication which is latent during the chronic phase is severely impaired. The ability of the immune system is similarly impaired from containing other causes for disease, such as bacteria, micro-organisms, etc.

Cavert is unsure that the loss of FDCs and the consequent trapping of virus they accomplish plays a great role in increasing plasma viral load. However, Panaleo says the outcome of the destruction of the lymphoid tissue is severe immunosuppression.

At the recent IDSA conference in San Francisco ending September 16, 1997, Dr. Mark Feinberg reported promising but preliminary data from a small study where individuals were treated with a 4-drug regimen consisting of indinavir+nevirapine+two nucleosides. One of the participants had 7 CD4s and 2 million copies/ml of viral load at baseline. Her plasma viral load was rendered undetectable and it appears as if the architecture in her lymph tissue (LT) was being restored. This gives encouragement for the potential restoration of the immune system. A more detailed report of this and other IDSA reports will be available soon on the NATAP web site.

A review of several reports on the effect of a potent protease inhibitor 3-drug regimen on LT viral load, which were presented at the 4th Retrovirus Conference in Jan 1997, is available in NATAP Reports Feb '97 Special Issue coverage of the Conference. The issue is available on our web site for your referral (see the NATAP Reports section). Also available on the web site is a review of data presented by Marianne Harris in November '96 of the effect of nevirapine/AZT/ddI in the Incas study on virus activity in the lymph tissue. The nevirapine report can be found in the section called Conference Reports, Birmingham Report. Essentially, all these studies report that when there was virus suppression to below "detection" in plasma, viral activity in lymph tissue is well suppressed. Residual virus is found in the lymph tissue but the individuals were only studied for at most 6 months. Follow-up research is required to see if what happens to the residual virus.

As a side note, when therapy fails or when a person stops therapy that is being effective (undetectable plasma viral load) and plasma viral load rebounds, virus will probably re-populate the lymph tissue; and the progress made in possibly restoring the immune system will be back to square one.

The article in Science discusses the preliminary findings by Cavert and others of the treatment effect of a potent protease inhibitor (ritonavir) plus two nucleosides (AZT+3TC) on the viral load in lymph tissue which were first presented by Daan Notermans at the Birmingham conference in November 1996. NATAP's report of this presentation, as well as the report on this study at the 4th Retroviral Conference, are also available on the NATAP web site as well as in our February '97 issue of NATAP Reports covering the Retrovirus Conference.

Cavert reports that previously, it's been reported that lymph tissue (LT) virus pools are little affected by monotherapy with nucleoside analogue drugs and only modestly affected by 2 or 3 RTIs. (4, 5, 6, 7). In this study, 34 previously untreated HIV-infected individuals with 50 CD4 and 30,000 HIV RNA copies/ml were randomized to receive ritonavir+AZT/3TC in an open-label study. Some individuals received the triple regimen from the start of the study. The remaining individuals received ritonavir initially followed 3 weeks later with AZT/3TC to test the hypothesis that initial reductions in viral replication would forestall the development of drug resistance against RTIs. After 6 months of treatment plasma viral load decreased at least 2.9 log in both groups from a median baseline of 5.3 log (about 200,000 copies/ml).

Tonsil LT biopsies from 10 individuals were taken 2 weeks before and 2 days, 22 days and 24 weeks after treatment started. The time points were specifically chosen to detect anticipated early rapid changes in the MNC compartment and the more gradual changes in the FDC-associated pool. The later time points were chosen based on previous work describing the prolonged retention of the virus-antibody immune complexes by the FDCs, the apparent stability of the FDC-HIV association, and the large pool size of the FDC associated HIV reservoir. From these previous observations, the authors expected a much slower decay of the FDC associated virus pool.

At baseline the amount of copies of viral RNA per gram on FDCs was equal to the concentration that may be found at an earlier stage of disease (4). The patients in the study were a little more advanced than on average, but Cavert says the total body viral load probably does not change greatly in persons who do not receive treatment after one gets beyond primary infection until late-stage AIDS. Essentially, at baseline there were a higher number of cells which produced higher levels of virus replication. It is these higher level virus producing cells that may be responsible for increasing viral production in plasma and for increasing the size of the virus pool in the FDC-associated cells compartment. So control of these cells may be more important than initially being able to control the cells chronically infected with low levels of copy number. However, continued follow-up will evaluate the treatment effect on these low level copy number cells.

After 2 days a rapid drop was observed in the frequency of MNCs with the highest concentration of viral RNA, those with >75 copies of HIV RNA per cell. Cells actively producing virus were in less frequency. By day 2 more than 75% of the infected MNCs (mostly consisting of CD4s) contained only 20-60 copies of HIV RNA. This represented a considerable reduction in the amount of virus being produced into the system. Most virus production comes from these cells. Cells with less copies may be chronically infected but don't appear to contribute as much to daily viral replication in plasma.

From days 2 to 22, both the MNC and FDC pools continue to decay in parallel, but the FDC pool decayed at a slower rate. In the third week many MNCs were found with >20 and <75 copies of HIV RNA.

Because these cells disappear so quickly the authors say it affirms their previous proposal that this population of cells (actively producing virus) are in the late stages of viral life cycle. This population with the highest concentration of virus turns over more quickly as a result of treatment.

Between 3 and 24 weeks of treatment, the number of MNCs with >20 copies RNA per cell fell to undetectable amounts in all 10 individuals (mean decrease in cells per gram >2.3 log units), as did the amount of FDC associated HIV RNA in most individuals (mean decrease in RNA copies per gram >3.4 log units. By conducting more sensitive testing, they discovered residual virus in FDCs in 6/10 individuals; and, in 5 individuals they observed cells with low levels of MNCs, that is with fewer than 10 copies of RNA per cell. At week 24, in one patient they could not detect any residue RNA in either cellular compartment down to a sensitivity of <300 cells per gram for cells with at least one copy of RNA per cell, but viral DNA was found in this patient.

In all 6 for whom frozen biopsy specimens were still available HIV DNA was detectable. Their assay did not distinguish between integrated or unitegrated proviral DNA

In summary, the investigators concluded "after 6 months of treatment LT still harbor residual infected cells, and in some of these cells there is evidence of low levels of viral gene expression. The lifespan and functioning of HIV disease in MNCs in which we can detect residue virus requires further investigation. But, as long as there are latently infected cells (proviral DNA) there is reason to continue treatment."

Despite the residual virus, the number of copies of HIV RNA per gram from LT was 3.4 log units over just 6 months of therapy. If continued at the same rate of decay seen between 3 weeks and 6 months, if integrated proviral DNA is not a factor, and if there aren't other unforeseen factors (such as increased or decreased pace of decay), viral RNA would be eliminated within an average of 30 months of triple therapy.

The investigators say, further research will be necessary to ascertain if it's possible to purge completely HIV from LT, and to ascertain the effect of integrated proviral DNA. However, within 6 months, >99% of the actively producing viral lymphoid cells responsible for immune depletion have been eliminated.


  1. Kinetics of Response in Lymphoid Tissue to Antiretroviral Therapy of HIV-1 Infection (Science, Vol 276, 9 May 1997, Cavert et al
  2. Fauci et al, Ann of Int Med 1996; 124: 654-663
  3. Pantaleo & Fauce, Ann Rev Immun 1995; 13:487-512
  4. AT Haase et al, Science, 274, 985, 1996
  5. OJ Cohen et al, Jnl Infect Dis, 173, 849, 1996
  6. OJ Cohen et al, Proc Natl Acad Sci USA, 92, 6017, 1995
  7. A Lafouillade et al, Jnl Infect Dis, 174, 404, 1996
  8. Embretson et al, Nature 1993, 362:359-362
  9. Chun et al, Nature, Vol 387, 8 May 1997