[Deep lung--cellular reaction to HIV]

The course of HIV infection is accompanied by a wide individual variability. The complex and large interplay between host and viral factors is crucial in the disease's evolution. The lung has been recognised from the beginning of the disease as one of the main targets of infectious and non-infectious complications of AIDS. In this setting both anatomic and immunologic particularities of this organ play an important role. The hallmark of HIV is progressive immune dysfunction. Despite the intensive research into the pathogenesis, several questions remain to be answered on the dynamic effects of HIV on pulmonary cells. Previous studies in which we have participated showed the early presence of lymphocytic alveolitis from the asymptomatic phase of infection. Since then, many collected data has brought new insights into the immune and biochemical mechanisms involving HIV cell entry, as well as target cells, cytokines and other cellular mediators. In this context, the discovery that specific chemokine receptors could act as co-receptors for HIV, allowed a better understanding of the mechanisms underlying viral cellular entry and tropism. On this issue several authors have reported that in addition to the CD4 molecule, most strains of HIV use the chemokine receptor CCR5 for viral attachment and entry into the host cells. This receptor seems to be very important in disease transmission, whereas CXCR4 receptor tends to be used by the viral strains that emerge later in the disease in addition to or instead of the CCR5.

To evaluate the pulmonary cellular dynamics in AIDS patients regarding the viral load in bronchoalveolar lavage fluid (LLBA), as well as cellularity and tropism through CCR5 and CXCR4 receptors.

14 AIDS patients were enrolled in this study, with a mean age of 39 +/- 14.3 years (9 males and 5 females) all HIV1, heterosexuals, 6 smokers and 8 non-smokers, none of them drug addicts. These patients were referred to bronchoscopy with BAL, for clinical suspicion of opportunistic lung infections. These patients were later divided into two groups: Group I (recent diagnosis) and Group II (non-recent diagnosis). While all patients had AIDS, group I had also recent diagnosis of opportunistic infections and had not received yet anti-retroviral therapy whilst group II had a long-term disease evolution with several opportunistic episodes and anti-retroviral therapeutic.

BAL was performed both in the middle bronchus in diffuse or in other segmentar bronchus, depending on radiographic abnormalities. Plasma viral load was performed through PCR-RT in blood samples with EDTA, centrifuged and frozen (-80 masculine Celsius) in the first 4 hours after being collected. The viral load in BALf was quantified in 9 patients using the automatized Cobas Ampliprep/Cobas Amplicor HIV1 Monitor TM Test, version 1.5 Roche Diagnostic Systems. The results were expressed in a numeric scale, with a dynamic variation of 50-750.000 copies of RNA HIV1/cm3 and later converted into a logarithmic scale. In 10 patients an immunological study was carried out in BALf and blood to quantify the lymphocyte populations and subsets (CD3, CD4, CD8, CD19, CD56 and CD56CD8) as well as the receptors CD3CCR5, CD4CCR5, CD8CCR5, CCR5Mø, CXCR4, CD3CXCR4, CXCR4CD14 and co-stimulatory molecule CD28, CD3CD28, CD4CD28, CD8CD28 through monoclonal antibodies - CD8FITC, CD19FITC, CD3PE, CD56PE, CD4PECY5-Lymphogram Cytognos; CCR5PE, CXCRFITC-R & D Systems; CD8Cy5 and CD3Cy5-DaKo, CD4PE, CD14PE, CD28FITC- Immunotech; CD4FITC-CLB, CD8Percp- Beckton Dickinson and CD3 APC - Beckton Dickinson, by flow cytometry (Facs Calibur-Beckton-Dickinson) with 3 or 4 fluorescences - FL1- -FITC, FL2-PE, FL3-PECY, FL4-APC. In the statistical analysis, we used the Student t-test, and li- near correlation.

Presence of HIV1 in BALf (2.95 log +/- 3.08 log), in small levels compared with plasma viral loads (5.89 log +/-5.90 log) (Table IV). There was great variability of viral loads in BALf as there was in blood independent of the time elapsed between diagnosis and the exam. As for the lymphocytic populations and subsets in blood (Table V) determined in 13 patients, there was a significant fall of total lymphocytes as well as of their subsets, although more marked in CD4 cells; 42.9% had CD4 levels < 50 cels/mm3 and only 2 patients (n masculine 12, 13) had CD4> 250 cels/mm3. The CD19 was reduced with an individual distribution similar to the CD4 subset. In most cases, the fall of CD8 accompanied the decrease of CD4 and CD19 (patients-n masculine 7 and 8). The lymphocyte populations and subsets in BALf (10 patients) (Table VI) showed a percentual distribution similar to that observed in blood (Table VII) for CD3, CD19, CD4 and CD8 lymphocytes, although the percentage of T cells was higher than in blood (94.5 +/- 5 /84.1 +/-10.4) as opposed to B cells (2.2 +/-3 /10.4 +/- 9.6). In BALf CD8 T cells were higher than in blood (77.7 +/- 17.6 /67.6 +/- 4.2), which was not observed for the CD4 lymphocytes (8.1 +/- 9.5 BALf vs.10.4 +/- 9.6 in blood). The natural killer activity expressed by CD56 T cells had important individual variations in both biological fluids: higher levels in blood than in BALf (9.1 +/-8 /2.9 +/-1.9). The cytotoxic activity of CD56CD8 was similar in blood and in BALf (2.2 +/- 2 / 1.7+/- 1.2) while the individual distribution seemed more homogeneous in BALf (Table VI) than in blood (Table VII). The double-negative (DN) cells had slightly higher values in BALf (7.6 +/- 4.5 vs 5.6 +/- 5.3). Curiously, in BALf we observed a higher percentage of less differentiated cells (13 +/- 13.6) (Table VI). The analysis of the receptors CCR5 and CXCR4 showed in general terms different behaviour concerning the two biological means (Tables VI and VII). Thus, the CCR5 CD3 was higher in blood (10.9+/- 13.2) than in BALf (8.4 +/- +/- 3.5) while the CCR5 CD4 and CCR5 CD8 had an increased expression in BALf in relation to blood ( 2 +/- 2.3 and 4.9 +/- 3.7 / 0.9 +/- 0.7 and 4.1 +/- 4.0 respectively). Concerning the expression of this receptor on monocyte macrophage lineage a marked higher value was attained in BALft (77.8 +/- +/- 41 in BALf vs. 18.7 +/- 15 in blood). On the contrary the total expression of CXCR4 was higher in BALf (31 +/- 19.9) than in blood (16.4 +/- +/- 8.1). This tendency extended equally to the T lymphocytes (26.6 +/- 19.8 vs. 10.7 +/-7.6) and also to the monocyte-macrophage lineage in an exuberant manner (84.5 +/- 30.2 / 4.8 +/- 4.6). The co- stimulatory activity of CD28 showed higher values in blood (22.8 +/- 16.2) than in BALf (15.9 +/- +/- 10.1) for total T cells, CD4 and CD8 lymphocytes 22.5 +/-16.7; 7.8 +/- 8.3; 13.3 +/- 8.3 / 16.5 +/- +/- 10.5; 2.9 +/- 2.8; 10.8 +/- 8.0 respectively).

1. HIV infection is responsible for important and extensive abnormalities in lung host defences. 2. The complex interaction between host and aggressor as well as the immune response particularly represented by natural killer and cytotoxic activities, apoptosis, and opportunistic diseases or others, therapeutics and other factors may contribute to the difficulty in obtaining homogenous medical samples within research. There are also ethical issues that restrict a purely scientific approach to these patients. 3. These results point to a pulmonary response to HIV in a compartmentalised fashion according to the dynamic cellular elements involved and receptors in which the latter had distinct profiles related to the biological fluids. In this context, the lung compartimental response is particularly dependent on alveolar macrophages activity which is from the beginning the cornerstone of this process and is the last cellular defense mechanism in this territory when all others are profoundly affected. 4. The dynamics of chemokines receptors may be very important in therapeutic approach as the blockage of the CCR5 receptor does not seem to trigger an increased expression of CXCR4 strains. In fact, we found that CXCR4 remained high in monocyte-macrophage cells throughout infection and its expression was increased in T-lymphocytes in Group II patients as opposed to CCR5 behavior in BALf which significantly decreases. However, in blood, CCR5 expression increased, unlike CXCR4. 5. Due to high co-existing opportunistic infections (71.4%) we cannot ignore the hypothesis that this increased expression of CXCR4 was a result of the modulation induced by opportunistic agents. 6. Finally, this striking individual variability undoubtly has clinical implications. This makes a case-by-case management strategy the correct approach.