Lymphocytes proliferate in blood and lymph nodes following interleukin-2 therapy in addition to highly active antiretroviral therapy

Background: Substantial redistribution of lymphocytes occurs upon the initiation of highly active antiretroviral therapy (HAART) and immune-based HIV therapies.

Objective: To evaluate the relative contribution of apoptosis and proliferation to changes in lymphocyte populations in peripheral blood and lymph node resulting from interleukin-2 (IL-2) therapy in patients receiving stable HAART.

Cited Here...: Lymphocyte apoptosis was analyzed on various subtypes using fluorescence activated cell sorting with an annexin-V antibody in peripheral blood and by the TUNEL (terminal uridine nucleotide end labelling) method in corresponding lymph node sections. Lymphocyte proliferation was evaluated using an antibody against the cell cycle-associated marker Ki-67 (MIB-1) in peripheral blood and lymph nodes.

Cited Here...: A transient increase in apoptosis was seen in peripheral blood and lymph nodes during a cycle of subcutaneous IL-2. A pronounced proliferative effect of IL-2 (from 6.4% of total lymphocytes in patients only treated with HAART to 23.4% in those treated with HAART + IL-2) was detected in peripheral blood, affecting the CD4, CD8 and CD16/56 subsets to a similar extent. Remarkably, the proliferative effect also occurred in lymphoid tissues. While the lymph node structure gradually disintegrated over 24 months in some individuals, the amount of proliferating lymphocytes, including CD4 cells, B cells and follicular dendritic cells, greatly increased upon IL-2, while HIV RNA load in lymph nodes remained unaffected.

Conclusion: These results show that IL-2 leads to lymphocyte proliferation in peripheral blood and lymph nodes without an impact on viral load in lymphoid tissue. These results have important implications for attempts to reconstitute the immune system in HIV disease.

Introduction

Interleukin-2 (IL-2), an important T cell cytokine, has been evaluated together with highly active antiretroviral therapy (HAART) for the treatment of HIV infection in several randomized, controlled studies, which showed sustained CD4 lymphocyte elevations and stable HIV RNA levels [^1-3]. In addition, IL-2 in conjunction with HAART has been shown to reduce the pool of resting CD4 lymphocytes containing replication-competent HIV [⁴]. Whether this elevation of CD4 cells allows immune restoration and translates into clinical benefit is currently being studied in several international trials.

A number of investigators have reported a correlation of apoptosis in peripheral blood mononuclear cell (PBMC) with plasma HIV load [^5-8], while the lymphoid compartment has only recently been analysed [^9-11]. However, data on the proliferation and apoptosis of peripheral blood lymphocytes in HIV-positive patients following HAART with and without IL-2 are sparse. Recently, IL-2 has been found to reduce lymphocyte apoptosis in vitro [¹²], but in vivo data including studies on lymph nodes are lacking. Therefore, the present study evaluates the relative contribution of apoptosis and proliferation to lymphocyte changes in concomitant samples of peripheral blood and lymph node sections upon the addition of IL-2 to HAART.

Materials and methods

Population

A group of 25 HIV-1-infected patients (24 men and one woman) agreed to have lymph node biopsies performed. At the time of surgical excision a concomitant blood sample was also obtained. Five patients consented to sequential lymph node extirpations. All patients were receiving long-term stable HAART (> 6 months).

Interleukin-2 administration

Patients who were receiving 6 months or more of stable HAART (saquinavir hard gel capsules, zidovudine and didanosine) received recombinant human IL-2 (9 × 10⁶ IU/day for 5 days) every 6-8 weeks as subcutaneous injections [²]. To assess proliferation and apoptosis in peripheral blood lymphocytes in 13 patients, blood samples were taken prior to (day -1 and day 0) and on days 3, 7, 14 of the third cycle of IL-2 treatment. Lymph node samples were obtained prior to, at 6 months (i.e, after three cycles of IL-2) and, if possible, at 24 months (13 cycles) after the start of adjunctive IL-2 therapy in a subset (n = 5) of patients. IL-2 therapy continued for a minimum of 2 years in cycles of therapy approximately every 8 weeks, with individual dosages after that time.

Sampling

Lymph nodes were biopsied surgically in the axillary region (10 patients) or in the inguinal region (15 patients). Biopsies were performed 6 or 7 weeks after completion of the third cycle of IL-2. In five subjects, subsequent excisional biopsies were performed at the same anatomical site at 0, 6 and 24 months. Lymph nodes were bisected directly after extirpation and used for viral load analysis and immunohistochemistry experiments. Flow cytometric analysis of peripheral blood cells was performed at the same time for the assessment of lymphocyte apoptosis and proliferation.

Immunohistochemistry

Tissues were fixed overnight in neutral-buffered formalin, embedded in paraffin and stained with haematoxylin and eosin in routine histology. As controls, five lymph nodes from patients with follicular hyperplasia that was not related to HIV infection were selected from our repository and treated similarly. For immunohistochemistry, dewaxed paraffin sections were added to 0.1 mol/l buffered sodium citrate (pH 6), placed in a microwave at 350 W and boiled for 5 min three times before chilling to room temperature. The sections were incubated with a primary antibody (CD3, Coulter, Krefeld, Germany; CD4, Novocastra, Hamburg, Germany; CD8, Dako, Hamburg, Germany; CD45RA, Dako; CD57 (corresponds to CD56 but can also be used in paraffin sections), Becton Dickinson, Heidelberg, Germany; CD20, Coulter; and CD23, Medac, Freiburg, Germany) according to the manufacturer's instructions. Because three-colour immunohistochemical analysis was not possible, only one marker could be analysed in lymph nodes sections in addition to either TUNEL (terminal uridine nucleotide end labelling) or cell cycle-associated marker Ki-67 (MIB-1) staining. Binding of antibodies was visualized by the alkaline phosphatase anti-alkaline phosphatase (APAAP) technique, using New Fuchsin (Dako) as chromogen [¹³].

Analysis of proliferation in lymph nodes

Proliferating cells were visualized in situ using MIB-1, which is exclusively expressed in proliferating cells during the late-G₁, S and G₂ phases [^14,15]. Double-staining of sections prestained with the CD cluster antibodies was performed with the MIB-1-antibody (1:10, Dianova, Hamburg, Germany) [¹⁶]. Bound MIB-1 antibody was visualized with nitroblue tetrazolium/5-bromo-4-chloro-3-inodolyl phosphate (Boehringer Mannheim, Germany) resulting in black nuclear staining.

Analysis of apoptosis in lymph nodes

Formalin-fixed and paraffin-embedded lymph node sections were deparaffinized and incubated with TDT (terminal deoxynucleotidyl transferase) and dUTP-biotin following proteinase K (20 μg/ml) digestion for 15 min and quenching of endogenous peroxidase activity by 2% H₂O₂ for 5 min at room temperature. The biotin-labelled UTPs at the 3′-OH ends were visualized with a peroxidase-conjugated monoclonal antibody against anti-digoxigenin according to the manufacturer's instructions (Oncor, Gaithersburg, Maryland, USA). Upon incubation with 3,3′-diaminobenzidine as substrate, a brown-grey precipitate developed. The TUNEL-stained sections were subjected to cell surface staining with monoclonal antibodies against CD4, CD8, CD20, CD45RA and CD57 using the APAAP method. Finally, sections were counterstained with haematoxylin and eosin. The stained apoptotic cells were counted under a light microscope at × 400 magnification using a measuring ocular eyepiece. Ten randomly selected quadrants of 0.0625 mm² were counted and averaged. Multiplication by 16 gave the average number of apoptotic cells (±SD) in a square millimetre.

Flow cytometry

To measure early apoptosis, fluorescence activated cell sorting (FACS) analysis was performed using the annexin-V method with a monoclonal fluorescein isothiocyanate (FITC)-labelled antibody (Coulter). Samples were stained and analysed for CD3/CD4, CD3/CD8 and CD3/CD56 lymphocytes using the Lysis II software (Becton Dickinson). Blood was anti-coagulated with ethylenediamine tetraacetic acid (EDTA) and mononuclear cells were isolated by Ficoll gradient and centrifugation at 900 ×g for 15 min. To analyze proliferation, PBMC were incubated with the respective surface markers as phycoerythrin conjugates and PC5 (phycoerythrin-cyanin 5.1) to stain CD3 (Coulter) prior to permeabilization (Fix and Perm Permeabilization Kit, An der Grub Bio Research, Germany) and subsequent staining with the FITC-MIB-1 monoclonal antibody. To visualize apoptotic lymphocytes, PBMC were incubated with 10 mg/ml annexin-V-FITC antibody (Coulter) for 30 min at 4°C prior to three washes and FACS analysis.

Quantification of HIV RNA in lymph node samples

Lymph node specimens with fatty tissue removed were thawed and homogenized in 8 mol/l guanidinium hydrochloride (Pierce, Rockford, Illinois, USA) and 0.3 mol/l sodium acetate (Fisher Scientific, Edmonton, Canada) for extraction of HIV RNA [¹⁷]. After the addition of lauroyl sarcosine salt (Sigma, Taufkirchen, Germany) and polyadenylic acid (Sigma), the extracted RNA was selectively precipitated with 0.5× volume of ethanol. Following centrifugation (20 min, 10 000 ×g), the nucleic acid pellet was washed with 70% ethanol and HIV RNA was measured using the branched DNA assay (Quantikine HIV-RNA 2.0, Bayer Diagnostics, Leverkusen, Germany) as previously described. Results are expressed as the number of HIV RNA copies/μg RNA. The sensitivity limit was generally in the order of 50 copies/μg RNA.

Results

Correlation of apoptosis in lymph nodes with levels of circulating CD4 cells

The degree of apoptosis and proliferation of various lymphocyte subtypes was analyzed in lymph nodes of patients undergoing HAART with and without repetitive cycles of IL-2. In 25 concomitant peripheral blood mononuclear cell and lymph node samples from patients taking HAART without IL-2 (Table 1), a correlation between the circulating CD4 cell counts in peripheral blood and TUNEL-positive cells in lymph nodes was found (Fig. 1; r = -0.463). In contrast, there was only a weak correlation of lymph node (r = 0.227) and peripheral blood viral load (r = 0.288) with TUNEL-positive cells in lymphoid tissue (data not shown).

Fig. 1

Table 1

Effect of interleukin-2 on apoptosis

When the rate of apoptosis was evaluated in lymph nodes before and 6 weeks after three cycles of IL-2 in five evaluable patients (Table 1), an increase in the level of apoptosis was seen (Fig. 2). Upon further quantitative analysis, a 2.8-fold increase was detected for CD8 cells (Fig. 2), a 2.1-fold increase for CD4 cells and a 1.8-fold increase for CD57 cells. Apoptotic B cells were also increased to a similar extent (data not shown). As depicted in a representative patient's lymph node section (Fig. 3), an increase in apoptotic CD4 cells can be seen at the 6 and 24 months time points, while the lymph node architecture in this patient showed signs of disintegration.

Fig. 2

Fig. 3

When a cycle of IL-2 was administered to 13 consecutive patients (Table 1), the rate of apoptosis in peripheral lymphocytes increased from 19.7 to 27.0%, as evidenced by triple-colour flow cytometry (Fig. 4a). All analysed lymphocyte subtypes (CD4, CD8 and CD16/56) showed a transient rise in apoptosis irrespective of the patient's CDC (Centers for Disease Control and Prevention) stage, with baseline levels being reached by day 14.

Fig. 4

Effect of interleukin-2 on proliferation

In order to distinguish lymphocyte redistribution from true proliferation induced by concomitant IL-2 therapy, triple-colour staining of peripheral blood lymphocytes and immunostaining of lymph node sections using the proliferation-associated marker MIB-1 was performed. Whereas 6.4% of total peripheral blood lymphocytes were MIB-1 positive in HIV-positive patients with stable HAART (n = 12) at day 0, the addition of therapy with IL-2 resulted in 23.4% of lymphocytes being MIB-1 positive at day 7 (3.7-fold increase;Fig. 4b). Remarkably, the CD4, CD8 and CD16/56 subsets showed a parallel proliferative increase, with baseline levels being reached by day 14. The CD4 cell compartment showed the most dramatic increase, from 9.3 to 39.2% proliferating cells, equalling a 4.2-fold augmentation (P < 0.001). The CD45RA population (containing naive T cells) also increased 3-fold (data not shown). Elevated circulating CD4 cell counts could still be measured at week 8 when the next IL-2 cycle was administered. The expansion of the CD8 cell fraction was 3.8-fold (from 8.5 to 32.6%;P < 0.005), while the CD16/56 fraction (containing natural killer cells) showed the promptest response and a 2.2-fold increase (from 22.2 to 47.9%;P < 0.01).

The pronounced proliferative effect of IL-2 in peripheral blood was also paralleled in lymphoid tissues, where total MIB-1-positive lymphocytes were counted in sequential lymph node specimens (n = 5) obtained before and 6 weeks after three cycles of IL-2. Whereas total lymphocytes increased 1.5-fold in lymph nodes, proliferating CD4 cells increased 1.4-fold (not significant), CD8 cells increased 1.9-fold (P < 0.05) and CD57 cells increased 3-fold (P < 0.005) (Fig. 5). As an example, the marked proliferative capacity of IL-2 in lymph node sections is shown where proliferating cells (black staining) were significantly increased at 6 and 24 months of therapy (Fig. 6). Proliferating CD4 cells can be identified by the red surface and black nuclear staining (Fig. 6).

Fig. 5

Fig. 6

Lymph node architecture

Follicular hyperplasia was prominent in 83% of the extirpated lymph nodes (Fig. 3 and 6). The follicles were diffusely distributed rather than confined to the outer cortex [^13,14]. The germinal centre microenvironment appeared disturbed. For example, there were many infiltrating CD8 T cells within the follicular dendritic cell network (data not shown). While the lymph node structure gradually disintegrated in some of the lymph nodes, the number of proliferating lymphocytes, including CD4 cells (large arrows in Fig. 6), greatly increased, as shown after 6 and 24 months of IL-2 therapy (Fig. 6). Moreover, B cells (Fig. 7) and CD23 follicular dendritic cells (Fig. 8) also showed marked proliferation in germinal centres.

Fig. 7

Fig. 8

Calculation of lymphocyte proliferation in peripheral blood

The augmentation of lymphocytes induced by one cycle of subcutaneous IL-2 (see Methods) was estimated in all 13 patients assuming a blood volume of 5 l and baseline cell counts of lymphocytes 2000 × 10⁶ cells/l, CD4 cells 400 × 10⁶ cells/l, CD8 cells 1116 × 10⁶ cells/l and CD16/56 cells 127 × 10⁶ cells/l. After one cycle of IL-2, the following average blood parameters were detected: lymphocytes 3200 × 10⁶ cells/l, CD4 cells 680 × 10⁶ cells/l, CD8 cells 1841 × 10⁶ cells/l and CD16/56 cells 277 × 10⁶ cells/l. Based on these figures, an increase of 3000 × 10⁶ cells/l total lymphocytes (640 × 10⁶ cells/l on day 0 and 3700 × 10⁶ cells/l on day 7) was calculated. For CD4 cells, a 7-fold net increase equalling 1100 × 10⁶ cells/l (190 × 10⁶ cells/l on day 0 and 1300 × 10⁶ cells/l on day 7) was derived. Since circulating lymphocytes represent only a small proportion of total body lymphocytes, the amount of lymphocytes generated by a single IL-2 cycle is likely much higher.

HIV RNA in lymph nodes during intermittent therapy with interleukin 2

While it is known that IL-2 has no adverse effects on HIV load in peripheral blood [^1-3], the HIV RNA load extracted from lymph nodes of five patients with stable HAART was measured by branched DNA quantification before and after IL-2 therapy. HIV RNA load remained unchanged with 4266 ± 7585 Eq/μg RNA prior to and 2041 ± 1660 Eq/μg RNA after 6 months of IL-2.

Discussion

This study attempted to quantify apoptosis and proliferation in lymph nodes and peripheral blood in HIV-positive patients taking HAART plus IL-2. There was a correlation between circulating CD4 lymphocytes in peripheral blood and TUNEL-positive CD4 cells (and other lymphocyte subtypes) in lymph nodes. It has been shown previously that CD4 cell apoptosis correlated with plasma viral load [^7,8,18] and the activation status, with HLA-DR+CD38+CD45R0+ cells being prone to apoptosis [^6,19,20]. Some investigators found a correlation among cellular death, stage of disease and CD4 cell counts [⁵], while others did not [²¹]. Following the initiation of HAART, there was a rapid reduction in lymphoid tissue apoptosis and susceptibility of peripheral CD4 T cells to Fas-mediated apoptosis, correlating with a significant decrease in viral load and increase in peripheral T cells [²⁰].

When the additional effect of IL-2 on apoptosis was assessed, a moderate increase was shown for the CD4, CD8 and CD16/56 subtypes that persisted for up to 14 days. This in vivo effect contrasted with in vitro results, where IL-2 was shown to reduce apoptosis by blocking the downregulation of bcl-2 [²²]. However, this difference may be explained by the microenvironment in lymph nodes and blood compared with cultured cells, stromal cells being absent in the latter. To balance apoptosis of lymphocytes, a substantial rate of lymphocyte proliferation has to occur, since IL-2 is known to elevate T cell counts substantially [^1-3]. Therefore, proliferation was also assessed in peripheral blood leukocytes, using the proliferation-associated marker MIB-1. Substantial 2- to 4-fold increases in lymphocyte proliferation rates of all analysed subsets were seen when IL-2 was administered. Such an effect has not been described for HAART alone and, therefore, is attributable to the effects of IL-2 [²³]. Tissot et al. showed that patients at various stages of HIV disease treated with HAART showed the same low lymphocyte proliferation rate as observed in HIV-negative controls, while plasma viral load decreased and CD4 cell counts increased upon commencing HAART. Interestingly, the rate of proliferation of lymphocytes was even reported to decrease after the institution of HAART in these patients. Lymphocyte proliferation exclusively increased during the acute phase of opportunistic infection, when patients generally had CD4 cell counts < 100 × 10⁶ cells/l. Whereas the study of Tissot et al. only looked at the peripheral blood compartment, our investigations focused on lymph nodes, where we found elevated levels of lymphocyte production that further increased upon the administration of IL-2. While CD4 cell production has been postulated to be the main mechanism of CD4 cell homeostasis and recovery in HIV-1 infection [²⁴], telomere shortening in these cells has not been observed, as would be expected if cells were actively cycling [²⁵]. Other potential mechanisms involved in CD4 T cell recovery following the institution of HAART include cell mobilization from hidden reservoirs, release of cells sequestered in inflamed tissues and increased CD4 cell survival in the circulation after reduction of the viral load [²⁶]. Indeed, peripheral blood lymphocytes represent a very small part of the total lymphocyte pool, totalling only up to 5% [^27,28]. The observed kinetics of CD4 cell recovery during antiretroviral treatment appears to progress in two phases: a first phase of relatively rapid increase of memory cells followed by a much slower rise of naive cells [^29,30]. In that regard, our results clearly point to marked lymphocyte proliferation in peripheral blood as well as in lymphoid tissues. Therefore, simple redistribution from lymphoid organs or from hidden cell reservoirs is unlikely to represent an important mechanism for lymphocyte homeostasis [⁴]. However, an additional effect of redistribution of lymphoid cells cannot be excluded [³¹]. The observed proliferation rates in peripheral blood are in agreement with earlier studies of Sachsenberg et al., who showed a five times shorter doubling time of T cells in HIV-infected individuals compared with healthy adults [³²]. These studies revealed an inverse correlation between proliferating CD4 cells and the amount of CD4 T cells, but no correlation in the CD8 compartment. The mean increase in the CD4 cell growth fraction in HIV infection was around 6-fold higher compared with that in healthy adults [^11,32].

While lymphocyte proliferation during IL-2 treatment appeared to be similar in all analyzed subtypes, the proliferation kinetics were fastest in CD3+CD56+ natural killer lymphocytes, which would be consistent with the expression of intermediate affinity IL-2 receptors on these cells whereas CD4 and CD8 cells predominantly express the low-affinity IL-2-receptor (CD25) [³³]. The observed proliferation of lymphocyte subtypes in peripheral blood was paralleled by a similar expansion of these cells, including naive T cells, in lymph nodes. Since T cell proliferation alone could not account for the entire MIB-1-positive fraction, CD20 B cells and CD23 follicular dendritic cells were evaluated; these also cycled to a substantial extent. This would be expected since these cells also carry IL-2 receptors [^3,33]. However, our study does not allow conclusions as to whether the proliferation is quantitatively larger in the peripheral blood or the lymphoid compartment.

Generally, follicular hyperplasia is present in HIV-positive patients with generalized lymphadenopathy. It is well known that lymph nodes of HIV-infected individuals represent an extensive reservoir for HIV, with virions being trapped by follicular dendritic cells in germinal centres. In addition, various degrees of follicular dendritic cell damage and a marked elevation in CD8 lymphocytes have been described in both the T-dependent zone and the germinal centres (an unusual site for CD8 T cells) in HIV infection [^13,34].

In summary, the present study measured apoptosis and proliferation in peripheral blood and lymph nodes and showed the non-selective effect of additional IL-2 on various IL-2 receptor-bearing lymphocyte subtypes. Lymphocytes of HIV-infected patients treated with HAART are subject to apoptosis and proliferation in peripheral blood and lymphoid tissue. Concomitant therapy with HAART and IL-2 caused a substantial increase in lymphocyte proliferation in peripheral blood (and to a lesser extent in lymph nodes) that greatly exceeded the observed level of apoptosis; IL-2 did not impact on lymphoid tissue viral load. Such a substantial lymphocyte proliferation has not been described for HAART alone. However, an additional redistribution of lymphocytes induced by IL-2 cannot be excluded. These results have important implications for attempts to achieve immune reconstitution with HAART and concomitant IL-2, which is currently being studied in large phase III trials (SILCAAT and ESPRIT).

Acknowledgements

We would like to thank Ms Andrea Nolte for histological sectioning; Hagen Apel for valuable help with photography and Nicole-C. Bartosch for typing the manuscript and editorial assistance.