B cell non-Hodgkin's lymphoma (NHL) is seen commonly in people who are HIV-infected (HIV+) . While the incidence of AIDS-associated Kaposi's sarcoma has decreased dramatically since the introduction of HAART, the incidence of AIDS-NHL has not decreased to the same extent . There are several subtypes of AIDS-NHL: Burkitt's lymphoma (BL), large cell lymphoma (LCL), immunoblastic plasmacytoid lymphoma (IBL), central nervous system (CNS) lymphoma, and primary effusion lymphoma (PEL). These NHL subtypes differ in their pathology, molecular lesions, histology, and tumor Epstein–Barr virus (EBV) infection status . Two major mechanisms are believed to lead to the development of AIDS-NHL: (i) the loss of immunoregulatory control of EBV-infected B cells in advanced HIV disease, leading to uncontrolled EBV infection and transformation of B cells, and (ii) the HIV infection-associated B cell hyperactivation, which may result in the accumulation of molecular lesions (c-MYC:IgH translocation, BCL-6 overexpression/mutations) in activated B cells, leading to the development of NHL .
B cell activation, as well as infection with and/or exposure to several oncogenic viruses (EBV, hepatitis C virus, HIV), leads to the induction of a cytidine deaminase necessary for IgH class switch recombination (CSR) and somatic hypermutation (SHM): activation induced cytidine deaminase (AID) ([3–7] and Epeldegui et al. unpublished data). AID is an APOBEC-like enzyme that is exclusively expressed in germinal center (GC) B cells in secondary lymphoid tissues, and is required for IgH CSR and SHM . It is clear that AID plays a direct role in the induction of these DNA-modifying processes, and that errors in these processes contribute to the genesis of NHL [3,9]. Interestingly, recent studies indicate that AID is expressed in NHL, but only in those tumors that are of GC origin [9–12]. Therefore, AID over-expression in B cells, resulting from chronic B cell hyperactivation associated with HIV infection  and/or by direct induction with HIV virion ( and Epeldegui et al. unpublished data), has the potential to contribute to lymphomagenesis .
Since AID expression is induced by exposure to oncogenic viruses and/or B cell activation, and is also observed in NHL cells, it is reasonable to think that AID expression may precede the development of NHL in the setting of HIV infection. Therefore, AID expression was quantified in viably frozen peripheral blood mononuclear cells (PBMC), isolated prior to NHL diagnosis from HIV+ subjects in the Los Angeles Multicenter AIDS Cohort Study (MACS). This is the first report showing significantly elevated levels of AID expression in PBMC specimens obtained in the months to years preceding the diagnosis of AIDS-NHL.
Material and methods
All study subjects were participants in the Los Angeles MACS. These subjects were homosexual men who had study visits at 6-month intervals since 1984/1985 to examine the natural history of HIV infection and AIDS. At each visit, detailed histories and blood samples were obtained, and a local repository of serum and PBMC was maintained at the University of California at Los Angeles (UCLA). Clinical information from each visit was obtained from the participants and by review of external registries, and confirmed by medical record review. The protocols and questionnaires used in the MACS (available at www.statepi.jhsph.edu/macs/macs.html) have been approved by the institutional review board at each center.
AIDS-NHL cases (n = 16) were subjects who were HIV-seropositive upon entry into the MACS, had been diagnosed with lymphoma of known subtype prior to 30 June 1994, and had viably-frozen archived PBMC available in the UCLA repository at up to three visits: (i) a visit preceding lymphoma diagnosis by ≤ 1 year (index visit); (ii) a visit > 1 year but < 5 years prior to lymphoma diagnosis; and (iii) a visit ≥ 5 years prior to lymphoma diagnosis. Cases had tumors classified as small, non-cleaved/BL-like (BL, n = 6), diffuse LCL, n = 5), and primary lymphoma of the CNS (n = 5). At the index visit, the median absolute CD4 T cell count for AIDS-NHL cases was 262 cells/μL (range, 6–790 cells/μL).
Two control groups were selected in parallel with the AIDS-NHL subjects: AIDS controls (no NHL) and HIV-uninfected controls. For each AIDS-NHL case, controls were selected from MACS subjects who had PBMC samples available in the UCLA repository within two contemporary MACS visits (approximately ±1 year) of the visit at which the case's PBMC samples were collected, to ensure consistent age and handling of frozen PBMC. AIDS controls (n = 17) met all three of the following criteria: (i) an AIDS diagnosis (according to the 1993 definition of the Centers for Disease Control and Prevention, excluding cases defined by CD4 T cell count alone) ; (ii) no reported malignancy; and (iii) matched by absolute CD4 T cell number (±100cells/μl) to the AIDS-NHL cases at the index visit. At the index visit, the AIDS controls had a median absolute CD4-positive T cell count of 233 cells/μl (range, 19–563 cells/μl). HIV-uninfected controls were also selected from MACS subjects who were HIV-seronegative (HIV-, n = 16). The mean age at the index visit was 41.7, 40.4 and 34.9 years for the pre-AIDS-NHL cases, AIDS controls and HIV-negative controls, respectively.
RNA isolation and quantitative real time RT–PCR
RNA isolation and construction of RNA standard curves were performed as described previously . Real-time PCR measurements were done in duplicate for the samples and in triplicate for the standard curve. AID mRNA levels detected by TaqMan RT–PCR were normalized for total RNA content by dividing the AID copy number by the GAPDH copy number. This AID/GAPDH ratio was then multiplied by 10 000 to provide a relative numerical index of AID expression that is adjusted for RNA content [(AID/GAPDH) × 10 0000]. AID expression was considered detectable when the numerical index was >2.0.
Frozen vials were thawed and PBMC were washed twice in culture media [RPM1, 10% fetal calf serum (FCS), 1% l-glutamine, 1% P/S] and once in phosphate-buffered saline (PBS). PBMC were counted and 3 × 106 cell pellets were prepared. Pellets were disrupted in 350 μl of RLT buffer+1%β-mercapethanol. Disrupted cells were then frozen at −80°C.
B cell isolation
Frozen vials were thawed and PBMC were washed twice in culture media (RPM1, 10% FCS, 1% l-glutamine, 1% P/S) and once in PBS. PBMC were counted and resuspended at 25 × 106 cells/ml in cold PBS containing 0.1% bovine serum albumin (BSA) and 2 mM EDTA. CD19 immunobeads (25 μl; Dynal Biotech, Invitrogen, Carlsbad, California, USA) was added per 25 × 106 cells and incubated at 2–8°C for 20 min on a rotator. CD19+ cells were isolated using a magnet, and washed twice with PBS containing 0.1% BSA and 2 mM EDTA. Beads were eliminated by adding 100 μl of Detachabead (DAB, Dynal Biotech) per 1 × 106cells in culture medium, while incubating for 40 min on a rotator. Following this, detached cells were washed two to three times with culture medium, after which cells were pelleted.
Longitudinal analyses were performed to assess the major research questions of differential AID detectability (binary data) and AID expression level (log-transformed continuous data) in the three study groups using the generalized estimating equations (GEE) method  implemented in the Genmod procedure of SAS, version 9.1 (SAS Institute, Cary, North Carolina, USA). Due to the relatively small sample sizes and distributional features of the expression levels, several alternative analyses were conducted, including mixed effects analysis and permutation test-corrected (i.e., for nesting of studies within subjects) non-parametric procedures. Results (not shown) from these alternative analyses were strongly consistent with the GEE method.
AID expression is elevated in PBMC preceding the development of AIDS-NHL
AID expression was quantified using TaqMan RT–PCR in PBMC collected pre-NHL diagnosis from subjects who developed AIDS-NHL, as well as in PBMC collected around the same times from AIDS and HIV-negative controls. Detectable AID expression at any visit was seen in 50% of the AIDS-NHL subjects (38% of samples), compared to 29% of AIDS controls (19% of samples; P = 0.16), and 6% of HIV− controls (3% of samples; P = 0.01).
Longitudinal analyses of all time points for each subject indicated that the level of AID expression was elevated more than fivefold pre-NHL compared to AIDS controls (P = 0.04), and more than 10-fold compared to HIV-negative controls (P = 0.01) (Fig. 1). For instance, several subjects who went on to develop AIDS-NHL had PBMC AID levels > 1000, while none of the AIDS control PBMC had levels of this magnitude. The difference was even more marked when compared to HIV-negative controls, in whom detectable AID expression was rare, and who never showed AID expression levels > 5 copies.
Remarkably, detectable AID expression was seen in some subjects long before AIDS-NHL diagnosis (Fig. 2). In one case, elevated AID expression was seen in PBMC collected as long as 8.5 years prior to NHL diagnosis, with an AID expression level >10 000 (Fig. 1).
AID expression was higher in subjects who went on to develop AIDS-NHL of the BL subtype, when compared to those who developed other subtypes of AIDS-NHL
Of the 16 AIDS-NHL cases: six developed BL, five LCL, and five CNS lymphoma. When segregated by NHL subtype, marked differences in PBMC AID expression were noted. Of the six subjects who developed BL, five (83%) had detectable PBMC AID expression in at least one visit prior to NHL diagnosis (Fig. 2). In contrast, subjects who went on to develop LCL and CNS lymphoma had detectable AID expression less frequently: three of five (60%) and none of five, respectively.
Also, when NHL subjects were segregated into two groups (BL versus non-BL), it was clear that subjects who went on to develop BL had higher levels of PBMC AID expression than the ones who developed other forms of NHL (Fig. 3). Utilizing longitudinal analyses taking into account multiple time points, the level of AID expression seen in pre-NHL PBMC specimens was significantly higher (approximately 100-fold) in those who developed BL, when compared to those who developed non-BL subtypes (P = 0.002).
From previous work, data were available on these AIDS-NHL and control subjects for serum levels of cytokines associated with immune system and/or B cell activation [16–20]. Cytokine levels were measured at the MACS visit closest to, but preceding, NHL diagnosis, for which serum was available in the UCLA repository. Serum cytokine data were utilized if they were obtained within two visits (1 year) of the visit closest to NHL diagnosis at which the PBMC sample used for AID measurement was collected. Elevated pre-NHL levels of serum interleukin (IL)-6 and sCD23 were seen only in subjects who went on to develop BL, and not in those who developed other AIDS-NHL subtypes, as reported previously . Consistent with the longitudinal analyses, significantly elevated levels of AID expression were observed in PBMC collected within a year to BL diagnosis (P = 0.01), but that was not the case for the subjects who developed other types of lymphoma (non-BL) (P = 0.6), when compared to AIDS controls (Table 1). BL and non-BL subjects included in Table 1 did not differ significantly in CD4 T cell number at the study visit at which cytokines and AID expression was measured, nor in duration of HIV infection prior to AIDS-NHL diagnosis (5.7 and 6.1 years, respectively), making it unlikely that other differences seen were due to HIV disease status/progression.
AID expression, preceding AIDS-NHL, is enriched in highly purified B cells isolated from PBMC
The most likely cells to be expressing AID in PBMC are B cells, as AID is normally expressed exclusively in GC B cells, and is not expressed by most non-B cells. To determine if the cells in pre-NHL PBMC that express AID are B cells, B cells were isolated from PBMC of a subject who went onto develop BL. PBMC were thawed and B cells were purified by positive selection using CD19 beads; 98% of the purified cells were CD19+. AID expression was seen to be 25-fold higher in these CD19+ cells (enriched B cells) than in the PBMC from which were they were purified (data not shown). Given that enriching for B cells clearly enriches AID expression, this suggests that B cells are responsible for the bulk of the AID expression seen in pre-NHL PBMC specimens.
The results presented here show that AID expression occurs more frequently, and at higher levels of expression, in PBMC of HIV+ subjects who went on to develop NHL. In previous work by others, AID expression was detected in NHL tumors [9–11], but this is the first report showing elevated AID expression in circulating cells collected prior to NHL diagnosis. It appears that B cells are the source of the greater part of the elevated AID expression seen in these pre-NHL PBMC. Given what is known about the function of AID in mediating B cell activation-associated DNA modifying events, these AID expressing B cells may be undergoing IgH CSR and SHM, events that normally occur in GC B cells, and not in circulating B cells. Errors in these DNA modifying events could lead to DNA lesions that contribute to the development of NHL.
A marked difference in the levels of AID expression was noted, depending on AIDS-NHL subtype, with significantly higher levels of AID expression seen in those subjects who developed BL. It was previously reported by us that IL-6 and sCD23 were also preferentially elevated in circulation of patients who developed BL ; interestingly both of these molecules also play a role in IgH CSR. Recently it was seen that expression of AID induces not only IgH CSR, but also increases the frequency of the occurrence of the c-myc:Ig translocation that is the hallmark of BL, and is believed to occur due to errors in CSR . Therefore, chronic AID overexpression may be contributing to the development of these seminal MYC:IgH translocations, leading to the formation of malignant clones that later become clinically-apparent BL cancers. This difference in the pattern of AID expression may provide insights on how these different NHL subtypes develop.
It is interesting to note that elevated AID expression was generally not seen to occur in those who developed CNS lymphomas, tumors that are virtually all EBV-positive, and are associated with the loss of EBV-specific cellular immunity, rather than with chronic B cell hyperactivation .
It is not known whether the cells that express AID pre-NHL are premalignant B cells, or simply circulating activated B cells. Given that AID is expressed by NHL of GC origin, and that AID activity is thought to contribute to the lymphomagenic molecular lesions seen in many AIDS-NHL, these AID+ cells may represent the tumor cell clone that eventually develops in these subjects. Alternatively, AID expression in PBMC may reflect the presence of elevated levels of circulating activated B cells, and not circulating tumor cells. Elevated levels of circulating phenotypically-aberrant B cells, displaying markers of cellular activation and immaturity, are seen in HIV infection . Chronic B cell activation occurs in HIV infection, and is known to be especially elevated prior to NHL diagnosis. Ongoing studies aim to determine when, prior to AIDS-NHL diagnosis, tumor-specific B cells are first detected, and to correlate this with relative levels of B cell AID expression.
We would like to thank the participants in the MACS, who have made this and may other studies possible. We would also like to thank Philip Arlen and Jerome Zack (UCLA) for their invaluable assistance in the development of the real time PCR for AID and GAPDH. We wish to acknowledge Najib Aziz for his invaluable help. Lastly we would like to thank Jose D. Ruisanchez for his help and support during this work. This work was supported by grants from the Universitywide AIDS Research Program of the University of California (D04-LA-403 and ID05-LA-047), and by National Institutes of Health (NIH) grants CA57152, CA96888, and CA73475. The Multicenter AIDS Cohort study (AI28697) is funded by the National Institute of Allergy and Infectious Diseases, with additional supplemental funding form the National Cancer Institute and National Heart, Lung and Blood Institute.
1. Bernstein WB, Little RF, Wilson WH, Yarchoan R. Acquired immunodeficiency syndrome-related malignancies in the era of highly active antiretroviral therapy. Int J Hematol 2006; 84:3–11.
2. Gaidano G, Capello D, Carbone A. The molecular basis of acquired immunodeficiency syndrome-related lymphomagenesis. Semin Oncol 2000; 27:431–441.
3. Epeldegui M, Widney DP, Martinez-Maza O. Pathogenesis of AIDS lymphoma: role of oncogenic viruses and B cell activation-associated molecular lesions. Curr Opin Oncol 2006; 18:444–448.
4. He B, Raab-Traub N, Casali P, Cerutti A. EBV-encoded latent membrane protein 1 cooperates with BAFF/BLyS and APRIL to induce T cell-independent Ig heavy chain class switching. J Immunol 2003; 171:5215–5224.
5. Machida K, Cheng KT, Sung VM, Lee KJ, Levine AM, Lai MM, et al
. Hepatatis C virus induced a mutator phenotype: enhanced mutations of immunoglobulin and proto-oncoges. Proc Nalt Acad USA 2004; 101:4262–4267.
6. He B, Qiao X, Klasse PJ, Chiu A, Chadburn A, Knowles DM, et al
. HIV-1 envelope triggers polyclonal Ig class switch recombination through a CD40-independent mechanism involving BAFF and C-type lectin receptors. J Immunol 2006; 176:3931–3941.
7. Gourzi P, Leonova T, Papavasiliou FN. Viral induction of AID is independent of the interferon and the Toll-like receptor signaling pathways but requires NF-kappaB. J Exp Med 2007; 204:259–265.
8. Muramatsu M, Sankaranand VS, Anant S, Sugai M, Kinoshita K, Davidson NO, et al
. Specific expression of activation-induced cytidine deaminase (AID), a novel member of the RNA-editing deaminase family in germinal center B cells. J Biol Chem 1999; 274:18470–18476.
9. Smit LA, Bende RJ, Aten J, Guikema JE, Aarts WM, van Noesel CJ. Expression of activation-induced cytidine deaminase in confined to B-cell non-Hodgkin's lymphomas of the germinal-center phenotype. Cancer Res 2003; 63:3894–3898.
10. Greeve J, Philipsen A, Krause K, Klapper W, Heidorn K, Castle BE, et al
. Expression of activation-induced cytidine deaminase in human B-cell non-Hodgkin's lymphomas. Blood 2003; 101:3574–3580.
11. Lenz G, Nagel I, Siebert R, Roschke AV, Sanger W, Wright GW, et al
. Aberrant immunoglobulin class switch recombination and switch translocations in activated B cell-like diffuse large B cell lymphoma. J Exp Med 2007; 204:633–643.
12. Kotani A, Kakazu N, Tsuruyama T, Okazaki IM, Muramatsu M, Kinoshita K, et al
. Activation-induced cytidine deaminase (AID) promotes B cell lymphomagenesis in Emu-cmyc transgenic mice. Proc Natl Acad Sci USA 2007; 104:1616–1620.
13. Martinez-Maza O, Breen EC. B-cell activation and lymphoma in patients with HIV. Curr Opin Oncol 2002; 14:528–532.
14. CDC. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adult
15. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986; 42:121–130.
16. Breen EC, Epeldegui M, Boscardin WJ, Widney DP, Detels R, Martinez-Maza O. Elevated levels of soluble CD44 precede the development of AIDS-associated non-Hodgkin's B-cell lymphoma. AIDS 2005; 19:1711–1712.
17. Breen EC, Fatahi S, Epeldegui M, Boscardin WJ, Detels R, Martinez-Maza O. Elevated Serum Soluble CD30 Precedes the Development of AIDS-Associated Non-Hodgkin's B Cell Lymphoma. Tumour Biol 2006; 27:187–194.
18. Breen EC, van der Meijden M, Cumberland W, Kishimoto T, Detels R, Martinez-Maza O. The development of AIDS-associated Burkitt's/small noncleaved cell lymphoma is preceded by elevated serum levels of interleukin 6. Clin Immunol 1999; 92:293–299.
19. Widney DP, Breen EC, Boscardin WJ, Kitchen SG, Alcantar JM, Smith JB, et al
. Serum levels of the homeostatic B cell chemokine, CXCL13, are elevated during HIV infection. J Interferon Cytokine Res 2005; 25:702–706.
20. Widney D, Gundapp G, Said JW, Van der Meijden M, Bonavida B, Demidem A, et al
. Aberrant expression of CD27 and soluble CD27 (sCD27) in HIV infection and in AIDS-associated lymphoma. Clin Immunol 1999; 93:114–123.
21. Ramiro AR, Jankovic M, Callen E, Difilippantonio S, Chen H-T, McBride KM, et al
. Role of genomic instability and p53 in AID-induced c-myc-Igh translocations. Nature 2006; 440:105–109.