Disruption of Direct 3-Dimensional (3D) Telomere-TRF2 (Telomere Related Factor 2) Interaction Is a Hallmark of Primary Hodgkin (H) and Reed-Sternberg (RS) Cells

Introduction: Numerical and complex structural chromosomal aberrations are regularly identified in primary mononuclear H and multinucleated RS cells and in Hodgkin's lymphoma (HL) derived H cell lines. The transition from H to RS cells is associated with dynamic progressive 3D telomere dysfunction, major changes in the telomere-protecting shelterin complex and formation of giant "zebra" chromosomes. Analogous findings are observed in an in vitro model for post-germinal center B-cell, Epstein-Barr-virus (EBV)-associated HL (Blood. 2015;125 (13):2101-2110). In this experimental system the EBV encoded oncogene LMP1 mediates multinuclearity through downregulation of TRF2. Thus, we hypothesized that the 3D interaction of telomeres and TRF2 is directly associated with the formation of H and RS cells. To this end, we developed a combined quantitative 3D TRF2-telomere immuno FISH protocol (3D TRF2/Telo-Q-FISH) and applied it to monolayers of primary H and RS cells enriched from diagnostic lymph node biopsies allowing 3D analysis of the entire nuclear content (often not achieved using laser microdissection of H and RS cells, given that their nuclei are generally >10 um in diameter and that this technique is performed on 5 um sections).

Methods: 3D TRF2/Telo-Q-FISH(Blood. 2015;125 (13):2101-2110; data supplement) was performed on cultured BJ-5ta normal diploid fibroblasts and B-cell rich suspensions (negative selection) of diagnostic lymph nodes, cultured for 12-36 hours. Quality of H and RS cells was assessed by Wright-Giemsa staining. Lymphocytes, H and RS cells of 10 classical HL were analyzed, 3 of them containing LMP1+ tumor cells (EBV-associated HL).

Results: Whereas the BJ-5ta normal diploid control fibroblasts and most internal control lymphocytes showed quantitatively and qualitatively intact 1:1, tight physical 3D association of TRF2 with telomeres, with only few very short telomeres (<20%), the H and RS cells of all 10 cases consistently displayed telomere-TRF2 steric disruption, with a significant increase of very short telomeres (49-71%). Interestingly, two opposite disruption patterns of the 3D telomere-TRF2 interaction were identified.

The 3 LMP1+ cases were characterized by abundant telomeres without TRF2 interaction (de-protected). TRF2 spots were rare and the quantitative ratio of telomere/TRF2 signals was > 6:1. This loss of TRF2 expression showed gradual progression from H to RS cells.

The 7 LMP1- cases showed a different 3D telomere-TRF2 pattern. Three cases with clinically aggressive disease (stages IVA, IVB and IIIB) were characterized by abundant TRF2 signals, most of them not associated with telomeres. In two of them huge multinucleated RS "ghost" cells without any or only few telomeres but numerous TRF2 signals and internuclear DNA-bridges were observed. In one of these two and in the third case bi-nucleated RS cells with grossly different nuclear telomere content of nucleus one and nucleus two were present. Most importantly, these 3 cases showed an inversion of the quantitative ratio of telomere/TRF2 signals <1:3 and the progressive loss of telomere signals from H to RS cells. The remaining 4 cases had a mixed pattern with an overall moderate loss of TRF2 signals including TRF2 de-protected telomeres but also TRF2 signals not associated with telomeres (quantitative, telomere/TRF2 signals ratio of about 1.5-2:1).

Conclusions: Disruption of the 3D TRF2-telomere interaction was demonstrated in 100% of primary H and RS cells. However, two opposite pathogenetic mechanisms appear to be operative: in EBV-associated HL telomere de-protection is due to an impressive loss of TRF2 signals physically linked to telomeres, in agreement with the hypothesis of our in vitro model for EBV-associated HL. In clinically aggressive EBV-negative HL massive attrition of telomere signals is associated with an overwhelming increase in TRF2 signals that are not associated with telomeres. In both scenarios, the telomeres are predominantly very short and the crucial 3D interaction is disrupted, thereby qualifying HL as a shelterin-associated disease with the telomere-shelterin complex playing the role of a "plaque tournante" in the pathogenesis of HL. Further molecular dissection of the shelterin-telomere interaction may offer new therapeutic targets for refractory/relapsing disease.