Site-specific expression of polycomb-group genes encoding the HPC-HPH/PRC1 complex in clinically defined primary nodal and cutaneous large B-cell lymphomas - PubMed (original) (raw)

Comparative Study

Site-specific expression of polycomb-group genes encoding the HPC-HPH/PRC1 complex in clinically defined primary nodal and cutaneous large B-cell lymphomas

Frank M Raaphorst et al. Am J Pathol. 2004 Feb.

Abstract

Polycomb-group (PcG) genes preserve cell identity by gene silencing, and contribute to regulation of lymphopoiesis and malignant transformation. We show that primary nodal large B-cell lymphomas (LBCLs), and secondary cutaneous deposits from such lymphomas, abnormally express the BMI-1, RING1, and HPH1 PcG genes in cycling neoplastic cells. By contrast, tumor cells in primary cutaneous LBCLs lacked BMI-1 expression, whereas RING1 was variably detected. Lack of BMI-1 expression was characteristic for primary cutaneous LBCLs, because other primary extranodal LBCLs originating from brain, testes, and stomach were BMI-1-positive. Expression of HPH1 was rarely detected in primary cutaneous LBCLs of the head or trunk and abundant in primary cutaneous LBCLs of the legs, which fits well with its earlier recognition as a distinct clinical pathological entity with different clinical behavior. We conclude that clinically defined subclasses of primary LBCLs display site-specific abnormal expression patterns of PcG genes of the HPC-HPH/PRC1 PcG complex. Some of these patterns (such as the expression profile of BMI-1) may be diagnostically relevant. We propose that distinct expression profiles of PcG genes results in abnormal formation of HPC-HPH/PRC1 PcG complexes, and that this contributes to lymphomagenesis and different clinical behavior of clinically defined LBCLs.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Immunohistochemical detection of PcG proteins in primary nodal and primary cutaneous LBCLs. Shown are immunohistochemical staining patterns of HPC-HPH/PRC1 PcG complex proteins (BMI-1, RING1, and HPH1) and EED-EZH/PRC2 PcG complex proteins (EED and EZH2). A–E: Primary nodal LBCLs. Large tumor cells express components of the HPC-HPH/PRC1 complex (BMI-1, RING1, and HPH1) and components of the EED-EZH/PRC2 complex (EED and EZH2). F–J: Primary cutaneous LBCLs originating from the trunk. K–O: Primary nodal LBCLs originating from the leg. Shown are two BMI-1neg examples of primary cutaneous LBCLs; large neoplastic cells in the primary cutaneous LBCLs from the trunk variably express RING1 (G) and are HPH1neg (H). By contrast, neoplastic cells in primary cutaneous LBCLs from the leg are RING1neg (L) and HPH1pos (M). Both primary cutaneous LBCLs use EED (I, N) and EZH2 (J, O). P–T: Secondary cutaneous deposit from a primary nodal LBCL; neoplastic cells display a PcG expression profile that is identical to the profile in primary nodal LBCL (A–E). Note that cells in the reactive infiltrate variably express the HPC-HPH/PRC1 complex proteins BMI-1, RING1, and HPH1, which serves as an internal positive control in cases in which the tumor cells are negative. The EED-EZH/PRC2 PcG complex proteins EED and EZH2 are infrequently detected in reactive lymphocytes. The PcG expression pattern in reactive lymphocytes reflects the mutually exclusive expression pattern of these PcG genes in healthy resting and cycling cells (see Figures 2 and 3), and indicates that the presence of BMI-1, RING1, and HPH1 in tumor cells is abnormal. Original magnifications: ×63.

Figure 2

Figure 2

Triple immunofluorescence for BMI-1, EZH2, and MIB-1 in primary nodal and primary cutaneous LBCLs. BMI-1, EZH2, and the proliferation marker MIB-1 were detected by red, green, and blue immunofluorescence, respectively. A–C: Primary nodal LBCLs. Large cycling (MIB-1pos) neoplastic cells express BMI-1 (A) and EZH2 (B), leading to co-expression of BMI-1 and EZH2 in the same nucleus (C). D–F: Primary cutaneous LBCLs originating from the trunk. G–I: Primary cutaneous LBCLs originating from the legs. In primary cutaneous LBCLs, BMI-1 is undetectable in large cycling (MIB-1pos) neoplastic cells (D and G) whereas EZH2 is present (E and H). F and I: Combination of the BMI-1 and EZH2 signals. Note that the reactive infiltrating cells express BMI-1 in the absence of EZH2. In addition, small EZH2pos cells in the reactive infiltrate are dividing cells because the EZH2 signal overlaps with the signal for MIB-1 (B, E, H). Occasional small EZH2pos cells with undetectable MIB-1 expression may reflect cells that are in transition from being in cycle to being in rest, or they could reflect nonlymphoid cells with a different PcG expression profile.

Figure 3

Figure 3

Triple immunofluorescence for BMI-1, HPH1, and MIB-1 in primary nodal and primary cutaneous LBCLs. BMI-1, HPH1, and the proliferation marker MIB-1 were detected by red, green, and blue immunofluorescence, respectively. A–C: Primary nodal LBCLs. Large cycling (MIB-1pos) neoplastic cells express BMI-1 (A) and HPH1 (B), leading to co-expression of BMI-1 and HPH1 in the same nucleus (C). Co-expression of BMI-1 and HPH1 also occurs in the reactive infiltrate, but is limited to resting (MIB-1neg) cells (see comments in legend of Figure 2, and in the text). D–F: Primary cutaneous LBCLs originating from the trunk. In the example shown, large cycling (MIB-1pos) neoplastic cells are BMI-1neg (D) and HPH1neg (E). By contrast, BMI-1 and HPH1 are co-expressed (F) in resting (MIB-1neg) cells of the reactive infiltrate. G–I: Primary cutaneous LBCLs originating from the legs. In the example shown, large cycling (MIB-1pos) neoplastic cells express HPH1 (H) and are BMI-1neg (G). Note that HPH1pos neoplastic lymphocytes in primary nodal LBCLs (A–C) and primary cutaneous LBCLs originating from the legs contain domains that strongly stain for HPH1. In primary nodal LBCLs, these domains appear to be BMI-1neg (insets in A–C). The faint signal for BMI-1 in tumor cells shown in G is not representative, and most likely reflects weak background staining that was occasionally observed in cutaneous tissue sections.

References

    1. Jacobs JJ, van Lohuizen M. Polycomb repression: from cellular memory to cellular proliferation and cancer. Biochim Biophys Acta. 2002;1602:151–161. - PubMed
    1. Raaphorst FM, Otte AP, Meijer CJ. Polycomb-group genes as regulators of mammalian lymphopoiesis. Trends Immunol. 2001;22:682–690. - PubMed
    1. Smale ST. The establishment and maintenance of lymphocyte identity through gene silencing. Nat Immunol. 2003;4:607–615. - PubMed
    1. Lessard J, Sauvageau G. Polycomb group genes as epigenetic regulators of normal and leukemic hemopoiesis. Exp Hematol. 2003;31:567–585. - PubMed
    1. McGinnis W, Krumlauf R. Homeobox genes and axial patterning. Cell. 1992;68:283–302. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources