HACE1 is a tumor suppressor gene candidate in natural killer cell neoplasms

Abstract

HACE1 is an E3 ubiquitin ligase located in 6q21, the genomic region frequently deleted in natural killer (NK) cell malignancies. Here, we report HACE1 as a candidate tumor suppressor gene silenced through a combination of deletion and cytosine phosphate guanine island hypermethylation. We detected deletion of HACE1 in malignant NK cell lines (6 of 9, 67%) and primary biopsies (5 of 15, 33%) by quantitative PCR, with most of the specimen showing cytosine phosphate guanine island hypermethylation in the remaining allele, leading to low mRNA transcription. The ectopic expression of HACE1 in an HACE1-null NK cell line led to apoptosis and G2/M cell cycle arrest. Moreover, HACE1 expression was up-regulated in IL-2-activated normal NK cells and NK cells cocultured with an engineered NK cell target, K562 Clone 9.mbIL21, suggesting its role in the regulation of NK cell homeostasis. In conclusion, HACE1 is another potent tumor suppressor gene located within the 6q21 region, and loss of function of multiple tumor suppressor genes within 6q21 may be a critical determinant of NK cell lymphomagenesis.

Figure 1

HACE1 is silenced through a combination of hemizygous deletion and CpG island hypermethylation in NKCL samples. A and B: NK cell lines and NKCL cases with or without deletion were determined by the HACE1/RPL13A ratio for each sample. NKCL cases with HACE1/RPL13A <0.75 that of human resting primary PB NK cell DNA were defined to have deletion. HACE1/RPL13A for each sample represented the average of two independent experiments for both HACE1 and RPL13A. The deletion status of the NKCL cases by aCGH was shown with the ± signs at the bottom of the figure. The plus sign indicates a hemizygous deletion in the minimal common region of 6q21 according to the two previous reports that applied bacterial artificial chromosome-aCGH on NK cell lines and NKCL cases or Cartes d’Identité des Tumeurs CGH on NKCL cases. The dashed horizontal bar represents the cutoff used to determine the deletion status of the NK cell lines and NKCL cases. C: Quantification of HACE1 mRNA in NK cell lines. The HACE1 mRNA expression in NK cell lines was determined with RT-qPCR. Relative expression values were normalized to the values of the resting NK cells. Resting and 4-day IL-2–activated primary PB NK cells were used as control. Data are means ± SDs of two experiments. D: Hypermethylation of the three CpG islands, CpG-29, CpG-177, and CpG-88, in NKCL cases was determined with MSCC. The differential methylation of the HpaII sites was shown after normalization of the HpaII count sites to the count sites of the 48-hour IL-2–activated peripheral blood NK cells. Gray (4- to 10-fold) and black (>10-fold) boxes show the hypermethylated HpaII sites. E: HACE1 mRNA expression in seven NKCL cases. HACE1 mRNA expression in NKCL cases was normalized to the HACE1 expression in resting PB NK cells. Resting and 4-day IL-2–activated PB NK cells were used as the controls. The HACE1 deletion status was marked with plus and minus signs on the x axis. F: The total number of hypermethylated HpaII sites (ie, hypermethylated HpaII sites in CpG-177, CpG-88, and CpG-29 altogether) (left panel) and the number of hypermethylated HpaII sites in each CpG island compared with normal 48-hour IL-2–activated NK cells (right panel) were indicated for the same NKCL cases with the same sample order as in E. Fourfold count was set as the threshold for hypermethylation. G: MSP-qPCR results of the seven malignant NK cell lines are shown. Resting NK cells (NK0) were used as the negative, and M.SssI methylated lymphocyte DNA (Met) was used as the positive control, respectively. −ΔCt values are calculated as follows: −(Ct_{NK cell line} − Ct_NK0).

Figure 2

Reconstitution of HACE1 in a HACE1-null NK cell line induces apoptosis and cell cycle arrest. A: Vector or HACE1-transduced KAI3 cells were stained with Annexin V-PE and tested with fluorescence-activated cell sorting. B: Ectopic expression of HACE1 caused apoptosis in the KAI3 cell line (vector: 9.74% ± 0.92%, n = 2; HACE1: 89.04% ± 0.21%, n = 2). The rate of apoptosis was determined by measuring the proportion of Annexin V⁺ cells in the GFP⁺ population 2 days after transduction. Data are means ± SDs of two independent experiments. *P < 0.001 versus vector. C: Vector- or HACE1-transduced KAI3 cells were stained with Hoechst 33342, and the cell cycle profile on the GFP-gated population of vector- or HACE1-transduced KAI3 cells was determined with fluorescence-activated cell sorting. D: Ectopic HACE1 expression causes G₂/M cell cycle arrest in the KAI3 cell line [(%HACE1 − %Vector); (ΔG₀/G₁ = −23%, ΔS = 1%, ΔG₂/M = 12.9%)]. Data are means ± SDs of two independent experiments. E: HACE1 mRNA expression was determined in IL-2–activated PB NK cells. RPL13A was used as the housekeeping gene for normalization. F: HACE1 mRNA expression in cocultured primary NK cells is shown. *P < 0.05 versus Resting NK cells. G: Ectopic expression of HACE1 in KAI3 cells. RT-qPCR was applied on GFP⁺ KAI3 cells sorted 42 hours after transduction with vector or HACE1. RPL13A was used as the housekeeping gene for normalization. Data are means ± SDs of two independent experiments. *P < 0.001 versus vector.