RSPO2 abnormal transcripts result from read-through in liver tumours with high ß-catenin activation and CTNNB1 mutations

RSPO2/RANKL-LGR4 signaling regulates osteoclastic pre-metastatic niche formation and bone metastasis

Therapeutics targeting osteoclasts are commonly used treatments for bone metastasis; however, whether and how osteoclasts regulate pre-metastatic niche and bone tropism is largely unknown. In this study, we report that osteoclast precursors (OPs) can function as a pre-metastatic niche component that facilitates breast cancer (BCa) bone metastasis at early stages. At the molecular level, unbiased GPCR ligand/agonist screening in BCa cells suggested that R-spondin 2 (RSPO2) and RANKL, through interacting with their receptor LGR4, promoted osteoclastic pre-metastatic niche formation and enhanced BCa bone metastasis. This was achieved by RSPO2/RANKL-LGR4 signal modulating WNT inhibitor DKK1 through Gαq and β-catenin signaling. DKK1 directly facilitated OP recruitment through suppressing its receptor low-density lipoprotein-related receptors 5 (LRP5) but not LRP6, upregulating Rnasek expression via inhibiting canonical WNT signaling. In clinical samples, RSPO2, LGR4 and DKK1 expression showed positive correlation with BCa bone metastasis. Furthermore, soluble LGR4 extracellular domain (ECD) protein, acting as a decoy receptor for RSPO2 and RANKL, significantly alleviated bone metastasis and osteolytic lesions in mouse bone metastasis model. These findings provide unique insights into the functional role of OPs as key components of pre-metastatic niche for BCa bone metastasis, indicate RSPO2/RANKL-LGR4 signaling as a promising target for inhibiting BCa bone metastasis.

Predictive Patterns of Glutamine Synthetase Immunohistochemical Staining in CTNNB1-mutated Hepatocellular Adenomas

Some hepatocellular adenoma (HCA) subtypes are characterized by different CTNNB1 mutations, leading to different beta-catenin activation levels, hence variable immunostaining patterns of glutamine synthetase (GS) expression, and different risks of malignant transformation. In a retrospective multicentric study of 63 resected inflammatory (n=33) and noninflammatory (n=30) molecularly confirmed CTNNB1-mutated b-(I)HCA, we investigated the predictive potential of 3 known GS patterns as markers for CTNNB1 exon 3, 7/8 mutations. Pattern 1 (diffuse homogenous) allowed recognition of 17/21 exon 3 non-S45 mutated b-(I)HCA. Pattern 2 (diffuse heterogenous) identified all b-(I)HCA harboring exon 3 S45 mutation (20/20). Pattern 3 (focal patchy) distinguished 12/22 b-(I)HCA with exon 7/8 mutations. In exon 3 S45 and 7/8 mutations, both b-HCA and b-IHCA showed a GS+/CD34- rim with diffuse CD34 positivity in the center of the lesion. Interobserver reproducibility was excellent for exon 3 mutations. Comparative analysis of GS patterns with molecular data showed 83% and 80% sensitivity (b-HCA/b-IHCA) and 100% specificity for exon 3 non-S45. For exon 3 S45, sensitivity was 100% for b-(I)HCA, and specificity 93% and 92% (b-HCA/b-IHCA). For exon 7/8, sensitivity was 55% for both subtypes and specificity 100% and 96% (b-HCA/b-IHCA). Preliminary data from 16 preoperative needle biopsies from the same patients suggest that this panel may also be applicable to small samples. In surgically resected HCA, 2 distinct GS patterns can reliably predict CTNNB1 exon 3 mutations, which are relevant because of the higher risk for malignant transformation. The third pattern, although specific, was less sensitive for the identification of exon 7/8 mutation, but the GS+/CD34- rim is a valuable aid to indicate either an exon 3 S45 or exon 7/8 mutation.