Cholesterol promotes the formation of dimers and oligomers of the receptor tyrosine kinase ROR1

Abstract

Receptor tyrosine kinase-like orphan receptor 1 (ROR1) is a member of the receptor tyrosine kinase (RTK) family that plays a crucial role during organogenesis of bone and neural systems by regulating noncanonical Wnt signaling. Misregulation of ROR1 is additionally a causative factor for carcinogenesis in solid and liquid tumors. However, we have a poor understanding of how ROR1 activity is regulated. We employed a recently developed single-molecule method termed SiMPull-POP to study the oligomeric state of ROR1. RTK function is typically triggered by ligand binding, which promotes self-assembly of RTKs to form dimers and in some cases oligomers. However, our data indicate that ROR1 does not follow this paradigm. Instead, ROR1 forms dimers and oligomers in a process that is not affected by the presence of the ROR1 ligand Wnt5a. Additional experiments indicate that the transmembrane domain of ROR1 has a strong tendency to self-assemble, suggesting that this domain modulates ROR1 dimerization. Investigation into a regulatory mechanism for ROR1 self-assembly led to evaluation of the role of the lipid cholesterol, which plays pleiotropic roles in Wnt signaling. Cholesterol was found to promote the assembly of ROR1, and our results point to the transmembrane domain as the region where cholesterol exerts this regulatory effect. Taken together, our results indicate that ROR1 self-assembles in human cells; however, unlike other RTKs, this process is not stabilized by ligand binding but is instead facilitated by membrane cholesterol.

Keywords: ROR1; cholesterol; protein assembly; receptor tyrosine kinase; single-molecule fluorescence.

Figure 1

ROR1 forms dimers and oligomers. Oligomeric distribution of ROR1 isolated in DIBMALPs prepared from HEK293T cells, as determined by SiMPull-POP. N = 3. Data are mean ± S.D. ROR1, receptor tyrosine kinase-like orphan receptor 1; DIBMALP, DIBMA lipid particle.

Figure 2

Dimerization of ROR1 TMD in HEK293T cells.A, schematic representation of the bimolecular fluorescent complementation (BiFC) assay. B, amino acid sequences fused to VFP halves used in the assay and their predicted ΔG (ΔG_app) values for membrane insertion in kcal/mol. C, relative fluorescence units (RFU) of each tested homo-oligomer in the BiFC assay. Mean ± S.D. for at least 13 independent experiments are shown. The individual value of each experiment is represented by a dot. The VN-GpA/VC-GpA homodimer is used as a positive control and to normalize values across experiments. The Lep H2 TMD was included as a negative control. The homo-oligomers that produced fluorescence levels significantly higher than the H2 TMD homo-oligomer (two-tailed homoscedastic t test) are highlighted in green. The lower panels display Western blot data for the VN and VC constructs’ expression levels detected by α-c-Myc antibody and α-HA antibody, respectively. An α-GPADH antibody was used as a loading control. RFU values were normalized to expression levels. ROR1, receptor tyrosine kinase-like orphan receptor 1; GpA, glycophorin A.

Figure 3

Cholesterol removal promotes monomeric ROR1.A, cholesterol levels after treatment with MβCD. B, oligomeric distribution of ROR1. C, BiFC of ROR1 TMD homo-oligomers in control conditions and after MβCD treatment. All experiments were performed in HEK293T cells. Mean ± S.D. are shown with individual experiments represented by a square or dot. For statistical analysis, an unpaired t test was used for data shown in (A), a two-tailed homoscedastic t test in (B), and a two-way ANOVA followed by a multiple comparison unpaired t-tests in (C). ROR1, receptor tyrosine kinase-like orphan receptor 1.

Figure 4

ROR1-GFP partitions into GPMV L_o domains more efficiently than the RTK EphA2-GFP.A, representative images of GPMVs from control and MβCD-treated HeLa cells transfected with ROR1-GFP or (B) EphA2-GFP (scale bars represent 5 μm) (left). L_d partition coefficient (K_p) of ROR1-GFP and EphA2-GFP in phase-separated GPMVs (right). The K_p was determined by Equation 1. Total number of GPMVs analyzed per condition were 68 (ROR1, Control), 37 (ROR1, MβCD), 27 (EphA2, Control), and 22 (EphA2, MβCD). Data are mean ± S.D., with K_p values for at least three biological replicates. An unpaired t test was run to compare control and MβCD conditions; no statistical differences were found. Individual GPMVs analyzed are represented by a solid dot. RTK, receptor tyrosine kinase; ROR1, receptor tyrosine kinase-like orphan receptor 1.