Prohibitins 1 and 2 interact with NTSR2.
As previously mentioned, NTSR2 and TrkB have been identified as a protein complex involved in resistance to cell death in CLL-B cells (12). In order to further understand this oncogenic interaction, we expressed both of these proteins in HEK293T cells.
HEK293T cells were co-transfected with a plasmid encoding biotin transferase BirA and a plasmid encoding TrkB and a version of NTSR2 with an AviTag™ added at the N-terminus (T-N-AviTag™).
They were then treated with or without BDNF, since it was shown that it enhances the NTSR2-TrkB interaction (12), and with biotin to enable the biotinylation of the AviTag™ by BirA. Following transfection, cell lysates were incubated with streptavidin-coated magnetic beads to isolate the tagged NTSR2 (14). We started by validating the experimental approach, as shown on Fig. 1, the transfection of HEK293T cells with the T-N-AviTag™ plasmid enables the efficient isolation of NTSR2. We checked for the co-precipitation of TrkB, which was indeed observed following the pull-down of the tagged version of NTSR2.
Then, to study the interactome of NTSR2, the isolated proteins were sent out on-beads to be analyzed by mass spectrometry (14). Mass spectrometry results showed the co-precipitation of TrkB, confirming its interaction with NTSR2 in our cellular model. Interestingly, in each of the four replicates we sent for analysis, we also observed the presence of prohibitins 1 and 2 (Table 1).
Table 1
Mass spectrometry analysis of the NTSR2 interactome in HEK293T cells. Coverage (%) indicates the percentage of the protein covered by the identified peptides; # Peptides indicates the number of distinct peptide sequences that are within the protein sequence; Score Sequest HT is the sum of the individual scores of each peptide (the higher this score, the better the identification). Four replicates were analyzed.
Replicate number | Accession | Description | Coverage [%] | # Peptides | Score Sequest HT |
1 | Q99623 | Prohibitin-2 OS = Homo sapiens OX = 9606 GN = PHB2 PE = 1 SV = 2 | 36 | 9 | 76.88 |
P35232 | Prohibitin OS = Homo sapiens OX = 9606 GN = PHB PE = 1 SV = 1 | 38 | 8 | 133.35 |
Q16620 | BDNF/NT-3 growth factors receptor OS = Homo sapiens OX = 9606 GN = NTRK2 PE = 1 SV = 1 | 18 | 12 | 166.79 |
O95665 | Neurotensin receptor type 2 OS = Homo sapiens OX = 9606 GN = NTSR2 PE = 1 SV = 2 | 10 | 3 | 136.15 |
P23560 | Brain-derived neurotrophic factor OS = Homo sapiens OX = 9606 GN = BDNF PE = 1 SV = 1 | 6 | 1 | 4.55 |
2 | Q99623 | Prohibitin-2 OS = Homo sapiens OX = 9606 GN = PHB2 PE = 1 SV = 2 | 53 | 14 | 90.43 |
P35232 | Prohibitin OS = Homo sapiens OX = 9606 GN = PHB PE = 1 SV = 1 | 47 | 11 | 120.45 |
Q16620 | BDNF/NT-3 growth factors receptor OS = Homo sapiens OX = 9606 GN = NTRK2 PE = 1 SV = 1 | 17 | 13 | 101.19 |
O95665 | Neurotensin receptor type 2 OS = Homo sapiens OX = 9606 GN = NTSR2 PE = 1 SV = 2 | 15 | 5 | 57.84 |
P23560 | Brain-derived neurotrophic factor OS = Homo sapiens OX = 9606 GN = BDNF PE = 1 SV = 1 | 4 | 1 | 5.92 |
3 | P35232 | Prohibitin-2 OS = Homo sapiens OX = 9606 GN = PHB2 PE = 1 SV = 2 | 18 | 4 | 12.00 |
Q99623 | Prohibitin OS = Homo sapiens OX = 9606 GN = PHB PE = 1 SV = 1 | 13 | 2 | 8.09 |
Q16620 | BDNF/NT-3 growth factors receptor OS = Homo sapiens OX = 9606 GN = NTRK2 PE = 1 SV = 1 | 11 | 6 | 21.54 |
O95665 | Neurotensin receptor type 2 OS = Homo sapiens OX = 9606 GN = NTSR2 PE = 1 SV = 2 | 25 | 7 | 69.23 |
P23560 | Brain-derived neurotrophic factor OS = Homo sapiens OX = 9606 GN = BDNF PE = 1 SV = 1 | 7 | 2 | 4.62 |
4 | P35232 | Prohibitin-2 OS = Homo sapiens OX = 9606 GN = PHB2 PE = 1 SV = 2 | 17 | 4 | 12.35 |
Q99623 | Prohibitin OS = Homo sapiens OX = 9606 GN = PHB PE = 1 SV = 1 | 17 | 3 | 8.63 |
Q16620 | BDNF/NT-3 growth factors receptor OS = Homo sapiens OX = 9606 GN = NTRK2 PE = 1 SV = 1 | 10 | 5 | 16.04 |
O95665 | Neurotensin receptor type 2 OS = Homo sapiens OX = 9606 GN = NTSR2 PE = 1 SV = 2 | 25 | 6 | 76.34 |
P23560 | Brain-derived neurotrophic factor OS = Homo sapiens OX = 9606 GN = BDNF PE = 1 SV = 1 | 8 | 2 | 7.81 |
Previous work within our team evidenced the implication of prohibitin 2 in adverse prognosis in DLBCL (8), additionally, given the literature describing the potential role of prohibitins in hematological malignancies (7), we decided to focus on better understanding their involvement in the NTSR2-TrkB interaction in chronic lymphocytic leukemia.
To confirm the mass spectrometry results, proteins isolated on beads were also analyzed by western blotting. As shown on Fig. 2A, both prohibitins 1 and 2 were found among the eluted proteins co-precipitated with NTSR2, in higher relative quantity when cells were treated with BDNF, the natural ligand of TrkB (Fig. 2B).
As shown in Figs. 2C to 2F, proximity ligation assay was performed on both transfected HEK293T cells and leukemic B cells from CLL patients (patients IDs 1, 2 and 3) to once again confirm the interaction of prohibitins with NTSR2, determine whether they interacted with NTSR2 alone or also with TrkB, and evaluate the effect of BDNF treatment on the observed interactions. Interestingly, results showed that both prohibitins 1 and 2 interact with NTSR2 but not with TrkB and that their interaction is significantly enhanced in the presence of BDNF, further highlighting its role in this oncogenic platform.
Prohibitins impact the expression of NTSR2 and its interaction with TrkB.
To further confirm the interest of studying prohibitins in CLL, we compared the level of mRNA expression of PHB1 and PHB2 in B cells from three CLL patients (patients IDs 2, 3 and 4) versus PBMCs from three healthy donors. As shown on Fig. 3A, both PHB1 and PHB2 are overexpressed in malignant cells compared to their healthy counterparts, PHB1 expression shows a 1.7-fold increase and PHB2 expression a 3.4-fold increase.
Having established the overexpression of prohibitins in CLL patients B cells, we set out to explore the impact of prohibitins on the NTSR2/TrkB oncogenic platform. In order to achieve this, we performed siRNA-mediated inhibition of PHB1, PHB2 or PHB1 and 2 simultaneously on B cells from CLL patients. Unfortunately, CLL-B cells did not withstand the siRNA transfection. Despite trying various transfecting agents and protocols, we were unable to obtain an efficient inhibition of prohibitins while maintaining a satisfying cell mortality rate. We therefore opted for the CLL cell line MEC-1, which expresses both NTSR2 and TrkB (12), to pursue our experiments. Data showing the effective inhibition of both mRNA expression and protein levels of prohibitins following selective inhibition by siRNA in MEC-1 cells is available in Supplementary Fig. 1. Following selective inhibition of prohibitins, no significant expression change in NTRK2, the gene encoding TrkB, was observable. However, mRNA expression of the NTSR2 gene was affected in all experimental conditions (PHB1, PHB2 and PHB1/2 siRNA), independently of BDNF treatment (Figs. 3B & 3C). This decrease in NTSR2 mRNA levels upon PHBs inhibition was also observed at the protein level, as shown on Figs. 3D and 3E. TrkB protein levels, similarly to mRNA levels, were unaffected (Fig. 3D and 3F).
We then evaluated how the inhibition of prohibitins affected the ability to interact of NTSR2 and TrkB. Following siRNA transfection, MEC-1 cells were harvested, and their protein extract was subjected to immunoprecipitation using protein A magnetic beads functionalized with an anti-NTSR2 antibody. Co-immunoprecipitation of TrkB was then evaluated through western blot analysis and their level of interaction was quantified using the densitometry ratio TrkB / NTSR2. As shown on Figs. 3G and 3H, PHB1 and PHB2 inhibition induced a decrease in the level of interaction between NTSR2 and TrkB, and that inhibition was found regardless of the presence of BDNF in the culture medium.
In conclusion, prohibitins do not only impact the mRNA expression level of NTSR2, but they also modulate its ability to interact with TrkB.
To investigate through which mechanisms prohibitins were able to regulate the expression of NTSR2, we proceeded to several co-immunoprecipitations on the MEC-1 cell line using anti-PHB1 and anti-PHB2 antibodies. We tested for the presence of various transcription factors, among them nuclear factor-kappa B (NF-κB), which regulates the expression of anti-apoptotic Bcl-2 members (15) and is implicated in the in vitro survival of CLL-B cells and clinical disease progression of CLL (16). It has also already been associated with prohibitins, notably PHB1 which has been shown to inhibit NF-κB activity in intestinal epithelial cells (17). We also checked for an interaction with specific protein 1 (Sp1) which has been linked to leukemogenesis (18) and cytokine signaling (19) in CLL. Finally, we tested for the presence of CAMP responsive element binding protein 1 (CREB1) as it has previously been associated with acute lymphoblastic leukemia (20) and acute myeloid leukemia (21) and can bind the TGACGTCA sequence, which is present in the promoter region of NTSR2 (GeneHancer, genecards.org).
Neither NF-κB nor Sp1 were found among the proteins isolated by co-immunoprecipitation, however, CREB1 was co-immunoprecipitated with PHB2, as visible on Fig. 4, showing that they are able to interact.
Prohibitin-binding compound fluorizoline affects downstream NTSR2/TrkB signaling.
As previously described by Abbaci et al. (12), the NTSR2/TrkB interaction induces the phosphorylation of Src as well as an increase in the expression of anti-apoptotic proteins Bcl-2 and Bcl-xL. Fluorizoline is a synthetic compound able to bind to PHB 1 and 2 (22) and which induces CLL-B cells apoptosis in vitro (23). In order to confirm the role prohibitins play in this interaction, we investigated the effect of fluorizoline-induced inhibition of prohibitin activity on downstream NTSR2/TrkB signaling. MEC-1 cells were treated overnight with either 2.5 µM or 5 µM of fluorizoline or left untreated. They were then harvested, and their protein extracts were analyzed by western blot. We verified that fluorizoline inhibited the NTSR2/TrkB interaction, as previously observed with the siRNA-mediated inhibition, fluorizoline induced a decrease in their level of interaction, as shown on Fig. 5A.
Following fluorizoline treatment, we observed an increase in the amount of Src protein, and therefore a decrease in the P-Src/Src phosphorylation ratio (Figs. 5B, 5E). We also observed a decrease in the amount of anti-apoptotic proteins Bcl-2 and Bcl-xL (Figs. 5C, 5F, 5D, 5G). As expected, upon fluorizoline treatment, we observed a decrease in NTSR2/TrkB signaling characterized by a diminished P-Src/Src ratio and lower amounts of anti-apoptotic proteins Bcl-2 and Bcl-xL. These findings once again highlight the impact of prohibitins on this interaction and show that the inhibition of their activity can effectively inhibit its oncogenic properties.