NCX1 expression is associated with disease prognosis and bortezomib sensitivity in human MM
Previously, our group found that NCX1 was highly expressed in myeloma cell lines (RPMI8226, KMS11, U266, MM1S) and human MM BM tissues[15]. To further investigate the impact of NCX1 on MM prognosis and BTZ sensitivity, we proceeded to collect BM tissues from 42 newly diagnosed myeloma patients who had received two or more courses of BTZ treatment. The clinical characteristics and parameters in the 42 newly diagnosed MM patients are summarized in Tables 1 and 2. We examined the protein expression of NCX1 in BM samples from 42 MM patients and 18 IDA by IHC. Consistently, NCX1 expression was significantly higher in MM BM than in corresponding IDA BM samples (Fig. 1a, b).
Table 1
Clinical feature of 42 newly diagnosed MM patients from CHINA.
Clinical feature
|
|
No. of cases(%)
|
Gender
Male
Female
Age (years)
<60
≥60
ISS Stage
Stage I
Stage II
Stage III
M component at diagnosis
IgG type
IgA type
IgM type
IgD type
Light chain type
Kappa type
Lambda type
|
|
21(50%)
21(50%)
15(36%)
27(64%)
7(17%)
16(38%)
19(45%)
18(43%)
11(26%)
0(0%)
2(5%)
11(26%)
23(55%)
19(45%)
|
Abbreviations: |
ISS: International Staging System. |
Table 2
Clinical and pathological characteristics of 42 newly diagnosed MM patients.
Clinical parameters
|
|
Mean ± SD
|
Albumin (g/L)
Hemoglobin (g/L)
Calcium (mmol/L)
β2M (mg/L)
LDH (U/L)
Creatinine (µmol/L)
IgG (g/L)
IgA (g/L)
IgM (g/L)
Serum kappa (g/L)
Serum lambda (g/L)
Urine kappa (mg/L)
Urine lambda (mg/L)
|
|
35.71 ± 6.37
95.13 ± 20.45
2.41 ± 0.27
6.82 ± 4.12
216.42 ± 108.50
129.97 ± 144.68
25.47 ± 30.45
7.31 ± 17.9
0.30 ± 0.29
26.96 ± 32.59
10.57 ± 21.87
1219.50 ± 2426.72
717.82 ± 1799.36
|
Abbreviations: |
β2M, beta-2-microglobulin; LDH, lactate dehydrogenase. |
Next, to explore the clinical and pathological role of NCX1 in MM, we correlated NCX1 expression with clinical parameters in MM patients. We found that the protein expression of NCX1 was positively corrected with serum calcium levels, beta-2-microglobulin (β2M)and the percentage of BM CD138 + cells (Fig. 1c-e). Interestingly, NCX1 expression was increased in BM tissues of MM patients with ISS stage 2 or 3 compared with ISS stage 1 (Fig. 1f). Subsequently, we explored the relationship between the protein expression of NCX1 and the survival of MM patients treated with BTZ by Kaplan-Meier survival curve. As shown in Fig. 1g, MM patients with low NCX1 protein expression had better median overall survival (OS) (24.6 months vs. 14.8 months, p < 0.05). Taken together, these findings suggest that NCX1 expression was elevated in MM BM tissues and correlated with disease progression in those patients receiving BTZ treatment.
NCX1 inhibition potentiates the sensitivity of bortezomib in MM cells
In previous study, we have determined that inhibition of NCX1 is able to inhibit MM cell proliferation and induce apoptosis. To further investigate the effect of NCX1 on the sensitivity of MM cells to BTZ, we added the NCX1-specific inhibitor KB-R7943 or knocked down NCX1 in two MM cell lines, RPMI8226 and KMS11 cells, respectively, and exposed them to 10 nM BTZ. The transfection efficiency of knockdown NCX1 in RPMI8226 and KMS11 cells was shown in Additonal 1a. First, CCK8 assay results showed that compared with each single agent alone, a more pronounced inhibition of cell proliferation was observed in RPMI8226 and KMS11 cells treated with KB-R7943 plus BTZ (Fig. 2a). In addition, a similar effect was observed in NCX1 knocking-down combined with BTZ treatment (Fig. 2b), indicating that inhibition of NCX1 synergized with BTZ to suppress MM cell viability. Second, we performed flow cytometry to assess MM cells apoptosis, and the results suggested that inhibition NCX1 using KB-R7943 or knockdown NCX1 enhanced the pro-apoptotic efficacy of BTZ in RPMI8226 and KMS11 cells (Fig. 2c-2h). Third, cell cycle results showed that inhibition of NCX1 increased BTZ-induced G2/M arrest and S-interphase shortening in RPMI8226 and KMS11 cells (Fig. 2i-2j). These data indicated that NCX1 inhibition could potentiate the BTZ sensitivity of MM cells.
High extracellular calcium([Ca2+]o) promotes NCX1 expression and autophagic flux in MM cells
It is well known that cellular calcium homeostasis and autophagy play important roles in BTZ sensitivity of MM cells[23, 24]. Our previous studies showed that extracellular calcium ([Ca2+]o) can activate NCX1, promote extracellular calcium influx, and increase the viability of MM cells[19]. Here, we would like to further investigate the potential effects of high calcium microenvironment activating NCX1 expression on autophagy and BTZ sensitivity in MM cells. As shown in Fig. 3a-3d, under basal conditions, [Ca2+]o increased NCX1 expression and induced a significant increase in autophagy marker proteins (ATG7, ATG5, and LC3B-II), and caused an increase in the clearance of p62, an autophagy cargo receptor protein[25], suggesting enhanced autophagy in RPMI8226 and KMS11 cells. Next, we transduced MM cells with mRFP-GFP-LC3 tandem fluorescent protein lentivirus to evaluated the extent of autophagosome and autolysosome formation. The numbers of autophagosome (yellow dots) and autolysosome (red dots) per cell were both significantly increased after incubation in the medium with higher calcium concentration, and more free red dots than yellow dots were seen (Fig. 3e-3i), suggesting that [Ca2+]o increases autophagic flux. Subsequently, transmission electron microscopy (TEM) showed that autophagosomes and autolysosomes were increased in RPMI8226 and KMS11 cells treated with CaCl2 (Fig. 3j-m). Moreover, as shown in Fig. 3k and 3l, the increased [Ca2+]o inhibited the sensitivity of BTZ in MM cells.
Inhibition of NCX1 reverses the effect of higher extracellular calcium increasing autophagic flux in MM cells
To further confirm the role of NCX1 in autophagy, we inhibited the expression of NCX1 using KB-R7943, and observed whether NCX1 inhibition can reverse the increase of autophagic flux induced by [Ca2+]o. As shown in Fig. 4a and 4c, KB-R7943 reversed the increase in autophagic marker proteins (ATG7, ATG5, and LC3B-II) expression and p62 clearance induced by [Ca2+]o. Then, to verify whether autophagic marker proteins reversed by KB-R7943 were due to decreased autophagic flux, we treated MM cells with CaCl2 and KB-R7943 or chloroquine (CQ). CQ is a late-stage autophagy inhibitor that prevents autophagosome-lysosome fusion[26]. We found that KB-R7943 or CQ can reverse the increased of autophagic flux induced by CaCl2 (Fig. 4e-h). Further confirmation of autophagic flux inhibition was obtained by TEM. NCX1 inhibition or autophagy inhibitor markedly elevated the number of autophagosomes, and reversed the autolysosomes increased by CaCl2 (Fig. 4i-l).
Of note, BAPAT, an intracellular calcium chelator[27], also showed inhibitory effects on increased autophagy marker proteins, autophagic flux and autophagosomes/autolysosomes induced by CaCl2, suggesting that NCX1 regulates autophagy by interfering with calcium homeostasis (Additional 2).
Besides, we also confirmed that CaCl2-reduced BTZ sensitivity in MM cells could be reversed by KB-R7943 or BAPAT (Fig. 4m,n). These results demonstrate that NCX1 appears to regulate autophagy and BTZ sensitivity by disturbing calcium homeostasis.
Inhibition of autophagy sensitizes high NCX1 MM cells to BTZ
Previous studies have reported that targeted inhibition of autophagy is an effective strategy to increase BTZ sensitivity in MM[28, 29]. To further explore the correlation between NCX1 and autophagy in the sensitivity of MM to BTZ, we overexpressed NCX1 in RPMI8226 and KMS11 cells by lentivirus (Additional 1b). Then, MM cells overexpressing NCX1 were treated with BTZ with or without autophagy inhibitor, CQ, for 48 hours. Contrary to the above results that inhibition of NCX1 increased BTZ sensitivity in MM cells (Fig. 2), overexpression of NCX1(oeNCX1) decreased the sensitivity of MM cells to BTZ. However, compared with oeNCX1 MM cells treated with BTZ or CQ alone, the combination of BTZ and CQ significantly increased the inhibition of cell viability in a concentration dependent manner (Fig. 5a, b). Meanwhile, colony formation assay was performed to determine the synergistic effect of BTZ and CQ on inhibiting the proliferation of oeNCX1 MM cells. As shown in Fig. 5c, e, compared with BTZ alone, CQ combined with BTZ showed a more obvious inhibitory effect on colony formation in oeNCX1 MM cells. In addition, the synergistic effect of BTZ and CQ was also confirmed on promoting apoptosis of oeNCX1 RPMI8226 or KMS11 cells. As shown in Fig. 5g, h, compared with oeNCX1 RPMI8226 or KMS11 cells treated with BTZ alone, increased apoptosis in those cells treated with the combination of BTZ and CQ. These outcomes displayed that autophagy inhibition reversed the BTZ-resistant effect of oeNCX1 MM cells.
NCX1 induces autophagy through non-classic NFκB signaling pathway in MM cells
Recent evidence highlighted NFκB-induced autophagy has a tumorigenic effect in most human cancers[30]. Considering that NFκB is the key target of BTZ[31], we proposed to determine whether NFκB can affect NCX1 induced-autophagy activation in MM cells. To explore this, we first investigated the effects on canonical and non-canonical NFκB pathways by overexpressing and knocking down NCX1 in MM cell lines. We found that overexpression of NCX1 significantly increased the expression of non-canonical NFκB-associated proteins (P100, P52, and RelB), but did not affect the canonical NFκB-associated proteins (P105, P50, P-P65, and P65) (Fig. 6a, c). In contrast, knocking down of NCX1 in RPMI8226 and KMS11 cell lines significantly reduced the expression of non-canonical NFκB-related proteins, but also had no effect on canonical NFκB-related proteins (Fig. 6b, d). In addition, we found that CaCl2 promoted the expression of non-canonical NFκB-associated proteins, which can be reversed by NCX1 specific inhibitor KB-R7943 (Fig. 6e-h).
To determine whether non-canonical NFκB mediates NCX1-enhanced autophagy, we added NFκB pathway inhibitor SN52 in NCX1-overexpression MM cells. Then, western blot was performed to detect the expression of P100, P52, RleB, ATG7, ATG5, P62 and LC3B-II/I. P100, P52 and RelB proteins were significantly reduced after using SN52 in MM cells. As expected, ATG7, ATG5, and LC3B-II/I were also decreased and the autophagy substrate P62 was increased in NCX1-overexpression MM cells following the addition of SN52 (Fig. 6i, j). Moreover, as revealed in Fig. 6k and 6l, SN52 attenuated NCX1 and BTZ- induced P52 and RelB nuclear translocation in MM cells. Consistently, TEM results showed that SN52 could reverse the effect of NCX1 /BTZ on increasing the number of autophagic bodies (Fig. 6m-p). In addition, by detecting the viability of MM cells, we found that SN52 reversed the inhibitory effect of NCX1/Ca2+ on BTZ sensitivity (Fig. 6q, r).
NFκB2 encodes p100/p52 protein, we constructed NFκB2-shRNA lentiviral system for targeted inhibition in NCX1-overexpression MM cells. Besides knocking down p100/p52 levels, sh-NFκB2 also impaired NCX1-induced autophagy activation (Fig. 6s, t). Of note, knocking down NFκB2 reduced the expressions of NCX1 in MM cells (Fig. 6s, t). We next assessed whether NFκB2 activation promoted the transcription of NCX1. Sequence analysis by JASPAR predicted that NFκB2 has binding sites at the NCX1 promoter. Importantly, this was confirmed by the luciferase reporter assay (Fig. 6u), suggesting the formation of positive feedback loop between NCX1 and activation of the non-canonical NFκB signaling pathway. These results collectively demonstrate the critical role of non-canonical NFκB signaling pathway in NCX1-regulated autophagy and BTZ sensitivity.
NCX1 inhibition sensitizes MM cells to bortezomib in vivo
Lastly, we investigated the effectiveness of KB-R7943 or NCX1 knockdown in combination with BTZ using a MM xenograft NCG mouse model. In this model, NCG mice were divided into two groups (20 in each group), one group was subcutaneously injected with KMS11-shCON and KMS11-shNCX1 cells, and the other group was injected with RPMI8226 cells. Subsequently, KB-R7943, BTZ or their combination were injected intraperitoneally when tumors were palpable subcutaneously (at day 10 after MM cell injection) (Fig. 7a). As demonstrated in Fig. 7b, c, the combination of NCX1-knockdown or KB-R7943 with BTZ caused a greater reduction in tumor growth than either single treatment alone. Tumor growth patterns in mice showed that inhibition of NCX1 and BTZ treatment effectively weakened tumor growth (Fig. 7c, d). Notably, no significant changes in body weight were observed during the treatment period(Fig. 7e), indicating no evidence of toxicity caused by the applied of NCX1-knockdown, KB-R7943 or BTZ and their combination. Moreover, we determined the expression of NCX1, CD138, Ki67, ATG5 and ATG7 in tumor sections by immunohistochemical staining. We observed that combination therapy significantly suppressed the expression of Ki67 compared with single treatment, and knocking down NCX1 not only inhibited the expression of autophagy related proteins ATG5 and ATG7, but also reversed the increased expression of ATG5 and ATG7 induced by BTZ (Fig. 7f,g). Overall, these results recapitulate the in vivo observations, suggesting that targeted inhibition of NCX1 may enhance the anti-MM activity of BTZ in vivo by inhibiting autophagy..