MYCNOS performs chemoresistant functions in vitro and in vivo of SCLC

To explore the potential mechanisms of chemoresistance in SCLC, we first performed RNA-seq analysis in SCLC chemoresistant cell line H69AR and its parent cell line H69. In the sequencing results, we found MYCNOS gene is highly expressed in H69AR (Fig. 1A). As a natural antisense transcript of MYCN, it plays a crucial role in promoting MYCN expression [34]. According to our previous research, MYCN plays an important role in SCLC drug resistance [31], we speculate that MYCNOS may also be a key gene for SCLC drug resistance. Then we confirmed the result in H69AR/H69 and another pair of SCLC resistant/sensitive cell lines H446DDP/H446 by qRT-PCR and western blot (Fig. 1B). Additionally, we verified that MYCNOS was highly expressed in chemoresistant patient-derived xenograft (PDX-Resistant) compared PDX-Sensitive models from the same SCLC patient (Fig. 1C), and mainly located in the nucleus of SCLC (Fig. 1D).

Fig. 1: MYCNOS exhibits chemoresistance and indicates poor prognosis in SCLC.

A Analysis of differentially expressed genes in H69AR and H69 by RNA-seq. B The expression of MYCNOS mRNA and protein in SCLC cells was examined by qRT-PCR and Western blot, respectively (Student t-test). C, D MYCNOS was assessed by qRT-PCR and IHC in paraffin-embedded sections of PDX-Resistant and PDX-Sensitive tissue samples from the same SCLC patient. E The chemosensitivity of SCLC cells was treated by DDP or VP-16 for 24 h and then detected by CCK8 assay following MYCNOS knockout or upregulation. F The Chemosensitivity of SCLC cells after upregulation of MYCNOS in MYCNOS-SE-deleted cells was evaluated. G, I The chemosensitivity of SCLC cells detected by Subcutaneous xenograft experiment after up and downregulation of MYCNOS (N = 5 mice for each group). The growth curve of tumor volumes and tumor weights were determined. H, J MYCNOS was assessed by qRT-PCR in the tumor. K, L Expression of MYCNOS in the 62 paraffin embedding SCLC tissues and the GSE60052 dataset compared to the normal lung tissues (non-tumor tissue adjacent to cancer). M Expression level of MYCNOS in SCLC tissues and the normal lung tissues measured by IHC staining. N, O Kaplan–Meier analysis of the overall survival in the 62 paraffin embedding SCLC tissues and the GSE60052 dataset based on MYCNOS. *P < 0.05; **P < 0.01 (Student t-test).

Next, we established MYCNOS up-regulated and down-regulated SCLC cell lines (Fig. S1A–D). Our result showed that following MYCNOS knockdown, the IC50 values of H446DDP and H69AR cells significantly decreased upon treatment with cisplatin (DDP) and etoposide (VP-16) for 24 h (Fig. 1E). Conversely, when MYCNOS up-regulated, an opposite effect was observed (Fig. 1F). Furthermore, the effects on cell proliferation of MYCNOS were also explored. EDU and CCK8 proliferation experiment results demonstrated that MYCNOS promotes cell proliferation of SCLC (Fig. S1E–J). Flow cytometry analysis demonstrated that the apoptotic response of cells to DDP and VP-16 was regulated by MYCNOS. After the knockdown of MYCNOS, cell apoptosis increased by 12.31% and 13.42% respectively in H69AR cells treated by DDP and VP-16, while in H446DDP cells, it was 7.04% and 8.61% respectively. In contrast, when MYCNOS was up-regulated, cell apoptosis decreased by 7.25% (DDP) and 8.59% (VP-16) respectively in H69, and 9.25% (DDP)and 8.29% (VP-16) respectively in H446 (P < 0.01) (Fig. S2A–D). Additionally, we observed changes in the cell cycle after MYCNOS modulation. After downregulating MYCNOS in H69AR and H446DDP cells, then treated with DDP, which promoted a significant accumulation of cells at G1 phase (increase by approximately 12.97% and 10.16% respectively), while the number of cells in the S phase decreased significantly (decrease by approximately 12.11% and 8.25% respectively). For VP-16 treated cells, downregulation of MYCNOS accelerated the transition of cells from the S phase (decrease by approximately 10.48% in H69AR) to the G2 phase, or arrest in the G1 phase (decrease by approximately 12.29% in H446DDP), resulting in more cell arrest in the G2 phase (increase by approximately 12.97% and 13.18% respectively) (Fig. S2E, G). Conversely, after upregulating MYCNOS in H69 and H446, the same treatment reduced G1 (decrease by approximately 9.31% and 10.59% respectively for DDP-treated cells) and G2 blocked cells (decrease by approximately 10.78% and 11.74% respectively for VP-16 treated cells) (Fig. S2F, H). We then used a nude mouse xenograft model to investigate the ability of MYCNOS to confer chemoresistance in SCLC. As shown in Fig. 1G, I, tumor growth was inhibited or promoted in the shMYCNOS or MYCNOS-OE group treated with PBS or drugs (DDP and VP-16) compared with the controls. qRT-PCR analysis of MYCNOS expression found it to be significantly lower or higher in tumor tissues formed from the shMYCNOS group or MYCNOS-OE group than those from controls (Fig. 1H, J). These results suggested that MYCNOS not only functions to confer chemoresistance in SCLC but also affects SCLC proliferation.

To investigate the clinicopathological features of MYCNOS expression in SCLC, qRT-PCR was performed in 62 tumor samples from SCLC patients. MYCNOS expression level was significantly higher in SCLC tumor tissues than those in normal counterparts (non-tumor tissue adjacent to cancer) (Fig. 1K). And we got the same results in the SCLC dataset GSE60052 including RNA-seq of 79 SCLC and 7 normal controls from China (Fig. 1L). Next, higher expression of MYCNOS proteins was observed in SCLC tissues through immunohistochemistry, and it was mainly located in the nucleus of SCLC (Fig. 1M). Further, Kaplan–Meier survival analysis showed that high MYCNOS expression was correlated with poorer patient survival (Fig. 1N, O). Table 1 summarizes the correlation between MYCNOS expression and clinicopathological parameters of SCLC patients. The data indicated that higher expression of MYCNOS in limited disease SCLC (LD-SCLC) than in extensive disease SCLC (ED-SCLC) (P = 0.036). However, no significant difference was observed with respect to gender (female and male), age (≤65 years and >65 years), and smoking history in our study. Taken together, our study indicated that MYCNOS overexpression in SCLC tissues was correlated with the stage and survival of SCLC patients.

Table 1 Association of MYCNOS expression and clinicopathological characteristics in 62 SCLC patients.MYCNOS is regulated by MYCNOS-SE as a chemoresistance-associated transcription element to SCLC

After determining the function and high expression of MYCNOS, we subsequently dived into the potential mechanism governing the elevated expression of MYCNOS. Previous studies have shown that epigenetic regulation plays an important role in tumor chemoresistance [35]. We found that Liu et al. used JQ1 to treat human and mouse SCLC cell line-derived xenograft tumors in previous studies, showing significant inhibition of tumor growth [36]. JQ1 is a small molecule inhibitor of BRD4, which can block the binding of BRD4 to H3K27ac [37]. To ascertain whether epigenetic modulation is involved in MYCNOS-mediated chemoresistance, we probed the influence of BRD4 inhibitors on this process. The results of qRT-PCR and western blot showed that JQ1 specifically down-regulated the expression of MYCNOS in a time- and dose-dependent manner (Fig. 2A–D). Since JQ1 can lower the expression of the super-enhancer-driven target gene by inhibiting transcription co-activation and elongation from testing the effect on the target gene [38]. Then we performed the H3K27Ac ChIP-seq and RNA-seq in H69AR cells, as described in our former published research, SE-associated genes were finally obtained, which may be associated with the chemoresistance of SCLC [27] (Fig. S3A–E). Among them, one segment of enhancers has attracted our attention, and we hypothesize that it significantly contributes to the upregulation of MYCNOS (Fig. S3A). Additionally, the Hi-C contact map of the MYCNOS locus in human SW480 cells was generated in the 3D Genome Browser (http://3dgenome.org) and highlighted that the super-enhancer region had direct interactions with the promoter region of MYCNOS (Fig. S3F) [39]. We designated this super-enhancer as MYCNOS-SE and prioritized its investigation in the context of SCLC. Furthermore, other published ChIP-seq data were analyzed, there is no H3K27Ac peak in the MYCNOS-SE region in GM12878 (human lymph blastocytes), K562 (human chronic myeloid leukemia cells), MCF-7 (human breast cancer cells), and HepG2 (human liver cancer cells) (Figure S3G). This observation underscores the potential specificity of MYCNOS-SE’s role in SCLC.

Fig. 2: MYCNOS is epigenetically regulated by MYCNOS-SE-E5 as a chemoresistance-associated transcription element in SCLC.

A–D qRT-PCR and Western blot were used to detect the levels of MYCNOS after H69AR and H446DDP cells were treated with various concentrations of JQ1 for 48 h or 1000 nM for the indicated times. (one-way ANOVA test). E The location of MYCNOS-SE, MYCNOS, and component enhancers is shown. F Luciferase reporter assays were performed to measure the enhancer activity of each component enhancer (one-way ANOVA test). G The expression of MYCNOS was examined by qRT-PCR and western blotting analyses in knockout MYCNOS-SE and wild-type cells. H The expression of GACAT3 was examined by qRT-PCRPCR, and MYCNOS protein expression was analyzed by western blotting in Knockout SE and wild-type cells. I The chemosensitivity of SCLC cells after deletion of MYCNOS-SE was explored by CCK8 assay. J The Chemosensitivity of SCLC cells after Upregulation MYCNOS in deletion of MYCNOS-SE cells was explored by CCK8 assay. K, L H69AR, and H446DDP cells were treated with or without 6 nM JQ1 for 6 h, the cells were then subjected to ChIP analysis using antibodies against H3K27ac. K, L The association with the promoter region (K) and SE region (L) of MYCNOS was quantified by qPCR. M, N Luciferase reporter assay was conducted to measure MYCNOS promoter activity (M) and E1 super-enhancer (N) activity in H69AR and H446DDP cells treated with 6 nM JQ1 for 6 h. *P < 0.05; **P < 0.01 (Student t-test).

According to our data, MYCNOS-SE is located on chromosomes (Hg38 15939172–15951595). We further segmented the super-enhancer region of MYCNOS into five constituents (E1–E5) (Fig. 2E), and constructed the MYCNOS-SE-E1, MYCNOS-SE-E3, MYCNOS-SE-E4, and MYCNOS-SE-E5 (containing the DNA sequences of the E1–E5 respectively) to generate the luciferase reporters for the luciferase reporter assay. Notably, we refrain from considering the E2 component’s knockout in subsequent studies due to its overlap with exon 1 of MYCNOS, which would result in concurrent MYCNOS knockout. Strong transcription-enhancing activity was observed in H69AR and H446DDP cells transfected with E1–E5 plasmids respectively compared to control plasmids, especially those transfected with the E5 plasmid (Fig. 2F). Subsequently, we aimed to delete the E5 constituents (chr2:15949201–15951595) which may play the most important role in vivo function of MYCNOS-SE from the endogenous locus. We used the CRISPR-Cas9 nuclease system to generate MYCNOS-SE-E5 knockout H69AR and H446DDP cells. The E5 constituents deletion led to a near-complete loss of MYCNOS mRNA and protein expression (Fig. 2G). As we all know, super-enhancers often regulate the expression of nearby genes. To ascertain the specificity of MYCNOS-SE’s regulation on MYCNOS, we evaluated RNU5E-7P and GACAT3 which are close to it on the same chromosome. However, RNU5E-7P is a pseudogene, we chose GACAT3 (chr2:16013928–16087201) for the next verification. The result showed the expression of GACAT3 is not affected in any way when the MYCNOS-SE-E5 knockout (Fig. 2H). These findings suggest that MYCNOS-SE directly regulates MYCNOS, thereby influencing MYCN expression.

Further, we detected whether MYCNOS-SE played an important role in the chemoresistance of SCLC. The results of the CCK8 assays demonstrated that the IC50 values of H69AR and H446DDP were significantly decreased in the MYCNOS-SE-E5 knockout group when treated with chemotherapeutic drugs, including DDP and VP-16, compared to the wild-type group (Fig. 2I). Similarly, the data from flow cytometry revealed that MYCNOS-SE-E5 knockout increased cell apoptosis, 9.9% (DDP) and 15.25% (VP-16) in H69AR, 11.35% (DDP) and 14.1% (VP-16) in H446DDP (P < 0.01) (Fig. S4A, C). Furthermore, we observed the cell cycle arrest of DDP-treated H69AR and H446DDP cells mainly occurred at the G1 phase (increase by approximately 10.51% and 18.64% respectively), and the cell cycle arrest of VP-16-treated primarily happened at the G2 stage (increase by approximately 9.6% and 12.92% respectively) (P < 0.01) (Fig. S4B, D). As we know, SE has very powerful biological functions, we next explored its potential biological functions in SCLC. As anticipated, the EDU proliferation and CCK8 assays indicated that MYCNOS-SE knockout significantly inhibited the proliferation of SCLC cells (Fig. S4E, G). To further examine whether MYCNOS-SE played its role through direct regulation of MYCNOS, we conducted rescue experiments. Initially, the expression of MYCNOS was up-regulated in H69AR and H446DDP with MYCNOS-SE-E5 knockout. Then CCK8 assay results suggested that cotransfection partially rescued the chemoresistance impaired by MYCNOS-SE-E5 knockout (Fig. 2J). Likewise, the trends in cell proliferation were also partially rescued (Fig. S4F, H), as well as the inhibition of cell apoptosis and cell cycle arrest (Fig. S5A, B). Taken together, these results suggest that MYCNOS-SE does play a crucial role in SCLC, especially in chemoresistance.

As anticipated, the ChIP-qPCR assay also demonstrated the significant association of H3K27ac with the promoter of MYCNOS and the MYCNOS-SE region, which was disrupted by JQ1 treatment (Fig. 2K, L). Which showed a significant association of BRD4 with the promoter and super-enhancer region of MYCNOS-SE. Subsequently, we assessed the impact of JQ1 on enhancer activity utilizing a luciferase reporter assay. Consistent with the results of ChIP-qPCR, the luciferase activity of MYCNOS promoter and MYCNOS-SE was repressed by JQ1 treatment (Fig. 2M, N). To sum up, these results underscore the sensitivity of MYCNOS expression, driven by MYCNOS-SE, to BRD4 inhibition.

To further validate the role of MYCNOS-SE in conferring chemoresistance in vivo, we employed a nude mouse xenograft model. The tumor growth was inhibited in the MYCNOS-SE-E5 knockout group treated with PBS or drugs (DDP and VP-16) (Fig. S5E, H, I). We also found the expression of MYCNOS significantly lower in the MYCNOS-SE-E5 knockout group (Fig. S5F). Complementing these findings, immunohistochemistry implied that the expression of MYCNOS in tumor tissues of nude mice was significantly reduced after MYCNOS-SE-E5 knockout (Fig. S5G). Collectively, these findings solidify the regulatory role of MYCNOS-SE over MYCNOS, positioning it as a chemoresistance-associated transcription element in SCLC.

MYCNOS-SE regulates MYCNOS expression synergistically by recruiting transcription factors CTCF and KLF15

To determine how MYCNOS-SE regulates MYCNOS and subsequently impacts the chemoresistance of SCLC, we conducted ATAC-seq analysis in H69 and H69AR. Five transcription factors, which potentially bind to MYCNOS-SE and activate MYCNOS enhancers, were identified: CTCF, KLF14, KLF15, NRF1, and ARNT2. Among them, CTCF, KLF15, and ARNT2 presented ATAC peaks and overlapped with a region characterized by an enrichment of H3K27ac marks, indicating that these three transcription factors are likely to be key regulators of MYCNOS (Fig. 3A). The results of qRT-PCR and western blot suggested that only CTCF and KLF15 were highly expressed in H69AR and H446DDP compared to their respective controls, H69 and H446, whereas the expression of ARNT2 did not differ markedly between the two pairs of cells (Fig. 3B, C). Based on these findings, we speculate that CTCF and KLF15 are most probably involved in chemoresistance of SCLC. Consequently, CTCF and KLF15 were chosen for further investigation.

Fig. 3: MYCNOS-SE synergistically regulates MYCNOS expression by recruiting transcription factors CTCF and KLF15.

A A joint analysis of CHIP-seq and ATAC-seq data was conducted to screen for transcription factors recruited by MYCNOS-SE. B qRT-PCR detected the expression of CTCF, KLF15, and ARNT2 in SCLC cell lines. C Western blot detected the expression of CTCF and KLF15 in SCLC cell lines. D H69AR and H446DDP were subjected to ChIP analysis using antibodies against CTCF and KLF15. The association with the promoter region and SE of MYCNOS was then quantified by qPCR. E–H H69AR and H446DDP were subjected to ChIP analysis using antibodies against H3K27ac, following the knockdown of CTCF and KLF15. The association with the promoter region (E, G) and SE (F, H) of MYCNOS was quantified by qPCR. I, J Luciferase reporter assay detected the activity of MYCNOS promoter and SE region with CTCF and KLF15 knockdown. P < 0.05; **P < 0.01 (Student t-test).

According to the ATAC-seq and ChIP-seq analysis, we discerned that CTCF binds to both the promoter region and the E3 super-enhancer region of MYCNOS. Furthermore, the E1 constituent’s region of MYCNOS-SE contained another binding site of CTCF. Similarly, we observed that the promoter region and the super-enhancer region of MYCNOS contained multiple binding sites of KLF15. The top two scores binding sites of KLF15 from the promoter region and the super-enhancer region (located in the E1 and E4 constituents) of MYCNOS respectively were selected for further study. To elucidate the direct regulation of MYCNOS by CTCF and KLF15 at the transcriptional level via MYCNOS-SE, we initially performed ChIP-qPCR and observed a significant association of both CTCF and KLF15 with the promoter and super-enhancer regions of MYCNOS (Fig. 3D). We next constructed up-and down-regulated SCLC cell lines of CTCF and KLF15 (Fig. S6A, B). Then we conducted ChIP with an antibody against H3K27ac followed by qPCR. The results showed that CTCF knockdown increased H3K27ac enrichment at the promoter and super-enhancer regions of MYCNOS (Fig. 3E, F), whereas KLF15 knockdown decreased H3K27ac enrichment at the promoter and super-enhancer regions of MYCNOS (Fig. 3G, H). Here, we noticed the downregulation of CTCF did not curtail the enrichment of H3K27ac at the MYCNOS promoter and super-enhancer regions as expected. Such a result may be induced by the correlation between CTCF and chromatin accessibility. Knockdown of CTCF led to enhanced chromatin accessibility, leading to more H3K27ac enrichment [40, 41]. Nevertheless, CTCF’s dual role as a transcription factor and chromatin regulator is also evident in our study. Our data showed that CTCF and KLF15 were significantly associated with the promoter and super-enhancer regions of MYCNOS. Downregulation of CTCF and KLF15 genes resulted in inhibited MYCNOS expression at both mRNA and protein levels (Fig. S6C–F), whereas the downregulation of MYCNOS did not affect the expression of these two transcription factors (Fig. S6G–I). This indicated that MYCNOS serves as a downstream target gene of CTCF and KLF15.

To further clarify the direct transcriptional regulation of MYCNOS by CTCF and KLF15 via the MYCNOS-SE, dual luciferase reporting experiments were conducted. The results showed that the luciferase activity of the MYCNOS promoter and super-enhancer was suppressed upon knockdown of CTCF and KLF15 (Fig. 3I, J). Analysis of our ATAC-seq and CHIP-seq, we subsequently generated mutant luciferase reporter with the mutant sequence of the conserved CTCF and KLF15-binding sequence. The results suggested that the luciferase activity in the mutant group significantly diminished compared to the wild-type group (Fig. 4A–E). In conclusion, these results suggested that CTCF and KLF15 are key transcription factors to maintain the MYCNOS-SE activity, thereby promoting MYCNOS transcription.

Fig. 4: CTCF and KLF15 interact with the SE/promoter region and with each other in promoting their transcription functions.

A, B Schematic representation of a 631 bp and 717 bp region of the binding sites*2 (from chr2:15,942,870–15,943,501) and binding sites*1 (from chr2:15,939,172–15,939,889) respectively, or the mutant alleles of CTCF within the MYCNOS promoters and SE regions. The H69AR and H446DDP cells were transfected with the indicated plasmids for 48 h. Luciferase reporter assay was performed to detect the activity of the MYCNOS promoter and SE region. C–E Schematic representation of a 24,898 bp region (the binding site in the promoter region, spanning chr2:15,918,350–15,943,248), a 717 bp region of the binding sites*1 (the binding site in the SE region, spanning chr2:15,939,172–15,939,889) and a 1981 bp region of binding sites*2 (the binding site in the SE region, spanning chr2:15,943,500–15,945,481), or the mutant alleles of KLF15 within MYCNOS promoters and SE regions. The H69AR and H446DDP cells were transfected with the indicated plasmids for 48 h. Luciferase reporter assay detected the activity of MYCNOS promoter and SE region. F–K Following knockdown or overexpression of CTCF and KLF15, qRT-PCR and Western blot were used to observe their impacts on each other. L Luciferase reporter assay was performed to measure the activity of MYCNOS promoter after transfection with CTCF or KLF15 plasmids. *P < 0.05; **P < 0.01 (Student t-test or one-way ANOVA test).

As far as we know, super-enhancer generally recruits multiple transcription factors and performs its function through the synergism between transcription factors. We next explore the interplay between CTCF and KLF15. Both qRT-PCR and Western blot analyses revealed that upregulation or downregulation of either CTCF or KLF15 impacted the expression levels of the other and their target gene, MYCNOS (Fig. 4F–K). Furthermore, the dual luciferase reporting experiments revealed that concurrent transfection of CTCF and KLF15 luciferase plasmids significantly enhanced the luciferase activity within the MYCNOS promoter region compared to transfection with either plasmid alone (Fig. 4L). The above results suggest a positive correlation between CTCF and KLF15, which synergistically activates MYCNOS transcription. Furthermore, we performed qRT-PCR and immunohistochemical analysis to evaluate the potential chemoresistant role of CTCF and KLF15 in SCLC. We observed the expression levels of CTCF and KLF15 in chemoresistant PDX models of SCLC were higher than those in sensitive tissues, and both mainly located in the nucleus of SCLC (Fig. S6J, K).

We further examined the expression of CTCF and KLF15 in 62 SCLC tumor tissues and their normal counterparts (Fig. S7A). Additionally, we analyzed the data of GSE60052 for further insights (Fig. S7B). Notably, CTCF expression level was markedly elevated in SCLC tumor tissues (P < 0.01). Conversely, the expression level of KLF15 varied between our SCLC tumor tissues and GSE60052, potentially owing to limitations in sample size. Kaplan–Meier survival analysis revealed that the SCLC patients with high expression of CTCF and KLF15 had a shorter OS compared to those with high expression of either alone (Fig. S7C, D). Furthermore, a significant positive correlation was discerned between the expression levels of CTCF and KLF15 was also observed (Fig. S7E). Immunohistochemical analyses confirmed enhanced expression of CTCF and KLF15 proteins in SCLC tissues, where they were predominantly localized within the nucleus of tumor cells (Fig. S7F). Taken together, these results indicated that CTCF and KLF15 are overexpressed in drug-resistant samples of SCLC and are recruited by MYCNOS-SE, then promote the transcription of MYCNOS.

MYCNOS promotes MYCN expression through the notch pathway mediating chemoresistance in SCLC

Based on the above findings, we further explored the possible downstream signaling pathway that mediated by MYCNOS. Through analysis of our RNA-seq and CHIP-seq data, we enriched some tumor-related signaling pathways and graphed the top signaling pathways, including the notch signaling pathways (Fig. 5A). Our previous study reported that MYCN regulates HES1 via the Notch pathway, thereby contributing to the chemoresistance of SCLC [31]. In this study, we also found MYCN and HES1 mainly located in the nucleus of SCLC, and both highly expressed in chemoresistant PDX models, which means that both of them are related to SCLC chemoresistance (Fig. S7G–J). Interestingly, MYCNOS as a cis-antisense gene of MYCN, and previous studies have shown that there is always a mutual regulation relationship between them, since MYCNOS-SE can regulate MYCNOS, we next investigated whether MYCN is also affected after MYCNOS-SE-E5 knockout. It was discovered that the knockout of MYCNOS-SE resulted in reduced expression of MYCN, while upregulation of MYCNOS reversed the decreased expression of MYCN (Fig. 5B–D). Which suggested that MYCNOS-SE direct regulation of MYCNOS, then affects the expression of MYCN. Therefore, we speculate that MYCNOS-SE promoted the transcription of MYCNOS, thereby increasing the expression of MYCN through the notch signaling pathway, which is implicated in the chemoresistance of SCLC.

Fig. 5: MYCNOS interacts with MYCN and promotes chemoresistance through the Notch Pathway in SCLC.

A Top 20 enriched pathways from KEGG analysis for SE target genes. B Expression of MYCN after MYCNOS-SE knockout. C Expression of MYCN after transfection of the MYCNOS plasmid in MYCNOS-SE knockout cells. D.Expression of MYCNOS after transfection of the MYCN plasmid in MYCNOS-SE knockout cells. E–G qRT-PCR and Western blot detected the expression of MYCN and HES1 in SCLC cell lines with MYCNOS upregulation or downregulation. H Luciferase reporter assay measuring MYCN promoter activity after MYCNOS knockdown. I Proximity ligation assay [52] Of endogenous MYCN and MYCNOS in H69AR cells. J Reciprocal coimmunoprecipitation of endogenous MYCN and MYCNOS in H69AR cells. K, L Western blot and qRT-PCR were performed to assess the effects on MYCN expression after point mutation or deletion that abolishes the coding potential of MYCNOS. M Correlation between MYCNOS and MYCN in SCLC tissues (GSE60052 dataset). N Correlation between MYCNOS and MYCN in SCLC tissues (CCLE database). O Co-amplified of MYCNOS and MYCN in SCLC cell lines (CCLE database). P, Q Chemosensitivity of SCLC cells after MYCN and HES1 reversal while knocking down MYCNOS or MYCNOS-SE knockout. P < 0.05; **P < 0.01 (Student t-test).

We next observed the regulation of MYCN and HES1 by MYCNOS in SCLC cells. The results demonstrated that the upregulation of MYCNOS increased the mRNA and protein expression of MYCN and HES1, whereas the downregulation of MYCNOS significantly reduced the expression of both (Fig. 5E–G). Additionally, after overexpressing or downregulating MYCN, we found that it could not affect the expression of MYCNOS, and the rescue experiment results showed that the reduction of HES1 expression caused by downregulation of MYCNOS could be reversed by overexpression of MYCN (Fig. S7K–M). This further indicated that MYCNOS indirectly regulated HES1 through positive regulation of MYCN. It is important to note that MYCN is not expressed at the protein level in H446DDP and H446 in our previous studies. It was found that the downregulation of MYCNOS inhibited MYCN promoter activity (Fig. 5H), which aligns with the previous studies on human neuroblastoma [42, 43]. Proximity ligation assay (PLA) and CO-IP experiment demonstrated that direct combination between MYCNOS and MYCN (Fig. 5I, J). In addition, these effects were not affected by point mutation or deletion that abolished the coding potential of MYCNOS (Fig. 5K, L), indicating that its regulation of MYCN transcriptional activity and expression levels does not depend on whether it has protein-coding functions. Notably, previous studies suggested that MYCNOS is often co-amplified and co-expressed with MYCN [44]. To further investigate the correlation between MYCNOS and MYCN in SCLC, we analyzed the GSE60052 dataset and the CCLE database. The results showed a positive correlation between MYCNOS and MYCN, along with co-amplification in SCLC cell lines (Fig. 5M–O). In summary, MYCNOS binds to MYCN and regulates it at both the transcriptional and post-transcriptional levels in SCLC. Next, rescue experiments were performed to clarify the role of the notch pathway mediated by the MYCNOS. CCK8 assay results suggested that cotransfection could partially rescue the impaired chemoresistance induced by shMYCNOS (Fig. 5P). Similarly, overexpression of HES1 plasmids in MYCNOS-SE knockout cell lines also partially rescued the chemoresistance deficit caused by MYCNOS-SE knockout (Fig. 5Q). These findings indicated that MYCNOS initially promotes MYCN expression, which subsequently upregulates the notch pathway key gene HES1, thereby contributing to chemoresistance in SCLC.

MYCNOS is a promising efficacy marker for NOTCH inhibitors

Since MYCNOS plays a role through the notch pathway, we next investigated the impact of a notch pathway inhibitor on this regulatory mechanism. Consistent with our prior research [31], we first observed the expression of MYCNOS and MYCN at the protein level in SCLC PDX models [31] (Fig. 6A). Then we employed NOTCH inhibitor, inhibitor Mastermind Recruitment-1 (IMR-1), to treat various SCLC cell line [45]. The results showed that the IC50 values for H69 and H69AR cells, which exhibit high expression MYCNOS, were significantly lower upon treatment with IMR-1 compared to those of other cell lines with low-expression MYCNOS (Fig. 6B). Based on these findings, we posit that MYCNOS may emerge as a more promising and predictive biomarker than MYCN, which is absent in the majority of SCLC cell lines.

Fig. 6: MYCNOS as a more promising efficacy marker for NOTCH inhibitors than MYCN.

A MYCNOS protein expression in different cell lines and PDX samples B The Chemosensitivity of different SCLC cell lines. C The Chemosensitivity of SCLC Cells following MYCNOS overexpression. D Sensitivity of IMR-1 detected by subcutaneous xenograft experiment following MYCNOS upregulation in H446 (15 mg/kg). The growth curve of tumor volumes and tumor weights were determined. E, F Representative images of tumors from mice on day 92 after re-engraftment of PDX1- R-resistant tumor tissue and PDX2-S, along with the growth curve of tumor volumes and survival time of experimental PDX model mice receiving the indicated treatments. G qRT-PCR detected the expression of HES1 in each group of mice (one-way ANOVA test). H Hypothesis diagram of MYCNOS-SE and its interaction with two transcription factors. P < 0.05; **P < 0.01 (Student t-test).

To further verify our conjecture, we overexpressed MYCNOS in H446 that did not express MYCN, then observed the efficacy of IMR-1both in vitro and in vivo. The response to IMR-1 in the MYCNOS overexpression group was notably enhanced compared to the control group (Fig. 6C, D). Additionally, we selected two PDX models with distinct MYCNOS/MYCN expression profiles to investigate the therapeutic efficacy of IMR-1 in SCLC. The results implied that the tumor shrank significantly in MYCNOS high expression with the treatment of the IMR-1 group compared to the MYCNOS low-expression PDX model. However, the most pronounced efficacy observed with combined chemotherapy (DDP and VP-16) is the best (Fig. 6E, F). qRT-PCR analysis revealed a significantly lower expression of HES1 in tumor tissues derived from the IMR-1 treated group and the combined therapy group compared to controls (Fig. 6G). HE staining of various organ tissues from the PDX models showed that neither IMR-1 monotherapy nor combined chemotherapy exhibited any toxicity to the primary organs (Fig. S7N).