Abstract
There is limited understanding of the phenomenon of reversible drug resistance, which is characterized by tumor cells regaining sensitivity when the drug is changed or withdrawn after a period of drug resistance. This phenomenon is usually not associated with genetic alterations of tumor cells. In this study, reversible resistant state was induced by alectinib in EML4-ALK mutant lung cancer cell. By performing RNA sequencing on reversible drug-resistant cell line to examine changes in transcriptional profile, significant change in CCN1 was detected after withdrawal and repeated administration of alectinib. Targeting CCN1 resulted in inhibition of tumor cell proliferation and angiogenesis, and restoration of sensitivity to alectinib in reversible drug-resistant cells. Further studies revealed that CCN1 could affect the expression of VEGFA by affecting AKT phosphorylation, and the change of NF-κB could impact the activation of CCN1-AKT-VEGFA pathway. Suppressing NF-κB or CCN1 receptor could improve the sensitivity to alectinib, further suggesting that NF-κB and CCN1 might play a key role in overcoming reversible drug resistance.
Subject terms: Cancer therapy, Lung cancer
Introduction
Lung cancer is the leading cause of cancer mortality worldwide [1]. Non-small cell lung cancer (NSCLC) accounts for ~85% of lung cancers and exhibits a remarkable frequency of oncogenic driver alterations. Over the past two decades, the discovery of oncogenic driver alterations and the development of matched molecularly targeted therapies have contributed to a substantial reduction in NSCLC mortality [2, 3]. A variety of ALK gene alterations, including ALK fusion, point mutations, and amplification, have been identified across various tumor types. The oncogenic ALK can be activated by ALK fusions in several cancers, such as NSCLC, anaplastic large cell lymphoma (ALCL), as well as by point mutations in neuroblastoma and anaplastic thyroid cancer [4, 5]. Additionally, ALK amplification as a negative prognostic factor has also been reported in certain cancers [6, 7].
Chromosomal rearrangements involving ALK have been identified as an oncogenic driver in 3–7% of patients with NSCLC and have been associated with non-smoking status, younger age, and adenocarcinoma histology [8]. Although the development of ALK tyrosine kinase inhibitors (TKIs) has revolutionized the treatment landscape for ALK-driven NSCLC patients, acquired resistance is inevitable and remains a critical unmet challenge [9, 10].
Much effort has been devoted to elucidating the mechanisms of acquired resistance in order to develop subsequent therapies. The resistance mechanisms of ALK TKIs can be categorized as either “on-target” such as ALK aberrations, or “off-target” involving the activation of bypass pathways, downstream pathways, or phenotypic transformation [10, 11]. Despite the improved understanding of resistance mechanisms, little is known about the changes that occur during the evolution of acquired resistance. Previous studies have reported the “reversible resistance phenomenon” in the clinical setting, which manifested by the fact that some patients resistant to targeted agents could regain sensitivity after a period of treatment discontinuation [12–15]. The “reversible resistance phenomenon”, which could be reverted, may indicate an incompletely resistant state resulting from non-genomic mechanisms. The study of reversible resistance would undoubtedly promote a better understanding of resistance transformation and optimize clinical medication.
In this study, we established an ALK-mutant lung cancer cell line that was unstably resistant to alectinib, the second-generation ALK inhibitor. Genomic sequencing did not identify known resistant mutations. Therefore, RNA sequencing was performed to investigate potential changes in the expression landscape. We found that CCN1, which encoded a secreted matricellular protein, showed obvious changes when alectinib was withdrawn or rechallenged. Further experiments revealed that CCN1 was required for growth and angiogenesis ability via AKT-VEGFA signaling in unstable resistant cells. Targeting CCN1 could reverse resistance to alectinib. Our study provides new insights into the therapeutic targets or resistance biomarkers of ALK mutant NSCLC.
Materials and methods
Cell culture and reversible drug resistance cell line construction
The human lung cancer cell line H3122 was obtained from the American Type Culture Collection (ATCC). Short tandem repeat (STR) analysis was performed to authenticate the cells, and routine screening for mycoplasma contamination was conducted. All cells were cultured under standard conditions using RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (P/S).
To induce reversible resistance, H3122 cells were continuously exposed to sequentially increasing concentrations of alectinib. After reaching a concentration of 2 μM, the cells were maintained in culture medium containing 2 μM alectinib for 3 months. The reversible drug resistance of the alectinib-treated cell line was confirmed by assessing drug sensitivity both during the alectinib maintenance period and 2 weeks after alectinib withdrawal.
Drug holiday-administration-holiday experimental cycles
In order to clarify the mechanism of reversible drug resistance, we performed a cycle of drug holiday-administration-holiday experiment. Briefly, the reversibly resistant cell lines were pre-maintained at 2 μM alectinib, and cells were collected at baseline. The period duration of drug withdrawal, rechallenge, and re-withdrawal was each 2 weeks, and cells were collected on day 3, day 7, and day 14 of each period, respectively. Cell samples harvested during the process were used for subsequent RNA sequencing.
Cell transfection
Lentiviral expression plasmids were synthesized from Tsingke Biotechnology (Beijing, China). These included the pLVX-CMV-PURO vectors subcloned with CCN1. The CCN1 plasmid was transfected into HEK293T cell line together with the packaging and envelope plasmids (psPAX2, pMD2.G) using Lipofectamine™ 3000 Reagent (#L3000008, Invitrogen). Lentivirus-containing cell culture supernatant was then collected, filtered through a 0.45 μm syringe, and immediately added to the target cells along with polybrene. Pooled target cell populations were then selected with 2 μg/ml of puromycin. CCN1 siRNA was synthesized from RIBO Biotechnology (Guangzhou, China) and transfected into cells using Lipofectamine™ RNAiMAX Reagent (#13778075, Invitrogen). The sequence of siCCN1 is CCAGAAATGTATTGTTCAA.
Whole-genome sequencing and RNA sequencing
Whole-genome sequencing (WGS) was performed in H3122 and H3122 alectinib resistant (AR) cells to identify gene alterations that emerged following long-term treatment with alectinib. Transcriptome changes during holiday-administration-holiday experiments were confirmed by RNA sequencing. Genomic DNA and total RNA were extracted using the QIAamp DNA Mini Kit (#51306, Qiagen) and MiniBEST Universal RNA Extraction Kit (#9767, TaKaRa) according to the manufacturer’s instructions, respectively. WGS libraries were constructed using MGIEasy Fast FS Library Prep Set (#940-001197-00, MGI Tech, China). RNA-seq libraries were constructed using MGIEasy Fast RNA Library Prep Set (#940-000890-00, MGI Tech, China). WGS and RNA sequencing procedures were performed on a DNBSEQ-T7RS sequencer (MGI Tech, China) with 150-bp paired-end reads according to the manufacturer’s recommendations. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive [16] in National Genomics Data Center [17], China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA-Human: HAR009333), which are publicly accessible at https://ngdc.cncb/ac.cn/gsa-human.
Sequencing data analysis
The WGS sequenced reads were mapped to the reference genome (GRCh37) using BWA [18] after removing adaptor and low-quality reads using fastp [19]. Duplicated reads were identified by unique identifiers and were ignored in subsequent analysis. Single-nucleotide variants (SNVs) and indels were identified using MuTect2 [20] and annotated using VEP [21]. Structural variation (SV) was identified using Manta [22], and copy number variations (CNVs) were identified using GATK [23]. For RNA sequencing, the expression of the transcripts was quantified against the reference genome (GRCh37) using Salmon [24]. maSigPro [25, 26] was used to analyze the time-course expression data to find genes most associated with drug treatment.
Real-time quantitative PCR (qPCR)
Total RNA was extracted from tumor cells using MiniBEST Universal RNA Extraction Kit (#9767, TAKARA) and reverse transcribed using PrimeScript™ RT reagent Kit with gDNA Eraser (#RR047A, TAKARA). TB Green™ Premix Ex Taq™ (Tli RNaseH Plus) (#RR820A, TAKARA) was used as a fluorescent DNA binding dye, and RT-PCR was performed on a Biorad system (Applied Biosystems, California, USA). Gene expression was normalized to that of GAPDH. The following primers were used to amplify CCN1 [5′-AATGGAGCCTCGCATCCTATA-3′(forward) and 5′- TTCTTTCACAAGGCGGCA-3′(reverse)], CCN2 [5′-CCCTCGCGGCTTACCGACTGG-3′(forward) and 5′-CACAGGTCTTGGAACAGGCGC-3′(reverse)] and GAPDH [5′- GTCTCCTCTGACTTCAACAGCG-3′(forward) and 5′-ACCACCCTGTTGCTGTAGCCAA-3′ (reverse)].
Cell viability and colony formation assays
Cell viability was assessed using the Cell Counting Kit-8 kit (#C0005, Dojindo). Cells were collected from the suspension medium, spun down at 300 g for 5 min, and resuspended in fresh RPMI medium. 1000 cells per well were plated in 96-well plates. CCK-8 solution (10 μl) mixed with medium was added to each well on a given day. After incubation at 37 °C for 2 h, the absorbance was measured at 450 nm. For the drug sensitivity analysis, 5000 cells per well were plated in 96-well plates. Cells were treated with ten different concentrations of inhibitors in serial dilutions or vehicle alone to a final volume of 100 μl per well. After 72 h, Cell Counting Kit-8 was added to each well. The half-maximal inhibitory concentration (IC50) values were calculated using GraphPad Prism 10 at 50% inhibition. For colony formation assays, 1000 cells were seeded in 6-well plates. After 24 h, 100 nM of alectinib was added to maintain the resistant state of H3122 AR cells in the whole process. Cell colonies were fixed with methanol and stained with 0.3% crystal violet before being photographed. Each experiment was replicated three separate times.
Western blotting
Total protein was extracted using RIPA lysis buffer with protease inhibitor cocktail. Proteins from each group were loaded onto SDS-PAGE gels and separated before transfer to PVDF membranes. After blocking with skim milk for 1 h at room temperature, the membrane was incubated with primary antibodies against each target protein overnight at 4 °C. The membrane was then incubated with secondary antibodies for 1 h at room temperature. Primary antibodies included ALK (#3633), pALK (#3341), pAKT (#9271), AKT (#4691), pERK (#4370), ERK (#4695), STAT3 (#4904), pSTAT3 (#9143), CCN1 (#14479), CCN2 (#10095), FAK (#3285), pFAK (#8556), NF-κB (#8242), pNF-κB (#3033), VEGFR2 (#2479), YAP (#14074), TGF-β (#3711) from Cell Signaling Technology and VEGFA (#ab46154) from Abcam. Blots were probed with antibodies against GAPDH (Proteintech, #10494-1-AP) as a loading control.
Scratch wound healing assays
Cells with different treatments were cultured in six-well plates at 37 °C. Scratch wounds were made by using the fine end of 10 μl pipette tips. Images of migrated cells were taken at different times and analyzed using the Image J software.
Tube formation assay
HUVECs were cultured for tube formation assay. Supernatants from tumor cells with different treatments were collected and stored at −80 °C. The µ-Slide plates (#81506, Ibidi) were coated with Matrigel (#354234, Corning) and incubated at 37 °C for 30 min to allow gelation. HUVECs were diluted with various supernatants and plated at a density of 10,000 cells per well. Cells were incubated at 37 °C with 5% CO2, and images were captured within 6 h. The tubular structures were photographed under light microscope and analyzed using the Image J software.
Statistical analysis
GraphPad Prism 10 was used to generate graphs and statistics. Unpaired Student’s t-test was used to analyze the differences between the two groups. The results are presented as means ± standard. Each experiment was replicated three separate times. Statistical significance was defined as a two-sided P value < 0.05.
Results
High-concentration of alectinib induced reversible drug resistance in EML4-ALK mutant lung adenocarcinoma cells
To construct adenocarcinoma cells resistant to alectinib, high concentration of alectinib was used to incubate human H3122 cells carrying EML4-ALK fusion. The concentration of alectinib was gradually increased and maintained at 2 µM when the cancer cells could survive and grow with TKI treatment. After the cells could well tolerate with 2 µM of alectinib, the mean IC50 of alectinib in alectinib-treated H3122 cells (H3122 AR) reached 614.1 ± 154.7 nM, over 10-fold higher than that of the parental H3122 cells (10.71 ± 3.869 nM) (Fig. 1A). WGS was performed in H3122 and H3122 AR cells to identify gene alterations following long-term treatment with alectinib and investigate potential genomic-resistance mechanisms. Although a great number of genetic mutations were identified, most alterations did not change the amino acid sequence (Tables S1 and S2; Fig. S1A). Only 62 nonsynonymous mutations were detected and did not include any ALK-dependent mutations leading to resistance or any off-target mechanisms such as HER2 amplification (Table S3 and Fig. S1B). Alectinib was withdrawn to determine whether the reversible resistant phenotype was constant. The IC50 decreased after 1 week of drug withdrawal (326.0 ± 85.84 nM) and became lower after 2 weeks (80.28 ± 23.07 nM) (Fig. 1A). The improved sensitivity after alectinib withdrawal suggested that resistance to alectinib was unstable. Therefore, we confirmed that the resistant state of H3122 AR was reversible and did not result from genetic aberrations.
Fig. 1. CCN1 changed as ALK inhibitor treatment altered.
A Alectinib sensitivity detection in H3122 cells, H3122 AR cells, and AR cells after alectinib withdrawal. B Alectinib sensitivity detection in H3122 AR before and after alectinib withdrawal, and alectinib rechallenge. C MaSigPro analysis for samples collected at different times during alectinib holiday-administration-holiday period. D Top 50 genes in clusters 5, 7, and 9 with similar RNA changes. E CCN1 RNA level detection at different times during alectinib holiday-administration-holiday period. F CCN1 protein level detection at different times during alectinib holiday-administration-holiday period. Error bars reflect mean ± standard deviation.
RNA sequencing found that CCN1 changed as ALK inhibitor treatment altered
To confirm the observation of the reversible resistance to alectinib, we performed drug holiday-administration-holiday experimental cycles to the H3122 AR cells. As shown in Fig. 1B, after 2 weeks of drug withdrawal, increased sensitivity to alectinib was observed (IC50 fell from 940.4 ± 168.9 nM to 68.80 ± 13.36 nM). Then IC50 returned to 861.9 ± 196.3 nM after 2 weeks of alectinib rechallenge and decreased to120.6 ± 10.71 nM after repeated alectinib withdrawal (Fig. 1B). Therefore, the reversible resistance phenomenon was determined in response to TKI treatment. Given that such resistance is not associated with genetic alterations, non-genetic regulation might play an important role during the process. Thus, RNA samples were collected across the drug holiday-administration-holiday experiment on day 0, day 3, day 7, and day 14 for transcriptome analysis. We hypothesized that there should be differences in the transcriptome of the cells during the drug holiday-administration-holiday experimental cycles, with these differences remaining consistent over time. According to the gene expression profiles throughout the time series, we employed MaSigPro to categorize the genes into clusters, ultimately identifying 9 distinct clusters exhibiting varied temporal patterns. Further analysis of these clusters led to the identification of three gene clusters (as shown in Fig. 1C, clusters 5, 7, and 9) that were most likely implicated in drug resistance. These clusters demonstrated a tendency to be down-regulated during drug interruption and up-regulated upon drug rechallenge, suggesting a possible direct role in mediating drug resistance. Given the similar temporal profiles of these three clusters, we consolidated them to investigate the genes that might be important on this reversibly resistant phenomenon. Among the top 50 genes in these 3 clusters, CXCL family, such as CXCL8, CXCL2, and matrix metalloproteinases (MMPs), such as MMP9, MMP10, emerged. Two members of the cellular communication network (CCN) family, CCN1 and CCN2, were noticed (Fig. 1D). Previous studies have reported that CCN1 and CCN2 are associated with tumor growth, migration, and drug resistance [27–29]. However, the role of CCN family in lung cancer remains controversial. qPCR confirmed the expression changes of CCN1 and CCN2 (Figs. 1E and S2A), agreeing with the transcriptomic data. CCN1 protein level also mirrored the changes on RNA level (Fig. 1F). However, CCN2 protein did not change accordingly (Fig. S2B). Taken together, we found that CCN1 was an important factor that tended to be downregulated after drug withdrawal, upregulated after drug rechallenge, and downregulated again after drug withdrawal in both RNA and protein levels.
Tumor cell proliferation and angiogenesis can be inhibited and drug sensitivity to alectinib can be enhanced by inhibiting CCN1 in H3122 AR cells
Considering that the resistant state was rapidly changed without alectinib induction, genetic knockdown of CCN1 by siRNA was applied to investigate CCN1 function. By knockdown of CCN1, the colony forming capacity fell by 39% in H3122 AR cells (p = 0.0011, Fig. 2A, B). Significant inhibition of cell growth viability was also observed (Fig. 2C). Tube formation assay showed that CCN1 knockdown decreased the vascular length of H3122 AR cells (p = 0.0019, Fig. 2D, E). However, there was no significant difference in migration ability between siCCN1 H3122 AR cells and control cells (p = 0.6535, Fig. S3). In addition, we found that knockdown of CCN1 could improve the sensitivity to alectinib in H3122 AR cells (median IC50, 310.7 ± 11.32 nM vs 63.22 ± 11.66 nM, p < 0.0001, Fig. 2F, G). CCN1 mainly acts by binding to the integrin receptor [29, 30]. Therefore, SB273005, an integrin-targeting agent, was used for combination therapy. As shown in Fig. 2H, high concentration of SB273005 did not obviously affect the survival of H3122 AR cells. However, when combined with SB273005, H3122 AR cells became more sensitive to alectinib, indicating the synergistic effect of the combination regimen (Fig. 2I). In conclusion, genetic knockdown of CCN1 could affect the ability of growth and angiogenesis in H3122 AR cells. Improved sensitivity was demonstrated by targeting CCN1.
Fig. 2. CCN1 could affect the ability of growth and angiogenesis, and alectinib sensitivity in H3122 AR cells.
A Colony forming experiment of H3122 AR cells after CCN1 knockdown. B CCN1 knockdown significantly decreased the colony forming capacity of H3122 AR cells. C The growth curve of H3122 AR cells after CCN1 knockdown. D Tube formation assay of H3122 AR cells after CCN1 knockdown. E CCN1 knockdown significantly decreased the angiogenesis capacity of H3122 AR cells. F, G CCN1 knockdown could improve the alectinib sensitivity in H3122 AR cells. H High concentration (10 μM) of SB273005 did not obviously affect the survival of H3122 AR cells. I The addition of SB273005 could improve the sensitivity of alectinib in H3122 AR cells. Error bars reflect mean ± standard deviation.
CCN1 affects tumor cells through the AKT-VEGFA pathway
CCN1 could affect the sensitivity to alectinib, growth, and angiogenesis activity of resistant H3122 cells. The underlying mechanism was not clear. KEGG pathway analysis was performed on maSigPro clusters with a similar tendency. Several pathways, such as VEGF pathway, Hippo pathway, and adherens junction pathway, were identified (Figs. 3A, B and S4). Previous studies have shown that CCN1 could promote angiogenesis by activating VEGFA and VEGFR2 [31, 32]. In addition, CCN1 could activate FAK to facilitate cell adhesion and migration [33, 34]. Combined with previous studies, we detected potential proteins that could be associated with CCN1. Several proteins were detected during alectinib withdrawal, rechallenge, and re-withdrawal (Fig. 3C). Consistent with RNA sequencing, CCN1 tended to be downregulated after drug withdrawal and upregulated after drug rechallenge. Abnormality of ALK can activate the downstream RAS-MAPK, PI3K-AKT, and JAK-STAT3 signaling pathways through autophosphorylation. When treated with alectinib, phosphorylation of ALK and STAT3 was inhibited, then recovered after drug discontinuation. However, other downstream proteins, AKT phosphorylation, and ERK phosphorylation had a similar trend with CCN1. VEGFA and VEGFR2, involved in angiogenesis, and FAK phosphorylation, involved in cell adhesion, also changed with CCN1. Further work was done to investigate the underlying regulatory relationships.
Fig. 3. CCN1 affects tumor cells through the AKT-VEGFA pathway.
A, B KEGG analysis for samples collected at different time during alectinib holiday-administration-holiday period (cluster 5, cluster 7). C Western blotting of samples collected at different time during alectinib holiday-administration-holiday period. D Western blotting of H3122 AR cells after CCN1 knockdown. E Western blotting of H3122 AR transfected with CCN1 plasmid. F Western blotting of H3122 AR treated with exogenous CCN1. G Western blotting of H3122 AR treated with MK2206 2HCI, an AKT inhibitor.
CCN1 knockdown was performed to investigate potential regulatory mechanisms in H3122 AR cells. To exclude the influence of ALK, related proteins were also examined. Decreased CCN1 caused obvious decrease in AKT phosphorylation, while no significant changes were observed in the levels of ALK phosphorylation, STAT3 phosphorylation, ERK phosphorylation, suggesting that AKT phosphorylation could be mediated by CCN1 independent of ALK (Fig. 3D). As for the VEGF pathway, a prominent decrease in VEGFA was seen, while VEGFR2 became higher. There was no change in the phosphorylation level of FAK. Similarly, pharmaceutical inhibition of CCN1 by SB273005 demonstrated a significant decrease in the phosphorylation of AKT and VEGFA (Fig. S5). Our results indicated that CCN1 could regulate the activation of AKT phosphorylation independent of ALK. Combining the suppressed angiogenesis activity with the decreased VEGFA level after CCN1 knockdown, we hypothesized that CCN1 might affect blood vessel formation by regulating VEGFA. Overexpression of CCN1 in H3122 AR also induced the expression of phosphorylation AKT and VEGFA, supporting the regulatory function (Fig. 3E). However, overexpression of CCN1 did not confer resistance to alectinib (IC50, NC 189.9 ± 13.51 nM vs CCN1 OE 184.1 ± 22.52 nM, p = 0.7215, Fig. S6), suggesting a complicated resistance mechanism in H3122 AR. CCN1 mainly functions as a secreted protein. By exogenous CCN1, increased phosphorylation of AKT and VEGFA was observed in H3122 AR cells (Fig. 3F). On the contrary, exogenous CCN1 did not affect the expression of phosphorylation AKT and VEGFA (Fig. S7) in the parental H3122 cell line, suggesting that H3122 AR might be more susceptible to CCN1 treatment. To explore the regulatory relationships between AKT and VEGFA, MK2206 2HCI, an AKT inhibitor, was used to suppress the phosphorylation AKT, which caused a significant decrease in VEGFA, indicating that VEGFA could be regulated by AKT expression (Fig. 3G). Taken together, we found that in H3122 AR, CCN1 could affect the expression of VEGFA by affecting the phosphorylation level of AKT. Although knockdown of CCN1 could improve the sensitivity to alectinib, overexpression of CCN1 did not cause alectinib resistance, suggesting that the underlying resistance mechanism is complex and CCN1 could serve as a potential therapeutic target.
CCN1 was activated by NF-κB in unstably resistant cells
To identify the potential upstream factor that could cause CCN1 activation, we detected a number of proteins and found that NF-κB phosphorylation shows the same trend as CCN1 during alectinib holiday-administration-holiday period (Fig. 4A). We further used Caffeic Acid Phenethyl Ester (CAPE), an NF-κB inhibitor, to treat H3122 AR. As shown in Fig. 4B, suppression of NF-κB phosphorylation resulted in decreased CCN1, AKT phosphorylation and VEGFA, supporting that NF-κB could act as an upstream factor (Fig. 4B). Furthermore, CAPE reached a mean IC50 of 5.902 ± 1.302 μM, and the addition of CAPE could improve the sensitivity to alectinib in H3122 AR cells (472.8 ± 98.66 nM vs 142.7 ± 21.45 nM, p = 0.0048, Fig. 4C, D). Taken together, our results showed that in unstable drug-resistant H3122 AR cells, changes in NF-κB phosphorylation after drug rechallenge and withdrawal could be a factor leading to CCN1 alteration, which in turn affected pAKT and VEGFA. Inhibition of NF-κB expression may improve the sensitivity to alectinib in H3122 AR cells.
Fig. 4. CCN1 was activated by NF-κB in H3122 AR cells.
A Western blotting of samples collected at different time during alectinib holiday-administration-holiday period. B Western blotting of H3122 AR treated with Caffeic Acid Phenethyl Ester (CAPE), an NF-κB inhibitor. C CAPE sensitivity in H3122 AR cells. D The addition of CAPE could improve the sensitivity of alectinib in H3122 AR cells. Error bars reflect mean ± standard deviation.
Conclusion
ALK rearrangements define a distinct molecular subtype of NSCLC with a favorable clinical response to ALK TKIs. Although ALK inhibitors have produced remarkable outcomes in advanced NSCLC harboring ALK rearrangement, acquired resistance is inevitable, leading to disease progression. Several mechanisms of resistance to ALK-targeted therapy have been identified, including kinase domain mutations, amplifications, and ALK-independent pathway activation [8–10]. Despite the development and application of DNA sequencing in clinical practice, the mechanisms of resistance remain unknown in a substantial proportion of patients, particularly those with non-genomic resistant mutations [35, 36]. It is crucial to investigate the resistance mechanisms in patients in order to rationally select subsequent therapies. Previous studies have reported the occurrence of “reversible resistance,” in which patients resistant to an ALK inhibitor could regain the sensitivity to the same agent after a period of treatment discontinuation. The reversible state of resistance suggests the absence of genetic aberrations. Little is known about the potential mechanisms, so further studies are needed to improve clinical management.
In this study, alectinib was used to construct drug-resistant EML4-ALK mutant H3122 cells and confirmed the existence of reversible drug resistance, i.e., high concentrations of alectinib could maintain the drug-resistant state of the cells, and the cells became sensitive to alectinib after drug removal. In the absence of genetic alterations causing resistance, RNA sequencing was performed to explore changes in the expression profile. CCN1 changes were found in both RNA and protein upon withdrawal or rechallenge with alectinib [27–29]. CCN1 is a member of the CCN family of growth factors, which includes CTGF, NOV, WISP-1, WISP-2, and WISP-3. CCN1 plays an essential role in cell proliferation, survival, adhesion, and angiogenesis [30–32]. High expression of CCN1 has been detected in various cancers such as breast cancer, gastric cancer, and ovarian cancer. Previous studies have indicated that CCN1 can mediate resistance to antitumor therapies in various cancers through diverse mechanisms. CCN1 can activate the RAS-MAPK and PI3K-AKT signaling pathways by stimulating integrin αvβ3 or upregulate Bcl-xL and cIAP1 to cause drug resistance [37, 38]. In breast cancer, the TAZ-TEAD-CCN1/CCN2 signaling axis plays a crucial role in taxol resistance [39]. CCN1 can transcriptionally downregulate ERα expression, leading to tamoxifen resistance [40]. The response to imatinib in chronic myeloid leukemia can be regulated by CCN1 through the NF-κB/Bcl-2 pathway [41]. Furthermore, overexpression of CCN1 can inhibit carboplatin-induced apoptosis by decreasing Bax expression and increasing Bcl-xL, Mcl-1, and Bcl-2 levels in ovarian cancer cells [42]. Despite accumulating studies conducted, the function of CCN1 in lung cancer was not fully understood and remains controversial. Thus, the role of CCN1 in lung cancer remains unclear and deserves further investigation.
In our studies, we found that inhibition of CCN1 could also significantly decrease the proliferation and angiogenesis capacity of H3122 AR cells. CCN1 has been reported to affect tumor growth through its potent angiogenic activity. Our study found that knockdown of CCN1 could cause the downregulation of VEGFA via the AKT pathway, which could lead to the suppression of tumor growth and angiogenesis. In addition, inhibition of CCN1 could render reversibly drug-resistant cells more sensitive to alectinib. Although improved sensitivity to alectinib was observed by inhibiting CCN1, overexpression of CCN1 did not confer resistance to alectinib, indicating that the potential resistance mechanisms in H3122 AR cells are complicated. However, the improved sensitivity to alectinib by inhibition of CCN1 expression or receptor still suggests that CCN1 could serve as a therapeutic target. Further experiments revealed that activated NF-κB could induce CCN1 upregulation, thereby affecting AKT phosphorylation and VEGFA expression.
Although our study demonstrated that sensitivity to alectinib could be restored by inhibiting CCN1, the development of specific targeting agents for CCN1 remains relatively unexplored. Given that CCN1 can interact with integrins such as αvβ3 to mediate diverse functions, blocking these interactions with integrin inhibitors could be a viable strategy. Furthermore, since CCN1 is a secreted protein, the development of novel monoclonal antibodies specifically targeting CCN1 holds significant promise [29, 31].
CCN1 is an immediate-early gene that can be rapidly transcriptionally activated in response to various stimuli [31]. As CCN1 encodes a secreted matricellular protein whose aberrant expression can be detected through liquid biopsies such as serum or saliva samples, recent studies suggest that CCN1 may be useful as a biomarker or therapeutic target in certain diseases. The potential of CCN1 as a biomarker is currently being investigated in diseases such as juvenile idiopathic arthritis (NCT05534347) and acute kidney injury (NCT05242705). Zhang et al. reported that salivary CCN1 mRNA was upregulated in lung cancer compared to matched controls [43]. A pilot study also found that plasma CCN1 levels were significantly elevated in lung cancer patients compared with healthy controls. In our study, the changes in resistance status associated with CCN1 expression may indicate the value of CCN1 as a resistance biomarker [44]. Further research using clinical samples is needed.
Our study had several limitations. In this study, only one ALK TKI was employed to explore the unstable resistant mechanism. Further research is demanded to investigate additional targeted inhibitors, such as lorlatinib. The reversible resistant state was only determined in H3122 AR, and more cell models are needed to confirm the CCN1 function in unstable resistance. In addition, the rapid change of resistant status after alectinib withdrawal makes the animal experiment difficult.
In conclusion, our study firstly demonstrated that CCN1 plays an important role in regulating the reversible resistant state of lung cancer cells. Suppression NF-κB or CCN1 receptor could improve the sensitivity to alectinib, suggesting promising combination targets that deserve further investigation.
Supplementary information
Author contributions
Conceptualization: Jing Li, Li Zhang, Wenfeng Fang; Data curation: Yihua Huang, Jie Huang, Jianhua Zhan, Maojian Chen; Formal analysis: all authors; Investigation: all authors; Methodology: all authors; Software: Yihua Huang; Supervision: Jing Li, Li Zhang, Wenfeng Fang; Validation: Jing Li, Li Zhang, Wenfeng Fang, Yihua Huang; Original draft: Yihua Huang; Review & editing: Jing Li, Li Zhang, Wenfeng Fang.
Funding
This work was partly supported by the Chinese National Natural Science Foundation Projects (82241232, 82272789 to L.Z.), the Guangdong Pearl River Talents Program (2021QN02Y168 to J.L.), Guangdong Basic and Applied Basic Research Foundation (2022A1515011873 to J.L.) and Young Talents Program of Sun Yat-sen University Cancer Center (YTP-SYSUCC-0040 to J.L.).
Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive [16] in National Genomics Data Center [17], China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA-Human: HAR009333) which are publicly accessible at https://ngdc.cncb/ac.cn/gsa-human.
Competing interests
All authors declare no competing interests.
Ethics approval and consent to participate
Ethics approval was not needed in this study because we used only cell lines. Animals and humans were not included.
Footnotes
Edited by Nickolai Barlev
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Yihua Huang, Jie Huang, Jianhua Zhan, Maojian Chen.
Contributor Information
Wenfeng Fang, Email: fangwf@sysucc.org.cn.
Li Zhang, Email: zhangli@sysucc.org.cn.
Jing Li, Email: lijing3@sysucc.org.cn.
Supplementary information
The online version contains supplementary material available at 10.1038/s41419-025-07601-4.
References
- 1.Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. 2024;74:12–49. [DOI] [PubMed] [Google Scholar]
- 2.Yang SR, Schultheis AM, Yu H, Mandelker D, Ladanyi M, Buttner R. Precision medicine in non-small cell lung cancer: current applications and future directions. Semin Cancer Biol. 2022;84:184–98. [DOI] [PubMed] [Google Scholar]
- 3.Tan AC, Tan DSW. Targeted therapies for lung cancer patients with oncogenic driver molecular alterations. J Clin Oncol. 2022;40:611–25. [DOI] [PubMed] [Google Scholar]
- 4.Hallberg B, Palmer RH. The role of the ALK receptor in cancer biology. Ann Oncol. 2016;27:iii4–iii15. [DOI] [PubMed] [Google Scholar]
- 5.Mazzeschi M, Sgarzi M, Romaniello D, Gelfo V, Cavallo C, Ambrosi F, et al. The autocrine loop of ALK receptor and ALKAL2 ligand is an actionable target in consensus molecular subtype 1 colon cancer. J Exp Clin Cancer Res. 2022;41:113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bavi P, Jehan Z, Bu R, Prabhakaran S, Al-Sanea N, Al-Dayel F, et al. ALK gene amplification is associated with poor prognosis in colorectal carcinoma. Br J Cancer. 2013;109:2735–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pietrantonio F, Maggi C, Di Bartolomeo M, Facciorusso MG, Perrone F, Testi A, et al. Gain of ALK gene copy number may predict lack of benefit from anti-EGFR treatment in patients with advanced colorectal cancer and RAS-RAF-PI3KCA wild-type status. PLoS ONE. 2014;9:e92147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cooper AJ, Sequist LV, Lin JJ. Third-generation EGFR and ALK inhibitors: mechanisms of resistance and management. Nat Rev Clin Oncol. 2022;19:499–514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gainor JF, Dardaei L, Yoda S, Friboulet L, Leshchiner I, Katayama R, et al. Molecular mechanisms of resistance to first- and second-generation ALK inhibitors in ALK-rearranged lung cancer. Cancer Discov. 2016;6:1118–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Schneider JL, Lin JJ, Shaw AT. ALK-positive lung cancer: a moving target. Nat Cancer. 2023;4:330–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Recondo G, Mezquita L, Facchinetti F, Planchard D, Gazzah A, Bigot L, et al. Diverse resistance mechanisms to the third-generation ALK inhibitor lorlatinib in ALK-rearranged lung cancer. Clin Cancer Res. 2020;26:242–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Browning ET, Weickhardt AJ, Camidge DR. Response to crizotinib rechallenge after initial progression and intervening chemotherapy in ALK lung cancer. J Thorac Oncol. 2013;8:e21. [DOI] [PubMed] [Google Scholar]
- 13.Cappuzzo F, Morabito A, Normanno N, Bidoli P, Del Conte A, Giannetta L, et al. Efficacy and safety of rechallenge treatment with gefitinib in patients with advanced non-small cell lung cancer. Lung Cancer. 2016;99:31–7. [DOI] [PubMed] [Google Scholar]
- 14.Hochmair M, Weinlinger C, Prosch H. Intracranial remission with brigatinib rechallenge as fifth-line ALK inhibition therapy in a lung cancer patient. Anticancer Drugs. 2019;30:1058–60. [DOI] [PubMed] [Google Scholar]
- 15.Araki T, Kanda S, Obara M, Agatsuma T, Kakizaki Y, Hama M, et al. EGFR-TKI rechallenge in patients with EGFR-mutated non-small-cell lung cancer who progressed after first-line osimertinib treatment: a multicenter retrospective observational study. Respir Investig. 2024;62:262–8. [DOI] [PubMed] [Google Scholar]
- 16.Chen T, Chen X, Zhang S, Zhu J, Tang B, Wang A, et al. The Genome sequence archive family: toward explosive data growth and diverse data types. Genom Proteom Bioinforma. 2021;19:578–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Database Resources of the National Genomics Data Center, China National Center for Bioinformation in 2022. Nucleic Acids Res. 2022;50:D27–38. [DOI] [PMC free article] [PubMed]
- 18.Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25:1754–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34:i884–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Cibulskis K, Lawrence MS, Carter SL, Sivachenko A, Jaffe D, Sougnez C, et al. Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol. 2013;31:213–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.McLaren W, Gil L, Hunt SE, Riat HS, Ritchie GR, Thormann A, et al. The Ensembl variant effect predictor. Genome Biol. 2016;17:122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chen X, Schulz-Trieglaff O, Shaw R, Barnes B, Schlesinger F, Kallberg M, et al. Manta: rapid detection of structural variants and indels for germline and cancer sequencing applications. Bioinformatics. 2016;32:1220–2. [DOI] [PubMed] [Google Scholar]
- 23.McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20:1297–303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. 2017;14:417–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nueda MJ, Tarazona S, Conesa A. Next maSigPro: updating maSigPro bioconductor package for RNA-seq time series. Bioinformatics. 2014;30:2598–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Conesa A, Nueda MJ, Ferrer A, Talon M. maSigPro: a method to identify significantly differential expression profiles in time-course microarray experiments. Bioinformatics. 2006;22:1096–102. [DOI] [PubMed] [Google Scholar]
- 27.Chen CC, Lau LF. Functions and mechanisms of action of CCN matricellular proteins. Int J Biochem Cell Biol. 2009;41:771–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Jia Q, Xu B, Zhang Y, Ali A, Liao X. CCN family proteins in cancer: insight into their structures and coordination role in tumor microenvironment. Front Genet. 2021;12:649387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Jun JI, Lau LF. Taking aim at the extracellular matrix: CCN proteins as emerging therapeutic targets. Nat Rev Drug Discov. 2011;10:945–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Monnier Y, Farmer P, Bieler G, Imaizumi N, Sengstag T, Alghisi GC, et al. CYR61 and alphaVbeta5 integrin cooperate to promote invasion and metastasis of tumors growing in preirradiated stroma. Cancer Res. 2008;68:7323–31. [DOI] [PubMed] [Google Scholar]
- 31.Lau LF. CCN1/CYR61: the very model of a modern matricellular protein. Cell Mol Life Sci. 2011;68:3149–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Park MH, Kim AK, Manandhar S, Oh SY, Jang GH, Kang L, et al. CCN1 interlinks integrin and hippo pathway to autoregulate tip cell activity. Elife. 2019;8. [DOI] [PMC free article] [PubMed]
- 33.Tan TW, Yang WH, Lin YT, Hsu SF, Li TM, Kao ST, et al. Cyr61 increases migration and MMP-13 expression via alphavbeta3 integrin, FAK, ERK, and AP-1-dependent pathway in human chondrosarcoma cells. Carcinogenesis. 2009;30:258–68. [DOI] [PubMed] [Google Scholar]
- 34.Chuang JY, Yu NY, Chiang IP, Lai CH, Lin CD, Tang CH. Cyr61 increases matrix metalloproteinase-3 expression and cell motility in human oral squamous cell carcinoma cells. J Cell Biochem. 2012;113:1977–86. [DOI] [PubMed] [Google Scholar]
- 35.Perez-Gonzalez A, Bevant K, Blanpain C. Cancer cell plasticity during tumor progression, metastasis and response to therapy. Nat Cancer. 2023;4:1063–82. [DOI] [PMC free article] [PubMed]
- 36.Mikubo M, Inoue Y, Liu G, Tsao MS. Mechanism of drug tolerant persister cancer cells: the landscape and clinical implication for therapy. J Thorac Oncol. 2021;16:1798–809. [DOI] [PubMed] [Google Scholar]
- 37.Menendez JA, Vellon L, Mehmi I, Teng PK, Griggs DW, Lupu R. A novel CYR61-triggered ‘CYR61-alphavbeta3 integrin loop’ regulates breast cancer cell survival and chemosensitivity through activation of ERK1/ERK2 MAPK signaling pathway. Oncogene. 2005;24:761–79. [DOI] [PubMed] [Google Scholar]
- 38.Wang MY, Chen PS, Prakash E, Hsu HC, Huang HY, Lin MT, et al. Connective tissue growth factor confers drug resistance in breast cancer through concomitant up-regulation of Bcl-xL and cIAP1. Cancer Res. 2009;69:3482–91. [DOI] [PubMed] [Google Scholar]
- 39.Lai D, Ho KC, Hao Y, Yang X. Taxol resistance in breast cancer cells is mediated by the hippo pathway component TAZ and its downstream transcriptional targets Cyr61 and CTGF. Cancer Res. 2011;71:2728–38. [DOI] [PubMed] [Google Scholar]
- 40.Kim H, Son S, Ko Y, Lee JE, Kim S, Shin I. YAP, CTGF, and Cyr61 are overexpressed in tamoxifen-resistant breast cancer and induce transcriptional repression of ERalpha. J Cell Sci. 2021;134:jcs256503. [DOI] [PubMed]
- 41.Song Y, Lin Q, Cai Z, Hao T, Zhang Y, Zhu X. Cysteine-rich protein 61 regulates the chemosensitivity of chronic myeloid leukemia to imatinib mesylate through the nuclear factor kappa B/Bcl-2 pathway. Cancer Sci. 2019;110:2421–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Lee KB, Byun HJ, Park SH, Park CY, Lee SH, Rho SB. CYR61 controls p53 and NF-kappaB expression through PI3K/Akt/mTOR pathways in carboplatin-induced ovarian cancer cells. Cancer Lett. 2012;315:86–95. [DOI] [PubMed] [Google Scholar]
- 43.Zhang L, Xiao H, Zhou H, Santiago S, Lee JM, Garon EB, et al. Development of transcriptomic biomarker signature in human saliva to detect lung cancer. Cell Mol Life Sci. 2012;69:3341–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Ackar L, Casjens S, Andreas A, Raiko I, Bruning T, Geffken M, et al. Blood-based detection of lung cancer using cysteine-rich angiogenic inducer 61 (CYR61) as a circulating protein biomarker: a pilot study. Mol Oncol. 2021;15:2877–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive [16] in National Genomics Data Center [17], China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA-Human: HAR009333) which are publicly accessible at https://ngdc.cncb/ac.cn/gsa-human.