Downregulation of specific FBXW7 isoforms with differential effects in T-cell lymphoblastic lymphoma


FBXW7 is a driver gene in T-cell lymphoblastic neoplasia acting through proteasome degradation of key proto-oncogenes. FBXW7 encodes three isoforms, α, β and γ, which differ only in the N-terminus. In this work, massive sequencing revealed significant downregulation of FBXW7 in a panel of primary T-cell lymphoblastic lymphomas characterised by the absence of mutations in its sequence. We observed that decreased expression mainly affected the FBXW7β isoform and to a lesser extent FBXW7α and may be attributed to the combined effect of epigenetic changes, alteration of upstream factors and upregulation of miRNAs. Transient transfections with miRNA mimics in selected cell lines resulted in a significant decrease of total FBXW7 expression and its different isoforms separately, with the consequent increment of critical substrates and the stimulation of cell proliferation. Transient inhibition of endogenous miRNAs in a T-cell lymphoblastic-derived cell line (SUP-T1) was capable of reversing these proliferative effects. Finally, we show how FBXW7 isoforms display different roles within the cell. Simultaneous downregulation of the α and γ isoforms modulates the amount of CCNE1, whilst the β-isoform alone was found to have a prominent role in modulating the amount of c-MYC. Our data also revealed that downregulation of all isoforms is a sine qua non condition to induce a proliferative pattern in our cell model system. Taking these data into account, potential new treatments to reverse downregulation of all or a specific FBXW7 isoform may be an effective strategy to counteract the proliferative capacity of these tumour cells.


The F-Box and WD repeat domain containing 7 gene (FBXW7) encodes a conserved E3 ubiqutin-protein ligase that forms a part of the SCF complex (SKP1, CUL1 and F-box protein), which is involved in the degradation of critical oncoproteins [1]. In humans, the FBXW7 gene is located within the 4q32 chromosomal region, which is commonly deleted in a broad spectrum of human tumours [2, 3].

FBXW7 encodes three protein isoforms (α, β and γ) resulting from alternative splicing [3, 4] that contain conserved interaction domains at the C-terminus but isoform-specific domains at the N-terminal region [3, 5]. Intriguingly, FBXW7 isoforms localise to different cell compartments, showing different preferences for specific sets of substrates [1]. Degradation of NOTCH1, c-MYC, CCNE1 and SREBP1 is largely dependent upon the nucleoplasmic FBXW7α, which in turn is able to stabilise EGFR and PGC-1α. Specific substrates for the cytoplasmic FBXW7β include PGC-1α and MCL1. The nucleolar localisation of FBXW7γ grants this isoform the control over the turnover of c-MYC and CCNE1 [6]. The overall oncogenic potential of its substrates has identified FBXW7 as a key tumour suppressor gene mutated in many types of cancers [1, 7, 8], and in particular in T-cell lymphoblastic neoplasms [9,10,11,12,13,14].

In accordance to its cellular relevance, the expression of FBXW7 is tightly regulated by several coding and non-coding genes [15,16,17,18]. As such, rather than mutational changes, aberrant regulation of FBXW7 expression has been reported in multiple malignancies [19]. In particular, FBXW7 mRNA levels have been found decreased in breast cancer, melanoma and glioma [20,21,22]. However, the role of FBXW7 deregulation in haematological malignancies has not yet been properly understood. It is important to stress that the differential regulation of FBXW7 isoforms by various stress stimuli contributes to the conviction that these isoforms should not be categorised as one protein [23]. Consistent with that complexity, evaluating the role of FBXW7 in tumorigenesis would require deeper investigation considering not only its mutational spectrum, but also determining the expression levels of its different isoforms.

T-cell lymphoblastic neoplasms are aggressive malignancies, which mainly develop in children but may also affect adults. Usually, these malignancies exhibit extensive marrow and blood affectation (acute T-cell lymphoblastic leukaemia, T-ALL). Less frequently, these malignancies can manifest as a mass lesion in the mediastinum, with less than 25% marrow blasts (T-cell lymphoblastic lymphoma, T-LBL) [24]. The molecular basis of these neoplasms has been well established in T-ALL [25, 26], but to a lesser extent in T-LBL [10, 27] mainly due to the scarcity of samples.

Here, we have focused on conducting a detailed study of the FBXW7 gene in T-LBL using next-generation sequencing (NGS) and considering its three isoforms separately. We have also performed the study of the main factors that control the expression of these isoforms, including a mutational screening and the study of the expression of upstream factors, together with the epigenetic analysis, and the expression of miRNAs that could contribute to the deregulation of this gene. The potential carcinogenic impact of FBXW7 alterations, via critical oncoprotein activation, has been assessed through in vitro experiments with cell lines derived from T-cell lymphoblastic neoplasia.


Mutational screening in primary T-LBLs and cell lines

DNA sequencing revealed the absence of functional mutational alterations in the FBXW7 sequence in all primary samples of the discovery and extended cohorts (Supplementary Methods). However, mutations were found in some genes of the FBXW7 pathway. Those genes encoding FBXW7 upstream factors capable of directly controlling its expression, such as CEBPδ, TP53, REL, HES5 or PSEN2 [8, 19, 28] were firstly analysed. From those, only TP53 that is capable of inducing the expression of FBXW7β and FBXW7γ isoforms [17, 23, 29], exhibited a functional polymorphism (p.Pro72Arg) (Variation ID: 12351;, in all but three tumour samples (samples 408, 730 and 829). A missense mutation (p.Val143Leu), which might be pathogenic [30], appears in only one tumour (sample 192) with a very limited number of reads (intra-tumour allelic frequency 0.02). The mutational status of indirect upstream factors, such as TAL1, NOTCH3 and NF-κβ [31,32,33,34,35] was also analysed. TAL1, which can repress FBXW7 expression through the induction of miR-223 [34], was found to be wild-type in all but three samples. Tumour 408 exhibited a frameshift mutation (p.Gly273fs) presumably deleterious by two predictive algorithms. Tumours 554 and 460 harboured missense mutations, yet involving a very limited number of reads. Finally, regarding FBXW7 target genes, the four main genes encoding specific FBXW7 substrate proteins, such as NOTCH1, c-MYC, CCNE1 and c-JUN [8] were studied. From those, only NOTCH1 exhibited several deleterious missense mutations in 50% of the cases (Supplementary Tables S1 and S2).

Regarding cell lines, JURKAT cells are heterozygous for an inactivating FBXW7 missense mutation (p.Arg505Cis) [36] and a missense mutation in NOTCH1 (p.Pro724Leu) [37]. They also harbour two mutations on TP53, one is homozygous for the functional polymorphism p.Pro72Arg and the other is a heterozygous missense mutation described as pathogenic (p.Arg196Ter). On the other hand, SUP-T1 cells are wild-type for FBXW7. However, NOTCH1 is activated by a TRB-NOTCH1 translocation and TP53 accumulates several deleterious missense mutations (IARC TP53 Database version R16) (Supplementary Table S3).

Downregulation of FBXW7 in human T-LBL samples

RNA-Seq data (log2FC) revealed that total FBXW7 mRNA levels were significantly reduced when compared with controls in 37.5% of the discovery cohort samples. When the results were broken down by isoforms, the expression of the β-isoform was significantly reduced in 87.5% of these tumours, whilst the α-isoform expression was significantly reduced in 25% of them. The behaviour of the γ-isoform turned out to be more erratic, and only one tumour showed significant downregulation, whereas another one exhibited significant upregulation (Fig. 1a; Supplementary Table S4A).

Fig. 1

Evaluation of FBXW7 expression. a Heatmap representation depicting changes of the expression of total FBXW7-mRNA and its three main isoforms in the discovery cohort as determined by RNA-seq analysis. The numbers in each box represent the fold changes (log2FC) calculated as the ratio of the FPKM values (fragments per kilobase of transcript per million mapped read) between tumour samples and normal thymuses. Negative values indicate downregulation, and positive values upregulation. Columns show the values of each tumour sample identified by its ID code at the bottom of the map. Colour key interpretation is indicated in the upper right part of the figure. Transcripts sequences are identified according to Ensembl references ( References: total FBXW7, ENSG00000109670; FBXW7α, ENST000000281708; FBXW7β, ENST00000263981 and FBXW7γ, ENST00000296555. b Expression levels of the total FBXW7 and its isoform by qRT-PCR in the discovery cohort (left) and in the extended cohort (right). Quantitative data are represented as the mean ± standard deviation (SD) of three independent experiments normalised with respect to the normal thymus values. Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001). c Left, representative western blots of FBXW7β and five FBXW7 target proteins in available control and T-LBL samples. ICN1 is the NOTCH1 intracellular domain (n = 2). Right, densitometric analysis of western blots from three independent experiments. Non-degraded T-LBL sample data were normalised with respect to the data obtained from the control thymus 405. Quantitative data are represented as the mean ± standard deviation (SD). Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001)

These results were validated using qRT-PCR in all the samples of the discovery cohort (Fig. 1b). Additional analyses were performed in the extended cohort, where FBXW7 downregulation was seen in 60% of the analysed samples, essentially attributed to the reduced expression of the β-isoform. This further confirmed the data obtained in the discovery cohort.

Deregulation of the FBXW7β isoform was confirmed at the protein level in most of the tumour samples with enough material available, confirming the mRNA data functionality (Fig. 1c).

Deregulation of upstream factors potentially controlling the expression of FBXW7

RNA-seq analysis performed on the discovery cohort revealed the deregulation of critical transcripts of the TP53 gene. A significant increase in the dominant-negative Δ133p53 isoform was found in half of the samples of this cohort. Interestingly, three samples also showed significant decreases of the full-length transcripts, which encode the p53α isoform, thus reducing TP53 transcriptional activity. Contrary to expectations, HES5, which directly represses the transcription of FBXW7β [38], was downregulated in all tumours with the exception of samples 554 and 460. The gene CEBPδ, which specifically inhibits the expression of FBXW7α isoform [15, 39, 40] was downregulated in two tumours. Among the upstream factors that indirectly control the expression of FBXW7, only TAL1 [34] and NOTCH3 were upregulated in 62.5% and in 25% of T-LBL samples, respectively. On the contrary, REL [32, 33] was downregulated in 50% of the tumours. Finally, the expression of PSEN2, which is able to downregulate the FBXW7α isoform [41], was found to be significantly increased in three tumours (Supplementary Table S4A). Data from the extended cohort revealed deregulations of these upstream factors in a similar way (Supplementary Fig. S1). Thus, the deregulation presented by TP53 (proven to affect p21), NOTCH3 and TAL1 (proven to affect MYB) may indicate that they could be major contributing factors to downregulate the expression of FBXW7 isoforms in primary T-LBL samples.

Microarray-based comparative genomic hybridisation (aCGH)

We analysed the region 4q32 for copy-number changes as indicated in Supplementary methods. However, we found no evidence of deletions directly affecting FBXW7 in none of the analysed T-LBL samples (Supplementary Fig. S2).

Epigenetic analysis of FBXW7 isoforms

The use of Methyl Primer Express v1.0 software, allowed us to identify three candidate promoter areas located upstream of the transcription start sites (TSS) of FBXW7α (DMR1 and DMR2) and FBXW7β (DMR3) susceptible of displaying differential methylation patterns in tumour samples (Fig. 2a). The density of methylated CpG sites in these regions determined by pyrosequencing, revealed that only DMR3 presented different methylation levels among the T-LBL tumours (Fig. 2b, upper). Despite the limited number of available primary samples, our data showed a significant correspondence between reductions in the expression of the β-isoform and γ-isoform and the degree of DMR3 methylation (Fig. 2b, bottom). Accordingly, samples 192 and 521 present the highest methylated levels on DMR3 region and the lowest FBXW7β mRNA levels by RNA-seq (Fig. 1a). In order to experimentally verify this correlation, we treated two T-cell lymphoblastic-derived cell lines that present high levels of DMR3 methylation (JURKAT and MOLT4) with 5-aza-2′-deoxycytidine (AZA), a DNA methyltransferase inhibitor. As expected, this treatment induced significant demethylation on DMR3, causing an increase on FBXW7β expression and, to a lesser extent, on the FBXW7γ expression (Fig. 2c, left). Changes in FBXW7β expression post AZA treatment were confirmed at the protein level (Fig. 2c, right). Interestingly, tumour 346, which exhibited the highest levels of FBXW7γ, was the only tumour showing significant hypomethylation in this region (Fig. 2).

Fig. 2

Epigenetic analyses in the three candidate differential-methylated regions. a The physical map of the FBXW7 gene with indication of the candidate regions for DNA methylation named DMR1 (black), DMR2 (green) and DMR3 (dark blue) around the transcriptional start site of the three main isoforms. Black boxes represent exons. Vertical magenta lines represent CpG sites. Horizontal blue-coloured lines indicate the regions whose sequences were amplified by pyrosequencing. b The levels of CpG methylation in the selected CpG sites in available primary samples and cell lines obtained by pyrosequencing. Analyses were performed only when DNAs had the required purity. ND, not done. Pearson’s correlation analyses of DMR2 and DMR3 are added below. c Results of treatments with 5-aza-2′-deoxycytidine (5-AZA) in JURKAT and MOLT-4 cell lines. Upper left, changes of methylation density in DMR2 and DMR3 sites after treatment. Lower left, changes in methylation mainly affect β- and γ-isoforms expression levels (n = 3). Upper right, representative western blot of FBXW7β after treatment (n = 3). Lower right, densitometric analyses of three independent experiments (n = 3). Quantitative data are represented as the mean ± standard deviation (SD) normalised with respect to untreated cells. Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001)

Deregulation of specific miRNAs in T-LBL samples

Another alternative explaining FBXW7 downregulation could be the upregulation of specific miRNAs. The levels of miRNA expression were initially determined by massive small RNA sequencing in the eight T-LBL samples of the discovery cohort (Supplementary Table S4B). Only 32 miRNAs were found to meet the initial selection requirements described in the section Methods. Most of them were capable of recognising the three FBXW7 isoforms. Among them, six miRNAs (hsa-miR-195-5p, hsa-miR-223-3p, hsa-miR-101-3p, hsa-miR-497-5p, hsa-miR-15b-5p and hsa-miR-92a-3p), exhibiting both overexpression in at least one tumour and more than 100 reads in all samples (indicated with an asterisk in Fig. 3a), were considered as the most influential ones. Other miRNAs, such as hsa-miR-1248 and hsa-miR-29b-1-5p, which are upregulated in practically all samples, were rejected due to the low levels of their relative expression (Supplementary Fig. S3, Tables S2 and S4C).

Fig. 3

miRNA analysis data. a Heatmap representation depicting the changes of the expression of selected microRNAs between tumour samples of the discovery cohort and normal foetal thymuses as determined by small-RNA-seq analysis. The numbers in each box represent the fold changes (log2FC) calculated as the ratio of the FPKM values (fragments per kilobase of transcript per million mapped read) between tumour samples and normal thymuses. An asterisk was added to the value number on the box when the miRNA counts are higher than 100 reads. Negative values indicate downregulation, and positive values upregulation. Thirty-two selected miRNAs are listed on the right side of the figure. Targeted FBXW7 isoforms are indicated on the left side according to the grey bar on the left-hand side. Columns consist of the values of each sample identified by its ID code on the bottom of the map. Colour key interpretation is indicated in the upper right part of the figure. b Selected miRNA information. Upper, a representative scheme of the FBXW7 3′UTR with indication of the seed sequence and the recognition sites of the three most influential miRNAs (pink to hsa-miR-223-3p; blue, hsa-miR-195-5p and green, hsa-miR-101-3p). Dark tone indicated poorly conserved sites. Below, a summary of the main properties of the selected miRNA on 3′UTR region: position of the seed-sequence joining; status of the interaction, conserved or poorly conserved (in mammals). Site type refers to the strength of the interaction. In total, 8 mer indicates an exact match to positions 2–8 of the mature miRNA, whereas 7 mer-A1 indicates an exact match to positions 2–7 of the mature miRNA, and consequently it has less strength than 8 mer. PCT (probability of conserved targeting) values >0.75 indicate a higher probability to interaction between miRNA and mRNA. (Data obtained from the website tool TargetScan Release 7.2.

Experimental validation of the effects of miRNA deregulation on the FBXW7 expression, its targets and cell physiology

To assess the functional consequences of miRNA deregulation, we selected only the most representative member of each miRNA family (according to the miRGate agreement values). The suitable miRNA candidates for performing in vitro assays were hsa-miR-195-5p, hsa-miR-223-3p and hsa-miR-101-3p. The effects on FBXW7 expression were studied using simple or double combinations with miRNA mimics (Fig. 3b). The chosen cell lines for these experiments had to show low levels of the selected miRNA expression and present no mutation at the FBXW7 3′-UTR that could compromise the miRNA–mRNA interaction (Supplementary methods). From the six cell lines initially tested, only SUP-T1 (with high levels of FBXW7 expression) and JURKAT (with low levels of FBXW7 expression) met those requirements (Supplementary Figure S4A-D).

JURKAT cells, despite having low FBXW7 expression, were used as methodology control as they had been previously used to test the effect of enforced mimic-hsa-miR-223 expression over FBXW7 and its target genes [33]. Transient transfection of the mimic-hsa-223-3p caused a reduction of the FBXW7α and FBXW7β mRNAs (Fig. 4a, left). An increase in c-MYC and/or NOTCH1 proteins was also detected (Fig. 4b, left). In SUP-T1 cells, the transient transfection of mimic-hsa-miR-101-3p caused a significant reduction of the β-isoform. Furthermore, when this mimic acted in combination with any of the other two selected ones, a decrease in FBXW7α expression was also found significant (Fig. 4a, right). There seem to be non-significant effects over the γ-isoform at the mRNA level when transiently transfecting the cells. At the protein level, all mimic transfections induced increasing levels of c-MYC, which coincide with the downregulation of the β-isoform and less frequently with the decrease of α-isoform (mimic-hsa-miR-223+mimic-hsa-miR-195-5p). CCNE1 showed a modest but significant increment after mimic-hsa-101-3p transfection (Fig. 4b, right). Regarding NOTCH1, the TRB-NOTCH1 activating translocation present in SUP-T1 cells could be masking the effect of FBXW7 downregulation over ICN1 (Supplementary Table S3). This target was therefore discarded. Of note, only SUP-T1 cells showed an increment in cell proliferation after transient transfection with the mimics (Fig. 4c, supplementary Fig. S5). Finally, none of the two cell lines showed a significant reduction on cell death (data not shown).

Fig. 4

The effects of transient transfections of mimics in JURKAT and SUP-T1 cell lines. a Changes in the expression levels of total FBXW7 and each of its isoforms between the different miRNA mimics transduction combinations (n = 3). b Representative western blots of FBXW7β and five FBXW7 target proteins. ICN1 is the NOTCH1 intracellular domain (active form of the NOTCH1 protein). The results of densitometric analyses of all independent experiments are displayed below in an histogram (n = 3). c Cell-viability test data as determined by trypan-blue counts in three independent experiments. Each line shows the mean ratio between alive and dead cells taken at 6, 24, 48 and 72 h after transfections (n = 3). All quantitative data are represented as mean ± standard deviation (SD) of three independent experiments normalised with respect to those obtained in cells transfected with the negative control mimic miRNA (CN). The mimics and the combinations used in these experiments are referred to as the miR symbol followed by the numbers corresponding to the miRNAs. Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001)

As SUP-T1 cells were the only cell line presenting a significant increase in cell proliferation after mimic transfection, they were selected to carry on further with the miRNA analysis. The next step was to transfect these cells with either the mimics or the specific inhibitors of the selected endogenous miRNAs to verify their implication on FBXW7 expression, the consequent regulation of its targets and the overall cell physiology. Three single transfections and one combined transfection with the inhibitor hsa-miR-101-3p (inh-hsa-miR-101-3p from now on) and the inhibitor hsa-miR-195-5p (inh-hsa-miR-195-3p from now on) were performed in parallel with mimic transfections for comparison. Out of all the conditions tested, the transfection of inh-miR-hsa-101-3p caused the highest increase in total FBXW7-mRNA levels (Fig. 5a). Furthermore, the upregulation of the α- and β-isoforms was detected after transfection of inh-miR-hsa-101-3p (alone or in combination with inh-miR-hsa-195-5p), coinciding with a decrease in the c-MYC and CCNE1 protein levels (Fig. 5b) and a reduction in cell proliferation (Fig. 5c).

Fig. 5

Experimental validation of the effects of mimics on the expression levels of FBXW7 by treating SUP-T1 cells with inhibitors of specific endogenous miRNAs. a Expression levels of total FBXW7 and its isoforms by qRT-PCR after three independent experiments. b Representative western blots showing the affectation of FBXW7α, FBXW7β and four of the FBXW7 target proteins after transfections with mimics and inhibitors. Both antibodies used in the immunoprecipitation of FBXW7α are specific for the α-isoform (n = 3). c Viability tests (n = 3). On the top, the trypan-blue counts are shown. Each line shows the ratio between living and dead cells at 6, 24, 48 and 72 h after transfections. At the bottom, MTT assay is represented in a histogram where its bar shows the normalised data of the optical density ratio between 560 and 750 nm. All quantitative data are represented as mean ± standard deviation (SD) of three independent experiments normalised with respect to those obtained in cells transfected with the negative control mimic miRNA (CN). The mimics and the inhibitors used in these experiments are referred to as the miR or inh symbol followed by the numbers corresponding to the miRNAs. Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001)

Differential roles for FBXW7 isoforms

To study the specific role of each FBXW7 isoform, we generated four SUP-T1 cell lines each stably expressing a lentiviral vector harbouring one of the FBXW7 isoforms lacking the 3′UTR, therefore preventing mimics binding to that region (Fig. 6). The stable cell lines Lv225-FBXW7α, Lv225-FBXW7β and Lv225-FBXW7γ harboured exogenous expression vectors of FBXW7α, FBXW7β and FBXW7γ, respectively. Lv225-NEG was the negative control. Each transduced culture expressed the correspondent exogenous isoform untargeted by the mimics, whilst the endogenous forms are susceptible of being targeted by them. Further transfection of those cells with mimic-hsa-miR-101-3p and mimic-hsa-miR-195-5p proved that the presence of the exogenous isoform reverted the effect that mimics induced on the endogenous mRNA for that given isoform, whilst the expression of the other two isoforms was found reduced (Fig. 6a). Regarding FBXW7 downstream targets, in Lv225-NEG cells, c-MYC and CCNE1 protein levels were found increased (Fig. 6b). In the other cultures, the endogenous FBXW7 isoform downregulation generated an isoform-dependent upregulation of the FBXW7 targets. As such, c-MYC was upregulated in Lv225-FBXW7α where the β- and γ-isoforms are downregulated, and CCNE1 was upregulated in Lv225-FBXW7β where α- and γ-isoforms are downregulated (Fig. 6b). Surprisingly, in Lv225-FBXW7γ cells, the endogenous β-isoform remained unaffected by the mimics both at the mRNA and protein levels (Fig. 6a, b). Furthermore, only these cells showed an increase in PGC-1α despite FBXW7α being pointed out as the responsible isoform for PGC-1α stabilisation and FBXW7β for its degradation [42] (Fig. 6b). However, the expression levels of this FBXW7 target could also be stabilised by the p38 MAP kinase pathway activation [43]. Therefore, we checked if this pathway could be induced by the downregulation of its inhibitor, MKP-1, which is in fact a target of hsa-miR-101-3p [44]. Our findings evidenced that MKP-1 protein remained unaffected in Lv225-FBXW7γ cells after mimic-hsa-miR-101-3p transfection, indicating that PGC-1α increase must be caused by the downregulation of the α-isoform in this cell line, given that the β-isoform does not vary after mimic transfections (Supplementary Fig. S6).

Fig. 6

Effects of mimic transfections on mRNA and protein expression levels in SUP-T1 cells stably expressing FBXW7 isoforms. a Expression levels of total FBXW7 and each isoform by qRT-PCR after three independent experiments. The quantitative data are represented as mean ± standard deviation (SD) of three independent experiments normalised with respect to those obtained in cells transfected with the negative control mimic miRNA (CN). The mimics and the combinations used in these experiments are refered to as the miR symbol followed by the numbers corresponding to the miRNAs. Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001). b Representative western blots of each generated SUP-T1 cell line showing the affectation of protein levels of α- and β-isoforms and four of FBXW7 targets in transduced SUP-T1 cells after being transfected with mimics. c Representative western blot of each generated SUP-T1 cell line indicating the relative amount of α- and β-isoforms and four of FBXW7 targets. FBXW7α protein levels were determined by immunoblotting after specific immunoprecipitation using specific antibodies for the α-isoform

Strikingly, mimic transfections were only able to induce a proliferative pattern in Lv225-NEG cells, whereas in the rest of the transduced cell lines, this induction was reversed, indicating that all isoforms should play a role in triggering proliferation (Fig. 7) through the simultaneous increase of c-MYC and CCNE1. In order to check this, we generated a c-MYC knockdown by transducing SUP-T1 cells with a retroviral construction carrying a c-MYC shRNA (pRS-shRNA-cMYC), generating cells with a stable reduction of c-MYC expression (Supplementary Fig. S7A, B). Our results indicate that downregulation of FBXW7 induced by mimics in these cells is no longer able to trigger cell proliferation, confirming the importance of c-MYC on the induction of cell proliferation (Supplementary Fig. S7C, D). Unfortunately, the knockdown of CCNE1 did not generate viable cell lines (data not shown), impeding the study of its implication in cell proliferation.

Fig. 7

Viability tests of transduced SUP-T1 cells transfected with mimics. a Cell-viability test by trypan-blue counts after three independent experiments (n = 3). Each line shows the mean ratio between living and dead cells 6, 24, 48 and 72 h after the transfection. b MTT assays with selected mimic combinations 48 and 72 h after transfection after three independent experiments (n = 3) are represented in a histogram where each bar shows the normalised data of the optical density ratio between 560and 750 nm. Quantitative data are represented as mean ± standard deviation (SD) of three independent experiments normalised with respect to those obtained in cells transfected with the negative control mimic miRNA (CN). The mimics and the combinations used in these experiments are referred to as the miR symbol followed by the numbers corresponding to the miRNAs. Significant differences were determined using Student's t test (*P < 0.05, **P < 0.01, ***P < 0.001)

Finally, to verify the interactions between the different FBXW7 isoforms and its two main targets involved in cell proliferation (CCNE1 and c-MYC), we transfected SUP-T1 cells with a siRNA duplex able to bind the exon 10 of FBXW7 (common sequence to the three isoforms) or to the first exon of each isoform (specific sequence for each isoform). The transfection of these siRNAs showed significant upregulation of CCNE1 and c-MYC 24 h after the downregulation of FBXW7α (48 h after transfection) and FBXW7β (72 h after transfection), respectively. We also observed upregulation of CCNE1 96 h after transfecting cells with a specific siRNA against FBXW7γ (Supplementary Fig. S8).


The role of FBXW7 as a tumour suppressor gene has been sustained by many studies that revealed a high frequency of inactivating FBXW7 mutations in a wide range of cancers, including T-cell lymphoblastic neoplasia [8]. However, the incidence of mutations ranges from 8 to 30% in T-ALL [45, 46], and around 18% in the T-LBL subtype [10]. Therefore, the way FBXW7 acts in the remaining 70% of these haematological malignancies is poorly understood. Several lines of evidence suggest that the deregulation of wild type FBXW7 could be playing an important role in the initiation of such tumours [47]. In fact, the conditional ablation of Fbxw7 in mouse haematopoietic system promotes T-ALL development [48,49,50].

Our data revealed that deregulation of FBXW7 by the downregulation of its expression is a frequent event in T-LBL. This is executed by the deregulation of a vast set of factors functioning in different and multiple combinations. In fact, there are no tumours presenting the same combination of alterations on the FBXW7 upstream factors (Fig. 8a). The low incidence of upstream factors mutation rate suggests that inactivating mutations might not be a significant factor in disabling functional FBXW7 in these tumours.

Fig. 8

A summary of the main points of the FBXW7 study. a Factors and mechanisms involved in the deregulation of FBXW7. This figure has been built based on gene expression data obtained by RNA-Seq or qRT-PCR (Fig. 1a, b, Supplementary Table S4 and figure S1) in the 18 tumour samples of the discovery and extended cohorts. Dark red colour indicates significant upregulation, while dark blue colour denotes significant downregulation. Light red and blue colours indicate non-significant trends. The last line of the table shows the methylation density values (%) of CpG sites in DMR3 region. Plus and minus signs before FBXW7 isoforms denote activation or repression of isoforms. ND, not done. b Summary of the results obtained in the panel of SUP-T1 cells stably expressing each isoform without 3′UTR. The top panel shows the cellular status after transduction and the lower panel shows the results after miRNA transfection. Blue colour and downward arrows indicate significant downregulation, whereas red colour and upward arrows indicate significant upregulation of the studied gene/protein or biological processes. Black colour and = symbol indicate no significant variation

Regarding epigenetic mechanisms, some authors have discussed that the inactivation of the FBXW7β could be related to promoter hypermethylation of the area termed as DMR3 in this work. This link was reported in breast cancer, glioma and thymoma, although hypermethylation of DMR3 seems not to be a major mechanism of downregulation of this isoform in glioma [20, 22, 51]. However, according to our data, the methylation density of this region in primary T-cell lymphoblastic lymphomas could be a good indicator of the expression levels of the β- and γ-isoforms.

In addition to transcriptional mechanisms, we hypothesised that the expression of FBXW7 could be also influenced by the combinatorial effect of upregulated miRNAs. Using restrictive criteria, we selected three miRNAs (hsa-miR-195-5p; hsa-miR-101-3p and hsa-miR-223), which are significantly upregulated in our sample series of human primary T-LBL tumours. Using SUP-T1 cells as a system for functional studies, we demonstrate that these miRNAs were capable of inducing modest [52] yet significant effects on the expression levels of the FBXW7 isoforms and its target proteins. Regarding the effects of the mimics on FBXW7γ, our results were contradictory. Significant decrease of this isoform after mimic transfection only was detected in Lv225 cells. This indicates that for studies of the γ-isoform, qRT-PCR could not be the best approach to study the downregulation of this isoform by miRNAs probably due to the low stability of its mRNA [8]. Regretfully, there is not another alternative nowadays due to the absence of a specific antibody. However, the data from the assays on Lv225 cells and siRNA assays evidencing an increase of CCNE1 levels depending on FBXW7γ downregulation, support that miRNAs are able to reduce FBXW7γ expression levels.

Taking all these data into account, upregulation of miRNAs together with upstream factors deregulation and the epigenetic hypermethylation of DMR3 should be the main factors involved in the global decline of FBXW7 expression observed in primary T-LBL tumours in the context of high inter-tumour complexity [53].

Focusing on the selected miRNAs, a growing body of evidence showed that upregulation of hsa-miR-223 is key in mouse and human T-cell lymphoma and leukaemia, therefore reinforcing the importance of this miRNA in T-cell lymphomagenesis [54, 55]. This miRNA has the ability to reduce the levels of FBXW7 expression both in mouse [54] and human [34]. To our knowledge, the involvement of hsa-miR-195-5p and hsa-miR-101-3p [56] in modulating FBXW7 expression in T-LBL is a complete novelty in the present work. Interestingly, hsa-miR-101-3p apparently caused a bigger effect on FBXW7β 3′UTR (Fig. 4c). However, other miRNAs upregulated in our tumours, which were taken off the experimental approaches in the present work (Fig. 3a) could also be contributing to downregulation of FBXW7.

Data from SUP-T1 cells stably expressing each isoform without 3′UTR or transfected with different siRNAs against total or each FBXW7 isoforms revealed differential roles for α, β and γ that may differ from previous reports. We found that α- and γ-isoforms were able to cooperatively modulate the amount of CCNE1, since we observed a significant increase in the level of this protein when both endogenous isoforms are simultaneously decreased (Lv225-NEG and Lv225-FBXW7β cells) [35] or after transfecting specific siRNAs for both isoforms. We also observed that the β-isoform appears to have a predominant role on c-MYC degradation in this system. This interaction has been reported as of less importance than that of the modulation of c-MYC mediated by FBXW7α and FBXW7γ [42, 57]. Interestingly, most of the analysed primary T-LBL samples showed changes in the expression levels of c-MYC and CCNE1 proteins that are in accordance with the changes detected in the FBXW7β protein levels.

The viability tests revealed that only SUP-T1 cells transduced with the empty lentiviral expression vector (Lv225-NEG) were able to induce an increment in cell proliferation and viability. These data indicate that downregulation of all isoforms is needed to the increment of both cell functions in this system. A corollary is that the upregulation of both c-MYC and CCNE1 is needed in these cells to generate a proliferative pattern (Fig. 8B and Supplementary Fig. S8). Consequently, treatments of these haematological malignancies to reverse downregulation of FBXW7, acting directly over them or indirectly over the upstream factors responsible for their deregulation, may be an effective strategy to counteract the proliferative capacity of tumour cells.

Taken together, our results indicate that different combinations of upregulated miRNAs (the six most influential ones and, to a lesser extent, other miRNAs initially discarded) should be acting redundantly together with multiple combinations of deregulated upstream factors and epigenetic hypermethylation of the DMR3 region to modulate the expression of the FBXW7 isoforms in a tumour-dependent manner. Importantly, deregulation of FBXW7 by upregulation of miRNAs is able to significantly affect the integrity of its target proteins, thus contributing to an increment in cell proliferation and viability.

Materials and methods

T-LBL primary tumours

Eighteen primary T-LBL tumours were diagnosed according to the World Health Organisation Classification of Haematological Malignancies and the European childhood lymphoma pathology panel recommendations [58] (Supplementary Table S5). Institutional review board approval was obtained for these studies (references CEI 31-773 and CEI-70-1260). The participants provided written informed consent in accordance with the Declaration of Helsinki. Additional information is included in Supplementary Methods.

Whole-transcriptome sequencing (RNASeq)

Massive sequencing of mRNAs, image analysis, per-cycle basecalling and quality score assignment were made with Illumina Real Time Analysis software (Illumina, San Diego, CA). In the small RNAseq (miRNAs), image analysis and per-cycle basecalling was performed using Illumina Real Time Analysis software (RTA1.9) (Supplementary Methods). Raw sequencing data and transcript expression quantification are available as a superseries in GEO (Gene Expression Omnibus) under the following ID GSE109234.

Selection of miRNAs

The levels of miRNA expression were initially determined by massive small RNA sequencing in the eight T-LBL samples of the discovery cohort (Supplementary Table S4B). Selection of putative miRNAs for the three FBXW7 isoforms was performed using miRGate [59]. Migrated miRNAs were classified according to their agreement values and then filtered according to three inclusion criteria:

  1. 1.

    miRNAs with agreement values higher or equal than the third quantile (1.1953168) or whose interaction with 3′UTR of FBXW7 mRNA had been experimentally validated.

  2. 2.

    miRNAs with relative expression values >10 read counts in at least one control or T-LBL sample. Since only the most abundant miRNAs in a cell really mediated target suppression [60] (see Density Plot in Supplementary Fig. S9).

  3. 3.

    miRNAs exhibiting fold changes greater than 1.5 or lesser than −1.5 in at least one T-LBL sample.

Western blot

All information regarding the FBXW7α immunoprecipitation, western blot procedure and the generation of the FBXW7 isoform controls (Supplementary Fig. S10) are indicated in Supplemental Methods.

Transfection of cell lines with non-coding RNAs

SUP-T1 and JURKAT cells were transfected with 100 nM of the negative control #1 and specific mimics and inhibitors (the latest only in SUP-T1 cells) (Ambion, Cambridside, UK). The mimic-hsa-miR-1 was used as a positive control of transfection according to the manufacturer’s instructions through its highly specific action over human PTK9 3′UTR. Alternatively, SUP-T1 cells were transfected with 100 mM of siRNA duplexes against different FBXW7 regions and the scrambled negative control [61, 62] (Sigma Aldrich, MO, USA) (Supplementary Table S6). BLOCK-iT Alexa Fluor Red Fluorescent Control (Invitrogen, Carlsbad, CA, USA) was used as a positive control to evaluate the transfection efficiency by flow cytometry (λ = 562 nm) 24 h after transfection. GenePulser MXCellTM (Bio-Rad, Hercules, CA, USA) was used for all transfections with adjusted conditions [63]. In mimic assays, RNA and proteins were extracted 24 h and 48 h after transfections, respectively. siRNA effects were evaluated at protein level from 48 to 96 h after transfection.

Generation of stably cell lines

SUP-T1 cells were transduced with lentiviral vectors harbouring the mRNA of FBXW7 isoforms without their 3′UTR: Lv225-NEG (empty lentiviral vector), Lv225-FBXW7α, Lv225-FBXW7β and Lv225-FBXW7γ (GeneCopoeia, Rockville, MD, USA). SUP-T1-shRNA-cMYC knockdown cells, stably expressing reduced c-MYC expression levels, were also generated using the retroviral vector pRetroSuper-shRNA-cMYC (Addgene, Watertown, MA, USA). Lentiviral and retroviral with their packing vectors were initially transfected on HEK-293T.

Statistical analyses

Differences in the expression of mRNA and miRNA (RNA-Seq) were assessed by calculating the log2 fold changes (log2FC) of the expression levels in tumour and control samples. Fold change values greater than 1 or less than −1 (indicate two orders of magnitude differences), were considered as thresholds of significant deregulation. Student's t test was used to compare the results from different analyses (qRT-PCR, Wb) between tumours and controls and the in vitro assays from at least three independent experiments (n = 3), where each one was obtained from three technical repeats. After these analyses, we checked the expression dataset for normality distribution using the Kolmogorov–Smirnov test and the homogeneity of variances by Levene’s, using SPSS Statistics 25 (IBM Corporation, Somers, NY, USA). Pearson's correlation coefficient and test were used to determine the statistical dependence between two variables. All statistical analyses were performed using the R (Foundation for Statistical Computing, Vienna, Austria) or SPSS software. All statistical test values below 0.05 were considered statistically significant.


  1. 1.

    Welcker M, Clurman BE. FBW7 ubiquitin ligase: a tumour suppressor at the crossroads of cell division, growth and differentiation. Nat Rev Cancer. 2008;8:83–93.

    CAS  Article  Google Scholar 

  2. 2.

    Knuutila S, Aalto Y, Autio K, Björkqvist AM, El-Rifai W, Hemmer S, et al. DNA copy number losses in human neoplasms. Am J Pathol. 1999;155:683–94.

    CAS  Article  Google Scholar 

  3. 3.

    Spruck CH, Strohmaier H, Sangfelt O, Müller HM, Hubalek M, Müller-Holzner E, et al. hCDC4 gene mutations in endometrial cancer. Cancer Res. 2002;62:4535–9.

    CAS  PubMed  Google Scholar 

  4. 4.

    Ho MS, Tsai P-I, Chien C-T. F-box proteins: the key to protein degradation. J Biomed Sci. 2006;13:181–91.

    CAS  Article  Google Scholar 

  5. 5.

    Crusio KM, King B, Reavie LB, Aifantis I. The ubiquitous nature of cancer: the role of the SCF(Fbw7) complex in development and transformation. Oncogene. 2010;29:4865–73.

    CAS  Article  Google Scholar 

  6. 6.

    Xu W, Taranets L, Popov N. Regulating Fbw7 on the road to cancer. Semin Cancer Biol. 2016;36:62–70.

    CAS  Article  Google Scholar 

  7. 7.

    Akhoondi S, Sun D, von der Lehr N, Apostolidou S, Klotz K, Maljukova A, et al. FBXW7/hCDC4 is a general tumor suppressor in human cancer. Cancer Res. 2007;67:9006–12.

    CAS  Article  Google Scholar 

  8. 8.

    Davis RJ, Welcker M, Clurman BE. Tumor suppression by the Fbw7 ubiquitin ligase: mechanisms and opportunities. Cancer Cell. 2014;26:455–64.

    CAS  Article  Google Scholar 

  9. 9.

    Baldus CD, Thibaut J, Goekbuget N, Stroux A, Schlee C, Mossner M, et al. Prognostic implications of NOTCH1 and FBXW7 mutations in adult acute T-lymphoblastic leukemia. Haematologica. 2009;94:1383–90.

    CAS  Article  Google Scholar 

  10. 10.

    Bonn BR, Rohde M, Zimmermann M, Krieger D, Oschlies I, Niggli F, et al. Incidence and prognostic relevance of genetic variations in T-cell lymphoblastic lymphoma in childhood and adolescence. Blood. 2013;121:3153–60.

    CAS  Article  Google Scholar 

  11. 11.

    Callens C, Baleydier F, Lengline E, Ben Abdelali R, Petit A, Villarese P, et al. Clinical impact of NOTCH1 and/or FBXW7 mutations, FLASH deletion, and TCR status in pediatric T-cell lymphoblastic lymphoma. J Clin Oncol. 2012;30:1966–73.

    CAS  Article  Google Scholar 

  12. 12.

    Larson Gedman A, Chen Q, Kugel Desmoulin S, Ge Y, LaFiura K, Haska CL, et al. The impact of NOTCH1, FBW7 and PTEN mutations on prognosis and downstream signaling in pediatric T-cell acute lymphoblastic leukemia: a report from the Children’s Oncology Group. Leukemia. 2009;23:1417–25.

    CAS  Article  Google Scholar 

  13. 13.

    Neumann M, Vosberg S, Schlee C, Heesch S, Schwartz S, Gökbuget N, et al. Mutational spectrum of adult T-ALL. Oncotarget. 2015;6:2754–66.

    Article  Google Scholar 

  14. 14.

    Park M-J, Taki T, Oda M, Watanabe T, Yumura-Yagi K, Kobayashi R, et al. FBXW7 and NOTCH1 mutations in childhood T cell acute lymphoblastic leukaemia and T cell non-Hodgkin lymphoma. Br J Haematol. 2009;145:198–206.

    CAS  Article  Google Scholar 

  15. 15.

    Balamurugan K, Wang J-M, Tsai H-H, Sharan S, Anver M, Leighty R, et al. The tumour suppressor C/EBPδ inhibits FBXW7 expression and promotes mammary tumour metastasis. EMBO J. 2010;29:4106–17.

    CAS  Article  Google Scholar 

  16. 16.

    Feng D-D, Zhang H, Zhang P, Zheng Y-S, Zhang X-J, Han B-W, et al. Down-regulated miR-331-5p and miR-27a are associated with chemotherapy resistance and relapse in leukaemia. J Cell Mol Med. 2011;15:2164–75.

    CAS  Article  Google Scholar 

  17. 17.

    Kimura T, Gotoh M, Nakamura Y, Arakawa H. hCDC4b, a regulator of cyclin E, as a direct transcriptional target of p53. Cancer Sci. 2003;94:431–6.

    CAS  Article  Google Scholar 

  18. 18.

    Mao J-H, Perez-Losada J, Wu D, Delrosario R, Tsunematsu R, Nakayama KI, et al. Fbxw7/Cdc4 is a p53-dependent, haploinsufficient tumour suppressor gene. Nature. 2004;432:775–9.

    CAS  Article  Google Scholar 

  19. 19.

    Wang L, Ye X, Liu Y, Wei W, Wang Z. Aberrant regulation of FBW7 in cancer. Oncotarget. 2014;5:2000–15.

    PubMed  PubMed Central  Google Scholar 

  20. 20.

    Akhoondi S, Lindström L, Widschwendter M, Corcoran M, Bergh J, Spruck C, et al. Inactivation of FBXW7/hCDC4-β expression by promoter hypermethylation is associated with favorable prognosis in primary breast cancer. Breast Cancer Res. 2010;12:R105.

    CAS  Article  Google Scholar 

  21. 21.

    Cheng Y, Chen G, Martinka M, Ho V, Li G. Prognostic significance of Fbw7 in human melanoma and its role in cell migration. J Invest Dermatol. 2013;133:1794–802.

    CAS  Article  Google Scholar 

  22. 22.

    Gu Z, Inomata K, Mitsui H, Horii A. Promoter hypermethylation is not the major mechanism for inactivation of the FBXW7 beta-form in human gliomas. Genes Genet Syst. 2008;83:347–52.

    CAS  Article  Google Scholar 

  23. 23.

    Sionov RV, Netzer E, Shaulian E. Differential regulation of FBXW7 isoforms by various stress stimuli. Cell Cycle. 2013;12:3547–54.

    CAS  Article  Google Scholar 

  24. 24.

    de Leval L, Bisig B, Thielen C, Boniver J, Gaulard P. Molecular classification of T-cell lymphomas. Crit Rev Oncol Hematol. 2009;72:125–43.

    Article  Google Scholar 

  25. 25.

    Aifantis I, Raetz E, Buonamici S. Molecular pathogenesis of T-cell leukaemia and lymphoma. Nat Rev Immunol. 2008;8:380–90.

    CAS  Article  Google Scholar 

  26. 26.

    Van Vlierberghe P, Ferrando A. The molecular basis of T cell acute lymphoblastic leukemia. J Clin Invest. 2012;122:3398–406.

    Article  Google Scholar 

  27. 27.

    Burkhardt B. Paediatric lymphoblastic T-cell leukaemia and lymphoma: one or two diseases? Br J Haematol. 2010;149:653–68.

    CAS  Article  Google Scholar 

  28. 28.

    Khoury MP, Bourdon J-C. The isoforms of the p53 protein. Cold Spring Harb Perspect Biol. 2010;2:a000927.

    Article  Google Scholar 

  29. 29.

    Grinkevich VV, Nikulenkov F, Shi Y, Enge M, Bao W, Maljukova A, et al. Ablation of key oncogenic pathways by RITA-reactivated p53 is required for efficient apoptosis. Cancer Cell. 2009;15:441–53.

    CAS  Article  Google Scholar 

  30. 30.

    Petitjean A, Mathe E, Kato S, Ishioka C, Tavtigian SV, Hainaut P, et al. Impact of mutant p53 functional properties on TP53 mutation patterns and tumor phenotype: lessons from recent developments in the IARC TP53 database. Hum Mutat. 2007;28:622–9.

    CAS  Article  Google Scholar 

  31. 31.

    Correia NC, Durinck K, Leite AP, Ongenaert M, Rondou P, Speleman F, et al. Novel TAL1 targets beyond protein-coding genes: identification of TAL1-regulated microRNAs in T-cell acute lymphoblastic leukemia. Leukemia. 2013;27:1603–6.

    CAS  Article  Google Scholar 

  32. 32.

    Huang H, Ma L, Li J, Yu Y, Zhang D, Wei J, et al. NF-κB1 inhibits c-Myc protein degradation through suppression of FBW7 expression. Oncotarget. 2014;5:493–505.

    PubMed  PubMed Central  Google Scholar 

  33. 33.

    Kumar V, Palermo R, Talora C, Campese AF, Checquolo S, Bellavia D, et al. Notch and NF-kB signaling pathways regulate miR-223/FBXW7 axis in T-cell acute lymphoblastic leukemia. Leukemia. 2014;28:2324–35.

    CAS  Article  Google Scholar 

  34. 34.

    Mansour MR, Sanda T, Lawton LN, Li X, Kreslavsky T, Novina CD, et al. The TAL1 complex targets the FBXW7 tumor suppressor by activating miR-223 in human T cell acute lymphoblastic leukemia. J Exp Med. 2013;210:1545–57.

    CAS  Article  Google Scholar 

  35. 35.

    Zhou Z, He C, Wang J. Regulation mechanism of Fbxw7-related signaling pathways (Review). Oncol Rep. 2015;34:2215–24.

    CAS  Article  Google Scholar 

  36. 36.

    Malyukova A, Dohda T, von der Lehr N, Akhoondi S, Akhondi S, Corcoran M, et al. The tumor suppressor gene hCDC4 is frequently mutated in human T-cell acute lymphoblastic leukemia with functional consequences for Notch signaling. Cancer Res. 2007;67:5611–6.

    CAS  Article  Google Scholar 

  37. 37.

    Kalender Atak Z, De Keersmaecker K, Gianfelici V, Geerdens E, Vandepoel R, Pauwels D, et al. High accuracy mutation detection in leukemia on a selected panel of cancer genes. PLoS ONE. 2012;7:e38463.

    Article  Google Scholar 

  38. 38.

    Sancho R, Blake SM, Tendeng C, Clurman BE, Lewis J, Behrens A. Fbw7 repression by hes5 creates a feedback loop that modulates Notch-mediated intestinal and neural stem cell fate decisions. PLoS Biol. 2013;11:e1001586.

    CAS  Article  Google Scholar 

  39. 39.

    Kourtis N, Moubarak RS, Aranda-Orgilles B, Lui K, Aydin IT, Trimarchi T, et al. FBXW7 modulates cellular stress response and metastatic potential through ​HSF1 post-translational modification. Nat Cell Biol. 2015;17:322–32.

    Article  Google Scholar 

  40. 40.

    Yumimoto K, Akiyoshi S, Ueo H, Sagara Y, Onoyama I, Ueo H, et al. F-box protein FBXW7 inhibits cancer metastasis in a non-cell-autonomous manner. J Clin Invest. 2015;125:621–35.

    Article  Google Scholar 

  41. 41.

    Rocher-Ros V, Marco S, Mao J-H, Gines S, Metzger D, Chambon P, et al. Presenilin modulates EGFR signaling and cell transformation by regulating the ubiquitin ligase Fbw7. Oncogene. 2010;29:2950–61.

    CAS  Article  Google Scholar 

  42. 42.

    Trausch-Azar JS, Abed M, Orian A, Schwartz AL. Isoform-specific SCF(Fbw7) ubiquitination mediates differential regulation of PGC-1α. J Cell Physiol. 2015;230:842–52.

    CAS  Article  Google Scholar 

  43. 43.

    Puigserver P, Rhee J, Lin J, Wu Z, Yoon JC, Zhang CY, et al. Cytokine stimulation of energy expenditure through p38 MAP kinase activation of PPARgamma coactivator-1. Mol Cell. 2001;8:971–82.

    CAS  Article  Google Scholar 

  44. 44.

    Wei X, Tang C, Lu X, Liu R, Zhou M, He D, et al. MiR-101 targets DUSP1 to regulate the TGF-β secretion in sorafenib inhibits macrophage-induced growth of hepatocarcinoma. Oncotarget. 2015;6:18389–405.

    PubMed  PubMed Central  Google Scholar 

  45. 45.

    O’Neil J, Grim J, Strack P, Rao S, Tibbitts D, Winter C, et al. FBW7 mutations in leukemic cells mediate NOTCH pathway activation and resistance to gamma-secretase inhibitors. J Exp Med. 2007;204:1813–24.

    Article  Google Scholar 

  46. 46.

    Thompson BJ, Buonamici S, Sulis ML, Palomero T, Vilimas T, Basso G, et al. The SCFFBW7 ubiquitin ligase complex as a tumor suppressor in T cell leukemia. J Exp Med. 2007;204:1825–35.

    CAS  Article  Google Scholar 

  47. 47.

    Tremblay CS, Curtis DJ. The clonal evolution of leukemic stem cells in T-cell acute lymphoblastic leukemia. Curr Opin Hematol. 2014;21:320–5.

    CAS  Article  Google Scholar 

  48. 48.

    King B, Trimarchi T, Reavie L, Xu L, Mullenders J, Ntziachristos P, et al. The ubiquitin ligase FBXW7 modulates leukemia-initiating cell activity by regulating MYC stability. Cell. 2013;153:1552–66.

    CAS  Article  Google Scholar 

  49. 49.

    Matsuoka S, Oike Y, Onoyama I, Iwama A, Arai F, Takubo K, et al. Fbxw7 acts as a critical fail-safe against premature loss of hematopoietic stem cells and development of T-ALL. Genes Dev. 2008;22:986–91.

    CAS  Article  Google Scholar 

  50. 50.

    Onoyama I, Tsunematsu R, Matsumoto A, Kimura T, de Alborán IM, Nakayama K, et al. Conditional inactivation of Fbxw7 impairs cell-cycle exit during T cell differentiation and results in lymphomatogenesis. J Exp Med. 2007;204:2875–88.

    CAS  Article  Google Scholar 

  51. 51.

    Gu Z, Mitsui H, Inomata K, Honda M, Endo C, Sakurada A, et al. The methylation status of FBXW7 beta-form correlates with histological subtype in human thymoma. Biochem Biophys Res Commun. 2008;377:685–8.

    CAS  Article  Google Scholar 

  52. 52.

    Lai EC. Two decades of miRNA biology: lessons and challenges. RNA. 2015;21:675–7.

    CAS  Article  Google Scholar 

  53. 53.

    Katt ME, Placone AL, Wong AD, Xu ZS, Searson PC. In vitro tumor models: advantages, disadvantages, variables, and selecting the right platform. Front Bioeng Biotechnol 2016;4;

  54. 54.

    Mavrakis KJ, Van Der Meulen J, Wolfe AL, Liu X, Mets E, Taghon T, et al. A cooperative microRNA-tumor suppressor gene network in acute T-cell lymphoblastic leukemia (T-ALL). Nat Genet. 2011;43:673–8.

    CAS  Article  Google Scholar 

  55. 55.

    Mussolin L, Holmes AB, Romualdi C, Sales G, D’Amore ESG, Ghisi M, et al. An aberrant microRNA signature in childhood T-cell lymphoblastic lymphoma affecting CDKN1B expression, NOTCH1 and growth factor signaling pathways. Leukemia. 2014;28:1909–12.

    CAS  Article  Google Scholar 

  56. 56.

    Hu J, Wu C, Zhao X, Liu C. The prognostic value of decreased miR-101 in various cancers: a meta-analysis of 12 studies. Onco Targets Ther. 2017;10:3709–18.

    Article  Google Scholar 

  57. 57.

    Welcker M, Orian A, Jin J, Grim JE, Grim JA, Harper JW, et al. The Fbw7 tumor suppressor regulates glycogen synthase kinase 3 phosphorylation-dependent c-Myc protein degradation. Proc Natl Acad Sci USA. 2004;101:9085–90.

    CAS  Article  Google Scholar 

  58. 58.

    Swerdlow SH, Campo E, Pileri SA, Harris NL, Stein H, Siebert R, et al. The2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood. 2016;127:2375–90.

    CAS  Article  Google Scholar 

  59. 59.

    Andrés-León E, González Peña D, Gómez-López G, Pisano DG. miRGate: a curated database of human, mouse and rat miRNA-mRNA targets. Database (Oxf). 2015;2015:bav035.

    Google Scholar 

  60. 60.

    Mullokandov G, Baccarini A, Ruzo A, Jayaprakash AD, Tung N, Israelow B, et al. High-throughput assessment of microRNA activity and function using microRNA sensor and decoy libraries. Nat Methods. 2012;9:840–6.

    CAS  Article  Google Scholar 

  61. 61.

    van Drogen F, Sangfelt O, Malyukova A, Matskova L, Yeh E, Means AR, et al. Ubiquitylation of cyclin E requires the sequential function of SCF complexes containing distinct hCdc4 isoforms. Mol Cell. 2006;23:37–48.

    Article  Google Scholar 

  62. 62.

    Guo D, Ye J, Dai J, Li L, Chen F, Ma D, et al. Notch-1 regulates Akt signaling pathway and the expression of cell cycle regulatory proteins cyclin D1, CDK2 and p21 in T-ALL cell lines. Leuk Res. 2009;33:678–85.

    CAS  Article  Google Scholar 

  63. 63.

    Jin HY, Gonzalez-Martin A, Miletic AV, Lai M, Knight S, Sabouri-Ghomi M, et al. Transfection of microRNA Mimics Should Be Used with Caution. Front Genet. 2015;6:340.

    PubMed  PubMed Central  Google Scholar 

Download references


The authors thank the Spanish Biobanks for providing us with the T-LBL samples to elaborate this work. Dr. Grim (Fred Hutchinson Cancer Research Centre) and Dr. Van Santen (CBMSO) for providing us with the CMV-FBXW7 constructs and packing vectors for retroviral transduction, respectively. The Cytometry, Cell Culture and Genomic services of the CBMSO for technical support. We thank Dr. Iria González-Vasconcellos (CBMSO-UAM) for the critical reading of this paper.


Spanish Ministry of Economy and Competitiveness (SAF2015-70561-R; MINECO/FEDER, EU; BES-2013-065740); the Autonomous Community of Madrid, Spain (B2017/BMD-3778; LINFOMAS-CM); the Spanish Association against Cancer (AECC, 2018; PROYE18054PIRI) and the Instituto de Salud Carlos III (ISCIII) (ACCI-CIBERER-17). Institutional grants from the Fundación Ramón Areces and Banco de Santander to the CBMSO are also acknowledged.

Author information



Corresponding author

Correspondence to José Fernández-Piqueras.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Vázquez-Domínguez, I., González-Sánchez, L., López-Nieva, P. et al. Downregulation of specific FBXW7 isoforms with differential effects in T-cell lymphoblastic lymphoma. Oncogene 38, 4620–4636 (2019).

Download citation

Further reading


Quick links