YME1L overexpression exerts pro-tumorigenic activity in glioma by promoting Gαi1 expression and Akt activation

Identifying novel glioma-driven signaling molecules and explor-ing the corresponding molecularly targeted therapies

In this study, we will show that YME1L overexpression exerts pro-tumorigenic activity in glioma by promoting Gαi1 expression and Akt activation. First, The Cancer Genome Atlas (TCGA) database was first consulted to retrieve YME1L RNA sequencing data in human glioma. As shown, in the human glioma tissues ("Tumor," n = 166), the number of YME1L mRNA transcripts is significantly higher than that in the normal brain tissues ("Normal," n = 1,157) (P < 0.001, Fig. 1A). Of the normal brain tissues, 1,152 of them were retrieved from the Genotype-Tissue Expression (GTEx) database and five tissues were from TCGA database (tumor-surrounding normal brain tissues) (Fig. 1A). The subgroup analyses based on clinical characteristics showed that high YME1L mRNA expression in human glioma tissues was correlated with IDH (isocitrate dehydrogenase) mutation (P < 0.001, Fig. 1B). It was not correlated with age ( Fig. 1C) and sex (Fig. 1D) of the patients.
To confirm the significance of the bioinformatics observations, we tested YME1L expression in local human glioma tissues ("T") and surrounding normal brain ("N") tissues, from a total of 16 grade III-IV glioma (HGG) patients (see our previous studies such as Liu et al. (2018) and Wang et al. (2021). The real-time quantitative reverse transcription PCR (qRT-PCR) assay results in Fig.  1E showed that YME1L mRNA expression in glioma tissues was (A)TCGA database shows YME1L expression (RNA-Seq) in glioma tissues ("Tumor", n = 166) and in the normal brain tissues ("Normal", n = 1,157). (B-D)The subgroup analyses of YME1L mRNA expression and clinical characteristics of glioma patients were shown. (E) Human glioma tissues ("T") and the paired normal brain tissues ("N") derived from a total of 16 HGG patients were homogenized and dissolved in the tissue lysis buffer, YME1L mRNA and protein expression was tested by qRT-PCR and (F-G) Western blot assays, respectively, with results quantified. (H) The human tissue immuno-fluorescence images of YME1L (green fluorescence) and the MitoTracker red in the glioma slide and the adjacent normal brain slide of one representative glioma patient (Patient 3). (I) Expression of listed proteins in mitochondrial lysates and mitochondria-null lysates of two representative glioma patients (Patients 3 and 4) was shown. (J) The P1 primary human glioma cells, stably expressing the applied YME1L shRNA ("lv-shYME1L-seq1/2", two different sequences) or the lenti-CRSIPR/Cas9-YME1L-KO-puro construct ("koYME1L") were established. Control P1 glioma cells were transduced with the lentiviral scramble shRNA plus the CRSIPR/Cas9 empty vector ("lv-shC + Cas9-C"). Expression of YME1L mRNA and (K) listed proteins (in mitochondrial lysates and mitochondria-null lysates) was shown. (L) Cells were further cultured for applied time periods, cellular functions, including cell viability (CCK-8 OD), (M) cell proliferation (EdU staining assays) and (N) cell migration were tested by the mentioned assays. "Pare" stands for the parental control cells. The data were presented as mean ± standard deviation (SD). *P < 0.05 vs. "Normal"/"N"/"Pare". "ns" stands for non-statistical difference (P > 0.05). (M-N) The in vitro experiments were repeated five times with similar results obtained. Scale bar = 50 μm (H). Scale bar = 100 μm. significantly higher than that in the normal tissues. Testing YME1L protein expression, using Western blot assays, further confirmed YME1L protein upregulation in glioma tissues of four representative glioma patients (Patients 1-4, Fig. 1F). Western blot quantification results further confirmed that YME1L protein upregulation is significant in glioma tissues (P < 0.001 vs. "N" tissues, Fig. 1G).
The tissue immuno-fluorescence images, Figs. 1H and S1A, show that YME1L protein (green fluorescence) is co-localized with the mitochondrial marker MitoTracker (red fluorescence) in both glioma slides and the adjacent normal brain slides of two representative glioma patients (Patients 3 and 4). More importantly, YME1L fluorescence intensity in the human glioma slides was significantly higher than that in the adjacent normal brain tissue (Figs. 1H and S1A).
Furthermore, an examination of mitochondrial lysates isolated from fresh human glioma tissues of four representative patients (Patients 1-4) confirmed that YME1L was enriched in the mitochondria fraction (Figs. 1I and S1B), as indicated by VDAC1 (voltage-dependent anion-selective channel 1), a mitochondrial marker protein (Figs. 1I and S1B). Lamin-B1 is a nuclear marker protein and α-tubulin is a cytosol marker protein (Figs. 1I and S1B). Once again, mitochondrial YME1L protein expression in glioma tissues was significantly elevated (Figs. 1I and S1B). Conversely, YME1L protein expression was almost not detected in the mitochondria-null lysates of human tissues (Figs. 1I and S1B). These results show that YME1L protein is upregulated and localized to the mitochondria of glioma tissues.
YME1L expression in the glioma cells was tested next. The established glioma cell lines, A172 and U251, as well as the primary human glioma cells that were derived from three different patients, "P1", "P2", and "P3" , were tested. YME1L mRNA expression in the glioma cells was significantly higher than that in the primary human astrocytes (Fig. S1C). Moreover, YME1L protein upregulation was detected in the immortalized and primary glioma cells (Fig. S1D). Whereas in the primary astrocytes, YME1L protein expression is low (Fig. S1D). These results together showed that YME1L is upregulated in human glioma.
To silence YME1L expression, the P1 primary human glioma cells (Liu et al., 2018;Wang et al., 2021) were individually transduced with two different lentiviral YME1L small hairpin RNA (shRNA) (lv-shYME1L-seq1 and lv-shYME1L-seq2, with different sequences). Stable cells were established following selection by puromycin. Alternatively, the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) method was utilized. A lenti-CRSIPR/Cas9-YME1Lknockout (KO)-puro construct was transduced to the P1 glioma cells. Single stable cells were established by puromycin selection and YME1L KO screening, namely koYME1L cells. As compared to control P1 glioma cells with the lentiviral scramble shRNA (shC) plus the CRSIPR/Cas9 empty vector ("lv-shC + Cas9-C"), YME1L mRNA expression was dramatically downregulated in shYME1L-expressing cells and koYME1L cells (Fig. 1J). Western blot testing the mitochondrial fraction lysates confirmed specific depletion of YME1L protein in the mitochondria of P1 human glioma cells by the applied shRNA and KO strategies (Fig. 1K). While its expression was not detected in mitochondria-null lysates of P1 glioma cells with or without the applied genetic treatments (Fig. 1K).
The primary human glioma cells that were derived from two other patients ["P2" and "P3" (Liu et al., 2018;Wang et al., 2021)] as well as the immortalized cell lines (A172 and U251) were cultured and infected with lv-shYME1L-seq1-expressing lentivirus. Stable cells were again established by puromycin selection. The qRT-PCR assay results, Fig. S1E, confirmed that YME1L mRNA levels were robustly decreased in the glioma cells with YME1L shRNA. shRNA-induced silencing of YME1L largely inhibited the viability (CCK-8 OD) of the primary and established glioma cells (Fig. S1F).
Moreover, cell proliferation, tested by the EdU-positive nuclei ratio (Fig. S1G), and in vitro cell migration ("Transwell" assays, Fig. S1H) were potently inhibited by the YME1L shRNA. Together, YME1L silencing or KO resulted in significant anti-glioma cell activity, inhibiting cell survival, proliferation, and migration.
Whether YME1L depletion could provoke apoptosis activation in glioma cells was studied next. As shown, the caspase-3 activity and the caspase-9 activity were both significantly increased in the YME1L-silenced or the koYME1L P1 glioma cells (Fig. S2A). Figure S2B showed that YME1L shRNA or KO resulted cleavages of caspase-3 and poly(ADP-ribose) polymerase (PARP) in P1 glioma cells. In addition, the histone-associated DNA fragments were robustly increased in P1 glioma cells after YME1L silencing or KO (Fig. S2C). The terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)-positive nuclei ratio was significantly increased in P1 glioma cells with YME1L depletion (Fig. S2D). As expected, the lentiviral scramble shRNA plus the CRSIPR/Cas9 empty vector ("lv-shC + Cas9-C") did not induce significant apoptosis activation in P1 glioma cells (Fig. S2A-D).
In other primary human glioma cells (P2 and P3) and immortalized cell lines (A172 and U251), YME1L silencing by lv-shY-ME1L-seq1 (see Fig. 1) similarly induced caspase-3 activation (Fig. S2E) and cell apoptosis (evidenced by nuclear TUNEL ratio increase, Fig. S2F). In the primary human astrocytes, infection with the lv-shYME1L-seq1 resulted in robust YME1L silencing as well (Fig. S2G). Yet, the cell viability (tested by the CCK-8 OD, Fig.  S2H) and cell apoptosis (tested by the TUNEL-positive nuclei ratio, Fig. S2I) were not significantly affected by YME1L silencing in the astrocytes. These results implied a glioma cell-specific effect by YME1L depletion.
We next hypothesized that further increasing YME1L expression should exert cancer-promoting activity in glioma cells. The lentivirus encoding the wild-type YME1L cDNA was transfected to P1 glioma cells. Puromycin was then utilized to select stable cells: OE-YME1L-sL1 and OE-YME1L-sL2 (two lines). Fig. S4A confirmed that the YME1L mRNA expression was significantly elevated in OE-YME1L P1 glioma cells (versus vector control cells/"Vec"). In addition, significant YME1L protein elevation in total cell lysates was detected (Fig. S4B). YME1L protein upregulation was detected only in the mitochondria of P1 glioma cells with the YME1Loverexpressing lentiviral construct (Fig. S4B). Again no YME1L protein expression was detected in the mitochondria-null lysates (Fig. S4B).
Next, TCGA database results were retrieved and differentially expressed gene (DEGs) analyses were performed to examine co-expression genes with YME1L in glioma tissues. By employing the Pearson Correlation Coeffcient analyses, the co-expression volcano map was shown ( Fig. 2A). The top 20 DEGs that were upregulated in YME1L-high glioma tissues were presented (Fig.  2B). One key gene is GNAI1 (encoding Gαi1 protein, Fig. 2B). Our previous studies have shown that Gαi1 associated with multiple RTKs in human glioma, required for downstream Akt activation and glioma tumorigenesis (Liu et al., 2018;Wang et al., 2021). Conversely, Gαi1 silencing, knockout, or mutation largely inhibited glioma cell growth (Liu et al., 2018;Wang et al., 2021).
To study the potential effect of YME1L on glioma cell growth in vivo, the P1 glioma cells were s.c. injected to the nude mice. Within 3 weeks of cell inoculation, P1 glioma xenografts were established ("Day-0," with tumor volume close to 100 mm 3 ). The xenograft-bearing nude mice were then randomly assigned into three groups and were subject to intratumoral injection of adeno-associated virus (aav)-packed shRNA, including aav-shY-ME1L-seq1, aav-shYME1L-seq2 or aav-shC. The aav injection was performed daily for 14 consecutive days. The tumor growth curve results, recording tumor volumes every 6 days ("Day-0" to "Day-42"), showed that injection of shYME1L aav potently inhibited P1 glioma xenograft growth in nude mice (Fig. S5A). The volumes of aav-shYME1L-injected tumors were significantly lower than those with aav-shC injection (Fig. S5A). The estimated daily tumor growth was calculated and the following formula was utilized: (Tumor volume at "Day-42" − Tumor volume at "Day-0")/42. Results showed that P1 glioma xenograft growth was largely inhibited after injection aav-shYME1L (Fig. S5B). P1 glioma xenografts were all isolated and weighted at "Day-42.". We found that aav-shYME1L-injected xenografts were significantly lighter than aav-shC-injected xenografts (Fig. S5C). The mice body weights, on the other hand, were not significantly different between the three groups ( Fig. S5D).
At experimental "Day-5" and "Day-10," 3 h after the aav injection, one mouse in each group was killed after anesthesia, and tumor resections were performed. A total of six glioma xenografts were obtained and tumor lysates were tested. As shown, YME1L mRNA levels were dramatically decreased in the aav-shYME1L-injected tumors (Fig. S5E). YME1L protein silencing was detected as well (Fig. S5F). In addition, levels of Gαi1 and p-Akt were decreased in YME1L-silenced xenografts (Fig. S5F). Therefore, in line with the in vitro signaling findings, aav-shRNA-induced silencing of YME1L inhibited Gαi1 expression and Akt activation in P1 glioma xenografts. On the contrast, levels of cleaved-caspase-3 and cleaved-PARP were increased in YME1L-silenced xenografts (Fig. S5G), indicating apoptosis activation.
We have previously shown that Gαi1 mRNA and protein expression was significantly elevated in human glioma tissues, being more dramatic in high-grade gliomas (Liu et al., 2018). Overexpressed Gαi1 is associated with multiple RTKs, required for downstream Akt activation and glioma cell growth (Liu et al., 2018). Conversely, Gαi1 shRNA, dominant negative mutant interference, complete KO, or expressing the anti-Gαi1 miR-200a inhibited Akt activation and glioma cell growth (Liu et al., 2018). Moreover, Gαi1/3 mediation of neuroligin-3-induced downstream signaling is essential for neuronal-driven glioma intracranial growth . These results verified that Gαi1 should be an important therapeutic target of human glioma.
YME1L is important for Gαi1 expression in glioma cells. Gαi1 expression and downstream Akt activation were decreased after YME1L silencing or KO, but were augmented with YME1L overexpression in primary glioma cells. In vivo, Gαi1 expression and Akt activation were largely inhibited in YME1L-silenced or YME1L-KO glioma xenograft tissues. Importantly, Gαi1 re-expression, by Ad-Gαi1, restored Akt activation and largely inhibited YME1L KO-induced anti-glioma cell activity. In addition, restoring Akt activation, by caAkt1, also alleviated YME1L KO-induced proliferation inhibition and apoptosis in glioma cells. These results clearly supported that YME1L-driven glioma cell progression is mediated, at least in part, by mediating Gαi1-Akt cascade. The underlying signaling mechanisms warrant further cauterizations. In conclusion, by promoting Gαi1 expression and Akt activation, YME1L overexpression exerts significant pro-tumorigenic activity in glioma. YME1L should be an important therapeutic target of human glioma.