Disruption of rack1 suppresses SHH‐type medulloblastoma formation in mice

Abstract Introduction Medulloblastoma (MB) is a malignant pediatric brain tumor that arises in the cerebellar granular neurons. Sonic Hedgehog subtype of MB (SHH‐MB) is one of the major subtypes of MB in the clinic. However, the molecular mechanisms underlying MB tumorigenesis are still not fully understood. Aims Our previous work demonstrated that the receptor for activated C kinase 1 (Rack1) is essential for SHH signaling activation in granule neuron progenitors (GNPs) during cerebellar development. To investigate the potential role of Rack1 in MB development, human MB tissue array and SHH‐MB genetic mouse model were used to study the expression of function of Rack1 in MB pathogenesis. Results We found that the expression of Rack1 was significantly upregulated in the majority of human cerebellar MB tumors. Genetic ablation of Rack1 expression in SHH‐MB tumor mice could significantly inhibit MB proliferation, reduce the tumor size, and prolong the survival of tumor rescue mice. Interestingly, neither apoptosis nor autophagy levels were affected in Rack1‐deletion rescue mice compared to WT mice, but the expression of Gli1 and HDAC2 was significantly decreased suggesting the inactivation of SHH signaling pathway in rescue mice. Conclusion Our results demonstrated that Rack1 may serve as a potential candidate for the diagnostic marker and therapeutic target of MB, including SHH‐MB.

of SHH signaling leads to excessive proliferation of GNPs which will lead to SHH-MB formation. 8,14,15 Attempts were carried out recently to recognize possible new therapeutic targets, particularly targeting Smoothened (SMO), 16,17 which is the upstream regulator of the SHH signaling pathway. 18 However, despite many targets of MB have been identified, many of those show resistance and progression of disease after the initial inhibitory effect in SHH-MB. 19,20 Rack1 (the receptor for activated C kinase 1) is a spheroidal scaffolding protein composed of seven WD40 repeating domains. 21 The structural characteristics of Rack1 enable it to bind a variety of proteins and play plenty of regulatory functions in multiple cells.
According to previous studies, Rack1 has close relevance to tumor formation. Its expression is significantly increased in liver cancer, non-small cell lung cancer, and glioma, but decreased in gastric cancer and colon cancer. 22 Rack1 deficiency can inhibit the proliferation and migration of neuroblastoma in vitro. 23 Nevertheless, the potential function of Rack1 in the cerebellar MB formation remains elusive.
Our previous study has demonstrated that Rack1 is essential for SHH signaling activation in GNPs during postnatal mammalian cerebellar development. 24 Ablation of Rack1 in GNPs results in a significant deficiency of proliferation and migration in the developing cerebellum, 24 raising the possibility of whether ablation of Rack1 could suppress SHH-MB formation. In this study, we firstly observed a significant increase in Rack1 expression in human cerebellar MB tumors.
Subsequently, using a SHH-MB genetic mouse model, Atoh1-Cre; SmoM2 +/− , we demonstrated that ablation of Rack1 could significantly inhibit the growth of tumor cells with a dramatically extended lifespan in rescue mice. Meanwhile, Rack1 deficiency did not cause significantly increased cell death owing to apoptosis or autophagy analysis, but dramatically reduced the proliferation of MB tumor cells. Mechanistic studies indicate that the expression of crucial molecules downstream of SHH signaling pathway, including Gli1 and HDAC2, was significantly decreased. These results suggest that the tumor suppression effect in Rack1-deficient SHH-MB mice might be related to the reduced hyperactivation of the SHH signaling pathway. Therefore, our study indicates that Rack1 may serve as a potential candidate for the diagnostic marker and therapeutic target of MB, including SHH-MB in the clinic.

| Human MB tissue array
Human paraffin-embedded tissue array was purchased from Alivabio (DC-Bra01011b). The patients' sex, age, and histological data were included in Table 1. Tumor tissue array contained 65 points including 20 MB patients (4-49 years old), one healthy human cerebellum, and one patient's adrenal gland. The diameter for each point of tissue is 1.5 mm.

| Mouse lines
The Atoh1-Cre, Ai9, Rack1 F/F , and SmoM2 +/+ lines were generated as previously described. 25 All mice were genotyped by PCR assay using murine tail DNA.
Sequences of the oligo nucleotides used for the genotyping were provided in Table 2.
And OCT (Tissue-tek, 4583) was used to embed the brain, and frozen sections were performed. The immunofluorescent staining (IF) of frozen cerebellar sections was performed using standard techniques as previously described. 24 Briefly, frozen sections (40 μm) were washed for 10 min with 0.3% Triton X-100/PBS (PBST) for three times and then covered with 3% BSA in PBST (0.3% Triton X-100) for 1 h. The sections were then incubated with primary antibodies overnight at 4°C. Subsequently, the sections were washed TA B L E 1 Histological data of human paraffin-embedded tissue array The H&E staining of frozen cerebellar sections was performed using standard techniques as previously described. 24 Briefly, slides were stained in hematoxylin solution for 8 min, then wash in running tap water for 5 min. Subsequently, differentiate in 1% acid alcohol for 30 s and wash running tap water for 1 min. Counterstain in eosinphloxine solution for 30 s to 1 min, and dehydrate through 95% alcohol, two changes of absolute alcohol, 5 min each. Finally, the slides were cleared in two changes of xylene and mounted with xylenebased mounting medium. (pH 8.5) for 10 min at room temperature. After three times wash in PBS, continue with standard immunohistochemistry protocol as described above.

| Western blot analysis
The experiments were performed as previously described. 24 Briefly, cerebellar tissues were lysed by RIPA lysis and extraction The film signal was electronically scanned and analyzed by Image Pro Plus.

| Statistical analysis
The cerebellar immunofluorescence images were taken from 7 × 7 slices for confocal scanning or imaged with 20× (zoom 2.0) objective multiplelayer scans. Western blot results and cerebellar section areas were statistically analyzed using GraphPad Prism 8.0 and NDP view. The data between the two independent groups were shown as mean ± SEM for at least three independent experiments. The experimental data were analyzed using the non-parametric t-test. p < 0.05 (*); p < 0.01(**); p < 0.001(***), or p < 0.0001 (****) was considered a statistically significant difference.

| Upregulated expression of Rack1 in some MB samples
Our previous studies have demonstrated that Rack1 plays critical roles in regulating granule neuron precursors (GNPs) proliferation and migration during mammalian cerebellar development. 24 Figure 2B). Accordingly, we showed that the average lifespan of the rescued mice model was prolonged over 2-fold (range from 58 to 89 days; Figure 2B). Together, these results indicated that suppression of Rack1 expression in GNPs could significantly inhibit SHH-MB tumor growth and increased the survival rate of rescued mice.

| Ablation of Rack1 suppresses MB proliferation instead of induction of the cell death
To

| Rack1 deficiency in MB causes the inactivation of the SHH signaling
Given that SHH signaling was significantly suppressed in Rack1deficient GNPs, 24   Our previous studies have reported that Rack1 plays a crucial role in regulating the proliferation and migration of GNPs by inhibiting the degradation of HDAC1/2 and therefore promoting the activation of Gli1/2 during postnatal cerebellar development. 24 It has also been demonstrated that Rack1 promotes non-small-cell lung cancer (NSCLC) tumorigenicity by activating Smoothened to mediate Gli1-dependent transcription in cancer cells. 46 In this study, we also found that the expression of HDAC2 and Gli1 were both significantly downregulated in rescue mice model, further indicating the crucial role of Rack1-mediated SHH signaling activation in promoting SHH-type MB tumorigenesis. In addition, we also found the significantly downregulated expression of cdk4 in both Rack1 cKO and rescue mice. In agreement with our studies, previous studies have also shown that SHH signaling drives MB growth via CDK6, and inhibition of CDK4/6 significantly attenuates the growth of SHH-MB. 47 Moreover, previous studies have identified multiple mutant genes involved in brain tumor pathogenesis. [48][49][50][51] Given that the significantly upregulated expression of Rack1 in different subtypes of MB tumors based on our immunostaining and gene profiling assay using GEO DataSet (Figure 1 and Figure S4), it would be interesting to ask whether Rack1 gene mutations are involved in the pathogenesis of brain tumors. Interestingly, according to the Cancer Genome Atlas (TCGA) database, we found that the gene mutation rate of Rack1 in brain tumor samples was about 15.38% (8/52) based on the gene profiling assay (data not shown). These results indicated that Rack1 mutagenesis might be involved in the pathogenesis and progression of several types of brain tumors, such as gliomas and neuroblastoma, 52,53 therefore, more detailed mutation sites and the underlying signaling mechanisms of Rack1 in brain tumors need further characterization and investigation.
In conclusion, we demonstrated that the levels of Rack1 were upregulated in majority of MB tumor samples, and genetic ablation of Rack1 in SHH-MB tumor mice significantly reduced MB proliferation, reduced the tumor size, and prolonged the survival of tumor rescue mice. We propose that Rack1-mediated Gli1 signaling and cell cycle activation in MB tumor cells might be essential for SHH-MB tumorigenesis. Rack1 could serve as a potential diagnostic marker and therapeutic target for MB patients in clinic.