Cyclin C promotes pancreatic developmentand suppresses cancer initiation by maintenance of the autophagy-lysosome pathway
Sara E. Hanley, Kathy Q. Cai, Stephen D. Willis, David C. Stieg, Andres J. Klein-Szanto, Kerry S. Campbell, Randy Strich
TL;DR
Cyclin C supports pancreatic health by regulating autophagy and preventing cancer, and its absence increases sensitivity to proteasome inhibitors.
Contribution
Cyclin C's dual role in pancreatic development and cancer suppression via autophagy and mitochondrial function is newly identified.
Findings
Cyclin C is essential for autophagy-related gene transcription in mouse embryonic fibroblasts.
Loss of Ccnc in the pancreas leads to islet atrophy, acinar cell damage, and accelerated precancerous lesions.
Ccnc−/− cells with autophagy deficiency are hypersensitive to proteasome inhibitors.
Abstract
Cyclin C (CCNC) is a component of the mediator complex that regulates gene transcription. In stressed cells, cyclin C also translocates to the mitochondria to induce fission and stimulate programmed cell death. The present study found that cyclin C is required for autophagy lysosome pathway gene transcription in mouse embryonic fibroblasts. In vivo, pancreatic ablation of Ccnc caused islet atrophy and acinar cell damage. However, Ccnc pancreatic ablation caused more dramatic phenotypes than autophagy mutants alone, including increased mortality and accelerated precancerous lesion formation. Previous studies found that autophagy-deficient pancreatic cells expressing oncogenic Kras undergo Tp53-dependent cell death. However, KrasG12D;Ccnc−/− pancreata did not undergo cell death, suggesting a role for the mitochondrial cyclin C function. Finally, loss of CCNC activity rendered cells…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAutophagy in Disease and Therapy · Cancer-related Molecular Pathways · Hedgehog Signaling Pathway Studies
Introduction
Pancreatic ductal adenocarcinoma (PDAC) is the most common pancreatic cancer being diagnosed in about 64,000 individuals in the US in 2021 and resulting in about 50,000 deaths.1 Prognosis is extremely poor (5-year relative survival rate of <5%) since most patients are diagnosed with advanced disease. Sporadic PDAC is often associated with KRAS activation with cells undergoing acinar-to-ductal metaplasia (ADM), leading to pancreatic intraepithelial neoplasia (PanIN) arising in these ducts. Normally, ADM lesions undergo proliferation but then re-differentiate back to functional acinar cells following repair. However, in the presence of activated KRAS, de-differentiated ADM cells are locked into a proliferation program, generating PanINs.2^,^3 More advanced disease is associated with inactivation of tumor suppressors such as CDKN2/INK4A and/or p53.4 Chronic pancreatitis is often a preamble to PDAC by causing acinar cell damage and ADM. Acinar pathology involves inflammatory injury and can be caused by multiple agents including alcohol, genetic, and/or environmental factors.5
The autophagy-lysosomal pathway (ALP) is a proteolysis system that identifies and destroys various cargos including toxic protein aggregates, damaged organelles, or recycles cytoplasmic content in response to nutrient deprivation.6 Two major types of autophagy have been described. First, bulk or non-selective autophagy is induced in response to starvation and usually degrades cytoplasmic constituents to provide metabolites to maintain viability.6 Selective autophagy uses receptors that target specific cellular components including damaged organelles or protein aggregates.7 In the pancreas, ALP is essential for normal β cell formation, insulin production,8 and the prevention of many disease states.9 For example, autophagy prevents pathologies such as pancreatitis and cancer initiation.10 In murine models, ablation of autophagic genes (Atg5 or Atg7) causes ADM or can trigger the intrinsic regulated cell death (iRCD) pathway.11^,^12
Under normal conditions, cyclin C and its cognate kinases Cdk8 or Cdk19 are components of the Mediator kinase module (MKM)13 that regulates transcription through association with RNA polymerase II mediator complex.14 Recent transcriptome analysis revealed that, depending on the locus, cyclin C represses or activates the transcription of ∼3,000 genes in mouse embryonic fibroblasts (MEFs).15 Many of these genes are involved in stress responses or differentiation pathways. In addition, we discovered a second role for cyclin C (Ccnc) that is independent of transcription. In cells subjected to oxidative damage or anti-cancer drugs, a portion of cyclin C (but not Cdk8 or Cdk19) exits the nucleus and associates with the mitochondria.16 There, cyclin C directly interacts with the fission dynamin-like GTPase Drp1 to stimulate mitochondrial fragmentation.16^,^17 In addition, cyclin C is required for iRCD by stimulating mitochondrial outer membrane permeability (MOMP)16 through recruitment of the pro-apoptotic protein Bax.18 Finally, mouse knockout studies revealed that cyclin C is required for normal development with embryos arresting around day 9.5.19 Conversely, Cdk8 deletions arrest prior to implantation while Cdk19 null embryos grow into adulthood.20 Therefore, developmental roles of the MKM appear more complex than simply altering transcription.
In the present study, we found that a conditional Ccnc knockout in the pancreas resulted in reduced autophagy, defective islet development, and premature animal death. Combining the Ccnc knockout with the activated Kras^G12D^ allele resulted in a rapid acceleration of ADM and PanIN formation compared to Kras^G12D^ activation or Ccnc^−/−^ deletion alone. In addition, cyclin C is also required for p53-induced cell death in transformed pancreatic cells. These results support a model that cyclin C supports pancreatic development and suppresses pancreatic cancer initiation through maintaining autophagy and triggering cell death in damaged cells to prevent tumorigenesis.
Results
Cyclin C is required for steady-state and induced autophagic gene transcription
We previously determined the Ccnc-dependent oxidative stress transcriptome in MEFs.21 The RNA sequencing (RNA-seq) data have been previously deposited at GEO repository and are publicly available. The accession number is listed in the key resources table. These RNA-seq data indicated that cyclin C is required for the steady-state transcription of several genes sets including those involved in autophagy. To verify these results, quantitative reverse-transcription PCR (RT-qPCR) was performed on a subset of these genes with mRNA prepared from wild-type and Ccnc^−/−^ MEF cells under nonstress conditions. As a control, we measured the mRNA levels of Sqstm1/p62, an autophagy gene22 that we previously found not controlled by cyclin C. Genes involved in autophagosome assembly (Becn1, Uvrag, Agt9b, and Map1LC3B) exhibited reduced steady-state expression in the absence of Ccnc (Figure 1A). Agt101 was the exception for genes examined. To determine if cyclin C is also required for autophagic gene induction, cells were treated with a sub-lethal concentration of the proteasome inhibitor MG-132,23 which has been shown to induce ALP in certain instances.24^,^25 These studies revealed that Ccnc is also required for the induction of these autophagy genes (Figure 1A). We repeated these experiments with genes involved in autophagosome regulation (Trp53inp1, Gabarap, Dram1, and Tollip) (Figure 1B) or autophagosome-lysosome fusion (Plekhm1, Tecpr1, Stx17, Snap29, Vps33a, and Vps16) (Figure 1C). Steady-state levels of their mRNA decreased in Ccnc^−/−^ cells, although the extent of this reduction varied. Most genes failed to be induced upon MG-132 treatment with Plekhm1 transcript levels being the exception (Figure 1C). These results confirmed a positive role for cyclin C in supporting both steady state and induced mRNA levels of autophagy genes.Figure 1. Cyclin C is required for normal autophagic gene transcription(A and B) RT-qPCR analysis of mRNA levels of genes involved in autophagosome assembly (A) or autophagy regulation (B) in Ccnc^+/+^ and Ccnc^−/−^ MEF cells with or without MG-132 treatment (5 μM, 1 h). The horizontal line at 0 represents untreated wild-type mRNA quantitation. Values above or below this line represent the log_2_ increase or decrease in mRNA abundance, respectively.(C) RT-qPCR quantitation of mRNA of loci involved in autophagosome maturation-lysosome fusion. Asterisks indicate p values ∗ ≤0.05, ∗∗ ≤0.01, ∗∗∗ ≤0.005, ∗∗∗∗ ≤0.001 from Student’s t test. Error bars = ±SD. N = 3 biological replicates.
Ccnc supports normal pancreatic islet development and insulin production
Previous studies ablated the autophagy genes Atg512 or Atg711 in the murine pancreas and observed defective pancreatic β cell development, reduced insulin production, and acinar cell damage. Therefore, we chose the pancreas as an organ system to provide a specific readout for ALP activity. We chose the Pdx1-cre mouse model system, which induces the cre recombinase in early pancreatic progenitor cells.26 Pdx1-cre mice were mated to a Ccnc-floxed (Ccnc^f^) mouse16 generating Ccnc^f/f^;Pdx1-cre (CC) mice. Mice harboring Atg5 or Atg7 ablation exhibited reduced viability although ∼40% of the animals exhibited a normal lifespan.27 We found that all CC animals displayed a dramatic reduction in viability compared to control mice with a median lifespan of 10 weeks (Figure 2A). By 8 weeks, the CC animals all exhibited reduced weight (Figure 2B), severe lethargy, and reached humane endpoints by 15 weeks of age. We found no differences between males and females for this phenotype. Reduced weight in the CC animals suggested the possibility of pancreatic dysfunction. Therefore, we examined islet formation by immunohistochemistry (IHC) and measured serum glucose levels in CC animals. Compared to control animals, serum glucose levels were elevated in CC animals (Figure 2B), nearly identical to those previously described after Atg5 ablation.12 Compared to wild-type controls (Figure 2C), histopathology revealed dysmorphic islet appearance and reduced insulin production in CC pancreata (Figure 2D). These results are consistent with a model that Ccnc is required for autophagy to maintain β cell integrity in the pancreas.Figure 2. Cyclin C supports pancreatic development and function(A) Kaplan-Meier survival plot of Kras^G12D/+^;Pdx1-cre (KC), Ccnc^−/−^;Pdx1-cre (CC, 6:5 M:F), and Kras^G12D/+^;Ccnc^−/−^;Pdx1-cre (KCC, 7:8 M:F) animals as indicated. p value given for indicated survival curves.(B) Weight and plasma glucose levels are listed for 8-week-old animals with the indicated genotypes. Glucose concentrations were obtained over multiple time points (see STAR Methods for details).(C–F) Corresponding images of 11- and 12-week-old animals with the indicated genotypes stained with H&E or tested for insulin production via IHC. Open arrows indicate normal islet morphology and insulin production. Bars, 100 μM. p values using Student’s t test and experimental trials are indicated.
Finally, previous studies revealed that autophagy prevented acinar cell damage that caused pancreatitis or ADM.11^,^12 To test whether either result was also observed in CC animals, pancreata were obtained and analyzed by H&E staining. These studies revealed ADM in CC mice although with only partial penetrance (five out of 33 mice, ages between 5 and 13 weeks). The ADM lesions ranged from small focal regions (n = 2, Figures 3A and 3B) to extensive regions covering most of the pancreas (n = 3, Figures 3C and 3D). This result is similar to those obtained with pancreatic ablation of Atg5,12 indicating that cyclin C prevents ADM, but its loss alone is not sufficient to drive PanIN formation. These results indicate that cyclin C is required for normal islet formation, insulin production, and animal lifespan. In addition, we observed acinar cell injury consistent with loss of ALP function.Figure 3. Cyclin C suppresses ADM/PanIN production in the pancreas(A and B) H&E staining of pancreatic sections from 8-week-old CC animals. Asterisks indicate normal acinar tissues; arrows indicate ADM regions. Bars indicate 100 μM(C and D) H&E images of extensive ADM regions in two CC animals.(E and F) Magnified images from two 8-week-old KCC pancreata indicating ADM (blue arrows) and PanIN (white arrows) regions. Bars, 100 μM(G and H) The percentage of regions with ADM and PanIN pathologies within the total pancreatic area is shown for 8- (G) or 11- to 13-week-old (H) animals with the indicated genotypes. The number of animals of each genotype are indicated. Asterisks indicate statistical differences from Student’s t test (p < 0.05).(I) Percentage of regions exhibiting ADM, PanIN1, and PanIN2 lesions in 8- and 17-week-old KCC animals.(J–L) Quantitation of ADM/PanIN areas 26–29 weeks for the indicated genotypes. IHC analysis with the indicated antibodies probing pancreatic tissues expressing Kras^G12D/+^ in either (K) 11-week-old Ccnc^+/+^, (L) 8-week-old Ccnc^+/-^, or (M) 8-week-old Ccnc^−/−^ animals. Blue arrowheads indicate regions of PanIN hyperplasia. (A–D) Bars, 50 μM, (E) and (F) indicate 100 μM. (K–M) Bars, 100 μM.
Oncogenic KrasG12D expression stimulates PanIN production in Ccnc-ablated animals
Constitutive activation of the KRAS oncogene is observed in ∼90% of PDAC and is a common feature of mouse models.28 A current model posits that the presence of oncogenic Kras drives dedifferentiated ADM cells toward PanIN development.2^,^3 To determine if introduction of oncogenic Kras (Kras^G12D^) altered the phenotype observed in CC animals, the Lox-stop-Lox (LSL)-Kras^G12D^ allele was introduced through mating. Cre expression removes the transcriptional “stop” DNA sequence allowing expression of Kras^G12D^.26 The alleles were confirmed by PCR analysis of genomic DNA.29 In contrast to CC animals, LSL-Kras^G12D/+^;Pdx1-cre (KC) animals did not exhibit a shortened lifespan (Figure 2A). Consistent with this observation, KC animals displayed normal weight, serum glucose levels (Figure 2B), and islet formation (Figure 2E). Interestingly, the LSL-Kras^G12D^;Ccnc^−/−^;Pdx1-cre (KCC) double-mutant animals displayed a reduced lifespan although not as severe as the CC single-mutant animals (Figure 2A). However, serum glucose levels and body weight in the double-mutant KCC animals were not significantly different than the wild-type control (Figure 2B). Surprisingly, IHC analysis revealed that the KCC animal displayed defective islet formation (Figure 2F) as observed in the CC mutant mice. Formally, these results indicated that Kras^G12D^ expression suppressed the elevated serum glucose phenotype observed in CC animals but only partially suppressed the shortened life span (see discussion).
Next, we determined whether introduction of the activated Kras^G12D^ allele, in combination with Ccnc ablation, simulated or suppressed ADM and PanIN formation by determining the percent of a pancreatic specimen containing these lesions (see Figures S1A and S1B for example). Examination of pancreata from 8-week-old CC single-mutant or Pdx1-cre;LSL-Kras^G12D+^;Ccnc^f/+^ heterozygous animals revealed a modest appearance of ADM and more advanced PanIN lesions (quantified in Figure 3G). The accumulation of ADM or PanIN lesions was also similar for these two genotypes at 12–14 weeks (Figure 3F, quantified in Figure 3H). However, a significant difference was observed in the area containing ADM and PanIN lesions in 26- to 29-week-old KCC animals compared to the CC control (Figure 3I). These results are consistent with previous studies reporting that CC animals exhibited widespread and higher grade ADM and PanIN lesions at 8–12 months of age.30
Repeating these experiments with the KCC double-mutant mice revealed an 8-fold increase in both ADM and PanIN lesions (Figure 3E, right, quantified in Figure 3G). By 12–14 weeks, nearly all of the pancreata underwent ADM and PanIN conversion (Figure 3F, quantified in Figure 3H). The analysis of individual cell types in these lesions indicated that ADM was most prevalent, while PanIN1 and PanIN2 regions were more rare (Figure 3I). These findings indicate that Ccnc suppresses the formation of precancerous lesions in the Kras^G12D^ pancreatic cancer model. However, adenocarcinoma formation and/or metastasis were not observed in these animals. This may be due to the relatively short lifespan of these animals or the proposed role for autophagy in late-stage pancreatic tumor development (see discussion).
To assess whether the suppression of pre-neoplastic lesions also required the MKM component Cdk8, we generated LSL-Kras^G12D^;Cdk8^f/f^;Pdx1-cre animals by mating. Surprisingly, these animals exhibited no accelerated ADM or PanIN lesion formation compared to Kras^G12D^ expression alone (Figures S1C and S1D, quantified in Figure 3H). We verified the efficient cre-directed recombination event to generate the Cdk8 null allele in the pancreas using tissue-derived genomic PCR analysis (Figure S1E). Taken together, these results indicate that Ccnc ablation suppresses precancerous lesion formation in the presence of an activated Kras allele, but the tumor suppressor function of Ccnc may be independent of Cdk8 (see discussion).
KCC pancreata exhibit aggressive cancer markers
To confirm that the PanIN lesions described earlier were of ductal epithelial origin, tissue sections were probed with epithelial lineage antibody β-catenin. KC tissues exhibited modest staining (Figure 3J), while the staining in Kras^G12D^;Ccnc^+/-^ heterozygous animals exhibited more robust staining and increased ADM formation (blue arrowheads, Figure 3K). KCC pancreata exhibited strong β-catenin staining, confirming the presence of ADM and PanIN lesions (Figure 3M). Next, we investigated the aggressiveness of the lesions by staining for the proliferation marker Ki-67. Again, the KCC pancreatic tissue exhibited higher proliferation rates than KC alone or the Ccnc^+/-^ heterozygous sample (red arrowheads, middle, Figures 3K–3M). Finally, acinar injury leading to pancreatitis is a major contributor of ADM and PanIN formation.31 Previous studies reported that upregulation of the H3 lysine 27 methyltransferase EZH2 is important for the dedifferentiation program necessary for restoring normal acinar function following cerulein-induced injury.32 To determine if KCC pancreata exhibited this damage response, EZH2 levels were followed in PanIN regions. These studies found elevated EZH2 expression in Kras^G12D/+^;Ccnc^+/-^ heterozygote and KCC homozygous samples (Figures 3L and 3M). These results suggest that preventing cell damage is important for Ccnc-dependent ADM/PanIN suppression. The tumor suppressor Tp53 is upregulated in damaged cells and can indicate the activation of the iRCD pathway.33 To examine this possibility further, we examined ADM/PanIN regions for the presence of elevated Tp53 and activated caspase-3 as a marker for iRCD. Interestingly, Tp53 staining was observed in ductal cells (fuchsia arrowheads) undergoing proliferation (as evidenced by Ki67), but no indication of caspase-3 cleavage was observed (Figure S2). This contrasts with studies that found autophagy-deficient pancreata expressing Kras^G12D^-induced iRCD.27 Taken together, these studies indicate that even partial loss of Ccnc activity enhances ADM and PanIN generation in the presence of an activated Kras. However, Tp53 upregulation does not cause growth cessation or induce cell death. These results indicate that although the damage pathways have been initiated, either the damage is not sufficient to induce cell death or that cyclin C plays a role in triggering this response (see discussion).
Ccnc is required for normal autophagy
Autophagic flux is difficult to quantify in vivo. Therefore, to further investigate the role cyclin C plays in supporting autophagy in the pancreas, a cell line (193) was developed from pancreatic tissue derived from a 14-week-old KCC animal. The cell line was validated for the presence of the Ccnc deletion allele (Figures S3A and S3B) and elimination of protein expression (Figure S3C). As a comparison, we chose another murine pancreatic cell line (470) derived from a 15.5-month-old KC mouse that displayed extensive PanINs although PDAC lesions were not noted. Previous studies revealed a dependence of cyclin C stability on the presence of Cdk8 and Cdk19.34^,^35 However, loss of cyclin C did not significantly affect Cdk8 (Figure S3D) or Cdk19 (Figure S3E) levels in these pancreatic cells. Finally, given the frequency at which p53 is mutated in human cancers, we determined the levels of Tp53 in these cell lines as well as a downstream effector p21. These studies revealed a modest increase in p53 and p21 levels in KCC versus KC cell lines (Figures S2F and S2G). Taken together, these results indicate that cell lines 193 and 470 represent reasonable models for further study. A more comprehensive analysis of the transcriptome in these cell lines would be required to fully understand the impact that gain of function (Kras^G12D^) and loss of function (Ccnc^−/−^) have on transcription.
To define the role of cyclin C in the autophagic response in pancreatic cells, the appearance of autophagic protein LC3B puncta, a marker of active autophagy, was determined in 193 and 470 cell lines by IHC. The 193 and 470 cell lines were treated with the drug chloroquine (CQ), a potent stimulator of the ALP.36 CQ treatment prevents autophagosome fusion to the lysosome, thus inhibiting final degradation of the autophagic cargo. This failure is sensed by the cell resulting in induction of early autophagic genes. Following CQ treatment, LC3B puncta were observed in the 470 cell line but not in 193 Ccnc^−/−^ cells (Figure 4A). To directly monitor ALP activity, we next used western blot analysis to distinguish between the unmodified (LC3BI) and lipidated (induced) LC3B species (LC3BII). Under uninduced conditions, LC3BI and LC3BII levels were similar in both cell lines (Figure 4B, left two lanes). LC3BII levels normally increase following treatment by CQ and/or the mTOR inhibitor Torin 1.37 Elevated LC3BII levels were observed in the 470 Ccnc^+/+^ cell line following CQ treatment (Figure 4B). However, no increase in L3CBII was observed in the 193 Ccnc^−/−^ cells treated with either drug or in combination. These results indicate that the 193 cell line is defective for autophagy induction. To confirm that the reduced autophagic response was due to reduced Cdk8 and/or Cdk19 activity, 470 Ccnc^+/+^ cells were treated with the Cdk8/Cdk19 inhibitor Senexin A.38 Treating 470 cells with Senexin A reduced LC3BII accumulation following CQ and Torin 1 treatment similar to the Ccnc^−/−^ 193 cell line (Figure 4C).Figure 4. Cyclin C is required for autophagic induction in pancreatic cell line(A) IHC analysis for LC3B puncta in the Kras^G12D/+^;Ccnc^+/+^ (470) and Kras^G12D/+^;Ccnc^−/−^ (193) cell lines following chloroquine (CQ, 50 μM, 24 h) treatment. Zoomed panels are on the right.(B) Western blot analysis of the unmodified (LC3BI) and lipodated (LC3BII) species in whole-cell extracts prepared from 470 or 193 cell lines with the indicated treatments of autophagy-inducing CQ (50 μM, 24 h) and/or Torin 1 (250 nM, 24 h). β-actin served as a loading control. Molecular weight markers are indicated on the left (kDa).(C) Western blot analysis probing for LC3BI and LC3BII in extracts prepared from CQ/Torin 1-treated 470 and 193 cell lines exposed to the Cdk8/Cdk19 inhibitor Senexin A as indicated.(D) Viability studies for 470 and 193 cell lines treated with chloroquine (CQ, 50 μM, 24 h) or Torin 1 (250 nM, 24 h). Non-viable, PI-permeable cells were counted by fluorescent cell analysis. The percent of the population that is PI positive (mean ± SD) is presented. N = 3 independent cultures. ∗∗p = 0.01.(E) Western blot probing for cleaved PARP in extracts prepared from line 470 cell line treated with autophagic inhibitors indicated. β-actin served as a loading control. Molecular weight markers are on the left (kDa).(F–I) LC3B and p62 levels were examined by IHC in KC and KCC animals as indicated. Cell types are indicated. Box indicates region of insert zoom.
Next, we determined whether the 193 or 470 cell lines were sensitive to an autophagy inhibitor. Previous studies revealed that elevated autophagy observed in PDACs makes these cells sensitive to inhibitors.36^,^39 Both cell lines were treated with either Torin 1 or CQ, and their viability was measured using fluorescence cell analysis of propidium iodide (PI) uptake by dead cells. These experiments revealed no change in viability in either Torin 1-treated cell line (Figure 4D). However, the Ccnc^+/+^ 470 cells exhibited a significant increase in cell death following CQ treatment while the 193 cells did not. The resistance of 193 cells to CQ is consistent with our finding that these cells grow independent of ALP function. To further assess the nature of the cell death, PARP cleavage was measured in 470 cells treated with CQ or CQ + Torin 1. These experiments revealed enhanced PARP cleavage indicating that the cells were dying via iRCD mechanisms. Taken together, these results confirm the role of cyclin C in autophagy gene expression and execution of the process. As autophagy is required for maintaining normal acinar homeostasis, these findings indicate that pancreatic damage in Ccnc-ablated animals is caused, at least in part, by reduced autophagy (see discussion). Finally, we examined autophagic flux in vivo by monitoring both LC3B and p62 levels in KC and KCC pancreatic tissues. In wild-type animals, consistent LC3B expression was observed in acinar cells as is expected given the high protein synthesis rate observed for this tissue (Figure S4A). In tissues obtained from KC and KCC animals, we focused on the PanIN lesions as the 193 and 470 cell lines are from ductal origin. When comparing KC with KCC animals, both LC3B and p62 had similar levels of punctate staining in acinar cells (Figures 4F and 4H). However, punctate LC3B staining was reduced in PanIN lesions in the KCC-derived tissue (Figure 4G). In contrast, the levels of p62 were variable in PanIN lesions between the KC and KCC animals. The LC3B results indicate that autophagy is reduced in PanIN from KCC animals, whereas acinar tissues retain stable expression of punctate staining.
The 193 cell line exhibits elevated endogenous ROS and colony-forming ability
Several studies found that autophagy reduces reactive oxygen species (ROS) production and increases genomic stability in pancreatic tissues.25^,^36 Depending on the concentration, ROS can serve as a signaling molecule able to enhance cell growth40 or induce damage and even cell death.41^,^42 Superoxide increases dihydroethidium (DHE) fluorescence providing an approximation of cellular ROS levels. We examined DHE fluorescence in immortal MEFs, Kras^G12D/+^ (470), and Kras^G12D/+^;Ccnc^−/−^ (193) cell lines. As expected, the immortal MEF line exhibited reduced DHE oxidation compared to the transformed Kras^G12D/+^ 470 cells (Figure 5A). Interestingly, the Kras^G12D/+^;Ccnc^−/−^ 193 cells exhibited even higher DHE oxidation than the 470 line. We next examined mitochondrial ROS levels in these cell lines using the MitoSox oxidant-sensitive dye. Consistent with the DHE results, we found elevated ROS in the mitochondrial compartment in 193 cells compared to the 470 cell line (Figure 5B). Taken together, these results indicate that the Kras^G12D+^;Ccnc^−/−^ 193 cells are carrying a higher ROS burden than Kras^G12D+^ cells alone.Figure 5. Transformed phenotypes of Kras^G12D/+^;Ccnc^−/−^ cell lines(A) Endogenous ROS levels were determined by DHE staining followed by fluorescence cell analysis. Wild-type (WT) immortalized MEF cells are used as a control for non-transformed cell type. Kras^G12D/+^;Ccnc^+/+^ (470) and Kras^G12D/+^;Ccnc^−/−^ (193) traces are indicated. Biological replicates for the 193 cell line are shown.(B) Two biological replicates of the 470 and 193 cell lines were stained with MitoSox and analyzed as described in (A).(C–E) (C) The MPT-2B (Kras^G12D/+^;Trp53^+/−^;Ccnc^+/+^), (D) 470 (Kras^G12D/+^;Ccnc^+/+^), and (E) 193 (Kras^G12D/+^;Ccnc^−/−^) cells were plated in soft agar, and colony-forming ability was imaged. Representative images following three weeks in culture are shown at the indicated magnification.(F) Colony-forming efficiency of MPT-2B was set at 100% for comparing 470 and 193 cell lines. N = 3 independent trial, graph indicates mean +/- SD. Student's T-test was applied, Asterisks indicate p = 0.05.
We next addressed whether the 193 and 470 cell lines were able to form colonies in soft agar, a common trait of transformed cells. As a positive control, we tested the colony-forming ability of the MPT-2B (Kras^G12D/+^;Tp53^+/−^;Ccnc^+/+^), a mouse cell line that demonstrated tumor-forming ability following orthotopic injections. As expected from the aggressive nature of this tumor, the MPT-2B readily produced large colonies in soft agar, which was used as a standard for colony formation (100%). The 470 cell line exhibited strong colony formation (Figure 5C, quantified in Figure 5E) but not to the extent of the MPT-2B cells. The 193 cell line produced relatively small colonies (Figure 5D) at lower frequency. Furthermore, no evidence of pancreatic tumor development was observed 8 months after mice were injected orthotopically with the 193 cells (data not shown). Therefore, despite rapid development of aggressive PanIN lesions in KCC mice, a cell line derived from these mice exhibited a weak transformed phenotype. This may be due to the relatively early time (15 weeks) of tissue harvesting precluding the accumulation of additional mutations. Alternatively, the reduced autophagy in this cell line may retard more advance cancer stages as suggested previously36 (see discussion).
Ccnc is required for normal 26S proteasome function
The requirement of cyclin C for the ALP raised the question about how this impacted ubiquitin-proteasome system (UPS) activity. Reduced UPS activity has been shown to upregulate the ALP to help maintain proteostasis (reviewed in Dikic43). However, studies have both supported23 or refuted44 that autophagy inhibition stimulates UPS activity. To determine if cyclin C loss alters UPS activity in our pancreatic cell lines, extracts prepared from 193 to 470 cells were incubated with a model substrate (LLVY-AMC) that fluoresces when cleaved by the proteasome. These studies revealed that the relative velocity of the proteasome activity was 113 RFU/min for 470 cells (Figure 6A). Interestingly, Ccnc^−/−^ 193 cell extracts exhibited significantly reduced proteasome activity (62 RFU/min). Next, we determined whether Ccnc^−/−^ cells were still sensitive to MG-132 inhibition. In these experiments, we treated cells with the same concentration of MG-132 used to induce autophagy. MG-132 treatment reduced UPS activity in both the Ccnc^+/+^ (64 RFU/min) and Ccnc^−/−^ (27 RFU/min) cell lines (Figure 6B). These results indicated that cyclin C is also required for normal proteasome function as well as ALP.Figure 6. Ccnc maintains normal UPS function(A) LLVY-AMC cleavage assays were performed with extracts prepared from 193 (Ccnc^−/−^) and 470 (Ccnc^+/+^) cell lines.(B–D) The indicated cell lines were treated with the proteasome inhibitor MG132 (B), CQ (C), and Senexin A (D). The control curves for (C) and (D) are transferred from the experiment in (A).(E) Western blot analysis of the indicated protein components of the 20S and 19S complexes. Molecular weight markers (kDa) are indicated on the left.In all panels, N = 3 with two technical replicates. The data reflect mean ± SD. Student’s t test, ∗∗p < 0.01; ∗∗∗∗p < 0.001.
One possible explanation of our results is that reduced autophagy gene expression has a negative impact on proteasome function in this background. If correct, two predictions should be true. First, inhibiting autophagy in Ccnc^+/+^ cells should phenocopy the Ccnc^−/−^ results. Therefore, proteasome activity was measured in 470 Ccnc^+/+^ cells treated with CQ. These studies revealed a similar reduction in proteasome activity as observed in Ccnc^−/−^ cells (Figure 6C). Second, inhibiting Cdk8/Cdk19 activity with the drug Senexin A should also reduce proteasome function in Ccnc^+/+^ cells. Senexin A treatment caused a significant reduction in proteasome activity although not to the extent observed for the Ccnc^−/−^ cell line (Figure 6D). These results are consistent with the model that Ccnc is required for autophagy gene transcription, which reduces both autophagy and proteasome activity. Alternatively, Ccnc may directly support UPS activity by mediating transcription of proteasome component genes. To address this question, we monitored the levels of several components of the 20S core particle and the 19S regulatory lid in the 193 and 470 cell lines. These experiments did not reveal any significant difference in levels for the selected proteins from either complex (Figure 6E), although this does not rule out other proteasome components being downregulated. Taken together, these results reveal a new role for MKM transcriptional control in supporting both ALP and UPS activity although the latter role may be indirect.
Reducing MKM activity sensitizes tumor cell lines to proteasome inhibitors
Proteotoxic stress stimulates both the UPS and ALP systems.45^,^46 Therefore, we tested the sensitivity of the Ccnc^−/−^ cell line to the UPS inhibitors bortezomib (BZ) or MG-132 utilizing Annexin V staining to monitor iRCD. BZ or MG-132 treatment of the 470 (Ccnc^+/+^) cell lines resulted in a 3- and 5-fold increase in cell death versus untreated control cells (Figure 7A). Repeating these experiments with the 193 cell line revealed an additional 3-fold increased sensitivity to BZ or MG-132 treatment compared to 470 cells. These results indicate that loss of Ccnc activity sensitizes cells to proteasome inhibitors. To determine if this hypersensitivity was expressed in other tumor types, CCNC was deleted via CRISPR protocols in HeLa (cervical) and HCT116 (colon) human cancer cell lines (Figure S4). CCNC^−/−-^ HeLa cells exhibited a 3-fold increase in cell death following BZ treatment compared to the CCNC^+/+^ parental cells (Figure 7B). A similar, but less dramatic, effect was observed in BZ-treated HCT116 CCNC^−/−^ cells (Figure 7C). Taken together, these studies indicate that inactivating CCNC results in a general hypersensitivity to proteasome inhibitors.Figure 7. Reduced MKM function enhances sensitivity to proteasome inhibitors(A) 470 (Kras^G12D/+^;Ccnc^+/+^) and 193 (Kras^G12D/+^;Ccnc^−/−^) cells were treated with the proteasome inhibitors bortezomib (BZ, 100 nM) or MG-132 (250 nM), and the percent of the population that was Annexin V positive and PI negative was determined. N=3 biological replicates in all studies.(B) Ccnc^+/+^ or Ccnc^−/−^ HeLa cells were treated with BZ and analyzed as before.(C) Ccnc^+/+^ and Ccnc^−/−^ HCT-116 cells were treated with the CDK8/19 inhibitor AS-2863619 (AS, 1 μM) and/or BZ as indicated. HCT-116^Ccnc−/−^ cells were treated with only BZ.(D) MPT-2 cells were treated with the CDK8/19 inhibitor Sel-120 (Sel, 1 μM) and/or BZ.(E) HepG2 cells were treated as described in (C).(F) MPT-2B or non-transformed pancreatic epithelial cells were treated as indicated. In all panels, data are represented as mean ± SD. Student’s t test with p values is calculated as ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.005; ∗∗∗∗p < 0.001.(G) Model for Ccnc function in pancreatic tissue differentiation and disease. Autophagy supports β cell differentiation, islet formation, and insulin production, thus protecting the animal from hyperglycemia and shortened life. In addition, autophagy protects acinar cells from proteotoxic stress by removing misfolded proteins and aggregates. ALP dysfunction induces ADM where it can redifferentiate back to the acinar lineage. In the presence of activated Kras^G12D^, proliferating ADM cells are pushed toward PanIN formation. This step has two outcomes with Kras^G12D^ supporting advanced PanIN or PDAC development. Ccnc-dependent iRCD can be activated, blunting disease progression.
We next tested whether CDK8/CDK19 inhibitors phenocopied the CCNC^−/−^ allele with respect to BZ hypersensitivity. First, HCT-116 CCNC^+/+^ cells were treated for 24 h with the CDK8/CDK19 inhibitor AS-2863619 (AS)47 prior to BZ addition. As a control, AS treatment alone had no significant impact on HCT-116 viability (Figure 7C). However, BZ addition increased AS-treated HCT-116 cell death to levels observed for the CCNC^−/−^ derivative (Figure 7C). Finally, these studies were repeated with two additional cell lines, MPT-2B and the liver cancer cell line HepG2. For the MPT-2B cells, a different CDK8/CDK19 inhibitor (Sel-12048) was used. In both experiments, inhibiting Cdk8/Cdk19 activity increased BZ sensitivity (Figures 7D and 7E). Importantly, treating SV-40 immortalized normal pancreatic epithelial cells with this regimen did not significantly increase toxicity over proteasome inhibitor alone (Figure 7F). These findings point to a general enhanced sensitivity of transformed to non-transformed cells. The analysis of additional cell types revealed that a thyroid cell line (D44549) or normal human lung fibroblasts (IMR90) cells were modestly sensitized to BZ following treatment with the Cdk8/Cdk19 inhibitor (Figures S5A and S5B). Finally, we found that the 193 cell line was more sensitive than 470 cells to BZ over multiple drug concentrations (Figure S5C). Importantly, this study also revealed several-fold elevated sensitivity to BZ by the 193 cells compared to proliferating normal pancreatic epithelial cells. This difference was more striking when quiescent pancreatic epithelial cells were compared to the 193 cell line. Taken together, these results indicate that inhibiting CDK9/CDK19 represents a reliable strategy to specifically induce tumor cell hypersensitivity to proteasome inhibitors.
Discussion
The present study revealed two roles for cyclin C in preventing pancreatic pathogenesis. First, cyclin C is required for both basal and induced autophagic gene transcription (Figure 7G). Autophagy protects normal β cell islet development and insulin production8^,^50 and prevents acinar cell damage associated with initiating precancerous lesions.11^,^12^,^51 A previous study found that p53 induced the transcription of the macroautophagy gene DRAM1.52 Similarly, we found that Ccnc is required for both steady-state and induced Dram1 transcription. As Ccnc-Cdk8 has previously been identified as a co-activator with p53 for p21 transcription,53 this may indicate that a similar regulatory mechanism is present at DRAM1 and perhaps other autophagy gene promoters. Second, cyclin C promotes regulated cell death in autophagy-defective, Kras-driven transformed pancreatic cells. Therefore, cyclin C prevents the damage required for pancreatic cancer initiation on the front end and helps kill any damaged cells that do escape to suppress tumorigenesis. Finally, we find that Ccnc supports both ALP and UPS activity, rendering Ccnc^−/−^ cells hypersensitive to proteasome inhibitors, providing a new avenue to attack pancreatic cancers.
The roles that MKM components play in murine development are complicated. For example, deleting Cdk8 results in embryo arrest at the preimplantation stage.54 Conversely, Cdk19^−/−^ mice exhibit a normal lifespan.55 In the middle of these extremes, Ccnc^−/−^ mice arrest at day 9.5 days post-coitum at the onset of organogenesis (Li et al.19 and our unpublished results). However, pancreatic Kras^G12D^;Cdk8^−/−^ animals did not exhibit ADM or PanIN lesions above Kras^G12D^ expression alone, suggesting that CDK8 activity is replaced by CDK19.14^,^35 However, as Cdk19^−/−^ animals develop normally, its requirement alone seems unlikely. Therefore, deleting both kinases would most likely be required to phenocopy the transcriptional defect in Ccnc^−/−^ cells.
The phenotypes associated with loss of cyclin C function were more severe than deleting Atg5 or Atg7. For example, Atg5 or Atg7 pancreatic ablation led to shortened lifespan (median ∼16 weeks) although 40% of the animals exhibited a normal lifespan.27 Animals harboring Ccnc pancreatic deletion did not survive past 15 weeks. In addition, KCC animals exhibited an accelerated rate of ADM and PanIN lesions compared to pancreata harboring Kras^G12D^;Atg5^−/−^ alleles. These results suggested that cyclin C plays a role(s) in pancreatic biology in addition to maintaining autophagy. This second role may be related to the mitochondrial function of cyclin C contributing to restricting neoplastic growth (Figure 7G). Activated Kras increases mitochondrial-derived ROS through elevated metabolism associated with transformed growth.56 In many cancers, ROS are maintained at levels that enhance mutagenesis and proliferation but below those that induce iRCD.42 Our previous studies revealed that Ccnc is required for ROS-induced iRCD.18 Therefore, cyclin C loss could allow continued Kras^G12D^-driven cell division even in the presence of normally iRCD-triggering ROS levels. Consistent with this possibility, we find that the Kras^G12D^;Ccnc^−/−^ cell line exhibits higher endogenous and mitochondrial ROS concentrations than a Kras^G12D^ tumor cell line. In addition, evidence of iRCD observed in Kras^G12D^;Atg5^−/−^ pancreata is lacking in KCC organs. Taken together, these results indicate a second role for cyclin C in suppressing tumorigenesis by inducing iRCD in response to elevated ROS.
Previous studies have reported that the MKM has varied roles in development and disease progression. For example, Cdk8 overexpression is associated with colorectal cancer progression57 and angiogenesis in pancreatic cancers.58 Alternatively, other studies revealed a tumor suppressor function for Ccnc in acute lymphoblastic leukemia (ALL)19 and in the Pten^thyr−/−^ mouse thyroid tumor model.59 Given the strong phenotype observed in the KCC animals, it is surprising that mutations in CCNC, CDK8, or CDK19 have not been directly associated with PDAC.60 However, the CCNC (6q16.2) and CDK19 (6q21) loci reside on the long arm of Chr. 6 that is often deleted in cancers,61^,^62^,^63^,^64^,^65 including loss of heterozygosity (LOH) in 69% (38/55) of PDAC.61 As these cytogenetic losses are large,60^,^66^,^67 many loci are contained in these deletions. Our findings argue that CCNC at least contributes to the tumor suppressor activity in the 6q region while acknowledging that additional players are likely present. It is not clear why 6q is susceptible to loss although a fragile site (FRA6F) has been mapped to region 6q16-6q21.68 Additional reports of breakpoints have been seen around FRA6F in other cancers as well (Morelli et al.69 and references therein). Therefore, chromosomal instability may be inherent to this location and contribute to PDAC evolution through loss of CCNC.
A connection between hyperglycemia and cancer has been proposed, as tumor growth is stimulated by excessive serum glucose.70^,^71 Consistent with this possibility, mice harboring Kras^G12D/+^;Atg7^−/−^;Tp53^−/−^ pancreatic tumors exhibited increased glucose uptake, resulting in elevated glycolysis to support anabolic pathways.27 Interestingly, CC animals exhibit elevated serum glucose but did not produce more advanced PanIN lesions. However, KCC mice, which produce PanINs at an elevated rate, do not display elevated serum glucose despite exhibiting poorly developed islets. These findings suggest that the presence of extensive PanIN lesions in KCC mice metabolizes glucose at an elevated rate, thus reducing serum glucose levels. This would suggest that the metabolic reprogramming associated with PDAC may be manifested earlier in precancerous lesions lacking Ccnc.
PDACs are usually discovered following metastasis making surgical options difficult, thus elevating the reliance on chemotherapeutic approaches. However, the success rates of current regimens are disappointing. Therefore, new approaches are necessary to improve patient outcome. Due to the unregulated growth associated with transformation, elevated concentrations of damaged or misfolded proteins generating proteotoxic stress stimulate both UPS and ALP activity.45^,^46 Increased proteostasis is important for cancer cell survival as KRAS-transformed cells become addicted to this activity with cells becoming hypersensitive to proteasome inhibition.72 Although proteasome inhibitors have proven useful in some hematopoietic malignancies, they have been less successful in solid tumors including pancreatic cancers (reviewed in Murugan and Voutsadakis73). One possible explanation for this result is that solid tumors require an additional sensitization step. Our finding that the activity of both the ALP and UPS is reduced following Ccnc ablation or Cdk8/Cdk19 inhibition may sensitize tumors to proteasome inhibitors. Therefore, our results suggest that even if CCNC deletion is not directly involved in tumorigenesis, the pharmaceutical inhibition of CDK8 and CDK19 may increase proteasome inhibitor efficacy.
Limitations of the study
The rapid reduction in viability of the Kras^G12D^;Ccnc^−/−^ mice prohibited analysis of frank tumor formation and metastasis. The use of mosaic or heterozygous Ccnc mutant alleles may be required for this type of study. A comprehensive analysis of the transcriptome in the cell lines described in this study would be useful for understanding the full context of the transcriptional consequences of deleting Ccnc and activating Kras.
Resource availability
Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Randy Strich ([email protected]).
Materials availability
Animals and cell lines described in this study will be made available upon request and proper consent.
Data and code availability
Data generated in this study will be provided upon request. The Ccnc^−/−^ transcriptome dataset has been deposited previously (GEO accession is [GSE126450](GSE126450)). Methods to interpret the data are detailed in STAR Methods. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.
Acknowledgments
The authors appreciate expert technical support from the FCCC Histopathology, Cell Culture, Laboratory Animal, Bioinformatics & Biostatistics, and Mouse Genotyping Facilities, specifically the work of Catherine (Cass) Renner, Jirong (Jenny) Zhang, Fang Jin, Suraj Peri, Irina Shchaveleva, Justin Rambert, Tyanah Yisrael, Lauren Heisey, Mark Tague, Simon Tarpinian, and Rita Michielli. We also thank Dr. Jonathan Chernoff for the SV40-immortalized mouse pancreatic epithelial cell line and Dr. Edna Cukierman for the normal pancreatic cell line. We thank Drs. Edna Cuckierman and Igor Astsaturov and members of the Marvin and Concetta Greenberg Pancreatic Cancer Institute at FCCC for advice. This work was supported by grants from the 10.13039/100000002National Institutes of Health awarded to R.S. (GM113052) and the FCCC Comprehensive Cancer Center Support Grant (CA06927) in support of the Histopathology, Cell Culture, Laboratory Animal, and the Mouse Genotyping Facilities at FCCC. Additional support was provided by the Boye Foundation and the 10.13039/100001774New Jersey Health Foundation (to R.S.) and from the Martin and Concetta Greenberg Pancreatic Cancer Institute at FCCC and Pennsylvania DOH 10.13039/100005622Health Research Formula Funds (to K.S.C.).
Author contributions
S.E.H., investigation and formal analysis; K.Q.C., investigation and formal analysis; S.D.W., investigation and formal analysis; D.C.S., investigation and formal analysis; A.J.K.-S., formal analysis, writing – review and editing; K.S.C., writing – review and editing, supervision, funding acquisition, and resources; R.S., writing – original draft, review and editing, supervision, funding acquisition, and resources.
Declaration of interests
The authors declare no competing interests.
STAR★Methods
Key resources table
REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesCaspase 3 cleaved (IHC, 1:50)Cell Signaling TechASP175, 9661AB_2341188Trp53 (IHC, 1:200)Vector Labs.CM5, VP-P956AB_2335917LC3 lipodated (IHC, 1:100)Nanotools5F10, 0231-100AB_2722733β-catenin (IHC, 1:200)Santa Cruz BiotechnologySC-59737AB_781850β-actin (Western, 1:2000)MerckA1978AB_476692Ki-67 (IHC, 1:300)AbcamAb16667AB_302459Insulin (IHC, 1:300)BD Bioscience565688AB_2739330EZH2 (IHC, 1:500)Cell Signaling5246AB_10694683LC3B (Western, 1:1000)Cell SignalingCat. #3868AB_2137707Tp53 (Western, 1:1000)Santa CruzCat. #SC-126AB_628082Ccnc (Western, 1:1000)Cell SignalingCat. #68179NACdk8 (Western, 1:1000)Santa CruzCat. #SC-1521AB_2260300Cdk19 (Western, 30:1,000)Dev. Studies Hybridoma BandAB_2133570AB_2133570Alexa Fluor 488-conjugated secondary antibody (Western, 1:5000)Thermo ScientificA11008AB_143165alkaline phosphatase-conjugated rabbit secondary antibody (Western, 1:5000)AbcamAb97061AB_10680575IHC Secondary AbRoche OmniMap DAP Rabbit detection kit760–149Chemicals, peptides, and recombinant proteinsMG-132CalbiochemCAT#474790CAS:133407-82-6Torin 1Santa Cruz BiotechCAT#133407-82-6 CAS:1222998-36-8ChloroquineCayman Chemical CoCAT#30708CAS:50-63-5CDP StarThermoFisherCAT#T2216CAS:160081-62-9Critical commercial assaysGlucose Colorimetric Assay KitCell Biolabs, Inc.#STA-680Annexin V Cell Death AssayBD Biosciences556419Bradford ProteinBio-Rad5000001Suc-LLVY-AMCSanta Cruzsc-495769Deposited dataRNAseq Ccnc−/− dataGEO: [GSE126450](GSE126450)Ref.21Experimental models: Cell linesMPT-2B derived from a 16 week old female [LSL-Kras^G12D^;Pdx1-cre;Trp53^+/−^] mouseFCCCThis study. Mycoplasma freeValidated using genomic PCR; IHC with Troma-1 to ensure ductal origin193 derived from a 15 week old female mouse [LSL-Kras^G12D^;Pdx1-cre;Ccnc^−/−^]FCCCThis study. Mycoplasma freeValidated using genomic PCR; IHC with Troma-1 to ensure ductal origin470 derived from a 77 week old male mouse [LSL-Kras^G12D^;Pdx1-cre]. No PDAC lesions were observed.FCCCThis study. Mycoplasma freeValidated using genomic PCR; IHC with Troma-1 to ensure ductal origin697 derived from a 49 week old male mouse [LSL-Kras^G12D^;Pdx1-cre]. No PDAC lesions were observed.FCCCThis study.Mycoplasma freeSV40-transfected C57Bl/6 pancreatic epithelial cellsA gift from Jon Chernoff, FCCCN/AMouse embryonic fibroblast (MEF)RowanSOMRef.16Mycoplasma freeHuman Hela-ovarian cancerATCCN/AMycoplasma freeHuman HCT-116-colon cancerATCCN/AMycoplasma freeHuman HepG2-liver cancerATCCN/AMycoplasma freeHuman IMR90 normal fibroblastATCCN/AMycoplasma freeHuman poorly differentiated thyroid cancer D445Kras^G12D^;Trp53^−/−^Ref.49Experimental models: Organisms/strainsC57BL/6NTac Cdk8^tm1a(EUCOMM)Hmgu^Institut Clinique de la Souris: ICShttp://www.ics-mci.frNAB6.129S4-Gt(ROSA)26Sortm2)FLP∗)Sor/JJackson LabsCat. #012930IMSR_JAX:012930129/SvJae/C57Bl/6 LSL-Kras^G12D/+^;Pdx1-cre^+/−^ crossed at least 12 times to the C57Bl/6 background prior to starting these experiments.Beatson Institute for Cancer Research, GlasgowRef.26^,^29NAB6N.Cg-Krasetm4tyj/CjDswJJackson Labs, Bar Harbor Maine USA019104IMSR_JAX:019104Oligonucleotides, genotyping primersCcnc2 see Figure S3TAATCGACCAGACAGTACGGGAGTCRef.16NASdl2 see Figure S3GGTAGTTTATCTGAACTGATGAAAACACATCRef.16NALOX1 see Figure S3GGAAGCAGAAGCAACAGGAATCTGRef.16NAKRAS WT see Figure 3GTCGACAAGCTCATGCGGGTGRef.29NAKRAS G12D see Figure 3AGCTAGCCACCATGGCTTGAGTAAGTCTGCARef.29NAKras Universal see Figure 3CCTTTACAAGCGCACGCAGACTGTAGARef.29NAPdx1-cre F see Figure 3CTGCATAGTACGCKTATACCCTGTRef.29NAPdx1-cre R see Figure 3GCAGGTCGAGGGACCTAATARef.29NACdk8 F see Figure S1CGTAGGTAGCAATCTGGTCGGGGTThis StudyNACdk8 R See Figure S1CAGGTACACAGGCTGGATTTGCACThis StudyNAAtg9b F See Figure 1GGTTCCTGGCATCACATCCAThis StudyNAAtg9b R See Figure 1CAGCGAAGGAGGAAGGTTGTThis StudyNAAtg101 F See Figure 1CCTTGCGCAAGGTTGTTGGGGAThis StudyNAAtg101 R See Figure 1CACCTCCCAGGGGATGCACTCAThis StudyNAGabarap F See Figure 1GAAAAAGCCCCCAAAGCTCGThis StudyNAGabarap R See Figure 1CACTGGTGGGTGGAATGACAThis StudyNABecn1 F See Figure 1CACTGGTGGGTGGAATGACAThis StudyNABecn1 R See Figure 1TCGAGAGACACCATCCTGGCGAThis StudyNAUvrag F See Figure 1TGGGGTGGAAAGGAAGAGGCCTThis StudyNAUvrag R See Figure 1GTTGCGGGCATGGATCTGCTGTThis StudyNADram1 F See Figure 1ACTTGGTGTCCTTGGCGCTTGGThis StudyNAMap1LC3B F See Figure 1CAGCGCCGGAGCTTTGAACAAAThis StudyNAMap1LC3B R See Figure 1CTTGGTCTTGTCCAGGACGGGCThis StudyNAGapdh FGGTTGTCTCCTGCGACTTCAThis StudyNAGapdh R See Figure 1CCTCGGCCCCTCCTGTTATThis StudyNACCNC gRNA Exon 1 See Figure S4TCAGGGCAAGCTGTGTTCCATGGThis StudyNACCNC gRNA Exon 2 See Figure S4TCTCAGAGGAAGAATATTGGAAGThis StudyNASoftware and algorithmsImageJ v1.54Ref.74https://imagej.nih.gov/ij/
Method details
Animals
All animal experiments were conducted in accordance with protocols reviewed and approved by the institutional animal care and use committee (IACUC) review at Fox Chase Cancer Center (#99-22) and RowanSOM (#2020-1241). All mouse strains examined were in the C57Bl/6 background with genetically manipulated animals generated by mating transgenic strains. Males and females were included in these studies. Mice not in the C57Bl/6 background were backcrossed at least ten generations into the C57BL/6 background prior to use. The Cdk8 floxed mouse Cdk8^tm1a(EUCOMM)Hmgu^ was purchased from the Mouse Clinical Institute as part of the Constitutive Knockout/Conditional Knockout (KO-cKO pc) project at the Institut Clinique de la Souris. This mouse was first mated with a strain harboring the FLP recombinase under the control of the Rosa promoter [B6.129S4-Gt(ROSA)26Sortm2(FLP∗)Sor/J] to remove the selection marker (neomycin resistance) from the Cdk8 floxed allele. Expression of the cre recombinase deleted exon 5, resulting in a frameshift mutation. The Pdx1-Cre;LSL-Kras^G12D^ strain was originally obtained from Dr. Owen Sansom (Beatson Institute for Cancer Research, Glasgow). The floxed Ccnc (Ccnc^fl/fl^) strain was described previously.16 Littermates with the indicated genotypes were used as controls. Both sexes were included in this work and sex was found not a determining factor in these results. Animals were euthanized at predetermined timepoints in consultation with the staff veterinarian at both institutions.
Cell lines
MPT-2B and 470 cell lines were cultured at 5% CO_2_ in DMEM supplemented with 10% FBS. The 193 cell line was cultured in low calcium medium (DMEM/F-12 1:1 [Gibco, special order medium with no calcium] supplemented with 5% chelated horse serum, 20 ng/mL EGF, 100 ng/mL cholera toxin, 10 μg/mL insulin, 0.5 μg/mL hydrocortisone, 0.04 mM calcium chloride, dihydrous, 100 U/ml penicillin, 100 μg/mL streptomycin, 10 μg/mL ciprofloxacin, and 0.25 μg/mL amphotericin B). These three cell lines were generated from pancreata of mice with the indicated genotypes in the FCCC Cell Culture Facility. These cell lines all tested negative for mycoplasm at various stages in the investigation.
Mouse genotyping
The floxed Ccnc (Ccnc^fl^),16 Kras^G12D^ and the pancreas-specific Cre (Pdx1-Cre) alleles have been previously described.26^,^30 Genotyping at Rowan University was accomplished using genomic tail DNA purified using the Phire genotyping kit and a split PCR method was run for 35 cycles (Thermo Scientific, Rockford, IL, USA). Genotyping at FCCC was performed by the Mouse Genotyping Facility from genomic tail DNA using Promega GoTaq Green Master Mix or Sigma JumpStart Taq ReadyMix. The genotyping primers are listed in the key resources table.
Immunohistochemistry
H&E staining was performed as previously described.75 Ki-67 staining was conducted essentially as previously described.76 Immediately after euthanizing the mice, pancreata were collected and all connective tissue was removed. The tissues were then washed twice in formalin and fixed in formalin overnight. Then the tissues were stored in 70% ethanol for future examination. All the tissues were embedded in paraffin and sectioned at 6 μm. Sections were subjected to antigen retrieval in 0.1 mM sodium citrate and counterstained with hematoxylin. For autophagy analysis, sections were stained with either anti-LC3B (NOVUS #NB600-1384) or anti-p62 (ABclonal #A19700). All stained sections were imaged at 40-800× magnifications and analyzed using ImageJ software.74
Western blot analysis
Mouse cells were homogenized in RIPA buffer (150 mM NaCl, 50 mM Tris, 1% Nonidet P-40 substitute, 0.5% sodium deoxycholate, 0.1% SDS, pH 8) containing 1% protease inhibitor cocktail (PIC) (Sigma P8340, St. Louis, MO USA), 1% EZBlock phosphatase inhibitor cocktail IV (BioVision, Milpitas, CA USA), 10 mM NaF, 10 mM β-glycerol phosphate, and 2 mM Na_3_VO_4_. Homogenates were incubated for 2 h at 4 °C and centrifuged at 14,000 × g for 20 min at 4 °C to separate soluble proteins from aggregates and cell debris. Protein concentrations were determined by Bradford assay (Bio-Rad, Hercules, CA USA). Samples were dissolved in sample buffer (SB) (100 mM Tris, 4% SDS, 20% glycerol, 2 mg/mL bromophenol blue, pH 6.8) supplemented with 100 mM DTT, boiled for 5 min, and separated by SDS-PAGE then probed with the indicated antibodies. Western blot signals were visualized using CDP-STAR (CDP∗) substrate and signals were visualized using an iBright FL1500 Imaging System (Thermo).
RT-qPCR analysis
Total RNA was prepared from the cell samples using Monarch® Total RNA Miniprep Kit. On-column DNase treatment was performed to eliminate contaminating DNA during RNA extraction. Total RNA (500 ng) was converted to cDNA using the ThermoFisher™ Maxima cDNA Synthesis kit. The cDNA from each sample (1/100 dilution) was subjected to qPCR amplification using ThermoFisher™ PowerSYBR™ Green PCR Master Mix and a StepOne™ Real Time PCR System. These assays were conducted with three independent preparations assayed in duplicate. Gapdh was used as the internal standard for comparative (ΔΔC_T_) quantitation.77 Statistical significance was determined via Student’s t test analysis. p values < 0.05 were considered significant. Primer sequences for RT-qPCR analysis can be found in key resources table.
CRISPR CCNC knockout protocol
CCNC was depleted in the HeLa and HCT116 cell lines by dually transfecting CRISPR Cas12a plasmids (pTE4398, Addgene) containing gRNA directed to Exon 1 and Exon 2 (pSW496 and pSW497, respectively). Transfections were performed by electroporation using the Neon Electroporator (ThermoFisher) as described by the manufacturer. Briefly, trypsin/EDTA treated cells harvested at 50% confluent were washed then resuspended at 1x10^6^ cells/mL in R buffer. Cells were electroporated in 10 μL tip at 1250 V with 20 ms pulse width. The transfectants were selected for 7 days with puromycin and then clonally diluted into 96 well plates. Individual clones were then analyzed by Western blot for the presence of CCNC.
Soft agar assays
The cell lines indicated in tissue culture medium were resuspended in 1.5 mL of 0.3% noble agar in Hank’s buffer (Millapore-Sigma H6648) at 37°. The cell suspensions were immediately plated into a 24-well plate with 0.5 mL of 0.6% bottom agar medium already solidified. Once the suspensions solidified (30 min), the plates were incubated at 37° for three weeks. 100 μL of medium was added to each well as necessary. The cells were fixed with methanol and stained with 0.15% crystal violet prior to imaging. The percentage of the cells plated forming macroscopic colonies was calculated.
Plasma glucose analysis
Plasma was obtained at 2-week intervals from mice starting at 4 weeks of age by retroorbital bleed into heparinized tubes. Plasma was cryopreserved at −80°C and glucose concentrations were performed at 1:100 dilution using a Colorimetric Glucose Assay Kit from Cell Biolabs, Inc. (#STA-680; San Diego, CA) according to manufacturer’s instructions.
Proteasome activity assays
26S proteasome activity was determined in adherent cells grown to ∼75% confluency and lysed using buffer A (25 mM Tris, pH 7.4, 10 mM MgCl2, 10% glycerol) supplemented with fresh 1 mM ATP, 1 mM DTT, and 5 μL (for 500 μL of buffer) of protease inhibitor and phosphatase inhibitor. Cells were centrifuged, resuspended in 75 μL of lysis buffer A, and incubated for 30 min at 4^0^ with gentle shaking. Cell debris was removed by centrifugation (15K x g) for 4 min at 4^0^ and the resulting lysates were retained. Each assay contained 25 μg of protein as determined by Bradford protein assays. To monitor proteasomal-independent fluorescence, the proteasome inhibitor MG-132 (72.5 μM) was added to each control well and non-specific cleavage of the substrate subtracted from the experimental samples. Finally, the substrate buffer (lysis buffer A, 200 μM Suc LLVY, 1 mM DTT, and 1 mM ATP) was added to each well. Fluorescence was immediately monitored via a plate reader. Fluorescent readings (excitation = 360, emission = 400) were taken every 5 min for 90 min at 37^0^ with gentle agitation. All assays were performed with three biological repeates with two technical replicates. In these assays, Senexin A (0.4 mM) and chloroquine (100 μM) was added as indicated 24 h prior to cell harvesting.
Quantitation and statistical analysis
The Student’s t test was applied to determine statistical differences for the data presented. p values < 0.05 were considered significantly different in these assays. The Kaplan-Meier survival curves were calculated using GraphPad. Actual calculations of p values are provided in each figure as denoted in the legend.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Siegel R.L.Miller K.D.Wagle N.S.Jemal A.Cancer statistics, 2023 CA Cancer J. Clin.732023174810.3322/caac.2176336633525 · doi ↗ · pubmed ↗
- 2De La OJ.P.Emerson L.L.Goodman J.L.Froebe S.C.Illum B.E.Curtis A.B.Murtaugh L.C.Notch and Kras reprogram pancreatic acinar cells to ductal intraepithelial neoplasia Proc. Natl. Acad. Sci. USA 1052008189071891210.1073/pnas.081011110519028876 PMC 2585942 · doi ↗ · pubmed ↗
- 3Habbe N.Shi G.Meguid R.A.Fendrich V.Esni F.Chen H.Feldmann G.Stoffers D.A.Konieczny S.F.Leach S.D.Maitra A.Spontaneous induction of murine pancreatic intraepithelial neoplasia (m Pan IN) by acinar cell targeting of oncogenic Kras in adult mice Proc. Natl. Acad. Sci. USA 1052008189131891810.1073/pnas.081009710519028870 PMC 2596215 · doi ↗ · pubmed ↗
- 4Hruban R.H.Iacobuzio-Donahue C.Wilentz R.E.Goggins M.Kern S.E.Molecular pathology of pancreatic cancer Cancer J.7200125125811561601 · pubmed ↗
- 5Kleeff J.Whitcomb D.C.Shimosegawa T.Esposito I.Lerch M.M.Gress T.Mayerle J.Drewes A.M.Rebours V.Akisik F.Chronic pancreatitis Nat. Rev. Dis. Primers 320171706010.1038/nrdp.2017.6028880010 · doi ↗ · pubmed ↗
- 6Feng Y.He D.Yao Z.Klionsky D.J.The machinery of macroautophagy Cell Res.242014244110.1038/cr.2013.16824366339 PMC 3879710 · doi ↗ · pubmed ↗
- 7Grumati P.Dikic I.Ubiquitin signaling and autophagy J. Biol. Chem.29320185404541310.1074/jbc.TM 117.00011729187595 PMC 5900779 · doi ↗ · pubmed ↗
- 8Marasco M.R.Linnemann A.K.beta-Cell Autophagy in Diabetes Pathogenesis Endocrinology 15920182127214110.1210/en.2017-0327329617763 PMC 5913620 · doi ↗ · pubmed ↗