Delay of EGF-Stimulated EGFR Degradation in Myotonic Dystrophy Type 1 (DM1)

Myotonic dystrophy type 1 (DM1) is an autosomal dominant disease caused by a CTG repeat expansion in the 3′ untranslated region of the dystrophia myotonica protein kinase gene. AKT dephosphorylation and autophagy are associated with DM1. Autophagy has been widely studied in DM1, although the endocytic pathway has not. AKT has a critical role in endocytosis, and its phosphorylation is mediated by the activation of tyrosine kinase receptors, such as epidermal growth factor receptor (EGFR). EGF-activated EGFR triggers the internalization and degradation of ligand–receptor complexes that serve as a PI3K/AKT signaling platform. Here, we used primary fibroblasts from healthy subjects and DM1 patients. DM1-derived fibroblasts showed increased autophagy flux, with enlarged endosomes and lysosomes. Thereafter, cells were stimulated with a high concentration of EGF to promote EGFR internalization and degradation. Interestingly, EGF binding to EGFR was reduced in DM1 cells and EGFR internalization was also slowed during the early steps of endocytosis. However, EGF-activated EGFR enhanced AKT and ERK1/2 phosphorylation levels in the DM1-derived fibroblasts. Therefore, there was a delay in EGF-stimulated EGFR endocytosis in DM1 cells; this alteration might be due to the decrease in the binding of EGF to EGFR, and not to a decrease in AKT phosphorylation.


Introduction
Myotonic dystrophy type 1 (DM1) is an inherited disease characterized by progressive muscle weakness and wasting. DM1 is the most common muscular dystrophy; its prevalence varies between 1 and 35 per 100,000 people [1,2]. It is an autosomal dominant disease caused by a nucleotide repeat expansion of cytosine-thymine-guanine (CTG) in the 3 untranslated region (UTR) of the dystrophia myotonica protein kinase (DMPK) gene [1,3].

Immunofluorescence Microscopy
Cells were fixed with 4% paraformaldehyde (PFA) and permeabilized with either 0.1% Triton X-100 (Sigma-Aldrich, T9284) in PBS for 5 min [20] or 0.01% saponin (Fluka BioChemika, Merck KGaA, Darmstadt, Germany, #47036) in BSA (1 mg/mL) for one hour. Saponin was only used for endocytosis/lysosome labeling. Triton permeabilization was followed by a one-hour incubation in BSA. After permeabilization, the cells were incubated with primary antibodies against EEA1 (C45B10) ( D1306) [20]. Images were visualized using an Olympus IX51 inverted microscope equipped with a DP71 camera and with a confocal microscope (A1 confocal imaging system mounted on an inverted Eclipse Ti microscope (Nikon Corp., Tokyo, Japan). The scale of 8-bit images was set from pixels to microns, the threshold was adjusted, and the particle circularity was fixed between 0 and 1. The particle number and the average area occupied in µm 2 by those particles were analyzed with ImageJ software 1.53f51, National Instiutes of Health, USA.

EGFR Degradation
The cells were seeded in 6-well plates containing complete DMEM (as described above). The following day, HFs were washed with PBS and starved overnight in FBS-free DMEM supplemented with 0.1% BSA and L-glutamine (termed binding DMEM) [22]. The cells were then treated with or without hEGF (100 ng/mL) in binding DMEM for 0, 15, 30, or 60 min at 37 • C to monitor EGFR internalization and degradation. Depending on the aim of the experiment, the cells were pretreated with leupeptin or BAF.A1. To stop cell treatment, the plates were washed with ice-cold PBS and immediately stored at −80 • C until further use. For WB, the thawed cells were harvested on ice using lysis buffer (1% Triton X-100, 10% glycerol, 50 mM HEPES (pH 7.4), 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, and protease inhibitors) [22] and centrifuged at 13,000 rpm for 15 min. The supernatant was quantified with a bicinchoninic acid (BCA) kit and loaded onto 4-20% gels.

Binding, Endocytosis, and Trafficking of Alexa Fluor 555 EGF
HFs were seeded on cover slips in 24-well plates. After 24 h, the cells were washed with PBS and starved in binding DMEM (as described above) for 1-2 h at 37 • C. Next, the cells were kept on ice for 10 min and subsequently incubated with 100 ng/mL Alexa-EGF-555 (Invitrogen, Willow Creek Road, OR, USA, E35350) that was diluted in binding medium. After 30 min of incubation on ice, unbound ligand was removed by washing the cells with cold PBS. Time 0 was immediately fixed with 4% PFA to stop the reaction. Warm binding medium was added to the other cover slips, and the internalization of the ligand was analyzed by incubating cells at 37 • C for 5, 10, 20, 30, and 60 min followed by PFA fixation at the indicated times. Images were visualized using an Olympus IX51 inverted microscope equipped with a DP71 camera. The EGF+ vesicles were analyzed with ImageJ software.

Acidic Compartment Staining with LysoTracker Red
Plated cells were detached with trypsin and incubated for 15 min at 37 • C in complete DMEM containing 100 nM Lysotracker Red (LTR, Invitrogen L7528) [24]. LTR is a fluorescent probe that accumulates in acidic organelles (lysosomes and late-endosomes). The percentage of LTR + fluorescence signal (n = 10,000 cells) was determined by flow cytometry (Beckman Coulter FC500-MPL).

N-Acetylglucosaminidase Activity
The activity of β-N-acetylglucosaminidase (NAG) was determined using a colorimetric kit assay (Sigma-Aldrich, CS0780). Ten microliters of cell lysate were incubated with 90 µL of substrate solution (1 mg/mL 4-nitrophenyl-N-acetyl-β-D-glucosaminide [NP-GlcNAc] in 90 mM citrate) for 30 min at 37 • C. The hydrolysis of NP-GlcNAc releases the product p-nitrophenol. The reaction was stopped by adding 200 µL of sodium carbonate solution and incubating at 37 • C for 10 min. The absorbance of the yellow product at 405 nm was measured by the TECAN microplate reader. The experiment was performed according to the manufacturer instructions. The results are presented in U/mL.

Transmission Electron Microscopy (TEM)
HFs were grown on coverslips and subsequently fixed in 2% PFA (16% EM grade, Electron Microscopy Sciences, Hatfield, PA, 15710S) and 2% glutaraldehyde (TAAB Laboratory Equipment Ltd., Aldermaston, England) in 0.1 M PHEM buffer (Electron Microscopy Sciences, 11162) for 30 min at room temperature. The samples were then osmicated and further processed for resin embedding [27]. Resin blocks were sectioned (Leica Microsystems, Wetzlar, Germany; UC7), and 70 nm ultrathin serial sections were collected on formvar-coated slot grids and stained with uranyl acetate and lead citrate. The samples were observed under a transmission electron microscope (Tecnai G2 Spirit, FEI) and imaged with an Orius SC1000B charge-coupled device camera using Digital Micrograph software (both Gatan).

Statistical Analysis
The collected data were analyzed using Microsoft Excel and GraphPad Prism 8.0.2. Student's t test, one-way ANOVA (Tukey's multiple comparison tests), and two-way ANOVA (Tukey's multiple comparison tests) were used to establish significant differences between control (CTRL) and DM1 cells and differences in the treatment effects. A p value < 0.05 indicated statistical significance.

Increase in Autophagy Flux and Cell Death in DM1 Cells
Increased autophagy has been associated with DM1 muscle atrophy [8]; therefore, we assessed autophagy in CTRL and DM1 HFs by treating the cells with 1 µM rapamycin (RAPA) or 100 nM bafilomycin A1 (BAF. A1) for 2 h ( Figure 1A-E). RAPA is an autophagy inducer that increases the autophagosome formation (LC3-II marker) via the downregulation of mTORC1, whereas BAF.A1 inhibits autophagy allowing the accumulation of LC3-II and the sequestration of substrates in the autophagosomes. RAPA ( Figure 1A,B) or BAF.A1 ( Figure 1D,E) significantly increased the level of LC3-II in CTRL and in DM1 HFs. Indeed, RAPA activates the formation of autophagosomes in both cell lines by significantly decreasing the phosphorylation levels of S6 ( Figure 1A,C), a downstream target of mTOR. However, there were no significant differences in p-S6 levels between CTRL and DM1 cells under basal and RAPA conditions. BAF.A1 inhibits the lysosomal V-ATPase and lysosome-mediated degradation of LC3-II, thereby allowing its accumulation [28]. The difference in LC3-II between CTRL and DM1 cells was not significant under basal and RAPA conditions, but it was higher with BAF.A1 treatment ( Figure 1D,E). This difference in the amount of LC3-II suggests an increase in the autophagy flux in DM1 HFs. To evaluate autophagic degradation, SQSTM1/p62 is used as an autophagy substrate due to its selective degradation [29]. In both cell lines, the analysis of SQSTM1/p62 immunostaining indicates a markedly increase in dots per cell with BAF.A1 treatment ( Figure 1F,G), although this increase was lower in DM1 cells. There was also a slight reduction in SQSTM1/p62 levels between CTRL and DM1 cells under basal conditions. However, the decrease in SQSTM1 protein levels in DM1 compared with CTRL cells correlates with a diminution in SQSTM1 mRNA in the DM1 cells ( Figure 1H). Therefore, the reduction in SQSTM1 protein cannot be only attributed to enhanced autophagy degradation in DM1 cells.
The measuring of the autophagic flux of DM1 HFs treated with RAPA and/or BAF.A1 confirmed a significant increase in LC3-II protein levels with BAF. A1 and with the combined treatment (BAF.A1 and RAPA) (Figure 2A,B). Moreover, the difference in LC3-II levels between BAF.A1 and the combined treatment confirmed that autophagy is induced in DM1 cells. In fact, RAPA completely decreased the phosphorylation of S6 (Figure 2A,C). Increased autophagy flux in DM1, evidenced here ( Figure 1D,E and Figure 2A,B) and by others, has been shown to be concomitant with an increase in apoptosis [5]. To test this event, cells were treated with RAPA for 24 h and then stained with annexin V and propidium iodide (PI). We observed that DM1 cells displayed a two-fold increase in annexin-V-positive cells (apoptosis) and PI-positive cells (necrosis) when compared with CTRL cells ( Figure 2D,E). Furthermore, RAPA treatment maintained DM1-induced apoptosis ( Figure 2D). Unexpectedly, it slightly decreased necrosis in DM1 ( Figure 2E), although this decrease did not really reduce the significant ratio of necrotic cell death between CTRL and DM1 cells. Taken together, these data indicate that autophagy and cell death were activated in DM1-derived fibroblasts. The measuring of the autophagic flux of DM1 HFs treated with RAPA and/or BAF.A1 confirmed a significant increase in LC3-II protein levels with BAF. A1 and with the combined treatment (BAF.A1 and RAPA) (Figure 2A,B). Moreover, the difference in LC3-II levels between BAF.A1 and the combined treatment confirmed that autophagy is induced in DM1 cells. In fact, RAPA completely decreased the phosphorylation of S6 ( and by others, has been shown to be concomitant with an increase in apoptosis [5]. To test this event, cells were treated with RAPA for 24 h and then stained with annexin V and propidium iodide (PI). We observed that DM1 cells displayed a two-fold increase in annexin-V-positive cells (apoptosis) and PI-positive cells (necrosis) when compared with CTRL cells (Figure 2D,E). Furthermore, RAPA treatment maintained DM1-induced apoptosis ( Figure 2D). Unexpectedly, it slightly decreased necrosis in DM1 ( Figure 2E), although this decrease did not really reduce the significant ratio of necrotic cell death were assessed by immunoblotting and their densitometry was normalized to the loading control, GAPDH. Data are the mean ± SD of two replicates (* p < 0.05, *** p < 0.001, **** p < 0.0001 versus untreated cells, ¤¤ p < 0.01 versus BAF.A1treated CTRL cells, with two-way ANOVA-Tukey's test), arbitrary units (a.u). Each group (CTRL or DM1) consisted of three cell lines. (F,G) CTRL and DM1 cells were treated with BAF.A1 for 4 h. (F) Cells were immunolabeled with SQSTM1 antibody (Red) and nuclei stained with DAPI (blue), scale bar represents 10 µm. The original magnification is ×60. G/Quantification of SQSTM1 dots per cell was determined using ImageJ software (n = 100 cells/condition). Data are the mean ± SD of two replicates (** p < 0.01 versus untreated cells and ¤ p < 0.05 versus BAF.A1-treated CTRL cells, with two-way ANOVA-Tukey's test). (H) Represents the SQSTM1 mRNA expression level determined by qPCR. Data are the mean ± SD of two replicates (* p < 0.05 respect to CTRL cells). Each group (CTRL or DM1) consisted of three cell lines. All experiments were performed at least three times. between CTRL and DM1 cells. Taken together, these data indicate that autophagy and cell death were activated in DM1-derived fibroblasts.

Enlarged Endosomes in DM1 Cells
The delivery of endocytic cargo to lysosomes is initially routed to early endosomes (RAB5-positive vesicles) and then transferred to late endosomes (RAB7-positive vesicles) [30]. To better understand the endocytic degradative pathway in DM1 HFs, we first characterized the endosomes through the detection of the early endosomal antigen-1 (EEA1), a RAB5-effector that binds phosphatidylinositol-3 phosphate [PI(3)P] via its FYVE domain [31]. The EEA1 protein levels detected by Western blot were similar between CTRL and DM1 cells ( Figure 3A,B). We distinguished two populations of EEA1-positive structures by immunofluorescence in both groups: a perinuclear endosomal subpopulation and a distal subpopulation ( Figure 3C). Although the total number of EEA1-positive structures did not change between CTRL and DM1 cells (119 ± 12.09 vs. 120 ± 9.33, p = 0.67, respectively), the perinuclear early endosomes were morphologically larger in the DM1 HFs than in CTRL HFs. Indeed, the average area occupied by EEA1 structures is greater in the DM1 cells than in the CTRL cells ( Figure 3D). This enlarged endosome subpopulation was also detected by transmission electron microscopy ( Figure  3E). The conversion of RAB5 to RAB7 ensures the maturation of early endosomes into late endosomes [32]. The RAB7 protein levels ( Figure 3F,G) did not change between cell lines, but the immunofluorescence analysis of lyso-bisphosphatidic acid (LBPA), another late endosomal marker [33], showed an increase in LBPA-positive structures in DM1 cells

Enlarged Endosomes in DM1 Cells
The delivery of endocytic cargo to lysosomes is initially routed to early endosomes (RAB5-positive vesicles) and then transferred to late endosomes (RAB7-positive vesicles) [30]. To better understand the endocytic degradative pathway in DM1 HFs, we first characterized the endosomes through the detection of the early endosomal antigen-1 (EEA1), a RAB5effector that binds phosphatidylinositol-3 phosphate [PI(3)P] via its FYVE domain [31]. The EEA1 protein levels detected by Western blot were similar between CTRL and DM1 cells ( Figure 3A,B). We distinguished two populations of EEA1-positive structures by immunofluorescence in both groups: a perinuclear endosomal subpopulation and a distal subpopulation ( Figure 3C). Although the total number of EEA1-positive structures did not change between CTRL and DM1 cells (119 ± 12.09 vs. 120 ± 9.33, p = 0.67, respectively), the perinuclear early endosomes were morphologically larger in the DM1 HFs than in CTRL HFs. Indeed, the average area occupied by EEA1 structures is greater in the DM1 cells than in the CTRL cells ( Figure 3D). This enlarged endosome subpopulation was also detected by transmission electron microscopy ( Figure 3E). The conversion of RAB5 to RAB7 ensures the maturation of early endosomes into late endosomes [32]. The RAB7 protein levels ( Figure 3F,G) did not change between cell lines, but the immunofluorescence analysis of lyso-bisphosphatidic acid (LBPA), another late endosomal marker [33], showed an increase in LBPA-positive structures in DM1 cells when compared with CTRL cells (Figure 3H,I). All these data indicate that the endosome morphology in DM1 cells is different from that of CTRL cells. when compared with CTRL cells (Figure 3H,I). All these data indicate that the endosome morphology in DM1 cells is different from that of CTRL cells.

Enhanced Acidic Vesicles and Enzymatic Activities in DM1 Cells
Lysosomes are acidic organelles that receive material from autophagosomes and endosomes to be degraded [29]. We observed that levels of lysosomal-associated membrane protein (LAMP) 1 ( Figure 4A-D) and LAMP2 ( Figure 4E,F) were not significantly different between CTRL and DM1 cells. Despite the comparable expression levels of ly-

Enhanced Acidic Vesicles and Enzymatic Activities in DM1 Cells
Lysosomes are acidic organelles that receive material from autophagosomes and endosomes to be degraded [29]. We observed that levels of lysosomal-associated membrane protein (LAMP) 1 ( Figure 4A-D) and LAMP2 ( Figure 4E,F) were not significantly different between CTRL and DM1 cells. Despite the comparable expression levels of lysosomal proteins and the number of lysosomes (142.5 ± 11.78 vs. 129.8 ± 10.97, p = 0.43 with Student's t-test), the lysosome size was increased in DM1 cells ( Figure 4C), as shown by the average % area significantly occupied by LAMP1 + structures ( Figure 4D). Consequently, we stained cells with LTR, a fluorescent probe that selectively accumulates in acidic vesicles, especially in lysosomes and in late endosomes [34]. Interestingly, there were significantly more LTR + signals in the DM1 cells ( Figure 4G), as previously demonstrated in the DM1 fly model and in DM1 myoblasts [5]. We think that the increment in LTR signal is related to the increased lysosomal size in DM1 cells. Lysosomal enzymes are essential for the degradation of lysosomal cargoes, and an increased expression of cathepsin B (CTSB) protein level has been reported in human DM1-NSCs [6]. We investigated whether there were differences in the levels of cathepsin (CTS) enzymes between CTRL and DM1 cells. Pro and intermediate enzyme forms were classified as immature CTS. Variations in CTS protein levels were clearly observed within each group of cell line; however, the mature forms of CTSB ( Figure S1A,B), CTSC ( Figure S1D,E), CTSL ( Figure S1G,H), and CTSD ( Figure 4H,I) were not different between the two cell types. A similar result was observed when determining the mCTS/immature ratio ( Figure S1C,F,I). When measuring the enzyme activities, CTSB ( Figure 4J) and β-HEX ( Figure 4K) levels were not different between the two cell types. However, the NAG ( Figure 4L) and CTSD ( Figure 4M) activities were increased in DM1 cells when compared with CTRL cells; similar results were observed in dystrophic muscles [35]. Therefore, DM1-derived fibroblasts exhibited increased lysosomal size and a significant increase in CTSD activity, suggesting a degree of lysosome functionality.

Defect in Fluorescent EGF Binding and Delay of EGFR Degradation in DM1 Cells
Endosomes ( Figure 3C) and lysosomes ( Figure 4C) are enlarged in DM1 cells when compared with CTRL cells; therefore, we examined their impact in the degradative endocytic pathway by incubating fibroblasts with Alexa-555-EGF (555-EGF) for 30 min on ice and followed the evolution of endocytosed 555-EGF by immunofluorescence microscopy at different time points (Figure 5A). At time zero, 555-EGF binding was significantly reduced in DM1 cells compared with CTRL cells ( Figure 5B). During the incubation of the fibroblasts at 37 • C, the amount of 555-EGF was quantified at different times from 5 to 60 min. Given the difference in 555-EGF binding, the percentage of 555-EGF + in each group was considered to be 100 at time zero. Firstly, we observed an increase in 555-EGF + vesicles in CTRL and DM1 HFs at 10 and 20 min, respectively ( Figure 5C). Subsequently, the amount of endocytosed 555-EGF begin to decrease and we observed a significant delay in 555-EGF degradation in DM1 cells compared with CTRL at 20 min. Such a delay disappeared around 60 min of incubation in DM1 cells. We then conclude that there was a decrease in the fluorescent EGF binding to EGFR in DM1, followed by a delay in 555-EGF trafficking and degradation during the first 20 and 30 min of endocytosis.
age of EGFR in each group was considered to be 100 at time zero. This percentage started to decrease progressively in CTRL and DM1 cells from 15 min to 60 min ( Figure 5E). Additionally, the degradation of EGFR was slower in DM1 cells than in CTRL cells at 15 min (p = 0.02). Interestingly, this difference in EGFR protein level tended to disappear from 30 min (p = 0.07) and become similar at 60 min in CTRL cells and in DM1 cells. Overall, these results sustain a delay in EGF-stimulated EGFR degradation in DM1 cells between 0 and 30 min. To corroborate the defect in the endocytosis trafficking of 555-EGF in DM1 cells, we assessed the EGFR protein level in HFs upon EGF ligand stimulation. Serum-starved cells were treated with 100 ng/mL EGF for 15, 30, and 60 min ( Figure 5D). EGF treatment stimulates and triggers the activation and internalization of the EGF receptor (EGFR) via the endocytic pathway. At time zero, the level of EGFR protein between CTRL and DM1 cells was similar (100 ± 15.81 vs. 101.44 ± 32.44, respectively). Nevertheless, the percentage of EGFR in each group was considered to be 100 at time zero. This percentage started to decrease progressively in CTRL and DM1 cells from 15 min to 60 min ( Figure 5E). Additionally, the degradation of EGFR was slower in DM1 cells than in CTRL cells at 15 min (p = 0.02). Interestingly, this difference in EGFR protein level tended to disappear from 30 min (p = 0.07) and become similar at 60 min in CTRL cells and in DM1 cells. Overall, these results sustain a delay in EGF-stimulated EGFR degradation in DM1 cells between 0 and 30 min.

Activation of EGFR Signaling Pathway upon EGF Stimulation in DM1 Cells
EGF-induced EGFR activation stimulates downstream signaling pathways such as AKT and ERK1/2 [36], which results in an increase in their phosphorylation levels and EGFR endocytosis [37]. To study whether EGFR signaling is activated in DM1 cells, serumstarved cells were treated with EGF for 0, 20, and 60 min. Upon EGF stimulation, the phosphorylation levels of AKT (Ser473) and ERK1/2(Thr202/Tyr204) significantly increased in both cell lines at 20 min ( Figure 6). This increase was markedly reduced at 60 min, suggesting that the phosphorylation observed at 20 min was followed by a phase of decay, which results in an attenuation of the EGFR signaling. It is important to note that despite the decrease in EGF binding to EGFR, the EGFR downstream signaling was activated at 20 min in DM1 cells as compared with time zero. However, there was a sustained decrease in p-AKT in DM1 cells as compared with CTRL cells at 20 and 60 min. These data allow us to conclude that EGF-activated EGFR is able to signal from endosomes and its delay in the endocytic trafficking does not affect the downstream (AKT and ERK1/2) signaling pathways in DM1 cells, but highlights a difference in p-AKT levels between cell lines.
Data are the percentage ± SEM of 555-EGF + relative to time zero (* p < 0.05 DM1 versus CTRL cells at 20 min, with a multiple Student's t-test), reflecting the trafficking and degradation of the fluorescent EGF. (D,E) CTRL and DM1 were starved overnight and then incubated at different time points (0, 15, 30, and 60 min) with 100 ng/mL EGF. The EGFR protein level was normalized to GAPDH, and the percentage was determined relative to time zero. (E) Data are the mean percentage ± SD of EGFR degradation at the indicated times (* p < 0.05, with multiple Student's t-tests).

Activation of EGFR Signaling Pathway upon EGF Stimulation in DM1 Cells
EGF-induced EGFR activation stimulates downstream signaling pathways such as AKT and ERK1/2 [36], which results in an increase in their phosphorylation levels and EGFR endocytosis [37]. To study whether EGFR signaling is activated in DM1 cells, serum-starved cells were treated with EGF for 0, 20, and 60 min. Upon EGF stimulation, the phosphorylation levels of AKT (Ser473) and ERK1/2(Thr202/Tyr204) significantly increased in both cell lines at 20 min ( Figure 6). This increase was markedly reduced at 60 min, suggesting that the phosphorylation observed at 20 min was followed by a phase of decay, which results in an attenuation of the EGFR signaling. It is important to note that despite the decrease in EGF binding to EGFR, the EGFR downstream signaling was activated at 20 min in DM1 cells as compared with time zero. However, there was a sustained decrease in p-AKT in DM1 cells as compared with CTRL cells at 20 and 60 min. These data allow us to conclude that EGF-activated EGFR is able to signal from endosomes and its delay in the endocytic trafficking does not affect the downstream (AKT and ERK1/2) signaling pathways in DM1 cells, but highlights a difference in p-AKT levels between cell lines.   A,B) and p-ERK1/2 (A,C) were normalized to GAPDH. Values are the mean ± SD of two replicates (** p < 0.01, **** p < 0.0001 versus time zero and ¤¤ p < 0.01, ¤¤¤ p < 0.001, ¤¤¤¤ p < 0.0001 versus 20 min with two-way ANOVA-Tukey's test).

Endocytosed EGFR Is Sorted and Degraded into Lysosomes in DM1 Cells
To determine whether EGF-stimulated EGFR is sorted to lysosomes, serum-starved cells were treated with EGF (100 ng/mL) and co-labeled with EGFR and LAMP1 antibodies ( Figure 7A). EGFR was significantly perinuclear in unstimulated DM1 cells ( Figure 7A,B). This result can explain, at least in part, the lower binding of EGF to EGFR in DM1 cells compared with the control cells. This perinuclear distribution persisted and was unchanged upon EGF treatment in DM1 cells, unlike in CTRL cells. In agreement with this intracellular distribution of EGFR, the fluorescence intensity of EGFR significantly decreased upon EGF stimulation for 60 min in CTRL and DM1 cells ( Figure 7C). We then evaluated the colocalization of EGFR and LAMP1. Despite the significant difference between CTRL and DM1 cells under basal conditions, the colocalization of EGF-stimulated EGFR with LAMP1 was similar in both cell types at 60 min ( Figure 7D). Subsequently, serum-starved cells were pretreated with the lysosomal inhibitor BAF.A1 (100 nM) for one hour and then stimulated for another hour with EGF. A significant accumulation of EGFR was observed in the BAF.A1-treated cells compared with EGF alone ( Figure 7E,F), indicating that BAF.A1 reverses EGF-induced EGFR degradation in CTRL and in DM1 cells. Taken together, these results indicate that the degradation of EGF-stimulated EGFR occurs via the endosomallysosomal pathway in both cell lines, although the binding of EGF to its receptor is reduced in DM1 cells.
To determine whether EGF-stimulated EGFR is sorted to lysosomes, serum-starved cells were treated with EGF (100 ng/mL) and co-labeled with EGFR and LAMP1 antibodies ( Figure 7A). EGFR was significantly perinuclear in unstimulated DM1 cells (Figure 7A,B). This result can explain, at least in part, the lower binding of EGF to EGFR in DM1 cells compared with the control cells. This perinuclear distribution persisted and was unchanged upon EGF treatment in DM1 cells, unlike in CTRL cells. In agreement with this intracellular distribution of EGFR, the fluorescence intensity of EGFR significantly decreased upon EGF stimulation for 60 min in CTRL and DM1 cells ( Figure 7C). We then evaluated the colocalization of EGFR and LAMP1. Despite the significant difference between CTRL and DM1 cells under basal conditions, the colocalization of EGF-stimulated EGFR with LAMP1 was similar in both cell types at 60 min ( Figure 7D). Subsequently, serum-starved cells were pretreated with the lysosomal inhibitor BAF.A1 (100 nM) for one hour and then stimulated for another hour with EGF. A significant accumulation of EGFR was observed in the BAF.A1-treated cells compared with EGF alone ( Figure 7E,F), indicating that BAF.A1 reverses EGF-induced EGFR degradation in CTRL and in DM1 cells. Taken together, these results indicate that the degradation of EGF-stimulated EGFR occurs via the endosomal-lysosomal pathway in both cell lines, although the binding of EGF to its receptor is reduced in DM1 cells. The perinuclear distribution of EGFR that was determined by manual counting, n = 80-100 cells (* p < 0.05, ** p < 0.01 two-way ANOVA Tukey's test). (C) The fluorescence intensity of EGFR represented as integrated density from ImageJ software (** p < 0.01, **** p < 0.0001 two-way ANOVA Tukey's test). (D) Reflects the co-localization between EGFR/LAMP1 using Pearson's coefficient (** p < 0.01, **** p < 0.0001 with two-way ANOVA). All experiments were performed at least two times. (E,F) Serum-starved cells (CTRL and DM1) were pre-treated for 60 min with 100 nM BAF.A1 and then stimulated with 100 ng/mL EGF for one hour. (F) Represents the level of EGF-induced EGFR degradation normalized to GAPDH. Data are the mean ± SD of two replicates, (* p < 0.05, with two-way ANOVA Tukey's test).

Discussion
Human fibroblasts from DM1 patients displayed an increase in autophagic flux, despite the downregulated SQSTM1 expression. Most important, lysosomal enzyme activities such as CTSD and NAG were augmented. Moreover, DM1 cells exhibited enlarged perinuclear endosomes and an increase in lysosomal size, which was in correlation with the significant increase in LTR fluorescence signal. Additionally, EGF-stimulated EGFR was internalized and degraded in lysosomes, and such a degradation was prevented with BAF.A1 treatment in CTRL and DM1 cells. However, the binding of EGF to EGFR and EGFR trafficking were significantly reduced in DM1 cells. In spite of the activation of the EGFR signaling pathway, EGFR endocytosis was slowed in DM1 cells.
Autophagy was upregulated in DM1-derived fibroblasts (Figure 1), and the impairment of this catabolic process has been widely reported in different DM1 models [5][6][7]9]. However, fibroblasts from three investigated DM1 patients differentiated into myotubes displayed a blocked autophagic flux under the combination of chloroquine treatment and/or starvation [9]. At the same time, chloroquine did not generate additional autophagic vacuoles over the already induced basal autophagy in DM1-NSCs [6]. In our models, LC3-II (an indicator of autophagosome formation) was significantly accumulated with BAF.A1 treatment ( Figure 1D,E), which was accompanied by an increase in SQSTM1/p62 puncta ( Figure 1F,G). Autophagy is negatively regulated by the Class I PI3K/AKT/mTOR pathway [38], and its exacerbation contributes to muscle atrophy. In muscle atrophy [10] and eventually in DM1, the phosphorylation level of AKT is downregulated and correlated with the reduction in DMPK and muscleblind-like 1 (MBNL1) proteins [16]. Rapamycin, an mTOR inhibitor, promoted LC3 lipidation in the DM1-derived fibroblasts ( Figure 1A,B) but not in muscle tissue from HSA LR mice [9]; however, it was reported to sufficiently improve muscle function. Hence, this effect was attributed to an mTOR-independent mechanism. In our study, rapamycin downregulated S6(Ser235/236) phosphorylation levels (Figure 2A,C) and did not reduce apoptosis in DM1 HFs after 24 h ( Figure 2D). Consistent with these data, autophagy inhibition through Tor overexpression prevents CTG-induced muscle atrophy in DM1 model flies [5]. Another study reinforced that autophagy inhibition improved the proliferation of DM1 skeletal muscle satellite cells via the overexpression of MBNL1, which activates mTOR [39]. Although autophagy outcomes have been controversial among the studied DM1 models, there is evidence that AKT/mTOR pathway dysregulation [5,6,9] is involved in the pathogenesis of myotonic dystrophy.
AKT phosphorylation is also regulated by the activation of tyrosine kinase receptors, such as the EGFR. Upon stimulation, EGFR triggers a downstream signaling cascade that involves ERK1/2 and AKT. The internalization and degradation of EGFR in lysosomes promotes signal attenuation [12]. To date, there are not sufficient data to fully clarify the role of endocytosis in DM1. Increased endocytosis was associated with dystrophic muscle fibers, as evidenced by the formation of vesicles from transverse (T) tubules [35,40]. T-tubules are essential for the coordination of calcium release and muscle contraction. Endocytosis ensures the formation of T-tubules from the sarcolemma by membrane sequestration and stabilizes its integrity through tight junctions with the sarcoplasmic reticulum [41]. Therefore, T-tubule alterations in DM1 [42,43] allow us to think that endocytosis may be altered in skeletal muscles cells. In this study, we demonstrated that DM1 HFs contained enlarged endosomes ( Figure 3C-E) and lysosomes ( Figure 4C,D). This phenotype suggested that the endocytosis pathway could be altered. Interestingly, the binding of EGF to EGFR was reduced and the receptor internalization was also decreased ( Figure 5). Despite the enlarged perinuclear endosomes in DM1 cells, the endocytic trafficking was slowed during early EGFR trafficking, but it subsequently reached the EGFR level of the CTRL cells in the later phase of EGF stimulation (e.g., 1 h). A previous study attributed the decrease in EGFR degradation to complete AKT inhibition leading to an accumulation of non-degraded EGF-EGFR complex in EEA1-positive structures [12]. Accordingly, the decrease in AKT phosphorylation [16] could explain the endosome morphology in DM1 cells. Additionally, a crosstalk between autophagy and the endocytic pathway has been established, because autophagy inhibition has been shown to induce damaged endosomes, reduce EGFR recycling, and prevent EGF-mediated signaling [44]. The induction of autophagy in DM1 cells allows us to think that endosomes are not damaged in our experimental model. In addition, there was no co-localization between EEA1 and LAMP1 (data not shown).
On the other hand, EGFR endocytosis depends on the expression and localization of the EGFR receptor and on its binding to the ligand [37]. We have not observed a significant difference in EGFR protein levels between cell lines, although it fluctuated in serum-starved CTRL and DM1 cells. In both cell lines, the EGFR-bound EGF resulted in a significant activation of AKT and ERK1/2, which was subsequently down-modulated (Figure 6), suggesting a degradation of EGFR [45]. Therefore, primary human fibroblasts are a good model to study EGFR-mediated signaling. We hypothesized that the decrease in EGF binding might delay the internalization of the receptor in DM1 cells but did not impede its sorting and its degradation in lysosomes, which was clearly prevented with BAF.A1 pre-treatment ( Figure 7E,F). Similar results were expected with the addition of 50 µM leupeptin, but without success. This may be due to inadequate concentrations or very short treatment duration. Even though EGFR endocytosis was delayed in DM1 cells, we think that the significant increase in LBPA proteins and CTSD activity might be a compensatory mechanism that mediates the sorting and the degradation of EGF-EGFR complexes.
We conclude that the delay in EGFR trafficking is not due to decreased receptor expression or inhibition of AKT in EGF-treated DM1-derived fibroblasts. However, it is the consequence of a perinuclear distribution of the receptor, which significantly reduces the binding of the ligand. Therefore, it will be interesting to stimulate DM1 cells with low EGF concentrations to promote EGFR recycling and to check whether the perinuclear distribution of EGFR will decrease and the ligand binding rate will be restored. The insulin receptor is a tyrosine kinase receptor which is endocytosed upon insulin binding to regulate metabolism [46]. Low insulin responsiveness has been attributed to the aberrant alternative splicing of insulin receptor in DM1 muscle cells [47]. Although descriptive, we think that our findings will provide new insights into the low-insulin-induced metabolic effects in DM1. Moreover, a subcellular distribution of insulin receptor might be a contributing factor to insulin resistance [46]. Nevertheless, further studies are required to fully elucidate EGFR activity in DM1, because altered EGFR signaling is associated with human cancers [48]. Several publications have indexed the overall risk of cancers in DM1 patients [4,49], with the most prevalent being skin cancers, specifically basal cell carcinoma [50,51]. To that end, we cannot exclude endocytic pathway alterations from the mechanism underlying the onset of cancers and/or metabolic diseases [52] in DM1 patients. This study was performed in DM1 primary human fibroblasts; therefore, it would be essential to elucidate the role of endocytosis in DM1 skeletal muscle cells. Impaired endocytosis could be involved in the etiology of DM1 by altering calcium homeostasis. This hypothesis seems to be supported by several studies that have reported a morphological alteration of the sarcoplasmic reticulum and/or the tubular system in myotonic dystrophy [42,43]. Funding: This research was supported by the Isabel Gemio Foundation (P18-13) and was also partially supported by the "Fondo Europeo de Desarrollo Regional" (FEDER) from the European Union.