Biochimie 121 (2016) 123e130

Contents lists available at ScienceDirect

Biochimie journal homepage: www.elsevier.com/locate/biochi

Research paper

Alteration of cathepsin D trafficking induced by hypoxia and extracellular acidification in MCF-7 breast cancer cells Oussama Achour a, Yahya Ashraf b, Nicolas Bridiau a, Meriem Kacem a, Nicolas Poupard a, de ric Sannier a, Nathalie Lamerant-Fayel c, phanie Bordenave-Juchereau a, Fre Ste c Claudine Kieda , Emmanuelle Liaudet-Coopman b, Jean-Marie Piot a, Thierry Maugard a, Ingrid Fruitier-Arnaudin a, * a Universit e de La Rochelle, UMR CNRS 7266, LIENSs, Equipe Approches Mol eculaires, Environnement-Sant e, D epartement de Biotechnologies, Avenue Michel Cr epeau, 17042 La Rochelle, France b Institut de Recherche en Canc erologie de Montpellier (IRCM), Institut National de la Sant e et de la Recherche M edicale (INSERM) U896, Universit e Montpellier1, Montpellier, France c Centre de Biophysique Mol eculaire, UPR CNRS 4301, Orl eans, France

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 February 2015 Accepted 9 November 2015 Available online 12 November 2015

The microenvironment that surrounds tumor cells is characterized by hypoxic conditions and extracellular acidity. These hostile conditions induce crucial changes in cell behavior and can promote the secretion of many soluble factors such as growth factors, cytokines and enzymes. The lysosomal aspartylendopeptidase cathepsin D (CD) is a marker of poor prognosis in breast cancer and is associated with a metastatic risk. In this study, the transport of CD was investigated in a model of breast cancer cells line (MCF-7) cultivated under hypoxia and acidification of media. CD secretion was assessed using Western blot analysis and protease activity was measured in conditioned culture media. We demonstrate that cultured MCF-7 cells secrete an active 52 kDa pCD precursor and report that under hypoxia there was an increased amount of pCD secreted. More surprisingly, extracellular acidification (pH 6 and 5.6) induced the secretion of the fully-mature and active (34 kDa þ 14 kDa) double chain CD. Our findings reflect the fact that chemical anomalies influence the secretion path of CD in a breast cancer cell model, resulting in altered trafficking of the mature form. This important result may provide new arguments in favor of the role of extracellular CD in the degradation of the matrix proteins that constitute the breast tumor microenvironment.  te  Française de Biochimie et Biologie Mole culaire (SFBBM). All rights © 2015 Elsevier B.V. and Socie reserved.

Keywords: Cathepsin D Pro-cathepsin D Tumor microenvironment Hypoxia Acidification Proteolytic activity

1. Introduction The tumor microenvironment (TME) is a heterogeneous and complex medium that consists of cells, soluble factors, signaling molecules, an extracellular matrix and mechanical cues that surround and feed the tumor [1]. TME conditions are considerably different from those found in normal tissues. Indeed most tumors

Abbreviations: pCD, pro-cathepsin D; TME, tumor microenvironment; CD, cathepsin D; ppCD, pre-pro-cathepsin D; MCA, 7-methoxycoumarin-4-acetyl; DNP, dintrophenyl; SA, sodium acetate; MES, 2-(N-morpholino)ethanesulfonic acid); MOPS, 3-(N-morpholino)propanesulfonic acid. * Corresponding author. E-mail address: [email protected] (I. Fruitier-Arnaudin).

develop a patho-physiological microenvironment characterized by low oxygen levels (hypoxia) [2], low glucose concentration [3] and micro-acidic conditions [4]. In addition, a variety of hydrolytic enzymes are over-expressed by both cancer and stromal cells during the different stages of tumor progression and are often hypersecreted in the TME [5]. Cathepsin D (CD, EC 3.4.23.5) is a lysosomal aspartyl endopeptidase present in all healthy cells except erythrocytes [6]. CD is synthesized as a pre-pro-cathepsin D (ppCD) in the rough endoplasmic reticulum. After cleavage of the signal peptide, the 52 kDa pro-cathepsin D (pCD) is glycosylated at asparagine residues 70 and 199 with two N-linked oligosaccharides [7]. This glycoprotein is transferred to the Golgi apparatus, where the mannose of the oligosaccharide is phosphorylated [8,9]. The pCD is targeted to the

http://dx.doi.org/10.1016/j.biochi.2015.11.007  te  Française de Biochimie et Biologie Mole culaire (SFBBM). All rights reserved. 0300-9084/© 2015 Elsevier B.V. and Socie

124

O. Achour et al. / Biochimie 121 (2016) 123e130

late endosome by means of the mannose-6-phosphate receptor (M6P-R), and there it is rapidly converted to an active, 48 kDa, single-chain intermediate, before being transported to the lysosomes where it becomes the fully active mature protease composed of a 34 kDa heavy chain and a 14 kDa light chain [10] (see Fig. 8). Numerous physiological functions have been described for CD based on its ability to cleave structural and functional proteins and peptides [11]. Moreover, CD can also act independently of its proteolytic activity. It has been linked with several pathologies, including cancer, and it is associated with tumor progression, angiogenesis and metastasis [11]. In some tumors, pCD escapes the normal targeting mechanisms and can be hyper-secreted into the extracellular space [15]. Thus, an increase in pCD has been observed in several human neoplastic tissues in the breast [12] and thyroid [13], and in prostate sarcoma [14]. This hypersecretion was associated with tumor invasive phenotypes and suggests that pCD is involved in the proteolysis and degradation of extracellular matrix proteins [11]. Although pCD (52 kDa) is supposed to be proteolytically inactive, some studies identified its ability to auto-activate under acidic extracellular conditions, giving rise to a catalyticallyactive 51 kDa pseudo-cathepsin D [16,17]. Based on these data, CD is considered a promising target for cancer therapy and, in our laboratory, we are currently focusing on CD as an entry point for innovative cancer treatments [18,19]. Our objective is to design a new molecular technology to visualize and deliver drugs specifically to breast tumors by exploiting the unique pathological patterns of the breast TME. Moreover, the ability to evaluate specific proteolytic activity in vivo would have considerable clinical applications such as an earlier detection of cancer and the targeting of drugs to localized areas or tissues. Recently, we validated a novel CD peptide substrate which is a part of a new device for TME targeting [18]. We demonstrated its ability to be recognized and cleaved by both auto-activated 51-kDa recombinant CD and mature 34kDa þ14-kDa double chain CD at pH conditions similar to those found in the TME (i.e. pH 6.8). In order to further develop this new device, we must take into consideration the complexity of the TME using an in vitro model where CD secretion and proteolytic activity can be assessed. Moreover, some data in the literature remain ambiguous and unclear, especially those concerning the effect of chemical anomalies developed by the TME, such as hypoxia or extracellular acidification, on the secretion of active forms of CD. To attain our objectives, the human MCF-7 breast cancer cell line was cultivated in either hypoxic or acidic conditions, mimicking the TME created by solid mammary tumors, and the impact of these types of chemical environment on the secretion and proteolytic activity of CD were explored.

2.2. Cell culture The MCF-7 human mammary carcinoma cells were cultured at 37  C in a complete medium: Opti-MEM® Reduced-Serum Medium with glutaMAX™ supplemented with 2% heat-inactivated fetal bovine serum (FBS) (56  C, 1 h), 0.1% Fungizone and 0.2% Gentamicine. Cells were cultured in 6 well plates in a final volume of 1 mL of culture medium with a starting cell number of 1  104 cells per well. For hypoxia experiments, the 6 well plates were placed in an airtight chamber flushed with a gas mixture containing 94% nitrogen, 5% carbon dioxide and 1% oxygen. Equal temperatures were maintained for all treatment groups by sealing and placing the chamber inside the incubator used for normoxic conditions (5% carbon dioxide and atmospheric air). For acidification conditions, the pH of the complete medium was adjusted to 7.5, 6.5, 6.0 or 5.6 with 2-(N-morpholino)ethanesulfonic acid (MES) buffer before the inoculation of MCF-7 cells. Conditioned culture media were taken after 24 h, 48 h and 72 h, centrifuged at 700 g for 5 min, filtered with a 0.22 mm syringe filter unit (Millex-GV, PVDF 33 mm, Merck Millipore) and stored at 80  C until analysis. Cells were harvested from each well with trypsin, centrifuged at 700 g for 5 min, and counted visually in hemocytometer chambers. Trypan blue was used to assess the number of viable cells and the cells viability. The cells viability in percent was calculated using the following equation:

Cell viabilityð%Þ ¼ ðNumber of viable cells=Total Number of cellsÞ  100 2.3. Kinetics assays of CD using the fluorogenic substrate Cathepsin D activity was measured in conditioned culture medium using a fluorogenic peptide. Hydrolysis of the fluorogenic peptide (final concentration of 5 mM) was performed in white 96 half-well plates (Corning® #3693) using a BMG Labtech Fluostar Omega spectrofluorometer. The reaction was conducted with 5 mL of conditioned culture medium in a final volume of 100 mL and at 37  C. Three different buffers were used to adjust the reaction pH: Sodium acetate buffer (for pH 3.7, pH 4.5 and pH 5.6), Sodium MES Buffer (for pH 5.6, pH 6 and pH 6.5) and Sodium MOPS (for pH 6.5 and pH 6.8). The fluorescence was measured at lem ¼ 390 nm and lex ¼ 330 nm with an interval of 20 s. A negative control was obtained by the addition of complete culture medium and a positive control by the addition of active pro-cathepsin D and mature double chain Cathepsin D (final concentration of 1 ng/mL). Moreover, Pepstatin A (5 mM) was systematically used as an aspartyl protease inhibitor for all conditions (20). 2.4. Western blot

2. Materials and methods 2.1. Materials All reagents, unless otherwise specified, were purchased from Sigma Aldrich® (St Quentin Falavier, France). Recombinant pCD (52 kDa), which is more than 95% pure, was purchased from R&D systems® (USA). CD activity was measured using a fluorogenic peptide (R[K-DNP]LRFFLIPK[G-MCA]) supplied by Sigma Aldrich® (MCA: 7-Methoxycoumarin-4-Acetyl and DNP: dinitrophenyl). The MCF-7 cell line was purchased from ATCC. Opti-MEM® ReducedSerum Medium with glutaMAX™, fetal bovine serum (FBS), Fungizone and Gentamicine were purchased from Life Technologies (France).

Mouse anti-human cath-D was obtained from BD Biosciences. Sheep anti-mouse IgG-peroxidase conjugates was supplied by GE Healthcare. Rabbit anti-human actin was obtained from SigmaeAldrich and goat anti-rabbit IgG-peroxidase conjugates was from Cell Signaling Technology. Samples (50 mL) taken from the MCF7 cell culture medium were separated by SDS-PAGE on 15% gels (prestained molecular masses; Precision Plus Protein Standards; Bio-Rad, Hercules, CA, USA) and electroblotted onto nitrocellulose membranes. These membranes were first incubated with the primary antibody (1 mg/ml, in PBS, 0.1% Tween, and 5% dried milk for 1 h at room temperature), then with the secondary IgG-peroxidase conjugate for 1 h at room temperature. Proteins were detected by chemiluminescence (ECL Plus Western Blotting detection system;

O. Achour et al. / Biochimie 121 (2016) 123e130

Amersham Biosciences). MCF-7 cell lysate was used as positive control for actin, the 52 kDa pCD, the mono-chain 48 kDa CD, and the mature doublechain 34 þ 14 kDa CD. Cells were grown to 80% confluency in DMEM medium supplemented with 10% FBS. Cells were lysed in lysis buffer (20 mM TriseHCl [pH 7.4], 137 mM NaCl, 1 mM MgCl2, 1 mM CaCl2, 0.5% Nonidet P-40, 1 mM phenylmethylsulphonyl fluoride, 1 mM dithiothreitol, 1 mM vanadate, 10% glycerol, and 100 kallikrein units/ml Trasylol). After gentle shaking for 20 min at 4  C, cell extracts were obtained by centrifugation in a microfuge at 13 000 rpm X g for 15 min. MCF-7 conditioned medium (7 days) was used as the 52 kDa pCD positive control that was semi-quantified on a Western-blot by comparing the band intensity with that of a commercial 52 kDa pCD. 2.5. Statistical analysis The data are presented as the means ± SE of at least one triplicate runs. For cell count, cell viability and cathepsin D activity, statistical analysis was performed using Sigma Plot software (Systat Software Inc., San Jose, USA). Replicates (n ¼ 3) were averaged and the values were tested for normality (ShapiroeWilk). Data originating from three individual culture well (N ¼ 3) were analyzed by a one way ANOVA test. 3. Results and discussion 3.1. Effect of hypoxia on CD trafficking in breast cancer We first investigated the effect of hypoxia on CD secretion by MCF-7 breast cancer cells. MCF-7 cells were cultivated under normoxic or hypoxic conditions for 24 h, 48 h or 72 h. After incubation, adherent cells were removed by trypsin treatment and the number of cells was counted in a hemocytometer chamber using trypan blue exclusion to assess cell viability. Fig. 1 presents the effect of normoxic and hypoxic culture conditions on MCF-7 viable cell number (A) and viability (B). After 24 h of incubation, the number of living cells was 1.1  106 (±0.04  106) cell/mL for plates cultured in normoxic conditions and 1.083  106 (±0.062  106) cell/mL for plates cultured in hypoxic conditions. While the number of living cells cultured in normoxic conditions increased progressively to attain 1.74  106 (±0.199  106) cell/mL after 72 h, the number of living cells cultured in hypoxic conditions remained stable after 48 h before decreasing

to 0.82  106 (±0.074  106) cell/mL after 72 h of incubation (Fig. 1A). Beside, the cell viability in percent (Fig. 1B) remains relatively stable after the different culture time. At an early stage of culture, MCF-7 cell proliferation did not seem to be affected by the hypoxic stress. However, cells that were cultured in hypoxic conditions seemed to be less resistant at the later stage of culture. In order to study the secretion of CD into the extracellular medium and to identify which CD isoform was released under hypoxia, Western blot analysis was performed with the conditioned media obtained from cells cultured under normoxic or hypoxic conditions using a CD-specific antibody that can recognize all CD forms. The Western blot results showed the presence of only the 52 kDa form of CD, corresponding to the pCD pro-enzyme (Fig. 2). The absence of the mature 34 kDa CD and actin in the conditioned culture media indicated that cell lysis, which would release cellular mature lysosomal CD did not occur in our experimental conditions. These data confirm previous studies describing the over-secretion of the pCD form by MCF-7 cells under normoxic conditions [17]. As expected, the intensity of the pCD band increased over time due to the accumulation of secreted pCD. Interestingly, the band obtained after 72 h of incubation under hypoxic conditions was more intense compared to that obtained under normoxic conditions, suggesting a higher quantity of pCD released under hypoxia. The 52 kDa pCD form is usually considered to be catalytically inactive, although it has been demonstrated that it can have a proapoptotic effect [20]. Hypoxia is known to trigger apoptosis in many cell lines [21,22] which may be one of the reasons of the hypersecretion of pCD that act as a proapoptotic signal. At acidic pH, pCD can auto-activate into a 51 kDa pseudo-form that has enzymatic activity [15,16]. This auto-activation, which until now has been observed in cellulo, involves the intra-molecular proteolysis of 26 amino acid residues to yield an active pseudo-CD form [23]. Consequently, we next investigated the proteolytic activity of pCD released into the conditioned medium under hypoxia or normoxia conditions using a CD fluorogenic substrate at acidic pH 3.7 [18]. Fluorescence intensity was monitored over time and the product concentration was calculated using a standard curve obtained with pure MCA, which is the fluorescent molecule used to label the peptide substrate. Fig. 3A shows the change in concentration of the product obtained from the hydrolysis of the fluorogenic peptide in the conditioned media under hypoxia or normoxia at 24, 48 and 72 h. An increase in product concentration is observed over reaction time which indicates the presence of an enzymatic activity that degrades the fluorogenic peptide. Moreover, no cleavage was observed at pH

120

2.5e+6

**

Normoxic conditions Hypoxic conditions

**

2.0e+6

A

* 1.5e+6

1.0e+6

5.0e+5

B

100

Cell viability (%)

Number of viable cells / mL

125

80 60 40 20 0

0.0 24 h

48 h

Culture time (h)

72 h

24 h

48 h

72 h

Culture time (h)

Fig. 1. Effect of normoxic and hypoxic culture conditions on MCF-7 viable cell number (A) and on cell viability (B). MCF-7 cells were cultured in normoxic or hypoxic conditions for 24e72 h. Adherent cells were lifted by trypsin treatment and counted by hemocytometer. The percentage of dead cells was determined by trypan blue exclusion. The error bar represents the standard deviation of at least one triplicate runs. Statistical significance was assessed by the one way ANOVA test (n ¼ 3). The asterisk indicates that p is less than 0.05 and the double asterisk denotes that p is less than 0.01.

126

O. Achour et al. / Biochimie 121 (2016) 123e130

Fig. 2. Western blot analysis of conditioned media obtained from MCF-7 cells cultured under hypoxic or normoxic conditions for 24 h, 48 h or 72 h. A: MCF-7 cell lysate (10 mg); B: MCF-7 conditioned medium (5 ng of Cathepsin D).

Fig. 3. Evolution over time in the concentration of the product obtained from the hydrolysis of a fluorogenic peptide by the enzyme in conditioned culture media at pH 3.7 (A) and the corresponding initial rate (B). The error bar represents the standard deviation of at least three values obtained from separate conditioned culture media. Statistical significance was assessed by the one way ANOVA test (n ¼ 3). The asterisk indicates that p is less than 0.05 and the double asterisk denotes that p is less than 0.01.

3.7 when the reaction was carried out with the negative control (i.e. complete culture medium in the absence of MCF-7 cells) confirming that the enzymatic activity results from cultured MCF-7 cells. As well, all the experiments were performed in the presence or absence of Pepstatin A, which is a specific inhibitor of aspartyl proteases [24] and no cleavage in the reactions performed in the presence of Pepstatin A (data not shown). These data thus indicate that the observed catalytic activity was due to an aspartyl protease released extracellularly by MCF-7 cells. Moreover, at acidic pH, the fluorogenic peptide that was used is known to be specifically cleaved by aspartyl proteases such as Cathepsin D and E [25]. Cathepsin E is only produced by certain types of cells [26] and, to our knowledge, production of Cathepsin E by MCF-7 cells has never been reported. Added to the Western blot results that showed the presence of pCD in the extracellular media, these data strongly suggest that pCD is the protease responsible for the observed enzymatic activity at pH 3.7. The initial rates were measured as the slope of the linear segment of the kinetics plot for the product. Fig. 3B shows the initial rate of production of the fluorogenic peptide hydrolysis product using the conditioned culture media. The initial rate of substrate cleavage by conditioned culture medium obtained in hypoxic conditions was always significantly higher than that of conditioned culture medium obtained in normoxic conditions, confirming the presence of a higher quantity of pCD. For example, in hypoxic conditions the cleavage rate was 0.72 (±0.04) mM/h after

72 h of incubation, which was approximately 1.5 fold higher than the cleavage rate in normoxic conditions at this time point. The increase in the initial rate of substrate hydrolysis may be due either to an increase in the quantity of pCD active form in the extracellular culture medium, and/or to the presence of an activator or the absence of an inhibitor in these media. However, the results of the Western blot analysis, which is a semi-quantitative method, confirmed that the quantities of pCD increased over culture time and that the hypoxic conditions induced a high level of secretion of pCD after 72 h of incubation. This observation reinforces the first hypothesis concerning the increase in the quantity of active pCD. Moreover, pCD activity did not correlate with the number of cells obtained after culture. In fact, although culture in hypoxic conditions resulted in a lower number of cells (see Fig. 1), the quantity of pCD seemed to be higher in conditioned culture medium (see Fig. 2). This observation strongly suggests that hypoxia induced an over-production of pCD in the extracellular medium. This over-production may be due to many factors such as mRNA over-expression and/or up-regulation of translation and/or hypersecretion of the pCD form in the extracellular medium. Concerning the hypothesis of mRNA over-expression, Bando et al. reported the up-regulation of about 600 genes out of 12 625 genes tested in MCF-7 cells under hypoxia [27]. However, the CD gene was not mentioned as one of the up-regulated genes in this work. Other studies have shown that hypoxia-inducible factor (HIF-1),

O. Achour et al. / Biochimie 121 (2016) 123e130

which plays a crucial role in the regulation of genes involved in the hypoxia adaptive pathway [28], induces the up-regulation of certain proteases, including CD, in a human colon carcinoma cell line, HCT116 [29], and a human prostatic carcinoma [30] transfected by a recombinant plasmid containing HIF-1 genes. On the other hand, hypoxia inhibited the secretion of CD by the GH4C1 Pituitary Adenoma Cell Line [31]. In pituitary adenoma in vivo, the expression of CD was independent of HIF-1 factor expression [32]. Hypoxia is a stress that modifies cell metabolism, including acid-extrusion by active transport phenomena [33]. The pH of the conditioned culture medium was measured and it was found that hypoxia resulted in a slightly higher acidification of the medium than normoxia (after 72 h of incubation, the pH values were 7.28 (±0.05) and 7.17 (±0.02) for the normoxic conditioned culture medium and the hypoxic conditioned culture medium, respectively). This acidification may be one reason for the pCD hypersecretion. 3.2. Effect of acidification on CD trafficking in breast cancer We next investigated the effect of acidity on the secretion of CD by MCF-7 cells. MCF-7 cells were cultivated in media acidified to pH 7.5, 6.5, 6 and 5.6 using MES buffer. It has been shown that the pH of the tumor microenvironment is lower than what is found in normal tissue and can be as low as 5.6 in some cases [34]. Fig. 4 presents the effects of the extracellular pH on MCF-7 cell growth and viability. At pH 7.5 the number of cells per mL increased progressively over time to reach 2.15  106 (±5  104) cell/mL after 72 h. When the culture medium pH was decreased, the number of viable cells per mL decreased. After 72 h of culture, the number of viable cells per mL was 1.46  106 (±10  104) cells/mL at pH 6.5 and only reached 6.6  104 (±2.8  104) cells/mL. On the other hand, the number of viable cells continued to increase after 72 h at pH 7.5, remained stable at pH 6.5 and decreased at pH 6 and 5.6. These data allow us to conclude that the MCF-7 breast cell line proliferates better in media at neutral pH and that the acidification of the medium reduced cell growth and resistance. To investigate the release of extracellular CD at acidic pH, Western blot analyses were performed on the acidified conditioned media using a CD-specific antibody (Fig. 5). Only the pCD form was detected in media obtained after culture at pH 7.5 and pH 6.5, with a band intensity that increased with incubation time. Interestingly, a certain amount of the mature form of CD was detected in the media at a more acidic pH (6 and 5.6). The

127

intensity of the bands increased with time, except at pH 5.6, which could be explained by the low number of viable cells (see Fig. 4). The presence of mature CD in the extracellular medium may be due to cell lysis, which releases the intracellular content into the extracellular media. However, in addition to the fact that observations under a microscope did not reveal the presence of cell debris in the medium (data not shown), we noted the absence of both the 48 kDa mono-chain CD that was observed in the cell lysate control, and b-actin, which is one of the most abundant proteins in the cytoskeleton. Moreover, there was no correlation between the intensity of the bands obtained in the Western blot analysis and the total number of cells obtained with each condition. All of this information would tend to exclude the hypothesis of mature CD liberation after cell lysis. Moreover, some studies previously described the controlled release of lysosomal mature CD under specific conditions [35e37]. In certain cases, conventional lysosomes can acquire the machinery for regulated exocytosis, and there is evidence that they can, in many cell types, respond to stimuli such as a rise in intracellular free Ca2þ concentration by fusing with the plasma membrane [36,37]. Furthermore, the acidification of the culture medium has been shown to affect the size, the number and the localization of lysosomes in some tumor cell lines, including breast cancer cell lines. In tumor cells exposed to an acidified extracellular medium, the lysosomes were displaced to the cell periphery by extracellular acidosis and this might facilitate lysosomal exocytosis, increasing the release of lysosomal enzymes, as was demonstrated for mature cathepsin B [38,39]. We then assessed the proteolytic activity of CD released into the conditioned media at different pH (Fig. 6). The amount of product was measured as the slope of the linear segment of the kinetics plot of the product. The activity of CD released into the conditioned culture medium was measured using a fluorogenic peptide at pH 3.7, which is the optimal pH for CD activity. We focused on the fact that only 5 mL of conditioned medium were added to each reaction well (100 mL final volume). At this ratio, the pH of the conditioned medium did not affect the pH of the final reaction media. Pepstatin A was systematically used for all conditions in order to confirm the aspartic protease activity. Fluorescence intensity was monitored over time and the product concentration was calculated using a standard curve obtained with pure MCA, which is the fluorescent molecule used for labeling the peptide substrate. Fig. 6 shows the initial rate of production of the fluorogenic peptide hydrolysis product with all the released CD

Fig. 4. Effects of the acidification of the culture medium on MCF-7 cell number (A) and viability (B). MCF-7 cells were cultured in medium buffered with MES at pH 7.5, 6.5, 6 and 5.6 from 24 to 72 h. Adherent cells were removed by trypsin treatment and counted with a hemocytometer. The percentage of viable cells was determined by trypan blue exclusion. The error bar represents the standard deviation of at least one separate triplicate run. Statistical significance was assessed by the one way ANOVA test (n ¼ 3). The asterisk indicates that p is less than 0.05 and the double asterisk denotes that p is less than 0.01.

128

O. Achour et al. / Biochimie 121 (2016) 123e130

Fig. 5. Western blot analysis of conditioned culture media obtained after cultivation of the MCF-7 cells at pH 7.5, 6.5, 6 and 5.6 for 24 h, 48 h or 72 h. (A), MCF-7 whole cell lysate (10 mg of total protein); (B), MCF-7 conditioned medium (7 days) (5 ng of cathepsin D).

conclusions concerning the amount of active CD since the different forms of CD that are present in the conditioned culture medium are characterized by different levels of activities. In fact, mature CD is known to be more active than pCD [18]. 3.3. Characterization of proteolytic activity of extracellular pCD

Fig. 6. Rate of product production measured in the conditioned culture medium of MCF-7 cells cultured at different extracellular pH for the indicated times (24, 48 or 72 h). The error bar represents the standard deviation of at least three values obtained from separate conditioned culture media. Statistical significance was assessed by the one way ANOVA test (n ¼ 3). The asterisk indicates that p is less than 0.05 and the double asterisk denotes that p is less than 0.01.

forms in conditioned culture media. At physiological pH 7.5, CD proteolytic activity increased progressively with time due to the accumulation of extracellular pCD protein. At acidic pH (6.5, 6 and 5.6), CD proteolytic activity was roughly stable over time and seemed to be independent of the pH of the culture medium. Indeed, there was no significant difference between the rates measured in conditioned culture medium obtained after cultivating MCF-7 at pH 6.5, 6 or 5.6. Interestingly, for the medium taken after 24 h, the rates were higher in low pH media (pH 6.5, 6, and 5.6) than for physiological pH media (pH 7.5). Nevertheless, we were not able to draw

In a third part, we found interesting to study some element of the enzyme kinetic of the pCD form secreted under hypoxic condition. In fact, we observed, while assessing the activity of this form (see Fig. 3) that the product appeared after a latency period that seemed to be the same at all kinetics obtained with the tested culture conditioned media. This latency period may correspond to the pre-steady-state i.e. before the enzymeesubstrate complex is formed at the same rate that it decomposes. Another possibility is that this latency corresponds to the auto-activation period of pCD when it is produced in an inactive 52 kDa form. In order to verify if the pCD present in the reaction media was active or if it was activated after an acidification at pH 3.7, the conditioned reaction medium was incubated at pH 3.7 for one hour prior to the addition of substrate, which has been shown to lead to a complete autoactivation of pCD [17,18]. An example of the change in the products during the reaction catalyzed by 72 h hypoxia-conditioned culture media, pre-activated or not, is presented in Fig. 7A. Preactivated commercial 52 kDa pCD was also used and the results are presented in the same figure. Apart from a slight increase in the reaction rate with the pre-activated conditioned culture medium, a similar latency period was observed whether the reaction was catalyzed by pre-incubated conditioned culture medium or not. Moreover, the same latency period was observed with commercial recombinant pCD that was pre-activated by incubation at pH 3.7 for one hour. These data support the hypothesis that the latency period does not reflect a pre-activation period but almost certainly the pre-steady-state period, which seems to be exceptionally long for

O. Achour et al. / Biochimie 121 (2016) 123e130

129

Fig. 7. Evolution over time in the concentration of the hydrolysis product of the fluorogenic peptide, using pre-activated or standard conditioned culture media, at pH 3.7 (A), and the effect of reaction pH on pCD activity (B). The error bar represents the standard deviation of at least three values obtained from separate conditioned culture media.

MCF-7 Breast Cancer Cell

Hypoxic condi ons of culture

CDsc (48 kDa)

CDdc (34 + 14 kDa)

Endosome

Lysosome

Auto-ac va on

Over-secre on of pCD pCD (52 kDa)

Extracellular acidic condi ons of culture Secre on

Intracellular normal pathway Over-Secre on

CDdc (34 + 14 kDa) AcƟve

Altera on of mature CD trafficking

pCD (52 kDa)

AcƟve ?

ppCD (54 kDa)

Golgi apparatus

Endoplasmic reƟculum

pCD (51 kDa) Nucleus

AcƟve Cell membrane

Fig. 8. Schematic representation of CD trafficking in MCF-7 cells cultivated under abnormal conditions of breast cancer microenvironment. ppCD: pre-pro-cathepsin D (54 kDa); pCD: pro-cathepsin D (52 kDa); CDsc: single chain cathepsin D (48 kDa); CDdc: double chain cathepsin D (34 þ 14 kDa).

this enzyme. This latency period was observed in many previous studies with the pro-form of CD [18,40]. The slight increase in the reaction rate may reflect the activation of a certain quantity of enzyme but most of the pCD seems to be active already. This observation suggests that the secreted pCD, which is a 52 kDa form, is secreted in either an active form or that it can be activated even if the cells are placed at a neutral physiological pH. In fact, studies that found an inactive 52 kDa pCD were performed using hemoglobin as the substrate [41e43]; hemoglobin is a large protein that cannot access the active site of the 52 kDa pCD, which is characterized by the presence of a pro-peptide that probably obstructs the active site. More investigations will be needed to determine whether the pCD is secreted in an active form or an inactive form which is subsequently activated in the extracellular media. In the latter case, the activation mechanism could either be through auto-activation by intra-molecular proteolysis or by another mechanism. Nevertheless, our data may provide new arguments in favor of the role of pCD in the proteolysis and degradation of extracellular matrix proteins in the TME. The activity of pCD was demonstrated at pH 3.7, which has been described as the optimal pH for CD activity. However, this pH value is lower than the reported pH of TME, which lies between 6.8 and 5.6 [34]. For this reason, we investigated the proteolytic activity of pCD released from MCF-7 cells at a higher pH. Fig. 7B shows the effect of the pH of the reaction medium on the

rate of substrate cleavage by pCD. As expected, the highest activity was observed at pH 3.7 and this decreased when the reaction medium pH was less acidic. The activity was completely lost at pH 6. However, we and others have previously demonstrated that pCD is proteolytically active at a higher pH (6.8) in specific heterogenic conditions such as in the presence of phospholipid micelles [17,18,44]. 4. Conclusions In this work, we investigated the effects of hypoxia and extracellular acidification on the secretion (release) of CD by MCF-7 breast cancer cells (see Fig. 8). Hypoxia increased the level of pCD detected in the extracellular culture medium compared to normoxic conditions. On the other hand, when MCF-7 cells are cultivated in acidic conditions, they are able to release the mature form of CD in the extracellular media, whereas the amount of extracellular pCD seems to be reduced. These new data provide a better understanding of the ‘environmental’ conditions that may affect CD secretion in the tumor microenvironment and provide novel arguments in favor of the role of CD on the proteolysis and remodeling of extracellular matrix proteins. In the context of the development of new targeting strategies for tumor diagnosis and active drug delivery, these new results should be taken into consideration and suggest that all the form of CD can be used to target of breast tumor microenvironment.

130

O. Achour et al. / Biochimie 121 (2016) 123e130

Acknowledgments This study was supported by a PhD grant from the Ligue Contre le Cancer of Charente-Maritime for O. Achour (2011e2014). The ^le de Ligue Contre le Cancer of Charente-Maritime and the PRES (Po rieur) of Limousin PoitouRecherche et d'Enseignement Supe Charentes are acknowledged for financial support through the research project “Conception et validations de nanovecteurs in double fonctionnalite  pour la the rapeutique et le telligents a diagnostic des tumeurs”. References [1] M.A. Swartz, N. Iida, E.W. Roberts, S. Sangaletti, M.H. Wong, F.E. Yull, et al., Tumor microenvironment complexity: emerging roles in cancer therapy, Cancer Res. 72 (2012) 2473e2480, http://dx.doi.org/10.1158/0008-5472.CAN12-0122. [2] B.G. Wouters, S.A. Weppler, M. Koritzinsky, W. Landuyt, S. Nuyts, J. Theys, et al., Hypoxia as a target for combined modality treatments, Eur. J. Cancer Oxf. Engl. 38 (2002) 240e257, 1990. [3] J. Li, R. Ayene, K.M. Ward, E. Dayanandam, I.S. Ayene, Glucose deprivation increases nuclear DNA repair protein Ku and resistance to radiation induced oxidative stress in human cancer cells, Cell Biochem. Funct. 27 (2009) 93e101, http://dx.doi.org/10.1002/cbf.1541. [4] M.F. McCarty, J. Whitaker, Manipulating tumor acidification as a cancer treatment strategy, Altern. Med. Rev. J. Clin. Ther. 15 (2010) 264e272. [5] F. Fan, A. Schimming, D. Jaeger, K. Podar, Targeting the tumor microenvironment: focus on angiogenesis, J. Oncol. 2012 (2012) 281261, http://dx.doi.org/ 10.1155/2012/281261. [6] A. Minarowska, M. Gacko, A. Karwowska, Ł. Minarowski, Human cathepsin D, Folia Histochem. Cytobiol. Pol. Acad. Sci. Pol. Histochem. Cytochem. Soc. 46 (2008) 23e38, http://dx.doi.org/10.2478/v10042-008-0003-x. [7] S.C. Fortenberry, J.S. Schorey, J.M. Chirgwin, Role of glycosylation in the expression of human procathepsin D, J. Cell Sci. 108 (Pt 5) (1995) 2001e2006. [8] A. Hasilik, E.F. Neufeld, Biosynthesis of lysosomal enzymes in fibroblasts. Phosphorylation of mannose residues, J. Biol. Chem. 255 (1980) 4946e4950. ~o, I. Sakwa, S. Tiede, et al., [9] K. Kollmann, S. Pohl, K. Marschner, M. Encarnaça Mannose phosphorylation in health and disease, Eur. J. Cell Biol. 89 (2010) 117e123, http://dx.doi.org/10.1016/j.ejcb.2009.10.008. [10] N. Zaidi, A. Maurer, S. Nieke, H. Kalbacher, Cathepsin D: a cellular roadmap, Biochem. Biophys. Res. Commun. 376 (2008) 5e9, http://dx.doi.org/10.1016/ j.bbrc.2008.08.099. [11] P. Benes, V. Vetvicka, M. Fusek, Cathepsin Demany functions of one aspartic protease, Crit. Rev. Oncol. Hematol. 68 (2008) 12e28, http://dx.doi.org/ 10.1016/j.critrevonc.2008.02.008. [12] S. Aziz, S. Pervez, S. Khan, N. Kayani, M. Rahbar, Immunohistochemical cathepsin-D expression in breast cancer: correlation with established pathological parameters and survival, Pathol. Res. Pract. 197 (2001) 551e557, http://dx.doi.org/10.1078/0344-0338-00126. taye , C. Millet, D. Margerit, P. Ingrand, J.M. Goujon, et al., [13] J.L. Kraimps, T. Me Cathepsin D in normal and neoplastic thyroid tissues, Surgery 118 (1995) 1036e1040. [14] V. Vetvicka, J. Vetvickova, M. Fusek, Effect of procathepsin D and its activation peptide on prostate cancer cells, Cancer Lett. 129 (1998) 55e59. [15] G. Nicotra, R. Castino, C. Follo, C. Peracchio, G. Valente, C. Isidoro, The dilemma: does tissue expression of cathepsin D reflect tumor malignancy? The question: does the assay truly mirror cathepsin D mis-function in the tumor? Cancer Biomark. 7 (2010) 47e64, http://dx.doi.org/10.3233/CBM2010-0143. [16] M.M. Gopalakrishnan, H.-W. Grosch, S. Locatelli-Hoops, N. Werth, E. Smolenov a, M. Nettersheim, et al., Purified recombinant human prosaposin forms oligomers that bind procathepsin D and affect its autoactivation, Biochem. J. 383 (2004) 507e515, http://dx.doi.org/10.1042/BJ20040175. bois, M. Gary[17] V. Laurent-Matha, P.F. Huesgen, O. Masson, D. Derocq, C. Pre Bobo, et al., Proteolysis of cystatin C by cathepsin D in the breast cancer microenvironment, FASEB J. Off. Publ. Fed. Am. Soc. Exp. Biol. 26 (2012) 5172e5181, http://dx.doi.org/10.1096/fj.12-205229. [18] O. Achour, N. Bridiau, M. Kacem, R. Delatouche, S. Bordenave-Juchereau, F. Sannier, et al., Cathepsin D activity and selectivity in the acidic conditions of a tumor microenvironment: utilization in the development of a novel cathepsin D substrate for simultaneous cancer diagnosis and therapy, Biochimie 95 (2013) 2010e2017, http://dx.doi.org/10.1016/j.biochi.2013.07.010. [19] M. Cohen, I. Fruitier-Arnaudin, R. Sauvan, D. Birnbaum, J.-M. Piot, Serum levels of Hemorphin-7 peptides in patients with breast cancer, Clin. Chim. Acta 337 (2003) 59e67, http://dx.doi.org/10.1016/j.cccn.2003.07.011.

[20] O. Schestkowa, D. Geisel, R. Jacob, A. Hasilik, The catalytically inactive precursor of cathepsin D induces apoptosis in human fibroblasts and HeLa cells, J. Cell. Biochem. 101 (2007) 1558e1566, http://dx.doi.org/10.1002/jcb.21269. [21] A.E. Greijer, E. van der Wall, The role of hypoxia inducible factor 1 (HIF-1) in hypoxia induced apoptosis, J. Clin. Pathol. 57 (2004) 1009e1014, http:// dx.doi.org/10.1136/jcp.2003.015032. [22] C.-N. Lee, W.-F. Cheng, M.-C. Chang, Y.-N. Su, C.-A. Chen, F.-J. Hsieh, Hypoxiainduced apoptosis in endothelial cells and embryonic stem cells, Apoptosis Int. J. Program. Cell Death 10 (2005) 887e894, http://dx.doi.org/10.1007/ s10495-005-2946-0. [23] G.R. Richo, G.E. Conner, Structural requirements of procathepsin D activation and maturation, J. Biol. Chem. 269 (1994) 14806e14812. [24] C.G. Knight, A.J. Barrett, Interaction of human cathepsin D with the inhibitor pepstatin, Biochem. J. 155 (1976) 117e125. [25] Y. Yasuda, T. Kageyama, A. Akamine, M. Shibata, E. Kominami, Y. Uchiyama, et al., Characterization of new fluorogenic substrates for the rapid and sensitive assay of cathepsin E and cathepsin D, J. Biochem. (Tokyo) 125 (1999) 1137e1143. [26] N. Zaidi, H. Kalbacher, Cathepsin E: a mini review, Biochem. Biophys. Res. Commun. 367 (2008) 517e522, http://dx.doi.org/10.1016/j.bbrc.2007.12.163. [27] H. Bando, M. Toi, K. Kitada, M. Koike, Genes commonly upregulated by hypoxia in human breast cancer cells MCF-7 and MDA-MB-231, Biomed. Pharmacother. 57 (2003) 333e340. [28] T. Kodama, N. Shimizu, N. Yoshikawa, Y. Makino, R. Ouchida, K. Okamoto, et al., Role of the glucocorticoid receptor for regulation of hypoxia-dependent gene expression, J. Biol. Chem. 278 (2003) 33384e33391, http://dx.doi.org/ 10.1074/jbc.M302581200. [29] B. Krishnamachary, S. Berg-Dixon, B. Kelly, F. Agani, D. Feldser, G. Ferreira, et al., Regulation of colon carcinoma cell invasion by hypoxia-inducible factor 1, Cancer Res. 63 (2003) 1138e1143. [30] Y. Luo, D.-L. He, L. Ning, S.-L. Shen, L. Li, X. Li, et al., Over-expression of hypoxia-inducible factor-1a increases the invasive potency of LNCaP cells in vitro, BJU Int. 98 (2006) 1315e1319, http://dx.doi.org/10.1111/j.1464410X.2006.06480.x.  pez-Barrera, G.M. De La Escalera, C. Clapp, Hyp[31] G. Cosío, M.C. Jeziorski, F. Lo oxia inhibits expression of prolactin and secretion of cathepsin-D by the GH4C1 pituitary adenoma cell line, Lab. Investig. J. Tech. Methods Pathol. 83 (2003) 1627e1636. [32] D. Yoshida, K. Kim, M. Yamazaki, A. Teramoto, Expression of hypoxiainducible factor 1alpha and cathepsin D in pituitary adenomas, Endocr. Pathol. 16 (2005) 123e131. [33] A. Hulikova, A.L. Harris, R.D. Vaughan-Jones, P. Swietach, Regulation of intracellular pH in cancer cell lines under normoxia and hypoxia, J. Cell. Physiol. 228 (2013) 743e752, http://dx.doi.org/10.1002/jcp.24221. [34] K. Engin, D.B. Leeper, J.R. Cater, A.J. Thistlethwaite, L. Tupchong, J.D. McFarlane, Extracellular pH distribution in human tumours, Int. J. Hyperthermia. Off. J. Eur. Soc. Hyperthermic Oncol. North Am. Hyperth. Group 11 (1995) 211e216, http://dx.doi.org/10.3109/02656739509022457. [35] I. Shafat, I. Vlodavsky, N. Ilan, Characterization of mechanisms involved in secretion of active heparanase, J. Biol. Chem. 281 (2006) 23804e23811, http:// dx.doi.org/10.1074/jbc.M602762200. [36] N.W. Andrews, Regulated secretion of conventional lysosomes, Trends Cell Biol. 10 (2000) 316e321, http://dx.doi.org/10.1016/S0962-8924(00)01794-3. [37] A. Rodríguez, P. Webster, J. Ortego, N.W. Andrews, Lysosomes Behave as Ca2þ-regulated exocytic vesicles in fibroblasts and epithelial cells, J. Cell Biol. 137 (1997) 93e104, http://dx.doi.org/10.1083/jcb.137.1.93. [38] K. Glunde, S.E. Guggino, M. Solaiyappan, A.P. Pathak, Y. Ichikawa, Z.M. Bhujwalla, extracellular acidification alters lysosomal trafficking in human breast cancer cells, Neoplasia N. Y. N. 5 (2003) 533e545. [39] J. Rozhin, M. Sameni, G. Ziegler, B.F. Sloane, Pericellular pH affects distribution and secretion of cathepsin B in malignant cells, Cancer Res. 54 (1994) 6517e6525. [40] M. Beckman, C. Freeman, C.R. Parish, D.H. Small, Activation of cathepsin D by glycosaminoglycans, FEBS J. 276 (2009) 7343e7352, http://dx.doi.org/ 10.1111/j.1742-4658.2009.07444.x. [41] F. Capony, M. Morisset, A.J. Barrett, J.P. Capony, P. Broquet, F. Vignon, et al., Phosphorylation, glycosylation, and proteolytic activity of the 52-kD estrogen-induced protein secreted by MCF7 cells, J. Cell Biol. 104 (1987) 253e262, http://dx.doi.org/10.1083/jcb.104.2.253. [42] M. Morisset, F. Capony, H. Rochefort, Processing and estrogen regulation of the 52-kilodalton protein inside MCF7 breast cancer cells, Endocrinology 119 (1986) 2773e2782, http://dx.doi.org/10.1210/endo-119-6-2773. [43] H. Rochefort, F. Capony, P. Augereau, V. Cavailles, M. Garcia, M. Morisset, et al., The estrogen-regulated 52K-cathepsin-D in breast cancer: from biology to clinical applications, Int. J. Rad. Appl. Instrum. B 14 (1987) 377e384, http:// dx.doi.org/10.1016/0883-2897(87)90015-8. [44] S. Watabe, N. Yago, Phospholipids activate cathepsin D, Biochem. Biophys. Res. Commun. 110 (1983) 934e939.

Alteration of cathepsin D trafficking induced by hypoxia and extracellular acidification in MCF-7 breast cancer cells.

The microenvironment that surrounds tumor cells is characterized by hypoxic conditions and extracellular acidity. These hostile conditions induce cruc...
565B Sizes 0 Downloads 9 Views