ACTH ELISA 专题出版物焦点：压力的气味阻断

2021 年 6 月 14 日

MD Bioproducts

<压力气味阻断：为了应对恐惧和压力，下丘脑促肾上腺皮质激素释放激素神经元会诱导血压激素增加。

Eun Jeong Lee、Luis R.Saraiva、Naresh K.Hanchate、Xiaolan Ye、Gregory Asher、Jonathan Ho、Linda B. Buck

神经科学 Pub 日期：2021-02-25几千年来，气味一直被用来缓解恐惧和压力，但人​​们对它们是否具有这种作用却知之甚少。为了应对恐惧和压力，下丘脑促肾上腺皮质激素释放激素神经元（CRHN）会诱导血液应激激素增加。在这里，我们发现某些结构和感知上不同的气味可以阻止小鼠应激激素对三种潜在压力源的反应：身体束缚、捕食者气味和男性与男性的社会对抗。两种气味剂都会激活下丘脑腹内侧核 (VMH) 中 CRHN 突触前的 GABA 能抑制神经元。刺激这些神经元会抑制约束引起的 CRHN 激活和应激激素增加，类似于阻断气味剂。相反，它们的沉默可以防止气味剂阻断这两种反应。值得注意的是，我们还观察到终纹床核（BNST）中 CRHN 突触前神经元应激源激活的气味阻断。总之，这些发现表明，选定的气味剂确实可以阻断应激反应，并且气味阻断可以通过两种途径发生：阻断气味信号直接抑制 CRHN 的直接途径和抑制 CRHN 突触前神经元应激源激活的间接途径并防止它们向 CRHN 传输压力信号。

继续阅读并下载该出版物

 

分享此内容：较旧的帖子较新的帖子新！物种特异性 IL-17a 检测

2021 年 7 月 26 日

MD Bioproducts

为什么使用检测交叉反应性进行测试您的样品何时可以从物种特异性检测和物种特异性对照中受益？ MD Biosciences Bioproducts 很自豪地推出针对 IL-17a 的物种特异性检测。这些检测方法作为新 MonELISA 系列产品的一部分发布。 MonELISA 产品设计为使用相同的通用试剂按照相同的方案运行。试剂盒以特定于检测的成分试剂盒和足以容纳 5 个板的辅助试剂的单独试剂盒的形式提供。用户可以混合搭配任意数量的 MonELISA 试剂盒，从而受益于降低的检测成本、更小的冰箱空间和出色的检测性能。

犬、ELISA、猫、IL-17a、炎症ons 研究，鼠标，ST2，T1/ST2分享此内容：较旧的帖子较新的帖子Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript. AbstractProteases are involved in many aspects of tumor progression, including cell survival and proliferation, escape from immune surveillance, cell adhesion and migration, remodeling and invasion of the extracellular matrix. Several lysosomal cysteine proteases have been cloned and shown to be overexpressed in cancer; yet, despite the great potential for development of novel therapeutics, we still know little about the regulation of their proteolytic activity. Cystatins such as cystatin M are potent endogenous protein inhibitors of lysosomal cysteine proteases. Cystatin M is expressed in normal and premalignant human epithelial cells, but not in many cancer cell lines. Here, we examined the effects of cystatin M expression on malignant properties of human breast carcinoma MDA-MB-435S cells. Cystatin M was found to significantly reduce in vitro: cell proliferation, migration, Matrigel invasion, and adhesion to endothelial cells. Reduction of cell proliferation and adhesion to an endothelial cell monolayer were both independent of the inhibition of lysosomal cysteine proteases. In contrast, cell migration and matrix invasion seemed to rely on lysosomal cysteine proteases, as both recombinant cystatin M and E64 were able to block these processes. This study provides the first evidence that cystatin M may play important roles in safeguarding against human breast cancer. IntroductionLysosomal cysteine proteases are involved in many biological processes such as intracellular protein catabolism, pericellular matrix remodeling, antigen processing, virulence, and programmed cell death (for recent reviews, see Chapman et al., 1997; Villadangos et al., 1999; Bromme and Kaleta, 2002). Impaired regulation of expression and activity of lysosomal cysteine proteases has been implicated in cancer progression (Keppler and Sloane, 1996; Roshy et al., 2003). Several groups have hypothesized that the increased proteolytic activities observed in malignant tumors are due to upregulation of proteases and/or downregulation of their endogenous inhibitors (Liotta and Stetler-Stevenson, 1991; Brunner et al., 1994; DeClerck and Imren, 1994). Cystatins are considered physiological inhibitors of lysosomal cysteine proteases (Turk and Bode, 1991; Abrahamson, 1994). They control the catalytic function of target proteases by forming reversible, high-affinity complexes (Barrett et al., 1986; Alvarez-Fernandez et al., 1999).Cystatin M was first identified by differential display as a transcript that was downregulated in a metastatic breast cell line when compared to a matched primary tumor cell line (Sotiropoulou et al., 1997). The same inhibitor was independently cloned from embryonic lung fibroblasts and named cystatin E (Ni et al., 1997). Here, we will refer to cystatin E/M as cystatin M. Cystatin M is a cell-secreted cystatin of 121 amino acids that exists in both glycosylated (17鈥塳Da) and unglycosylated (14.4鈥塳Da) forms (Ni et al., 1997; Sotiropoulou et al., 1997). Since cystatin M is expressed in normal and premalignant breast epithelial cells but not in metastatic breast cancer cell lines, it was hypothesized to be a novel metastasis suppressor (Sotiropoulou et al., 1997).Metastasis is a multi-step process in which tumor cells escape from the primary tumor and establish colonies at local and distant sites (Poste and Fidler, 1980). Metastatic tumor cells must detach, acquire motile function, and penetrate the basement membranes and other connective tissues. Thus, degradation of these supportive and confining structures has been hypothesized to be a critical component of the metastatic process (Liotta and Stetler-Stevenson, 1991). Studies have shown that lysosomal cysteine proteases can degrade components of connective tissues and basement membranes in vitro, including various collagens, elastin, tenascin, laminin, and fibronectin (Maciewicz et al., 1990; Buck et al., 1992; Guinec et al., 1993; Bromme et al., 1996; Sameni et al., 2000; Mai et al., 2002). Secretory cystatins, like cystatin M, could potentially regulate the function of exocytosed cysteine proteases (Linebaugh et al., 1999; Hulkower et al., 2000). Loss of expression of cystatin M could therefore lead to proteolysis of the glycoprotein scaffolds that maintain tissue architecture, and this could facilitate invasion and metastasis of cancer cells.The long-term goal of our studies is to determine if cystatin M is a bona fide metastasis suppressor. As a first step towards that aim, we have stably transfected the highly tumorigenic and metastatic human breast cancer MDA-MB-435S cell line with a cystatin M expression vector and studied the effects of cystatin M expression on the malignant properties of these cells. We report for the first time that ectopic expression of cystatin M reduced cell proliferation, migration, matrix invasion, and tumor-endothelial cell adhesion. Surprisingly, however, when mock-transfected cells were treated with a broad-spectrum inhibitor of lysosomal cysteine proteases (E64), no effect was seen on cell proliferation and tumor-endothelial cell adhesion, while the same agent inhibited both cell migration and matrix invasion. We conclude from these studies that cystatin M may have distinct functions in the safeguard against malignant progression.ResultsExpression pattern of cystatins C and M in human cancer cell linesCystatin M expression has been analysed before in various normal human tissues as well as in neoplastic skin disorders and established breast cancer cell lines (Ni et al., 1997; Sotiropoulou et al., 1997; Zeeuwen et al., 2002). Little is known about its expression pattern in other cancers. Previously, we had shown that another secreted cystatin, cystatin C, was quite ubiquitously expressed in human cancer cells (Keppler et al., 1994). Therefore, we compared by semiquantitative RT鈥揚CR analysis the levels of expression of cystatin C and cystatin M mRNAs in a number of different human cancer cell lines (breast: BT-20, BT-549, and MDA-MB-435S; prostate: DU-145, LNCaP, and PC-3; colon: CaCo-2 and HT-29; glioma: U-87MG; and melanoma: WM-164, WM-793, WM-852, WM-902B, WM-1205Lu, and WM-1341D). The diploid, spontaneously immortalized and nontumorigenic, human breast epithelial MCF-10A cell line served as a 鈥榥ormal鈥?counterpart in this screen. The cystatin M mRNA was detected only in a few cell lines, including immortalized MCF-10A and three cancer cell lines (BT-20, DU-145, and PC-3). Low levels of expression were also detectable in BT-549, LNCaP, CaCo-2, HT-29, and U-87MG cells. As for most established human breast cancer cell lines such as MDA-MB-435S (Sotiropoulou et al., 1997), little or no cystatin M transcript could be detected in any of the melanoma cell lines used in this study (Figure 1, cyst. M). In contrast to the cystatin M mRNA, the cystatin C mRNA was detected in all cell lines analysed, with the exception of one, BT-20 (Figure 1, cyst. C). RT鈥揚CR analysis of the 尾2-microglobulin mRNA was performed to verify equal sampling of total RNA from one cell line to another (Figure 1, 尾2-micro).Figure 1Standard RT鈥揚CR analysis of cystatin M and cystatin C expression in human cell lines. Total RNA was isolated from human cell lines and subsequently reverse-transcribed and analysed by PCR amplification for expression of cystatin M (upper panel), cystatin C (middle panel) and, as a loading control, 尾2-microglobulin (lower panel). Cell lines are as follows: breast (MCF-10A, BT-20, BT-549, and MDA-MB-435S), prostate (DU-145, LNCaP, and PC-3), colon (CaCo-2 and HT-29), glioblastoma (U-87MG), and melanoma (WM-164鈥揥M-1341D)Full size imageTransfection studiesMDA-MB-435S cells express cystatin C, but little or no cystatin M message (Figure 1, compare MDA-435S in the middle and upper panels, respectively). These cells are highly tumorigenic, invasive, angiogenic, and metastatic in immunocompromised mice (Price and Zhang, 1990). Therefore, they seemed an appropriate model in which to test the antiproliferative, anti-invasive, antiangiogenic, and/or antimetastatic properties of cystatin M. As a first step towards that aim, we transfected MDA-MB-435S cells with the cystatin M cDNA expression vector or the 鈥榚mpty鈥?control vector. Like the cystatin M vector, the 鈥榚mpty鈥?control vector has an intact CMV-promoter, a transcriptional start site, followed some 300鈥塨p downstream by a patent polyadenylation signal. PCR analysis of genomic DNA isolated from individual Zeocin-resistant clones showed stable integration of the cystatin M cDNA in cystatin M-transfected clones (Figure 2a, CM-12, CM-13, CM-15, and CM-17). The 422-bp PCR product could not be detected in parental (untransfected) or two mock-transfected control clones (Figure 2a; parental, mock-1, and mock-2). The bands of higher sizes ( 422鈥塨p), seen only with genomic DNA containing the transfected cystatin M cDNA, most likely represent oligomers of the amplified product, as they are not seen in control DNA.Figure 2Characterization of MDA-MB-435S transfectants. The cystatin M expression plasmid or the empty vector were transfected into MDA-MB-435S cells. (a) Several zeocin-resistant clones were isolated and characterized by PCR for stable integration of the cystatin M cDNA into genomic DNA. The expected 422-bp PCR product was detected only in genomic DNA isolated from cystatin M clones (CM-12, CM-13, CM-15, and CM-17), but not from control clones (parental, mock-1, and mock-2). (b) The same clones were analysed by immunoblotting for constitutive secretion of cystatin M (upper panel) and cystatin C (lower panel). Clones CM-12, CM-13, CM-15, and CM-17 expressed and secreted two forms of cystatin M, that is, the 14.4-kDa unglycosylated and the 17-kDa N-glycosylated form (Cyst. M, upper panel). The parental, mock-1, and mock-2 control clones produced no detectable cystatin M in the media. As additional control, all clones were shown to secrete similar levels of cystatin C (Cyst. C, lower panel)Full size imageImmunoblot analysis of cell-conditioned media showed that clones CM-12, CM-13, CM-15, and CM-17 expressed and secreted two forms of cystatin M (Figure 2b, upper panel, cyst. M). Two forms of cystatin M, that is, a 17-kDa glycosylated and a 14.4-kDa unglycosylated form, have been identified previously (Ni et al., 1997; Sotiropoulou et al., 1997). Cystatin M protein could not be detected in the conditioned media of control clones (Figure 2b, upper panel, parental, mock-1, mock-2). Cystatin C is a closely related secreted inhibitor that has a similar molecular mass to cystatin M and may have a partially overlapping inhibitory profile (Turk and Bode, 1991; Abrahamson, 1994). Immunoblot analysis revealed that ectopic expression of cystatin M did not affect expression and secretion of cystatin C (Figure 2b, lower panel, cyst. C), suggesting that there was not a compensatory downregulation of cystatin C in cells that overexpress cystatin M.Cystatin M reduced cell proliferation in vitroSeveral studies have shown that cystatin C from different species can modulate cell proliferation (Sun, 1989; Leung-Tack et al., 1990; Tavera et al., 1992; Taupin et al., 2000; Konduri et al., 2002). Therefore, we determined the effect of cystatin M on proliferation of MDA-MB-435S cells. We observed a significantly (P 0.01) slower rate of proliferation of the cystatin M transfectants as compared to that of the mock transfectants (Figure 3a). If secreted cystatin M was responsible for the growth differences in these cells, then conditioned media from the cystatin M transfectants should also slow the growth of mock-transfected cells. This was indeed the case, as cystatin M-conditioned media significantly (P 0.001) slowed the rate of proliferation of mock-transfected cells when compared to the same cells grown in the presence of mock-conditioned media (Figure 3b). Our results are thus consistent with reduced proliferation in the cystatin M 鈥搕ransfectants being due to secreted cystatin M.Figure 3Effect of cystatin M on proliferation of MDA-MB-435S cells. (a) Comparison of growth rates between MDA-MB-435S cell clones stably transfected with either the empty vector (mock-1 and mock-2) or the cystatin M expression vector (CM-12, CM-13, and CM-17). Cells (500) of each clone were seeded in seven 96-well plates and fed every 48鈥塰 with DMEM+10% FBS+200鈥?i>渭g/ml zeocin. Plates were collected at days 1, 3, 5, 7, 9, 11, and 13, and cell number measured with the CyQuant proliferation kit. Each point represents the mean卤s.e.m. of six wells. The experiment was repeated several times with similar results. Statistical analysis comparing mock-transfected and cystatin M-transfected clones was performed using one-way ANOVA (*P猢?/span>0.01). (b) Effect of cell-conditioned media on proliferation of mock-1. Conditioned media (24鈥塰) were collected from parental and mock-transfected cells and pooled (mock medium); the same was done with the media from CM-12, 13 and 17 cells (CM medium). Mock-1 cells were then seeded as above and grown in either mock or CM medium (n=6). Cells were fed with the pooled media and plates collected and processed as described above. Statistical analysis comparing the effect of mock and CM media was performed using two-factor ANOVA (*P猢?/span>0.001). (c) Effect of recombinant GSTCM on cell proliferation. Mock-1 cells were seeded as above, and grown in the presence of various concentrations of recombinant GSTCM or GST (n=6). Cells were fed with fresh medium and recombinant protein every 48鈥塰, and plates collected and processed as described above. Statistical analysis comparing control-treated and GSTCM/GST-treated mock-1 cells was performed using one-way ANOVA. (d) Inhibition of papain activity by GSTCM. The ability of GSTCM to inhibit papain was tested using various concentrations of recombinant GSTCM or GST (as control). Active site titrated papain (20鈥塶M) was preincubated with GSTCM/GST for 20鈥塵in at room temperature. The enzyme was then incubated at 37掳C in the presence of 100鈥?i>渭 M of Z-Phe-Arg-AMC, and initial reaction velocities recorded semicontinuously. Residual enzyme activity is expressed as relative fluorescence units (RFU)/min. (e) Effect of a broad-spectrum inhibitor of lysosomal cysteine proteases on cell proliferation. Mock-1 cells were seeded as above and grown in the absence (control) or presence of an excess concentration (10鈥?i>渭 M) of the inhibitor E64 or its membrane-permeant derivative E64d (n=6). Cells were fed with fresh medium and inhibitor every 48鈥塰, and plates collected and processed as described above. Statistical analysis comparing control-treated and inhibitor-treated mock-1 cells was performed using two-factor ANOVAFull size imageThese results suggested that recombinant cystatin M should also reduce cell proliferation. We tested this hypothesis with a GST-fusion protein in which GST is fused to the N-terminus of cystatin M (GSTCM). Recombinant GST was used as a control. Neither recombinant protein had any effect on proliferation of the cells (Figure 3c). This was surprising because GSTCM expressed in bacteria inhibited papain with a similar nanomolar Ki (Figure 3d) as cystatin M expressed in insect Sf9 cells (Ni et al., 1997). The correct folding of GSTCM was further substantiated by the fact that GST had no apparent papain-inhibitory potential when tested under similar conditions (Figure 3d). In addition, the broad-spectrum inhibitor of lysosomal cysteine proteases, E64, or its membrane-permeant ester derivative, E64d, had no effect on cell proliferation (Figure 3e). Previously, we had established that the 10-渭 M dose of E64/E64d used was able to completely inhibit lysosomal cysteine proteases present in a subconfluent 100-mm Petri dish. Therefore, lysosomal cysteine proteases do not seem to be involved in proliferation of MDA-MB-435S cells. Additionally, it might suggest that either post-translational modifications or an accessible amino-terminus might be required for cystatin M to function as a regulator of cell proliferation.Cystatin M reduced cell migrationOne of the hallmarks of malignant cancers is their ability to invade/infiltrate the surrounding normal tissues. MDA-MB-435S cells are highly motile and invasive (Thompson et al., 1992; Liu et al., 1997). To test whether cell migration was affected by overexpression of cystatin M, we performed Transwell migration assays using FBS as the chemoattractant. During the initial optimization of this assay in our laboratory, we determined that 1% heat-inactivated FBS was maximally stimulatory for MDA-MB-435S cell migration (data not shown). We observed a time-dependent accumulation of migratory cells on the bottom surface of the 8-渭m filters. Little or no migration was seen in the absence of chemoattractant over a 13-h incubation period (data not shown). Migration of the cystatin M-overexpressing clone CM-13 was significantly (P 0.05) and 猢?/span>60% reduced in comparison to migration of the parental and mock-1 control clones (Figure 4a). Migration of the mock-1 clone was reduced 猢?/span>70% in the presence of either 200鈥塶M of recombinant GSTCM or 10鈥?i>渭 M of E64 (Figure 4b). This was in contrast to what we observed with cell proliferation, which was not sensitive to either 2鈥?i>渭 M of GSTCM or 10鈥?i>渭 M of E64. These assays indicate that FBS-stimulated migration of MDA-MB-435S cells required the participation of one or several lysosomal cysteine proteases inhibited by cystatin M and/or E64.Figure 4(a) Effect of cystatin M on cell migration. Parental, mock- and cystatin M-transfected cells (5 脳 104) in 500鈥?i>渭l of serum-free DMEM were seeded into Transwell culture inserts on top of 8-渭m filters. The bottom wells contained 750鈥?i>渭l of DMEM supplemented with 1% FBS as chemoattractant. Cells were incubated for various times at 37掳C in this gradient of FBS. All other conditions were as described in the Materials and methods section. Cells that had migrated to the bottom surface of the filter were counted on the whole surface using an inverted microscope at 脳 100 magnification. Results are expressed as mean卤s.e.m. of triplicate experiments for each clone (parental, mock-1, and CM-13) and for each time point (3, 6, 9, and 13鈥塰). Little or no migration was observed in the absence of the chemoattractant. Statistical analysis comparing migration of the cystatin M clone to that of parental and mock control clones was performed using one-way ANOVA (*P猢?/span>0.05). (b) Effect of recombinant cystatin M and E64 on cell migration. Mock-1 cells were incubated as described above, in the absence (Ctrl) or presence of 200鈥塶M of recombinant GST/GSTCM or 10鈥?i>渭 M of E64. Cell migration was assessed after 13鈥塰. *Indicates a highly significant difference with mock control and GST-treated mock cells (P猢?/span>0.001)Full size imageCystatin M blocked Matrigel invasionTo determine if cystatin M affected the ability of MDA-MB-435S cells to invade, we performed Matrigel invasion assays (Hendrix et al., 1987). At 60- and 72-h time points, invasion of CM-13 cells that overexpressed cystatin M was significantly lesser ( P 0.001) than invasion of the parental and mock-1 control clones (Figure 5a). Invasion of the mock-1 clone was nearly completely suppressed by a single addition of 200鈥塶M of recombinant GSTCM or 10鈥?i>渭 M of E64 to the media in the top well (Figure 5b). Our data therefore suggested that specific lysosomal cysteine protease(s) might play a rate-limiting role in serum-stimulated migration and invasion.Figure 5(a) Effect of cystatin M on Matrigel invasion. The experimental set up was the same as in the legend to Figure 4, except that (1) Matrigel-coated filter inserts were used instead of uncoated filters (Matrigel was a preparation with reduced growth factor content); (2) incubation times were much longer, that is, 13, 24, 48, 60, and 72鈥塰, because of the barrier function of Matrigel. *Indicates a highly significant difference with parental and mock controls (P猢?/span>0.001). (b) Effect of recombinant cystatin M and E64 on Matrigel invasion. Mock-1 cells were incubated as described above, and in the absence (Ctrl) or presence of 200鈥塶M of recombinant GST/GSTCM or 10鈥?i>渭 M of E64. Invasion was assessed after 72鈥塰. *Indicates a highly significant difference with mock control and GST-treated mock cells (P猢?/span>0.001)Full size imageCystatin M reduced tumor-endothelial cell adhesionThe ability of circulating tumor cells to adhere firmly to the microvascular endothelium at a distant site is considered a crucial step in establishing clinically successful metastases (Liotta and Stetler-Stevenson, 1991; Al-Mehdi et al., 2000). To test whether adhesion of MDA-MB-435S cells to endothelial cells was affected by cystatin M, we performed in vitro adhesion assays using HUVEC monolayers. Adhesion of MDA-MB-435S cells overexpressing cystatin M (Figure 6, CysM) to HUVECs was significantly decreased (P 0.05) compared to adhesion of mock-1 cells (Figure 6, Ctrl). Adhesion of the mock-1 cells could be inhibited to a similar extent (30%) as that of the CM-13 cells in the presence of 200鈥塶M of recombinant GSTCM (Figure 6, GSTCM). In contrast, 10鈥?i>渭 M of E64 did not inhibit the adhesion of mock-1 cells (Figure 6, E64). These results show that 鈥?as for tumor cell proliferation 鈥?tumor cell adhesion to endothelial cells in vitro does not seem to require the participation of lysosomal cysteine proteases. In contrast to tumor cell proliferation, however, adhesion of MDA-MB-435S cells to HUVECs was sensitive to GSTCM. Cystatin M has been shown to be a double-headed inhibitor capable of targeting both papain- and legumain-type lysosomal cysteine proteases (Alvarez-Fernandez et al., 1999). Further studies are required to determine whether or not a legumain-type cysteine protease(s) might be assisting tumor cells in their adhesion to endothelial cells.Figure 6Effect of cystatin M on tumor-endothelial cell adhesion. Mock-1 (Ctrl) and CM-13 (CysM) cells were labeled with 1-渭 M BCECF-AM, washed in HBSS, and added (3 脳 105 cells per well) to confluent monolayers of HUVECs grown in 24-well plates. Transfected clones were allowed to adhere for 15鈥塵in at 37掳C. Nonadherent cells were removed and endothelial monolayers washed three times with HBSS to remove loosely bound tumor cells. Adherent cells were then lysed and fluorescence intensity determined. Adhesion of mock-1 cells to endothelial cells was also studied in the presence of 200鈥塶M of recombinant GSTCM or 10鈥?i>渭 M of E64, and compared to the adhesion of untreated Mock-1 cells (Ctrl). Results were expressed as mean卤s.e.m. (n=8) and compared statistically using one-way ANOVA. Adhesion measurements were normalized to that of the mock clone to clearly identify the differences in tumor cell adhesionFull size imageDiscussionMultiple genetic and epigenetic alterations are involved in the development of a malignant tumor. Loss of expression of certain genes in tumor cells is often linked to the potential of their gene products to act as tumor or metastasis suppressors. Our overall working hypothesis is that cystatin M might be such a suppressor gene. Indeed, differential display of RNAs isolated from a matched pair of a primary human breast tumor and its metastatic lesion (Sager, 1997) identified cystatin M (CST6) (Sotiropoulou et al., 1997). Although cystatin M is expressed in normal and premalignant breast epithelial cell lines, its expression is often lost in breast cancer cell lines (Sotiropoulou et al., 1997). In this report, we analysed the mRNA expression pattern of cystatin M and cystatin C in a panel of human tumor cell lines of different degrees of differentiation and potential for tumor formation, tissue invasion, and metastasis. When compared to cystatin C, which is ubiquitously expressed (see also Keppler et al., 1994), cystatin M was found to be abundantly expressed by only three out of 15 tumor cell lines. These comprised one breast cancer (BT-20) and two prostate cancer cell lines (DU-145 and PC-3). Regulation of the steady-state levels of the two cystatin mRNAs seemed therefore to be quite different. Together with their quite different inhibition profiles towards lysosomal cysteine proteases (Abrahamson, 1994), this suggested that cystatins C and M might likely play different roles in tumor progression.In a recent study of a case of oropharyngeal squamous cell carcinoma, cystatin M mRNA and protein were found to be significantly increased in a metastatic lesion when compared to the primary tumor (Vigneswaran et al., 2003). Cystatin M immunostaining in the metastatic cervical lymph node lesion was restricted to a few focally keratinized areas. Therefore, downregulation of expression of cystatin M during progression of cancers may only be relevant for a few cancers such as breast cancer and melanoma. Alternatively, immunostaining of cystatin M in a subpopulation of cells that have metastasized suggests that the cystatin M gene might be re-expressed in oropharyngeal tumor cells that are undergoing differentiation at the secondary site. This would be in agreement with earlier studies showing that cystatin M is crosslinked to the stratum corneum during terminal differentiation of keratinocytes (Zeeuwen et al., 2001). Moreover, cystatin M expression was found to be increased 20-fold in human head and neck squamous carcinoma cells, upon induction of cell differentiation by the low calcemic vitamin D3 analog EB1089 (Lin et al., 2002). Therefore, further studies are needed to determine the extent of expression and downregulation/re-expression of cystatin M in normal and tumor tissues, respectively.To address the potential biological significance of the loss of expression of cystatin M in tumor cells, we constitutively expressed cystatin M in the highly malignant human breast cancer cell line MDA-MB-435S (Price and Zhang, 1990). The initial RT鈥揚CR screen had established that this cell line expressed little or no cystatin M. Ectopic expression of cystatin M in MDA-MB-435S cells reduced their in vitro growth rate, migration, invasion through Matrigel, and adhesion to an endothelial cell monolayer. Reduced proliferation of MDA-MB-435S cells was independent of the inhibition of lysosomal cysteine proteases because neither the broad-spectrum inhibitor E64 nor its membrane-permeant ester derivative E64d were able to mimic the effect of cystatin M expression. In addition, the bacterially expressed recombinant GSTCM fusion protein was unable to reduce the proliferation of MDA-MB-435S cells even at concentrations of 2.0鈥?i>渭 M. The fusion protein, however, was fully functional as an inhibitor of cysteine proteases, since it inhibited papain with a Ki (1.35鈥塶M) very similar to that of the native inhibitor (0.39鈥塶M) (Ni et al., 1997).The lack of effect of bacterially expressed cystatin M on cell proliferation suggested that this effect could depend on post-translational modifications of the protein. Indeed, several cystatins have been shown to undergo N- or O-glycosylation, Ser/Thr-phosphorylation, dimerization, or to bind calcium ions (Bell et al., 1989; Laber et al., 1989; Ekiel et al., 1997; Merz et al., 1997; Taupin et al., 2000). Further studies are required to determine which of these modifications occur in cystatin M and are important for its antiproliferative function. Others have shown that cystatin C is able to induce DNA synthesis and mitosis in various cell types including fibroblasts, mesangial cells, and neuronal stem cells (Sun, 1989; Leung-Tack et al., 1990; Tavera et al., 1992; Taupin et al., 2000). In rat neuronal stem cells, stimulation of cell proliferation by cystatin C was also shown to be independent of the inhibition of cysteine proteases (Taupin et al., 2000). Together with the present data, this suggests that cystatin C and cystatin M may have opposite effects on cell proliferation.The ability of cystatin M to inhibit migration and invasion of MDA-MB-435S cells through Matrigel was consistent with previous reports on the inhibition of migration and invasion by stefin A (cystatin A), cystatin C, and E-64 (Boike et al., 1992; Kobayashi et al., 1992; Redwood et al., 1992; Sexton and Cox, 1997; Kolkhorst et al., 1998; Coulibaly et al., 1999; Konduri et al., 2002). However, the extent of inhibition of invasion by cystatin M observed in the present study, that is, 猢?/span>95%, suggested that cystatin M may target some cysteine protease(s) that was rate-limiting in the proteolytic cascade, leading to matrix dissolution and invasion. The potential physiological target(s) of cystatin M could be one of the 11 papain-type cysteine proteases (Bromme and Kaleta, 2002; Mason et al., 2002; Puente et al., 2003), or it could be a legumain-type cysteine protease such as Asn-endopeptidase (Alvarez-Fernandez et al., 1999). With a Ki of 1.6鈥塸M, cystatin M indeed very efficiently inhibits Asn-endopeptidase (=mammalian legumain) (Alvarez-Fernandez et al., 1999). This latter highly selective protease has recently been shown to be able to activate the zymogen of matrix metalloproteinase-2 (proMMP-2 or progelatinase A) (Chen et al., 2001). Consistent with this view is also the fact that overexpression of Asn-endopeptidase in transformed human embryonal kidney HEK-293 cells leads to increased migration, invasion, and proMMP-2 activation (Liu et al., 2003). Preliminary studies from our laboratory using microarray and RT鈥揚CR analysis show that MDA-MB-435S cells did indeed express Asn-endopeptidase. However, Asn-endopeptidase is not sensitive to E64 even at concentrations of up to 2鈥塵M (Kembhavi et al., 1993; Chen et al., 1997). Therefore, if the rate-limiting factor targeted by cystatin M is indeed Asn-endopeptidase, it would suggest that cystatin M and E64 acted at two different levels within the same pathway. For example, Asn-endopeptidase might be required for the activation of proMMP-2, which in turn is required to loosen up the extracellular matrix, thereby stimulating endocytosis/phagocytosis and digestion of extracellular matrix components by lysosomal proteases (Keppler et al., 1996).Cathepsin B, a papain-type cysteine protease, participates in both pericellular and intracellular digestion of matrix proteins in breast cancer BT-20 and BT-549 cells, respectively (Sameni et al., 2000). Both E64 and the highly selective cathepsin B inhibitor CA074 block pericellular degradation of collagen-IV by BT-20 cells and intracellular degradation of collagen-IV by BT-549 cells by more than 50%. In addition, serine and matrix metalloproteinase inhibitors blocked pericellular degradation of collagen-IV by 鈭?/span>50%. This suggests that cathepsin B, similar to Asn-endopeptidase, may participate in a pericellular proteolytic cascade. Indeed, other studies show that cathepsins B and L are involved in the conversion of pro-uPA into active uPA (urinary-type plasminogen activator/urokinase), and that inhibition of invasion by E64 is due to interference with pro-uPA activation at the tumor cell surface (Goretzki et al., 1992; Kobayashi et al., 1992; Guo et al., 2002). Inhibitors might thus regulate cysteine proteases involved in the activation of both MMP-2 and uPA.Besides the detachment from the primary tumor, the dissolution of extracellular matrices, and the invasion of surrounding normal tissue, successful metastasis depends on a number of other critical steps. Metastatic tumor cells also need to survive the shear stress encountered during hematogenous dissemination, continuously escape immune surveillance, arrest at a distant site, and, finally, initiate cell divisions within a completely different microenvironment. Whereas most of these steps cannot be analysed in vitro, we have nevertheless attempted to study the role of cystatin M in the adhesion of MDA-MB-435S cells to vascular endothelial cells such as HUVECs. Ectopic expression of cystatin M was found to reduce adhesion to this endothelium by 30% when compared to control cells. Although HUVECs are by no means representative of microvascular endothelial cells at sites of secondary colonization, our data are the first to report on the inhibitory effect of a cystatin on tumor-endothelial cell adhesion. Unlike cell proliferation and cell migration/invasion that were either not sensitive to exogenous administration of E64 or GSTCM or responded strongly to both agents, respectively, tumor-endothelial cell adhesion showed a split response with E64 having no significant effect and GSTCM reducing adhesion to the same extent as ectopic expression of the inhibitor. Under physiological conditions of blood flow and shear stress, we can reasonably expect that the antiadhesive effect of ectopically expressed cystatin M will be much more pronounced.In summary, we have shown that ectopic expression of cystatin M in the highly proliferative, invasive, and metastatic MDA-MB-435S cells seriously compromised their in vitro malignant properties. Some of the effects of cystatin M, on cell proliferation in particular, were found to be unrelated to the inhibition of lysosomal cysteine proteases, and suggest a novel mechanism of action. Future studies are now aimed at elucidating this mechanism. In order to clearly establish that cystatin M has a bona fide tumor- and/or metastasis-suppressing function, it will be crucial to analyse its role in tumor formation, invasion, angiogenesis, and metastasis in vivo. Such in vivo studies in mice are currently in progress in our laboratory.Materials and methodsCell culture and transfection conditionsBT-20, BT-549, MDA-MB-435S, CaCo-2, HT-29, and U-87MG cells were purchased from ATCC and maintained according to their protocols. DU-145, PC-3, and LNCaP cells were a kind gift from Dr Michael Cher (Wayne State University, Detroit, MI, USA). DU-145 cells were maintained in DMEM+10% FBS. PC-3 and LNCaP cells were maintained in RPMI+5% FBS. All media were supplemented with 100鈥塙/ml of penicillin and 100鈥?i>渭g/ml of streptomycin (Sigma, St Louis, MO, USA). MCF-10A, a human diploid breast epithelial cell line, was obtained from the Barbara Ann Karmanos Cancer Institute Cell and Tissue Core (Wayne State University, Detroit, MI, USA). This cell line was maintained in DMEM/F12 containing 5% horse serum, 1鈥?i>渭 M insulin (Sigma, St Louis, MO, USA) and 20鈥塶g/ml EGF (BD Collaborative Research, Bedford, MD, USA). The human melanoma cell lines WM-164, WM-793, WM-852, WM-902B, WM-1205Lu, and WM-1341D were a generous gift from Dr Meenhard Herlyn (Wistar Institute, Philadelphia, PA, USA). They were grown in MCDB-153/L-15 (Sigma, St Louis, MO, USA) supplemented with 2% FBS, 5鈥?i>渭g/ml bovine insulin, and 1.68鈥塵M CaCl2. Human umbilical vein endothelial cells (HUVECs) were isolated and cultured as previously reported (Kevil et al., 1998). HUVECs were seeded onto fibronectin-coated 24-well tissue culture plates in EGM media, according to the recommendations from Clonetics Cell Systems (Cambrex, East Rutherford, NJ, USA), and cultured to confluency. All experiments were performed using passage 1 HUVECs.To eliminate a potential source of heterogeneity, the MDA-MB-435S cell 鈥榣ine鈥?was first cloned by limiting dilution and a clone representative of the overall parental phenotype (e.g., morphology and cystatin expression) isolated and used for transfection experiments. Transfections were performed by the liposome method using Lipofectamine Plus (Gibco, Rockville, MD, USA). Briefly, 2鈥?i>渭g of DNA was transfected into 8 脳 105 cells, according to the manufacturer\'s instructions. At 48鈥塰 after transfection, cells were split into two 100-mm dishes and the selection drug Zeocin (Invitrogen, Carlsbad, CA, USA) was added at 400鈥?i>渭g/ml. The media were changed every 48鈥塰. After 2 weeks, colonies appeared. In all, 24 colonies from the mock-transfected and cystatin M-transfected cells were expanded and maintained in DMEM supplemented with 10% FBS, antibiotics, and 200鈥?i>渭g/ml Zeocin.Semiquantitative RT鈥揚CR analysisRNA was isolated from human cell lines with Trizol (Gibco, Rockville, MD, USA) and reverse-transcribed. Briefly, 1鈥?i>渭g of DNase-treated total RNA was annealed with 0.5鈥?i>渭g oligo-dT15 and reverse-transcribed in a 20鈥?i>渭l volume containing 1 脳 reverse transcription buffer, 0.1鈥塵g/ml BSA, 40鈥塙 of RNasin (Promega, Madison, WI, USA), 1鈥塵M dNTPs, and 200鈥塙 of MMLV-reverse transcriptase (Promega, Madison, WI, USA) at 37掳C for 90鈥塵in. A 2鈥?i>渭l aliquot was used for subsequent PCR reactions using the Taq PCR Core kit from Qiagen (Valencia, CA, USA). For cystatin C (59掳C, 261-bp PCR product), cystatin M (61掳C, 422-bp PCR product), and 尾2-microglobulin (56掳C, 150-bp PCR product), the amplification conditions were as follows: 25 cycles of 94掳C for 30鈥塻, 59, 61 or 56掳C for 30鈥塻, 72掳C for 3鈥塵in, followed by a final 5-min extension at 72掳C. PCR products were resolved on a 2% agarose gel, stained with SYBR Green, imaged, and quantified using a STORM 840 system from Molecular Dynamics. The primer sequences were as follows: cystatin C forward, 5鈥?IndexTermAGGAGGGTGT GCGGCGTG-3鈥? cystatin C reverse, 5鈥?IndexTermGCCAAGGCAC AGCGTAGAT-3鈥? cystatin M forward, 5鈥?IndexTermTGGTCGCATT CTGCCTCCTG-3鈥? cystatin M reverse, 5鈥?IndexTermCTCGGGGACT TATCACATCT GC-3鈥? 尾2-microglobulin forward, 5鈥?IndexTermTTAGCTGTGC TCGCGCTACT CTCTC-3鈥? 尾2-microglobulin reverse, 5鈥?IndexTermGTCGGATGGA TGAAACCCAG ACACA-3鈥?Cystatin M expression plasmidsConstitutive mammalian expression vectorThe full-length human cystatin M cDNA was PCR-amplified with the following primers: forward, 5鈥?IndexTermttaGGTACCA TCATGGCGCG TTC-3鈥?and reverse, 5鈥?IndexTermtttGAATTCG GACTTATCAC ATCTGC-3鈥? The underlined sequences indicate restriction enzyme sites for KpnI and EcoRI (New England Biolabs, Beverly, MA, USA) in the forward and reverse primers, respectively. PCR was performed as follows: 94掳C for 3鈥塵in, then 30 cycles of 94掳C for 30鈥塻, 60掳C for 30鈥塻, and 72掳C for 1鈥塵in, followed by a final extension of 72掳C for 5鈥塵in to produce a 475-bp product. The PCR product was digested with KpnI and EcoRI at 37掳C for 1鈥塰, purified with a PCR purification kit (Qiagen, Valencia, CA, USA) and ligated into the mammalian expression vector pTracer-CMV2 (Invitrogen, Carlsbad, CA, USA). This vector uses the CMV promoter to drive expression of the inserted gene. Several bacterial colonies were picked and plasmids were isolated and sequenced to verify the integrity of the construct.Inducible bacterial expression vectorThe DNA sequence encoding full-length mature human cystatin M was amplified by PCR and subcloned into the IPTG-inducible bacterial expression vector pGEX2T (Amersham-Biosciences, Piscataway, NJ, USA), as described previously (Sotiropoulou et al., 1997). IPTG-driven expression from this plasmid will generate a glutathione-S-transferase 鈥?cystatin M fusion protein (GSTCM). Several bacterial colonies were picked and plasmids were isolated and sequenced to verify the integrity of the fusion construct.Production and purification of recombinant cystatin MThe empty vector (pGEX2T) and the fusion construct (pGEX2T/cystatin M) were transformed into Origami (Novagen, Madison, WI, USA), an E. coli strain that has been engineered to increase the likelihood of disulfide-bond formation and correct folding of recombinant proteins. Recombinant GST and GSTCM were expressed and purified as described earlier (Sotiropoulou et al., 1997). Correct folding of recombinant GSTCM was assessed by titration of papain activity on Z-Phe-Arg-AMC (Bachem, Torrance, CA, USA), as previously described (Keppler et al., 1997). A similar titration was performed with recombinant GST and served as a negative control. Before addition to cell cultures, purified recombinant GST and GSTCM were filter-sterilized by passage through 0.22-鈥?i>渭m pore-size Millex-GV low protein binding durapore PVDF membranes (Millipore Corp., Bedford, MA, USA).SDS鈥揚AGE and immunoblot analysisParental, mock-transfected, and cystatin M-transfected clones of MDA-MB-435S were seeded in 100-mm dishes. When 70鈥?0% confluent, cells were washed with serum-free media and incubated in serum-free media for 24鈥塰. Since cystatin M is a secreted protein (Ni et al., 1997; Sotiropoulou et al., 1997), conditioned media were collected and concentrated by TCA precipitation of total proteins. Proteins were resuspended in Laemmli buffer, the volume of which was adjusted to cell number. Samples normalized this way were electrophoresed on a 15% SDS鈥揚AGE gel, electroblotted onto nitrocellulose membranes, and probed with rabbit polyclonal antibodies raised against cystatin C (a kind gift from Dr Magnus Abrahamson, University of Lund, Sweden) (Abrahamson et al., 1986) and cystatin M (Sotiropoulou et al., 1997). The bound antibodies were visualized with an HRP-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), and enhanced chemiluminescence (Pierce, Rockford, IL, USA).Quantitation of cell growthA total of 500 cells of either mock-transfected or cystatin M-expressing clones were seeded in each of six wells in seven 96-well microtiter plates. Cells were fed every 48鈥塰. Plates were collected on days 1, 3, 5, 7, 9, 11, and 13. The media were removed and plates were stored at 鈭?0掳C until all plates were collected. Cell number was determined with the CyQuant system (Molecular Probes, Eugene, OR, USA) on a Tecan SpectraFluor Plus fluorescent plate reader, using an FITC-based filter set. Relative fluorescence units were converted to cell number using a calibration curve established with known cell numbers.In separate experiments, 500 cells of the mock-transfected clone were seeded in each of six wells in five 96-well microtiter plates containing either 24-h conditioned medium from mock-transfected cells or cystatin M-transfected cells. Cells were fed with the appropriate medium every 48鈥塰. Plates were collected on days 1, 3, 5, 7, and 9. Cell number was determined as described above. Experiments were repeated three times. The growth of mock-transfected cells was also studied in the presence and absence of various doses of recombinant GST and GSTCM, or E64 and its membrane-permeant ester-derivative E64d (Buttle et al., 1992).Migration and Matrigel invasion assaysTranswell culture inserts with their companion 24-well plates (BD Falcon, Bedford, MD, USA) were used for the assessment of cell migration and extracellular matrix invasion. The culture inserts consist of an 8-渭m pore-size PET membrane upon which cells can be seeded and grown. In all, 50鈥?00 cells in 500鈥?i>渭l of serum-free DMEM were seeded on top of the PET filters. The bottom wells received 750鈥?i>渭l of DMEM supplemented with 1% FBS as chemoattractant. Control wells received DMEM only (no chemoattractant). After various periods of incubation at 37掳C in the CO2 incubator, cells in the top well that had not migrated were wiped off the top surface of the PET membrane, using a cotton swab. Cells on the bottom surface of the filter, that is, cells that had migrated through the filter, were fixed in 2% paraformaldehyde and stained using Mayer\'s hematoxylin and a 1% solution of eosin Y (EMD Biosciences, Madison, WI, USA). Culture inserts were placed on top of a hemocytometer in a drop of immersion oil and migrated cells counted using an inverted microscope at 脳 100 magnification. Triplicate experiments were performed for each clone and each time point. Results are expressed as means卤s.e.m. and compared statistically using ANOVA. The migration of mock-transfected cells was also studied in the presence and absence of various doses of recombinant GST and GSTCM, or E64.For the Matrigel invasion studies, culture inserts with an 8-渭m pore-size PET membrane precoated with growth-factor-reduced Matrigel (BD Biocoat, Bedford, MD, USA) were used. Matrigel, a reconstituted basement membrane, blocks non-invasive cells from migrating through the membrane. In contrast, invasive cells such as MDA-MB-435S cells are able to degrade and invade into Matrigel and eventually cross the membrane through the 8-渭m pores (Hendrix et al., 1987). Matrigel was allowed to rehydrate for 1鈥塰 in serum-free medium at 37掳C in the CO2 incubator. Culture inserts were then transferred to a new 24-well plate and loaded with 50鈥?00 cells per well in serum-free medium. All other conditions were as described for the migration assays.Tumor-endothelial cell adhesion assayTumor cell adhesion to the vascular endothelium was performed as previously described for monocytes (Kevil et al., 2001). Briefly, mock-1 and CM-13 cells were labeled for 20鈥塵in with the fluorescent dye BCECF-AM (Molecular Probes, Eugene, OR, USA), at a final concentration of 1鈥?i>渭 M, in Hank\'s balanced salt solution (HBSS). After three washes in HBSS, tumor cell clones were added to HUVEC monolayers at 37掳C for 15鈥塵in. Nonadherent cells were removed and endothelial monolayers washed three times with HBSS to remove loosely bound cells. Adherent cells were then lysed with 1-mM NaOH and fluorescent intensity determined using a Tecan Genios Plus plate reader. Adhesion of mock-1 cells to HUVECs was also studied in the presence and absence of recombinant GSTCM (200鈥塶M) or E64 (10鈥?i>渭 M), and compared to the adhesion of CM-13 cells. Eight experiments (n=8) were performed for each clone and each treatment. Results were expressed as means卤s.e.m. and compared statistically using ANOVA. Adhesion measurements were normalized to that of the mock clone to clearly identify differences in tumor cell adhesion.Statistical analysesOne-way ANOVA analyses were performed for the comparison of transfection clones. Two-factor ANOVA analyses were performed for the comparison of mock-transfected clones grown in the presence or absence of various compounds or media. ReferencesAbrahamson M . (1994). Methods Enzymol., 244, 685鈥?00.Abrahamson M, Barrett AJ, Salvesen G and Grubb A . (1986). J. Biol. Chem., 261, 11282鈥?1289.Al-Mehdi AB, Tozawa K, Fisher AB, Shientag L, Lee A and Muschel RJ . (2000). Nat. Med., 6, 100鈥?02.Alvarez-Fernandez M, Barrett AJ, Gerhartz B, Dando PM, Ni J and Abrahamson M . (1999). J. Biol. Chem., 274, 19195鈥?9203.Barrett AJ, Rawlings ND, Davies ME, Machleidt W, Salvesen G and Turk V . (1986). Proteinase Inhibitors, Barrett AJ and Salvesen G (eds). Elsevier Science Publishers: Amsterdam. Google Scholar聽 Bell ET, Featherstone JD and Bell JE . (1989). Arch. Biochem. Biophys., 271, 359鈥?65.Boike G, Lah T, Sloane BF, Rozhin J, Honn K, Guirguis R, Stracke ML, Liotta LA and Schiffmann E . (1992). Melanoma Res., 1, 333鈥?40.Bromme D and Kaleta J . (2002). Curr. Pharm. Des., 8, 1639鈥?658.Bromme D, Okamoto K, Wang BB and Biroc S . (1996). J. Biol. Chem., 271, 2126鈥?132.Brunner N, Pyke C, Hansen CH, Romer J, Grondahl-Hansen J and Dano K . (1994). Cancer Treat Res., 71, 299鈥?09.Buck MR, Karustis DG, Day NA, Honn KV and Sloane BF . (1992). Biochem. J., 282, 273鈥?78.Buttle DJ, Murata M, Knight CG and Barrett AJ . (1992). Arch. Biochem. Biophys., 299, 377鈥?80.Chapman HA, Riese RJ and Shi GP . (1997). Annu. Rev. Physiol., 59, 63鈥?8.Chen JM, Dando PM, Rawlings ND, Brown MA, Young NE, Stevens RA, Hewitt E, Watts C and Barrett AJ . (1997). J. Biol. Chem., 272, 8090鈥?098.Chen JM, Fortunato M, Stevens RA and Barrett AJ . (2001). Biol. Chem., 382, 777鈥?83.Coulibaly S, Schwihla H, Abrahamson M, Albini A, Cerni C, Clark JL, Ng KM, Katunuma N, Schlappack O, Glossl J and Mach L . (1999). Int. J. Cancer, 83, 526鈥?31.DeClerck YA and Imren S . (1994). Eur. J. Cancer, 30A, 2170鈥?180.Ekiel I, Abrahamson M, Fulton DB, Lindahl P, Storer AC, Levadoux W, Lafrance M, Labelle S, Pomerleau Y, Groleau D, LeSauteur L and Gehring K . (1997). J. Mol. Biol., 271, 266鈥?77.Goretzki L, Schmitt M, Mann K, Calvete J, Chucholowski N, Kramer M, Gunzler WA, Janicke F and Graeff H . (1992). FEBS Lett., 297, 112鈥?18.Guinec N, Dalet-Fumeron V and Pagano M . (1993). Biol. Chem. Hoppe Seyler, 374, 1135鈥?146.Guo M, Mathieu PA, Linebaugh B, Sloane BF and Reiners Jr JJ . (2002). J. Biol. Chem., 277, 14829鈥?4837.Hendrix MJ, Seftor EA, Seftor RE and Fidler IJ . (1987). Cancer Lett., 38, 137鈥?47.Hulkower KI, Butler CC, Linebaugh BE, Klaus JL, Keppler D, Giranda VL and Sloane BF . (2000). Eur. J. Biochem., 267, 4165鈥?170.Kembhavi AA, Buttle DJ, Knight CG and Barrett AJ . (1993). Arch. Biochem. Biophys., 303, 208鈥?13.Keppler D, Sameni M, Moin K, Mikkelsen T, Diglio CA and Sloane BF . (1996). Biochem. Cell Biol., 74, 799鈥?10.Keppler D and Sloane BF . (1996). Enzyme Protein, 49, 94鈥?05.Keppler D, Sordat B and Sierra F . (1997). Mech. Ageing Dev., 98, 151鈥?65.Keppler D, Waridel P, Abrahamson M, Bachmann D, Berdoz J and Sordat B . (1994). Biochim. Biophys. Acta, 1226, 117鈥?25.Kevil CG, Patel RP and Bullard DC . (2001). Am. J. Physiol. Cell Physiol., 281, C1442鈥揅1447.Kevil CG, Payne DK, Mire E and Alexander JS . (1998). J. Biol. Chem., 273, 15099鈥?5103.Kobayashi H, Ohi H, Sugimura M, Shinohara H, Fujii T and Terao T . (1992). Cancer Res., 52, 3610鈥?614.Kolkhorst V, Sturzebecher J and Wiederanders B . (1998). J. Cancer Res. Clin. Oncol., 124, 598鈥?06.Konduri SD, Yanamandra N, Siddique K, Joseph A, Dinh DH, Olivero WC, Gujrati M, Kouraklis G, Swaroop A, Kyritsis AP and Rao JS . (2002). Oncogene, 21, 8705鈥?712.Laber B, Krieglstein K, Henschen A, Kos J, Turk V, Huber R and Bode W . (1989). FEBS Lett., 248, 162鈥?68.Leung-Tack J, Tavera C, Gensac MC, Martinez J and Colle A . (1990). Exp. Cell Res., 188, 16鈥?2.Lin R, Nagai Y, Sladek R, Bastien Y, Ho J, Petrecca K, Sotiropoulou G, Diamandis EP, Hudson TJ and White JH . (2002). Mol. Endocrinol., 16, 1243鈥?256.Linebaugh BE, Sameni M, Day NA, Sloane BF and Keppler D . (1999). Eur. J. Biochem., 264, 100鈥?09.Liotta LA and Stetler-Stevenson WG . (1991). Cancer Res., 51, 5054s鈥?059s.Liu C, Sun C, Huang H, Janda K and Edgington T . (2003). Cancer Res., 63, 2957鈥?964.Liu Z, Brattain MG and Appert H . (1997). Biochem. Biophys. Res. Commun., 231, 283鈥?89.Maciewicz RA, Wotton SF, Etherington DJ and Duance VC . (1990). FEBS Lett., 269, 189鈥?93.Mai J, Sameni M, Mikkelsen T and Sloane BF . (2002). Biol. Chem., 383, 1407鈥?413.Mason RW, Stabley DL, Picerno GN, Frenck J, Xing S, Bertenshaw GP and Sol-Church K . (2002). Biol. Chem., 383, 1113鈥?118.Merz GS, Benedikz E, Schwenk V, Johansen TE, Vogel LK, Rushbrook JI and Wisniewski HM . (1997). J. Cell. Physiol., 173, 423鈥?32.Ni J, Abrahamson M, Zhang M, Fernandez MA, Grubb A, Su J, Yu GL, Li Y, Parmelee D, Xing L, Coleman TA, Gentz S, Thotakura R, Nguyen N, Hesselberg M and Gentz R . (1997). J. Biol. Chem., 272, 10853鈥?0858.Poste G and Fidler IJ . (1980). Nature, 283, 139鈥?46.Price JE and Zhang RD . (1990). Cancer Metastasis Rev., 8, 285鈥?97.Puente XS, Sanchez LM, Overall CM and Lopez-Otin C . (2003). Nat. Rev. Genet., 4, 544鈥?58.Redwood SM, Liu BC, Weiss RE, Hodge DE and Droller MJ . (1992). Cancer, 69, 1212鈥?219.Roshy S, Sloane BF and Moin K . (2003). Cancer Metastasis Rev., 22, 271鈥?86.Sager R . (1997). Proc. Natl. Acad. Sci. USA, 94, 952鈥?55.Sameni M, Moin K and Sloane BF . (2000). Neoplasia, 2, 496鈥?04.Sexton PS and Cox JL . (1997). Melanoma Res., 7, 97鈥?01.Sotiropoulou G, Anisowicz A and Sager R . (1997). J. Biol. Chem., 272, 903鈥?10.Sun Q . (1989). Exp. Cell. Res., 180, 150鈥?60.Taupin P, Ray J, Fischer WH, Suhr ST, Haakansson K, Grubb A and Gage FH . (2000). Neuron, 28, 385鈥?97.Tavera C, Leung-Tack J, Prevot D, Gensac MC, Martinez J, Fulcrand P and Colle A . (1992). Biochem. Biophys. Res. Commun., 182, 1082鈥?088.Thompson EW, Paik S, Brunner N, Sommers CL, Zugmaier G, Clarke R, Shima TB, Torri J, Donahue S and Lippman ME . (1992). J. Cell. Physiol., 150, 534鈥?44.Turk V and Bode W . (1991). FEBS Lett., 285, 213鈥?19.Vigneswaran N, Wu J and Zacharias W . (2003). Oral Oncol., 39, 559鈥?68.Villadangos JA, Bryant RA, Deussing J, Driessen C, Lennon-Dumenil AM, Riese RJ, Roth W, Saftig P, Shi GP, Chapman HA, Peters C and Ploegh HL . (1999). Immunol. Rev., 172, 109鈥?20.Zeeuwen PL, Van Vlijmen-Willems IM, Egami H and Schalkwijk J . (2002). Br. J. Dermatol., 147, 87鈥?4.Zeeuwen PL, Van Vlijmen-Willems IM, Jansen BJ, Sotiropoulou G, Curfs JH, Meis JF, Janssen JJ, Van Ruissen F and Schalkwijk J . (2001). J. Invest. Dermatol., 116, 693鈥?01.Download referencesAcknowledgementsWe wish to acknowledge the generous gift of human melanoma cell lines from Dr Meenhard Herlyn (The Wistar Institute, Philadelphia, PA, USA) and of cystatin C antibodies from Dr Magnus Abrahamson (University of Lund, Lund, Sweden). Our warmest thanks are also due to Dr Avraham Raz, for his lively input throughout this work. Part of this study was supported by a Virtual Discovery Grant (DK) from the Barbara Ann Karmanos Cancer Institute and Research Grants CA91785 (DK) and CA36481 (BFS) from the National Cancer Institute.Author informationAffiliationsDepartment of Pharmacology, Wayne State University School of Medicine, Detroit, 48201, MI, USARavi Shridhar,聽Bonnie F Sloane聽 聽Daniel KepplerDepartment of Cellular Biology Anatomy, Louisiana State University Health Sciences Center, Shreveport, 71130, LA, USAJun Zhang,聽Jin Song聽 聽Daniel KepplerDepartment of Pathology, Louisiana State University Health Sciences Center, Shreveport, 71130, LA, USABlake A Booth聽 聽Christopher G KevilFeist-Weiller Cancer Center, Louisiana State University Health Sciences Center, Shreveport, 71130, LA, USAChristopher G Kevil聽 聽Daniel KepplerDepartment of Pharmacy, University of Patras, Rion, Patras, 26500, GreeceGeorgia SotiropoulouBarbara Ann Karmanos Cancer Institute, Wayne State University School of Medicine, Detroit, 48201, MI, USABonnie F Sloane聽 聽Daniel KepplerAuthorsRavi ShridharView author publicationsYou can also search for this author in PubMed聽Google ScholarJun ZhangView author publicationsYou can also search for this author in PubMed聽Google ScholarJin SongView author publicationsYou can also search for this author in PubMed聽Google ScholarBlake A BoothView author publicationsYou can also search for this author in PubMed聽Google ScholarChristopher G KevilView author publicationsYou can also search for this author in PubMed聽Google ScholarGeorgia SotiropoulouView author publicationsYou can also search for this author in PubMed聽Google ScholarBonnie F SloaneView author publicationsYou can also search for this author in PubMed聽Google ScholarDaniel KepplerView author publicationsYou can also search for this author in PubMed聽Google ScholarCorresponding authorCorrespondence to Daniel Keppler.Rights and permissionsReprints and PermissionsAbout this articleCite this articleShridhar, R., Zhang, J., Song, J. et al. Cystatin M suppresses the malignant phenotype of human MDA-MB-435S cells. Oncogene 23, 2206鈥?215 (2004). https://doi.org/10.1038/sj.onc.1207340Download citationReceived: 29 September 2003Revised: 05 November 2003Accepted: 07 November 2003Published: 15 December 2003Issue Date: 18 March 2004DOI: https://doi.org/10.1038/sj.onc.1207340Keywordscystatinscathepsinslysosomal proteasesprotease inhibitorstumor progression Nanhong Tang, Qun Xie, Xiaoqian Wang, Xiujin Li, Yanlin Chen, Xu Lin Jianyin Lin Archives of Pharmacal Research (2011) Jon J Briggs, Mads H Haugen, Harald T Johansen, Adam I Riker, Magnus Abrahamson, 脴ystein Fodstad, Gunhild M M忙landsmo Rigmor Solberg BMC Cancer (2010) Eunkyung Ko, Seong-Eun Park, Eun Yoon Cho, Yujin Kim, Jung-Ah Hwang, Yeon-Su Lee, Seok Jin Nam, Saik Bang, Joobae Park Duk-Hwan Kim Breast Cancer Research (2010) M R Morris, C Ricketts, D Gentle, M Abdulrahman, N Clarke, M Brown, T Kishida, M Yao, F Latif E R Maher Oncogene (2010)