Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Alterations in Gene Expression of Proprotein Convertases in Human Lung Cancer Have a Limited Number of Scenarios

Alterations in Gene Expression of Proprotein Convertases in Human Lung Cancer Have a Limited Number of Scenarios

  • Ilya V. Demidyuk, 
  • Andrey V. Shubin, 
  • Eugene V. Gasanov, 
  • Alexander M. Kurinov, 
  • Vladimir V. Demkin, 
  • Tatyana V. Vinogradova, 
  • Marina V. Zinovyeva, 
  • Alexander V. Sass, 
  • Irina B. Zborovskaya, 
  • Sergey V. Kostrov


Proprotein convertases (PCs) is a protein family which includes nine highly specific subtilisin-like serine endopeptidases in mammals. The system of PCs is involved in carcinogenesis and levels of PC mRNAs alter in cancer, which suggests expression status of PCs as a possible marker for cancer typing and prognosis. The goal of this work was to assess the information value of expression profiling of PC genes. Quantitative polymerase chain reaction was used for the first time to analyze mRNA levels of all PC genes as well as matrix metalloproteinase genes MMP2 and MMP14, which are substrates of PCs, in 30 matched pairs of samples of human lung cancer tumor and adjacent tissues without pathology. Significant changes in the expression of PCs have been revealed in tumor tissues: increased FURIN mRNA level (p<0.00005) and decreased mRNA levels of PCSK2 (p<0.007), PCSK5 (p<0.0002), PCSK7 (p<0.002), PCSK9 (p<0.00008), and MBTPS1 (p<0.00004) as well as a tendency to increase in the level of PCSK1 mRNA. Four distinct groups of samples have been identified by cluster analysis of the expression patterns of PC genes in tumor vs. normal tissue. Three of these groups covering 80% of samples feature a strong elevation in the expression of a single gene in cancer: FURIN, PCSK1, or PCSK6. Thus, the changes in the expression of PC genes have a limited number of scenarios, which may reflect different pathways of tumor development and cryptic features of tumors. This finding allows to consider the mRNAs of PC genes as potentially important tumor markers.


Proprotein convertases (PCs) is a protein family which includes nine highly specific subtilisin-like serine endopeptidases in mammals (reviewed in [1]). The key function of these enzymes is processing and/or activation of numerous proteins and peptides. Endogenous substrates of PCs include neuropeptides, peptide hormones, growth and differentiation factors, adhesion molecules, extracellular matrix proteins, receptors, enzymes, blood coagulation factors, and plasma proteins. In addition, pathogenic viruses and bacteria can use host PCs to switch on their proteins such as viral coat proteins or bacterial toxins. Since the activation of proproteins in the right time and place is clearly crucial for homeostasis, PCs are involved in the control of various physiological processes in health and disease. PCs as a processing system feature a combination of specificity and redundancy [2]: each protein of the group has unique structural and functional properties; at the same time, the properties of PCs overlap. The specificity and redundancy are observed not only at the levels of substrate specificity and cellular localization but also for the temporal/tissue profiles and, possibly, for the regulation mechanisms of gene expression. In this context, the identification of individual physiological properties and natural partners of enzymes of this group is not an easy matter, which can be properly solved only when PCs are considered as an integrated system.

Many substrates of PCs are associated with malignant diseases. For instance, the direct involvement in tumor progression and metastasis has been demonstrated for insulin-like growth factor 1 (IGF-1) and its receptor (IGF-1R), transforming growth factor β (TGF-β), vascular endothelial growth factor C (VEGF-C), and matrix metalloproteinases (MMPs) (reviewed in [3]). By activating the key cancer-associated proteins, PCs have an effect on cell proliferation, motility, and adhesion as well as tumor invasion, which suggest PCs as promising therapeutic targets [4].

The first data on the association of PCs with cancer were published in 1987 [5]. Since then, numerous studies analyzed the expression of PCs in cancer and the correlations between the PC expression levels and cancer properties using various experimental approaches [6][21]. Overall, the data obtained demonstrated altered levels of PC mRNAs in cancer. Correlations between PC expression profiles and cancer aggressiveness [9], [12], [16], survival rate [18], and neuroendocrine differentiation of cancer cells [6], [7], [13] were shown. This allows us to propose the expression status of the PC system as a possible marker for cancer typing and prognosis.

Lung cancer is the most widespread oncological disease, which causes 1.4 million deaths annually [22]. Not surprisingly, PC expression data were first obtained for this cancer type [5][7]. These as well as more recent publications [8], [11][13] demonstrated altered expression of FURIN, PCSK1, PCSK2, and PCSK6 genes in lung cancer. (Hereafter, gene symbols follow the recommendations of the HUGO Gene Nomenclature Committee, The corresponding common protein designations are given in Table 1.) High FURIN expression was found in non-small cell lung carcinomas (NSCLCs) vs. small cell lung carcinomas (SCLCs) [5], [8] and correlated with the aggressiveness of lung cancer cell lines [12]. PCSK1 and PCSK2 expression is largely detected in cancers with neuroendocrine features, in particular, SCLC [6][8], [11], [13]. At the same time, PCSK6 expression is not necessarily observed in lung cancer, although it is more common in NSCLC than in SCLC [8]. Thus, the response of the PC system varies with lung cancer types. Gene expression surveys involving microarray analysis (including whole-transcriptome ones) demonstrate a substantial heterogeneity of lung cancer samples [13], [23][26], and the revealed differences correlate with the patients' survival rate [23], [24], [26]. Overall, this proposes lung cancer as a test system to assess the information value of the approach based on expression profiling of PC genes.

Table 1. Gene designations and TaqMan Gene Expression Assays used in real-time PCR.

In this context, the present work for the first time evaluated mRNA levels of all PC genes in lung cancer using reverse transcription followed by a quantitative real-time polymerase chain reaction (qPCR). In addition, we studied the gene expression of two matrix metalloproteinases (MMP2 and MMP14), which are key factors in cancer invasion and metastasis [27] and substrates of PCs [28][30].

Materials and Methods

Ethics Statement

The research was approved by the Institutional Review Board of Blokhin Cancer Research Center (Moscow, Russia), and written informed consent was obtained from each patient involved in the study.

Collection of tissue samples

Specimens of cancer tumor tissues and of adjacent tissues without histological pathology (further referred to as normal tissue) were taken from 30 patients with diagnosed small cell lung carcinoma and non-small cell lung carcinoma (tumor stage I–III) during surgery (Figure 1, Table S1). In every case localization of a primary tumor node was determined. If a tumor originated from the smallest bronchi in peripheral segments of lung and had no connection with bronchi lumen, its localization was considered peripheral. If a tumor originated from a large bronchus, its localization was considered central. The normal tissue specimens were taken from the edge of resections (the distance between tumor and normal tissues was no less than 20 mm). All patients were under medical supervision in the Blokhin Cancer Research Center (Moscow, Russia) during a period from May 2004 to November 2005. None of these patients received radio or chemical therapy to the moment of the investigation.

Figure 1. Description of the specimens studied and heat map presentation of the ratio values of gene expression in tumor vs. adjacent tissues without histological pathology.

SCC, squamous cell lung carcinoma; AdC, adenocarcinoma; AdC/SCC, both AdC and SCC cells were found in the tumor tissue; SCLC, small cell lung carcinoma; P, peripheral tumor location; C, central tumor location; Y, tumor with keratinization; N, tumor without keratinization; ‘-’, no data. The heat map is shown in log2 scale. Gray cells indicate specimens with undetectable mRNA in both tumor and normal tissues.

Each specimen was split into two portions. The first one was immediately frozen in liquid nitrogen for mRNA isolation. The second portion was used for histological examination after hematoxylin and eosin staining of the paraffin sections. Tumor tissue specimens contained more than 70% of malignant cells. In normal tissue specimens, no malignant cells were found. For SCC samples presence of keratinazation was determined. Existence of keratinization allowed to refer a sample to a group of well-differentiated cancer.

RNA isolation and purification

The total RNA was isolated from homogenized tumor or normal tissues by guanidine isothiocyanate lysis and acid-phenol extraction with subsequent removal of polysaccharide admixtures [31]. Additional purification was performed by RNA precipitation using an RNeasy Mini kit (Qiagen, USA). Further treatment with DNase I (Promega, USA) was done according to the supplier's recommendations. The obtained RNA samples were characterized electrophoretically in 1% agarose gel. RNA concentration was determined by spectrophotometry.

Double-stranded cDNA synthesis

Oligonucleotides AAGCAGTGGTATCAACGCAGAGTACGCrGrGrG and AAGCAGTGGTATCAACGCAGAGTACT30VN (V = C, or G, or A) (Syntol, Russia) were used in the reverse transcription reaction. For the first strand cDNA synthesis, 1 µg of isolated RNA was incubated with reverse transcriptase PowerScript (Clontech, USA) as described by Y. Zhu et al. [32]. The obtained reaction mixture was used for the second strand synthesis followed by PCR using the Advantage 2 DNA polymerase (Clontech, USA) and primer AAGCAGTGGTATCAACGCAGAGT under the following conditions: 95°C for 1.5 min; up to 17 cycles of 95°C for 20 s; 65°C for 20 s; and 72°C for 3 min. To obtain equal amounts of all amplification products, the number of cycles varied (commonly, 15 cycles).

Real-time PCR

Real-time PCR was performed using the primers and probes of the TaqMan Gene Expression Assays system (Applied Biosystems, USA) (Table 1). TaqMan Pre-Developed Assay Reagent GAPDH 20× (Applied Biosystems, USA) was used to quantify the reference gene, glyceraldehyde 3-phosphate dehydrogenase (GAPDH). PCR was conducted using a Chromo4 Dyad Disciple cycler (BioRad, USA) according to the supplier's recommendations with the following program: 50°C for 2 min; 95°C for 10 min; 45 cycles of 95°C for 15 s and 60°C for 60 s; the reaction volume was 20 µl. Every sample was tested at least twice in duplicates. The threshold cycle was defined using the Opticon Monitor 3 software (BioRad, USA).

Experimental data processing

The experimental data obtained for genes under study were normalized to GAPDH mRNA levels using the formula: , and the results were averaged (Table S1). The values for tumor and normal tissues were designated as ExprT and ExprN, respectively. The ExprT to ExprN ratio (RatioT/N) and 95% confidence intervals were calculated for each gene.

In some samples, mRNAs of certain genes were not detected in tumor or normal tissues in one of two independent experiments. In these cases, the Expr and Ratio values were calculated from the data of the other experiment. In some samples, real-time PCR failed to detect mRNAs of certain genes in tumor or normal tissues in both experiments; in these cases the CT was set equal to 42 in the RatioT/N calculations. If mRNAs were undetectable in tumor and normal tissues in both experiments, the RatioT/N values were not calculated.

Statistical analyses

Wilcoxon matched-pairs rank-sum test was used to evaluate the significance of difference between the mRNA levels of genes in tumor and normal tissues. Kruskal-Wallis one-way analysis of variance was performed to evaluate the influence of tumor type, stage, and TNM characteristics on mRNA levels of the studied genes. Spearman rank correlation coefficients were calculated to evaluate the relationship between pairs of gene expression profiles. Cluster analyses of gene expression patterns and expression profiles were performed for the Expr and RatioT/N values by the Ward method using Spearman rank correlation coefficients as the distance measure. All above statistical analyses were performed using the Statistica 8.0 software (StatSoft, USA). Heat maps were built using the Matrix2png software tool [33].

Results and Discussion

In this work, quantitative PCR was used for the first time to analyze mRNA levels of all PC genes (listed in Table 1) as well as matrix metalloproteinase genes MMP2 and MMP14 in 30 matched pairs of samples of human lung cancer tumor and adjacent normal tissues (Table S1, Figure 1). Expression of MBTPS1, PCSK7, MMP2, and MMP14 was observed in all tumor and normal tissue samples; and PCSK5, in almost all samples. Conversely, PCSK4 mRNA was detected in two tumor samples only. The expression profiles of other genes were more complex. PCSK9 transcript was found in 29 normal tissue samples but only in 18 tumor ones. PCSK6 expression was detected in about two thirds of normal and tumor tissues; and PCSK2, in 15 and 11, respectively. The most pronounced differences between normal and tumor tissue expression was observed for FURIN and PCSK1. The samples where mRNAs of these genes were detected were twice more frequent in tumor than in normal tissues, while their expression was undetectable in a substantial fraction of both tumor (21/30 for PCSK1 and 8/30 for FURIN) and normal samples (26/30 and 20/30, respectively). Overall, these findings demonstrate significant differences in the expression of individual PC genes in the human lung, on the one hand, and high variation in their expression patterns (i.e., combinations of expression levels of the genes) between individuals, on the other hand.

The data obtained are, in general, in good agreement with published results. PCSK4 was shown to be largely limited to testicular and ovarian germ cells [34][36]. PCSK1 and PCSK2, the principal activators of prohormones and proneuropeptides within the regulated secretory pathway, are largely detected in neural and endocrine cells [37]. All other genes studied (encoding both PCs and MMPs) are commonly reported as ubiquitously or widely expressed. As well as other authors, we found mRNA of MBTPS1, PCSK5, MMP2, and MMP14 in all or practically all the samples of tumor and normal lung tissues (e.g. [38], [39], datasets GDS1650, GDS1673, and GDS2491 in the Gene Expression Omnibus database at In case of PCSK6, in full conformity to other published data ([8], GDS1650, GDS1673, and GDS2491), we found the mRNA not in all, but major portion of the samples analyzed. Our results concerning PCSK9 are also in agreement with the published data (GDS1673), although the current information about expression of this gene in lung is scarce. We found FURIN mRNA in approximately half of all samples, which is in agreement with data obtained by microarray technology (GDS1650, GDS1673, and GDS2491), but in contradiction to published data acquired by Nothern blot analysis [5], [8]. The reason of this discrepancy may be explained by features of the methods used. The largest inconsistency concerns PCSK7. There are few data about its expression in lung. The data presented so far are obtained with microarray technology and do not match to each other. We found PCSK7 mRNA in all our samples, Gruber with coworkers found it in 14 out of 40 of not-diseased lung samples ([40] and GDS1673), and Stearman with colleagues did not find it in any of 20 tumor and 19 normal lung tissue samples ([41] and GDS1650). It does not seem possible to explain reasons of the distinctions, but it seems most probably to be due to differences in the experimental platforms used: qPCR and various generations of Affymetrix chips.

Thus, the obtained and published data demonstrate a high variation in PC gene expression between individuals. This gives no grounds to expect that mRNA levels of PCs in tumor or normal tissue alone can have any prognostic value or can be used for cancer typing. Indeed, no significant differences between the expression levels of the genes analyzed or their expression patterns have been revealed for groups of normal or cancer samples with similar clinical features. Likewise, cluster analysis failed to reveal groups of samples with similar gene expression patterns.

At the same time, we have found moderate but significant (p<0.05) pairwise correlations between the expression profiles of the genes studied (Table 2). Note that the sets of correlated profiles substantially differed for tumor and normal tissues. Assuming that the revealed correlations indicate the coordinated regulation of gene expression, one can propose that the regulation of expression of PCs and MMPs is modified in lung cancer. It is important to note, this applies to the great majority of PC genes. Moreover, the correlations between the profiles of changes in the expression in tumor vs. normal tissues (differential expression profiles) can be attributed to the mechanisms underlying the expression changes common for several genes.

Table 2. Correlations between expression profiles of the genes studied.

Analysis of mRNA levels of the studied genes demonstrated significant differences between tumor and normal tissues: the average level of FURIN mRNA increased (p<0.00005); mRNA levels of PCSK2 (p<0.007), PCSK5 (p<0.0002), PCSK7 (p<0.002), PCSK9 (p<0.00008), and MBTPS1 (p<0.00004) decreased; while PCSK1 mRNA level showed a tendency to increase (Figure 1). Thus, the expression of seven out of eight PC genes (except PCSK6), whose mRNA is detectable in the lung, demonstrated unidirectional changes in lung cancer in our samples. Although the expression of PCs was analyzed in many publications, this is an original finding, since mRNA levels of most PC genes in tumor vs. normal tissues have not been quantified previously. At the same time, the high level of FURIN expression in NSCLC [5], [8] and other cancer types [10], [16], [18] has been reported previously.

The role of MMPs in cancer progression as regulators of the tumor microenvironment is currently receiving much attention (reviewed in [42], [43]). In this study we analyzed expression of two MMP genes of different types: secreted MMP2 и membrane-anchored MMP14. These proteases are the major MMPs involved in cancer cell invasion and proliferation, tumor angiogenesis and vasculogenesis, cell adhesion and migration as well as in immune surveillance. Taking this into account, the absence of significant differences between MMP2 and MMP14 expression levels in tumor and adjacent tissues without histological pathology can look surprising. However, this result is in agreement with ample evidence for high levels of their expression both in cancer and stromal cells in NSCLC [38], [39], [44][52]. At the same time, a direct comparison of MMP2 and MMP14 expression in cancer tumor vs. adjacent normal tissues was reported only in two publications, and their conclusions are at variance. The former, similar to our study, observed significant differences for neither MMP2 nor MMP14 [46]. The latter publication demonstrated an elevated expression of MMP14 in cancer relative to normal lung specimens [39]. Most likely, this discrepancy is due to different specimen types analyzed: squamous cell carcinomas (SCCs) prevailed in the former study [46] and in our work, while more adenocarcinomas (AdCs) were analyzed in the latter report [39].

In our view, the most prominent result was obtained by comparing the patterns of changes in the expression of PC genes between tumor and normal tissues. Cluster analysis divided studied samples into four groups (Figure 2), which did not correlate with the available clinical features of tumors. Three of these groups (C1, C2, and C3) cover 80% of samples. Each group is rather homogeneous and has a single key gene: FURIN in C1, PCSK1 in C2, and PCSK6 in C3. Usually, the key gene expression is elevated in cancer; although, it can be unaltered or slightly decreased against the background of a substantial decrease in mRNA levels of other PCs (Figure 1). Undetectable PCSK6 expression in most samples is an extra character of C1, while C3 features undetectable expression of PCSK1 and/or FURIN in more than a half of samples. C4 is more heterogeneous. The samples in this group share similar expression changes of PCSK5, PCSK7, PCSK9, and MBTPS1 as well as undetectable mRNAs of FURIN and PCSK1 in most cases. Thus, the changes in the expression of PC genes in lung cancer have a limited number of scenarios, which may correspond to previously undetected NSCLC types.

Figure 2. Clustering of PC genes' expression data.

Specimens numbering corresponds to Fig. 1. The RatioT/N values in the heat map were row-normalized and shown in linear scale. Gray cells indicate specimens with undetectable mRNA in both normal and cancer tissues. Branch length reflects the distance between the dendrogram nodes. The clusters found are marked as C1, C2, C3 and C4.

It is of interest that the enzymes encoded by the key genes of the revealed groups belong to different types of PCs [53]. Furin, PCSK1, and PCSK6 have different kinds of C-terminal extensions. These PCs have different expression profiles: PCSK1 is localized in neural and endocrine cells, PCSK6 occurs widely, and furin is ubiquitous. Finally, they exhibit different secretion patterns: PCSK1 follows the regulated secretory pathway, while furin and PCSK6 are constitutively secreted. In this context, one can propose that different scenarios of alteration in the expression of PC genes induce different changes in the range of activated substrates.

The data available to date can provide only hints about the origin of the revealed groups. For instance, C2 with active PCSK1 can correspond to NSCLCs with signs of neuroendocrine differentiation [54][65], which can point to the origin of these tumors. The formation of C1 (FURIN) and C3 (PCSK6) can be mediated by the E2F1 transcription factor, which specifically upregulates PCSK6 but not FURIN or PCSK5 [66]. However, these data provide no reliable explanation of the mechanisms underlying the typical scenarios of changes in the transcription of PCs in lung cancer. Still, at least two radically different considerations can be advanced. On the one hand, the observed effects can stem from the differences that existed before malignant transformation, e.g., genotype differences between individuals or cell type differences within an individual. On the other hand, it can be due to the alterations emerged during cancer formation, in particular, local disorders in the expression of individual PC genes or varieties of the global dysregulation of gene expression. Moreover, the observed events can result from the effect of several factors at the same time.

Overall, analysis of the patterns of changes in the expression of PC genes in individuals allowed us to reveal several NSCLC types and to demonstrate that the expression changes have a limited number of scenarios, which may reflect different pathways of tumor development and cryptic features of tumors. This finding warrants further investigation, and allows to consider the mRNAs of PC genes as potentially important tumor markers.

Supporting Information

Table S1.

Characteristics of specimens and gene expression data.



We acknowledge Prof. Evgeny D. Sverdlov for invaluable contribution to the conceiving of the study and for critical reading of the manuscript.

Author Contributions

Conceived and designed the experiments: IVD VVD TVV IBZ SVK. Performed the experiments: AMK MVZ A.V. Sass IBZ. Analyzed the data: IVD A.V. Shubin EVG VVD SVK. Contributed reagents/materials/analysis tools: VVD A.V. Sass IBZ. Wrote the paper: IVD A.V. Shubin EVG VVD TVV SVK.


  1. 1. Seidah NG (2011) What lies ahead for the proprotein convertases? Ann N Y Acad Sci 1220: 149–161.
  2. 2. Creemers JW, Khatib AM (2008) Knock-out mouse models of proprotein convertases: unique functions or redundancy? Front Biosci 13: 4960–4971.
  3. 3. Bassi DE, Fu J, Lopez de Cicco R, Klein-Szanto AJ (2005) Proprotein convertases: “master switches” in the regulation of tumor growth and progression. Mol Carcinog 44: 151–161.
  4. 4. Chretien M, Seidah NG, Basak A, Mbikay M (2008) Proprotein convertases as therapeutic targets. Expert Opin Ther Targets 12: 1289–1300.
  5. 5. Schalken JA, Roebroek AJ, Oomen PP, Wagenaar SS, Debruyne FM, et al. (1987) fur gene expression as a discriminating marker for small cell and nonsmall cell lung carcinomas. J Clin Invest 80: 1545–1549.
  6. 6. Creemers JW, Roebroek AJ, Van de Ven WJ (1992) Expression in human lung tumor cells of the proprotein processing enzyme PC1/PC3. Cloning and primary sequence of a 5 kb cDNA. FEBS Lett 300: 82–88.
  7. 7. Scopsi L, Gullo M, Rilke F, Martin S, Steiner DF (1995) Proprotein convertases (PC1/PC3 and PC2) in normal and neoplastic human tissues: their use as markers of neuroendocrine differentiation. J Clin Endocrinol Metab 80: 294–301.
  8. 8. Mbikay M, Sirois F, Yao J, Seidah NG, Chretien M (1997) Comparative analysis of expression of the proprotein convertases furin, PACE4, PC1 and PC2 in human lung tumours. Br J Cancer 75: 1509–1514.
  9. 9. Hubbard FC, Goodrow TL, Liu SC, Brilliant MH, Basset P, et al. (1997) Expression of PACE4 in chemically induced carcinomas is associated with spindle cell tumor conversion and increased invasive ability. Cancer Res 57: 5226–5231.
  10. 10. Cheng M, Watson PH, Paterson JA, Seidah N, Chretien M, et al. (1997) Pro-protein convertase gene expression in human breast cancer. Int J Cancer 71: 966–971.
  11. 11. Rounseville MP, Davis TP (2000) Prohormone convertase and autocrine growth factor mRNAs are coexpressed in small cell lung carcinoma. J Mol Endocrinol 25: 121–128.
  12. 12. Bassi DE, Mahloogi H, Al-Saleem L, Lopez De Cicco R, Ridge JA, et al. (2001) Elevated furin expression in aggressive human head and neck tumors and tumor cell lines. Mol Carcinog 31: 224–232.
  13. 13. Bhattacharjee A, Richards WG, Staunton J, Li C, Monti S, et al. (2001) Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses. Proc Natl Acad Sci U S A 98: 13790–13795.
  14. 14. Matei D, Graeber TG, Baldwin RL, Karlan BY, Rao J, et al. (2002) Gene expression in epithelial ovarian carcinoma. Oncogene 21: 6289–6298.
  15. 15. Fu Y, Campbell EJ, Shepherd TG, Nachtigal MW (2003) Epigenetic regulation of proprotein convertase PACE4 gene expression in human ovarian cancer cells. Mol Cancer Res 1: 569–576.
  16. 16. Lopez de Cicco R, Watson JC, Bassi DE, Litwin S, Klein-Szanto AJ (2004) Simultaneous expression of furin and vascular endothelial growth factor in human oral tongue squamous cell carcinoma progression. Clin Cancer Res 10: 4480–4488.
  17. 17. Tzimas GN, Chevet E, Jenna S, Nguyen DT, Khatib AM, et al. (2005) Abnormal expression and processing of the proprotein convertases PC1 and PC2 in human colorectal liver metastases. BMC Cancer 5: 149.
  18. 18. Page RE, Klein-Szanto AJ, Litwin S, Nicolas E, Al-Jumaily R, et al. (2007) Increased expression of the pro-protein convertase furin predicts decreased survival in ovarian cancer. Cell Oncol 29: 289–299.
  19. 19. Shibru D, Hwang J, Khanafshar E, Duh QY, Clark OH, et al. (2008) Does the 3-gene diagnostic assay accurately distinguish benign from malignant thyroid neoplasms? Cancer 113: 930–935.
  20. 20. Estilo CL, O-charoenrat P, Talbot S, Socci ND, Carlson DL, et al. (2009) Oral tongue cancer gene expression profiling: Identification of novel potential prognosticators by oligonucleotide microarray analysis. BMC Cancer 9: 11.
  21. 21. D'Anjou F, Routhier S, Perreault JP, Latil A, Bonnel D, et al. (2011) Molecular validation of PACE4 as a target in prostate cancer. Transl Oncol 4: 157–172.
  22. 22. (2009) WHO Fact Sheet N 297.
  23. 23. Beer DG, Kardia SL, Huang CC, Giordano TJ, Levin AM, et al. (2002) Gene-expression profiles predict survival of patients with lung adenocarcinoma. Nat Med 8: 816–824.
  24. 24. Raponi M, Zhang Y, Yu J, Chen G, Lee G, et al. (2006) Gene expression signatures for predicting prognosis of squamous cell and adenocarcinomas of the lung. Cancer Res 66: 7466–7472.
  25. 25. Kuner R, Muley T, Meister M, Ruschhaupt M, Buness A, et al. (2009) Global gene expression analysis reveals specific patterns of cell junctions in non-small cell lung cancer subtypes. Lung Cancer 63: 32–38.
  26. 26. Valk K, Vooder T, Kolde R, Reintam MA, Petzold C, et al. (2010) Gene expression profiles of non-small cell lung cancer: survival prediction and new biomarkers. Oncology 79: 283–292.
  27. 27. Sato H, Takino T (2010) Coordinate action of membrane-type matrix metalloproteinase-1 (MT1-MMP) and MMP-2 enhances pericellular proteolysis and invasion. Cancer Sci 101: 843–847.
  28. 28. Sato H, Kinoshita T, Takino T, Nakayama K, Seiki M (1996) Activation of a recombinant membrane type 1-matrix metalloproteinase (MT1-MMP) by furin and its interaction with tissue inhibitor of metalloproteinases (TIMP)-2. FEBS Lett 393: 101–104.
  29. 29. Remacle AG, Rozanov DV, Fugere M, Day R, Strongin AY (2006) Furin regulates the intracellular activation and the uptake rate of cell surface-associated MT1-MMP. Oncogene 25: 5648–5655.
  30. 30. Koo BH, Kim HH, Park MY, Jeon OH, Kim DS (2009) Membrane type-1 matrix metalloprotease-independent activation of pro-matrix metalloprotease-2 by proprotein convertases. FEBS J 276: 6271–6284.
  31. 31. Chomczynski P, Mackey K (1995) Modification of the TRI reagent procedure for isolation of RNA from polysaccharide- and proteoglycan-rich sources. Biotechniques 19: 942–945.
  32. 32. Zhu YY, Machleder EM, Chenchik A, Li R, Siebert PD (2001) Reverse transcriptase template switching: a SMART approach for full-length cDNA library construction. Biotechniques 30: 892–897.
  33. 33. Pavlidis P, Noble WS (2003) Matrix2png: a utility for visualizing matrix data. Bioinformatics 19: 295–296.
  34. 34. Seidah NG, Day R, Hamelin J, Gaspar A, Collard MW, et al. (1992) Testicular expression of PC4 in the rat: molecular diversity of a novel germ cell-specific Kex2/subtilisin-like proprotein convertase. Mol Endocrinol 6: 1559–1570.
  35. 35. Torii S, Yamagishi T, Murakami K, Nakayama K (1993) Localization of Kex2-like processing endoproteases, furin and PC4, within mouse testis by in situ hybridization. FEBS Lett 316: 12–16.
  36. 36. Tadros H, Chretien M, Mbikay M (2001) The testicular germ-cell protease PC4 is also expressed in macrophage-like cells of the ovary. J Reprod Immunol 49: 133–152.
  37. 37. Seidah NG, Gaspar L, Mion P, Marcinkiewicz M, Mbikay M, et al. (1990) cDNA sequence of two distinct pituitary proteins homologous to Kex2 and furin gene products: tissue-specific mRNAs encoding candidates for pro-hormone processing proteinases. DNA Cell Biol 9: 415–424.
  38. 38. Yamamura T, Nakanishi K, Hiroi S, Kumaki F, Sato H, et al. (2002) Expression of membrane-type-1-matrix metalloproteinase and metalloproteinase-2 in nonsmall cell lung carcinomas. Lung Cancer 35: 249–255.
  39. 39. Atkinson JM, Pennington CJ, Martin SW, Anikin VA, Mearns AJ, et al. (2007) Membrane type matrix metalloproteinases (MMPs) show differential expression in non-small cell lung cancer (NSCLC) compared to normal lung: correlation of MMP-14 mRNA expression and proteolytic activity. Eur J Cancer 43: 1764–1771.
  40. 40. Gruber MP, Coldren CD, Woolum MD, Cosgrove GP, Zeng C, et al. (2006) Human lung project: evaluating variance of gene expression in the human lung. Am J Respir Cell Mol Biol 35: 65–71.
  41. 41. Stearman RS, Dwyer-Nield L, Zerbe L, Blaine SA, Chan Z, et al. (2005) Analysis of orthologous gene expression between human pulmonary adenocarcinoma and a carcinogen-induced murine model. Am J Pathol 167: 1763–1775.
  42. 42. Kessenbrock K, Plaks V, Werb Z (2010) Matrix metalloproteinases: regulators of the tumor microenvironment. Cell 141: 52–67.
  43. 43. Gialeli C, Theocharis AD, Karamanos NK (2011) Roles of matrix metalloproteinases in cancer progression and their pharmacological targeting. FEBS J 278: 16–27.
  44. 44. Nakagawa H, Yagihashi S (1994) Expression of type IV collagen and its degrading enzymes in squamous cell carcinoma of lung. Jpn J Cancer Res 85: 934–938.
  45. 45. Polette M, Nawrocki B, Gilles C, Sato H, Seiki M, et al. (1996) MT-MMP expression and localisation in human lung and breast cancers. Virchows Arch 428: 29–35.
  46. 46. Nawrocki B, Polette M, Marchand V, Monteau M, Gillery P, et al. (1997) Expression of matrix metalloproteinases and their inhibitors in human bronchopulmonary carcinomas: quantificative and morphological analyses. Int J Cancer 72: 556–564.
  47. 47. Thomas P, Khokha R, Shepherd FA, Feld R, Tsao MS (2000) Differential expression of matrix metalloproteinases and their inhibitors in non-small cell lung cancer. J Pathol 190: 150–156.
  48. 48. Cox G, Jones JL, Andi A, Waller DA, O'Byrne KJ (2001) A biological staging model for operable non-small cell lung cancer. Thorax 56: 561–566.
  49. 49. Ishikawa S, Takenaka K, Yanagihara K, Miyahara R, Kawano Y, et al. (2004) Matrix metalloproteinase-2 status in stromal fibroblasts, not in tumor cells, is a significant prognostic factor in non-small-cell lung cancer. Clin Cancer Res 10: 6579–6585.
  50. 50. Hoikkala S, Paakko P, Soini Y, Makitaro R, Kinnula V, et al. (2006) Tissue MMP-2 and MMP-9 are better prognostic factors than serum MMP-2/TIMP-2–complex or TIMP-1 in stage I-III lung carcinoma. Cancer Lett 236: 125–132.
  51. 51. Leinonen T, Pirinen R, Bohm J, Johansson R, Kosma VM (2008) Increased expression of matrix metalloproteinase-2 (MMP-2) predicts tumour recurrence and unfavourable outcome in non-small cell lung cancer. Histol Histopathol 23: 693–700.
  52. 52. Eren B, Sar M, Oz B, Dincbas FH (2008) MMP-2, TIMP-2 and CD44v6 expression in non-small-cell lung carcinomas. Ann Acad Med Singapore 37: 32–39.
  53. 53. Seidah NG, Mayer G, Zaid A, Rousselet E, Nassoury N, et al. (2008) The activation and physiological functions of the proprotein convertases. Int J Biochem Cell Biol 40: 1111–1125.
  54. 54. Linnoila RI, Mulshine JL, Steinberg SM, Funa K, Matthews MJ, et al. (1988) Neuroendocrine differentiation in endocrine and nonendocrine lung carcinomas. Am J Clin Pathol 90: 641–652.
  55. 55. Berendsen HH, de Leij L, Poppema S, Postmus PE, Boes A, et al. (1989) Clinical characterization of non-small-cell lung cancer tumors showing neuroendocrine differentiation features. J Clin Oncol 7: 1614–1620.
  56. 56. Skov BG, Sorensen JB, Hirsch FR, Larsson LI, Hansen HH (1991) Prognostic impact of histologic demonstration of chromogranin A and neuron specific enolase in pulmonary adenocarcinoma. Ann Oncol 2: 355–360.
  57. 57. Carles J, Rosell R, Ariza A, Pellicer I, Sanchez JJ, et al. (1993) Neuroendocrine differentiation as a prognostic factor in non-small cell lung cancer. Lung Cancer 10: 209–219.
  58. 58. Linnoila RI, Piantadosi S, Ruckdeschel JC (1994) Impact of neuroendocrine differentiation in non-small cell lung cancer. The LCSG experience. Chest 106: 367S–371S.
  59. 59. Pujol JL, Boher JM, Grenier J, Quantin X (2001) Cyfra 21-1, neuron specific enolase and prognosis of non-small cell lung cancer: prospective study in 621 patients. Lung Cancer 31: 221–231.
  60. 60. Hiroshima K, Iyoda A, Shibuya K, Toyozaki T, Haga Y, et al. (2002) Prognostic significance of neuroendocrine differentiation in adenocarcinoma of the lung. Ann Thorac Surg 73: 1732–1735.
  61. 61. Pelosi G, Pasini F, Sonzogni A, Maffini F, Maisonneuve P, et al. (2003) Prognostic implications of neuroendocrine differentiation and hormone production in patients with Stage I nonsmall cell lung carcinoma. Cancer 97: 2487–2497.
  62. 62. Howe MC, Chapman A, Kerr K, Dougal M, Anderson H, et al. (2005) Neuroendocrine differentiation in non-small cell lung cancer and its relation to prognosis and therapy. Histopathology 46: 195–201.
  63. 63. Ionescu DN, Treaba D, Gilks CB, Leung S, Renouf D, et al. (2007) Nonsmall cell lung carcinoma with neuroendocrine differentiation–an entity of no clinical or prognostic significance. Am J Surg Pathol 31: 26–32.
  64. 64. Segawa Y, Takata S, Fujii M, Oze I, Fujiwara Y, et al. (2009) Immunohistochemical detection of neuroendocrine differentiation in non-small-cell lung cancer and its clinical implications. J Cancer Res Clin Oncol 135: 1055–1059.
  65. 65. Sterlacci W, Fiegl M, Hilbe W, Auberger J, Mikuz G, et al. (2009) Clinical relevance of neuroendocrine differentiation in non-small cell lung cancer assessed by immunohistochemistry: a retrospective study on 405 surgically resected cases. Virchows Arch 455: 125–132.
  66. 66. Yuasa K, Suzue K, Nagahama M, Matsuda Y, Tsuji A (2007) Transcriptional regulation of subtilisin-like proprotein convertase PACE4 by E2F: possible role of E2F-mediated upregulation of PACE4 in tumor progression. Gene 402: 103–110.